Western Plainchant
Lu&iJL       iImMWI tot uf 0 mm*
' -   - •      t'       f f
t
« - 7 r
3      ' ' '*
'f
*/.r:: •
London, British Library, Harley 4951, fo. 298", showing the start of the section for mode 3 in the tonary (cf. Ex. 111.14.1-2). By permission of the British Library.
Western Plainchant
A Handbook
DAVID HILEY
CLARENDON PRESS • OXFORD
1993
Oxford University Press, Walton Street, Oxford 0x2 6np Oxford Xew York Toronto Delhi Bombay Calcutta Madras Karachi Kuala Lumpur Singapore Hong Kong Tokyo Xairobi Dar es Salaam Cape Town Melbourne Auckland Madrid and associated companies in Berlin Ibadan
Oxford is a trade mark of Oxford University Press
Published in the United States by Oxford i'nirersity Press Inc., Xew York
V David Hiley 1993
First published 1993 Reprinted 1993
All rights reserved. Xo part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press. Within the UK, exceptions are allowed in respect of any fair dealing for the purpose of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, or in the case of reprographic reproduction in accordance with the terms of the licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside these terms and in other countries should be sent to the Rights Department, Oxford [.tuversity Press, at the address above
British Library Cataloguing in Publication Data Data available 1SBX 0 19 816289 8
Library of Congress Cataloging in Publication Data Western plainchant: a handbook / David Hiley. Includes bibliographical references and index. 1. Chants (Plain, Gregorian, etc.)-- History and criticism. I. Title.
ML3082.115-1   1992   182.3T22   dc20   92 13020 ISBX 0 19 816289 8
Typeset by Hope Services (Abingdon) Ltd. Printed in Great Britain on acid-free paper by St. Edmundsbury Press, Bury St. Edmunds
for Marjorie and Henry
PREFACE
I have been up to London to get the book I am writing, out of the British Museum. I have got a lot of it out, and I shall go again presently to get some more; and when I have got it all, there will be another book ... So many people were there, getting out their books. It doesn't seem to matter everything's being in books already: I don't mind it at all. There are attendants there on purpose to bring it to you. That is how books are made, and it is difficult to think of any other way. I mean the kind called serious . . .
No doubt the author of any reference book such as the present one could echo the sentiments of Miss Charity Marcon, in Ivy Compton-Burnett's Daughters and Sons. A great deal of my book is indeed got out of others, as the bibliography and references in the text make clear. What justification is there for this, and what is the purpose of the book?
In the first place I wished to provide a book of reference both for those coming new to plainchant and for those needing guidance in the specialist literature. The book starts with the assumption, reasonable in this secular age, that many things about the liturgy and its plainchant, even quite basic matters, are unfamiliar to the reader. At every stage in the encounter with plainchant one comes up against specialist terminology and concepts which constitute a real obstacle. That is in the nature of the subject, for ecclesiastical ritual is essentially exclusive, remote from everyday experience, reserved for specially trained personnel. To start with essentials does not mean, however, that difficulties have been avoided. I have not, I hope, confused inexperience with lack of intelligence. The reader will encounter here many complex problems, both those for which scholars have found solutions and others which remain obscure. I have also illustrated techniques of research and given examples, not just of the music in plainchant sources, but also of their make-up, the way they deploy their material, and their notation.
Those with access to well-equipped libraries will find here sufficient references to further specialist literature. But I have also tried to make the book self-explanatory, and well enough illustrated, so that it will be useful also to anyone interested in and able to read music.
Such a book fulfils a need primarily because of the enormous expansion of plainchant studies in the last few decades. The last major work of synthesis in the English language, Apel's Gregorian Chant, is now over thirty years old. It has not, of course, been my intention to try and replace Apel, let alone the more comprehensive Einführung in die gregorianischen Melodien by Peter Wagner from before the First World War, both of which remain indispensable. But much has been explored and discovered since their day. The best modern survey, the article 'Plainchant' by
viii Preface
Kenneth Levy and John Emerson in The New Grove Dictionary, is necessarily brief (but with an excellent bibliography). The writings of scholars such as the late Bruno Stäblein—his Schriftbild der einstimmigen Musik and articles in Die Musik in Geschichte und Gegenwart—together practically constitute a textbook on chant. But an up-to-date one-volume work is clearly required.
The compression and omissions entailed in such a work are the least pleasant things facing an author. For my own undoubted sins of omission I have tried to make amends in the bibliography, by citing literature to which I could not do justice in the main text. The bibliography should go some way towards being a reference tool in itself. In writing the book I tried to take into account literature up to 1990. I take this opportunity to mention the recent appearance of a new chant bibliography with over 4,200 items by Thomas Kohlhase and Günther Michael Paucker, in Beiträge zur Gregorianik, 9-10 (1990). To one work which appeared when the main text was all but complete, but whose contents I knew intimately, I have made no reference. I acted as co-editor with Richard Crocker of the new edition of The New Oxford History of Music, ii: The Early Middle Ages to 1300 (Oxford, 1989), and it was not always easy to avoid unconscious borrowing from it. At any rate, that volume now takes its place beside Wagner, Apel, and Stäblein's Schriftbild as an essential part of the chant scholar's library.
Historical writing about plainchant is a relatively young phenemonon. Prince-Abbot Martin Gerbert's De cantu et musica sacra a prima ecclesiae aetate usque ad praesens tempus of 1774, the ancestor of all musicological writing on plainchant, was motivated by the desire to stimulate reforms in the church music of his own time. Knowledge of the past would make men conscious of the malpractices of the present: 'I desire ... to lay open to view the astounding abuses in these matters, which in my opinion are the gravest in the discipline of our church' (quoted in p. lviii* of Othmar Wessely's introduction to the 1968 reprint of Gerbert's work). The work of the great restorers of the late nineteenth and early twentieth centuries was imbued with the spirit of reform, to be achieved, as with Gerbert, through recovery of a former ideal state. Work which is of incalculable value for musicologists sprang from these reforming efforts: the series of facsimiles Paleographie musicale, the volumes of Im Graduel romain, the studies of notational practices made by Dom Eugene Cardine and his pupils (I cite here only a few examples from the work of monks of the French Benedictine monastery of Solesmes). However, the fact that more than musicology was involved inevitably affected the selection of information and the use to which it was put. This type of counterpoint between scholars and their material is of course common to all scholarship. In the case of chant studies the consequences of different viewpoint and purposes can be appreciated by comparing Cardine's 'Vue d'ensemble sur le chant gregorien' and Levy and Emerson's 'Plainchant'.
The researches of the Benedictine fathers also had a definite practical purpose: the provision of model, definitive chant-books for use in the Roman Church. Consequently, their work was concerned with those chants sung in the Roman liturgy of their own time, but not with parts of the repertory, such as sequences and tropes, which were no
Preface ix
longer admitted, though important in the Middle Ages. In recent years mostly lay scholars have devoted considerable energy to those repertories, and it is perhaps not fanciful to see this partly as a reaction against the bias of earlier chant studies. In my own case, it led among other things to a decision to transcribe all musical examples from original sources, rather than rely on modern service-books, slight though the differences between them may sometimes be.
A great deal of the satisfaction gained from writing a book such as this comes from sorting out problems in one's own mind, for, as the renowned author of a monograph on Lassus succinctly expressed it: 'Nothing clears up a case so much as stating it to another person.' Yet even more satisfying has been the ever-increasing wonderment at the variety and richness of the chant repertory, a delight to anyone working in the area. I hope that the musical examples, at least, will help further the appreciation of these treasures.
Like Miss Marcon, I owe a considerable debt to the British Library, but also to the Anselm Hughes Library of Royal Holloway and Bedford New College, University of London, where I once taught; to the University of London Library at Senate House, and more recently to the Universitätsbibliothek of Regensburg University, where I now teach. Many of my musical examples were transcribed from microfilms, work with which has in many cases been made possible by grants from the Central Research Fund of the University of London. My principal debt is to the numerous fellow chant scholars from whom I have learned over the years, many of whom could have written a better book than this. It is the fate of textbooks to be used for target-practice, as it were, by better-informed teachers, but I hope my colleagues will find here something they can set before their students with reasonable confidence. Above all, I hope the book will straighten the path of those who, like myself a couple of decades ago, are trying to find their way in an initially foreign, often bewildering, but always fascinating and rewarding musical world.
I am most grateful to the libraries which kindly supplied photographs for the plates. Plates 1, 4-6, 8-9, 11-14, and 18 are published by permission of the British Library; plates 2, 7, and 16 by permission of the Bodleian Library; plate 10 by permission of the Syndics of Cambridge University Library; plate 15 by permission of the Master and Fellows of University College, Oxford; and plate 17 by permission of Edinburgh University Library.
In conclusion I wish to thank those who have played an especially important part in the production of the book: Malcolm Gerratt, who launched the project and fanned it along for several years; Bruce Phillips, who brought it into the safe haven of Oxford University Press, and the staff of OUP, especially Leofranc Holford-Strevens and Bonnie Blackburn, who did far more for the book than an author has a right to expect from his copy-editor. Greatest of all is my debt to my wife Ann, without whose patience and encouragement I should never have reached the end of the long voyage.
D.H.
CONTENTS
Analytical Table xii
List of Illustrations xxi
List of Tables xxii
List of Music Examples xxiv
Note on Music Examples xxx
Abbreviations xxxi
Bibliography xxxiii
I. Plainchant in the Liturgy 1
II. Chant Genres 46
III. Liturgical Books and Plainchant Sources 287
IV. Notation 340 Plates 403
V. Plainchant and Early Music Theory 442
VI. Plainchant up to the Eighth Century 478
VII. The Carolingian Century 514
VIII. Gregorian Chant and Other Chant Repertories 524
IX. Persons and Places 563
X. Reformations of Gregorian Chant 608
XL The Restoration of Medieval Chant 622
Index of Tex t and Music Incipits 631
Index of Manuscript and Printed Sources 638
Index of Names and Terms 643
ANALYTICAL TABLE
I. PLAINCHANT IN THE LITURGY
1.1. Introduction 1
1.2. Liturgy and Worship 2
1.3. The Church Year 7
1.4. The Daily Round 16
1.5. Mass 22
1.6. Office 25
(i) The Night Office (Matins, Vigils, or Nocturns) in Secular Use 25
(ii) The Night Office in Monastic Use 27
(iii) Lauds in Secular Use 28
(iv) Lauds in Monastic Use 28
(v) Vespers in Secular Use 28
(vi) Vespers in Monastic Use 29
(vii) Compline 29
(viii) Prime, Terce, Sext, and None 30
1.7. Processions 30
1.8. Ceremonies of Holy Week 32
(i) General 33
(ii) Palm Sunday 33
(iii) Maundy Thursday 34
(iv) Good Friday 36
(v) Holy Saturday or Easter Eve 38
(vi) Easter Sunday 39
1.9. The 'Feast of Fools' and Related Customs 39
1.10. Other Services: Baptism, Confirmation, Ordination, Coronation,
Marriage, Burial, Dedication 42
II. CHANT GENRES
>' ILL   Introduction 46
11.2. Recitation Formulas for Prayers and Lessons 47
(i) General 47
(ii) Prayers 49
(iii) Lessons 54
11.3. Tones for Psalms and Other Chants 58
(i) Psalms 58
(ii) Other Psalm Tones: The Parapteres, Tonus peregrinus 62
(iii) Tones for the Canticles Magnificat and Benedictus 64
(iv) Tones for the Psalm Verses of Introits and Communions 64
(v) Tones for Responsory Verses 65
Analytical Table xiii
(vi) Tones for the Invitatory Psalm 66 (vii) Benedictus es domine Dens pairiim nostmm in the Saturday Mass
of the Ember Weeks 68
(via) Te Deum laudamus 68
11.4. The Great Responsories of the Night Office 69
(i) Introduction 69
(ii) Repertory, Texts, and Form 69
(iii) Music 71
(iv) Centonization 74
(v) Melismas 76
11.5. Graduals and Tracts 76
(i) Introduction 76
(ii) Graduals in a: The 'Iustus ut palma' Group 77
(iii) Graduals in F 80
(iv) Other Graduals 81
(v) Tracts 82
(vi) Tracts in Mode 2 82
(vii) Tracts in Mode 8 83
11.6. Short Responsories 85
11.7. Antiphons 88
(i) Introduction 88
(ii) Ferial or Psalter Antiphons 91
(iii) Antiphons for the Psalms of Vespers, the Night Office, and Lauds 92
(iv) Antiphons for the Magnificat and Benedictus 96
(v) The Great O-Antiphons 98
11.8. Invitatory Antiphons 99
11.9. Processional Antiphons 99
(i) Introduction 100
(ii) Rogation Antiphons 101
(iii) Palm Sunday Antiphons 102
(iv) Antiphons for Other Occasions 102
11.10. Marian Antiphons 104
11.11. Introits 109
(i) Introduction 109
(ii) Introits in Mode 3 110
(iii) Comparison with Office Antiphons and Responsories 114
11.12. Communions 116
(i) General 116
(ii) Groups of Communions with Psalm and Gospel Texts 117
(iii) Some F-mode Communions 118
(iv) Communions and Responsories 120
11.13. Offertories 121
(i) Introduction 121
(ii) Texts 122
(iii) The Melodies of the Offertory Respond 122
(iv) Verse Melodies 126
xiv
Analytical Table
11.14. Alleluias 130
(i) Introduction 130
(ii) The Earlier and Later Styles 132
(iii) Rhymed Alleluias and Late Medieval Melodies 136
11.15. Hymns 140
(i) Introduction 140
(ii) Texts 141
(iii) Music 142
(iv) Processional Hymns 146
11.16. Chants for the Ordinary of Mass 148
11.17. Kyrie eleison 150
(i) Kyrie eleison as a Litany 150
(ii) Kyrie eleison after the Introit at Mass 151 (Hi) Early Melodies 152
(iv) Italian Melodies 154
(v) Melodic Types 155 K(v'i) Later Melodies 155
11.18. Gloria in excelsis Deo 156
(i) Introduction 156
(ii) Recitation Types 157
(iii) Through-composed Melodies 159
11.19. Sanctus 161
(i) The Oldest Melodies 161
(ii) Other Melodies 162
11.20. Agnus Dei 165
11.21. Credo 168
11.22. Sequences 172
(i) Introduction 172
(ii) Early Sequences with Parallel-Verse Structure 173
(iii) Notation: Performance; Partially Texted Melodies 180
(iv) Short Aparallel Sequences 181
(v) Italian Sequences 183
(vi) The Early History of the Sequence 185
(vii) Rhymed Sequences 189
11.23. Tropes 196
(i) Introduction 196
(ii) Added Melismas in Introits 197
(iii) Added Melismas in Glorias 199
(iv) Added Responsory Melismas 200
(v) Prosulas for Offertories and Alleluias 201
(vi) Responsory Prosulas 204
(vii) Other Prosulas 209
(viii) Kyries with Latin Text, Kyrie Prosulas, and Kyrie Tropes 211
(ix) Benedicamus Chants with Extended Text, Prosulas, and Tropes 213
(x) Introit, Offertory, and Communion Tropes; Sequence Tropes 215
(xi) Gloria, Sanctus, and Agnus Tropes 223
(xii) Farsed Lessons, Creeds, and Paternoster 233
Analytical Table
xv
V 11.24. Latin liturgical songs 238
(i) Introduction 238 (li) Versus in Early Aquitanian Manuscripts; versus with 'Double
Cursus' 239
(iii) Twelfth-century Songs: Textual and Musical Style 241
(iv) Twelfth-century Songs: Liturgical Function 248
11.25. Liturgical Dramas 250
(i) Liturgy and Drama 251
(ii) The Quern queritis Dialogue 252
(iii) Easter Ceremonies from the Eleventh Century Onward 255
(iv) Christmas and Epiphany Ceremonies; Rachel's and Mary's Laments 263
(v) Rhymed Ceremonies; the Fleury Playbook; the Ludus Danielis 266
11.26. Offices with Verse Texts 273
11.27. Metre, Accent, Rhythm, and Rhyme in Liturgical Texts 279
(i) Metre and Stress 280
(ii) Rhyme 284
(iii) Prose Rhythm (Cursus) and Prose Rhyme 285
III. LITURGICAL BOOKS AND PLAINCHANT SOURCES
111.1. Introduction 287
111.2. Ordines Romani 289
111.3. Sacramentaries and Lectionaries 291
(i) Sacramentaries 291
(ii) Lectionaries 293
111.4. Graduals (Mass Antiphoners) and Cantatoria 295
(i) Introduction 295
(ii) Graduals without Notation 296
(iii) Notated Graduals 299
111.5. Antiphoners (Office Antiphoners) 303
(i) The Earliest Antiphoners 304
(ii) Types of Manuscript and Calendric Organization 305 (iii) Comparison of Sources 307
111.6. Psalters, Hymnals, Collectars, Office Lectionaries 308
(i) Psalters 308
(ii) Hymnals 310
(iii) Collectars 311
(iv) Office lectionaries 313
111.7. Sequentiaries, Tropers, and Kyriales 313 II 1.8. Processionals 317 III.9. Missals 319 III. 10. Breviaries 320
111.11. Compendia 321
111.12. Pontificals and Rituals 324
(i) Pontificals and Benedictionals 324
(ii) Rituals, Manuals, or Agenda 324
111.13. Ordinals and Customaries 325
xvi
Analytical Table
III. 14. Tonanes 325
(i) Definition and Function 326
(ii) Type-Melodies for the Eight Modes 331
(iii) How Psalm Tones were Specified 333
III. 15. Identifying and Describing Chant-Books 335
IV. NOTATION
IV. 1.   Introduction 340
(i) Preliminary 340
(ii) The Signs'in Montpellier H. 159 341
(iii) Neume 345 IV.2.   Regional Styles 346
(i) French and German Notation 347
(ii) Palaeofrankish, Laon, Breton, and Aquitanian Notations 347 (in) Types Related to French-German Notation 351
(iv) Other Italian Notations 352
(v) Examples 353 IV.3.   Liquescence, Oriscus, Quilisma; Other Special Signs 357
(i) Signs for Liquescence 357
(ii) Quilisma 358
(iii) Oriscus 359
(iv) Virga strata, Pressus, Pes stratus, Pes quassus, Salicus 360 IV.4.   The Origins of Chant Notation 361
(i) Introduction 362
(ii) Early Examples 362
(iii) Early References to Notation 364
(iv) Parallel Systems 364
(a) Prosodic Accents 365
(b) Punctuation 366
(c) Ekphonetic Notation 367
(d) Byzantine Notation 369
(v) The 'Cheironomic' Theory 370
(vi) The Early Transmission of Chant 370
(vii) Some Conclusions 371 IV.5.   The Notation of Rhythm 373
(i) Rhythmic Elements in Early Notations 374
(ii) Rhythm in Simple Antiphons 380
(iii) Cardine's 'Gregorian Semiology' 382
(iv) The Evidence of Theorists 384
(v) Conclusions 385 IV.6.   Pitch-Notation 386
(i) From the Ninth Century to William of Dijon 386
(ii) Guido of Arezzo 388
(iii) Staff-Notation in Different Lands 389 IV.7. Theorists' Notations 392 IV.8.   Printed Chant-Books 395
Analytical Table xvii
IV.9.   Modern Transcription 398
(i) The Liturgical Context 399
(ii) Transcription 400 (iii) Transcription from Staffless Notations 400
PLATES
Frontispiece ii
Plate  1.   London, British Library, Harley 1117 406
Plate 2.   Oxford, Bodleian Library, Lat. liturg. d. 3 408
Plate  3.    London, British Library, Egerton 857 410
Plate 4.   London, British Library, Harley 110 412
Plate 5.   London, British Library, Add. 30850 414
Plate 6.   London, British Library, Add. 19768 416
Plate 7.   Oxford, Bodleian Library, Canonici liturg. 350 418
Plate 8.   London, British Library, Arundel 156 420
Plate 9.   London, British Library, Royal 8. C. xiii 422
Plate 10.   Cambridge, University Library, LI. 2. 10 424
Plate 11.   London, British Library, Add. 10335 426
Plate 12.    London, British Library, Egerton 3511 428
Plate 13.   London, British Library, Add. 29988 430
Plate 14.   London, British Library, Add. 17302 432
Plate 15.   Oxford, University College 148 434
Plate 16.   Oxford, Bodleian Library, Lat. liturg. a. 4 436
Plate 17.   Edinburgh, University Library 33 438
Plate 18.   London, British Library, printed book IB. 8668 440
V. PLAINCIIANT AND EARLY MUSIC THEORY
V.l.   Introduction 442
V.2.   The Legacy of Antiquity 443
V.3.   A Pitch System for Plainchant 447
(i) Introduction 447
(ii) Hucbald of Saint-Amand 448
(iii) The Enchiriadis Group of Treatises 452 V.4.   The Modes 454
(i) General 454
(ii) Aurelian of Reome 456
(iii) Regino of Prüm 458
(iv) Nomenclature 459
(v) Greek Names and Octave Species 461
(vi) Italian Theory: The Dialogus de musica 463
(vii) Guido of Arezzo 466
(viii) South German Writers 470 (ix) Modal Theory in South Germany 472
'' (x) Later Syntheses    - 476
xv111
Analytical Table
VI. PLAINCHANT UP TO THE EIGHTH CENTURY
VI. 1.   Introduction 478
VI.2.   The Churches of Christendom 480
VI.3.   The Early Church 484
VI. 4. Office Chants before the Eighth Century 487 VI.5.   Mass Chants before the Eighth Century 494
(i) The Gradual                            ' 495
(ii) The Introit and Other Chants at the Start of Mass 496
(iii) The Chants at Communion 498
(iv) The Offertory 499
(v) The Chants beside the Lessons 500 VI.6.   Gregory the Great 503
(i) Introduction 503
(ii) Gregory and the Deacons 504
(iii) Gregory and the Alleluia 504
(iv) Isidore of Seville; the Liber pontifical'is; the List of Chant 'Editors'
in Ordo Romanus XIX 505
(v) The Anglo-Saxon Tradition 506
(vi) The Biographies by Paul Warnefrid and John Hymmonides 508
(vii) The 'Gregorian' Sacramentary 509
(viii) The Prologue Gregorius praesul 510 X'(ix) The Reception of the Gregory Legend in the Ninth Century and
Later; the Dove; the Modes 511
(x) Gregory and the Lenten Communions 511
(xi) Conclusions 513
: VII. THE CAROLINGIAN CENTURY
VII. 1. Introduction 514 VI1.2. The Establishment of Roman Chant in Francia 514 VI1.3. The Frankish Expansion of the Chant Repertory 518 VII.4. The Codification of Plainchant * 520 VII.5.   The Place of Music in the Carolingian Renaissance 521
VIII. GREGORIAN CHANT AND OTHER CHANT REPERTORIES
VIII.l.   Introduction 524
VIII.2.   The Influence of Byzantium 525
(i) Introduction 525
(ii) The System of Eight Modes 527
(iii) Antiphons for the Adoration of the Cross and Other Chants in Old
Italian Repertories 527
(iv) The Trisagion 528
(v) The Frankish 'missa graeca' 528
(vi) The Communion Omnes qui in Christo 529
(vii) The Byzantine Alleluias 529
Analytical Table xix
(viii) Textual Concordances without Musical Similarity 529
(ix) The Veterern hominem Antiphons for the Octave of Epiphany 530
VIII.3.   Old Italian Chant I: Rome 530
(i) Introduction 530
(ii) Sources and Studies of Old Roman Chant 531
(iii) Examples: Communions 532
(iv) Graduals 533
(v) Antiphons 533
(vi) Offertories 534
(vii) Alleluias 538
(viii) Oral Tradition 539 VIII.4. Old Italian Chant II: Milan 540 VIII.5. Old Italian Chant III: Benevento 549 VIII.6. Gallican Chant 552 VI11.7. Old Spanish (Mozarabic) Chant 557 VIII.8.   Conclusions 560
IX. PERSONS AND PLACES
IX. 1.   Introduction 563
IX.2.   Amalarius of Metz 569
IX.3.   St Gall 571
IX.4.   Cluny 574
IX.5.   William of Dijon 578
IX.6.   England before and after the Norman Conquest 580
IX.7.   North Italian Traditions 585
IX.8.   Benevento and Montecassino 589
IX.9.   Rome and the Franciscans 594
IX. 10. Aquitaine and Saint-Martial at Limoges 596
IX. 11. The Hispanic Peninsula after the Reconquest 601
IX. 12. Northern and Eastern Europe 604
(i) Scandinavia and Iceland 604
(ii) Poland 605
(iii) Czechoslovakia 606
(iv) Hungary 607
,    X. REFORMATIONS OF GREGORIAN CHANT
X.l.   Introduction 608
X.2.   The Cistercians 609
X.3.   The Dominicans 611
X.4.   Chant in Other Religious Orders 613
(i) The Carthusians 613
(ii) The Premonstratensians 614 X.5. The 'Medicean' Gradual (1614-15) 615 X.6.   Neo-Galhcan Chant 618
xx
Analytical Table
XI. THE RESTORATION OF MEDIEVAL CHANT
XI. 1.   The Return to the Source 622
XI.2.   Solesmes and the Vatican Edition 624
XI.3.   Practical Editions and Scholarly Research 628
ILLUSTRATIONS
Plates
For list	sec above, p.xvii.	
Maps		
IV.2.1.	Centres where Laon notation was used	350
IX.1.1.	The archdioceses of late medieval Europe	565
IX.1.2.	Ecclesiastical map of Italy	566
IX.1.3.	Ecclesiastical map of Germany	567
IX.1.4.	Ecclesiastical map of France	568
IX.4.1.	Monasteries following the use of Cluny, Saint-Benigne at Dijon, or Bee	577
Figure		
V.4.1	Guidonian hand after Munich, Bayerische Staatsbibliothek, elm 14965b	469
TABLES
1.3.1. The Church year: the fixed part dependent upon Christmas 12
1.3.2. The Church year: the movable part dependent on Easter 13
1.3.3. Kalendár for October: Cambridge, Corpus Christi College 391 14-15
1.3.4. Saints widely celebrated in the Roman Church 17
1.4.1. The liturgical horarium at Cluny in the late eleventh century 20
1.4.2. Psalms allotted to office hours, ferial cursus 21 1.6.1. Psalms at ferial Vespers 29
I. 9.1. Beauvais Circumcision office, First Vespers 43
II. 7.1. Antiphon themes or prototype melodies 91 II. 11.1. Reciting notes in introit melodies 115
11.23.1. Tropes for the Easter introit Resunexi in Winchester sources 217
11.23.2. Gloria trope sets starting O laudabilis rex 226-1
11.23.3. Farsed Apostles'Creed in Laon 263 234
111.4.1. Interlocking Temporale and Sanctorale in Roman use 300
111.4.2. Disposition of Temporale and Sanctorale in Frankish use 301
111.4.3. Separate Temporale and Sanctorale in later medieval use 302 111.5.1. Deployment of material in two antiphoners 306 III.8.1. Contents of the Castle Acre processional 318
III. 15.1. Responsories in the Night Office of St Andrew 337
IV. 1.1. Notational signs in Montpellier, Faculté de Médecine H. 159 342-3 IV.2.1. French and German notational signs 348 IV.2.2. Palaeofrankish, Laon, Breton, and Aquitanian notational signs 349 IV.2.3. Signs related to French-German notation 352 IV.2.4. Other Italian notations 353 IV.3.1. Special signs in St Gall and Laon notation 361 IV.4.1. Prosodic accents 366 IV.4.2. Byzantine ekphonetic signs 368 IV.4.3. Accents, ekphonetic signs, and neumes (Byzantine) 369 IV.6.1. Special signs at the semitone step in Montpellier, Faculté de Médecine H. 159 388
IV. 6.2. Late forms derived from Laon and German notation 390
V. 2.1. Greater and Lesser Perfect System, diatonic genus 446 V.4.1. Greek and medieval modal names 462
V. 4.2. Writers on music theory in south Germany in the eleventh century 472
VI. 5.1. Chants at the start of mass in four rites 497 VI.5.2. Chants beside the lessons in four rites 500
Tables xxiii
IX.3.1.   Sources grouped according to their musical variants 574
IX.10.1. Aquitanian sources tor. 1200 597
IX. 10.2. Post-Pentecost alleluia series in Limoges 601-2
MUSIC EXAMPLES
Ex. I	1.2.1.	Versicles and responses	48
Ex. I	1.2.2.	From the Blessing of the Font on Easter Eve	50
Ex. I	1.2.3.	From the Exultet	52
Ex. I	1.2.4.	From a litany on Rogation Tuesday	53
Ex. I	1.2.5.	Pater noster	54
Ex. I	1.2.6.	From the Gospel on the feast of St John the Evangelist	55
Ex. I	1.2.7.	From the Genealogy according to St Matthew	57
Ex. I	1.2.8.	From the Genealogy according to St Luke	58
Ex. I	1.3.1.	The eight psalm tones	59-60
Ex. I	1.3.2.	Antiphon Scitote quia prope est, with psalm-tone	61
Ex. I	1.3.3.	Cadences from the Commemoratio brevis	62
Ex. I	1.3.4.	Tonus peregrinus	63
Ex. I	1.3.5.	Antiphon with 'parapter' psalm tone	63
Ex. I	1.3.6.	Psalm verses for introits	64
Ex. I	1.3.7.	Verses of the responsories Salvátorem expectamus, Audite verbum,	and
		Ecce virgo concipiet	65
Ex. I	1.3.8.	Invitatory tone	67
Ex. I	1.4.1.	Responds of four responsories	72
Ex. I	1.5.1.	Gradual Tecum piincipium	78
Ex. I	1.5.2.	From Gradual/Ww« nostra V. Eaqueus contntus est	80
Ex. I	1.5.3.	From Gradual Exaltabo te V. Domine Deus mens	81
Ex. I	1.5.4.	Tract Jubilate Domino omnis terra	84
Ex. I	1.6.1.	Short responsory Super te lerusalem	86
Ex. I	1.6.2.	Short responsory Resurrexit Dominus	87
Ex. I	1.6.3.	Ferial antiphons Credidi and Portio me a	87
Ex. I	1.7.1.	Psalter antiphons	92
Ex. I	1.7.2.	Antiphons with the melody Gevaert theme 29	93
Ex. I	1.7.3.	Antiphons with the melody Gevaert theme 36	94
Ex. I	1.7.4.	Antiphons	95
Ex. I	1.7.5.	Magnificat antiphons	97
Ex. I	1.8.1.	Invitatory antiphon Ouoniam Dens magnus and part of responsory	
		Tolle anna tua	99
Ex. I	1.9.1.	Processional antiphon Ego sum Deus	101
Ex. I	1.9.2.	From the processional antiphon Collegenint pontifices	102
Ex. I	1.9.3.	Processional antiphon Ego sum alpha et w	103
Ex. I	I.10.1.	Marian antiphon Sub tuum presidium	105
Ex. I	1.10.2.	Marian antiphon Speciosa facta es in two versions	106
Ex. I	1.10.3.	From Marian antiphons Alma redemptons mater and Salve regina	107
Ex. I	1.10.4.	Marian antiphon Au re i Hominis Mana	107
		Music Examples	xxv
Ex. I	1.10.5.	Marian antiphon Trinitatis thalamuni	108
Ex. I	1.11.1.	Mode 3 introit openings	111
Ex. I	Ill .2.	Mode 3 introit cadences	112
Ex. I	1.11.3.	Recitation around c in mode 3 introits	113
Ex. I	1.11.4.	From introit Ecce oculi	113
Ex. I	1.11.5.	Antiphon-Do/w/wí/ó- legifer noster and introit Timete Dominum	114
Ex. I	1.12.1.	Mode 6 communions	118
Ex. I	1.12.2.	Communion Dicit Dominus implete hydrias	119
Ex. I	1.12.3.	Diffusa est gracia as communion and responsory	120
Ex. I	1.13.1.	Offertory respond Ad te Domine levari	123
Ex. I	1.13.2.	Offertory respond In die sollempnitatis	124
Ex. I	1.13.3.	Offertory respond lubilate Deo universa terra	125
Ex. I	1.13.4.	From offertory respond Tollite port as	127
Ex. I	1.13.5.	Second verse of offertory Tollite port as	128
Ex. I	1.13.6.	Verses of offertory lubilate Deo universa terra	129
Ex. I	1.14.1.	Alleluia Dominus dixit ad me	132
Ex. I	1.14.2.	Another melody i or Alleluia Dominus dixit ad me	133
Ex. I	1.14.3.	Alleluia Attendite popule mens	134
Ex. I	1.14.4.	Alleluia Nuptie facte sunt in Chana	135
Ex. I	1.14.5.	Alleluia Veni sponsa Christi	136
Ex. I	1.14.6.	Alleluia Ante thronum trinitatis	138
Ex. I	1.14.7.	Alleluia 0 Maria rubens rosa	139
Ex. I	1.14.8.	Trope verses for Alleluia 0 Maria rubens rosa	139
Ex. I	1.15.1.	Hymn Christe qui lux es	143
Ex. I	1.15.2.	Hymn Somno refect is artubus	143
Ex. I	1.15.3.	Two melodies for hymn Ecce iam noctis	144
Ex. I	1.15.4.	Hymn Fange lingua	145
Ex. I	1.15.5.	Hymn Sanctorum tneritis	145
Ex. I	1.15.6.	From processional hymn Ardua spes mundi	147
Ex. I	1.17.1.	Litany	151
Ex. I	1.17.2.	Kyrie 55, Vatican ad lib. VI	153
Ex. I	1.17.3.	Kyrie 68, Vatican XIV	154
Ex. I	1.17.4.	Kyrie 155, Vatican XV	154
Ex. I	1.17.5.	Kyrie 77 and Kyrie 52	155
Ex. I	1.17.6.	Kyrie 97	156
Ex. I	1.17.7.	Kyrie 139	156
Ex. I	1.18.1.	Cadential figures (a-d) and intonation figure (e) from Gloria 39	158
Ex. I	1.18.2.	Gloria 2	159
Ex. I	1.18.3.	Gloria 5	160
Ex. 1	1.19.1.	Greek Sanctus and Latin Sanctus 66	163
Ex. ]	1.19.2.	Sanctus 68	164
Ex. ]	1.19.3.	Sanctus 216	165
Ex. I	1.19.4.	Sanctus 98	165
Ex. 1	1.20.1.	Old Roman Agnus; Agnus 226	167
Ex. I	1.20.2.	Agnus 144	168
Ex. 1	1.21.1.	Credo	170
xxvi		Music Examples	
Ex. I	1.21.2.	From Credo by Lully	171
Ex. I	[.22.1.	Sequence Clans vocibus	174
Ex. I	1.22.2.	From sequence Clans vocibus	175
Ex. I	[.22.3.	Start of sequence Eia recolamus	175
Ex. I	[.22.4.	Sequence Organicis canamus modulis	176
Ex. I	1.22.5.	Sequence Sancti baptisté	177
Ex. I	1.22.6.	Sequence Pin a Deum	179
Ex. I	1.22.7.	Alleluia Ostende and associated sequences	182
Ex. I	1.22.8.	Sequence Sancte cmcis celebremus	184
Ex. I	Í.22.9.	Sequence Victimepaschali laudes	189
Ex. I	1.22.10.	Sequence Congaudentes exultemns	190-1
Ex. I	1.22.11.	Sequence Hodieme lux diei	192
Ex. I	1.22.12.	Sequence Paranymphus salutat virginem	193
Ex. I	1.22.13.	Sequence Affluens deliciis	195
Ex. I	1.23.1.	Trope verses Dilectus iste Domini, etc. for introit in medio ecclesie	198-9
Ex. I	1.23.2.	Verse Ostende nobis for offertory Benedixisti, with prosulas Da nobis	
		famulis and Da nobis potenti	202
Ex. I	1.23.3.	Alleluia Concussum est mare with iprosxúz. Angelus Michael	203
Ex. I	1.23.4.	Responsory Beatus .Xicholaus with prosula Oportet devota mente	206
Ex. I	1.23.5.	Responsory 0 mater nostra with prosula Eterna virgo memorie	207
Ex. I	1.23.6.	From Gloria 56 (Vatican IV) with trope verses 0 gloria	
		sanctorum, etc., Regnum prosula 0 rex glorie	208
Ex. I	1.23.7.	Kyrie Te Chris te supplices	210
Ex. I	1.23.8.	Kyrie Eons bonitatis	212
Ex. I	1.23.9.	Kyrie 55 with trope verses Chrisle redemptor, etc.	213
Ex. I	1.23.10.	Benedicamus chant Eia nunc puen	214
Ex. I	1.23.11.	Benedicamus chants	215
Ex. I	1.23.12.	Trope verses Ecce iam Johannes, etc. for introit /;/ medio ecclesie	218-19
Ex. I	1.23.13.	Trope verse Dens paterfilium suum and start of introit Piter natus est	222
Ex. I	1.23.14.	Introductory verses I'Iodic cantandus est, etc. and start of introit Puer	
		natusest	222
Ex. I	1.23.15.	Gloria 39 with trope verses 0 laudabilis rex, etc. and Sccptrum	
		prosula Lumen eternum	228-30
Ex. I	1.23.16.	Sanctus 56 (Vatican III) with trope verses Summe pater, etc. and	
		Osanna prosulas Omnes tua gratia and Martyr Cluisti gratia	230-1
Ex. I	1.23.17.	Agnus 15 with trope verses Pro cunctis deductus, etc.	232
Ex. I	1.23.18.	Agnus 81a with trope verses Ad de x tram pat lis, etc. and Humanuni	
		genus, etc.	233
Ex. I	1.23.19.	Farsed Credo	235-6
Ex. I	1.23.20.	Introductory verse iMiides Deo for a farsed epistle	237
Ex. I	1.23.21.	Start of farsed lesson with verses Ce que Ysaies nos escrit, etc.	237
Ex. I	1.24.1.	Start of versus Sancte Paule pastor bone	241
Ex. I	1.24.2.	Benedicamus song Lux omni festa populo	243
Ex. I	1.24.3.	Latin song Letabundi iubilemus	244
Ex. I	1.24.4.	Latin song Natus est, natus est	245
Ex. I	1.24.5.	Benedicamus song Thesaurus nove grade	246
Music Examples
xxvii
Ex.	II.24.6.	Latin song Plebs Domini hac die	246
Ex.	II.24.7.	Latin song Ave mater salvatoris	247
Ex.	II.24.8.	Macaronic song Be den hoi mats	248
Ex.	11.25.1.	Quern queritis	253
Ex.	II.25.2.	Visitatio sepulchri	256-7
Ex.	II.25.3.	Thomas scene	258-9
Ex.	II.25.4.	Thomas scene	260
Ex.	11.25.5.	Visitatio sepulchri	262-3
Ex.	II.25.6.	Lamentatio Rachelis	266
Ex.	II.25.7.	From the Tours Easter play	268-9
Ex.	II.25.8.	Song Adam novus veterem from Fleury Peregrinus play	270
Ex.	II.25.9.	From the Ludus Danielis	271
Ex.	11.25.10.	From the Ludus Danielis	272
Ex.	11.25.11.	From the Ludus Danielis	272
Ex.	II.26.1.	Responsory Videns Rome vir beatus	275
Ex.	II.26.2.	Antiphon Gregorius vigiliis	276
Ex.	II.26.3.	Antiphons from the office of St Dominic	277
Ex.	II.26.4.	Antiphons from the offices of St Dominic and St Thomas of Canterbi	jry 278
Ex.	II.26.5.	Responsory Post sex annos	279
Ex.	III.14.1.	Extract from a tonary, part of the third tone	327
Ex.	III.14.2.	Type-melodies from a tonary	332
Ex.	III.14.3.	From the Gloria tones	334
Ex.	IV.1.1.	Communion Ego sum vitis vera	344
Ex.	IV.2.1.	Antiphon Beatus ille servus	355
Ex.	IV.2.2.	Communion Ouis dabit ex Siori	356
Ex.	IV.3.1.	From gradual Exultabunt sancti V. Cantate Domino	358
Ex.	IV.3.2.	Examples of oriscus	359
Ex.	IV.4.1.	Openings of mode 3 introits (St Gall notation)	372
Ex.	IV.5.1.	From offertory Benedictus es Domine	375
Ex.	IV.5.2.	Normal and modified signs in St Gall notation (selection)	376
Ex.	IV.5.3.	Normal and 'longer' signs in Laon notation (selection)	377
Ex.	IV.5.4.	A standard phrase from mode 5 graduals (Apel Fl) with Laon and St	
		Gall notation	378
Ex.	IV.5.5.	Possible gradations of duration in pes and clivis (Laon and St Gall	
		notations)	378
Ex.	IV.5.6.	Mensural transcription of passages horn Alleluia Pascha nostrum	379
Ex.	IV.5.7.	Mensural transcription of passages from Alleluia Pascha nostrum	380
Ex.	IV.5.8.	Antiphons with the same melody	381
Ex.	IV.6.1.	Alphabetic notation a-p	387
Ex.	IV.7.1.	Hucbald's letter-notation	393
Ex.	IV.7.2.	Dasia signs in the Enchiriadis group of treatises	393
Ex.	IV.7.3.	Boethius' pitch-letters	394
Ex.	IV.7.4.	Pitch-letters in the Dialogus de musica and Guido of Arezzo	395
Ex.	IV.7.5.	Hermannus Contractus' interval notation	396
Ex.	IV.8.1.	Communion Pascha nostrum	398
xxviii		Music Examples	
Ex.	V.2.I.	Bocthius' division of the monochord	445
Ex.	V.2.2.	Species of intervals	447
Ex.	V.3.I.	Antiphon Kcce vere Israelita notatcd by Ilucbald on six 'strings'	449
Ex.	V.3.2.	Hucbald's arrangement of ascending tetrachords	450
Ex.	V.3.3.	Introit Statuit ei as described by Ilucbald	450
Ex.	V.3.4.	Hucbald's letter-notation	451
Ex.	V.3.5.	Examples in protus mode from Musica enchinadis	453
Ex.	V.4.I.	Psalm-verse cadences and openings of mode 1 introits cited by Aurelian	
		of Reome	457
Ex.	V.4.2.	Division of the monochord in Dialogus de musica	464
Ex.	V.4.3.	Notes required to notate antiphon Domine qui ope rati', various versions	465
Ex.	V.4.4.	Ranges of chants in the eight modes, Dialogus de musica	465
Ex.	V.4.5.	'Similitudes' of notes to the various modes, Dialogus de musica	466
Ex.	V.4.6.	Hymn i't queant laxis, Guido of Arezzo	467
Ex.	V.4.7.	Antiphon Diligite Dominion	468
Ex.	V.4.8.	The modi vocum, Guido of Arezzo	470
Ex.	V.4.9.	Notes required to notate communion Defructu, two versions	474
Ex.	V.4.10	Modal qualities of notes, Hermannus Contractus	475
Ex.	VIII.3.1.	Old Roman and Gregorian versions of communion Rerelabiturgloria	
		Domini	532
Ex.	VIII.3.2.	Gradual verses Cell enarrant and In manibus portabunt te	534
Ex.	VIII.3.3.	Antiphons in Old Roman and Gregorian versions	535
Ex.	VIII.3.4.	Old Roman and Gregorian versions of offertory verse Qui propitiatur	536
Ex.	VIII.3.5.	Old Roman and Gregorian versions of verse Quoniam angelis from	
		offertory Scapulis suis	537
Ex.	VIII.3.6.	First verse of Alleluia Para turn cormeum	538
Ex.	VIII.4.1.	Transitorium Convertimini omnes	543
Ex.	VIII.4.2.	Responsorium cum infantibus 67)cjrf;?/ in te	545
Ex.	VIII.4.3.	Melodiae of Ambrosian Alleluia VI	546
Ex.	VIII.4.4.	Gregorian, Old Roman, and Ambrosian versions of introit/ingressa	
		Invocavit me	548
Ex.	VIII.5.1.	Ambrosian and Old Bcneventan versions of communion llymnum canite	551
Ex.	VIII. 6.1.	Processional antiphon Cum aitdtsset populus	555
Ex.	VIII.6.2.	Part of offertory $a?ictificavit Moyses in Old Spanish, Gregorian,	
		Ambrosian, and Old Roman versions 556-7	
Ex.	IX.7.1.	Trope verses Qui dicat patri, etc. for introit Resurrexi	587
Ex.	IX.7.2.	Sequence Clara gaudia festa paschalia	588
Ex.	IX.8.1.	Rogation antiphon Peccavimus Domine	592
Ex.	IX.8.2.	Start of farsed lesson with verses Ad laudem regis glorie, etc.	593
Ex.	IX. 10.1.	Responsory 0 sancte Dei apostole 599	-600
Ex.	X.3.I.	From gradual Pxurge Domine in Gregorian and Cistercian-Dominican	
		versions	612
Music Examples xxix
Ex. X.5.1.        Medieval and Mcdiccan versions of gradual Otunes de Saba 617
Ex. X.6.1.        Extracts from Graduelde Paris, 1754 619-20
Ex. X1.1.1.      Opening of gradual I'lxurge Domine non prevalent homo 625
NOTE ON MUSIC EXAMPLES
Slurs are used only for notes joined in the same sign in the original notation.
h   = oriscus /v = neume at the semitone step ('mi-neume')
w   = quilisma ' = apostropha
r° ^ = liquescent note
Pitch-letters follow Guidonian, not modern practice.
m * '
r A BCD EEG ab c d e f g a I 'etc.
ABBREVIATIONS
AcM        Acta tnusicologica
AfAlw      Archiv für Musikwissenschaft
AH Analecta hymnica medii aevi, ed. Guido Maria Dreves, Clemens Blume, and
Henry Marriott Bannister, 55 vols. (Leipzig, 1886-1922); Register, ed. Max Lütolf, 3 vols. (Berne, 1978). Text editions of hymns: 2, 4, 11-12, 14, 16, 19, 22-3, 27, 43, 48, 50-2; of rhymed offices: 5, 13, 17-18, 24-6, 28, 45; of sequences: 7-10, 34, 37, 39-40, 42, 44, 53-5; of tropes: 47, 49. For individual volumes see Bibliography
AM Antiphonale monasticum pro diurnis horis (Tournai, 1934)
AMS Rene-Jean Hesbert, Antiphonale missarum sextuplex (Brussels, 1935) [Monza, Basilica S. Giovanni, CIX: 'Modoetiensis'; Zürich, Zentralbibliothek, Rheinau 30: 'Rhenaugiensis'; Brussels, Bibliotheque Royale, 10127-10144: 'Blandiniensis'; Paris, Bibliotheque Nationale, lat. 17436: 'Compendiensis'; Paris, Bibliotheque Nationale, lat. 12050: 'Corbiensis'; Paris, Bibliotheque Sainte-Genevieve, 111: 'Silvanectensis']
Ali Antiphonale sacrosanctae Romanae ecclesiae (Rome, 1912)
AS Antiphonale Sarisburiense: A Reproduction in Facsimile of a Manuscript of the
Thirteenth Century, ed. Walter Howard Frere (London, 1901-24) BzG        Beiträge zur Gregorianik
CAO Corpus antiphonalium officii, ed. Rene-Jean Hesbert (Rerum ecclesiasticarum documenta, Series maior, Fontes, 7-12; Rome, 1963-79). For individual volumes see Bibliography
CCM       Corpus consuetudinum monasticarum, ed. Kassius Hallinger (Siegburg, 1963- ).
For individual volumes see Bibliography CM Consuetudines monasticac,  ed.   Bruno  Albers.  For  individual  volumes see
Bibliography
CS Edmond de Coussemaker, Scriptorum de musica medii aevi novam seriem . . . , 4
vols. (Pans, 1864, 1867, 1869, 1876) CSM       Corpus scriptorum de musica (Rome,  1950- ). For individual volumes see
Bibliography
CT Corpus troporum (Studia Latina Stockholmiensia; Stockholm, 1975- ). For
individual volumes see Bibliography DACL      Fernand Cabrol, Henri Leclercq, and Henri Marrou (eds.), Dictionnaire d'archeologie
chretienne et de liturgie, 15 vols, in 30 (Paris, 1907-53) DMA      Divitiae musicae artis, ed. Joseph Smits van Waesberghe (Buren, 1975- ). For
individual volumes see Bibliography EG Etudes gregoriennes
EL Ephemerides liturgicae
EMH      Early Music History
GR Graduate sacrosanctae Romanae ecclesiae (Rome, 1908)
xxxii Abbreviations
GS Martin Gerbert, Scriptores ecclesiastici de miisica sacra potissimum, 3 vols.
(Sankt-Blasien, 1784) [to be used in conjunction with Bernhard 1989] GT Graduate triplex, ed. Marie-Claire Billecocq and Rupert Fischer (Solesmes, 1979)
[Graduate Romanům of 1974 with neumes of Laon 239, Einsiedeln 121, St Gall
359 and 339, etc.]
HMT       Hans-Heinrich Eggebrecht (ed.), Handwörterbuch der musikalischen Terminologie
(Mainz, 1971- ) JAMS      Journal of the American Musicological Society jfPMMS   Journal of the Plainsong & Mediaeval Music Society Kmjb      Kirchenmusikalisches Jahrbuch
LR Liber responsorialis pro festis I. classis et communi sanctorum juxta rit um
monasticum (Solesmes, 1894) LU Liber usualis missae et officii pro dominicis et festis I. vel II. classis (Rome, 1921)
MD Musica disciplina
Mf Die Musikforschung
MGG       Friedrich Blume (ed.), Die Musik in Geschichte und Gegenwart: Allgemeine
Enzyklopädie der Musik, 17 vols. (Kassel, 1949-86) MGH       Monumenta Germaniae historica. For individual volumes see Bibliography MMMA   Monumenta monodica medii aevi (Kassel, 1956- ). For individual volumes see
Bibliography
MMS       Monumenta musicae sacrae, ed. René-Jean Hesbert. For individual volumes see
Bibliography MQ Musical Quarterly
NCE The New Catholic Encyclopedia, 15 vols. (New York, 1967; supplementary vols. 1974, 1979)
NG Stanley Sadie (ed.), The New Grove Dictionary of Music and Musicians, 20 vols.
(London, 1980)
PalMus Paléographie musicale: Les principaux manuscrits de chant grégorien, ambrosicn, mozarabe, gallican [premier série, deuxiěme série]. For individual volumes sec Bibliography
PL Patrologiae cursus completus, series latina, ed. J.-P. Migne, 221 vols. (Paris, 1844-64)
RB Revue bénédictine
RCG       Revue du chant grégorien
RG Revue grégorienne
RIM        Rivista italiana di musicologia
RISM      Repertoire international des sources musicales (Munich and Duisburg). For
individual volumes see Bibliography SMII       Studia musicologica Academiae scie?itiarum Hungaricae
VGA       Veröffentlichungen der Gregorianischen Akademie zu Freiburg in der Schweiz
BIBLIOGRAPHY
Abate, G., 'II primitivo breviario francescano', Miscellanea francescana, 60 (1960), 47-240. Abert, Hermann, Die aesthetischen Grundsätze der mittelalterlichen Melodiebildung (Halle, 1905).
-Die Musikanschauung des Mittelalters und ihre Grundlagen (Halle, 1905).
Abrahamsen, Erik, Elements romans et allemands dans le chant gregorien et la chanson
populaire en Danemark (Publications de l'Academie gregorienne de Fribourg (Suisse), 11;
Copenhagen, 1923).
Acta sanctorum collecta . . . a Sociis Bollandianis (Antwerp, 1643- ; 3rd edn., Paris and Brüssels, 1863- ).
Acta sanctorum ordinis sancti Benedict/, ed. Jean Mabillon, Luc d'Achery, et al., 9 vols. (Paris, 1668-1701).
Adler Festschrift 1930 = Studien zur Musikgeschichte: Festschrift für Guido Adler (Vienna, 1930).
Agustoni, Luigi, Gregorianischer Choral: Elemente und Vortragslehre mit besonderer
Berücksichtigung der Neumenkunde (Fribourg, 1963). -'Die Frage der Tonstufen SI und MF, BzG 4 (1987), 47-101.
-and Göschl, Johannes Berchmans, Einfühlung in die Inteipretation des Gregorianischen
Chorals, 1: Grundlagen (Regensburg, 1987). All 2 (1887) = Hymnarius Moissiacensis: Das Hymnar der Abtei Moissac im 10. Jahrhundert
nach einer Handschrift der Rossiana, ed. Guido Maria Dreves. All 4 (1888) = Hymni inediti: Liturgische Hymnen des Mittelalters aus handschriftlichen
Breviarien, Antiphonalien und Processionalien, ed. Dreves. AH 5 (1889), 13 (1892), 18 (1894), 24 (1896), 25-6 (1897), 28 (1898), 45 (1904) = Historiae
tiiythmicae: Liturgische Reimoffizien des Mittelalters aus Handschriften und Wiegendnicken,
cd. Dreves.
All 7 (1889) = Prosarium Lemovicense: Die Prosen der Abtei St. Martial zu Limoges, aus
Troparien des 10., 11. und 12. Jahrhunderts, ed. Dreves. AH 8-9 (1890), 10 (1891) = Sequentiae ineditae: Liturgische Prosen des Mittelalters aus
Handschriften und Frühdrucken, ed. Dreves. AH 11 (1891), 12 (1892), 19 (1895), 22 (1895), 23 (1896), 43 (1903) = Hymni inediti:
Liturgische Hymnen des Mittelalters aus Handschriften und Wiegendrucken, ed. Dreves. AH 13 see AH 5.
AH 14 (1893) = Hymnarius Severinianus: Das Hymnar der Abtei S. Severin in Neapel nach
den Codices Vaticanus 7172 und Parisinus 1092, ed. Dreves. AH 16 (1894) = Hymnodia Hiberica: Spanische Hymnen des Mittelalters aus liturgischen
Handschriften und Druckwerken römischen Ordos, ed. Dreves. AH 17 (1894) = Hymnodia Hiberica: Liturgische Reimoffizien aus spanischen Brevieren. Im
Anhange: Carmina Compostellana, die Lieder des s.g. Codex Calixtinus, ed. Dreves. AH 18 see AH 5. AH 19, 22, 23 see AH 11.
xxxiv
Bibliography
AH 24-6, 28 see AH 5.
AH 27 (1897) = Hymnodia Gotica: Die mozarabischen Hymnen des alt-spanischen Ritus aus
handschriftlichen und gedruckten Quellen, ed. Clemens Blume. AH 34 (1900), 37 (1901), 39 (1902), 42 (1903), 44 (1904) = Sequentiae inedhae: Liturgische
Prosen des Mittelalters aus Handschriften und Frühdrucken, ed. Blume. AH 40 (1902)  - Sequentiae inedhae: Liturgische Prosen des Mittelalters aus Handschriften
und Frühdrucken, ed. Henry Marriott Bannister. AH 42 see AH 34. AH 43 see AH 11. AH 44 see AH 34. AH 45 see AH 5.
AH 47 (1905) = Tropi graduales: Tropen des Missale im Mittelalter. I. Tropen zum Ordinarium Missae. Aus handschriftlichen Quellen, ed. Blume and Bannister.
AH 48 (1905), 50 (1907) = Hymnographi Latini: Lateinische Hymnendichter des Mittelalters aus gedruckten und ungedruckten Quellen, ed. Dreves.
AH 49 (1906) = Tropi graduales: Tropen des Missale im Mittelalter. II. Tropen zum Proprium Missae. Aus handschriftlichen Quellen, ed. Blume.
AH 50 see AH 48.
AH 51 (1908) = Thesauri hymnologici hymnarium: Die Hymnen des Thesaurus Hymnologicus
H. A. Daniels und anderer Hymnen-Ausgaben. I. Die Hymnen des 5.-11. Jahrhunderts
und die irisch-keltische Hymnodie aus den ältesten Quellen, ed. Blume. AH 52 (1909) = Thesauri hymnologici hymnarium . . . II. Die Hymnen des 12.-16.
Jahrhunderts aus den ältesten Quellen, ed. Blume. AH 53 (1911) = Thesauri hymnologici prosarium: Liturgische Prosen erster Epoche aus den
Sequenzenschulen des Abendlandes, insbesondere die dem Notkerus Balbulus zugeschriebenen,
nebst Skizze über den Ursprung der Sequenz, aufgrund der Melodien aus den Quellen des
10.-16. Jahrhunderts, ed. Blume and Bannister. AH 54 (1915) = Thesauri hymnologici prosarium: Liturgische Prosen des Ubergangsstiles und
der zweiten Epoche, insbesondere die dem Adam von Sanct Victor zugeschriebenen, aus
Handschriften und Frühdrucken, ed. Blume and Bannister. AH 55 (1922) = Thesauri hymnologici prosarium: Liturgische Prosen zweiter Epoche auf Feste
der Heiligen nebst einem Anhange: Hymnologie des Gelderlandes und des Haarletner
Gebietes aus Handschriften und Frühdrucken, ed. Blume. Aikää, Ermo, 'Ein marianischer Gloria-Tropus', paper read at the First European Science
Foundation International Workshop on Tropes, Munich, 1983. Albarosa, Nino, 'La scuola gregonana di Eugene Cardine', RIM 9 (1974), 269-97; 12 (1977)
136-52.
-'The Pontificio Istituto di Musica Sacra in Rome and the Semiological School of Dom
Eugene Cardine', JPMMS 6 (1983), 26-33.
-'Paleografi non semiologi?', in Huglo 1987, 101-5.
Albers see CM.
Albrecht (Hans) Gedenkschrift 1962 = Wilfried Brennecke and Hans Haase (eds.), Hans
Albrecht in Memoriam (Kassel, 1962). Albrecht (Otto) Festschrift 1980 = John Walter Hill (ed.), Studies in Musicology in Honor
ofOtto E. Albrecht (Kassel, 1980). Alfonzo, Pio, L'antifonario deU'Uffizio Romano: Note sulle origini della composizione dei
testi (Monografie liturgiche, 3; Subiaco, 1935).
Bibliography xxxv
-/ responsori biblici deU'Uffizio Romano (Lateranum, ns, 2/1; Rome, 1936).
Allworth, Christopher, 'The Medieval Processional: Donaueschingen MS 882', EL 84 (1970), 169-86.
Almeida, Fortunato de, História da igreja em Portugal, 8 vols, in 4 (Coimbra, 1910-22; 2nd edn., 1930).
Amelli, Ambrogio M., 'L'epigramma di Paolo Diacono intorno al canto Gregoriano e
Ambrosiano'. Memorie storiche forogiuliesi, 9 (1913), 153-75. Amiet, Robert, 'Un bréviaire note du xive siěcle de la primatiale Saint-Jean de Lyon', EG
15 (1975), 201-6.
-'Le Liber misticus de Combret, au diocese d'Elne', EG 20 (1981), 5-68.
Andersen, M. Geert, and Raasted, Jorgen, Inventar over Det kongelige Biblioteks
fragmentsamling [Inventory of the Fragment Collection in the Royal Library] (Hjaelpemidler,
6; Copenhagen, 1983).
Anderson, Gordon A. (ed.), The Las Huelgas Manuscript, Burgos, Monasterio de las
Huelgas, 2 vols. (Corpus mensurabilis musicae, 79; Rome, 1982). Anderson (Gordon)  Gedenkschrift  1984 = Gordon Athol Anderson (J929-1981) in
memoriam, 2 vols. (Henryville-Ottawa-Binningen, 1984). Anderson, Warren, 'Alypius', NG.
Andoyer, Raphael, 'Le Chant romain antégrégorien', RCG 20 (1911-12), 69-75, 107-14.
-'ĽAncienne Liturgie de Bénévenť, RCG 20 (1911-12), 176-83; 21 (1912-13), 14-20,
44-51, 81-5, 112-15, 144-8, 169-74; 22 (1913-14), 8-11, 41-4, 80-3, 106-11, 141-5, 170-2; 23 (1919-20), 44-50, 116-18, 151-3, 182-3; 24 (1920-1), 48-50, 87-9, 146-8, 182-5.
Andrieu, Michel, 'Reglement d'Angilramne de Metz (768-791) fixant les honoraires de quelques fonctions liturgiques', Revue des sciences religieuses, 10 (1930), 349—69.
-I^es Ordines Romani du haut moyen age (Spicilegium sacrum Lovaniense, 11, 23-4,
28-9; Louvain, 1931-61).
-Le Pontifical romain au moyen áge (Studi e testi, 86-8, 99; Rome, 1938-41).
Andrieu Festschrift 1956 = Melanges en ľhonneur de M. Michel Andrieu (Strasburg, 1956).
Angerer, Joachim Fridolin, Die liturgisch-musikalische Erneuerung der Melker Reform: Studien zur Etforschung der Musikpraxis in den Benediktinerklöstern des 15. Jahrhunderts (Osterreichische Akademie der Wissenschaften, Philologisch-historische Klasse, 287; Vienna, 1974).
-'Die Consuetudines Monasticae als Quelle für die Musikwissenschaft', in Haberl
Festschrift 1977, 237.
-Lateinische und deutsche Gesänge aus der Zeit der Melker Reform (Forschungen zur
älteren Musikgeschichte, 2; Vienna, 1979). Angles, Higini, 'Epištola farcida del martiri de Sant Esteve', Vida cristiana, 9 (1922), 69-
75.
-(ed.), El codex musical de Ims Huelgas (Música a veus dels segles XIII-XIV), 3 vols.
(Biblioteca de Catalunya, Publicacions del Departament de Müsica, 6; Barcelona, 1931). -La música a Catalunya fins al segle XIII (Biblioteca de Catalunya, Publicacions del
Departament de Musica, 10; Barcelona, 1935). -'La  müsica  medieval  en  Toledo hasta el siglo  XI',  Gesammelte Aufsätze zur
Kulturgeschichte Spaniens, 7 (Spanische Forschungen der Görresgesellschaft, 1st ser.;
Münster, 1938), 1-68. -História de la música medieval en Navarra (Pamplona, 1970).
xxxvi Bibliography
Angles, IIigini and Subira, Jose, Catdlogo musical de la Biblioteca uacioual de Madrid, 1:
Manuscritos (Barcelona, 1946). Angles Festschrift 1958-61 = Misceldnea en homenaje a Monsenor Higinio Angles, 2 vols.
(Barcelona, 1958, 1961). Antifonario visigotico mozärabe de la catedral de Leon (Monumcnta Ilispaniae sacra, scrie
litürgica, v/2; Madrid, 1953) (see also Brou and Vives 1959). Antiphonale Ilispaniae vetus (s.X-XI). Biblioteca de la Universidad de Zaragoza (Saragossa,
1986).
Antiphonale missamm juxta ritum sanctae ecclesiae Mediolanensis [cd. G. M. Sunol] (Rome, 1935).
Antiphonale monasticum secundum traditionem Helveticae Congregatioms Benedictinae (1943).
Antiphonale Pataviense (Wien 1519), ed. Karlheinz Schlager (Das Erbe deutscher Musik, 88; Kassel, 1985).
Antiphonale Silense: British Libran' Mss. Add. 30.850, ed. Ismael Fernandez de la Cuesta (Madrid, 1985).
Antiphonar von St. Peter [Vienna, Nationalbibliothek, series nova 2700], 2 vols. (Codices
selecti, 21; Graz, 1969-73, 1974). Apel, Willi, Gregorian Chant (Bloomington, Ind., 1958).
Archer, Peter, The Christian Calendar and the Gregorian Reform (New York, 1941). Arlt, Wulf, Ein Eestoffizium des Mittelalters aus Beauvais in seiner liturgischer! und
musikalischen Bedeutung, 2 vols. (Cologne, 1970). -'Einstimmige  Lieder des  12. Jahrhunderts und Mehrstimmiges in französischen
Handschriften des 16. Jahrhunderts aus Le Puy', Schweizer Beiträge zur Musikwissenschaft, 3
(1978), 7-47.
-'Zur Interpretation der Tropen', Forum musicologicum, 3 (1982), 61-90.
-and Stauffacher, Matthias, Engelberg Stiftsbibliothek Codex 314 (Wintcrthur, 1986).
Arnese, Raffaele, I codici not at i della Biblioteca nazionale di Napoli (Florence, 1967). Arnoux, G., Musique platonicienne—äme du monde (Paris, 1960).
Ashworth, Henry, 'Did St. Augustine Bring the Gregorianum to England?'7^7.72 (1958), 39-43.
-'The Liturgical Prayers of St. Gregory the Great', Traditio, 15 (1959), 107-61.
-'Further Parallels to the "Iladrianum" from St. Gregory the Great's Commentary on the
First Book of Kings', Traditio, 16 (1960), 364-73.
Asketorp, Bodil, 'The Musical Contents of Two Danish Pontificals from the Late Middle-Ages', JPMMS 7 (1984), 28-46.
Atchley, E. G. C. F., Ordo Rom an us primus (London, 1905).
Atkinson, Charles M., 'The Earliest Agnus Dei Melodv and its Tropes', JAMS 30 (1977), 1-19.
-'Ö amnos tu theu: The Greek Agnus Dei in the Roman Liturgy from the Eighth to the
Eleventh Century', Kmjb 65 (1981), 7-30.
-'Zur Entstehung und Überlieferung der "Missa Gracca'", AfMw 39 (1982), 113-45.
-'The Parapteres: Nothi or Not?', MO 68 (1982), 32-59.
-'On the Interpretation of Modi, quos abusive tonos dieimus', in Gallacher and Damico
1989, 147-61.
-'The Doxa, the Pisteuo, and the Ellinici Erat res: Some Anomalies in the Transmission of
the Chants of the "Missa Graeca'", Journal of Musicology, 7 (1989), 81-106.
Bibliography
xxxvii
-'From "Vitium" to "Tonus acquisitus": On the Evolution of the Notational Matrix of
Medieval Chant', in Cantus Planus 1990, 181-97. -'Parapter', Handwörterbuch der musikalischen Terminologie, ed. Hans-IIeinrich Eggebrecht (Mainz, 1971- ). Au da, Antoine, L'Ecole musicale liegeoise au X' siede: Ktienne de Liege (Brussels, 1923). -lxjs Modes et les tons de la musique et specialement de la musique medievale (Academie
royale de Belgique. Classe des beaux arts. Memoircs, iii, pt. 1; Brussels, 1930). -Contribution ä Vhistoire de l'origine des modes et des tons gregoriens (Grenoble, 1932) =
RCG 36 (1932), 33-9, 72-7, 105-11, 130-2. Augustine, St, De musica, a Synopsis, ed. W. F. J. Knight (London, 1949). -On Music: De musica.  The Fathers of the Church, a New Translation, trans.
R. C. Taliaferro (Writings of Saint Augustine, 2; New York, 1947), 153-379. Avenary, Hanoch, 'The Northern and Southern Idioms of Early European Music: A New
Approach to an Old Problem', AcM 49 (1977), 27-49. Avery, Myrtilla, The Exultet Rolls of South Italy (Princeton, NJ, 1936). Babb, Warren, Hucbald, Guido, and John on Music, ed., with introductions, by Claude V.
Palisca, index of chants by Alejandro Enrique Planchart (New Haven, Conn., 1978). Bagnall see Yardley.
Bailey, Terence, The Processions of Sarum and the Western Church (Toronto, 1971). -77?^ Intonation Fonnulas of Western Chant (Toronto, 1974).
-'Accentual and Cursive Cadences in Gregorian Psalmody', JAMS 29 (1976), 463-71.
-'Modes and Myth', Studies in Music from the University of Western Ontario, 1 (1976),
43-54.
-'Ambrosian Psalmody: An Introduction', Studies in Music from the University of
Western Ontano, 2 (1977), 65-78.
-'De modi's musicis: A New Edition and Explanation', Kmjb 61—2 (1977-8), 47-60.
-'Ambrosian Psalmody: The Formulae', Studies in Music from the University of Western
Ontario, 3 (1978), 72-96. -Commemoratio brevis de tonis et psalmis modulandis: Introduction, Critical Edition,
Translation (Ottawa, 1979).
-The Ambrosian Alleluias (Englefield Green, 1983).
-'Ambrosian Chant in Southern Italy', JPMMS 6 (1983), 1-7.
-The Ambrosian Cantus (Ottawa, 1987).
-'Milanese Melodic Tropes', JPMMS 11 (1988), 1-12.
-and Merkley, Paul, The Antiphons of the Ambrosian Office (Ottawa, 1989).
--The Melodic Tradition of the Ambrosian Office-Antiphons (Ottawa, 1990).
Baldovin, John F., The Urban Character of Christian Worship: The Origins, Development,
and Meaning of Stational Liturgy (Orientalia Christiana analecta, 228; Rome, 1987). Baltzer, Rebecca A., 'Johannes de Garlandia', NG.
Bamberg 6 = Die Handschrift Bamberg Staatsbibliothek Lit. 6 (Monumenta palaeographica
Gregoriana, 2; Münsterschwarzach, [1986]). Bannister, Henry Marriott, Monument! vaticani di paleografia musicale latina, 2 vols.
(Codices e vaticanis selecti, phototypice expressi, 11; Leipzig, 1913). -see also AH.
Baralli, Raffaello, 'A proposito di un piccolo trattato sul canto ecclesiastico in un manoscritto dei sec. X-XF, Rassegna gregoriana, 4 (1905), 59—66.
xxxvu1
Bibliography
Bärdos, Korn£l, Volksmusikartige Variierungsteclittik in den ungarischen Passionen, 15. bis 18. Jahrhundert (Musicologia Hungarica, ns, 5; Budapest, 1975).
Barezzani, Maria Teresa Rosa, Im notazione neumatica di un codice bresciano (secolo XI) (Cremona, 1981).
Baroffio, Bonifazio, Die Offertorien der ambrosianischen Kirche: Vorstudie zur kritischen Ausgabe der mailändischen Gesänge (Diss., Cologne University, 1964).
——'Die mailändische Überlieferung des Offertoriums Sartctificavit', in Stäblein Festschrift 1967, 1-8.
-'Ambrosianische Liturgie', in Feilerer 1972, 191-204.
-'Le origini del canto liturgico nella Chiesa latina e la formazione dei repertori italici',
Renovatio, 13 (1978), 26-52.
-'Benevent', MGG (supplement).
-'Ambrosian Rite, Music of the', NG.
-and Steiner, Ruth, 'Offertory', NG.
Batiffol, Pierre, Histoire du breviaire vornan (3rd edn., Paris, 1911); trans.: History of the
Roman Dreviavy (London, 1913). Baudot, Jules, 'Benedictionnaire', DACL.
Bäumer, Suitbert, Geschichte des Breviers, 2 vols. (Freiburg, 1895); trans, and rev. R. Biron,
Histoire du breviaire, 2 vols. (Paris, 1905). Baumstark, Anton, Comparative Liturgy (London, 1958).
Bautier, Robert-Henri, Gilles, Monique, Duchez, Marie-Elisabeth, and Huglo,
Michel, Odorannus de Sens: Opera omnia (Paris, 1972). Bautier-Regnier, Anne-Marie, 'A propos du sens de neuma et de nota en latin medieval',
Revue beige de musicologie, 18 (1964), 1-9. Bayart, Paul, Les Offices de Saint Winnoc et de Saint Oswald d'apres le manuscrit 14 de la
Bibliotheque de Bergues, Annales du Comite Flamand de France 35 (Lille, 1926). Beare, William, Latin Verse and European Song: A Study in Accent and Rhythm (London,
1957).
Becker, Gustav Heinrich, Catalogi bibliothecavum antiqui (Bonn, 1885).
Becker, Hansjakob, Die Responsorien des Kartäuserbreviers (Münchener theologische
Studien, II. Systematische Abteilung, 39; Munich, 1971). -Das Tonale Guigos I: Ein Beitrag zur Geschichte des liturgischen Gesangs und der Ars
Musica im Mittelalter (Münchener Beiträge zur Mediävistik und Renaissance-Forschung,
23; Munich, 1975).
Bede, A History of the English Church and People, ed. Leo Sherley-Price (Harmondsworth,
1955; 2nd edn., 1968)/
-Opera historica, ed. C. Plummer, 2 vols. (Oxford, 1896).
Benoit-Castelli, Georges, 'Le "Praeconium paschale'", EL 67 (1953), 309-34.
-'Un processional anglais du xivcmc siecle: Le processional dit "de Rollington'", EL 75
(1961), 281-326.
Benson, Robert L., Constable, Giles, and Lanham, Carol D. (eds.), Renaissance and
Renewal in the Twelfth Century (Cambridge, Mass., 1982). Benton, John F., 'Nicolas of Clairvaux and the Twelfth-Century Sequence, with Special
Reference to Adam of St. Victor', Traditio, 18 (1962), 149-79. Berger, Hugo, Untersuchungen zu den Psalmdijferenzen (Kölner Beiträge zur Musikforschung,
37; Regensburg, 1966).
Bibliography
xxxix
Bernard, Madeleine, 'L'Officium Stellae Nivernais', Revue de musicologie, 51 (1965), 52-65.
-Bibliotheque Sainte-Genevieve (Repertoire de manuscrits medievaux contenant des
notations musicales, 1; Paris, 1965). -Bibliotheque Mazarine (Repertoire de manuscrits medievaux contenant des notations
musicales, 2; Paris, 1966). -Bibliotheque  de  I'Arsenal,  Bibliotheque  nationale  (musique),  des beaux-arts, de
1'Vniversite et petits fonds (Repertoire de manuscrits medievaux contenant des notations
musicales, 3; Paris, 1974). -'Un recueil inedit du xne siecle et la copie aquitaine de l'office versifie de saint Bernard',
EG 16 (1977), 145-59.
-'Les Offices versifies attribues ä Leon IX (1002-1054)', EG 19 (1980), 89-164.
Bernhard, Michael, Wortkonkordanz zu Anicius Manlius Severinus Boethius De Institutione
Musica (Bayerische Akademie der Wissenschaften, Veröffentlichungen der Musikhistorischen
Kommission, 4; Munich, 1979). -Studien zur Epistola de hannonica institutione des Regino von Prüm (Bayerische
Akademie der Wissenschaften, Veröffentlichungen der Musikhistorischen Kommission, 5;
Munich, 1979).
-Clavis Gerberti: Eine Revision von Martin Gerberts Scriptores ecclesiastici de musica
sacra potissimum (St. Blasien 1784) 1 (Bayerische Akademie der Wissenschaften, Veröffentlichungen der Musikhistorischen Kommission, 7; Munich, 1989).
[Berry, Mary] Mother Thomas More, 'The Performance of Plainsong in the Later Middle Ages and the Sixteenth Century', Proceedings of the Royal Musical Association, 92 (1965— 6), 121-34.
-'Gregorian Chant. The Restoration of the Chant and Seventy-five Years of Recording',
Early Music, 7 (1979), 197-217.
-arts. 'Cistercian Monks, 'Dominican Friars', 'Sarum Rite, Music of the', NG.
Bescond, Albert, I^e Chant gregorien (Paris, 1972).
Besseler Festschrift 1961 = Festschrift Heinrich Besseler zum sechzigsten Geburtstag (Leipzig, 1961).
Bewerunge,  IL,  'The Metrical  Cursus in the Antiphon  Melodies', Zeitschrift der
internationalen Musikgesellschaft, 13 (1910-11), 227. Biblioteca sanctorum, 13 vols. (Rome, 1961-70).
BiLLECOCQ, Marie-Claire, 'Lettres ajoutees ä la notation neumatique du codex 239 de Laon', EG 17 (1978), 71-44.
Binford-Walsh, Hilde M., 'The Ordering of Melody in Aquitanian Chant: A Study of
Mode One Introit Tropes', in Cantus Planus 1990, 327-39. Birkner, Günther, 'Eine "Sequentia sancti Johannis confessoris" in Trogir (Dalmatien)',
Musik des Ostens, 2 (1963), 91. Bischoff,  Bernhard, 'Otloh',  in Karl Langosch (ed.), Die deutsche Literatur des
Mittelalters. Vetfasserlexicon, 5: Nachtragband (Berlin, 1955), 658-70. ——'Panorama der Handschriftenüberlieferung aus der Zeit Karls des Großen', in Braunfels,
ii (1965), 233-54 = Bischoff 1966, 5-38. -Mittelalterliche Studien: Ausgewählte Aufsätze zur Schriftkunde und Literaturgeschichte, 3
vols. (Stuttgart, 1966, 1967, 1981). -'Frühkarolingische Handschriften und ihre Heimat', Scriptorium, 22 (1968), 306-14
[index to Bischoff 1965].
xl
Bibliography
Bischoff, Bernhard, Die südostdeutschen Schreibschulen und Bibliotheken in der Karo-
lingerseit, 2 vols. (3rd cdn., Wiesbaden, 1974, 1980). -Paläographie des römischen Altertums und des abendländischen Mittelalters (Grundlagen
der Germanistik, 24; Berlin, 1979; 2nd edn., 1986).
-Kalligraphie in Bayern: 8.-12. Jahrhundert (Wiesbaden, 1981).
-see also Carmina Burana.
Bishop, Edmund, Liturgica histotica (Oxford, 1918).
Bishop, William Chatterly, The Mozarabic and Ambrosian Rites: Four Essays in Comparative Liturgiology (Alcuin Club Tracts, 15; London, 1924).
Bjork, David, 'Early Repertories of the Kyrie Eleison', Ktnjb 63-4 (1979-80), 9-43.
-'On the Dissemination of Quem quaeritis and the Visitatio Sepulchri and the Chronology
of their Early Sources', Comparative Drama, 14 (1980), 46-69.
-'Early Settings of the Kyrie Eleison and the Problem of Genre Definition', JPMMS 3
(1980), 40-8.
-'The Kyrie Trope', JAMS 33 (1980), 1-41.
-'The Early Frankish Kyrie Text: A Reappraisal', Viator, 12 (1981), 9-35.
Björkvall, Gunilla, 'Offertory Prosulas for Advent in Italian and Aquitanian Manuscripts', in Cantus Planus 1990, 377-400.
-and Steiner, Ruth, 'Some Prosulas for Offertory Antiphons', JPMMS 5 (1982), 17-35.
-see also CT.
Blachley, Alex, 'Some Observations on the "Germanic" Plainchant Tradition', in Sanders
Festschrift 1990, 85-117. Blanchard, P., 'La Correspondance apocryphe du pape S. Damase et de S. Jerome sur le
psautier et le chant de 1"Alleluia'", EL 63 (1949), 376-88. Blezzard, Judith, Ryle, Stephen, and Alexander, Jonathan, 'New Perspectives on the
Feast of the Crown of Thorns', JPMMS 10 (1987), 23-47. Bloch, Herbert, Monte Cassino in the Middle Ages, 3 vols. (Rome, 1986). Blume see All.
Boe, John, 'A New Source for Old Beneventan Chant: The Santa Sophia Maundy in MS
Ottoboni lat. 145', AcM 52 (1980), 122-33. -'Gloria A and the Roman Easter Vigil', Ml) 36 (1982), 5-37.
-'The Beneventan Apostrophus in South Italian Notation, a.d. 1000-1100', EMH 3
(1983), 43-66.
-'The "Lost" Palimpsest Kyrics in the Vatican Manuscript Urbinas latinus 602', JPMMS
8 (1985), 1-24.
-'Hymns and Poems at Mass in Eleventh-Century Southern Italy (other than Sequences)',
in Congress Bologna 1987, iii. 515-41. -Beneventanum Troporum Corpus II: Ordinary Chants and Tropes for the Mass from
Southern Italy, A.D. 1000-1250. Part 1: Kyrie eleison (Madison, Wise, 1989); Part 2:
Gloria in excelsis (Madison, Wise, 1990). -'Italian and Roman Verses for Kyrie leyson in the MSS Cologne-Geneve, Bibliotheca
Bodmeriana 74 and Vaticanus latinus 5319', in Leonardi and Menesto 1990, 337-84. Boethius, De institutione anthmetica libri duo. De institutione musica libri quinque, ed.
G. Friedlein (Leipzig, 1867). -'Fundamentals of Music' by Anicius Manlius Severinus Boethius, trans, with an
introduction and notes by C. M. Bower, ed. bv Claude V. Palisca (New Haven, Conn.,
1989).
Bibliography
xli
Bomm, Urbanus, Der Wechsel der Modalitätbestimmung in der Tradition der Meßgesänge im IX. bis XIII. Jahrhundert und sein Einfluß auf die Tradition ihrer Melodien (Diss., Göttingen, 1928; Einsiedeln, 1929).
Bona, J., Herum liturgie a mm libri duo (Rome, 1671; Paris, 1672).
Bonastre, F., 'Estudis sobre la verbeta: la verbeta a Catalunya durant els segles XI-XVI'
(Diss., Tarragona, 1982). Bonniwell, William R., A History of the Dominican Liturgy (New York, 1944). Boor, Helmut de, Die Textgeschichte der lateinischen Osterfeiern (Tübingen, 1967). Borella, Pietro, // rito ambrosiano (Brescia, 1964).
van den Borren Festschrift 1945 = S. Clercx and A. van der Linden (eds.), Hommage ä
Charles van den Borren (Antwerp, 1945). Borsai, Ilona, 'Coptic Rite, Music of the', NG.
Bosse, Detlev,  Untersuchung einstimmiger mittelalterlicher Melodien zum 'Gloria in
excelsis deo' (Regensburg, 1955). Botte, Bernard, lx?s Origines de la Noel et de l'Epiphanie (Louvain, 1932). Boutemy Festschrift 1976 = Guy Cambier (ed.), Ilommages ä André Boutemy (Collection
Latomus, 145; Brussels, 1976). Bouyer, Louis, La Vie de la liturgie (Paris, 1956); trans.: Life and Liturgy (London, 1956). -Le Rite et l'homme, sacralité naturelle et liturgie (Paris, 1962); trans.: Rite and Man
(London, 1963).
Bower, Calvin M., 'Boethius' "The Principles of Music", an Introduction, Translation, and Commentary' (Diss., George Peabody College, 1967; UMI 67-15005).
-'Natural and Artificial Music: The Origins and Development of an Aesthetic Concept',
MD 25 (1971), 17-33.
-'Boethius and Nicomachus: An Essay Concerning the Sources of De Institutione
Musica', Vivarium, 16 (1978), 1-45.
-'An Alleluia for Mater', in Herz Festschrift 1982, 98-116.
-'The Modes of Boethius', Journal of Musicology, 3 (1984), 252-63.
-'Boethius' "De institutione musica". A Hand-List of Manuscripts', Scriptorium, 42
(1988), 205-51.
-'The Grammatical Model of Musical Understanding in the Middle Ages', in Gallacher
and Damico 1989, 133-45. Boyce, James John, 'The Office of St Mary of Salome', JPMMS 11 (1988), 25-47.
-'The Office of the Three Marys in the Carmelite Liturgy', JPMMS 12 (1989), 1-38.
-'The Medieval Carmelite Office Tradition', AcM 62 (1990), 119-51.
Boyle, Leonard E., Medieval Latin Palaeography: A Bibliographical Introduction (Toronto
Medieval Bibliographies, 8; Toronto, 1984). Bradshaw, Paul, Daily Prayer in the Early Church (Alcuin Club Collections, 63; London,
1981).
Braganqa, Joaquim O., Missal de Mateus: Manuscrito 1000 da Biblioteca Publica e Arquivo Distrital de Braga (Lisbon, 1975).
-Processionale-Tropario de' Alcobaca (Lisbon, 1985).
Bragard see CSM 3.
Branner, Robert, 'Two Parisian Capella Books in Bari', Gesta, 8 (1969), 14-19.
-Manuscript Painting in Paris during the Reign of Saint Louis (Berkeley, Calif., 1977).
Braunfels, Wolfgang (ed.), Karl der Große: Lebenswerk und Nachleben, 5 vols. (Düsseldorf, 1965-8).
xlii Bibliography
Breviári um Nidrosiense (Parisius 1519) (fac. edn., Oslo, 1964).
Briggs, Henry B., The Musical Notation of the Middle Ages Exemplified by Eacsimilcs of Manuscripts Written between the 10th and 16th Centuries Inclusive (London, 1890).
Brockett, Clyde Waring, Antiphons, Responsories, and other Chants of the Mozarabic Rite (Brooklyn, 1968).
-'Unpublished Antiphons and Antiphon Series Found in the Gradual of St. Yrieix', Ml)
26 (1972), 67-94.
-'Easter Monday Antiphons and the Peregrinus Play', Kmjfb 61-2 (1977-8). 29-46.
Bronarski, L., 'Die Quadripartita rlgura in der mittelalterlichen Musiktheorie', in Wagner
Festschrift 1926, 27-43. Brou, Louis, 'Les Impropéres du Vcndredi-Samť, RG 20 (1935), 161-79; 21 (1936), 8-16;
22 (1937), 19, 44-51.
-'ĽAlléluia gréco-latin "Dies sanctificatus" de la messe du jour de Noél: Origine et
evolution ďun chant bilingue et protéiforme', RG 23 (1938), 170-5; 24 (1939), 1-8, 81-9, 202-13.
-'Le Répons "Ecce quomodo moritur" dans les traditions romaine et espagnole', RB 41
(1939), 144-68.
-'Le Psallendum de la messe et les chants connexes ďaprés les sources manuscrites', EL 61
(1947), 13-54.
-'Etudes sur la liturgie mozarabe: Le Trisagion de la messe d'aprés les sources
manuscrites', EL 61 (1947), 309-34. -'Les "Benedictiones" ou cantique des trois enfants dans ľancienne messe espagnole',
Hispánia sacra, 1 (1948), 21-33. -'Les Chants en langue grecque dans les liturgies latines', Sacris erudiri, 1 (1948), 165—
80; 4 (1952), 226-38.
-'UAntiphonaire wisigothique et ľantiphonaire grégorien au debut du vmc siécle. Essai
de musicologie comparée', Anuario musical, 5 (1950), 3-10. -'ĽAlléluia dans la liturgie mozarabe. Etude liturgico-musicale ďaprés les manuscrits',
Anuario musical, 6 (1951), 3-90.
-'Sequences et tropes dans la liturgie mozarabe', Hispánia sacra, 4 (1951), 1-15.
-'Notes de paléographie musicale mozarabe', Anuario musical, 7 (1952), 51-76; 10
(1955), 23-44.
-'Etudes sur le missel ct le bréviaire "mozarabes" imprimés', Hispánia sacra, 11 (1958),
1-50.
-'L'Ancien Office de saint Vaast, évéque d'Arras', EG 4 (1961), 7-42.
-and Vives, José, Antifonario visigótico mozarabe de la Gate dral de I^eón: Edición del
texto, notas e indices (Monumenta hispaniae sacra. Serie litúrgica, v/1; Barcelona and
Madrid, 1959).
Brovelli, F., 'Per uno studio dei messali francesi del XVIII" secolo, saggio di analisi', EL 96 (1982), 279-406.
Brown, Julian, Patterson, Sonia, and Hiley, David, 'Further Observations on Wľ,
JPMMS 4 (1981), 53-80. Browne, Alma Colk, 'The a-p System of Letter Notation', MD 35 (1981), 5-54. Brunhölzl, Franz, 'Zur Antiphon "Alma redemptoris mater'", Studien und Mitteilungen zur
Geschichte des Benediktinischen Ordens und seiner Zweige, 78 (1967), 321-4. -Geschichte der lateinischen Literatur des Mittelalters, i (Munich, 1975).
Bibliography
xliii
Brunner, Lance W., 'A Perspective on the Southern Italian Sequence: The Second Tonary
of the Manuscript Monte Cassino 318', EMU 1 (1981), 117-64. -'The Performance of Plainchant: Some Preliminary Observations of the New Era', Early
Music, 10 (1982), 317-28.
-'Catalogo delle sequenze in manoscritti di origine italiana', RIM 20 (1985), 191-276.
Bryant, David, 'Aquileia', NG.
Büchner, M., 'Die "Vita Chrodegangi": Eine kirchenpolitische Tendenzschrift aus der Mitte des 9. Jahrhunderts, zugleich eine Untersuchung zur Entwicklung der Primatial- und Vikariatsidee', Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Kanonische Abteilung, 16 (1927), 1-36.
Bujíc, Bojan, 'Zadarski neumatski fragmenti v Oxfordu/Neumatic Fragments of Zadar in
Oxford', Muzikološki zbornik, 4 (1968), 28. Bukofzer, Manfred, 'Speculative Thinking in Medieval Music', Speculum, 17 (1942), 165-
80.
Bulst, Neithart, Untersuchungen zu den Klosterreformen Wilhelms von Dijon (962-1031)
(Pariser historische Studien, 11; Bonn, 1973). -'Rodulfus Glabers Vita domni Willelmi abbatis. Neue Edition nach einer Handschrift
des 11. Jahrhunderts (Paris, Bibl. nat., lat. 5390)', Deutsches Archiv für Erforschung des
Mittelalters, 30 (1974), 450-87. Bulst Festschrift 1960 = Hans Robert Jauss (ed.), Medium aevum vivum: Festschrift für
Walter Bulst (Heidelberg, 1960). Buschinger, D., and Crépin, A. (eds.), Musique, littérature et societě au MoyenAge (Paris,
1980).
Bužga, Jaroslav, 'Kantional (tschechisch)', MGG. Cabaniss, A., Amalarius of Metz (Amsterdam, 1954).
Cabrol, Fernand, Les Livres de la liturgie latine (Paris, 1930); trans.: The Books of the Latin
Liturgy (London, 1932). -'Mozarabe (La liturgie)', DACL.
Calcidius, Platon, Timaeus. A Calcidio translatus comentarioque instructus, ed. J. Waszink
(London and Leiden, 1962). Caldwell, John, 'The De Institutione Arithmetica and the De Institutione Musica', in
Gibson 1981, 135-54.
Callewaert, Camille, 'S. Grégoire, les scrutins et quelques messes quadragésimales', EL 53
(1939), 191-203 = Callewaert 1940, 659-71. -Sacris erudiri (Steenbrugge, 1940).
Campbell, Thomas P., and Davidson, Clifford (eds.), The Fleury Playbook: Essays and Studies (Kalamazoo, Mich., 1985).
Canal, J. M., 'Sobre el autor del antifonario cisterciense', EL 74 (1960), 36-47.
Cantus Planus 1990 = International Musicological Society Study Group 'Cantus Planus'. Papers Read at the Third Meeting, Tihany, Hungary, 19-24 September 1988, ed. László Dobszay, Peter Halász, Jánoš Mezei, and Gábor Prószéky (Budapest, 1990).
Cantus selecti ex libris Vaticanis et Solesmensibus excerpti (Solesmes, 1949).
CAO 1 (1963) = Manuscripti 'cursus romanus' [Bamberg, Staatsbibliothek, lit. 23; Paris, Bibliothěque Nationale, lat. 17436; Durham, Cathedral Chapter Library, B. iii. 11; Ivrea, Biblioteca Capitolare, 106; Monza, Basilica S. Giovanni, C. 12/75; Verona, Biblioteca Capitolare, 98].
xliv
Bibliography
(AO 2 (1965) = Mamiscripti 'atrsus monasticus' [St. Gall, Stifsbibliothek, 390-391; Zürich, Zentralbibliothek, Rheinau 28; Paris, Bibliotheque Nationale, lat. 17296; Paris, Bibliotheque Nationale, lat. 12584; London, British Library, Add. 30850; Benevento, Archivio Capitolare, 21].
CAO 3 (1968) = Invitatoria et antiphonae.
CAO 4 (1970) = Responsona, z:ersus, liymtii, varia.
CAO 5 (1975) = Pontes earurnque prima ordinatio.
CAO 6 (1979) = Secunda et tertia ordinationes.
CAO-ECE = Corpus Antiphonalium Officii—Ecclesiarum Centralis Europae: A Preliminary Report, ed. László Dobszay and Gábor Prószéky (Budapest, 1988).
CAO-ECE IIA Salzburg (Temporale), ed. László Dobszay (Budapest, 1990).
Capelle, B., 'Le Kyrie de la messe et le pape Gélase', RB 46 (1934), 126-44.
Cardine, Eugene, 'La Psalmodie des introits', RG 26 (1947), 172-7, 229-36; 27 (1948), 16-25.
-'La Corde recitative du 3C ton psalmodique dans l'ancienne tradition sangallienne', EG 1
(1954), 47-52.
-'Le Chant grégorien est-il mesuré?', EG 6 (1963), 7-38; trans. A. Dean: Is Gregorian
Chant Measured Music? (Solesmes, 1964). -Graduel neurné (Solesmes, 1966).
-Semiologia gregoriana (Rome, 1968); French trans.: 'Sémiologie grégorienne', EG 11
(1970), 1158; trans. Robert M. Fowels: Gregorian Semiology (Solesmes, 1982). -'Vue d'ensemble sur le chant grégorien', EG 16 (1977), 173-92.
Cardine Festschrift 1980 = Johannes Berchmans Göschl (ed.), IJt mens concordet roci: Festschrift Eugene Cardine zum 75. Geburtstag (St. Ottilien, 1980).
Cannina Burana: Faksimile-Ausgabe der Iis. Clin 4660 + 4660a, ed. Bernhard Bischoff (Brooklyn, 1967).
Cassiodorus, Institutiones divinanan et humananun rerum, ed. R. A. B. Mynors (Oxford, 1937).
Castiglioni Festschrift 1957 = Studi in onore di Möns. Carlo Castiglioni (Milan, 1957). Castro, Eva, 'Le Long Chemin de Moissac ä S. Millan (Lc troparium de la Real Acad. I list-,
Aemil 51)', in Leonan'- and Menesto 1990, 243-63. Cattaneo, Enrico, 'i  ranti della frazione e comunione nella liturgia ambrosiana', in
Mohlberg Festschrift ' '48, i. 147-74.
-Note storiche sul i>   to ambrosiano (Milan, 1950).
Cavallo, Guglielmo, .. toli di Exultet dell'Italia meridionale (Bari, 1973).
CCM 9 (1976) = Cons:   udines Floriacenses saeculi tertii deeimi, ed. Anselmo Davril.
CCM 10 (1980) = Libi    Yamitis aevi Odilonis Abbatis, ed. Peter Dinter.
CCM 12 (2 vols., 1975, K87) — Consuetudines Fructuarienses-Sanblasianae, ed. Luchesius
G. Spätling and Pete )inter. Censorinus, De die i. • /// Uber ad O. Caerellium, accedit Anonymi cujusdam Epitoma
diseiplinarum (Eragr   itum Censorini), ed. Nicolas Sallmann (Leipzig, 1983).
-Lť jour natal, trat    lion annotée, trans. Guillaume Rocca-Serra (Paris, 1980).
Chadwick, Menry, Bo   úus: The Consolations of Music, Logic, Theology, and Philosophy
(Oxford, 1981).
Chailley, Jacques, 'L . Jocument sur la danse eeclésiastique', AcM 21 (1949), 18-24. -'Le Mythe des mc s s grecs', AcM 28 (1956), 137-63.
Bibliography xlv
---'Les Anciens Tropaires et séqucntiaires de l'Ecolc de Saint-Martial de Limoges (xc-xic
siede)', EG 2 (1957), 163-88. -- L'Imbroglio des modes (Paris, 1960).
-L'Ecole musicale de Saint-Martial de Limoges jusqu'a la fin du XT siěcle (Paris, 1960).
-.4//« Musica (Tratte de musique du /Ac siěcle). Edition critique commentée avec une
introduction sur I'origme de la nomenclature modale pseudo-grecque au Moyen-Age (Paris,
1965).
Chambers, E. K., The Mediaeval Stage, 2 vols. (Oxford, 1903). Charlesworth, K., The Odes of Solomon (2nd edn., Missoula, Mont., 1977). Chartier, Yves, 'L'Epistola dc armonica institutione de Réginon de Prüm' (Diss., Ottawa, 1965).
-'La Musica d'Hucbald de Saint-Amand (traité de musique du xic siěcle). Introduction,
établissement du texte, traduction et commentaire' (Diss., Sorbonne, 1973).
-'Hucbald de Saint-Amand et la notation musicale', in Huglo 1987, 145-55.
Chavanon, Jules, Adémar de Chabannes: Chronique (Collection des Textes; Paris, 1897).
Chavasse, Antoine, 'Le Caréme romain et les scrutins prébaptismaux avant le ixc siěcle', Recherches de science religieuse, 35 (1948), 325-81.
-'Les Plus Anciens Types du lectionnaire et de l'antiphonaire romain de la messe', RB 62
(1952), 3-94.
-'Cantatorium et Antiphonale missarum. Quelques procédés de confection. L'Antiphonaire
des dimanches aprěs la Pentecňte. Les graduels du Sanctoral', Ecclesia orans, 1 (1984), 15-55.
Chevalier, Ulysse, Repertorium hymnologicum. Catalogue des chants, hytnnes, proses,
sequences, tropes en usage dans 1'église latine depuis les origines jusqu'a nos jours, 6 vols.
(Louvain and Brussels, 1892-1920). -Prosolarium Anicensis: Office en vers de la Circoncision en usage dans 1'église du Puy
(Bibliothěque liturgique, 5; Paris, 1894). -Sacramentaire et marlyrologe de l'abbaye de Saint-Rémy: Marty rologe, calendiier,
ordinaires et piosaire de la metropole de Reims (VIIF-XHF siecle), publiés ďaprěs les
manuscrits de Paris, Ijondres, Reims et Assise (Bibliothěque liturgique, 7; Paris, 1900). -Ordinaire et coutumier de 1'église cathédrale de Bayeux (XIIT r'ěcle), publiés ďaprěs les
manuscrits originaux (Bibliothěque liturgique, 8; Paris, 1902). Chibnall, Marjorie, The Ecclesiastical History of Orderte Vital. ■   6 vols. (Oxford, 1969-
80).
Choisselet, Daniele, and Vernet, Placide (eds.), Les Ecclesia-ira Officio cisterciens du
XIF siecle (Oelenberg, 1989). Chomton, L., Histoire de 1'église Saint-Bénigne de Dijon (Dijon, >00). Claire, Jean, 'L'Evolution modale dans les repertoires liturgie . s occidentaux', RG 40
(1962), 196-211, 229-45.
-'La Psalmodie résponsonale antique', RG 41 (1963), 8-29, 4 J  >2, 77-102.
-'L'Evolution modale dans les récitatifs liturgiques', RG 41 (1 1 5), 127-51.
-'Les Repertoires liturgiques latins avant 1'octoéchos. I. L'offio :érial romano-franc', EG
15 (1975), 5-192.
-'The Tonus peregrinus—A Question Well Put?', Orbis music    Studies in Musicology,
7 (1979-80), 3-14.
-'Les Psaumes graduels au cceur de la liturgie quadragésimale'. . ,'G 21 (1986), 5—12.
xlvi
Bibliography
Claire, Jean, 'Lc Rituel quadragesimal des catechumencs ä Milan', in Gy Festschrift 1990, 131-51.
Clark, j. M., The Abbey of St. Gall as a Centre of Literature and Art (Cambridge, 1926). Clemoes Festschrift = Michael Lapidgc and Helmut Gneuss (eds.), learning and Literature
in Anglo-Saxon England (Cambridge, 1985). CM 1 (Stuttgart and Vienna, 1900) = Consuetudines Farfenses. CM 2 (Montecassino, 1905) = Consuetudines Cluniacenses antiquiores. CM 3 (Montecassino, 1907) = Statuta Murbacensia, Capitula Aquisgranensia. CM 4 (Montecassino, 1911) = Consuetudines Fructuarienses (etc.). CM 5 (Montecassino, 1912) = Redactio consuetudinem saeculi duodecimi (etc.). Cocheril, Maur, L'Evolution historique du kyriale cistercien (Port-du-Salut, 1956). -'Le "Tonale Sancti Bernardi" et la definition du "ton"',  Citeaux: Commentarii
cistercienses, 13 (1962), 35-66. Codex Albensis, ein Antiphonar aus dem 12. Jahrhundert, cd. Zoltän Falvv and L. Mezcv (Monu-
menta Hungariae musicae, 1; Budapest, 1963) [fac. of Graz, Universitätsbibliothek, 211]. Cody, Aelred, 'The Early History of the Octoechos in Syria', in Garsoian, Mathews, and
Thomson 1982, 89-113. Colette, Marie-Noel, 'La Notation du demi-ton dans le manuscrit Paris, B.N. Lat. 1139 et
dans quelques manuscrits du Sud de la France', in Leonardi and Menesto 1990, 297-311. Colgrave, Bertram, The Earliest Life of Gregory the Great by an Anonymous Monk of Whitby
(Lawrence, Kansas, 1968). Collins, A. Jefferies, Manuale ad usum percelebris ecclesie Sarisburiensis (Henry
Bradshaw Society, 91; Chichester, 1960). -The Bridgettine Breviary of Syon Abbey: From the MS. with English Rubrics F. 4. 11 at
Magdalene College, Cambridge (Henry Bradshaw Society, 96; Worcester, 1969). Combe, Pierre, Histoire de la restauration du chant gregorien d'apres des documents inedits:
Solesmes et Vedition vaticane (Solesmes, 1969). Congress Berkeley 1977 = Daniel Heartz and Bonnie Wade (eds.), International Musicological
Society: Report of the Twelfth Congress Berkeley 1977 (Kassel, 1981). Congress  Berlin   1974  =   Helmut  Kühn  and  Peter  Nitsche  (eds.),  Gesellschaft für
Musikforschung: Bericht über den internationalen musikwissenschaftlichen Kongreß Berlin
1974 (Kassel, 1977).
Congress Bologna 1987 = Atti del XIV congresso della Societä Internazionale di Musicologia: Trasmissione e reeezione delle forme di cultura musicale, Bologna, 27 agosto-1 settembre 1987, ed. Angelo Pompilio, Donatella Restani, Lorenzo Bianconi, and F. Alberto Gallo, 3 vols. (Rome, 1990).
Congress Cluny 1949 = A Cluny: Congres seien tifique: Fetes et ceremofiies liturgiques en
Vhonneur des saints abbes Odon et Odilon 9-11 jutilet 1949 (Dijon, 1950). Congress Cologne 1958 = G. Abraham, S. Clercx-Lejeune, H. Federhofer, and W. Pfannkuch
(eds.), International Musicological Society: Bericht über den siebenten internationalen
musikwissenschaftlichen Kongress Köln (Kassel, 1959). Congress Copenhagen 1972 — International Musicological Society: Report of the Eleventh
Congress Copenhagen 1972 (Copenhagen, 1974). Congress Hamburg 1956 = W. Gerstenberg, H. Husmann, and II. Heckmann (eds.),
Gesellschaft für Musikforschung: Bericht über den internationalen musikwissenschaftlichen
Kongreß Hamburg 1956 (Kassel, 1957).
Bibliography
xlvii
Congress Kassel 1962 = G. Reichert and M. Just (eds.), Gesellschaft für Musikforschung: Bericht über den internationalen musikwissenschaftlichen Kongreß Kassel 1962 (Kassel, 1962).
Congress Leipzig 1925 = Beucht üben den I. musikwissenschaftlichen Kongreß der deutschen Musikgesellschaft in Leipzig vom 4. bis 8. Juni 1925 (Leipzig, 1926).
Congress New York 1961 = Jan LaRue (ed.), International Musicological Society: Report of the Eighth Congress New York 1961 (Kassel, 1962).
Congress Salamanca 1985 = Espaňa en la Müsica de Occidente: Actas del Congreso Internacionál celebrado en Salamanca (29 de octubre-5 de noviembre de 1985), ed. Emilio Casares Rodicio, Ismael Fernandez de la Cuesta, and José López-Calo, 2 vols. (Madrid, 1987).
Congress Strasburg 1982 = Marc Honegger, Christian Meyer, and Paul Prévost (eds.), La
Musique et le rite sacré et profane: Actes du XIIE Congrés de la Société Internationale de
Musicologie Strasbourg 1982 , 2 vols. (Strasburg, 1986). Congress Todi 1958 = Spiritualita cluniacense, 12-15 ottobre 1958 (Convegni del Centro di
studi sulla spiritualita medievale, 2; Todi, 1960). Congress Utrecht 1952 = International Musicological Society, Report of the Fifth Congress
Utrecht 1952 (Amsterdam, 1953). Congress Vienna 1954 = Zweiter internationaler Kongreß für katholische Kirchenmusik Wien
4.-10. Oktober 1954: Bericht (Vienna, 1956). Congress Vienna 1956 = Erich Schenk (ed.), Gesellschaft für Musikforschung: Bericht über
den internationalen musikwissenschaftlichen Kongreß Wien, Mozartjahr 1956 (Graz and
Cologne, 1958).
Connolly, Thomas H., 'The Introits and Communions of the Old Roman Chant' (Diss.,
Harvard University, 1972). -'Introits and Archetypes: Some Archaisms of the Old Roman Chant', JAMS 25 (1972),
157-74.
-'The Graduale of S. Cecilia in Trastevere and the Old Roman Tradition', JAMS 28
(1975), 413-58.
-'Musical Observance of Time in Early Roman Chant', in Otto Albrecht Festschrift 1980,
3-18. — 'Psalm H', NG.
Constable, Giles, Medieval Monasticism: A Select Bibliography (Toronto Medieval
Bibliographies, 6; Toronto, 1976). Constitution on the Sacred Liturgy of the Second Vatican Council, in Austin P. Flannery (ed.),
Documents of Vatican II (Grand Rapids, Mich., 1975; 2nd edn., 1984), 1-282. Contreni, John, The Cathedral School of Laon from 850-930: Its Manuscripts and Masters
(Munich, 1978).
Conybeare, Frederick C. (ed.), Philo: About the Contemplative Life or the Fourth Book of the
Treatise Concerning Virtues (London, 1895). Corbin, Solange, La Musique religieuse portugaise au Moyen-Age (Paris, 1952). -'Le Cantus Sibyllae: Origines et premiers textes', Revue de musicologie, 34 (1952), 1-
10.
-'Le Manuscrit 201 d'Orléans: Drames liturgiques dits de Fleury', Romania, 74 (1953),
1-43.
-'Valeur et sens de la notation alphabétique ä Jumiéges et en Normandie', in Jumiéges:
Congrés scientifique du Xllľ centenaire, ii (Rouen, 1955), 913-24.
xlviii Bibliography
Corbin, Solange, Im Deposition liturgique du Christ ou Vendredi Saint: So Place dans l'histoire des tites et du theatre religieux. Analyse des documents portuguais (Paris, 1960).
-L'Eglise ä la conquete de sa musique (Paris, 1960).
-Die Neumen (Palaeographie der Musik, i/3; Cologne, 1977).
-'Neumatic Notations, I-IV, NG.
Cornford, Frances Macdonald, Plato's Cosmology: The Timaeus of Plato Translated with
a Running Commentary (London, 1937). Coussemaker, Edmond de, Histoire de l'harmonie au moyen-äge (Paris, 1852).
-Drarnes liturgiques du moyen-äge (Paris, 1861).
-see also CS.
Cowdrey, Herbert E. J., The Cluniacs and the Gregorian Reform (Oxford, 1970).
-The Age of Abbot Desiderius: Montecassino, the Papacy and the Normans in the Eleventh
and Early Twelfth Centuries (Oxford, 1983). Crocker, Richard L., 'The Repertoire of Proses at Saint Martial de Limoges (Tenth and
Eleventh Centuries)' (Diss., Yale University, 1957; UMI 68-4915). -'The Repertory of Proses at Saint Martial de Limoges in the 10th Century', JAMS 11
(1958), 149-64.
-'Musica rhythmica and Musica metrica in Antique and Medieval Theory', Journal of
Music Theoty, 2 (1958), 2-23. -'Discant, Counterpoint and Harmony', JAMS 15 (1962), 1-21.
-'Pythagorean Mathematics and Music', Journal of Aesthetics and Art Criticism, 22
(1963), 189-98, 325-35.
-'The Troping Hypothesis', MO 52 (1966), 183-203.
-'Aristoxenus and Greek Mathematics', in Reese Festschrift 1967, 96-110.
-'A New Source for Medieval Music Theory', AcM 39 (1967), 161-71.
-'Some Ninth-Century Sequences', JAMS 20 (1967), 367-402.
-'Hermann's Major Sixth', JAMS IS (1972), 19-37.
-'The Sequence', in Schrade Gedenkschrift 1973, 269-322.
-The Early Medieval Sequence (Berkeley, Calif., 1977).
-'Alphabet Notations for Early Medieval Music', in Charles Jones Festschrift 1979, ii. 79-
104.
-'Matins Antiphons at St. Denis', JAMS 39 (1986), 441-90.
-arts. 'Agnus Dei', 'Credo', 'Gloria in excclsis Deo', 'Kyrie eleison', 'Sanctus', 'Sequence
(i)\ 'Versus', NG.
-and Hiley, David (eds.), New Oxford History of Music, ii. The Early Middle Ages to
1300 (2nd edn., Oxford, 1989). Cserba, Simon M. (ed.), Hieronymus de Moravia O. P., Tractatus de musica (Freiburgcr
Studien zur Musikwissenschaft, 2; Regensburg, 1935). CSM 1 (1950) = Johannis Affligemensis De musica cum tonario, ed. Joseph Smits van
Waesberghe.
CSM 2 (1951) = Aribonis De musica, ed. Joseph Smits van Waesberghe. CSM 3 (7 vols., 1955-73) = Jacobi Leodiensis Speculum rnusicae, cd. Roger Bragard. CSM 4 (1955) = Guidonis Aretini Micrologics, ed. Joseph Smits van Waesberghe. CSM 14 (1970) = Walte/i Odington Summa de speculatione rnusicae, ed. Frederick F. Hammond.
CSM 21 (1975) = Aureliani Reomensis Musica disciplina, ed. Lawrence Gushee. CSM 23 (1975) = Willehelmi Hirsaugensis Musica, ed. Denis Harbinson.
Bibliography
xlix
CSM 24 (1974) - Epistola S. Bemardi de revisione cantus Cisterciensis el Tractatus scriptus ab auctore incerto Cisterciense, Cantum quern Cisterciensis ordinis ecclesiae cantare consueverant, ed. Francis J. Gucntncr.
CSM 25 (1975) = Ameri Practica artis musice, ed. Cesarino Ruini.
CSM 29 (1976) = Petrus de Cruce AtnbianensifsJ: Tractatus de Tonis, ed. Denis Harbinson. CT 1 (SLS 21, 1975) = Tropes du propre de la messe 1: Cycle de Noel, ed. Ritva Jonsson. CT 2 (SLS 22, 1976) = Prosides de la messe T. Tropes de l'alleluia, ed. Olof Marcusson. CT 3 (SLS 25, 1982) = Tropes du propre de la messe 2: Cycle de Päques, ed. Gunilla
Björkvall, Gunilla Iversen, and Ritva Jonsson. CT 4 (SLS 26, 1980) = Tropes de lAgnus Dei, ed. Gunilla Iversen. CT 5 (SLS 32, 1986) = Les Deux Tropaires d'Apt, mss. 17 et 18, ed. Gunilla Björkvall. CT 6 (SLS 31, 1986) = Prosules de la messe 2: I^es prosules limousmes de Wolfenbüttel, ed.
Eva Odelman.
CT 7 (SLS 34, 1990) = Tropes du Sanctus, ed. Gunilla Iversen.
Curran, M., The Antiphonaiy of Bangor and the Early Irish Monastic Liturgy (Dublin, 1984). Cutter, Paul F., 'The Question of the "Old-Roman" Chant: A Reappraisal', AcM 39 (1967), 2-20.
-'The Old-Roman Chant Tradition: Oral or Written?', JAMS 20 (1967), 167-81.
-'The Old-Roman Responsories of Mode 2' (Diss., Princeton University, 1969; UMI 70—
08358).
-'Die altrömischen und gregorianischen Responsorien im zweiten Modus', Kmjb 54
(1970), 33-40.
-'Oral Transmission of the Old-Roman Responsories?', MQ 62 (1976), 182-94.
-Musical Sources of the Old-Roman Mass (Musicological Studies and Documents, 36;
Neuhausen-Stuttgart, 1979). [-] 'Responsory', NG.
Dauelsen Festschrift 1978 = Thomas Kohlhase and Volker Scherliess (eds.), Pestschrift
Georg von Dadelsen zum 60. Geburtstag (Neuhausen-Stuttgart, 1978). Dahlhaus Festschrift 1988 = II. Danuser, H. de la Motte-Haber, S. Leopold, and N. Miller
(eds.),  Das  musikalische  Kunstwerk:   Geschichte-Asthetik-Theorie.   Festschrift Carl
Dahlhaus zum 60. Geburtstag (Laaber, 1988). Dalton, J. N. and (vol. iv) Düble, G., Ordinale Exon (Henry Bradshaw Society, 37, 38, 63,
and 79; London, 1909, 1909, 1926, and 1941). Damilano, Piero, 'Laudi latine in un antifonario bobbiese del trecento', Collectanea historiae
musicae, 3 (1963), 15-57. -'Sequenze bobbiese', R1M 2 (1967), 3-35.
D'Angers, O., 'Le Chant liturgique dans l'Ordre de Saint-Frangois aux origines', Etudes
franciscaines, 25 (1975), 157-306. David, Lucien, and Handschin, Jacques, 'Un point d'histoire gregorienne. Guillaume de
Fecamp', RCG 39 (1935-6), 180-3; 40 (1936-7), 11-17 (see also Handschin 1936-8). Davies, John G. (ed.), A Dictionary of Liturgy and Worship (London, 1972). -A New Dictionary! of Liturgy and Worship (London, 1986).
Davril, Anselme, 'Ä propos d'un breviaire manuscrit de Cluny conserve ä Saint-Victor-sur-
Rhins', RB 93 (1983), 108-22. -'Johann Drumbl and the Origin of the Quern quaeritis: A Review Article', Comparative
Drama, 20 (1986), 65-75. Deanesly, Margaret, Augustine of Canterbury (London, 1964).
1 Bibliography
Dechevrens, Antoine, Etudes de science musicale, 3 vols. (Paris, 1898).
-Composition musicale et composition littéraire á propos du chant grégonen, 2 vols. (Paris,
1910).
Delalande, Dominique, Vers la version authentique du Graduel grégorien: Le Graduel des
Prěcheurs (Bibliothěque d'histoirc Dominicaine, 2; Paris, 1949). Delaporte, Yves, 'Fulbert de Chartres et l'ecole chartraine de chant liturgique au xic siecle',
EG 2 (1957), 51-81.
-'L'Office fécampois de saint Taurin', L'Abbaye bénédictine de Fecamp, out rage
scientifique du XIIF centenaue, 658-1958 (Fecamp, 1959-60), ii. 171-89, 377. Deléglise, Francois, '"Illustris civitas", office rimé de saint Théodule (xinc siěcle)',
Vallesia, 38 (Sion, 1983), 173-308. Deshusses, Jean, Ijj Sacramentaire grégorien: Ses principales formes ďaprés les plus anciens
manuscrits (Spicilegium Friburgense, 16; Fribourg, 1971; 2nd edn., 1979), and 24 (1979). -and Darragon, B., Concordances et tableaux pour Vétude des grands sacramentaires
(Spicilegii Friburgensis Subsidia, 9-14; Fribourg, 1982-3). Dewick, E. S., and (vol. ii) Frere, W. II., The I^ofric Collector (Henry Bradshaw Society,
45 and 56; London, 1914, 1921). Dickinson, Francis Henry (ed.), Missale ad usum insignis et praeclarae ecclesiae Sarum
(Burntisland, 1861-83). van Dijk, S. J. P., 'St Bernard and the Instituta Patmm of St Gall', MD 4 (1950), 99-109. -'The Legend of "the Missal of the Papal Chapel" and the Fact of Cardinal Orsini's
Reform', Sacris erudiri, 8 (1956), 76-142.
-'The Authentic Missal of the Papal Chapel', Scnptonum, 14 (1960), 257-314.
-'The Urban and Papal Rites in Seventh- and Eighth-Century Rome', Sacris enidiri, 12
(1961), 411-87.
-'The Old-Roman Rite', Studia patristica, 5, Texte und Untersuchungen, 80 (1962),
185-205.
-'Papal Schola "versus" Charlemagne', in Smits van Waesberghe Festschrift 1963, 21-30.
-'Gregory the Great Founder of the Urban "Schola cantorum'", EL 77 (1963), 335-56.
-Sources of the Modern Roman Liturgy: The Ordinals by Ilaymo of Faversham and
Related Documents (1243-1307), 2 vols. (Leiden, 1963). -'Recent Developments in the Study of the Old-Roman Rite', Studia patristica, 8, Texte
und Untersuchungen, 93 (1966), 299-319. -'The Medieval Easter Vespers of the Roman Clergy', Sacris emdin, 19 (1969-70), 261-
363.
-(completed by Joan Hazelden Walker), The Ordinal of the Papal Court from Innocent III
to Boniface VIII, and Related Documents (Spicilegium Friburgense, 22; Fribourg, 1975).
-and Walker, Joan Hazelden, The Origins of the Modern Roman Liturgy: The Liturgy of
the Papal Court and the Franciscan Order in the Thirteenth Century (London, 1960).
Dix, Gregory, The Shape of the Liturgy (London, 1945).
DMA A.I (1975) = De numero tonorum litterae episcopi A. ad coepiscopum E. missae ac
Commentum super lottos episcopi E. (ad 1000). DMA A.II (1981) = Adaboldi Episcopi Ultraiectensis: Per musicae instrumentalis artem ad
hannoniam musicae mundanae (ad 1025). DMA A.Ill (1975) = Tres tractatuli Guidonis Aretini. [1] Guidonis Prologus in antiphonarium. DMA A.IV (1985) = Tres tractatuli Guidonis Aretini. [2] Guidonis Aretini Regulae
rhythmicae.
Bibliography
Ii
[DMA   A.V   =  Tres  tractatuli  Guidonis  Aretini.   [3]   Guidonis  Aretini  Epistola ad
Michaelem—did not appear]. DMA A.VIfl (1978) = Bernonis Augiensis abbatis de arte innsica disputationes traditae, pars
A: Bernonis Augiensis De mensurando monochordo. DMA A.VI6 (1979) = Bernonis Augiensis abbatis de arte tnusica disputationes traditae, pars
B: Quae ratio est inter tria opera de arte tnusica Bernonis Augiensis. DMA A. VII (1977) = Musica Domni Ileinria: Augustensis magistri. DMA A.VIII« (1988) = Summula: Tractatus metricus de musica.
DMA A.Xa (1979) = Codex Oxoniensis Bibliothecae Bodleianae, Rawlinson C 270, pars A:
De vocum consonantiis ac De re musica (Osberni Cantuariensis?). DMA A.Xb (1979) = Codex Oxoniensis Bibliothecae Bodleianae, Rawlinson C 270, pars B:
Al 7/ tractatuli a quodam studioso peregrino ad annum MC collecti. Dobson, Eric J., and Harrison, Frank Ll., Medieval English Songs (London, 1979). Dobszay, Läszlo, 'The System of the Hungarian Plainsong Sources', SMH 27 (1985), 37-65. -'The Program "CAO-ECE"', SMH 30 (1988), 355-60.
-'Experiences in the Musical Classification of Antiphons', in Cantus Planus 1990, 143-56.
-'Plainchant in Medieval Hungary', JPMMS 13 (1990), 49-78.
-and Pröszeky, Gabor, Corpus Antiphonalium Officii—Ecclesiarum Centralis Europae: A
Preliminary Report (Budapest, 1988).
Dold, Alban, Das älteste Liturgiebuch der lateinischen Kirche: Ein altgallikanisches Lektionar des 5.-6. Jhs. aus dem Wolfenbüttler Palimpsest-Codex Weissenburgensis 76 (Texte und Arbeiten, 26-8; Beuron, 1936).
-Neue St. Galler vorhieronymianische Propheten-Eragmente der St. Galler Sammelhandschrift 1398b zugehörig: Anhang L Ein neues Winitharfragment mit liturgischen Texten, 2. Irische Isidorfragmente des 7. Jahrhunderts (Texte und Arbeiten, 31; Beuron, 1940).
Donato, Giuseppe, GH elementi costituitivi dei tonari (Messina, 1978). Donovan, Richard B., The Liturgical Drama in Medieval Spain (Toronto, 1958). Dören, Rombaut van, Etude sur ('influence musicale de l'abbaye de Saint-Gall (VHP au XT
siede) (Brussels, 1925). Dreves see AH.
Drinkwelder, Otto, Ein deutsches Sequentiar aus dem Ende des 12. Jahrhunderts (VGA 8;
Graz and Vienna, 1914) [Berlin, Staatsbibliothek Preußischer Kulturbesitz, Mus. 40048]. Dronke, Peter, 'The Beginnings of the Sequence', Beiträge zur Geschichte der deutschen
Sprache und Literatur, 87 (1965), 43-73. -'Types of Poetic Art in Tropes', in Silagi 1985, 1-24.
Droste, Diane Lynne, 'The Musical Notation and Transmission of the Music of the Sarum
Use, 1225-1500' (Diss., Toronto, 1983; Canadian theses 59826). Drumbl, Johann, 'Zweisprachige Antiphonen zur Kreuzverehrung', Italia medioevale e
umanistica, 19 (1976), 41-55.
-'Ursprung des liturgischen Spiels', Italia medioevale e umanistica, 22 (1979), 45-96.
-Quem queritis: Teatro sacro dell'alto medioevo (Rome, 1981).
Dubois, Jacques, Le Martyrologe d'Usuard (Subsidia hagiographica, 40; Brussels, 1965). Duchesne, Louis, Origines du culte chretien: Etude surla liturgie latine avant Charlemagne
(5th edn., Paris, 1919); trans.: Christian Worship: Its Origin and Evolution. A Study of the
I^atin Liturgy up to the Time of Charlemagne (London, 1919).
Iii
Bibliography
Duchesne, Louis, lxj Liber pontificalis, 2 vols. (Paris, 1886 and 1892); rev. edn. bv Cvrille
Vogel, 3 vols. (Paris, 1955, 1955, 1957). Duchez, Marie-Elisabeth, Imago mundi: La naissance de la theorie music ale occidentale
dans les comment aires carolingiens de Martianus Ca pell a (Paris, 1979). -'La Representation spatio-verticale du caractere musical grave-aigu et l'elaboration de la
notion de hauteur cle son dans la conscience occidentale', AcM 51 (1979), 54-73. -'Description grammaticale et description arithmetique des phenomencs musicaux: Le
tournant du ixc siecle', Sprache und Erkenntnis im Mittelalter: Akten des VI. Internationalen
Kongresses für mittelalterliche Philosophie Bonn 1977 (Miscellanea medievalia, 13/2; Berlin
and New York, 1981), 561-79. -'Des neumes ä la portee', in Huglo 1987, 57-60.
Dumas, A., and Deshusses, Jean, Liber sacramentorum Gellonensis (Corpus Christianorum,
159-159A; Turnhout, 1981). Dyer, Joseph, 'The Offertories of Old Roman Chant: A Musico-liturgical Investigation'
(Diss., Boston University, 1971; UMI 71-26401). -'Singing with Proper Refinement from De modo bene cantandi (1474) by Conrad von
Zabern', Early Music, 6 (1978), 207-27. -'Augustine and the "Hymni ante oblatium": The Earliest Offertory Chants?', Revue des
etudes augustiniennes, 27 (1981), 85-99. -'The Offertory Chant of the Roman Liturgy and its Musical Form', Studi musicali, 11
(1982), 3-30.
-'Latin Psalters, Old Roman and Gregorian Chants', Kmjb 68 (1984), 11-30.
-'Monastic Psalmody of the Middle Ages', RB 99 (1989), 41-74.
-'The Singing of Psalms in the Early-Medieval Office', Speculum, 64 (1989), 535-78.
-'On the Monastic Origins of Western Music Theory', in Cantus Planus 1990, 199-225.
Ebel, Basilius, Das älteste alemannische Ilymnar mit Noten, Kodex 366 (472) Einsiedeln
(XII. Jahrhundert) (VGA 17; Einsiedcln, 1930). Eberle, L., The Rule of the Master (Cistercian Studies Series, 6; Kalamazoo, Mich., 1977). Echternacher Sakramentar und Antiphonar: Vollständige Facsimile-Ausgabe im Originalfonnat
der Handschrift 1946 . . . aus dem Besitz der Hessischen Landes- und Hochschulbibliothek
Dannstadt. With Essays bv Kurt Hans Staub, Paul Ulveling and Franz Unterkircher
(Codices Selccti, 74; Graz, 1982). Edwards, Kathleen, The English Secular Cathedrals in the Middle Ages (Manchester, 1949;
2nd edn., 1967).
Edwards, Owain Tudor, 'A Fourteenth-Century Welsh Sarum Antiphonal: National Library of Wales ms. 20541', JPMMS 10 (1987), 15-21.
-Matins, Lauds and Vespers for St David's Day: The Medieval Office of the Welsh Patron
Saint in National Library of Wales MS 20541 E (Woodbridge and Wolfeboro, 1990).
Eeles, Francis Charles, The Holy rood Ordinale: A Scottish Version of a Directory of English Augustinian Canons with Manual and Other Liturgical Forms (The Book of the Old Edinburgh Club, 7; Edinburgh, 1916).
Eggen, Erik, The Sequences of the Archbishopric of Nidaros (Bibliotheca Arnamagnasana, 212; Copenhagen, 1968).
Ehrensberger, Hugo, Bibliotheca liturgica manuscripta: Nach Handschriften der großherzoglich badischen Hof- und iMndesbibliothek (Karlsruhe, 1887).
Ekberg, Gudrun, 'Ekphonetic Notation', NG.
Bibliography
liü
Ekenberg, Anders, Cur cantatur? Die Funktionen des liturgischen Cesangs nach den
Autoren der Karolingerzeit (Stockholm, 1987). Ellinwood, Leonard, Musica Hennanni Contract/' (Eastman School of Music Studies, 2;
Rochester, NY, 1936).
Emerson, John A., 'The Recovery of the Wolffheim Antiphonal', Annates musicologiques, 6 (1958-63), 69-97.
-'Fragments of a Tropcr from Saint-Martial de Limoges', Scriptorium, 16 (1962), 369-
72.
-'Über Entstehung und Inhalt von MüD', Ktnjb 48 (1964), 33-60.
--'Two Newly Identified Offices for Saints Valeria and Austriclinianus by Ademar de
Chabannes (Ms Paris, Bibl. Nat., Latin 909, fols. 79-85v)\ Speculum, 40 (1965), 31-46. -'Sources, MS, II: Western Plainchant', .VC
Engel Festschrift 1964 = Pestschrift Hans Fngel zum siebzigsten Geburtstag (Kassel, 1964).
Epstein, Marcy J., 'Ludovicus decus regnantium: Perspectives on the Rhymed Office', Speculum, 53 (1978), 283-333.
Erbacher, Rhabanus, Tonus peregrinus: Aus der Geschichte eines Psalm tons (Münsterschwarzach, 1971).
Escudier, Denis, 'La Notation musicale de St. Vaast: Etude d'une particularity graphique',
in Huglo 1987, 107-18. Evans, Joan, Monastic Life at Cluny, 910-1157 (Oxford, 1931).
Evans, Paul, 'Some Reflections on the Origins of the Trope', JAMS 14 (1961), 119-30.
-'The Tropi ad sequentiain , in Strunk Festschrift 1968, 73-82.
-The Early Trope Repertory of Saint Martial de Limoges (Princeton, NJ, 1970).
-'Northern French Elements in an Earlv Aquitanian Troper', in Husmann Festschrift
1970, 103-10.
Evans, Roger William, 'Amalarius of Metz and the Singing of the Carolingian Offices'
(Diss., New York University, 1977; UMI 77-19546). Falconer, Keith, 'Early Versions of the Gloria Trope Pax sempiterna Christus', jfPMMS 7
(1984), 18-27.
-'A Kyrie and Three Gloria Tropes in a Norwegian Manuscript Fragment', Svensk
Tuiskrift for Musikforskning, 67 (1985), 77-88.
-'Some Early Tropes to the Gloria' (Diss., Princeton University, 1989; UMI 89-20341).
Fallows, David, 'Sources, MS, III, 2: Secular Monophony, Latin', NG.
Falvy, Zoltän, Drei Reimoffizien aus Ungarn und ihre Musik (Musicologia Hungarica, ns,
2; Budapest, 1968). -see also Codex Albensis.
Farmer, David Hugh, The Oxford Dictionary of Saints (Oxford, 1978).
Fassler, Margot, 'Musical Exegesis in the Sequences of Adam and the Canons of St. Victor'
(Diss., Cornell University, 1983; UMI 83-28706). -'Who was Adam of St. Victor? The Evidence of the Sequence Manuscripts', JAMS 37
(1984), 233-69.
-'The Office of the Cantor in Early Western Monastic Rules and Customaries: A
Preliminary Investigation', EMH 5 (1985), 29-51. -'Accent, Meter, and Rhythm in Medieval Treatises "De rithmis'", Journal of Musicology,
5 (1987), 164-90.
Feicht Festschrift 1967 = Zofia Lissa (ed.), Studia Hieronymo Feicht septuagenario dedicata (Warsaw, 1967).
li v Bibliography
Felder, Hilarin, Die liturgischen Reimoffizien auf den heiligen Franciscus und Antonius.
Gedichtet und componiert durch Julian von Speier (Fribourg, 1901). Fellerer, Karl Gustav, 'Zur Choralpflege und Chorallehre im 17./18. Jahrhundert', Kmjb
56 (1972), 51-72.
-(ed.), Geschichte der katholischen Kirchenmusik, i (Kassel, 1972).
-'Zur Choral-Restauration in Frankreich um die Mitte des 19. Jahrhunderts', Kmjb 58-9
(1974-5), 135-47.
-Kirchenmusik im 19. Jahrhundert (Studien zur Musik des 19. Jahrhunderts, 2;
Regensburg, 1985).
-'Gregorianik im 19. Jahrhundert', in Feilerer 1985, 9-95.
-arts. 'Caecilianismus', 'Choralreform', MGG.
Fellerer Festschrift 1962 = Heinrich Huschen (ed.), Festschrift Karl Gustav Fellerer zum
sechzigsten Geburtstag am 7. Juli 1962 (Regensburg, 1962). Fenlon, Iain (ed.), Cambridge Music Manuscripts 900-1700 (Cambridge, 1982). Fernandez de la Cuesta, Ismael, Manuscritos y fuentes musicales en Espaňa: Edad media
(Madrid, 1980).
-'La irrupción del canto gregoriano en Espaňa', Revista de musicología, 8 (1985), 239-48.
-see also Antiphonale Silense.
Férotin, M., Le Liber ordinum en usage dans 1'église zvisigothique et mozarabe dEspagne du cinquiěme á Tonziéme siěcle (Monumenta ecclesiae liturgica, 6; Paris, 1904).
Ferretti, Paolo, Principi teoretici e practici di canto gregoriano (Rome, 1905; 3rd edn., 1913).
-// cursu metrico e il ritmo delle melodie gregoriane (Rome, 1913).
-'Etude sur la notation acquitaine ďapres le Graduel de Saint-Yrieix', PalMus 13,
54-211.
-'I manoscritti gregoriani dell'Archivio di Montecassino', in Casineňsia: Miscellanea di
studi cassinensi pubblicati in occasione del XTV centenario della fondazione della badia di
Montecassino, i (Montecassino, 1929), 187-203. -Estetica gregoriana: Trattato delle forme musicali del canto gregoriano, i (Rome, 1934);
trans. A. Agaésse: Esthétique grégorienne, ou traité des formes musicales du chant
grégorien, i (Solesmes, 1938).
-Estetica gregoriana dei recitativi liturgici, ed. P. Ernetti (Venice, 1964).
Fiala, V., and Irtenkauf, W., 'Versuch einer liturgischen Nomenklatur', Zeitschrift für
Bibliotheksivesen und Bibliographie. Sonderheft 1: Zur Katalogisierung mittelalterlicher und
neuerer Handschriften (Frankfurt, 1963), 105-37. Fickett, Martha Van Zandt, 'Chants for the Feast of St. Martin of Tours' (Diss., Catholic
University of America, 1983; UMI 84-01475). ■ Fisher (Kurt von)  Festschrift  1973  =  Hans-Heinrich Eggebrecht and Max Liitolf
(eds.), Studien zur Tradition in der Musik: Kurt von Fischer zum 60. Geburtstag (Munich,
1973).
Fisher (Kurt von) Festschrift 1973 = Hans-Heinrich Eggebrecht and Max Lütolf (eds.), Studien zur Tradition in der Musik: Kurt von Fischer zum 60. Geburtstag (Munich, 1973).
Fischer, Ludwig (ed.), Bernhardi cardinalis et ÍMteranensis ecclesiae prioris, ordo officiorum ecclesiae Lateranensis (Munich, 1916).
Fischer, Pieter see RISM.
Fischer (Wilhelm) Festschrift 1956 = Festschrift Wilhelm Fischer zum 70. Geburtstag,
Bibliography
lv
überreicht im Mozartjahr 1956 (Innsbrücker Beiträge zur Kulturwissenschaft, Sonderheft 3; Innsbruck, 1956).
Fleming, Keith, 'The Editing of Some Communion Melodies in Medieval Chant Manuscripts' (Diss., Catholic University of America, 1979; UMI 79-20544).
Flint, Valerie I. J., 'Are Heinricus of Augsburg and Honorius Augustodunensis the Same Person?' lUi, 92 (1982), 148-68.
Floros, Constantin, Universale Neumenkunde, 3 vols. (Kassel, 1970).
-Einführung in die Neumenkunde (Wilhelmshafen, 1980).
Flotzinger, Rudolf, 'Zu Herkunft und Datierung der Gradualien Graz 807 und Wien
13314', SMH 31 (1989), 57-80. Floyd, Malcolm, 'Processional Chants in English Sources', JPMMS 13 (1990), 1-48. Foley, E., 'The Cantor in Historical Perspective', Worship, 56 (1982), 194-213. Fontaine, G., 'Presentation des missels diocesains frangais du xvne au xix° siecles', La
Maison-Dieu, 141 (1980), 97-166. Fortescue, Adrian, The Ceremonies of the Roman Rite Described (London, 1917; rev.
J. B. O'Connell, 12th edn., 1962). Fournier, Dominique M., 'Sources scripturaires et provenance liturgique des pieces de chant
du graduel de Paul VI. (I. Ancien Testament)', EG 21 (1986), 97-114. -'Sources scripturaires et provenance liturgique des pieces de chant du graduel de Paul
VI. (II. Les Psaumes)', EG 22 (1988), 109-75. Franca, Umberto, Le antifone bibliche dopo Pentecoste (Studia anselmiana, 73; Rome, 1977). Freistedt, Heinrich, Die liqueszierenden Noten des gregorianischen Chorals (Fribourg,
1929) .
Frenaud, Georges, 'Les Temoins indirects du chant liturgique en usage ä Rome aux ixc et xe
siecles', EG 3 (1959), 41-74. Frere, Walter Howard, The Winchester Troper from MSS of the Xth and Kith Centuries
(Henry Bradshaw Society, 8; London, 1894). -Bibliotheca musico-liturgica: A Descriptive Handlist of the Musical and Latin-liturgical
MSS of the Middle Ages Preserved in the Libraries of Great Britain and Northern Ireland, i
(London, 1894 and 1901), ii (1932). -The Use of Sarurn: 1. The Sarum Customs as Set Forth in the Consuetudinary and
Customary (Cambridge, 1898); //. The Ordinal and Tonal (Cambridge, 1901).
-Pontifical Services (Alcuin Club Collections, 3-4; London, 1901).
-The Principles of Religious Ceremonial (London, 1906: 2nd edn., 1928).
-(ed.), Pars antiphonarii (London, 1923) [fac. of Durham Cathedral Library B. iii. 11].
-Studies in Early Roman Liturgy, i: The Calendar (Alcuin Club Collections, 28; London,
1930) .
-Studies in Early Roman Liturgy, ii: The Roman Gospel-Lectionary (Alcuin Club
Collections, 30; London, 1934). -Studies in Early Roman Liturgy, iii: The Roman Epistle-Lectionary (Alcuin Club
Collections, 32; London, 1935). -and Brown, Langton E. G., The Hereford Breviary (Henry Bradshaw Society, 26, 40,
and 46; London, 1904, 1911, and 1915). -see also AS; Graduale Sarisburiense.
Froger, Jacques, 'L'Alleluia dans l'usage romain et la reforme de saint Gregoire', EL 62 (1948), 6-48.
lvi
Bibliography
Froger, Jacques, 'Lc Chant de l'introit', EL 62 (1948), 248-55.
-Les Chants de la messe au VIIL et IXC siecles (Tournai, 1950) = RG 26 (1947), 165-72,
218-28; 27 (1948), 56-62; 98-107; 28 (1949), 58-65, 94-102. -'L'Edition critique de l'Antiphonale Missarum romain par les moines de Solesmes', EG 1
(1954), 151-7.
-'L'Epitre de Notker sur les "lettres significatives'", EG 5 (1962), 23-72.
-La Critique des textes et son automatisation (Paris, 1968).
-'Les Pretendus Quarts de ton dans le chant gregorien et les symboles du ms. II. 159 de
Montpellier', EG 17 (1978), 145-79. -'The Critical Edition of the Roman Gradual by the Monks of Solesmes', JPMMS 1
(1978) , 81-97.
-'La Methode de Dom Hesbert dans le volume V du Corpus Antiphonaliurn Officii', EC} 18
(1979) , 97-143.
-'Le Fragment de Lucques (fin du vmc siecle)', EG 18 (1979), 145-55.
-'Le Lieu de destination et de provenance du "Compendiensis"', in Cardine Festschrift
1980, 338-53.
-'La Methode de Dom Hesbert dans le volume VI du Corpus Antiphonaliurn Officii', EG
19 (1980), 185-96. Fuller, David, 'Plainchant musical', NG.
Fuller, Sarah, 'Aquitanian Polyphony of the Eleventh and Twelfth Centuries' (Diss.,
University of California at Berkeley, 1969; UMI 70-13051). -'An Anonymous Treatise dictus de Sancto Martiale: A New Source for Cistercian Music
Theory', MD 31 (1977), 5-30.
-'Theoretical Foundations of Early Organum Theory', AcM 53 (1981), 52-84.
Gajard, Joseph, Im Methode de Solesmes, ses principes constitutifs, ses regies pratiques
d'interpretation (Tournai, 1951). -'Les Recitations modales des 3C et 4C modes dans les manuscrits beneventains et
aquitains', EG 1 (1954), 9-45. Gallacher, Patrick J., and Damico, Helen (eds.), Hermeneu tics and Medieval Culture
(Albany, NY, 1989).
Gallo, F. Alberto, 'Philological Works on Musical Treatises of the Middle Ages. A
Bibliographical Report', AcM 44 (1972), 78-101. Gamber, Klaus, Sakramentartypen: Versuch einer Gruppierung der Handschriften und
Fragmente bis zurfahrtausendwende (Texte und Arbeiten, 49-50; Beuron, 1958).
-'Das Lectionar und Sakramentar des Musäus von Massilia', RB 69 (1959), 118-215.
-'Fragment eines mittelitalienischen Plenarmissale aus dem 8. Jh.', EL76 (1962), 335-41.
-Codices liturgici Latini antiquiores (Spicilegii Friburgensis, Subsidia, 1; Freiburg, 1963;
2 vols., 2nd edn., 1968; Subsidia [supplement and indices], 1988). Garand, M., 'Le Missel clunisien de Nogent-le-Rotrou', in Boutemy Festschrift 1976, 129-
51.
Garsoian, Nina G., Mathews, Thomas F., and Thomson, Robert W. (eds.), East of Byzantium: Syria and Armenia in the Formative Period (Dumbarton Oaks, 1982).
Gastoue, Amedee, 'Le Chant gallican', RCG 41 (1937), 101-6, 131-3, 167-76; 42 (1938), 5-12, 57-62, 76-80, 107-12, 146-51, 171-6; 43 (1939), 7-12, 44-6; also published as a book (Grenoble, 1939).
Gatard, A., 'Ambrosien (chant)', DACL.
Gautier, Leon, Histoire de la poesie liturgique au moyen age, i: I^es Tropes (Paris, 1886).
Bibliography
lvii
Gavel, M. H., 'A propos des crreurs d'accentuation latine dans les livres liturgiques', EG 1 (1954), 83-148.
Gelineau, Joseph, Chant et musique dans le culte chretien, principes, lois et applications (Paris, 1962); trans. C. Howell: Voices and Instruments in Christian Worship (Collegeville, Minn., 1964).
Gennrich, Friedrich, Grundriß einer Formenlehre des mittelalterlichen Liedes als Grundlage
einer musikalischen Formerdehre des Liedes (Halle, 1932). Gerbert, Martin, De cantu et musica sacra a prima ecclesiae aetate usque ad praesens
tempus, 2 vols. (Sankt Blasien, 1774; repr. Graz, 1968 with introduction and index by
Othmar Wessely). -see also GS.
Gevaert, Franc;ois-Auguste, Les Origines du chant liturgique de l'eglise latine (Ghent, 1890).
-La Melopee antique dans le chant de l'eglise latine (Ghent, 1895).
Gibson, Margaret (ed.), Boethius (Oxford, 1981).
Gilissen Festschrift 1985 = Jacques Lcmaire and Emile van Balberghe (eds.), Calames et cahiers: Melanges de codicologie et de paleographie offerts ä Leon Gilissen (Brussels, 1985).
Gillingham, Bryan (ed.), Paris, Bibliotheque Nationale, fonds latin 1139 (Publications of Mediaeval Manuscripts, 14; Ottawa, 1987).
-Paris, Bibliotheque Nationale, fonds latin 3719 (Publications of Mediaeval Manuscripts,
.15; Ottawa, 1987).
-Paris, Bibliotheque Nationale, fonds latin 3549 and London, B. L., Add. 36,881
(Publications of Mediaeval Manuscripts, 16; Ottawa, 1987). -Cambridge, University Library, Ff. i. 17(1) (Publications of Mediaeval Manuscripts, 17;
Ottawa, 1989).
Gindele, Corbinian, 'Chordirektion des gregorianischen Gesangs im Mittelalter', Studien und Mitteilungen zur Geschichte des Benediktiner Ordens und seiner Zweige, 63 (1951), 31-44.
-'Doppelchor und Psalmvortrag im Frühmittelalter', Mf 6 (1953), 296-300.
Gingras, G., Egeria: Diary of a Pilgrimage (Ancient Christian Writers, 38; New York, 1970).
Gjerl0W, Lilli, Adoratio Crucis, the Regularis Concordia and the Decreta Lanfranci: Manuscript Studies in the Early Medieval Church of Norway (Oslo, 1961).
-Ordo Nidrosiensis ecclesiae (Libri liturgici provinciae Nidrosis medii aevi, 2; Oslo,
1968).
-'Votive Masses Found in Oslo', EL 84 (1970), 113-28.
-Antiphonarium Nidrosiensis ecclesiae (Libri liturgici provinciae Nidrosis medii aevi, 3;
Oslo, 1979).
-Liturgica Islandica (Bibliotheca Arnamagnaeana, 35-6; Copenhagen, 1980).
Gleeson,  Philip, 'Dominican Liturgical Manuscripts before 1254', Archivum fratrum
praedicatorum, 42 (1972), 81-135. Gmelch, Joseph, Die Vierteltonstufen im Meßtonale von Montpellier (Eichstätt, 1911). Gneuss, Helmut, Hymnar und Hymnarien im englischen Mittelalter (Tübingen, 1968). Goede, N. de, The Utrecht Prosarium (Monumenta musica Neerlandica, 6; Amsterdam,
1965) [Utrecht, Bibliotheek der Rijskuniversiteit, 417]. Göllner, Theodor, 'Unknown Spanish Passion Tones in Sixteenth-Century Hispanic
Sources', JAMS 28 (1975), 46-71.
lviii
Bibliography
GoMBOSi, Otto, 'Studien zur Tonartlehre des frühen Mittelalters', AcM 10 (1938), 174-t>4;
11 (1939), 28-39, 128-35; 12 (1940), 21-52. -'Key, Mode, Species', JAMS 4 (1951), 20-6.
Göschl, Johannes Berchmans, Semiologische Untersuchungen zum Phänomen der gregorianischen Liqueszenz: Der isolierte dreistufige Epiphonus praepunctis, ein Sonderproblem der Liqueszenzforschung, 2 vols. (Forschungen zur älteren Musikgeschichte, 3; Vienna, 1980).
-'Der gegenwärtige Stand der semiologischen Forschung', BzG 1 (1985), 43-102.
Gottesdienst der Kirche: Handbuch der Liturgiewissenschaft, ed. Hans B. Meyer, Hansjörg auf der Mauer, Balthasar Fischer, Angelus A. Häußling, and Bruno Kleinheyer. Vol. iii: Gestalt des Gottesdienstes: Sprachlich und nichtsprachliche Ausdrucksfonnen (Regensburg, 1987).
Graduate Arosiense impressum, ed. Toni Schmid (Laurentius Petri Sällskapets Urkundsserie, 7; Malmö and Lund, 1959-65).
Graduate de tempore et de Sanctis juxta ritum Sanctae Romanae Ecclesiae cum can tu Pauli V. Pont. Max. jussu reformalo (Rome, 1898).
Graduate Pataviense (Wien 1511): Faksimile, ed. Christian Väterlein (Das Erbe deutscher Musik, 87; Kassel, 1982).
Graduel romain . . . Chant restaure par la Commission de Reims et de Cambrai d'apres les anciens manuscrits (Paris, 1874).
Graduel romain: Edition critique par les moines de Solesmes. II: l^es Sources (Solesmes, 1957); IV: Le Texte neumatique, i: Le Groupement des manuscrits (Solesmes, 1960); //: Les Relations genealogiques des manuscrits (Solesmes, 1962).
Graduel romain . . . publiee par la Commission Ecclesiastique de Digne (Marseilles, 1872).
Graduate Romanum . . . restitutum et editum Pauli VI (Solesmes, 1974).
Graduate Sarisburiense = Walter Howard Frere (ed.), Graduate Sarisburiense: A Reproduction in Facsimile of a Manuscript of the Thirteenth Century, with a Dissertation and Historical Index Illustrating its Development from the Gregorian Antiphonale missa?~um (London, 1894).
Gr£goire, Reginald, Les Homeliaires du moyen age: Inventaire et analyse des manuscrits (Rerum ccclesiasticarum documenta, Series maior, 6; Rome, 1966).
-'Repertorium liturgicum italicum', Studi medievali, 3rd ser., 9 (1968), 463-592; 11
(1970), 537-56; 14 (1973), 1123-32.
-Homeliaires liturgiques medievaux (Biblioteca degli 'Studi medievali', 12; Spoleto,
1980).
Grisar, Hartmann, 'Die Gregorbiographie des Paulus Diakonus in ihrer ursprünglichen Gestalt, nach italienischen Handschriften', Zeitschrift für katholische Theologie, 11 (1887), 158-73.
Gros, Miquel S., 'El Processoner de la Catedral de Vic—Vic, Mus. Episc, MS 117
(CXXIV)', Miscehlänia litürgica catalana, 2 (1983), 73-130. Grotefend, Hermann, Zeitrechnung des deutschen Mittelalters und der Neuzeit, 2 vols.
(Hannover, 1891-8).
-Taschenbuch der Zeitrechnung des deutschen Mittelalters und der Neuzeit (Hannover
and Leipzig, 1898; 10th edn., 1960). Guentner see CSM 24.
Guidetti, Giovanni Domenico, Directorium chori ad usum sacrosanctae basilicae vaticanae et aliarum cattedralium (Rome, 1582).
Bibliography
lix
Gümpel, Karl-Werner, 'Zur Interpretation der Tonus-Definition des Tonale Sancti Bernardi', Akademie der Wissenschaften und der Literatur, Jahrgang 195912 (Wiesbaden, 1959), 25-51.
Gushee, Lawrence A., 'The Musica Disciplina of Aurelian of Reöme, a Critical Text and
Commentary' (Diss., Yale University, 1963; UMI 64-11873). -'Questions of Genre in Medieval Treatises on Music', in Schrade Gedenkschrift 1973,
365-433. -see also CSM 21.
Gy, Pierre-Marie, 'Collectaire, rituel, processional', Revue des sciences philosophiques et
theologiques, 44 (1960), 441-69. -'Typologie et ecclesiologie des livres liturgiques medievaux', La Maison-Dieu, 121
(1975), 7-21.
-'Les Tropes dans l'histoire de la liturgie et de la theologie', in Iversen 1983, 7-16.
-'L'Influence des chanoines de Lucques sur la liturgie du Latran', Revue des sciences
religieuses, 58 (1984),' 31-41. -'Les Repons de l'office nocturne pour la fete de S. Martin', in Giustino Farnedi (ed.),
Traditio et progressio: Studi liturgici in onore del Prof Adrien Nocent, OSB (= Studia
anselmiana, 95 (1988)), 215-23. Gy Festschrift 1990 = Rituels: Melanges offerts au Pere Gy OP, ed. Paul de Clerck and Eric
Palazzo (Pans, 1990).
Gyug, Richard F., 'Tropes and Prosulas in Dalmatian Sources of the Twelfth and Thirteenth Centuries', in Leonardi and Menesto 1990, 409-38.
Haapanen, T., Verzeichnis der mittelalterlichen Handschriftenfragmente in der Universitätsbibliothek zu Llelsingfors (Helsinki, 1922-32).
-Die Neumenfragmente der Universitätsbibliothek llelsingfors: Eine Studie zur ältesten
nordischen Musikgeschichte (Helsinki, 1924).
Haas, Max, Byzantinische und slavische Notationen (Palaeographie der Musik, 1/2; Cologne, 1973).
-'Probleme einer "Universale Neumenkunde'", Forum musicologicum, 1 (1975), 305-22.
-'Studien zur mittelalterlichen Musiklehre. I. Eine Übersicht über die Musiklehre im
Kontext der Philosophie des 13. und frühen 14. Jahrhunderts', Forum musicologicum, 3
(1982), 223-456.
Haberl (Ferdinand) Festschrift 1977 = Franz A. Stein (ed.), Festschrift Ferdinand Haberl zum 70. Geburtstag: Sacerdos et cantus gregoriani magister (Regensburg, 1977).
Haberl, Franz Xaver, 'Geschichte und Wert der offiziellen Chorbücher', Kmjb 27 (1902), 134-92.
Haddan, Arthur West, and Stubbs, William, Councils and Ecclesiastical Documents
Relating to Great Britain and Ireland, 3 vols. (Oxford, 1869-71). Hain, Karl, Ein musikalischer Palimpsest (VGA 12; Fribourg, 1925). Hallinger, Kassius, Gorze-Kluny: Studien zu den monastischen Lebensformen und
Gegensätzen im Hochmittelalter (Studia anselmiana, 22/2324/25; Rome, 1950-1). -see also CCM.
Halperin, David, 'Contributions to a Morphology of Ambrosian Chant' (Diss., Tel Aviv University, 1986).
Hämel, Adalbert, 'Überlieferung und Bedeutung des Liber Sancti Jacobi und des Pseudo-Turpin', Sitzungsberichte der philosophisch-historischen Klasse der Bayerischen Akademie der Wissenschaften (1950), no. 2, 1-75.
lx Bibliography
Hammerstein, Reinhold, Die Musik der Engel: Untersuchungen zur Musikanschauung des
Mittelalters (Berne, 1962). Hammond see CSM 14.
Handschin, Jacques, 'Ein mittelalterlicher Beitrag zur Lehre von der Sphärenharmonie',
Zeitschrift für Musikwissenschaft, 9 (1926-7), 193-208. -'Über Estampie und Sequenz', Zeitschrift für Musikwissenschaft, 12 (1929), 1-20; 13
(1930-1), 113-32.
-'Sequenzprobleme', Zeitschrift für Musikwissenschaft, 17 (1935), 242-50.
-'L'Organum ä l'eglise et les exploits de l'abbe Turstin', RCG 40 (1936-7), 179-82; 41
(1937-8), 14-19, 41-8 (see also David and Handschin 1935-7).
-'Eine alte Neumenschrift', AcM 21 (1950), 69-97; 25 (1953), 87-8.
-'Trope, Sequence, and Conductus', The Sezc Oxford History of Music, ii, ed. Anselm
Hughes (London, 1952), 128-74.
-'Sur quelques tropaires grecs traduits en latin', Annales musicologiques, 2 (1954), 27-60.
-The Timaeus Scale', MD 4 (1950), 3-42.
Handschin Gedenkschrift 1962 = II. Angles, G. Birkner, Ch. van den Borren, Fr. Bcnn, A. Carapetyan, and H. Husmann (eds.), In memoriam Jacques Handschin (Strasburg, 1962).
Hänggi, A., Der Rheinauer Uber Ordinarius (Zürich Rh 80, Anfang 12. Jh.) (Spicilegium Friburgense, 1; 1957).
Hannick, Christian, arts. 'Armenian Rite, Music of the', 'Christian Church, Music of the
Early', 'Ethiopian Rite, Music of the', 'Georgian Rite, Music of the', NG. Hansen, Finn Egeland, // 159 Montpellier (Copenhagen, 1974).
-'Editorial Problems Connected with the Transcription of II 159, Montpellier: Tonary of
St. Benigne of Dijon', EG 16 (1977), 161-72.
-The Grammar of Gregorian Tonality, 2 vols. (Copenhagen, 1979).
Hanslik, R., Benedicti Regula (Corpus scriptorum ecclesiasticorum latinorum, 75; Vienna,
1960; rev. edn., 1977). Hanssens, Jean-Michael, Amalarii episcopi opera liturgica omnia (Studi e testi, 13840;
Rome, 1948-50). Harbinson see CSM 23.
Hardison, O. B., Christian Rite and Christian Drama in the Middle Ages (Baltimore, Md., 1965).
Harrison, Frank Ll., Music in Medieval Britain (London, 1958; 2nd edn., 1963).
-'Benedicamus,  Conductus,  Carol:  A  Newly-Discovered Source', AcM 37 (1965),
35-48.
Hartzell, K. Drew, 'An Unknown English Benedictine Gradual of the Eleventh Century',
Anglo-Saxon England, 4 (1975), 131-44. -'An Eleventh-Century English Missal Fragment in the British Library', Anglo-Saxon
England, 18 (1989), 45-97. Haug, Andreas, Gesungene und schriftlich dargestellte Sequenz: Beobachtungen zum
Schriftbild der älteste?? ostfränkischen Sequenzenhandschriften (Neuhausen-Stuttgart,
1987).
Hayburn, Robert F., Papal legislation on Sacred Music, 95 A.D. to 1977 A.D. (Collegeville, Minn., 1979).
Heard, Edmund Brooks, 'Alia musica: A Chapter in the History of Medieval Music Theory' (Diss., University of Wisconsin, 1966; UMI 66-13798).
Bibliography
lxi
IIeiming, Odilo, 'Zum monastisehen Offizium von Kassianus bis Kolumbanus', Archiv für
Liturgiewissenschaft, 7 (1961), 89-156.
--'Das Corpus Ambrosiano-Liturgicum. Ein Bericht', EL 92 (1978), 477-80.
IIeisler, Maria-Elisabeth, 'Die Problematik des "germanischen" oder "deutschen" Choraldialekts', SMH 27 (1985), 67-82.
-'Studien zum ostfrankischen Choraldialekt' (Diss., Frankfurt, 1987).
IIeitz,  Carol, Recherches sur les rapports entre architecture et liturgie a l'epoque
carolingienne (Paris, 1963). I Isländer,  S., Ordinarius Lincopensis och dess liturgiska fbrebilder [The Ordinal of
Linköping and its liturgical models] (Bibliotheca theologica practicae, 4; Lund, 1957). Henderson, Isobel, 'Ancient Greek Music', New Oxford History of Music, i, ed. Egon
Wellesz (Oxford, 1957), 336-403. Henderson, William G., Missale ad usum insignis ecclesiae Eboracensis (Surtees Society,
59-60; London, 1874).
-Missale ad usum percelebris ecclesiae Herfordensis (Leeds, 1874).
-(ed.), Manuale et processionale ad usum insignis ecclesiae Eboracensis (Surtees Society,
63; London, 1875).
-(ed.), Processionale ad usum insignis ac praeclarae ecclesiae Sarum (Leeds, 1882).
Herlinger, Jan, The Lucidarium oj Marchetto oj Padua: A Critical Edition, Translation, and
Commentary (Chicago, 1985). Hermesdorff, Michael, Graduate juxta usum ecclesiae cathedralis Trevirensis dispositum.
Quod ex veteribus codicibus originalibus accuratissime conscriptum et novis interim
ordinatis seu indultis festis auction (Trier, 1863). -Graduate ad nonnam cantus S. Gregorii, auf Grund der Eorschungs-Resultäte und unter
Beihilfe der Mitglieder des Vereins zur Erforschung alter Choralhandschriften nach den
ältesten und zuverlässigsten Quellen (Trier, 1876-82). 1 Ierz Festschrift 1982 = Robert L. Weaver (ed.), Essays on the Music of J. S. Bach and Other
Divers Subjects: A Tribute to Gerhard Herz (Louisville, Ky., 1982). IIerzo, Anthony Marie, 'Five Aquitanian Graduals: Their Mass Propers and Alleluia
Cycles' (Diss., University of Southern California, 1967; UMI 67-10762). IIesbert, Rene-Jean, 'La Messe "Omnes gentes" du vnc dimanchc apres la Pentecöte et
l'"Antiphonale Missarum" romain', RG 17 (1932), 81-9, 170-9; 18 (1933), 1-14. -'Les Dimanchcs de Careme dans les manuscrits romano-beneventains', EL 48 (1934),
198-222.
-'L"'Antiphonale Missarum" de l'ancien rit beneventain', EL 52 (1938), 28-66, 141-58;
53 (1939), 168-90; 59 (1945), 69-95; 60 (1946), 103-41; 61 (1947), 153-210. -'Un curieux antiphonaire palimpseste de l'office. Rouen, A. 292 (IXC s.)', RB 64 (1954),
28-45.
-'L'Antiphonaire d'Amalar', EL 94 (1980), 176-94.
-'L'Antiphonaire de la Curie', EL 94 (1980), 431-59.
-'The Sarum Antiphoner—Its Sources and Influence', JPMMS 3 (1980), 49-55.
-'Les Antiphonaires monastiques insulaires', RB 112 (1982), 358-75.
-'Les Matines de Päques dans la tradition monastique', Studia monastica, 24(1982), 311-
48.
-see also AMS, CAO, and MMS.
Hiley, David (with Julian Brown and Sonia Patterson), 'Further Observations on Wl', JPMMS 3 (1980), 53-80.
lxii Bibliography
IIiley, David 'The Norman Chant Traditions—Normandy, Britain, Sicilv', Proceedings of the
Royal Musical Association, 107 (1980-1), 1-33. -'The Liturgical Music of Norman Sicily: A Study Centred on Manuscripts 288, 289,
19421 and Vitrina 20-4 of the Biblioteca Nacional, Madrid' (Diss., London, 1981; British
Theses D70780/82).
-'Some Observations on the Relationships between Trope Repertories', in Iversen 1983,
29-38.
-'Quanto c'e di normanno nei tropari siculo-normanni?', RIM 18 (1983), 3-28.
-'The Plica and Liquescence', in Anderson Gedenkschrift 1984, 379-91.
-'Ordinary of Mass Chants in English, North French and Sicilian Manuscripts', JPMMS
9 (1986), 1-128.
-'Thurstan of Caen and Plainchant at Glastonbury: Musicological Reflections on the
Norman Conquest', Proceedings of the British Academy, 72 (1986), 57-90.
-'The Rhymed Sequence in England: A Preliminary Survey', in Huglo 1987, 227-46.
-'The Chant of Norman Sicily: Interaction between the Norman and Italian Traditions',
SMH 30 (1988), 379-91.
-'Cluny, Sequences and Tropes', in Leonardi and Menesto 1990, 125-38.
-'Editing the Winchester Sequence Repertory of ca. 1000', in Cantus Planus 1990, 99-
113.
-arts. 'Neo-Gallican Chant', 'Neuma', 'Notation, III, 1: Western, Plainchant', 'Pontifical',
NG.
Höfler, J., 'Rekonstrukcija srednjeveskega sekvenciarija v osrednji Sloveniji/Reconstruction
of the medieval sequencer in Central Slovenia', Muzikoloski zbornik, 3 (1967), 5. Hofmann-Brandt, Helma, Die Tropen zu den Responsorien des Officium, 2 vols. (Diss.,
Erlangen-Nürnberg University, 1971). Hohler, Christopher, 'The Proper Office of St. Nicholas and Related Matters with
Reference to a Recent Book', Medium aevum, 36 (1967), 40-8.
-'A Note on Jacobus', Journal of the Warburg and Courtauld Institutes, 35 (1972), 31-80.
-'Reflections on Some Manuscripts Containing 13th-Century Polyphony', JPMMS 1
(1978), 2-38.
-review of Babb, Hucbald, Guido, and John on Music, in JPMMS 3 (1980), 57-8.
Holder, Stephen, 'The Noted Cluniac Breviary-Missal of Lewes: Fitzwilliam Museum
Manuscript 369', JPMMS 8 (1985), 25-32. Hollaender, Albert, 'The Sarum Illuminator and his School', Wiltshire Archaeological and
Natural History Magazine, 1 (1942-4), 230-62. Holman, Hans-Jorgen, 'The Responsoria prolixa of the Codex Worcester F 160' (Diss.,
Indiana University, 1961; UMI 61-04447).
-'Melismatic Tropes in the Responsories for Matins', JAMS 16 (1963), 36-46.
Holschneider, Andreas, Die Organa von Winchester: Studien zum ältesten Repertoire
polyphoner Musik (Hildesheim, 1968). -'Instrumental Titles to the Sequentiae of the Winchester Tropers',  in Westrup
Festschrift 1975, 8-18.
-'Die instrumentalen Tonbuchstaben im Winchester Troper', in Dadelsen Festschrift
1978, 155-66.
Holtz, Louis, 'Quelques aspects de la tradition et de la diffusion des "Institutiones"', in Flavio Magno Aurelio Cassiodoro: Atti delta Settimana di Studi su Cassiodoro (Cosenza-Rossano-Squillace, 1984), 281-312.
Bibliography
lxiii
Uoman, Frede.ric W., 'Final and Internal Cadential Patterns in Gregorian Chant', JAMS 17 (1964), 66-77.
Hoppin, Richard H., Cypriot Plainchant of the Manuscript Torino, Biblioteca Nazionale J.
II. 9 (Musicological Studies and Documents, 19; Rome, 1968). Hourlier, Jacques, 'Le Domaine de la notation messine', RG 30 (1951), 96-113, 150-8. -'Remarques sur la notation clunisienne', RG 30 (1951), 231-40.
-'Le Breviaire de Saint-Taurin: Un livre liturgique clunisien ä l'usage de l'Echelle-Saint-
Aurin (Paris B. N. lat. 12601)', EG 3 (1959), 163-73.
-'Notes sur l'antiphonie', in Schrade Gedenkschrift 1973, 116-43.
-'L'Origine des neumes', in Cardine Festschrift 1980, 354-61.
-and Huglo, Michel, 'La Notation paleofranque', EG 2 (1957), 212-19.
--'Etude sur la notation beneventaine', PalMus 15, 71-161.
Hrabanus Maurus, Artium liberalium ordo et natura (Book 3 of De clericorum instituttone),
PL 107, 377-420. -De computo, PL 107, 669-728.
Hucke, Helmut, 'Untersuchungen zum Begriff "Antiphon" und zur Melodik der Offiziums-
antiphonen' (Diss., Freiburg im Br., 1951).
-'Musikalische Formen der Offiziumsantiphonen', Kmjb 37 (1953), 7-33.
-'Die Entwicklung des christlichen Kultgesangs zum Gregorianischen Gesang' Römische
Quartalschrift, 48 (1953), 147-94. —:—'Die Einführung des Gregorianischen Gesangs im Frankenreich', Römische Quartalschrift,
49 (1954), 172-87.
-'Improvisation im Gregorianischen Gesang', Kmjb 38 (1954), 5-8.
-'Die Tradition des Gregorianischen Gesanges in der römischen Schola cantorum', in
Congress Vienna 1954, 120-3. -'Graduale', EL 69 (1955), 262-4.
-'Gregorianischer Gesang in altrömischer und fränkischer Überlieferung', AfMw 12
(1955), 74-87.
-'Die Entstehung der Überlieferung von einer musikalischen Tätigkeit Gregors des
Großen', Mf 8 (1955), 259-64. -'Die Gregorianische Gradualeweise des 2. Tons und ihre ambrosianischen Parallelen.
Ein Beitrag zur Erforschung des Ambrosianischen Gesangs', AfMw 13 (1956), 285-314.
-'Cantus gregorianus', in Hermann Schmidt 1956-7, ii. 901-50.
-'Eine unbekannte Melodie zu den Laudes regiae', Kmjb 42 (1958), 32-8.
-'Zu einigen Problemen der Choralforschung', M/ 11 (1958), 385-414.
-'Zum Problem des Rhythmus im Gregorianischen Gesang', in Congress Cologne 1958,
141-3.
-'War Gregor der Große doch Musiker?', Mf 18 (1965), 390-3.
-'Tractusstudien', in Stäblein Festschrift 1967, 116-20.
-'Die Texte der Offertorien', in Husmann Festschrift 1970, 193-203.
-'Le Probleme de la musique religieuse', La Maison-Dieu, 108 (1971), 7-20.
-'Das Responsorium', in Schrade Gedenkschrift 1973, 144-91.
-'Die Herkunft der Kirchentonarten und die fränkische Überlieferung des Gregorianischen
Gesangs', in Congress Berlin 1974, 257-60.
-'Karolingische Renaissance und Gregorianischer Gesang', Mf 28 (1975), 4-18.
-'Der Übergang von mündlicher zu schriftlicher Musiküberlieferung im Mittelalter', in
Congress Berkeley 1977, 180-91.
lxiv
Bibliography
Hucke, Helmut, 'Die Cheironomie und die Entstehung der Newumenschrift', Mf 32 (1979), 1-6.
-'Towards a New Historical View of Gregorian Chant', JAMS 33 (1980), 437-67.
-'Zur Aufzeichnung der altrömischen Offertoricn', in Cardine Festschrift 1980, 296-313.
-'Die Anfänge der Bearbeitung', Schweizer Jahrbuch für Musikwissenschaft, ns, 3
(1983), 15-20.
-'Die Anfänge der abendländischen Notenschrift', in Elvers Festschrift 1985, 271-88.
-'Gregorianische Paläographie als Überlieferungsgeschichte', in Huglo 1987, 61-5.
-'Choralforschung und Musikwissenschaft', in Dahlhaus Festschrift 1988, 131-41.
-'Gregorianische Fragen', M/41 (1988), 304-30.
-'Responsorium', MGG.
-arts. 'Gradual (i)', 'Gregorian and Old Roman Chant', 'Gregory the Great', 'Tract', AG.
-and Huglo, Michel, 'Communion', NG.
Hudovsky, Zoran, 'Missale beneventanum MR  166 della Biblioteca metropolitana a
Zagrabia', Jucunda laudatio, 3 (1965), 306.
-'Benedictionale MR 89 of the Metropolitan Library in Zagreb', SMH 9 (1967), 55-75.
-'Neumatski rukopis Agenda Pontificalis MR 165 Metropolitanske knjiznice u Zagrebu',
Arti musices, 2 (1971), 17-30. Hughes, Andrew, Fifteenth Century Liturgical Music: I, Antiphons and Music for Holy Week
and Easter (Early English Church Music, 8; London, 1968).
-Medieval Music: The Sixth Liberal Art (Toronto, 1974; 2nd edn., 1980).
-Medieval Manuscripts for Mass and Office: A Guide to their Organization and
Terminology (Toronto, 1982).
-'Modal Order and Disorder in the Rhymed Office', MD 37 (1983), 29-52.
-'Late Medieval Rhymed Offices', JPMMS 8 (1985), 33-49.
-'Rhymed Office', NG.
FIughes, Anselm, Anglo-French Sequelae, Edited from the Papers of the Late Dr. He my
Marriott Bannister (Burnham, 1934). -The Portiforium of Saint Wulst an (Corpus Christi College, Cambridge, Ms. 391) (Henry
Bradshaw Society, 89 and 90; Leighton Buzzard, 1958, 1960).
-The Bee Missal (Henry Bradshaw Society, 94; Leighton Buzzard, 1963).
Hughes, David G., 'Further Notes on the Grouping of the Aquitanian Tropcrs', JAMS 19
(1966), 3-12.
-'Music for St. Stephen at Laon', in Merritt Festschrift 1972, 137-59.
-'Variants in Antiphon Families: Notation and Tradition', in Congress Strasbourg 1982,
li. 29-48.
-'Evidence for the Traditional View of the Transmission of Gregorian Chant', JAMS 40
(1987), 377-404.
Huglo, Michel, 'La Tradition occidentale des melodies byzantines du Sanctus', in Johner
Festschrift 1950, 40-6. -'La Melodie grecque du "Gloria in excelsis" ct son utilisation dans le Gloria XIV, RG 29
(1950), 30-40.
-'Origine de la melodic du Credo authentique de l'Edition Vaticanc', RG 30 (1951), 68-78.
-'Les Antiennes de la procession des reliques: Vestiges du chant "Vieux Roman" dans le
pontifical', RG 31 (1952), 136-9. -'Un tonaire du graduel de la fin du VHP siecle (Paris, B. N. lat. 15139)', RG 31 (1952),
176-86, 224-33.
Bibliography
lxv
-'Les Noms des neumes et lcur origine', EG 1 (1954), 53-67.
-'Le Chant vieux-romain: Manuscrits et témoins indirccts', Sacris erudiri, 6 (1954), 96-
124.
-'Vestiges d'un ancien repertoire musical dc Haute-Italie', in Congress Vienna 1954,
142-5.
-'Antifone antiche per la "fractio panis'", Ambrosius. Rivista dipastorale ambrosiana, 31
(1955), 85-95.
-'Les Prcces des graduels aquitains empruntés ä la liturgie hispanique', Hispánia sacra, 8
(1955), 361-83.
-'Le Tonaire de Saint-Bénigne de Dijon', Annales musicologiques, 4 (1956), 7-18.
-'Trois anciens manuscrits liturgiques d'Auvergne, ii: Le Graduel d'un prieuré clunisien
d'Auvergne', Bulletin historique et scientifique de VAuvergne, 11 (1957), 81-104 at 92-100.
-'Origine et diffusion des Kyrie', RG 37 (1958), 85-7.
-'Le Domaine de la notation bretonne', AcM 35 (1963), 53-84.
-'La Chironomie medievale', Revue de musicologie, 49 (1963), 155-71.
-'Le Chant des Beatitudes dans la liturgie hispanique', Hispánia sacra, 17 (1964), 135-
40.
-'Les Chants de la Missa greca de Saint-Denis', m Wellesz Festschrift 1966, 74-83.
-'Reglement du xmc siěcle pour la transcription de livres notes', in Stäblein Festschrift
1967, 121-33.
-'Un théoricien du xic siěcle: Henri d'Augsbourg', Revue de musicologie, 53 (1967), 57-9.
-'Un troisiěme témoin du tonaire carolingien', AcM 40 (1968), 22-8.
-'UAuteur du Dialogue sur la Musique attribuée ä Odon', Revue de musicologie, 55
(1969), 119-71.
-'Les Listes alléluiatiques dans les témoins du graduel grégorien', in Husmann Festschrift
1970, 219-27.
-Lj?s Tonaires: Inventaire, analyse, comparaison (Paris, 1971).
-'Der Prolog des Odo zugeschriebenen "Dialogus de Musica'", AfMw 28 (1971), 134-46.
-'Comparaison de la terminologie modale en orient et en Occident', in Congress
Copenhagen 1972, ii. 758-61. -'L'Introduction en Occident des formules byzantines d'intonation', Studies in Eastern
Chant, 3 (1973), 81-90. -'Tradition orale et tradition écrite dans la transmission des melodies grégoriennes', in
Kurt von Fischer Festschrift 1973, 31-42. -'Le Graduel palimpseste de Plaisance (Paris, B. N., lat. 7102)', Scriptorium, 28 (1974),
3-31.
-'Le Développement du vocabulaire de l'Ars Musica ä ľépoque carolingienne', I^atomus,
34 (1975), 131—51.
-'Liturgia e musica sacra aquileiese', in G. Folena (ed.), Storia della cultura veneta dalle
origini al trecento (Vicenza, 1976), 312-25. -'L'Auteur du traité de musique dedié ä Fulgence d'Affligem', Revue beige de musicologie,
31 (1977), 5-20.
-'Les Livres liturgiques de la Chaise-Dieu', RB 87 (1977), 62-96, 289-348.
-'Aux origines des tropes d'interpolation: Le trope méloform d'introíť, Revue de
musicologie, 64 (1978), 5-54. -'Les Remaniements de l'Antiphonaire grégorien au ixe siěcle: Hélisachar, Agobard,
lxvi
Bibliography
Amalaire', in Culto cristiano, politica imperiale carolinga. XX'III Coiwegno internazionalc di studi sulla spiritualita medievale (Todi, 1979), 87-120.
-'Abelard, poete et musicien', Cahiers de civilisation medievale XC-XIT siěcles, 22
(1979), 349-61.
-'On the Origins of the Troper-Proser', JPMMS 2 (1979), 11-18.
-'Les Débuts de la polyphonie ä Paris: Les premiers organa parisiens', Forum
musicologicum, 3 (1982), 93-164. -'Le Répons-Graduel de la messe: Evolution de la forme. Permanence de la fonction',
Schweizer Jahrbuch für Musikwissenschaft, ns, 2 (1982), 53-73. -'Remarques sur la notation musicale du bréviaire de Saint-Victor-sur-Rhins', RB 93
(1983), 132-6.
-'LAncien Chant bénéventain', Ecclesia orans, 2 (1985), 265-93.
-'Analyse codicologique des drames liturgiques de Fleury', in Guissen Festschrift 1985,
61-78.
-'La Notation wisigothique est-elle plus ancienne que les autres notations européennes?',
in Congress Salamanca 1985, 19-26. -'La Penetration des manuscrits aquitains en Espagne', Revista de musicologia, 8 (1985),
249-56.
--(ed.), Musicologie medievale: Notations et sequences, Table ronde du CNRS á l'IRHT
d'Orleans-La Source, 10-12 Septembre 1982 (Paris and Geneva, 1987).
-Les Livres de chant liturgique (Typologie des sources du Moyen Áge occidental, 52;
Turnhout, 1988).
-'Bibliographie des editions et études relatives ä la théorie musicale du Moyen Age (1972-
1987)', AcM 60 (1988), 229-72. -'Bilan de 50 années de recherches (1939-1989) sur les notations musicales de 850 ä 1300',
AcM 62 (1990), 224-59. -arts. 'O Antiphons', 'Offertory Chant', NCE.
-arts. 'Antiphon', 'Antiphoner', 'Breviary', 'Cluniac Monks', 'Epistle', 'Exulteť, 'Färse',
'Gallican Rite, Music of the' [trans, and rev. from 'Altgallikanische Liturgie' in Feilerer 1972, 21933], 'Gerbert d'Aurillac', 'Gospel', 'Gradual (ii)', 'Litany', 'Missal', 'Odo', 'Processional', 'Tonary', NO.
-et al., Fonti e paleografia del canto ambrosiano (Milan, 1956).
-see also RISM B/III/3.
Hunt, Noreen, Cluny under St. Hugh, 1049-1109 (London, 1967).
Huschen, Heinrich, 'Regino von Prüm, Historiker, Kirchenrechtler und Musiktheoretiker', in Feilerer Festschrift 1962, 205-23.
-arts. 'Ars musica', 'Artes liberales', 'Augustiner', 'Benediktiner', 'Dominikaner', 'Franziskaner', 'Harmonie', 'Kartäuser', 'Prämonstratenser', 'Zisterzienser', MGG.
Huschen Festschrift 1980 = Detlef Altenburg (ed.), Ars musica, musica scientia: Festschrift Heinrich Huschen zum 65. Geburtstag (Beiträge zur rheinischen Musikgeschichte, 126; Kassel, 1980).
Husmann, Heinrich, 'Die St. Galler Sequenztradition bei Notker und Ekkehard', AcM 26 (1954), 6-18.
-'Sequenz und Prosa', Annales musicologiques, 2 (1954), 61-91.
-'Das Alleluia Multifarie und die vorgregorianische Stufe des Sequenzengesanges', in
Schneider Festschrift 1955, 17-23.
Bibliography
Ixvii
-'Alleluia, Vers und Sequenz', Annales musicologiques, 4 (1956), 19-53.
-'Iustus ut palma. Alleluia und Sequenzen in St. Gallen und St. Martial', Revue beige de
musicologie, 10 (1956), 112-28.
-'Die Alleluia und Sequenzen der Mater-Gruppe', in Congress Vienna 1956, 276-84.
-'Zum Großaufbau der Ambrosianischen Alleluia', Anuario musical, 12 (1957), 17-33.
-'Alleluia, Sequenz und Prosa im altspanischen Choral', in Anglěs Festschrift 1958-61,
407-15.
-'Sinn und Wesen der Tropen', AfMw 16 (1959), 135-47.
-'Ecce puerpera genuit. Zur Geschichte der teiltextierten Sequenzen', in Besseler
Festschrift 1961, 59-65.
-Grundlagen der antiken und orientalischen Musikkultur (Berlin, 1961).
-'Das Graduale von Ediger. Eine neue Quelle der rheinischen Augustinerliturgie', in
Feilerer Festschrift 1962, 224-34. -'Die Sequenz Duo tres. Zur Geschichte der Sequenzen in St. Gallen und St. Martial', in
Handschin Gedenkschrift 1962, 66-72. -'Zur Stellung des Meßpropriums der österreichischen Augustinerchorherren', Studien
zur Musikwissenschaft, 25 (1962) (Festschrift für Erich Schenk), 261-75. -'Studien zur geschichtlichen Stellung der Liturgie Kopenhagens. Die Oster- und
Pfingstalleluia der Kopenhagener Liturgie und ihre historischen Beziehungen', Dansk
Aarbog for Musikforskning, 1962, 3-58, and 1964-5, 3-62. -'Zur Uberlieferung der Thomas-Offizien', in Smits van Waesberghe Festschrift 1963,
87-8.
-'Notre-Dame und Saint-Victor. Repertoire-Studien zur Geschichte der gereimten
Prosen', AcM 36 (1964), 98-123, 191-221. -'Zur Geschichte der Meßliturgie von Sitten und über ihren Zusammenhang mit den
Liturgien von Einsiedeln, Lausanne und Genf, AfMw 22 (1965), 217-47. -'Die Handschrift Rheinau 71 der Zentralbibliothek Zürich und die Frage nach Echtheit
und Entstehung der St. Galler Sequenzen und Notkerschen Prosen', AcM 38 (1966), 118-
49.
---'Das Einsiedelner Gradual-Sakramentar St. Paul/Kärnten 25.2.25 (Seine Stellung und
Bedeutung)', in Feicht Festschrift 1967, 89-95. -'Hymnus und Troparion. Studien zur Geschichte der musikalischen Gattungen von
Horologion und Tropologion', Jahrbuch des Staatlichen Instituts für Musikforschung,
Preußischer Kulturbesitz (Berlin, 1971), 7-86. -'Das Brevier der hl. Klara und seine Bedeutung in der Geschichte des römischen
Chorals', Studi musicali, 2 (1973), 217-33. -'Ein Missale von Assisi: Baltimore, Walters Art Gallery W. 75', in Huschen Festschrift
1980, 255-62.
-'Zur Herkunft des "Andechser Missale" Clm 3005', AfMw 37 (1980), 155-65.
-'Notre-Dame-Epoche', MGG.
-'Syrian Church Music', NG.
-see also RISM B/V/l.
Husmann Festschrift 1970 = Heinz Becker and Reinhard Gerlach (eds.), Speculum musicae artis: Festgabe für Heinrich Husmann zum 60. Geburtstag (Munich, 1970).
Mutter, Josef, Česká notace I. Neumy (Prague, 1926), II. Nota choralis (Prague, 1930).
-Notationis Bohemicae antiquae specimina selecta e codicibus Bohemicis, I. Neumae, 2.
Nota choralis (Prague, 1931).
lxviii Bibliography
Hymnarium Oscense (s. XI), 1. Edition facsimil. 11. Estudios, ed. Antonio Durän, Ramön
Moragas, Juan Villareal (Saragossa, 1987). Irtenkauf, Wolfgang, 'Das Seckauer Cantionarium vom Jahre 1345 (Iis. Graz 756)', Aßfiv
13 (1956), 116-41. -'Reimoffizium', Al GG.
Isidore, Etymologiarum sive originum libri XX, ed. W. M. Lindsay (Oxford, 1911). Iter Helveticum, ed. Pascal Ladner (Spicilegii Friburgensis Subsidia; Fribourg, 1976- ). Iter Helveticum 1 (SFS 15, 1976) = Josef Leisibach, Die liturgischen Handschriften der
Kantons- und Universitätsbibliothek Freiburg. Iter Helveticum 2 (SFS 16, 1977) = Josef Leisibach, Die liturgischen Handschriften des
Kantons Freiburg (ohne Kantonsbibliothek). Iter Helveticum 3 (SFS 17, 1979) = Josef Leisibach, Die liturgischen Handschriften des
Kapitelsarchiv in Sitten. Iter Helveticum 4 (SFS 18, 1984) = Josef Leisibach, Die liturgischen Handschriften des
Kantons Wallis (ohne Kapitelsarchiv Sitten). 1 versen, Gunilla (ed.), Research on Tropes (Stockholm, 1983). -see also CT.
Jacob, A., 'A propos de l'edition de l'Ordinaire de Tongres', Revue d'histoire ecclesiastique, 65 (1970), 789-97.
Jacobsthal, Gustav, Die chromatische Alteration im liturgischen Gesang der abendländischen Kirche (Berlin, 1897).
Jakobs, Hermann, Die Hirsauer: Ihre Ausbreitung und Rechtsstellung im Zeitalter des
Investiturstreits (Kölner historische Abhandlungen, 4; Cologne and Graz, 1961). Jammers, Ewald, 'Die Antiphonen der rheinischen Reimoffizien', EL 43 (1929), 199-219,
425-51; 44 (1930), 84-99, 342-68. -Das Karisoffizium 'Regali natus' (Sammlung musikwissenschaftlicher Abhandlungen,
14; Strasburg, 1934).
-Der gregorianischen Rhythmus: Antiphonale Studien (Strasburg, 1937).
-Die Essener Neumenhandschriften der Ixindes- und Stadtbibliothek Düsseldorf (Ratingen,
1952).
-'Die paläofrankische Neumenschrift', Scriptorium, 7 (1953), 235-59.
-Musik in Byzans, im päpstlichen Rom und im Frankenreich: Der Choral als Alusik der Textaussprache (Abhandlungen der Heidelberger Akademie der Wissenschaften, Philosophisch-historische Klasse, 1962/1; Heidelberg, 1962).
-'Einige Anmerkungen zur Tonalität des gregorianischen Gesanges', in Feilerer Festschrift
1962, 235-44.
-Tafeln zur Neumenkunde (Tutzing, 1965).
-'Der Choral als Rezitativ', AfAlw 22 (1965), 143-68.
-'Abendland und Byzanz, II: Kirchenmusik', [1969] Reallexikon der Byzantinistik, ed.
Peter Wirth.
-Das Alleluia in der Gregorianischen Alesse (Liturgiewissenschaftliche Quellen und
Forschungen, 55; Münster, 1973).
-'Cantio', MGG.
-'Cantio', NG.
Jan, K. von (ed.), Alusici scriptores graeci, 2 vols. (Leipzig, 1895-9).
Janson, Tore, Prose Rhythm in Aledieval Ixitin from the 9th to the Uth Century (Studia Latina Stockholmiensia, 20; 1975).
Bibliography
lxix
J ic a n n et e a u, Jean, Lms motlos gregoriatios: História—análisis—estética (Studia Silensia, 11;
Abadia de Silos, 1985). Jeffery, Peter, 'An Early Cantatorium Fragment Related to MS. Laon 239', Scriptorium, 36
(1982), 245-52.
-'The Oldest Sources of the Graduate: A Preliminary Checklist of MSS Copied before
about 900', Journal of Musicology, 2 (1983), 316-21. -'The Introduction of Psalmody into the Roman Mass by Pope Celestine I (422-432):
Reinterpreting a Passage in the Liber Pontificalis', Archiv für Liturgiewissenschaft, 26
(1984), 147-55.
Jesson, Roy, 'Ambrosian Chant: The Music of the Mass' (Diss., Indiana University, 1955;
UMI 00-12833). -'Ambrosian Chant', in Apel 1958, 465-83.
Johner, Dominicus, Wort und Ton im Choral: Ein Beitrag zur Aesthetik des gregorianischen
Gesanges (2nd edn., Leipzig, 1953). Johner Festschrift 1950 = Franz Tack (ed.), Der kultische Gesang der abendländischen
Kirche (Cologne, 1950). Johnstone, John G., 'Beyond a Chant: "Tui sunt caeli" and its Tropes', Studies in the
History» of Music, 1: Music and Language (New York, 1983), 24-37. Jones, Charles W., The Saint Nicholas Liturgy and its Literary Relationships (Berkeley,
1963).
Jones (Charles) Festschrift 1979 = Margot U. King and Wesley M. Stevens (cds.), Saints,
Scholars, and Heroes: Studies in Medieval Culture in Honor of Charles W. Jones, 2 vols.
(Collegeville, Minn., 1979). Jones, Cheslyn, Wainwright, Geoffrey, and Yarnold, Edward (eds.), The Study of
Liturgy (London, 1978). Jonsson,  Ritva, História: Etudes sur la genese des offices versifies  (Studia Latina
Stockholmiensia, 15; 1968). -'Corpus Troporum', JPMMS 1 (1978), 98-115.
-and Treitler,  Leo, 'Medieval Music and Language: A Reconsideration of the
Relationship', Studies in the History of Music, I: Music and language (New York, 1983), 1-23.
-see also CT.
Jounf.l, Pierre, Ij? Culte des saints dans les basiliques du Latran et du Vatican au douziéme siede (Rome, 1977).
Jungmann, Josef A., Missarum sollemnia: Eine genetische Erklärung der römischen Messe, 2 vols. (Vienna, 1948; 5th edn., 1962, 6th edn., 1966); trans. Francis A. Brunner from 2nd edn., 1951: The Mass of the Roman Rite, 2 vols. (New York, 1951, 1955); abridged version by Charles K. Riepe, 1 vol. (New York, 1959).
-Liturgisches Erbe und pastorale Gegenwart: Studien und Vorträge (Innsbruck, 1960);
trans.: Pastoral Liturgy (London, 1962).
Jungmann Festschrift 1959 = Balthasar Fischer and J. Wagner (eds.), Paschatis sollemnia: Studien zu Osterfeier und Osterffömmigkeit. Festschrift J. A. Jungmann (Basel, 1959).
Kähmer, Inge, Die Offertoriums-Uberlieferung in Rom Vat. lat. 5319 (Diss., University of Cologne, 1971).
Kainzbauer, Xaver, 'Der Tractus Tetrardus: Eine centologische Untersuchung', BzG 11 (1991), 1-132.
lxx Bibliography
Kantorowicz, Ernst H., Laudes regiae: A Study in Liturgical Acclamations and Mediaeval
Ruler Worship. With a Study of the Music of the Ixiudes and Musical Transcriptions by
Manfred F. Bukofzer (Berkeley and Los Angeles, 1946). Karnowka, G.-H., Breviarium Passaviense: Das Passauer Brevier im Mittelalter und die
Breviere der altbayrischen Kirchenprovinz (Münchener Theologische Studien, 11/44;
St. Ottilien, 1983).
Karolingisches Sakramentar: Codex Vindobonensis 958 (Codices selecti, 25; Graz, 1971). Keil, Heinrich, Grammatici latini, 8 vols. (Leipzig, 1857-80).
Kelly, Thomas Forrest, 'Melodie Elaboration in Responsory Melismas', JAMS 27 (1974), 461-74.
-'New Music from Old: The Structuring of Responsory Proses', JAMS 30 (1977), 366-
90.
-'Introducing the Gloria in excelsis , JAMS 37 (1984), 479-506.
-'Melisma and Prosula: The Performance of Responsory Tropes', in Silagi 1985, 163-80.
-'Montecassino and Old Beneventan Chant', EMH 5 (1985), 53-83.
-'Beneventan and Milanese Chant', Journal of the Royal Musical Association ,112 (1987),
173-95.
-'Neuma triplex', AcM 60 (1988), 1-30.
--The Beneventan Chant (Cambridge, 1989).
Kennedy, V. L., 'For a New Edition of the Micrologus of Bernold of Constance', in Andrieu Festschrift 1956, 229-41.
Ker, Neil R., Medieval Libraries of Great Britain: A List of Surviving Books (2nd edn., London, 1964); supplement ed. Andrew G. Watson (London, 1987).
Kienle, Ambrosius, 'Notizen über das Dirigieren mittelalterlicher Gesangschöre', Vierteljahrsschrift für Musikwissenschaft, 1 (1885), 158-69.
King, Alec Hyatt, Four Hundred Years of Music Printing (London, 1964; 2nd edn., 1968).
King, Archdale A., The Rites of Eastern Christendom (Rome, 1947-8).
-Liturgies of the Religious Orders (Milwaukee, Wise, 1955, London, 1956).
-Liturgies of the Primatial Sees (Milwaukee, Wise, and London, 1957).
-Liturgy of the Roman Church (Milwaukee, Wise, and London, 1957).
-Liturgies of the Past (Milwaukee, Wise, and London, 1959).
Klauser, Theodor, 'Eine Stationsliste der Metzer Kirche aus dem 8. Jhd. wahrscheinlich ein Werk Chrodegangs', EL 44 (1930), 162-93.
-'Die liturgischen Austauschbeziehungen zwischen der römischen und der fränkischdeutschen Kirche vom achten bis zum elften Jahrhundert', Historisches Jahrbuch der .Corres-Gesellschaft, 53 (1933), 169-89.
-Das römische Capitulare Evangeliorum, i: Typen (Liturgiewissenschaftliche Quellen
und Forschungen, 28; Münster, 1935).
-'Der Ubergang der römischen Kirche von der griechischen zur lateinischen Liturgiesprache', in Miscellanea G. Mercati (Studi e testi, 121; Rome, 1946), 467-82.
-Kleine abendländische Liturgiegeschichte (Bonn, 1965); trans.: A Short History of the
Western Liturgy (London, 1969).
Klöckner, Stefan, 'Analytische Untersuchungen an 16 Introiten im I. Ton des altrömischen und des fränkisch-gregorianischen Repertoires hinsichtlich einer bewußten melodischen Abhängigkeit', BzG 5 (1988), 3-95.
Knowles, David, The Monastic Constitutions of Lanfranc (London, 1951).
-The Monastic Order in England (Cambridge, 1940; 2nd edn., 1963).
Bibliography
lxxi
-Christian Monasticism (New York, 1969).
Kohrs, Klaus Heinrich, Die aparallelen Sequenzen: Repertoire, liturgische Ordnung, musikalischer Stil (Beiträge zur Musikforschung, 6; Salzburg, 1978).
Konrad, Karel, Geschichte des alttschechischen Kirchengesanges (Prague, 1881).
KowALEWicz, H., Cantica medii aevipolono-latina, i: Sequentiae (Bibliotheca latina medii et recentioris aevi, 15; Warsaw, 1964).
Krämer, Sigrid, Handschriftenerbe des deutschen Mittelalters (Mittelalterliche Bibliothekskataloge Deutschlands und der Schweiz, 5 (Ergänzungsband), 1. Aachen-Kochel, 2. Köln-Zyfflich; Munich, 1989).
Krieg, Eduard, Das lateinische Osterspiel von Tours (Würzburg, 1956).
Kroon, Sigurd, Ordinarium missae (Lunds universitets ärsskrift, ns, sect. 1, vol. 39, no. 6; Lund, 1953).
Kurzeja, Adalbert, Der älteste Liber Ordinarius der Trierer Domkirche: London, Brit.
Mus., Harley 2958, Anfang 14. Jh. (Liturgiewissenschaftliche Quellen und Forschungen,
52; Münster, 1970). Kyriale seu ordinarium missae (Rome, 1905).
Labhardt, Frank, Das Sequentiar Cod. 546 der Stiftsbibliothek von St. Gallen und seine
Quellen (Publikationen der Schweizerischen Musikforschenden Gesellschaft, Ser. 2, 8; 2
vols., Berne, 1959, 1963). La Fage, Adrien de, Essais de diphtherographie musicale ou notices, descriptions, analyses,
ex traits et reproductions de manuscrits relatifs ä la pratique, ä la theorie et ä l'historie de la
musique, 2 vols. (Paris, 1864) [includes edition of Paris 7211]. Laistner, Max Ludwig W., Thought and Letters in Western Europe A.D. 500-900 (London,
1931).
Lambilotte, Louis, Antiphonaire de saint Gregoire: facsimile du manuscrit de Saint-Gall
(Brussels, 1851, 2/1867). Lambres, BenoIt, 'Le Chant des Chartreux', Revue beige de musicologie, 24 (1970), 17-42. -'L'Antiphonaire des Chartreux', EG 14 (1973), 213-18.
Lamothe, Donat R., and Constantine, Cyprian G. (eds.), Matins at Cluny for the Eeast of
St. Peter's Chains (London, 1986). Landwehr-Melnicki, Margaretha, Das einstimmige Kyrie des lateinischen Mittelalters
(Regensburg, 1955). -see also MMMA and Melnicki.
Laum, Bernhard, Das alexandrinische Akzentuationssystem unter Zugrundelegung der theoretischen Lehren der Grammatiker und mit Heranziehung der praktischen Verwendung in den Papyri (Studien zur Geschichte und Kultur des Altertums, 4. Ergänzungsband; Paderborn, 1928).
Launay, Denise, 'Un esprit critique au temps de Jumilhac: Dom Jacques Le Clerc,
benedictin de la Congregation de Saint-Maur', EG 19 (1980), 197-219. Lawley, Stephen W. (ed.), Breviarium ad usum insignis ecclesiae Eboracensis (Surtees
Society, 71, 75; London, 1880, 1883). Lechner, Joseph, and Eisenhofer, Ludwig, Liturgik des römischen Ritus (Freiburg, 1953);
trans.: The Liturgy of the Roman Rite (Freiburg and Edinburgh, 1961). Leclercq, Henri, arts. 'Gallicane (liturgie)', 'Liturgies neo-gallicanes', DACL. Leclercq, Jean, L'Amour des lettres et le desir de Dieu: Initiation aux auteurs monastiques
du moyen äge (Paris, 1957); trans. Catharine Misrahi: The Love of Learning and the Desire
for God (New York, 1961; 2nd edn., 1974).
lxxii Bibliography
Ledwon, Jacob Carl, 'The Winter Office of Sant'Eutizio di Norcia: A Study of the Contents and Construction of Biblioteca Vallicelliana Manuscripts C 13 and C 5' (Diss., State University of New York at Buffalo, 1986; UMI 86-19339).
Leeb, Helmut, Die Psalmodie bei Ambrosius (Wiener Beiträge zur Theologie, 17; Vienna, 1967).
-Die Gesänge im Gemeindegottesdienst von Jerusalem (vom 5. bis 8. Jahrhunderl)
(Wiener Beiträge zur Theologie, 28; Vienna, 1979). Lefěvre, Placide F., 'La Nouvelle Edition du Processionale Praemonstratense de 1932',
Analecta praemonstratensia, 9 (1933), 170-5.
-'Le Nouvel Antiphonaire de Prémontré', Analecta praemonstratensia, 13 (1937), 63-9.
-L'Ordinaire  de Prémontré ďaprés  des manusetits du KIT et  du XIIP siécle
(Bibliothěque de la 'Revue ďhistoire eeclésiastique', 22; Louvain, 1941). -Coutumiers liturgiques de Prémontré du XI IF et du XIV siécles (Bibliothěque de la
'Revue ďhistoire eeclésiastique', 27; Louvain, 1953). -La Liturgie de Prémontré: Historie, fonnularie, chant et ceremonial (Bibliotheca
analectorum praemonstratensium, 1; Louvain, 1957). -Les Ordinaries des collégiales Saint-Pietre ä Louvain et Saints-Pierre-et-Paul á
Anderlecht d'apres des manuscrits du XIV siécle (Bibliothěque de la 'Revue ďhistoire
eeclésiastique', 36; Louvain, 1960). -L'Ordinaire de la collégiale autrefois cathédrale de Tongres, d'apres un manuscrit du
XV siécle (Spicilegium sacrum Lovaniense, Etudes et documents, 345; Louvain, 1967-8). -'L'Antiphonale psalterii d'apres le rite de Prémontré', Analecta praemonstratensia, 44
(1968), 247-74.
-'Les Répons prolixes aux heures diurnes du Triduum sacrum dans la liturgie canoniale',
Analecta praemonstratensia, 48 (1972), 5-19. Lefferts, Peter M., 'Cantilena and Antiphon: Music for Marian Services in Late Medieval
England', in Sanders Festschrift 1990, 247-82. Legg, John Wickham, Missale ad usum ecclesiae Wertmonasteriensis (Henry Bradshaw
Society, 1; London, 1891; 5 (1893); and 12 (1897).
-77?^ Processional of the Nuns of Chester (Henry Bradshaw Society, 18; London, 1899).
-77k' Sarum Missal Edited from Three Early Manuscripts (Oxford, 1916).
Le Holladay, Richard, 'The Musica Enchiriadis and Scholia Enchiriadis: A Translation
and Commentary' (Diss., Ohio State University, 1977). Lejay, Paul, 'Ambrosien (rit)', DACL.
Lemarié, J., 'Les Antiennes "Veterem hominem" du jour Octave de l'Epiphanie et les
antiennes ďorigine grecque de l'Epiphanie', EL 72 (1958), 3-38. -Le Bréviaire de Ripoll: Paris B. N. lat. 742 (Scripta et documenta, 14; Montserrat,
1965).
Leon Tello, Francisco José, Estudios de historia de la teoria musical (Madrid, 1962). Leonardi, Claudio, and Menesto, Enrico (eds.), La tradizione dei tropi liturgici (Spoleto, 1990).
Leroquais, Victor, l^es Sacramentaires et les missels manuscrits des bibliothéques publiques de France, 3 vols. (Paris, 1924).
-Les Bréviaires manuscrits des bibliothéques publiques de France, 5 vols. (Paris, 1934).
-L^e Bréviaire-missel du prieuré clunisien de Lewes (Paris, 1935).
-Les Pontificaux manuscrits des bibliothéques publiques de France, 3 vols. (Paris, 1937).
Bibliography
lxxiii
Leroquais, Victor, Psautiers manuscrits latins des bibliotheques de France, 2 vols. (Macon, 1940-1).
Le Roux, Mary Prothase, 'The "De harmonica institutione" and "Tonarius" of Regino of Prüm' (Diss., Catholic University of America, 1965; UMI 66-00318).
Le Roux, Raymond, 'Aux origines de l'office festif: Les antiennes et les psaumes de Matines et de Laudes pour Noel et le Ier Janvier', EG 4 (1961), 65-170.
-'Les Graduels des dimanches apres la Pentecöte', EG 5 (1962), 119-30.
-'Les Repons de Psalmis pour le Matines de l'Epiphanie ä la Septuagesime', EG 6 (1963),
39-148.
-'Guillaume de Volpiano. Son cursus liturgique au Mont-Saint-Michel et dans les abbayes
normandes', Millenaire monastic/lie du Mont-Saint-Michel, 1 (Paris, 1967), 417-72. -'Repons du Triduo sacro et de Päques', EG 18 (1979), 157-76.
Levy, Kenneth J., 'The Byzantine Sanctus and its Modal Tradition in East and West',
Annales musicologiques, 6 (1958-63), 7-67.
-'The Italian Neophytes' Chants', JAMS 23 (1970), 181-227.
-'"Lux de luce": The Origin of an Italian Sequence', MO 57 (1971), 40-61.
-'The Trisagion in Byzantium and the West', in Congress Copenhagen 1972, 761-5.
-'A Gregorian Processional Antiphon', Schweizer Jahrbuch für Musikwissenschaft, ns, 2
(1982), 91-102.
-'Toledo, Rome and the Legacy of Gaul', EMH 4 (1984), 49-99.
-'Old-Hispanic Chant in its European Context', in Congress Salamanca 1985, i. 3-14.
-'Charlemagne's Archetype of Gregorian Chant', JAMS 40 (1987), 1-30.
-'On the Origin of Neumes', EMH 7 (1987), 59-90.
-arts. 'Byzantine Rite, Music of the', 'Ravenna Rite, Music of the', 'Trisagion', NG.
-and Emerson, John, 'Plainchant', NG.
Libergradualis (Tournai, 1883; 2nd edn., 1895).
Liber hymnarius cum invitatoriis & aliquibus responso/iis (Solesmes, 1982). Liber vesperalis juxta ritum sanctae ecclesiae Mediolanensis [ed. G. M. Sunol] (Rome, 1939).
Lickleder, Christoph, Choral und figurierte Kirchenmusik in der Sicht Franz Xaver Witts anhand der Fliegenden Blätter und der Musica Sacra (Documenta Caeciliana, 3; Regensburg, 1988).
Lipphardt, Walter, 'Studien zur Rhythmik der Antiphonen', Mf3 (1950), 47-60, 224-34.
-'Das Moosburger Cantionale', Jahrbuch für Liturgik und Hymnologie, 3 (1957), 113-17.
-'Das Herodesspiel von Le Mans nach den Handschriften Madrid, Bibl. Nac. 288 und 289
(11. und 12. Jhd.)', in Smits van Waesberghe Festschrift 1963, 107-22. -Der karolingische Tonarvon Metz (Liturgiewissenschaftliche Quellen und Forschungen,
43; Münster-Westfalen, 1965).
-lateinische Osterfeiern und Osterspiele, 6 vols. (Berlin, 1975-81).
-'Mensurale Hymnenaufzeichnungen in einem Hymnar des 15. Jahrhunderts aus
St. Peter, Salzburg (Michaelbeuern Ms. Cart. 1)', in Cardine Festschrift 1980, 458-87. -'Liturgische Dramen', MGG.
Lloyd, Richard Wingate, 'Cluny Epigraphy', Speculum, 7 (1932), 336-49.
LoEW, Elias A very, The Beneventan Script: A History of the South Italian Minuscule.
Second edition prepared and enlarged by Virginia Brown, 2 vols. (Rome, 1980). Lohr, I., Solmisation und Kirchentonarten (Basel, 1943).
lxxiv
Bibliography
Loriquet, Henri, Pothier, Joseph, and Colette, Armand K., Le Graduel de I'eglise cathedrale de Rouen auXIIF siecle, 2 vols. (Rouen, 1907) [Paris, Bibliotheque Nationale, lat. 904].
Lot, Ferdinand (ed.), Hariulf: Chronique de I'Abbaye de Saint-Riquier (V siecle-1104)
(Collection de Textes; Paris, 1894). Lowe, Elias Avery [= Loew], 'Two New Latin Liturgical Fragments on Mount Sinai', RB
74 (1964), 252-83.
Lunden, T., Den heliga Birgitta och den helige Petrus av Skanninge: Officium parvum beate Marie Virginis [The Lady-offices of St Bridget and the Venerable Peter of Skanninge] (Acta universitatis Upsaliensis, Studia historico-ecclesiastica Upsaliensia, 27; 1976).
Lutolf, Max (ed.), Das Graduate von Santa Cecilia in Trastevere (Cod. Bodmer 74), 2 vols. (Cologne-Geneva, 1987).
Lutz, Cora E., Joannis Scotti Annotationes hi Marcianus (Cambridge, Mass., 1939).
-'Remigius' Ideas on the Classification of the Seven Liberal Arts', Traditio, 12 (1956), 65-
86.
-'The Commentary of Remigius of Auxerre on Martianus Capella', Medieval Studies, 19
(1957), 138-56.
-(ed.), Remigii Autissiodorensis commentum in Martianum Capellam, 2 vols. (Leiden,
1962-5).
-Schoolmasters of the Tenth Century (Hamden, Conn., 1977).
Maas see RISM B/III/1.
McArthur, A. A., The Evolution of the Christian Year (London, 1953).
McCarthy, M. C, The Rule for Nuns of St. Caesarius of Aries: A Translation with a Critical
Introduction (The Catholic University of America Studies in Medieval History, ns, 16;
Washington, DC, 1960). McGee, Timothy J., 'The Liturgical Placements of the Quern quaeritis Dialogue',,JAMS 29
(1976), 1-29.
Machabey, Armand, Genese de la tonalite musicale classique des origines au XV siecle (Paris, 1955).
Maciejewski, T., 'Gradual/ Chehnna [The Chclmno Gradual]', Musica mediiaevi, 4 (1973), 164-245.
McKinnon, James W., 'The Meaning of the Patristic Polemic against Musical Instruments', Current Musicology, 1 (1965), 69-82.
-'The Tenth Century Organ at Winchester', The Organ Yearbook, 5 (1974), 4-19.
-'The Fifteen Temple Steps and the Gradual Psalms', Imago musicae, 1 (1984), 29-49.
-'On the Question of Psalmody in the Ancient Synagogue', EMH 6 (1986), 159-91.
-'The Fourth-Century Origin of the Gradual', EMH 7 (1987), 91-106.
-Music in Early Christian Literature (Cambridge, 1987).
-'The Patristic Jubilus and the Alleluia of the Mass', in Cantus Planus 1990, 61-70.
McKitterick, Rosamund, 77z<? Prankish Church and the Carolingian Reforms, 789-895 (London, 1977).
-The Frankish Kingdoms under the Carolingians, 751-987 (London, 1983).
McRoberts, David, 'Some 16th-Century Scottish Breviaries and their Place in the History of
the Scottish Liturgy', The Innes Review, 3 (1952), 33-48. -'Catalogue of Scottish Medieval Liturgical Books and Fragments', The Innes Review, 3
(1952), 49-63 (published separately, Glasgow, 1953).
Bibliography
lxxv
-'The Medieval Scottish Liturgy Illustrated by Surviving Documents', Transactions of
the Scottish Ecclesiological Society, 15 (1957), 22-40. Macrobius, Commentarius in Somnium Scipionis, ed. James Willis (Leipzig, 1970);
Commentary on the Dream of Scipio, trans. William Harris Stahl (New York, 1952). Madrignac, Andre, 'Les Formules centons des Alleluia anciens', EG 20 (1981), 3-4; 21
(1986), 27-45.
Magistretti, Marco (ed.), Beroldus sive ecclesiae Ambrosianae Mediolanensis kalendarium
et ordines saec. xii (Milan, 1894). -(ed.)', Manuale Ambrosianum ex codice saec. xi olim in usum canonicae Vallis
Travaliae, 2 vols. (Milan, 1904, 1905). Mallet, Jean, and Thibaut, Andr£, Les Manuscrits en ecriture beneventaine de la
Bibliotheque capitulaire de Benevent, i: Manuscrits 1-18 (Paris, 1984). Manitius, Maximilianus, Geschichte der lateinischen Literatur des Mittelalters, 3 vols.
(Handbücher der klassischen Altertumswissenschaft; Munich, 1911-31). Mansi, Giovanni Domenico, Sacrorum conciliorum nova et amplissima collectio, 31 vols.
(Florence and Venice, 1757-98); 53 vols. (Paris, Leipzig, and Arnheim, 1901-27). Marcusson see CT.
Markovits, M., Das Tonsystem der abendländischen Musik im frühen Mittelalter (Publikationen der Schweizerischen Musikforschenden Gesellschaft, ser. 2, 30; Berne, 1977).
Marossz£ki, Solutor Rodolphe, Les Origins du chant cistercien: Recherches sur les reformes du plain-chant cistercien au XIIe siede (Analecta Sacri Ordinis Cisterciensis, 8; Rome, 1952).
Martene, Edmond, De antiquis ecclesiae ritibus, 3 vols. (Rouen, 1700-2).
Martianus Capella, De nuptiis Philologiae et Mercurii, ed. James Willis (Leipzig, 1983);
The Marriage of Philology and Mercury: Martianus Capella, trans. William Harris Stahl
(New York, 1977).
Martimort, Aim£-Georges (ed.), L'Eglise enpriere: Introduction ä la liturgie (Paris, 1961; 3rd edn., 1965); first two of the four parts trans.: The Church at Prayer: Introduction to the Liturgy (New York, 1968) and The Church at Prayer: The Eucharist (New York, 1973).
-'Origine et signification de l'alleluia de la messe romaine', in Quasten Festschrift 1970,
811-34.
-(ed.), L'Eglise en priere, edition nouvelle, 4 vols. (Paris, 1983), trans. Matthew J.
O'Connell: The Church at Prayer, New Edition, 4 vols. (Collegeville, Minn., and London,
1987, 1987, 1988, and 1986). -'A propos du nombre des lectures ä la messe', Revue des sciences religieuses, 58
(1984), 42-51.
Marxer, Otto, Zur spätmittelalterlichen Choralgeschichte St. Gallens: Der Codex 546 der
St. Galler Stiftsbibliothek (VGA 3; St. Gallen, 1908). Mas, Josiane, 'La Notation catalane', Revista de musicologia, 11 (1988), 11-30. Mathias, R. X., Die Tonarien (Diss., Graz, 1903).
Meier, Bernhard, 'Modale Korrektur und Wortausdeutung im Choral der Editio Medicaea',
Kmjb 53 (1969), 101-32. Melnicki, Margareta, and Stäblein, Bruno, 'Graduale (Buch)', MGG. -see also MMMA and Landwehr-Melnicki.
Menager, Armand, 'Apergu sur la notation du manuscrit 239 de Laon. Sa concordance avec
les codices rythmiques sangalliens', PalMus 10, 177-211. -'Etude sur la notation du manuscrit 47 de Chartres', PalMus 11, 41-131.
lxxvi Bibliography
Mercenier, F., 'La Plus Ancienne Priere ä la Vierge "Sub tuum praesidium'", Questions
liturgiques et paroissiales, 25 (1940), 33. Merlette, B., 'Ecoles et bibliotheques ä Laon du declin de l'antiquite au developpement de
l'universite', Enseignement et vie intellectuelle, Actes du 95'' Congres des Societes Savantes
(Paris, 1975), i. 21-53. Merritt Festschrift = Laurence Berman (ed.), Words and Music: The Scholar's Vtezv. A
Medley of Problems and Solutions Compiled in Honor of A. Tillman Merritt by Sundry
Hands (Cambridge, Mass., 1972). Mettenleiter, Dominicus, Musikgeschichte der Stadt Regensburg (Regensburg, 1865). Meyer, Christiane RISM B/II 1/3.
Meyer, Kathi, 'The Eight Gregorian Modes on the Cluny Capitals', Art Bulletin, 34 (1952), 75-94.
MGH (Antiquitates) Poetae Ixitini aevi Carolini, i, ed. Ernst Dümmler (1881). MGH (Antiquitates) Poetae Latini aevi Carolini, iv/1, ed. Paul von Winterfeld (1899). MGH (Epistolae) Epistolae Karolini aevi, ii, ed. Ernst Dümmler (1895). MGH (Epistolae) Epistolae Karolini aevi, iii, ed. Ernst Dümmler, Karl Hampe, et al. (1898-9).
MGH (Leges) Capitularia regum Francorum, i, ed. Alfred Boretius (1883).
MGH (Leges) Concilia aevi Karolini, i/2, ed. Albert Werminghoff (1906).
MGH (Scriptores) Annales et chronica aevi Salici, ed. Georg Heinrich Pertz (1844).
MGH (Scriptores) Die Chronik von Montecassino (Chronica monasterii Casinensis), ed.
Hartmut Hoffmann (1980). MGH (Scriptores) Liber Pontificalis, i, ed. Theodor Mommsen (1898). MGH (Scriptores) Passiones vitaeque sanctorum aevi Merovingici et antiquiorum aliquot, \,
ed. Bruno Krusch (1896). MGH (Scriptores) Scriptores rentm Sangallensium: Annates, chronica et historiae aevi
Carolini, ed. Georg Heinrich Pertz (1829). Miazga, Tadeusz, Die Melodien des einstimmigen Credo der römisch-katholischen lateinischen
Kirche: Eine Untersuchung der Melodien in den handschriftlichen Uberlieferungen mit
besonderer Betiicksichtigung der polnischen Handschriften (Graz, 1976). -Pontyfikaly Polskie (Graz, 1981).
Miller, John II., Fundamentals of the Liturgy (Notre Dame, Ind., 1959). -'Liturgy', NCE.
Milveden, Ingmar, 'Manuskript, Mönch und Mond. Ein Hauptteil des Cod. Upsal. C 23 in quellenkritischer Beleuchtung', Svensk tidskrift för musikforskning, 46 (1964), 9.
-'Neue Funde zur Brynolphus-Kritik', Svensk tidskrift för musikforskning, 54 (1972), 5-
51.
Missale notatum Strigoniense ante 1341 in Posonio, ed. Janka Szendrei and Richard Rybaric
(Musicalia Danubiana, 1; Budapest, 1982). Missale plenarium Bib. Capit. Gnesnensis Ms. 149, ed. Krzysztof Bieganski and Jerzy
Woronczak (Antiquitates musicae in Polonia, 1112; Warsaw and Graz, 1970, 1972). Misset, E., and Weale, W. H. J., Analecta liturgica II: Thesaurus hymnologicus 1-2
(Bruges and Lille, 1888, 1892).
-and Aubry, Pierre, Les Proses d'Adam de Saint Victor (Paris, 1900).
Mittelalterliche Bibliotheks-Kataloge Deutschlands und der Schweiz (Munich, 1918- ) (see
also Krämer 1989). Mittelalterliche Bibliotheks-Kataloge Oesterreichs (Vienna, 1915- ).
Bibliography
lxxvii
M MM A 1 (1956) = Hymnen I. Die mittelalterlichen Hymnenmelodien des Abendlandes, cd. Bruno Stäblein.
MMMA 2 (1970) = Bruno Stäblcin (introduction) and Margareta Landwehr-Melnicki
(edition), Die Gesänge des altrömischen Graduate Vat. lat. 5319. MMMA 3 (1970) = lntroitus-Tropen I. Das Repertoire der südfranzösischen Tropare des 10.
und 11. Jahrhunderts, ed. Günther Weiß. MMMA 7 (1968) = Alleiuia-Melodien 1, bis J100, ed. Karlheinz Schlager. MMMA 8 (1987) = Alleiuia-Melodien II, ab 1100, ed. Karlheinz Schlager. MM S 1 (Mäcon, 1952) = Le Prosaire de la Sainte-Chapelle [part of Ban, Biblioteca
Capitolare, 1].
MMS 2 (Mäcon, 1954) = Les Manuscrits musicaux de Jumiéges.
MMS 3 (Touen, 1960) = Le Prosaire d'Aix-la-Chapelle [part of Aachen, Bischöfliche
Diözesanbibliothek, 13 (XII)]. MMS 3 (Rouen, 1960) = Le Trop aire-pros aire de Dublin [part of Cambridge, University
Library, Add. 710].
MMS 5 (Pans, 1981) = Le Graduel de St. Denis [Paris, Bibliothéque Mazarine, 384]. Moberg, Carl Allan, Uber die schwedischen Sequenzen: Eine musikgeschichtliche Studie, 2
vols. (VGA 13; Uppsala, 1927). -Die liturgischen Hymnen in Schweden (Uppsala, 1947).
Mocquereau, Andre, 'Origine et classement de differentes écritures neumatiques: 1. Notation oratoire ou chironomique, 2. Notation musicale ou diastématique', PalMus 1, 96-160.
-'Neumes-accents liquescents ou sémi-vocaux', PalMus 2, 37-86.
-l^e Nombre musical grégorien ou tythmique grégorienne, 2 vols. (Rome, 1908-1927).
-'La Pensée pontiŕicale et la restauration grégorienne', RG 5 (1920), 181-9; 6 (1921), 9-
18, 46-53.
Modf.rini, Ave, Im notazione neumatica di Nonantola, 2 vols. (Cremona, 1970). Moeller, E., Corpus benedictionum pontificalium, 4 vols. (Corpus Christianorum, 162; Turnhout, 197-19).
Mohlberg, L. Cunibert (ed.), Missale Gothicum (Rcrum ecclesiasticarum documenta, Series maior, Fontes, 5; Rome, 1961).
-Eizenhöfer, L., and Siffrin, P. (eds.), Sacramentariurn Veronense (Rerum ecclesiasticarum documenta, Series maior, Fontes, 1; Rome, 1956).
----(eds.), Missale Gallicanum vetus (Rerum ecclesiasticarum documenta, Series
maior, Fontes, 3; Rome, 1958).
---(eds.), Liber sacramentorum Romanae aecclesiae ordinis anni circuit (Sacramentariurn Gelasianum) (Rerum ecclesiasticarum documenta, Series maior, Fontes, 4; Rome, 1960).
Mohlberg Festschrift 1948 = Miscellanea liturgica in honorem L. Cunibert Mohlberg, 2 vols.
(Bibliotheca ephemerides liturgicae, 23; Rome, 1948). Molitor, Raphael, Die nach-Tridentinische Choralreform zu Rom, 2 vols. (Leipzig,
1901-2).
-Deutsche Choral-Wiegendrucke: Ein Beitrag zur Geschichte des Chorals und des
Notendruckes in Deutschland (Regensburg, 1904). Möller, Hartmut, 'Research on the Antiphoner—Problems and Perspectives', JPMMS 10
(1987), 1-14.
-'Zur Reichenauer Offiziumstradition der Jahrtausendwende', SMH 29 (1987), 35-61.
lxxviii Bibliography
Möller, Hartmut, 'Deutsche Neuměn—St. Galler Neuměn. Zur Einordnung der Echternacher Neumenschrifť, SMII 30 (1988), 415-30. -(ed.), Das Quedlinburger Antiphonar (Berlin, Staatsbibliothek Preußischer Kulturbesitz
Mus. ms. 40047), 3 vols. (Tutzing, 1990). -'Die Prosula "Psalle modulamina" (Mü 9543) und ihre musikhistorische Bedeutung', in
Leonardi and Menesto 1990, 279-96. Mone, F. J., Lateinische und griechische Messen aus dem ziveiten bis sechsten Jahrhundert
(Frankfurt, 1850; PL 138, 862-82). Moneta Caglio, Ernesto, 'I responsori "cum infantibus" nella liturgia ambrosiana', in
Castiglioni Festschrift 1957, 481-574. Morawski, Jerzy, Polska li/yka muzycna w sredniowieczu: Repertuarsekwencyjny cystersów
(XIII-XVI w.) [résumé: 'Polish musical lyric in the Middle Ages: The Cistercians'
sequence repertory (13th—16th c.)'] (Warsaw, 1973). -'Recherches sur les variantes regionales dans le chant grégorien', SMH 30 (1988), 403-
14.
Morin, Germain, 'Fragments inédits et jusqu'ä present uniques d'Antiphonaire Gallican', RB 22 (1905), 329-56.
-'Les Plus Anciens Comes ou lectionnaire de l'Eglise romaine', RB 27 (1910), 41-74.
-'Liturgie et basiliques de Rome au milieu du viic siěcle', RB 28 (1911), 296-330.
-'Le Plus Ancien Monument qui existe de la liturgie gallicane: Le lectionnaire palimpseste
de Wolfenbüttel', EL 51 (1937), 3-12.
-(ed.), Sancti Caesarii Arelatensis opera omnia (Maredsous, 1942).
Mühlmann, Wilhelm, Die Alia Musica (Gerbert, Scriptores l): Quellenfrage, Umfang, Inhalt
und Stammbaum (Leipzig, 1914). Müller, Herman, Die Musik Wilhelms von Hirschau. Widerherstellung, Ubersetzung und
Erklärung seines musik-theoretischen Werkes (Frankfurt, 1883).
-Hucbalds echte und unechte Schriften über Musik (Leipzig, 1884).
Muller, H. F., 'Pre-history of the Medieval Drama: The Antecedents of the Tropes and the
Conditions of their Appearance', Zeitschrift für romanische Philologie, 44 (1924-5), 544-
75.
Müller-Blattau Festschrift 1966 = Zum 70. Geburtstag von Joseph Müller-Blattau
(Saarbrücker Studien zur Musikwissenschaft, 1; Kassel, 1966). Murphy, Joseph Michael, 'The Communions of the Old Roman Chant' (Diss., University of
Pennsylvania, 1977; UMI 78-06625). Murray, Gregory, Gregorian Chant according to the .Manuscripts (London, 1963). La Musique dans la liturgie {La Maison-Dieu, 108; 1971).
Nejedlý, Zdeněk, Zpěv předhusitský/Geschichte des vorhussitischen Gesanges in Böhmen (Prague, 1904; 2ndedn., 1954) [vol. i of Djiny husitského zpěvu, 3 vols. (Prague, 1904-13; 2nd edn., 1954-6)].
Netzer, H., LTntroduction de la messe romaine en France sous les Carolingiens (Paris, 1910).
Neufville, J. see Vogue, Adalbert de.
Nicholson, Edward Williams Byron, Early Bodleian Music 3: Introduction to the Study of Some of the Oldest Latin Manuscripts in the Bodleian Library Oxford (London, 1913).
Nicolau, M. G., L'Origine du 'cursus' rythmique et les debuts de Vaccent ďintensité en latin (Paris, 1930).
Bibliography
lxxix
Niemöller, Klaus Wolfgang, Die Musica gregoriana des Nicolaus Wollick: Opus aureum, Köln 1501, pars 1111 (Beiträge zur rheinischen Musikgeschichte, 11; Cologne, 1955).
Norberg, Dag, Introduction á Vétude de la versification latine medievale (Studia Latina Stockholmiensia, 5; 1958).
Norden, Eduard, Die antike Kunstprosa (4th edn., Leipzig, 1923).
NoRTiER, Genevieve, 'Les Bibliothěques médiévales des abbayes bénédictines de Normandie',
Revue Mabillon, 187 (1957), 1-34, 57-83, 135-71, 214-44; 188 (1958), 1-19, 99-127,
165-75, 249-57; new edition (Paris, 1971). Norton, Michael Lee, 'The Type II Visitatio Sepulchri: A Repertorial Study' (Diss., Ohio
State University, 1983; UMI 83-11783). -'Of "Stages" and "Types" in Visitatione Sepulchri', Comparative Drama, 21 (1987), 34-
61, 127-44.
Nowacki, Edward, 'The Syntactical Analysis of Plainchant', in Congress Berkeley 1977, 191-201.
-'Studies on the Office Antiphons of the Old Roman Manuscripts' (Diss., Brandeis
University, 1980; UMI 80-24546). -'The Gregorian Office Antiphons and the Comparative Method', Journal of Musicology,
4 (1985), 243-75.
-'Text Declamation as a Determinant of Melodic Form in the Old Roman Eight-Mode
Tracts', EMH 6 (1986), 193-225. -'The Performance of Office Antiphons in Twelfth-Century Rome', in Cantus Planus
1990, 79-92.
Odelman, Eva, 'Comment a-t-on appelé les tropes? Observations sur les rubriques des tropes
des xe et xic siěcles', Cahiers de civilisation medievale, 18 (1975), 15-36. -see also CT.
Oesch, Hans, Guido von Arezzo: Biographisches und Theoretisches unter besonderer Berücksichtigung der sogennanten odonischen Traktate (Publikationen der Schweizerischen Musikforschenden Gesellschaft, ser. 2, 4; Berne, 1954).
-Berno und Hermann von Reichenau als Musiktheoretiker (Publikationen der Schweizerischen Musikforschenden Gesellschaft, ser. 2, 9; Berne, 1961).
Offertoriale triplex, ed. Rupert Fischer (Solesmes, 1985) [Ott 1935 with neumes of Laon 239 and Einsiedeln 121 or St Gall 339].
Omlin, Ephrem, Die Sankt Gallischen Tonarbuchstaben: Ein Beitrag zur Entwicklungsgeschichte der Offiziumsantiphonen in Bezug auf ihre Tonarten und Psalmkadenzen (VGA 18; Regensburg, 1934).
Onnerfors, A., 'Zur Offiziendichtung im schwedischen Mittelalter—mit einer Edition des Birger Gregersson zugeschriebenen "Officium S. Botuida'", Mittellateinisches Jahrbuch, 3 (1966), 55-93.
Orel, Dobroslav, Kancionál Franusäv z roku 1505: Příspěvek k vývoji notace a k dějinám zpěvu jednohlasého i vícehlasého doby Jage Hov cň v Cechách (Prague, 1922).
-Hornof, Vladimír, and Vosyka, Václav, Český Kancionál (Prague, 1921).
Ossing, H., Untersuchungen zum Antiphonale Monasteriense (Alopecius Druck 1537): Ein Vergleich mit den Handschriften des Münsterlandes (Kölner Beiträge zur Musikwissenschaft, 39; Regensburg, 1966).
Ott, Karl, Offertoriale sive versus offertoriorum (Tournai, 1935).
Ottosen, Knud, UAntiphonaire latin au Moyen-Age: Reorganisation des series de répons de
lxxx
Bibliography
VAvent classes par R.-J. Hesbert (Rcrum ecclesiasticarum documenta, extra sericin; Rome, 1986).
Ottösson, Robert A., Sancti Thorlaci Episcopi officia rhythmica et proprium missae in AM
241a folio (Bibliotheca Arnamagnaeana Supplementum, 3; Copenhagen, 1959). Oxford Classical Dictionary, ed. Nicholas G. Hammond (2nd edn., Oxford, 1970). Oxford Dictionary of the Christian Church, ed. Frank L. Cross and Elizabeth A. Livingstone
(2nd edn., Oxford, 1974). Page, Christopher, 'The Earliest English Keyboard', Early Music, 7 (1979), 309-14. Palisca, Claude V., arts. 'Anonymous Theoretical Writings', 'Theory, Theorists', iYG\ PalMus 1 (Solesmes, 1889) = le Codex 339 de la Bibliotheque de Saint-Gall (Xe siecle):
Antiphonale missarum sancti Gregorii. PalMus 2 (Solesmes, 1891) = le Repons-graduel Justus ut palma, reproduit en facsimile
d'apres plus de deux cents antiphonaires manuscrits du IX1' au XVIF siecle. PalMus 3 (Solesmes, 1892) = Le Repons-graduel Justus ut palma: Deuxieme partie. PalMus 4 (Solesmes, 1894) = Le Codex 121 de la Bibliotheque d'Einsiedeln (IX'-XF siecle):
Antiphonale missarum sancti Gregorii. PalMus 5 (Solesmes, 1896) = Antiphonarium Ambrosianum du Musee Britannique (XII''
siecle), Codex Additional 34209. PalMus 6 (Solesmes, 1900) = Antiphonarium Ambrosianum du Musee Britannique (XIF
siecle), Codex Additional 34209: Transcription. PalMus 7 (Solesmes, 1901) - Antiphonarium tonale missarum, XT siecle: Codex II. 159 de la
Bibliotheque de l'Ecole de Medecine de Montpellier. PalMus 8 (Solesmes, 1901-5) = Antiphonarium tonale missarum, XL siecle: Codex II. 159 de
la Bibliotheque de l'Ecole de Medecine de Montpellier. Phototypies. PalMus 9 (Solesmes, 1906) = Antiphonaire monastique, XII1 siecle: Codex 601 de la
Bibliotheque Capitulaire de Lucques. PalMus 10 (Solesmes, 1909) = Antiphonale missarum sancti Gregorii, IX'-Xe siecle: Codex
239 de la Bibliotheque de Imoh. PalMus 11 (Solesmes, 1912) = Antiphonale missarum sancti Gregorii, X1' siecle: Codex 47 de
la Bibliotheque de Chart res. PalMus 12 (Solesmes, 1922) = Antiphonaire monastique, XIIT siecle: Codex F. 160 de la
Bibliotheque de la Cathedrale de Worcester. PalMus 13 (Solesmes, 1925) = Le Codex 903 de la Bibliotheque Nationale de Parts (XL
siecle): Graduel de Saint-Yrieix. PalMus 14 (Solesmes, 1931) = Le Codex 10673 de la Bibliotheque Vaticane fonds latin (XL
siecle): Graduel Beneven tain. PalMus 15 (Solesmes, 193757) = Le Codex VI. 34 de la Bibliotheque Capitulaire de Benevent
(XL'-Xir siecle): Graduel de Benevent avec prosaire et tropaire. PalMus 16 (Solesmes, 1955) = Le Manuscrit du Mont-Renaud, X' siecle: Graduel et
antiphonaire de Noyon. PalMus 17 (Solesmes,  1958)  = Fragments des manuscrits de Chartres, reproduction
phototypique: Presentation par le Chanoine Yves Delaporte. PalMus 18 (Berne, 1969) - Le Codex 123 de la Bibliotheque Angelica de Rome (XT siecle):
Graduel et tropaire de Bologne. PalMus 19 (Berne, 1974) = le Manuscrit 807, Universitätsbibliothek Graz (X1F siede):
Graduel de Klosterneuburg.
Bibliography
lxxxi
PalMus 20 (Berne, 1983) = L<? Manuscrit VI-33, Archivio Arcivescovile Benevento: Missel de Bénévent (debut du XT siecle).
PalMus 11/1 (Solesmes, 1900) = Antiphonaire de I'office monastique transcrit par Hartker: MSS. Saint-Gall 390-391 (980-1011).
PalMus 11/2 (Solesmes, 1924) = Canlatorium, IX'' siěcle: N° 359 de la Bibliothěque de Saint-Gall.
Pannain, Guido, 'Liber musicae. Un teorico anonimo del XIV secolo', Rivista musicale
italiana, 27 (1920), 407-40. Paredi, A., 'Milanese Rite', NCE.
Parvio, M., Missale Aboense secundum ordinem Fratrum Praedicatorum (Helsinki, 1971). Pa scher, Joseph, Das liturgische Jahr (Munich, 1963).
PATier, Dominique, 'Un office rythmique tchěque du xiv"m' siěcle: Étude comparative avec
quelques offices hongrois', SMH 12 (1970), 41-129.
-'L'Office rythmique de sainte Ludmila', EG 21 (1986), 49-96.
Perego, Camillo, La regola del canto fermo arnbrosiano (Milan, 1622).
Pesce, Dolores, 'B-Flat: Transposition or Transformation V, Journal of Musicology, 4 (1985—
6), 330-49.
-The Affinities and Medieval Transposition (Bloomington, Ind., 1987).
Pfaff, Richard W., New Liturgical Feasts in Later Medieval England (Oxford, 1970).
-Medieval Latin Liturgy: A Select Bibliography (Toronto Medieval Bibliographies, 9;
Toronto, 1982).
Phillips, Nancy, 'Musica and Scolica enchiriadis: Its Literary, Theoretical, and Musical
Sources' (Diss., New York University, 1984; UMI 85-05525).
-'The Dasia Notation and its Manuscript Tradition', in Huglo 1987, 157-73.
-and Huglo, Michel, 'The Versus Rex caeli—Another Look at the So-called Archaic
Sequence', JPMMS 5 (1982), 36-43. --'Le "De musica" de saint Augustin et l'organisation de la durée musicale du ixc au
xii1' siěcles', Recherches augustiniennes, 20 (1985), 117-31. Picone, Carmelo, 'II "salicus" con lettere espressive nel codice di Laon 239', EG 16 (1977), 7-
143.
Pikulik, Jerzy (ed.), Etat des recherches sur la musique religieuse dans la culture polonaise (Warsaw, 1973).
-'Les Offices polonais de saint Adalbert', in Pikulik 1973, 306-72.
-'Sekwencje Polskie/Polish sequences', Musica medii aevi, 4 (1973), 7—128; 5 (1976), 6—
194.
-Indeks sekwencji w polskich rtfiopisach muzycznych [résumé: 'Sequence index in the
Polish   musical   manuscripts'].   Sekwencje  zespoiu  r$kopisów   tarnowskich [résumé:
'Sequences from the Tarnow manuscript set'] (Warsaw, 1974). -'Indeks špiewów ordinarium missae w gradualach polskich do 1600 R.' [résumé: 'Index
des chants polonais de l'ordinaire de la messe au moyen äge'], Muzyka religijna w Polsce:
Materiály i studia, 2 (1978), 139-272. -Spiewry alleluia o najšwiptszej Maryi Pannie w polskich gradualach przedtrydenckich
[résumé: 'Les chants alleluia en 1'honneur de la Vierge Marie dans les graduels polonais
antétridentins'], Muzyka religijna w Polsce: Materiály i studia, 6 (1984) [complete
volume].
-'Les Tropes du Kyrie et du Sanctus dans les graduels polonais médiévaux', in Leonardi
and Menesto 1990, 325-35.
lxxxii
Bibliography
Pineau, C, Le Plain-chant musical en France au XVIT siecle (n.p., 1955). Pinell, J. M., 'Los textos de la antigua liturgia hispänica: Fuentes para su estudio', in Rivera Recio 1965, 109-64.
Pitman, Grover Allen, 'The Lenten Offertories of the Aquitanian Manuscripts' (Diss.,
Catholic University of America, 1973; UMI 73-12885). Planchart, Alejandro Enrique, The Repertory of Tropes at Winchester, 2 vols. (Princeton,
1977).
-'The Transmission of Medieval Chant', in Iain Fenlon (ed.), Music in Medieval and
Early Modern Europe: Patronage, Sources and Texts (Cambridge, 1981), 347-63.
-'About Tropes', Schweizer Jahrbuch für Musikwissenschaft, ns, 2 (1982), 125-35.
-'Italian Tropes', Mosaic, 18/4 (1985), 11-31.
-'On the Nature of Transmission and Change in Trope Repertories', JAMS 41 (1988),
215-49.
-'The Interaction between Montecassino and Benevento', in Leonardi and Menesto 1990,
385-407.
-and Fuller, Sarah, 'St Martial', NG.
Planer, John Harris, 'The Ecclesiastical Modes in the Late Eighth Century' (Diss.,
University of Michigan, 1970; UMI 71-15270). Plocek, Vaclav, Catalogus codicum notis musicis instructorum qui in Bibliotheca publica rei
publicae Bohemicae socialisticae in Bibliotheca universitalis Pragensis servantur, 2 vols.
(Prague, 1973).
Pocknee, Cyril E., The French Diocesan Hymns and their Melodies (London, 1954). Podlaha,  Adolf Patera-Antonin, Handschriften-Katalog der Bibliothek des Prager
Metropolitankapitels (Prague, 1910). Polheim, Karl, Die lateinische Reimprosa (Berlin, 1925; 2nd edn., 1963). Ponchelet, Rene, 'Le Salicus en composition dans le codex St. Gall 359', EG 14 (1973), 7-
125.
Ponte, Joseph P., 'Aureliani Reomensis Musica disciplina: A Revised Text, Translation and
Commentary' (Diss., Brandeis University, 1961; UMI 62-01207). -Aurelian of Reöme (ca. 843): The Discipline of Music (Musica disciplina) (Colorado
College Music Press, Translations, 3; 1968). Poras, Gr£goire, 'Recherches sur les conditions du chant liturgique pendant le haut moycn
äge', EG 21 (1986), 23-5. Porter, W. S., The Gallican Rite (London, 1958).
Pothier, Joseph, I^es Melodies gregoriennes d'apres la tradition (Tournai, 1880). Potiron, Henri, Im Theorie harmonique de trois groupes modaux et Vaecord final des 3e et 4'' modes (Monographies gregoriennes, 6; Tournai, 1925).
-La Modalite gregorienne (Monographies gregoriennes, 9; Tournai, 1928).
-L'Origine des modes gregoriennes (Tournai, 1948).
-La Notation grecque et Boece (Tournai, 1951).
-Boece, theoricien de la musique grecque (Paris, 1954).
-'La Modalite de la Messe V, RG 33 (1954), 45-8.
-'La Notation grecque dans l'Institution harmonique d'Hucbald', EG 2 (1957), 37-50.
-'La Terminologie grecque des modes', EG 7 (1967), 57-61.
-'Valeur et traduction de la notation grecque', EG 15 (1975), 193-9.
Powers, Harold S., 'Mode', NG.
Prado, German, Manual de liturgia hispano-visigotica o mozärabe (Madrid, 1927).
Bibliography
lxxxiii
Prado, German, Historia del rito mozárabe y toledano (Santo Domingo de Silos, 1928).
-Cantus lament at'ion urn pro ultimo triduo hebdomadae majoris juxta hispanos codices
(Paris, 1934).
Prim, J., '"Chant sur le livre" in French Churches in the 18th Century', JAMS 14 (1961), 37-49.
Processionale ad usum Sarum (London: Richard Pynson, 1502) (fac, Clarabricken, 1980). Processionale monasticum ad usum congregationis gallicae ordinis Sancti Benedicti (Solesmes, 1893).
Procter, Francis, and Wordsworth, Christopher (eds.), Breviarium ad usum insignis
ecclesiae Sarum, 3 vols. (Cambridge, 1882, 1879, 1886). Psalterium monasticum (Solesmes, 1981).
Quasten Festschrift 1970 = P. Granfield and Josef Jungmann (eds.), Kyriakon: Lastschrift
Johannes Quasten, 2 vols. (Münster, 1970). Quentin, Henri, Les Martyrologes historiques du moyen age (Paris, 1908). Raasted, J0rgen, Intonation Formulas and Modal Signatures in Byzantine Musical
Manuscripts (Monumenta musicae Byzantinae, Subsidia, 7; Copenhagen, 1966). Raby, F. J. E., A History of Christian-Latin Poetry from the Beginnings to the Close of the
Middle Ages (Oxford, 1927; 2nd edn., 1953). Rado, Polycarp, Libri liturgici manuscript! bibliothecarum Hungariae et limitropharum
regionům* (Budapest, 1973). Rajecky, Benjamin, and Rado, Polycarp, Hymnen und Sequenzen (Melodiarum Hungariae
medii aevi, 1; Budapest, 1956; rev. edn. 1982 with new supplementary vol.). Rand, Edward Kennard, Founders of the Middle Ages (Cambridge, Mass., 1929). Randel, Don M., The Responsorial Psalm Tones for the Mozarabic Office (Princeton, 1969).
-'Responsorial Psalmody in the Mozarabic Rite', EG 10 (1969), 87-116.
-An Index to the Chant of the Mozarabic Rite (Princeton, 1973).
-'Antiphonal Psalmody in the Mozarabic Rite', in Congress Berkeley 1977, 414-22.
-'El antiguo rito hispánico y la salmodía primitiva en occidente', Revista de musicologia, 8
(1985), 229-38. -'Mozarabic Rite, Music of the', NG.
Ranke, E., Codex Fuldensis NT latine, interprete Hieronymo (Marburg and Leipzig, 1868). Rankin, Susan K., 'The Music of the Medieval Liturgical Drama in France and in England'
(Diss., University of Cambridge, 1981). -'The Mary Magdalene Scene in the Visitatio Sepulchri Ceremonies', EMH 1 (1981),
227-55.
-'A New English Source of the Visitatio Sepulchri', JPMMS 4 (1981), 1-11.
-'From Memory to Record: Musical Notations in Manuscripts from Exeter', Anglo-Saxon
England, 13 (1984), 97-112. -'The Liturgical Background of the Old English Advent Lyrics: A Reappraisal', in
Clemoes Festschrift 1985, 317-40.
-'Musical and Ritual Aspects of Quern queritis', in Silagi 1985, 181-92.
-'Neumatic Notations in Anglo-Saxon England', in Huglo 1987, 129-44.
Rasmussen, Niels Kroch, Les Pontificaux du haut moyen-äge ([Diss., Paris]; Aarhus,
1977).
Ratcliff, E. C, Expositio antiquae liturgiae Gallicanae (Henry Bradshaw Society, 98; London, 1971).
lxxxiv
Bibliography
Reckow, Fritz, Der Musiktraktat des Anonymus 4: Edition und Interpretation der Organum
purum-Lehre (Beihefte zum Archiv für Musikwissenschaft, 4-5; Wiesbaden, 1967). Reese Festschrift 1967 = Jan LaRue (ed.), Aspects of Medieval and Renaissance Music,
a Birthday Offering to Gustave Reese (New York, 1967). Rehle, Sieghild, 'Missale Beneventanum (Codex VI 33 des Erzbischöflichen Archivs von
Benevent)', Sacris erudiri, 21 (1972-3), 323-405. -and Gamber,  Klaus, Sacramentarium Antonis: Die Fragmente des Salzburger
Exemplars (Textus patristici et liturgici, 8 and 10; Regensburg, 1970, 1973). Reichert, Georg, 'Strukturprobleme der älteren Sequenz', Deutsche Vierteljahrsschrift für
Literaturwissenschaft und Geistesgeschichte, 23 (1949), 227-51. Reier, Ellen Jane, 'The Introit Trope Repertory at Nevers: MSS Paris B. N. lat. 9449 and
Pans B. N. n. a. lat. 1235' (Diss., University of California at Berkeley, 1981; UM I 82-
12077).
Reimer, Erich (ed.), Johannes de Garlandia: De mensurabili musica. Kritische Edition mit Kommentar und Interpretation der Notationslehre (Beihefte zum Archiv für Musikwissenschaft, 10-11; Wiesbaden, 1972).
-'Musicus und Cantor. Zur Sozialgeschichte eines musikalischen Lehrstücks', AfMw 35
(1978), 1-32.
Renaudin, Andre, 'Deux antiphonaires de Saint-Maur-des-Fosses: B. N. lat. 12584 et
12044', EG 13 (1972), 53-119. Renoux, Athanase, 'Un manuscrit du Lectionnaire armenien de Jerusalem (cod. Jerus. arm.
121)', U Museon, 74 (1961), 361-85; 75 (1962), 385-97. -Ije Codex armenien Jerusalem 121 (Patrologia orientalis, 35/1 and 36/2; Turnhout, 1969,
1971).
Ribay, Bernard, 'Les Graduels en IIA', EG 12 (1988), 43-107. Riemann, Hugo, Notenschrift und Notendruck (Leipzig, 1896).
-Geschichte der Musiktheorie im IX.-XIX. Jahrhundert (2nd edn., Berlin, 1920); trans.
Raymond H. Haggh: History of Music Theory, Books I and II (Lincoln, Nebr., 1962). Righetti, Mario, Manuale di storia liturgica, 4 vols. (Milan, 1945-53; 3rd edn., 1964-9). RISM B/III = The Theory of Music from the Carolingian Era up to 1400: Descriptive
Catalogue of Manuscripts. RISM B/III/1 (1961) = Joseph Smits van Waesberghe, Pictcr Fischer, and Christian Maas,
Austria, Belgium, Switzerland, Denmark, France, Luxembourg, Netherlands. RISM B/III/2 (1968) = Pieter Fischer, Italy.
RISM B/III/3 (1987) = Michel Iluglo and Christian Meyer, Federal Republic of Germany. RISM B/V/l (1964) = Heinrich Husmann, Tropen- und Sequenzenhandschriften. Rivera Recio, J. F. (ed.), Estudios sobre la liturgia mozdrabe (Toledo, 1965). Robert, Leon, 'Les Chants du celebrant', RG 41 (1963), 113-26.
Robert, Michel, 'Le Graduel du Mont Saint-Michel', Millenaire rnonastique du Mont-Saint-
Michel, 1 (Paris, 1967), 379-82. Robertson, Anne Walters, 'Benedicamus Domino: The Unwritten Tradition',,JAMS 41
(1988), 1-62.
-The Service Books of the Royal Abbey of Saint-Denis: Images of Ritual and Music in the
Middle Ages (Oxford, 1991). " Roederer, Charlotte D., 'Can We Identify an Aquitanian Chant Style?', JAMS 27 (1974),
75-99.
Bibliography
lxxxv
Rojo, C, and Prado, G., Ell canto mozärabe: Estudio histörico-critico de su antigüedad y
estado actual (Barcelona, 1929). Rönnau, Klaus, Die Tropen zum Gloria in excelsis Deo, unter besonderer Berücksichtigung
des Repertoires der St. Martial-Ilandschnften (Wiesbaden, 1967). -'Regnum tuum solidum', in Stäblein Festschrift 1967, 195-205.
Rousseau, O., Histoire du mouvement liturgique: Esquisse historique depuis le debut du XIXe
siecle jusqu'au pontifical de Pie X (Paris, 1945). Ruini see CSM 25.
Russell, Carlton Thrasher, 'The Southern French Tonary in the Tenth and Eleventh
Centuries' (Diss., Princeton University, 1966; UMI 66-07182). Russell, Tilden A., 'A Poetic Key to a Pre-Guidonian Psalm and the Echemata', JAMS 34
(1981), 109-18.
Rutter, Philip, 'The Epiphany Trope Cycle in Paris, Bibliotheque Nationale, fonds latin
1240', in Leonardi and Menesto 1990, 313-24. Sablayrolles, Maur, 'A la recherche des manuscrits gregoriens espagnoles: Iter Hispanicum',
Sammelbände der Internationalen Musik-Gesellschaft, 13 (1911-12), 205-47, 401-32,
509-31.
Sachs, Klaus-Jürgen, 'Gerbertus cognomento musicus: Zur musikgeschichtlichen Stellung des Gerbert von Rheims', AfMw 29 (1972), 257-74.
Sackur, Ernst, Die Cluniacenser in ihrer kirchlichen und allgemeingeschichtlichen Wirksamkeit bis zur Mitte des elften Jahrhunderts, 2 vols. (Halle, 1892, 1894).
Sacred Music and Liturgy [De sacra musica et sacra liturgia]. Instruction of the Sacred Congregation of Rites September 3rd, 1958, trans. J. B. O'Connell (London, 1959).
Sallmann, Nicolas, 'Censorinus' "De die natali". Zwischen Rhetorik und Wissenschaft', Hermes, 111 (1983), 233-58.
Salmon, Pierre, L'Office divin: Histoire de la formation du breviaire (Lex orandi, 27; Paris, 1959); trans.: The Breviary through the Centuries (Collegeville, Minn., 1962).
-'Un breviaire-missel du xie siecle: Le manuscrit Vatican lat. 7018', in Melanges Eugene
Tisserant, vii (Studi e testi, 237; Rome, 1964), 327-43.
-L'Office divin au moyen age (Lex orandi, 43; Paris, 1967).
-l^es Manuscrits liturgiques latins de la Bibliotheque Vaticane (Studi e testi, 251, 253,
260, 267, and 270; Rome, 1968-72). Salvat, M., 'Un traite de musique du xinc siecle: Le De musica de Barthelemi L'Anglais', in
Buschinger and Crepin 1980, 345-60. Sanders Festschrift 1990 - Peter M. Lefferts and Brian Seirup (eds.), Studies in Medieval
Music: Festschrift for Ernest H. Sanders = Current Musicology, 45-7 (1990). Sandon, Nick, The Use of Salisbury: The Ordinary of the Mass (Lustleigh, 1984).
-The Use of Salisbury, 2: The Proper of the Mass in Advent (Lustleigh, 1986).
-The Use of Salisbury, 3: The Proper of the Mass from Septuagesima to Palm Sunday
(Lustleigh, 1991). Scharnagl, August, 'Offertorium', MGG.
Scheiwiler, A., Das Kloster St. Gallen: Die Geschichte eines Kulturzentrums (Einsiedeln and Cologne, 1938).
Schildbach, Martin, Das einstimmige Agnus Dei und seine handschriftliche Uberlieferung
vom 10. bis zum 16. Jahrhundert (Diss., Erlangen-Nürnberg University, 1967). Schlager, Karlheinz, Thematischer Katalog der ältesten AIleluia-Melodien aus Handschriften
lxxxvi
Bibliography
des 10. und 11. Jahrhunderts, ausgenommen das ambrosianische, alt-spanische und altrömische Repertoire (Erlanger Arbeiten zur Musikwissenschaft, 2; Munich, 1965).
-'Ein beneventanisches Alleluia und seine Prosula', in Stäblein Festschrift 1967, 217-25.
-'Anmerkungen zu den zweiten AUeluja-Versen', AfMw 24 (1967), 199-219.
-'The Microfilm Archive of Medieval Music Manuscripts at the Institut für Musikwissenschaft of Erlangen-Nürnberg University', JPMMS 2 (1970), 61-4.
-'Choraltextierung und Melodieverständnis im frühen und späten Mittelalter', in Cardine
Festschrift 1980, 314-37.
-'Trinitas, Deitas, Unitas—A Trope for the Sanctus of Mass', JPMMS 6 (1983), 8-14.
-'Tropen als Forschungsbereich der Musikwissenschaft—Vom Lebenslauf eines Melismas',
in Wersen 1983, 17-28.
-'Te Deum', MGG.
-'Alleluia. I', NG.
-see also Antiphonale Pataviense and MMMA 7-8.
Schmid, Hans (ed.), Musica et Scolica Enchiriadis una cum aliquibus tractatulis adiunctis (Bayerische Akademie der Wissenschaften, Veröffentlichungen der Musikhistorischen Kommission, 3; Munich, 1981).
Schmid, Toni, 'Smärre liturgiska bidrag VIII. Om Sankt Swithunusmässen i Sverige', Nordisk Tidsktiftför Bok- och Biblioteksväsen, 31 (Uppsala, 1944), 25-34.
-see also Graduate Arosiense.
Schmidt, Hans, Untersuchungen zu den Tractus des zweiten Tons aus dem Codex St. Gallen
359 (Diss., Bonn, 1955). -'Die Tractus des zweiten Tons in Gregorianischer und stadtrömischer Uberlieferung', in
Schmidt-Görg Festschrift 1957, 283-302.
-'Untersuchungen zu den Tractus des zweiten Tones', Kmjb 42 (1958), 1-25.
-'Gregorianik—Legende oder Wahrheit?', in Huschen Festschrift 1980, 400-11.
Schmidt, Hermann, Hebdomada sancta, 2 vols. (Rome, 1956, 1957).
Schmidt-Görg Festschrift 1957 = Dagmar Weise (ed.), Festschrift Joseph Schmidt-Görg
zum 60. Geburtstag (Bonn, 1957). Schmitz, Arnold, 'Ein schlesisches Cantional aus dem 15. Jahrhundert', Archiv für
Musikforschung, 1 (1936), 385-423. Schneider Festschrift 1955 = Walther Vetter (ed.), Festschrift Max Schneider zum achzigsten
Geburtstage (Leipzig, 1955). Schoenbaum, Camillo, 'Hymnologische Forschung in der Tschechoslowakei', Jahrbuch für
Liturgik und Hymnologie, 5 (1960), 157-65. Schrade, Leo, 'Die Darstellung der Töne an den Kapitellen der Abteikirche zu Cluny (Ein
Beitrag zum Symbolismus in mittelalterlicher Kunst)', Deutsche Vierteljahrsschrift für
Literatur- und Geisteswissenschaft, 7 (1929), 229-66 = Schrade 1967, 113-51.
-De scientia musicae studia atque orationes, ed. Ernst Lichtenhahn (Berne, 1967).
Schrade Gedenkschrift 1973 = Wulf Arlt, Ernst Lichtenhahn, and Hans Oesch (eds.),
Gattungen der Musik in Einzeldarstellungen: Gedenkschrift fürl^eo Schrade (Berne, 1973). Schubiger, Anselm, Die Sängerschule St. Gallens vom 8. bis 12. Jahrhundert: Ein Beitrag
zur Gesangsgeschichte des Mittelalters (Einsiedeln, 1858). Schueller, Herbert M., The Idea of Music: An Introduction to Musical Aesthetics in
Antiquity and the Middle Ages (Kalamazoo, Mich., 1988). Schuler, Ernst August, Die Musik der Osterfeiern, Osterspiele und Passionen des
Mittelalters (Kassel, 1951).
Bibliography
lxxxvii
Schuler, Manfred, 'Die Musik an den Höfen der Karolinger', AfMw 27 (1970), 23-40. Schwyzer, Eduard, Griechische Grammatik (Handbuch der Altertumswissenschaft, Il/l/l; Munich, 1939).
Seay, Albert, 'An Anonymous Treatise from St. Martial', Annales musicologiques, 5 (1957), 7-42.
Seebass, Tilman, 'Musik und Musikinstrumente in der Tonarillustration: Organologische,
stilistische und ikonographische Studie anhand der Handschrift Paris fonds latin 1118'
(Diss., Basel University, 1970). Servatius, Viveca, Cantus Sororurn: Musik- und liturgiegeschichtliche Studien zu den
Antiphonen des birgittinischen Eigenrepertoires, nebst 91 Transkriptionen (Uppsala, 1990). Sesini, Ugo, La notazione comasca nel cod. Ambrosiano E. 68 supra (Milan, 1932). Sevestre, Nicole, 'The Aquitanian Tropes of the Easter Introit—A Musical Analysis',
JPMMS 3 (1980), 26-39. Sidler, Hubert, 'Studien zu den alten Offertorien mit ihren Versen' (Diss., Fribourg, 1939). Sigl, Maximilian, Zur Geschichte des Ordinarium Missae in der deutschen Choralüberlieferung,
2 vols. (Regensburg, 1911). Silagi, Gabriel (ed.), Liturgische Tropen: Referate zweier Colloquien des Corpus Troporum
in München (1983) und Canterbury (1984) (Münchener Beiträge zur Mediävistik und
Renaissance-Forschung, 36; Munich, 1985). Skeat, W. W., The Book of Lindisfarne: The Holy Gospels in Anglo-Saxon, Northumbrian,
and Old Mercian Versions, 4 vols. (Cambridge, 1871-87). Smith, John A., 'The Ancient Synagogue, the Early Church and Singing', Music and
Letters, 65 (1984), 1-16. Smits van Waesberghe, Joseph, Muziekgeschiedenis der Middeleeuwen, 2 vols. (Tilburg,
1938-42).
-'Some Music Treatises and Their Interrelation: A School of Liege c. 1050-1200?', MD 3
(1949), 25-32, 95-118.
-'The Musical Notation of Guido of Arezzo', MD 5 (1951), 15-53.
-'Guido of Arezzo and Musical Improvisation', MD 5 (1951), 55-64.
-Cymbala (Bells in the Middle Ages): Edition of Texts and Introduction (Rome, 1951).
-'La Place exceptionelle de l'Ars Musica dans le developpement des sciences au siecle des
Carolingiens', RG 31 (1952), 81-104; trans. II. Schulze: 'Die besonderer Stellung der Ars
Musica im Zeitalter der Karolinger', in Smits van Waesberghe 1976, 48-70. -De musico-paedagogico et theoretico Guidone Aretino eiusque vita et moribus (Florence,
1953).
-'Neues über die Schola Cantorum zu Rom', in Congress Vienna 1954, 111-19.
-'Musikalische Beziehungen zwischen Aachen, Köln, Lüttich und Maastricht vom 11. bis
zum 13. Jahrhundert', in Beiträge zur Musikgeschichte der Stadt Aachen (Cologne, 1954),
5-13.
-A Textbook of Melody: A Course in Functional Melodic Analysis (Rome, 1955).
-Expositiones in Micrologum Guidonis Aretini (Amsterdam, 1957) [Liber argumentorum,
Liber specierum, Metrologus, Commentarius in Micrologum Guidonis Aretini]. -'Les Origines de la notation alphabetique au moyen-äge', Anuario musical, 12 (1957), 3-
14.
-'Over het onstaan van Sequens en Prosula en beider oorspronkelijke uitvoeringswijze',
Feestaflevering ter gelegenheid van de zestige verjaardag van Prof. Dr. K. Ph. Bernet
lxxxviii Bibhography
Kempers = Orgaan K. N. T. V., Officeel Maandblad van de Koninklijke Nederlandsche Toonkunstenaars Vereeniging (Amsterdam, 1957), 41-57; shorter version: 'Zur ursprünglichen Vortragsweise der Prosulen, Sequenzen und Organa', in Congress Cologne 1958, 251-4.
-'L'Etat actuel des recherches scientifiques dans le domaine du chant gregorien', in Actes
du troisieme congres international de musique sacree, Paris l'''-8"'e juillet 1957 (Paris, 1959), 206-17.
-'Die Imitation der Sequenztechnik in den Hosanna-Prosulen', in Feilerer Festschrift
1962, 485-90.
-'"De glorioso officio . . . dignitate apostolica . . ." (Amalar); zum Aufbau der Groß-
Alleluia in den päpstlichen Osterverspern', in Wellesz Festschrift 1966, 4873 = Smits van
Waesberghe 1976, 117-46.
-Musikerziehimg (Musikgeschichte in Bildern, 3, pt. 3; Leipzig, 1969).
-'Neue Kompositionen des Johannes von Metz (um 975), Hucbalds von St. Amand und
Sigeberts von Gembloux?', in Husmann Festschrift 1970, 285-303. -'Studien über das Lesen (pronunciare), das Zitieren und über die Herausgabe
lateinischer musiktheoretischer Traktate', AJMw 28 (1971), 155-200; 29 (1972), 64-86. -'Gedanken über den inneren Traditionsprozess in der Geschichte der Musik des
Mittelalters', in Kurt von Fischer Festschrift 1973, 7-30. -'Wie Wortwahl und Terminologie bei Guido von Arezzo entstanden und überliefert
wurden', AJMw 31 (1974), 73-86. -Dia-pason, de omnibus: Ausgewählte Aufsätze von Josef Smits van Waesberghe:
Festgabe zu seinem 75. Geburtstag (Buren, 1976). -see also CSM 1, 2, and 4; DMA, RISM B/III/1.
Smits van Waesberghe Festschrift = Pieter Fischer (ed.), Organicae voces: Festschrift J.
Smits van Waesberghe (Amsterdam, 1963). Snow, Robert, 'The Old-Roman Chant', in Apel 1958, 484-505.
Söhner, L., Die Geschichte der Begleitung des gregorianischen Chorals in Deutschland
vornehmlich im 18. Jahrhundert (Augsburg, 1931). Sowa, Heinrich, 'Zur Handschrift Clm 9921', AcM 5 (1933), 60-5, 107-20. -Quellen zur Transfonnation der Antiphonen: Tonar- und Phythmus-Studien (Kassel,
1935).
Sowulewska,   HalinA;  'Les  Relations de  proximite entre  les graduels polonais des
Premonstres et les manuscrits europeens', SMH 27 (1985), 123-9. Spanke, Hans, 'Das Moosburger Graduale', Zeitschrift für romanische Philologie, 50 (1930),
582-95.
-'St. Martial-Studien. Ein Beitrag zur frühromanischen Metrik', Zeitschrift für französische
Sprache und Literatur, 54 (1930), 282-317, 385-422; 56 (1931), 450-78.
-'Rhythmen- und Sequenzstudien', Studi medievali, ns, 4 (1931), 286-320.
-'Aus der Vorgeschichte und Frühgeschichte der Sequenz', Zeitschrift für deutsches
Altertum und deutsche Literatur, 71 (1934), 1-39.
-Beziehungen zwischen romanischer und mittellateinischer Lyrik mit besonderer Berücksichtigung der Metrik und Musik (Abhandlungen der Gesellschaft der Wissenschaften zu Göttingen, Philologisch-historische Klasse, 3rd ser., 18; Berlin, 1936).
-'Sequenz und Lai', Studi medievali, ns, 11 (1938), 12-68.
-'Die Kompositionskunst der Sequenzen Adams von St. Victor', Studi medievali, ns, 14
(1951), 1-30.
Bibliography
lxxxix
Spunar, Pavel, 'Das Troparium des Prager Dekans Vit (Prag Kapitelbibliothek, Cim 4)',
Scriptorium, 9 (1957), 50-62. Srawley, J. H., The Early History of the Liturgy (Cambridge, 1913; 2nd edn., 1947). Stäblein, Bruno, 'Zur Geschichte der Chorälen Pange lingua-Melodie', in Johner Festschrift
1950, 72-5.
-'Von der Sequenz zum Strophenlied. Eine neue Sequenzemelodie "archaischen" Stiles',
Mf7 (1954), 257-68, 511. -'Zur Frühgeschichte der Sequenz', AfHvlw 18 (1961), 1-33.
-'Die Unterlegung von Texten unter Melismen. Tropus, Sequenz und andere Formen',
in Congress New York 1961, 12-29. -'Das sogenannte aquitanische "Alleluia Dies sanetificatus" und seine Sequenz', in Hans
Albrecht Gedenkschnft 1962, 22-6. -'Die Schwanenklage. Zum Problem Lai—Planctus—Sequenz', in Feilerer Festschrift
1962, 491-502.
-'Der Tropus "Dies sanetificatus" zum Alleluia "Dies sanetificatus'", Studien zur
Musikwissenschaft, 25 (1962) (Festschrift für Erich Schenk), 504-15. -'Notkenana', AfMw 19-20 (1962-3), 84-99.
-'Zum Verständnis des "klassischen" Tropus', AcM 35 (1963), 84-95.
-'Modale Rhythmen im Saint-Martial-Repertoire?', in Blume Festschrift 1963, 340-62.
-'Zwei Textierungen des Alleluia Christus Resurgens in St. Emmeram Regensburg', in
Smits van Waesberghe Festschrift 1963, 157-67. -'Die Sequenzmelodie "Concordia" und ihr geschichtlicher Hintergrund', in Engel
Festschrift 1964, 364-92.
-'Zur Musik des Ludus de Antichristo', in Müller-Blattau Festschrift 1966, 312-27.
-'Der "altrömische" Choral in Oberitalien und im deutschen Süden', Mf 19 (1966), 3-9.
-'"Gregorius Presul", der Prolog zum römischen Antiphonale', in Vötterle Festschrift
1968, 537-61.
-'Theses 6qualistes et mensuralistes', Encyclopedic des musiques sacres, 2 (Paris, 1969),
80-98.
-'Zwei Melodien der altirischen Liturgie', in Feilerer Festschrift 1973, 590-7.
-Schriftbild der einstimmigen Musik (Musikgeschichte in Bildern, 3, pt. 4; Leipzig,
1975).
-'Pater noster-Tropen', in Haberl Festschrift 1977, 247-78.
-'Einiges Neue zum Thema "archaische Sequenz'", in Festschrift Dadelsen 1978, 352-83.
-arts. 'Agnus Dei', 'Alleluja', 'Antiphon', 'Antiphonar', 'Brevier', 'Cantatorium', 'Canticum',
'Choral', 'Communio', 'Credo', 'Deutschland. B. Mittelalter. I. Der römische Choral im Norden', 'Epistel, A. Katholisch', 'Evangelium, A. Katholisch', 'Exultet', 'Frühchristliche Musik', 'Gallikanische Liturgie', 'Gemeindegesang, A. Mittelalter', 'Gloria in excelsis Deo', 'Graduale (Gesang)', 'Gregor I.', 'Gregorianik', 'Hymnar', 'Hymnus, B. Der lateinischen Hymnus', 'Improperien', 'Introitus', 'Invitatorium', 'Kyriale', 'Kyrie', 'Litanei', 'Messe, A. Die lateinische Messe', 'Missale', 'Passion, A. Die einstimmige lateinische Passion', 'Pater noster', 'Präfation', 'Psalm, B', 'Saint-Martial', 'Sequenz (Gesang)', 'Tropus', 'Versus', MGG.
-see also MMMA.
Stäblein Festschrift 1967 = Martin Ruhnke (ed.), Festschrift Bruno Stäblein zum 70.
Geburtstag (Kassel, 1967). Stahl, William Harris, Martianus Capeila and the Seven Liberal Arts (New York, 1971).
xc
Bibliography
Steer, Georg, '"Carmina Burana" in Südtirol: Zur Herkunft des elm 4660', Zeitschrift für
deutsches Altertum und deutsche Literatur, 112 (1983), 1-37. Steglich, Rudolf, Die Quaestiones in musica; ein Choraltraktat des zentralen Mittelalters
und ihr mutmaßlicher Verfasser Rudolf von St. Trond (1070-1138) (Publikationen der
Internationalen Musikgesellschaft, Beihefte 11/10; Leipzig, 1911). Stein, Franz A., Das Moosburger Graduate (Diss., Freiburg im Br., 1956). Steinen, Wolfram von den, 'Die Anfänge der Sequenzendichtung', Zeitschrift für
schweizerische Kirchengeschichte, 40 (1946), 190-212, 241-68; 41 (1947), 19-48, 122-62.
-Notker der Dichter und seine geistige Welt, 2 vols. (Berne, 1948).
Steiner, Ruth, 'Some Questions about the Gregorian Offertories and their Verses', JAMS 19
(1966), 162-81.
-'The Prosulae of the MS Pans BN lat. UlS', JAMS 22 (1969), 367-93.
-'The Responsories and Prosa for St. Stephen's Day at Salisbury', MO 56 (1970), 162-82.
-'Some Melismas for Office Responsories', JAMS 26 (1973), 108-31.
-'The Gregorian Chant Melismas of Christmas Matins', in Essays in Honor of Charles
Warren Eox (Rochester, NY, 1979), 241-53. -'The Canticle of the Three Children as a Chant of the Roman Mass', Schweizer Jahrbuch
für Musikwissenschaft, ns, 2 (1982), 81-90.
-'Antiphons for the Benedicite at Lauds', JPMMS 7 (1984), 1-17.
-'The Music for a Cluny Office of St. Benedict', in Verdon and Dally 1984, 81-113.
-'Reconstructing the Repertory of Invitatory Tones and their Uses at Cluny in the Late
11th Century', in Huglo 1987, 175-82. -arts. 'Cantatorium', 'Compline', 'Cursus', 'Gregorian Chant', 'Hymn, II. Monophonic
Latin', 'Introit', 'Invitatory', 'Lord's Prayer', 'Prosula', 'Psalter', 'Te Deum. 1.2.', 'Trope',
NG.
-and Levy, Kenneth, 'Liturgy and Liturgical Books', NG.
Stenzl,  Jörg, Repertorium der liturgischen Musikhandschriften der Diözesen Sitten,
Lausanne und Genf, 1 (VGA, ns, 1; Fribourg, 1972). Stephan, Rudolf, 'Lied, Tropus und Tanz im Mittelalter', Zeitschrift für deutsches Altertum
und deutsche Literatur, 87 (1956), 147-62. Stevens, John, Words and Music in the Middle Ages: Song, Narrative, Dance and Drama,
1050-1350 (Cambridge, 1986). -'Medieval Drama, II. Liturgical Drama', NG.
Stevenson, Joseph (ed.), Chronicon man aste rii de Abingdon, 2 vols. (Rerum Britannicarum
medii aevi scriptores (Rolls Series), 2; London, 1858). Stotz, Peter, Sonderformen der sapphisehen Dichtung (Munich, 1982). Stratman, C. J., Bibliography of Medieval Drama (Berkeley, Cal., 1954; 2nd edn., 1972). Strecker, Karl, Introduction to Medieval Latin, trans, and rev. Robert B. Palmer (Dublin
and Zürich, 1957).
Strehl, Reinhard, 'Zum Zusammenhang von Tropus und Prosa "Ecce iam Christus'", Mf\l (1964), 269-71.
Strömberg, Bengt, Den pontifikala liturgin in Lund och Roskilde under medeltiden [The pontifical liturgy in Lund and Roskilde during the Middle Ages] (Studia theologica Lundensia, 9; Lund, 1955).
Strunk, Oliver, 'Intonations and Signatures of the Byzantine Modes', MQ 31 (1945), 339-55 = Strunk 1977, 19-36.
-Source Readings in Music History (New York, 1950).
Bibliography
xci
-'The Antiphons of the Oktoechos', JAMS 13 (1960), 50-67 = Strunk 1977, 165-90.
-'The Latin Antiphons for the Octave of the Epiphany', in Melanges Georges Ostrogorsky,
ii (Recueil de travaux de l'lnstitut d'Etudes byzantines, 8; Belgrade, 1964), 417-26 =
Strunk 1977, 208-19.
-'Tropus and Troparion', in Husmann Festschrift 1970, 305-11 = Strunk 1977, 268-76.
-'Die Gesänge der byzantinisch-griechischen Liturgie', in Feilerer 1972, 12847; trans.:
'The Chants of the Byzantine-Greek Liturgy' in Strunk 1977, 297-330. -Essays on Music in the Byzantine World (New York, 1977).
Strunk Festschrift 1968 = Harold S. Powers (ed.), Studies in Music History: Esaysfor Oliver
Strunk (Princeton, NJ, 1968). Stuart, Nicholas, 'Melodic "Corrections" in an Eleventh-Century Gradual (Paris, B. N.,
lat. 903)', JPMMS 2 (1979), 2-10. Stubbs, William, Memorial of Saint Dunstan (Rerum Britannicarum medii aevi scriptores
(Rolls Series), 63; London, 1874). Suchier, W., 'Die Entstehung des mittellateinischen und romanischen Verssystems',
Romanistisches Jahrbuch, 3 (1950), 529-63. Sunol, Gregoire Marie, Introduction ä la paleographie musicale gregorienne (Tournai,
1935); rev. and enlarged trans, of Gregori Maria Sunyol, Introducciö a la paleografia
musical gregoriana (Montserrat, 1925). -see also Antiphonale missarum and Liber vesperalis.
Symons, Thomas, Regularis Concordia: The Monastic Agreement of the Monks and Nuns of
the English Natio?t (London, 1953). Szendrei, Janka, 'Zur Notations- und Vortragsweise der Prosulen nach den ungarischen
Handschriften', in Magyar Könyvszemle (Budapest, 1972), 157-65. -A magyar közepkor hangjegyes forräsai [resume: 'Notierte Quellen des ungarischen
Mittelalters'] (Budapest, 1981). -Közepkori hangjegyiräsok Magyarorszdgon [resume: 'Mittelalterliche Choralnotationen
in Ungarn'] (Budapest, 1983). -'Beobachtungen an der Notation des Zisterzienser-Antiphonars Cod. 1799** in der
Österreichischen Nationalbibliothek', SMH 27 (1985), 273-90.
-'The Introduction of Staff Notation into Middle Europe', SMH 28 (1986), 303-19.
-'Die Geschichte der Graner Choralnotation', SMH 30 (1988), 5-234.
-'Choralnotationen in Mitteleuropa', SMH 30 (1988), 437-46.
-see also Missale notatum Strigoniense.
Szigeti, K., 'Denkmäler des gregorianischen Chorals aus dem ungarischen Mittelalter', SMH 4 (1963), 129-72.
Szöverffy, Josef, Die Annalen der lateinischen Hymnendichtung, 2 vols. (Berlin, 1964-5). -Repertorium hymnologicum novum (Berlin, 1983).
Tack, Franz, Der gregorianische Choral (Das Musikwerk,  18; trans. Everett Helm:
Gregorian Chant, Anthology of Music, 18; Cologne, 1960). Taft, Robert, The Liturgy of the Hours in East and West: The Origins of the Divine Office
and its Meaning for Today (Collegeville, Minn., 1985). Tarchnischvili, Michel, Le Grand Lectionnaire de l'eglise de Jerusalem (V-VIIT siecles)
(Corpus scriptorum Christianorum orientalium, 188-9, 204-5; Louvain, 1959, 1960). Terrizzi, Francesco (ed.), Missale antiquum S. Panormitanae ecclesiae (Rerum ecclesiast-
icarum documenta, series maior, 13; Rome, 1970). Thannabaur,  Peter Josef, Das einstimmige Sanctus der römischen Messe in der
xcii Bibliography
handschriftlichen Überlieferung des 11.-16. Jahrhunderts (Erlanger Arbeiten zur Musikwissenschaft, 1; Munich, 1962).
-'Anmerkung zur Verbreitung und Struktur der Hosanna-Tropen im deutschsprachigen
Raum und den Ostländern', in Stäblein Festschrift 1967, 250-9.
-'Sanctus', MGG.
Thibaut, Jean Baptiste, Origine byzantine de la notation neumatique de l'eglise latine (Paris, 1907).
-Monuments de la notation ekphonetique et neumatique de l'eglise latine (St Petersburg,
1912) .
-Monuments de la notation ekphonetique et hagiopolite de l'eglise grecque (St Petersburg,
1913) .
Thodberg, Christian, Der byzantinische Alleluianonzyklus (Monumenta musicae Byzantinae,
Subsidia, 8; Copenhagen, 1966). Thompson, Edward Maunde, An Introduction to Greek and Ixitin Palaeography (Oxford,
1912).
Tillyard, H. J. W., Handbook of the Middle Byzantine Notation (Monumenta musicae
Byzantinae, Subsidia, 1/1; Copenhagen, 1935). Tolhurst, J. B. L., The Monastic Breviary of Hyde Abbey, Winchester (mss. Rawlinson
Liturg. e. 1*, and Gough Liturg. 8, in the Bodleian Library, Oxford) (Henry Bradshaw
Society, 69; London, 1932; 70 (1933), 71 (1934), 76 (1938), 78 (1940), and 80 (1943) ). Traub, Andreas, 'Hucbald von Saint-Amand. De harmonica institutione', BzG 7 (1989), 3-
101.
Treitler, Leo, 'Musical Syntax in the Middle Ages: Background to an Aesthetic Problem',
Perspectives of New Music, 4 (1965-6), 75-85. -'The Aquitanian Repertories of Sacred Monody in the Eleventh and Twelfth Centuries'
(Diss., Princeton University, 1967; UMI 67-9613). -'On the Structure of the Alleluia Melisma: A Western Tendency in Western Chant', in
Strunk Festschrift 1968, 59-72. -'Homer and Gregory: The Transmission of Epic Poetry and Plainchant', MO 60 (1974),
333-72.
-contribution to symposium '"Peripherie" und "Zentrum"', in Congress Berlin 1974, 58-74.
-'"Centonatc" Chant: Übles Flickwerk or Epluribus umtsY', JMIS 28 (1975), 1-23.
-'Transmission and the Study of Music History', in Congress Berkeley 1977, 202-11.
-'Oral, Written, and Literate Process in the Transmission of Medieval Music', Speculum,
56 (1981), 471-91.
-'Observations on the Transmission of Some Aquitanian Tropes', Forum mustcologicum,
3 (1982), 11-60.
-'The Early History of Music Writing in the West', JAMS 35 (1982), 237-79.
-'From Ritual through Language to Music', Schweizer Jahrbuch für Musikwissenschaft,
ns, 2 (1982), 109-24/
-'Reading and Singing: On the Genesis of Occidental Music-Writing', EMH 4 (1984),
135-208.
-'Paleography and Semiotics', in Huglo 1987, 17-27.
Turco, Alberto, Tracce delta modalitä arcaica nella salmodia del Temporale e del
Sanctorale (Milan, 1972). -'Les Repertoires liturgiques latins en marche vers 1'octoechos. La psalmodie gregorienne
des fetes du Temporal et du Sanctoral', EG 18 (1979), 177-233.
Bibliography
xcni
Turco, Alberto, 'Introito "Tibi dixit" c "Alleluia. Dies sanctificatus"', in Cardinc Festschrift 1980, 257-67.
-'Melodie-tipo e timbri modali nell'Antiphonale romanům', Studi gregoriani, 3 (1987),
191-241.
Turner, Derek Howard, The Missal of the New Minster, Witichester (Le Havre, Bibliotheque Municipale, MS330) (Henry Bradshaw Society, 93; Leighton Buzzard, 1962).
Tyrer, John Walton, Historical Survey of Holy Week: Its Services and Ceremonial (Alcuin Club Collections, 29; London, 1932).
Udovich, JoAnn, 'The Magnificat Antiphons for the Ferial Office', JPMMS 3 (1980), 1-25.
-'Modality, Office Antiphons, and Psalmody: The Musical Authority of the Twelfth-Century Antiphonal from St.-Denis' (Diss., University of North Carolina at Chapel Hill, 1985; UM I 85-27331).
Uhlfelder, Myra L. (trans.), Joannes Scotus, Erigena: Periphyseon. On the Division of Nature, with summaries by Jean A. Potter (Library of Liberal Arts, 157; Indianapolis, Ind., 1976).
Ullmann, Peter, 'Bericht über die vergleichende Repertoire-Analyse der Breviere aus
Ungarn', SMH 27 (1985), 185-92. Underwood, Peter, 'Melodic Traditions in Medieval English Antiphoners', JPMMS 5
(1982), 1-12.
Undhagen, Carl-Gustav (ed.), Birger Gregerssons 'Birgitta-officium' (Svenska Fornskrifts-
ällskapet, ser. 2; Latinská skrifter, 6; Stockholm, 1960). Vagaggini, Cipriano, // senso teologico delta liturgia: Saggio di liturgia teologica generale, 2
vols. (Rome, 1957; 4th edn., 1965); trans.: Theological Dimensions of the Liturgy, 2 vols.
(Collegeville, Minn., 1959). Valous, Guy de, lxj Monachisme clunisien des origines au XV' siecle: Vie Interieure des
monastěres et organisation de Vordre, 2 vols. (Ligugé and Paris, 1935; 2nd edn., 1970). Van der Werf, Hendrik, The Límergence of Gregorian Chant, i, 2 vols. (Rochester, NY,
1983).
Van Deusen, Nancy M., 'An Historical and Stylistic Comparison of the Graduals of Gregorian and Old Roman Chant' (Diss., Indiana University, 1972; UMI 73-09783).
-Music at Nevers Cathedral—Principle Sources of Medieval Chant, 2 vols. (Henryville,
1980).
-'The Sequence Repertory at Nevers Cathedral', Forum musicologicum, 2 (1980), 44-59.
-'Style, Nationality and the Sequence in the Middle Ages', JPMMS 5 (1982), 44-55.
Vaničky, Jaroslav, 'Frater Domaslev (Domaslaus), der älteste bekannte Sequenzdichter
Böhmens', Jahrbuch für Liturgik und Hymnologie, 5 (1960), 118-22. Variae preces ex liturgia, tum hodierna tum antiqua (Solesmes, 1896). Väterlein see Graduate Pataviense.
Vecchi, Giuseppe, Uffici drammatici padovani (Florence, 1954).
-(ed.), Troparium sequentiarium Nonantulanum, Cod. Casanat. 1741 (Monumenta lyrica
medii aevi italica, I. Latina, i; Modena, 1955). Velimirovič, Miloš, arts. 'Echos', 'Russian and Slavonic church music', NG. Verdon, Timothy Gregory, and Dalley, John (eds.), Monasticism and the Arts (Syracuse,
NY, 1984).
Verheijen, Luc, La Regle de Saint Augustin, 2 vols. (Paris, 1967).
Vidakovič, A., 'I nuovi confini della scrittura neumatica musicale nell'Europa sud-est', Studien zur Musikwissenschaft, 24 (1960), 5-12.
xc1v
Bibliography
ViLLETARD, Henri, Office de Pierre de Corbeil (Office de la Circoncision) improprement appele 'Office des Fous' (Paris, 1907).
-Im Danse ecclesiastique ä la metropole de Sens (Paris, 1911).
-Odorannus de Sens et son uvre musicale (Paris, 1912).
-Office de St Savinien et de St Potentiell, premiers eveques de Sens: Catalogue somrnaire
des livres liturgiques de Fanden diocese de Sens (Paris, 1956). Vincent, Alexandre-Joseph-Hydulphe, 'Emploi des quarts de ton dans le chant gregorien,
constate dans l'Antiphonaire de Montpellier', Revue archeologique, 11 (1854), 262-72. Vivell, Coelestin, Initia tractatuum musices ex codicibus editorum (Graz, 1912). -Frutolfi Breviarium de musica et tonarius (Akademie der Wissenschaften in Wien,
Philosophisch-historische Klasse. Sitzungsberichte 188, Abhandlung 2; Vienna, 1919). Vives, J., Oracional visigötico (Monumenta Hispaniae sacra, Serie liturgica, 1; Barcelona,
1946).
Vogel, Cyrille, 'Les Echanges liturgiques entre Rome et les pays francs jusqu'ä l'epoque de Charlemagne', in chiese nei regni deU'Europa occidentale e i low rapporti con Roma fino all' 800 (Settimane di studi del Centro italiano di studi sull'alto medioevo, 7; Spoleto, 1960), 185-95.
-'La Reforme cultuelle sous Pepin le Bref et sous Charlemagne', in Erna Patzelt, Die
karolingische Renaissance (Graz, 1965), 171-242.
-'La Reforme liturgique sous Charlemagne', in Braunfels, ii (1965), 217-32.
-Introduction aux sources de l'histoire du culte chretien au moyen äge (Biblioteca degli
studi medievali, 1; Spoleto, 1966; 2nd edn., 1975; 3rd edn., 1981); trans, and rev. William
G. Storey and Niels Krogh Rasmussen, Medieval Liturgy: An Introduction to the Sources
(Washington, DC, 1986). -and Elze, Reinhard (eds.), Le Pontifical romano-germanique du dixieme siecle (Studi e
testi, 2267, 269; Rome, 1963, 1972). Vogüe, Adalbert de, La Regle du Maitre (Sources chretiennes, 105-7; Paris, 1964). -and Neufville, Jean (eds.), La Regle de Saint Benoit, 7 vols. (Sources chretiennes,
181-6; Paris, 1971-7).
Vollaerts, J. W. A., Rhythmic Proportions in Early Medieval Ecclesiastical Chant (Leiden, 1958, 2/1960).
Vötterle Festschrift 1968 - Richard Baum and Wolfgang Rehm (eds.), Musik und Verlag:
Karl Vötterle zum 65. Geburtstag (Kassel, 1968). Waddell, Chrysogonus, 'Monastic Liturgy: Prologue to the Cistercian Antiphonary', in
The Works of Bernard of Clairvaux: Treatises—/ (Cisterican Fathers Series, 1; Spencer,
Mass., 1970), 153-62.
-'The Origin and Early Evolution of the Cistercian Antiphonary: Reflections on Two
Cistercian Chant Reforms', in The Cistercian Spirit: A Symposium in Memory of Thomas Merton, ed. M. Basil Pennington (Cistercian Studies Series, 3; Spencer, Mass., 1970), 190-223.
-'The Early Cistercian Experience of Liturgy', in Rule and Life: An Interdisciplinary
Symposium, ed. M. Basil Pennington (Cistercian Studies Series, 12; Spencer, Mass., 1971), 77-116.
-'The Two Saint Malachy Offices from Clairvaux', in Bernard of Clairvaux: Studies
Presented to Dom Jean Leclercq (Cistercian Studies Series, 23; Washington, DC, 1973), 123-59.
-'Peter Abelard's Letter 10 and Cistercian Liturgical Reform', in Studies in Medieval
Bibliography
xcv
Cistercian History, 2, ed. J. R. Sommerfeldt (Cistercian Studies Series, 24; Kalamazoo, Mich., 1976), 75-86.
-'St Bernard and the Cistercian Office at the Abbey of the Paraclete', in The Chimaera of
his Age: Studies on Bernard of Clairvaux, ed. E. J. Elder and j. Sommerfeldt (Studies in Medieval Cistercian History, 5; Cistercian Studies Series, 63; Kalamazoo, Mich., 1980).
-'The Reform of the Liturgy from a Renaissance Perspective', in Benson, Constable, and
Lanham 1982, 88-109.
-The Twelfth-Century Cistercian Hymnal (Cistercian Liturgy Series, 1-2; Gethsemani
Abbey, 1984).
-'The Pre-Cistercian Background of Citeaux and the Cistercian Liturgy', in Goad and
Nail: Studies in Medieval Cistercian History, 10, ed. E. Rozanne Elder (Kalamazoo, Mich., 1985), 119.
-'Epithalamica: An Easter Sequence by Peter Abelard', MO 72 (1986), 239-71.
Waeltner, Ernst Ludwig, and Bernhard, Michael, Wortindex zu den echten Schriften
Guidos von Arezzo (Bayerische Akademie der Wissenschaften, Veröffentlichungen der
Musikhistorischen Kommission, 2; Munich, 1976). Wagener, H., Die Begleitung des gregorianischen Chorals im neunzehnten Jahrhundert
(Regensburg, 1964). Wagner, Peter, Einführung in die gregorianischen Melodien, 3 vols.
Wagner I = Ursprung und Entwicklung der liturgischen Gesangsformen (Fribourg, 1895; 2nd edn., 1901; 3rd edn., 1911); trans, of 2nd edn. by Agnes Orme and E. G. P. Wyatt: Introduction to the Gregorian Melodies, 1: Origin and Development of the Forms of Liturgical Chant up to the End of the Middle Ages (London, 1907).
Wagner II = Neumenkunde. Paläographie des liturgischen Gesanges (Leipzig, 1905; 2nd edn., 1912).
Wagner III = Gregorianische Formenlehre (Leipzig, 1921).
-'Un piccolo trattato sul canto ecclesiastico in un manoscritto del secolo x-xi', Rassegna
gregoriana, 3 (1904), 481-4. -'Germanisches und Romanisches im frühmittelalterlichen Kirchengesang', in Congress
Leipzig 1925, 21-34. -'Aus der Frühzeit des Liniensystems', AfMw 8 (1926), 259-76.
-'Der mozarabische Kirchengesang und seine Überlieferung', Gesammelte Aufsätze zur
Kulturgeschichte Spaniens, 1 (Spanische Forschungen der Görresgesellschaft, 1st ser.;
Münster, 1928), 102-21. -'Untersuchungen zu den Gesangstexten und zur responsorialen Psalmodie der altspanischen
Liturgie', Gesammelte Aufsätze zur Kulturgeschichte Spaniens, 2 (Spanische Forschungen
der Görresgesellschaft, 1st ser.; Münster, 1930), 67-113. -Das Graduate der St. Thomaskirche zu Leipzig (14. Jahrhunderts) (Publikationen älterer
Musik, 5-6; 1930, 1932).
-'Zur mittelalterlichen Tonartlehre', in Adler Festschrift 1930, 29-32.
-Die Gesänge der Jakobsliturgie zu Santiago de Compostela aus dem sogennanten Codex
Calixtinus (Fribourg, 1931). Wagner Festschrift = K. Weinmann (ed.), Festschrift für Peter Wagner (Leipzig, 1926). Walker, G. S. M., Sancti Columbani Opera (Dublin, 1957). Wallace, Wilfrid, The Life of St. Edmund of Canterbury (London, 1893). Wallace-Hadrill, J. M., The Frankish Church (Oxford, 1983).
xcv1
Bibliography
Warren, Frederick E., The Liturgy and Ritual of the Celtic Church (Oxford, 1881); rev.
Jane Stevenson (Woodbridge, 1987). -The Leofnc Missal (Oxford, 1883).
-The Antiphonary of Bangor (Henry Bradshaw Society, 4 and 10 (London, 1893, 1895).
Waszink, J., Studien sum Timaioskommentar des Calcidius (Leiden, 1964). Weakland, Rembert, 'Hucbald as Musician and Theorist', MO 42 (1956), 66-84.
-'The Beginnings of Troping', MO 44 (1958), 477-88.
-'The Compositions of Hucbald', EG 3 (1959), 155-62.
-'Milanese Rite, Chants of, .\Y7-;.
Weinmann, K., Das Konzil von Trient und die Kirchenmusik (Leipzig, 1919). Weinrich, Lorenz, 'Abelard', NG.
Weisbein, Nicolas, 'Le "Laudes crucis attollamus" de Maitre Hugues d'Orleans, dit le Primat', Revue du moyen age latin, 3 (1947), 5-26.
Weiss, Günther, '"Tropierte Introitustropen" im Repertoire der südfranzösischen Handschriften', Mf 17 (1964), 266-9.
-'Zum Problem der Gruppierung südfranzösischer Tropare', AflVlw 21 (1964), 163-71.
-'Zum "Ecce iam Christus'", Mf 18 (1965), 174-7.
-'Zur Rolle Italiens im frühen Tropenschaffen', in Stäblein Festschrift 1967, 287-92.
-see also MMMA 3.
Wellesz, Egon, Eastern Elements in Western Chant (Copenhagen, 1947).
-A History of Byzantine Music and Hymnography (Oxford, 1949; 2nd edn., 1961).
-'Gregory the Great's Letter on the Alleluia', Annates musicologiques, 2 (1954), 7-26.
Wellesz Festschrift 1966 = Jack Westrup (ed.), Essays Presented to Egon Wellesz (Oxford, 1966).
Westrup Festschrift 1975 = F. W. Sternfeld, Nigel Fortune, and Edward Ollcson (eds.),
Essays on Opera and English Music in Honour of Sir Jack Westrup (Oxford, 1975). Weyns, Norbert I., 'Le Missel premontre', Anale eta praemonstratensia, 43 (1967), 203-25. -Antiphonale missarum praemonstratense (Bibliotheca analectorum praemonstratensium,
12; Averbode, 1973). W hite, Alison, 'Boethius in the Medieval Quadnvium', in Gibson 1981, 162—205. Whitehill, W. M., Carro GarcTa, J., and Prado, G., Liber Sancti Jacobi: Codex
Calixtinus (Santiago de Compostela, 1944). Wiesli, Walter, Das Quilisma im Codex 359 der Stiftsbibliothek St. Gallen, erhellt durch das
Zeugnis der Codices Einsiedeln 121, Bamberg lit. 6, Laon 239 und Chart res 47: Eine
paläographisch-semiologische Studie (Bethlehem Immensee, 1966). Wilkinson, J., Egeria's Travels (London, 1971).
Willis, G.G., Essays in Early Roman Liturgy (Alcuin Club Collections, 46; London, 1964).
-Further Essays in Early Roman Liturgy (Alcuin Club Collections, 50; London, 1968).
Wilmart, Andre, 'Le Comes de Murbach', RB 30 (1913), 25-69.
-'Le Recueil des poemes et des prieres de Saint-Pierre Damien', RB 41 (1929), 342-57.
-Le Monachisme clunisien des origines au XV siede (Paris, 1935; 2nd edn., 1970).
Wilson, Henry Austin (ed.), The Pontifical of Magdalen College, ivith an Appendix of
Extracts from Other English Mss. of the Twelfth Century (Henry Bradshaw Society, 39;
London, 1910).
-(ed.), The Gregorian Sacramentary under Charles the Great (Henry Bradshaw Society,
49; London, 1915).
Bibliography
xcvii
Winkler, G., 'Über die Kathedralvesper in den verschiedenen Riten des Ostens und
Westens', Archiv für Lituigiewissenschaft, 16 (1974), 53-102. Winterfeld, Paul von, 'Rythmen- und Sequenzstudien, I: Die lateinische Eulaliasequenz
und ihre Sippe', Zeitschrift für deutsches Altertum und deutsche Literatur, 45 (1901),
133-47.
Wolf, Johannes, 'Ein anonymer Musiktraktat des elften bis zwölften Jahrhunderts',
Vierteljahrsschrift für Musikwissenschaft, 9 (1893), 186-234. Woods, Isobel, '"Our awin Scottis use": Chant Usage in Medieval Scotland', Journal of the
Royal Musical Association, 112 (1987), 21-37. Wordsworth, Christopher, Ceremonies and Processions of the Cathedral Church of
Salisbury (Cambridge, 1901).
-and Littlehales, Henry, The Old Service Books of the English Church (London, 1904).
Wormald, Francis, English Kalendars before A.D. J100, i (Henry Bradshaw Society, 72;
London, 1934).
-English  Benedictine Kalendars after A.D.  J100,  i. Abbotsbury-Durham (Henry
Bradshaw Society, 77; London, 1939). ——English Benedictine Kalendars after A.D. 1100, ii. Ely-St. Neots (Henry Bradshaw
Society, 81; London, 1946). Wright, Craig, Music and Ceremony at Notre Dame of Paris, 500-1500 (Cambridge, 1989). [Yardley], Anne D. Bagnall, 'Musical Practices in Medieval English Nunneries' (Diss.,
Columbia University, 1975; UMI 75-25648). Yardley, Anne Bagnall, 'The Marriage of Heaven and Earth: A Late Medieval Source of
the Consecratio virginutn, in Sanders Festschrift 1990, 305-24. Yearley, Janthia, 'A Bibliography of Planctus in Latin, Provengal, French, German,
English, Italian, Catalan and Galician-Portuguese from the Time of Bede to the Early
Fifteenth Century', JPMMS 4 (1981), 12-52. -'The Medieval Latin Planctus as a Genre' (Diss., York, 1983; British Theses D48872/
84).
Young, Karl, The Drama of the Medieval Church, 2 vols. (Oxford, 1933).
I
Plainchant in the Liturgy
I.l. INTRODUCTION
Frere 1906; Fortescue 1917; Miller 1959; Miller, 'Liturgy', NCE; Lechner and Eisenhofer 1961; Gelineau 1962; Righetti 1964; Martimort 1965; La Musique dans la liturgie, 1971; Cheslyn Jones et al. 1978; Steiner and Levy, 'Liturgy and Liturgical Books', NG\ Pfaff 1982; Martimort 1983; Ekenberg 1987; Gottesdienst der Kirche, lii, 1987.
Plainchant is liturgical music, music to be performed during the celebration of a divine service. The performance of the music is not, generally speaking, an end in itself but part of a religious ritual. Sometimes music assumes a prominent place within the performance of the ritual, sometimes a very minor one. Occasionally it has been cultivated with an exuberance and extravagance that seems to go beyond the needs of the liturgy; but this is not its normal role. Its function is to add solemnity to Christian worship. Liturgical texts which are sung, whether chanted on a monotone or to a highly melismatic melody, are more solemn, inspiring, and impressive, and a more worthy vehicle for human prayer and praise of God, than spoken words. (For a fuller discussion of its function see Gelineau 1962, Gottesdienst derKirche, iii, 1987, and La musique elans la liturgie, 1971; for the views of early medieval writers see Ekenberg, 1987.)
Practically the whole of the plainchant repertory is music sung with a text. This is another reason why the music cannot always be discussed as a thing in itself: one has to see whether, and how, it articulates the texts being sung. Moreover, the texts were not usually chosen for their musical potential, in the sense of being particularly easy to sing, or having characteristics which showed to best advantage in musical performance. The nature of the music chosen for a particular text is determined by a sense of what was proper, that is appropriate, for the liturgical occasion. This is shown by the fact that some texts, particularly verses from the Book of Psalms, were used again and again in the liturgy; and they were sung in different ways according to what was proper at the particular point in the liturgy where they were being sung. The liturgical context determined the type of music that was to be sung, just as it had determined the texts.
2 /. Plainchant in the Liturgy
Fox the most part, in this book I shall not discuss the theological and ritual bases of liturgical music. I describe the plainchant that has been sung, the state in which it has come down to us, and the way it was understood, not as ritual but as music. The book is written by a musician and not a liturgist, much less a theologian. But in order to understand the function of the music, why it was sung, and why it assumed the forms it did, the place of music in the liturgical services of the church must also be described. The first chapter of this book therefore gives a brief account of the liturgy of which plainchant is a part. In the sections which follow, I try to explain what a 'liturgy' is, why it is performed, and what it means; then comes an account of the yearly cycle within which are organized the services which go to make up the liturgy, followed by sections on the individual services themselves. This chapter is restricted to the Roman liturgy, that most widely used—in various forms—in Western Europe. Later in the book some mention is made of other Western liturgies and their chant, but, given the space available, discussion of these has inevitably had to be severely-restricted.
Much of what follows describes a sort of 'norm' of liturgical practice. But every church in every age has had its own idiosyncrasies of practice, and it is difficult to steer a middle way between over-simplification of the variety that has always existed and the confusion which may result if too many of the variations are described. During this century it has become increasingly convenient to refer to the practice laid down in Vatican books as a standard. The Liber usualis, an amalgam of several official Roman service-books, which contains a large proportion of the plainchant needed for mass and the services of the office, is probably the compendium most used for the study of plainchant today. But in this and other books we see only one among many possible liturgies, a twentieth-century one, moreover, not a medieval one. From time to time some attempt will be made to indicate a few of the differences between uses, in liturgical as in musical matters.
1.2. LITURGY AND WORSHIP
Cheslyn Jones et al. 1978, 1-29.
Liturgy may be defined as the communal forms of worship of the Christian church. Christians have from the earliest days of the church ('church' in the sense of 'all Christian people') gathered together to praise and adore God, to pray to him, thank him, ask his assistance, and especially to relive in a symbolic way the events of Christ's life on earth, most importantly the Last Supper which Christ shared with his disciples. The liturgy is the more or less formal, organized way in which this is done. So the study of liturgy (sometimes called 'liturgiology') broadly includes the history of these formal acts of worship and their theological and social (ethical, anthropological) significance, the texts used, the music, the ceremonial actions, the special clothes (vestments) worn by those who officiate at the ceremonies, and the nature of the
2. Liturgy and Worship 3
buildings in which worship take place (church architecture, internal decoration, and furniture).
Christians believe that God saved mankind from its natural state of sin through his son, Jesus Christ, who by his death paid the penalty for man's sins, and ensured eternal life for all mankind. In this, Christ was made a sacrifice for mankind. After Christ had risen to heaven, the Holy Spirit (the third Person, with God the Father and Christ the Son, of the Holy Trinity) came first to the Apostles, and then to the whole church, to continue the work of salvation ('saving'). By meeting together for worship Christians further the work of the Holy Spirit in a special way, since liturgical acts of worship bring them into particularly close contact with the events at the root of their religious beliefs. The acts of worship are an encounter with Christ which renews the Christian people, unites them with him, and makes them ready to receive God's grace (unmerited favour of being saved). As the Constitution on the Sacred Liturgy of the Second Vatican Council (1963) puts it:
Thus not only when things are read 'which were written for our instruction' (Rom. 15: 4), but also when the Church prays or sings or acts, the faith of those taking part is nourished, and their minds are raised to God so that they may offer him their spiritual homage and receive his grace more abundantly. (Para. I. I. 33.)
The most important ritual act within the liturgy is the Eucharist, Holy Communion, or Lord's Supper. The word 'eucharist' derives from the Greek word eukharistia, meaning thanksgiving. It refers to the thanks Christ gave at the Last Supper, and also to the thanks of the church for God's work of redemption. At the last meal with his disciples
Jesus took bread, and blessed it, and brake it, and gave it to the disciples, and said, Take, eat; this is my body. And he took the cup, and gave thanks, and gave it to them, saying, Drink ye all of it; For this is my blood of the new testament, which is shed for many for the remission of sins. (Matt. 26: 26-8.)
And in St Luke's account, Christ commands the disciples: 'This do in remembrance of me' (Luke 22: 19). It is through a re-enactment of this scene, a reliving of the Last Supper, that Christians come into closest union with God. In the eucharist the minister, or celebrant (that is the priest who is the chief actor in the ceremony, administering the bread and wine) is Christ's representative. But more than this:
Christ is always present ... in the Sacrifice of Mass not only in the person of his minister . . . but especially in the eucharistic species ... he is present when the Church prays and sings, for he has promised 'where two or three are gathered together in my name there am I in the midst of them' (Matt. 18: 20). (Constitution on the Sacred Liturgy, I. I. 7.)
The word 'eucharist' is also used to mean the sacred elements, or species, that is, the bread and wine, which Christians eat and drink at the culmination of the service. The word 'service' is used here in its usual liturgical sense of 'one self-contained communal act of worship', so that the 'service of Holy Communion' would mean the actual taking of the eucharist by the Christian congregation, together with all the prayers, readings,
4 /. Plainchant in the Liturgy
and music surrounding that act. The service of Compline, to take another example, is likewise made up of a series of prayers, readings, and chants; and so for other services.
The word 'liturgy', interestingly, derives from the Greek leitourgia, originally meaning some gratuitous act of service to the public, such as the giving of money to pay for a military project, or public entertainment. In the Christian West the word has been used in several different ways. It can refer to the whole complex of forms of public worship, in its broadest sense. But it is also used to refer to specific parts of that whole, such as the part for a particular day (for example, 'the Good Friday liturgy'), or the forms used in a particular part of the universal church (for example, 'the Sarum liturgy', meaning the forms of worship used at Salisbury cathedral). In the East, 'liturgy' is usually restricted to meaning the Holy Communion service alone, and some books on 'the liturgy' restrict themselves according to the same convention (for example, Srawley 1947). Partly because of this, I shall use another common term for the Holy Communion service, the 'mass'. As the Latin form of the service developed in the fourth century onwards, it acquired the words of dismissal 'Ite missa est' ('Go, you are dismissed'), and from this the title for the whole service is taken.
As well as the consuming of the bread and wine—the 'body' and 'blood' of Christ— praying, singing, and reading sacred texts are important parts of the liturgy. Mass contains all these things. The other services in the liturgy—usually known as the canonical hours, office hours, or simply the office—which do not include communion, are composed almost entirely of praying, singing, and reading. The reading of sacred texts is important for bringing to mind the history of God's work. The writings now collected in that part of the Bible known as the Old Testament, compiled by Jewish teachers before Christ's ministry on earth, are used beside those of Christian writers (the Gospels of the four evangelists, the letters of St Paul, etc.), for Christians believe that the earlier history of Israel contains signs and prefigurations of later events. Eventually the writings of Christian figures from times much later than the apostolic age (the first century or so of Christianity) also came to be read at liturgical services. These were often patristic writings, the writings of the Church Fathers of the second to sixth centuries (that is, roughly to the time of Pope Gregory I 'the Great', d. 604).
The sense of divine history to which I have just alluded affects the character and content of the liturgy according to the time of year. Thus, although the act of communion does not vary in essentials from one day to the next, the prayers, readings, and also the texts of the chants vary in accordance with a cycle of commemorations of the events of the life of Christ and his most important followers. At the time of year when Christ was born, the texts, and some of the ceremonial actions, refer to Christ's birth; similarly for Christ's death, resurrection, and ascension into heaven, and the coming of the Holy Spirit. The whole year is full of special commemorations of this sort. For example, 10 August is traditionally the day in the year when St Laurence, a deacon of the church at Rome, was martyred by being roasted on a grid during the persecutions of Valerian in 258. (King Philip II of Spain, in thanks for a military victory won at Saint-Quentin on 10 August 1557, built
2. Liturgy and Worship 5
his great palace, monastery, church, and college of El Escorial, supposedly in the shape of this (inauthentic) instrument of execution, a striking and macabre example of the strength of Christian historical symbolism.) In the Middle Ages passages relating his life and death would have been read during Matins, some of the prayers at mass would invoke his intercession, and some of the chants would also refer to him. Thus at mass, the alleluia would probably use the verse Levita Laurentius: 'The Levite Laurence has wrought a pious work, who by the sign of the cross enlightened the blind, and distributed to the poor the riches of the church.' This short text would also be used for some chants of the office (sung to different music, as befitting the particular liturgical function of the chant in question). Not all the texts would have been special to St Laurence, since he was but one of very many martyrs, though undoubtedly one of the more important in medieval eyes. The introit at mass uses a text suitable for almost any saint: Confessio et pulchritudo (from Ps. 96): 'Glory and worship are before him: power and honour are in his sanctuary', and indeed this introit was used for other saints as well.
The annual cycle of commemorations and symbolic re-enactments is of such importance that the next section of this book is devoted to it.
The brief indication of the liturgy's significance sketched above is only one way of looking at the matter, albeit one suggested by modern scholarship (Crichton in Cheslyn Jones et al. 1978, Bouyer 1956 and 1962, Vagaggini 1957) and the official teaching of the Church. The liturgy obviously played different parts in the lives of different Christians: priest and people, monk and clerk. It meant different things to different persons at different times and places. Communion, the actual consuming of the bread and wine, was strikingly rare in the Middle Ages, and later times as well—to the extent that it might take place only on Easter Day. On other occasions, although the form of the mass would have been followed, and all the texts and music heard, the sacred elements would merely have been displayed to those present, the only communicant being the priest. This sort of variation in practice obviously has consequences for the music needed during the ceremony (in this case, see Atkinson 1977). Just as striking as the differences in liturgical practice are the differences in interpretation of the liturgy. In the Middle Ages in particular, numerous allegorical explanations of the significance of liturgical actions were composed—often of what were in origin rather simple, functional actions. (See the account of allegorical interpretations of the mass, and a rapid sketch of the changes in the character of the ceremony, in Jungmann 1962, Missarum, Pt. I, chs. 9 and 11-14, and Pt. II, ch. 2.)
For example, the influential Amalarius of Metz (d. c.853) saw symbolic significance in every person, text, action, vestment, and property found in the liturgy. I quote here from Jungmann's presentation of Amalarius's summary of the contents of his Expositio or Eclogae (813-14) of the mass (for an extensive presentation of the much longer discussion of the mass by Amalarius in his De ecclesiasticis officiis, see Hardison 1965, Essay II; note, however, the reservations expressed by McKitterick 1977 148 ff.: Amalarius may have occupied a somewhat extreme position). The parts
6 /. Plainchant in the Liturgy
of the mass mentioned in this extract represent Christ's life 'from the first coming of the Lord to the time when he hastened to Jerusalem to suffer':
The Introit represents the choir of the Prophets (who announced the advent of Christ just as the singers announce the advent of the bishop) . . . , the Kvrie eleison represents the Prophets at the time of Christ's coming, Zachary and his son John among them; the Gloria in excelsis Deo indicates the throng of angels who proclaimed to the shepherds the joyous tidings of our Lord's birth (and indeed in this manner, that first one spoke and the others joined in, just as in the Mass the bishop intones and the whole church joins in); the first collect represents what our Lord did in His twelfth year . . . ; the Epistle represents the preaching of John, the Rcsponsorium the readiness of the Apostles when our Lord called them and they followed Him; the Alleluia their joy of heart when they heard his promises or saw the miracles He wrought . . . , the Gospel his preaching. (Jungmann 1962, Missarum, Pt. I, 118.)
One need not necessarily agree with Hardison's interpretation of Amalarius—he sees him as a striking example of the mentality which cultivated liturgical drama—to realize that the possible relevance of the ideas in Amalarius' writings to liturgical texts (tropes, and dramatic ceremonies) has to be considered. (For another example of the possible connection between theological/allegorical ideas and liturgical music, see Schlager 1983, Trinitas'.)
It is not only the theological or allegorical significance of the liturgy that has consequences for musical matters, but also a great deal of its practical detail. I have already alluded to the matter of frequency of communion, more precisely the number of communicants at mass. While this is of crucial importance in recent liturgical history, it has practical consequences for the singing of chants during communion: how many are required? To take another example, the introit chant, the first of mass, has at one time, it seems, consisted of the singing of most or all of a psalm, with an antiphon, then been reduced to a single psalm verse plus antiphon, then on high feasts been supplemented by numerous trope verses (making, in effect, a chant equivalent in length to one of the longer psalms). The introit is traditionally a chant sung during the entrance of the officiant (the priest who administers communion) and his assistants. Are we to imagine entrance routes of varying length, varying numbers of persons in the procession, as calling forth these varying forms of introit chant? Some answers to questions such as these will be found in the course of this book.
For many it is not easy today to gain a sense of the power of the liturgy, its slow rhythm from day to day, week to week, year to year, its seasons of grief and penitence, hope and joy. Those who attend church regularly are in a better position than most, for, albeit on a reduced scale, and interrupted by the numerous activities of the rest of the week, they may experience the sense of unity with the whole Christian church in communion, and follow in Bible readings, prayers, and the texts of musical compositions the annual cycle of Christ's birth, passion, and resurrection, supported by the witness of the Old Testament, and uplifted by the ancient lyric texts of the psalms. Yet, for those with imagination, even the mere perusal of liturgical books can
3. The Church Year
7
awaken a sense of the immensity of the material involved, the majesty of the slow procession through the psalter, the steady flow of the great responsories, and again, the colourful juxtaposition of diverse forms and styles in the chants of mass. How much more strongly, then, must those whose lives were (and are) a continuous celebration of the liturgy have experienced these things! The reader is urged, therefore, to try to keep the musical material discussed in these pages in a liturgical perspective. Beyond that, a day spent attending the services in a monastic community teaches one more than many books. For while a chant may be discussed and dissected here as an object of study in itself, it must not be forgotten that it was composed in the creation of a complete way of life, the performance of the 'opus Dei', the work of God.
1.3. THE CHURCH YEAR
Frere 1930; McArthur 1953; Pascher 1963; Andrew Hughes 1982; Vogel 1986, 304-14.
For special parts of the year, Hermann Schmidt 1956-7 and Jungmann Festschrift 1959 (Easter), Botte 1932 (Christmas and Epiphany), Willis 1964 (Ember Days).
For information on the saints found in Western kalendars and service-books, there are numerous volumes, both great (the series Acta sanctorum and Biblioteca sanctorum) and small (Farmer 1978). The lists in Grotefend 1898 and Dalton and Doble 1941 are also very useful.
For different days of worship in the Christian year, different prayers and lessons are intoned and different chants are sung. At the beginning of a new year (usually reckoned to be the start of the Advent season, which looks forward to Christ's birth), the whole mighty cycle begins again, repeating the prayers, lessons, and chants of the previous year. In order to appreciate the rhythm and variety of the huge body of liturgical material which has come down to us, some sense of the way the church year is organized is essential.
Although Christian worship is organized in a year's cycle, weekly units and cycles are also most important. Sunday, the day in the week when Christ rose from the dead, was of such significance that much of the liturgy revolves around Sunday services. To one seeking absolute regularity in adherence to the dates of a 365- (or 366-)day year, the Christian year is therefore an odd arrangement of fixed and (mostly) variable dates. It has its own compelling logic, however, attuned to the natural rhythm of the seasons, the waxing and waning of the moon, the summer and winter solstices, the week and the day.
Many individual elements within the yearly cycle are organized on an annual basis, in the sense that the special liturgies of the days in question come round only once in the year. Even these, however, are not always fixed to an unvarying calendar date. Some of the days are given to the commemoration of events in the life and ministry of Christ and his followers, and also to remembering the witness of later Christians, particularly those martyred during the persecution of the early Church and those who
8 /. Plainchant in the Liturgy
carried Christianity into new lands. Holy days which commemorate saints are usually fixed to particular dates in the year, mostly that when the saint passed to his or her heavenly reward, or occasionally when the saint's relics were translated (solemnly removed) from one place to another. An important, and generally more ancient part of the yearly cycle, however, is governed by the date of Easter, when Christ rose from the dead. This date varies from year to year, so that those parts of the Christian year which prepare for, and follow on, from Easter are said to be 'movable'. In fact, the fixed-date part is governed by the solar cycle, the movable part by the lunar cycle: two different temporal cycles have to be reconciled.
Even when reference is being made to Christmas and Easter, however, the weekly unit may still be an organizing factor. Thus services may take place on 'the Sunday after', such and such a date or day. Since the calendar dates of Sundays are not the same from one year to the next, any day in the Christian year which is fixed on a particular date—such as Christmas Day, 25 December, or a saint's day—will fall sometimes on a Sunday and sometimes not, and some recognition of this will often be found in the liturgy. Another basic layer of liturgical material is formed by the weekly recitation of the office, where a seven-day supply of material is simply repeated week after week.
Hence many items in the liturgy are said to be 'proper', used only on one day, or certain days of the year, or 'ordinary', used on every or nearly every day. There is in fact a very large number of intermediate gradations between completely proper and totally ordinary, and it is impossible to give here more than a vague idea of the main outlines of the picture. (Andrew Hughes 1982 is a strikingly comprehensive and serious attempt to account for the patterns of repetitions throughout the year in later medieval books.) What should be borne in mind is the great variety of the possible arrangements. Formularies may be proper not simply to single feast-days, but also to seasons of the year, or to one day in the week (that is, not sung on the other six). Much office material is subject to complex variation of this sort: thus there is a different hymn for each day of the week for Lauds, but these may be displaced by a hymn proper to the season at certain times of the year. Particularly during Lent, the 'triduum' (the last three days before Easter Sunday), and the Easter season generally (Easter Sunday to Whit Sunday), many 'ordinary' items will disappear or change place. Music may also contribute to the 'properness' of a piece. While the text may remain the same for a certain period of the year, or even for the whole of it, the musical setting may vary from occasion to occasion. The best-known instances of this are the various melodies for some of the ordinary texts of mass, Kyrie, Gloria, Credo, Sanctus, and Agnus Dei. Some of these, however, were associated with particular occasions—Christmas, Easter, feasts of the Blessed Virgin Mary—and were to that extent proper. The usual hymn for Prime, lam lucis ortu sidere, changes its melody according to season.
Rather few events in Christ's life are commemorated on unchanging dates: his birth (Christmas Day) has traditionally been celebrated on 25 December in the Roman church since the fourth century. (It was the pagan festival of the winter solstice when
J. The Church Year
9
first adopted by the church.) Practically all other fixed days are those of saints, the Blessed Virgin Mary, and so on. The movable period dependent upon Easter contains such days as Ash Wednesday (the start of the Lenten fasting period), Palm Sunday (when Christ entered Jerusalem, the start of the week leading to his crucifixion and resurrection), Ascension Day (when he was taken up into heaven), and Whit Sunday (when the Holy Spirit came upon the Apostles). Since every Sunday in the church year had its own special prayers, lessons, and chants, the Sundays were usually reckoned in relation to the great feast-days, either those with a fixed date or those which were movable. Different places, periods, and liturgical books had their own ways of setting out the material required by this complex arrangement of holy days.
For a correct performance of the liturgy it was essential to know the correct calendar date, the days when Sunday occurred, and, most importantly, when Easter fell. It became customary in the Middle Ages to compile tables, sometimes known as 'compotus' or 'computus' tables, which set out this information in ready-reckoner form. These usually contain code letters and numbers for days and years which assist in calculating the repetition of calendric cycles over a span of decades or even centuries. Computus tables are sometimes found bound with liturgical books, and many liturgical books are provided with a kalendar which sets out the yearly cycle of saints' days and other feasts. (The spelling 'kalendar' is here employed for the actual document; 'calendar' is reserved for the general notion of a system of fixed dates.)
The calculation of the relationship between the seven-day week and 365- (or 366-)day year and the date of Easter necessitates a knowledge of the solar cycle (yearly alternation of summer, with longer days, and winter, with longer nights) and the lunar cycle (waxing and waning of the moon). The 365-day year, with 366 days every fourth year, is an arrangement promoted by Julius Caesar, hence the term 'Julian calendar'. The days of the month were reckoned in Roman use from the Kalendae, the first day (originally the first day of the new moon), backwards. The time of the full moon was the Idus, and nine days before that came the time of the quarter moon, the Nonae. To gain his 'leap year', Julius Caesar doubled the sixth before the Kalends of March, that is, the sixth day before the first day of March, 24 February. Such a year was also known, therefore, as 'bissextile'.
The lunar cycle coincides with the solar cycle every nineteen (solar) years: that is, nineteen years after a given new moon another new moon will appear, during which period 235 lunations will take place. Each of the nineteen years in the cycle (known as the 'Metonic' cycle) was given—in Greek antiquity and therefore in Roman and medieval times—a 'golden number' (Archer 1941, 4). To find the golden number of any year, the number of the year is divided by 19 and 1 is added to the remainder. Thus for 1990 the golden number is (1990/19),. + 1 = 14 + 1 = 15.
The Christians also assigned to each day of the year an alphabetical letter from A to G, in order from 1 January (A) onward. Saint's days, always fixed to one date in the year, thus always corresponded to one particular letter. The letters were known as 'dominical' or 'Sunday' letters, for the relation between the occurrence of Sundays and the date of the day could be deduced therewith.
10 /. Plainchant in the Liturgy
The predominant medieval system of finding the date of Easter was established at the Council of Nicaea in 325. The date depends first of all on the vernal equinox, then as now 21 March. Next one has to know which lunar cycle (28 or 29 days, beginning with a new moon) has its fourteenth day on or after 21 March. Easter falls on the first Sunday after that fourteenth day. Thus in 1990 the relevant new moon fell on Monday, 26 March, fourteen days after being Monday 9 April, Easter Sunday thus falling on 15 April. The earliest possible date for Easter Sunday is 22 March and the latest 25 April.
It would in theory have been possible for the Christians to have adopted a fixed date for Easter Day, that of the year of Christ's resurrection. Its occurrence on Sunday (the day after the Jewish Sabbath), however, and its relationship to the Jewish Passover (determined in a manner adapted by the Christians for Easter) made of it a movable feast. The correct calculation of Easter was one of the most important accomplishments of a medieval priest. It could symbolize the difference between true and deviant faith, as in the disputation between representatives of Roman and Celtic-practice at the Synod of Whitby in 664 (Bede, trans. Sherley-Price, 186 ff.: the British used an 84-year cycle, and celebrated Easter on the fourteenth day if that was a Sunday). It was a vital part of the Carolingian ecclesiastical reforms, as witness Hrabanus Maurus' treatise De computo, of 820, revised from previous versions (possibly of Irish origin).
Not only Easter itself but also the previous and subsequent weeks were associated in the movable portion of the church year. Before Easter came the penitential season of Lent, notionally a fast of forty days, corresponding to periods of fasting by Moses, Elijah, and Christ himself. The actual number of days varied: in fourth-century Jerusalem there were eight weeks of five-day fasts (Saturday and Sunday were exempt); in the Eastern churches in the Middle Ages seven such weeks, plus Holy Saturday (that is thirty-six days); in the West six weeks, excluding only Sundays (thirty-six days). From the seventh century in the West, Ash Wednesday and the subsequent three days were added, making forty. The forty days were known also as Quadragesima, and that term was also given to the first Sunday of the period. By analogy, the previous Sundays became known as Quinquagesima (which is indeed fifty days before Easter), Sexagesima, and Septuagesima. The latter is a term first encountered in the Gelasian Sacramentary. Although not the start of the Lenten fast proper, it was marked in the Middle Ages by a change to purple vestments and the exclusion of the word 'alleluia' from all services. The terms Septuagesima, etc., were suppressed in Roman usage in 1969.
After Easter Sunday there are seven weeks, fifty days, until the Sunday of Pentecost. The important stage on the way is Ascension Day, forty days after Easter. Pentecost was, like other feasts in the Christian year, originally a Jewish holy day. The term 'Pentecost' (literally 'fifty days') at first referred to the whole period, but was soon restricted to the day itself.
The length of the time preceding Septuagesima, that is the period after Epiphany, must therefore be adaptable in order to allow for the varying date of Easter. If Easter
J. The Church Year
11
is early, then there may be no more than one Sunday after Epiphany. If it is late, then up to six will be celebrated. Similarly, the fixed start of Advent will bring to an end the series of Sundays after Pentecost, whose number varies according to whether Easter is early or late.
From about the ninth century the first Sunday after Pentecost was often given over to a commemoration of the Holy Trinity. The long series of Sundays through the summer season of the year might therefore be reckoned as Sundays after Pentecost or after Trinity. (From the fourteenth century the Thursday after Trinity was usually celebrated as the Feast of Corpus Christi, officially instituted by Urban IV in 1264.) Interestingly, the enumeration of these Sundays is generally different according to whether a mass or an office book is involved. In missals and graduals it was customary to set out a cycle of Sundays after Pentecost (or Trinity), numbered 1 to 23 (24, 25, or however many were deemed to be necessary by the compiler of the book; the later Sundays might be omitted). Office books, however, usually tied the Sundays to specific dates of the calendar, and it was the early Sundays that were omitted if necessary.
Before the fixed date of Christmas comes the Advent season, which begins with the Sunday nearest St Andrew's Day, 30 November. At least four Sundays therefore precede Christmas. After Christmas, Epiphany on 6 January is celebrated as the day when Christ was 'manifested' to the Gentiles, that is, to the three Magi. It had additional significance as the day of Christ's baptism (that was its primary significance in the East, whence the feast originally came), and of the miracle Christ performed at the wedding feast at Cana.
At four times in the year special days of fasting and abstinence are observed, the so-called Ember Days (so called because of the ashes sometimes marked on the head at this time). Although at first their date was somewhat variable, and only three of them were observed, by the sixth or seventh century there were four, known often as the 'quatuor tempora'. They fell on the Wednesday, Friday, and Saturday of four weeks, after St Lucy (13 December; or, alternatively, during the third week of Advent), Quadragesima Sunday (that is, during the first week of Lent), Whit Sunday, and the Exaltation of the Holy Cross (14 September). The liturgy during these days is somewhat similar to those of Lent.
There are several other days of special observance, which require separate description in later chapters. Among these are the Rogation Days, the Monday, Tuesday, and Wednesday before Ascension Day (which is on Thursday). The preceding Sunday, actually the fifth after Easter, is thus known as Rogation Sunday. On the Rogation Days special prayers of intercession, including the 'greater Litany', are chanted (cf. Latin 'rogare': to ask), and processions are held. Processions will be discussed later, but it may be stated here that Palm Sunday is naturally distinguished by special processional rites, as befits the day commemorating Christ's triumphal entry into Jerusalem. Later in that week, Holy Week, the liturgies of Maundy Thursday, Good Friday, and Holy Saturday are unique, as well as very ancient in many respects. Some of the special ceremonies in these days concern the blessing and
12 /. Plainchant in the Liturgy
bringing into use of special symbolic objects: candles on the feast of the Purification of the Blessed Virgin Mary (2 February), penitential ashes on Ash Wednesday, palm branches on Palm Sunday, the New Fire, paschal candle, and font on Easter Eve.
When no special day occurred, the liturgy of the day reverted to a ferial form. (The Latin word 'feria' actually means 'feast'; in Latin Christian use it came to mean a day of worship; it was inevitably most often used when no other obvious designation could be given to a day, and therefore came to mean 'non-feast day'. Weekdays from Monday to Friday are 'feria ii' to 'feria vi'; Saturday is usually 'Sabbato' rather than 'feria vii', and Sunday, though actually 'feria i', is always called 'Dominica'.) In the celebration of mass, this meant repeating the formularies of the previous Sunday or feast-day, with some reduction of solemnity; in the office a regular weekly or ferial set of services, independent of any feast, was resumed.
Some especially solemn parts of the year, however, had proper formularies (prayers, lessons, chants, etc.) for each day of the week. These are the days of Lent, and the week after Easter. It should be said, however, that as time went on there was a tendency to provide more proper material (that is, prayers, chants, and particularly lessons special to a particular day); it is not uncommon to find lessons and prayers assigned to particular days in Advent (Wednesday, Friday, Saturday outside the Ember Week) and elsewhere in late medieval books.
Tables 1.3.1. and 2 give lists of the two sections of the church year described briefly so far.
The two components of the church year sketched briefly above—the fixed part associated with Christmas and the movable part associated with Easter—were joined by a large number of commemorative days for saints and other holy persons. The decision as to which saints were to be remembered by special services was to a certain extent a local one, but all liturgical books have services for days of considerable antiquity, when the heroes of the early church are honoured. Then there will be remembered those who brought Christianity to the country, diocese, or area relevant
Table 1.3.1 The Church Year: the fixed part dependent upon Christmas
1st Sunday in Advent—as near St Andrew (30 Nov.) as possible 2nd Sunday in .Advent
Ember Days: Wednesday, Friday, Saturday—in the week after St Lucy (13 Dec.) 3rd Sunday in Advent 4th Sunday in Advent
Vigil of Christmas Day, or Christmas Eve (24 Dec.) Nativity, or Christmas Day (25 Dec.)
Octave of the Nativity, or Feast of the Circumcision, or New Year's Day (1 Jan.) Epiphany (6 Jan.)
Sundays after Epiphany (up to 6 as required)
Ember Days: Wednesday, Friday, Saturday—in the week after Exaltation of the Holy Cross (14 Sept.)
3. The Church Year Table 1.3.2. The Church year: the movable part dependent on Taster
13
Septuagesima Sunday (9th before Easter) Sexagesima Sunday (8th before Easter) Quinquagesima Sunday (7th before Easter) Ash Wednesday Ember Friday Ember Saturday
Quadragesima Sunday (1st in Lent, 6th before Easter)
2nd Sunday in Lent (5th before Easter)
3rd Sunday in Lent (4th before Easter)
4th Sunday in Lent (3rd before Easter)
Passion Sunday (5th Sunday in Lent, 2nd before Easter)
Palm Sunday
Maundy Thursday
Good Friday
Holy Saturday
Easter Sunday
1st Sunday after Easter (Low Sunday)
2nd Sunday after Easter
3rd Sunday after Easter
4th Sundav after Easter
5th Sunday after Easter
Ascension Day (Thursday)
Sunday after Ascension
Pentecost Sunday, or Whit Sunday
Ember Days: Wednesday, Friday, Saturday
Trinity Sunday
Sundays after Pentecost or after Trinity (up to 25 as required)
to where the book is used. Finally, there may be commemorations peculiar to the particular church for which the book is compiled.
Frere (1930) is an account of how the cycle of sanctoral developed, and it is instructive to look through the kalendars edited by Wormald (1934, 1939, 1946), for example, to see the typical intermingling of universal and local commemorations. As an example I give in Table 1.3.3. in an abbreviated form the entries of the month of October from the kalendar in an eleventh-century English manuscript, the so-called 'Portiforium of St Wulstan', Cambridge, Corpus Christi College 391. (Most of the text of this manuscript is edited in Anselm Hughes 1958-60, the kalendar in abbreviated form in Dewick and Frere 1921 and Wormald 1934. A facsimile of the October page is given by Dewick and Frere, pi. I.) To help the reader I have given the dates 1 to 31 in the first column. These do not appear in the original manuscript. In the second column (the original first column) appears a series of roman numerals which indicate when the full moon will appear in particular years of the nineteen-year
14 /. Plainchant in the Liturgy
cycle. The golden number for 1 October is 16: in the sixteenth year of the cycle the full moon will fall on that day. (This information is particularly important, of course, for the period when Easter may be celebrated.) The next two columns of letters in the original are not of liturgical significance and are omitted here. Then come the dominical letters for each date. After that is given the date not in modern but in Roman fashion, beginning with a large KL for the Kalendae, then the days before NON (Nonae), before ID (Idus), and before the Kalendae at the beginning of the next month.
Many of the saints' names inscribed for October on this kalendar might be found on almost any Latin Christian kalendar. Others are more local. I have separated these out in the lists given after the kalendar in Table 1.3.3.
Some of these feasts are of especial importance. SS Simon and Jude have a Vigil marked: that is, the previous day will also be marked by liturgical material in their honour. (Some saints have the further distinction of an Octave, one week later, when their liturgy will be repeated once more.) On 31 October a Vigil is marked for the next day, 1 November, which is All Saints' Day. Some feasts are marked 'lc. xii' (twelve lessons) in red ink, to show their special solemnity: the Night Office will be celebrated with the full number of lessons and reponsories. Among the additions (preceded by an asterisk) the Translation of St Oswald of Worcester is entered in red ink (indicated by R in brackets in the table). (This way of indicating specially important feasts is the origin of the popular expression 'red-letter day'.)
Table 1.3.3. Kalendar for October: Cambridge, Corpus Christi College 391 * denotes material added in a later hand
a[f.]qUAT f.T OCTOBER SEMENTIS TEMPORE l1bram
1	XV i	A	KL OCT	Sanctorum Remigii. Uedasti. Germani. * xii lc. * Obitus Laefgaeuaľ mater Godithe.
2	V	B	VÍ NON	Sancti Leodegarii episcopi.
3	xiii	C	V	
4	ii	D	ÜÜ NON	
5		E	iii	
6	X	F	Ü NON	* Sancte Fidis. virginis et martiris. xii lc. * Obitus /Eadwi decani.
7		G	NON	Sancti Marci pape. Marcelli et Apulei.
8	xviii	A	viii ID	* TRANSLATIO SANCTI OSVVALDI ARCHIEPISCOPI. et Sancti Demetrii martiris. (R)
9	vii	B	vii	Sanctorum Dionisii. Rustici. et Eleutherii.
10		C	VÍ ID	Sancti Paulini episcopi.
11	XV	D	V	
12	iiii	E	ÜÜ ID	Sancti Uuilfridi episcopi. * lc. xii
13		F	iii	
3. The Church Year 15
14	xii	G	ü id	Sancti Calesti pape.
15	i	A	idus	* Comniemoratio Sanctorum quorum reliquie hie
16		B	xvü kl nov	
17	ix	c	xvi	
18		D	xv kl	Sancti Luce euangeliste et Sancti Iusti martyris. sol in scorpionem
19	xvii	E	xiiii	
20	vi	F	XÜ1 kl	
21		G	xii	* Ordinatio sancti dvnstani archiepiseopi.
22	xiiii	A	xi kl	
23	iii	B	x	
24		C	ix kl	* Obitus Henrici episeopi
25	xi	D	viii	*Sanctorum Crispini et Crispiniani.
26		E	vü kl	
27	xix	F	vi	vigília
28	viii	G	v kl	apostolorum symonis. et iude.
29		A	iiii	
30	xvi	B	iii kl	
31	v	C	ii	Sancti Quintini martyris. vigília.
nox horas .xiii. dies .x.
.Votes. The feasts may be classified as follows:
(i) feasts of the Blessed Virgin Mary, Apostles, and other biblical persons: 18 Luke the Evangelist; 28 the Apostles Simon and Jude
(ii) early martyrs, bishops (usually of Rome, i.e. popes), and Church Fathers: 7 Pope Mark (336), the martyrs Marcellus and Apuleius; 14 Pope Calixtus I (d. 222)
(iii) saints local to North France or England: 1 Rcmi of Reims, Vedast of Arras, and Germanus of Auxerre; 2 Ledger of Autun; 9 Denis of Paris and his companions Rusticus and Eleuthcrius; 10 Paulinus of York (d. 644); 12 Wilfrid of York (c.633-709); 18 Justus of Beauvais; 31 Quentin of Saint-Quentin
(iv) Additions: 1 Obit of La.'fga-va, mother of Lady Godiva, benefactress of Worcester and Bishop Wulfstan; 6 Faith, apparently a 3rd-c. martyr; obit of dean .Eadwi; 8 Translation of St Oswald of Worcester, carried out by Wulfstan in 1089; also Demetrius, a 3rd-c. martvr; IS Feast of the Relics at Worcester, instituted by Wulfstan on the Octave of St Oswald; 21 Dunstan of Canterbury (909-88); 24 Obit of Bishop Henry (d. 1189); 25 the 3rd-c. Roman martyrs Crispin and Crispinian; one legend has it that they fled persecution to F"aversham in Kent
All the indications of twelve lessons for certain feasts arc in a later hand.
This kalendar was clearly consulted for some time after its compilation, for it has received numerous additional entries, all of more or less local significance. These also are listed after Table 1.3.3. The obits (entries recording the deaths of prominent persons) are unlikely to be of liturgical moment. Rather the kalendar has been used as a convenient place to note these secular events.
There is a certain amount of astronomical information on the kalendar, the number of hours of day and night, the position of the sun in the zodiac. Not all kalendars include this, especially later medieval examples, but particularly from the thirteenth century onward kalendars were often much more specific about the degree of
16 /. Plainchant in the Liturgy
solemnity of the feast: not simply how many lessons were to be sung, but its grade according to liturgical dress (for example, whether or not copes were to be worn), the number of 'rulers' (leading singers) for the choir, and so on. The various possibilities were subsumed under a series of headings: feast of first class, second class, double, semi-double, etc., by which were understood all the possible distinctions of ritual at a particular institution. (Different churches had different systems of grading. For much useful information on the consultation of later kalendars, see Andrew Hughes 1982.)
Setting out the feasts to be celebrated in a form convenient for quick consultation, kalendars are often useful for determining the provenance of a manuscript, according to the local elements they may contain. The indication 'lc. xii' (twelve lessons) shows that Cambridge, Corpus Christi College 391 was used at a monastic institution, and its numerous Worcester peculiarities leave its provenance in no doubt. The rest of the manuscript comprises computus tables, and material for the Benedictine office: a psalter, hymnal, canticles, collectar, blessings, and a series of offices for saints, Sundays, and so on. (See III.6).
A list only of the more important saints' days, such as might be found in almost any medieval kalendár, is given in Table 1.3.4. Some of these feasts (Margaret, Katherine, Nicholas, Thomas of Canterbury) became popular only later in the Middle Ages. (For some later feasts see Pfaff 1970.)
1.4. THE DAILY ROUND
Andrew Hughes 1982.
In the Roman Church a fairly constant daily pattern or succession of services was established by the eighth century, although certain days of the year differed from the majority in having special services of an individual character. This generalization holds good despite the inevitable modifications which occurred as time went on and the varying preferences of different regions and churches.
The most important service of the day was Mass, which usually took place in the morning. (The actual time varied from one season of the year to the other, and also, of course, from place to place. The same is true of all the other daily services.)
The start of the day, from a liturgical point of view, was anciently the evening before, when the Vespers service took place, notionally at sunset. (This convention was taken over by early Christians from the Jewish reckoning of time.) It was followed by Compline, a brief service before the community retired for the night. On Sundays and important feasts the liturgical day was reckoned to extend further and to include the next Vespers and Compline service, and this meant that for all the subsequent ferias Vespers and Compline were not the evening before, so to speak, but at the end of the day. (Compline had very little proper material, special to one day rather than another, so its assignment was not a sensitive matter.) If two feast-days fell on consecutive days, then usually Vespers of the second feast took precedence.
4. The Daily Round Table 1.3.4. Saints widely celebrated in the Roman Church
17
JANUARY
14 Felix M
16 Marcellus P & M
20 Fabian & Sebastian MM
21 Agnes V&M
22 Vincent M
25 Conversion of Paul AP *
28 Octave (Second Feast) of Agnes
FEBRUARY
2 Purification of the BVM
5 Agatha V & M
22 Peter's Chair (Cathedra) *
MARCH
12   Gregory P
21 Benedict AB *
25 Annunciation of the BVM
APRIL 4   Ambrose EP & D *
23 George M
MAY
1   Philip & James APP
3 Invention of the Holy Cross *
6 John before the Latin Gate
JUNE
24 John the Baptist
26 John & Paul MM
29 Peter & Paul APP
30 Paul AP
JULY
10 Seven Brothers MM
11 Translation of Benedict AB * 20   Margaret *
22 Mary Magdalene *
25 James AP *
AUGUST
1
3 6 10 15 24
29
14
21
22
29
30
Peter's Chains (Vincula) Invention of Stephen M and companions* Transfiguration * Laurence M
Assumption of the BVM Bartholomew AP * Augustine EP & D * Beheading of John the Baptist *
SEPTEMBER
8 Nativity of the BVM Exaltation of the Holy Cross Matthew AP & EV* Maurice and his companions MM * Michael the Archangel Jerome D *
OCTOBER
9 Denis EP & M and his companions * 18 LukeEV*
28   Simon & Jude APP*
NOVEMBER
1 All Saints *
8 Four Holy Crowned Martyrs
11 Martin EP
22 Cecilia V&M
25 Katherine of Alexandria V & M *
30 Andrew AP
DECEMBER
6 Nicholas EP *
13 Lucy V&M
21 Thomas AP *
26 Stephen M
27 John the Evangelist
28 Holy Innocents
20   Thomas of Canterbury M *
31 Silvester P
* not in the 'Hadrianum', the Roman sacramentary sent to Charlemagne in the late 8th c.
AB    = Abbot EV = Evangelist
AP    = Apostle M   = Martyr
BVM = Blessed Virgin Mary     P    = Pope"
D      = Doctor V   = Virgin
EP    = Bishop
18 /. Plainchant in the Liturgy
During the night, the Night Office was sung, the longest of the services other than Mass. The service developed from the early Christian Vigils service, and was known by that name occasionally up until the eleventh century. Its most common name is Matins, but this name was also used in medieval times for the service at daybreak, Lauds, and is therefore avoided here. Yet again, the name Nocturns is occasionally found, reflecting the composition of the service as a number of 'nocturns'.
After Lauds, notionally at daybreak, there followed a succession of Day Hours, or Little Hours. These were Prime, Terce, Sext, and None (or Nones), which in theory took place at the first, third, sixth, and ninth hour of the day (beginning, in summer at least, at our 6 a.m.; the 'hours' were shorter, and began later, in winter).
The main mass of the day, often called the magna missa or high mass, usually came between Terce and Sext on Sundays and feasts, after Sext on ferial days. There were, however, often other masses of one kind or another (low mass), especially in religious communities. Medieval monasteries frequently celebrated a morning mass before the daily meeting in chapter. A daily mass for the dead was also often celebrated. Later in the Middle Ages, when many important churches had acquired altars and chantries specially endowed so that mass could be said (not usually sung) at them daily, the number of masses celebrated was staggeringly large. (Harrison 1963, 56 cites the Lincoln statutes of 1531, which indicate forty-four masses daily.) Almost all of these were of little musical significance, said by a single priest throughout. An important exception to this generalization was the Mass of the Blessed Virgin Mary, which from the thirteenth century became a weekly celebration, on Saturday, and was often embellished by the polyphonic setting of some of its chants. Such a mass, often taking place in a specially built Lady Chapel dedicated to the Virgin, is known as a votive mass. Sometimes a weekly cycle of votive masses was organized, for example: for the Holy Trinity on Sunday, the Holy Spirit Monday, Angels Tuesday, All Saints Wednesday, Corpus Christi Thursday, the Holy Cross Friday, and the Blessed Virgin Mary on Saturday. So ardent was veneration of the Blessed Virgin in the later Middle Ages that a complete cycle of Hours of the BVM, parallel with the main cycle, was often sung or said.
Some monastic observances at other times of the day were in many ways as formalized as a liturgical service. One such occasion was the daily meeting which monastic communities held in their chapter-houses. It usually included the reading of the martyrology, containing accounts of the saints to be commemorated on the day in question. Several versicles and responses were said, with Paternoster and Kyrie, and readings might include passages from the Rule of St Benedict, or from the gospels. After the business matters had been attended to, the meeting ended with a series of psalms for the dead, recited in the chapter-house or on the way to the church for mass. Little if anything in the meeting seems to have involved singing.
Among other monastic additions to the office (their description takes more than 100 pages in Tolhurst 1943) may be mentioned:
(i) the Office of the Dead, that is Vespers, the Night Office, and Lauds, said in parallel with the main or canonical office; from the first word of its first
4. The Daily Round
19
antiphon the Vespers service was known as 'Placebo', and the Night Office as 'Dirige';
(ii) a similarly formed Office of All Saints;
(iii) the fifteen Gradual Psalms (Pss. 119-33) said privately by each monk seated ready for the start of the Night Office;
(iv) the Trina Oratio, a threefold devotion performed three times daily, before the Night Office, before Prime in summer, and before Terce in winter, and after Compline, whose main constituent was the saying of three groups of penitential psalms, each group introduced by a prayer (oratio);
(v) the Psalmi Familiäres, psalms said after each hour for the royal family.
The exact time of the services depended on their content (shorter days in winter might mean shorter day services; ferial services were shorter than festal ones), on whether a period of fasting was in force, and on the other obligations of the community. The extent to which a monk's waking hours would be occupied with religious devotions or services, in church or elsewhere, may be seen from modern editions of any medieval monastic 'horarium' (timetable), for example those of tenth-and eleventh-century England, in the Regularis Concordia and the statutes of Lanfranc, respectively, published by Knowles (1963, 714, 448 ff.; Knowles 1969, 217, based on Knowles 1951). Table 1.4.1 shows the usual timetable at eleventh-century Cluny, the foremost Benedictine monastery of its time. For most days of the year, the succession of services followed this order. Several special days, such as the triduum, the three days preceding Easter Sunday, followed a somewhat different pattern. On many high feasts a solemn procession was made, most commonly before mass. An account of the most important of these special days is given later. First an outline of the more usual services is presented.
As with the constituent parts of the day, so the contents of each individual service varied from occasion to occasion, being especially dependent on the solemnity of the day. In the descriptions of each service in the sections below, as usual in this book, only a generalized account of 'normal' practice is therefore given. (Most of these services accumulated varied amounts of prefatory material in the form of prayers, versicles, and responses, but this is not described.)
Mass, the most important and complex of the services, is described first. Although it had some elements analogous to features of the office hours, its essential nature and history resulted in many unique forms. Whereas the office hours are static, contemplative services, Mass has elaborate opening ceremonies, and the elevated ritual of the eucharist itself. Its chants are exceptionally varied. The introit and communion resemble an antiphon with psalm verses simply intoned. The offertory is an elaborate chant with verses in highly ornate musical style. After the opening rituals, with the Kyrie litany chant, and the extended Gloria hymn, come lessons and melismatic chants, almost the only moments of repose in a seemingly dramatically unfolding performance. Two of its important chants, the Sanctus and Agnus Dei, are actually part of a long and elaborate series of prayers.
The most regular part of the office services, by contrast, was the singing of psalms.
Table 1.4.1. The liturgical horarium at Cluny in the late eleventh century (after Hunt 1967, 101-3)
Winter
(extra Lenten observances in brackets)
Trina Oratio last 32 psalms
Night office
4 psalmi familiäres
(2 psalms prostrate)
(Procession to Church of Our Lady)
Lauds of All Saints, Lauds of the Dead, extra psalmody
ImucIs
extra psalmody 4 psalmi familiäres (2 psalms prostrate)
Prime (at dawn) 4 psalmi familiäres
(2 psalms prostrate, 7 penitential psalms)
Litany
4 psalms
Chapter 'private' masses
Terce
4 psalmi familiäres (2 psalms prostrate)
Missa matutinalis
Sext
4 psalmi familiäres (2 psalms prostrate) Litany
Missa maior
midday meal
None
4 psalmi familiäres (2 psalms prostrate)
Vespers
4 psalmi familiäres (2 psalms prostrate)
(Procession to Church of Our Lady) Vespers of All Saints, Vespers of the Dead
evening meal
(Vigils of the Dead)
Compline Trina Oratio
Summer
Trina Oratio
Night office
2 or 4 psalmi familiäres
iMuds
2 or A psalmi familiäres Trina Oratio
Prime (at dawn)
2 or A psalmi familiäres
Litany 7 psalms
Missa matutinalis
Chapter 'private' masses
Terce
2 or 4 psalmi familiäres
Missa maior Sext
2 or 4 psalmi familiäres
midday meal None
Vespers
evening meal
Compline Trina Oratio
4. The Daily Round
21
A weekly round of the Office would hear the singing of all 150 psalms, if uninterrupted by feasts of one kind or another, which would have their own proper psalms. Most of the first two-thirds of the psalms were sung during the Night Office, the last third during Vespers. Lauds had the last three psalms, which, since they begin with the word 'Laudate', are supposed to have given the service its name. Others were used at various hours because of their allusions to particular times of day, as Ps. 62 at Lauds: 'O God thou art my God: early will I seek thee'; or Ps. 4 at Compline, with its ninth verse: 'I will lay me down in peace, and take my rest . . .'. The rather complex deployment of the psalms across the week's services is outlined in both secular and monastic cursus by Andrew Hughes 1982, 52. A simplified outline is given in Table 1.4.2. When it is stated below that a certain number of psalms are sung at such and such a service, it should be understood that 'psalm' might actually mean more than one psalm according to the numerical series, if a particularly brief psalm were involved. Correspondingly, longer psalms might be split into sections.
Table 1.4.2. Psalms allotted to office hours, ferial cursus Secular cursus:
Night Office: 1-3, 6-20, 26-41, 43-9, 51, 54-61, 63, 65, 67-88, 93-108 Vespers: 109-16, 119, 121-32, 134-41, 143-7 Prime, Terce, Sext and None: 21-5, 53, 117-18, 120 Lauds: 5, 42, 50, 52, 62, 64, 66, 89, 91-2, 99, 142, 148-50 Compline: 4, 30, 90, 133
Monastic cursus:
Night Office: 20-34, 36-41, 43-9, 51-5, 57-61, 65, 67-74, 76-86, 88, 92-108
Vespers: 109-16, 128-32, 134-41, 143-7
Prime, Terce, Sext and None: 1-2, 6-19, 118-27
Lauds: 5, 35, 42, 50, 56, 62-4, 66, 75, 87, 89, 91, 117, 142, 148-50
Compline: 4, 90, 133
On the numbering of the psalms in medieval usage, as opposed to the English Authorized Version of the Bible, sec Andrew Hughes 1982, 51, 224-6.
The descriptions below begin with the Night Office, because it is the most substantial musically, distinguished from the other services by its lessons and associated great responsories. Next in size, and somewhat similar to each other, are Vespers and Lauds, while Compline and the so-called Little Hours of Prime, Terce, Sext, and None are comparatively simple and brief.
All except the four Little Hours contain not only psalms but one or more canticles, passages from the Old Testament or other sources in the nature of 'songs', and sung exactly like psalms. The best-known examples are the three New Testament canticles sung towards the close of Vespers, Compline, and Lauds, respectively: the Magnificat (Luke 2: 46-55, the Song of the Blessed Virgin Mary, when Gabriel announced to
22
/. Plainchant in the Liturgy
her that she was to bear Christ), the Nunc Dimittis (Luke 2: 29-32, the Song of Simeon, when he had seen Christ), and the Benedictus (Luke 1: 68-79, the Song of Zacharias, when his son John the Baptist was born). Fourteen others were sung on a weekly basis at Lauds, and one other (varying with the season) was sung in the Night Office in monastic uses on Sundays and feasts.
All the offices contain at least one responsory. These chants, consisting of respond and psalm verse, with alternating solo and choral sections, are of two kinds. Those of the Night Office, the 'great' responsories, are elaborate, melismatic chants, among the chief glories of the chant repertory. The other hours usually have a 'short' responsory, syllabic in style, whose simple character has occasioned the belief that they may contain vestiges of the most ancient recoverable psalmodic practice. At Vespers on high feasts it was sometimes the practice to use one of the great responsories. (They were commonly thus employed also in processions, at least in the later Middle Ages.)
The other musical element in the office services to which attention may be drawn in advance is the hymn, one of which was generally sung at each of the hours.
The musical genres just mentioned are those that are discussed in separate chapters later in this book. Naturally, less attention is paid to the numerous versicles and responses, brief dialogues between priest (or other person) and choir, which are intoned to simple formulas at many points in the services.
1.5. MASS
Jungmann 1962; Andrew Hughes 1982.
The actual start of mass was preceded by various ceremonies of preparation. On Sundays, these included the blessing and sprinkling of salt and water on the main altar of the church, with prayers, responses, and the antiphon with psalm verse Asperges me V. Miserere mei and Gloria (Vidi aquam V. Confitemini in Paschal time, from Easter Sunday to Whit Sunday). If several altars were to be sprinkled, the ceremony might involve a procession. After the priest and his assistants had vested themselves, to the accompaniment of more prayers and the recitation of Ps. 42, they moved to the altar.
The introit accompanied this solemn entry. Because it was part of the proper, a different introit being sung on each important day of the year, it was usually the first chant copied for a particular day in a gradual or missal.
The Kyrie and Gloria followed, their texts remaining basically the same throughout the year and therefore being reckoned part of the ordinary of mass. From a musical point of view, the distinction is not strictly appropriate, since a variety of melodies were available for both chants, and particular melodies became associated with particular feast-days, or classes of feasts (important, not so important, feasts of the Blessed Virgin Mary, etc.). Furthermore, tropes of one kind or another might make the Kyrie and Gloria proper to a particular feast.
5. Mass
23
The priest then intoned the collect proper to the day. This ended the opening section of the mass.
There followed the lessons and responsorial chants of mass. The lessons were the epistle and the gospel, between which were sung various combinations of gradual, alleluia, sequence, and tract, depending on the liturgical season. The gradual was a more or less constant item, except for the Saturday of Easter week and the days up to Whit Sunday, when an extra alleluia was sung. The alleluia was not sung from Septuagesima to the end of Lent, nor on penitential Ember Days outside Lent, though this was relaxed for the Pentecost Ember Days. The tract was sung only in penitential seasons: on Septuagesima and succeeding Sundays until Easter, and on Mondays, Wednesdays, and Fridays during Lent, beginning with Ash Wednesday. The sequence was sung only on the highest feasts, and not usually on those falling in Lent.
This part of mass was considerably extended on the Ember Saturdays. Four lessons were each followed by a gradual (an alleluia at Pentecost), then a reading from Daniel was followed by the canticle (sometimes labelled 'tractus' or 'hymnus') with refrain (Benedictus es Domine V. Et laudabilis in Advent, Benedictus es in firmamento V. Hymnum dicite in Lent, Omnipotentem semper adorant V. Et benedicunt in September. Modern practice has Benedictus es on all three occasions.) After that came the epistle, another tract, and the gospel. The Wednesday of the fourth week of Lent also had extra lessons: this was the so-called 'Day of the Great Scrutiny', when catechumens (see below) were examined before being admitted to baptism.
The gospel was intoned by the deacon. The Credo was then sung (from the eleventh century, and only on the most solemn days).
In the early centuries the mass up until the gospel had been regarded as a 'fore-mass' service, for all Christians to attend. What followed, however, was for baptized Christians only. By the central Middle Ages no trace remained of the dismissal of the catechumens (those undergoing preparation for baptism), which would previously have taken place here. The change in the character of the liturgy is noticeable, nevertheless, the entrance ceremonies having been succeeded by lessons and responsorial chants, which now gave way to the solemn prayers and chants of the eucharistie ritual.
The gifts were brought to the altar as the offertory chant was sung, and the music might last through the priest's preparation of the altar, reception of and prayers over the gifts, including the silent prayer known as the secret (proper to the day). There followed the preface, which began with short phrases sung in dialogue between priest and congregation and was continued by the priest alone, a long text with sections proper to the season. It led directly into the Sanctus chant, sung by the choir. The text of the preface always requires the Sanctus as its natural conclusion, for example:
(Preface, ending) . . . Et ideo cum angelis et archangelis, cum thronis et dominationibus, cumque omni militia caelestis exercitus, hymnum gloriae tuae canimus, sine fine dicentes: Sanctus, Sanctus, Sanctus Dominus Deus Sabaoth . . .
24
/. Plainchant in the Liturgy
And therefore with angels and archangels, with thrones and dominations, and the whole host of the heavenly army, we sing the hymn of thy glory, evermore singing: Holy, Holy, Holy Lord God of Hosts . . .
Sanctus and Benedictus are musically linked, both concluding with 'Hosanna in excelsis', but from the later Middle Ages it was customary to split off the Benedictus and sing it during a later part of the prayers of this part of mass, for example after the consecration of the bread and wine. The prayers began with the canon, a long series of brief formulas which include Christ's words 'This is my body . . . This is the cup of my blood . . .' The Paternoster (Lord's Prayer) was sung shortly after, and the bread was broken. During the next group of prayers a small portion of the bread was mingled with the wine, and the Agnus Dei chant was sung by the choir. Further solemn prayers were sung while the bread and wine were consumed by the priest (they were rarely received by anyone else in the Middle Ages). The communion chant was sung during the clearing-up actions, after which the priest intoned the postcommunion prayer of thanksgiving (proper). During the dismissal of the congregation, the priest would intone another proper prayer, the Super Populum (literally 'over the people'), during Lent.
Tte missa est' ('Go, you are dismissed'), the words which gave the mass its name (missa)f was the usual form of dismissal sung by the priest, answered 'Deo gratias' ('Thanks be to God') by the choir. However, it was generally used only when the Gloria was sung, that is during Christmas, Epiphany, Easter, and the summer season. On other occasions the conclusion was 'Benedicamus Domino R. Deo gratias', and this formula was also used when another service followed immediately upon mass, as would be the case on several of the highest feasts with extended liturgies (Andrew Hughes 1982, 93).
Since early times it has been the custom, for various practical reasons, to preserve the bread (occasionally also the wine, also bread onto which a drop of wine has been placed) consecrated at mass but not consumed there. This is known as reservation of the sacrament. It may be kept in one of several places: the sacristy, outside the church; an aumbry in the wall of the church; a pyx suspended over the altar; or a tabernacle on the altar.
In the Middle Ages, and for centuries afterwards, communion by any person other than the priest was relatively unknown, outside one or two of the most important days of the church year. For theological reasons (and perhaps, for many, because of the lack of closer contact with the sacred elements) there developed from the thirteenth century on the practice of elevation. After it had been consecrated, the celebrant raised on high the host (that is, the bread—elevation of the wine was a later trend) for all to see. The singing of the Benedictus chant, musically a part of the Sanctus, was often delayed to coincide with this solemn moment.
Veneration for the sacred host eventually resulted in ritual forms, in particular Benediction. Here the host was exhibited outside mass, and a procession might be made with it, culminating in a general blessing with the host. Favourite accompanying
6. Office 25
chants were hymns such as Adoro te devote and Tan turn ergo sacramentum (the fifth verse and following of Pange lingua glon'osi). As a result of their great popularity, many of these have multiple melodies. (The chief result of the increase in veneration for the host was the establishment of the feast of Corpus Christi, introduced in Liege in 1246, prescribed for the whole church by Urban IV in 1264.)
1.6. OFFICE
(i) The Night Office (Matins, Vigils, or Nocturns) in Secular Use
(ii) The Night Office in Monastic Use
(iii) Lauds in Secular Use
(iv) Lauds in Monastic Use
(v) Vespers in Secular Use
(vi) Vespers in Monastic Use
(vii) Compline
(viii) Prime, Terce, Sext, and None Tolhurst 1943; Andrew Hughes 1982.
(i) The Night Office (Matins, Vigils, or Nocturns) in Secular Use
Apart from the differences between festal and ferial forms of the Night Office, there are other important differences between secular and monastic forms. (By 'secular', 'Roman', or 'canonical' is understood the use of non-monastic churches such as most cathedrals, collegiate chapels, and parish churches.) As with most of the offices, the singing of psalms and antiphons forms an important part of the service, but the Night Office is distinguished by a group of great responsories, sung in association with the lessons. These numerous and lengthy responsorial chants are at once among the chief glories of the chant repertory and yet among its least known parts, since in modern times the singing of the Night Office in its medieval manner has all but ceased.
Before the Night Office began, Paternoster, Ave Maria and the Apostles' Creed (Credo in Deum pattern omnipotentem) were said silently.
The Night Office started with versicles and responses. The first was usually Deus in adiutorium meum intende R. Domine ad adiuvandum me festina, with Gloria, which began each of the office hours (it was often omitted during the triduum, that is the last three days before Easter Sunday, and during Easter Week). This was followed by Domina labia mea aperies R. Et os meuirn annuntiabit laudern tuam.
The Invitatory psalm (Ps. 94, Venite exultemus Dominum) was sung, with its antiphon, the invitatory. A hymn followed (less commonly found in early medieval secular uses than in monastic ones). A further versicle and response led to the first Nocturn. This was a group of psalms and antiphons, lessons and responsories, with more brief versicles and responses. The number of nocturns depended on the solemnity of the day, less important days having only one.
26 /. Plainchant in the Liturgy
Psalms and antiphons began each nocturn. There were different conventions about the number of psalms and antiphons. The singing of an antiphon preceded the whole group of psalms and concluded the series, but it might not be performed between individual psalms in the group. The same was true of the Gloria ('Glory be to the Father . . .'), added as a final verse to all or only to the last of the psalms. Some of the commoner schemes were the following (A = Antiphon, Ps = Psalm, Gl = Gloria patri):
On ferial days: A, two Ps, Gl, A—six times (total of 12 Ps, 6 A) On Sundays: A, four Ps, Gl, A—three times (total of 12 Ps, 3 A) On major feasts: A, Ps, Gl, A—three times (total of 3 Ps, 3 A)
A further versicle and response were sung, commonly to a tone which, like the opening Deus in adiutorium, recurred at other hours of the day.
The Paternoster was said silently until the final verse, which was said aloud. Then the lector asked for the blessing of the officiating priest. At the end of the lesson he would intone a further versicle.
After each lesson (with versicles), one of the great responsories was sung. In secular use three lessons and thus three responsories were sung in each nocturn. The usual form of the responsory was respond-verse-respond, but for the last responsory of the nocturn the Gloria would be used as a second verse, as it were, making the scheme R, V, R, Gl, R. (The repeats of the respond were usually progressively shortened.)
The lessons might not all be biblical. The first nocturn (on some days the only nocturn) had Old Testament readings (except during the Easter season), from different books according to season. The others commonly had sermons (commentaries on the appropriate liturgical theme by one of the Fathers of the Church, such as St Augustine of Hippo or St Gregory the Great) or homilies (commentaries, again by a Church Feather, on a passage from the Gospels, which would therefore itself be intoned at the start of the lesson). The lessons on a saint's day might be drawn from his vita or legenda (account of his life, death and miracles—the terms mean literally 'life' and 'that which is to be read').
The number of nocturns to be sung according the above patterns depended upon the solemnity of the day. Sundays and feast-days usually had three. An important exception was Easter Sunday, which had only one (but some later medieval books give it three), and this was also sometimes the pattern for the rest of the Easter season up to and including Whit Sunday.
Although in general antiphons and responsories may be said to be proper, in the sense that each day usually had its own particular chants, there was a considerable amount of reuse within weeks or seasons. For example, for those parts of the year where nine responsories were sung on Sunday and three on each weekday, the Sunday-nine might be re-employed during the week, in sets of three. (See Andrew Hughes 1982 for some of the complex schemes evolved.)
The Night Office ended with the singing of the Te Deum on Sundays and on most feasts except in Advent and Lent. In the later Middle Ages, the last responsory might
6. Office 27
be repeated if the Te Deum were not sung. If Lauds did not immediately follow the Night Office, then the closing formula (priest) Benedicamus domino V. (choir) Deo gratias would be sung. (The conventions about when it was sung are somewhat unclear: the preceding statement is derived from Andrew Hughes 1982, 66.)
(ii) The Night Office in Monastic Use
Monastic use (the Benedictine pattern is referred to here, but not all monastic uses were exactly the same) differed in many important details from the secular norm outlined above, though the basic pattern was similar. Only the chief dissimilarities are pointed out here.
The larger differences are that although the number of nocturns was generally three on Sundays and feasts, there were two on ferias (not one as in secular use). Furthermore, the nocturns were not of an identical pattern.
After the opening versicle and response, Ps. 3 was intoned, without any antiphon. The numbers of psalms and antiphons, lessons and responsories in each nocturn were substantially different from secular patterns. The antiphon-psalm pattern was the following:
Ferias, both nocturns:
either: A, Ps, Gl, A—six times
or: A, six Ps, Gl, A
Sundays and feasts:
first nocturn: A, Ps, Gl, A—six times second nocturn: A, Ps, Gl, A—six times third nocturn:     A, three canticles, Gl, A
(total of 13 A)
On summer ferias, the rest of the first nocturn had but one lesson and responsory, and the second nocturn only a very brief lesson, no more than a verse of scripture, commonly called a capiluluni (chapter). There followed a versicle and response, the short Kyrie litany, Paternoster, and a closing collect.
Winter ferias were similar except that in the first nocturn there were three lessons, each followed by a responsory.
On Sundays and feasts, the three nocturns each had four lessons with their responsories. Furthermore, after the Te Deum, and versicles and responses, a passage from the Gospel was intoned, making an entirely different climax to the service. On Christmas Day the gospel was the Genealogy of Christ from St Matthew's Gospel, on Epiphany the Genealogy from St Luke. The hymn Te decet laus was sung, then followed the blessing and final collect, with Benedicamus Domino when appropriate.
In some early antiphoners, for example the Hartker Antiphoner (St Gall, Stiftsbibliothek 390-391; PalMus II/l), the first two nocturns each have three lessons and responsories, the third nocturn also sometimes three of each, more often four. This is understood to be a survival of earlier arrangements wherein monastic use was not as strongly differentiated from secular as was later the case.
28
/. Plainchant in the Liturgy
(iii) Lauds in Secular Use
Lauds and Vespers were two similar services, at sunrise and sunset, respectively. Of the two, it may be said that Vespers has the larger body of material proper to particular days of the week, or to various seasons of the year.
In secular use, Lauds began with versicles and responses, the first of which was, as usual with the office services, Deus in adiutorium. There followed five psalms, either each with its Gloria and antiphon, or else with only one antiphon and the Gloria said only after the last psalm.
In the weekly or ferial cycle, the psalms on Sunday were repeated during the week, except for the second, which was different on each of the other six days. Regarding psalms it should be understood that the third psalm was Pss. 62 and 66 together, the fourth was one of the lesser canticles, and the fifth was Pss. 148—50. Like the second psalm, the lesser canticle changed daily, and also according to season; there were fourteen in all (listed by Andrew Hughes 1982, 365). Feast-days would occasion the selection of other psalms and another canticle, but the pattern remained the same.
There followed a brief chapter, with choral response 'Deo gratias', a hymn, a versicle and response, and then the major canticle of Lauds, the Benedictus, with its antiphon (a relatively long and elaborate one). Final blessings and prayers followed and the Benedicamus Domino.
(iv) Lauds in Monastic Use
In monastic use Lauds had almost the same form as in secular use. The psalms were chosen, or arranged, differently; but the notional number of five (including the lesser canticle) was the same. Monastic Lauds, however, included a short responsory after the little chapter.
(v) Vespers in Secular Use
As mentioned above (1.4), Vespers on most Sundays and important feasts was celebrated twice, first in its ancient position as a vigil ceremony anticipating the succeeding holy day, secondly at the end of the same day. Most of the material of the two would be identical, though with a certain reduction in solemnity, by using common rather than proper items, omissions, and shortenings, at Second Vespers. (Andrew Hughes 1982, 69 ff. gives detailed information.)
Not only did Vespers have a structure similar to that of Lauds, but some of the items for any one day might be sung at both Lauds and Vespers: for example, the large antiphon for the major canticle, the psalm antiphons, chapter, and hymn.
Instead of the opening versicle Deus in adiutorium, omitted anyway around Easter, a ninefold Kyrie was sung during Easter week, the so-called 'Paschal Kyrie' (Vatican I, Melnicki melody 39).
The psalms for the weekly cycle form a continuous numerical sequence at Vespers,
6. Office 29
and there is none of the repetition during the week of Sunday items that characterizes ferial Lauds. The notional number of five psalms is the same, but a lesser canticle forms no part of the group. During Easter week, only three psalms were sung, with one Gloria and antiphon; they were followed by the Easter gradual Hec dies, with verse changing day by day, and an alleluia varied similarly through the week.
After the chapter on the more important days of the year, a responsory was sung (as in monastic Lauds, but here not on days of lesser importance). Occasionally this was one of the great responsories, rather than a short one.
Particularly notable among the antiphons for the Vespers canticle, the Magnificat, are those for ferial days before Christmas, the great 'O' antiphons, so called because their texts all begin with 'O', all of which use the same melody.
(vi) Vespers in Monastic Use
The essential difference between secular and monastic Vespers was the presence of only four psalms in monastic use. Only a smaller number of psalms could therefore be sung during the weekly cycle, the others being taken into the Little Hours (Table 1.6.1.).
Table 1.6.1. Psalms at ferial Vespers
	Secular	Monastic
Sunday	109-13	109-12
Monday	114-16,119-20	113-16, 128 (115+116 as 1 psalm)
Tuesdav j	121-5	129-32
Wednesday	126-30	134-7
Thursday	131-2, 134-6	138-40 (138 as 2 psalms)
Friday	137-41	141, 143 as 2 psalms, 144 first half
Saturday	143-7	144 second half, 145-7
		(119-27 to the Little Hours)
At the end of both Lauds and Vespers it was sometimes customary to say or sing a series of antiphons, called 'suffrages', for the Blessed Virgin Mary and the patron saints of the church.
(vii) Compline
Compline, the last service of the evening, and the four short hours of the day, Prime, Terce, Sext, and None, were similar in length and content, although their histories were not identical.
Versicles and responses opened the service as usual, followed by four psalms, sung with one antiphon only and Gloria: A, 4 Ps, Gl, A. The hymn came next, then the
30 /. Plainchant in the Liturgy
chapter was intoned, and a short responsory was sung. After the versicle and response Custodi nos R. Sub umbra ('Keep me as the apple of thine eye . . .') came the canticle Nunc dimittis with its antiphon. On more important days the closing items would include the preces, beginning with Kyrie invocations (only intoned, as usual with prayers), Lord's Prayer and Apostles' Creed (both said silently until the last verse). There might also be confession and absolution. After the Benedicamus the service ended (at least from the later Middle Ages) with a votive antiphon to the Blessed Virgin Mary, and a collect for the same.
Rather little of Compline changed during the year, or even during the week. In Easter week, however, Ps. 90 fell out of the cursus, and the Nunc dimittis was moved to take its place. There followed the Easter gradual Hec dies V. In resurrectione tua, as at Vespers.
Monastic use differed in that there were but three psalms, sung without an antiphon (but usually each with its Gloria). More surprisingly, the Nunc Dimittis was not part of monastic Compline, except when sung during Easter week as a psalm. The service was thus even more like one of the Little Hours than was secular Compline.
(viii) Prime, Terce, Sext, and None
In each of these Little Hours there were generally three psalms with only one antiphon, but usually each with its Gloria: A, Ps + Gl, Ps + Gl, Ps + Gl, A. On Sunday and some feasts, however, Prime had nine psalms, except during the Easter season. Prime was also different in that after the psalms the Athanasian Creed Quincunque vult was usually sung, with an antiphon, except during the triduum and Easter.
Unlike other hours, these had the hymn after the opening formularies. After the psalms came the chapter with response, short responsory, but no canticle. There were the usual series of versicles and responses and other prayers at either end of the service, similar to the arrangement at Compline.
Monastic use had four psalms on Sunday at Prime (actually four sections of Ps. 118), and short responsories were not usually sung, but otherwise there was little difference from secular arrangements in the Little Hours.
1.7. PROCESSIONS
Bailey 1971.
Liturgical processions were a common feature of medieval worship on feast-days. Their use was particularly subject to local variation, since they were necessarily adapted to the design of the church buildings and the geography of the town where they took place, and the identity of the relics which often featured prominently in the proceedings. But although irregular and variable in frequency and form, processions
Processions
31
were highly organized rituals, whose rubrics are set out in medieval books with as much care and exactitude as those of any other service. The main purpose of the procession was to visit some holy place within the main church (such as an altar) or without (another church in the town). Chants would be sung on the way there and on the way back. There a station was made, that is, the persons in the procession halted. Prayers would be said, chants sung, and usually some special ceremonial action performed, such as the sprinkling of the altar with holy water or the veneration of the relics of a saint preserved at the place of the station. Sometimes mass would be celebrated.
Thus, on the feast of the Purification of the Blessed Virgin Mary (2 February), it was the custom to make a procession with lighted candles, symbolizing Christ as the light of the world. The occasion in the gospels from which this ceremony takes its theme is the Virgin's visit to the temple in Jerusalem for the Jewish purification ritual. In Luke 2: 2-39 it is recounted how the child Jesus was presented in the temple (as ritual also demanded), and the aged Simeon recognized him as the Christ, declaring him to be 'a light to lighten the Gentiles'. After Terce, therefore, in medieval liturgies, an array of candles was blessed by the priest, with prayers, sprinkling them with holy water, and censing them. They were distributed to all those present, while the Nunc Dimittis canticle (Simeon's words themselves) was sung, with the antiphon Lumen ad revelationem gentium. The deacon turned to the people and sang the versicle 'Let us proceed in peace', and they answered 'In the name of Christ. Amen.'
The procession then set off, a typical order having the thurifer first, swinging the censer, then the subdeacon with the cross, with two acolytes with candles on either side, then all the clergy in order of rank, ending with the priest, and deacon on his left. During the procession, grand ceremonial antiphons would be sung, such as the Aciorna thalamum suum (translated from a Byzantine original: see Wellesz 1947, 60 ff.—the processional ceremony apparently derives from Byzantine usage). The procession would perambulate the churchyard or cloister, and on returning the priest and clergy would change vestments for the celebration of mass, which followed immediately.
Very different was the atmosphere on Ash Wednesday, the beginning of Lent. As usual on a feria, Mass would follow Sext, but it was preceded by the ceremony of the laying of ashes on the heads of those present (originally of sinners who thenceforth wore sackcloth and were excluded from Holy Communion, until their period of penitence ended with a ceremony of reconciliation on Maundy Thursday). The service might begin with the singing of the Seven Penitential Psalms (Pss 6, 32, 38, 51, 102, 130, 143) with an antiphon. The priest would intone various prayers over a bowl of ashes, and versicles and responses would be sung. He would sprinkle holy water over the ashes; then he would first receive them on his own forehead in the form of a cross and then place them in the same fashion on the foreheads of the others present. The antiphon Exaudi nos Deus, with an intoned psalm verse Salvum mefac, often accompanied this action, and also ceremonial antiphons such as Immutemur habitu in cinere et cilicio ('Let us change our garments for sackcloth and ashes') and
32 /- Plainchant in the Liturgy
Emendernus in melius ('Let us amend . . .') with verse Adiuva nos Domine. A procession would form, and move to the west door of the church, where the penitents would be expelled, one by one, kissing the hand of the official delegated for the task. The door would be shut and the clergy and choir would return to the east end of the church for mass. During the processional movement, there and back, ceremonial antiphons such as those mentioned would be sung.
Some processions were regular and oft-repeated ceremonies, for example the procession before the main mass held on Sundays and major feasts at many medieval churches. Others, being proper to a particular saint, would often contain special items of considerable musical interest, for example a responsory with a prosula, or even a chant set in polyphony. Much of the material sung or said in processions might be borrowed from other services of the day. A favourite practice in the later Middle Ages was to borrow one of the great responsories of the Night Office (which was why the Solesmes Processionale monasticum of 1893 has long been a useful source for the study of the responsories, poorly represented in modern editions). Several processional responsories or antiphons (the terms seem often to have been interchangeable) were of a different musical character, having possibly survived from Gallican liturgical uses of the time before the Carolingian move towards Roman practice. They were frequently long chants in ornate style, sometimes with a less lengthy verse (hence the designation 'responsorial').
Bailey (1971) lists several other points during the day, apart from the procession before mass, when processions were held in Sarum use: during Vespers in Easter week and on Saturdays through the summer; after Vespers on the eve of a saint with an altar in the church, and on the four saints' days after Christmas (SS Stephen, John the Evangelist, Holy Innocents, Thomas of Canterbury); during the Night Office in Easter week; after the Night Office on Easter Day; and after None on Wednesdays and Fridays in Lent, the Rogation Days, the triduum, and on the eve of Whit Sunday.
Some of the most ancient and solemn liturgies of the church year, such as those of the triduum, include processions, and these are mentioned in their place in the next chapters. The unique colour and splendour of many of the processions—with their special vestments, crosses, censers, and ranks of choir and clergy—seen against the steady regularity of the office hours, must have contributed greatly to the awe-inspiring solemnity of the medieval liturgy.
1.8. CEREMONIES OF HOLY WEEK
(i) General
(ii) Palm Sunday
(iii) Maundy Thursday
(iv) Good Friday
(v) Holy Saturday or Easter Eve
(vi) Easter Sunday
8. Ceremonies of Holy Week 33
Tyrcr 1932; Young 1933; Hermann Schmidt 1956-7; Hardison 1965; Andrew Hughes 1982.
(i) General
The ceremonies of Holy Week are among the most ancient and universal of the Church. Here the normal sequence of mass and office hours is frequently disturbed, and the content of the services is usually not like that of mass and the office hours on other days.
Lent, the period of the church year before Easter, is popularly understood to be a penitential season. The penance implied was in ancient times a very real observance, undertaken freely or by command of an ecclesiastical authority, and involved the exclusion of the person concerned from mass until a ceremony of reconciliation on Maundy Thursday. (The ceremony of exclusion has been mentioned above, 1.7.) Lent was also the time of preparation for the catechumens, those being trained and instructed for baptism. After a series of 'scrutinies' on various Lenten weekdays (the most important, the Great Scrutiny, took place on the Wednesday of the fourth week in Lent), to ascertain who was suitable, baptism finally took place on Easter Eve. Numerous rituals of deep symbolism surrounded the final days before Easter, such as the extinguishing of candles for Maundy Thursday and then the lighting of the great Paschal Candle, the light of Christ, on Easter Eve. Coinciding as it did with the passage from winter to spring, this progress from penitential preparation and the death of Christ to his resurrection, and the rebirth in Christ of those newly admitted to communion, gave the Passiontide and Easter liturgies immense power and importance. When Christ had died on Good Friday, no consecration of bread and wine for communion, his body and blood, was possible until the Vigil Mass (or Paschal Mass) on Easter Eve. A previously consecrated eucharist had to be used, with corresponding alteration of the ceremony, which was therefore called the Mass of the Presanctified.
Some of the general differences in liturgical matters between Lent and the rest of the year were indicated briefly above, when the essential structure of each service was indicated (1.5-6). At a ceremonial level, a new note was struck on the fifth Sunday of Lent, known as Passion Sunday, when crucifixes, images, and pictures in the church were veiled in purple, and the Gloria patri was omitted from psalms, invitatories, and introits. On the next Sunday, Palm Sunday, the most important procession of the year was made, a symbolic re-enactment of Christ's entry into Jerusalem.
(ii) Palm Sunday
The procession on Palm Sunday took place between Terce and Mass, as usual on a Sunday. The main point of the action is that the priest, as Christ's representative, will make a ceremonial entry into his church from the outside, as if entering Jerusalem. Quite often the procession would cover a considerable distance, visiting other
34 /. Plainchant in the Liturgy
churches or special places, where a consecrated host might have been placed: this would then represent Christ on his entry. The amount of music normally provided in medieval service books—processional antiphons and hymns, and so on—is correspondingly generous. There is of course great variety between different books in the exact form of the procession and correspondingly in the choice of chants (see Edmund Bishop, 1918 for discussion of the practice at Rouen, Salisbury, and Hereford). What is sketched below is merely a typical order of ceremony.
The special objects used in the procession, palms, branches of olive trees, or other branches, must of course first be blessed, consecrated for the purpose. This was commonly done to the accompaniment of the short ceremonial antiphon Hosanna filio David (without verse), with prayers, sprinkling with holy water, and censing. The consecrated palms were then distributed, while verses from Psalms 23 and 46 were sung, with the antiphons Pueri Hebreorum port antes and Pueri Hebreorum vestimenta after every pair of verses. The gospel might then be read, before the procession formed up and left the church.
Some of the most notable of processional antiphons and hymns were sung during this ceremony, including the antiphons Cum appropinquaret Dominus, Cum audisset populus, Collegerunt pontifices V. Unus autem ex ipsis, and the refrain-hymns En rex vend and Gloria laus et honor. The antiphon Ingrediente Domino V. Cum audisset was commonly sung at the moment of re-entry into the church. A final station might be made before the principal cross of the church.
Mass would then begin. Two of its most important special features were the singing of the longest tract in the repertory, Deus Deus meus, with as many as fourteen verses (from Ps. 21), and the intoning of the Passion, that is the gospel account of Christ's betrayal and death, on this day from Matthew 26 and 27. The Passion was intoned from Mark's Gospel on Tuesday, from Luke on Wednesday, and from John at the Mass of the Presanctified on Good Friday, in each case to a tone different from that usual for gospel lessons.
(iii) Maundy Thursday
On this day Christ ate the Last Supper with his disciples, and instituted the Eucharist. He washed the feet of the disciples, and the words of the first antiphon sung during the medieval repetition of this ceremony, Mandatum novum, gave its name to the day itself.
The Night Office and Lauds on this and the next two days acquired the name of Tenebrae (literally 'shadows') because of the custom of extinguishing the lights during the two services. The Night Office would begin with only one set of fifteen candles before the altar. A candle was extinguished after each of the psalms, until the Benedictus of Lauds was reached, after which the one remaining candle was hidden behind the altar.
The Little Hours might be said and not sung, without lights.
On these days the Night Office had nocturns of three lessons even in monastic use,
8. Ceremonies of Holy Week 35
and on Holy Saturday only one nocturn (although later in the Middle Ages there was a tendency to make it a three-nocturn service.) The first three lessons of the Night Office (the only three on Saturday) were taken from the Lamentations of Jeremiah (they were not sung to a tone other than the usual one, however, although the blessings for the reader might be omitted).
Mass on Maundy Thursday, which took place after None (notionally 3 p.m.), was especially complicated by the ceremonies of the Reconciliation of Penitents and the Blessing of the Holy Oils. The former preceded mass, and involved the repeated genuflection and prostration of the penitents as they made their way from the west door of the church in procession. The Seven Pentitential Psalms were said; and the antiphon Venite venite was sung between each verse of Ps. 33 : 'Come ye children, and hearken unto me: I will teach you the fear of the Lord.' The short Kyrie invocation was sung during the series of prayers, versicles, and responses.
Mass followed, and during the Gloria in excelsis the bells might be rung for the last time until Easter Eve. The Blessing of the Holy Oils had one special musical item, the refrain hymn O redemptor sume carnem, to accompany the bringing of the oils from the sacristy to the choir, and their return after the blessing. The point at which this took place was usually just before the consecration of the hosts, not only of the host for this mass but also that required for the next day. The oils were of a special type known as the chrism, hence the name for this mass of Missa Chrismalis. They were to be used for the anointing of the baptized in two days' time. The host which was to be 'reserved' for future use was carried to a special place either before the end of this part of the ceremony or after the Vespers section. This translation and reservation was carried out with considerable solemnity, with a properly formed procession, and, of course, chants, such as the hymn Pange lingua. The host might actually be buried in a sepulchre, thus anticipating the burial of Christ himself; a more solemn burial, or deposition, would take place on Good Friday.
As soon as the communion chant was over, Vespers began, without even the Deus m adiutorium versicle and response. Vespers consisted only of five psalms and the Magnificat canticle, all with antiphons. The postcommunion prayer for the end of mass was postponed until after this reduced Vespers, with Ite missa est to conclude.
Compline would follow later, but first the altars of the church would be stripped and washed with holy water (which would first be blessed), the whole procedure accompanied by some sequence of psalms with antiphons, responsories borrowed from the Night Office, or other ceremonial antiphons, and prayers.
Either now, as a separate ceremony, or occasionally at some point in the mass (such as after the gospel), the ceremony of the foot-washing, 'pedilavium' or 'Mandatum', was performed. It might be performed at any convenient spot, in church or perhaps in the chapter-house. Usually twelve chosen men (the number of disciples) had their feet washed by the officiating priest. The ceremony might be preceded by the relevant gospel reading, perhaps even a sermon, and it was followed by prayers and versicles. During the action of washing a series of ceremonial antiphons was sung, some with intoned psalm verses. The first in the series was Mandatum novum do vobis, but the
36 /. Plainchant in the Liturgy
choice thereafter varied widely. (An antiphon found in some uses, Venit ad Petrum, is occasionally provided with a long melisma on the word caput. In Sarum use the antiphon is last in the series, and the Sarum version of the melisma is famous as the cantus firmus of a polyphonic setting of the ordinary of mass once attributed to Dufay.)
After Compline a watch might be kept by the reserved sacrament, (iv) Good Friday
The Night Office and Lauds had the same form as on Maundy Thursday, having the character of Tenebrae, and including further readings from the Lamentations.
Again as on Maundy Thursday, Mass took place after None, and contained even more special rituals, notably the Adoration of the Cross. As usual during the triduum, normal ceremonial was largely stripped away, and none of the usual opening items— introit, Kyrie, or Gloria—was sung. The service began straightaway with a series of lessons and responsorial chants. The order is: lesson, tract, prayer, lesson, tract, Passion (St John). (The tracts are often labelled 'responsory', and thus stand somewhat apart from the rest of the repertory, as do the 'cantica' of Holy Saturday.)
Nine solemn collects followed, prayers following a special formula. First the priest, beginning 'Oremus et pro ..." ('Let us pray for . . .'), announced the purpose of the prayer. He then intoned the single word 'Oremus', and he or the deacon sang 'Flectamus genua' ('Let us bow the knee'). All prayed in silence until the command 'Levate' ('Rise').
The Adoration (or Veneration) of the Cross now took place. A cross with the image of the crucified Christ, veiled in purple, was brought with processional solemnity from the sacristy or wherever it had been kept. The priest uncovered it a little and the antiphon Ecce lignum crucis was sung three times, at successively higher pitches. After a few moments of silent adoration, the cross was completely unveiled, and the principal act of adoration took place. This involved genuflection before the cross and the kissing of the feet of the Saviour on the cross. The first two chants sung during this action were refrain chants of a special kind, known as the Improperia (Reproaches) after the character of the text. Each main verse had Christ, as it were, saying what he had done for his people (drawing on events in the Old Testament), and reproaching them with their cruel ingratitude. In both chants, some sort of alternatim scheme was used for the performance, either between soloist and choir or between two choirs.
The first had as refrain a group of alternating Greek and Latin invocations, 'Agios . . . Sanctus . . .', known as the Trisagion. As set out in modern Roman missals, the scheme is as follows:
two cantors in mid-choir: Popule mens ... V. Quia eduxi te de terra Egypti . . . choir 1: Agios o Theos choir 2: Sanctus Deus choir 1: Agios ischyros
8. Ceremonies of Holy Week 37
choir 2: Sanctus fortis
choir 1: Agios athanatos eleison imas
choir 2: Sanctus immortahs miserere nobis
two soloists from choir 2: Quia eduxi te per desertion . . .
choirs 1 and 2: Agios . . . Sanctus . . . (etc. as before)
two soloists from choir 1: Quid ultra debui facere tibi . . .
choirs 1 and 2: Agios . . . Sanctus . . . (etc. as before)
The second of the Improperia had a simpler refrain, namely the respond Popule meus (without verse) of the first chant. This was sung in answer to each of a long series of verses beginning 'Ego . . .', chanted to a simple tone by cantors from either side of the choir in turn:
cantors of choir 1: VI full: Popule meus . . . cantors of choir 2: V2 full: Popule meus . . . —and so on
The antiphon Crucem tuam adoramus was then sung, with simply intoned verse Deus misereatur nostri. The final chant of the group was the refrain hymn Crux fidelis V. Pange lingua gloriosi, where the refrain is divided into two, and each part in turn performed between the verses:
Crux fidelis (complete)
VI Pange lingua
Crux fidelis (first part)
V2 De parentis
Crux fidelis (second part)
V3 Hoc opus
—and so on
Mass now began, sometimes with a procession to bring the presanctified host from where it had been kept, with accompanying antiphons such as Salvator mundi salva nos. The prayers for the eucharist were changed from those of other days to reflect the difference of the Blessed Sacrament, and the ritual was said rather than sung, except for the chanting of appropriate responsories or antiphons from the office of the day. The same might be used for a solemn procession to return the Blessed Sacrament to the sacristy or other place where it was kept, or even to 'bury' it again in a special tabernacle or 'sepulchre'. Many local versions of the ceremony of burying the host and/or a cross (the 'Depositiae hostiae/crucis') are known (described in Young 1933, chs. IV-V), together with their exhumation before the Night Office early on Easter morning.
All lights were extinguished. Vespers was omitted. There remained only Compline to sing.
38
/. Plainchant in the Liturgy
(v) Holy Saturday or Easter Eve
At the Night Office, which followed the secular cursus even in monastic use, the Lamentations once more provided the lessons. Night Office and Lauds together again formed the Tenebrae services. The Little Hours and Vespers were sung in their turn, but the whole focus of the day was towards its end, and the momentous ceremonies of the Blessing of the New Fire, the sacrament of Baptism, and the Mass of the Paschal Vigil, usually timed so as to start at midnight.
The lighting of the Paschal Candle, symbolizing the light of Christ, was surrounded by complex and solemn ritual, involving the striking of fire from a flint, the placing of five grains in the candle, the lighting of the candle and, from it, all the others held by those present. Each action was accompanied by sprinklings and censing where necessary, and, as the Paschal Candle was lit and shown to the people, the solemn intonation 'Lumen Christi' was sung three times at successively higher pitches, and answered 'Deo gratias'. In some uses a procession would be made with the candle, calling for a chant such as the refrain hymn Inventor rutili found in Sarum books.
The chief glory of the ceremony was the Exultet chant, which now followed, so called from its first words, 'Exultet iam angelica turba celorum', and also known as the Praeconium paschale (Proclamation of Easter) or Benedictio cerei (Blessing of the Candle). Its text bears some resemblance to a preface at mass, while musically it is an immensely long and elaborate intonation, using several different formulas.
A short form of mass ensued, starting not with the opening chants and prayers but straight away with lessons (perhaps as many as twelve) and responsorial chants. In number they differed from one use to another, but the sequence: lesson—silent prayer—collect—lesson, and so on, with a tract (sometimes called here a canticle) replacing a lesson at intervals, seems generally to have been followed.
Attention then turned to the font, where baptism was to take place. The procession there and back was usually accompanied with the chanting of litanies, or a canticle, or a refrain hymn such as Rex sanctorum angelonim, found in Sarum use for the return procession. An elaborate ritual of blessing the water of baptism was carried out, including the dipping of the paschal candle three times into the water, symbolizing the descent of the Holy Spirit into it, and the mingling with it of baptismal oil. Although the prayer of blessing was textually like the Exultet, no special chant was sung.
The sacrament of baptism, with its solemn questions and promises, required no musical support. Those baptized, sometimes referred to henceforth as 'neophytes', were now ready to receive their first communion.
The mass which was then celebrated had no introit chant, for priest and clergy were already present. The Kyrie formed the natural conclusion to the processional litany, the Gloria was sung, and between the lessons the Alleluia V. Confiternini Domino quoniam bonus and the tract (or canticle) Laudate Dominum. There was no offertory chant or Agnus Dei or communion chant. As on the previous days, Vespers
9. The 'Feast of Fools' 39
intervened before the final formularies of mass, having but a single psalm, Ps. 116 Laudate Dominum, with antiphon Alleluia. There was no chapter, hymn, or versicle, and after the Magnificat, the postcommunion prayer of mass and Ite missa est were sung.
A short Compline, lacking hymn, chapter, and short responsory, was then sung, (vi) Easter Sunday
In the early Middle Ages the Night Office had but one nocturn, monastic use following (or reverting to) the secular form with three psalms and three lessons, with their respective antiphons and responsories. Its last responsory, commonly Dum (Cum) transisset Sabbatum V. Et valde mane, frequently introduced the enactment of the famous Visitatio sepulchri ceremony (discussed in 11.25). There was sometimes a special ceremony to discover and display a cross and/or host buried on Good Friday (Young 1933, chs. IV-V), which might take place before or after the Night Office. As for all such ceremonies, chants specially selected or composed were performed, and occasionally they would amount to a regular Visitatio representation.
The other services of the day proceeded more or less according to their common form, although affected in character by the festive and joyful nature of the day.
1.9. THE'FEAST OF FOOLS'AND RELATED CUSTOMS
Chevalier 1894; Chambers 1903; Villetard 1907; Wagner 1931; Whitehill et at. 1944; Arlt 1970; Hohler 1972; Arlt 1978.
At this point I interrupt the generalized account of liturgical usages in order to describe some customs which, in contrast to those so far treated, were by no means universally practised. Their outstanding musical importance necessitates this excursus. The customs are those associated with the so-called 'Feast of Fools'.
While the most important part of the year from the church's point of view is undoubtedly Easter, in the later Middle Ages the liturgy of the Christmas season was often enlivened by practices which are of special musical interest. Because of the licence involved, such a festal liturgy was often known as a 'Feast of Fools'. It was most often performed on New Year's Day, the Feast of Circumcision. While such liturgies are based on the normal services of the daily round, they are ornamented by numerous extra items, particularly the rhyming Latin songs known as conductus or versus. Five principal sources for a Feast of Fools liturgy, or something closely comparable, have survived: from Sens, Beauvais, Laon, Santiago, and Le Puy.
The official in a medieval cathedral who had charge of matters concerning the performance of the liturgy was usually the precentor (derived from prae + cantor, literally 'foremost singer'), whose responsibilities included the conduct and welfare of
40 /. Plainchant in the Liturgy
the choir and the management of chant-books. There was naturally variety from place to place as to the exact delegation of authority in these and other matters (see the survey in Kathleen Edwards, 1967); and in many institutions the practical rehearsal and execution of music was in the charge of a succentor (from sub 4- cantor). Within the choir itself, leading singers were detailed to sing solo parts of those chants which required them. The number of leaders (or rectores, 'rulers', as they were sometimes called) reflected the solemnity of the feast, more soloists (up to four) indicating the most important rank of feast (see Harrison 1963, 51 ff. and 104 ff. for useful information, again about English arrangements).
On certain days of the year, however, the normal hierarchy of dignitaries and officials in church might be reversed. On St Nicholas's Day (6 December) or on Holy Innocents' Day (28 December) some churches might elect a 'Boy Bishop' to preside over the liturgy of the day, and the boy choristers directed the services, detailing others to perform all the duties which boys were accustomed to carry out, such as intoning the first lesson of the Night Office, holding the book for the officiating priest, bearing crosses, censers, and other ceremonial equipment (Chambers 1903, ch. XV).
A similar transference of authority might take place on other days of the Christmas season, and in some cathedrals (mostly French) it was the custom for the deacons to order the services of St Stephen's Day (26 December—St Stephen was a deacon of the early church), priests on St John's Day (27 December), and the subdeacons on 1 January, the Octave of the Nativity or Feast of the Circumcision (when Christ was circumcised), known also, because of the extravagant licence with which the liturgy was performed, as the Feast of Fools, the Feast of the Ass, and so on. Another name was the festům bacali, after the baculurn, the rod of office normally held by the precentor but on this day wielded by the subdeacon. Furthermore, one oft-quoted writer of the late twelfth-century, John Beleth (see Chambers 1903, i. 275; Arlt 1970, Darstellungsband, 40), speaks also of tripudia (literally 'dances') performed by the deacons, priests, choirboys, and subdeacons respectively, on the four above-mentioned days after Christmas. The subdeacons' feast might also be on Epiphany (6 January) or its octave.
Apart from the strictures issued by ecclesiastical authorities against it (see Chambers 1903, ch. XIII), we have little other evidence about dancing (and other horseplay) which marked the Feast of Fools, although a Sens precentor's book of the fourteenth century, Sens, Bibliothěque Municipale 6, notates a melisma in a processional responsory for the feast of the Invention of St Stephen (3 August— St Stephen was the patron saint of Sens cathedral) with the instruction that the precentor shall dance to it (Chailley 1949). There do survive, however, five principal sources of music for Feast of Fools liturgies, or a representative portion of it, as well as many minor sources. One is from Sens, already mentioned, and later Sens tradition ascribed its compilation to Pierre of Corbeil, archbishop of Sens. Paris records also attribute to Pierre the version of the Circumcision liturgy in use in Paris, though no copy of a special festal office has survived from Paris. It is known that the conduct of the feast at the cathedral of Paris (that is, of Notre-Dame) was reformed in order to
9. The 'Feast of Fools' 41
control some of its abuses in the closing years of the twelfth century. It is therefore possible that what we see in the copy of the Sens office, is, as it were, a 'respectable' version of the Feast of Fools.
The Sens copy of a Circumcision office, Sens, Bibliotheque Municipale 46, dates from the early thirteenth century (ed. Villetard 1907). This is also the date of a similar office from Beauvais, with a somewhat richer provision of special festal material, London, British Library, Egerton 2615 (ed. Arlt 1970—the manuscript also contains the famous Play of Daniel, probably for performance on the same day, and has been bound with some polyphonic compositions; some of these were for performance during the Circumcision liturgy; some of them were composed in Paris). Between them, these two sources may indicate something of the (lost) Parisian Circumcision liturgy. A late twelfth-century source from Laon cathedral, Laon, Bibliotheque Municipale 263, contains much similar material rubricated for feasts of the Christmas period, concentrating especially on the feast of the Epiphany.
Outside North France the sources are sparser. This is not to say that manuscripts containing music typical of the Feast of Fools are lacking—such songs may be found in considerable quantities in, for example, the Norman-Sicilian tropers Madrid, Biblioteca Nacional 288, 289, and 19421; the Aquitanian tropers and song-books Paris, Bibliotheque Nationale, lat. 1139, 3549, and 3719; and the song-books Cambridge, University Library Ff. i. 17 (English?) and London, British Library, Add. 36881; all these are twelfth-century sources. But none of them sets out the material in systematic liturgical order or indicates the succession of items which will make up the complete festal liturgy. This occurs, outside North France, only in the well-known 'Jacobus' or 'Codex Calixtinus' of Santiago de Compostela (twelfth century; for St James Day, 25 July; ed. Wagner 1931, and Whitehill et al. 1944; see also Hohler 1972), and the Circumcision office of Le Puy (late medieval copies and supplements of an original going back, like the material in the other sources, to the twelfth century; text ed. Chevalier 1894; see also Arlt 1978).
All these sources contain chants and other material for the secular cursus of the office. The most important characteristic they share is the inclusion of conductus or versus, Latin songs, at various points in the services, either as replacements for a regular liturgical form or, more frequently, as extra festal items. Troped and farsed items are abundant, sequences replace hymns on several occasions. Among many services worthy of fuller description, the Epiphany completorium infinitum (endless Compline) of Laon should be mentioned, where, after a particularly lavish provision of conductus, the direction is given to conclude the office: 'tot Benedicamus quot novit quisque canamus' ('Let us sing all the Benedicamus [songs] we know') (see Arlt 1970, Darstellungsband, 220 ff.; also Steiner, 'Compline', NG). Polyphony is occasionally rubricked in the Beauvais office (the manuscript is now bound together with some polyphony), and appears in supplements to 'Jacobus' and the Le Puy office.
As an indication of the way in which special material was grafted on to the normal liturgy, Table 1.9.1. sets out the course of the last service of the Beauvais office,
42 /. Plainchant in the Liturgy
Second Vespers (Arlt 1970, Editionsband, 139 ff). On the left are placed the regular liturgical forms, on the right the songs, tropes, and so on.
Despite the licence which undoubtedly marked the celebration of the Feast of Fools, these offices do not, in their sober manifestation on medieval parchment or in modern edition, seem unduly disorderly. To those, however, who were accustomed to the regular unfolding of the liturgy, day by day and week by week, such efflorescences of markedly un-Gregorian chant (their style is discussed 11.24) could not fail to have made a vivid, even disconcerting impression. Perhaps if we try to imagine correspondingly exotic irregularity in the ceremonial action and the visual presentation of the liturgy, we may be better able to appreciate the reasons for the episcopal displeasure evidently aroused from time to time. When the thirteenth-century Bayeux ordinal Bayeux, Bibliotheque Municipale 121 says that the four special Christmas feasts are to be celebrated 'quam sollennius possunt' (ed. Chevalier 1902, 65, cited by Arlt 1970, 48; 1978, 7) the term used, 'as solemnly as they [the clergy] are able', should not be interpreted in the modern sense as 'serious, sober'. The adjective sollemnis is conventional in ordinals as an indication of the character of the liturgies of feast-days, and should rather convey the idea of ritual splendour, richness, and variety; and, on the Feast of Fools, outbursts of extravagant and exotic music as well.
1.10. OTHER SERVICES: BAPTISM, CONFIRMATION, ORDINATION, CORONATION, MARRIAGE, BURIAL,
DEDICATION
The account of the liturgy given in the preceding chapters has concerned itself with the regular cycle of services as performed day by day and year by year, with some added information about the special forms those services might take on certain days of unusual significance. The present chapter contains a few brief remarks on ceremonies of a different character, largely independent of the normal liturgical cycle, performed to sanctify a special occasion concerning a person, a church, or an ecclesiastical object of some sort. Many of them are episcopal functions: that is, the services are carried out by a bishop, and the prayers and chants sung are mostly found in the pontifical, the bishop's service-book.
Not all these services require the performance of music as an essential element. The role of plainchant varies according to how much ceremonial action is involved; where a procession of some sort takes place, as in the service for the dedication of a church, then it is not surprising to find a large provision of ceremonial antiphons. Some of the most important of all Christian rites, however, have little musical support.
Among those to be mentioned here are the services marking the sacraments. A sacrament was defined by Thomas Aquinas as 'the sign of a sacred thing in so far as it sanctifies men', and there are usually said to be seven: the eucharist, baptism, confirmation, penance, ordination, matrimony, and extreme unction (the anointing of
Tabic 1.9.1. Beauvais Circumcision office, First Vespers
Arlt Regular item No.
Replacement item
Addition
10
11
12
13
14
15
17
19
YersicleDťH.v in adiutoriitm
['Alleluia' acclamation after 'Amen']
A. F.cce an mat I io vobis Ps. Dixit dominus ('all antiphons are to be started with "falseto"')
A. Hodie intacta virgo Ps. Conjitebor
A. Virgo verba concepit Ps. Beat its vir
A. Virgo hodie fidelis Ps. De profundis
A. Sesciensmater Ps. Memento
Chapter Popidus gent tum
R. Confirmation est cor virginis (no V. given)
16 [hymn]
A. Qui de terra
Magnificat [Benedicamus versicle]
Alleluia \ eni sonde Spiritus, 'or Veni doctorprevie with Organum'*
sequence l^tabundus exultet fidelis chorus
verse Cordepatrisgenitus . . . Benedicamus domino, Superomnes alias ... referam us gra tias
Verse Lux hodie Conductus 'when the ass is brought along', Onentis partibus
verse sung behind the altar,
I lac est clara dies
verse sung before the altar,
Salve festa dies
prose sung by full choir,
[atemur gaitdiis
verse Christus mcuiens 'in the
pulpit with Organum'*
'in the pulpit with organum'*, with supplementary trope verses, one series set out as a prosula performed alternating texted by leaders and melismatically by choir
verse Ave virgo speciosa
*organum present in another part of manuscript.
44 /. Plainchant in the Liturgy
the sick). The eucharist has of course been the subject of extended discussion above. The baptismal rite in its ancient place during the mass of Easter Eve has also been described briefly, and it was stated that processional music was sung during the progress to the font and on the return. A similar ceremony took place when the baptism was held on other days. It was one that had grown out of the early practice of adult baptism, whereas infant baptism was much more common in the Middle Ages, as now. The ritual surrounding this was considerably shorter. Confirmation was in many respects the same as baptism in its ritual requirements, and in general took place alongside baptism on Easter Eve. Unlike baptism, it remained a service which only a bishop could perform.
Public rites of penance consisted largely of formulas and prayers pronounced by the bishop, with accompanying actions. They centred upon the ceremony when the penance was laid upon those who had confessed their sins, on Ash Wednesday, and the complementary rite when the penitents were reconciled on Maundy Thursday. During Lent intercessions for the penitents were said, and it is possible that the tract Domine non secundum V. Domine ne memineris V. Adiuva nos Deus, sung first on Ash Wednesday and then on each Monday, Wednesday, and Friday during Lent, is a relic of such prayers.
The ordination rites of the church—whereby a candidate is admitted to one of the so-called 'major' orders of bishop, priest, deacon, or subdeacon, or one of the 'minor' orders of acolyte, exorcist, reader, or door-keeper, or to the monastic profession—are largely matters of prayers, exorcisms, and other intoned forms delivered by the bishop, and chants to cover an action are rarely necessary. The tonsuring of clergy would be accompanied by the singing of a psalm with its antiphon. The more important the rank being conferred, the more elaborate the ceremony, with perhaps a full preface prayer, a series of interrogations, blessings of the badges of office (bishop's ring and staff), and so on.
Since kings ruled by divine right, their coronation was a full-scale liturgical ceremony, with considerably more opportunity for ritual action, and hence music, than most special ceremonies of this type. Processional antiphons (Firmetur manus tua) and litanies would accompany the entrance of the monarch, moments such as the anointing with oil would have their ceremonial antiphon (Unxerunt Salomonem Sadoch sacerdos) and the singing of the Te Deum was a constant feature of a ceremony which naturally took several forms at diverse times and places.
In contrast to the marriage service, which in its normal form has no claims to attention as far as plainchant is concerned, the ceremonial surrounding a funeral is impressively elaborate. The Mass for the Dead, popularly known as the Requiem after the opening words of its introit, was something which might be sung, or at least said, at any time of the year as a votive commemoration. And in the later Middle Ages the performance of a complete Office for the Dead was also not uncommon, especially in monastic houses. The actual occasion of a person's death and burial might also call for the performance of these services. Before that, however, there would have been a procession to bring the corpse to the church for the final exequies, accompanied by
/ 0, Other Services
45
the singing of penitential psalms with appropriate antiphons and the addition of the versicle Requiem eternam. On entering the church the responsory Subvenite sancti Dei V. Suscipiat te V. Requiem eternam was sung.
After mass (and any preceding services of the day) the absolution ceremony begins with the responsory Libera me Domine, with several verses. Versicles and prayers then lead to the procession to the grave, during which the ceremonial antiphon In paradisum is sung, with psalms as necessary. The grave itself may be blessed, and it and the body are sprinkled with holy water and censed. The Benedictus canticle, with antiphon Ego sum resurrectio, is the last important musical item, for the rest of the versicles and prayers, and psalms for the return to the church, are mostly said.
Finally, perhaps the most extended use of ceremonial antiphons is to be found in the ceremony for the dedication or consecration of a church, with its several perambulations and blessings, culminating, like several of these special services, in mass. Again, several forms were known, but the following are the usual features. The clergy assemble to the singing of psalms and versicles, and an antiphon (Zachee festinans descende) is sung at the entrance to the church. All then walk round the church three times, singing the major litany. Psalm 23 Domini est terra is also sung, with verse 7 as its antiphon Tollite portas, and this and the other final verses provide the text for the entrance ritual. The bishop knocks on the door of the church, and sings 'Lift up your heads O ye gates and be ye lift up ye everlasting doors: and the king of glory shall come in!' From inside comes the question: 'Who is the king of glory?' The bishop walks round the church again, and the same sequence is repeated. After a third perambulation bishop and clergy finally sing the answer: 'Even the Lord of hosts, he is the king of glory.' A litany may accompany the approach to the altar, and then the bishop traces with his staff the letters of the Greek and Latin alphabets across the corners of the church, to the singing of ceremonial antiphons (see Plate 10). The sanctifying of the building requires various substances—salt, ashes, water, hyssop—and their exorcizing, blessing, and sprinkling or other administration, in various parts of the church, which also calls for the singing of antiphons, or complete psalms with antiphon. The church will have holy relics of its patronal saint, and their installation in the high altar, or wherever they are to reside, requires further censing and other actions, prayers, and chants. When the ceremony is complete, the mass for the dedication, beginning with introit Terribilis est locus iste, is celebrated. This mass will be sung each year on the same day, together with a full office.
II
Chant Genres
II.1. INTRODUCTION
In this chapter I describe the chief genres, forms and styles, of chant in the Roman liturgy.
The main genres are arranged according to their musical type. I start with the simple recitation formulas for prayers and lessons (11.2), and then go on to psalm tones and the more elaborate recitations for invitatory and responsory verses (II.3). After that I discuss the much more complex chants which retain elements of recitation patterns by marking syntactical units with typical melodic gestures: responsories, graduals, and tracts (II.4-5). Short responsories are then considered (II.6). The next sections are about chants which generally use free-ranging melodies of varying complexity: first office antiphons (II.7), and then invitatory antiphons, processional antiphons, Marian antiphons, and other special types (11.8-10). After that I consider the 'antiphons' of mass: introits and communions (11.11-12) and offertories (11.13). Alleluias follow in 11.14, then there is a switch to quite different genres, hymns (11.15) and chants for the ordinary of mass (11.16-21). The next sections treat various later medieval chants: sequences (11.22), the multitude of different types of trope (11.23), and liturgical Latin songs of various kinds, often with texts in verse (11.24). So-called liturgical dramas (11.25) and offices with verse texts (11.26) conclude the chapter, after which I have appended an excursus about some literary aspects of poetic texts (11.27), intended to be of assistance to the non-specialist.
Describing the musical make-up of medieval chants involves consideration of many features, not all of which can be given their due in the space available here. Some of the most discussed heretofore relate to form, in the sense of the layout of a melody: repeat structures, recitation patterns such as those of office psalmody which can sometimes be found in other chants, deployment of standard musical phrases for the delivery of varying texts. The function of chant as text declamation is fundamentally important, and has also often been treated, though the 'ground rules' for all genres of chant have not yet been defined. (Approaches of these types may be found in, for example, Johner 1953, Jammers 1965, 'Choral'.) Although the modality of chants has also been much discussed, some of their tonal aspects are less well understood: for example, the identity and role of structurally important and less important notes
2. Recitation Fonnulas
47
(which may not necessarily be directly related to the final). Nowacki (1977) has recently suggested ways of coping with questions of underlying structure and its embellishment. I have also found useful the distinctions employed by Hansen (1979) in what is to date the only comprehensive attempt at a tonal analysis of the complete repertory of proper of mass chants (as recorded in Montpellier H. 159). Hansen regards as 'secondary' those notes only approached and quitted by step, whereas 'main tones' are those approached and quitted by a leap. (Hansen employs supplementary criteria as well.) On this basis most of the chants can be divided into pentatonic or predominantly pentatonic melodies, and melodies where interlocking chains of thirds are structurally important. These distinctions override classifications according to liturgical function, mode, final, ambitus, and so on, so that chants of different genres and mode may have the same (pentatonic or tertian) tonal backbone. (Other promising computer-assisted analyses have been carried out, on very different repertories, by Halperin 1986 and Binford-Walsh 1990.)
It is nevertheless safe to say that the musical analysis of plainchant is in many respects still at a preliminary stage. In what follows different examples are therefore discussed in different ways, emphasizing now one possible approach, now another.
11.2. RECITATION FORMULAS FOR PRAYERS AND LESSONS
(i) General
(ii) Prayers
(iii) Lessons
Leon Robert 1963; Huglo, 'Epistle', 'Exultet', 'Gospel', 'Litany', NG; Stablein, 'EpisteP, 'Evangelium', 'Exultet', 'Litanei', 'Passion', 'Pater noster', 'Prafation', MGG; Steiner, 'Lord's Prayer', NG.
(i) General
A great deal of the music performed during office and mass is extremely simple in character, being for the most part (sometimes entirely) a monotone, with melodic nuances at significant points in the text: usually at the starts and ends of phrases, and particularly at final cadences.
Even the simplest versicles and responses often illustrate some of the general principles by which recitations usually work. For the most part the versicle is sung to one repeated note, which may of course be sung as many times as required by the length of the text. Cadences may be marked by a fall in the voice, or a more elaborate gesture. Occasionally the voice rises from a lower pitch in order to attain the main reciting note, as it is usually called, or 'tenor'. This is what happens at the beginning of the ubiquitous office versicle Dominus vobiscum, quoted in its Sarum form in Ex. II.2.1. Here there is a double nuance at the cadence. It introduces a collect, sung completely monotone until the simplest of falling final cadences. Then Dominus
48 II. Chant Genres
Ex. II.2.1. Versicles and responses (AS 4-5; Frere 1898-1901, pt.II, 210 and lxxvj)
y
Priest        Dominus uobiscum.
f    *  , « ^
Choir Et cum spiritu tu-o.
y
Priest O-remus.     Excita quesumus ... per omni-a secula seculorum.
y
Choir A-men. (Priest and Choir 'Dominus' and 'Et cum' as above)
a m m
f0 •   . é
y
Boy Benedicamus domino.
-•—•-
y
Choir De-o eracias.
vobiscum and its response are repeated, followed by Bene.dicam.us Domino and its response: these take a different reciting note and have a lightly ornamented cadence.
In the melodic formulas used for singing prayers and lessons, a hierarchy of major and minor divisions in the text is often observed, corresponding roughly to our full stops, semicolons, and so on. Interrogative sentences may be signalled with their own special cadence. Some types of chant—one thinks immediately of the verses of the responsories of the Night Office (see 11.3)—are sung to much more elaborate and musically sophisticated recitation formulas. Even these, however, retain one important feature of recitation technique: the length of the text to be sung is reflected in the number of times the reciting note is repeated.
In some of the examples discussed below, it will be noticed that an appreciation of the accentuation of particular words is important for the correct performance of the chant. There will be a departure from the reciting note at, say, the penultimate accented syllable, in order to mark a phrase-end. Sometimes accented syllables will be marked by a pitch higher than that of the reciting note—or better, there is a double reciting note, the higher note for accented syllables and the lower note for unaccented ones. There seems little doubt that these refinements of delivery were not universally practised, and that the cadential or other formula would often have been sung at the appropriate moment without respect for text accent. Similarly, most recitation
2. Recitation Formulas
49
formulas proceed without higher pitches for accented syllables. Thus the melodic inflexions for phrase-ends form a sort of musical punctuation. They mark syntactical and sense units, but do not necessarily reflect anything of the natural rise and fall of the human voice.
Although there is no doubt that the practice of reciting prayers and lessons in this way is an old one, medieval copies of recitation formulas are generally scarce, since they were known well enough for codification to have been unnecessary. Furthermore, most surviving music-books were made by and for trained musicians, and recitations of this type were not among their duties. On the other hand, more ornate and unusual lesson tones, such as those for the genealogies, are among the very earliest chants found with musical notation.
(ii) Prayers
The most important prayers to be recited in the liturgy are those of mass: the collect, secret, and postcommunion prayers, and the prayers of the eucharistic rite. The first three are relatively brief, and may be classed with many other occasional prayers for ritual actions, such as sprinkling holy water and blessings of various kinds. The Liber usualis gives a number of tones, plain or more varied to suit less or more important liturgical occasions. (There is as yet no convenient way of finding out how varied medieval practice was.)
The Preface prayer of mass (also sung, with a different type of conclusion, at special ceremonies of consecration of various kinds) is a much longer item, which begins with versicles and responses before launching into a long text varying with the liturgical occasion. More than one medieval tone for the prayer is known, but the one given in Ex. II.2.2 is typical. In it most clauses are sung to one formula, final clauses to another. As often in such formulas, the singer must know where the two final accented syllables fall. For the non-final clauses, he makes his inflections as follows: on the syllable after the penultimate accented syllable he quits the reciting note (c) for b \ on the last accented syllable he goes even lower, so that a sub-tonal ending on b can be made (ab b). Final clauses have a slightly more ornate formula, depending only on the final accented syllable. Three syllables earlier the singer quits the reciting note to introduce a turning cadence which will settle on a (aG, Ga b ab a). Practically without exception, end syllables in these texts will be unaccented. If more than one syllable follows the accented one, then the a may simply be repeated {ab a a).
One prayer whose text has the character of a preface but whose music is much more elaborate is that for the consecration of the Paschal Candle on Easter Eve, the chant known from its opening word as the Exultet. It begins with a section not found in other prefaces, then come the versicles and responses, then the main part of the prayer. The first section is sung to a melody of four phrases repeated five times. The main part often uses the normal preface formulas, but introduces far more frequent cadences, and other more elaborate inflections. The degree of elaboration varies from one tradition to another (see Huglo, 'Exultet', NG for a summary of available
50
//. Chant Genres
Ex. II.2.2. From the Blessing of the Font on Easter Eve (Oxford, Bodl. Lib. e Mus. 126, fo. 41r)
1
Per omni-a secula seculorum.
A-men.
* it wr t-
Dominus uobiscum.
Et cum spiritu tu-o.
Sursum corda.
Habemus ad dominum.
0   0   0 0—*—0-
•  m ŕ, p '  £•  m       fr«  •    ém £0
Gratias a-gamus domino De-o nostro.     Dignum et iustum est.
>   ' « fit <
Vere dignum et iustum est equum et salutare
*—»-M-0 0
nos tibi semper et u-bique gratias a-gere
* *  0 +
domine sancte pater omnipotens e-terne Deus
t IT
0  m J0—0-
qui inuisibili potenci-a sacramentorum tu-orum mirabiliter o-peraris effectu
0  * ml A 0 ■>
*' m
(two more phrases, then)     e-tiam ad nostras preces aures tu-e pi-etatis inclinas.
Deus ... (final clause)   Ut omnis homo hoc sacramentum regeneracionis
ingressus in uere innocentie
T   f    *      « t
in nouam infanciam renascatur.
2. Recitation Formulas
51
information and bibliography; also Stablein, 'Exultet', MGG for a survey of melodies). The start of the chant is given in Ex. 11.2.3 in two versions, from Paris and Salisbury respectively. Although both sources are from the thirteenth century, and not widely separated geographically, there are considerable differences between their versions. The Paris version is overall a fifth lower than the English one, but that is a relatively superficial distinction. More important is the simplicity of the Paris version compared to Salisbury. The contrast in elaboration is already apparent in the first part of the melody, the four-phrase unit which will be repeated several times. Then, when the recitation formula is employed in the main section of the prayer, the Salisbury version almost conceals the reciting note(s) under a wealth of melodic detail. The Paris version has F as its reciting note, with E for secondary cadences and D for main ones. For Salisbury, the possibility of a repeated c (or b) is hardly considered, and freely ranging phrases between G and d are often given. A simple fall cb becomes cbbcb; ab becomes acbbc. The assumption here is that the Salisbury version notates an ornate elaboration of some simple earlier pattern, an elaboration that has been developed in performance over the years. The solemnity of the occasion has resulted in a far higher degree of surface decoration than was normally practicable for such a prayer.
The prayers mentioned earlier were sung by the officiating priest, or celebrant, at mass. The Exultet differed from other prefaces in being sung usually by the deacon, hence the possibility of musical sophistication.
One other type of prayer, in which the whole assembly takes part, rather than having the priest speak for all, is the litany. Litanies were frequently sung in procession, for example during the ceremonies for consecrating a church. They were performed during the progress to and from the font for the solemn baptism ritual of Easter Eve (shortly after the Blessing of the Candle in which the Exultet is sung), and again, if that were the custom, in the echo of that service on Whitsun Eve. On the Rogation Days (the three weekdays before Ascension Day, which always falls on a Thursday) litanies were sung in procession before mass: this series was sometimes known as the Lesser litanies, while the Greater litanies were those sung in procession on St Mark's Day, 25 April. The normal form of such litanies is that of a long series of short verses, with a refrain. Invocation and refrain were sung first by soloists and then by the whole choir or assembly. Normally there will be introductory verses, often including the words 'Kyrie eleison, Christe eleison', then a series of saints (including the patron saints of the church) will be invoked (response 'Ora pro nobis': 'Pray for us'). Safety from ills and misfortunes will be requested ('Libera nos domine': 'Save us O Lord') and the saving events of Christ's life and ministry recalled; finally, Christ's aid for the church and people is requested ('Te rogamus audi nos': 'Hear us, we beseech thee'). The final versicles may include an Agnus Dei formulary and Kyries once again. Each section, with its refrain, is sung to a different musical formula. The amount of repetition might vary according to the custom of the church. The Easter Eve litanies were sometimes arranged in three series, called respectively by some such name as the Sevenfold (Septiformis), Fivefold (Quinquepartita) and
52
//. Chant Genres
Ex. II.2.3. From the Exultet (left: Paris, Bibl. Nat. lat. 1112, fo. 96r; right: Graduate Sarishuriense, 105)
Paris Bibliothěque Nationale lat.1112 f.96  Graduate Sarisburiense 105
m  ft *    «    «  * * « « *H~i a "Ü a
f
E-xul-tet iam angelica turba celorum        E-xultet iam angelica turba ce-lo-rum
e-xultent diuina misteri-a
e-xultent diui-na miste-ri-a
et pro tanti regis uictori-a
et pro tan-ti re-gis uic-to-ri-a
tuba intonet salutaris.
tu-ba   in-tonet sa-lu - ta   - ris.
I
■t. * » ■ t
Vere qui-a dignum et iustum est
Ve - re qui-a dignum   et     iustum est
• ■ ■ «    • «    *   »     i C a
•* P •---
inuisibilem Deum patrem omnipotentem.     inuisibilem Deum omnipotentem patrem.
É
,  f. P f. ^
«—#-
O ue-re be-a - ta nox
O
be
a-ta nox
• * #   f> g» • *
que expoli-a-uit E-gypti-os
que expoli-a-uit E-gypti-os
P
•   *^ JÍ*-•-P—r—
di-ta-uit He-bre-os.
di-ta-uit Hebre-os.
t f*'f.  fa faP *
Nox in qua terrenis celesti-a   iunguntur.   Nox in qua terrenis celesti-a      iun-gun - tur.
O-ramus er-go   te do-mine
O-
-ramus te
do-mine ...
2. Recitation Formulas
53
Threefold Litany (Tripartita Letania) according to the number of those leading the singing.
Ex. 11.2.4 gives the opening of the Rogation Tuesday litany from a fifteenth-century York processional. As with most sources, the alternation and repetition scheme is not specified. Apart from the 'Ora pro nobis' phrase, which will presumably recur after each saint has been invoked, there are four refrain verses which, at first at least, are used in turn after each of the invocations. It is not clear whether or not this pattern is to continue throughout the subsequent series, which is not notated. (For a variety of other litany chants, many not using recitation formulas, see Stablein,
Though the chanting of the saints' names or other invocations clearly involves a reciting note (c in Ex. 11.2.4), the litanies may be relatively highly inflected. The number of notes other than the reciting note, combined with a relatively small
Ex. 11.2.4. From a litany on Rogation Tuesday (Oxford, Bod. Lib. e Mus. 126, fo. 58v)
Litanei', MGG.)
Ky-rie   - ley-son,
Christe - ley-son.
Domi-ne   mise-re-re,     Christe miserere.
Miserere nobis pi-e   rex do-mi-ne Ihe-su Christe.
Chri - ste au-di nos.
Sancta Ma-ri-a o - ra pro no-bis.
(Kyrie...Christe... as above)
Sancta De-i   ge-nitrix o - ra pro no-bis.
(Domine miserere as above)
Sancte Ga-bri-el o - ra   pro no-bis.
(Miserere nobis as above)
Omnis chorus angelo-rum o - ret pro no-bis.
(Christe audi nos as above)
54
//. Chant Genres
number of syllables, means that the reciting note is not repeated very much. The same is true of most melodies for the Lord's Prayer (see Leon Robert 1963, 127, and Stablein, 'Pater noster', MGG, for various melodies). A widely known melody is transcribed from the fourteenth-century Cluniac breviary-missal of Lewes, Cambridge, Fitzwilliam Museum 369 in Ex. 11.2.5. For the first part of the chant, the reciting note (the only one repeated) is b, but from 'sicut et nos' a becomes more prominent. The cadences are either on G (less important) or a (full stops).
Ex. II.2.5. Pater noster (Cambridge, Fitzwilliam Museum 369, fo. 2417297v)
		
j£__m     r-__              m_^__ *__		^ • * —
fA\     •  *-    •   •      •     •    •    ■ -	m   -  * •               0    #• «	• —w  w   yw--~-
		
Pater noster qui es in celis      sanctifi-cetur   nomen tu-um.       Adueniet regnum tuum.
				
A        a     a.       a a				*      ä                    a            ^ _
OT)    #    *-t-M-Jt-• ,	—* •—*  * »0	—•*——	m	
- Fi-at uoluntas tu-a	sicut in celo	et in	terra.	Panem nostram coti-di-a-num
da nobis hodi-e.      Et dimitte nobis debita nostra      sicut et nos dimittimus debi-to-
ribus nostris.      Et ne nos inducas in temptaci-o-nem.      Sed libera nos a malo. Amen, (iii) Lessons
The same principles of recitation as in those for prayers, and some of the same variety of musical elaboration, may be seen in the tones for chanting lessons. The tone for the short chapter of the office hours given in modern Roman books is a monotone without any initial rise, with flex, metrum, and full-stop figures. The longer lessons of the Night Office are sung in the same way, with the full stop coming at the lower fifth, for example:
reciting note c, full-stop figure c-G-d-F, or reciting note a, full-stop figure d-F-G-D
The epistle sung at mass, with reciting note c, has the following figures in modern Roman books (for a survey of various medieval formulas, see Stablein, 'EpisteP, MGG):
metrum: a-c-b-b-c
interrogation: reciting tone b, a—b—bc
full-stop: cd-b-d-b
The end of the whole lesson is signalled by an accented ac inflection, then a fall to b for the reciting note and a final close bc-c. The gospel tone given as 'ancient' in
2. Recitation Formulas
55
modern Roman books (the other appears not to be older than the sixteenth century, being similar to the prescriptions of Guidetti's Directorium chori of 1582) is similar to this (see Stäblein, 'Evangelium', MGG for further medieval tones).
At various times and places, and even to a certain extent in modern service-books, more elaborate tones have been used for lessons on particularly solemn occasions in the year. The best-known instances are the tones used for the lessons of the first nocturn of the Night Office during the triduum, the three days before Easter Sunday. (See Prado 1934 for some remarkable Spanish examples.) Gospels at Christmas and Easter were occasionally also sung to a more elaborate tone (see Huglo, 'Gospel', NG), as for example the one from the Moosburg gradual, Munich, Universitätsbibliothek 2° 156 given in Ex. II.2.6. Here the hierarchy of main and secondary closes is not clear, and
Ex. 11.2.6. From the Gospel on the feast of St John the Evangelist (Munich, Univ.-Bibl. 2° 156, fo. 226r)
Domi-nus     uo - bis-cum.        Et cum spiri-tu        tu - o.
*   '  . * 9   0   f  . * *        * 0 *   0    . •••'"Im**
Sequentia sancti e-uange - li-i        secundum Jo-hannem.        Glori-a ti-bi domine.
P
.        .0      0    0 0     0     90m     0± . M0* 0~j
-0-M*- -0-Jf
-f---»-9-W-"-m-f-•-
In illo tern-pore        Dixit Ihe-sus Petro.        Se - que-re me.
* ' . y •     a . * *=
Conuersus Pe-trus        uidit ilium discipulum quern diligebat Ihesus sequentem.
* M 0 0 M0m
-0-*-'-«-f-0-
0   m»9 O
pr—» w   , w-
Qui et recubuit in cena super pectus e-ius   et dixit.        Domine quis est qui   tradet te?
0      '.»0   0 0   0 «** 0±
-•-■-*-•—j0-«-•-y—0 -
Hunc ergo cum uidisset Petrus        di-cit Ihesu.
0 • • • 0 y ^=
Domine hie autem quid?        Dicit e - i Ihesus.
« ' * ' . t * . » . y • • # . q . . .
Sic enim uolo manere donee ue-ni-o        quid   ad te?        Tu   me sequere.
56
//. Chant Genres
questions do not have a distinct inflexion. The reciting note is generally G, reached from an initial E or from a longer preliminary figure G-a-E—F-G, sometimes with pauses on the E. The endings are E-(F-E)-DEG-G and FGa-GE-D-E', until the conclusion of the lesson.
The intoning of the Passion—the Gospel on four days of Holy Week—was also performed in unusual ways. As far back as the ninth century copies of the texts were marked with the letters c, t, and s at the point where the evangelist, Christ, and other protagonists respectively were speaking. It appears that these letters are to be interpreted as in the series of notational aids known as significative letters (see below, IV.3), though in the later Middle Ages they were understood as labels for the actual persons. Other letters (see Stablein, 'Passion', MGG) are also to be found occasionally:
c =celeriter ('quickly'), later cronista ('evangelist') or cantor
m = mediocriter or media ('at medium pitch, medium voice') is also found t =trahere or tenere ('slow, sustained'), later written as a cross for Christ
i = iusum, inferius ('at a lower pitch') and b = bassa ('low voice') are also found s — sursum ('at a higher pitch'), later synagoga ('the people') or succentor
a = alta ('high voice') is found, and sometimes Is = levare sursum ('raise high') for the people is contrasted with Im — levare mediocriter ('raise moderately') for the disciples
The pitch-levels indicated by the letters are indeed to be found as three different reciting notes (usually F, c, and/) in copies in staff-notation, which are of course later than the oldest 'marked-up' copies of the text. In the late Middle Ages, if not earlier, different singers sang the different parts of the text, as appears from the fourteenth-century Sarum missal Parma, Biblioteca Palatina 98, where the 'parts' are marked secunda vox, Hi vox, and so on (MGG 10, Tafel 61.2, after col. 960—this is the only passion in the source; furthermore, only the parts for the singers other than the evangelist are copied in full, mere incipits for the evangelist being given; the Sarum tone is edited in Andrew Hughes 1968, 184-6). An idea of the variety of tones used, within a general adherence to the principles of recitation technique and distinction between the persons of the story, can be seen from Stablein's tables ('Passion', MGG). Particular attention was often directed towards Christ's words, for example the cry 'Eli, eli, lama sabacthani', and it is thus not surprising to see neumes entered specially above these words in the Jumieges evangeliary Rouen, Bibl. Municipale 310 (A. 293) (facsimile in MMS 2, pi. XLVII). For other extravagant passion tones, see Gollner 1975 (late Spanish sources).
Two other gospels of special ritual significance, and thus special musical character, were those which recounted the genealogy of Christ. That appearing at the start of St Matthew's Gospel was sung on Christmas Eve, and that according to St Luke on Epiphany. Monastic use, which had a gospel reading at the end of the Night Office, had the genealogies at that point in the liturgy. Gospels were not read there in secular uses as a general rule, but on these days an exception was made (they were intoned
2. Recitation Formulas
57
before rather than after the Te Deum). Usually the verses were grouped in pairs, with a different tone for each half of the unit. Occasionally more elaborate schemes are encountered in medieval sources (see Stablein, 'Evangelium', MGG). Ex. 11.2.7 gives a common tone, taken from a Reims missal of the thirteenth century. Ex. 11.2.8 has a more complex melody for the St Luke genealogy, from a book of similar date but enigmatic provenance (it has elements of both Reims and Paris usage), Assisi, Biblioteca Comunale 695. Here two melodies are sung in alternation—they can hardly be called recitations any longer, though they are of course adaptable by note-repetition for names of differing numbers of syllables. Each melody has five distinct phrases; the shorter phrases of the St Luke's text lend themselves well to this treatment.
Ex. II.2.7. From the Genealogy according to St Matthew (Reims, Bibl. Mun. 224, fo. xi/16v)
1
Liber generati-o-nis Ihesu Christi fi—1 i — i Dauid fi - li-i Abraham.
A-braham genu-it Y-sa-ac. Y-sa-ac au-tem ge-nu-it la-cob.
Iacob autem genu-it Iudam et fratres ei-us.
Iudas autem ge-nu-it Phares et Zaram de Thamar.
-—-——k—0-0-p-0-0X-m-0-
Phares autem genu-it Esron. Esron au-tem ge-nu-it A-ram. ...
-•—•—»
.. Iacob autem genuit Ioseph uirum Mari-e
P
#  ? 0p • «zte:
-0~
De qua natus est Ihesus qui uocatur Christus.
58 //. Charit Genres
Ex. II.2.8. From the Genealogy according to St Luke (Assisi, Bibl. Com. 695, fo. 37')
u -						
/                                                                                                                           im „						
	_M_			•    m "		
		# •	•              #     * •			•             •*   •« «
P • ... Qui n	fuit Dauid,	qui fuit Ihesse,	qui fuit Obeth, ___XT	qui fuit	Bo-oz,	- -a-^- qui fuit Salmon, ...
			m			
/	_ _					
	■ • *	*     »     »    #    ■ •	•        •           • a	• *	•      *■ M	
					9 w	
... Qui fuit Na-ason,   qui fuit Aminadab,   qui fuit Aram,   qui fuit Esrom,   qui fuit Pharens, ...
11. 3. TONES FOR PSALMS AND OTHER CHANTS (i) Psalms
(ii) Other Psalm Tones: The Paraptcres, Tonus peregrinus
(iii) Tones for the Canticles Magnificat and Benedktus
(iv) Tones for the Psalm Verses of Introits and Communions
(v) Tones for Responsory Verses
(vi) Tones for the Invitatorv Psalm
(vii) Benedictus es Dotnine Dens pat mm nostmm in the Saturday Mass of the Ember Weeks
(viii) Te Deum laudamus (i) Psalms
Bailey 1976, 'Accentual', 1979; Stablein, 'Psalm, B', MGG; Connolly, 'Psalm, II, AT;.
The cycle of psalms chanted daily during the office is performed to tones in a manner similar to the intonation of prayers and lessons. From an early date (the late eighth or early ninth century), perhaps under the influence of Byzantine practice (see below, 111.14 and V.4), there were eight tones in the Gregorian system. The choice of tone for the performance of a particular psalm depended upon the mode of the antiphon with which the psalm was coupled in the liturgy of the day. In the course of a year the same psalm might be sung more than fifty times, nearly always with a different antiphon, and thus to a different tone. For most tones a variety of cadences (differentiae, diffinitiones) was available, in theory at least to cater for the different pitches with which the antiphon would start on its repetition after the psalm. Medieval sources regularly differ on the number of differentiae they provide, and their assignment of particular differentiae to particular antiphons. (Sometimes even the mode of the antiphon, hence the tone to be used for the singing of the psalm, was the subject of differing opinions: see V.4.) Thus the selection of differentiae given in Ex. II.3.1 below is not the same as that in modern Vatican books.
It is important to remember that the tone may be concluded on a variety of notes, and is in no way bound to finish on the same final note as the antiphon, that is, on what is normally reckoned to be the final of the mode. It is easy to confuse 'mode' and
3. Tones for Psalms and Other Chants Ex. II.3.1. The eight psalm tones (Piacenza, Bibl. Cap. 65, fos. 264v-267v)
«  ■  ■ *«
■  ■ ■
Glori-a pa-tri et fi-li-o        et spiritu-i sancto
59
9 •
Sicut e-rat in principi-o et nunc et semper        et   in   secula seculorum a - men.
th   • • _      • «	• • •    5* *~	• • m      «• m	• ■ •    • •	■    •   M           * **
vy         • ■ •	• •*^ *•	• a •	• a •	•m    w ■
		
	---	t   1   ,         g» J=j
iJ£-*-1	1-1-" »	_——C_LL
2- i
t   ♦ i r
v - - V* •
Glori-a pa-tri et fi-li-o        et spiritu-i sancto
_   ■   ■    a    a  a  a a «^   »•     • £•
#  •  • ■ » « • • «, ^
Sicut e-rat in principi-o et nunc et semper        et in secula seculorum a-men.
3.
				
		•		
cm    » •				* •
				
Glori-a pa-tri et fili-o        et spiri-tu-i sancto
rm »   »    a   m   a m
Sicut e-rat in principi-o et nunc et sem-per        et in   secula seculo-rum a-men.
* ■ > -—»-
Glori-a pa-tri et fi-li-o        et spiri-tu-i sancto
Sicut e-rat in principi-o et nunc et semper        et in secula secu-lorum a-men.
-m—^—m—w&m-=-	i-1-1 ---7^.- ---~m—m -m-wwm--	-9—~m—3f—21—WB—~—	-m—mil. "m ^\-	—w—m—w—m—mb—— "
Q)   , u '        42 f.			| .......^ |	
60
(Ex. II.3.1. cont.)
//. Chant Genres
firr:-m—mrt—■——r—^—	—■——>—vOm—-	-m * m 9 m-
tfQ—' ** * * ' «>t *m	—' V * * " II- l-#—1	---.—.——--___*
cadence for Custodiebant testimonia
Glori-a pa-tri et fi-li-o        et spiritu-i sancto
Sicut e-rat in principi-o et   nunc et semper        et in secula seculorum a-men.
6.  (A\    , , n'  I   I   i <; t1       *   • *± frf* tu
Glori-a pa-tri et fi-li-o        et spiritu-i sancto
>    *   *   •    « M*
•    M   0* «H
Sicut e-rat in principi-o et nunc et semper        et in secula secu-lorum a-men. . jm •        *   *   • •• £•        •    • * * . m —
Glori-a pa-tri et fi-li-o        et spiritu-i sancto
Sicut e-rat in principi-o et nunc et semper        et in secula seculorum a-men.
P
• * • f
• • * • m w+
• * • »U r%   I   • •
—0-*-
* * • *•
Glori-a pa-tri et fi-li-o        et spiritu-i sancto
U                                                                                   =__m___«».-		
A. ...   .—•—«    «   •   ■  « « *o   *--"—f-*-	■   m m m    m  m t	»       »n .-
g\>    • '-	_mL___ -------	-«B—0-
Sicut e-rat in principi-o et nunc et semper        et in secula seculorum a-men.
~p-=-1			
Jt    m m m »1 -	• • 0 « • •	■ • - *i ^	
a-— ■"     •    •   mu la			
sro—                -■■ - i			I---—U
3. Tones for Psalms and Other Chants
61
'tone', and it should therefore be borne in mind that while 'mode' is an abstract quality, having to do with the tonality of a chant, a tone is a sort of chant in itself, a melodic formula capable of supporting the performance of an almost infinite variety of psalm verses.
Ex. II.3.1 gives the eight psalm tones, and their cadences, as they appear in the tonary of a thirteenth-century compendium from Piacenza (Piacenza, Biblioteca Capitolare 65). (A tonary, which usually gives a list of chants in modal order, is often a more convenient source for tones than is a full antiphoner. Few antiphoners set out tones in full one after another. Indeed, to find out what tones and endings were used in a particular antiphoner it is necessary to check through each antiphon in the whole manuscript, as was done, for example, for the Italian antiphoner published in PalMus 9.) The cadences are given as six-syllable formulas, to fit the words seculorum amen (usually abbreviated 'euouae'). At the end of the cadences for tone 4 is one for the antiphon Custodiebant testimonia, a chant which requires both b and bb and which must therefore be notated with final on a instead of E.
If one were to turn to a medieval antiphoner in order to see how these tones were used, one might well be puzzled at first. At the appropriate point in the liturgy the antiphoner will probably give, not the text of the psalm to be sung, notated in accordance with the psalm tone, but simply a text incipit and the differentia. For example, in the antiphoner Karlsruhe, Badische Landesbibliothek, Aug. perg. 60, for Vespers of Christmas we find as first antiphon Scitote quia prope (LU 365, AM 237), a mode 8 antiphon (see Ex. II.3.2). It is followed by a text incipit for Ps. 112, Laudate pueri, and a differentia for the eighth tone. The full text of the psalm would be known to the performers, the monastic choir of Petershausen on the shore of Lake Constance, where this manuscript was used. The text could be found in a liturgical psalter of the church, as one of the psalms for Vespers on a Sunday. But even there it would probably not be marked up for singing, that is, the text would not contain any indication of when the singers should leave the reciting tone and make a cadence, for each verse of the psalm as required; they would be expected to know the practice by heart.
Ex. II.3.2. Antiphon Scitote quia prope est, with psalm-tone cadence (Karlsruhe, Badische Landesbibl. Aug. perg. 60, fo. 17r)
Scito-te qui-a prope est regnum De-i     a - men dico uobis     qui-a non tardabit.
Ps. Laudate pueri.
Sometimes no psalm text incipit is given: the appropriate psalm would be ascertained from the liturgical psalter. Sometimes no cadence formula is given: the
62
//. Chant Genres
mode of the antiphon must then be decided upon, and the matching psalm tone used, with an appropriate cadence.
Modern books such as the Liber usualis, Antiphonale Romanum and Antiphonale monasticum give clear instructions on how the tone, the musical formula, is to be adapted for each verse of the psalms, whose texts are marked accordingly: the syllables where a change is to be made are given in bold or italic type. As far as the medieval practice goes, however, we are generally reliant on didactic texts, such as the Commemoratio brevis de tonis et psalmis modulandis of the ninth to tenth century. As Bailey (1976, 'Accentual') has pointed out, medieval practice cannot have been uniform. Some sources indicate that attention was paid to the accentuation of the text being sung, while others apply the cadence formulas mechanically. In Ex. II.3.3, taken from Bailey's edition of the Commemoratio brevis (1979, 52-3), the cadence has been adjusted so that the last accented syllable shall be sung to G, any further syllables to F:
Ex. II.3.3. Cadences from the Commemoratio brevis (ed. Bailey, 1979, 52-3)
vj) ^ m  • m m ^    m    »m m m  m  ^   m   • *
..... fecit dominus ..... sanctum e-ius ..... iu-sti-ti-am su-am
If text accent were to be respected, then both the median cadence and the final cadence of the doxology would require adustments:
. . . et fi-li-o . . . spi-ri-tu-i san-cto
. . . nunc et sem-per    . . . se-cu-16-rum. A-men
We can look back at Ex. 11.3.1 and see if there is a change at Fi-lio (that would be an accentual cadence) or at fi-Li-o (a mechanically applied 'cursive' cadence). If the final change comes at spi-Ri-tui, the cadence is accentual, if at spiri-TU-i it is cursive. In this case the test is met in the median cadences 2-5 and 8, but in the final cadences only of tones 5 and 7. Here Piacenza 65 always observes accentual cadences.
Tones for psalms, for the canticles, and for introit verses are given in parallel by Connolly ('Psalm, II', NG). As well as these, Stablein gave responsory verse tones ('Psalm, B', MGG). Both authors draw upon both the Vatican edition and the Commemoratio brevis.
(ii) Other Psalm Tones: the Parapteres, Tonus peregrinus
Erbachcr 1971; Bailey 1977-8; Atkinson 1982, 'Parapteres'; Atkinson, 'Parapter', HMT.
The two halves of psalm verses are sung to the same reciting note in the eight common psalm tones, and most other tones, though both responsory verses and invitatory psalm tones have a change of reciting note. A rarely used psalm tone, the so-called
3. Tones for Psalms arid Other Chants 63
'tonus peregrinus' ('wandering tone'; the Latin term is first found in the writings of German theorists of the fourteenth century), also uses two different reciting notes. It is given from the late-tenth-century Aquitanian tonary Paris, Bibliotheque Nationale, lat. 1118 in Ex.II.3.4.
Ex. II.3.4. Tonus peregrinus (Paris, Bibl. Nat. lat. 1118, fo. 113v)
• m
Si-cut e-rat in principi-o et nunc et semper        et in secula seculorum a-men.
$ *.....
Seculorum a-men.
It is not the only tone of this sort. A number of them, called 'parapteres' (paracteres, medii toni, etc., possibly from Greek para + apto, 'join alongside'), are mentioned in several early medieval treatises, from Aurelian of Reome onward. They are usually cited in conjunction with a small group of antiphons of irregular modality, 'modulating antiphons', as they have been called. The most prominent groups of antiphons are those like Nos qui vivimus, and those designated by Gevaert (1895) as 'Theme 29', which, as the tenth-century treatise De modis musicis puts it, 'are not ended in the same way as they began'. (See 11.7 for examples and discussion.) The psalm tone therefore reverses the modulation in the antiphon itself, so that there is no abrupt change of tonality.
Among the other tones, the Commemoratio brevis cites the one given in Ex. II.3.5 (the antiphon is completed from the Petershausen antiphoner). The higher second 'alleluia' of the antiphon is answered by the higher recitation on a in the psalm tone. Then the tone moves down to F, which is where the antiphon will begin when repeated after the psalm verses.
Ex. II.3.5. Antiphon with 'parapter' psalm tone (Karlsruhe, Badische Landesbibl. Aug. perg. 60, fo. 17v; Commemoratio brevis, ed. Bailey, 1979, 52-3)
Notum fecit dominus     allelui-a     salutare su-um     alle-lu - ia.
<fr t«>.........•......... . . =
Cantate domino canticum nouum     qui-a mirabili-a fecit dominus.
As Bailey and Atkinson have both suggested, in these 'irregular' tones we seem to have vestiges of a more flexible psalmodic practice not limited by, and probably anterior to, the familiar eight tones. (The notion is reinforced by a comparison with Old Roman practice: see Dyer 1989, 'Singing', and below, VIII.3.) As to the tonus
64
//. Chant Genres
peregrinus, it has been remarked that antiphons requiring it (such as Nos qui vivimus) are often assigned as antiphons for the Benedicite at Lauds, which has led Steiner (1984, 'Antiphons') to ask whether the tonus peregrinus might possibly have been a special Benedicite tone. On the other hand, the tonus peregrinus was also regularly used during Vespers for Ps. 113, In exitu Israel (on Sunday in modern secular use, Monday in monastic use); it is even possible that the name of the tone was suggested by the psalm, which speaks of Israel's departure from the land of Egypt. The text Nos qui vivimus comes from this psalm.
(iii) Tones for the Canticles Magnificat and Benedictus
Where they are to be found in medieval sources, which is rarely, they are slightly more ornate than the common psalm tones. Most have a few more two-note groups than the psalm tones, but this is not so in all cases or in all manuscripts.
(iv) Tones for the Psalm Verses of Introits and Communions
Somewhat more ornate again are the tones for singing psalm verse(s) and doxology with the introit and communion at mass. The introit tones, in contrast to the usual practice for office psalm tones, have a new intonation after the median cadence. In this they resemble the tones for responsory verses.
At this point it becomes possible to illustrate without great difficulty the way in which text and tone were joined, for a number of medieval graduals copy out in full the psalm verse to be sung. Ex. II.3.6 gives some psalm verses from introits in the Chartres cathedral gradual Provins, Bibliotheque Municipale 12. The verses begin identically, and the first two are also alike after the median cadence, despite the fact that in the second 'iusticia' would normally be accented on the second syllable: the musical figure is independent of accentuation. In the first half of the verse, the rise to d is made on an accented syllable, and that means that in the second example only one c follows, because only three syllables remain altogether, instead of the four in the first
Ex. II.3.6. Psalm verses for introits (Provins, Bibl. Mun. 12, fos. 203v, 178r, 186r)
Jf'        I'm m m	m m—v	a Cm	f m m m	m  m    m m	m_■—SjRa—^—	m\~L—*ei._
		•m •	--		- "o m	*° m*—
Benedic a-	-nima me-a	domino	et omni-a qui	intra me sunt	nomini sancto	e-ius.
¥        r"m m m	m    m m m	*    • •	m ■ •  *   -> Cm		a m m m m m£l	_n——a—f*-_
			*a m	—9-*-	-	
V In te domi	ne speraui	non confundar in e-ternum		in iustici-a tu	-a libera me et	e-ripe me.
	■   m mm*!					
Et Petrus ad se con - uersus dixit.
J. Tones for Psalms and Other Chants 65
verses. The final cadence is applied mechanically, however, so that no change is made in applying the cadence formula: contrast the accentuation of 'nomini sancto eius' and 'et eripe me':
no-mi-ni san-cto ei-us et  e - r l-pe me
The third verse is very short and the two halves of the psalm tone are elided.
The same tones appear to have been used for the communion as for the introit. The communion psalm verse was in any case frequently borrowed from the introit.
(v) Tones for Responsory Verses
Far more elaborate are the tones used for the verses of office responsories. The same principles nevertheless hold true. Ex. 11.3.7 gives the verses for the first three mode 3 responsories from the Petershausen antiphoner. Obviously, text accentuation determines where certain groups of notes will be placed. The figure cdcc is used for the first accented syllable, so that Preoccupemus begins with three unaccented cs, A solis ortu with only one, and Tollite portas with none at all. The rest of the word '(Tol)lite' has two unaccented syllables, so the figure abcaaG is split. The median cadence is the same in all three verses, but two different approaches are used: in
Ex. II.3.7. Verses of the responsories Salvatorem expectamus, Audite verbum, and Ecce virgo concipiet (Karlsruhe, Badische Landesbibl. Aug. perg. 60, fos.2v-3r)
Pre-oc-cu - pe-     -mus      fa-ci-em e-ius in con-fessi - o- -ne
et    in psalmis iubi- -le-mus   e- -i.
A   so - lis or-tu et oc-ca- -su
P
ab a-qui - lo- -ne     et    ma- -ri.
,.r«       ;a . m . . >;* . . . . ? . . . .7 m • fr* »i
Tol - li-    -te portas principes nostras et e - le-ua-mini porte   e-ter-na - les
et in - tro - i- -bit.
66
//. Chant Genres
Preoccupemus and Tollite portas there is rather a lot of text to sing, mostly on a, with a couple of liquescent neumes in Preoccupemus and with accented syllables highlighted in Tollite; a more concise figure appears in A solis ortu. The second half of each verse recitation is on c\ in Preoccupemus there is a short intonation figure. The final cadence is very ornate, stretching over five syllables (that is the whole text in Ecce virgof). The music is applied here mechanically, for the three verses have different accent patterns, but the music is always the same, without any extra single notes or splitting of neumes.
It will be noticed that different reciting notes are used in the two halves of the verse, a and c. That is usually the case in the tones for responsory verses. (Copies from various sources may be found in Stablein, 'Psalm B', MGG; Connolly, 'Psalm II', NG; Cutter, 'Responsory', NG; Processionale monasticum, and AS, the latter reprinted with a useful note in Hucke 1973.)
Most but not all tones—whether simple psalm tones or those of responsory verses—have a rising intonation and a falling cadence. Several rising cadences (the word 'cadence' is etymologically inappropriate) may be seen in Ex. 11.3.1 above. A glance at the responsory verse tones in one of the editions just cited will reveal several falling openings.
As will be discussed later (V.4), the classification according to a system of eight modes is a relatively late development in Western chant. The possibility has just been mentioned that there were more than eight psalm tones, and the same possibility exists for the tones of responsory verses. A systematic enquiry has not yet been published, but it may be mentioned that Frere noticed a second verse tone for mode 8 responsories (AS, Intro., 60, facs. 171, 174, et.), and Hucke ('Responsorium', MGG, 320) reported that other verse tones were occasionally to be found, even amongst the old core repertory. Bearing in mind the fact that the notation of office chants does not seem to have been undertaken systematically before the later tenth century, one might well ask whether these other tones are the relics of a once greater number, or alternatively whether the restriction to eight was becoming relaxed by the end of the millennium.
(vi) Tones for the Invitatory Psalm
Frere in AS; Stablein, 'Invitatorium', MGG; Steiner, 'Invitatory', NG.
Psalm verses usually consist of two hemistichs, and psalm tones therefore usually consist of two elements. At the start of the Night Office, however, Ps. 94, Venite exultemus, is sung in a different way, in units of five phrases. The text is not that of the so-called Gallican psalter, used for the common psalmody, but that of the so-called Roman psalter (see VIII.6).
There are both simple and ornate Venite tones. Sometimes the five elements are disposed as two times two phrases, with a recitation formula like a psalm tone, then a concluding phrase. In several of the tones, however, reciting notes are rarely to be
3. Tones for Psalms and Other Chants 67
heard. Some of the melodies, as we may call them, are quite ornate. The impression is not that of a 'tone' in the sense we have just been using, but rather of a long melody with a number of elastic points where expansion for a long verse, or contraction for a short one, may easily be effected.
Ex. II.3.8 gives the first three verses of a Venite from the Saint-Denis antiphoner Paris, Bibliotheque Nationale, lat. 17296. In the first phrase c can be heard as a reciting note (in verse 5 it is used more frequently), in the second phrase the note is a. Later verses have further repeated cs at the start of the third phrase, which is then supplanted by d. In the fourth phrase there are again two reciting notes, c at the beginning, then a, as in phrase 2, while c is the most important note in phrase 5. The mutual attraction of a and c is discernible throughout. The figure a-G-a—c—b, moving from a to c and leading back down again, contains the melodic essence of the piece (it is marked with a bracket in Ex. II.3.8). In the course of each verse it appears four times; practically everything else is recitation around a or c, before the final melisma elaborates the kernel figure for a last time. It is noticeable, however, that this figure appears sometimes at a break in the text ('salutari nostro'), sometimes elsewhere ('exultemus'), so it cannot be regarded as a cadential figure; neither does any other obvious cadential figure appear: the melodic material is disposed rather freely over the whole verse.
Ex. II.3.8. Invitatory tone (Pans, Bibl. Nat. lat. 17296, fo. 347r)
i--1
Ve - ni - te   e-xul-temus domino iu-bi-le-mus Deo   sa-lu-ta-ri nostro.
V                           i*    m m					
A	*   »o •	•			*#
rm •			°   mi a		
					
Preoc-cu-pe-mus fa	-ci-em e-	ius	in confes	-si - 0 -	ne.
i	i	l	J		
	_ ^~*\		_____		
if       .    r%    m    m m	*»*			r-__	
<fo-*« * -	• m   + •			-0*-	
Y- Et   in psalmis iu -	bi-   -le -	mus	e-	-i.	
Few Venite tones are available in modern editions. Eight may be found in Liber responsorialis, others in PalMus 12. No comprehensive survey of the medieval repertory has been made, but it seems that invitatory antiphons were reckoned to belong only to modes 2—7, and consequently only six Venite tones were required. In practice some modes, especially mode 4, had several alternative tones. (For Cluniac practice, see Steiner 1987.)
68
//. Chant Genres
(vii) Benedictus es Domine Deus patrum nostrum in the Saturday Mass of the Ember Weeks
Another canticle with its own tone—in this case one only—is the Benedictus es sung between the graduals and the tract at mass on Saturday in three of the four Ember Weeks of the year. The same melody is used for each verse. In modern service-books it is entitled 'Hymnus'. As a song of praise in a general sense it may qualify for this appellation, but the number of syallables in each verse differs, so that the melody, or tone, is stretched out by means of repeated notes, or shortened by elisions, as required. In its mobility the melody resembles some of the Venite tones. As in the biblical text itself, the second half of each verse forms a refrain.
(viii) Te Deum laudamus
Schlager, 'Te Deum', MGG; Steiner, 'Te Deum, 1-2', XG.
Like the Gloria of the mass, the Te Deum is a song of praise whose text is a compilation of heterogeneous parts: parallel verses at the start, a quotation from the Sanctus, a doxology (verses 11-13), then a section praising Christ, and finally verses drawn from the psalms.
Only one melody is known, in several variant versions. It is built upon psalmodic formulas, which change when a new section of text begins. This raises the possibility that the musical formulas are as old as their respective texts, at least in essence. An adequate study based on a comparison of manuscript sources is still needed.
Steiner has outlined the tonal problems of the melody, which is usually assigned to mode 4. From the conflicting or evasive versions in the available sources it looks as if the melody originally had both F# and F§ {b^ and bb at a higher transposition, /sfcj and Eb at a lower). F§ definitely seems right for the latter part of the chant, and earlier/''can be avoided altogether if desired (cadencing G-G-E instead of G-F%—F).
We may take the position with reciting note a, final E, for the purposes of discussion. The following information would have to be modified for individual sources. The reciting note in each phrase is circled.
The first part uses the psalmodic formula:
G-b-c-b~@-b-a || E-G-®-G-ab-baG-G
After some less regular verses another formula takes over:
E-G-®-G-a-b-a || E-G-®-G-a-b-G-E
At 'Aeterna fac' the whole tonal level sinks a fourth, which is where the decision about Ft] or F# becomes important:
C-DE-®-F-D-F-EDC || C-E-F-©-a-F-GFE-E
Sources like the Worcester Antiphoner (PalMus 12, 5) which start on D, with a reciting note G, now move up a tone to join the other versions.
4. Great Responsories of the Night Office 69
At 'Per singulos dies' the previous level is regained. It should be emphasized that these formulas are handled with a good deal more flexibility than simple psalm tones.
11.4. THE GREAT RESPONSORIES OF THE NIGHT OFFICE
(i) Introduction
(ii) Repertory, Texts, and Form
(iii) Music
(iv) Centonization (v) Melismas
Frere in AS; Hucke, 'Responsorium', MGG; Hucke 1973; [Cutter], 'Responsory', NG.
(i) Introduction
The great responsories of the Night Office, like the graduals and tracts of mass considered in the next section, are long, ornate chants in which the same material can be found in several different pieces of the same tonality. This is comparable to what we have just seen in simpler chants. Underneath the surface detail of the ornate chants one may still recognize passages of recitation, and common ways of opening and closing the piece can be discerned, comparable to the intonations and cadences of simple psalmody. The melodies are, however, much more elaborate than the common psalm tones, and there are far more of them, in the case of responsories over a thousand in some medieval manuscripts. They may nevertheless be grouped in families, according to the musical material they use. And if we regard each melody-type as roughly equivalent to a simple psalm tone, then the disparity in numbers is not so startling.
Responsories, graduals, and tracts have fared well as far as musical analysis goes. Frere made a pioneering study of office responsories in the introduction to AS, identifying and labelling recurring musical figures to provide a taxonomy of a great part of the repertory. His shorthand method of characterizing pieces was extended by Apel (1958) to graduals and tracts. In the meantime Wagner (III) and Ferretti (1938) had also published analyses of the repertory. More recently Hucke has devoted several important studies to them ('Responsorium', MGG; 1955, 'Gregorianischer'; 1956; 1973; and further remarks, 1980, 'Towards'). Here it will be sufficient to illustrate the general principles at work.
(ii) Repertory, Texts, and Form
The responsories are not as well known as the graduals and tracts of mass, for they are far more numerous and less easily accessible in modern editions. There are the fascimiles AS, PalMus 9, and PalMus 12, and the texts of other antiphoners (and breviaries) have been published (notably the twelve in CAO). But because the Night
70
//. Chant Genres
Office is no longer sung with the elaborate responsory melodies, Vatican versions are not available. The Liber responsorialis contains over 300, the Provessionale monasticum over 100, and others are in LU.
Most medieval manuscripts contain two or three times this number. The Old Roman antiphoners contain over 600, and so does one of the earliest notated Frankish antiphoners, that of the monk Hartker of St Gall (St Gall, Stiftsbibliothek 390-391: PalMus II/l). Later medieval sources may contain over 1,200. As Hucke ('Respons-orium', MGG) has pointed out, a basic core of the repertory is to be found in practically all medieval sources, but the order in which the responsories are assigned to the nocturns is very variable. (The same is true for office antiphons.) Comparing the order of responsories in various sources has therefore become a favourite way of detecting relationships between sources.
Responsory texts are mostly selected with respect to the readings of the Night Office which they follow. Thus there are blocks of verses excerpted from the Prophets in Advent, from the psalms (in numerical order) after Epiphany, from the Heptateuch in Lent, and so on. Groups of them were sometimes copied with titles denoting their source, as for example 'de Adam', 'de Noe' in Lent, or 'Historie' during the summer months (when Tobit, Judith, Esther, the Maccabees, and so on, were read).
Responsories consist of two main parts, here called the respond and the verse. The verses, usually sung to a rather elaborate tone, have just been discussed. The order of performance is basically respond-verse-respond, to which a Gloria and further repeat of the respond is sometimes added, giving the form R-V-R-G-R. When the respond is repeated it is usually shortened, in that only its second half is sung. For the repeat after the Gloria it may be further shortened, so that only the final phrase is sung.
The assignment of particular verses to particular responds is generally unstable. Early in the ninth century Amalarius of Metz reported that in Rome many responsories were sung with more than one verse, but in the earliest Frankish sources practically all have one only. Extra verses then become more frequent in the later Middle Ages. The variety between sources in choice of verses seems to reflect early practice, whereby the cantor selected his verses at will. This is an important point of difference with so-called antiphonal psalmody, where a complete psalm is sung.
Usually only the first part of the doxology was sung ('Gloria patri et filio et spiritui sancto'), though some sources stretch the same music over the complete text. The Rule of St Benedict calls for the doxology for the last responsory of each nocturn, the practice followed in most medieval books, though once again the practice varied from church to church.
Of the 634 responsories in Hartker's antiphoner (figures in Ferretti 1938, 246, and Cutter, 'Responsory', iVG), about an eighth are in mode 4, about the same as mode 2, slightly more than in mode 3, and slightly less than in mode 1. Most (about a fifth each) are in modes 7 and 8, while the F modes have relatively few.
4. Great Responsories of the Night Office
71
(iii) Music
We can gain an idea of the musical make-up of responsories by placing side by side four pieces from the Night Office for St Stephen (Ex. II.4.1). Remembering how responsory verses were performed, we can see here also standardized points of arrival and departure, and occasionally recitation notes (/*'); but there is greater freedom of movement than in the verse tones, and even room for excursions into foreign territory.
Frere used a system of labelling cadences which I have followed here, where possible (cf. AS, Introduction, 33-4). But in Frere's source (an antiphoner of Sarum use) the second phrase ends on E, whereas in the Klosterneuburg antiphoner transcribed here it ends on F. The same happens in three of the four responds in the third phrase.
The beginning and the end find the four responds in closest agreement, that is phrase Oa and phrase El.
In the second phrase, the first two responds are almost identical, the third respond has the same cadence, but the fourth respond merely has the same range and final note.
The third phrase is similar in all responds, though more loosely so than at the beginning of the responsory; the longer texts have extra recitation (a ac a, etc), while the shortest text, 'domine Iesu Christe', forgoes this altogether. It is interesting that all the four responds converge on the notes aGFGaGaGF (as for 'sanctorum' in the fourth of the responds), but two of them wander on beyond it: 'Christe' needs an extra four notes, 'ait' an extra five.
The fourth phrase sees the responds going along quite different paths. Three of them remain in the lower range, but the third respond rises up rather dramatically to c once again, perhaps in response to the words of Stephen himself: 'Behold, I see the heavens opened' (Acts 7: 56).
The second respond now has only one more phrase, the final phrase which is very similar in all four responds. In front of this, two of the responds have a preparatory phrase, cadencing on G. The third respond uses this phrase almost in passing, aiming instead for a melisma on 'stantem', which carried the music over into the final phrase.
These four are not the only responds using this musical material, as a glance at Frere's analyses will show. The final cadence is also used in mode 3 pieces, and so is that of the third phrase (which resembles the median cadence in the responds for this mode). Some phrases which were unique among our four responds may be found in other mode 4 responsories, such as 'Christi martyri' in the fourth respond or 'accipe spiritum' in the first. Other phrases remain unique: 'ecce video' and the 'stantem' melisma in the third respond. In view of the structural importance of the note F, one might expect some correspondence with F-mode responsories, for example, in the opening. But this is not the case; even though some F-mode responsories start with a phrase reciting on F and cadencing on D, they have a quite different approach.
72 //. Chant Genres
Ex. II.4.1. Responds of four responsories (Klosterneuburg, Stiftsbibl. 1013, fos. 34v—36r)
m <y m -r
V«—-
in-uocantem et di
-cen - tem
La-pidabant
Stephanum
*.c* * ■ •
Videbant omnes Stephanum
qui erant in
con-     -ci-lt- -o
ft
*1 ■
*TT
In-tuens 0«-
♦ » •
in ce - lum
be- -a-tus
Stepha- -nus
•   *   ■ r-T
«  ' *
Pa-tefacte sunt ianue ce- -li
Christi martyri   be - a - to Stephano
•    ■   ■    • ■
ac-cipe spiri-tum meum
4ft do-mi-ne Ie-su     Chri- ste
X3n
^ et in-tu - e-bantur    uul-tum   e - ius
tanquam uultum ange - li
P
ui-
-dit glori-am De - i  et a- -it
ecce ui-
-deo   ce-los   a-   -per - tos
t  t' I t
in-uen - tus est pri-
Dm
qui in nu-mero       sancto- -rum
-mus
*-:-~—	-—:-
!n)-rm hi-■ -/h\09O—#—	-mm-£»-«i» '**•'*'•.*<
et ne sta-   -tuas      il- -lis	-""#m—*-—--» #* '0 hoc   ad    pec- -ca-tum
J2-.-	--
-	-rfs—■—b—•*» r*\
r-	-"j—■—■—■-0 0* *0 -stantes in - ter       il - los L
~ -—    <~m •	
■-■ «^      *• • • ^   et fi-li-um ho-mi - nis	stan-                    -tem a dextris    uir-tu-tis      De - i
	--^-
ft)-r"h m m m m -f*m*m*m "3	-*+X-Sm'\<m*"\ s- 1
V ^,*m                ,*-1 S|<- Et i-deo trium-phat in ce- -lis	1-«-"-mm* »» 1 co - ro- -na-tus
4. Great Responsories of the Night Office 73
The similarities and dissimilarities between responsories are easy enough to pinpoint, especially with Frere's analyses to hand. What does this tell us about the way these pieces were composed?
To answer this question, we should know among other things who sang the pieces. For the period immediately preceding the committing of these pieces to writing, that is, the eighth and ninth centuries, when the repertory was settling into the form we have in the earliest manuscripts, the situation is not altogether clear. The practice of the present day is that the leader(s) of the choir intone the first word (only) of the respond, and the choir completes the respond. The verse is sung by a soloist, or small group of singers. The repeat of the respond is sung by the choir. This practice is by no means a modern convenience, for it can be seen in the Parisian polyphonic settings of the twelfth to thirteenth century. If we allow the principle that soloists usually sing more difficult music than the choir, then this way of performing responsories seems somewhat contradictory. It would surely be easier to train a choir to sing the verses, adapting the well-known texts to the appropriate tone. Wre could imagine the cantor directing his singers where to make the appropriate inflexions, just as he might do for simpler psalmody. The responds, by contrast, are far less uniform, and one would expect them to have required more rehearsal.
As Helmut Hucke has pointed out (1977, 186; 1980, 'Towards', 452), in the Lucca antiphoner (PalMus 9) antiphons and responsories are usually marked with a cross at the point where the solo part ends and the choir takes over. (In responsories it is often, but not always, the same place as where the repeat after the verse starts.) The cross is equivalent to the asterisk found in modern chant-books, but it usually occurs far later than the modern asterisk. The choir apparently sang very little of the antiphon and only a small part of the responsory. In Ex. II.4.1 above I have placed crosses as they appear in the Lucca manuscript, although the Klosterneuburg manuscript has no such marks. (For the third respond the place of the cross is not clear: as far as I can see, one has been erased and another added). This is reassuring, for the crosses indicate that here the choir sings only the most frequently used phrases for mode 4 responsories, labelled Gl and El. Admittedly, more extended rehearsal would still be needed for the third respond.
The division of the respond between soloist and choir may, however, be a relatively late practice. There are several indications that at an earlier time soloist and choir both sang the complete respond. Amalarius of Metz, writing r.830, reported both the Roman and the Frankish practice of his time (ed. Hanssens, iii. 55). The Romans, he says, sang responsories as follows:
Praecantor: Respond Succentores: Respond Praecantor: Verse Succentores: Respond Praecantor: Gloria Succentores: Respond second part
74
//. Chant Genres
Praecantor: Respond Succentores: Respond
According to Amalarius, the Gloria had but recently been added in Rome (but in Ordo Romanus VI, of the eighth century, a Frankish monk said that all responsories were sung with the Gloria). The Franks, on the other hand, usually sang only the second part after the verse. (Complete repeats are nevertheless called for on some occasions in modern service-books.) At any rate, Amalarius does not speak of a solo start and choir continuation of the respond.
The choir's task is easier if the soloist has already sung the respond. Furthermore, in the Winchester polyphonic settings of the late 10th century the whole respond appears to be for the soloist.
The singing of the respond proceeded along clearly understood lines. The tonality (in this case mode 4) would be known, which brought with it numerous conventions guiding the performance: the notes used for reciting passages with longer text, how to start and end phrases, which intermediate goals to aim for (D at the end of the first phrase, EIF second and third), which phrases were opening phrases (Oa) and which closing phrases (Gl-El). Two decisions depend on a close reading of the text: how many phrases must be distinguished, and which syllables should carry musical figures reserved for accented syllables.
(i v) Ce n to n iza t io n
Stablein ('Graduale (Gesang)', MGG, 650-1) and Treitler (1975) have warned us against seeing conventional turns of phrase as fixed entities which a composer took 'off the peg', so to speak, and inserted at the appropriate moment. These are not (to adopt another modern analogy) 'identikit' compositions, put together from pre-existing segments. The music of each responsory is a new creation, which, by virtue of the fact that it is a responsory (and not, say, a gradual) and in mode 4 (not 3), follows conventions appropriate to those categories.
The term 'cento' has often been used to describe the musical make-up of these chants. The Latin word was commonly used in Roman times to mean a garment made of several pieces sewn together, a patchwork. But it also had a special literary meaning, being used as the title of a poem made up from verses of other poems. It seems to have come into modern circulation because John Hymmonides 'the Deacon' used it to describe the work of St Gregory himself: 'Antiphonarium centonem cantorum studiosissimus nimis utiliter compilavit' (PL 75, 90). If we understand the expression in its literary sense, 'cento antiphoner' describes very well the texts sung in the liturgy. Three of the four responsories just discussed have texts selected and adapted from Acts 6 and 7: I give here the Vulgate texts, printing the chant excerpts in capitals. The chant adds other words as well, and adapts others, for example through different conjugational endings, hence the half-capitalized words here.
4. Great Responsories of the Night Office 75
Acts 7: 58-9: Et lapidabant Stephanum invocantem et dicentem: 'Domine Jesu, suscipe spiritum meum.' Positis autem gcnibus, clamavit voce magna, dicens: 'Domine, ne statuas illis hoc peccatum.'
Acts 6: 15: Et iNTUEntes eum omnes, qui scdebant in concilio, viDErunt faciem eius tanquam faciem angeli.
Acts 7: 55: Cum autem esset plenus Spiritu sancto, iNTEndens in caelum, vidit gloriam Dei, et Jesum stantem a dextris Dei; et ait: 'Ecce, video caelos apertos et filium hominis stantem a dextris Dei.'
At least some of these words are actually read as lessons during the office.
The use of conventional turns of phrase in singing these texts seems rather different, a procedure for which the term 'cento' is not really appropriate. These are neither pre-composed scraps to be sewn into place, nor quotations of an already existing composition.
Having come so far, we still do not know how the first decisions of all were taken, that is, how it was decided that this particular matrix of musical material, one of the mode 4 complexes, was the right one for these particular texts. It is conceivable that one piece led to another, for the close proximity of these four in the liturgy is surely not accidental. Other bunches of responsories linked both by musical similarity and by liturgical proximity may easily be picked out of Frere's tables, for example those using Frere's 'theme a' in mode 2, in Passiontide, the summer Histories, and for St Laurence (AS, Introduction, 7). Nevertheless, of the earliest stages of text selection and mode of performance we have no detailed knowledge.
There is perhaps a natural tendency to regard those responsories which use conventional turns of phrase as relatively old, those which are more independent as more recent in date. One associates a simple system with beginnings, deviations as later decadence. But our knowledge of the musical shape of the responsories dates only from the end of the tenth century, the date of the earliest notated antiphoners (the Mont-Renaud manuscript, PalMus 16, and the antiphoner copied by Hartker of St Gall, St Gall 390-391, PalMus II/l). Independent melodies occur even amongst the earliest attested pieces. To some extent comparison of the sources brings out different chronological layers. On this basis Frere did not hestitate to assign pieces to the ages of gold (ending soon after the time of St Gregory), silver (up to the tenth century), bronze, and clay. Later responsories occasionally have new music for their verses.
Frere's theme a in mode 2 (AS, Introduction, 5ff.; Apel 1958, 332 ff.) and his theme a in mode 8 (AS, 52 ff.; Apel, 337 ff.) are two ways of singing responsories which deserve a designation other than 'matrix of musical material'. So many chants use them in such similar ways that they might well be called melodies. Even so, there are deviations from the commonest patterns. Apel actually divides the mode 8 group into four sub-groups. One must also remember that these 'melodies' have been identified as such not because of their melodic characteristics but because they have
76
//. Chant Genres
been adapted for several different texts. They stand, in fact, near one end of a seamless continuum between similarity and dissimilarity.
(v) Melisrnas
The great responsories of the Night Office are generally much less melismatic than the graduals and tracts of mass. In the earliest musical sources the graduals in particular already have long melisrnas on single syllables, and these are generally absent from responsories. Those which do appear are not present in all sources, or 'wander' from one responsory to another, or enter the written tradition at a relatively late stage. For this reason I have considered it best to discuss them as melodic additions to the earliest recoverable state of the responsories, and placed them with the sections on tropes (see II.23.iv).
II.5. GRADUALS AND TRACTS
(i) Introduction
(ii) Graduals in a: The Mustus ut palma' Group
(iii) Graduals in F
(iv) Other Graduals
(v) Tracts
(vi) Tracts in Mode 2
(vii) Tracts in Mode 8
On graduals: Stablein, 'Graduate (Gesang)', MGG; Hucke 1955, 'Gregorianischer',
1956, 'Gradual (i)', NG On tracts: Hans Schmidt 1957, 1958; Hucke 1967, 'Tract', Y<7; Nowacki 1986;
Kainzbauer 1991.
(i) Introduction
Graduals and tracts are more ornate chants than responsories, but much of the melodic embellishment is stylized, and they may more easily be classified in families sharing the same melody.
They can be divided into distinct tonal groups. In Stablein's list of the 115 graduals in the earliest sources ('Graduale (Gesang)', MGG) they break down as follows, according to final (omitting four cases where Stablein registers doubt):
D (mode 1 only) 15 E (modes 3 and 4) 11 F (mode 5 only) 46 G (modes 7 and 8) 15 a ('mode 2') 24
Only the gradual has such a high proportion of pieces in the F mode. Tracts (listed in Hucke, 'Tract', NG) are even more selective:
5. Graduals and Tracts
11
I) (mode 2 only) 6 G (mode 8 only) 15
Apart from their different musical materials, graduals and tracts are performed in a different way. Graduals were sung like the great office responsories, but with a complete repeat of the respond after the verse. Tracts were sung by soloists throughout (a practice sometimes referred to as 'direct psalmody'), without any section repeats.
Amalarius of Metz (Ordinis missae expositio I, 6; ed. Hanssens, iii. 302), writing about 830, reports that in the gradual there was a full performance of the respond by soloists, repeated in full by the choir. This seems to be supported by the copies in early cantatoria (books containing only the soloists' chants), where they are notated in full. By the beginning of the eleventh century, however, similar books (such as Oxford, Bodleian Library, Bodley 775) give only the first word of the respond, implying that the soloist simply intones the opening, the rest being sung by the choir. Later still the choir was allowed to sing the closing word or two of the verse, the practice set out in modern Solesmes/Vatican books. We know this, for example, through the polyphonic gradual settings from Paris in the twelfth and thirteenth centuries, where only the soloist's parts are set in polyphony.
(ii) Graduals in a: The 'Iustus ut palma' Group
The most closely unified group of graduals throws up some interesting questions. This is the family which includes Iustus ut palma, made famous by its presentation in over 200 facsimiles in PalMus 2 and tabulated there by Mocquereau, by Sunol (1935, fold-out), Ferretti (1938, 165), and Ribay (1988). Six of them are sung during the Ember Week of Advent. Both respond and verse of these graduals end on a. This pitch is usually reckoned as representing an upward transposition of a fifth from I) (in order to avoid writing low lib and both and /'"ft), and the group is thus usually assigned to the second mode. The awkwardness of classification indicates that we have here a body of chant which was composed before (and therefore disregards) the eight-mode system.
Ex. II.5.1 is a transcription of Tecum principium V. Dixit dominus from the Aquitanian manuscript Paris, Bibliotheque Nationale, lat. 776. I have marked the individual phrases with the labels given by Apel (1958, 360), whose tables provide a convenient way of surveying the musical make-up of the gradual repertory. Tecum principium conveniently includes nearly all the musical material of the 'Iustus ut palma' family.
In the respond, the first and last phrases (Al and A3) are those of practically all others in the family. (A3 was so well known that the scribe did not trouble to write it out in full: it is completed here from the first gradual of the year with this melody, Tollite portas.) Most graduals have only one phrase between these two, either A2 or Fl; both appear here, with a linking phrase cl. Haec dies, however, has a different opening phrase (A4), then cl, Fl, and A3.
78 //. Chant Genres
Ex. II.5.1. Gradual Tecum principium (Paris, Bibl. Nat. lat. 776, fo. 12r)
A,
zm»±
Tecum principi
-um in
di-
-e uir - tu-       -tis tu-
in splendori-   -bus sancto-
-rum
cl
-—.-
a   ■ *J- 0m 00 m      mm m s- m
* wm   m   mwmM m    a *m
ex u - te - ro
an - te lu-ci-
-ferum
» ••••
ge- -nu-i
te.
*—•-»
Dixit do-minus do-mino me-
Aio.
*****.....*   Vfr.   »■ m»m*T^m*
m ■
se- -de ad dextris me-Cio
-is
do-nec po - nam    in - i - mi - cos
tu- -os
sca-bel-
-lum
pe-     -dem tu-o - rum.
In the verse, all but a handful of graduals have the first phrase given here, D10. The others have a different D-phrase, so that the tonal direction of the melody was always the same. The second phrase, A10, is also practically always the same. For some verses, no more phrases were required, and the final invariable A12 concluded the piece. Most are four-phrase verses, however, and here one of the two F-phrases was employed: Tecum principium has Fl, which has already been heard in the respond. Fl was generally employed when a fifth phrase, CIO, was also present, as here; otherwise F10 was preferred. Just occasionally CIO is present without an F-phrase to follow. The double use of Fl goes hand in hand with the fact that A12 at the end of the verse is similar to A3 at the end of the respond, and cl in the respond ends exactly like CIO in the verse. The overall form is therefore something like AB CB (repeat of respond AB again).
5. Graduals and Tracts
79
The ways in which the melody was used for this particular text, Tecum principium, are typical. The first phrase of text is exceptionally long, and thus is adapted in a way found only for two other graduals with similarly long opening phrases, In omnem terram andiVe avertas. (Exsultabunt sancti, by contrast, is exceptionally short and is therefore sung to yet another adaptation of the 'normal' form.) It is interesting that instead of simply repeating a reciting note to cope with the extra syllables, a different melodic excursion was included (extending roughly from '-pium' to 'virtu-'). Elsewhere repeated notes were indeed employed, for the number of syllables is not excessive: a and c in the respond, d, c, and lastly a in the verse. A3 and A12 begin with accented syllables, and here an unaccented preliminary F was sometimes sung.
It might be supposed that the melody is so constant because the phrases of text are all short (with the exception of the first), so that there was little need for variation, or because the melody is so ornate, necessitating careful regulation of detail. Yet on the one hand the text phrases in responsories are similarly brief, and on the other hand other graduals are similarly ornate but not so uniform in melody. For various reasons the 'Iustus ut palma' graduals have been regarded as very old: their uniformity, in which they display a similarity to the technique of singing tracts, which were also believed to be ancient; their use on days of long standing in the church year. On the other hand, Hucke has argued that the fashion for this melody might be relatively recent. A few of its characteristic turns of phrase can also be found in other graduals (compare CIO with C12 in the F-mode group, F10 with F17 in the F-mode group and F10 in the D-mode group). Hucke thinks this is because phrases from the newly popular melody were taken into the singing of older graduals. The argument cuts both ways, however, and depends on how exclusive the material for the different melodies or melodic complexes was reckoned to be. I myself incline to agree with Hucke, for it seems inherently more probable that a new melody might be used to meet a sudden demand for more graduals (perhaps in response to a liturgical and/or musical reform), than that an old and consistent practice should be superseded by something much less regular. But in the absence of firm historical data one cannot be dogmatic.
One thing is indisputable: the much more ornate surface detail of these and other graduals when compared with the office responsory. Most of this solemn splendour is concentrated on melismas for particular syllables, rather than being distributed evenly. End syllables are particularly favoured, whether accented or (more often) not. The musical matter is of a peculiar kind, reiterating particular notes or decorating them in a seemingly superfluous way. There is no firm direction to these musical ruminations: they are rather a means of making the performance more impressive and ceremonious. Some figures simply revolve around a central tone: 'princiPIum' and 'uteRO'. Others move further but return whence they came. Take for example the setting of 'die'. The melody moves from d to a, but introduces for ornamental effect a palindromic dccaGaccd before finally falling on to a. One may regard this either as an ornamentation of d or a double descent to a, but either way the redundancy is clear. The same happens on a much larger scale in the verse at 'meo' (twice down to G and back up to e) and 'meis' (twice down to a from e). 'Meis' is also interesting because of
80
//. Chant Genres
the oscillation between c and a in the middle. Such reiteration of c, and occasional touches down to a, are found not in graduals only, but also in tracts, introits, and offertories. The notes favoured for reiteration are c,f, and#, that is, those above the semitone step. The other notes touched upon, as it were, to provide extra impetus for the reiteration, are always a minor third lower: a, d, and e respectively.
(iii) Graduals in F
The largest group of graduals to be notated in the same mode is those with final on F. The melodic links between these graduals are much looser, however, than in the 'Iustus ut palma' group. One has only to scan Apel's table (346-7) to appreciate this, especially in the responds, where only the trio Christus fact us est, Fxiit sermo, and Ecce sacerdos are consistently similar. The verses are more easily assignable to families, but only one phrase, the final one (Apel's F10), is found almost throughout the repertory. Apel discerns eight groups in all, with several other graduals not assignable to any group. Between several groups there is almost no similarity at all.
Some general tendencies are nevertheless worth noting. Over half the verses start with a phrase ending on a. The next phrase is then either an F-, an a-, or a r-phrase. There are regular ascents to high/, even before the descent to phrase F10, whereas the respond stays much more consistently in the lower range F-c. In the lower range bb appears regularly, whereas bht is normal for most of the verse. Correspondence of musical material between respond and verse is restricted, when present at all, to the cadential melisma.
Although it is tempting to look for another basic melody, such as the 'Iustus ut palma' melody, at the root of the graduals in F, it cannot really be found. The variety of material used for performing these chants is much greater, the correspondences between the families of graduals consequently more informal. The responds give the impression of almost total lack of dependence upon standard turns of phrase. At the same time, one would not mistake the pieces for anything but graduals. For a start, F-mode pieces are relatively rare in other chant genres. Secondly, the ornamental character already noticed permeates these graduals as well. Somewhat more prominent than in the 'Iustus ut palma' group are recitations on or around one note, as in the closing phrases oiAnima nostra (Ex. 11.5.2). The final turn down to at the end is almost perfunctory, made by using the only other common cadence figure in
Ex. II.5.2. From Gradual Anitna nostra V. Laqiteus contritus est (Paris, Bibl. Nat. lat. 776,
fo.	17V)				
u		m  **      * «•»■ ••••			
/	mm m	•  a          »         •   #• •		•         ■ **	• ••              m m* m m~s mm
					a
					
	et nos	li-be-ra-	-ti	su- -mus	ad-iu-to-ri-um no-b
y		*±         at __i M—  M—   ^mw— —__		______a__	_—-m%__         M____mm *____
/	••	mm      •» •	mm *«, mt		#• # •*         mm* mm -
		•—		• m	am      •   • a •»
					
-strum in no-mine   do- -mi-ni     qui fecit celum et ter - ram.
5. Graduals and Tracts
81
these graduals (Apel's f 11, found also at the end of the respond). (The manuscript has neither clefs nor coloured lines, and consequently does not distinguish the semitone step in the scale. I have assumed bbs by analogy with other manuscripts.)
(iv) Other Graduals
There are other smaller groups of graduals in D, E, and G. The D-mode graduals resemble the F-mode ones in that their verses resemble each other more closely than their responds, and also move in a generally higher range than the responds. The E-and G-mode graduals, on the other hand, are more loosely related to one another, and their verses are not much closer related than the responds. In all these graduals, then, one has the feeling that the singers were expected to range freely through the appropriate material, exercising their art with a degree of freedom not detectable in the 'Iustus ut palma' group or in the tracts shortly to be discussed. Once again many hallmarks of graduals are present, but the melodic gestures do not appear with such regularity. The striking melisma of Ex. II.5.3., where the melody seems to go into suspended animation, is found in only four graduals, its final phrase in only two (but the last six notes are a common way of cadencing on G in many chant genres).
Ex. II.5.3. From Gradual Exaltabo te V. Domine Deus mens (Paris, Bibl. Nat. lat. 776, fo. 54v)
^    .   ^ > .....■ ««.."'. ***«*" '.S SS «t»nt
sal - ua - sti me
The phrase given in Ex. II.5.3 includes some repeated figures (shown by brackets), but such closed forms are rare in graduals. They are even rarer if we set aside examples in this floating, reiterative manner. Clamavemnt iusti has two very mobile melismas in its verse, one with repeat structure AABBC, one unpatterned. Videmnt ornnes has an example with AAB form in its verse. Such patterned melismas are common in alleluias, and are generally understood to be a relatively late development.
We have rather little evidence about the age of the music of the graduals, as is indeed the case with most chants for the proper of mass as it appears in the oldest music manuscripts. Hucke has suggested that the latest of the E-mode graduals may be Iuravit dominus, sung on a number of feasts introduced into the Roman kalendar in the second quarter of the seventh century: no gradual for a feast introduced later than this has an E-mode melody. The important F and a melodies would then be later than this. It would nevertheless be over-optimistic to believe that the melodies as we have them from the ninth century were sung in quite that way two centuries earlier. The known versions must be the result of a long process of stylization, adjustment to changing circumstances (not least the learning of Roman chant by the Franks), and, if Hucke is right, cross-fertilization from other families of melodies.
82
//. Chant Genres
(v) Tracts
As already mentioned, tracts are sung by a soloist or small group of soloists, and include no return to a previous respond or other refrain section. They also have several verses. Those in mode 8 have up to five verses; the three mode 2 tracts designated as such in early sources are much longer: Eripe me (composed after the other two, according to Amalarius of Metz) has 11 verses, Qui habitat has 13, and Deus Deus meus 14. The technique of following a general pattern or mode (in the sense of 'manner', 'way') of singing a verse which we have already seen in the case of office responsories and graduals is also practised here, but now the correspondences may be observed not simply between different tracts but between different verses of the same tract. Since within the two tonal families they display considerable regularity of technique, they have been favourite subjects for discussion of centonization (a term now discredited) or, better, the development of ways of singing elaborate chants in the absence of written music (see especially Treitler 1974, 'Homer', where Deus Deus meus and other tracts are used for illustration). As usual, one may find one's bearings with the aid of Apel's tables.
The twenty-one tracts in the earliest sources comprise the following groups: eleven mode 8 tracts with psalm texts; four mode 8 cantica sung at the Paschal Vigil; three mode 2 tracts with psalm texts; three mode 2 tracts designated as graduals.
The first three cantica of the Paschal Vigil have texts which form a sequence with the lessons of this unusual mass ceremony (see above, I.8.v); they may even be regarded as 'sung lessons' (rather than simply intoned ones; of course their texts are much shorter than those of the other lessons). The fourth tract, Sicut cervus, accompanies a solemn procession to the font.
The three tracts called graduals in early manuscripts were presumably performed like graduals, that is, with the first verse repeated as a respond after each subsequent verse. This seems to have had an interesting effect on their use of cadence formulas: one particular formula (Apel's Dn) is usually reserved for the last cadence of all, but in these three pieces it is also used for earlier verses. This is presumably because the cadence of the first verse, of the respond, was now the final one (D15); there was no longer any need to reserve Dn for signalling the end of the performance.
(vi) Tracts in Mode 2
As may easily be seen from Apel's table, five of the mode 2 tracts follow the same general pattern in most of their verses. (De necessitatibus uses some turns of phrase characteristic of the group, but does not follow the pattern.) They usually have four sections, cadencing on D, C, F, and D respectively. Within these guide-lines there is nevertheless considerable variety of musical material.
All end the last verse in the same way; at the start of the first verse, there are two different opening phrases. But the opening phrases for other verses, D-phrases that is,
5. Graduals and Tracts
83
number no less than ten; Deus Deus meus alone has eight different ones; to be fair, D10 usually functions as a preliminary phrase before D5.
The second section is less variable, though frequently only a cadence formula from the standard phrase is heard in its cadences; there are two alternative phrases. Most constant of all is the F-phrase, with no regular alternatives, though some small deviations. For the last phrase, ending on D, there are five main formulas, four of them used in Deus Deus mens, Qui habitat, and Eripe me, the other used in Domine audivi and Domine exaudi.
The variety of procedure, within the tonal guide-lines indicated, is quite remarkable, and it is clear that these tracts were no more bound in a strait-jacket than other chants. This is shown by the fact that no combination of the various D, C, F, and D phrases appears identically in more than one verse. (If we reduce the requirements simply to cadence formulas, a few identical sequences do emerge, but they are still remarkably rare.)
(vii) Tracts in Mode 8
The eleven tracts and four cantica in mode 8 do not display the regular succession of cadence points seen in the mode 2 tracts. Practically all opening and closing phrases have G-cadences, and usually there is an F-cadence somewhere between. For some verses two F-phrases suffice, three G-cadences are sometimes found, once four G-cadences, without any F cadences at all.
The number of different G- and F-phrases is restricted, more so than in the other tracts. Setting aside the invariable closing G-phrase for the very end of the tract, one may summarize Apel's table as follows:
5 G-phrases for the start of the tract, of which only one is used more than twice; 9 other G-phrases, of which only three are used more than thrice; one of them
(Apel's Gl) opens verses, another (G2) usually closes them, or appears as a
penultimate phrase in the very last verse of the tract; 4 F-phrases;
1 f-phrase, a high-ranging melisma used at the start of two verses in Commovisti.
Within the group, the cantica for the Paschal Vigil form a particularly homogeneous set.
Ex. 11.5.4 is the tract for Quinquagesima Sunday, Iubilate domino, a setting of verses from Ps. 99. When the psalm was sung during the Night Office the division of verses would usually have been somewhat different (the text would also have been that of the so-called Gallican psalter, rather than the older translation used here). Verse 3 would run from 'Scitote' to 'non ipsi nos'. But here there is a short verse 3 and a longer verse 4. The source here transcribed, Paris, Bibliotheque Nationale, lat. 776, indicates no break between verses 2 and 3, but in view of the usual function of phrase G2 as a verse-terminator, I have made the break indicated in other sources.
We can speculate about the way chosen for singing this tract. About the first and last phrases there can have been little pause for consideration, for these were the most
84 //. Charit Genres
Ex. II.5.4. Tract Iubilate Domino omnis terra (Paris, Bibl. Nat. lat. 776, fo. 32v)
V.l
Iu-bi - la- -te do - mino it_„_
om - nis terra
......'
.««■«S    ' "«« l'l'«   .« .™«
ser-ui - te       do-mi - no
in
lae-ti-
-ti
V.2
V.3
V.4
5i
♦—•—#~
In - tra - te in conspectu e-
iL
-ius    in ex - ul - ta - ti-93'     b b_
-o- -ne.
7  *' %* ' '*.....
m   * M*a—n
• ***** »*>V=
Sci - to - te   quod     do-mi - nus
ip - se    est de-
-us.
**.*« :      «**"6 .    UK *.« &...../*.*.*.*-. "**,..
5a
IP-
-se   fe  -  cit nos
et
non ip - se nos
nos au-tem po - pulus       e- -ius
-_-		
	'"S., ^"»'«,	
et   o-ues pas-   -cu - e
e-ius.
conventional parts of the chant. Since the first verse was divided into four phrases, it would be appropriate to move to an /''-cadence next, and because the text is very short (only two words!) F2 is better than the longer Fl. For a pair of short G-phrases, G4 + G3 is often used, G3 being a terminal phrase. They come round again for verse 3. Verse 2, on the other hand, has two longer phrases, so Gl + G2 is preferred. In fact G4 is similar to G2, the latter having a longer end-melisma in keeping with its terminal function.
The first phrase of verse 3, 'Scitote quod dominus', is shorter than 'Intrate in conspectu eius' in verse 2, but not yet as short as 'servite domino' in verse 1. To cope with 'Scitote' the opening of G2 is brought into play. The melisma at the end of the verse, in phrase G3, is usually an end-melisma, but is occasionally split to accommodate unaccented syllables, as here and in verse 1.
The last verse could have started with the popular pair of phrases Gl + Fl, were the text not too short on this occasion. In any case, over half the tracts have unique music at this point, before using standard closing phrases. At the half-verse, for the
6. Short Responsories 85
melisma on the second 'nos', a conventional cadential formula is used, already heard in verse 1 in a similar position. The last melisma of all was so well known that it was not copied out in full by the scribe of the source used here: it is completed from the tract of this family for the preceding Sunday, Commovisti.
With rare exceptions, these tracts move in a very restricted range, hardly exceeding the sixth between F and d. Some could easily be classed in mode 7. They lack some of the splendour of the graduals, for internal melismas are rare and the pulsating reiterations of a structurally important tone are largely absent. Because of their procedural consistency several writers have seen similarities with simple psalm tones, pointing to intonations and recitations, flexes and mediants, and so on. Such an analogy should not be pressed too far, for then the temptation arises to strip the music down to some sort of basic tone, and see the tract as the result of historical development out of imagined simple beginnings. These are not 'variations upon a theme (in G or in D)' for there is no pre-existing theme, and the level of decoration, the degree of solemnity, might have been an essential part of the tract from the start. The conventions of articulation (starting phrases, terminal melismas) are a natural response to the need to mark off the major breaks in the text, found in very many chant genres and not necessarily deriving from simple psalmody. The tonality and range provided a musical frame of reference, and phrases of similar length and balance were sung in similar ways. The result is a mode of delivery flexible enough for the performance of multiple biblical texts, but musically characteristic enough of a particular liturgical moment to be recognizable for what it is and fulfil its proper liturgical function.
II.6. SHORT RESPONSORIES
Huckc, 'Responsorium', MGG; Claire 1962, 1975.
As well as the great responsories of the Night Office (the responsotia prolixa), much shorter responsories (responsotia brevia) were sung after the short lesson (capitulum, chapter) of the Little Hours (Prime, Terce, Sext, None, and Compline). In monastic use a short responsory was sung after the short lesson of Lauds and Vespers as well. The repertory is small, for single responsories did duty for whole seasons of the year, and rather few feasts had proper ones. In most churches, very few melodies seem to have been used, though some sources have more, florid versions of the usual simple melodies, newly composed ones, or melodies adapted from great responsories (see, for example, the Worcester antiphoner (PalMus 12), and the Nevers manuscript Paris, Bibliotheque Nationale, nouv. acq. lat. 1235 consulted by Wagner (III, 217-23). No survey of the repertory is available. The conventions for singing the doxology are obscure: some sources contain no cues at all; other specify Glorias of differing length (for example, up to 'sancto', or 'semper').
The musical style of most short responsories is very simple, chiefly syllabic,
86
//. Chant Genres
consisting of melodic formulas which are easily adaptable for a variety of texts. The form is that of all responsorial chants: solo respond, repeated by the choir, solo verse, choir respond (or the last part thereof), solo doxology, choir respond.
Hucke thought that the most popular melodies might be the result of a ninth-century Frankish recension. In one sense this is no doubt true, as it is of a large part of the chant repertory, but the simplicity of the melodies has encouraged speculation that a much older tradition lies behind them. A peculiarity of some of the melodies is that the formula used for singing the first part of the respond is the same as that for the whole of the verse. Musically, therefore, it is the second part of the respond which constitutes a refrain. This may be seen in Ex. 11.6.1, a transcription of the first of the short responsories in the Lucca antiphoner (PalMus 9). The doxology is divided into two parts, the first rhyming musically with the first half of the respond (and the verse), the second with the second part of the respond.
Ex. II.6.1. Short responsory Super te lerusalem (Lucca, Bibl. Cap. 601, p. 6)
		
		
fm	■a	
\y mm	m *"]•••	
R. Super	te lerusalem	*        •           # • o-ri-e-tur dominus.
			- —
A-			
rt IT"-Tmt-■—M—■-■-~M-M~		-m—m-	-■-1*M-M--
_i-W._■-■_X_X_m._ V. Et glori-a e-ius in te	-*w-ui-	• ' - debitur.	-• " m        mm ' • R. O-ri-e-tur dominus.
, M •        m    m   • m m m    m . m *± p +^
Glori-a pa-tri et fi-li-o        et spiritu-i sancto.
	— -	
-.-,		
	-m-m-JTT—k--m--m-~M-aO-^m #	
Sicut e-rat in principi-o        et nunc et semper	—■-■—9 m—m-m—■—■__ et in secula seculorum a-men.	
		
	mm* R. Super te ...	
The melody of Ex. 11.6.1 was the usual one for the half of the year from Advent onward. During the rest of the year a single F-mode melody in one of the three different variants sufficed. One of these is shown in Ex. 11.6.2, transcribed from the Worcester antiphoner (PalMus 12). (The section of the verse 'alleluia, alleluia' is always written out complete, but presumably one did not then repeat the second half of the respond.) Once again the two phrases of the doxology end on the same notes as the respond, though here the shorter text is preferred.
In both the melodies quoted, the overall form is therefore:
6. Short Responsories
87
Respond: A B
Verse, second part of Respond: A B
Doxology (cadence notes): . . . a . . . b
Respond: A B
Ex. II.6.2. Short rcsponsory Resurrexit Dominus (Worcester, Chapter Lib. F 160, p. 142)
i
'   '* m      mm* *h—r
m * m M   • *
R.   Resurrexit do-mi-nus
al-le - lu - ia     al-le - lu
P
•   » *
o 0
V.   Sic-ut di-xit uo-bis
al-le - lu - ia     al-le - lu - \i
*-0-
m m
Dlo-ri-a   patri et fi-li-o
et spi-n-tu-i sancto.
It would be interesting to know how widespread this pattern was, and whether, at some earlier time, the respond was simply the short second phrase.
The similarity of such simple melodies to some antiphons has often been remarked. Ferretti, for example (1938, 265) thought that the mode 6 antiphon melody for such pieces as Ego sum vitis vera and Notum fecit dominus (Gevaert theme 39) had given rise to the melody of Ex. 11.6.2. Claire's work has tended to suggest the opposite: that the short responsories preserve relics of ancient melodic types, and families of antiphons were modelled on them. Some antiphons would derive from the second part (B above), some from the whole melody (AB). Claire draws parallels not only with Gregorian examples but also with Old Roman and Milanese ones. Particularly suggestive are the similarities with ferial antiphons, especially with the versions of these antiphons in manuscripts from Metz, Aachen, and Lyons. Two antiphons from Aachen, Dombibliothek, 20 (Ex. II.6.3) may be compared with Ex. II.6.1 above (compare the versions in Claire 1975, nos. 67bis and 44). Most other sources transform these into G-mode antiphons. Whether or not one agrees with Claire's
Ex. II.6.3. Ferial antiphons Credidi and Portio mea (Aachen, Dombibl. 20, fos 43r, 47r)
			
A     * *			
rM.         • a	m	M	- • •
			
Credidi propter locutus sum.
Porti-o me-a   domi-ne        sit in terra uiuen-ti-um.
88 //. Chant Genres
hypothesis, it is clear that these short responsories have no connection with any eight-mode or eight-tone system.
II.7. ANTIPHONS
(i) Introduction
(ii) Ferial or Psalter Antiphons
(iii) Antiphons for the Psalms of Vespers, the Night Office, and Lauds
(iv) Antiphons for the Magnificat and Benedictus
(v) The Great O-Antiphons
Gevaert 1895; Frere in AS; Alfonzo 1935; Stablein, 'Antiphon', MGG; Hucke 1951, 1953, 'Formen'; Turco 1972, 1979, 1987; Hourlier 1973; Claire 1975; Franca 1977; Udovich 1980; Huglo, 'Antiphon', NG; Crocker 1986; Dobszay 1990, 'Experiences'.
(i) Introduction
There are more pieces called antiphons than anything else in the chant repertory. The great majority belong to the singing of the office, where they are coupled to the daily, weekly, and yearly cycles of psalms and canticles. But some have no connection with psalms and are sung to accompany processions, or as free-standing votive anthems, most often in honour of the Blessed Virgin Mary. They are of a different musical character and are therefore discussed in a later section, as are the antiphons for the Venite of the Night Office, known as invitatories.
The many hundreds of pieces in the main body of office antiphons may be divided roughly into three groups.
1. Generally short and simple in style are antiphons for the ferial office, that is, for the office hours on ordinary days (including Sunday) when no feast intervenes. Nearly 100 antiphons of this type, with texts drawn from the psalms they accompany, were generally required.
2. The bulk of the repertory was sung on days with their own special liturgy, the dozens of feast-days of various types throughout the year. Over 1,000 antiphons in this category are usually to be found in medieval books, composed for Vespers, the Night Office, and Lauds (proper antiphons were not usually required for the Little Hours). Repertorially, there is a general distinction between secular and monastic books, because of the different numbers of pieces required: only at Lauds, where five antiphons were sung in both uses, is much agreement to be found, mostly in the Temporale.
3. Antiphons for the two canticles of Vespers and Lauds, the Magnificat and Benedictus respectively (called gospel canticles because of their literary source), are generally longer and sometimes musically more elaborate than the others, at least on feast-days. Their texts are usually taken from the gospel at mass of the day, otherwise
7. Antiphons 89
they look to the Old Testament lessons of the Night Office, or, on saints' days, to the vita or life of the saint read in chapter and during the office itself.
Antiphons in these three categories will usually number about 1,500 in most medieval manuscripts. But there is considerable variety between sources, so that the total number used across Europe was enormous. The twelve sources whose texts were edited by Hesbert in CAO have well over 4,000 between them.
Antiphons are settings of prose texts; they follow no regular mode of delivery such as a psalm tone, but naturally, in view of the great numbers required, often display melodic identities, similarities, or standard responses to appropriate texts. Some melodies were very popular and were used with minimal variance for numerous different texts. Other melodies, or complexes of melodic material, were used with much greater flexibility; for example, Frere said of mode 7 antiphons: 'there is much similarity of material and method, which does not amount to a unity of theme'—there is of course room for disagreement as to what does constitute 'unity of theme' (see Nowacki 1977 for an analytical discussion).
The melodic style of most antiphons is relatively simple, with clear-cut phrases. There is no need of the melismas found in florid responsorial chants to mark crucial cadences or other structural features of the text. Within phrases one may find repeated notes, anacruses, and so on, in order to 'stretch' a melodic phrase over a longer text, but this is rarely so extended as to remind one of a psalmodic recitation. When a text has a relatively large number of phrases, however, extra phrases of music will be provided. (Something of this has already been seen in the responsorial chants.) A number of melodies seem to have been composed deliberately for two-phrase texts, three-phrase, four-phrase, and so on.
Since each antiphon preceded and succeeded the singing of a psalm or canticle, the choice of psalm tone was preferably tonally compatible with the antiphon melody. Lists of antiphons were drawn up in medieval tonaries (see III. 14), where the antiphons are grouped according to mode and according to the psalm tone and differentia that they command. These groupings naturally bring together antiphons which are similar melodically. They do not, however, constitute a reliable thematic catalogue, for, if antiphons are to take the same tone and differentia, it is sufficient for their opening and their final to be similar: what happens in between is another matter.
Taking the tonary of Regino of Prüm (d. 915) as a starting point, the Belgian musicologist Gevaert published a melodic catalogue of over 1,000 antiphons under forty-seven 'themes' (Gevaert 1895). Except through passing remarks of Gevaert s, however, it is not easy to get an idea of how stable the 'themes' are in practice, since Gevaert usually cites only the opening of each antiphon. Apel's discussion of mode 7 antiphons demonstrates the value and drawbacks of Gevaert's presentation. So also does a comparison with Frere's analysis in AS: Frere identified fifty themes, established on the basis of the whole melody, not the incipits which Gevaert (like the medieval tonaries) cited, though Frere too was guided by the groupings of the Sarum tonary. Frere's themes are presented with comments and illustrations of their
90
//. Chant Genres
stability, or lack of it. In the index to the Sarum antiphoner which follows the introductory analyses, Frere marked about two-fifths of the 1,600 or so antiphons with the melodic labels assigned in his introduction.
The complex reality behind the 'themes' can be judged from, for example, some of the mode 8 antiphons. Gevaert's theme 12 antiphons are distributed across Frere's themes Villa, c, and j. Antiphons of Gevaert's themes 13 and 15 are both in Frere's theme VI116, but some go into VIII/. But if we actually check through the antiphons in Gevaert's theme 13, we find surprisingly few assigned to a theme by Frere at all. Gevaert has thirteen antiphons in his first epoch, that is, with texts from the psalter (including ferial antiphons listed in a footnote), and six of these are assigned by Frere to VIII/?. But of Gevaert's twenty-nine second-epoch antiphons (with texts from other books of the Bible) and two third-epoch antiphons, not one is assigned in Frere. Some, it is true, were not sung in Sarum use, and others had variant forms of the melodies, but the lack of melodic similarity in Gevaert's group, beyond the opening, is still strikingly demonstrated.
For another example one may look at the transcriptions of twenty-two mode 8 antiphons in parallel given by Ferretti (1938, opposite p. 112). Of these, which are certainly closely related melodically, eight do not appear in the Sarum antiphoner. Three are not assigned to a theme by Frere, one is assigned to theme Villa, six to VHIe and four to VIII;'. of the twelve that appear in Gevaert's catalogue, eight are classified under theme 16 and four under theme 12 (not the same four as Frere's VI11».
Later commentators have been more wary of seeing identity between antiphons. Huglo ('Antiphon', NG) mentions only seven 'prototype' melodies whose basic shape was largely unaffected by adaptation for different texts. Table 11.7.1 compares his choice with Gevaert's and Frere's.
The most sophisticated morphology of the repertory so far has been achieved by Hucke, who distinguishes 'Lieder' (songs, the relatively unchanging melodies) from 'Strophen' (variable successions of standard phrases or groups of phrases) and from recitation types. Hucke's analysis has the merit of not being rooted in the concept of fixed tunes, and identifies features such as initia and other motifs which are independent of thematic families. The fact that only two-fifths of the antiphons were assigned to families by Frere, impressive though the number is, shows that this procedure alone cannot give a satisfactory account of the repertory. (Dobszay 1990, 'Experiences' has reported on a forthcoming new classification. Comparative studies with the Old Roman repertory, analysed by Nowacki 1980, and the Milanese repertory, analysed by Bailey and Merkley 1990, will undoubtedly shed further light on the Gregorian antiphons.)
The work of Claire (1975) and Turco (1972, 1979, 1987) has recently opened up important new avenues of inquiry, relating standard antiphon melodies to apparently ancient psalmodic practice (see also Jeanneteau 1985). There is space here only to mention once again the psalter antiphons.
7. Antiphons 91
Table 11.7.1. Antiphon themes or prototype melodies
Mode			Number	of themes	Prototype (Huglo)
			Gevaert	Frere	
D (Protus)	authentic	1	8	7	
	plagal	2	3	4	
	either		4		Gevaert 9, Frere lie (the
					melody of the 'O-antiphons')
E (Deuterus)	authentic	3	5	5	
	plagal	4	5	3	Gevaert 29, Frere IV b
F (Trims)	authentic	5	3	2	Gevaert- , Frere Va
	plagal	6	3	5	Gevaert 39, Frere VT6
G (Tetrardus)	authentic	7	9	5	Gevaert 23, Frere VI Ic
	plagal	8	7	19	Gevaert 13/18, Frere VI116//
					Gevaert 12/16, Frere VII le/j
(i i) Ferial or Psalter An tiphons
As mentioned in the previous section, Claire has suggested that some of the simple melodies for the chanting of antiphons during the ferial office may derive from the short responsories. Both would in fact spring ultimately from ancient responsorial psalmody, documented in, for example, the sixth-century Psalter of Saint-Germain-des-Pres (Paris, Bibliotheque Nationale, lat. 11947; see Huglo 1982, 'Repons-Graduel'). Another correspondence with old responsorial practice is that the texts of the ferial antiphons are taken from the psalm which the antiphon accompanies, often the first verse (this is also the case in over a quarter of the graduals of mass).
Many of the antiphons have a very limited ambitus, and their tonality was somewhat unstable, for one finds different versions in different modes in various medieval sources. This may also be an indication of antiquity, or at least of an origin before the advent of the eight-mode system. Some half-dozen melodies seem to have been particularly popular (though not always identical in all sources), those which appear in Gevaert (G) and Frere (F) as follows: Gl, ¥\c; G2, Fid (these two are quite similar to one another); G14, FVIII6; G34, F Hie; G40, FVIc; and G44, FVIIIw. It is not uncharacteristic that the melody cited in the previous section as Ex. II.6.3, ending on b (in a sort of transposed E mode), should in the Sarum antiphoner on which Frere's analysis is based be a G-mode melody (VIIIn).
Ex. II.7.1 gives three psalter antiphons from manuscript Aachen, Dombibliothek 20, all employing one of the mode 1 melodies (Gevaert 1, Frere Ic). In this source they are assigned to various psalms of the Tuesday cycle. Secundum magnam misericordiam is an arrangement of the first verse of Ps. 50, Miserere mei. Sitivit in te is the second verse of Ps. 62, Deus Deus meus. Adiutorium nostrum is the last verse of
92 //. Chant Genres
Ex. II.7.1. Psalter antiphons (Aachen, Dombibl. 20, fo. 44r)
Secundum magnam misericordi-am tu-am     miserere me-i Deus.     E u o u a e. Miserere.
•   mmm£m»mm__ m m m 0 m 9
m m m m m -mr
Sitiuit in te        a-nima me-a   De-us meus.        E u o u a e.        Deus Deus-
Adiuto-rium nostrum        in nomine   do-mi-ni.        E u o u a e.        Nisi quia.
Ps. 123, /vVs/ <7«za dominus. Each is followed by the same differentia for the psalm tone and an incipit for the psalm itself (that of Deus Deus meus is in fact the mediant cadence, for the first half-verse consists of no more than the three-word incipit).
The three antiphons display in miniature several features of the genre as a whole. Unlike a psalm tone, where the reciting note or tenor is repeated as often as necessary for the number of syllables, the antiphon melody is more mobile, and repeated notes, extensions, and contractions may be employed wherever it is felt necessary. Not even the cadence, often the least variable part of a melody, is constant here. The basic shape of the melody may be summed up in terms of trichords: F(G)a {aGF) EFG FED (brackets enclose notes not always present).
The two-phrase structure is fairly universal among psalter antiphons, although in the briefest examples a division is hardly necessary.
(iii) Antiphons for the Psalms of Vespers, the Night Office, and Lauds
With all due regard for the variability of the melodic material itself and the sources in which it has come down to us, one may point to the melody Gevaert theme 29, Frere IV6, as relatively stable. It was a favourite for texts of four phrases. It is sometimes notated with final on E, but usually with final on a, and regarded as mode 4 transposed. The reason for this is that both b and bb are required, which would be F\ and F% if the E final were preferred, and F% was not part of the pitch-series used to notate chant. The first three phrases usually sound as if from a normal mode 3 melody (with b§), whereas the final Phrygian cadence falls not on E but on a, approached through bb. Some sources nevertheless choose c instead of b or bb, or notate at the lower pitch with Ft) throughout. And different antiphons may show different approaches to the tonality. Ex. 11.7.2 demonstrates both the formal stability and the tonal difficulty of the melody. The melody is strongly represented in Advent and Passiontide, and the examples are taken from the latter season (in the Aachen manuscript they follow one another directly). (For other examples, presented synoptically, see Stablein, 'Antiphon', MGG, 539, all on E.)
7. Antiphons 93 Ex. II.7.2. Antiphons with the melody Gevaert theme 29 (AS and Aachen, Dombibl. 20)
AS 203
Magister di-cit Aachen f.72v [9          1 H	-a-•-■—>-•   • • tempus meum prope est	apud te faci-o pascha	cum discipulis meis.
,.-	—•—■——mm*	i—•-	1 ^ ^'iij
AS 20
9			•
p> •* 1 Desideri-o Aachen f.72v -8-■-	desideraui hoc pascha	-•-•-• *« manducare uobiscum	f» *-i-LI- antequam pati-ar.
frTT" m"mmrm	■ m • *  *m    * m	-a-	
«P    . • * *-	desidera-ui pascha	1-•-*   *  •  4i *'i 1	1      •      • fr • "
The most common melody, or melodic complex, in the regular E-mode is found among the antiphons of Gevaert's theme 36, Frere IIIc. Here the number of phrases is less regular, usually four or five. It has a relative among the psalter antiphons, and there is perhaps a temptation (which should be resisted) to see in these examples a steady expansion from simple beginnings (Ex. II.7.3). Cunctis diebus, the psalter antiphon, has but two phrases. Picking up the G—a opening of the second of these phrases, Fidelis servus brings in a central phrase, which cadences on G, in between the b and E cadences of the outside phrases. For a fourth phrase the a-c, b-G steps at the end of the first phrase may be developed: in Herodes enim this happens in the second phrase, in Hie est discipidus ille in the third. The last phrase may also generate an extension, as in Nigra sum.
One simple way of generating music for a longer text is to repeat the melody, which happens with some members of Gevaert 18, Frere VI116, and Gevaert 12, Frere VIII/. Another group (Gevaert 39, Frere VI6), displays a simple ABAC form, where B and C are settings of 'alleluia', making an intermediate and final cadence respectively.
A large number of antiphons, however, draw upon a stock of phrases in a way somewhat akin to the procedures we have seen in responsorial chants. This has been demonstrated, for example, for a group of mode-1 antiphons by Ferretti (1938, 113-16; compare the antiphons having Ferretti's opening 1 with Gevaert theme 6 and Frere la). Some of the phrases were clearly associated with openings, some with antiphon endings; others may have a mediant function. Hucke pointed to the universality of this technique through many chant genres (1953, 'Formen'; see the example on pp. 16-19). A glimpse of it can be seen in Ex. II.7.3 above, where the outside phrases are constant but the inner ones vary.
94 //. Chant Genres
Ex. II.7.3. Antiphons with the melody Gevaert theme 36 (Lucca, Bib!. Cap. 601)
92__
— />*■_*«
P
-*>—•—
Cunctis di-e-bus uite nostre saluos nos fac domine. 540
P
•  *    -    ■    * -   -   - .=»
• • *-*-•—■—■—■—■——■—•-•—•—»
Fidelis seruus et prudens quern constitu-it dominus super fami-li-am su-am. 454
t  " » «  ' <i I        •   •   «fr *•* « A ,
Herodes enim tenuit et ligauit Iohannem et posuit in carcerem propter Herodi-adem. 55
" 0    0   ****** mamammmm*m9 tt«mmmm*9
Hie est discipulus ille qui testimonium perhibet de his     et scimus quia uerum est
testimonium eius.
543
0m*     *     *   '  *        0 9 I   I • I ,        •   •   • t> ' •   0 •    •   •   m • • 9-
Nigra sum sed formosa      fili-a Iherusalem      i-de-o dilexit me       rex et introduxit me
in cubiculum suum.
Ex. 11.7.4 shows something of the same technique persisting in a relatively late group of antiphons. They were notated in some sources, including the Lucca antiphoner used here, on c, in others on F with bb throughout.
Gaudeamus omnes is from the liturgy of the Octave of the Nativity; exactly the same form of this melody was used for Nesciens mater and Virgo hodie fidelis, on the same feast. One may distinguish five phrases, labelled A to E. The third phrase C uses the same small decorative turn cefed as the second phrase B. The sense of the text carries us over the break between third and fourth phrase, though other antiphons in the family show a caesura here.
Pro eo quod non credidisti, for St John the Baptist, has a longer first phrase, which is given an ornamental opening O; the second phrase is as in Gaudeamus, but phrase C is absent and E is much shorter; D now echoes the opening; the last phrase has little room to expand up to the high g.
In Modicum et non videbitis (Easter week) the initial fall to G (O) is no longer ornamental. The second phrase (X) is now quite different, for the composer seems to want to reflect the parallelism of the text in his music. The D and E phrases are still
7. Antiphons Ex. II.7.4. Antiphons (Lucca, Blbl. Cap. 601)
38_A _______ Ä B
95
*a * 9 a r:
Gaude-a-mus omnes fi - deles        saluator   nos - ter natus est in mundo
409 0
m
0  > • * •  m  a s a
ft*
Pro e - o 236 o_
quod non credidisti uerbis meis e-ris ta - cens et non pote-ris loqui
A X
* • a* a a fr ' »5E
•   f9 Š
3.
* *9 a
Modicum et non ui-de-bitis me dicit dominus
i-te-rum modi-cum et ui-debitis me
-fi-:-	m ■	-Ä-	-S£-		*   * •
X -			•   0   0 •		0               ma 9m
Y\\           ■ •m					
ÍU •					
Que 640 o
5.
IE
muli-er habens dragmas decern     et si per-di - de-rit dragmam u-nam
-A_,_A   .     .     . .j.
■»I«!»»'»!!-
0 * m  • m  »m •» r
O
ad-mira-bile commertium   cre-ator generis hu-mani
» * •
»—0-
-*—m-
C_ D _
w   ' t i i • i »'* *• ť«    « * • * *
a-ni-ma-tum corpus sumens
3=*
a a * **' ij a fij
rr
hodi-e processit  pro - les   magnifi-ci germi-nis
et perseuerat pudor uirgi-ni-tatis. JE,-
	■V -, + • 0 B 0 s?at-	-B0-*0                           0 W
%-		-♦ ■-
m  m   m -mrnw-
m 0  ' • j
usque in di-em
na-ti-ui-ta-tis e-ius.
quia uado ad patrem
J)
al-le-lu - ia al-le - lu-ia. ti ___Ex^
f *
0*0 m
'tat
nonne accendit lucer - nam      et e-uertit domum
et querit diligenter donee inueniat.
-6-                       M            f0*0-   ml m	0     0             0 1	—ä-—£#-	—-
ft-m  m   0   m  *   0   m   0     *     * 9	0    9          m    m* m	ěl 10	BA
		■ •	' 0
Kv			
de uirgi-ne nasci dig-na-   -tus est      et procedens homo   si-ne se-mi-ne
-* m*
* 000
■ ' ■
9 ■  B *'
lar-gitus est nobis suam de - i - tatem.
96
//. Chant Genres
recognizably related to those in the other antiphons, E now consisting of two distinct units, the two 'alleluia' calls. But E also now resembles the opening phrase.
In Que mulier (post Pentecost) all the melodic material seen so far is used, with E again dividing into two units.
Finally, O admirabile commercium (Circumcision) needs an extended first phrase, makes a second out of it in fact. D is also more extended than before. E has the same form as in Gaudeamus.
Although details vary from one piece to another, there is obviously a clear sense of the character and direction of each phrase. A is a decorated recitation on r; B makes a decorative half-cadence which includes high /, then takes the high / as the starting point for a descent back to c. C recites on c then makes the same decorative cadence more conclusively on d. The function of D is to turn the recitation down to a low G. E reverses that movement, perhaps with enough energy to touch on high g. With quite simple ideas like this in mind, the singer could easily add more material of the same kind, omit or substitute an internal phrase.
(iv) Antiphons for the Magnificat and Benedictus
Much analysis remains to be done before we can see the full extent of the material shared by different antiphons. Perhaps a layering of the repertory, by season, or even chronologically, might then be discernible. The task demands not only analysis of the melodies themselves but also careful comparison of the repertories in sources of different provenance and date.
Considerable further difficulty arises because of the variety of musical readings in different sources. Even when sources have the same melody (disagreement even at the basic level is not uncommon) there is considerable difference of detail (see the parallel transcriptions by Udovich 1980).
At the moment it appears that for much of the yearly cycle antiphons for the Magnificat of Vespers and the Benedictus of Lauds shared melodic material and general style with the bulk of other office antiphons. Nevertheless, it is rare for these antiphons to have less than four phrases. Some antiphons, however, many probably of relatively late date, are far more extended and more florid, particularly those for the Sanctorale. An example will make this clear.
Ex. 11.7.5 gives three Magnificat antiphons of this type, all in mode 1. All three relate part of the story of the martyrs in question, Andrew (30 November), Agnes (21 January), and Agatha (5 February). In each we are told where the saint is (Andrew comes to the place where the cross has been erected, Agnes stands amidst the flames, Agatha in prison) and then each saint speaks: Andrew addresses the cross itself, Agnes glorifies God, Agatha prays for strength and eventual reward in heaven. Despite the similarities in the text (particularly at the start of the Agnes and Agatha antiphons) the musical resemblances are not usually literal. As we should expect, however, the antiphons make use of phrases typical for this tonality, adapted to the text as necessary. Much of the movement is concerned with rising from D to a and
7. Antiphons 97 Ex. 11.7.5. Magnificat antiphons (Lucca, Bibl. Cap. 601) 504 _pcF*_^_*D
^     »*Qk m   *• 0     9        ' ° *• «3 9—9-
Cum perue - nis - set be - a-tus An-dreas ad lo-cum u-bi crux pa - ra - ta e-rat »ca. _&.F<f _FC_
Í
excla - mauit et di-xit       O   bo - na crux   di - u de - si - de - ra-ta
Í
^—_p « *^-• • •-
et iam concupiscen-ti   a - nimo pre-pa-rata    se-curus et gaudens   ueni - o  ad te
1
i - ta ut et tu  e-xultans     susci-pi-as me   disci-pulum   e-     -ius qui pepen - dit in te.
i
339__
—_. r
-- - ----r
Stans bea-ta   Agnes      in me-di-o flammarum expan-sis ma-nibus
o - ra - bat   ad do-minum   omni-potens a-doran-de      co-len - de tremende
a.0_c»r>__
' '    fr'fS    £ +   .    '    '    9    *    • 9±    9 9 9^
be-ne-di-co te        et glo-ri-fi-co nomen tu - um in     e - ternum.
364      (DcFfc)  Ca-P (P)»,C-
^   *   *    9j   f   9..    . ... *   * *
Stans  be - a-ta   A - ga-tha    in   medi-o   car - ce-ris     ex - pan-sis manibus tota mente
p
o- ra-bat ad dominum     do-mi-ne le - su Christe        ma-gis - ter   bo - ne PC Fa,_(£,)». p_
^  •    *     'M«   I ^ ^ Í*   *» *"g    " 9 j* 9
gra-ti - as    a - go ti-bi   qui me fe-cis-ti uin-cere       tor-men-ta car-ni-fi-cum POP*. P>.G CaP
^,9  ' V       <^ " f^,« *  *  *   * £,   ^» " **
iu-be me domi-ne   ad     tu-am      in-m'arcessi-bilem glo-ri-am   feli-ci-ter perue-nire.
98
//. Chant Genres
falling back again. Sometimes the descent is to C, sometimes there are pauses around a or G. The following may be picked out here (the labels represent important pitches in each phrase and are also used in the transcriptions).
(i) DFDCFa + a-D and c-a-D. The penultimate phrase in all three descends from a to C, and the final phrase is an arch from C back to a and down to the final D. All three begin with a melodic shape seen at its simplest in the Agnes antiphon: DFDCFa. For Andrew there is then a long descending phrase back toD. Agatha only arrives at a at the beginning of the second phrase, but the overall progress has been the same as for Andrew. There is an exact parallel to the two opening phrases of the Andrew antiphon in Agatha's 'gratias . . . carnificum'. The opening phrase recurs yet again in Agatha's 'iube . . . tuam'. The rise from C to a and back to D is compressed into one unit for the final phrase; and it can be seen again in the Andrew antiphon for 'et iam . . . preparata', and twice in the Agnes antiphon; in the latter it forms an alliance with a descending a—C phrase, so that 'orabat . . . adorande' and 'colende . . . te' are the same; 'orabat ad dominum' is exactly as in the Agatha antiphon.
(ii) a—C. In Andrew 'o bona crux' forms part of a longer descent from a to C by 'desiderata'. This is also the purpose of 'discipulum eius'. We have already seen it twice in the Agnes ('omnipotens adorande', 'benedico te'), and noted its role as a penultimate phrase.
(iii) a—c-a. Andrew: 'exclamat et dixit', 'securus et gaudens'; Agatha: 'domine Iesu Christe'.
(iv) a-G. Andrew: 'o bona crux', 'venio ad te', 'suscipias me'; Agnes: 'in medio flammarum'.
Although the antiphons initially appear somewhat ornate, the basic melodic gestures are in fact quite simple. Similar phrases (indeed identical ones at the opening) may be seen in Ferretti's table of mode 1 formulas (Ferretti 1938, 113-14), though they are generally syllabic. If compared with their more numerous simpler sisters, these antiphons make more leisurely progress, and the whole effect is more solemn, as befits their role in the liturgy.
Magnificat and Benedictus antiphons were occasionally directed to be sung as processional chants. There is also a repertorial link with the antiphonae ante evangelium, possibly non-Roman survivals (see VI.5.v). All this may indicate that melodies of the Gallican rite have survived in the guise of Magnificat and Benedictus antiphons; but further study and analysis are required.
(v) The Great O-Antiphons
A special group of big, six-phrase Magnificat antiphons, all sung to the same unique mode 2 melody and beginning with the word 'O' (O Sapientia, O Adonay, O Radix Iesse, and so on) were sung on the seven days leading up to Christmas Eve. Others were later written in imitation of them, but the original seven appear to have been composed as a group, for they are all addressed to Christ, and may be linked by an
8. Invitatory Antiphons 99
acrostic: reading in reverse order the initial letters of each second word one finds the text 'ERO CRAS', which is interpreted as 'Tomorrow I shall be [with you].' The largest bell of the church was rung while they and the Magnificat were sung, and they were assigned in turn to the most prominent members of the ecclesiastical hierarchy: abbot, prior, cellarer, and so on.
II.8 INVITATORY ANTIPHONS
Frere in AS; Stablein, 'Invitatorium', MGG; Steiner, 'Invitatory', NG.
The invitatory at the start of the Night Office comprised the singing of Ps. 94, Venite exsultemus, and an accompanying antiphon. Just as the Venite was sung to tones independent of the eight simple psalm tones, so the antiphons form a musical class of their own. In some respects they have more in common musically with the great responsories of the Night Office than with other office antiphons.
Ferretti (1938, 220-1) reckoned that twenty-nine invitatory antiphons belonged to the earliest layer of the repertory. A typical antiphoner will contain seventy or eighty. There is enormous variety in medieval sources as to the choice of antiphons and the Venite tone they command, and little systematic research on the repertory has yet been accomplished. This instability argues for a relatively late, expanding corpus; yet other features seem archaic. As reported above (II.3.vi), there are no antiphons (or very few) in modes 1 and 8. Furthermore, the antiphon seems to have been repeated after each verse of the Venite, the complete antiphon at the start and after verses 1, 3, 5, the second half of the antiphon after verses 2, 4, and the doxology. It is usually thought that in ancient practice all psalms were performed this way, or with complete repeats after each verse. Against this, it may be pointed out that certain processional hymns of recent and non-Roman origin were also sung thus in the Middle Ages (II.15.iv).
Ex. 11.8.1. Invitatory antiphon Quoniam Deus magnus and part of responsory Tolle anna tua (Bamberg, Staatsbibl. lit. 25, fo. 47r)
Quo-ni - am De-      -us   magnus      do - mi - nus
et    rex     ma- -gnus    su - per om  - nes        de-       -os.     Veni- -te.
^   jAii  *t    ft ' *     ' * z
R. Tol-   -le     ar - ma tu-        -a     pha-re-tram   et ar-       -cum ...
100
//. Chant Genres
Most invitatory antiphons consist of some four phrases set in an ornate melodic style. As with antiphons in general, they sometimes use relatively stable typical melodies, sometimes draw upon material characteristic of the tonality but capable of more flexible use (Frere gives an account of the groupings in the Sarum antiphoner). An interesting feature of the mode 7 antiphons is their use of material also found in great responsories of the same mode. Ex. 11.8.1 gives a Lenten invitatory antiphon and the start of the succeeding responsory (this fortuitous succession does not, of course, occur in all sources).
II.9. PROCESSIONAL ANTIPHONS
(i) Introduction
(ii) Rogation Antiphons
(iii) Palm Sunday Antiphons
(iv) Antiphons for Other Occasions
Bailey, 1971. (i) Introduction
The antiphons for processions and the votive antiphons for the Blessed Virgin and other occasions, many composed in the later Middle Ages, require further study even more than do some of the other antiphon repertories just discussed. For processional antiphons a study by Bailey has fortunately prepared the ground. (In the absence of other editions, the facsimiles PalMus 12, 13, and 15 and Vecchi 1955 may be consulted.)
Medieval books are likely to have processional antiphons for the following occasions: (i) as part of the ritual accompanying the chanting of the litanies on 25 April (St Mark's Day) and on the rogation days on the Monday, Tuesday, and Wednesday before Ascension Day (Thursday); (ii) for the liturgy of Palm Sunday; (iii) in smaller groups as required for other days with special liturgies, such as those of Holy Week or the Blessing of the Candle on 2 February (Purification of the Blessed Virgin Mary); (iv) for the less elaborate processions instituted as a regular feature of many feast-days of the year, mostly before the high mass or after None. Most sources have about forty antiphons altogether.
Processional antiphons are generally ornate chants, some very long indeed, with lengthy melismas. Some have an equally ornate verse and were performed responsorially, that is, with a repeat of all or part of the first section, the respond, after the verse. Others appear with an incipit for the chanting of a psalm, but it is not clear how many verses would have been sung (presumably in alternation with the antiphon).
There is no doubt that the repertory contains chants of quite different origins. Over eighty can be found in Roman sources and may have originated in Rome, while
9. Processional Antiphons 101
Frankish composition may be suspected for many others, and old Gallican relics may also be present. Yet the basic analyses of style which might help differentiate these three types, if they are really present, have not been carried out.
(ii) Rogation An tiphons
The earliest sources contain nearly 100 rogation antiphons, invoking God's aid in various times of trouble, or referring to the procession itself and the saints' images carried during it. Later sources usually content themselves with about twenty. Bailey's table of antiphons in twenty early sources, including those from Spain, Milan, and Rome, lists over 150 items (Bailey 1971, 122-7).
The antiphons with Roman counterparts usually display standard Gregorian features, the conventional cadences (for example EGF FE or baGa aG), occasional groups of repercussive notes (reminding one of introits or offertories) and the same gapped scale that provides a framework for many Gregorian chants. Occasionally there are hints of a different manner, for example in repeated musical phrases, but this is not a regular feature. (Compare the Gregorian version of Non vos demergat, Bailey, 57, which has a repeated ornate recitation formula, with the Roman one, MMMA 2, 573, without repetition. Other repetitions may be found in the 'alleluia' endings, but these are usually later additions).
Ego sum Deus, Ex. 11.9.1, is an antiphon of this sort. One would expect a conventional Zs-cadence after 'eos', but it is made imperfect, and a twofold alleluia follows. Up until here the antiphon has the same general outline as the Roman version (MMMA 2, 545), but the 'alleluia' is quite different. In fact it is a 'wandering' addition, which also turns up at the end of several other antiphons in this mode.
Bailey has pointed out several examples of musical material found in more than one antiphon of the same mode, for example between Ego sum deus and Populus Sion convertimini (Bailey, 147). No network of interrelationships seems to be present, however, of the sort that has been determined for some groups of office antiphons and responsorial chants.
Ex. II.9.1. Processional antiphon Ego sum Deus (Provins, Bibl. Mun. 12, fo. 163r)
		
		_
fn\     #• •                    *        •* ■•■Hi		
W.—*-•—		--* % 0
E-     -go sum De-       -us pa-trum ue-stro-     -rum    dicit      do-   -mi - nus
-fl-:-iSi—:-Oi-~-:-	
a       *h m\         -  ■_.          «■-• ■■_	■••   r m
fm           »             •   •!   •                 _ ■Ml» m	f_* at  m ___             ■Tai^-i. j-^*S_
\sy                              '               •   • »	
uidens ui - di   af - flic - ti - o-      -nem      et     des - cendi
-r)-                "         "                                                          - m----		
fa-7*^—.^■>i-—iSl—7?-^—	m a> 'a	
	•	• «•    • • 9m*H tj...
li-    -be - ra-      -re e  -  os      al-le- -lu - ia    al-le- -lu - ia.
102
//. Chant Genres
One rogation antiphon, Deprecamur te, has achieved special fame, since Bede reported that Augustine and his followers sang it as they first approached Canterbury, carrying a cross and an image of the Saviour, on their mission to England in 597 (Bede, trans. Sherley-Price, 70). The antiphon has been edited several times (for example, Stablein, 'Antiphon', MGG, 542; MMMA 2, 565), and its various Frankish and Italian versions have been compared by Levy (1982).
(iii) Palm Sunday Antiphons
Of quite different dimensions and style are some of the grand antiphons for the Palm Sunday procession, such as Collegetunt pontifices V. Unus autem, Cum appropinquaret Dominus lesus, and Cum audisset populus. (Most sources have some half-dozen chants.) Collegerunt (Ex. II.9.2) is occasionally found as an offertory for mass, possibly a reflection of earlier Gallican liturgical practice. It begins with a sweeping melisma, repeated almost at once, and other repeats are present (marked with letters). A richly ornamented recitation may be discerned, whereby the common cadence formula at 'Romani' and 'locum' is almost lost. Repetitions such as these may be a natural event in the setting of a long text, in melismatic style, when no vocabulary and syntax of the type developed for, say, graduals and tracts is available. We have already seen it in some later office antiphons, and the suspicion arises that it is a non-Gregorian characteristic.
Ex. IL9.2. From the processional antiphon Collegerunt pontifices (Provins, Bibl. Mun. 12, fo. 110r)
* *~ A fi
Col-        -le- -ge-     -runt ... et    di-       -ce - bat
ue- -ni - ant        Ro-ma-      -ni     et tol-     -lant no-
m
-strum lo-   -cum ...    ab il-lo er-go di - e cogi-tauerunt
(iv) Antiphons for Other Occasions
Yet other stylistic features appear in the big ceremonial antiphons from other times of the year. Both Ave gratia plena and Adorna thalamum suum (Purification) were translated from Greek kontakia when the ceremony of blessing the candle was instituted in Rome by Sergius I (678-701). Byzantine melodies for these texts are not
9. Processional Antiphons 103 Ex. II.9.3. Processional antiphon Ego sum alpha et a> (Paris, Bibl. Nat. lat. 903, fo. 74r)
E-go sum al-fa et CO        primus et no-uissimus        i-ni-ci-um et fi - nis
P
qui an-te mundi princi-pi-um        et in seculum secu-li        ui - uo in aeter - num.
-»-"*-'     * •-
A - quam me -e   que uos fe-ce-runt        cla-uis con-fixe sunt
propter uos flagellis cesus sum        spi-nis   co-ro-natus sum.
-»-"»    9^-W-• •   * •-
A - quam pe-ti - i pen-dens        et a - ce-tum por-re-xerunt in escam me-am fel de-derunt        et in la-tus lance-am.
9- * —# -f-
Mor-tu-us et se-pultus sum        resurre-xi uo - biscum sum
• »• V      V * - "   *"   * m^ *   **o *
ui-de-te qui-a   e-go ip-se   sum        et non est de - us pre-ter me.
Al-le- -lui- -a.
						
			--»\	m m		
* V. E	—•—m— -go sum	mm   m -uestra	—•—# 0 redem	-pti-o	e-go sum	rex uester
"» ^ JOB   99*,   * •     • • •* 1
m  mm  mm   m • **-
"----^3)        Wly,       - -■-V-• *"VV
e-go uos resusci-ta-bo        per dex-te-ram me-am. Videte.
104
//. Chant Genres
known, but the Latin antiphons are in any case not related to each other musically. Adorna thalamum has a very obvious repeat structure, since practically all lines follow the same highly ornate recitation formula (Stablein, 'Antiphon', MGG, 542).
Different again are such antiphons as those usually assigned to Sundays after Epiphany or during Lent: Ecce carissimi, Cum venerimus, and In die quando (Bailey, 30-3). In these three big D-mode antiphons practically every cadence is made from the tone under the final or the dominant, C—D—D or G—a-a, a cadence often thought to be of Gallican origin, certainly not Gregorian, much used in sequences. In spite of an overall uniformity, caused by limited number of melodic goals, the only formal repetition occurs in one or two melismas. And not the least problem of deciding on the provenance of such pieces lies in the fact that the melismas may take on different forms in different sources, or be omitted altogether, and the cadences may take more normal Gregorian forms.
Easter processional antiphons include Stetit (Sedit) angelus V. Crucifixum in came (two versions are discussed in Roederer, 1974) and Christus resurgens V. Dicant nunc Iudei. The two are nevertheless of different tonality, proportions, and musical character, Stetit angelus being in G, with a long respond and short verse and a very mobile, free-ranging melody, Christus resurgens being in D, with respond and verse equally long, cramped in range, and making use of the Gallican cadence.
One remarkable antiphon, Ego sum alpha et w (Bailey, 42), is a meditation by Christ himself (Ex. 11-9.3). Here yet another distinctive voice makes itself heard. Elaborate melismas are absent, with only a short alleluia with (in this source) brief repetition. The Gallican cadence is obvious, and the music of several lines is the same (7-10, 11-14, 15-18 form loosely three strophes; 19 and 21 start similarly); the alleluia has an AAB form. Comparison with other sources shows again, however, that not all these features are constant across the manuscript tradition (cf. Bailey, 42).
For many processions it was usual, at least in the later Middle Ages, to borrow one of the responsories from the Night Office of the day, or, since Vespers on feast-days might also have a responsory, that one instead. (The pieces in the Liber responsorialis are almost all such responsories.)
11.10. MARIAN ANTIPHONS
Harrison 1963, 81-8; Huglo, 'Antiphon', AG.
During the twelfth and thirteenth centuries the practice arose in many churches of singing an antiphon to the Blessed Virgin Mary as a devotional act in itself, independent of, though often attached to, one of the other services of the day. For example, in Roman and Franciscan usage since the thirteenth century, four antiphons have been sung at the end of Compline, one for each of four seasons in the church year: Regina caeli, Alma redemptoris mater, Ave regina caelo?~um and Salve regina. Harrison indicates the large number of different practices found in medieval England,
10. Marian Antiphons 105
and this is no doubt typical of medieval Europe in general. Antiphons to the Saviour might also form part of the devotion. Collective names for such chants are votive, devotional, or commemorative antiphon. Similar antiphons might also be composed for other occasions, but those for the Virgin are the most numerous, because of the immense enthusiasm for her worship and the multiplication of services in her honour: a complete cycle of office hours, weekly or even daily mass, and so on.
In practice it is hard to draw a clear line between antiphons fulfilling these various functions. An antiphon in solemn (that is, ornate) style might serve as a Magnificat antiphon in one source, as the antiphon for a commemoration at the end of Vespers or Compline in another, or elsewhere be assigned to a separate ceremony.
Much work remains to be done in comparing the repertories of different sources, studying the rubrics in chant-books, ordinals, and customaries, and in analysing the style of the chants themselves. Few are available in print (some are published in Variae preces, Cant us selecti, and Processionale monasticum.)
Antiphons had of course been sung as part of the normal office hours on feasts of the Virgin since early times. Most early Marian antiphons are, as we would expect, simple, largely syllabic pieces indistinguishable musically from the rest. Such is, for example, Sub tuum praesidium (Ex. II.10.1), whose text can be traced back to the third century (Mercenier 1940) and which was sung in the Milanese as well as the Roman rite. A favourite source for such texts was the Song of Songs.
Ex. II.10.1. Marian antiphon Sub tuum presidium (Paris, Bibl. Nat. lat. 1139, fo. 126v)
$    .   .   '   ' '      *   ' " " # <^^-
Sub tu-um presi-di-um confu-gimus   de-i genitrix
i
• • •
nostras deprecati-o-nes ne despici-as   in ne-cessitatibus
P
m - ' m
set a periculis libera nos       sem-per      uirgo   be - ne-di-cta.
Of more ample dimensions are antiphons which doubled as Magnificat or Benedictus antiphons. One such, 0 virgo virgin um, was borrowed from the series of impressive O-antiphons. It was of course a simple matter to copy a psalm-tone cadence at the end of the antiphon if required, or omit it. The same antiphons might also be sung in procession, as for example, Alma redemptoris mater and Ave regina caelorum in Sarum use.
Older texts might be reset in a more ornate style to make them more appropriate for votive use. Ex. II. 10.2 gives Speciosa facta es in two versions, the first as a simple office antiphon (in the Lucca antiphoner for the Night Office at Purification), the
106 //. Chant Genres
Ex. II. 10.2. Marian antiphon Speciosa facta es in two versions (Lucca, Bibl. Cap. 601, p. 347; AS' 529)
Lucca
4n-r* m-	-£m a—•-	-M      j.-                                               " """
	—tß    0 *—--	
Speci-o-sa facta es      et su-a-uis      in delici-is tu-is      sancta De-i ge-nitrix.
—-^-m'   *   * »'S ^ IN-^--m* m**0
Speci-o - sa     fa-cta es et su-a - uis        in de-li-ci - is uir-gi-ni-ta-tis
P
sancta   de-i ge-ni-trix        quam ui-den   -   tes fi-li-e Sy-on i-1
_«   &    «      *    P   '   mU   ^ J*    . 1    fl mf»*
uernan-tem in flo-ri-bus ro - sa - rum        et      li - li - is con - ual-li-um
be-a - tis-si - mam     pre - di-ca - uerunt        et re-gi-ne     lauda-ue - runt e-am.
second reset and extended for processional use (in the Sarum antiphoner for Vespers of the Nativity of the Blessed Virgin). The great difference in musical style goes beyond the degree of decorative figuration. The simple Lucca version uses the popular 'theme 29' melody already seen in Ex. II.7.2 above. The Sarum melody gives the differentia for psalm tone 6 at the end, but because a bb is required (for the un-Gregorian bb—c—c cadences at 'liliis convallium') the chant is notated upon c. In almost every phrase the major triad c-e-g is outlined, usually followed by a stepwise descent g—f-e—d—c (bracketed in the example). All the cadences fall on c, e, or g.
Something of the same musical quality pervades Alma redemptoris mater (Ex. II. 10.3), possibly the oldest of the four best-known Marian antiphons (Brunholzl 1967), though not older than the ninth century. Its text, in hexameters, calls upon the Virgin's aid, in a manner not unlike that of many rogation antiphons. Regina caeli, with its joyful alleluias, is clearly an Easter antiphon. Its 'F-major' melody (F-mode but with bb throughout) features short melismas with repeat structure. Ave regina caelorum has a text in rhymed verse, in octosyllabic couplets, and is yet again in the 'major' mode, exactly like Speciosa facta, often pitched on c with an occasional bb; and as in Speciosa facta the triadic flavour of the melody is noticeable.
Salve regina is in the D-mode (Ex. II.10.3). The parallelism of the first two phrases, the rapid sweep through the whole octave d-D at 'misericordes oculos', and
10. Marian Antiphons
107
Ex. II. 10.3. From Marian antiphons Alma redemptoris mater zná Salve regina (PalMus 12, 303 and 352)
A!- -ma redempto-ris ma-ter
'^■2> •—
Sal-ue gemen-tes ...       il-los tu - os  mise-ri - cor - des o - culos ...
the insistence on a restricted number of melodic goals (nearly all cadences fall on D) mark it out as a relatively late piece of music, of the same post-Gregorian vintage as the other three. In all of them, the scale is clearly divided triadically, with a strong insistence on the fifth and octave.
Another feature of the antiphons, and particularly of Salve regina, is the frequency of short attributive, exclamatory, or supplicatory phrases: 'Salve regina . . . spes nostra . . . O clemens, O pia, O dulcis virgo Maria'. This type of phrase lent itself well to the clearly oriented musical phrases. Ex. II. 10.4 is another piece of this type. (Although we have just seen examples in this tonality notated on F and c, the sub-final here is always flat, so G has been chosen as final. A mode-6 differentia is nevertheless given once again. Transcription of the piece is not entirely without its problems.)
Text rhyme was an obvious way to bring such ejaculatory phrases into harmony. The history of the Marian antiphon here followed the same course as the office in
Ex. II. 10.4. Marian antiphon Aurei nominis Maria (Paris, Bibl. Nat. lat. 1139, fo. 138r)
-ä-9--
Au-re - i     no - mi-nis   Ma-ri - a       ma-ter   mi -se - ri - cor-di - e
m  0   ••'   • ť'*«t
pa-ra-di - si por - ta     pecca-to-rum refu - gi - urn speci - a- -le
P
O bo-    -na    O   pul-   -chra   O singu - la- -ris   post Deum spes no-stra
-9-— *■         írr                   ■■ -........		
A-•=-■ * ' ' 0	—z—A—a	*   ** rfS   ffk ^ .—
(f)—0-	—0—*4—■	-U   • • 0i 0-
Ma-ri-a Ma-ri - a      tu il - la magna   Ma-ri - a
					
A _n					
	..i .	*  m   mm   * m*	mm m		
	•			m *	
if U					
nos      De  - o  fi-li-o   tu-o     re-      -con-ci - li - a.
108 //. Chant Genres
Ex. 11.10.5. Marian antiphon Trinitatis thalannim (Paris, Bibl. Nat. lat. 1139, fo. 119r)
P P
«9 V * ***** " * * ** ^* ^
Tri - ni - ta - tis ta - la-mum an-ge - lo - rum glo-ri - am casti - ta-tis balsamum et lapsorum  ue - ni - am
m
et precetur
lau  -  de     totus        et  de - uo - tus      co-rus is-te uene-re-tur   Dei matrem
0
ut  nos    i - ta sub hac ui-ta mundos   om-ni  ui - ti-o     caros reddat fili-o
-m—*9-m—*—*r~z-*       ~*r
quod ip - si - us prece pi - us pi - e   matris      regno pa-tris
ip-        -so du-ce     ue-ra  lu - ce       fru-i     do-net     et co - ro-net
ut.   di - lec-tos et  e-lec-tos        lau - re - a tri-um-   -pha- -li.
general, and there are stylistic links with the Benedicamus songs, conductus, versus, and cantiones which became popular from the 11th century onward. Ex. II. 10.5 gives an example of this type. (The D mode is indicated by the psalm-tone differentia for mode 1 at the end. Tonally it often behaves according to the pattern outlined in Ex. II. 10.3 (see especially phrases 2-4 and 13-14), though a countering C-E-G is sometimes noticeable (lines 15-17). Another feature seems more archaic, however: none of the numerous short phrases is repeated.)
The most popular of the antiphons, not surprisingly, received attributions to such worthy musicians as Hermannus Contractus. Very few of these attributions have stood up to critical scrutiny (for information on individual pieces, see Szoverffy 1964-5 and 1983).
11. In traits
109
11.11. INTROITS
(i) Introduction
(ii) Introits in Mode 3
(iii) Comparison with Office Antiphons and Responsories
Cardine 1947; Frogcr 1948, 'Introit'; Stablein, 'Introitus', A/G'G; Connolly 1972, 'Introits and Communions', 'Introits and Archetypes'; Steiner, 'Introit', AG; Hucke 1988, 'Fragen'.
(i) Introduction
The introit may be considered as a special festal type of antiphon, the more so because in earlier times it seems to have been sung with several psalm verses, perhaps even a complete psalm. In the earliest books with chant texts, of the eighth and ninth centuries, only one psalm verse usually remains. In some cases, however, another verse is to be found, the Versus ad repetendum, giving the overall form (I = Introit antiphon, Ps = Psalm verse, Gl = doxology, VadR = Versus ad repetendum):
I-Ps-I-Gl-I [-VadR-I]
These Versus ad repetendum appear in both the earliest Frankish sources and the Old Roman ones, also for the communion.
Practically all introit texts were taken from the Bible, two-thirds from the psalms. Those for the first seventeen Sundays after Pentecost have texts in the numerical order of the psalms.
Introit melodies are not easy to typify. They do not fall into melodic families, nor, apart from a few openings and cadences, are standard phrases to be found which link chants across the repertory. This means that they seem much more individual than most office antiphons. Of course, there are far fewer introits in most medieval chant-books, less than 150 in the earliest sources. Antiphoners and breviaries will contain ten times that number of antiphons, which makes multiple use of melodies or parts of melodies inevitable. With the introits it is almost as if there existed a conscious desire to^make each chant recognizable in its own right, rather than simply part of a class or genre.
Connolly (1972, 'Introits and Archetypes') made the interesting observation that some Old Roman introit melodies are closely related to each other, whereas the Gregorian melodies for the same texts are not. Yet a similarity is discernible between each individual Old Roman melody and its Gregorian counterpart. This suggested to Connolly that introits once sung to the same basic melody (something like the Old Roman version) 'grew apart' on the way to achieving their Gregorian state. It could equally be argued that melodies once different (as the Gregorian ones are) grew more similar between the time of the Gregorian redaction (ninth century) and the date of the earliest Old Roman sources (eleventh century).
If standard phrases cannot be identified, how are we to come to grips with the
110
//. Chant Genres
substance of the melodies? What makes them appropriate for their liturgical function? The following should of course be regarded merely as preliminary remarks about only a small part of the repertory, which may suggest lines of approach for the rest. In describing melodies I have found it useful to concentrate on such features as the pitches most favoured (usually some sort of gapped scale is utilized, where certain notes, often E and b, are avoided or approached and quitted only by step), ornamental figures, the role of recitation, and the degree of mobility in individual phrases (their range and rapidity of movement).
Among the 140-50 introits in the earliest sources, there are about two dozen F-mode melodies, about thirty G-mode, about four dozen D-mode, and about four dozen E-mode melodies. The distribution throughout the year is not very even. For example, modes 1, 2, and 3 provide the majority of introits for saints' days. D-mode melodies are absent from Passion Sunday to Pentecost. Melodies in the same mode rarely occur in close proximity, an exception being the three mode 3 melodies on the Ember Days at Whitsuntide (characteristically, however, these are no more closely related melodically than others in the same mode).
(ii) Introits in Mode 3
In the oldest sources twenty-six melodies are assigned to mode 3. These may give some idea of the cohesion of the introit repertory as a whole. All quotations are taken from Hansen's transcription of the Dijon tonary, Montpellier, Faculte de Medecine H. 159. In this source the chants appear in tonal order, so that it is relatively easy to compare melodies for their likenesses and dissimilarities.
The openings are the most conventional feature of the mode-3 introits. Nearly all begin on G and rise to c in the first phrase, perhaps with an additional preliminary clause starting on E or F. Ex II.11.1 shows progressively more involved openings which follow this basic pattern:
(a) simple rise G-c, the first syllable being acented;
(b) rise through a\ but Benedictte retains G (in this source: the treatment of such cliches occasionally differs from manuscript to manuscript);
(c) a decorated form of (b), where the second syllable in a three-syllable word is accented;
(d) preliminary E, ED, or FED; the last two examples, Ego autem and Ego clarnavi, use the decorative turn in (c);
(e) the rise to c may be delayed within the lower tetrachord D-G;
if) shows a flourish around the final c; the substitution of adc for the simple ac is very common in other phrases also.
The choice of opening depends on the importance of the opening words in the text as a whole: in (a) the opening word is one of the most important in the phrase, whereas in (e) the key words come later. Then there are the usual adjustments for accentuation. Many of the ground-rules for introit openings have been outlined by Hucke (1988, 'Fragen').
11. In twits 111 Ex. 11.11.1. Mode 3 introit openings (Montpellier, Faculte de Medecine H. 159)
A     f'm *	
-	-0-
In-tret
Ti-bi
Ka-ri-tas Om-ni-a
	.  P  9 *	-pr-i-	# * 49     0m     mi_,
V? * -1 In de-o	W In no-mi-ne	w—w •- - Libera-tor me(-us)	w   w w Bene-di-cite
I 0 w
Lo-que - tur Ti-me    - te
1—■—	I'm a a	-rm—m—		rm   a a
$ 07, • •	• a 1-» -1	1   -» ? * \	1 ,M* p d	-m  mmm f,  a '0-
Ul
fen
0 *
Cm §*'
/fa »
-#-0
Dum cla-marem   ad  do-minum     Vo-cem iu-cun-di-tatis     Sacer - do - tes tu-i do(mine)
					
-"^"^—	-^-		-n—	.-0*0	»0 *
(0) <<(a '»1	~0*~	-0*—	-*_		
Ec - ce Cog-no-Sperent	0 -	cu -vi in	- li te	do -do -do -	mi-ni mi-ne mi-ne
The cadences are not quite so conventional, except in the last few notes. Practically every cadence touches on the notes E—G—F-E in that order, naturally with repetitions or extra insertions; it seems important that the final E be anticipated or prepared. As we shall shortly see, many phrases tend to hang around either high c or middle G, and most cadences proceed from one or the other of these two. The cadences, illustrated in Ex. II. 11.2, are grouped as follows.
(a) a rapid descent from c;
(b) more gradual descent from c; apart from the first example, 'mihi caro', these have the most common closing figure EGFF FE; some show an initial rise to high c before the run-in to the cadence;
(c) bb in the closing phrase;
(d) cadences which proceed from an orientation around G also favour the EGFF FE ending, otherwise aGFGFE E.
These examples do not, of course, exhaust all the openings and cadences of the mode 3 introits, only those which may easily be compared. What is noticeable, in fact, is that so many chants are dissimilar. A good deal of the likeness between responsorial chants was due to the presence of cadential melismas, and these are absent here. Of course, the phrases are moulded in accordance with the words being set to music, and
112 II. Chant Genres
Ex. 11.11.2. Mode 3 introit cadences (Montpellier, Faculte de Mcdecine H. 159)
Intret oracio, Liberator Timete dominum
hi   h        - =
0* 000
0 *
-<5»j
do-In Deo
-mi - ne.
o- -mni   bo - no.
Si íniquitates
«") g0m
'0
mi- -hi Dum clamarem
ca  - ro.
De - us
Is - ra - hel.
Vocem iocunditate
0   !'» *
al -le-
lp-se te Karitas Dei
-nu-tn Loquetur
et.
-lu
ía.
/5k-áW-x--^i-	-mbiJ*---g\ • \-—b-—í95-T>-^-
(eh  .  ••™0MiM0 pmm ^	-*-*-méHé   m*m-*  »m —^00 ■}
al-le-Ecce oculi
-lu
con - uer-tun - tur
Sacerdotes tui
ad ip - sum.
Dura sanctificatus
ftk-■■■m**S-x-----	--7*^rV-~^mr-	-T^m—z—s___—--——
g)                     *•»            phi 0}	-• *q—	-•   '0»m«m* »2
al - le-
-lu - ia.
Chri - sti
tu- -i.
spi-n-tum
no
uum.
Tibi dixit	In nomine	Ego autem	Coqnoui
-q-■-			
Äk--	-_ M	—-"tm-	
(rt)    m   • 0Um0ff,		-•    m'0m»^ t	1-  ' Wi, 0 1
tu-um a
me.
pa-
-t ris.
su- o-
-rum.
re-pel-
-las.
variety is only to be expected, but even so, surprisingly few cadences are actually identical apart from the last seven or eight notes.
Such similarity as exists concerns either small ornamental turns around one of the structurally important notes, or the general character of the phrase. By the latter I mean such features as the following: many phrases begin with a rise up to c, then dwell on c before falling down to a lower note, b, a, G, and E being the most common. Within this broad outline a great deal of variety is naturally possible. One cannot, however, characterize these phrases much more specifically. Among the mode 3 introits, phrases of this type outnumber all others, many introits having more than one such. A good proportion centre on G. Other phrases with a clear focal pitch are much rarer, and phrases which move directly from one pitch to another, or avoid a point of repose in some more convoluted way, are also relatively uncommon.
Ex. II. 11.3 gives examples of phrases which dwell on c, almost as on a recitation tone, and fall to G. There seem to be several figures which may be used to decorate the c: ccc, ccca, cdc, cdcc, perhaps acbc and cabcbc as well. The ways of attaining c from a lower pitch have already been seen (Ex. II. 11.1). There is little consistency in the descent to G, though nearly all progress through c-a—G, seen most simply in 'dominus', 'illis', and 'laude tua'. The last three examples are more highly ornamented than the rest.
11. Introits
113
Ex. II. 11.3. Recitation around c in mode 3 introits (Montpellier, Faculte de Medecine II. 159)
■P-              -   T-1	
/L—* "* C * * * WV——	ra  a   a   a'am  rm   m m'äa
tfy-■-**#—*—	—«-«-■-**#—
I
ha- -bi-tans in il-Vocem iocunditatis
-lis In-tret o-ra- -ci - o me - a Liberator Timete
am aau
^3
■#—
li - be-rauit do - minus     e-xal  - t<
bis
me     inquiren - tes au-tem do - minum
Repleatur		Sacerdotes tui _ i-:=»=:---1
	a a a aaa a ,f.h< ^	
V «»g i* *-		
Re-ple - a-tur os meum laude tu- -a propter Da- -uid ser - uum tu- -um Dum sanctificatus Vocem iocunditatis
al aa.
-quam
mun - dum     nun-ci - a - te us-   -que ad extre-
•mum terre
The recitation, if we may call it that, may not restrict itself to one note, but be a sort of oscillation between two poles. In Ex. II. 11.4 the twin poles of G and c seem to have equal attractive force. It is no coincidence that b is rare in the last example, and that we have been discussing openings and cadences with G, a, and c as the most important pitches. This reflects the pentatonic orientation of very many chants, where K and b are often avoided or approached and quitted only by step. In effect, then, a large proportion of the phrases in mode 3 introits adumbrate the rise G—a-c, dwell on c as a recitation note, then fall to b or descend pentatonically to a, G, or E. The same sort of thing may be seen in other introits, except that the selection of structurally important tones will be different. For the plagal modes (mode 4 to an even greater extent than the others), F is important for recitation.
Ex. II.11.4. From introit Ecce oculi (Montpellier, Faculte de Medecine H. 159, p. 35)
	
m         mmm                     ' m\	___r—- ___                                f ™t    **     a* _^_
rm               *    ■ *  * a     'a   r adh	j-    km   m    at    Mm   mUM    Cm   m                *D 9mT
"y                             '     am   a a	W       WWW             W    £w    W                        v w
su-per ti - mentes    e-     -um    sperantes in  mi-se - ri - cordi-a e - ius
u___■_«ö_«9_		
fin-,Q — m-.y <m\*ml ml	^ ram*	
[Pd-•-* *6——	—i*-a-	—■—a-♦*-**•" a}—
al - le - lu--fl-	-ia      ut e-	-ri-piat	a	mor-te	
-ji--		/■» aaa			
rm         a   a*   "ia a~m		_ m		* m	mä9 g> Q      -m jf*_____
		war		v *	•    • maa raaaa
a - ni - mas e - o-        -rum    quoni - am ad-iu - tor et protec - tor noster est
114
//. Chant Genres
(iii) Comparison with Office Antiphons and Responsories
While the office antiphons share with introits many of the same structural tones and melodic goals, their mainly syllabic style makes them less static. The introit can achieve in a short melisma the melodic movement that would need several syllables in an office antiphon. In the introit there is therefore more time to dwell on recitation notes. Nor would the ornamentation of recitation notes be appropriate in an office antiphon.
Ex. II. 11.5 gives the mode 3 antiphon Dominus legifer noster, a member of a melodic family already cited (Ex. II.7.3), and underneath it the introit Timete dominum. It is not suggested here that the two melodies are directly related; but their melodic outlines, at their very simplest, are at least comparable. The example is
Ex. II.11.5. Antiphon Dominus legifer noster (AS, k) and introit Timete Dominum (Montpellier, Faculte de Medecine H. 159, p. 44)
r» x.
Do-mi-nus legifer noster
t r*^* jl t hi 0-m föh ^ , p • j*mz
Ti - me - te do-mi-num om - nes sancti e-ius
mm*     m t
do-minus rex noster
ft   t   * «i   *   »        •}   £• IjmjV-
-0-0-
quo - ni-am ni-hil     de-est ti-menti-bus   e- -urn
ip-se ue-ni-et
0,   0    0    0    0    0*    0     0   0   0   0   0m *mta*0*
di - uites   e-gu - erunt   et e-xu-ri-e- -runt       inqui-ren - tes autem do - minum
0 <h
et sal-ua-bit nos.
0 Bji0   0m*iUm   *0   0"«"" «j   C~0&j m
non de -  fi  - ci-       -ent  om-       -ni   bo - no.
11. Introits
115
intended to display the difference in musical manner between the two. The first phrase of each rises to c, dwells there, dips temporarily to a, then cadences from c down to b. The second phrases are similarly comparable, except that the introit cadences on E instead of G. The next introit phrase, a recitation on G, has no counterpart in the antiphon, but the last two phrases are comparable.
The ornamental figures in the introits occasionally recall those of responsories. Although it is generally true that Gregorian chant genres do not share melodic formulas, some introits have figures used elsewhere. For example, the more ornate cadences in Ex. 11.11.2 and 3 may be found in office responsories. Compare:
Ex. II. 11.2 (d) Ecce oculi, etc. with Frere in AS, Intro., 31 (E2);
Ex. II. 11.2 (c) Karitas del with Frere, 34 (E3);
Ex. II. 11.3 Sacerdotes tut with Frere, 30 (g2). This does not mean, however, that there is a generic link between introit and responsories (and even if there were, which would come first?). We are dealing here with the common coin of Gregorian chant.
Nevertheless, most of the ornate cadential figures which provide strong links between responsories are absent from introits, and the responsories are richer in short melismas of 7-11 notes. Furthermore, the groups of repeated cs or Fs which characterize introits are rare in responsories. But these are in any case surface details. Wrhat of the melodic backbone? Scanning Frere's examples of mode 3 and mode 1 responsories (AS, Introduction, 29-32 and 17-28 respectively), one does indeed find similarities. Yet—and despite the obvious presence of recitation—most phrases in responsories are more mobile than those in introits, that is, they move more readily through the range of a fifth or more.
Much more space would be needed to characterize successfully the introits of other modes. Since it plays an important part, a few remarks on recitation may conclude this section. The role of either E or c as reciting notes in all modes deserves further study. Only these two notes carry groups of two, three, or more repeated notes as an
Table 11.11.1. Reciting notes in introit melodies Mode    Favoured reciting notes       Psalm verse
lower higher
1	F	a	a
2	F	F	F
3	G or a	c	c
4	F	G (rarely a)	a
5	a	c	c
6	F	G, a or c	F, a
7	c	d	d
8	G (rarely a)	c	c
116 //. Chant Genres
ornamental figure (the so-called repe reus si). The favoured notes in each mode, usually divided between higher- and lower-lying phrases, are given in Table 11.11.1 (together with the reciting note for the corresponding psalm tone). Since F is the final, and a and c are structurally important notes, mode 5 introits are particularly triadic. Mode 7 introits, on the other hand, display a tension between c and d, the lower phrases hanging around c, the higher ones not able to sustain any reciting note higher than d.
11.12. COMMUNIONS
(i) General
(ii) Groups of Communions with Psalm and Gospel Texts
(iii) Some F-mode Communions
(iv) Communions and Responsories
Stablein, 'Communio', MGG; Murphy, 1977; Hucke and Huglo, 'Communion', NG. (i) General
The text of one of the earliest communion chants was Gustate et videte quoniam suavis est dominus, from Ps. 33, documented in descriptions of many early rituals. (This verse survives for the mass of the eighth Sunday after Pentecost, though we have no means of knowing how its music relates to the chant of the early churches.) At some time a cycle of chants was developed so that each mass had its own proper communion. Like the introit, the communion chant appears with a psalm verse or verses in the earliest chant-books (list of sources in Huglo 1971 Tonaires, 401-2), but these fell out of use after the turn of the millenium.
Communions are among the most puzzling of chants. There are inconsistencies and irregularities in almost every aspect of the repertory.
Their texts are practically all biblical, but only a little over two-fifths are taken from the psalms. Most of the rest come from the Gospels, and often seem to sum up the theme of the mass in question; quite often they are therefore a verse from the gospel of the day, or at least from the same chapter of the New Testament. No other chant genre has this type of selection of texts.
Communions in the oldest sources are roughly as numerous as introits, but the proportions assigned to different modes are different (where the sources agree, for the modal tradition of many communions is unstable). D-mode melodies (over forty) are the most numerous, closely followed by G-mode (around forty). F-mode melodies (around three dozen) overtake E-mode pieces (less than thirty).
Early descriptions of the performance of the chant, and the fact that early sources have a psalm verse, permit one to call the communion chant an antiphon with psalm. In this it is directly comparable with the introit. Yet in musical character it resembles the introit only intermittently. Some communions are very short, others as long as the
12. Communions
117
most ample introits; some are like simple office antiphons, others much more elaborate. Bomm (1929) discusses no less than forty-six communions over whose modality medieval manuscripts or theorists disagree (see also Fleming 1979). In some of these cases quite different melodies are in fact at issue. One group of Lenten communions with gospel texts appear to have had very simple melodies at first, which were frequently replaced by later editors in order to bring the chants musically into line with the somewhat more ornate style prevailing elsewhere. If this is what happened (it has been argued that the simple melodies may originate in the seventh century), it supports the idea that the communion repertory is composed of several stylistic layers (for the Lenten communions see VI.6.x below). Another peculiarity of the repertory is that a number of communions share a text with an office responsory, and sometimes the music of the two chants seems to be related.
Such is the variety among communion chants that one suspects they may have been culled from several different sources. Perhaps there was something haphazard about the compilation of the repertory, new antiphons being brought in from, say, the office, when a new mass formulary was introduced. Or perhaps older communions kept their honoured place when new pieces in more modern style were composed. The following remarks can pretend to do no more than highlight a few groups in this multifarious repertory.
(ii) Groups of Communions with Psalm and Gospel texts
There are several seasons of the year when communion chants with psalm texts predominate, others when a block of gospel texts appears. On two occasions the psalm texts are arranged in numerical order: for the weekdays of Lent and for the Sundays after Pentecost. These arrangements could be retrospective, that is, the original order could have been non-numerical (Murphy 1977, 144 ff. argues, however, that theme plays a role here as well).
In the case of the post-Pentecost Sundays, the numerical ordering affects the introits and offertories as well. In the communions the selection of texts is not quite consistent, for in both Lent and post-Pentecost series some gospel texts intervene.
The sources of the texts for most communions, as for other chants, can easily be checked in Wagner (see the table at the end of Wagner I). Besides the two long series of psalm texts in Lent and after Pentecost, there are other groups of psalm and gospel texts:
Psalms:
Sundays from Septuagesima to Quadragesima Sunday 4 (mode 1, 8, 1, 3, 5, 1, 4)
Gospels:
the three Sundays after Epiphany (mode 1, 6, 7)
Easter Sunday, and ferias, to the fourth Sunday after Easter (mode 6, 6, 7, 8, 7, 1, 2, 6, 2, 8, 8)
Sunday after Ascension to the end of Pentecost week (mode 4, 5, 7, 8, 5, 3)
118
//. Chant Genres
From a musical point of view, however, there seems to be little stylistic unity within these groups. There are, of course, the common phrase openings and endings (though not the lengthy melismas which mark cadences in responsorial chants) which can be found throughout the chant repertory. But consistency of form is absent. As with introits, many phrases may be construed as intonation-recitation-cadence, but beyond this it is difficult to go, for the ways of elaborating this simple idea (and other basic shapes) are infinite.
(iii) Some F-mode Communions
The group of Eastertide communions with gospel texts includes three in mode 6 which may serve as examples of elaborated recitation (Ex. II. 12.1). The only common cadence which the three use very often is FGF F ('veritatis', 'Petro' and the succeeding 'alleluia', 'alleluia' at the end of Mitte manum). Another cadence common in the F mode, FGbbaGFG GF, appears only once, at the end of Pascha nostrum, and then in modified form. There are three C-cadences ('epulemur', the first alleluia at the end of Pascha nostrum, and 'fidelis', the last two being the same), otherwise
Ex. II.12.1. Mode 6 communions (Montpellier, Faculte de Medecine H. 159) 69
Pascha     nostrum       im-mo-la-tu(s)      est Christus        al-le - lu - ia
i - ta- -que e-pu - le - mur in a-   -zi-mis sin-ce-ritatis et ue-ri-ta-tis
al-le-        -lu - ia      al-le-        -lu - ia    al-le- -lu-
71 (transposed down a 5th)____
Sur-re-xit   do - minus     et ap-pa-   -ru-it  Pe-tro     al- -le- -lu-ia.
71 (transposed down a 5th)_
|)b .   &   *   .   m%   Q   .   ........L, ^
Mit-te  manum tu-am   et  cognosce lo-ca clauorum al-le-lu- -ia
et no-li es-se incre-dulus sed fi-de  -  lis     al-le-lu - ia   al - le- -lu-ia.
12. Communions
119
practically all cadences are on F. It is also difficult to construe many of the phrases as anything other then decorated recitation around F. For example, 'itaque epulemur' has a momentary dip DCD, immediately balanced by touching on aga, so that the central tone F is not seriously challenged. In the first phrase of Mitte manum, G features as an alternative to F, mostly for accented syllables; a is secondary, using some of excess energy of the rise from F to G in ornamental turns. The basic shape of the phrase is bipartite, with both sections rising to G:
FFG   G   F G || F  F GFFFFFG Mit-te ma-num tu-am || et cog-nos-ce lo-ca cla-vo-rum
Flad the use of standard phrases been the norm, the 'alleluia' phrases in these communions could easily have been identical. Those in Pascha nostrum are least striking. Surrexit dominus has already started out more adventurously. Its 'alleluia' then moves unexpectedly out of the F-bb tetrachord down to the C-F range. There are repeated groups of notes in both these clusters: ab aG—ab aG . . . FDEDC—FDE DC. In the absence of a notational sign for Eb, the whole chant is transposed up a fifth in some sources. In Mitte manum (transposed for the same reason), the long F— G recitation just cited leads into an 'alleluia' which cadences on G by way of the trichord Eb-G-bb. The final 'alleluia' phrases are different again.
Dicit dominus implete, for the second Sunday after Epiphany, has features in common with these, but develops an interesting sequel which seems to be a direct response to the text being set. The underlying notes in the first phrases seem to be:
Dicit dominus F—a . . . -G
implete hydrias aqua F . . .
et ferte architriclino F-G-a . . . —G-F
dum gustasset architriclinus F'-G-a—a—G-F-G-G—F
Ex. II. 12.2. Communion Dicit Dominus implete hydrias (Montpellier, Faculte de Medecine II. 159, p. 61)
									
A									_t\tm   .-I__. _
fm"   rm *								i°  ■   *n '	
\(\) m	w		m*      /a m nt u mmm		mm*m	m			
V Dicit	do-	-mi -	nus     implete hy-dri -	• as	a -	qua	et	fer-te archi	-tri-cli- -no
									
A\-ip*-			-*■" ■-Cm K\-->S-		■2—»5-			—„U-m*	^_-
w—* * *		mm-fa	m *m-		*0—		m'm	—0-" 0-	—°   m* m-
dum gus-tas - set     archi-tri - cli-   -nus     a-qua ui-num factum di-       -cit sponso
ser-ua-sti ui-num bo-num   us-   -que adhuc hoc signum fe-cit Hie-sus pri-mum coram dis-ci - pu - lis su-is.
120
//. Chant Genres
The words of the guest, 'Thou hast kept the good wine until now', move right out of this range, which is why the communion is sometimes classed in mode 5 instead of the mode 6 suggested so far. A further surprise comes in the last line, a simple narrative statement: 'This beginning of miracles did Jesus before his disciples.' For the first time the chant is almost completely syllabic.
(iv) Communions and Responsories
Since both communions and responsories draw upon texts which have been read as lessons, instead of relying as heavily as other chant genres on the Book of Psalms, it is perhaps not surprising that they share a number of texts. In the Old Roman chant repertory some of the melodies are also shared (Murphy 1977, i. 481 ff. lists twenty-one cases of near identity). This is much less evident in the Gregorian repertory: perhaps the case could be argued for three or four melodies. One such is Diffusa est gratia, given in Ex. II. 12.3 in its Sarum version both as communion and responsory (the verse of the responsory is omitted). It need not trouble us unduly that standard F-cadences appear in the communion, standard Zs-cadences in the responsory. The similarities of underlying phrase shape seem largely independent of this, except perhaps at 'propterea', where the communion rises a fifth F-c and cadences on F, the responsory-rises a fifth D-A and cadences on E. The first phrase in both, 'Diffusa est gracia', dwells on F before descending to D. After 'propterea', already mentioned, the two melodies reunite during 'benedixit' and run similar courses to the end.
Communions, therefore, are highly inconsistent in style, and point in many different directions, musically and liturgically. The impression remains of a fragmented repertory, the investigation of whose layers of material is one of the most intriguing tasks facing scholarship.
Ex. II.12.3. Diffusa est gracia as communion (Montpellier, Faculte de Medecine II. 159, p. 68) and responsory (AS, 663)
communion
I
Di-fu - sa est gra - ci - a in la   -   bi-is   tu- -is
responsory
Dif-fu-sa est gra-ci-a in la-bi-is   tu- -is
j0   0"   i*0U.0   «.      g0   .   fr   ,   , _   ^ *■». 0»'<
pro-pte - re-        -a      be-ne-di-xit te De-      -us in    e- -ter - num.
pro-pte-re-     -a be-ne-di-xit te"   De-      -us in    e-      -ter - num.
13. Offertories
121
11.13. OFFERTORIES
(i) Introduction
(ii) Texts
(iii) The Melodies of the Offertory Respond
(iv) Verse Melodies
Ott 1935; Sidler 1939; Baroffio 1964; Steiner 1966; Hucke 1970; Dyer 1971; Kahmer 1971; Pitman 1973; Baroffio and Steiner, 'Offertory', NG; Dyer 1982; Levy 1984; Offertotiale triplex.
(i) Introduction
The splendid offertories of the medieval Gregorian repertory are still among the least known of chants. The long verses which were sung in the early Middle Ages, but which fell out of use in the twelfth to thirteenth centuries, do not form part of the modern Roman liturgy and were consequently not included in such books as the Graduate Romanům and the Liber usualis. Although edited by Ott (1935—his edition was recently reissued in Offertotiale triplex, with the addition of neumes from early sources), and despite Sidler's study of 1939, the verses are unfamiliar. But the first part of the chant (which I shall henceforth call the 'respond') is also relatively poorly understood, for, like the introit and communion, it does not rely on easily identified formulas. Discussion of the melodies has therefore mainly concerned their possible responsorial nature, and the repeat structures in the lengthy melismas which occur in both respond and verses.
For two reasons it has often been supposed that the offertory originally consisted of a psalm with antiphon, as many verses as were required to cover the liturgical action. The first part of the offertory, the respond, is sometimes labelled 'A(ntiphona)' in medieval sources. Furthermore, the function of the offertory somewhat resembles that of the introit and communion: a chant accompanying the solemn entrance (introit), the bringing of gifts to the altar (offertory), the clearing up after consumption of the sacred elements (communion). The melodies that are transmitted by the earliest sources are, however, nothing like office antiphons, or any other antiphons, but long, melismatic outpourings as impressive as anything in the Gregorian repertory. Some have no verse, but most have from one to three verses; four are occasionally found. After each verse all or part of the respond was repeated, as in the gradual or office responsory.
Because of their generally wide range and frequent change of register or even of tonality, not least between respond and verse, it is not always possible to distinguish between authentic and plagal modes. The Dijon tonary (Montpellier, Faculté de Médecine H. 196, PalMus 8, ed. Hansen 1974), for example, makes a simple fourfold division between chants in Protus mode (D; 28 melodies), Deuterus (E; 29), Tritus (F; 16) and Tetrardus (G; 31). (One or two of these are multiple textings of the same melody, but the number is about the average for early medieval
122
//. Chant Genres
sources.) The modern books do make the distinction, however, since they restrict themselves to the respond.
All examples in the present chapter are transcribed from Montpellier H. 196. Because of its notation of both b and 6b it is a particularly valuable early witness, and it often appears to indicate chromaticisms and apparent modulations not found in other sources. Hansen's edition of the manuscript is the best way into the repertory. Ott's edition is based partly on Montpellier H. 159 and Trier, Stadtbibliothek 2254, the so-called Bohn Codex (see Steiner 1966 on the dangers of using Ott as a basis for melodic analysis). Facsimiles of sources with pitch notation are PalMus 13 (Paris, Bibliotheque Nationale, lat. 903), 15 (Benevento, Archivio Capitolare 34), and 19 (Graz, Universitätsbibliothek 807), and Thibaut, 1912 (St Petersburg, Saltykov-Shchedrin Public Library O. v. I. 6).
(ii) Texts
Most offertories have texts taken from the psalms (all but fourteen of the 107 in AMS, according to Hucke 1970). Hucke's study shows that more than a dozen offertories have for their verses the first verses of the psalm, the respond being taken from elsewhere in the psalm: this is a principle of antiphonal psalmody, where the whole psalm is sung, beginning of course with the first verse, whereas the antiphon is independent. In fifteen cases it is the respond which uses the first verse of the psalm, whereas in over twenty cases both respond and verse(s) select from later verses of the psalm: this method of selection resembles responsorial psalmody, where the cantor who sang the verses could select his texts as he pleased. Two dozen offertories have as first verse the first verse of the psalm, but for other verses select freely. Since the same types of text can be found among graduals and office responsories, this in itself does not tell us anything definite about the early history—antiphonal or responsorial—of the offertory.
The non-psalmodic texts have received special attention from Levy (1984), who argued for a musical connection between some members of the group and the old Spanish offertory, the sacrifiaurn. Levy's hypothesis is that these are the descendents of the Gallican chant repertory (see further VIII.6). It is nevertheless difficult to see much musical difference between them and other offertories.
(iii) The Melodies of the Offertory Respond
The responds of the offertories vary considerably in length, but generally consist of at least four clauses, often eight or more. The melodic style is highly ornate, and melismas of considerable proportions sometimes appear (as they do in the verses). Even in the shortest offertory there will usually be at least a couple of syllables with melismas of ten or more notes. As in the introit, gradual, and communion, reiterations of a single pitch, F or c, are frequently present. Although standard phrases are absent from the offertory, some chants actually contain phrases borrowed
13. Offertories 123
from graduals: Baroffio and Steiner ('Offertory', NO) cite examples in the offertory Superflumina, which uses mode 1 gradual phrases (Apel 1958, 351, Dl, somewhat extended, and cl). As far as openings, cadences, and underlying structures are concerned, however, the offertory is most easily compared with the introit, often seeming rather like a more ecstatic and wide-ranging expansion of the introit style.
As in introits and communions, many phrases are highly ornate recitations. Several mode 2 pieces are formed of little else but a constant oscillation between D and F, some mode 6 ones hardly make a single significant departure from F. Nearly a third of all the notes (over 190) in Reges Tharsis (mode 5) are c.
Ex. II. 13.1. Offertory respond Ad te Domine levavi (Montpellier, Faculte de Medecine H. 159, p. 205)
		
As-rm 'iTm'mn0 i	i—£*m**0—•»-	-■-0m    m"m.m"*   m    m»-m    mTm m-
qy  . •'•'•zzzzz—'	c_» -■-■_	-*V0 » -■-—M-- -** -mLZ mw-m_
Ad te do - mi-      -ne        le-     -ua - ui   a- -ni-mam   me - am
-v- ■						
/ ■	• a*00				ma*	■   * b«-   * • *
fn\			m v m			
vU						
De-	us  me - us	in   te con-	fi -	do	non	e - ru-     -bes - cam
						
A.	m± mmm				' — w	• »J«
MS	• m	mmm wm	•			a*mmm ml
VSD				0)		- — v--
ne-que   ir — ri — de-       -ant me      in-i - mi - ci   me- -i
£      m    iT. ^   .   ?   *   *i   .   , ^ f   mi**0'0     S0**k0 Iri
V 0 ' - - U0   ' —
ex. - e - nim u - ni-uer-si  qui te ex-pectant     non confun-        -den- -tur.
Ad te domine (Ex. II. 13.1) is an offertory respond of this type. Since the melody requires both bb and bh{, and since the second verse will explore the lower region between C and G, the melody is notated with final on a. Most of the melodic movement is therefore between a and c. Although several syllables are centred on c, only one phrase actually cadences on c, 'non erubescam', transposing literally the a-cadence on 'animam meam' for the purpose. There are two cadences on G, and one on the next lowest note E (there is noF), the other four all being a-cadences.
The chromatic inflexion which appears briefly here is but a hint of the much bolder chromaticisms to be found elsewhere, a feature of offertories which evidently caused considerable problems for notators using pitch notation (see Sidler 1939 and Steiner 1966). Montpellier H. 159, the earliest source to distinguish between 6tj and bb, preserves especially colourful versions of some melodies. In In die sollempnitatis, Ex. II.13.2, 'In die' at the start and 'alleluia' at the end have 6t], but the whole central section of the piece is notated with bb. The opening and ending have the same character as Ex. 11.13.1, that is, D-mode transposed up a fifth, but elsewhere the music is that of the E-mode transposed up a fourth. The dominating pitches are
124
//. Chant Genres
Ex. II. 13.2. Offertory respond In die sollempnitatis (Montpellier, Faculte de Mcdccinc H. 159, p. 215)
g ,   . . » •Vn? »««.'. to       fo «1Kb,««« ±A =
In di- -e sollempni-ta - tis  ues-   -tre   di-cit   do- -mi  - nus
^#**"" fr«*b.i»"l   <i»f«>.   bft. M, b#*^ fewHj     brf. b^».'.lF=
in - du- -cam uos in     ter- -ram flu-en-
-tern lac et mel   al - le- -lu - ia.
therefore bb and d, a step higher than normal for transposed D-mode. (Some sources do notate the melody in D-mode, but without the Zsbs which would result if the Montpellier version were transposed literally down a fifth: see Bomm 1929, 166 ff. and Jacobsthal 1897, 222 ff.)
As far as surface ornamentation goes, these offertories represent the more modest end of the reportory, with only a slightly higher overall degree of decoration than introits. Many offertories, however, go well beyond this. One finds phrases of text being repeated, usually with more elaborate music, a phenomenon practically unique in the chant repertory. The repeated notes can be spun out into passages as long as are to be found in graduals (compare Ex. II.5.3), and lengthy melismas may be introduced. All these may be illustrated by the respond of lubilate Deo universa terra, a celebrated example which has been cited, at least in part, many times before, but which never fails to impress (Ex. II.13.3).
Tonally this composition is quite unproblematic. There is no modulation, and the verses (discussed below) do not move into a different range. E and b are mostly avoided or used with circumspection, so that the basic scale within which the music moves is pentatonic, CDFGacd; bb appears only between two as (or, in the verses, at the peak of a motif such as abbGF).
After the standard opening, found in all sorts of D-mode chants, the first phrase is mostly concerned with a and c, including the notes which lead into a from below, EGa, and those that lead to r, Gac, or, just as often, aGc, which gives extra elan to the achievement of the top note. The text is then repeated with much more extended music: 'Deo universa' is the same as the first time, but 'terra' is more elaborate. The first word is set to a superb descending then ascending melisma; there is a rapid descent to F (note that the ornamental GEE is preferred to, say, aGE) then to D (ornamented DCD), followed by a gradual rise made up of little starts and pauses, rather like the ascending flight of a bird, hovering then climbing higher, D-E-a-c-d and just touching on e and/before tumbling back to 'Deo' as it was before. (For a
13. Offertories 125
Ex. II. 13.3. Offertory respond Iubilate Deo universa terra (Montpellier, Faculte de Medecinc H. 159, p. 199)
., • #--—-
Iu-bi-la - te   De - o u - ni - uer - sa ter-
ra
• v 'V"^*--—
iu-bi - la- -te       De - o
0*0 *i g*   0' m* * *   *    *• * * »M«
u - ni-uer-      -sa   ter- -ra
psalmum di-    -ci - te       no- -mi  - ni    e- -ius
f0   •U*  m\ _    1-007*   *\ m\   rm  mmm r
* *  • **m *m9
ue-ni - te    et   au-di-     -te      et        narra-bo    uo - bis
gifi&i *r"" *b0.0 . a* <s     & *       £* '**    0) fi.00
om- -nes     qui ti - me - tis   De-um     quanta fe-cit   do-mi  - nus
puu i0Ht 'i*' * *7# # if «>g;     fo ^i^'st'V*^ mj^j0 ^
a- -ni-meme-e al - le- -lu- -ia.
similar case of elaborated repetition see Iubilate Deo omnis terra, cited by Baroffio and Steiner, 'Offertory', NG.)
The next phrase, 'psalmum dicite', is once again made up of FGa and Gac groups. The first rise to c, 'psalmum di-', and the second, 'nomi-', are actually the same, except for the extra ornament bdc the second time.
The section to be repeated after the verse now begins. The first phrase, 'venite . . . vobis', is not at all static in the way most of the music has been up until now. All syllables have more than one note, and the melodic line traces two broad curves. It is not really possible to speak of main and auxiliary notes here. The mobility persists to some extent until the end, for the repeated cs for 'omnes' (the music is the same as for 'no[mini]') are the last of their kind, and F takes over as the main centre of attraction.
— ^ ... ,%tii
126
//. Chant Genres
(iv) Verse Melodies
Most of the features observed in offertory responds are present in verses, often somewhat exaggerated. There are repetitions of words and musical phrases, chromaticisms, and lengthy melismas, the latter being more frequent in verses than in responds.
A number of verses conclude with the same music as occurs in the respond just preceding the repeat section. The simplest form would thus be:
a Respond, first part: soloist(s)
b Respond, second part: choir
c Verse: soloist(s), using cadence from a
b Respond, second part: choir
Where there is more than one verse, perhaps only one of them will display this feature. In Jubilate Deo universa terra (Exx. II. 13.3 and 6) both verses use the cadence from the respond.
There is frequently a hiatus between the ranges of verse and respond, perhaps even a change in tonality. Thus the D-mode Deus Deus mens (Hansen, no. 857) and its first verse Sitivit in te rise no higher than a, whereas the second verse, In matutinis, moves largely between G and c. Benedicam dominum is more clearly in mode 1 (authentic D-mode) throughout, as far as range goes; but the second verse, Notas fecisti, begins on bb, with a phrase which sounds as if it had been transposed up a fourth; and the whole of the second half of the verse moves in the range a-e, as if transposed up a fifth, before falling somewhat precipitately back to the final cadence on D. Sometimes different scribes in Montpellier H. 159 supplied different pitch notation, a clear sign of the difficulties caused by notating and singing melodies with unusual range. For example, in Ascendit Deus in iubilatione (Hansen, no. 866), the respond is unequivocally a mode 1 melody. But at the end of the first verse, Omnes gentes, the music starts to hang around bb, and ends with a cadence on F. The original letter notation had the subsequent two verses beginning on a and cadencing in identical fashion to the first verse on F. But a second set of pitch-letters was then added for most of the third verse and all of the fourth. The third verse starts off a fifth lower (on D), then about half-way through comes into unison with the first version, while at the end of the verse it rises above the first version, before cadencing like it on G. The third verse in its first version rose splendidly to highg. In its second version it is again notated a fifth lower at first, but the last phrase ('sub pedibus nostris') is a tone higher until the final cadence on F.
Tollite portas in many sources looks like a fairly typical mode 2 offertory, moderately elaborate, with only one large melisma at the end of the second verse. It is found in this form in, for example, Graz 807 (PalMus 19) and Benevento 34 (PalMus 15). Although verse 1 seems to end oddly—FG Gb aG GF—this is simply the lead into the repeated part of the respond, and reflects the corresponding passage in the respond itself. Verse 2, by contrast, occupies a higher register and would by itself be classified as mode 1; it ends on D.
13. Offertories 127
Montpellier H. 159 classifies this offertory in mode 8. But at first the melody is notated with the same intervals as in the other sources, only a fifth higher. It therefore looks rather like Ex. 11.13.1, and we should expect later 6bs to be the reason for the choice of pitch a fifth higher. But the 6bs appear in a quite unexpected manner. At the end of the respond, the melody suddenly dips down a tone, with bb replacing c as the usual 'reciting' pitch, and cadences on G, instead of the expected a (Ex. II.13.4). As Bomm (1929, 174 ff.) recognized, this is not likely to be a mistake, because (a) the same ending can be seen in St Petersburg O. v. I. 6, and (b) Frutolf of Michelsberg cites the D-mode ending in a way which suggests he is reproducing a correction for a well-known trouble-spot. (See also the discussion in Sidler 1939 43 ff.)
Ex. 11.13.4. From offertory respond Tollite portas (Benevento, Archivio Cap. 34, fo. 13r; Montpellier, Facultc de Medecine H. 159, p. 275)
Benevento
Montpellier
Tol-    -li-te por- -tas ..... rex        glo - ri - e.
Verse 1 in Montpellier H. 159 is notated a fifth higher than in the other sources. But verse 2 brings further complications. Here Montpellier is only a fourth higher than the other sources, until the closing comments of the final melisma, when it moves up a step and cadences on a. This seems illogical, for if it had continued only a fourth higher than the rest, another G-close would have occurred, as in the respond. This cadence is, however, not the last of the piece, for the last part of the respond will now be repeated as usual.
Ex. II. 13.5 gives the whole of verse 2 at the two contrasting pitches. Such examples make it abundantly clear that the adoption of pitch notation in the eleventh and twelfth centuries may often have resulted in the smoothing out of 'irregularities' in the tonal organization of many melodies.
The melisma in Ex. II. 13.5 has the form AAB, extremely common in alleluias, moderately so in offertories. In the Montpellier source, which in this respect is representative, there are 103 offertories, with 216 verses in all. All but a handful of responds, and over ninety of the verses, have no lengthy melismas at all. Some verses have more than one. There is a rough balance (somewhat over eighty examples of each) between unstructured melismas and those which have a repeat form of some sort. There are about fifty melismas with AAB form, but few other forms are to be found regularly; there are, for example, only five instances of AABBC form. Among the more extended schemes may be cited AAAB-C-AAAB (Super flumina), AABBCCD {Benedicam dominum), AABBCCCD (Deus enim firmavit), AABBCDDE (Benedictus es . . . in labiis), and AABCCDCC (Domine Deus meus).
128 //. Chant Genres
Ex. II.13.5. Second verse of offertory Tollite portas (sources as Ex. II. 13.4)
Benevento
Montpellier -
*00~0    0 0g'
Ipse super   ma - ri - a funda-        -uit e- -um   et super  flu-mi-   -na prepa-rauit
3 times
" mJ0m
p
2 times   „ _ _ ^
•     • 0*0* j|    * *• *     *       * *     *    mm/~0~      «     * mmlim
e- -um.
For examples of extended melismas, and of the carrying over of melodic material between respond and verses, we may return to Jubilate Deo universa terra, whose respond was given in Ex. II.13.3. The two verses appear as Ex. II.13.6. A blow-by-blow account of the verses is by now unnecessary. The reader will be able to locate without difficulty such features as the free recitation around particular notes (as at the start of verse 1). Sometimes the melodies become locked into repeated rs, with G and a in support (cf. the second 'tibi vota' in verse 1 and '[holocaujsta medullata' in verse 2). These ornate phrases lead into similar cadences: those for the second 'mea' and 'distinxerunt' in verse 1, and those for the second 'mea' and 'medullata' in verse 2 are all the same.
At the start of both verses the text is repeated: while the music in verse 1 is different the second time, in verse 2 the music is the same, except for the slightly more extended treatment of 'mea'. More surprisingly, this music is the same as the first line of the respond. Both verses have a lengthy melisma, and the two end identically. There is also some similarity at the start, where both melismas have brief repetition. Then verse 1 runs rather obviously from to c and back again. The centre part of both verses hovers around repeated rs, much more extended in verse 2, with brief descents to F for relaxation of tension.
These are purely musical outpourings, where attention to the text is suspended and sheer joy in singing seems to take over, ecstatic and improvisatory (at least when compared to the repetitious schemes found, for example, in many alleluias). It remains unclear, nevertheless, what occasioned the composition of these glorious melodies. They are not assigned regularly to the high feasts of the church year. The contrast with that other great musical high point of mass, the sequence, could not be more pointed. Whereas sequences were sung only on the greater feasts, the
13. Offer-tones 129 Ex. 11.13.6. Verses of offcrtorv Iubilate Deo universa terra (continuation of Ex. 11.13.3)
V.l (fo       •  'fih  »   nfi sjs *        ^ "V* '
Reddam ti-bi   uo - ta     me-    -a red- -dam
ti-       -bi uo-       -ta   me- -a
«7«'«'» "'*'«« >.        "« *«, .'^
que dis-       -tin - xe- -runt
la-
-bi-   -a   me- -a.
Lo-cu-tum est os me-um in  tri - bu-la-ti-o - ne   me- -a lo-cu-tum est os me-um       in tri-bu-lati-o - ne   me- -a
ho-lo-cau- -sta me-dul - la - ta
________________♦*»—
o-
-fe - ram ti- -bi.
assignments of the five offertories with especially extended melismas mentioned above are as follows:
Super flumina: twentieth Sunday after Pentecost* Benedicam dominion: Monday of second week in Lent Deus enim firmavit: second Mass of Christmas Day
130 //. Chant Genres
Benedictus es . . . in labiis: Quinquagesima
Domine Deus metis (in very few sources): twenty-fourth Sunday after Pentecost* (*the Sunday may vary between sources)
Jubilate Deo universa terra is universally assigned to the second Sunday after Epiphany.
11.14. ALLELUIAS
(i) Introduction
(ii) The Earlier and Later Styles
(iii) Rhymed Alleluias and Late Medieval Melodies
Stablein, 'Alleluia', MGG] Schlager 1965; Treitler 1968; Jammers 1973; Schlager, 'Alleluia. I', NG) Bailey 1983, Alleluias; MMMA 7, 8.
(i) Introduction
The alleluia is a responsorial chant in that the first part of the chant ('Alleluia') forms a choral respond to be repeated after the verse, while singing the verse is a soloist's task (there may be more than one verse, and more than one soloist). It is conventional to divide the respond into two parts, the setting of the word 'alleluia' itself, and the vocalization, melisma, or jubilus on -a. The method of performance indicated in modern books is:
Cantor: 'Alleluia'
Choir:   'Alleluia' + jubilus
Cantor: Verse (main part)
Choir:   end of Verse (which often includes a repeat of the jubilus) Cantor: 'Alleluia' Choir: jubilus
The early notated books known as cantatoria (because they contain only the music sung by the cantor, not that of the choir) contain complete alleluia melodies. This suggests the following manner of performance, where the 'Alleluia' call and the jubilus constitute an undivided respond:
Cantor: 'Alleluia' + jubilus
Choir: 'Alleluia' + jubilus
Cantor: Verse
Choir: 'Alleluia' + jubilus
Possibly a further two statements of the respond then followed, by the cantor and the choir respectively, as was originally the case with other responsorial chants (see 11.4). The even more elaborate performance schemes of Old Roman and Milanese usage are also suggestive (see VI11.3—4).
The outstanding work of Schlager in cataloguing and editing the medieval alleluia
14. Alleluias
131
repertory (Schlager 1965 and MM A 7-8) means that the bases for study have been more firmly established than for almost any other chant genre. Nevertheless, a consensus on many of its aspects can hardly be said to have been achieved. The alleluia has attracted considerable musicological attention, partly because it has seemed a more purely musical genre than most others (with its wordless jubilus), partly because of its mysterious early history and the continuous expansion of the repertory through the Middle Ages, partly because of its connection with the sequence (an extended alleluia or not?). All these matters have been the subject of controversy.
In the earliest books with chant texts, those edited by Hesbert (1935), there are just over 100 alleluia texts. Not all have their own unique melody, however; Schlager reckons that around sixty melodies were used (see the list in Schlager, 'Alleluia', NG). Prominent among the melodies used for more than one text are those for Dies sanctificatus (third Mass on Christmas Day, nine other texts in the early repertory), Dominus dixit ad me (first Mass on Christmas Day, eleven other texts) and Excita Domine (third Sunday of Advent, six other texts). (For melodies with several texts, see the synoptic presentation by Madfignac 1981-6.) Over the two centuries up to about 1100 (the cut-off point for Schlager's catalogue) the repertory expanded dramatically. Schlager records 410 melodies, plus a considerable number of others in adiastematic neumes which could not be catalogued: a sevenfold increase. The number of texts had increased at the same time to over 600.
Just as the same melody might be used for several different texts, so might the same text be sung to several different melodies, a phenomenon far more widespread among alleluias than any other chant genre, with the exception of hymns and late medieval rhymed sequences. So the total number of different combinations of melody and text is far higher than that of the 600 texts alone. After the eleventh century composition continued. Schlager points to the striking efflorescence of creativity in South Germany and Bohemia in the fifteenth century and the frequent use of rhymed texts.
A corollary of the continuous compositional activity is the variety between manuscripts in their selection of alleluias, both melodies and texts (which means that the alleluia repertory is a prime resource for identifying the liturgical use of a manuscript and its nearest relatives). This instability extends right back to the earliest sources, so that one is tempted to hypothesize that the repertory in the eighth century, that is the period immediately before the copying of the extant sources, was rather small. Apel (1958, 379) listed the alleluias of the temporale whose assignment remained constant across the eighth- to ninth-century sources edited by Hesbert in AMS: only twenty-two in all (eighteen, discounting repeats). The number of melodies for these is only eleven. The agreement in the sanctorale is minimal. The supposition that not many alleluias were needed is strengthened by the state of the Old Roman and Milanese chant repertories, which used very few melodies right into the twelfth and thirteenth centuries.
132
//. Chant Genres
(ii) The Earlier and Later Styles
Many of the melodies which have a constant assignment in the earliest sources share features of musical style, which suggests they belong to an early layer. The most important of these (Apel 1958, 391 discusses others) is a negative characteristic which becomes obvious when comparisons are made with the majority of later alleluias: the absence of repetition within the jubilus on -ia.
Ex. II.14.1. Alleluia Dominus dixit ad me (Paris, Bibl. Nat. lat. 776, fo. 12')
.11*5 mfMi
ii. * JL
JOG
***la
Al - le-
-lu-
-ia.
Do - minus   di-xit ad
me     fi-li - us   meus es
tu
,i_ l,
t   A   •» m "*     mm **l a   •   *s m
0 I **   l«* * 0*0   0   *< *m   0b b
e   -   go ho-
-di - e
«* *.«
•* 0*
§0
' 0a
»e - nu - i
te.
Ex. II. 14.1 gives All. Dominus dixit ad me with its presumed original melody (melody 281 in Schlager's catalogue) from an Aquitanian manuscript. So well known is the melody that the scribe does not copy it in full, and one has to refer back to the first appearance of the melody, on the first Sunday of Advent for All. Ostende nobis, for the ending. There are in all three melismas of some length: the jubilus, on 'ho-[die]', and on 'te' at the end of the verse. The only one with a hint of repetition is the last, which might be construed as having paired phrases at the end, one ending 'imperfectly' on F, the other making the 'perfect' cadence on G (bracketed in Ex. II. 14.1). In the respond, c is an obvious melodic goal. The melody keeps pushing up to it, then falling away, to G, to F, b, and finally G again. This is also what happens in the verse, with cadences on /; ('me'), G ('tu'), F ('hodie'), and G again ('te'). In the final melisma F exerts a strong pull, so that the G cadence seems like a point of balance or repose, rather than a point of departure towards c. There are of course other ways of hearing the piece, but the similarity with other types of chant discussed so far, particularly the offertories, seems clear. The melisma on 'hodie' is particularly suggestive of the offertory, with its graded descent from e to a, then d to G, and finally b to F. It is not surprising, either, to find that the final groups of notes at the ends of the respond and verse are also found in G-mode graduals (cf. G10 and Gl in Apel 1958, 356).
14. Alleluias
133
Ex. II.14.2. Another melody for Alleluia Dominus dixit ad me (Paris, Bibl. Nat. lat. 776, fo. 12r)
Al-le - lu - ia.
■f-			__ J Q			-
/ •	•  »              •           • #°			•		
rA\ m		•	•			m*b-     -            -m        *   m m
						
Do -	mi-         -nus di - xit ad	me	fi   - li-	us	me -	us es     tu      e- -go _ i——
Q liT*			a*			
						mm*              * «H
	•	•	• *»		>•	
vy				—w—	■*	
^ ho-			-di -	e	ge -	nu - i te.
Perhaps because this melody seemed old-fashioned, another setting of the same text was copied immediately afterwards in the source used here, Paris, Bibliotheque Nationale, lat. 776. The melody is unique to this manuscript (Schlager 1965, No. 190), and it turns out to bear all the hallmarks of later composition (Ex. II. 14.2). The music makes use of repetition to a high degree. Firstly, the jubilus has an obvious repeat (which should perhaps be made at the end of the verse as well). This music appears again for 'hodie', also repeated, and a good deal of the same phrase is used for 'Dominus' at the start of the verse. The music for 'Alleluia' is also used in the verse, from 'dixit' to 'meus', and one is even tempted to hear it behind the music for 'ego'. In other words, the music of the verse has been derived very largely from the respond. Since each phrase forms a melodic arch, no note establishes itself as a reciting pitch. (The manuscript has no clefs or coloured lines, and the choice of a-final, tantamount to D-mode with /?bs throughout, is a matter of opinion. Schlager transcribes the piece with /s-final.)
Among the sixty or so alleluia melodies assigned to the early (ninth-century) repertory by Schlager there is a rough balance between those with and those without melismas having a repeat structure. Probably other alleluias could therefore be added to Apel's eleven 'earliest of all' (eighth century or earlier?) melodies, such as those for the common of saints or for Sundays, which by their nature would have no firm assignment to one particular day.
A melody just emerging into the new era, so to speak, is that for/4//. Attendite, for one of the summer Sundays (Schlager 1965, No. 224), Ex. II. 14.3. (The semitone step is indicated often enough in Paris 903 to enable one to use Bb as a key signature; Montpellier H. 159 notates the melody with a flat throughout.) There is a hint of repetition at the start of the jubilus, but a more obvious example is reserved for the middle of the verse, on 'meus'. Even without this repetition, one would be inclined to see a later composer at work than in All. Dominus dixit ad me (Ex. II.14.1), because of the easy way the melody moves in scale passages through the range F to c.
134
//. Chant Genres
Ex. II. 14.3. Alleluia Attendite popule mens (Paris, Bibl. Nat. lat 903, fo. 122r) Al-    -le- -lu - ia.
		
	■ "a "a a	
		mM              ■ "a
		9 WB
		
Adten-       -di-     -te    po- -pu - le
_U_i_*■                      -»•                              -                     —         — —>__	
	
	
	
me- -us   in le - gem ue - stram.
One needs only to glance at the critical commentary to Schlager's edition to see how the shape of the longer melismas may vary from manuscript to manuscript. Simultaneously with the composition of new melodies, already existing ones were retouched, extended by repetition, made more regular. This affects not only the melismas but other phrases as well: for example, the start of the Alleluia for All. Attendite (Ex. II. 14.3) in Paris 776 is like the start of the verse:
D FDDCF Al-le-
Newer alleluias naturally exploited the full possibilities of repeat structures, and practically all of those which make up the rest of the repertory up to c. 1100 include a repeated melisma, or reuse of respond material in the verse, or both. The scheme AAB is extremely popular, and forms like miniature sequences appear, for example AABBCCD, used for the following melodies (in some sources at least):
All. Beatus sanctus Mart in us (Schlager 1965, No. 396)
All. Cum esset Stephanas (Schlager 1965, No. 102)
All. Swrexit Dominus et occurrens (Schlager 1965, No. 10)
Others display repetition with variation, as All. Ego sum pastor bonus (Schlager Melody No. 299, unique to Pistoia 120), which might be interpreted: AB A'B' A"B A"'B". The jubilus in/1//. Nuptie facte sunt (Schlager 1965, No. Ill, found only in a few Aquitanian sources) displays a subtle alternation of like and unlike phrases (Ex. II. 14.4). The small figure labelled 'x', or similar ones, appears four times in the jubilus. Four phrases in all end with an identical cadence, CDD, the so-called 'Gallican cadence' found in numerous sequences of this period. In the verse, the recurrent figure 'x' turns up once more, at the end of the 'Chana' melisma. Music from the 'alleluia' call is then reused at the end of the verse for 'mater eius', providing the lead into a repeat of the jubilus (indicated by a cue).
Thus, while the repetition in a short AAB scheme may be literal, the longer melismas tend to handle it in a more sophisticated way. In Ex. II. 14.4 different
14. Alleluias 135 Ex. II. 14.4. Alleluia Nuptie facte sunt in Ghana (Paris, Bibl. Nat. lat. 903, fo. 18r)
,_h_.   J, x (x)
- _' i   *      i-1 i--------1
Al-le-                             -lu - ia. _tr_ x _}s_ (oc)
Al-le- -lu - ia.
_tr_ x_ _ {f (x.)
' -1 i i     r i r--------1_
i
X
Nup-ti-e fac-te sunt    in   Cha- -na
c- C (c)
i i i-1 i-------1
y      * »**«,#Tl 1,%,^*^
Ga-li - le
et erat  i - bi Ihe-sus   et Mari- -a     ma - ter   e- -ius.
phrases were brought into relation by means of a common cadence, whereas Ex. II. 14.5, All. Veni sponsa Ch?isti (Schlager 1965, No. 35, likewise Aquitanian), shows similar beginnings leading to different endings, analogous to the first and second endings of later music. The jubilus consists of nine phrases, labelled 'b' to 'j'. The basic idea seems to be the alternation of phrases cadencing on a with those which move down to D. The a phrases are 'b' and 'e', which comes twice, the first time with a short, supplementary a-phrase 'F. The /^-phrases are 'c + d' (labelled separately because 'c' appears later alone) and 'g', while a longer ending includes a G-cadence ('h') before the final phrase ending on D. Within this pattern, however, there are others. The first notes of 'e' and 'g' are the same, and this causes 'e' and 'g' to be heard as a pair, with a half close (on a) and then a full close (on D). Three-note scale segments are common throughout the piece (a symptom of relatively late date): since there are pairs of them just before the cadence in both 'b' and'd', one is inclined to hear these phrases as a pair as well, in an antecedent-consequent relationship.
The verse begins like the 'alleluia' call, but the next phrase, 'sponsa Christi', stands by itself. The four phrases of 'accipe coronam' make up an A-A-A'-A" scheme: T is actually a version of 'c+d', transposed up a fifth; the next phrase uses 'c' literally, up a fifth, with a new extension 'm'; together they might be regarded as an extended version of T; the last phrase is made up of 'c+j', likewise transposed up a fifth. Is it stretching a point to call 'quam tibi' a variant of 'j'? and 'dominus' a variant of'm', each with a new start (the scale segment abed)?; 'j' then recurs at the end of 'preparavit'.
136 //. Charit Genres
Ex. II.14.5. Alleluia Yeni sponsa Christi (Paris, Bibl. Nat. lat. 903, fo. 104v)
j, _ (r c 6 6 /
#• li— li li -1 i
"^J      *•  •## j£—*---v--—- -*-
Al - le-        -lu - ia.
,_3_ _&__K__J_
i —i i i r~-11-1
Ve- -ni  spon-sa Chri-sti
._i   Gy. t+st)__i__~i d») et jt
n     1_w_ i i _ ii ii 1
((^)    w  g   »   • am'000H   m0m *     '»'i #»n  * * 'iffii   * * j  a —
ť . -
acci-pe co-ro- -nam
n +■ i/_ n. ■*■ mí i a,
i—        J it— ti------ -1 i -i
t..* i^., 0Üm0*mtmT^r ma t0   . =tg
quam ti - bi      do-   -mi-nus pre-pa-ra- -uit   in ae- -ternum.
The last phrase, 'in aeternum', takes up the 'alleluia' call again, leading into a repeat of the jubilus, indicated by a brief cue.
For some, an analysis like this may be over-ingenious, while others may disagree over detail. The important point is that repetitions and echoes of the sort described are unthinkable in earlier chants and mark a radical departure in compositional technique. The whole composition is held together not simply by its overall responsorial form but by a network of internal references and patterning. It is important that the references are internal, that is, they do not relate the piece to others of its class but contribute to the individual identity of this one 'work of art'.
(iii) Rhymed Alleluias and Late Medieval Melodies
Very many alleluias composed after the ninth century remained local compositions, such as the Aquitanian ones just given as Ex. II.14.2, 4, and 5. The same remains true of most compositions of the later Middle Ages, from the tw-elfth century onwards. This is mostly because the basic liturgy throughout the year had its consignment of alleluias from an early date, and while new initiative might win local acceptance, it was hardly likely to cause wholesale revisions of the repertory right across Europe. Many later alleluias are compositions for local saints, whose cult was peculiar to a restricted area, even to a single church. One of the few classes of feast-day which was celebrated with increasing and universal enthusiasm was feasts of the Blessed Virgin
14. Alleluias
137
Mary, and it is among the vast corpus of Marian alleluias that a few widely known modern pieces may be identified. Many Marian alleluias were retexted for other liturgical occasions (particularly for virgin saints, of course), some more than twenty times. As with the earlier repertory, some texts are found with more than a dozen different melodies in different sources.
One of these is All. Virga Iesse floruit (for which Schlager lists nearly 200 sources in MMMA 8, 822 ff.). Its text is notable for the incipient rhyme, best called assonance, common to many new liturgical texts of the twelfth century. In what seems to be the earliest version, the last line is not assonant:
Virga lesse floruit
virgo Deum et hominem gcnuit
pacem Deus reddidit
in se reconcilians yma suinmis
Other versions (MMMA 8, 564) replace the last line with:
qui poli summa condidit
orate (or 'Dei genitrix ora') pro nobis
Musically, this particular alleluia presents nothing radically new. Given the amount of repetition already present in earlier alleluias, however, it is not surprising that musical rhyme should occasionally be introduced in later pieces, either in support of, or in counterpoint with, the text rhyme. This occurs in, for example, All. Ante thronum trinitatis, another Mary alleluia, probably dating from the late thirteenth century. As a supplement to the alleluia with its usual text, transcribed by Schlager (35 ff.), I give the version adapted for St Barbara added in the sixteenth century to Paris, Bibliotheque Nationale, lat. 905 (from Rouen) in Ex. II. 14.6. The musical repetitions affect not just the ends of each eight-syllable line, but also some of the half-verses. The text has the following rhyme scheme:
-is -ata -is -ata -atis -iice -atis -iice
whereas the music proceeds as follows (the scheme is marked on the transcription):
a b c
d e c
a e c
e b (melisma with AAB structure, new ending)
The jubilus is by contrast relatively uninteresting (though it follows an AAB repeat pattern in some sources).
The melody is rather modest in scope. Phrases such as 'cunctis horis eiice' have an undoubtedly modern flavour, not so much because of the three-note groups as because of the rapid motion through the octave f—F. Every phrase begins or ends on F, c, or/, and most wing their way over the whole scale.
138 //. Chant Genres
Ex. II. 14.6. Alleluia Ante thronum trinitatis (Paris, Bibl. Nat. lat. 905, fol. 31 lv)
Al-   -le-     -lu - ya.
^       .'.7*       ■ " «. »tj . ............ * * \T
An-   -te      thro-num tri-ni - ta-tis Bar-       -ba - ra nunc premi - a - ta
*»..'  *.        " "     '   * . ' .   ** < «.
post   Ma-ri-am pi-e - ta-tis mater es et     ad - uo-ca - ta
^b y ».«*     * '   *"* * '........." «.
ser - uos     tu-   -os a pec-ca-tis cun-       -ctis ho-ris  e-       -ii - ce
*0  *0m m # • m00m m *' r     ** *009m
et       in   fi - ne cum be - a - tis   hinc precibus adii-ce.
But there are plenty of more flamboyant melodies among late medieval alleluias. An example which one could call celebrated if it were better known is All. Ora voce pia pro nobis virgo Maria (MMMA 8, 393, in a handful of South German sources), with a staggering range of two and a half octaves from G to ci. Somewhat less spectacular, though still exotic by comparison with early medieval melodies, is All. O Maria rubens rosa. The piece is of manifold interest.. Firstly, the numerous sources cited by Schlager (over 40: MMMA 8, 727 ff.) show that it was one of the most popular rhymed alleluias in south-east Europe; secondly, its E-mode melody is of a type characteristic for that- area (Schlager thinks it might even have been the starting-point for the type); and thirdly, in about a quarter of the sources trope verses are inserted in the verse.
The turn around E which constitutes the most prominent fingerprint of this group of melodies stands out clearly in Ex. II.14.7. Other features of the melody are perhaps best described as typically late-medieval, only applied here to an E-mode melody, rather than the more common F- or G-modes (or transposed to c): the readiness to run through the whole octave, or beyond (here practically always descending), the octave leap e to E, the succession of leaps of a fourth and a fifth (line 8), yet again outlining the modal octave. There are again numerous melodic repetitions. The source transcribed here has no trope verses. Asterisks show where they appear in other manuscripts, and they are given in Ex. II.14.8. (cf. MMMA 8, 351 ff.).
The repertory of rhymed alleluias shares compositional features with other rhymed chants, such as the responsories for late medieval rhymed offices (see 11.26). The parallel is not an exact one, however, because it was not the custom to compose whole
14. Alleluias 139
new masses with rhymed texts: the alleluia and sequence (not necessarily as a pair) are usually the only such items at mass.
Ex. II. 14.7. Alleluia O Maria mbens rosa (St Gall, Stiftsbibl. 546, fo. 327r)
$ . .....     "«....v * ..... •*'".'......"...       ^>"i...... I
Al-le- -lu-ia.
f!> .»... .     . " ".^..... .   ' "»".>^
O Ma - ri - a     ru-bens   ro- -sa  deli catum li-li- -um
_*3
dul-cis mi-   -tis  et formo-   -sa summum celi   gaudi- -um
V» 0
tu cel-sa Christi lu-ci-     -da                 tu ui-tis   co-pi-o- -sa -—*2——•—----—a-fa-—-—-. ji. .-•-—
# ■      *     t "' jj\
#   •    *      m0   *00    0    00000 0 "A.***** *
e - xau - di  nos mi-se-ros et o - ra pro no - bis pi-
9   '  **0   Í0 '     J , 0   •    *»0 '0, 0
et perdue   nos ad fi - Ii - um tu - um
uir-      -go   Ma-ri- -a
Ex. II.14.8. Trope verses for Alleluia O Maria rubens rosa (Munich, Univ.-Bibl. 2° 156, fo. 260r)
•         m   0   0   •   0   m mm00*'0 *  *  0         . 0*0^-0-
... Summum celi   gau-di-um et spes in te cre-den - ti-um O uir-ginum Ma-ri - a.
Al-le-lu-ia al - le-lu - ia ti - bi ca - nentes sub - le-ua tu ru-ti-lans au - ro - ra.
*2        Ma-ri - a candens li - li - um ... pla-ca tu-um   fi - li - um O dulcis et for-mo-sa. Alleluia (etc. as *1)
*3        ... Ut pe-tas au-xi-li-um        no-bis post hoc e-xi-li-um     tu mundi lu-mi-no-sa. Alleluia (etc. as *1)
140
//. Chant Genres
11.15. HYMNS
(i) Introduction
(ii) Texts
(iii) Music
(iv) Processional Hymns
Mobcrg 1947; MMMA 1; Szöverffy 1964-5; Gneuss 1968; Stäblein, 'Hymnus, B. Der lateinische Hymnus', MGG; Steiner, 'Hymn, II. Monophonic Latin', NG.
(i) Introduction
Despite the outstanding work of Stablein—the edition and commentary in MMMA 1 and the article for MGG—a repertorial survey of a sufficient number of sources from over all Europe is still not available and much musical analysis remains to be done. Of Moberg's projected edition only the valuable comparative tables were completed. Stablein edited the contents of a Milanese hymn collection and three French, one English, three German, and two Italian hymnaries, supplementing these melodies with others from additional French, English, German, and Italian sources. Over 550 melodies are edited. Lists of contents for all the sources Stablein consulted were not published, however, and it is even difficult to compare the contents of his main sources. It can nevertheless be ascertained that very few hymns (less than thirty) were sung in all areas of Europe to the same melody; many popular texts were set time and again to different melodies, and many popular melodies were used for a dozen or more different texts. Some melodies became strongly associated with a particular season or liturgical position. Most widely known, and probably among the oldest, are the hymns of the weekly office cycle. But as soon as one moves beyond these and the main feasts of the year, medieval sources display great variety of choice. The majority of hymnaries contain between 80 and 100 pieces; those which make provision for local and lesser saints may have up to twice this number.
Hymns were reported by St Augustine to have been promoted in the church of Milan by St Ambrose (d. 397) for singing during the long night vigils. Augustine attributed Aeterne return conditor, Deus creator omnium, lam surgit hora tertia, and Veni redemptor gentium to Ambrose. Only centuries later, alas, do the earliest Milanese musical sources appear, and there is no substantial agreement in the rest of Europe about melodies for these hymns. Indeed, lam surgit hora tertia was rarely sung at all.
The development of the Milanese repertory has been outlined by Huglo et al. (1956, 85-103), and also its relationship to the Gallican repertory (Huglo, 'Gallican Rite, Music of the', XG).
75. Hymns 141
The earliest notated hymnary appears to be the Kempten collection (edited by Stablein), from around the turn of the millenium, with alphabetic notation. This means that for the crucial period of the Carolingian renaissance we are reliant on texts alone for our knowledge of the history of the repertory. Gneuss (1968) has provided valuable indications as to how it developed. He sees an early repertory dating back at least to the sixth century ('Old Hymnal' type I) drawn upon in Milan, by the rule of St Benedict and the respective rules for nuns and monks of Caesarius and Aurelian of Aries; this also survives in two early English sources. A new recension of the 'Old Hymnal' (type II) seems to belong to the Carolingian period. Part of it is found, for example, in the same source as the earliest tonary, probably from Saint-Riquier, Paris, Bibliotheque Nationale, lat. 13159, written in the closing years of the eighth century. But in the second half of the ninth century a new repertory (the 'New Hymnal') gained rapid and widespread acceptance. It forms the basis of all subsequent collections, including those in modern service-books. All notated hymnaries, therefore, are based on the 'New Hymnal'. We do not know if musical changes accompanied the changes in text selection. And, partly because hymns were usually copied in independent collections (often with the psalter), we cannot yet relate the history of the hymn to that of other office chants. Discussion of the music is also made difficult by the wide variance between sources in their transmission of many melodies. Hymns are of course not peculiar in this respect, but the lack of a strong central tradition seems to have resulted in unusually wide differences. Stablein has noticed examples of the apparent 'modernization' of archaic melodies, the simplification of ornate ones, the regularization of note groups to reflect text rhythm, and differences of modal interpretation.
In what follows hymns are cited with their 'Old Hymnal' or 'New Hymnal' number, following Gneuss (24 ff., 60 ff.; if a distinction between the two types of Old Hymnal is possible, the appellation '1' or 'II' is also given). Examples are mostly taken from an English thirteenth-century manuscript of Sarum use, Oxford, Bodleian Library, Laud lat. 95, whose repertory is naturally related to the French and English sources edited by Stablein.
(ii) Texts
The most obvious textual and musical characteristic of hymns is their strophic form and metrical regularity. In view of this formal simplicity and the relatively small number of melodies required, it is hardly surprising that musical notation was not thought necessary in early sources. The last strophe of the hymn is usually a paraphrase of the doxology (the same doxology-strophe could be used for any number of hymns in the same metre). 'Amen' is usually sung at the end.
A large number of different poetic metres may be found, particularly among early hymns: asclepiads, distichs (usually in processional hymns), and so on (Stablein surveys the commonest metres in MGG). But the majority of hymns are written in one of two metres.
142 //. Chant Genres
Iambic dimeter x-^-x-^- four times) is the metre of the four hymns attributed to St Ambrose, and such poems were frequently known as 'Ambrosiani'. Strophes are of four lines, and there are typically four or eight strophes in a hymn. Stablein reckoned that two-thirds of the repertory uses this metre. (Metres are explained below, 11.27.)
Sapphic metre was the most popular of the antique metres used in medieval hymns. Each strophe has three full lines and a final half-line (as adonic):
—v_/---/ww-w-w (3 times)
—\u <j—
Such poetic compositions were often sufficiently highly regarded as artistic creations for their authors' names to have come down (not all ascriptions are regarded as secure; AH 50 and 51 are collections of ascribed texts). Best known, besides those of St Ambrose, are:
Prudentius (d. r.410): Inventor rutili, 0 crucifer bone, Pastis visceribus, Ales diei nun this, etc.
Venantius Fortunatus (d. c.600): Agnoscat ornne saeculum, Crux benedicta nitet, Pange lingua, Quern terra pontus aethera, Salve jest a dies, Tibi laus perennis auctor, Vexilla regis
Sedulius (first half of fifth century): A solis ortus cardine, Hostis Herodes impie
Not all these were originally designed as office hymns, some being extracts from longer poems, some written for special occasions. Thus Hostis Herodes impie is actually a later part of the same poem as A solis ortus cardine. The first three Prudentius hymns were sung during processions, not during the office.
(iii) Music
Many hymns are through-composed, without remarkable musical features. Others have simple repeat structures (aaba or abca, for example). In many cases—as a natural complement to some repeat structures—the highest notes of the strophe come in the third line, before the return to the tonic. In general hymns lack such musical features as repeated reciting notes, reminiscent of the delivery of prose texts. Quite often no strong tonal centre is apparent, and the lines may cadence on unexpected notes.
Christe qui lux es et dies (OH 30, NH 12) has been sung at Compline since the time of the 'Old Hymnal'. Its 'original' melody cannot be determined—if indeed it makes sense to think of one at all, for tunes are so readily interchangeable. It is given in Ex. II.15.1 with its commonest melody (No. 9 in Stablein's edition). It is the only text for this melody—an unusual situation. The last line of melody repeats the first. The highest note, G, occurs in the third line, though the range is so limited that this makes only a subtle impression. Even so, this small gesture is necessary, is prepared for in fact, by the repetitive second line. It is not uncharacteristic that the melody starts on a note above the final, although the D final does not seem seriously in doubt.
More unpredictable tonally is Sotnno refectis artubus (NH 14), usually sung during
75. Hymns 143 Ex. II. 15.1. Hymn Christe qui lux es (Oxford, Bodl. Lib. Laud lat. 95, fo. 140r)
Christe qui lux es et di-es        noctis tenebras detegis
lu-cisque lumen crede-ris lumen be - a - turn predicans.
Ex. II. 15.2. Hymn Somno refect is artubus (Oxford, Bodl. Lib. Laud lat. 95, fo. 137v)
Sompno refectis artubus spreto cubi-li surgimus ^   * •   •.....,
nobis pa-ter   ca-nenti-bus        ad-es-se te deposci
mus.
the Night Office on Monday, given in Ex. II. 15.2 with its commonest melody (Stablein No. 142; this is also the commonest melody for Primo dierum omnium, the equivalent Sunday hymn). The final E sounds very much like an 'imperfect' cadence, with perhaps D as the 'real' final. At any rate, that is the impression one has from the first two lines, which are the same but for their first-time/second-time cadence.
Both these hymns are 'Ambrosiani'. Whether their metre was reflected in performance in. any way seems doubtful. In iambic dimeter long syllables could occasionally be substituted for short ones. In a performance with long and short notes (in effect, in triple time), either the regular rhythm would be disturbed or one would have to ignore the variance in the metre. Another consideration is the pattern of stressed and unstressed syllables (not the same as long and short). This varies so often from line to line that any attempt to reflect it in performance (again, perhaps through long and short notes) seems misguided. This is not to deny that the reflection of metric or accentual patterns in the music (through groups as opposed to single notes, or with tonic accents) can sometimes be found, particularly in late medieval melodies. But clearly it constituted no universal principle of word-setting. (See the discussion of text rhythm in 11.27.)
The hymn Ecce iam noctis (NH 6), usually sung at Lauds on Sundays, was widely sung to melody 144 in Stablein's edition. The metre is sapphic. As in the previous two examples, the simplicity of the melody may reflect high age, or it may be due to the lesser solemnity of the liturgical occasion. In the Sarum hymnary used for the examples in this section, two melodies are given for the hymn. Melody 144 is used for the second strophe, while for the first strophe melody 107 is given, as reproduced in Ex. II. 15.3. The tonality of melody 144 is clearer than that of Ex. II. 15.2, for b (E in
144
//. Chant Genres
Ex. II.15.3. Two melodies for hymn Here iam noctis (Oxford, Bodl. Lib. Laud lat. 95, fo. 144v)
107 (k 0T9   f    m   ^   ^-- ,  «     i    # :»
Ec - ce iam no-ctis te-nu-a-tur um-bra lu - cis au - ro - ra ru-tilans co - rus-cat
^  '  '   *   ° ^
144
1
m'a
-»-0-
Ut de-us noster mise-ratus omnem
#  • «   .   ,   « ^
pellat languorem tribu-at sa-lu-tem S* m   m   m _ -.
-#—*-
ni-si-bus to-tis rogi-te-mus om-nes cuncti-po-tentem.
« **   M •   9    •  m •
V
doner et nobis pi-e-tate pa-tris regna polorum.
many sources which have the melody a fifth lower) is announced in the first line clearly enough for the repetition of the opening phrase at the end to be instantly recognizable.
Melody 107 was also fairly widely known, though not firmly attached to any particular text (Isle confessor in France, Virginisproles in Germany, and so on). It has a relatively modern air, partly because of the clear tonality, despite the way the melody moves easily through the whole octave, partly because of the 'Gallican' final cadence, the frequent steps of a third, and the repetition, not only of the first two lines but within the third also, where the tonal area above the final is explored for the only time. It is noticeable that groups of more than one note fall only on long syllables or at the ends of lines.
Because notation came so late to the hymn repertory it is impossible to know how-old many melodies may be. For Fortunatus' well-known hymn Pange lingua gloriosi proelium certaminis (Ex. II. 15.4) Stablein melody 56 was sung all over Europe, though half a dozen other melodies gained local acceptance (see Stablein 1950 for a possible predecessor of melody 56). This poem and the equally well-known Vexilla regis prodeunt appear to have been composed for Queen Radegunda at the convent of the Holy Cross in Poitiers, to whom the Emperor Justinian had given a fragment of the newly discovered True Cross. The poem has ten strophes. Like Vexilla regis it was taken up into the Passiontide liturgy, different parts often being sung on
15. Hymns
145
Ex. II. 15.4. Hymn Pange lingua (Oxford, Bodl. Lib. Laud lat. 95, fo. 141v)
6.........
Pange lingua glo-ri - o - si
et super crucis trophe-o C ^   E- ft_
* *
quali-ter redemptor orbis
pre-li-um certami - nis _F*_
die tri-umphum nobi-lem _6t
immola-tus ui-cerit.
different davs (the commonest divisions were made at strophe 6, Lustra sex qui iam peracta, and 8, Crux fidelis inter omnes). The popularity of the hymn entered a new phrase when St Thomas Aquinas (d. 1274) took the opening for his Corpus Christi hymn Pange lingua gloriosi corporis mysterium.
The text is in trochaic tetrameter with lines of 8 + 7 syllables (—x—^—x||-^-x-w-). The melody contains traces of traditional Gregorian turns of phrase: the opening is a common mode 3 figure (compare Ex. II. 11.1), c the usual reciting note for this mode; a then assumes its usual role as a goal for cadences above the final (with a link through D to the lower line 5 and an imperfect cadence for line 5 itself).
In Ex. II.15.4 the melodic variants for the next hymn in the Sarum manuscript, Lustra sex, are given over the stave. In this source the whole melody is given a fourth higher, to allow for both b§ and bb (f% and P\ when transposed for the transcription).
Hymns of the later Middle Ages betray the same modern stylistic features as other chants of the period. In some churches a freedom to supplement the repertory was clearly felt, as with other genres. Most noticeable in the stylistically later hymns are the wide range of the melodies and the freedom of rapid movement within that range, the clear emphasis on tonic (final) and dominant (upper fifth), and the division of phrases between the scale-segments they command. These features are already foreshadowed in Ex. II. 15.3, melody 107. They are more obviously present in Ex. 11.15.5, one of five melodies provided in the Worcester compendium for Sanctorum mentis (NH 119).
Ex. II. 15.5. Hymn Sanctorum mentis (Worcester, Cathedral Chapter Lib. F. 160; PalMus 12, 10*)
Sanctorum panga-mus	me-ri-tis so - ci-i	in-cli-ta gestaque	gau - di-a for - ti-a			
U V						
/						
fV\                m *		• *		- ^		
vy    • •			• m /»>			
nam gliscit	a-ni-mus	promere	cantibus	uic - to-rum	-*- * genus	op - ti - mum.
146
II. Chant Genres
The universally known melody for this hymn is Stablein's No. 159, which is mostly syllabic, tonally unfocused, though relatively mobile. It has the form ABC ABCD, with cadences on a, a, E, a, a, E, G (Worcester version, Stablein, p. 198) or a, a, E, a, a, E, E (Nevers version, Stablein, p. 102). In melody 420, given as Ex. II.15.5, there is no doubt about the primacy of C and G. After the repetition of the two opening lines an alternative tonal area is explored, E up to c with bb, rather than the upper segment of the C scale.
(iv) Processional Hymns
Hymns were often sung in processions. The repertory is not large (thirty-two in Stablein's edition, of which not all are actually processional hymns). A large proportion of processional hymns have a refrain. Often the first strophe would be repeated after each subsequent strophe. Pange lingua was sung in this way (to a melody different from the one just given), using Crux fidelis as the refrain strophe. In some hymns each alternate repetition of the refrain was of its second half alone (Rl 4-2 SI R2 S2 Rl+2 S3 R2, etc.). On the other hand, the refrains of some hymns are melodically and even metrically independent of the strophes.
Favoured metres are hexameters, distichs, and trochaic tetrameter. Fortunatus' Salve festa dies and Theodulf of Orleans's Gloria laus et honor are well-known examples in distichs.
The largest collection of processional hymns of this type that have come down to us is contained in St Gall, Stiftsbibliothek 381 (from St Gall, late tenth century), and smaller selections are found in other St Gall sources (e.g. St Gall 360). On pp. 23-50 and 142-66 of St Gall 381 there appear two series, of twelve and five items respectively, usually called versus (rather than hymnus), and frequently bearing a composer's name. Thus the first, Sacrata libri dogmata, is headed 'Versus Hartmanni, ante evangelium cum legatur cancndi' ('A versus by Hartmann, to be sung before the gospel is read'; facsimile in Stablein, 'Versus', MGG). The list of attributed pieces is as follows (see the remarkable 'footnote' surveying the repertory in Gautier 1886, 23-9; also, p. 159, an eleventh-century rhymed example from St Gall 382):
Hartmann (d. 925): Sacrata libri dogmata; Salve lacteolo decoration (Innocents); Cum natus esset Dominus (Innocents); Humili piece (for feast-days generally, found in many adaptations with verses in honour of various local saints; melody in W'agner III, 481); and Suscipe clementem plebs devotissima, one of the final three versus of the collection which are for the reception of a king
Fortunatus: Salve festa dies
Ratpert (d. 884): Ardua spes mundi (see Stotz, 1982; used as Humili prece); Laudes omnipotens ('ad eucharistiam sumenda'—'as the euchanst is taken up'); Aurea lux terra (for the reception of a queen); Annua sancte dei (St Gall) Notker (d. 912): Ave beati genninis ('about the Old Testament') Waldramm (d. r.900): Rex benedicte (for the reception of a king)
15. Hymns
147
Ratpert's/Wz/a spes mundi may serve as an example of the St Gall versus. (Rather few appear in later sources; all those that can be transcribed have been edited by Stablein, MM MA 1, 478 ff.; texts in AH 50, unfortunately not always making the refrain arrangement clear; see also examples in Wagner III, 480-2). In the Rouen cathedral manuscript (Paris, Bibliotheque Nationale, lat. 904, thirteenth century) from which Ex. II.15.6 is transcribed, both Ardua spes and Humiliprece are sung in the processions before mass on the Rogation Days leading up to Ascension Day (the weekdays following the fifth Sunday after Easter). In both chants verses in honour of Rouen saints are added to those already present; Ardua spes invokes Romanus and
Ex. 11.15.6. From processional hymn Ardua spes mundi (Paris, Bibl. Nat. lat. 904, fo. 136r)
«—0-
(three soloists)   Ar-du-a   spes mun-di        so-li-dator et   in - cly-te ce-li
•    •    ' «
Christe   e - xaudi nos
J /:<> * «;
(choir)   Ar-du-a ...
pro-pi-ti-us mi-se-re-re.
*   • *
(soloists)   Vir-go De-i   ge-ni-trix        ru-ti-lans in ho - no-re perhenni
• •— —# ^ #"
o-ra pro famu-lis
(choir)   Ar-du-a ...
• * ** #5. * #* •
sancta Ma-ri-a tu-is.
•   • S
Micha-el mi-se-re-re ci-to nostri
(soloists)    An-ge-le summe De-i
•   •  *   # #
ad-iuuet et Gabri-el atque pi-us Rapha-el.
-«-9—
(choir)   Christe ..
148 //. Chant Genres
Audoenus. The same melody is used for the refrain and the strophes (some other processional hymns have two different melodies).
11.16. CHANTS FOR THE ORDINARY OF MASS
In the study of chant, the term 'ordinary of mass' usually signifies those chants whose texts remain the same from one mass to the next, that is, Kyrie, Gloria, Credo, Sanctus, and Agnus Dei. There are a few other unvarying chants which are sung at festal masses: the antiphons Asperges me (outside Eastertide) and Yidi aquam egredientetn (during Eastertide) with their psalm verses; these are not considered further here. The special case of the versicle he missa est and its relation, the Benedicamus domino versicle at the close of office hours, are discussed briefly at the end of this section. Looking at the liturgv of mass as a whole, however, the ordinary comprises all texts which remain the same for each mass: not only the chants but the prayers as well.
Although the texts of the five main ordinary chants remain the same, they were set to numerous different melodies. These were often associated with particular feasts or grades of liturgical celebration. Early sources are not often specific about the assignment of the chants, but there seems little doubt that each church would have its own established customs in this respect, even before rubricated collections became common in the thirteenth century. Large numbers of trope verses are known for all the ordinary chants, excepting only the Credo. These often refer specifically to particular feasts and thus make the chant 'proper' rather than 'ordinary'. (Tropes for ordinary chants are discussed below, 11.23.)
Short sections are now devoted to each of the five chief ordinary chants. It should be noted in advance, however, that much research, transcribing, and analysing needs to be done before a rounded picture of these chants can be made up. For many years the best-known melodies have been those edited in the Solesmes/Vatican Kyriale sen ordinarium missae of 1905, which then passed into the Vatican Graduale and Solesmes Liber usualis. Eighteen sets were published, comprising Kyrie, Gloria, Sanctus, and Agnus, plus four Credo settings (later increased to seven) and other chants 'ad libitum'—eleven Kyries, three Glorias, three Sanctus, and two Agnus. These chants represent only a fraction of the medieval repertory, however, and the vast majority remain unavailable for study. It is fair to point out that many melodies had only a local circulation—many are known from but a single source—and the most popular are available in the Vatican selection. But until more are known it will be difficult to judge what is typical of the repertory as a whole, or of particular periods and areas, and what is eccentric.
The groundwork for future study has been established by the catalogues produced on the basis of the microfilm collection at Erlangen-Nuremberg University, by Landwehr-Melnicki, Bosse, Thannabaur, and Schildbach. The lack of British sources in the collection, and consequently in the catalogues, has been made good by Hiley
Chants for the Ordinary of Mass 149
(1986, 'Ordinary'). Yet the earliest sources of all, those of the tenth century, were poorly represented in these catalogues, and several questions about the distribution of the oldest recorded melodies remain to answered. The gap has been more or less filled in the case of the Kyrie by Bjork, the Gloria by Ronnau, and the Agnus by Atkinson. A comparable catalogue exists for Credo melodies, by Miazga, but its scope and emphasis is different. Early sources for the Credo are sparse, for it does not seem to have been sung by the schola until relatively late, and even then by no means universally. Florid choir settings are rare before the fifteenth century. Miazga's lists are dominated by the great mass of eighteenth-century settings, whereas the other catalogues hardly go beyond the Middle Ages; and Miazga's chief concern was the tradition of Poland: later West European sources are less well represented.
At one period or another it has been the custom for the whole congregation to sing the ordinary chants. In the early centuries this was certainly the case, and the question as to whether any melodies might have survived from this time has fascinated scholars for many years. Some simple melodies, such as might have been sung by the people, have indeed come down to us. Yet the sources which contain them are invariably late, later than those for more elaborate chants. A degree of caution is therefore necessary when judging the claims to antiquity of these chants. On the other hand, it has to be remembered that our early sources are nearly all cantors' manuscripts, containing music for the trained schola and especially for soloists. Should we expect them to record congregational chants? We are not being asked to subscribe to the belief that elaborate chants grew out of simple ones: that would be a naively simplistic view of plainchant history. The question is one of function. Chants undergo changes, or new chants are composed, when the liturgical conditions change. When the schola takes over the singing of a chant from the congregation (it is still acting in lieu of the congregation), then we should expect different music to develop. But if the congregation still sang the chants, at least occasionally, there would have been no pressure to change. In these circumstances chants might well have survived through centuries of unwritten tradition. As Wagner said of Agnus Vatican XVIII, one of the simple chants in question: 'The indication in the Solesmes books that it originated in the twelfth century is certainly wrong. It may well be that it does not occur in any earlier notated chant-book; but that is not decisive as far its age is concerned. Is our preface tone only as old as its earliest written occurrence?' (Wagner III, 448).
Stylistic analysis has seemed to offer insights into the problem. Thus Levy (1958-63) has made a bold attempt to link a Sanctus melody to other stylistically similar chants, Byzantine as well as Western, to reveal a 'modal area' from which the singing of the whole Byzantine ordinary is derived and the anaphora as well (that is, the prayers beginning with the preface, of which the Sanctus is a part, the memorial of the incarnation and words of institution). In the case of Kyrie and Agnus, litanies have been adduced as evidence.
In what follows some attention is therefore paid to possible early melodies. A few examples from the central repertory of the tenth to twelfth centuries are discussed
150
//. Chant Genres
(the biggest collections nearly all date from the twelfth century), and one or two quite different later chants (the fifteenth century marks a new high point in production) are also included.
Apart from those in the Vatican books, melodies have appeared rather haphazardly in facsimile or modern editions. The following may be consulted (mentioning facsimiles only of manuscripts in diastematic notation). Two Roman collections are available, the transcription from Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319 in MMMA 2, and the facsimile of Bodmer C. 74 edited by Lutolf (1987). The Beneventan manuscript Benevento, Biblioteca Capitolare VI. 34, is given in facsimile in PalMus 15. Transcriptions by Boe of the Beneventan repertory are in the course of publication. A facsimile of a Nonantola manuscript has been published by Vecchi (1955). French sources are poorly represented, only the Chartres melodies in PalMus 17 being available. On the other hand, two facsimiles of Sarum chants have been published, in Graduate Sarisburiense and MMS 4, and an edition has been made by Sandon (1984). For Eastern sources we have Marxer's transcriptions from St Gall, Stiftsbibliothek 546 (1908) and those from other manuscripts by Sigl (1911), and the facsimile of the Graduate Pataviense of 1511.
Few early sources indicate what music was used for singing the versicle he missa est and its response Deo gratias. It appears to have been popular to adapt a Kyrie melody for the purpose, or to share a melody with the office versicle Benedicamus domino (also sometimes sung at mass). The Benedicamus likewise relied heavily on borrowed music, for example, the melismas of office responsories. (See Harrison 1963, 74-6; Huglo 1982, 'Debuts'; Robertson 1988.)
11.17. KYRIE ELEISON
(i) Kyrie elcison as a Litany
(ii) Kyrie elcison after the Introit at Mass
(iii) Early Melodies
(iv) Italian Melodies
(v) Melodic Types
(vi) Later Melodies
Stablein, 'Kyrie', MGG; Landwehr-Melnicki 1955; Crocker, 'Kyrie elcison', XG; Bjork 1979-80; Boe 1989.
(i) Kyrie eteison as a Litany
'Kyrie eleison' is a litany formula and has been sung as a choral refrain in answer to petitions of one kind or another at least since the fourth century (as witnessed by Egeria in Jerusalem) and in the West at least since the fifth century. It can be found in numerous medieval litanies, as in these two examples:
//. Kyne eleison
151
Kyrie eleison. Christe eleison. Kyrie eleison. Pater de celis Deus miserere nobis. Fili redemptor mundi clemens miserere nobis. Spiritus sanctus Deus miserere nobis. Qui es trinus et unus Deus miserere nobis. Sancta virgo virginum ora pro nobis. [Then a series of saints' names, each followed by 'ora pro nobis'.]
Kyrie eleison. Christe eleison. Domine miserere. Christe miserere. Miserere nostri pie rex domine Iesu Christe. Christe audi nos. Sancta Maria ora pro nobis. [Then a series of saints' names, each followed by 'ora pro nobis'.]
Further 'Kyrie' invocations are sung at the end.
The invocation is a standard component of the preces sung in the Gallican liturgy, which survive in a few medieval sources (of which Paris, Bibliotheque Nationale, lat. 776 and 903 are the best known). Ex. II. 17.1 gives the opening of one of these. Here 'Kyrie eleison' is repeated after each verse. All the verses have the same basic melody. (In Paris 903 the Kyrie invocation is sung to a melody more closely resembling the verses than is the case here.)
Ex. II. 17.1. Litany (Cambrai, Bibl. Mun. 61, fo. 152v)
Ky-ri-e e-    -le- -yson.
	
	f* m
	m  m •
W-A*---—-*-—-% -	
Domi-ne De-us  om-nipo-     -tens patrum  nostro-      -rum. Ky-ri-e ...
One composition with this form appears among the Kyries of early Roman sources. This is the Kyrie with Latin verses beginning Devote canentes, in the manuscript Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319. Only Kyrie eleison is sung as an invocation, not Christe eleison. (In other sources, however, Christe is sung—see Boe 1990, 'Italian'; the Kyrie is edited by Melnicki in MMMA 2, 587 and by Boe 1989, 74.) As Boe has pointed out, this Kyrie may, in some form or other, be very ancient, for its seventh Latin verse is a petition against the Arian heresy, which in Rome at least was defeated as early as the fourth century.
(ii) Kyrie eleison after the Introit at Mass
The usual place for a litany in the East was after the lessons at mass, and how it came to occupy its Roman position after the introit is unclear. It is found there in Ordo Romanus I (early eighth century), no longer, it seems, a song for the whole congregation, but one for the schola, who sang invocations until the pontiff gave the signal to stop. Only the simple Greek text is mentioned, but the possibility cannot be excluded that further verses were sometimes sung. That is certainly the import of remarks in the letter of Gregory I (d. 604) to Bishop John of Syracuse: Gregory writes that in Rome Kyrie is sung as often as Christe, and that on non-festal days only
152
II. Chant Genres
the Greek text was sung, not the longer verses sung on other occasions; the schola sings first and the congregation answers.
Ordo Romanus IV, a Frankish recension (surviving in a Saint-Amand source) of previous Roman ordines of the late eighth century, has the number of petitions fixed at nine. This is presumably the form with which we are familiar:
Kytie eleison (Lord have mercy): three times Christe eleison (Christ have mercy): three times Kytie eleison (Lord have mercy): three times
(The ending eleison imas—have mercy on us—for the final phrase is occasionally found in early sources.) The Ordo says nothing about Latin verses. Our earliest sources with music unfortunately date from much later (as is so often the case). Here we find not only melodies with the Greek text alone but also compositions with Latin verses. These have been traditionally regarded as tropes, later additions to the composition. In the sense that the Greek invocations form a nucleus around which Latin verses could be added, this is not incorrect. Yet from the musical point of view the matter is not quite so simple. As Crocker (1966; see also 'Kyrie eleison', NG) pointed out, for several melodies the first recorded appearance has them with Latin verses. We have no proof that the melody was first conceived for the Greek text alone. And in view of Gregory's statement and the existence of numerous types of litany involving both Greek and Latin verses, it seems prudent to admit the possibility that new Kyries might be composed from the start with Latin text.
Discussion of the various ways in which combinations of Greek and Latin verses were made is postponed until the section on tropes (see II.23.viii).
(iii) Early Melodies
From Bjork's survey of the earliest sources (1979-80) it is clear that very few melodies were known all over Europe at the earliest period for which we have definite information, the tenth century. The best known were the following (numbers from Landwehr-Melnicki's catalogue, followed by the number in modern Vatican books): 55 (Vat. ad lib. VI), 68 (Vat. XIV) and 155 (Vat. XV). Early Rhenish and Eastern sources also have: 39 (Vat. I), 144, and 151 (Vat. XVIII), while Western sources have: 47 (Vat. VI), 102 (Vat. ad lib. I), and 124. The number increased rapidly in subsequent centuries, as Melnicki's catalogue shows. (It lists 226 melodies, and as more sources are surveyed more melodies may be added.) Many were of local significance only. The modern Vatican books have a selection of melodies of widely differing age and provenance (Huglo 1958).
It is not clear if there survive among the oldest recorded melodies any compositions from earlier centuries. It would not be easy for any simple melodies such as the congregation might have sung to have survived when the schola assumed the singing for themselves, and during the further period through to the earliest sources (themselves^designed to reflect the singing of the schola). Many of the early melodies
17. Kyne eleison
153
give the impression of being sophisticated, carefully designed pieces, which speak rather for composition by Frankish musicians.
Kyrie 55 (Vat. ad lib. VI) displays features typical of early Kyrie melodies (Ex. II. 17.2). The form of the piece is ABA CDC EFEx, where Ex signifies an extended version of E. The composer sets out very deliberately to explore different registers, so that B has a lower tessitura than A, D is lower than C, F than E. There is also a gradual rise through the three main sections. A rises to c, in C the melody-touches d for the first time, while E moves up to the higher octave. The final extended invocation has itself an AAB form, a not uncommon way of building towards the climax of the composition.
Ex. 11.17.2. Kyrie 55, Vatican ad lib. VI (Cambrai, Bibl. Mun. 61, fo.!55v)
8
'••m 0.1
Ky - ri - e
eley-son.      Ky - ri - e eieyson.
•%   00»   0^f00 ,mO *
Christe
£ C0H «.
eieyson.
Christe
eley - son.
0W*0 **o=i
00
'0 0 rm
Ky- -ri - e E.X
eley - son.
Ky-ri - e
eley-son.
V
Ky -   ri - e
elev-son.
This sort of melodic design, balancing lower against higher phrases and gradually pushing towards higher melodic goals, is also found in many early sequences, and when the Kyrie melody is sung with a Latin text, one syllable per note, like the sequence, the resemblance is even stronger. (One might compare the last phrase of Ex. II. 17.2 with the last phrase of the sequence melody known as 'Lyra', commonest text Kccepulchra, Anselm Hughes 1934, 54; or with the seventh and the last phrase of 'Hodie Maria virgo', text Aurea virga, Hughes, 45.)
Ex. II.17.3 shows melody 68 in Landwehr-Melnicki's catalogue (Vat. XIV). The overall scheme is simpler: AAA BBB CCCx. The last dozen or so notes of A and B are the same, but B begins by moving confidently upward to c, to be followed by the even higher-lying C section. Characteristically, the brief expansion in Cx consists of a partial repetition of what has preceded, including the so-called 'Gallican' cadence, cdd.
The expanded final invocation is particularly liable to varying treatment in different manuscripts. Ex. II. 17.4 shows melody 155 (Vat. XV), but not all sources have the
154                                        //. Chant Genres Ex. II.17.3. Kyrie 68, Vatican XIV (Cambrai, Bibl. Mun. 61, fo. 157r) A_ß_
Ky-    -ri-e eleyson. Christe eleyson.
c
-0-
Ky - ri- -e eleyson.
Ky-ri- -e eley-son.
Ex. II.17.4. Kyrie 155, Vatican XV (Cambrai, Bibl. Mun. 61, fo. 159v)
0 A B
Kyri-e          eley-son.      Ky-ri - e eley-son. n C D _
Christe               eleyson.   Christe eley-son. a E. F _
v
Kyri - e eleyson.       Kyri - e eleyson.
Kyri - e eley-son.
last verse in the form given here. Once again the form ABA CDC EFEx is used, but Ex actually combines E and F. A, B, C, D, and E all have the same ending. And yet again a consistent pattern of higher and lower phrases may be observed. In this case the progression from A to C to E is achieved simply by adding higher notes at the start of what is the same basic phrase throughout.
Composers evidently delighted in the possibilities offered by the thrice-three verse-scheme. All manner of repetition patterns and partial correspondence between verses may be discovered.
(iv) Italian Melodies
That these three melodies were so popular (and a number of others run them close when the greater number of sources from the eleventh and twelfth centuries is taken
17. Kyrie eleison 155
into account) seems to indicate that they answered well to the general sense of how a Kyrie should be sung. Their features are typical of a large number of early Kyrie melodies. But not all melodies have this character. Early Italian Kyries, for example, show little interest in repetition schemes and the upward surges so noticeable in Ex. II. 17.2. The above-mentioned Kyrie with Latin verses Devote canentes etc., sung to melody 77 in Landwehr-Melnicki's catalogue, has the same melody for all invocations (and all Latin verses), and a static, repetitious quality. So does melody 52, associated with the Latin verses Auctor celomm etc. These two melodies are given in Ex. II.17.5.
Ex. 11.17.5. Kyrie 77 and Kyrie 52 (Rome, Bibl. Vallic. C. 52, fos. 153r, 150') Kyrie 77 Kyrie 52
P
Ky-ri - e- -ley - son. Ky-ri - e- -ley   - son.
Since the earliest Italian sources date from the eleventh century (the earliest Roman one from as late as 1071), they are automatically excluded from the 'earliest' layer. This need not mean, however, that the melodies they contain are necessarily less ancient, only that they cannot be proven to be so. Stablein ('Kyrie', MGG, Ex. 9) prints one melody (Landwehr-Melnicki 84) from a Roman source that is so simple that it might be centuries old; or it might be a relatively late composition deliberately made simple for ferial use, as Landwehr-Melnicki 7 seems to have been.
(v) Melodic Types
Not surprisingly, many melodies in this large repertory bear a family resemblance to each other, or make use of similar melodic gestures. Thus thirteen melodies open Gab, or some variant thereof. Several other melodies besides Landwehr-Melnicki 55 (Ex. II. 17.2) fall to E after the G-opening (48, Vat. II, 102). Landwehr-Melnicki 39 (Vat. I), for example, sounds like a simplified version of 55; its Christe is the same as that of 55. Another family of short melodies has a first phrase ending on b, starting either on G orb, like Landwehr-Melnicki 155 (Ex. II. 17.4): Landwehr-Melnicki 144 and 151, or, rather longer, Landwehr-Melnicki 124 and 142.
(vi) Ixiter Melodies
From Landwehr-Melnicki's catalogue it can be seen that the composition of melodies continued unabated into the sixteenth century and beyond. Many of the newer items bear the marks of a later age: scale passages running through a fifth (usually from or towards the final), broken chord figures, and sudden moves to the upper octave. A number are couched in a somewhat sentimental 'F major' idiom, like the popular 'De angelis' melody (Landwehr-Melnicki 95, Vat. VIII, known mostly from French and
156
//. Chant Genres
Italian sources). A melody of this type found in eastern sources (from Germany, Switzerland, Austria, Bohemia, etc.) is Landwehr-Melnicki 97, reproduced here in Ex. II.17.6.
Ex. II. 17.6. Kyric 97 (St Gall, Stiftsbibl. 546, fo. 38v)
jL,-—	,w*w			9 *	—-m*k •'** *m*m w-z-
w— Ky-ri --6-	• e	"•§0 • e--* m	w -leyson. ~»m-	-W Christe	e-leyson. ----0-m-
I-		_ m	m*wm a		
)P- (Kyrie	as above, then)	Kyri -	* - e	• wwm w*	e - leyson.
Local compositions include even more extreme examples in this vein, as for example Landwehr-Melnicki 139, one of ten unique melodies in St Gall, Stiftsbiblio-thek, 546, compiled at the famous abbey by Joachim Cuontz in 1507 (Ex. II. 17.7).
Ex. II.17.7. Kyric 139 (St Gall, Stiftsbibl	. 546, fo.	38r)
J  J  J |jj<j U J !JJJ U   JJljJJ Iii J 1	il   i e*e	
->-"--		---^-
W-m—-w -n .-ww-	—w-	-mm-,-ww-m*mm ,#     *• *—
V -+< Ky-ri - e-	leyson.	-W M -' - - ~»*J Christe- -leyson.
~ r. •*-- /Um ť =
ě ir'ii       li.       ww    s"t_     •*   w* wm  ,m**    * a
Ky-ri - e- -leyson.
11.18. GLORIA IN EXCELS IS DEO
(i) Introduction
(ii) Recitation Types
(iii) Through-Composed Melodies
Bossc 1955; Stablein, 'Gloria in cxcclsis Deo', .1/(7(7; Crocker, 'Gloria in cxcclsis Deo', AG; Boe 1990, Gloria.
(i) Introduction
Some elements in the text of the Gloria in excelsis Deo are very ancient, dating back to early Christian times. Like the Te Deum, it is a non-biblical hymn of praise of irregular construction, an agglomeration of phrases of different date and origin. It seems to have been part of the office in the early centuries, but according to the Liber
18. Gloria in excelsis Deo
157
pontificalia it was introduced into mass in Rome on Sundays and saints' feast-days by Pope Symmachus (498-514). The earliest surviving version of the Gloria in Latin is no older than the so-called 'Antiphonary of Bangor', a late seventh-century Irish collection of texts mainly for the office. Variants in the text are still occasionally to be found in sources of the tenth century and later, that is, in the period from which we first have notated versions. Although originally a congregational chant, its execution was later reserved for the clergy, being intoned at the start by the pope, bishop, or priest as the occasion required. Among the surviving melodies some are easily singable by the whole congregation, as modern experience proves. But there is no way of knowing how old such melodies are. Some relatively ornate melodies seem more like music for soloists or the trained schola.
Bosse's catalogue lists fifty-two different melodies, to which a few may be added from sources not available to Bosse. Eighteen are to be found in the modern Vatican books. Bosse and Stablein both survey the different types of melody so far known. Some are potentially quite old: that is, they may easily be imagined to date from the period of oral transmission. One type of melody may be described as a highly ornate recitation, with a constantly reiterated central pitch and cadential melismas; another type of melody consists of the constant repetition of a single phrase, adapted as required to verses of differing length. Other types may be more recent: some are through-composed, others are freely composed but with constant resort to particular motifs, which serve to bind the musical fabric together. It is not, of course, possible to assign all melodies absolutely to a particular category.
(ii) Recitation Types
One set of melodies is dominated by a single recitation note with neighbouring tones, delivered in a highly inflected, elevated manner which raises them above the level of, say, the introit or communion psalmody heard elsewhere in the mass. Ends of verses are often marked by cadential flourishes. We could imagine the melodies in this small group to be music for the choir, or even for soloists.
The most important member of the group is Gloria 39 in Bosse's catalogue, sometimes referred to as 'Gloria primus' or 'Gloria A', unfortunately not included in the Vatican selection. It is the grandest and most expansive of Glorias. Boe (1982) has found a version of it in the Old Roman Easter Vigil mass. Since it is usually found with trope verses—it was sung more frequently with tropes than any other melody— it is given elsewhere in full (see Ex. 11.23.15). Suffice it here to point out its chief features. The melody has two recitation tones. In the first line and occasionally later the recitation is centred on a (typically in figures such as Gab a). Somewhat disconcertingly, the second line and most subsequent lines use F frequently (in such phrases as aGF Ga a), with bb as upper auxiliary. The most important cadential melismas are given in Ex. 11.18. la—d. The figure given as Ex. 11.18.le, which is also to be found at the start of the verse tone for mode 1 responsories, appears frequently from 'propter magnam' onwards.
158 II. Chant Genres
Ex. II.18.1. Cadential figures (a-d) and intonation figure (e) from Gloria 39
			
	__ —,		
fm                   i**m 4*	a	Ca at}	m   m*        9. 9i         J 9'9Sm 9*9
vy     99    9 99 9 99	*99 9 9	9*99 *0 -	wm w            9     9 »
a	b	9H^ C	d
Other examples of this type of melody may be seen in Bosse 13 and 22 (Vatican ad lib. II and III), centred around D, or the melody for the Greek Gloria Doxa en ipsistis theo (Huglo 1950, 'Gloria'). The latter begins with b as reciting note (but even more heavily ornamented than Gloria 39) with G a c as the common intonation. After several lines the cadences begin to fall on E, though most of the melodic movement still concerns the tetrachord G-c.
The simplest example of the type is Bosse 2, probably a south Italian melody (ed. Boe 1990, Gloria, 188) but given in Ex. II. 18.2 from one of its few northern sources (presumably brought back to Normandy after the Norman conquest of south Italy and Sicily). It will be seen at once that practically every phrase opens with the intonation F G a and ends a G. A great deal of what happens in between stays within the same narrow range. The use of a higher opening for 'Domine Deus', 'Domine fili', and 'Cum sancto spiritu' may also be mentioned. Since it comes from a part of Italy where much ancient, non-Frankish material is preserved, one is tempted to see here one of the oldest surviving Gloria melodies. The sources are, however, too late in date to prove any hypothesis conclusively. In its usual northern version, which corresponds to Vatican melody XI, it was catalogued by Bosse as melody 51.
If some features recall the singing of office psalms, or responsory verses, the above melodies nevertheless remain essentially free of any simple, repeated formula. They are too elaborate (with the exception of melody 2) and too varied to subsume under psalmodic practice. There are, however, several other melodies which come closer to simple psalmody, consisting of hardly more than the constant repetition of one musical idea or formula.
The best-known example of this type of melody is Bosse 43 (Vatican XV). Each verse begins E G a, recites on a with inflections down to G or up to b, and cadences a G a G E.
More adventurous is Bosse 25 (Vatican V), with more than one musical idea. In the top half of the range comes the phrase G b cd d b dcbaG ab aG G. This is followed by a phrase which descends into the lower half of the range: caG aGEFED. The final is achieved through the phrase D EFG GFa G. Each phrase is of course treated quite flexibly, to accommodate the varying numbers of syllables and their varied accentuation patterns. These phrases, although deployed in a fashion resembling simple psalmody, have no musical resemblance to psalm tones. Nor has Bosse 38 (Vatican VIII), the melody which makes a musical pair with the Kyrie 'de angelis'. They are both much later, and much less widely known, than Bosse 43.
18. Gloria in excelsis Deo 159 Ex. 11.18.2. Gloria 2 (Paris, Bibl. Nat. lat. 10508, fo. 20v)
£0  0   '   *,    »'   '     0     0 m     *     '    •*    m <S   ti*" »
Glo-ri-a in ex-celsis De-o.     Et in terra pax hominibus bone uoluntatis.     Laudamus te.
0   é 0»   ' W0   Í^IH, m í* •   * 7=7
0 *m '0 m *    0        9  w - m m   - -   -  -»•••■»        m m* •  m 0  0   * -*m-
Bene-di-cimus te.     A-doramus te.     Glori-fi-camus te. Grati-as a-gimus ti-bi
z^=3«z=«_«-^ 0  »   .  «       i"' » *' • f   «    í*     0        m   jt w a  >• * • =
propter magnam glori-am tuam.     Do - mine Deus rex celestis.     De-us páter omnipotens.
1»'"»  *   š*   •  0  0 P 0 t m f, I   I  .  , £0*00      £00* i^T
Do - mi-ne fi-li u-nigenite.     lesu Christe.     Domine Deus.     Agnus De-i.     Filius patris.
j$    '   je 0   0  '0 m -0—m        s 0 ' '  0 0        •   £ô 0   • •* s   07, • -
Qui tollis peccata mundi.     Miserere nobis.     Qui tollis peccata mundi. Susci-pe
0   0 0     Í0 0    ,     0 0    *   * a
0    í* •   0    *   014        m 0 * *
deprecati-o-nem nosíram.     Qui se - des ad dexteram patris.     Miserere nobis.
• f
000*0   £b 0
*    * b* * 0   m Í* *
Quoni-am tu solus sanctus.     Tu solus dominus.     Tu solus altissimus.     lesu Christe.
t"«    »'   • m   a   {•  m a   _   j, a &
Cum san - cto spiri-tu in glori-a De-i pa - tris A
men.
(iii) Through-composed Melodies
The majority of Gloria melodies are through-composed. This does not exclude the possibility of using recurrent motifs, which stamp the melody with a particular character without suggesting the technique of simple (or ornate) psalmody. Even when the motifs encompass whole phrases, their order is not predictable in the way that those of Bosse 25 are.
Few melodies actually lack recurrent motifs entirely. Bosse 24 (Vatican ad lib. I) is a well-known example. More typical is Bosse 56 (Vatican IV), the most popular of all Gloria melodies. The final, E, is heard at the end of practically every phrase, but it is not treated as a reciting note. A variety of musical ideas leads into it, from below (FD C D E) and above (GF Ga GF E), the latter sometimes preceded by a higher-lying motif (G a c a). The artistry of the composition lies in the judicious selection of one
160
II. Chant Genres
motif or another for the various verses of the text, balancing higher against lower, amplifying where necessary, as with the more expansive cadence heard four times: DFGaGFG E.
Another example of this type of melody is the little-known Bosse 5, found in Eastern European sources from around 1200 onwards (Ex. II. 18.3). There is no need to attempt to isolate all the motifs which make up the basic melodic material; in any case, some instances of their use might no doubt be felt to be fleeting coincidences rather than deliberate employment. The motif EDEG for 'excelSIS', for example, occurs many times throughout, sometimes split to accommodate different syllables. Are these all conscious attempts to bind the piece together?
Ex. II.18.3. Gloria 5 (Munich, Univ.-Bibl. 2° 156, fo. 157v)
__x     z_^ _
-•——i
« • • *      «C .    . *9*:    '       f*     *i «       ~ •* =
mm-*jr«-•- * 0*
Glo-ri - a in excel-sis   De-o.     Et in   ter-ra pax   ho-mi-ni-bus   bo-ne uo-lun-ta-tis.
X_J}_X_X_
Lauda - mus te.      Benedi-ci-mus te.     A-do-ra - mus te.      Glori-fi-ca-mus te.
—v »-—
>•     9   • •
Gra-ti-as a-gi-mus ti-bi propter magnam tu-am glori-am.     Domine De-us rex celestis
_X_
'     • H f. •      .      • m.....9 ' ' •   * • ' .  ' •} gé •
Deus pater omni-po-tens.     Domi-ne fi-li u-ni-geni-te   lesu Christe al-tis-si-me.
0   m        m .    •    *•   /ľa* •
* «   •   &  • 0 ^   , ,
Domine De-us agnus De - i       fi-li-us patris.     Qui tollis pec-ca-ta mundi _a_*_a_
•    ŕ»   m   - • •    m   m   f       m m
--•—m-*-
-•-0-
~0
miserere nobis.     Qui tollis peccata mun - di   suscipe deprecati-o-nem nostram.
-*—•
P
Qui se - des ad dex - teram pa-tris mi - se - rere nobis. Quoniam tu solus _a *_
0-0-
£0 0
sanctus.     Tu solus do - minus.     Tu solus al - tissimus lesu Christe.
Cum sancto spi-ri-tu in glo - ri - a   De - i     pa - tris        a - men.
19. Sanctus
161
It is perhaps more worth while to follow the falling figure bcbaG, which is first heard twice in the second verse, 'Et in terra pax hominibus' (labelled 'x' in Ex. II. 18.3). This has no rhetorical function: it may start or end a phrase, or stand in the middle. Among its many appearances are several where it is preceded by a fall and immediate rise of a fifth: at 'Adoramus te', 'Qui tollis' (the second one), and 'Tu solus dominus' (labelled V). The leap of a fifth may bridge two different phrases, as at 'dexteram partris/miserere' and perhaps also 'peccata mundi/Suscipe'. The figure 'x' is frequently succeeded by a rise to d, as at 'terra pax' (labelled V). And so on.
Glorias were usually sung with trope verses on high feasts from the tenth to the twelfth century (see II.23.xi), but because the trope verses were so often applied to different base melodies there seems to be no overwhelming reason for believing that tropes and melodies were composed at the same time. The number of known melodies is outnumbered two to one by the sets of trope verses. After Gloria tropes fell out of use in the thirteenth century, only one modest Marian example, Spiritus et alme, held its place. It and the Gloria melody it embellishes (Bosse 23) seem first to have become popular in the Paris region in the second half of the twelfth century.
Composition of new melodies was by no means as vigorous as for other ordinary of mass chants, though from Bosses tables it can be seen that a flurry of activity took place in the fifteenth century. Several fine melodies still remain generally unknown.
11.19. SANCTUS
(i) The Oldest Melodies
(ii) Other Melodies
Levy 1958-63; Thannabaur 1962; Thannabaur, 'Sanctus', MGG; Crocker, 'Sanctus', A'G."
(i) The Oldest Melodies
Much speculation has been occasioned by the problem of distinguishing a possible ancient melody, such as might have been sung from early times, when the Sanctus was a congregational chant, among the compositions which have come down to us in notated manuscripts from the tenth century onward. For many years it was confidently asserted that the simple melody Vatican XVIII, no. 41 in Thannabaur's catalogue, was a relic of this early period. Thannabaur then pointed out that the earliest source dates from the eleventh century, the melody becoming more widely known only in the thirteenth century. In fact, in Thannabaur's earliest source, Benevento 38, all the ordinary-of-mass chants are on added leaves of the twelfth century; and the melody appears there in a more elaborate version than was later to become usual (see MGG, 'Sanctus', Ex. 1; the ending should read Gaf FG). The next Italian source dates from the thirteenth century, and the rest are practically all fifteenth-century manuscripts. Levy (1958-63, 27) cites another twelfth-century source for the
162
II. Chant Genres
incipit, at least, of the elaborate version. On the other hand, Boe (1982) has linked it with other ordinary chants for the Old Roman Easter Vigil mass.
This is not to deny that the Sanctus was probably once sung to a simple melody. The text is a continuation of the preface, which was sung by the officiant to a recitation formula. Thannabaur 41, Vatican XVIII, is hardly more than a continuation of one of the preface tones. And the participation of the congregation in the singing of the Sanctus is recorded frequently throughout the Middle Ages. At various times and places, however, the chant was performed by clerics. And trained singers must also have played a part, at least in the singing of trope verses, and perhaps also in the execution of some of. the more elaborate melodies known from medieval sources.
It is therefore somewhat difficult to identify a possible early layer of melodies. One interesting candidate is the Greek Sanctus, or Agios, known from a group of tenth-and eleventh-century manuscripts, usually in conjunction with other ordinary-of-mass chants with Greek text (see the list in Atkinson, 1982, 'Missa Graeca'). Levy (1958—63) discovered a Byzantine relative of this and remarked that the same basic musical idea underlies the singing of the words 'sanctus, sanctus, sanctus' in the Te Deum. In fact, Levy pointed to the existence of a large family of melodies (or better, recitation formulas), both eastern and western, which may all have a common origin.
What unites them, to put it in most general terms, is their insistence on the scale segment Gab, with c and even d as upper options; E is sometimes used to start a phrase, but F is avoided. (The melodies are found in transposition with FGa or CDE as the basic notes.) The continuous oscillation within this narrow range is most familiar to us through the common tones for the preface and Lord's Prayer (see Stablein, 'Prafation' and 'Pater noster', MGG). And as far as ordinary of mass chants are concerned, we have just seen examples of this type of melody in Italian Kyries (Ex. II.17.4) and Glorias (Ex. II.18.2). Ex. II.19.1 gives a version of the Greek Sanctus and another Italian melody in this idiom. (Compare also Boe 1982, Ex. 3.)
(ii) Other Melodies
The majority of melodies in Thannabaur's catalogue are not of this type. Over 230 are listed there, and others are to be found in sources not accessible to Thannabaur. Relatively few become known right across Europe:
Thannabaur
Vatican
XVII
XVIII
earliest sources
32 41 49 116 177
202
203 223
IV VIII XII XI
II
XV
12/13th c. 10th c.
11th c. 12th c. 11th c. 12th c. 13th c. 12th c.
19. Sanctus 163
Ex. 11.19.1. Greek Sanctus and Latin Sanctus 66 (Modena, Bibl. Cap. O. I. 7, fos. 108r, 206v)
A - gy - os     A - gy - os     A - gy - os     o the - os     is-ky-ros   sa-ba - oth * »*0  t »«o **9
pli-nis   u-ra-nus     ke-y-gitin do-xi su        o-san-na emtas ipsistis
^ ^gě p ^ ^ í' í*   ééi      ■   »  »  » ■ 4» f» • •   •   ■ *# Í*
Eulo-gi-me-nos   o entho-me-nos     en o-nomati ky-ri-os     o-san-na emptis ipsistis.
San - ctus. (3 times)  Do-mi-nus De - us sa - ba-oht.
/• * ^ j& ,3 f. ň , mii?. #> g* #; Mfr.u0   '/ < «, ' « f>
Ple-ni sunt ce - li   et terra   glo -   ri - a    tu - a.                 Osan-na in excelsis.
Be-nedictus    qui ue - nit  in no - mi-ne domi-ni.                 Osanna in excelsis.
Melody 154 (Vatican I), widely known in France and Italy, is perhaps the only-other melody which can really be called popular. It follows that the repertory as a whole is very diffuse, like that of other ordinary of mass chants, most melodies having purely local currency. Some of the oldest melodies are not represented in the above list. In three tenth-century sources, for example (these and several other of the oldest sources are not covered by Thannabaur), we find the following:
Cambridge, Corpus Christi College 473 (Winchester): 111, 154, 155, 216, 223 Paris, Bibliotheque Nationale, lat. 1118 (Aquitaine): 89, 111, 216, 223 (plus one not in Thannabaur)
St Gall, Stiftsbibliothek 381 (St Gall): 154, 216 (plus one not in Thannabaur, possibly 153)
Like the repertories of Kyries and Agnus, Sanctus collections were rather small before the eleventh century, when a deliberate effort seems to have been made to provide a set of chants for the festal liturgy. After the twelfth century production slackened, until a new peak was reached in the fifteenth century.
Most early melodies show some relationship between the various verses. The two settings of 'Hosanna in excelsis' are naturally often the same, and material from the
164
//. Chant Genres
first verse may reappear later in the piece. The three 'Sanctus' acclamations often stand apart, though their music too may be redeployed for the text that follows.
Sanctus 68, transcribed in Ex. II. 19.2, is a fairly typical example. The music for the first 'Sanctus' recurs at 'Pleni sunt' and 'Benedictus'. The movement from G to d and back again at 'Sanctus dominus Deus' is heard again during 'Osanna' and 'qui venit'. Perhaps the short phrase for 'gloria tua' is an echo of 'Deus sabaoth'. The melody has a character quite different from that of the Greek Sanctus and the Italian melody given in Ex. II. 19.1. The first 'Sanctus' stays within the segment G—b, the second touches c, and then the third rises to d, a gradual unfolding which is carried one step further later on in 'Domini'. There are also literal repeats, in 'sabaoth' and the melisma after 'gloria tua'.
Ex. II. 19.2. Sanctus 68 (Paris, Bibl. Nat. lat. 10508, fo. 123r)
San-        -ctus.   San - ctus.   Sanctus. Dominus De-        -us   sa- -baoth. Pie  - ni sunt ce-       -li   et   ter - ra glori-a tu-a.
O-      -san - na      in ex-celsis.
Bene - die - tus qui      ue-   -nit   in no-mi-ne      do- -mi-ni.
(Osanna as above)
Melody 216 in Thannabaur's catalogue, found fairly widely among the earlier sources, has something of the same character, though it is cast in a more modest vein (Ex. II. 19.3). The three acclamations rise gradually, then 'Deus sabaoth' echoes the opening, adding the so-called 'Gallican' cadence. That cadence recurs at the end of each verse. 'Pleni' and 'Benedictus' take up the music of the third 'Sanctus', but the 'Osanna' verse is independent.
Twelfth-century sources such as those used for the transcriptions so far have the largest of all collections of Sanctus melodies, excepting such retrospective anthologies as St Gall 546. Ex. II. 19.4 is taken from this manuscript. Most of the new melodies of the fifteenth century and later are quite different in character from those seen so far, with arpeggiando figures and simple melodic repetitions. Thannabaur 98 is built almost entirely from the material of the second and third 'Sanctus' acclamation. The chant is notated mensurally in St Gall 546; each normal note is transcribed here as a crotchet, a double punctum becomes a minim, and the semibrevis a quaver.
20. Agnus Dei
Ex. II.19.3. Sanctus 216 (Paris, Bibl. Nat. lat. 10508, fo. 123v)
165
San - ctus.    San - ctus.  Sanctus.   Do-mi-nus     De - us   sa - ba - oth.
-ŕ*
*tui* *~0—£*-j-,-f-.-9-^-f4 ».—
T *-*-•    #H _^
Ple-ni sunt ce - li  et  terra     glo - ri - a    tu - a. O-saň-na
in excel-sis.
lie
* . . . fr  .j p
-m »   -    -      »   -    •   -     -    m    #- *#
Be - ne-di - ctus qui uenit   in  no  -  mi-ne   do-mi - ni.        (Osanna as above)
Ex. II. 19.4. Sanctus 98 (St Gall, Stiftsbibl. 546, fo. 68r)
j^i f 'jf r> j j j j ľ f frrj r f r f g ff jj jf f f f r r j j J
San-
-ctus san-
-ctus san-
-ctus
f r Mu r 17rŕ^Fŕj
j j j j f f r
*=3=*
Dominus De-    -us sa-
-ba-oth.   Pleni sunt celi et terra glo-ri-a tu-
ý Jf f f f r r j j
n j j r f r'
-a. O-
-sanna   in excel-
f O J', j j  Mr f,f, Hu ľ i7r j ff Jj nr p
-sis.
Benedictus Marie fi-li-us     qui  uenit in nomine do-
n f r nnm
■mi-ni. O-
-sanna    in excel-
-S1S.
11.20. AGNUS DEI
Stablein, 'Agnus Dei', MGG; Schildbach 1967; Atkinson 1977; Crocker, 'Agnus Dei', AG.
Investigation of the earliest melody or melodies for the Agnus Dei runs into the same problems as for the other ordinary-of-mass chants. According to the Liberpontificalis, Sergius I (687-701) introduced the chant into the Roman mass, as a piece sung by both clergy and people to accompany the breaking of the bread. Whether or not the
166
//. Chant Genres
attribution is secure, the approximate date can be supported indirectly. Ordo Romanus I (early eighth century) also understands it to be a fraction chant, but sung by the schola. Ordo Romanus III—the relevant part is Frankish and dates from the third quarter of the eighth century—indicates that the singing continues until the fraction ceremony is completed, rather in the way the early introit, Kyrie, and other chants were performed. Shortly after this, however, a number of Carolingian documents have the chant sung during the kiss of peace or communion itself. There arose concurrently the practice of using pieces of unleavened bread, and communion by the entire congregation was abandoned. The fraction ceremony diminished in significance. This seems to have resulted in a shifting of the performance of the Agnus towards communion and curtailment of its length, restricting it to the three petitions with which we are familiar. This is its form in the earliest extant document which is specific in such details, the Amiens sacramentary Paris, Bibliotheque Nationale, lat. 9432 (ninth century, second half). (Facsimile in Atkinson 1977, 10; Atkinson provides a detailed review of the evidence summarized here.)
The earliest musical sources are somewhat later (table in Atkinson, 13), and from these it is clear that only one melody had Europe-wide currency at an early date, no. 226 in Schildbach's catalogue, Vatican II. If we are looking for a hypothetical simple melody of the type which might have been sung by the congregation, repeated as often as necessary, according to the older practice, we might doubt whether this is it, or at least whether it appears in its former state. It has been suggested that two simpler melodies may be particularly ancient: Schildbach 101 (Vatican XVIII) has long been designated the oldest melody, partly because of its simplicity, partly because it follows easily from the preceding dialogue ('Pax domini . . . Et cum spiritu tuo'—see Wagner III, 449), partly because of musical identity (the text is different) with the 'Agnus Dei' section of the litany of the saints (Stablein; GR 198, LU 838). Stablein also directed attention to the Roman version of the melody (registered separately by Schildbach as no. 98), which appears in two Roman sources, Vatican Library, Vat. lat. 5319 and Archivio San Pietro B. 79. Whether the melody is much older than the earliest sources—which are much later than those for melody 226— remains open to question.
The oldest Roman source, Bodmer C. 74, has another relatively simple melody which belongs to the Italian type of chant discussed above in connection with the Kyrie, Gloria, and Sanctus. As Boe (1982) has shown, it belonged to the Old Roman Easter Vigil mass. Ex. 11.20.1 gives this melody (not in Schildbach), and melody 226. The greater range and variety of the latter are obvious.
Many Agnus melodies may be compared with Kyries in that they suggest a tripartite form. For the above examples only one verse is copied out, to be sung three times. Other melodies may have a contrasting central verse, giving an ABA form, or even three different verses (for example, Vatican XI, Schildbach 220). Furthermore, there is often a similarity of range for 'Agnus Dei' and 'miserere nobis', whereas 'qui tollis peccata mundi' will rise to a higher level, giving an ABA shape to the verse. Thus Vatican XVII, Schildbach 34, has the overall form ABA A'BA ABA, the only
20. Agnus Dei 167
Ex. 11.20.1. Old Roman Agnus (Cologny, Bibl. Bodmeriana C. 74, fo. 125v); Agnus 226 (Paris, Bibl. Nat. lat. 1119, fo. 249r)
Cologny__
P
a  • •   w £ims »g *T*    • 9 m-jE
Ag - nus De - i qui tollis pecca- -ta mundi misere- -re no- -bis. Paris
Ag - nus        De - i    qui  tol-lis      pec - ca-ta mun-di   mi-se-re - re  no- -bis.
difference between the second verse and the others being the lower start. Such symmetry is a feature of the later rather than earlier repertory (the earliest sources for melody 34 are of the thirteenth century). Among the considerable variants often found between sources many affect the form of the piece. Thus while many sources have but a single verse of melody 226 (as in Ex. 11.20.1), many have a contrasting second verse, most with the one published as Vatican II, others with that of Vatican XVI and XV.
The best known melodies were:
Schildbach Vatican
34        XVII (less well known in France and England) 101 XVIII 114 IX 136 IV 164 XVI 209 XV 220 II
Of these, 34, 114, and 136 are relatively late F-mode melodies, with a particularly clear delineation of the Vac triad. Such Agnus melodies are not uncommon, and because of this preference the modal 'profile' of the repertory is different from that of Kyries and Sanctus. The rough percentages given here are derived from the tables of Landwehr-Melnicki, Thannabaur, and Schildbach. (The figures of course vary from area to area and century to century, as the commentary by those three authors shows. Some melodies are naturally found in different modes in different sources, and transpositions between G-mode, C-mode, and F-mode with Bb are common. These are ignored here.)
D E F G Kyrie 30 17 21 31 Sanctus 31 20 23 26 Agnus     36   19   27 19
As an extreme example in the F-mode vein, Ex. II.20.2 gives melody 144, found in a few fifteenth- and sixteenth-century sources. The mensural notation indicates a
168 II. Chant Genres
Ex. II.20.2. Agnus 144 (St Gall, Stiftsbibl. 546, fo. 76v)
Ag-
-nus   De -  i      qui   tollis peccata mun-di mise-re-re
do- -na
i ffií f f r frfr f f j j ■> ;jj .
no-no-
-bis.   Ag - nus   De-       -i qui    tollis peccata mun-di
-bis pacem.
j »r r f J) j
3
mise-re-re no-
-bis.   (repeat first Agnus, with 'dona nobis pacem')
gentle, rocking rhythm. The first verse (repeated for the third) has an ABA form, the second begins differently, somewhat unsubtly taking up the music of 'Dei qui tollis', but then rejoins the music of the first.
For some melodies the earliest sources nearly all have trope verses. That is the case, for example, with melody 226. Following Crocker's prompting, one wonders if some melodies might not have been conceived from the first with trope verses. The case is different from that of the Kyrie, however; for the Agnus there is no previous record of singing extra verses, simply of repeating the same text over and over again. Atkinson has demonstrated that for melody 226, at least, the tropes are most probably later additions.
The text itself underwent a small change after about the turn of the millenium, when the third verse 'Agnus Dei . . . dona nobis pacem' begins to appear. Another common variant in the early sources, particularly in conjunction with melody 164, is the second verse 'Qui sedes ad dexteram patris miserere nobis', which seems to be a simple variant rather than an inserted trope verse.
11.21. CREDO
Huglo 1951; Stablein, 'Credo', MGG; Miazga 1976; Crocker, 'Credo', AG.
Three different Credo texts were known in the Middle Ages. The Apostles' Creed ('Credo in Deum patrem omnipotentem creatorem caeli et terrae') was often said as part of the preparatory prayers before the services of the office. The only known musical settings are farsed ones from special festal liturgies (see Ex. 11.23.19 below). Curiously, however, among the Greek ordinary-of-mass chants which are notated in a number of early sources (see VI11.2. v) is a Greek creed, and this is the Apostles' Creed. No source with diastematic notation is known. Likewise unknown in any notated source is the Athanasian Creed ('Quicumque vult salvus esse'), said at Prime. The Nicene Creed ('Credo in unum Deum patrem omnipotentem factorem caeli et
21. Credo
169
terrae') is so called because it is supposed to sum up the doctrinal beliefs established at the Council of Nicaea (325), though it was formulated somewhat later. It was originally part of the baptismal rite, the profession of faith of those about to be baptized. Eventually it was taken up in the mass, and is first found there in a Latin liturgy in Spain (Council of Toledo, 589). It achieved an official place (sung after the gospel) in the Frankish-Roman liturgy in a version by Paulinus of Aquileia (Council of Aachen, 798). It did not, however, become part of the liturgy in Rome itself until 1014, at the request of the German emperor Henry II, and then only for Sundays and those feast-days actually referred to in the text.
From the differing accounts and commentaries on the liturgy it seems that in the early Middle Ages it was sometimes recited by the congregation, sometimes sung by the clergy. Notated versions do not survive from before the eleventh century, and these are quite simple, either the syllabic chant known as Vatican I or variants of it. It was rarely copied with other ordinary-of-mass chants until the late Middle Ages. It was not usually troped, though farsed versions are known from the festal liturgies of Laon, Sens, and Beauvais. Only much later, particularly from the seventeenth century onward, were alternative musical settings composed in large numbers (many catalogued by Miazga).
Vatican I is built around two phrases, one lower in range, rising at the end EGa, the other higher in range, with bb (c in eastern sources), ending aGFG. These are stretched or contracted freely as the verses of the text require. Often the first is preceded by a supplementary phrase with the basic shape EFGFE, and the second may be succeeded by another with the shape EFGFaG. G plays the part of reciting note, though recitation is of secondary importance in accommodating the lengthier verses, and the effect is different from that of simple psalmody. Recitation-type melodies may, however, be found in the Hispanic and Milanese chant repertories (Rojo and Prado 1929, 123, PalMus 6, 316).
Huglo (1951) has suggested a Byzantine link with melody Vatican I, though the relationship must have survived many centuries of unrecorded transmission. Vatican II, V, and VI use variants of the same typical phrases as Vatican I. And Stablein has pointed out that a further melody in an Aquitanian source belongs to this group. This melody, reproduced here as Ex. 11.21.1, has the opening word in its plural form 'Credimus' (as does Vatican VI in its original form in Paris, Bibliotheque Nationale, lat. 887). It is copied in the mass for Whit Sunday.
There are numerous variants in the text from the standard form. The music seems to combine formulas for openings and cadences fairly freely, with recitation on both G and a. Many verses simply begin on G, but there is one popular opening gesture (marked 'a' on the transcription) and another used only twice ('b'). The two chief cadences have the skeleton aFaG ('x') and aFGE ('z') respectively, and a third cadence figure GaGFFJ ('y') is also sometimes used. The similarity to Vatican I lies in the tonality, particularly in the dominant role of E, G, and a, in the syllabic word-setting, and in the similarity of some motifs. The resemblance is in fact often closer to Vatican II.
170 //. Chant Genres
Ex. 11.21.1. Credo (Pans, Bibl. Nat. lat. 776, fo. 92v)
Credimus in u-num De-um     patrem om-nipo-tentem     factorem caeli et terrae
_tr_Z:_
**   * 0 b 0* 0m  0  *H _ m 0  m m   m*  m   m m~e>    0     , —
- -0-u- - - 0    wf   #-#- -
uisibilium homnium   et   inuisi-bi-li-um conditorem   et in unum dominům Ihesum Christum _Z._&_Oc_x._
000 0 0* 00 00g     t 00* Š 0 0   0 0 0 0 0t 00 0 0   0 0  0* m—
filium De-i uni-genitum     ex patre natum ante omni-a secula     Deum uerum
0*    m     •    0....... . * *        '     ' *' •* *
-■-•-"-0—0— -0- - -
de De-o uero     genitum non factum homo-u-si-o   patris   per quern omni-a facta sunt
0    0    •    0   0    00    0    0    0* 0 0   0  0~b   a       0*0 0 »***'* *   *   m  9 0 m  0 0
qui propter nos et propter nostram salutem descendit et incarna - tus est de spiritu sancto _2_z._a,_x-
•0 *
'0-0-
y
Natus ex Mari-a uirgine homo factus   passus est sub Ponti-o     cruci-fi-xus mortu-us
0*o 0
et sepultus terci-a di-e surre-xit Ascendit ad caelos sedet ad dexteram patris £___&,_x._S_
i
<tt* ■     *   t*  0   ,2   ,9  0^ ^    ^ > ,1* *   9   *    0j 0   0        00    0° 0    « 0*00
in-de uenturus in glo-ri-a iudica-re uiuos et mortuos cuius regni non erit finis _i_3_
et in spiritum sanctum dominům et uiuificantem   ex patre et fi-li-o procedentem
1
ty   ,   •* • 090*0 . 0 • 000 ,     «       t 0 . «2 „        .   f **" ' '   *    * ^ *
cum patre et fi-li-o a-dorandum et conglorificandum qui locutus est per prophetas __^_h._?k_
•    •    •      •       •        0 0     9     0    00000      •     0W  00t   M &-—
et unam sanctam catholicam atque apostolicam aecclesi-am confitemur unum baptisma __*_2*-£-
+»• 0   §90999     0M.0, 0****00 J0*
0-0 -*-*0-0-0—0-0-* 0"-
00 00
in remissionem peccatorum   expectemus resurrectionem mortuorum uitam futuri seculi Amen.
21. Credo
171
The other Credos of the Vatican selection are a mixed bunch. Vatican VII is from the Sens Circumcision office, stripped of its farse verses. Vatican III is a triadic F-mode melody of the fifteenth-century, in a vein which matches the Kyrie 'de Angelis'. Vatican IV, known as the 'Credo cardinalis', is of similar age; it is frequently found with mensural notation (see Tack 1960, 50).
The Vatican selection gives no hint of the vast numbers of Credo melodies composed particularly from the seventeenth century onwards. Miazga's catalogue lists over 500 variants of the Vatican I melody, and over 700 other melodies. His figures for the different centuries record 57 for the fifteenth century (the first when new composition becomes strikingly evident), 44 for the sixteenth, 110 for the seventeenth, and no less than 424 for the eighteenth century! It is not yet possible to see this enormous production in a proper perspective, for the whole history of plainchant in these centuries is still poorly known. The catalogues of the other ordinary of mass chants concentrate almost exclusively on the medieval period, and the numbers of melodies they register are not comparable. Miazga's research has also concentrated on eastern European, particularly Polish, and Italian sources, and much remains to be done for other lands. For France, for example, the Cinq messes en plainchant (1669) of Henry Du Mont (1610-84) are relatively well known, being sung in France until this century. Less well known is a similar mass by Lully, still in use in the nineteenth century, of which an extract is given here from the gradual printed under the aegis of the ecclesiastical commission of Digne (Ex. II.21.2). The word-painting at 'descendit' and 'ascendit' will not pass unnoticed.
Ex. II.21.2. From Credo by Lully (Graduel romain, Marseilles, 1872, p. 116*)
Qui propter nos homines,   et propter nostram salutem   descen - dit de coe-lis.
$ ' ' ■ ' T '   ^'' *' "    • '«.   >,  • * * - ^=^=
Et incarnatus est   de Spiritu Sancto     ex Maria Virgine:   ET HOMO FACTUS EST.
* . '       ' , ..... * *;■ • * * *
Crucifixus etiam pro nobis:   sub Pontio Pilato passus, et sepultus est.   Et resurrexit tertia
			
/L  k#--^-	-^ * fc» *	3       • b» M .	
fy     '    • *   .    • *•   m —	-í* *-	—m-«—s—"-	
di-e,   secundum Scripturas.   Et ascendit in coelum:   sedět ad dexteram Pa-tris.
172
//. Chant Genres
11.22. SEQUENCES
(i) Introduction
(ii) Early Sequences with Parallel-Verse Structure
(iii) Notation; Performance; Partially Texted Melodies
(iv) Short Aparallel Sequences
(v) Italian Sequences
(vi) The Early History of the Sequence
(vii) The Rhymed Sequence
Stablein, 'Sequenz', MGG; Crocker 1973; Crocker 1977; Crocker, 'Sequence', AG. Facsimiles: MMS 1, 3, 4; PalMus 15, 18; Vecchi, 1955.
Editions: Misset and Aubry 1900; Drinkwcldcr 1914; Mobcrg 1927; Anselm Hughes 1934; de Goede 1965; Eggeii 1968.
(i) Introduction
The origins of the sequence are so much disputed that I have chosen to present first an account of the sequences which appear in books of the ninth to eleventh centuries, the 'first epoch' of sequence-writing. The evidence about the early history of the sequence is reviewed in section vi below.
Sequences were sung at least from the ninth century onward after the alleluia at mass on feast-days (sometimes also elsewhere in the liturgy, for example as a substitute for the Vespers hymn). While an alleluia at a less important ceremony would be performed alleluia—verse—alleluia, on a high feast the pattern would be alleluia— verse—sequence. (See, for example, the rubrics in some Sarum books of the later Middle Ages: Dickinson 1861-83, cols. 9-10.,)
Most sequences were constructed in paired versicles, each line of music sung twice to different words. (In some sequences not all versicles are paired.) Some pieces, which may be called sequences because of their liturgical function (they follow the alleluia at mass on feast-days), are much shorter and are not constructed in parallel versicles (section iv).
The early sources of sequences are very disparate in both character and contents. The discussion of the early or 'first-epoch' sequence below is restricted to a few of the compositions which belong to the period up to about 1000. About 150 melodies were used in different parts of Europe up to this time. Many of them seem to have been local compositions which did not travel outside a particular area. The number of melodies known 'internationally', that is in both west and east Francia, is small, only about thirty. Since most melodies were sung with different texts in the different areas, the number of texts known internationally in the tenth century was practically nil, though the total number of texts was already quite considerable.
Little is known about the early stages of the Italian sequence tradition, but it seems as if the notion there of what constituted a sequence was rather different from that
22. Sequences 173
prevalent in the north. Italian compositions are treated briefly in a separate section below (v).
I have used the term 'sequence' to refer to the genre in general, with or without text. If I have wanted to be specific, I have referred to 'sequence melodies' and 'sequence texts'. Other writers (see Husmann 1954, 'Sequenz und Prosa') have preferred to reserve the word 'sequence' for the melody alone, calling the text 'prose'. Although this corresponds to the practice of some medieval manuscripts, it was not a universal custom. (I know of no witness earlier than the seventeenth century to the use of the word sequela for a sequence melody: see Gautier 1886, 14.)
(ii) Early Sequences with Parallel-Verse Structure
Some characteristics typical of many larger sequences of the early period are exemplified in Ex. 11.22.1. This sequence is regular in structure, in that its verses are all paired. As in several other sequences, one of the verses has within itself an AAB structure: verse 6, Virginum O regina. Again like several other sequences, there is a distinctive change of register part way through the piece: verse 1 moves between E and c, verse 2 pushes a little higher to d, consolidated in verse 3. Then in verse 4 the cadence is made on d itself, which provides a base for another move, up to high# in verse 6. Verse 7 returns to the register of verse 5 (a—f), before the final string of verses in the highest register.
It is worth taking note of the neumatic notation of the melody. From the text alone it is clear that in the paired versicles two-syllable words are very often matched with two-syllable words in the parallel versicle, three-syllable words with three-syllable, and so on. The note-groupings of the melody when it is notated without its text (this is the usual practice in early sequence collections, but it died out in the twelfth century) also reflect the syllable groups. (On this oft-discussed phenomenon see Reichert 1949 and Schlager 1980.) Ex. 11.22.2 shows verses 2 and 3 with each compound neume and its corresponding text.
(Vans vocibus is the most widespread 'western' text for the melody of Ex. 11.22.1; Ex. II.22.3 gives the start of Eia recolamus, the best-known 'eastern' one. Although Eia recolamus has no verse corresponding to verse 9 of Claris vocibus, the two are remarkably close in matters of note- and syllable-groupings, as can be seen from verses 2 and 3 once again. Eia recolamus has an 'extra' first verse (labelled 'A'), which texts the corresponding 'Alleluia' preceding Claris vocibus. This is a characteristic difference between western and eastern practice. The word 'alleluia' contains two instances of what is known as liquescence, where a special notational sign indicates a special manner of performance of the first / and the i. Exactly similar provision is made in the new text, for i in 'Eia' and gn in 'digna'.
While the correspondence between the western Claris vocibus and the eastern Eia recolamus in note- and syllable-groupings is close, this is not true for all the melodies known in both parts of northern Europe. Exx. II.22.4-5 give the western Organicis canamus and the eastern Sancti baptiste (the text is attributed to Notker Balbulus of St Gall), for the melody labelled 'Justus ut palma' in some early manuscripts. In order
174
II. Chant Genres
Ex. 11.22.1. Sequence Claris vocibus (Oxford, Bodl. Lib. Bodley 775, fos. 122v, 141'; London, Brit. Lib. Egerton 3759, fo. 73v)
Í
9   *?4  *9 ** '*••   fo 9
Alle- -lu - ia.
Claris uo-ci-bus in-cli-ta ca-ne turma sacra me-lo-di - mata. Vo-ci mens bene conso-na     sonent uerbis pneumata concordi - a.
# ' é
Di-ui-na robus-to tetrachor-da plectro docta manus pe-ri-te fa-ci-at. Resul-tet uirtu-tum pi - a li - ra De-o sonans nunc dramata dulcis-so-na.
—9— a    . a -ti
Ast ar-mo-ni - a hec di-ui-na so - no-re uir-tutum li-qui-dis-si-ma. Mixta cas-titas est quas intra     in se-de   lo - ca-ta mixto li - di-ca.
1
-9-9-
31
Cuius in te consis-to-ri - a sunt immu-ta-bi - li - a. Que mater es in-ui - o - la-ta      uir-goque pu-er-pe-ra.
*   •   •   * *
-9-9-
Id-cir - co tu-a De-um fu-e - re digna fer-re uisce-ra. Quem nunc celi-ca necne terre-a cuncta laudunt spati-a.
-*---«-«-9-•-•-.-m-*-*-,-*-
i
10
Virgi-num O re-gi-na te ca-nimus Ma-ri-a per quam fulsere cla-ra mundo Tu sa-lus orbis  al - ma tu ce - li por-ta facta    per te se-cu-lo  ui - ta omni
lu - mi - na. de - di - ta.
-9—•-
Ce-li-cis terre-a tu iungis di - ui-nis hu-ma-na. Pa - ra-di-si - a-ca per te nobis pa-tet ia - nu - a.
.      0     m     *      '     *     '     0     .      .-9      9      * •
•     m 0-
Ades-to fa-mu-lis pi-is-si - ma in-flu-a iam suspendens prece peri-cu - la. Au-di  fi-de-li-a preca-mi-na      in-pe-tratam ce-li-tus de-ferans ue-ni-am.
3
Ut qui-e-tem nobis temporum in-clitam. Hac in ui-ta nostra di-ri-ge o-pe-ra.
9   0 9
Post fu-ne-ra u-ra-ni-a nos due ad ha-bi - ta-cu-la. Quo le-temur omnes u-na tecum per cuncta se-cu-la.
P
. '...... «
Exuianient nunc omnigena   a-men redempta.
22. Sequences 175 Ex. II.22.2. From sequence Clans vocibus (Oxford, Bodl. Lib. Bodley 775, fos. 122v, 141r)
J1 /•• «.   h. ft
2 Diuina     robusto     tetrachorda plect Resultet uirtutum   pia   lira Deo
3 Ast armonia      hec diuina sonore Mixta castitas   est quas intra in sede
ft i\ ť
ro     docta manus perite faciat.
sonans nunc dramata dulcissona.
uirtutum liquidissima.
locata mixto lidica.
Ex. II.22.3. Start of sequence Eia recolamus (St Gall, Stiftsbibl. 381, p. 336; Rome, Bibl. Cas. 1741, fo. 83r)
i
-0**0 ' '   m    *   * fo
E - ia re-co - la-mus laudibus pi - is digna.
7  1    K cT* AÄ /-. r. ť
Hu-ius di - e - i car-mi-na       in qua nobis lux o - ritur gratis - si-ma. Noctis in-ter ne-bu-lo-sa       pe-re-unt nostri criminis umbra-cu-la. -lS___'^f      r/r    \      fAh_ Yl J*
.....
Ho-di - e se-cu - lo maris Stella est e - ni-xa no-ue sa-lu-tis gau-di-a. Quern tremunt baratra mors cruenta pa-uet ip-sa   a quo peri-bit mortu-a.
Gemit cap-ta pestis an-ti-qua co-lu-ber li - uidus perdit spo-li - a. Ho-mo lapsus o-uis abduc-ta      re-uo-ca-tur ad e-ter-na gau-di-a.
to facilitate discussion of the correspondences between the melodies I have numbered the lines in a special way. The eastern version, Sancti baptiste (Ex. II.22.5), is the more regularly parallel in structure, though there is a slight irregularity in verse 4, and 3X is like a modified repeat of one versicle only out of verse 3. The western version, Organicis canamus (Ex. II.22.4), has nothing to correspond with verse IX of Sancti baptiste—or should one say that verse 1 of Organicis corresponds to verse IX, not verse 1, of Sancti baptiste! Verse 4 of Organicis is complex. Its melody may be represented by the formula AAB, AAB, A AC, CD, where D is actually the same as the end of verse 3, transposed up a fourth. Compared to this one might say that Sancti baptiste ignores the first A, except for three notes in ascending order, which reappear at the end of the line. Stretching a point, one might thus represent the verse as A'ABA', ABA'. Finally, in Organicis verse 5 is but a single versicle, and runs straight into verse Z.
Some of the differences within corresponding verses are also of interest. The rise into a higher register so obvious in Organicis does not happen in Sancti baptiste. Both
176 //. Chant Genres
Ex. II.22.4. Sequence Organicis canamus modulis (Orleans, Bibl. Mun. 129, fo. 166r)
ffi    '     *       , .      '       •       * M      *       • S
Or - ga-ni-cis cana-mus modu-lis nunc Iohannes sollempni - a. Om-ni - ge-nis domi-no uo-cibus  reddentes odas de-bi-tas.
Qui in Sanctis su-is  mi-ra - bi-lis ni-mis......
Nam et in ip-sis quasi quibusdam mu-...-si-cis
.0*tf»'*»0* 0 0
multi-pli - ci uir-tutum flore e-os-dem deco-rat ac mi-ri-fi-ce ad-ornat. instrumentis di-gi - to propri - o fi-des   a-gi-tat      fi-des uirtu-tum so-norat.
Has numero-se percurrens singu-la Quam generat uir-tu-tem mater ilia
-•-*-—*—»-
permiscens singulis di - a - tessa-ron ... melli-flu-am me-lo-di-am. que a ~ l'-'s  decenter compo-si - ta      reddet su-auem simpho-ni-am.
3
Qua si-ne cuncta...      fi-unt dis-so-na      necnon et fri-uo-la. Qua  cum om-ni-a       fi-unt conso-na      necnon u - ti - li - a.
(4) ,   *   '   *   *-     ,  «•   '   '    *   \\:   '   '    *   '   '    ' '    , :]\
Qua ius-ti be-ne       mora - ti ri - te      pe-tentes excel-sa ... po-li   si-de - ra......
A-lacres decantant no-ua canti-ca in cytha-ra
Tre-i - ci
Quorum a-gentes festa     consor-ci-um me-re - amur in ce-lesti   pa-tri-
-men.
22. Sequences
Ex. II.22.5. Sequence Sancti baptisté (St Gall, Stiftsbibl. 546, fo. 124r)
177
Sancti        bap-tis-te Christi preco-nis.
&~ * * ' , * * *     * F=7
Solem-ni - a ce-le-brantes mori-bus ipsum se-quamur. Ut ad ui-am quam predixit asseclas su - os per-ducat.
De-uo-ti te sanctissi-me ho-minum Ap-parensque Zacha-ri-e Ga-bri-el
. •*-' *
a-mice Ihesu Christi fla - gita-mus ut gaudi - a per-ci-pi - a-mus.
re-pro-misit qui tu-am ce - lebrarent ob-sequi - is na-ti-ui-ta-tem.
IX
--—*#—a—«—■-*—•  ^ *——*——«—
Ut per hec festa e-terna gau-di - a ad - i-pis - camur. Qua san-cti de-i   sacris de - li - ci-is   le - ticongaudent.
F—»
Tu qui prepa-ras fi-de-li-um corda Te depos-cimus ut cri-mi-na nostra
—*----'-*—r-
Ne quid deui-um uel lu-bri-cus Deus in e - is in-ue-ni-at. Et  fa-ci-nora con-ti-nu-a pre-ce      stude-as ab-sol-ue-re.
Pla-catus       ut ip-se su-os      semper in - ui-se-re fi-de-les.
Et mansi-o-nes      in   e - is   fa-ce-re dig-netur.
3X /k   «   •   «   * :»      m   .        . =
Et ag-ni uel-le-re quern tuo di - gi - to.
Mundi monstraueras tol-le-re cri-mi-na nos uelit in-du - e - re.
é     *-,-X-*-----»-«-,-
Ut ipsum mere - a-mur an-gelis as-so - ci - i. In al - ba ueste se-qui per portam clarissi-mam.
Í
.   *   .   . ^   * «
A-mi-ce Christi Io-hannes.
178
//. Chant Genres
versions get as far as high d, but only Organicis presses on tog, hence the difference of a fourth at the final cadence. Almost equally noticeable is the way in which verses 1 and 2 in Organicis have a three-note cadence from the sub-final, CDD, which is absent in Sancti baptisté. The cadence is very common in sequences, somewhat more so in the west than the east, and is reputed to be of 'Gallican> origin. The text of Organicis has constant line-endings in 'a', which is usual in the western repertory, whereas Sancti baptisté does not, which is usual for Notker and some other eastern writers.
Just how the differences between the two versions arose is open to debate. There is close agreement about how verses such as 3 and 4 should proceed. Is the 'Gallican' cadence original or a later stylization? Many copies of Organicis (for example, the one presented by Crocker 1977, 288) have a double-versicle verse 5. Is that a later regularization of the melody? Crocker (1967, 'Sequences', and especially 1977) argues forcefully that regular parallelism was often imposed upon sequences at a later stage in their formation. Quite often it is the eastern versions which are less regularly parallel, and several of Crocker's presentations show how verses, single versicles, or segments within versicles, may have been added in the west, whereas the eastern versions preserve an earlier irregular state of the melody. (See especially Crocker 1977, chs. 3-4.)
The great variety of form displayed in the early sequence repertory makes the choice of what to mention here somewhat arbitrary. The obviously non-Gregorian character of the melodies has tempted some writers to suspect the influence of secular music, for example in the sequence with text Plangant cigni or Clangant cigni (and various later texts), entitled 'Planctus cigni' in some sources (see Stáblein 1962, 'Die Schwanenklage', also Stáblein 1975, 114; the melody is also in Anselm Hughes 1934, 63). This is one of many sequences which have a non-religious title; others are named after musical instruments (see Holschneider 1975). Sacred titles are usually alleluia incipits.
The repetition of segments of verses is very common in sequences. In some melodies, whole verses are repeated (e.g., Laudes deo, Crocker 1977, ch. 3; also the melody 'Hieronima' or 'Frigdola', Crocker 1977, ch. 6; and the melody of Laudum cannina, a sequence for St Benedict, given in Hughes 1934, 37). Special interest has focused on a few melodies where a group of verses is repeated as a block. An example of this is the well-known melody called variously 'Chorus', 'Concordia', etc., with texts such as the eastern Hanc concordi famidatu (attributed to Notker of St Gall) and the western Epiphaniam domino (the melody is discussed in Stáblein 1964 and Crocker 1977, ch. 5; see also Handschin 1954, 154). At its most repetitive (not all versions of the melody display this scheme), in Gaude eia unica, it has the form X, ABC, ABCB, A, Y (where X and Y are the non-repeated verses often found at the start and end of early sequences). A shorter example is the less well known Pura deum (Ex. 11.22.6) (also known with texts De sancto Iohanne and Pangat simid eia, all restricted to the area of north France and England; Stáblein 1978 thought Pange simid was the original, perhaps composed by Hucbald). The most obvious feature of the melody is the large-scale repeat involving all the central section of the melody. Added to the fact that the last half of verse 1 reappears in verse 4, this means that
22. Sequences
179
Ex. II.22.6. Sequence Pura Deum (Angers, Bibl. Mun. 97, fu. 104r)
Al-le-lu-    -ia. Pu-ra De-um laudet in-no-cen-ti - a.
Í
Innocens Christus su-a quam sacrat infan-ti - a. Par-uu-lorum ex mořte martyrům pre-ti-o - sa.
v        •        *   » * ' m ~r
Quos se - ui - ti - a.........He-ro-dis da - re ne-ci ius-serat.
Cum ex-tingue-re ni-ti-tur il-lum qui uitam da-re ue-nerat.
U         n. f						
Jt       «•' L						
			#			
?S-«b *	—w—	—w—		# #		
Natum namque regem audi-e-rat. Statim mi-li-tum pergunt agmina.
•   *   * _g_ •
•   •   *        *   • r        • —r
-*—r
V
Quern succe-de-re si-bi posse   in - a - ni  pa-uore for-mi-dabat Ac tyran-ni-ca complent iussa mili - a mactantes in - nocentum
qui-a non ce-li regem no-ue-rat. inter dul-ci - a matrum u-be-ra.
w (0) ,.....•   t   ,   *   9   . *
Mox in-ui - di - a ac-census et   i - ra fer-ui - da infantes   a  bi-matu et infra.
Q im-pi-e-tas O se-ua ni-mi-um fe-ri-tas O nunquam audi-ta cru-de-li-tas. —-i4-9*-
"T*—2—•—•—•—•——r
-Mů- -
Oc-ci-di iubet in totam Da-uidis Bethleem pari-ter ex e-ius per cuncta confi-ni - a. Inson-tes culpa non est que noxias fa-ci-at a-li-qua et mortis subeunt dis-crimi-na.
O Christi miran-da semper et in is-tis gra-ti - a. Pro se passis e-ter-na quibus est largitus premi - a.
—»—*—m ^_  ' *—•*   »w •   m-^—#—
Ip-si laus sit nunc et in se-cu-la.        A-men.
nearly all the music is recalled at one point or another. What is more, the music of the recapitulated middle section contains internal repetitions, and also relates to the outside verses: 3S is the same as 3V and these start like the other phrases in verse 3; 3W and 3X are the same; the cadence of 3W and 3X is that of verse 2.
Pura deum, or rather its other text Pange simul, was hailed by Winterfeld (1901) as an example of the so-called 'sequence with double cursus'. Whether it should be so
180
//. Chant Genres
linked with the other members of this group is open to question, in view of the fact that, unlike the others, it seems to have been at home in the liturgy from the start and contains many melodic resemblances to other liturgical sequences. The double-cursus genre is discussed below (II.24.ii).
Several things, then, play a part in making a sequence recognizable by its musical style. The parallel-verse structure, and the text-setting on the principle of one syllable per note, are the most important features, and then perhaps recurrent figures at cadences and a few other points. Repeated notes are rather rare, and there is thus a constant impression of movement, occasionally locked in circles of repeated motifs but more often pushing purposefully towards a clear melodic goal: the frequent surges up a fifth into a new range are the most obvious manifestations of this. The melodic goals referred to are usually limited to the final, with occasionally a higher final which supersedes the first, and sometimes also the lower fourth or fifth. This tonal single-mindedness creates an impression of rather insistent enthusiasm; the melodies rarely sound reflective for long, especially not when the texts are sung, for the declamation of the words in syllabic fashion tends to emphasize individual notes at the expense of the melodic phrase as a whole.
(iii) Notation; Performance; Partially Texted Melodies
Sequence melodies were recorded in a number of different ways. In some early sources the melodies alone are copied, without texts, for example in Chartres, Bibliotheque Municipale 47 (PalMus 11) and St Gall, Stiftsbibliothek 484 (Crocker 1977, pi. 5; Stablein, 'Sequenz', MGG, pi. 2). Another of the earliest major sources, by contrast, Paris, Bibliotheque Nationale, lat. 1240, has what became the commonest method: each syllable of text carries its own notational sign. In addition to the latter, most early West Prankish sources have a collection of melodies alone (see Crocker 1977, pis. I — II and Crocker, 'Sequence', A'G, 142; also Frere 1894, Winchester; Holschneider 1978; Stablein 1975, 115, 117).
East Frankish manuscripts usually place melody and text side by side in parallel columns, whereby each phrase of the melody (and the corresponding part of the text) is given a new line. The internal construction of the sequence is made particularly clear. (See Stablein, 'Sequenz', MGG, pi. 1 and Stablein 1975, 185, but especially Haug 1987).
A few sources, such as Rome, Biblioteca Angelica 123 (PalMus 18), have alternating verses of text and melody alone. A few? north French sources (such as Laon, Bibliotheque Municipale 263 and Cambrai, Bibliotheque Municipale 78) give the complete melody immediately after the complete text.
Whether the notation of melody and text separately in these ways reflects some sort of alternatim performance (text—melisma-—text—melisma, and so on) has been much debated (see e.g. Husmann, 1954, 'Sequenz und Prosa' and Smits van Waesberghe 1957 'Over het onstaan'). A systematic survey of rubrics in liturgical
22. Sequences 181
manuscripts might help clarify the matter, though such rubrics are rare before the thirteenth century. (The situation is thus less satisfactory than for the prosula, discussed by Kelly 1985, 'Melisma and Prosula'.)
The problem of the double notation, with and without text, is particularly critical in the case of a small number of sequences where in the midst of the melismatic version a few texted phrases are to be found. These phrases also appear embedded in the complete text for the whole sequence. If another text is provided for the sequence melody, then the same phrases are used again: they go with the melody. Their origin, function, and manner of performance are all puzzling and have inspired a good deal of speculation (see Blume in AH 49; von den Steinen 1946-7, 205-20; Husmann 1954, 'Sequenz und Prosa', 77-91; Stäblein 1961, 'Frühgeschichte', 8-33). The use of partial texts seems restricted to France (and north Spain) and England; the fashion for such pieces persisted longer in Aquitaine than elsewhere. (For an unusual Spanish example see Husmann, 1961, 'Ecce puerpera'.) It is not at all clear how an alternatim system of performance, or indeed any other system, would accommodate these special verses. (See, for example, Husmann 1954, 'Sequenz und Prosa', 86; on a peculiar method of notating the sequence Terribilis rex alme with partial text Gloria victoria, etc. in Beneventan sources, see Stäblein 1961, 'Frühgeschichte', 21.)
(iv) Short Aparallel Sequences
All early sequence collections are dominated by compositions where double versicles predominate, and where the number of verses is at least five and often ten or more: this means that the number of notes, even ignoring all repeats, often surpasses 200, and in the largest sequences, such as Fulgenspreclara (Anselm Hughes 1934, no. 23), may pass 300. But most early collections also include a smaller number of quite different melodies lacking parallel-versicle structure and usually no more than about seventy notes long. Without the clear structural features of the larger sequence, they sound simply like an alleluia jubilus, and indeed, although some of them have a small amount of internal phrase repetition, this does not go beyond what one would expect of an alleluia jubilus.
Among the 150 or so sequence melodies known up to c.1000, only twenty fall into the short aparallel category; and only one was widely known (that is in Winchester, Aquitaine, and eastern sources): that known as 'Excita domine', with text Qui regis sceptra, in the west, and as 'Laudate Deum [omnes angeli]', with text Angebrum ordo sacer, in the east (the titles refer to the alleluias which begin like the sequence; both have the same melody). Most have been surveyed by Kohrs (1978).
Ex. II.22.7 shows three short aparallel sequences which take as their starting point Alleluia Ostende. The first sequence was widely known in the early period, and was usually sung with the text Precamur nostras. The other two are known only from Paris, Bibliotheque Nationale, lat. 1084 (Aquitaine, late tenth century), from which manuscript all three are edited, with the titles given there. Ex. II.22.7 is ordered as follows:
182 //. Chant Genres
Ex. II.22.7. Alleluia Ostende and associated sequences (Paris, Bibl. Nat. lat. 776 (a-b) and 1084 (c-f) )
776 f.5v
Al - le-        -lu- -ia. 776 f.5y__
Osten - de nobis  do- -mine mise - ri - cor-di-am tu- -am et saluta - re
tu- -am da nobis.
1084 f.l97v 'Precamur
c
Al - le-     -lu- -ia.
1084 f.221r
-»-m-*- -—•-*-*-#
Precamur nostras Deus a - ni-mas et consci-en-ti - as dig-ne munda ut nunc nostra
Christus ue-niens corda si-bi in-ue-ni-at pa-ra-tam. 1084 f,197r 'Seq. de Ostende1
y
Al - le-       -lu- -ia.
1084 f.207v 'Ostende'__
Al - le-    -lu- -ia.
22. Sequences 183
(a) Alleluia, with jubilus
(b) Verse Ostende nobis
(c) Sequence (first)
(d) Text Precamur nostras for (c)
(e) Sequence (second) (/) Sequence (third)
The pitch of the neumes of Paris 1084 is not always certain (cf. Kohrs 1978, 150). But the essential fact is clear enough, that the only melodic correspondence between alleluia and sequences lies in their opening. There are no internal repetitions in the first sequence (r), but sequence (e) begins with a repeated phrase, and sequence (/) has two repeated phrases. Both (<?) and (/) end unexpectedly on an a, and it might be hypothesized that the melody should be completed by singing the jubilus of the alleluia. The effect of the text Precamur nostras is similar to that of an alleluia prosula.
(v) Italian Sequences
Brunner and Jonsson have recently emphasized that the regularity so conspicuous in (and occasionally imposed upon) sequences in northern lands was not as important a criterion for Italian composers or redactors. Several sequences of Italian origin display considerable irregularity of structure, and Italian versions of widely known sequences are often less regular than northern versions. Stablein (1964, 379) was tempted to speak of a 'degeneration of the sequence-principle [i.e. parallel-verse structure] in the direction of the litany or psalmody'. In Italian sequences one finds frequent singles, as opposed to the regular doubles of most northern pieces, or even a disregard for paired versicles. The strictly syllabic word-setting of the northern compositions seems also to be less important; two notes are often coupled for a single syllable, and notes are sometimes repeated to accommodate extra syllables.
Even more striking is the appearance of a small number of sequences which display traces of what might be called a variation technique. Here the same melody is used, with variations, for most verses of the composition. Levy (1971) was the first to draw attention to this, in an analysis of the sequences Lux de luce and Hodie dominus Iesus Christus. A third sequence of this type, Alma fulgens, has been cited by Brunner (in an unpublished paper, Milan, 1984; for a bibliography of sequences in Italian sources see Brunner 1985).
Many northern compositions were sung in Italy, and many sequences in the more usual manner were composed in Italy. The examples mentioned here are rare. Yet the Italian type is historically significant. It is related to the Italian practice of composing Kyrie verses with Latin text, where the same melody is used for all verses; and a number of troped Agnus Dei chants are also constructed in the same way (see below II.23.viii and xi).
Ex. II.22.8 shows Sancte crucis celebremus, which betrays its Italian origin by the
184 //. Chant Genres
Ex. II.22.8. Sequence Sancte cruris celebremus (Modena, Bib]. Cap. O.I.7, fo, 129"")
la  <fy   ,   ,   '««       '   '   '   • / ,
Sancte cru-cis ce-le-bre-mus de-uo-ci-o-ne ue-ne-randa.
ib e\) 0      ' ^
Signa - culum   triumphalem per quern sa-lu-tis sacramentum.
2a    (fa 0     •     *        0      *    0   m   0 t^m m00 0**0m£0»
-0-0-*-*--9-0-*-■-*-
Sumpsimus culpam que protoparenti     e-um e-ramus e-xules facti pa-tri-e.
2b |> , . P • f' ' *■ ' t '       » - f*   '   ' '" ' * * e • =
Redempti   ergo   graci-as a-gamus   per lignum que sancte redemit crucis.
Et me-ri-tis nos quamuis in-digni   ui-si-tauit gra-tu-i-ta de pi-e-ta-te.
3b    j[)   .    f '......~m     m* *
y
In do-mi-ni cruci-fi-xi laudes consonat uo-ce   mel-li-flu-a conci-namus can-ti-ca.
4a   (fa !> 0    •      0     0 0   *        0        *    •   0   *   in   0   •'   * *   0 *
O crux glo-ri-o-sa     O crux ue-ne-randa     que preci-um mundi    fe-re tu me-ru-is-ti.
4b    ^   '   '   m   '   0    * 0   *    *   •    •**  '   ' '*' ' ' 0   & ^ *
Salue presente   humilem ple-be   in   laude tu-a   ho - di - e congregatam.
== .   & • 0I  *0' £ , '\,  0''    »1 0
Rex mundi fu-is-ti   por-ta-re di - gni   ta - len-ta.
frequent two- and three-note groups assigned to some syllables. The constant deviations from exact parallelism are also typically Italian. This is not an example of the 'variation-versus', although there is an echo of such pieces in the way all verses except 2 have similar cadential phrases.
When northern composers composed a new text for an old melody, they usually adopted the note-count and note-grouping of the melody exactly. This was not always the case in Italy. When the melody of Sancte crucis celebremus was used in the Modena manuscript for another sequence, Sanctum diem celebremus, phrases were stretched or contracted freely and new material was added.
The early history of the sequence in Italy remains obscure, for practically no Italian
22. Sequences 185
sources survive from before r.1000. The inevitable impression is that Italy was less productive than other lands. Nevertheless, some churches were certainly active in sequence composition (see the information on south Italian sources in Brunner 1981), and they were sung even in the Old Roman repertory by the eleventh century (see Lütolf 1987 and MM MA 2). It is at least possible that later Italian sources preserve compositions dating back as early as any in the north. On the other hand, the stylistic discrepancies between northern and Italian manners of sequence composition may suggest that Italian musicians had to assimilate the sequence as a foreign genre, and occasionally fell back on native habits of composition which assorted uneasily with the 'true' sequence.
(vi) The Early History of the Sequence
The earliest mention of the sequence by name comes in the late eighth-century manuscript Brussels, Bibliotheque Royale 10127-10144, the 'Blandiniensis', a portable booklet containing among other things the texts of chants to be sung at mass, probably written in north France (ed. Hesbert mAMS). Here six alleluias in a list of twenty-five for the post-Pentecost Sundays are rubricated 'CUM SEQUENTIA'. All six are found in later manuscripts either as alleluias with extensive melismas at the end of the verse (melismas different from the alleluia jubilus), and/or as short aparallel sequences. (See Stäblein 1961, 'Frühgeschichte', 4-7; Kohrs 1978, 78 ff.; Crocker 1977, 393 ff.) The six alleluias are all for 'ordinary' Sundays of the summer months, when no special commemoration or feast was involved. Sequences for these days are generally rare in later sources. Were longer sequences already being sung on the more important feast-days of the year? The Mont-Blandin manuscript makes no mention of them.
About 830, Amalarius of Metz, describing the alleluia of mass, refers to 'Haec jubilatio quam cantores sequentiam vocant' (ed. Hanssens, iii. 304; PL 105, 1123; Crocker 1977, 392). These melodies might be either short or long ones. Sequentiae are also mentioned in almost the same terms in the late ninth-century Ordo Romanus II (ed. Andneu, ii 215) and by pseudo-Alcuin (PL 101, 1245).
Manuscript copies of sequences (melodies alone, or texts alone, or texts with notation) finally appear late in the ninth century, that is, at the same time as notated copies of most other chants, so that from this circumstance alone one cannot say that the sequences are a late development of the chant repertory. (See Stäblein, 'Sequenz', MGG, ex. 1; Stäblein 1961. 'Frühgeschichte', 7, with facs. facing 16; the melodies in Chartres 47, PalMus 11; and von den Steinen 1946-7, 252-63).
Again during the late ninth century the monk Notker Balbulus of St Gall (c.840-912) wrote a number of texts for sequences. (Documentary evidence exists for a few, but the rest are attributed only on stylistic grounds, the canon established by von den Steinen, 1948, having become generally accepted. See also Stäblein 1962-3; Crocker 1977 and Crocker, 'Sequence', NG.) The collection is usually reckoned to have been completed around 880, for it was dedicated to Bishop Liutward of Vercelli, counsellor and arch-chancellor of the emperor Charles the Fat, in 884.
186
II. Chant Genres
Several manuscripts preface the collection with a remarkable prooemium, apparently written by Notker to explain and dedicate his work to Liutward (critical edition and German translation in von den Steinen 1948, English translations Crocker 1977, 1 and 'Sequence', NG; see also the discussion in Husmann 1954, 'St Galler'). The gist of it is that Notker as a boy had difficulty remembering 'melodiae longissimae'—which in the context must mean sequence melodies—and wondered what could be done to make them easier to learn. A monk, fleeing from the sack of Jumieges in what is now Normandy, came to St Gall with a chant-book in which Notker saw that texts had been composed to fit the sequence melodies ('versus ad sequentias erant modulati'). Notker thought he could compose better texts. He showed his first attempts to his master Iso for correction (Iso seems to have been familiar with the technique of texting melismas), then went on to produce a whole collection. His master Marcellus had them copied out individually on rolls to be shared out among the choirboys for learning.
In support of Notker's story, it has been demonstrated that texts found in the west Frankish repertory lie behind some of Notker's compositions (Crocker 1977). Sometimes he reworked ideas in previous texts, sometimes he developed his own original ideas.
If von den Steinen's list of Notker's texts is accurate, there are forty texts to thirty-three melodies; eight texts are for eight short, aparallel melodies; none contain special verses (partially texted melodies). This compares with twenty-nine melodies in Chartres 47 (ten short aparallel, no special verses), and with thirty texts for twenty-nine melodies in Paris, Bibliotheque Nationale, lat. 1240 (no short aparallel sequences, four with special phrases; listed in von den Steinen 1946-7, 126 and Crocker 1958, 'Repertory', 154-5). Thus there is no uniformity between these crucial early witnesses as to the type of sequence they favour.
Nor is there any uniformity in the assignment of sequence melodies to particular feast-days (see, for example, Table IV in Crocker 1977, 404-5). The impression is that it was often the text which made a sequence proper to a particular feast (several texts are nevertheless neutral in this respect, being generally laudatory, celebratory, or christological without implying a connection with any particular feast). The melody, by contrast, could be employed for any feast as the author of the text thought fit.
In addressing the vexed question of the connection between alleluia and sequence, it is often forgotten that the alleluia melodies themselves can have had little liturgical fixity before the ninth century. Alleluia and sequence may well have followed parallel paths, both starting from a small repertory, perhaps only a dozen melodies for each genre, the verses of the alleluia attaching it to specific occasions, texts doing the same for the sequence. The composition of different melodies, verses, and texts did not proceed uniformly across Europe. By the time Notker's hymn-book was completed (r.880) and the early collections of west Frankish texts were made (c.930 for Paris 1240), considerable divergence existed between the liturgical assignments of sequence melodies in different areas of Europe. Whatever connections may have existed
22. Sequences 187
between particular sequences and alleluias, these gradually dissolved, although they are still detectable in a few cases in the late ninth- to tenth-century repertory that has survived.
The state of the Old Roman alleluia repertory (earliest musical codification only in the eleventh century, however) and the Milanese repertory (musical record only from the twelfth century, however) lends some support to this hypothesis. The Roman repertory had only three widely used melodies, with some twelve others used less often, eight of them once only. At Milan, the same melody was sung on practically all major feasts, supplemented by extra melodiae for festal occasions (see VI11.4). In comparison with these, the chief distinguishing Frankish-Gregorian developments would be the rapid multiplication of alleluia and sequence melodies in the ninth century and the provision of texts for the sequences. (On supposed melodic relationships between alleluia and sequence, see Husmann 1955, 1956 'Alleluia, Vers und Sequenz', 1956, 'Mater-Gruppe', 1956, Tustus ut palma'.)
Crocker has argued on several occasions against this train of thought (see, for example, 1977, 400). For him, the sequence is essentially a new creation of the ninth century, arising out of the musical impulses and ambitions of Frankish musicians and only gradually assimilated to the alleluia of mass. There is thus no need to establish connections with hypothetical alleluia repertories, or to explain away conflicting liturgical assignments of melodies. Furthermore, while allowing for the possibility of contrafact texts, it brings the composition of melody and text into same period, in fact makes them part of the same compositional process. Thus Crocker believes only the short aparallel melodies were used in the mass before the mid-ninth century. The larger sequences belong to a radically different creative effort.
An indication of how things might have proceeded is suggested by the case of the melody 'Mater', already discussed by Husmann but revisited by Bower (1982). Bower was able to identify an alleluia melody (Schlager catalogue 274) which shares significant amounts of musical material with the sequence, but which cannot be shown to be any older than the sequence. Sources for both date only from the later ninth century.
It is nevertheless possible to imagine a compromise explanation, one which sees the alleluia as the genre from which the sequence developed but does not deny the originality of much of the new repertory. The rapid and concurrent expansion of both alleluia and sequence repertories in the ninth century which I have envisaged, into the dozens of melodies known by <r.900, and the hundreds known by c.1000, allows both for a connection with the alleluia and energetic and inspired new composition. (None of these developments warrants the description of the sequence as a 'trope' of the alleluia.)
Crocker has argued that most sequences began life as the original creations of a poet/musician composing text and music together. Differences between one version of a melody and another would therefore be the result of deliberate revision. Yet the differences between eastern and western versions of many 'internationally' known melodies often suggests the end result of oral transmission of the melody alone (see
188
//. Chant Genres
Exx. 11.22.4-5 above). Here perhaps are some of the few, old melodies which circulated in different forms in different areas, before being texted. To provide a text would usually be to fix the shape of a melody in one of its different forms. Further texts composed in a particular area would then normally copy the familiar arrangement of verses and syllable-count in each verse.
Since sequence melodies often sound very un-Gregorian, explanations of their origin have often involved non-Gregorian music. Stablein believed that some at least incorporated a strong secular element, particularly E-mode melodies (most are in G or D) like 'Planctus cigni'. Parallels have naturally been drawn between the sequences and secular forms employing the double-versicle scheme: the lai and the estampie (see Handschin 1929; Spanke 1938; and Stablein 1962, 'Schwanenklage'), but lack of contemporary musical evidence about the secular forms remains a barrier. Byzantine influence is not now thought to have been a motivating factor, for although Byzantine pieces with occasional textual parallelism exist, their music is not syllabic but highly ornate. Stablein ('Sequenz', MGG, 531) also pointed to examples of Irish sacred texts with textual parallelism, but no music for them survives.
We do not know what alleluias in the Roman chant repertory sounded like in the eighth century. If some already had new music (called melodiae) for the repeat of the alleluia after the verse, as they do in the earliest written sources of the eleventh century, this might have stimulated the Franks to compose similar melismas.
The Milanese alleluias constitute a comparable case, for by the twelfth century, when they were written down for the first time, melodiae of astonishing length were being sung, displaying intricate repeat structures (transcription and structural analysis by Bailey 1983, Alleluias). For some alleluias there are two sets of melodiae, that is, music for two extended repeats after the verse, and some even have melodiae tertiae. Here we cannot say if outside influences were at work in Milan, or exactly how old the melodiae are. But their history might well run parallel to the Frankish sequence.
The alleluias (laudes) of the Mozarabic rite also included lengthy melismas with repeat structures, sung with the repeat of the alleluia after the verse (reproduced and discussed by Brou 1951, 'Alleluia'). Their age is uncertain, for the earliest surviving sources date probably from the eleventh century.
Was anything sung in the Gallican liturgy (used in Gaul before the Frankish imposition of Roman use) which could have been taken up in the sequence? Since no Gallican liturgical book with music survives, the question cannot ultimately be resolved. The chief source of information about the liturgy is the so-called Expositio antiquae liturgiae gallicanae (ed. Ratcliff 1971), formerly attributed to Germanus of Paris, now thought to have been written in Burgundy in the early eighth century. It describes a threefold alleluia (called laudes): 'habet ipsa Alleluia primam et secundam et tertiam', which immediately suggests the form of the Milanese alleluias. It is not clear, however, whether these Gallican 'alleluias' are long melodies or merely brief acclamations. Among the Milanese melodiae is the chant called francigenae (ed. Bailey, 127). This has been taken as an example of the Gallican ('Frankish') alleluia.
22. Sequences 189
But Bailey shows (a) that it is related melodically to another Milanese alleluia, and (b) it may have been composed no earlier than c. 1000. The possibility nevertheless remains that in all four churches, Mozarabic, Roman, Ambrosian and Gallican, the alleluia might already in the eighth century have been sung with extended melismas for the repeats of the alleluia after the verse.
None of these hypotheses about origins should distract our attention from the sequence as we have them in their ninth-century forms, remarkable monuments to the inventiveness and inspiration of their composers.
(vii) Rhymed Sequences
At some time in the eleventh century a modest little Easter sequence was composed in eastern France or western Germany which was to enjoy longer popularity than any of the pieces discussed so far. Victimae pas chair laudes is usually attributed to Wipo (d. c.1050), priest and chaplain to the emperors Conrad II and Henry III (see Szoverffy 1964-5, i. 372-4). It is only four verses long (3a is omitted in modern books because of its anti-Jewish sentiment). Its popularity may be due to the homely yet dramatic touch of including a question to Mary (usually interpreted as Mary Magdalene) and her answer in verse 2. Verses 2 and 3 are remarkable because of their internal- and end-rhyme, which links not the two half-verses but the phrases within the half-verse. It is one of the earliest sequences to exploit rhyme (Ex. 11.22.9).
Again presumably because of verse 2, it was taken up into the liturgical drama, and achieved a place within the cycle of Easter sequences in most uses of late medieval
Ex. II.22.9. Sequence Victime paschali laudes (Cambrai, Bibl. Mun. 61, fo. 174r)
pr—m—_^ #— - —•—9-*- —•—w-
Vic-time pascha-li lau-des immolent Christi-a-ni.
9 9
-------9—-9    -     9    ^    '-~     • 9
Agnus re-demit o-ues Christus in-nocens pa-tri recon-ci - Ii - a-uit pecca-to-res. Mors et ui-ta du-el-lo      confli-xe-re miran-do dux ui-te mortu - is regnat ui-uus.
0***9
Die nobis Ma-ri-a     quid ui - di-sti in ui - a     Sepulchrum Christi uiuentis An-ge-li-cos testes      su-da-ri-um et uestes     Surrexit Christus spes nostra ---*>'■•> —
et glo-ri-am ui - di re-sur-gentis. pre-cedit su-os in Ga-il-le-am.
-,---*-9
Credendum est magis so-li Ma-ri-e ue-ra-ci quam Iu-de-o-rum tur-be fal-la-ci. Scimus Christum surrexisse        a mortu-is  ue-re      tu nobis uictor rex mi-se-re-re.
190
//. Chant Genres
Europe. It proved popular enough to withstand the post-Tridentine pruning exercise in the sixteenth century and is one of only five sequences found in modern Vatican books.
Regular rhythm is not a feature of Victimae paschali, but can be found in some sequences composed shortly afterwards. Congandentes exultemus (Ex. 11.22.10) is found in manuscripts from about 1100 onwards and may have been composed in response to the bringing of the relics of St Nicholas of Myra to Bari in 1087. Most of the twelve verses are in regular accentual verse, but verses 4 and 5 are not. In verse 6 the lines are trochaic with 8 + 84-7 syllables each; the rhyme scheme for each double verse is AABCCB, which became the most popular scheme of all in subsequent centuries. Like many earlier sequences, Congandentes moves up to a higher register at verse 7, cadencing thenceforth always on a instead of D. The 'Gallican' cadence is likewise another feature common to many of both the older and the newer sequences.
Hodierne lux diei (Ex. 11.22.11) is one of the earliest completely regular sequences, probably written in the first half of the twelfth century. Since all verses are of the same length and metrical/rhythmic pattern, the effect is that of a ten-strophe hymn with five different melodies. It too moves gradually higher from its plagal opening verse, to a in verse 2, d in verse 3. Verse 4 relaxes a little before the climactic verse 5, two of whose phrases start on high d. The first phrase of verse 5 is the same as the second of verse 3, and the cadence of that phrase is already present at the end of the first phrase of verse 3. The cadence of verse 1, phrase 2 reappears in verse 5, phrase 2. Such melodic cross-references and echoes are not, of course, absent from the older sequence. But now that verses were composed of regular short phrases, the possibilities for exploiting them were greater. All lines end FEDCD. Another new feature, in comparison with the old sequence, is the frequent occurrence of two or more notes per syllable.
A large number of rhyming sequences have changes of metre. This might be seen both as the counterpart of the constantly changing line-lengths of the older sequence, and as a display of technical accomplishment. There are indeed several dazzling examples of metrical variety. They can be found in several of the oldest rhyming sequences, such as Stola iocunditatis (which still has one non-accentual verse), Mane prima sabbati (for Mary Magdalene), and Laudes cruris at toll am us (for the finding of the True Cross, perhaps written for the reception of a fragment of the True Cross at Notre-Dame, Paris, in 1109: see Husmann, 'Notre-Dame-Epoche', MGG, 1706; also Weisbein 1947). The melody of Laudes crucis became the most popular of all those for rhymed sequences, being used by Thomas Aquinas for the Corpus Christi sequence Lauda Sion salvatorem, also found in modern Vatican books. A vast number of contrafacta are known. Most verses proceed in the 8 + 8 + 7 scheme, but verse 3 is as follows:
3a 8 + 8 + 7 + 4 + 3+3+4 + 3+3+7 36   8 + 8 + 7+    7 +     7 +7
22. Sequences 191
Verses 9 and 10 are 8 + 8 + 8 + 7, verse 11 is 8 + 8 + 8 + 8 + 7. Not all contrafacta follow this plan exactly.
Paranymphus salutat virginem (Ex. 11.22.12) also has a 4 + 3 + 3 line, right at the beginning. (The scheme was very popular in the rhymed office.) Thereafter hardly two verses have the same scheme. The habit of ending a verse with a four-syllable word is best known from the Christmas sequence Letahundus. The inclusion
Ex. 11.22.10. Sequence Congaudentes exultemus (Madrid, Bibl. Nac. 19421, fol. 81v)
^  t 'o ,  .   *      *j '   0   t «9 ^ . ,
Congauden-tes ex-ul-temus uo-ca - li concor-di-a. Ad be-a - ti Nycho-la-i   uo-ti-ua sollem-ni-a.
*   0 «3 «
*  *   »   *   «f ^ * ^
Qui in cu-nis adhuc iacens seruando ie-iu-ni-a. A  pa-pil-lis  cepit summa prome-re-ri gau-di - a.
+ $ ***   » * •# • * * »«^» #
A-do-lescens amplexatur lit-te-rarum stu-di-a. A-li-e-nus et immunis ab om-ni las-ci-ui-a.
- —»2—r-_—•—'-•—0-•—s——-
'   *   ^   ' '    '    0 *   & ~  • *
Fe-lix confessor cuius fu - it dig-ni - ta-tis uox de ce-lo nun-ci-a.
Per quam prouectus presulatus sub-li-matur  ad summa fasti-gi - a.
-1-»-«-«—s—0—
E-rat in e-ius a - nimo pi - e-tas e - xi-mi-a Au-ro per e-um uir - ginum tol-li-tur in-fa-mi-a
et op-pressis impendebat mul-ta be-ne-fi-ci-a. atque patris  e - a - rundem le-uatur in-o-pi-a.
Quidam naute na-ui - gan - tes et contra fluctu-um se-ui-ci-am luctantes lam de ui-ta de-spe-ran - tes_in tan-ti po-si-ti   pe-ri-cu-lo clamantes
na-ui pe - ne disso-lu-ta. uo-ce dicunt omnes u-na.
-T**-*    0     - 9     0*00            . ».
» •   0 • jiZi,    * * * * 0 *     * ** 0
0  be - a - te Nycho-la-e nos ad portum maris trahe de mortis an-gu-sti-a.
Tra-he nos ad portum maris tu qui tot au-xi - li - a-ris pi - e - tatis gra-ti-a.
*    ,    0j 0    * 0    '    0   *0H   . ,     *    0    *H    0 ZE=MZ
192
//. Chant Genres
(Ex. 11.22.10 rout.)
Cum clamarent nec incassum   ec-ce quidam dicens adsum     ad uestra pre-si - di - a. Statim au-ra da-tur grata     et tempes,l;a| fit^ se t da-ta     qui-e-uerunt ma-ri - a.
- M   .   *  «7«  . =^ff
y -     .   <> '
Ex ip - si - us tumba ma-nat un-cti-o-nis co-pi-a. Que in-firmos omnes sa-nat per e-ius suffra-gi-a. %_ "
•n ,   *—r
Nos qui sumus in hoc mundo ui - ti - o - rum in profundo iam passi nau-fra-gi - a. Glo-ri - o-se Nycho-la - e      ad sa-lu-tis portum trahe       u-bi pax et glo- ri - a.
Ipsam nobis un-cti - o-nem im-petres a do - mi-no prece pi - a. Qua sa-nauit  le - si - o-nem   mul-torum pecca-minum in ma-ri - a.
4-*—V-—
Cu -ius festum ce-le-brantes gaudeant per se-cu-la.
Et co-ro-net e-ius Christus post uite curri - cu-la al-le-lu-ia.
Ex. 11.22.11. Sequence llodierne lux diet (Assisi, Bibl. Com. 695, fo. 51v)
*»—0  ^ »—r
Ho-di - er - ne lux di - e - i Decan-temus in hac di-e
-*—r
ce-le-bris in matris De-i semper uir-gi-nis Ma-ri - e
a-gitur me-mo-ri-a. laudes et pre-co-ni-a.
t   • §2
—*- -"-9~ _
Omnis ho-mo omnis ho-ra Psalle psalle ui-su to-to
ipsum o-ra et implo-ra cordis   o-ris uo-ce uo-to
e-ius pa-tro-ci-ni - a. a-ue ple-na gra-ci-a.
*2
3
pa-rens paris nesci - a. ge-ni-torem fi-li-a.
A-ue re - gi - na ce - lorum Fecun-da-ta si-ne ui-ro
in-ex-perta ui - ri thorum ge-nu-is-ti mo-do mi-ro
Florens ortus austro flante Fu-sa ce-li ro-retellus
porta clausa post et an-te fu-sum Gede - o-nis uellus
ui-a ui-tis in ui-a.
de-i - ta-tis plu -ui-a. -»3---
7f **~
Amen.
Salue splendor firmamenti Pla-ca ma-re maris Stella
tu ca - n - gi - no - se menu ne in-uoluat nos procel-la
de-super ir-ra-di-a. et tempestas ob-ui-a.
22. Sequences
193
of the markedly shorter verse 2 may derive from the earlier unrhymed sequence. A gradual rise in pitch is again noticeable. The final verse is strongly reminiscent of Hodieme lux diei. Much work remains to be done in tracing such interrelationships across the repertory.
Ex. 11.22.12. Sequence Paranymphus salutat virginem (Assisi, Bibl. Com. 695, fo. 75r) Pa-ranymphus     sa-lutat uir-gi-nem     no-ui partus      as-signans ordinem.
En inquit con-ci-pi-es par-uulumque pa-ri-es nunc pudo-ris sen-ti-es le-si-onem. lam preuen-ta gra-ti-a      sed de mo-do du-bi - a     que-rit re-i nes-ci-a ra-ti-onem.
O Ma-ri-a ne for-mides pre-be fi-dem qui-a fi-des potens in hoc o-pe-re. O Ma-ri-a  sis se-cu-ra     nu-tu De-i   pa-ri-tu-ra      si-neui-ri fe-de-re.
♦   « *
Verbum carni iun-gi-tur uir-gi-nis in u-te-ro.
Nec na-tu-ra tol - li - tur u-ni-us ab al-te-ro.
* ■        ...     • * * • * .
* *. • *; * • * * *# •*,,•*                 - •
O        felix  no-uitas O     mi - ra dig-na-ti-o      contracta de-i-tas
la-cet in pre-se-pi-o.
^5
O      pu-er sa-piens        O     uerbum ua-giens      O ma-iestas hu-mi-lis
nos iu-ua nos re-ge    nos uer-bo prote - ge      nobis car-ne si-mi-lis.
-fl-1-fcrS-	—-.-—		
A—-—*—«i a « f« ' i—	—-—*—<—	-~M-~s-	-_-—--------
fff)-* **-*=-■—			*   *         , —   #   * >-
O Ma-ri-a mater De-i spe respu-it in te re-i tu post Deum nostre spe-i Ihesu  pi - e Ihe-su fortis      Ihe-su nostre dux cohor-tis     fac nos es-se tu - e sortis
I
3E
-*—• • •
fi-du-ci-a.
-»—•—r
-*—r
in glo - ri-a  tu-ematrts gra-ti-a.
A-men.
The advent of the rhymed sequence is contemporaneous with the composition of many other genres in rhymed, accentual verse: Latin Benedicamus songs and
194
//. Chant Genres
conductus (versus), for example, and religious dramas such as the 'Sponsus' play. Many different French centres will have played a part. One may speculate how influential were individual composers such as Peter Abelard (1072-1142: see Waddell 1986). The new type of sequence made particularly deep inroads into the liturgical cycle in Paris in the twelfth and thirteenth centuries: no other sequence repertories of the later Middle Ages have as high a porportion of rhymed sequences as the Parisian ones. A key part in the development seems to have been played by Adam of Saint-Victor. Two phases of activity have been distinguished by Husmann (1964) and Fassler (1984). A new repertory of rhymed sequences was created in Paris (probably on the basis of a few already existing pieces), perhaps as early as the first half of the twelfth century, though the earliest surviving sources are of the next century. Many of the texts contain echoes of the theological writings of Hugh and Richard of Saint-Victor, which argues for a contribution by a personage connected with Saint-Victor, the Parisian Abbey of Augustinian canons regular. That they were sung in all the churches of Paris must mean, however, that they were in the liturgy of the cathedral of Notre-Dame. The repertory then became known outside Paris and mingled with other local uses (see Husmann 1964, 191 ff.).
The liturgical assignment of these sequences is not always identical at Saint-Victor and Notre-Dame, and a number of texts are found only in Saint-Victor manuscripts, presumably composed there. Furthermore, there appears to have been a systematic revision of the melodies in use at Saint-Victor. That is, the ones we find in notated books of the abbey differ from those of Notre-Dame and the rest of the city. Sometimes they can be shown to rework the older melody (see Fassler 1984, 252-4). A large number of the ostensibly new melodies are based on material in Laudes crncis attollanuis. This is the case not only in sequences which begin like Laudes crucis (see Apel 1958, 463), but in others where only internal phrases are reused (Fassler 1984, 258-60; incipits are therefore insufficient for checking interrelationships between repertories). Fassler believes that the resulting families of sequences were created in order to point up connections between ideas in the texts (Fassler, 1983). What makes this activity possible is of course the construction of the melodies in units of identical patterns, which, within limits, could be transferred from one text to another. (Misset and Aubry 1900 edit the earliest complete Saint-Victor source with music. Their melodic analyses do not distinguish between Victorine and Parisian melodies, but can be used with a list of specifically Victorine melodies to hand: see Husmann, IK), Fassler, 1984, 245).
As to the identity of Adam, Fassler connects him plausibly with the first period of sequence composition in Paris, and identifies him with the Adam who was precentor of Notre-Dame at least from 1107, who held a prebend at Saint-Victor from 1133. The problem remains of distinguishing sequences which originated in Paris (including Adam's) from others in similar style composed elsewhere. For another example of distinguished work in a similar vein by a contemporary of Adam one may point to the compositions of Nicolas of Clairvaux (see Benton 1962).
Rhymed sequences were composed in enormous numbers throughout the Middle
22. Sequences 195
Ages. The vast majority are quite unknown. (Labhardt 1959-63 is a relatively isolated study.) Selecting typical examples is inevitably arbitrary, but one final sequence may demonstrate another of the multifarious formal schemes to be found. Affluens dehcns (Ex. 11.22.13) appears to be a local composition. It has only three double-verses, but each  is composed of an unusually large number of phrases,  somewhat like
Ex. 11.22.13. Sequence Affluens deliciis (Munich, Univ.-Bibl. 2° 156, fo. 194r)
§
Af-fluens de-li-ci-is Ho-di-e cu-bi-culum
Dauid re-gis fi - li - a re-gis Hester susci-pit
K      —i--'    ■ ^
spon-si fertur brachi-is       ad ce-li se-di-li-a se da-re pe-ri-culum    quo hos-ti-lis ef-fi-cit
et a - mi - ca prope-rat sponsum quo ab-i - e-rat Aman instans fraudibus    pecca-ti ru-den-tibus
querens inter ly - li - a. mortem mundo conticit.
m
Per ce-li pa-la - ci - a Vox Rachelis in Ra-ma
cuncta transit hos-ti - a in-tra regis a-tri-a hie au-di-tur sed dragma   dul-ceti-bi ca-ni-tur
u-bi sceptrum aureum Christum os uir-gi-ne-um u-bi  te amulecti-tur     sponsus et al-lo-quitur
^  *   *   '   *   • +
os-cu-la-tur ho-di-e quo be-a - ta fru-e-ris
3
ut sit pax ec-cle-si-e. plus precunctis su-peris.
m
i' a ft *
----«-0-«-»-#-■#-Tf<---«-1-f-
-
Te transmittit hodi - e ..... tellus  ce-li  cu-ri-e ..... Dauid re - gis Thecu - i - tern
Ut fu-ga-ti re-uo-cemur    et prostra-ti sus-ci-temur     ad   e-ter-na gaudi - a .....
He-ly-se-i Su-na-mi-tem. u-bi es in glo- ri - a......
contemporary French lais (but without the first- and second-time endings of the lai; lais also have far more verses.) Melodically the piece betrays its origin by the constant use of the cadence figure EFEDE, which we have seen in an alleluia from this part of Europe (Ex. II. 14.7) and which is also to be found in numerous ordinary-of-mass chants (see for example the group given by Stablein, 'Agnus Dei', MGG, cols. 150 ff.), and other genres, as well as secular songs.
196
77. Chant Genres
11.23. TROPES
Introduction
Added Melismas in Introits Added Melismas in Glorias Added Rcsponsory Melismas Prosulas for Offertories and Alleluias Responsory Prosulas Other Prosulas
Kynes with Latin text, Kyrie Prosulas, and Kyrie Tropes Benedicamus Chants with Extended Text, Prosulas, and Tropes Introit, Offertory, and Communion Tropes; Sequence Tropes Gloria, Sanctus, and Agnus Tropes Farsed lessons, Credos, and Paternoster
Gautier, 1886; Paul Evans 1961; Crocker 1966; Paul Evans 1970, Early; Stablein, 'Tropus', MGG; Steiner, 'Trope', SG.
Facsimiles: Vecchi 1955; PalMus 15, 18
Text editions: AH 47, 49; CT; Planchart 1977
Music editions: Paul Evans 1970, Early; MM MA 3; Boe 1989 and 1990 (Gloria).
(i) Introduction
The great vitality and variety of trope composition in the Middle Ages creates considerable problems of definition and organization. I describe below three types of composition, all of which are essentially additions of one sort or another to preexisting chant:
(a) the addition of a musical phrase, a melisma, without additional text;
(h) the addition of a text, a prosula, without additional music;
(c) the addition of a new verse of chant, comprising both text and music.
(The Corpus Troporum research team in Stockholm has used the three terms (a) 'nieloform', (b) 'melogene', and (c) 'logogene', respectively, to signify these three types of trope. See Marcusson in CT 2, 7 and Jonsson 1978, 102; also Huglo 1978, 7.)
Odelman (1975), in a survey of the rubrics in the sources themselves, showed how the term 'trope' (in its various Latin forms) and others such as versus, laudes, prosa, prosula, verba, were applied to various types of chant for the mass (not however, for office chants such as responsories). Medieval usage was not uniform.
Whatever the type of trope, they were practically all composed to embellish chants for the great feast-days of the church year. Many reasons for the medieval interest in these pieces have been discussed (see for example Gy 1983), but first and foremost would seem to have been the simple desire to make more splendid and solemn the performance of the liturgy (particularly mass) on the most important days of the year. Whatever the significance of tropes as didactic, theological, poetic material, and so
23. Tropes 197
forth, their prime effect on the liturgy is to make it longer (performance of a troped introit can last over ten minutes), to give richer opportunities for solo singers, often in alternation with the choir (most tropes are for soloists, and the genres troped are mostly choral chants), and generally heighten the rich complexity of the ceremonial (although we know rather little about the detail of the ritual performance of troped chants).
(ii) Added Melismas in Introits
Melodic extensions of introit phrases have been partially catalogued, edited, and discussed by Huglo (1978). They are restricted to two groups of manuscripts: on the one hand an Aquitanian group dating from the tenth to the twelfth century, and on the other a German-Swiss group of the tenth and eleventh centuries. (For modern editions of some of the Aquitanian ones, see MMMA 3, supplement, pp. 22, 27, and 35; Evans 1961, 127; Sevestre in CT 1, 287; Sevestre 1980, 34. See also Weakland 1958).
There seems to have been only one melisma or set of melismas for any one introit in the Aquitanian sources. Not only might the several phrases of the introit antiphon be extended thus, but so too might the psalm verse(s) and doxology (at the final 'Amen'), where they are usually rather modest in scope. Texts have not been found for these additional melismas. The case is quite otherwise, however, with the German-Swiss sources. Furthermore, for several introits these sources contain more than one set of melismas (none coincides with an Aquitanian one).
The provision of extra melismas, and texts for them, in St Gall, Stiftsbibliothek 381 for the Easter introit Resurrexi exemplifies the situation. St Gall has no less than ten different sets of melismas. The largest number of phrases in a set is ten: that is, in one set of melismas there are melodic extensions for 'Resurrexi', 'alleluia', 'me', 'alleluia', 'est', 'tua', 'alleluia', 'alleluia' (in the introit antiphon), 'meam' (end of the psalm verse) and 'Amen' (end of the doxology). While that particular set of melismas has no texts, several others have a text for every melisma in the set, where the music has been treated according to the well-known principle of one syllable per note, familiar from the sequence repertory and the standard method of procedure in prosulas. Practically all these melismas and their texts fell out of use before being recorded in pitch-notation. The set for the St John the Evangelist introit In medio ecclesie is a rare exception (Ex. 11.23.1). Is it chance that the verse milibus argenti is a hexameter, that the right number of notes was available? or were text and music conceived together, as in the introductory verse Dilectus isle domini, which is in the form of an 'Ambrosian' hymn strophe? (For facsimiles of the German-Swiss melismas and texts see Gautier 1886; Weakland 1958; CT 1, pis. II-V; CT 3, pis. I-II; transcriptions in Handschin 1952, 166-7 and Stablein, 'Tropus', MGG, Ex. 1 and Abb. 1 and 2; also Stablein 1975, pi. 59 on 183.)
No other sources make so much of this type of introit trope as these early German-Swiss sources (the same is true of similar melismas for the Gloria, discussed next),
198
II. Chant Genres
Ex. II.23.1. Trop e verses Dilectus iste Donům, etc. (St Gall, Stiftsbibl. 484, p. 36; Pistoia, Bibl. Cap. C. 121, fo. 24r) for introit In medio ecclesie (Einsiedeln, Stiftsbibl. 121, p. 39; Graz, Univ.-Bibl. 807, fo. 18v)
Tro.
/   /  n  J /
/  p   n   /   j   / a
.  *   í'  *#  í» *
•9 9
Di-lectus  is - te do-mi-ni
% . ■   <h f. ja      & ~
Io - han-nes  est   a - pos-to - lus i*    F   /    J    ň   J     ň .?
ES
Int.
Tro.
Int.
Tro.
Int.
Int.
Int.
Ps.
Tro.
scrip-tis cu-ius est  mo-ni-tis pol-let   de-cus   ec-cle-si - ae.
r i
šmm     r*    *ěě   »    0    *i **a
f<i    99*    *     •     9     *** =
IN   ME-DI-O   EC-CLE-SI-E       A - PE - RU-IT OS    E - IUS. -//.    /   //   .    /   .    /    /   -    P -   .    /   -,/■     / /■
m
Os tu-um in-qui-ens a - pe-ri meque ip-sum il-lud pro certo sci-as im-ple-re.
ř ř
./vT
JI   A J)
,»* »I« £** s} í*r
ET IM - PLE - UIT   E   - UMDO-MI-NUS.
-#-H-
Qui om-ni-a que dl - xi potens est im-plere.
* a*a~f
SPI-RI-TU   SA - PI - EN-TI - E    ET   IN - TEL - LECTUS.
	/ /»	/   /• -			-|
					
/L-a-^2-				-a"\	_ -«yi -7~sr
<ř9—s—•—*—9-•—*—w •—a—a—a—*g a		a  m~\ ě	•o a		* *m* mm a a •# «a a*a
	6	o			
In-tel-lectum inquit dabo ti - bi et pneuma meum ponam super te. //m   /    ^      A    fl,    //    /~~   // X
- "__^     m__■_se_«~r-■-
STOLAM GLORI - AE    IN-DU-IT     E - UM.
/
a  # £j
BONŮM EST CONFITERI DO-MI-NO     ET   PSALLE-RE NOMI-NI   TU-O AL-TISS1-ME.
_______ _ - a    »-g a
—;—9—--------•- -— jr
Mi-li-bus ar-gen-ti   me-li-or  laus  is-ta uel au - ri
0 a^ *' '**a. ^
23. Tropes
199
(Ex. 11.23.1 com.
Dox.
U    'm V
a  m   a   * « *—*-
m   m  a -r
GLORI-A PATRI ET FILIO ET SPIRITU-I SANCTO     SICUT ERAT IN PRINCIPIO
-rr
ET NUNC ET SEMPER     ET IN SECULA SECULORUM. A-MEN.
which may mean that it originated at St Gall or a related centre. The textless melismas have been associated with the report of Ekkehard IV of St Gall (Casus St Galli, ch. 46: MGH, Sciiptores rerum Sangallensium, 101) that Tuotilo (a monk at St Gall in the late ninth and early tenth century) composed trope melodies for the rotta (probably a type of psaltery).
Weiss noticed that in a very few cases additional trope verses were also extended by melismas; that is another way of 'troping a trope'. (See Weiss 1964, 'Tropierte'; also MM MA 3, no. 10, p. 13-—the central melisma is not present in all sources—and the commentary to no. 153, p. 403.)
The principle of adding festal melodic extensions to traditional chants seems to recur throughout the history of plainchant. Weakland's and Huglo's belief that introit melismas were the oldest parts of the trope repertory cannot be substantiated, nor Stablein's idea that the melody of Ex. 11.23.1 above contains echoes of secular minstrelsy.
(iii) Added Melismas in Glorias
Other examples of such melismas may be found in the Gloria repertory, both in German-Swiss manuscripts and in French ones. In the German-Swiss sources, this type of Gloria trope predominates, indeed Ronnau (1967, Tropen, 197) stated that all the Gloria tropes he believed to have originated in St Gall were of this type; whereas for French ones added verses of text and music together are much more common. As in the case of the introit melismas, German-Swiss sources sometimes have texts for their Gloria melismas. The question always arises with this sort of double troping: is it certain that the melodic tropes preceded the provision of a text (or prosula, with one syllable per note)? could not both have been conceived together? Several of the St Gall sets of melismas have no texts, which suggests that texts were indeed additions.
The three French Gloria tropes of the type where both melisma and prosula exist have been discussed by Ronnau (1967, Tropen, 188-96, with transcriptions and facsimiles).
Another melisma connected with the Gloria trope repertory is of a different type. Within the trope verse Regnum tuutn solidum permanebit in eternum several sources have a long melisma on the syllable 'per-'. Many manuscripts also have a text for this melisma, or more than one text. (For discussion, see Ronnau 1967, Tropen, 179-87
200
//. Chant Genres
and Rönnau 1967, 'Regnum'.) These 'Regnum prosulas', as they are often called in modern literature, although traceable back to the earliest sources (Paris, Bibliotheque Nationale, lat. 1240), may safely be said to post-date the melisma to which they are fitted, since the melisma is actually borrowed from a sequence melody, the one called 'Ostende (maior)' in early French and English sources (text Salus eterna, etc.) and 'Aurea' in German-Swiss ones (text Clare sanctorum, by Notker Balbulus). Rönnau adduced several reasons for supposing that the Regnum verse and its melisma were very old (ninth century): association with Gloria 'A', Bosse no. 39; association with the trope set Laus tua; the fact that both early 'western' and early 'eastern' sources know both these chants and the Regnum verse. (For examples, see below, section xi.)
The Regnum melisma is heard climactically in one of the last trope verses of a Gloria, a final flourish not dissimilar in its effect to the melismas which so often crown the performance of a festal responsory. Like responsory melismas, the Regnum and other climax melismas were frequently texted, and are thus mentioned below in the sections dealing with prosulas.
(iv) Added Responsory Melismas
Holman 1961, 1963; Steiner 1973; Kelly 1974, 1988.
At least one important melisma can be dated securely to the early ninth century, on the report of Amalarius of Metz, writing c.840. He stated that:
In the last responsory [of matins of St John the Evangelist], that is In medio ecclesiae, in contrast to the practice for the other responsories, a neuma triplex is sung, and its verse and doxology are also prolonged by neumas, in contrast with usual custom . . . Therefore because the responsory is sung three times by the succentors [before and after the verse and after the doxology] the neuma is threefold . . . Furthermore, modern singers sing this neuma in the responsory Descendit de caelis [for Matins of Christmas Day] at the word 'fabricae mundi', with which word the neuma may very aptly harmonize.
(Ed. Hanssens, iii. 54-6; quoted in Hofmann-Brandt 1971, i. 12-13, Steiner 1979, 250; see also Handschin 1952, 142-5; Stäblein, 'Tropus', MGG, 811-12; Steiner 1970; Kelly 1988). In the sources that have come down to us, the melismas are usually found assigned to the responsory Descendit de cells, rather rarely to In medio ecciesie, and very occasionally to other responsories. The three melismas are usually arranged in an order of increasing length.
This is but the earliest traceable instance of a responsory melisma, or group of them, which (a) appears to be an addition to the original state of the responsory, for it is not found in all sources (though it could be argued that this is the result of suppression of an original practice because of its extravagance), and (b) is susceptible to transference from one responsory to another. In fact, such a triple set of melismas is very rare, and for parallels one has to look to the Ambrosian repertory, with its multiple responsory and alleluia melismas. By far the most common provision is a single melisma, usually on a syllable near the end of the respond, to be performed only
23. Tropes 201
during the final singing of the respond after the doxology. The responsory chosen is usually the last one of Matins, or the last responsory of an individual nocturn within Matins (the third and sixth in secular cursus, or the fourth and eighth in monastic use).
The extent of the repertory is difficult to assess, since this depends on extensive comparison of sources, in order to detect those melismas present in some sources but not others. (Holman 1961 and 1963 compares Worcester Cathedral, Chapter Library F. 160 with three other sources; see also David Hughes 1972.) That many of them might be considered as 'floating', assignable at will to any responsory of the appropriate mode, is suggested by the collection assembled by Jacobus of Liege as part of his tonary (ed. Bragard in CSM 3 vi. 256 ff.). Manuscripts vary widely in their choice of melismas and melodic details (Steiner 1973 and Kelly 1974 are excellent demonstrations of the instability of the repertory.)
(v) Prosulas for Offertories and Alleluias
Steiner 1969; Bjorkvall and Steiner 1982; Bjorkvall 1990; CT 2, 6.
Prosulas are the texts provided for melismas or predominantly melismatic music. The largest repertories that have come down to us are for alleluias, offertories, and responsories.
The texting of melismatic music, like the adding of melismas, seems to go back at least to the ninth century. One of the earliest of all examples of musical notation, Munich, Bayerische Staatsbibliothek, elm 9543, datable to the ninth century (perhaps even the first half of the century) transmits an alleluia prosula, Psalle modidamina for All. Christus resurgens. (See Smits van Waesberghe 1957, 'Over het onstaan' and 1959, and Stablein 1963, 'Zwei Textierungen'; on the early history of texting melismas, see Stablein 1961, 'Unterlegung'.)
Ex. II.23.2 uses three sources, Madrid, Biblioteca Nacional 288 (from Palermo, r.1100), with pitches derived from Montpellier, Faculte de Medecine H. 159 (Saint-Benigne at Dijon, c.1025), and Paris, Bibliotheque Nationale, lat. 776 (Albi, second half of the eleventh century). The example shows one of the verses (the second verse in each of the three sources used) for the offertory Benedixisti (third Sunday of Advent). On the last syllable of the verse appears a melisma (one of very modest length, chosen for that reason: many are far larger), and in both Madrid 288 and Paris 776 a prosula is also given for this melisma. Although the two texts differ, they are clearly composed in the same way. Both start with the phrase 'da nobis' of the offertory verse, and the prosula of Madrid 288 ends with 'da nobis'. (It could be argued that 'tenebris' is assonant with it.) As many syllables as possible in both texts have been made assonant with '-is'. The syllable groupings reflect the note groups of the music—the note groups are easier to see in the French neumatic notation of Madrid 288, but even in the Aquitanian notation of Paris 776 they are not difficult to observe. Paris 776 reflects the note groups more faithfully at 'in caelis in terris'. Both
202
//. Chant Genres
Ex. 11.23.2. Verse Ostende nobis for offertory Benedixisti, with prosulas Da nobis famulis and £)o nobispotenti (Madrid, Bibl. Nac. 288, fo. 121r; Montpellier, Faculte de Medecinc II. 159, p. 235; Paris, Bibl. Nat. lat. 776, fo. 8r)
Madrid 288
*m    0m~      ft   *    ' "*«  *#.y.       -     «. ft   "  —   ft   *' ^
»*•    «** "      9<\» 0\=iTt
i
r
Os-ten-de   no - bis do-mine mi-se-ri-cor-di-am   tu-am etsa-lu-
Paris 776
J' , > — „      „ -*   », iff.
« 0000
--0-*-"--- 30 »■■
Os-ten-de   no - bis   do-mine mi-se - ri-cor-di - am   tu-am etsa-lu-
Q    ^ ... 'J 0* ft. W & &
1^ —   — ----- -- ------«—•—t*-•-«-0 '»iit
-ta-re   tu- -um da   no - bis.
|j)  '« ■
-ta-re    tu- -um da   no  - bis.
*   • ä
•—#-
PROSA  Da   no - bis fa-mulis   ro-go Christe tu-is glo-ri - o - se pre-mi - a regni quae
w* m*
y
PROSA Da   no  -  bis po-ten-ti in caelis  in terris im-pe-rante uir-tu-te  tu - i quod
• *
promi-sis-ti Sanctis tu - is  da no-bis.
•   - t
—m-
holim nostris reful-sit in te-ne-bris.
texts expand in a modest way the meaning of the parent offertory verse, taken from Psalm 84 (85):
Ps. 84: 2   Shew us, O Lord, thy mercy: and grant us thy salvation.
(Madrid 288)   Give unto us thy servants, I ask Christ, the reward in glory of the kingdom which thou hast promised thy saints, give it unto us.
(Paris 776)   Give unto us, thou who reignest in might in heaven and in earth, your strength, which once shone in our darkness.
Ex. 11.23.3, an alleluia with prosulas, shows how a pre-existing text may be worked into the prosula, becoming completely assimilated into it. The prosula text is printed here with the pre-existing words (or parts of words) in capitals:
23. Tropes
203
Ex. II.23.3. Alleluia Concussum est mare with prosula Angelas Michael (Rome, Bibl. Ang. 123, fos. 139\ 252r; Modena, Bibl. Cap. O. I. 7, fo. 174r)
AL-LE-       -LU- -IA.
i
4 »
An-ge- -lus Mi-chael at-que Gabri-el si - mulque Raphael et omnes conciues -~ s—-
*-.--
pol-lorum si-de-ris ag-mina    nos     con-cedentem in se - cu - la.
CONCUS-   -SUM EST Concussum ac percussum est
MA-
-RE
Mare montes sa-xa et ar-ua
ET CONTRE-     -MUIT          Et contremu-it montes et ex-pauit &^ _jk —____£5_„_,_;_____
TER-
+- £* •
-RA       Draco-nem pes-ti-ferum serpes an-ti-quo
-—» ••~T =
qui   e - iectus est de ce - lo      di-mersus sub terra
,0 *IA
*}—r r
»* JO
U-BI   ARCHANGELUS   MI-CHA-EL   DESCEN-DE-BAT DE   CE- -LO
*—»
In   mons Garga-ni - co uic-to-ri-am Christo
pugnantes cum Sathan duri-us     et     ex-pu-lit   e-um ex-in-de.
204
//. Chant Genres
An-gE-LUs michAel atquc Gabriel simulquc Raphael et omnes concives polorum, sideris agmina, nos concedentem in saecula. concussum ac pcrcussum est mare, montes, saxa et arva, et contremuit mundus et expavit draconem pestiferum, serpcs antiquo qui eicctus est do caclo, dimersus sub terra, (ubi archangelus Michael descendebat) in mons Garganico, victoriam Christo expugnavit cum Sathan durius ct cxpulit eum exinde.
Alleluia. The sea was shattered and the earth was terror-shaken when the archangel Michael came down from heaven. Prosula. Angel Michael and Gabriel and Raphael as well and all the citizens of the heavens, the celestial host, together praise him who reigns for ever. The sea, the mountains, rocks, and land were shattered and battered, and the world was terror-shaken and grew afraid at the plague-bearing dragon, the ancient serpent who was cast out of heaven, engulfed below the earth. [When the archangel Michael came down from heaven] on Mount Gargano a victory for Christ he fought with Satan, more savagely, and drove him out from there. (Cf. Dronke 1985).
These examples show the two most important ways in which text was added to preexisting chants. Most offertory prosulas were provided for melismas in the verses (see Bjorkvall and Steiner 1982 for a useful range of examples). In the absence of a comprehensive survey of the repertory, generalizations would be premature. Some indication of how matters stand may be deduced from the repertory of Paris, Bibliotheque Nationale, lat. 1118, the largest Aquitanian repository of 'tropic' material. Steiner (1969, 371) lists twenty-two offertories with prosulas in Paris 1118, and comments that most of them are for feasts not ornamented by tropes (of the text-plus-music type; nor by sequences). The number of prosulas is also far inferior to the number of tropes.
A few offertory prosulas became popular as compositions independent of their parent offertory. Schlager (1983, 'Tropen') has discussed the most celebrated example, l^etemur gaudiis.
Alleluia prosulas are present in Paris 1118 in slightly fewer numbers (for twenty alleluias). Marcusson's edition of the texts (CT 2) has over 350 prosulas for eighty-six alleluias, from fifty-three sources. As with offertory prosulas, these numbers are far inferior to those of tropes.
The method of performance of offertory and alleluia prosulas is unclear.
Apart from isolated examples in secondary literature (see Schlager, 1967 for an interesting Beneventan alleluia prosula; also Steiner, 'Prosula', AG), offertory and alleluia prosulas have to be studied from manuscript facsimiles (PalMus 13, 15, 18, and Vecchi 1955).
(vi) Responsory Prosulas
Hofmann-Brandt 1971; Kelly 1977.
The texting of the threefold melismas for the Christmas responsories is apparently later than the provision of extra melismas. (The earliest surviving texts are in Paris
23. Tropes 205
1084 and 1118, both Aquitanian sources of the late tenth century; for editions from various sources, see Stäblein, 'Tropus', MGG, Ex. 9, Steiner 1979, Kelly 1988.)
An interesting report places another example in the tenth century. Hofmann-Brandt (1971, i. 17) cites an anonymous eleventh-century chronicler of the bishops of Eichstätt in Bavaria, who says that a distinguished musician called Reginold composed an office for St Nicholas, and, after becoming bishop of Eichstätt (966-91), composed a further office for Eichstätts first bishop, Willibald (d. 787). We are told that Reginold composed the office with long responsories, and placed melismas at the end of them ('in fine notulas apposuit'), and that he then put words beneath the melismas in the manner of a sequence: 'eisdemque notulis versiculos instar sequentiarum subiunxit'. (At least part of Reginold's very remarkable trilingual composition has survived: see Hofmann-Brandt 1971, i. 17-23.)
Apart from its early date, this report is interesting because it suggests that the melismatic 'cadenza' (the analogy is borrowed from Kelly 1974, 461) might have been thought distinct enough from the rest of the responsory to have been 'placed there' specially, like a jewel in the centre of a crown, one might say; and it suggests also that the provision of a text for the melisma might be more or less simultaneous.
Although Hofmann-Brandt was able to catalogue no fewer than 732 responsory prosulas, their dispersal among the surviving manuscripts is very thin, in the sense that few prosulas became widely known, and very few manuscripts have more than about twenty prosulas (the largest collection is that in Barcelona, Biblioteca de Catalunya M. 662, a Catalan antiphoner of the fourteenth to fifteenth century, which has forty; see Angles 1935, 234-9 for several examples; the manuscript has now been studied by Bonastre 1982). Over 180 of Hofmann-Brandt's 496 sources actually contain but a single prosula. Very many are the special feature of an office composed for a local saint and not known elsewhere. Associated with these pious observances, the responsory prosula as a genre survived well beyond the Middle Ages.
The prosula technique was restricted, as far as responsories are concerned, to discrete melismas. Clear repeat structures became increasingly popular as time went on, and it even happened that a freshly composed prosula with repeat structure might supplant a prosula in less popular style. The best-known example of this occurred in the St Nicholas office (see Charles Jones 1963 and Hofmann-Brandt). Here a modest melisma occurs on the word 'Sospes' at the end of the responsory Ex eius tumba. The melisma is an integral part of the responsory as it is transmitted in scores of sources. Very rarely, it appears to have been texted (with the prosula Sospes nunc efficitur— but this prosula is found only in two sources of the eleventh to twelfth century from Saint-Maur-les-Fosses, near Paris, Paris, Bibliotheque Nationale, lat. 12044 and 12584). From the eleventh century onward, however, we have sources for a replacement melisma and text, which was usually performed in place of the modest former melody during the final repeat of the respond (after the doxology; at least, this is its usual place, and the usual place for a prosula). The new text, Sospitati dedit egros, is in regular fifteen-syllable trochaic lines, with a melodic repeat scheme AA BB CC DD (the original 'Sospes' melisma has no repeat structure).
206 //. Chant Genres
Ex. II.23.4. Responsory Beatus Sicholaus with prosula Oportet devota mente (AS, 358)
R. Be-a - tus Ni-cho-la-     -us    iamtrium-pho    po-ti - tus nouit       su-is fa-
-mulus
pre-be - re   ce-le-       -sti-a   com-   -mo-da   qui to - to cor-   -de poscunt ei   - us lar-gi-     -ci-o-  -nem. *I1-    -li ni-mi-       -rum to-ta nos de-     -uo-ti-o - ne
o-por- -tet  co-mit - te  - re.
V. Ut a - pud        Christum e-ius patroci-ni - is   ad-iu-ue-mur sem - per. (Repeat from
*Illi)
PROSA       O-portet   de-uo-ta mente sin-ce-ri-ter.        Presu-li     tan-to commi-te - re. Pec-ca-to-res Christi-a-nos te-na-ci-ter.
t    '       *; *    f #;«»    , . ,
Glo-ri-a patri et fi-li  - o      et spi- ri-tu - i      san - cto.   (Repeat from
*Illi)
The new prosula—to all intents and purposes a newly inserted composition, text and music together—was extremely popular, and became the starting-point for numerous contrafacta (new texts to the same music). Since Sospitati dedit egros has been widely reproduced (for example, Stablein, 'Tropus', MGG, Tafel 89; AS 360) I have chosen to illustrate these contrasts of style with other examples. (See also the sparkling demonstration in Kelly 1977.) Exx. II.23.4 and II.23.5 give two prosulas from Cambridge, University Library Mm. 2.9 (a thirteenth-century Sarum antiphoner; facs. AS). In Ex. II.23.4 the prosula is a simple texting of a modest melisma from a different St Nicholas responsory, Beatus Sicholaus. Although short it has the repeat structure AAB.
In Ex. II.23.5 the composer of the prosula has 'manufactured' a repeat structure out of a previously non-repeating melisma. Each phrase of the melisma is texted twice, and the repetitions are doubled again by the manner of performance indicated in the manuscript, the chorus being instructed to repeat each phrase of the prosula, vocalizing to the vowel The vowel itself is derived from the original melisma, 'suscip-E', and each half verse of the prosula also rhymes with E.
23. Tropes 207 Ex. II.23.5. Responsory O mater nostra with prosula Kterne virgo memone (AS, W)
R.   O ma-ter nos - tra ter sancta quaterque be-a - ta.   *Cum prece de-uo-ta
-- ■» , -»-^"y'^v *........* fy ^ o.-
famulantum susci -pe uo-ta.
V.   lam       Christo        iun - eta sponsoque tu-o   soci - a- -ta. (Repeat from
*Cum prece)
n u •   *   * - S':" -
V * •    *    •    m    m .    •    •   ň _
XT
i
PROSA    E-ter-ne uir-go memo-ri-e     quam si - bi despondit rex glori - e. E.
Vir-gi-nis proles e-gre-gi-e     sponsusque uir-gi-nis eccle-si-e. E.
Tu gaudes ho-di-e de do-no gra - ti - e. E. Et cantas in ce - lo carmen le-ti - ci - e. E.
5=
»  _^  *-* *--»—*—*- *—i        0 +* t
Te laudantes in ter-ra res-pi-ce.      E. At-que nostra clementer susci-pe uota.
~f-
Glo-
-n -a
* r*
pa  - tri
et fi-li
o   et spi-ri-
-tu - i
~** * (Repeat from sancto.  *Cum prece)
Many prosulas are, in effect, small sequences, self-contained musical items which could easily take on a life of their own. One extremely popular piece, Inviolata Integra, which originated as a prosula for the Marian responsory Gaude Maria, was frequently used as a sequence at masses for the Blessed Virgin Mary. Quern ethera et terra, for the Christmas responsory Verburn caro factum est (ed. Irtenkauf 1956, 138), which was particularly popular among German and North Italian musicians, also achieved its independence.
By the later Middle Ages (and perhaps from the beginning) alternatim performance of responsory prosulas between singers performing the text and others singing the melisma was the norm. (See Kelly 1985, 'Melisma'; see Chailley 1949 for a rubric from a Sens manuscript instructing the cantor to dance during the prosula in a processional responsory.)
208
//. Chant Genres
Ex. II.23.6. From Gloria 56 (Vatican IV) with trope verses O gloria sanctontm, etc., Regnurn prosula O rex glorie (Madrid, Bibl. Nac. 19421, fo. 22v)
P
P
QUONI-AM TU SOLUS SANCTUS. TU SOLUS DOMINUS. Ce-les-ti-um terres - tri - um et infer-no-rum rex.
—*—*—i-«—0—0—
TU SOLUS ALTISSIMUS.
P
Regnurn tu-um so- -lidum.
-Hr-
0   '   '   * ^
' _ ^   *   '   "   *-"-"-"   ** +s •—*-
t T
O  rex glo-ri-e qui  es splendor......... ac de-cus eccle-si - e. E.
Qujtm decoras-ti tu - o quoque in cru-cepreci - o - so sangui-ne. E.
'3 i C
^Ž**
Hanc rege semper pi-is-si-me. Qui es fons mise - ri-cordi - e.
E. E.
Per-ma-nebit   in e-ternum.
IE-SU CHRI - STE.
—-a-a-»—
Christe ce-lo-rum.
* *
--»—9—-"-# ^       - -—-> * -«,»•—
CUM SANCTO SPI   -   RI-TU IN GLORI-A DE-I PA - TR1S.     A- -MEN.
P
. .-.-.—a-Í0—•-
j0* • 70-' •
Sal-ue uir-go uir- -gi-num.
'tě íi\**0. r» 0
0 0
-•-—*   »*       * *-
Ma-ri-a uir-go in-troce-de pro nobis ad do - minum. E. Et pi - a pre-ce ro-ga De-um poten-ter per omni - a. E.
c * •
~9--
a     •    0    ďj    0     0     • /h,*0 00 0—
Ut det nobis flo-ri-ge-ra se-de.
			
'*a—:	--	' »_       ~ A a	-#-
			-* < a, y ft * «—
Fru-i semper cum e - o.
Et potens es   in e-ternum.
23. Tropes
209
(vii) Other Prosulas
Although the Regnurn verse, its melisma on 'PER-manebit', and at least some of its prosulas were originally associated with the Gloria known as Gloria 'A' (Bosse 1955, no. 39), it was soon used with other Gloria melodies and with various sets of trope verses. Ex. II.23.6 gives the last section of Gloria 56 (in Bosse 1955; Vatican IV), with the trope set O gloria sanctorum, from Madrid, Biblioteca Nacional 19421 (Catania, Sicily, late twelfth century). This set of trope verses is adaptable to various saints' feasts, by the insertion of the required name: here 'Maria' is entered. The redactor of the manuscript has decided to use the Regnurn verse and melisma with the popular prosula 0 rex glorie. He precedes it with another 'wandering' element, the trope verse Celestium terrestrium, which is likewise found in many other sets of Gloria trope verses (see below, section xi). At this point the melody dips for the first time to low G (G\)—previously the range has been C—c—and the change of tessitura from this point on is quite noticeable (Gi-a). The main cadential note remains E, however, which is why no other pitch is possible for Celestium and Regnurn. (In Gloria 39 there is no hiatus in the pitch when these trope verses are used.) After each phrase of prosula, the melody is repeated as a vocalization on 'E', derived from 'PERmanebit' and matched by the rhyme in the prosula.
In the manuscript, this Gloria with Regnurn prosula is immediately followed by three more alternative prosulas with text in honour of, respectively, the Blessed Virgin Mary (obviously more apt for the Marian version of the 0 gloria sanctorum set of trope verses), Apostles, and the Blessed Virgin Mary again. Ex. II.23.6 shows only the first of these three alternatives. Interestingly, the principle of assonance has become rather confused. E is maintained for the melismas, though it no longer refers back to the first verse Salve virgo virginum, and the prosula has line-ends in u, a, e, and o.
The great majority of Sanctus prosulas are for the second 'Hosanna', again providing a climax for the composition. Like responsory prosulas they frequently assumed sequence form (see Thannabaur, 'Sanctus', MGG, Exx. 12-13; also Smits van Waesberghe 1962) and could even become independent liturgical items: Trinitas unitas deltas had an interesting career as a sequence, among other things (see Schlager 1983, 'Trinitas'), and Voci vita was used as a sequence at Hereford. (Two Hosanna prosulas form part of Ex. 11.23.16 below.)
Very rare indeed are prosulas where the Sanctus melody itself is retexted: Thannabaur (1967) cites three examples. The type is limited to eastern sources of the fourteenth century and later.
Two chants rarely embellished by any kind of trope were the gradual and tract. Stablein ('Graduale (Gesang)', MGG) gives examples of gradual prosulas, all from relatively late sources. Tract prosulas are found only in Italian manuscripts (see PalMus 15, fos. 75v, 90', 110r, and Vecchi 1955, fo. 74 for examples in facsimile).
Ex. II.23.7. Kyrie Te Chris t e supplices (Laon, Bibl. Mun. 263, fo. 20v)
fr ••...... . . ^
1. Te Christe suppli-ces e-xo-ramus euncti-potens ut nostri dig-neris e - ley - son. 3.   O bone rex super astra qui sedes et do-mi- ne qui euncta gubernas  e - ley - son.
1.  Ky - ri - e- -lei - son.
3.  Ky - ri - e- -lei - son.
W       • •  '  m  * *  •  9 * + • ■ •  ^ • * * •
2.    Te decet laus cum tripudi - o   pa-ter šumme un-de te pe-timus e - leyson.
2.   Ky - ri - e- -leison.
4.   Tu-a de-uo-ta plebs implorat iu-gi-ter  ut il - li   dig-neris e-lei-son...........
6.   O the - os   a-gy-e   saluans ui-ui-fi-ce   redemptor mundi e-......... -ley-son.
4.  Christe- -lei-son...........
6.   Christe- .......... -lei - son.
-»-
5.  Qui canunt ante te       tu pre-cibus an-nu-e       et no-bis semper e-ley-son.
% í** ''••;9 -j*0 *
5.   Christe- -lei-son.
7.   Clamat incessanter nostra con-ci-o   dicens e - leyson.     Ky - ri - e- -lei-son.
*   *   '   .   *    .   .    *h   ,        ' & .
Mi-se-re-re fi-li De-i   ui - ui  no-bis   e - lei - son.     Kyri-e- -lei - son.
3
In   ex-celsis ......... De-o ma-gna sit glo-ri-a   e-ter-no patri        di-camus in-de-si-
qui nos re - demit propri-o sangui-ne ut ui - ui - f i - caret  a mor-te
í
-nenter u - na uo-ce e - ley - son.       Ky - ri - e-
•1 »**m  Kt •£>
-lei - son.
23. Tropes
211
(viii) Kyries with Latin Text, Kyrie Prosulas, and Kyrie Tropes
Crocker 1966; Bjork 1979/80, 1980, 'Kyrie Trope', 1980, 'Early Settings', 1981.
Questions about what is added material and what constitutes the original form of the chant are particularly pressing in the case of many Kyries. Was each Kyrie with Latin text so composed in the beginning, or was the text added as a prosula to the ornate melody? If the earliest sources already have the Latin text, then we have no documentary evidence for supposing the untexted version existed earlier. Crocker and particularly Bjork have supported this reasoning with stylistic arguments.
Ex. II.23.7 is a transcription from Laon, Bibliotheque Municipale 263 (Laon, twelfth century) of a Kyrie with Latin text whose transmission goes back to both early west and east Frankish manuscripts. Although they are no certain test of prosula technique, the syllable- and note-groups do not suggest that this text has been fitted to a pre-existent melody. More important than that the syllable-groups do not follow the note-groups in this particular source (other sources show no better correspondence) is the fact that verses with the same melody (1 = 3, 4=6, 7 + 8=9) do not have the same syllable-groups. Against this could be adduced the e assonance in verses 5 and 6.
The Kyrie with text Kyrie fans bonitatis (Ex. II.23.8) is much more artful in these respects. The correspondence of note- and syllable-groups is quite striking, as also the constant e rhymes. This may simply reflect the taste of the eleventh century, as opposed to that of the ninth. But it possibly reflects also the fact that the text—we could justifiably call it a prosula—appears in the manuscript tradition later than the melody. The earliest sources, from Winchester and Arras, of the late tenth and early eleventh century, do not have the Latin text. (Previously available Variae preces, 165-7, Cantus selecti, 81*-82*, Graduate Sarisburiense, 2*v, together with several other Kyries of this type.)
Quite different in impact are the trope verses sometimes provided for these and other melodies, which were largely unknown until the examples in Bjork's articles appeared (principally Bjork 1980, 'Kyrie Trope', which contains practically all that can be reconstructed). Such trope verses contrast musically with the nine invocations of the main chant. Bjork (1980, 'Kyrie Trope', Table 2) lists some two dozen tropes, which are distributed thinly over early manuscripts from all areas of Europe. In early Swiss-German sources this type of Kyrie is preferred, while French ones contain a much greater proportion of Kyries with Latin text (or prosulas). Many fell out of use before being recorded in staff notation.
While many of these tropes consist of but a single introductory verse, others make up a set of eight verses deployed between the nine invocations of the main chant. A few sets contain three verses, distributed as in Ex. 11.23.9. The set appears here with Kyrie 55 (in Landwehr-Melnicki 1955), also the base melody of Ex. II.23.7. Melodically it both complements it and contrasts with it, sometimes awkwardly: the verse Iterum dicamus rises to d, preparing us for the higher-lying Chnste invocations, but then subsides on to the former low pitches. The last trope verse, Et submissis vultibus, again prepares for the move to higher pitches in the final Kyrie.
212 //. Chant Genres
Ex. II.23.8. Kyrie Funs bonitatis (Laon, Bibl. Mun. 263, fo. 26v)
* *  «   . - t   w   0    t       -   »   ,  « —
—- -#-H-m----
1. Ky-ri-efons bo-ni-ta-tis pa-ter in - ge-ni-te   aquo bo-na cuncta pro-cedunt e - ley - son.
2. Ky-ri-e qui  pa - ti   natum mundi pro cri m i-ne ip-sum ut sal-ua-ret mi-sis-ti e-lei-son.
3. Ky-ri-e qui sep - ti - for-mi  das dona pneumate   a quo celum ter-ra re-pletur e-lei-son.
1.   Ky-ri-e- -lei - son. (also after 2. and 3.
4. Christe   u-ni-ce De - i pat ris ge - ni-te quem de uirgi-ne nas-ci-turum mundo miri - fi - ce
5. Christe   a-gy-e   ce-li compos regi - e     melosglori-e   cu - i Semper adstans pro numine
6. Christe ce-li-tus adsis nostris precibus   pronis mentibus quem in terris deuo-te co-li-mus
sancti pre-di - xerunt prophete e - ley - son. an - ge - lorum decantat a-pex e - ley - son. ad  te pi-e Ihe-su clamantes e - lei  - son.
r*-r-**—v--*
4.   Christe- -lei - son.   (also after S. and 6.
I
* m
y---—,—„-„
7. Ky-ri-e spi - ritus alme co-herens pa-t ri natoque u - ni - us  u - si - e consis-tendo
8. Ki - ri - e qui bapti-zato   in Iorda - nis unda Christo effulgens speci-e colum-bi-na
9. Ky-ri-e  ig-nis diui-ne pecto-ra nostra succende ut digni pa-ri-ter proc!a-ma-re
y
flans ab u-troque e - le
ap-pa-ru-is-ti e - le
pos-simus semper e - le
- son.
- son.
- son.
7.   Ki - ri - e- -lei-son. (also after 8. and 9.)
A number of Italian Kyries with Latin verses are composed with the same melody throughout, that is, for all the acclamations and the alternating Latin verses. The result is not unlike the type of litany mentioned in section II.17.i above. (See Bjork 1980, 'Early Settings' and especially Boe 1989 for examples.) There is a connection here with other genres as cultivated by Italian musicians: sequences with the same melody for all verses have been mentioned above (II.22.v) and other troped ordinary
23. Tropes 213
Ex. II.23.9. Kvrie 55 with trope verses Christe redemptor, etc. (Paris, Bibl. Nat. lat. 903, fo. 166v)
Christe    redemptor mise-re-re   no-bis     kir - ri - e - lei - son
		
		
tff>-*       «*  1.-, —■» g» •—	-•—	-~M-' ♦      ~6m-Alt "
-Vg—**--	—•—*#—m—«-.	»—#*—--*—»f**—*—
ei - a  om-nes di-ci-te    mi-se-re-re do-mi-ne   kir - ri-e - lei - son.
Vo-ce corde proclamantes regem in - ui - si -bi - lem canen - tes   il - li.
^ t t0Sf j3 ^ ,t« u0a ,#a t * ** ***** *tf    ft*      if* *
1.3. KIRRI-E E - LEISON. 2. KIRRIE E - LEISON.
I-te-rum dicamus omnes Christe elei-son et rogemus Christum deum una uoce dicentes.
^   f'?  "r f'' 0S* f &  . .„ 'V*/"*. *
4.6.CHRISTE ELEISON. 5. CHRISTE E - LEISON.
Et submissis uul-ti-bus depreca-mur tri-ni-ta-tem re - gem aeter-num   canen-tes il-li.
*   "** *** »*0   * **l • • *«   ť1-rt^-
* • *        * V
* a—»-
7. KIR-RI-E ELEISON. 8. KIRRIE ELEI-SON.
,««1  ,  ««, ^ >#j w# % % #.j #q ^
f.l63v
*♦ *** *.     ^ ,
9. KIR - RI-E E-LEISON.
of mass chants followed the same pattern. Ex. 11.23.18 below is an Agnus Dei composed in this way.
(ix) Benedicamus Chants with Extended Text, Prosulas, and Tropes
Ark 1970; Huglo 1982, 'Debuts'.
Many early Benedicamus chants have not simply the two phrases of text 'Benedicamus domino. Deo gracias', but longer texts. Of the seventeen examples which Huglo dates in the tenth or eleventh centuries, eleven have extended text. Are these to be regarded
214
//. Chant Genres
as tropes of a pre-existing chant, or prosulas for a melismatic chant, or are they an especially elaborate type of Benedicamus, composed expressly in that way, perhaps for festal liturgies rather than ferias? In many cases we shall probably never know, since they survive as unica (for example, many of those in Paris, Bibliotheque Nationale, lat. 887, discussed by Arlt 1970, Darstellungsband, 161-6).
Some of the compositions can be regarded as prosulas in that the melody is borrowed from elsewhere, and a new text fitted. Several Kyrie melodies were used as Benedicamus chants in this way, at least from the eleventh century onward; so was the 'Flos filius' melody, a melisma from the responsory Stirps lesse. This is found widely, both with the simple Benedicamus text, with 'Benedicamus domino, alleluia, alleluia' (an Eastertide adaptation), and also with longer texts, for example: 'Benedicamus flori orto de stirpe lesse die hodierna, qua processit virga virgo domino' (Arlt 1970, Darstellungsband, 169).
Prosula seems to describe less well the technique of Eta nunc pueri (Arlt, 165), which is not consistently syllabic. Although it may be the texting of a pre-existent melody, it seems less likely that a Benedicamus melody without extended text lies beneath this composition, for then one would expect the word 'Benedicamus' to start the piece (Ex. 11.23.10).
Ex. 11.23.10. Benedicamus chant Eia nunc puen (Paris, Bibl. Nat. lat. 887, fo. 46r)
Ei-      -a     nunc   pu - e - ri   uo-ce precel - sa
		
		
I'll      ■        m *		_ *~ —____._
\Tu •-		
	- - -0---_-	^j*                *            * *   -r* •
be-ne-di-ca-mus regi mag-no o-       -lim na-to do- -mi-no.
De-    -o   le-ti ca-na-mus cuncti graci-a
e - o quod regit as-tra pontum et ar-ua nunc et in   se - cla    al-     -le- -lu - ia.
With Benedicamus . . . Verbi iungando (Arlt, 163) we have what seems to be a self-sufficient Benedicamus phrase, to which a couplet of eight-syllable lines has been added. (The process is repeated for Deo gracias.) If the Benedicamus melody were to be located elsewhere without Verbi iungando, then one might be tempted to speak of a trope. And indeed it is found, with a different continuation, in the same source as for Benedicamus . . . Verbi iungando. The situation is complicated, however, by the fact that this time the new text, Hodie surrexit, is placed first, using the putative
23. Tropes 215
Benedicamus melody, texting and extending it, and then 'Benedicamus domino' is spread over the whole new melody (Ex. 11.23.11).
In fact, this explanation of how the compositions arose seems strained. They, and Ex. 11.23.10, all come from the earliest substantial collection of such Benedicamus chants, Paris 887 (Aquitaine, eleventh century). I would prefer to believe that we are dealing with the result of a local campaign to extend the festal Benedicamus repertory, where several of the pieces have a family likeness not really attributable to a trope or prosula technique. They are newly composed from the start. It is only their prose texts, incorporating the words 'Benedicamus domino', and their predominantly traditional melodic style that tempt us to relate them to the trope repertory.
Ex. 11.23.11. Benedicamus chants (Paris, Bibl. Nat. lat. 887, fos. 46v, 46r) f.46v________
9 + .   ' '
-*- -9- *-«-
Be-ne-di-       -ca-mus   do-   -mi-no.  Verbi iungando carmina   re - ferentes
al-tis-si-mo. f.46r
^ J  *    '        '   '  ' '   »        -   *   »   »   ,   *l _^ , ,
O-di-e sur-rexit  le - o fortis Christus     fi-li-us De-i   omnes u - na
P
uo-        -ce   di - ca-mus al - le- -lu-ia.
V «   **  ''''»'> '*"''•*• +s *\. •''**!* ^r-r
Be-ne-di-ca- -mus do-mi-no   al - le- -lu-
la.
(x) Introit, Offertory, and Communion Tropes; Sequence Tropes
Paul Evans 1970 'Early'; MMMA 3; CT 1, 3; Planchart 1977.
The earliest recorded composition of introit tropes, if the testimony is trustworthy, is by Tuotilo of St Gall (known between 895 and 912), who is said by the St Gall chronicler Ekkehard IV (c.990-1060) to have composed the introductory verses Hodie cantandus est in his youth ('plane iuvenis'—MGH, Poetae I^atini aevi Carolini, iv. 1096; see also MGH, Sctiptores re mm Sangallensium, 80—see Ex. 11.23.14 below). Manuscript sources for them survive only from the tenth century onward (Paris, Bibliotheque Nationale, lat. 1240, Aquitaine, second quarter
216
II. Chant Genres
of the tenth century, is the earliest), but there seems little reason to doubt that they were well known in the ninth. It should nevertheless be noted that almost no introit tropes were known in both early 'western' and early 'eastern' sources, so that it is not possible to postulate a common basic layer on which later diverse collections were built.
Introit tropes were written in great numbers during the tenth and eleventh centuries, offertory and communion tropes in smaller quantities (many churches, including most Italian ones, appear to have ignored offertory and communion tropes almost entirely). The repertory is vast. Taking individual trope verses as units, there are 1,044 for the introit, 250 for the offertory, and 113 for the communion in CT I and III, that is, for the Christmas and Easter parts of the church year. There is a decline in sources from the twelfth century, and very few from the thirteenth century or later. This is critical in some areas, since very few north French and German trope collections with staff notation have survived.
Tropes for the introit, offertory, and communion (and the other chants discussed below in sections (xi) and (xii) ) take the form of extra verses placed before or among the phrases of the main chant. The arrangement in Table 11.23.1 is reasonably typical, for the Easter introit Resurrexi, in the tropers from Winchester, Cambridge, Corpus Christi College 473 and Oxford, Bodleian Library, Bodley 775 (late tenth and early eleventh century respectively) (for full texts see Frere 1894, Winchester and Planchart 1977). The trope incipits are given in lower case, main chant incipits in capitals (cf. St Gall, Stiftsbibliothek 376 in Gautier 1886, 139). Although in the Winchester sources the seventeen verses are laid out in this sytematic way, that is not the case in all sources. Some sources (for example, Paris 1118) have rather the character of an anthology. And the stylistic disparity between the trope verses so neatly arranged by the Winchester redactor shows that logic has been imposed upon what is an unstable repertory. Some of the verses are hexameters: the group for the second singing of the introit beginning Ecce pater, then the trope for the first psalm verse Eregit inferni, and then the group for the second singing of the introit beginning Virgine progenitus.
Trope verses in poetic metre of one kind or another are not at all uncommon. Hexameters dominate the field (in CT I, 40-2 it is stated that over 200 of the 766 verses in the volume are hexameters), and elegiac distichs, pentameters, and even verses like Ambrosian hymn strophes are also to be found. Stotz (1982) discusses examples in sapphics.
Ex. 11.23.12 illustrates some of this variety. It is a transcription of the tropes for the introit In medio ecclesiae, for St John the Evangelist's Day (27 December). A troped version of this introit has already been given above (Ex. 11.23.1), where St Gall and Pistoia manuscripts had an introductory metrical verse followed by several verses which seem to have originated as melismatic extensions of the introit melody. The Winchester collection of verses is quite different. There are almost as many verses as for the Easter introit above, except that the doxology and the second psalm verse have no trope. The first trope is a rhyming pair of trochaic fifteen-syllable lines; verse 5 is a
23. Tropes
217
Table 11.23.1. Tropes for the Tastet nitroit Resurrexi in Winchester sources
	Trope verse	Main chant
(Introit)	Psallite regi magno	RESURREXI
	Dormivi pater	POSUISTI
	Ita pater	MIRABILIS
	Qui abscondisti	ALLELUIA
(Psalm verse)	En ego verus sol	DOMINE
(Introit)	Ecce pater	RESURREXI
	Victor ut ad caelos	POSUISTI
	Quo genus humanuni	MIRABILIS
(Doxology)	Fregit inferni portas	GLORIA PATRI
(Introit)	Virgine progenitus	RESURREXI
	Quern non deservi	POSUISTI
	Ut per me	MIRABILIS
(Psalm verse)	Exsurge gloria mea	INTELLEXISTI
(Introit)	Postquam factus homo	RESURREXI
	In regno superno	POSUISTI
	Laudibus angelorum	MIRABILIS
	Cui canunt angeli	ALLELUIA
hexameter. Each trope or set of trope verses has its own history, part of which can be surmised from the sources known for these compositions (see CT I and Planchart 1977). The situation may be summarized as follows.
Trope verses 1-3 are found deployed in the same way in sources from all over Europe, some of the notable exceptions being the tropers from St Gall, St Emmeram at Regensburg, and Benevento.
Trope verse 4 is unique to England, appearing only in the two Winchester manuscripts and in London, British Library, Cotton Cal. A. xiv (early eleventh century). No transcription into modern notation is therefore possible.
Trope verses 5-7 appear in only a few French and German sources, and also in several Italian ones. The two Italian tropers from the Ravenna area (Padua, Biblioteca Comunale A. 47 and Modena, Biblioteca Capitolare 0.1.7) add two verses to the set, but can be used in conjunction with Paris, Bibliotheque Nationale, lat. 1084 (the only Aquitanian troper to be of help) to suggest pitches for the Winchester neumes.
Trope verses 8-10 are found in French sources only, apart from Ivrea, Biblioteca Capitolare 60 (which in any case has a high proportion of French material). Provins, Bibliotheque Municipale 12 (from Chartres, thirteenth century) and Paris, Bibliotheque Nationale, n.a.l. 1235 (from Nevers, twelfth century) are most convenient for deriving pitches.
Ex. 11.23.12. Trope verses Kcce iam Iohannes, etc. (Oxford, Bodl. Lib. Bodley 775, fo. 35N) for introit In medio ecclesie (Worcester, Cathedral Chapter Lib. F. 160, fo. 299r)
/   /   .    P   o   .   I   iJ i   J   I   P   J    P .
V---«    <.   fs ,
Ec-ce iam Iohan - nes adest ue-ne-randa  glo-ri - a
11 1
^ , >' *    Q ft . , -^ . « .
cu-i Christus am-pli-o-ra do-na credens mi-sti-ca.
Š)   rn sa*   m   m   m   *ji  s   aaa    'it   aaa  a   a   «3   a*a a
P    * *
IN  ME-DI-0 EC-CLESI-E       A-PE-RUITOS E - JUS. f     J     I   .    P    ...     /     i   I /
Quem uir - gi-ne-o    flo - re sa-crauit.
T2
ETIM-PLEBIT E - UM DO-MINUS SPI-RITU SA-PI - EN-TIE ET IN - TELLECTUS. p     J    I    "    I    I    I    P fil
Past o-rem no-bis t ri-bu - ens.
13
STOLA GLORI-E      IN - DUIT   E - UM.
Sacro fonte potauit et uerbi Dei gratiam in toto terrarum orbe diffudit. 1    J II
Psl _#—+>    a      f    *Z £* *   * a0 fa      t    **_*   *     »   t ~* f
BON U M EST CONFITERI DOMINO     ET PSALLERE NOM1NI TUO ALTISSIME.
r . \ \ J ť . . S . r- p p \ \
1
T 5
g)   *h .  fjTir tgt í**z=»   t & *   *i* (j~Vm~z
-»—#-
Gra-ti-a   cel-sa     De - i   lo-hannis pec-tus   ad-implens.   (II as above) J      I    I    p    J    p   i"-        I"     I   J    l"s      . J _
Un-de sa-lu - ti - fe - re      flux - erunt dog - ma-ta uitae.    (12 as above) J1     M      P     I    i1      I    P    .i     J    I    l ľ
Hocque do - cen - te   pa - t ris do - cu - it   uerbum ca-ro fac-tum.     (13 as above)
23. Tropes 219
(Ex.11.23.12 com.)
Ml
Dox.y—— fa a  a    a   a   a a a J    a *j     a ¥j   ft       aataaaaa #rz
GLORIA PATRI ET FILIO ET SPIRITUI SANCTO     SICUT ERAT IN PRINCIPIO
a    t?   Z*       a    a   a  a   a    a   a  t   fa     # m
ET NUNC ET SEMPER     ET IN SECULA SECULORUM AMEN.
. J p p. /i i1 J i1  j  i r f p p . . J . p p r- i J1 i i
ť . :> s. - a.  >r> •  « 'i „ „ . . ,» . j;
Fons et   o - ri - go     sa - pi - en - ti - e     ad pro-pa-ganda su - e di-uini - tatis archana.
i p i J p J p p . . J . p /i r- \ ß . .     01 as above)
T8
T9
>  - I* **
TP
* .*»...* .*        a   alt,   .v «"* * f=
Qui flu-en-ta   e - uan-ge-li - i   de ip - so   sa-cro   pec-to-re halusit. (12 as above)
o ( j J i p j $ i i !*• /» y  j i r- . , ^  p .
Vi r-gi - ni - tatis quoque    me - ri - to    ma-tri     uir-gi-ni   fi-li-um con - ferens.
|    J      I      I     | (13 as above)
Ps2 $tf> s   fa   a    a0 f*   a   aj £t      a  m   a    a     a    a   a m a   a a   a  ja , m J~
v
IUSTUS UT PALMA FLOREBIT     SICUT CEDRUS QUE IN LIBANO EST MULTIPLICABITUR.
JMJ      .   . J'    i J   r   J   P  P f .  j    J   I, h1 V ,
Til
£a   -m—m    fr1*   * 0*—aj   am   Í0* a   é**   •* ÍJ^-^kaj^-^1
■a-
T12 C
A- -moi an-ge - lo - rum et gau-di-um Christus lo-hannem di- -ligens. (II as <i\    .   J1     I   J    f>"   /•■      I    /•". i     1    \    ft"     r     I   I above)
Quo pan-de-retur   om-ni - bus   lux gen-tibus   uer - bi     De-i.     (12 as above)
f:    ,J    l   ß    .,   Z''       J-     .   1    )   i"    I    ß    J- .
/#g   a*0 fm -
- 9 U—r
Et   hune ad ae - ter-num      ho - di - e uocans con - ui - ui - urn.    (13 as above)
Principal secondary sources for the trope verses: Provins Bibliothěque Municipale 12 (1-2, 8-10); Paris Bibliothěque Nationale lat.1084 (3, 5-7); Rome Biblioteca Casanatense 1741 (11-13)
Trope verses 11-13 survive outside England only in Nonantola manuscripts. It is fortunate that these have diastematic notation.
If one were to guess at the place of origin of these verses, one might suggest north France for verses 1-3, Winchester itself for 4, north France again for 5-7 and France for 8-10; for 11-13 it seems difficult to make even a guess. This is speculation. What
220
//. Chant Genres
is more important is that it indicates that the Winchester redactor probably had more than one exemplar to hand; these in their turn probably derived from a variety of exemplars; moreover, new compositions would continually enter the repertory, as they did at Winchester itself (verse 4). The result is a fascinating variety of 'case histories'. (Planchart 1977 is a magisterial discussion of the case histories of the Winchester tropes; Planchart 1981 considers the methods and problems of such research; see also Planchart 1977, ii. 104-10 on the version of Ex. 11.23.12 in Husmann 1959, 138-9.)
Comparison of almost any two sources for the same trope reveals numerous small variants. These are not usually of major musical importance, except in so far as they suggest that the recorded versions rarely seem to be copied from any authoritative 'original' exemplar. Instead, the way in which the redactor of the source understands a trope, hears it with his inner ear, performs it in its due season, is more important at the moment the copy is made than exact adherence to an exemplar. Sometimes the deviation between sources is so wide as to suggest that the trope is being copied without any reference to an exemplar (see MMMA 3, 39, 123, or an extreme case, 294). And occasionally the music of two different sources will be so dissimilar that one may speak of two different melodies (see MMMA 3, 197, also the studies by Weiss 1964, 'Problem', and 1967). The grey areas between 'similar' and 'different' are sometimes a little hard to define (for a penetrating discussion see Treitler 1982 'Observations'). Occasionally similar music is used for divergent texts (MMMA 3, 62).
The ways in which trope verses introduce, amplify, explain the chants to which they are attached have been well surveyed by Husmann (1959), Stäblein (1963, 'Verständnis'), Paul Evans (1970, Early Trope) and Steiner ('Trope', NG), as well as in numerous articles on individual tropes. Husmann emphasized their role as invitations to begin the singing of a chant, commenting that this was something they shared with many Christian liturgical forms, eastern as well as western (those for the Gloria, Sanctus, and Agnus ace discussed in the next subsection; even the Te Deum was given an introductory trope at St Gall: see Gautier 1886, 170). Muller (1924-5, 566-7) pointed out their close similarity in this respect to the (spoken) prefaces at the beginning of the mass of the Gallican rite. Some Byzantine counterparts are discussed by Strunk (1970). Musical relations between trope and base chant have also been discussed by Stäblein, Evans, and Sevestre (CT I and 1980). Some of the more obvious points may be illustrated with reference once again to Ex. 11.23.12.
Introits customarily use verses drawn from the Book of Psalms, or sometimes another Old Testament book. Since these contain no direct references to New Testament events, all connections between an introit text and the day on which it is sung—Christmas, Easter, St Peter's Day, or whatever—are by special inference (it was central to Christian belief that the Old Testament contained numerous references to the coming of Christ). As far as the text is concerned, Ex. 11.23.12 is typical of many tropes in that it establishes an unambiguous connection between introit text and feast-day. The introit is adapted from Eccles. 15: 5-6 in the Apocrypha, verses which in their original context had nothing to do with preaching the Gospel. Once placed at
23. Tropes 221
the start of mass for St John the Evangelist, they immediately take on a special significance, and for the believer they become one more among the myriad threads woven into the great design in which the coming of Christ was the culmination of all the previous history of mankind. But the tropist goes beyond this inferred meaning of the introit, and spells out in clear terms whose feast-day it is, referring also to John's role as evangelist and as the beloved disciple to whom Christ on the cross entrusted his mother Mary.
Although the trope verses are musically compatible with the introit, they cannot be said to mirror its style to any significant extent. Nowhere in them do we find the reiterated Fs (repercussive notes) of the introit antiphon, nor do they echo the FFFC cadence of 'intellectus' nor the rise ac at 'eura', and in the psalm verses. The way in which both introit and tropes skip over E is simply a characteristic of chant melodies generally, not something which links these in any significant way. Stäblein (1963, 'Verständnis', 91) said that tropes were a sort of 'pseudo-' or 'neo-Gregorian' chant.
The texts of the trope verses in Ex. 11.23.12 are typical of many which complement, but do not lead strongly into the introit text at any point. Other texts do lead on, in such a way that their texts are in no way self-sufficient. For example, the trope set Dens pater filium suum for the Christmas introit Puer natus est reads as follows:
Deus pater filium suum hodie misit in Today God the Father sent his son into
mundum, de quo gratulanter dicamus the world, wherefore we sing rejoicing
cum propheta: (CT I, 78) with the prophet:
PUER NATUS EST NOBIS ET FILIUS DATUS UNTO US A CHILD IS BORN, AND UNTO US A
EST NOBIS. SON IS GIVEN.
This happens also to be one of many tropes in which a strong musical link seems to exist between trope and main chant. Several tropes for this introit begin with the same memorable rise of a fifth as the introit itself (cf. MMMA 3, 291-3, 299, 302-3; see also Stäblein 1963, 'Verständnis' for discussion of several other examples, and comments in Weiss 1965). Ex. 11.23.13 illustrates this. And yet, as Treitler (1982, 'Observations', 92) has pointed out, such a melodic profile is conventional among mode 7 melodies (cf. intvoits Aqua sapientiae, Oculi mei, Respice domine for the same rise of a fifth—the whole textual and melodic relationship between Puer natus est and the Deus pater trope set is explored in depth by Arlt 1982).
Altogether exceptional seems to be the use of a sequence, the short aparallel Ecce tarn Christus (melody 'Ostende (minor)', as an introit trope (see Strehl 1964 and Weiss 1965).
One set of introit tropes (if that is what one should all them) stands apart from the main chant. These are the Versus ante officium, relatively long compositions, consisting of several verses written for alternating sets of singers. In Cambrai, Bibliotheque Municipale 75 (from Saint-Vaast at Arras, early eleventh century) such pieces are labelled 'Ad processionem', implying that they were part of the pre-mass procession rather than the mass introit.
222
//. Chant Genres
Ex. 11.23.13. Trope verse Dens paterfilinm suum (Pans, Bibl. Nat. Lat. 1119, fo. 5r) and start of introit Puer natus est (Paris, Bibl. Nat. lat. 1132, fo. IT)
•    *--9 m   • m0
De-us   pater   fi-li-um   su - urn ho - di-ae   misit in mundum
*   V, *C
'* a
de quo gratu-lanter di - ca-mus cum prophe-ta «   •   *•»  ,  ,md   a,*.  •       ft> a,       *.*   I   a 4*^i
PU-ER NATUS EST NO-BIS     ET  FI-LI-US DATUS EST NO-BIS.
Ex. 11.23.14. Introductory verses Hodie cantandns est, etc. and start of introit Puer natus est (Graz, Univ.-Bibl. 807, fos. 167r, 14r)
f.!67r
—ti,ß 0       • ^» -**— ~*—*--
Ho-      -di - e   can-tandus est no-bis pu-er quern gignebat in-ef - fa - bi -li - ter
t>   /m a
an-te        tempo-ra pater   et   e - undem sub tempo - re   ge-ne-ra-uit   in-clita mater.
*    *o   »    *   * -*■» Ol
Quis est is-te   puer quern tarn magnis preconi-is dignum uoci-fe-ra-mi - ni
m    a    *    a-aj-
*-**--#-»-
di-ci-te   no-bis   ut col-lauda-to-res   es-se pos-si-mus.
^'    &.a, .    ;»   »S   ***   a   ;*T^    _*   , fa   a   i' ^£
Hie e- -nirn est quem pre-sa-gus   et   e-lectus   symnis-ta De-i   ad terras
*'a   a,    *   *   *Ö ~»   •   '*   •*•,•        _   a»~»a~*7»    s-    P   9 ^=
uen-tu-rum preuidens longe an - te       preno-ta-uit sic - que pre-di-xit. f.!4r
~9-	,—Ä-^~*J—.	----JV-	s~. -TV
a f*	' *   a   ««   ** * »	• « «J	at*   a    a a'aa
CTi m			a* a
		a	
V     PUER  NATUS EST NO - BIS   ET FI-LI-US   DATUS EST NO - BIS
23. Tropes 223
The most famous of them is one ascribed to Tuotilo of St Gall, Hodie cantandus est, which has three verses in the order 'Statement—Question—Answer'. This is a D-mode composition which contrasts strikingly with the G-mode Puer natus est. Some writers have thus been led to question its function as a direct introduction to Piter natus est (see the discussion of these versus by Planchart 1977, i. 234-6), while Stäblein (1966, 'Altrömische') went so far as to suggest that it was designed to introduce the Old Roman version of Puer natus est, with which it is tonally more compatible. (This implied that when Tuotilo composed the piece, Old Roman chant had not been fully replaced by Gregorian at St Gall, an idea that other scholars have been somewhat reluctant to accept.) Ex. 11.23.14 is a transcription of Hodie cantandus from Graz, Universitätsbibliothek 807 (Klosterneuburg, twelfth century; PalMus 19).
(Two other famous sets of introductory verses, Quern gueritis in sepulchro and Gregorius praesul, are discussed below, II.25.ii and VI.6.viii respectively.)
A rather small repertory of short introductory tropes for the sequence survives in the tenth- and eleventh-century Aquitanian tropers (Evans 1968).
(xi) Gloria, Sanctus, and Agnus Tropes
Ronnau 1967, Tmpen; Falconer 1989; CT 4 and 7.
Like those discussed in section (x), tropes for the Gloria, Sanctus, and Agnus consist of complementary verses, some serving as introductions to the singing of the chant, some interposed between phrases of the chant. Most of their special features follow naturally from the different chants they complement. Ronnau catalogued over 100 Gloria tropes (or sets of trope verses), of which at least one probably dates back into the ninth century; and Hucbald of Saint-Amand (d. 930) is credited with the composition of another. Although the largest collection of Gloria tropes is to be found in a twelfth-century manuscript, Madrid, Biblioteca Nacional 19421 from Catania in Sicily, they seem to have passed out of use in most centres by the thirteenth century, with the exception of one or two relatively modern compositions (such as Spintus et aime orphanorum, for the BVM). Sanctus and Agnus tropes may include some equally early compositions (on the earliest Agnus trope see Atkinson 1977) and groups of new compositions, especially for the BVM, continued to appear even late in the Middle Ages, many following the fashion for texts in regular rhyme and rhythm (some early Gloria, Sanctus, and Agnus texts are written in hexameters, distichs, etc.). From the catalogues of Thannabaur (1962) and Schildbach (1967) and other research, over 250 Sanctus tropes and over 140 Agnus tropes are known (one or two tropes are common to both categories).
While hardly any introit trope verses 'migrated', as it were, from one introit to another, many Gloria trope verses are found with more than one Gloria melody. And individual verses are associated now with one set of trope verses, now with another set. One such verse was used above in Ex. II.23.6, and it was also remarked there that a
224
//. Chant Genres
slight tonal disparity had resulted from the introduction of material originally designed for Gloria melody 39 into Gloria melody 56. In fact, the whole trope set O gloria sanctorum, in one arrangement or another, is found with Gloria 39, and its use with Gloria 56, as in the Catania manuscript used for Ex. II.23.6, is a peculiarity of a few-English and Norman sources; in Nonantola it appears with Gloria 11 (Vatican XIV).
Ronnau (1967, Tropen, 84-6 and 246-9, with transcriptions) introduced the term 'Wandervers' ('wandering verse') to describe such a verse when it was used in Gloria 39 in the oldest sources: he believed not only that Gloria 39 was the oldest melody in the repertory, but also that trope verses associated with it in the oldest sources and found among several different sets of trope verses were a sort of 'original' layer of trope verses, which were used freely with Gloria 39, to be joined later by various sets of other verses. But it remains unclear whether this is correct, or whether these verses are not simply more mobile than most in a generally unstable repertory, because they are among the oldest, and therefore had longer opportunity to travel.
Some of these points may be exemplified in Ex. 11.23.15, which is a complete transcription of the trope set O laudabilis rex as it appears in Paris, Bibliotheque Nationale, lat. 10508 (from Saint-Evroult in Normandy, twelfth century). The trope is found in sources right across France and in North Spain, Italy, and England as well. An inkling of the complexity of its transmission may be gained from Table 11.23.2. This shows how the veses are arranged in Paris 10508, in Oxford, Bodleian Library, Bodley 775 (Winchester, early eleventh century), Paris 1240 (the oldest of all sources, which has two versions, from Limoges, tenth century), and Paris 1121 (Saint-Martial at Limoges, early eleventh century). As one might expect, given Ronnau's criteria, Paris 1240 has by far the most 'wandering verses'. The latest source, Paris 10508, has only one. The number of verses in Paris 1240 is unusually large. The table gives a good idea of the mobility of the verses, and the different verses which make up the set in each source. Simply saying that a source has the Gloria trope O laudabilis rex (or any other trope) therefore simplifies matters rather drastically. Ronnau called the four verses at the beginning which are common to all versions 'constitutive verses'.
O laudabilis rex is a Gloria trope which makes use of musical phrases identified by Ronnau (1967, 222-38) as formulas used in several other tropes as well (called by him 'Cento-Tropen'; Stablein, 'Tropus', MGG, Ex. 14 is an example of a cento in the truer sense of the word, since its text is composed of distichs taken from Venantius Fortunatus' poem Tempora florigero). In Ex. 11.23.15 the phrase marked 'b' is one of these. The others do not correspond exactly with Ronnau's demonstration, but show how a number of other melodic ideas recur:
'b' describes a double curve abcbaGFGa;
'x' moves from a to F and back again, corresponding to the latter half of 'b'; it is
omnipresent in the Gloria melody itself; in 'y' bb takes the place of a as reciting tone, falling to F and returning to bb or a;
23. Tropes
225
usually it appears in a brief form baGaGFa; occasionally the reciting note is approached FGbb; V is a common mode 3 opening, here a whole tone lower
Much of the time, however, one has the impression of a highly ornate recitation around a or bb, too fluid for convincing labelling. Significantly, the two trope verses which cannot be explained in this way are two unique to Paris 10508: Misertus esto nostri and Nobis in terns (T6 and T8).
Like several other genres for which trope verses were provided, Glorias have introductory verses which stand apart from the rest (see Kelly, 1984).
Among an interesting group of rather abnormal compositions noticed by Stephan (1956) is a Gloria whose trope verses are the Christmas song Dies est leticie (in Prague, University Library VI. C. 20, a fifteenth- to sixteenth-century song-book from Prague), but this is evidently a late sport. The same is true of the Sanctus with the song Surrexit Christus hoclielChrist ist erstanden as trope (in Erlangen, Universitätsbibliothek 464, a fifteenth-century south German miscellany). A pair of full-blown vernacular tropes were added on fos. 299v-300v of Limoges, Bibliotheque Municipale 2 (a gradual of the fourteenth century from Fontevrault): the Sanctus trope Beau's peres and the Agnus trope Cist aigneaus. (The manuscript's compiler evidently had a special interest in vernacular chants, for two French epistle farses appear elsewhere in the source—see section (xii) below.) Even more extraordinary than these is the use of a German dance (if Stephan is right) as a textless Sanctus trope, apparently notated mensurally in Erlangen 464.
Most Sanctus and Agnus tropes (edited by Iversen in CT 7 and 4 respectively) follow the pattern of introductory intercalatory verses. The repertory of prosulas for the Hosanna of the Sanctus has already been mentioned above (vii). Because of the way in which the reciting of the eucharistic prayer normally leads without a break into the singing of the Sanctus, introductory trope verses for the Sanctus are rather few (Thannabaur 1962, signals only four out of over 240 tropes; see Steiner, 'Trope', for examples of both Sanctus and Agnus introductions).
Ex. 11.23.16 has supplementary trope verses for the three 'Sanctus' acclamations, and Hosanna prosulas in sequence form. The difference in musical character of the two types of trope is naturally very marked, though both seem to contain turns of phrase carried over from the main chant. The source used for the transcription, Madrid, Biblioteca Nacional 19421 (Catania, Sicily, second half of the twelfth century), records both prosulas one after the other, Otnnes tua gratia referring to Christ and Martyr Christi gratia, based upon it, for the commemoration of a martyr.
Table II.23.2. Gloria trope sets starting O laudabilis rex
Paris 1240, fo. 41v
Paris 1121, fo. 43v
Oxford 775, fo. 70r
Pans 10508, fo. 37r
GLORIA IN excels1s DEO ET IN TERRA PAX
() laudabilis rex
LAUDAMUS TE
Adonay benedicte
benedic1mus TE
O adoranda
ADORAMUS TE
Glorificande
GLORIFICAMUS TE
O bone rex (W)
GRATIAS AGIMUS
Plebs tua (W)
PROPTER MAGNAM
Sanctam maicstatem (W)
DOMINE DEUS
Pax salus et vita (W)
DEUS PATER
Da pacem famulis (VV)
DOMINE FILI
Pioquc tuo amorc
IESU CHRISTE
Aeternam cum Sanctis
() laudabilis rex
Adonay benedicte
C) adoranda
Glorificande
Pax salus ct vita (W)
Sanctam maicstatem (W) Da paccm famulis (W) Aeternam cum Sanctis Qui solus habes (W)
O laudabilis rex
Adonai benedicte
O adoranda
Glorificande
Pax salus et vita (VV)
() laudabilis rex
Adonay benedicte
() adoranda
Glorificande
Rex seculorum domine
Miscrtus esto
DOMINE DEUS
Magnus et fortis
AGNUS DEI
Rex pacificus (W)
F1LIUS PATRIS
Redemptor universi (W)
QUI TOLLIS
Suscipe nunc (W) Rex seculorum dominc
QUI TOLLIS
SUSCIPE DEPRECATION EM
Qui super astra (W) Caeli terraeque
QUISEDES
O decus omnium (W)
QUONIAM TU
Eros poli (W)
TU SOLUS DOMINUS
Prolis O rutilis (W)
TU SOLUS ALTISSIMUS
Path equalis Audi clemens
IESU CHRISTE
Qui solus abes (W)
CUM SANCTO SPIRITU
Qui unus idemque (W)
IN GLORIA DEI
Rex seculorum dominc Pax salus et vita (W)
Nobis in terris O ctcrni sapicntia Acterni sapientia Tu lux via
Tu lux via (.) virtus honor
Sceptrum tuum (+ prosula)
Rex rcgum (+ melisma)
No G lona melody is indicated in Paris 1240 or Pans 1121. Oxford 775 has Gloria 12, Paris 10508 has Gloria 39. (W) = wandering verse. The Sceptnoti titutn prosula has the same melody as the Regmtni melisma and its prosulas. The Rex region melisma (in Oxford 775) is one of rather few mclismas in the Gloria trope repertory which have a melody different from Regiium.
228 //. Chant Genres
Ex. 11.23.15. Gloria 39 with trope verses O laudabilis rex, etc. and Sceptrum prosula Lumen etermim (Pans, Bibl. Nat. lat. 10508, fos. 23v, 37r)
f.23v
00000**   "o 0
GLO-RI-A    IN    EX-CEL-SIS   DE  - O.
ET IN   TERRA PAX   HO-MI - NI-  -BUS BO-NE UO-LUN-TA- -TIS.
T1 ^ *•. •* *» • i* • * • **'u »; ** *° ^0**0*, £91*0 u&ft^
O lau-da-bi-lis rex do-rni-ne    De-us LAUDA-MUS TE. _x tr
T2
A - do-na-y bene-dic-te     De-us BE-NEDI-CI-MUS TE.
T3   ^*    '   '   >•*   *   0J,   0  ?   *   • .*> 0,t  '*'' *0.*~~^ 0»   •   P*0*. £0**9
O    a-do-ran-da  et  be-a - ta tri-ni-tas   De-us A-DO-RAMUS TE.
Q ,  Z \ * , -
T4   Y  -   0   *'  •••*'• ••A   m '*.» 0**0«. i0**0„0^0JL
Glo-ri-fi-cande   et metu-en - de   De-us GLORI-FICA-MUS TE.
T5    fo' It  00   »  *   0 0?    0   0*0 '"'ct
.*   0 t
Rex se-cu-lorum   do - mi-ne   Ie-su    Chris-te GRAT1 - AS AGIMUS Tl-
V9 0.rm"0  0*0.-0» &0G . U.&-0      fl*.0*0 n^^r~fi^±^.
-BI PROPTER MAG-NAM GLO - RI-AM     TU-      -AM. DO- -MINE
P
*    *• 0.    J0*0*. 0*   0 »0*1(0        f*0.   0*    *    * 0,0^^^-^^jZ
DE-US REX CE-LE- -STIS. DE - US PATER OM - NIPOTENS.
» &(*   *    *    *   i.   »»9        p   ^ s~ •* '0^^^*-^
DO-   -MI-NE FI-LI    U-   -NI -GE - NI -TE. T6   Mi - ser - tus es - to no-stri
i0*0 £0^0.      .* - ■ Q> 0  0  .1 0 *»  i*.u  ft ^c
* 0*00     00        0*00 00 *0 r™    *' 0*9
qui re-de- -mis-ti mundum tu-o   sacro    san - gui-ne.
;t  * 07
IE-     -SU       CHRI-STE. DO-MI-NE DE-US.
23. Tropes 229
ijllj *    * 0    0    m     m. I ,    m0m   9    0*m . — "
AG- -NUS DEI. FI - LI-US PATRIS.   QUI TOL- -LIS PECCA-TA MUN - DI
MI-SE-RE-RE      NO - BIS.                    QUI  TOL- -LIS PECCA-TA MUN-DI _3_._x_*_
T7        y      fi # ;» * 0 j. i*
Pax   sa - lus   et ui-ta ho-minum ti - bi   glo - ri - a.
S •     0     j.*}*!    ,0£l ;rJ*K^
SU-SCI-PE   DE-PRE-CA-TI - O-NEM NO- -STRAM.
T8
V
No-bis in ter-ris mi-se - re-re De-us   al - me. QUI SE-DES AD DEXTERAM PA-   -TRIS MI-SE-RE-RE      NO- -BIS.
					
A h	• •	■                     W        * M			
		w	9		
Vil)				* m	
T9 C
»—
O        e-ter-ni   sa-pi - en-ti-a pa-tris.
_<J---——--							
/L  P-A--^-r*i-					«'«.—^-»-i—-		-m*zi-
—#* *i—**	*v	•—»*—*—■-		_c_	-*—«%—**—*>±f—p—	-»—	* *'-
QUONI-AM	TU	SOLUS SANCTUS.	T10	Tu y	lux    ui - a     et spes	no-	st ra.
"0". -					___—		
/Lb m-		-		—	—*r>-»--a*'**	—"7V"	—i5f—*-
(q) —« •		—*--			-a—»-*m-'-		
v TU SOLUS	DO	-MINUS.	Til	o	uirtus ho-nor   De -	i	pa-tris
et glo-ri - a om-ni-um ui-ta mo-ri uo-lu-is-ti pro cunctis O bo-ne rex. ___3 _
TU SO-LUS AL-TIS-SI-MUS. T12 Sceptrum regni no-      -bi - le.
Lu-men... e-ternum qui splendor es sed de tu - o lu-mine. E. Sa-cre ec-cle-si - e   so-ci-as-ti   ad-mi-ra - bi -li do-te. E.
230
(Ex.II.23.15 cont.)
//. Chant Genres
' 0 » '
&   0*0 Í*
Digna - re proles ab-sol-ue-re di-lec-te. E.
* *
—•—0—*-
Sponse tu-e  di - uo pi - a - mi - ne.
Perma-ne-bit  in e-ternum
IE-     -SU CHRISTE.
> *       0*9 ** ji  * ? *  *  *   v,ft =^
CUM SANCTO SP1-RI-TU IN GLO-RI-A   DE-I    PA - TRIS.
A  - MEN.
Ex. 11.23.16. Sanctus 56 (Vatican III) with trope verses Sitmme pater, etc. and Osanna prosulas Omnes tua gratia and Martyr Christ i gratia (Madrid, Bib. Nac. 19421, fo. 90v)
••*•*•«  /~===gtffc^:~g f   *'  '   «3 r,  f* ,   f.   ^  p'fp  0 E
SAN-     -CTUS        Sum - me pater de quo mundi princi - pi - a con-stant.
1
~rsr
SANCTUS
Fi -li - us om-ni -po-tens per quern patris est pi-e   uel - le.
Ä5
7      * ^
SAN-     -CTUS        Spi - ri-tus   in    quo   par     uir-tus   si-ne fi - ne   re - fulget.
/-»   ft   <ž   »* #   ,   0*Ué 0 0}   0   0 -
DO-
-MINUS DE-US   SA-BAOTH.   PLENI SUNT CE-LI   ET TERRA GLO - RI-A TU-A.
*} & *
o-
-SAN-NA  IN EXCELSIS.
BE-NEDICTUS QUI UENIT IN NOMI-NE DO-MI-NI.
I
-*-0-
Omnes tu-a gra-ti-a      quos a morte  re-demis-ti per-pe-tu-a.
23. Tropes
231
(Ex. 11.23.19 com.)
Morte tu - a uis mortis cum principe proculcans ui-te nos re-pa-ras.
De-o pa-tri dans carum te pro nobis pre-ci-um et ui-uam hosti-am._
Tecum nos re-susci -ta. Tecum in ce-lis col-lo ca. Et re-gni lar - gi-re consor-ti - a. Te  er-go de-posci-mus.    Ut cum iudex ad-ue-ne-ris.   Cunctorum dis-cerne-re me-ri-ta.
P
3E
—'-•-•—» _^ a—a—•—'—a—•-
Nos cum ange-lis et Sanctis so-ci -es. O   -   SAN - NA IN EXCELSIS.
Cum quibus ti-bi ca-namus.........
—a-'    m   ^   a-a-«-*-"-•-*- -V—a~
Martyr Christi grati - a      cuius mortem sequens habes et gaudi - a.
P
Qui pro tu-is e - xoras-ti i - ni-micis ca-ri-ta - te plenus ge-mi-na. De-i  patris ...   fi-Ii-um Ihesum pro te astantem ad patris dex-teram.
Ui-dis -ti magna gra-ti - a......Cer-...-tus proti-nus dul-ci -a
Nunc er-...-go te poscimus martyr sancte ut de-uo-tis fa-mu-lis
1
—-,-m-----W
tecum il-lo ha-bi-turum gau-di-a. tu-a pos-tules ...... consor-ti - a.
Simul ut  an-ge-lis et Sanctis so-ci -i. O-      -SAN-NA IN EXCELSIS.
cum...quibus De-o ca-namus.........
Some Sanctus tropes, and Agnus tropes as well, 'migrate' from one melody to another. With the Sanctus repertory this is perhaps not altogether surprising, since although the number of known Sanctus melodies is very large, they display little tonal variety, for the melodies will normally be compatible with the intoned prayers which surround the chant.
Prosulas for Agnus melodies are to my knowledge unknown. Ex. 11.23.17 has both an introductory verse and verses before each 'Miserere nobis'. Its text is rhyming prose, where each first half-line of five to eight syllables rhymes with the next half, of six to fifteen syllables. This is not matched by musical rhyme, although certain melodic phrases are used several times. (It is admittedly often difficult to separate the 'common coin' of the melodic language from conscious melodic references.)
This source, as is very frequently the case with Agnus tropes in 'western' manuscripts, gives simply a 'miserere' incipit after the later (four) trope verses. It is
232
//. Chant Genres
not certain whether in these cases 'Agnus dei . . . mundi' should be supplied before each trope verse. This is in fact what usually appears in 'eastern' sources. But as it stands here, Ex. 11.23.17 becomes a short litany-type composition, with leaders singing verses to which the choir responds with a refrain. The five 'miserere' are not uncommon in early French sources, although three is the usual number.
Ex. 11.23.17. Agnus 15 with trope verses Pro cunctis deductus, etc. (Paris, Bibl. Nat. lat. 1134, fo. 106v)
P-i--"»-»-*<>   4V." ^#—44+. *——w—v—
Pro eun-ctis   de - ductus ad   im - mo-lan-dum fu - i - sti  ut ag-nus
- »M U»        - *^S- 4m
t   **  *•  »" *» ^     » v +~  " **V M''*mo-^r
re-dempta ple-be cap-ti-ua      te    le - ti   de-pos - ci - mus uo-ce        pre-cel- -sa.
A- -GNUSDE-I QUI TOL-   -LIS PEC-CA - TA MUN-DI
MI-SE-   -RE-   -RE   NO - BIS.
An-       -tiquus pla - stor et ae-ter-ne  ui-te dispo-si-tor MISERERE
No-stro-     -rum cri - mi-num   sor - des ab-lu - e   O      Ihesu   Christe MISERERE
Pri-     -o-rum spes an-ti-    -qua   nostrorum su-sci-pi-at tu-a   uo - ces clemen-ti - a
De-cus  an - ge-lo - rum et    rex se - cu - lo   -   rum MISERERE
ts
0 1—
MISERERE
Italian books prefer another arrangement, one which undoubtedly reflects a real difference in performance practice. Here the trope verse remains outside the Agnus verse. (Iversen in CT 4 makes clear these and other regional variations in practice.) Ex. 11.23.18 displays both this and another fashion cultivated by Italian musicians, that of reusing the music of the main chant for the trope verses. When this happens more than once for the same melodv, tropes with identical music result. The use of a
23. Tropes 233
pes (two-note ascending group) for 'secula' in the last trope verse of.W dextram pains is typical of South Italian recitation passages. This making of trope melodies from the music of the main chant is also known from some Italian Kyries (see section (viii) above) and is related to other Italian composition techniques (on sequences see II.22.v-).
Ex. 11.23.18. Agnus 81a with trope verses.-W dextram patris, etc. and Humanum genus, etc. (Bencvento, Arch. Cap. 34, fo. 18T, and 35, fo. 199r)
Benevento 34
'(•t.....e=
AGNUS DEI      QUI TOLLIS PEC-CATA MUNDI   MISE-      -RE  -  RE  NO - BIS.
-Ml-
Ad dextram patris re-sidens qui semper sal-ua (AGNUS DEI as above)
et  parce     tu-is alme.
41 ,7i
»• *•* ,
Quos tu-o sa-cro re-de-misti cru-ore absque (AGNUS DEI as above)
la-
-becusto-di pastor bo-ne.
& *• i*k-jr
Rex re-gum Deus rector an-ge-lorum sal-ua Benevento 35
nos in se-cu-la se-cu-lorum.
;« * ft
2E
»3 *
--#—•
AGNUS DE-I   QUI TOLLIS PECCATA MUNDI MISE-
-RE-
-RE NO - BIS.
■ [*» *• .] ň
C*K iti
«* ě*ě
ě
Hu-manum genus qui uenis-ti sal-uare quos sal-(AGNUS DEI as above)
-ua-       -sti prote-ge dig-neris.
-* * «0—•---
,T|W,
fir*—<t>ar
Qui morti claustra re-surgendo fregisti (AGNUS DEI as above)
sal - ua
et  par-     -ce quos re-de-misti.
Ne constent nobis pecca-torum uiti - a      quod re-        -sur-ga-     -mus ne pe-re-amus.
(xii) Farsed lxjssons, Creeds, and Paternoster
A different way of elaborating chants on special feasts was employed from the twelfth century onward for lessons and some other chants, especially in the highly individual liturgies of such feasts as Circumcision known from Beauvais, Sens, Laon, and other
234
//. Chant Genres
centres (see above, 1.9). Phrases of pre-existent chants were inserted into the lesson. Such an insertion is often referred to as farsa, both in the Middle Ages and in modern writings, although 'farse' can refer to other types of troping as well. The epistle of mass was the chant most often farsed: over forty examples are known (nearly all listed, with sources, and with the epistle tone they accompany, in Stablein, 'Epistel', MGG; see also Huglo, 'Epistle', NG). The Gospel, by contrast, was left unadorned (Stablein, 'Evangelium', MGG cites three examples only, all found in German and Swiss sources). The Paternoster, Nicene creed, and (from Compline) the Apostles' Creed were also farsed.
The range of farsed chants sometimes present in the festal offices can be appreciated by listing those in the manuscript containing the Sens Circumcision office:
for Circumcision: Paternoster and Apostles' Creed at Compline; Gloria, epistle,
creed at mass for St Stephen: epistle for St John: epistle for Innocents: epistle
(Farsed chants present in the festal offices of Santiago, Sens, and Beauvais are edited in Wagner 1931, Villetard 1907, and Arlt 1970 respectively.)
Table 11.23.3. Farse*
d Apostles' Creed in Ijxon 263
Farse verses
Source (all for Christmas season unless stated)
Solus qui tuetur Sine quo nichil Nat um ante secula Pro mundi remedio [Natus] ineffabiliter Sol de stella Ipse potestate Qui nulla perpetraret Gem it capta Tirannum trudens Unde descenderat Regno cuius iure Reddens vicem Sine quo preces Que constmitur Angeli quorum Quibus deum Inmortalitatem Quern repromisit
Antiphon (Martyr) Beatus vir
Sequence Nato canunt omnia
Sequence Natus ante secula
Benedicamus song Corde patris genitus
Hymn Christe redemptor omnium
Sequencel^tabundus
Hymn (St Peter) Aurea luce
Sequence (Sundays after Pentecost) Stans a longe
Sequence Eia recolamus
Hymn (Easter) Ad cenam agni
Sequence (Ascension) Rex omnipotent
Sanctus trope Peipetuo numine
Hymn (Advent) Verbum supermini
Sequence (Pentecost) Sancti spiritus
Hymn (Dedication) Urbs beata Ierusalem
Antiphon (Innocents) Angeli eorum
23. Tropes 235
The technique may be illustrated by Ex. 11.23.19, the Apostle's Creed as farsed among the Epiphany chants of Laon, Bibliotheque Municipale 263 (Laon, late twelfth century). The piece is also present in the Beauvais and Sens offices. Villetard's list of the sources of the borrowed phrases is given in Table 11.23.3).
Ex. 11.23.19. Farsed Credo (Laon, Bibl. Mun. 263, fo. 139r)
.   i   '   *   *    *    *     '   '  '       C        11: , =
CREDO IN DEUM PATREM OMNIPOTENTEM. So-lus qui tu-e-tur om-ni-a.
So-lus qui guber-nat om-ni-a.
.   a   9   *    '   *  ji   f. -j- =
CREATOREM CELI ET TERRE. Si - ne quo nichil est cre-a-tum.
a    a      *      *      •   a  *    a M
I
-+-^a}
v- — -
ET IN IHESUM CHRISTUM FILIUM EIUS UNICUM. Natum an-te se - cu-la.
*   M    -g»                                      *   *          *    •    'i *   * *     .   Cé   »*0   m .= **°    jfj      i a  «w w *
DOMINUM NOSTRUM.          Pro mundi re-me-di-o carnis o-pertum pal - Ii - o.
,    .     *    *     9     9 9  9  *   *g   áj £.   <• ä}    •   £•   .   m ,
QUI CONCEPTUS EST DE SPIRITU SANCTO. Na-tum in - ef - fa - bi - li-ter.
a'o -~)—#--
EX MARIA UIRGINE. Sol   de    stel - h
•   *   * '3»^=
*" PASSUS SUB PONTIO PILATO. Ip - se po-tes-ta-te tra-di-ta. -m *  t   *    m    a    a    Ca '   *o   m   *   a J—g-
CRUCIFIXUS MORTUUS ET SEPULTUS. Qui nul-la perpe-traret fa-ci-no-ra.
a     •     *   • *l a     * *#
DESCENDIT AD INFERNA. Ge-mit capta pestis an - ti-qua.
$)a''"'m'*'     •   jfo i\—a-
TERTIA DIE RESURREXIT A MORTUIS.
•     *•    ě    •    tj    ,    9 •
Ti-rannum trudens uincu-lo.
236
(Ex. 11.23.19 rout.)
II. Chant Genres
-#—m-
ASCENDIT AD CELOS.
Un-de descen-derat.
•    •     m    m  m   m    m    m —n
SEDET AD DEXTERAM DEI PATRIS OMNIPOTENTIS.    Reg-na cuius iu-re tenet pa - ri-li.
*    •    •     »   »   »   m   ä   * m
INDE UENTURUS IUDICARE UIUOS ET MORTUOS.
Reddens uicem per ab-di-tis
«  # *
iustis-que regnum pro bonis.
CREDO IN SPIRITUM SANCTUM.
Sine quo preces omnes casse creduntur et indigne De-i auribus.
sift 0=g^
SANCTAM ECCLESIAM CATHOLICAM. Que construitur in celis uiuis ex la - pi - di - bus.
V s    '    •    •   '? *'
SANCTORUM COMMUNIONEM.
An-ge-li quorum semper uident faci-em patris.
1    m *
* It
£ *
«—•—f
REMISSIONEM PECCATORUM.   Quibus Deum   of-fen-dimus   cor-de uerbis o - pe-ri - bus.
t *   i,   £• #?
CARNIS RESURRECTIONEM.
In-mor -ta - Ii - tatem cum Christo.
UITAM ETERNAM.        Quem repro-misit    De - us   di - li-gen-tibus se. A-
-men.
Probably the most widely known farsed epistle was the one sung all over Europe (see sources in AH 49, 171 and Stablein 1975, 61, n. 601) on Christmas Eve, Christmas Day, or Circumcision, whose introductory verses are Laudes Deo dicam per secula, seemingly made for the purpose, in alcaic metre, and sung to a simple repeating musical phrase (Ex. 11.23.20). (Some sources read Laudem and/or dicant.)
24. Latin Litu/gical Songs
237
Ex. 11.23.20. Introductory verse Laudes Deo for a farsed epistle (Limoges, Bibl. Mun. 2. fo.
26r)
Laudes   De -o it- '3 -
dicant  per  se - cu - la
rrVr*—■—	-—--	*       9 '		*	'm
vrp—	•    •    • •				^*
P qui	me plasma-uit	in   ma - nu	dex -	- te	- ra
et	re-for - ma-uit	cru-ce pur	-pu -	re	- a
san-	gui-ne na-ti	qui cunctos	re -	de	- mit
ab	or-tu so-lis	or - bis per	cli -	ma	- ta
us -	que ad mundi	par-tes oc	- ci -	du	- as
in lau - de cu-ius
cla - mo-res   ex - ci
tat.
Ex. 11.23.21. Start of farsed lesson with verses Ce que Ysaies nos escrit, etc. (Limoges, Bibl. Mun. 2, fo. 46v)
•     * *
m     M m-
___9     *      m* •-
~f-*-»--
1.......Ce que Y-sa-ies nos es-crit...       de ... 1'a-ue-nement Ihe-su Crist ...
2.......Or nous leuons en contre  lu - i si comme orroit encore.........
3.     Di-ex ap-parut ...c'est la pre-miesse        et l'estoille est I'autre lu - mie - re
3j 9
1: Ir jj (f
1. bi-en nos do-it   es-tre hu - i   en re - meenbran-ce
2. hu - i doit es-tre chascuns......... es - cle - ri - ez
3. par cui i uindrent............... li   troi   roi ...
1.   qui  en   di-eu   auons no  fi - en - ce ...
2.	li seinz	... .	..... iors	est re - pe-	riez						
3.	la    ti -	er -	ce lu-mie	-re est la	fo-	iz.					
	I: iff. JC										
											
				- *—K—*}-			9     W #-				
p 1.		•		-Wf-*w-		*	*---                "' - *»t	* *	**		
	car il	...	en hor-te	et semont		nos	me-ismes et	tout ...	le	mo -	rit.
2.	qui trois	meunieres de		clar - tez		nos	a dou... ciel	hu - i	a -	por -	tez.
LECTI-0   Y-SA-I - E PROPHE-TE.
238 //. Chant Genres
Other farsing verses 'made to measure', so to speak, rather than lifted from elsewhere, are a small number of vernacular insertions found in a few French and Spanish sources of the fourteenth century onward (see Huglo, 'Farse', NG). Limoges 2 (Fontevrault, fourteenth century) is the most important of these, with five sets. The opening of the Epiphany farse, which continues as an alternation between more verses of the farse (all with the same melody) and lines of the epistle is transcribed in Ex. 11.23.21.
Of several surviving Paternoster farses only Fidem auge in the Sens and Beauvais offices is relatively well known (edn. in Villetard 1907 and Arlt 1970). Stablein (1977) has surveyed over twenty examples, of which one is polyphonic, for two voices, and two-thirds are copied without musical notation. Laon 263 has two, all other sources one only. Four are transcribed by Stablein.
11.24. LATIN LITURGICAL SONGS
(i) Introduction
(ii) Versus in Early Aquitanian Manuscripts; Versus with 'Double Cursus'
(iii) Twelfth-Century Songs: Textual and Musical Style
(iv) Twelfth-Century Songs: Liturgical Function
Schmitz 1936; Stäblein, 'Saint-Martial', 'Versus', MGG; Emerson 1964; Harrison 1965; Treitler 1967; Arlt 1970; Crocker, 'Versus', AG; Planchart and Fuller, 'St Martial', SG.
Editions: Villetard 1907; Treitler 1967; Arlt 1970.
Facsimiles: Arlt and Stauffacher 1986; Gillingham 1987, 1989.
(i) Introduction
Benedicamus chants with extra text supplementing the basic 'Benedicamus domino: Deo gratias' have already been mentioned (II.23.ix). By the twelfth-century such compositions were often made with texts having rhyme and often regular rhythmic stress, features of new composition in many genres during that period. The texts are often strorjhic, and some have refrains. Very often the new style of the text (rhyme and regular stress-patterns) is matched by a new musicaj^tyje, emphasizing steps of a third (tertian nielotlic style), presenting the fifth and octave above the final as melodic goals, with free and rapid movement through the chosen range, a tendency to compartmentalize movement of individual phrases into one or other segment of the octave, and the coupling of stressed syllables to those notes which form the tonal
24. Latin Liturgical Songs
239
backborie_of the music. Some pieces refer to liturgical formulas other than the Benedicamus text; a large number are free^f HtujgicaLr Medieval names for
these compositions include versus, conductus, cantilena, and cantio. Hjyg they are called songs.
These songs share_some charactejisjtics wkh_ hymns and processional......hymns
(11.15). Like the songs, hymns have poetic texts in strophic lonu, and sometimes regular rhythm, with relatively sixnplernusic. Processiojiaj^ujijcjtjon and strophe-plus-refrain structure are further links. The new compositions did not, however, penetrate the traditional cycle of office hymns to any significant extent.
F'or many of the twelfth-century songs a fujn^tipn has to be surmised from circumstantial evidence. The same is also true of some other sacred songs in Aquitanian sources of the ninth to eleventh centuries, discussed briefly here before the twelfth-century repertory is addressed.
(ii) Versus in Early Aquitanian Manuscripts; Versus with double Cursus'
Paris, Bibliotheque Nationale, lat. 1154 is a manuscript from Aquitaine probably of the second half of the ninth century, whose contents include (fos. 98-143) an important _coflection of Latin poems, some of which were notated (probably not before the tenth century). Most of the pieces are called^f.m{5 in the manuscript, others ritmus, cannen, planetus (for two laments), hymnus (two of them are the processional hymns Pange lingua and Tellus ac ethra), and one jprosa (a liturgical sequence for St Martial, Concelebremus sacram). Several of the pieces are_abecedary hymns, that is, strophic compositions where each strophe begins with a different letter,, progressing.. from_/\,jhr.oug)l...th.e. alphapet. A wide range of classical metres is represented in the collection, and the use of £Sj-^aj£ejs strikingly common.
Only four of these pieces have been found elsewhere as liturgical items, and the only reason for mentioning the rest is that their compositional, techniques were presumably known to, and might have served as mojdels for,_.the writers of liturgical pieces. Those found in liturgical manuscripts are the two processional hymns and the sequence just mentioned, which are copied together at the end of the collection, and another sequence^Sjincte Paule. Many of the works are sacred at least in a general sense; some are topical, secular works (inventory and bibliography in Spanke 1931). Several of the poems are attributed to Gottschalk, Paulinus of Aquileia, and Boethius.
Scattered items of this sort are found in other sources. (See Fallows, 'Sources, MS, III, 2: Secular Monophony, Latin', NG.) In numbers they are more than outweighed by the fifty 'Goliard' songs in the 'Cambridge Songbook' (Cambridge, University Library Gg. v. 35, copied in the mid-eleventh century at St Augustine's abbey, Canterbury, from a Rhenish exemplar), certainly aj^cular repertory, but again one whose poetic techniques, may not bej^bj>ut.xej^ trie great majority of the pieces, no transcription is possible.
The importance of these works in the history of liturgical music has perhaps been
240
//. Chant Genres
exaggerated, since—-unless more of the works can be shown to have fulfilled a role in the liturgy—to view them as any more than possible models for liturgical compositions seems over-enthusiastic. A considerable body of literature has nevertheless arisen around some of them, in particular Satjcte Paule, and one other piece, Dulce carmen, labelled 'Vejrju^de^oo^Ma^uritiQ'. These are the two examples in Paris 1154 of what has been called (since Winterfeld, 1901) the 'sequence with double c.ursus' (Handschin's less happy term was 'archaic sequence'). These compositions are constructed at least partly in paired verses, ljkejhe sequence; but their later verses repeat the .musk of earlier ones, so that the text also must use the_s^Tue^t^njc^ure,: hence the term 'double cursus'. So fareight such pieces have been discovered. It should be pointed out that sets of four verses to the same melody, or even more, are almost as common as pairs (see the tabular summary of their structure in Phillips and Huglo 1982, with bibliography, to which can be added Stablein 1978; on 'recapitulations' in early liturgical sequence melodies see above, II.22.ii).
The most famous of these double-cursus versus is Rex caeli domine, cited in the j}inth:century treatise Musica enchiriadis and among the earliest.known_examples of jpolyphonic music (see Phillips and Huglo 1982 for corrections to the editions in Handschin 1929 and Stablein, 'Sequenz', MGG). Given the lack_of,,,connection htXyi&^.Mex. caeli (and most of the other double-cursus versus)jjidjh^.Iiturgy the assignation to such pieces of a critical role in the early development of the liturgical sequence by such writers as Dronke (1965) should be treated with caution. The same should be said of the early versus in general. There can be no doubt as to their artistic accomplishment: their texts display a considerable range of metre- and rhyme-schemes, and their music is frequently cast in short-breathed phrases to reflect the poetic designs. But in precisely these respects they are quite different in impact from those of the normal tenth- to eleventh-century liturgical sequence. Whereas the first-epoch sequence couples an expansive melodic exuberance with formal ruggedness, the versus prefers the deft interplay of small poetic and melodic units, without aiming for the monumentally of.the. sequence.
Ex. 11.24.1, the opening of Sancte Paule pastor, bone, shows how the versus in Paris 1154 (such as can be transcribed, that is) seem to proceed in jdjort versicles which are in^rjdaXe^jm^ l^Jin^in the poem is longer .than eleven syllables
(in Rex caeli no more than thirteen syllables), and the^ayerage length is much shorter. The overall niujdcaljorrr^ of the piece, following de Goede's reconstruction, is:
\ A A Bl Bl C C B2 B2 Dl D2 / A A    Bl Bl   C C   B2 B2   D3 D4
I B3
i
But this is only part of the story. As can be seen from the transcription (which is of AA B1B1 CC B2B2 in the above scheme) eadh„ verseis composed of smaller lines, Jn_dicated by_srnajljetters in the transcription. (The version of the 'D' section in Paris 1154 is corrupt according to Spanke, 1931, and thus the reconstruction presented by
24. Latin Liturgical Songs
241
Ex. 11.24.1. Start of versus Sonete Poule pastor bone (Paris, Bibl. Nat. lat. 1154, fol. 129v; Benevento, Arch. Cap. 34, fo. 208v)
a
m
Bl
B2
Sancte Pau-le pastor bo-ne      Vas e-lectum do-mini Quern Damasco propinquantem  Lux di - ui -na su - bi - to 2k__a.__
Predi-ca-tor gen-ti-um e-gregi-e Circumful-sit his uerbis interrogans
í
Sau-le Sau-le quid me cedis Tunc in terram corru - is-ti
E-go sum Hiesus benignus Nec suffe-re po-tu-is-ti
Omni-um condi-tor Qui te quoque condi-di ex ni-hi-lo Uocis uim     ce-li-ce      Mox fu-tu-rus doctor in ec-cle-si - am
Í-*-		c				
	, #—»—♦«—*—	--•—( _—w-0—<►*-	* *			-HI
Se-minans     fi-de-i      Dog-mata ca-tho-li-ce Ex qui-bus   nos su-mus     Licet no - uis-si - mis 4-?--
Nu-merum credentt-um Accu-mulans Consti-tu-ti tempo-ris Angus-ti-is
P
Nos audis supplican-ter Lux decus or-bis hu-ius Tu pastor no-bis da-tus Tu gregis Christi custos _a- _&_
At-que uoce fa-mu-lorum      Pu-ra mentě pos-tulantum E-ius pro sa-lu-te semper     Suppli-ca-re re-ginostro S _o_
Re-ci-ta iu-di-ci qui ce-lo-rum Continet in-sig-ni-a Cu-ius im - pe-ri-um      si-ne fi - ne Manet in per-pe-tu-um
de Goede 1965, LXI is hypothetical. For further information on this and other such pieces see also Stablein 1978.)
(iii) Twelfth-Century Songs: Textual and Musical Style
Some of the stylistic characteristics of the^ady secular versus (though not the double-cjursus idea) are also found in the songs of the twelfth century. Whether it was that poetic and musical resources previously considered appropriate only for secular songs were now brought into liturgical use is not clear. (Such poets as Abelard—see Weinrich, 'Abelard', NG, and Huglo 1979, 'Abelard'—contributed to both liturgical and secular repertories.) Another question, which might seem to stand the previous idea on its head, is whether the liturgical songs of the twelfth century exercised any influence on the vernacular lyrics of the troubadours. The matter is really beyond the
242
//. Chant Genres
scope of this book, but readers may like to consider for themselves whether or not the typical features of the Latin liturgical songs described here are strongly reflected in the vernacular repertories. (No writer has subjected both textual and musical features of both repertories to an adequate analysis. Starting-points in the various areas are provided by Spanke 1930-1 and 1936, Treitler 1967, Arlt 1970.)
Ex. 11.24.2 is a typical example of a twelfth-century liturgical song, one which includes the words 'Benedicamus domino' and 'Deo gratias', fitted neatly into the rhyme-scheme of the poem, in the extended third and sixth strophes of the piece. Mejodic movement is almost completely tertian for the first four lines of the piece. At first the triad a—c-e is stressed, then after the rapid descent to low D and C there is some gentle playing off of the D-F interval against the C-E-G triad upon which the piece finally settles. The iambic_.jh^tJbin of the text is so pervasive that slight irregularities (according to classical Latin usage) seem irrelevant. The quantities are less regular in alternating short and long. Yet it is noticeable that, two-note groups always iall on ...what, are jelt ,to,b_e stressed syllables in t;he second half of each strophe. Whether this reflects a performance in i_anliinioiisaip:J^ of course
open to question. What is beyond dispute is the regjjlar^rjhythmic nature of the text, which the music..highlights rather than disguises. (Regular note-grouping in the opposite sense, where it is the unstressed syllables which are given two-note groups, can also be found, for example in Dei sapientia, Arlt 1970, Editionsband, 148.)
Ex. II.24.2. Benedicamus song Lux omni festa papulo (Madrid, Bibl. Nac. 289, fo. 132v)
_-L_ a~ a   a a
a a
1. Lux om-ni fes-ta po-pu-lo b
re-cur-rit an - ni   cir-cu - lo 1: (_____
a a
qua nun-ci - an - te an-ge-lo
~a a,
ň—*
ex-or-taest  re-dempti - o.
Nostra-que li - be-ra - ci - o
ö—a~
ser-pentis ex   a - cu - le - o.
2.   a   Dum omnia silencio b   et nox inter altissimos c   Sermo tuus O genitor
continerentur medio perageret curriculo. regali uenit solio
3.   a   Sponsus uti de thalamo b   ita de matris utero
F
c   Pro seculi remedio
F
c   Quo circa nos in iubilo
elr
preceteris formosior processit orbis conditor. effectus est Deus homo. BENEDICAMUS DOMINO.
4.   a   O matris alme uiscera b   que genuerunt talia c   Beáta quoque uberá
24. Latin Liturgical Songs
repleta Dei gracia tanta sacrata pignora. que puer itle sugxerat.
243
5.   a   Cui tota celi curia a. c
b Cui talis est potencia c   celestia terrestria
tremens in laude consonat ut illi que sunt omnia genuflectuntur subdita.
a Cuius misericordia
b A morte nos perpetua
c Secundo nos eripiat
c Ut in polorum regia
et admiranda bonitas. aduentu primo liberet. ab infernali fouea. dicamus DEO GRACIAS.
This.simpje_and direct melodic idiom is the preferred style of what may be called the North French tradition in this international repertory, represented by the.sources with festal offices from Líumi, Sens, and Beauvais (1.9) and also the earlier tropers from Norman Sicily (Madrid, Biblioteca Nacionál 288, 289, and 19421), all of which manuscripts date from the twelfth or early thirteenth century. A^.uitanian..twelfth-century^jgurces occasionally record much more florid compositions. Letabundi iubilemus, Ex. II.24.3, is from Paris, Bibliothěque Nationale, lat. 1139, where it begins a series of pieces headed Tc incoant Benedicamus' ('Here begin the Benedicamus pieces'), although the text of this particular piece makes no reference to the ritual words. (In the Le Puy office it is rubricked as a 'conductus'.)
The text of Letabundi iubilemus is clearly of the newer type, with Hiie-lengths far joiQr£..^j.ried than in Ex. II.24.2 (' = stressed or half-stressed at line-ends; = unstressed):
Letabundi iubilemus Accurate celebremus Christi natalitia summa leticia cum gratia
produxit .' '
gratanter mentibus fidelibus
inluxit. .' '
The text consists of three,regular strophes, set in an ABA (da capo) musical form. But the textual similarity of the three strophes is quite obscured by the very ornate setting for the central strophe (in this source, though not in the Le Puy version), as if the compressed energy of the syllabic lines suddenly explodes in a virtuoso vocal display. Chains of thirds are much less prominent, partly because of the ornate
244
//. Chant Genres
surface of the melody, though the strong K—G and a-c-e chains of the first three lines, and a-c-e in 'quod in nobis' and 'arcanum', are heard clearly enough. However, if the note-groupings within the melismas reflect phrasing in performance, then other thirds may be audible: in '(Eructa)vit' for example, c-e,f—d, e—c, d-b, etc. And such progressions as that at 'vis acerbum', partially concealed by the text, are also of
Ex. II.24.3. Latin song U>tabundi iitbilemus (Paris, Bibl. Nat. lat. 1139, fo. 58')
►IT,
Le-ta-bun-di iu-bi - le-mus ac-cu-ra-te cele-bremus Christi na-ta-li - ci - a O   re dig-r.a   pre-di - ca - ri    cui non ualent conpara-ri   quanta uis mi - ra-cu-la
summa le - ti - ci -a fe-rit uirgun-cu-la
cum grati -a per se-cu-la
pro-du- -xit rec-to- -rem
gra-tanter menti-bus      fi-de-li-bus       in- -lu - xit.
conceptum e - di-dit     nec perdi-dit
pu-
-do-rem. (Fine)
E	- rue -	ta -	ui -	-t
V i	• • •* *		m        a 0 a    «*«     .*«.   • »• »»	
/V ■		< ** *	m*im      • •   *'*m •   •       • •#	
rA\		0M		
				
pa- -te- -r uer-bum
r*-*m
per-dit       ho-      -stis uis
-bum
T—*—TT*-*'*.•
«—'»
quod in no-bis     ha- -bu-
-it     quod di-u   la-tu-it       tunc pa-tu-it
— m*m      mm    m*m*'» *» —
3BE
'•m '•''<*.
ar  - ca-
-num
.      ,4*     »0 00
qui con-tra ga- -ri-unt in- -sa-ni- -unt ^    ..... .-^m^**^
ua-
-num. (Da Capo al Fine)
24. Latin Liturgical Songs 245
significance: G-b, a-c, b-d, c—e. An important element in the setting of the 'Eructaviť strophe is the roulade cde-fed-edc—dc in one permutation or another.
General speaking, the Benedicamus songs are the less adventurous in poetic metre and form, and few Benedicamus songs are in florid musical style, whereas songs wiTh" free texts often display rhythmic and melodic designs of considerable brilliance. The extravagance can be purely rjiythmic, as mj\gtus est, natus est (Ex. 11.24.4), where the repetitive melody highlights, by its very naivete, the irresistible drive of the poem. The piece has two large strophes, with numerous internal repeats, which link it indirectly to compositions like Ex. 11.24.1 above.
Ex. 11.24.4. Latin song Satus est, natus est (Madrid, Bibl. Nac. 289, fo. 144v)
3=				-fc-£-	-f-6- -1	-			-ITT"
					--	H-■—	-a-*-	—-—	--Hr-
		•		•		•		-w-	• *
	•		ft	•	•				* t
1.	Na-	tus	est	na-tus	est	na-tus est qui mun-di	ho - di di - lu	- e - it	do-mi-nus fa - ci - nus
2.	I -	gi-	tur	i - gi-	tur	i - gi - tur	munda	- na	fa-bri-ca
i
I
lam no - ua
concrepent can-ti-ca
quern pax
pa-ter factor om-ni-um in hoc mi - sit e--xi - li - urn ut fac-turam   re-di-me-ret et
est in ter-ra red-di-ta per pro-to-plaustum per-di-ta   or-ta pro-le summi pa-tris ca-
pa-ra-di-so red-de-ret -ro fac-ta car-ne matris
3
2.
nec mi - nu - it quod e - rat a-sumens quod non   e - rat
ci-pres-sus ex pla-ta-no ue-ni-ens   a Li-ba-no
sed carnis sumpto palli-o in uir-gi-nis pa-la-ci-o
est in-cli-na-ta de-i-tas ut   re-spi-ret huma-nitas
ut sponso de thalamo processit ex u-te-ro flos de lesse uir-gu-la fructu replet se-cu-la
O quanta le-ti-ci-a O et quanta glori-a quanta De-i gra-ci-a sunt in-ef-fa-bi -li-a
1. hunc predixit prophe-ti-a
nas-ci-turam ex Ma-ri-a
2. O na-ti-ui-tas miran-da O et di-es ue-neran-da
* * »3
quando flos is - te nas-ci-tur di - a - bo-lus confun-di-tur
O stel-la ma-ris in-cli - ta e-ternum florem prodi-ta
—
t*   '* r
-*2-
-W »-»-*-m ~l
1. et  mo - ri-tur mors   et   mo - ri-tur mors.
2. ut   ad-iu-uet nos    ut   ad - iu-uet nos.
246
//. Chant Genres
Poetic exuberance of a different kind is displayed mJJ.a laudis homo (Arlt 1970, Darstellungsband, 212), where the rg£etitioji-is--textual, suggesting an almost rapt amazernent at the wonder of the Incarnation.
Refrain songs are another characteristic feature of the repertory. Exx. II.24.5-7 show three examples. In Ex. II.24.5 the refrain consists simply of line endings.
Ex. II.24.5. Bcnedicamus song Thesaurus nove gracie (Madrid, Bibl. Nac. 289, fo. 134v)
$     '    '     " ' '
1. The-sau-rus no - ue gra - ci  - e est    re - ue - la - tus ho - di  - e
2. Tarn cla-ris in    na - ta - li - bus gau-den-dum con-stat om - ni - bus
3. Le - te-mur er - go pa - ri - ter psal - la - mus un - a - ni - mi - ter
m 41
3
1. ce - les - tis   flo - rem ger-mi - nis e - ma - nu - el pro - du - xit    al - uus  uir - gi - nis         in   is - ra - eJ .
2. se - cun - da   sa - lus    om - ni - urn        e - ma - nu - el cu - mu - ne   fa - cit   gau-di - urn        in   is - ra - el.
3. et   cor - de   uo - ci    con - so - no        e - ma - nu - el BE-NE - DI - CA - MUS DO-MI-NO in   Is - ra - el.
(most of strophes 4-6, for DEO GRACIAS, lost)
Ex. 11.24.6. Latin song Plehs Domini hoc die (Paris, Bibl. Nat. lat. 3549, fo. 167v)
' '    *   v —*~ v « --
1. Plebs do - mi-ni hac di - e        le-ta-mi-ni set pi - e        laus uir-gi-nis     laus uir-gi-nis
—*-4»-*-m—«   —-K—•»-*—*+w+9-'
Ma-ri-     -e        et cor-dibus      et  uo-ci-bus     et ac-ti-bus pro-ma-
-tur Ma - n-arn uox       Ma - nam cor      Ma - nam sensus mens uigur
i
f ^ * **———Tjr
proclament hac in di-     -e et    fi - li-um   Ma - ri-
24. Latin Liturgical Songs
247
(full text AH 20 no.149 p.l 19)
8.    Hoc sobrie de quanta stirps grácie gens sancta materie materie de tanta cum gaudio fit mentio iam lectio legatur.
Ex. II.24.6 has a refrain ('Mariam uox', etc.) equal in length to the strophes of the poem, and perhaps outweighing them musically by virtue of the rjjelismas.. (it is characteristic of many songs to place melismas on the ljasjLstressed.syllable of line; Ex. 11.24.6 has both short and long ones). The poem ends with an invitation to begin the reading of the lesson, and is thus one of several such songs whose probable function was to accornpany the procession of the subdeacon or deacon to the lectern to intone the Epistle or Gospel. (There is no indication for a repeat of the refrain after the rTnaTstrophe.)
Ex. II.24.7. Latin song Ave mater salvatoris (London, Brit. Lib. Add 36881, fo. 32v)
y								
A	u							
	v		'      ' M	0	-0		a     *     *    a *	
\v		m *	W			—w-		-w
1.		A-ue	mater sal-	ua	-to-	ris	nostri terminus do-	1 or i s
								
/ w		•li						
fiX\ \~			■        * m	#				
		• •	W			—^.		0
		Uirga	Ies-se cu-	i us	flo-	ris	-——J—- mater es et  fi - li -	a.
		Nostri	ter-minis	do	~lo-	r z s	confert nobis gaudi	-a .
2.    Moyses ardentem foris nostri terminus doloris uidit   rubům sed ardoris non  passus incendia. Nostri terminus doloris confert nobis gaudia.
3.   Angelici uerbum oris
nostri terminus doloris de supernis missum oris te repleuit gratia. Nostri terminus doloris confert nobis gaudia.
Ex. II.24.7 is an example of what has been called a 'Latin rondeau', by analogy with the French vernacular rondeau of Adam de la Halle and his contemporaries. It is easiest to represent its construction schematically. The music uses only two melodic units, in the order AAABAB, the text includes the first half of the refrain within the strophe, then the full refrain at the end:
strophe line 1 refrain line 1 strophe lines 2, 3 refrain lines 1, 2
248
//. Chant Genres
In French vernacular rondeaux the practice was usually to sing the full refrain at the beginning also.
One jothex.....important form was used more rarely in the Latin liturgical song
repertory than for secular Latin.songs (the so-called ^^ti^-Dame^cpnductus, for example): that of the sequence, or lai, as it has sometimes been called in this context. The paired-vers£.stmcture of the sequence is already visible in Ex. 11.24.3. The.best-known example is Alto consiljo, found in the Beauvais festal office and several other sources (for editions see Arlt 1970, Editionsband, 157; Stablein 1966, 'Musik'; Van Deusen 1982, 50.) What distinguishes such pieces from the more usual rhymed sequence is the rji\£td}rrn_c yariety in their texts and the tendency to build strophes out of a^su£cession of phrases of diverse length and rhythm.
(iv) Twelfth-Century Songs: Liturgical Function
Ex. 11.24.8 is another song from Paris 1139, one of three in the manuscript in the vernacular, or mixing L^tm^axui,Px.oyens.al (inventories in Spanke 1930-2 and Sarah Fuller 1969). Be deu hot mats neatly illustrates several of thejgroblems surrounding many of the songs found in the twelfth-century sources. The composition has been called (since Gautier 1886, 169) a trope of the Tu autem, that is, of the versicle sung after the end of a lesson in the office. Yet only the start of the text 'Tu autem domine miserere nobis R. Deo gratias' is quoted, and there is no musical correspondence (see Tones for the Lessons in LU). It is therefore questionable whether 'trope' is an appropriate term to use. By analogy with other examples, it seems possible that Be deu was a substitute for the Tu autem versicle.
Paris 1139 is the first of three twelfth-century Aquitanian books containing considerable numbers of liturgical songs (the others are Paris 3549 and 3719), and it contains pieces referring to several liturgical forms. De supernis affero nuntium (fo. 32r) invites the cantor to begin the singing of Deus in adiutorium, the opening versicle of all the office hours. Xunc clericorum concio (fo. 33v) bears an erased
Ex. II.24.8. Marcaronic song De deu hoi mais (Paris, Bibl. Nat. lat. 1139, fo. 44r)
• -r-
Be deu hoi ma-is finir nostra ra-zos un pauc soi las que trop fo aut lo sos
_b----
3 - „-    '    * ,
le-uen doi clerc que di-jen lo respos tu autem deus qui est pai-re glo-ri-os.
Nos te pre-iam que t' re-mem-bre de nos   quant tri-a-ras los mals d'antre los bos.
24. Latin Liturgical Songs 249
'Benedicamus' rubric, and in fact announces that a lesson is about to be read. And this is the first of many twelfth-century examples of songs evidently intended to introduce the reading of a lesson. (One has already been given above as Ex. II.24.6.)
The ritual words with which the lector asks the officiant's blessing before reading a lesson in the office is the versicle jfube domne benedicere (LU 119-21). And the words 'Jube domne' do indeed appear" at the end of one song, Congaudentes iubiletnus, in the Norman-Sicilian troper Madrid 289 (called a conduct us in the manuscript—it is the earliest source to carry this word as a rubric, which is equivalent to the word versus in Paris 1139). As with the 'Tjj autem' song, the music of the original tone is not used. Three Jube songs also appear in the so-called 'Codex Calixtinus' (Santiago de Compostela), dating from thej 160s, some twenty years later than Madrid 289 (ed. Wagner 1931, 39-40, 94-6, all called 'Conductum' and ending 'Lector lege et de rege, qui regit omne, die lube domne').
Of all songs apparently designed to introduce a lesson, ^Resonet in tone t was evidently the most popular. With its final strophe:
] Munda sit     pura sit     hec ergo concio Audiat sentiat      quid dicat lectio
it seems equivalent to a Jube song, and appears in front of an epistle with farse Laudes deo die am per secula in Madrid 289 and a Beneventan source of the same period in private hands (see Stablein 1975, 144 n. 8). It is also found in the song-book Cambridge, University Library Ff. i. 17 (Norman or English, twelfth-thirteenth century). But in two other south Italian sources it is amalgamated with Laudes deo and used as a farse throughout the epistle (partial facs. and edn., Stablein 1975, 144-5).
When a §ong is placed in front of the epistle in this way there can be no doubt that it functioned as a prelude to the reading of the lesson. But nearly all songs in these sources lackxubrics.to make their purpose plain. Even in the 'Codex Calixtinus' the songs are copied separately in supplements to the main body of liturgical material. Fortunately, there is,ample evidence ..in jhe Circumcision offices of Sens (Villetard 1907)waii,d,.^auyj^s.,.(Arlt 1970) as to the use of this type of material. (The story continues in the Le Puy office, text ed. Chevalier 1894; see also Arlt 1978.) Thus in the Sens office, Benedicamus songs, as one would expect, conclude first Vespers, Compline, Laiids, Prime, Terce, Sext, None, and second Vespers. There are no songs; for thejessons of Ma but there is a 'Conductus ad subdiaconum' before the epistleat,JVlass (which has the Laudes deo farse), and a 'Qond,u^tus ad evangelium' before the gospel. Neither^song includes rit.ualswords or contains an invitation to begin the reading, but their function of accdmpanying the approach to the lectern of the subdeacon and deacon respectively seems clear. TheJBeauyais office makes even more co|oj^feJj?royisJQn by having cojiduxtus. for each lesson of Matins ,as well (table in Arlt 1970, Darstellungsband, 97), and one of these includes an invitation to the reader: Nostrifestigaudiurn, for the fifth lesson, ends 'Legatur in gaudio lectio' (Arlt
250
//. Chant Genres
1970, Editionsband, 59). After each lesson, aj.B£nedi,c.tioiis.sung—the equivalent of, or substitute for, Jube domne—which actually lifts a section out of some appropriate chant, usually a sequence. Going one better than Sens, the Beauvais office splits the song for the gospel intoJ:w3rjarJ:s, so that the sjejconci part is s^ugg while the deacon reFumsTo^TrTe altar area after the gospel (Arlt 1970, Editionsband, 114-16). The capacity of the Beauvais redactor to farse hisjiturgy seems unlimited.
Prime sources for liturgical songs in later centuries are to be found in eastern manuscripts siicri as the 'Mopshur.g .Gradual' (Spanke 1930; Stein 1956; Lipphardt 1957) the 'Se^kauCantip^ (Irtenkauf 1956), and the Prague 'cantionales' (Orel 1922; Orel",' Hornof, and Vosyka 1921; Nejedly 1954-6; Buzga 'Kantional (tsechisch)', MGG). (See also Jammers, 'Cantio', MGG and A'G; Stablein 1975, 75 n. 768 lists some two dozen sources, with bibliography.) The terms cantionale and cantio used of these sources and their contents should not obscure the fact that the songs are generically and stylistically identical to their twelfth-century predecessors; indeed, the manuscripts preserve many of the s^me^ompQ.sitions, prolonging their useful life down to the end of the Middle Ages. As in the early days, Benedicamus songs are easily the most common of those that refer to a liturgical formula (see, for example, the fourteenth-cenfury collection in Aosta, Seminario Maggiore 9. E. 19, discussed by Harrison 1965).
The Italian repertories of Latin laude (if that is the correct term) are also JiXlLsii£§^lY. related, again including ciinxioxd^ances with the northern repertories, although their function—as part of non-liturgical religious exercises—r,se£s, ..tESni apart. Most are "written inJsaUaita form, which is absent from the northern repertories, despite the predilection for refrain songs. (See Damilano 1963 on the most substantial collection, Turin, Biblioteca Nazionale Universitaria, F. I. 4, an antiphoner from Bobbio.)
The further .the style of such pieces moves from that of the earliest parts of the 'Gregorian' repertory, the more it has inspired writers to surmise links with secular music and vernacular repertories. Given the scanty records of secular music (excepting the special courtly repertories of troubadour and trouvere songs), the difficulties of defining styles which might be typically secular are considerable. It can safely be asserted, however, that with the^sixnilati.Qri„.of,.lJb\e.lweJfthr.ceD.tury spngjnto the festal liturgies of many prominent churches, the c.hant..rep.ertory achieved a colour and variety unimaginable in previous centuries and still a source of surprised joy to those fortunate enough to encounter it.
11.25. LITURGICAL DRAMAS
(i) Liturgy and Drama
(ii) The Quern queritis Dialogue
(iii) Easter Ceremonies from the Eleventh Century Onward
25. Liturgical Dramas
251
(iv) Christmas and Epiphany Ceremonies; Rachel's and Mary's Laments
(v) Rhymed Ceremonies; the 'Fleury Playbook'; the Lucius Danielis
E.A. Schuler 1951; Lipphardt, 'Liturgische Dramen', MGG; Stratman 1972; Stevens, 'Medieval Drama, II. Liturgical Drama', XG; Rankin 1981, 'Music'; Norton 1983; Norton 1987.
Editions: Coussemaker 1861; Young 1933 (all ceremonies then known, texts only); Donovan 1958 (texts from Spain); Lipphardt 1975-81 (texts of all known Easter ceremonies); Rankin 1981, 'Music' (north French and English ceremonies); Campbell and Davidson 1985 (facsimile of Orleans, Bibliotheque Municipale 201).
(i) Liturgy and Drama
A great part of the liturgy is meditational; the sequence of office hours, and the first part of mass, the Ante-communion or Mass of the Catechumens. Prayer and praise are joined to readings from sacred texts to inspire devotion and recollection of God's work in the world. The second part of mass, the Mass of the Faithful, is of a different character, a commemoration of the Last Supper in which priest and congregation perform again the simple and essential actions first performed by Christ and his disciples. At a time when communion for most of those present was rare (as it was throughout the Middle Ages), one may ask how much the mass was thought of as a commemorative ritual act and how much as a spectacle performed to inspire the devotion of non-participants. Whatever the case may be, the principle 'Do this in remembrance of me' could be extended to justify other actions. Thus at the procession on Palm Sunday, branches were carried and the people called 'Hosanna', just as when Christ entered Jerusalem (see Amalarius' remarks: Hanssens, ii. 58-9, cited by Hardison 1965, 111). The washing of the feet at the Mandatum ceremony is another example. Here particular events are not merely the subject of meditation and interpretation, but re-enacted.
A number of symbolic acts run parallel to these literal representations. The Deposition ceremonies, where the host or a cross was buried on Good Friday, to be brought up again on Easter Day, are symbolic re-enactments of Christ's burial and resurrection (see Young 1933, ch. 4). Hardison gives ample demonstration of the propensity of Amalarius (a few other writers follow him) to interpret elements of the liturgy symbolically, though it is unlikely that all were originally intended to be symbolic.
Amalarius' writings are one manifestation of a desire detectable in many aspects of ninth- and tenth-century liturgical practice to make the liturgy more vividly present and topical. The addition of trope verses to introits and other chants is another example, where previously neutral texts are explicitly keyed to particular themes. It is perhaps against this background that the addition of a new representational act, centred on the Quern queritis dialogue, should be understood.
In view of the modern connotations of the words 'drama' and 'play', I have generally avoided them, for the roughly equivalent medieval title Indus is rare.
252 //. Chant Genres
Medieval terms, where present, are usually officium or ordo. This may be translated as 'ceremony', a word which remains evocative of the liturgical context.
(ii) The Quern queritis Dialogue
Some time in the ninth century an adjunct to the Easter liturgy was composed, a simple dialogue between an angel and the Marys visiting Christ's tomb on Easter morning, the famous Quern queritis dialogue. About its origin and the way in which it was first performed we are ignorant. That it achieved wide distribution already in the tenth century suggests that it was composed in the ninth, but in this, as in its place of origin, scholars are in disagreement (from an extensive literature on the subject may be cited de Boor 1967; Rankin 1985, 'Quern queritis'; Davril 1986). In the tenth-century sources it could clearly occupy one of several different places in the liturgy. Sometimes it apparently preceded the Introit Resurrexi of Mass on Easter Day. Sometimes it was part of the procession before mass, which could conveniently make a station by a 'sepulchre' where the dialogue would be sung. Since most early sources lack specific rubrics, it is usually impossible to distinguish between this usage and the previous. Another place was at the end of the Night Office on Easter morning, following the last responsory (usually Dum transisset sabbatum, which appropriately sets the scene), and preceding the Te Deum. (On the various liturgical assignments, see McGee 1976, and Bjork 1980, 'Dissemination'.)
The dialogue appears in many forms, where extra verses usually precede and/or succeed a relatively stable core. (What constitutes the 'original' core is also the subject of controversy.) The version in Ex. 11.25.1 has one extra verse before (Hora est) and one after (Alleluia) the Quern queritis exchange. Other sources have Quern queritis and Iesum Xazarenurn a fourth lower, in D-mode (as in the next example, Ex. II.25.2). The next verses, Xon est hie and he nuntiate . . . sutrexit, are then in the same D-mode, but sound relatively higher, like an exultant climax. The word 'dieente' is not always present, for it simply provides a link to the next verse. It is worth noting the musical similarities between Quern queritis and Iesum Xazarenurn: both begin and end with sub-phrases cadencing F-G (C~D in transposition), while in between comes a sub-phrase which begins by outlining the triad F—a-c.
As a glance at the text edition in CT 3, 217-23 will show, the first verse Hora est psallite can be found not only in Italian but also in French, Catalonian, and German sources, though the distribution is thin. The distribution of Alleluia resurrexit dominus is equally wide, and a little less thin. Some sources have additional verses after it, some have the introit Resurrexi. In the Pistoia source used for the transcription (two Pistoia manuscripts are the only ones with this particular combination of verses) there is no unequivocal rubric. It seems that one group of singers, representing the angels, has taken up a position behind the altar (in the closely similar ceremony from Piacenza they are led by the cantor: see Young 1933, i. 216), while the schola, representing the Marys, sings in the choir. For the final verse all come together in the choir. Just when this ceremony would have been sung is not
25. Liturgical Dramas 253 Ex. 11.25.1. Quern quentis (Pistoia, Bibl. Cap. 121, fo. 33r)
%   • •       ' *' ~ '     **** ,   $  & •
Ho-ra est psal-li - te    iu-bet domnus ca - ne - re       e-     -ia    di - ci - te. Tunc dicunt .ii. aut .iii. stantes retro altare quasi interrogando.
Quern que-ri - tis in se - pul - chro chris-ti - co - le. R scola
n 9 -    ■ t »—*-
&   r« «  . *  *   «- #) »  ^ p   i* * **~~T» •
le-sum Na-za-re-num cru-ci - fi - xum    O   cae-li - co - le.
et dicunt ipsi retro altare
/«     *   *    *    *   .--—+
Non est hie   sur-re-xit   sic-ut pre-di - xe - rat
i-    -te nun-ti - a - te    qui-a  sur-re-xit di-cen-tes. et dicunt ipsi uenientes in choro
Al - le - !u - ia   al - le - lu - ia resur-re-xit do- -mi-nus.
clear: in Piacenza it followed Terce, which means at the point when the procession before mass would have been made. It seems likely therefore that these chants were the last to be sung before mass itself, and would have been followed by the introit. (See the more explicit rubrics from other Italian churches given by Young, i. 215—16 and 228-9, McGee, 12-13.) But in this case they do not lead directly into Resurrexi as a trope.
It must be admitted that only one of the many verses which were used to conclude the little ceremony clearly needs a succeeding text: this is Eia carissimi verba canite Christi which in two sources leads into Resurrexi, in another into the Te Deum, and in a fourth manuscript has no indication of the next chant. Traditional liturgical
254
//. Chant Genres
antiphons (their text, that is, if not always their music) are sometimes included in the scene. A group of sources from the Rhine-Mosel area, and Winchester, have Surrexit dofninus de sepulchro (CIO 5079), some Swiss and south German ones have Surrexit enim sicut dixit (5081), Winchester has Cito euntes (1813) and Yenite et videte (5352). Another verse is clearly not sung by the Marys, for it sets the scene as well, being a literal setting of Mark 16: 3: 'Et dicebant ad invicem: Quis revolvet nobis lapidem ab ostio monumenti? Alleluia, alleluia.' It is in fact another office antiphon (CAO 2697) drafted into a new setting.
Much more explicitly representational than the Pistoia version and the other Italian ones just mentioned is the version known from late tenth-century England. The complete text with notation is preserved in the two tropers from Winchester, Cambridge, Corpus Christi College 473 and Oxford, Bodleian Library, Bodley 775, and a full description of the ceremony survives in two copies of the Regularis Concordia, guide-lines for English monastic life drawn up in the wake of a council held at Winchester r.973. This all-important description says specifically that the ceremony is performed 'in imitation of the angel seated on the tomb and of the women coming with perfumes to anoint the body of Jesus' (Regularis Concordia, ed. Symons 1953, 50). The monk representing the angel has a symbolic palm in his hand, the three brethren who are the Marys come 'as though searching for something'.
One should not over-emphasize the naturalism of the scene. The monks wear liturgical vestments and carry thuribles instead of real ointments. The ceremony is not a historical reconstruction of past events, but a vivid demonstration to the community that Christ is risen, there and then. When the scene is over Te Deum is sung, with the bells of the church ringing the while. What is intensely moving about this starkly simple scene has nothing to do with the character identification which is the essence of modern drama, but comes from the distillation of the Easter miracle into a few brief utterances. Realism is marginal, would even have been an impoverishment, for the more specific the representation, the more the symbolic overtones are lost. Nothing in the biblical narrative suggests that the angel holds a palm, but it brings to mind Jesus' triumphal entry into Jerusalem a week earlier, binds the scene to the ever-present sacred history. To dress the monks in women's garments would make them totally foreign to the religious community. When the angel sings 'Venite et videte locum' he is not merely enunciating the biblical text, but actually singing a liturgical antiphon with that text, making the ceremony part of the eternal liturgical cycle.
W7e are exceptionally fortunate in having so early a description of the ceremony, for earlier copies of the text and music (and many later ones) are quite unspecific in this respect. The question of whether the scene was acted more or less realistically from the beginning is, however, beside the point, for not even the most specific descriptions, such as the Winchester one, aim for realism.
Sources of the tenth century, and many later ones, do not go beyond the core dialogue, usually with a few accompanying verses. Already during this early period the verses were adapted for other seasons, Christmas (Quern queritis in presepe—
25. Liturgical Dramas 255
mainly French and North Italian sources) and Ascension (Quern creditis super astra ascendisse—mainly French sources).
(iii) Easter Ceremonies from the Eleventh Century Onward
From the later eleventh century onward the simple dialogue was often enlarged with extra material, while a few sources arrange new episodes from the Easter story. The latter tend to differ considerably from each other in text and music, which provides fascinating material for the study of interrelationships between the traditions and of differing concepts of representational ceremonies. Instead of attempting to mention each source of new material, I have chosen to describe briefly the reasonably typical contents of just one manuscript: Madrid, Biblioteca Nacional V. 20-4, one of the Norman-Sicilian manuscripts of the twelfth century.
Few sources have quite so much material altogether as this source. For the Night Office of Easter morning there is the traditional scene for the Marys and the angel, much expanded from the tenth-century versions (Ex. 11.25.2). At the start each Mary sings a short lamenting verse, which leads into a setting of a more extensive biblical extract than was earlier the case. The core dialogue Quern queritis is not literally biblical. The text here takes the narrative in Mark 16: 1-7, which was also the Gospel at Mass on Easter Day, as its starting point. Some is set without alteration, some
paraphrased:
Verse Mark 16
Quis revolvet 3
Ecce lapis revolutus cf. 4-5
Nolite timere cf. 6
Quern queritis —
Hiesum Nazarenum —
Non est hie 6
Venite et videte cf. 6
Ecce locus 6
Ite dicite 7
. . . precedet vos 7
Ibi eum videbitis 7
From then on the text is new composed, though the Marys repeat the words of the angel as they go to tell the good news.
The way in which the expansion of the older dialogue has been effected is sometimes all too clearly visible. The traditional antiphon Venite et videte already paraphrases the angel's invitation to see the place where Jesus had lain. Ecce locus, the biblical text unaltered, is almost superfluous, though we may imagine a pause between the two as the Marys approach the place more closely. (The lament at the start ends with an incipit for the Quis revolvet verse, as if the scribe was copying from two different exemplars and did not amalgamate them wholly successfully. I have omitted the incipit in the example.)
256 //. Chant Genres
Ex. II.25.2. Visitatio sepulchn (Madrid, Bibl. Nac. V. 20-4, fo. 102v)
In die resurrectionis domini uersus mulierum Priora
*«« «. ^ &  «3 * «=
- •** t—9-•*---
He - u mi-se-re cur conti-git ui-de-re mortem sal -ua - to - ris. Secunda
- *
He-u redempti-o Is-rahel ut quid tali-ter a-ge-re uo-lu-it. Tertia
«3  ^ . *  * *   . * 0 =
V ' ö   ^ *        •  ^  *   • ** -
He-u     con-so-la-ti - o  nostra ut quid mortem sustinu-it.
Insimul omnes            Uersus Marie mulierum
'* i*     *    ^ , , *        *3 # 0      * * + ^ $
O De-us      Quis re-uol-uet no-bis   la-pi-dem ab hosti-o mo-rru-men-ti.
Ec-ce la-pis re-uo-lu-tus et in-uenis sto-la candi-da co - o - per-tus. Angelus
• *
-_--^>-m-l-
P  'i   a iTs    ^r- *   .  "   •   a  *• ' **—r
No-li - te ti - me-re uos di - ci -te        quern que - ri - tis ad se-pulchrum christi-co-le
Mulieres
Hiesum Na-za - re - num  cru-ci - fi-xum que-ri - mus   O ce-li-co-le. Angelus
/   '    ^ \).'*•.*'   *J   . *   ' *   *' fi   0 ^   * &
Non est hic sur-re-xit si-cut predi-xe-rat ue - ni - te et ui - de - te locum    u - bi
po-si-tus    e-rat   do-minus.      Ec-ce   lo-cus  u-bi     po - su - e-runt e-um.
&   *   « ^ #    í'   '        f  '  • ^* '•    0  _ ^> -f   0 0^
I - te di-ci-te     dis-ci-pu-lis eius   quia  sur-re-xit de   se-pulchro     et ec-ce
. ' 0 .    . . • 0 ^ ? * • v*        * 0 0
prece-det uos in Ga-li-le-am i-bi e-um ui - de-bi-tis sic- -ut di - xit uo-bis. Mulieres
, i* * *> f' '0. .   . . *'* * P '   ^ & .      . .t. ^ r> ,
E-a - mus nun-ti - a - re   mirum quod ui - di-mus   et gaudium quod ac - ce-pi-mus
25. Liturgical Dramas 257
1
sur-re-xit ue-re sic-ut di-xit do-minus    pre-cedet uos in Ga-li-le-am  al - le - lu-ia
P
-j------«—m-'-----•   * * •
i - bi     e-um  ui - de-bitis     al - le-lu - ia al-le-lu-ia     al-le- -lu-ia. Chorus_Episcopus
De-o gra-ti-as      al-le-lu-ia al-le-lu-ia al-le-    -lu-ia. Te De-um lau-damus.
A few pages later in the manuscript further parts of the Easter story are represented. These are the encounter of two disciples with the risen Christ on the road to Emmaeus, the Teregrinus' (pilgrim) episode, so called because the two do not recognize Christ at first, believing him to be a pilgrim (Luke 24: 13-32). Although Christ disappears at the end of the scene, he then returns in order to show the disciples his wounds (Luke 24: 36—9). The next episode in the manuscript is Mary Magdalene's encounter with Christ (John 20: 11-18). Finally, doubting Thomas is portrayed (John 20: 24-9).
The distribution in other manuscripts of settings of these episodes is somewhat patchy. Peregrinus plays are fairly widespread. Mary Magdalene episodes textually and musically related to this one are rarely found outside north French and English sources; similar scenes are present in sources from Germany and from Bohemia hut were conceived quite independently. A scene with Thomas is given in only four other manuscripts: Tours, Bibliotheque Municipale 927 (from north France or England), Paris, Bibliotheque Nationale, n.a.l. 1064 (from Beauvais), Orleans, Bibliotheque Municipale 201 (the 'Fleury Playbook'), and Munich, Bayerische Staatsbibliothek, elm 4660a ('Carmina Burana'); in the Beauvais version the text is versified. Naturally there is a common recourse in these sources to dialogue drawn directly from the gospels. What tends to separate them is (a) the non-biblical material they include, (b) the musical setting, and (c) the casting of the text into verse.
There is no space here for a full demonstration of all these points. Fortunately, the Mary Magdalene episode has been lucidly analysed by Rankin (1981, 'Mary'). We have already seen that liturgical antiphons may be used in these scenes, understandable enough when so much of the biblical text was set in liturgical chants of this season (one needs only to browse through the Easter section of, say, the Antiphonale monasticum for this to become apparent; see Brockett 1977-8.) What is particularly interesting is that, as Rankin shows, some versions of the scene retain the traditional melodies for the antiphons thus drafted into use, while others replace them with newly composed music (Rankin 1981, 'Mary', Table 2 on p. 252).
The incorporation of pre-existing material raises questions about the musical
258 //. Chant Genres
consistency of the composition. The Thomas scene, as given in Madrid V. 20-4 (Ex. II.25.3), includes four liturgical antiphons:
	MI	CAO	mode
Vidimus dominum			D
Nisi videro			D
Data est michi	465	2099	G
Pax vobis! 0 Thoma			D
Misi digitum	480	3782	G
Quia vidisti me	479	4513	G
Surrexit dominus de sepulchro		5079	E
As mentioned above, the last item had previously been used in the simple dialogue scene. A conflict between it and the other traditional antiphons is understandable, but why were A7.v/ videro and Pax vobis not also made in G-mode?
Ex. II.25.3. Thomas scene (Madrid, Bibl. Nac. V. 20-4, fol. 107v)
Thomas autem non erat cum illis decern discipulis qui sunt in medio choro sed ueniens ex auerso ad illos decern discipulos et stabit. Qui surgentes dicant ei tribus uicibus:
i
Vi-di-mus do-mi-num. Thomas respondit:
I---«#,#-'000    I-,    _^ 0-
Ni-si   ui-de-ro   in manibus   e-ius    fi-xuram   cla - uo-rum   et    mittam  digitum meum
»   » 0.   0   * »
v 0       v   • &
in lo-cum cla-uo-rum        et mittam manum meam in la-tus e - ius non credam al-lelu-ia.
Tunc ueniat Ihesus et appareat omnibus discipulis dicens:
0 «tr - - - #    #->     0    0    m    a     0 -mr
-•- *    0     *0     B     0—0-0-a-*-0 -
Da-ta est mi-chi omnis po-testas in ce-lo et in ter-ra al-le-lu-ia al-le-lu-ia. Item dicat solummodo:
Pax uobis.        O Tho-ma   in-fer di -gi-turn tu-um hue   et     ui - de manus meas
1
_ *  0 ■      0  • 0  «3 ,  00. 0>      s  .   , jSi
p--*-*-m*—B0-41-"-*-0-***—0-0-»0
et   pedes meos      et af-fer manum tu-am   et mitte in   la-tus  meum     et no-li
25. Liturgical Dramas
259
1
—#
es - se in-credu-lus sed fi-de-lis al - le - lu - ia. Thomas uertat uultum suum ad populum dicat:
i
P       '       •       • m-
~9-*-
Mi-si     di - gi-tum meum in  fi-xu-ram cla-uorum   et manum me-um in  la-tus e-ius
* * '* '* fj *  t 4 » * f      ' • *
et di-xi       do-mi-nus me-us  et De-us me-us al-le-lu-ia. 'ribus uicibus dicat Dominus meus adorans. Et hoc facto dicat Ihesus Thome:
																					
I			m	m															m		
		t			—a—	•   • *		■ *r								W				#	
	#					—•-—	—w—	—— 0—	*	-#—	—#-0—	—w—	ě								
Qui-a  ui-disti me Thomas cre-di - di - sti        be-a-ti qui non ui-derunt et cre-di-derunt
al -le - lu -ia.
Tunc omnes discipuli uertant se ad populum. Insimul dicant alta uoce:
Sur-re-xit  do-mi-nus   de  se-pul-chro   qui pro nobis pepen-dit in li-gno
•6 *
■»—»-
al - le - lu-ia     al - le - lu - ia       al - le - lu - ia.
Such inconsistency, if we are to regard it as such, may still persist when the scene is versified and set to new music. The Beauvais version has the following tonal structure:
Pax vobis
Videte manus meus
Surrexit dominus de sepulchro
Vere Thoma (rhymed)
Nisi fixuram (rhymed)
Pax vobis
Thoma nunc vulnera (rhymed) O Ihesu domine (rhymed) Quia vidisti me
AM (not 478)
CAO mode
478
479
5079
4254
4513
E E E D D F D D G
260
//. Chant Genres
Ex. II.25.4. Thomas scene (Paris, Bibl. Nat. n. a. 1. 1064, fo. 10v)
Tunc ueniat Thomas qui defuerat. Et stanti in medio dicant ei duo pro aliis:
~9
Ue-re Thoma     ui-di-mus do-minum
qui destru-xit     mortis im-pe-ri-um. Quibus Thomas:
Ni-si fi-xu-ram clauorum ui - de - ro et di-gi-to uulnus pal-pa-ue-ro ... at-que manum    in  la-tus mi-se-ro      hoc sci-a-tis    nunquam credi-de-ro.
Tunc in medio ueniens dominus dicat omnibus:
Pax uobis    e-go sum     al-le-lu-ia        no-li-te ti-mere al-le-iu-ia.
Deinde dicat Thome:
_2: f_
EE3
Thoma nunc uulnera      conspi - ce cor-po-ris et ostendat ei
In-fer et di-gi-tum in lo-cum uul-ne-ris et iam in-credu-lus       in me   ne  fu - e - ris
ex-emplum fi-de-i       prebendo pos-te-ris.
The rhymed verses are all of a piece: syllabic D-mode items framed by liturgical antiphons in different modes. The first two items are musically similar, particularly in their use of the figure FEDEFG and the cadence GaG-E (found only in a handful of relatively late liturgical antiphons). Neither of these features is found in the antiphon Surrexit dominus. Ex. II.25.4 gives part of the scene, continuing from Surrexit dominus de sepulchro, already given in Ex. II.25.2. (The whole is edited by Coussemaker and Rankin.)
Rhymed verse is also present in the Madrid manuscript. At the start of the Peregrinus scene two openings are provided, one if the play is to be sung at Vespers on Easter Day, the other for Vespers on Easter Monday (the evening hour is appropriate because of the biblical setting). The first opening begins with two strophes, one for each disciple, of a rhyming hymn with eight syllables per line (text in Young 1933, i. 477). Another important feature of the play is the use of the sequence Yictimae paschali laudes during the Mary Magdalene scene. The question to Mary which comes in the middle of the sequence (already given above, Ex. II.22.9) can be worked
25. Liturgical Dramas
261
into a meeting between Mary and the disciples. The disciples sing the first verses of the sequence. Mary comes along and announces Christ's resurrection. The question-and-answer verses of the sequences follow, and the disciples sing the final pair. This is but one of many ways in which Victimae paschali laudes was worked into the Easter ceremonies. Sometimes only the question-and-answer verses were used; sometimes the sequence was made part of a scene with the three Marys.
In many eastern sources the basis for expansion away from the simple dialogue was rather different. The dialogue was recast textually and musically. Norton (1983, 1987) has identified the core of the new tradition as seven, instead of three verses, as follows:
Old tradition:
Quern queritis in sepulchro Iesum Nazarenum ... O celicole Non est hie surrexit
New eastern tradition: Quis revolvet nobis ab hostio lapidem Quern queritis O tremule mulieres Iesum Nazarenum . . . querimus Non est hie quern queritis Ad monumentům venimus Currebant duo simul (CAO 2081) Cernitis O socii
The new scene involves new action, for the angel's words Non est hie, telling the Marys to tell the good news to the disciples, are immediately acted upon. Ad monumentům is the Marys' announcement. Currebant duo tells how Peter and John run to the tomb. It is usually they who sing Cernitis O socii, an invitation to look at the grave-clothes and the empty tomb. Expansions of this core could include Victimae paschali laudes, increased dialogue for the disciples, and so on. The usual sequel is a chant of joy for the soloists or the schola, and often the Te Deum. Many later sources direct the vernacular hymn Christ ist erstanden to be sung.
The new music centres around E for the central verses, with Quis revolvet in A-mode and the last three verses in D. The transcription in Ex. II.25.5 is from one of several sources which transpose the central verses up a fourth to a (with 6bs). After the usual Dum transisset, the last responsory of the Night Office, a procession is to be made to a sepulchre, while the choir sings the antiphon Maria Magdalena. The seven core verses follow, then verses from Victimae paschali (Die nobis is repeated as a choral refrain). Christ ist erstanden and the Te Deum conclude the ceremony. Although the liturgical antiphon Currebant duo simul is in indirect speech, it is allotted here to Peter and John.
Of the author of the new, German, form of the dialogue, we are as ignorant as in the case of the old Quern queritis. De Boor (1967) has summarized the main traditions, of which Norton (1983) sees Augsburg as perhaps the source. Among the earliest versions is one from Aquileia (Udine, Biblioteca Arcivescovile 234, a gradual of the late eleventh century), which corresponds very closely with later ones from Augsburg. This can only be explained by Aquileia's having received the ceremony from Augsburg in the first place, during the period when the influence of the German
O   *s -m m
262 //. Chant Genres
Ex. II.25.5. Visitatio sepulchri (Antiphonale Pataviense, fo. 55r)
f Mulieres____
Quis revolvet no-bisab os-ti-o la-pi-dem quem tegere sanctum cernimus sepulchrum. Angelus__
]> . •? r* . a • • ' »_•__*» *—»—
Quem queri-tis   O tremu-le mu-li-e-res: in hoc tu-mu-lo gementes. Mulieres
Ie - sus Na -za - re -num cru-ci - fi - xum que - ri - mus. Angelus
Non est hic quem queri-tis    sed ci - to   e - un-tes nun-ci -a - te   dis-ci - pu - lis  e - ius
•—•-»
et Petro: qui-a sur - re - xit Ie-sus. Mulieres
i
F—»—»- •«^••^#-
Ad mo-nu-mentum ve - ni - mus gementes:   an-ge-lum do-mini   se-dentem vi-di-mus
3Ě
« ' *-* -« ,
et di - centem: qui-a sur-re-xit Ie-sus. Petrus et Ioannes
—,—s—^
Cur-rebant du-o   si-mul:   et il-le  a-li-us   dis-ci   -  pu - lus pre-cu-currit ci-ti-us
ft  !«L    $  *  '   r* *   * *   #'  * S-    *   **   * *
Pe-tro        et ve-nit pri-or   ad mo-numentum al-le-lu-ia.
Petrus et Ioannes cum sudario_______
*    '    '   "   **   f   *       ^ *   *    *"*    *    *   +>    *   »'   ***** ř
Cer-ni-tis   O   so - ci - i:       ec-ce  lin-the-a - mi-na    et   su - da- -ri-um:
—»- ° —*•—•—^ 4*—*——•-#—r -
et     cor-pus non est  in    se-pulchro in - ventum.
Chorus
Die no-bis Ma-ri-a     quid vi-dis-ti   in vi - a.
25. Liturgical Dramas 263
Maria_Chorus_
' v * *  ' i* ,     L_±_-t.....fi   . =
Sepulchrum Christi vi - ventis:     et  glo-ri - a   vi-di   re - sur-gentis. Die no-bis...
Maria Chorus
*   '   '    . *   '   «    ' T^
An-ge-li-cos testes:     su - da-ri-um et vestes. Die no-bis..
Maria__
m *       •        *       m       • Ck
Surrexit Christus spes mea:     pre-ce-det su-os in Ga-li-le-am. Chorus_Deinde populus
9 j ,
Credendum est... Scimus Christum... Christ ist erstanden... Postea
Te Deum laudamus.
empire upon the Italian church was strong, probably in this case when the Augsburg canon Heinrich was patriarch of Aquileia (1077-84). Its later rapid diffusion can then be followed through the various ecclesiastical reform movements in the German church of the twelfth century.
(iv) Christmas and Epiphany Ceremonies; Rachel's and Mary's Laments
The adaptation of the Quern queritis dialogue for Christmas is found in relatively few sources, usually with extra verses and some indication of how the ceremony was to be performed. In manuscripts of the thirteenth and fourteenth centuries from Rouen a much fuller enactment of the Christmas story is to be found, luckily also preserved with music in the thirteenth-century Rouen gradual Paris, Bibliotheque Nationale, lat. 904. (Christmas texts are edited by Young, ii, ch. XVII; the music of the Rouen ceremony and the closely related one in the so-called Fleury Playbook are edited by Coussemaker 1861.)
Although some texts of the Rouen Christmas ceremony coincide with liturgical items, chants do not appear to have been quoted. Given the date of the manuscript, it is not surprising to find some rhymed items, and there are signs that these are later additions to an older version of the ceremony. An angel announces Christ's birth to the shepherds, and a choir of seven boys 'standing in a high place' sings 'Gloria in excelsis deo' (but not to its usual liturgical antiphon melody). The shepherds then sing the versus Pax in terris nunciatur (facsimile Young, ii, opposite p. 16), a rhymed song with two strophes and the refrain 'Eya! Eya! Transeamus, uideamus . . .'. This is immediately followed by a chant with the biblical text Transeamus usque Bethleem, et videamus . . . , in more traditional chant style. The text is superfluous after the song just sung; perhaps it was pushed into second place when the new song was introduced.
264
//. Chant Genres
When the shepherds reach the manger two midwives ask them 'Quern queritis in presepe pastores dicite?', and when they have been shown the mother and child they sing a further short song, the two-strophe Salve virgo singularis. Curiously, although the whole of the rest of the ceremony is in G-mode, this little piece is copied in F, with Bb, perhaps another sign of late insertion.
Most interesting are the directions at the end of the ceremony and during the subsequent mass (which begins with a D-mode introit), which make it clear that the shepherds lead the choir, singing solo verses where proper. At the end another dialogue-chant is performed: the priest or bishop sings the verse Quern vidistis pastores to the-shepherds and they answer with the verse Natum vidimus. At the end of Lauds they also sing the Benedicamus song Verbum patris hodie.
The ceremonial nature of the shepherds' play, with its reminiscences of a liturgical procession and close integration with the subsequent mass, is also characteristic of the officium stellae ('ceremony of the star') in its Rouen version. It re-enacts the coming of the Magi to Herod and to Bethlehem, following the star. But this ceremony appears to have had quite different origins. Most of the early sources (which date from the first half of the eleventh century) are not liturgical books, and the ceremony contains relatively little of the biblical and liturgical material which one finds in Easter ceremonies, aiming rather to provide 'learned' new material in hexameters (or what might be called 'pseudo-hexameters', for they often offend against classical convention). The impression is of a ceremony conceived on an independent and original basis which different institutions inserted more or less loosely into the Epiphany liturgy. Thus some copies end with the cue Te Deum, some have no liturgical cue at all.
The various versions of the ceremony (there are over twenty) have been compared by Lipphardt (1963) and Drumbl (1981, 293-340). Most are either notated with adiastematic neumes or lack notation, but five copies with staff notation survive (for editions see Coussemaker 1861; Lipphardt 1963; Bernard 1965, 'Officium'). All the versions are closely related in content, though often they not only differ in small details of text and music but may also omit scenes from what appears to have been an original conceived on a grand scale. Thus the early source from Compiegne (Paris, Bibliotheque Nationale, lat. 16819) has parts not only for the three kings and Herod, but legates, a messenger and servant, scribes, and a knight, all in attendance on Herod, as well as the women at the manger and an angel. The action comprises numerous scenes: the Magi alone, their summons by messengers from Herod, their interview with Herod (which includes prophecy of Christ's kingship by the scribes), their arrival at the stable, the angel's warning, Herod's command to kill the children, and a final antiphon sung by the angel, Sinite parvulos ('Suffer the little children to come unto me . . .').
The Rouen redactor, on the other hand, evidently wished to integrate the ceremony as fully as possible into the liturgy. The processional element which is present to some extent in all versions (the three kings visiting various 'stations') is here strengthened by the inclusion of processional antiphons drawn from the liturgy. The occasion is the
25. Liturgical Dramas
265
usual time for a procession between Terce and Mass, on the feast of the Epiphany. The first king appears behind the high altar, the others on either side. After the first chants, a procession is formed 'as on Sundays', in other words with the usual complement of choir, assistants, and so on. The cantor begins the processional antiphon Magi veniunt V. Cum natus esset. The procession makes its way out into the nave to the altar of the Holy Cross, where a curtain conceals an image of the Blessed Virgin. A crown is suspended before the altar 'in modo stelle'. When the midwives have drawn back the curtain the kings pray and feign sleep, whereupon the boy, representing the angel, sings the warning to them to return another way. This instruction is indeed carried out literally, for the procession now returns to the choir via the font and the left entrance into the choir, again according to usual custom of the cathedral. During it, another processional antiphon is sung, Ttia sunt munera V. Salutis nostre auctorem, once again intoned by the cantor. Mass follows, with the kings leading the choir and singing solo verses where required.
Another Magi ceremony, from Limoges (Young, ii. 34), is placed after the offertory of mass, an obviously symbolic arrangement.
The Slaughter of the Innocents, traditionally commemorated on 28 December, is referred to in the Compiegne Epiphany ceremony and others, and at Laon forms a complete scene, but is one of many items apparently pruned for the special requirements of the Rouen ceremony. At Laon (unfortunately the manuscript has no notation) a moving lament for Rachel, symbolizing the mother of the children, is included, in dialogue with a 'Consolatrix'. In other sources another lament of Rachel has survived as an adjunct to the responsory Sub altare del, and there are two fullblown Innocents' plays as well, both entitled 'Ordo Rachelis'. The simpler lament (in Paris, Bibliotheque Nationale, lat. 1139) is musically unadventurous; the part of the angel is in the metre which became popular in rhymed offices and was also used for some rhymed sequences and songs (Ex. II.25.6).
Although Bernard (1965, 'Officium') has argued for a certain amount of musical differentiation between the characters in the Epiphany ceremony, this cannot be regarded as a common feature of liturgical dramas. The melismatic exclamations of the weeping Rachel found in the Fleury Playbook version are, however, obvious attempts to impress the tragedy on the attention of the listener (Stevens, 'Medieval Drama', NO, 34; Coussemaker 1861, 170). This lament draws upon one of Notker's sequence texts, Quid tu virgo mater, and, though to a lesser extent, its melody as well.
A much bigger repertory of laments exists largely outside the liturgy and is generally subsumed under the Latin termplanctus. As well as many secular planctus, Latin and vernacular (see the catalogue by Yearley 1981 and editions in Yearley 1983), there are a considerable number of laments of the Virgin Mary at the foot of the cross (German 'Marienklagen'). Although the circumstances of their performance are not always clear, it seems that such laments were often sung, at least in Germany and north Italy, beneath the crucifix on Good Friday. An appropriate liturgical place was after the reproaches during the Adoration of the Cross. Two early compositions of this sort,
266
//. Chatu Genres
Ex. II.25.6. Lantentatio Rachelis (Paris, Bibl. Nat. lat. 1139, fo. 32v) Lamentatio Rachelis
O     dul - ces fi-    -li- -i quos nunc pro - ge-     -nu - i.
O - lim die- -tu       ma - ter quod   no - men  te-     -nu - i.
O - lim per pig - no - ra uo - cor     pu - er-     -pe - ra.
c Mo - do sum mi   -  se - ra na - to - rum   ui-    -du - a.
He-u    mi -  chi mi   -  se - re cum pos - sim    ui-    -ue - re.
Cum   na - tus co - ram me in    De - o      per-    -de - re.
At - que la- -ce  -   ra - re pa-rum   de - trun - ca - re.
He - ro - des im - pi   - re fu - ro - re      re - pie - tus.
Ni - mi - urn su - per - bus per - dit    me - us      par-tus.
i
Angelus
5fe
3
9
No-li   Ra-chel defle-re pig-no-ra
Cur tris-ta-ris et tundis pec-to-ra
No-li   fle-re sed gaude po-ti-us
Cu-i   na - ti ui-uunt fe - Ii - ci-us. Er - go     gau- -de.
Summi  patris e-ter-ni     fi - li-us
Hie est il-le quern querit per-de-re
Qui uos fa-cit e-ter-ne ui - ue-re. Er - go     gau- -de.
which were often quoted in later laments, are Planctus ante nescia and Flete fideles animae (texts Young, i. 496 ff.; music Dobson and Harrison 1979, 84, 238; Gennrich 1932, 147; Stevens 1986, 131; Vecchi 1954, 56). Both are written in sequence form with paired strophes, but the length of the strophes and their varied length and rhythm is more reminiscent of the secular lai than the liturgical sequence. The Marienklage might be so extensive as to incorporate liturgical chants or vernacular songs.
Christ's passion was itself made the subject of dramatic representations, and it is not surprising to find laments being worked into the longer texts. For example, the Cividale Passion Play (Cividale, Museo Archeologico Nazionale CI, a fifteenth-century processional, music edited Coussemaker 1861, 285-97) borrows heavily from Flete fideles animae (the passages are identified by Young i. 512 n. 2). The manuscript is remarkable for its copious acting instructions, where almost every line is accompanied by a significant gesture.
(v) Rhymed Ceremonies; the Fleury Play book; the Ludus Danielis
Rhymed songs have already been referred to several times. In dramatic presentations they first appear in considerable numbers in an Aquitanian manuscript of the early twelfth century, Paris, Bibliotheque Nationale, lat. 1139. Benedicamus songs and
25. Liturgical Dramas
267
similar pieces from the manuscript have been cited in 11.24. The dramas are the following:
The Lamentatio (Ex. II.25.6), preceded by a responsory for the Holy Innocents, seems to have a liturgical place, and we may assume that the Quern queritis dialogue is sung on one of the usual occasions. The others remain unplaceable. The unique Sponsus play, where the story of the Wise and Foolish Virgins is enacted, is most remarkable. Its strophes are all of identical construction and are sung to one of four melodies (which do not, however, attach to specific characters). This, the presence of the vernacular, and the unusual subject matter, raise the possibility of non-liturgical origin. Perhaps it stands near the beginning of the tradition of the vernacular miracle play, played outside church. The music cannot, however, be distinguished from the repertory of Latin liturgical songs, of which this manuscript is one of the chief witnesses.
Whatever the relationship of Sponsus to the liturgy, rhymed verse and the vernacular can be found in several indubitably liturgical ceremonies. For example, the Easter ceremony in a fourteenth-century manuscript from the convent of Origny-Sainte-Benoite (Saint-Quentin, Bibliotheque Municipale 86), is composed predominantly of regular rhymed strophes, French as well as Latin (Coussemaker 1861, 256-79). For the two episodes where French is used (the episodes of the Marys buying spices from merchants, and Mary Magdalene's dialogue with an angel) the music settles into constantly repeated strophes. Such repetition, as in the Sponsus play, seems therefore to be associated with the vernacular and non-biblical material. That is not invariably the case, however. The Easter play in the manuscript Tours, Bibliotheque Municipale 927 (from north France or England, thirteenth century) is composed almost entirely in Latin verse (see Coussemaker 1861, 21-48; Krieg 1956). But through composition of the music is rather the rule than the exception. The metre of the text is irregular, or when regular does not remain constant for long; the line-ends are often assonant rather than rhyming, which suggests a date of composition earlier rather than later in the twelfth century. Ex. II.25.7, the passage where disconsolate soldiers admit to Pilate that Christ is risen from the tomb, shows how different the music is from the plays in unchanging strophes. The music shapes the scene in that two of Pilate's utterances, 'Vos Romani' and 'Legem non habuistis', have the same melody; and the soldiers' 'Pro quo gentiles' and 'Nos veritatem' also form a pair. Pilate's final strophes are in iambic trimeter.
One source particularly rich in versified Latin dramas, the famous Carmina Burana manuscript (Munich, Bayerische Staatsbibliothek, elm 4660a), has only adiastematic notation (see the texts edited in Young, i. 432, 463, 514, 518, ii. 172, 463; Steer 1983
fo. 32v Lamentatio Rachelis
fo. 53r Hoc est de mulieribus
fo. 53r Sponsus
fo. 55v [Ordo prophetarum]
Latin strophes Quern queritis dialogue Latin and Occitan strophes Latin strophes
268 //. Chant Genres
Ex. II.25.7. From the Tours Easter play (Tours, Bibl. Mun. 927, fo. 4')
Tunc milites surgant et redeant ad Pilatum tristi animo canendo
-*b—*—t>-
i»f 0j
**4*   ^   *   m--------"-----=--' *
\ mi-se-ri     quid fa-ci-mus     quid   di -ci-mus     qui - a    per-di-dimus
He-
- r . c 9 0
quern custo - di - mus.
De ce-lo ue-nit    an - gelus
qui di-xit mu-li - e - ri-bus
-'    *    '    «    0    *i   ^ 0
qui-a   sur-re-xit  do - mi-nus.
Deinde dicat Pilatus ad milites
~~9 _^ * *
Uos Roma-ni mi-li-tes
"   •   •   *        * *
pre-cium ac-ci-pi-te
et  om -ni-bus di - ci - te
i
:        M   4 ~ qp—* #;p»
quod uobis sub-la-turn est.
Milites simul respondeant
2
■   ■» *
# »0
9-*-*-9
Pro quo genti-les fu-i-mus     se-pulchrum custo-di-ui-mus     magnum sonum audi-ui-mus
--^*—^—*—s—
et in terram ce - ci - di-mus. Item dicat Pilatus
*   ^      ^ #   * •
Legem non ha-bu-istis
—•-v—»   ^ m—0-- 0
sed menti-ri po-tes-tis     quod disci-pu-li ue-ne-runt
^ ,  0 <
et e-um sustu-lerunt. Milites simul respondent
« 1
•   0%    0 *=3r
- -~*-9-9 •----•
Nos ue - ri-ta-tem di-cimus     de ce-lo ue-nit   an-ge-lus    qui di-xit mu-li - e - ri-bus
-----*0-—»--«-
qui-a su - re-xit   do - mi-nus.
25. Liturgical Dramas
269
Hoc audito Pilatus dicat militibus hos uersus 1a
_-x^f.---V-fc-p-b-^-
---f\-
*»#    *     .. M    *   m    f    ,    a *•»-
^        *-•---#» ^------
Hec er-go uo-lo ut sint uestra mu-ne-ra ne uos creda - tis a - liqua mendati-a Ad domos uestras      i - te nunc cum gaudi-o        et que ui-dis-tis   te-gi-te si-lenti - o
9     **    •    ^   » *
£
que  uos se-ducant       et pe - ri - re fa - ci-ant.
ne    ad   au-ditum        po-pu-li   e-ue-ni-at. Milites simul respondeant ad Pilatum
Tunc erit.
argues for provenance in South Tirol; facs. Bischoff, 1967). Even richer, however, is the equally famous Fleury Playbook (Orleans, Bibliotheque Municipale 201). The manuscript was kept for many years in the library of the celebrated Benedictine monastery of Saint-Benoit-sur-Loire, at Fleury, but there is no evidence that it originated there. The chief music scribe used an informal type of notation (similar to that, for example, in the Cambridge song-book Cambridge, University Library Ff. 1. 17 or Tours 927) which is practically impossible to localize, but which was doubtless used by many practising musicians of the time. (See the discussions by Corbin 1953 and Huglo 1985 'Analyse'; editions of all ten plays by Coussemaker 1861).
Some of the contents seem non-liturgical: four plays about miracles of St Nicholas, which make wide use of repeating musical strophes. At least for the Tres clerici play, where the same melody is used throughout the entire composition, one is tempted to suggest a connection with the method usually postulated for the performance of secular vernacular epics and romances (see Stevens 1986, 222ff.). The same technique is used for a play about another miracle, the resurrection of Lazarus. By contrast, the Conversion of St Paul, though written throughout in lines of 4 + 3 + 3 syllables, is basically through-composed, with only infrequent repetitions.
The remaining plays mix traditional and apparently new material. There is an Epiphany play, which begins with the episode of the shepherds coming to the stable, an Innocents play, an Easter play, and a Peregrinus play. These are among the most extensive ceremonies for their respective occasions. Three end with the Te Deum, Peregrinus with Salve festa dies. Rhyme, where present, is not invariably associated with regular rhythm. One of the few regular pieces is a small processional song in Peregrinus, in sequence form with lines of 7 + 5 syllables, to be sung between the first appearance of the risen Jesus to his disciples and the scene with doubting Thomas (Ex. 11.25.8). (The processional function of the song is even more clear in the Easter ceremony from Coutances: Young i. 408.)
Examples like this and the song which the shepherds sing at Rouen while processing to the stable fit well with our picture of the Latin liturgical songs of this period known as conductus (see above, 11.24). They are relatively rare in liturgical dramas, however, and only one play makes systematic use of them. This is the famous
270
//. Chant Genres
Ex. II.25.8. Song Adam norus veterem from Flcury Percgrinus play (Orleans, Bibl. Mun. 201, p. 228)
3
•i ---------
A-dam no - uus ue - te-rem Cre-a - to - rem re-co-lit
du-xit ad as-tra. iam ere - a - tu - ra.
i
Sed Ma - ri - a la - co - bi Et   Ma - ri - a   Sa - lo - me
cum Magda - le-na. ferunt un-guenta.
Qui-bus  di - xit an-ge-lus Re-sur-re - xit do-mi-nus
in ues-te al - ba. mor-te  cal - ca -ta.
3
Frac-ta linquens Tar-ta-ra Re-fert secum   spo-li - a
et spo - li - a - ta. uic-tor ad as-tra.
^3=
Se de-monstrat pos-te Di - lectis   dis-ci - pu-
- a lis
for-ma pre-cla - ra. in   Ga-li - le - a.
—'-»-*
Comes factus in-ere-pat Et scriptu - ra   re - se - rat
+7 *-'    •** •
la-tens in ui - a. pi - us   ar - cha - na.
pro-pri - a for - ma. lu - mi - na   cla - ra.
3
i
Con-ui-uans ag-nos - ci-tur Pa-nis   red-dit frac-ti-o
Si - bi  laus   et glo
n -
a.
Ludus Danielis of Beauvais, surviving in an early thirteenth-century manuscript (London, British Library, Egerton 2615), alongside a practically complete office for the Circumcision or 'Feast of Fools' on New Year's Day (see 1.9). In the first half of the piece nearly every entrance or exit of characters is accompanied by vigorously rhythmic conductus (so called in the source). The melodies of some can be found elsewhere: thus lubdemus regi nostra, sung as the sacred vessels from the temple are brought in to Belshazzar, turns up as Iubilemus cordis voce in the Laon Epiphany office; and the song which the nobles sing as they escort Daniel to the king, Hie vents Det famulus, has the same melody as the Benedicamus song Postquam celorum dominus in Paris 1139, a key repository of both festal songs and liturgical dramas. Hie vents Dei famulus is notable not only for its macaronic text but also musically for its
25. Liturgical Dramas 271 Ex. II.25.9. From the Lud'us Danielis (London, Brit. Lib. Egerton 2615, fo. 99r)
(Principes) _•_C_,
-*—*»
—«G—•-
*   Hie  ue-rus De - i     fa - mu-lus        quern laudat  om - nis po - pu-lus *-.a-.   «i    «-«-*—*X
Cu - ius   fa - ma   pru-den-ti - e est  no - ta   re - gis cu - ri - e
Cestui man-da li rois par nos. Daniel
	-"j      *      *** im
JL ------±*—-	-^-*-ft*-f*-V-
to-* ♦—	_4ffM---
-per  et   ex - u-     -lans        en-uois al   roi   par uos.
tertian framework and the delaying of the tonic c until the very end of the song (Ex. II.25.9).
Of the other elements mentioned up until now, there is little quotation of traditional chant (two Lenten responsories, Merito hec patimur and Emendemus in melius, when Darius' envious counsellors are about to be flung to the lions), and, although there are strophic songs other than the conductus, there is nothing to resemble the repetition technique of Sponsus and the miracle plays in the Fleury Playbook. The text is mostly rhymed with regular rhythm, set in music of admirable directness. Whereas much of the Tours and Fleury music moves at the leisurely pace of traditional chant, Daniel often slips into syllabic, patterned phrases. Consider, for example, the song of the envious counsellors as they inform Darius that Daniel is worshipping his own god (Ex. 11.25.10). In the first line there are three descending melodic thirds (bracketed); and F occurs three times on stressed syllables; the same happens in the third line, with three Gs. In Darius' reply the thirds recur.
This clearly rhythmic, patterned music is complemented by a few items in a less insistent style. When the counsellors have had their way, and Daniel has been condemned to the lions' den, he laments as shown in Ex. 11.25.11. The lines rhyme but their metre is obscured by the 'Heu' exclamations and by deviations from syllabic setting. Ignoring the exclamations, we have lines of verses of 5 + 9, 5 + 9, 7 + 8, and again 5 + 8 syllables. The sudden change into the very highest register is rhetorically inspired. So are the swoops down through the complete octave cc to c in lines 1 and 7-8 (running over the line-end), and perhaps the descent to a and bb for 'damnatio'. The compartmentalization of phrases into either the lower or the upper part of the octave (only c and g cadences are used) is typical of twelfth- and thirteenth-century pieces.
272
//. Chant Genres
Ex. 11.25.10. From the Lucius Danielis (London, Brit. Lib. Egerton 2615, fo. 105s)
(Consiliarii)
Nunquid Dari Qui  o - ra-re
-—*—«—
obser-ua-ri uel ro-ga-re
sta-tu-is quicquam a nu-mi
-0 - #
ti om-nibus nibus
Ni te de-um
il-lum re-um
-•—-*—w  * ^
da-remus le-o-nibus
m
Hoc e - dictum (Rex)
sic in-dictum     fu-it   a prin-ci-pibus.
—"-•-S-•-*-»--m—
Ve-re ius-si   me om-nibus
ad-o-ra-ri   a gen-tibus.
Ex. 11.25.11. From the Lucius Danielis (London, Brit. Lib. Egerton 2615, fo. 106r)
I
*0      'o -
1.   He-u       he-u       he-u       quo ca-     -su   sor - tis
2.   ue-nit   hoc dampna-ti - o mortis.
0      V*' ~**0     0_■
i
3.   He-u        he-u      he-u        sce-ius    in - fan-dum
4. cur   me da-bit ad   la - ce - randum.
^5.   Hec fera  tur-ba   fe - ris 6.   sic  me rexperde-re queris.
n —^»    0 —■ —^ 0
7.   He - u       qua morte     mo-ri     8.    me   co - gis    parce fu-ro-ri.
26. Offices with Verse Texts 273
The Lucius Danielis is a unique and, to modern sensibilities, particularly happy mix of styles. No other work of the period moves with such directness and energy. Even in its time it must have seemed extravagant, not least musically. Little of the liturgical spirit informs it, although it ends with Te Deum. It stands at the opposite end of the scale from the Peregrinus ceremonies, with their copious use of biblical and liturgical material. It seems to have had no successors, and indeed, the composition of new dramatic liturgical ceremonies of any kind fell off rapidly after the thirteenth century.
11.26. OFFICES WITH VERSE TEXTS
Jammers 1929-30; Irtenkauf, 'Reimoffizium', MGG; Szoverffy 1964-5 (s.v. 'Reimof-fizien'); Andrew Hughes, 'Rhymed Office', .\G, and 1985. Editions of texts: AH 5, 13, 17-18, 24-6, 28, 45
Editions of music: Feldcr 1901; Auda 1923; Bayart 1926; Jammers 1934; Ottosson 1959; Falvy 1968; Hoppin 1968; Bernard 1977 and 1980; Epstein 1978; Dcleglise 1983; Patier 1986; Edwards 1990.
The subject of the following remarks is the offices for saints with antiphons and responsories written in verse form. The topic is somewhat difficult to define and name. We are not dealing with a new liturgical genre, for the compositions in question form a normal part of the office; only their poetic technique distinguishes them from other antiphons and responsories. Nor are we dealing with a new type of music. While it is true that the chants of versified offices are usually arranged in modal order, offices with prose texts can also be found whose chants are in modal order. Other musical features of later medieval versified offices, discussed briefly below, are not exclusive to them, but can be found in other chants of the same period. Nor does the usual term 'rhymed office' seem adequate to contain the whole area, for rhyme is only one aspect of poetic technique which can be found in the offices. It is certainly true that the vast majority of versified offices were written from the twelfth century onward in accentual rhyming verse, but many earlier offices used rhyme more freely or not at all—one finds assonance, single rather than double rhyme, or rhymed prose; some texts are metrical rather than accentual. Thus the increasing and finally exclusive use of accentual rhyming texts is a matter of stylistic change, not of a new liturgical or musical form. It runs parallel to the use of accentual rhyming verse in conductus (which was a new genre), alleluias, sequences, and liturgical dramas.
The term historia is used for the rhymed as for the non-rhymed office.
The size of the repertory is very large, but rather few offices were widely known. The reasons are obvious. The most important saints, commemorated throughout Europe, already had offices with old prose texts. Saints of lesser, local importance might use chants from the Common of Saints, but many composers were inspired to compose a new office for the patron saint of the their church or diocese, using modern
274
//. Chant Genres
poetic techniques. The majority of offices with verse texts are therefore local demonstrations of piety. Only a few became widely known, such as those for St Thomas of Canterbury, St Francis, and St Dominic (the latter two naturally promulgated by their respective religious orders), which stimulated contrafacta for other saints, that is, the same music was utilized for new texts. Jammers (1929-30) surveyed ninety-nine offices in sources from the Rhine area. Of these seventeen were widely or fairly widely known outside that area, eight were promulgated by one of the religious orders, twelve were known in a restricted area, incuding at least part of the Rhineland, while the remaining sixty-two were sung in only a handful of churches or even just one. These figures seem typical of the general picture of distribution, as indeed they would be for many types of relatively new liturgical material.
One or two isolated metrical office antiphons and responsories, such as the antiphon Solve iubente Deo from the office for St Peter (text in hexameters, AM 993, AS 412; may be pre-Carolingian. Other metrical or rhyming or otherwise poetic antiphons and responsories are occasionally to be found among saints' offices of the ninth century. Hucbald of Saint-Amand (d. 930) is credited on indifferent authority with a number of texts (AH 13, 133 ff., 200 ff., 225 ff.; AH 26, 230 ff.). He appears to have a better claim to the antiphons of a St Peter office which has survived in transcribable notation (see Weakland 1959). This office is not versified. Another example, exceptional both because it was not composed in honour of a local saint and because it was adopted all over Europe, is the Trinity office composed by Bishop Stephen of Liege (901-20). (Parts of the Trinity office are in modern service-books, such as AM 535 ff.; facsimiles of a medieval one AS 286 ff.; a number of these early texts are discussed and edited by Jonsson 1968; there are facsimiles of eleventh-centurv metrical offices in Delaporte 1957 and 1959-60.)
It remains to be seen to what extent the music of these early offices differs from that of the rest of the repertory. Even at this early date the pieces may be arranged in modal order, as, for example, in the Trinity office:
modes
Vespers antiphons 1-5 Night Office antiphons 1-8, 7
Night Office responsories 1-8, 1
Lauds antiphons (which all have verses) 1-5
Different arrangements were naturally required for Vespers and the Night Office (but not for Lauds) in the monastic form of the office, which required four Vespers antiphons and twelve Night Office antiphons and responsories. (See Andrew Hughes 1983 for a number of modal schemes.)
Another relatively well known example of a saint's office with chants in modal order but not in verse is that for St Nicholas, said to have been composed by Reginold of Eichstätt before he became bishop of the diocese (966-91) (see Charles Jones 1963; Hohler 1967; AS 354 ff.).
Chants in rhyming accentual verse are occasionally to be found in offices of the
26. Offices with Verse Texts 275
tenth century, becoming more frequent in the eleventh. An early example is Gloriosa sanctissimi for St Gregory, attributed to Leo IX (1048-54), the former Bishop Bruno of Toul (fac. Bernard 1977). Ex. 11.26.1 gives the second responsory of this office, which is naturally in mode 2. Although the verse has a psalm verse for its text, it uses only the first half of the normal responsory verse tone. The respond has assonant rather than regularly rhyming lines, and their rhythm is varied.
Ex. 11.26.1. Responsorv Videns Rome vir heat us (Annccy, privately owned manuscript, fo. 139r)
R.    Vi-dens   Ro-       -me   uir be - a-       -tus        Anglo-     -rum for-te   pu - e-ros
be-ne     in-quid      be-ne    An-gli uul-tu   nitent ut     an- -ge - li.
O-   -portet   il- -lis monstra-ri i-   -ter  sa - lu- -tis
e- -ter- -ne.
-9* • *
V.   Quo-   -ni-am   do-mi-ni    est      re- -gnum
et   ip-se     do - mi-na-bi-tur   gen-        -ti-     -urn. O - por-tet...
The music of the respond, by contrast with that of the verse, owes practically nothing to the melodic idioms of mode 2 responsories. Among other un-Gregorian features one could pick out the two 'Gallican' cadences ('pueros', 'angeli') and the final cadence. The figure at 'angeli' is convincing enough in its function as a cadence, but it can be found internally as well, in the first line. Two turns as at 'beatus' can be found in the traditional repertory, but not three as at 'illis'. Then there are the C—E—G triads, at 'Anglorum' and twice at 'iter salutis'. The fact that all but two lines cadence on D is also typical for this period, so that the phrases tend to become preoccupied with the tonal space above and below D: compare 'bene inquid' and 'vultu nitent ut', 'bene Angli' and 'angeli'. The tendency to repetition may extend over line-ends, so that 'vir beatus Anglorum forte pueros' is similar to 'illis monstrari iter salutis e-'.
276
//. Chant Genres
Ex. II.26.2. Antiphon Gregorius rigiliis (Annecv, privately owned manuscript, fo. 140')
i
p—^
Gre - go - ri - us   ui - gi - li - is
con-fectus et   ie - iu - ni - is
-*r •** *'
* * *
et     si    mar-ce-bat   cor-    -po-re       ex spe ui - ge - bat    a - ru - me.   Seculorum amen.
Equally untraditional is the antiphon Gregorius vigiliis, the first of the Lauds antiphons but also in mode 2 (Ex. II.26.2). There are two 'Gallican' cadences in the first line alone, and two later ones. Even 'ex spe vigebat' in the last line sounds likely to close this way (cf. 'Gregorius'). It is not surprising that phrases which explore the same restricted area should sound alike: line 1 is practically identical to line 4. (For a comparison with other offices attributed to Leo, see Bernard 1980, with further facsimiles.)
The restriction of melodic goals continues as a feature of the offices in completely regular accentual verse of the later twelfth century onward. Ex. II.26.3 gives the mode 1 antiphons from the office of St Dominic: the antiphon for first Vespers, the first and last antiphons for the Night Office and the first of Lauds. All cadences except one are on D, a, or d. What is more, the starting notes are almost as uniform, with two-thirds starting on a, and the rest on C, D, E, and c, with one G. Nearly all movement is by step rather than by leap. There is no repetition of notes, one of the biggest differences between these and the oldest Gregorian antiphons. All is activity, though on a small scale. There is no sense of longer periods, so that in the first antiphon, for instance, the music divides naturally after the first four syllables of each line, even though there is no text rhyme. The difference between phrases therefore tends to be a matter of how far in which direction, and how fast, the melody will go, between rather rigidly observed poles and in rather uniform motion. Surprisingly few lines are closely similar (compare 'dotata gloria' in Gaude felix and the last line of Adest dies leticie). Indeed, there seems to be a deliberate desire to avoid repetition (though the room for manoeuvre is limited), when the constant recourse to the same few text patterns would have made repetition easy. Modern writers have nevertheless been tempted to speak of 'formulas' (Jammers) or 'variations' (Bernard). Compare, for example, the third antiphon of the Night Office for St Dominic with the same one for St Thomas of Canterbury, given in Ex. II.26.4.
Both antiphons make the usual opening progression through D-G-a—c, but whereas Documentis artium then takes another line to make a cadence on b, Cidtor agri has already done so and comes swirling down (without leaps) to E again. That is what Documentis artium also does in line 3, while Cidtor agri rises once again to c.
Deliberate musical rhyme may occasionally be found. Thus the St James office in the Codex Calixtinus includes some pieces in rhymed prose where musical rhyme
26. Offices with Verse Texts 277 Ex. II.26.3. Antiphons from the office of St Dominic (Rome, Santa Sabina XIV lit. 1)
f.296r
r»£* »
0*0^
Gau-de   fe-   -lix      pa-rens Ys-pa-ni-a        no - ue pro - le    dans mundo gau-di-a
i
4*
a »2 m
4? ™ %
do - ta-ta glo - ri -a
sed tu ma-gis
plau-de Bo-no-ni-a
tan - ti patns
M        ^   # r  -v ,## ,,j ^   ^ =
lau-
-da to-ta
ma-ter ec-cle  - si-a      no - ue lau - dis     a - gens sol-lem-ni-a.
f.296r
Pre - co   no - uus   et  ce-li - cus
mis - sus in fi - ne se - cu - li
pauper ful-sit   Do - mi-ni-cus
for-ma pre-ui-sus ca-tu-li.
f.297r
Li-ber car-nis uin-cu-lo
ce-lum in-tro - i - uit
P
- —007-0-
u - bi   pie - no po-cu-lo
»—r
gustat quod si-ti-uit.
f.297r
J) ff» ' fiL _ ,= V     9 **   * 9—*--0-
^~»0T
Ad-est  di - es     le - ti - ci - e
quo be - a - tus   Do - mi -ni - cus
P
—s —
-0*0-
au - lam    ce - les - tis   cu - ri - e
ci-uis   in - trat   mag-ni - fi - cus.
278
II. Charit Genres
Ex. II.26.4. Antiphons from the offices of St Dominic (Rome, Santa Sabina XIV lit. 1, fo. 296r) and St Thomas of Canterbury (Edinburgh, Univ. Lib. 123, fo. 156')
Dominic
0 £\
Do-cu - mentis    ar - ti - urn
e - ru-di-tus sa - tis
*—*j
trans-i - it   ad   stu - di - urn
summe     ue - ri - ta-tis.
Thomas	
A.                                     C*   "m    f* m	-J**—*—,---
Cm     af) •»•*••	*m Ci'm
w—'%** «;---	-*          *'»--•-
Cul - tor    a - gri do-mi-ni         tri - bu-los   a- -uel-lit
	
/            'o   m   m i-fl	
	
vT7 - —# •-	•      *       *     *.      •*» a
	
et uul-pes a  ui  - ne-is ar-cet et   ex-pel-lit.
supports or counterpoints the text rhyme. For example, the responsory Jacobe virgineifrater (ed. Wagner 1931, 73) has the following musical rhyme scheme (small letters indicate an identical cadence, capitals a whole identical line):
	text rhyme	musical :
Jacobe virginei frater	a	a
preciose Johannis	b	B
qui pius Ermogenem	c	a
revocasti corde feroccm	c	B
ex mundi viciis	d	a
ad honorem cunctipotentis	d	B
V. Tu preec continua	e	C
pro nobis omnibus ora	c	D
(Qui pius, etc.)		
Gloria patři almo	f	C
natoque flamini sancto	f	D
(Ad honorem, etc.)		
The shortness of breath evident in the antiphons just discussed, or perhaps better the concentration on short, self-contained lines, extends also to responsories. Since they are more ornate, in the sense that single syllables often have melismas, it frequently transpires that a self-contained musical phrase may cover no more than a single word, or even a single syllable (see Stablein 1975, 163). Ex. II.26.5 gives the fourth responsory of the monastic office of St Thomas. While one might argue for a single musical phrase encompassing both 'Post sex' and 'annos', others seem relatively self-contained: 'redit', 'vir', and 'stabilis', and obviously 'thesaurum'. These are each
27. Metre, Accent, Rhythm, and Rhyme
279
Ex. 11.26.5. Responsory Post sex	annos (Edinburgh, Univ. Lib. 123, fo. 156')
	
frrz---^ k\~- at^v "1 I*\-	--*r""S--jß\ mm -
^-7—^n.'v-	-f'm^ i.-/-*•••—■ -
^* «• • v* *	
R. Post sex an- -nos       re-        -dit    uir sta- -bi - lis
-e-1»—--						
f\         —     y» **d		-* *—■				
(A\               »V   *a   a "»a	a					
		•			•a •»	
V » da - re ter-	re	tes-	-te	uas	* m 'mm fra-	-gi-lis.
						
A          •       1 a				_ _ ___		
fm   *          • »*i			m) im	*# • > **i a v		
v VJ                                 * '		9		• a • • «0 ^	f 99mm	
					9 0	• •
Christo ua- -sis        thesau- -rum   fic- -ti-lis.
■y	
A             .______#_ä_'—-_	
	
	
V. Ne sit  lu - pis pre-da grex hu-mi - lis
						
■Jr                                                                                                                    • *•* >						
					^—^	a »
	—			• • •	•a     • *•	
se	pro	gre - ge		dat	*»    • •    a _J» « pas-tor no-bi-lis.	Christo...
						
/			—			
fm ^»			•ö    _-> i*	a -	j—    ^« * -—^	a a
			•jr^ a-	-P-*	—ft '* '* —*0a mU	
Glo-ri - a   pa-tri   et   fi — Ii — o     et  spi - ri- -tu - i     sancto. Christo...
comparable with lines of antiphons, but are now largely melismatic rather than broken up by changes of syllable.
Although the rhythmic schemes employed in these texts are the same as in rhymed sequences and Latin songs, there is no opportunity here for stressed syllables to dominate the tonality, as they do in syllabic or near-syllabic pieces.
11.27. METRE, ACCENT, RHYTHM, AND RHYME IN
LITURGICAL TEXTS
(i) Metre and Stress
(ii) Rhyme
(iii) Prose Rhythm (Cursus) and Prose Rhyme
Raby 1927; Gavel 1954; Beare 1957; Strecker 1957; Crocker 1958, 'Musica'; Norberg 1958; Oxford Classical Dictionary; Steiner, 'Cursus', NG; Fassler 1987.
' = stressed syllable . = unstressed syllable — = long syllable " = short syllable / ~ caesura
280
//. Chant Genres
(i) Metre and Stress
Most liturgical texts are prose, taken from (or composed to be compatible with) biblical prose texts such as the Book of Psalms. Verse has, nevertheless, an important place in the liturgy, and any survey of plainchant must of necessity refer to technical aspects of Latin verse: metre, stress or accent, rhythm, rhyme, verse-forms, and so on. I offer here a very brief introduction to some of these matters, bearing in mind that specialists themselves are divided over many issues referred to, such as the nature of accent and ictus, and the nature and progress of the change from metrical to accentual verse.
A distinction has to be borne in mind between metre (which concerns syllable lengths) and stress. In classical Latin the two were independent. Although there is disagreement as to the nature of stress (or accent) in Latin at various periods of the classical age, it is at least certain that it played no part in the organization of classical poetry. Later, organization of poetry according to patterns of stressed syllables became more common, and poems are found in which stress and metre are coordinated, so that all stressed syllables might also all be long ones. Organization by stress was occasionally so dominant that syllable-lengths in their turn were ignored. In discussions about the texts of liturgical songs of the twelfth and thirteenth centuries there has been a certain amount of controversy about whether a stressed syllable should necessarily be counted a long one. Metre and stress are obviously of importance for the transcription and performance of these songs.
In classical Latin stressed syllables fall: (a) on the first syllable of two-syllable words; (b) in longer words on the penultimate syllable if it is long, otherwise on the antepenultimate.
Classical poetry was organized not according to stress but by syllable-length, where groups of long and short syllables were reckoned in 'feet'. The feet encountered most frequently in liturgical poetry are the following:
trochee   —"   anapaest spondee —
iamb dactyl       —tribrach
It was common in classical times to accompany the recitation of poetic texts, in private or in teaching, with rhythmic movements, of the hand or a wand. Each foot was reckoned to have two parts: the arsis ('raising') and thesis ('setting down'), referring to upward and downward movement of the hand or stick. Marking of the feet in this way was referred to as marking the 'beat' (ictus, percussio, or plausus). Grammarians and authors of treatises on prosody do not agree as to how it was done, some calling each movement, arsis or thesis, an ictus, some assigning one 'beat' to the up + down combination, and some taking a pair of feet in one beat. There is also disagreement as to how the ictus was marked, if at all, in the declamation of the texts. And it is not known if the beats were spaced at regular intervals, so that, for example, a trochee took up the same time in performance as a spondee, perhaps in this way:
27. Metre, Accent, Rhythm, and Rhyme 281
trochee = spondee dotted crotchet + quaver = crotchet + crotchet
In classical Latin, ictus and stress were independent of each other, and neither seem to have been strongly heard in performance. (Beare 1957, 57-65 summarizes the problems and disagreement over ictus.)
The most common classical metres in liturgical poetry are the following (for others see Norberg 1958, ch. V):
(a) dactylic hexameters, where the first four feet are either dactyls or spondees, the fifth is a dactyl, and the last is a spondee (occasionally a trochee):
(or)--------
(b) the elegiac distich (distich = two verses, i.e. a couplet), where a hexameter is followed by a dactylic pentameter. The pentameter is in two halves: the first half has two feet, either dactyls or spondees, plus one long syllable; the second half has two dactyls and one further syllable:
—
(or) - - - -
(c) the sapphic stanza, whose last line is called an adonius:
1-3 trochee, spondee, dactyl, trochee, trochee or spondee
4 dactyl, trochee or spondee (the best-known example is probably the hymn Ut queant laxis, attributed to Paul the Deacon).
(d) the asclepiad, in its catalectic (wanting a final syllable) or minor form:
spondee, dactyl, long syllable, dactyl, trochee or spondee, long syllable
(e) various trochaic and iambic trimeters (having three pairs of feet) or tetrameters (having four pairs of feet); a very popular one was the trochaic tetrameter catalectic, or trochaic septenarius, called the rhythm of the Roman legions' marching-songs:
(or)     - -      - -
Hexameters have a long history in liturgical poetry. They are even to be found among the prayer texts in the so-called Mone Masses (Gallican, ed. Mone 1850 and Mohlberg, Eizenhofer, and Siffrin 1958), though not in Roman formularies. Many trope texts of the ninth century onward are in hexameters, and so are some texts in traditional genres, for example the Marian introit Salve sancta parens.
Among lyric stanzas, one associated with the name of St Ambrose was far more popular than any classical model among composers of office hymns. This is in iambic dimeter (i.e. having two pairs of iambs; spondees are allowed for the first and third iambs). Ambrose's poetry is still metrical, rather than rhythmical: stress is not regular and does not coincide with long syllables:
' ' ' ' ' ( ;
— —     ^   —   _        —      ^ _ — _ w _ w _ _
Aeterne rerum conditor,     noctem diemque qui regis
282
//. Chant Genres
et temporum das tempora,   ut alleves fastidium.
The same lack of correspondence between long and stressed syllables can be seen in Venantius Fortunatus' Pange lingue gloriosi (trochaic tetrameter catalectic):
Pange lingua gloriosi proelium certaminis
et super crucis tropaeo die triumphum nobi lern
qualiter redemptor orbis immolatus vicerit.
During the late imperial age, spoken Latin appears to have lost gradually its long-and short-syllable contrasts, and stressed syllables, previously only lightly felt, gained in intensity. A conflict between metre and stress pattern, previously of little or no significance, would presumably have become more noticeable. It also meant that, although poets continued to write metrical verses for centuries after, they would have had to acquire knowledge of correct quantities at second hand, from teachers and textbooks, rather than through natural experience of the language. And it became possible to write poetry which ignored quantities and was organized according to stress-patterns instead—possible, but for centuries not generally thought proper in educated circles. Norberg (1958, 92) cites the case of a ninth-century monk of Saint-Amand who asks that if his hexameters are not worthy to be called a 'carmen', they may at least be granted the name of 'rithmus', that is, a poem in accentual rather than metrical verse.
Rather than use the adjective 'rhythmic' to describe poetry organized according to stress-patterns, I have preferred the term 'accentual'. The word 'rhythm' is often used loosely nowadays with a wide variety of meanings; and rhythm us in antiquity and the early Middle Ages had a precise and special meaning not directly connected to stress or accent. As Crocker (1958, 'Musica') explains, rhythmus was the proportional relationship between the lengths of sounds (e.g. syllables, or feet, or groups of feet), whereas metre was concerned with the lengths of individual sounds (syllables).
The period roughly from the time of St Augustine (354-430) through to the ninth century saw increasing numbers of compositions where stress coincided with the metrical accent or ictus. And in other poems the metrical patterns are completely replaced by stress-patterns. The poetic feet are now:
trochee', anapaest..' spondee', iamb    .'   dactvl    '..   tribrach '.
27. Metre, Accent, Rhythm, and Rhyme 283
There are examples of accentual texts in the sections on hymns and versus and conductus above (11.15 and 24; Norberg, ch. VI has very many more). One of these is the Christmas epistle farse which begins with the strophe Laudes deo dicarn per secula (various sources have Laudem and/or dicant), which is in the accentual equivalent of alcaic metre (Ex. 11.23.20), that is, it is the stress-pattern which reflects the alcaic metre, not the quantities.
Slight deviations from stress-patterns are common in accentual poetry. Norberg (1958) uses a descriptive code indicating simply how many syllables there are in a verse, and where the last stress falls. 'Oxytone' (on final syllable) stresses are unknown in Latin poetry, though some bear a moderately weak one. The stress is therefore 'paroxytone' (falling on the penultimate syllable) or 'proparoxytone' (on the antepenultimate). Laudes deo is therefore '5p + 6pp'.
One pattern popular from about the eleventh century is 4p + 6pp ('.'. '..'..)> widely used in sequences, versus, liturgical dramas, and rhymed offices. Charles Jones (1963, 122-39) thought it German in origin, on the evidence of the Hildesheim source of the St Nicholas miracle play of the Three Daughters (London, British Library, Add. 22514: Young, ii. 311-14), which he believed to date from the early eleventh century. Hohler (1967, 47) objected that the manuscript is more likely to date from the early twelfth century, which means it is no earlier than Paris, Bibliotheque Nationale, lat. 1139, where other examples of the rhythm are to be found.
Another popular medieval pattern was the so-called 'Goliard' rhythm, 7pp 4- 6p (trochaic), which became widely practised in the twelfth century. It is used at the start of a versus in Paris 1139 (fo. 46v):
i ii iii
Incomparabi 1 iter   cum iocunditate
but was much more popular in secular verse, at least in the usual 'Goliard' strophes of two such verses with double rhyme. Incomparabdiler is more significant in that, like very many twelfth-century songs, it employs a variety of patterns within the same song, patterns which may be found elsewhere, to be sure, but not in this particular combination. The idea seems to be to have four main beats in each line: there is a clear pattern of alternating strong and secondary beats, with weak beats sometimes added as anacruses or subtracted at cadences.
Incomparabiliter
cum iocunditate '
gaudeamus pariter '
in hoc sollempnitate .'
In festis beatorum .'
huius et aliorum. .'
(See Norberg 1958, ch. VI for a wide range of exampl
I referred above to a 'rithmus' as a poem in accentual rather than metrical verse. Actually, even this matter is disputed, for in a number of poems neither metrical feet
es of other patterns.)
284
II. Chant Genres
nor recognizable stress-patterns seem to be present. It has been suggested (see Suchier 1950; Crocker 1958, 'Musica', 9) that the basic idea was to control the number of syllables in each line, regardless of quantity or stress, except perhaps at cadences (end of line or caesura). The theory of a 'syllable-counting' or 'numerical' poetry circumvents the need to explain apparent errors or irregularities in early medieval prosody.
It is a moot point whether or not the metrical accents (the 'ictus') or the stress-accents were reflected in musical performance or notation. Tropes written in hexameters and other metres usually have music too ornate to allow any appreciation of the quantities of the text. In syllabic text-settings stressed syllables could in theory-be highlighted by stress in performance. And since stress-patterns replaced patterns of long and short syllables, one might argue that it would be illogical to posit a system of musical rhythm where stressed syllables took long notes and unstressed syllables took short ones. It is true that some late-medieval sources do notate hymns, sequences, and other poetic genres mensurally, with long and short notes (for example the Las Huelgas manuscript, ed. Angles 1931 and Gordon Anderson 1982; and Michaelbeuern Cart. 1; see Lipphardt 1980). Yet these may be late exceptions which prove the rule.
(ii) Rhyme
Rhyme also underwent a substantial change in the early Middle Ages. It was used in classical poetry only for special effect, became common among hymn writers in the fifth and sixth centuries, held its own during the Carolingian renaissance (despite its 'unclassical' character), and reached a brilliant apogee in the songs for the festal liturgies of the twelfth and thirteenth centuries.
'Rhyme' here should be understood in the broad sense, including assonance (similarity of vowels) as well as identity of whole syllables. Not until the ninth century did pure monosyllabic rhyme become at all common, and disyllabic rhyme is hardly found before the eleventh (the sequence Yictimae paschali laudes is an early example). Neither assonance nor rhyme is particularly common among the items in medieval hymn collections. Ambrose's hymns do not use it, while Venantius Fortunatus' do so occasionally (Pange lingua, cited above, has rather rare assonances, while Vexilla regisprodeunt is regularly assonant). A solis ortus cardine of Sedulius in the fifth century is worth quoting as an example of an abecedary hymn (each strophe begins with a different letter of the alphabet) with a rhyme which is consistently present though not in a consistent pattern:
Some other uses of rhyme are less well known. It was possibly after Sedulius' example that rhyme in hexameters was used after the fifth century. Hexameters of the
A solis ortus cardine adusque terrae limitem Christum canamus principem, natum Maria virgine.
Beatus auctor saeculi servile corpus induit, ut carne carnem liberans non pcrderet quod condidit. (etc.)
27. Metre, Accent, Rhythm, and Rhyme
285
Carolingian renaissance employ it rarely, but during the ninth century it became more common. In a hexameter, rhyme is effected between the end-syllable and the syllable at the caesura, that is, at the most prominent word division in the verse (usually after the first two and a half feet). Such a hexameter is called 'leonine' (after the twelfth-century Parisian poet/composer Leoninus, erroneously believed to have invented the technique). At Reichenau the following trope verse was sung with the introit of Holy Innocents' Day (CT 1, 146):
Nos pueri puero / resonemus carmina Christo (ex ore . . .)
and the following distichs on the Octave (CT 1, 137, 146):
Mentibus intentis / resonemus carmina laudis
et cum psalmista / dicamus voce sonora (ex ore . . . )
Nunc fratres enesis / repetamus verbula dulcis
quae psaltes regi / sic fert cantando perenni (ex ore . . . )
(iii) Prose Rhythm (Cursus) and Prose Rhyme
Classical orators made use of metrical patterns in prose. A sentence-ending (clausula) employed a recognizable and approved succession of long and short syllables. When accentual rather than metrical verse became common, the metrical sentence-endings were transformed into accentual ones. In effect it was always the two final words (the separation is indicated by I below) that were affected. The commonest patterns appear to have been:
Metrical Accentual 'cursus planus' I —w        '. ' .'.
'cursus tardus, durus'or'ecclesiasticus'   —^ I—'. ' 'cursus velox' (most popular) I —w—    '.. '
'cursus trispondaicus' — I w"—      '. '
(For rhythm in classical prose, see Norden 1923; in medieval prose Nicolau 1930 and Janson 1975; see also Beare 1957, 193-205.)
Another oratorical device was prose rhyme, where pairs of cola (phrases, separated by speech-pauses) had end-rhyme. (See Norden 1923, Anhang I; and especially Polheim 1925.) In the liturgy it is best known from the prayers of the Gallican and especially Mozarabic rites (see Porter 1958, 49).
The extent to which these are relevant to chant texts is doubtful. In a massive demonstration in PalMus 4 Mocquereau tried to show that the accentual endings of many chant texts were reflected in their musical settings. Although scholars seem agreed that cursus survived through the period from the late imperial age to the Carolingian renaissance, and was not revived from a moribund state by ninth-century writers, it is by no means clear that the texts of the older 'Gregorian' repertory deliberately cultivate cursus. Mocquereau's thesis was that accented syllables in sentence-endings displaying cursus were matched by relatively higher notes.
286
//. Chant Genres
However, as Apel argues (1958, 297-301), the evidence is not convincing, especially not in the examples which use the cadences of psalm tones, introit verse tones, and responsory verse tones; these are formulas which are applied mechanically, and would seem by their nature incapable of reflecting different cursus rhythms.
Bewerunge (1910-11) and Ferretti (1913) believed that metrical, not accentual, cursus patterns were involved. As Apel again argues, Bewerunge's evidence is inconclusive and shows no systematic attempt by chant composers to match syllable-lengths with a greater or lesser number of notes. Ferretti looks for cursus patterns in the note-groupings themselves, independently of the text, but his demonstration is also unconvincing.
It would seem to be more profitable to investigate the occurrence of rhythmic prose and rhyme prose in the new liturgical prose texts of the ninth century and later, particularly in sequences and tropes. Their relevance to sequences was suggested by Polheim (1925, 350), a reference taken up by Crocker (1958, 'Repertory', 163). Except for the obvious insistence on line-endings in a in many sequences, end-rhyme appears to have played little or no part in their composition until the eleventh century. On the other hand, there is often a significant degree of internal assonance and alliteration. Crocker investigated instances of cursus in Notker's texts (Crocker 1977), where the rhythm of cursus planus (though not always its usual word-division) is often apparent (also occasionally cursus velox).
Examples of assonance, alliteration, and cursus planus may be found in, for
example, Notker's Johannes jfesu Christo, where the five note cadence F-G—a—a-G
occurs at the end of verses 4, 5, and 7. Five times out of six this goes with text in
cursus planus:
iii i i i
filii dei   esse perenni    dedit custodem
pater revelat   semper commenda
Assonance and alliteration in verse 5 are underlined:
Te Christus in cruce triumphans matri suae dedit custodem, Ut virgo virginem servares atque curam suppeditares.
Crocker (1958, 'Repertory') made the suggestion that art-prose, with the repertory of ornaments (tropi) such as those just outlined, was a formative influence in the early sequence, a suggestion that had the advantage of treating the prose texts of sequences on their own terms rather than as derivatives of a musical form (prosulas). Crocker also pointed to the possibility of 'isocolon' (or 'antithesis') technique, that is, identical clause lengths, in the repertory of tropi at the writer's disposal; and he has repeatedly stressed that there are numerous irregularities in the double-versicle structure of early sequences. The effect of these observations is to narrow the gap between art-prose and early sequence texts. Despite this, the double-versicle arrangement seems too insistent for the sequence to be regarded as art-prose pure and simple. However, the resources of the latter were no doubt occasionally exploited in the creation of sequence texts.
Ill
Liturgical Books and Plainchant
Sources
111.1. INTRODUCTION
Wordsworth and Littlehales 1904; GamBer 1968; Andrew Hughes 1982; Vogel 1986; Huglo 1988, Livres.
The books in which the plainchant has been recorded since Carolingian times are diverse in nature and complex in content. This is chiefly because different books contain material for different parts of the liturgical round and for the use of different personages involved in the celebration. The possible combinations are quite numerous, and at different times and places a multitude of different selections of material have been made. Some important distinctions which should be borne in mind are:
1. between books containing texts and music to be recited or sung, and books of instructions as to how the liturgy is to be celebrated;
2. between books for the mass, books for the office, and books for other ceremonies;
3. between books for the priest or other officiant, books for the cantor, and books for other persons.
The chief concern of this section of the book is naturally with sources containing music, which usually means the cantor's books. But chants might also be recorded elsewhere. For example, the gradual, precisely because it is a cantor's book, does not usually contain the music for the priest's recitation of the canon of mass. And the chants for ceremonies performed by a bishop are usually to be found in a pontifical, although most are choir chants. Then again, chant-books might be combined in various ways with books containing texts. The later medieval missal and breviary are combinations of this sort, though not the only ones:
combined in the missal: gradual (chants of mass) sacramentary (prayers of mass) epistolary and evangeliary (lessons of mass)
288 ///. Liturgical Books and Plain chant Sources
combined in the breviary: antiphoner (office chants) psalter hymnal
collectar (office prayers)
homiliary, lectionary, passionary (office lessons)
Some knowledge of liturgical books other than those with only music is desirable because they may contain valuable information about the liturgical context in which the chant was sung, or indicate by text incipit the state of the chant repertory. Older than any chant-book are a number of ordines, or books of instruction about how the liturgy was performed. These often describe the part played by the choir in the liturgy during a period (the eighth and ninth centuries) when first-hand musical evidence is lacking. Many later medieval ordinals, the descendants of the earlier ordines, specify exactly which chants are to be sung on which occasions. This may be invaluable if we have few or no other books from the church where the ordinal was used.
It goes without saying that some knowledge of the liturgy, and particularly of how the material is disposed over the church year, is essential when dealing with liturgical books. The reverse is also to some extent true. There is no better way to discover the intricacies of the medieval liturgy and enter into its modus operandi than by learning to find one's way around medieval liturgical manuscripts.
Vogel (1986) is an excellent introduction to the study of early medieval liturgical sources other than music-books and office-books (sizeable lacunae), with exhaustive bibliographies. Its strength lies in the coverage of early ordines, sacramentaries, lectionaries, and pontificals, but it ignores chant sources almost completely. Nor does it describe how the material in the books is put together, that is, which piece follows which, how the seasons of the liturgical year are disposed, in short, what the newcomer to medieval manuscripts might find when he or she first encounters the sources themselves. A heroic attempt at this was made, however, by Andrew Hughes (1982). Experience shows that Hughes's book is most useful as a reference tool and support for advanced studies. It deals chiefly with later medieval missals and breviaries (the latter are undoubtedly among the most complex of all service-books), and is a mine of information about the liturgy. It gives a number of brief tables of contents of selected sources (390-408). Gamber's Codices liturgici contains basic bibliographical information about over 1,500 early liturgical sources. Huglo's recent survey (1988, Livres) supplements his numerous articles on individual types of book in NG.
The best way to learn how to find one's way about a particular medieval liturgical manuscript and to read it fluently is to have a modern edition of a similar book to hand. Even those medieval books of roughly the same type, age, and provenance are not exactly the same, but they do not vary inordinately from one another. With the aid of a properly indexed modern edition of a missal or a breviary it is usually possible to locate oneself on the right feast-day and acquire a working knowledge of the heavily abbreviated or otherwise incomprehensible rubrics or titles of formularies or
2. Ordines Rotnani
289
individual items. Breviaries and missals, since they also contain chants, can thus act as a guide for antiphoners and graduals. The most recommendable books for use as references in this way are (a) modern Roman/Solesmes service-books before the post-Vatican II reforms, particularly LU, GR, Aft, and AM; (b) such editions as Breviarium ad mum Saturn (ed. Procter and Wordsworth 1879-86) and The Saturn Missal (ed. Legg 1916). Naturally, the usefulness of the modern edition will vary according to how closely related is the medieval use being compared with it. Since modern editions of medieval chant-books in their entirety are practically non-existent, either the modern Roman/Solesmes editions or well-indexed facsimiles have to be used when dealing with the music itself.
There is general agreement, but no universal standard, on the nomenclature of medieval liturgical sources. Since so many books combine heterogeneous material, it seems pointless to try to be over-specific in finding the right label. For example, books commonly referred to as tropers could contain such a diversity of material (certainly not just tropes) that the title is almost meaningless, and reference to a short list of the contents of each individual 'troper' is essential. Only then can one see what such books have in common with each other and where they differ. (For suggested typologies and nomenclatures, see Fiala and Irtenkauf, 1963, also the surveys in Ehrensberger 1887 and Cabrol 1930, and the list in Andrew Hughes 1982, 119-20.)
In the following sections most of the discussion centres on music-books and the earliest liturgical books. The last chapter is devoted to tonaries, which are not liturgical in the sense that they do not record a portion of the liturgical round as it would have been performed. As musical sources they nevertheless aided the performance of chant, and rather than place them in the chapter on theory I have brought them in here. Books of hours, being for private use rather than public worship, are not discussed.
III.2. ORDINES ROMANI
Andrieu 1931-61; Vogel 1986, 135-224.
In Andrieu's authoritative edition there are fifty Ordines Romani. They are descriptions of how various parts of the Roman liturgy should be performed. Their contents reach back to the mid-seventh century and forward into the ninth. Although based on the liturgical practice of the city of Rome itself, all surviving copies are Frankish, and some of the ordines have been adapted to accommodate northern usages. At the same time as the Franks were acquiring books of prayers (sacramentaries) and other material from Rome, they also required books of instruction on ceremonial. WTe have reports of Frankish pilgrims and churchmen seeking out liturgical books in Rome from as early as the seventh century, a practice which became royal policy in the mid-eighth century (Vogel, 147). Careful analysis has revealed which ordines are liturgically compatible with which sacramentaries. Some reflect papal usage (the
290 Liturgical Books and Plainchant Sources
Lateran church), some the practice of other churches in or about Rome (such as St Peter's at the Vatican). The biggest groups describe, respectively: the celebration of mass (Ordines I-X), special rituals throughout the year such as processions, rogations, the ceremonies of Holy Week, etc. (Ordines XX-XXXIII), and ordinations (XXXIV-XL). The ordines vary greatly in length. One of the earliest, Ordo Romanus I (trans. Atchley 1905), gives a very full description of the papal eucharist in the early eighth century, which can be shown to have been known in Francia by about 750. Others are much shorter.
Although the Ordines Romani contain no music, and indeed do not often give more than text incipits for the few chants they may mention, they contain important information for the musicologist. They may reveal the various roles of soloists and choir in the performance of chants. They are especially valuable when put side by side with other witnesses to early Roman usage, for example, the writings of Amalarius of Metz, and can be compared with the much later Roman chant-books for liturgical peculiarities. This is the case, for example, with Ordo Romanus XII, a brief description of the office which mentions a number of chants by incipit. (Andrieu's edition points out the correspondences with Amalarius' description.) One important source of ordines, Brussels, Bibliotheque Royale 10127-10144, also has a copy of the chant texts of mass to be sung throughout the year, in other words a gradual without music. The contents of the manuscript, the famous 'Blandiniensis' (so-called because it belonged to St Peter's abbey on Mont-Blandin at Ghent) are as follows (see Andrieu, i. 91-6). the folio numbers give an idea of the length of the texts.
r-79v		canonical texts, extracts of correspondence of popes
79v		Ordo Romanus 13: list of the books of the bible to be read during
		the Night Office
80'-	82r	end of a treatise on the ecclesiastical computus
82v-	84r	Bede on the ordering of Easter ferias
84'-	85r	Ordo Romanus 26: ordering of services from the fifth Sunday of
		Lent to Holy Saturday
85'-	86r	Ordo Romanus 3: six short Roman and Prankish points of liturgical
		practice
86'-	88v	Ordo Romanus 14: list of readings for the office at St Peter's, Rome
88v-	89v	Ordo Romanus 30b: arrangement of the office from Thursday of
		Holy Week to the Saturday after Easter
90r-	115r	gradual
115"	-12P	ordo for initiating a sick catechumen
12P	-124r	blessings for holy objects and substances
125r	-135r	proper prayers for eleven masses
Some of the material in the Ordines was eventually taken up in the rubrics of other types of book, and from the twelfth to thirteenth century ordinals appear with much fuller listing of items to be performed at services throughout the year. Much of the ceremonial described in the Ordines Romani was to be performed by the pope, or at
3. Sacrcunentaries and I^ctionaries 291
least a bishop, and was therefore subsumed in the pontifical. Andrieu's Ordo Romanus L was actually copied side by side with the Romano-German pontifical of the 950s, which became standard in the Empire and in Rome itself. It is a somewhat heterogeneous miscellany of current and archaic Roman ceremonies, together with commentary, a collector's book rather than purely for liturgical use.
111.3. SACRAMENTARIES AND LECTION ARIES
(i) Sacramcntarics
(ii) Lectionaries
(i) Sacramentaries
Gamber 1958,  1968; Hope in Cheslyn Jones et al. 1978, 224-8; Dcshusses and Darragon 1982-3; Vogel 1986, 31-134.
The sacramentary or liber sacramentorum was the book containing the texts to be recited by the officiating priest (bishop, pope) at mass, for administering sacraments, for consecrations, ordinations, and other rites. It was, in effect, a book of prayers, although it takes its title from the rites represented in it. The chief prayers of mass during the central Middle Ages were the Canon, with its proper prefaces for different times during the year, and the collect, secret and postcommunion prayers, which were proper for each mass. Ordination formularies and blessings were also included in the sacramentary, but not lessons, choir chant texts or rubrics covering ceremonial.
During the early centuries prayers were probably mostly improvised and there is little evidence of any formal collections of texts. The first sign of these appears at the end of the fourth century and the beginning of the fifth. For example, Musaeus, a priest of Marseilles (d. r.460) is said to have compiled a sacramentary, together with a lectionary and a book of responsories (that is, their texts) (Gennadius, Liber de viris illustribus: PL 58, 1104; see Morin 1937; Gamber 1959; Vogel 1986, 302-3, 321; McKinnon 1987, Music, 170). Most early collections seem to have been no more than leaflets containing only a few masses. The so-called Leonine or Verona Sacramentary (in the manuscript Verona, Biblioteca Capitolare, LXXXV, of the early seventh century) appears to be a conglomeration of such leaflets, or libelli miss arum, compiled perhaps for reasons of private piety rather than public worship; its material is Roman, from the fifth and sixth centuries (ed. Mohlberg et al. 1956).
The earliest full sacramentaries of which there is clear evidence date back to the seventh century. A few of the more important Roman and Roman-Frankish books may be mentioned here, because of both their intrinsic importance and the light they shed on the adoption of Roman use by the Franks in the eighth century. What actually survives of Roman use is a group of sacramentaries based on Roman books but written in Francia and containing certain amounts of Gallican material. The classes or types into which they have been divided are (following Vogel's characterizations) the following.
292 III. Liturgical Books and Plainchant Sources
1. The 'Old Gelasian' sacramentary, based on a book reflecting the use of the titular churches of Rome of the period 628-715. It survives in Rome, Biblioteca Apostolica Vaticana, Reg. lat. 316 (plus Paris, Bibliotheque Nationale, lat. 7193), written at Chelles f .750 (ed. Mohlberg et al. 1960).
2. The 'Frankish' or 'eighth-century Gelasian' sacramentary. The several sources preserve various versions based on an archetype perhaps compiled at Flavigny in Burgundy, perhaps during the reign of Pippin III 'the Short' (751-68). The book combines Old Gelasian material, a papal sacramentary adapted for presbyteral use at the Vatican, and the usual local Gallican formularies. (The Sacramentary of Gellone, perhaps the most important of many Frankish-Gelasian sacramentaries, has been edited by Dumas and Deshusses, 1981.)
3. The 'Gregorian' sacramentary. This papal sacramentary was probably compiled under Honorius I (625-38) (rather than Gregory the Great, 590-604, whose name it bears), and was supplemented with new material as required during the subsequent century and a half.
A copy of the Gregorian sacramentary in its late-seventh-century state appears to have been known to Alcuin of Tours (d. 804), chief liturgical adviser to Charlemagne; perhaps it came with other books he had sent from his home city of York in 797. It then became known to St Benedict of Aniane (d. 821), the-great reformer and adviser of Louis the Pious, and to Arno, bishop of Salzburg (785-821), former abbot of Saint-Amand and a friend of both Alcuin and Benedict; it was used in the compilation of the Sacramentary of Trent or Salzburg (ed. Rehle and Gamber 1970, 1973).
In its mid-eighth-century state the Gregorian sacramentary was copied and sent by Hadrian I (772—95) to Charlemagne between 784 and 791. This manuscript, which was kept at Aachen as an exemplar for further copies, has not survived, but a transcription made for Bishop Hildoard of Cambrai in 811-12 still exists (Cambrai Bibliotheque Municipale 164). (See the edition by Deshusses 1971-9.)
It is clear from this (i) that we owe our knowledge of these Roman sacramentaries to Frankish churchmen, (ii) a number of conflicting types were circulating simultaneously in Francia, to which we should add the material brought from Rome more or less privately in the form of libelli.
There also survive from eighth-century Francia four Gallican sacramentaries which contain small amounts of Roman material, the so-called 'Missale Gothicum', 'Missale Francorum', 'Missale Gallicanum vetus' and 'Missale Bobbiense'.
The sacramentary sent by Hadrian at Charlemagne's request was itself insufficient for Frankish purposes. Instead of the expected formularies for the cycle of Sundays after Christmas, Epiphany, Easter, and Pentecost, the 'Hadrianum' contained merely a pool of prayers to choose from as required. It also lacked prayers for consecrations and votive masses, which the FVanks were accustomed to seeing in previous sacramentaries. It contained, in fact, only what the pope himself would have required while celebrating in the Lateran and the stational churches of the city of Rome. Benedict of Aniane therefore compiled an official supplement (with an engagingly-frank preface: trans. Vogel, 87-8). The admixtures of Gelasian material in the
J. Sacrament dries and Lectionanes
293
subsequent generations of Gregorian sacramentaries bear witness to the less than total success with which Charlemagne promulgated 'Roman' use, at least as far as the sacramentary is concerned. Success was in any case hardly feasible, given the nature of the materials and the impossibility of publishing 'correct' books.
The relevance of all this to the history of plainchant may be stated as follows. Roman material of several different types came north both piecemeal and in more comprehensive codices. Is this how the chant repertory, at least for mass, was also transmitted? Since it seems unlikely that musical notation was used until later in the ninth century, we should presumably imagine books containing just chant texts, intermingling to a greater or lesser extent with previous practice. It may be that the date of the first recension of the gradual, containing chant texts for the year's cycle of masses, was compiled at about the same time as the first such sacramentary. We have no firm evidence for this, however. One can go no further than to suggest that the history of the two books may have run partly parallel.
The sacramentary as a separate book gradually went out of use after the late eleventh century, becoming part of the missal. Before and during that time, various combinations of sacramentary, lectionary, and/or gradual were made. These are discussed below where the missal is described. A number of early sacramentaries have chant-text incipits written in their margins, sometimes notated (two examples: Oxford, Bodleian Library, Bodley 579, the 'Leofric Missal', text ed. Warren, 1883; and Düsseldorf, Universitätsbibliothek D 1, facs. Jammers 1952; Stäblein 1975, 107).
No classification of post-Carolingian sacramentaries has been carried out—the task is of course immense—though comparisons of books from specific areas have been made (for example, by British scholars whose work appears in the volumes of the Henry Bradshaw Society).
(ii) I^ectionaries
Frere 1934, 1935; Klauser 1935; Chavasse 1952; Gamber 1968; Vogel 1986, 291-355.
The lessons to be read at mass were taken from the Bible and were known as 'pericopes'. They were recorded in one of several ways. The Bible could be marked up with crosses or other signs in the margin to indicate the passages to be recited. Lists of pericopes were also drawn up, known as capitularies, where incipits and explicits were recorded. Later the lessons were copied out separately in lectionaries (a general term for books with lessons for mass), perhaps divided between the epistolary (lessons from the Epistles, sung by the subdeacon) and the evangeliary (lessons from the Gospels, sung by the deacon). Another early name for the lectionary, or list of readings derived from a pre-existing lectionary, is comes (Liber comitis, Liber commicus).
In the early centuries scriptural passages would have been chosen more or less on an ad hoc basis, and not before the late fourth century is there good evidence that yearly cycles of lessons were organized. The example of Musaeus of Marseilles has just been
294 ///. Liturgical Books and Plainchant Sources
mentioned. Much of the other surviving evidence of this period also comes from Gaul: the information to be gleaned from the sermons of Bishop Caesanus of Aries (502-43) and the Expositio antiquae liturgiae gallicanae, and the earliest of all surviving Latin service-books: the palimpsest Gallican lectionary Wolfenbuttel, Herzog-August-Bibliothek, Weissenburg 76, dating from the fifth to sixth century (ed. Dold, 1936). The Wolfenbuttel source is doubly important because text incipits of chants to be sung after the lessons are also included (Dold, xcivff.). Also non-Roman is the next earliest source, the epistolary of Capua (ed. Ranke 1868), written
c. 545 for Victor of Capua, taken to England probably by Abbot Hadrian of Nisida, who was one of the companions of Archbishop Theodore of Canterbury, and later used by St Boniface). Another famous codex is the Lindisfarne Gospels, written c.700 at Lindisfarne but based on a Neapolitan exemplar (ed. Skeat 1871-87).
The non-Roman Latin liturgies had three readings at mass (from the Old Testament, the New Testament, and the Gospels, respectively), and this may have been the original usage in Rome as well. The earliest Roman witness is the capitulary or comes of Wiirzburg (r.700 but based on a model of the time of Gregory the Great,
d. 604; ed. Morin 1910 and 1911). For some masses this source has three lessons, but it seems possible that the extra lesson is a duplicate (see Martimort 1984). It was until recently widely held that the gradual and alleluia (or tract) of the medieval mass might formerly have been separated by another lesson (for otherwise there would be no point in having two chants here), but lately it has been argued that the gradual itself originated as a psalm reading (McKinnon 1987; see the discussion in VI1.5 below). From at least the seventh century, at any rate, two lessons separated by chant seems to have been the rule in Rome.
It seems that the selection of epistles and the selection of gospels were made independently of each other and independently of the chants and prayers of mass. At least, no system combining them all according to a logical principle has been discovered. The history of the organization of the lessons, which can be traced back to an earlier date than that of prayers and chants, may throw light on the latter. For example, the lessons from Septuagesima to Easter appear to have been organized before the time of Gregory the Great (d. 604) and remain constant thereafter. Chavasse (1952) has demonstrated that the development of the lectionary resembles that of the gradual, rather than the sacramentary.
Of the many different arrangements, Roman and non-Roman, for which witnesses survive from the eighth century, a Frankish-Gregorian type first represented by the Comes of Murbach (ed. Wilmart 1913) emerged as the basis of later medieval usage. In this case, unlike that of the sacramentary, no document of papal usage was solicited from Rome, and the readings were arranged to fit the liturgical occasions required by the Frankish-Gelasian sacramentary.
Lectionaries were combined with sacramentaries and/or graduals in a number of ways, culminating in the missals of the later Middle Ages. These are discussed briefly below (III.9). As with the sacramentary, no classification of post-Carolingian manuscripts (apart from late sources of the older types) has been accomplished.
4. Graduals and Cantatoria
295
III.4. GRADUALS (MASS ANTIPHONERS) AND CANTATORIA
(i) Introduction
(ii) Graduals without Notation
(iii) Notated Graduals
Stäblein, 'Cantatorium', MGG; Melnicki and Stäblein, 'Graduale (Buch'), MGG Graduel rornain, II: Ij>s Sources (1957); Graduel romain, IV: Im Texte neumatique (1960), 1962); Steiner, 'Cantatorium', NG; Huglo, 'Gradual (ii)', NG; Emerson, 'Sources, MS, IT, NG; Andrew Hughes 1982, 124-42, 157-9.
Facsimiles of graduals and noted missals: Bamberg, Staatsbibliothek, lit. 6: see Bamberg 6 Benevento, Archivio Capitolare 33: PalMus 20 Benevento, Archivio Capitolare 34: PalMus 15 Bratislava, etc., fragments: Missale notatum Strigoniense Chartres, Bibliotheque Municipale 47: PalMus 11
Darmstadt, Hessische Landes- und Hochschulbibliothek 1946: Echternacher Sakra-
mentar und Antiphonar Einsiedeln, Stiftsbibliothek 121 (sequentiary omitted): PalMus 4 Gniezno, Archiwum Archidiecezjalne 149: Missale planatium . . . Gnesnensis Graz, Universitätsbibliothek 807: PalMus 19 Laon, Bibliotheque Municipale 239: PalMus 10
Leipzig, Universitätsbibliothek 391 (sequentiary omitted): Wagner 1930-2
London, British Library, Add. 12194 (without sequentiary): Graduale Sarishuriense
Montpellier, Faculte de Medecine II. 159: PalMus 8
Paris, Bibliotheque Nationale 903 (troper and sequentiary omitted): PalMus 13 Paris, Bibliotheque Nationale 904: Loriquet et al. 1907 Passau, printed gradual of 1511: Graduale Pataviense Rome, Biblioteca Angelica 123: PalMus 18
Rome, Biblioteca Apostolica Vaticana, Vat. lat. 10673: PalMus 14 St Gall, Stiftsbibliothek 339: PalMus 1 St Gall, Stiftsbibliothek 359: PalMus 11/2
St Petersburg, Saltykov-Shchedrin Public Library O. v. I. 6: Thibaut 1912 Västeräs, printed gradual of <~.1513: Graduale Arosiense
Vienna, Nationalbibliothek, ser. nov. 2700 (with an office antiphoner): Antiphonar von St. Peter
private collection, the 'Mont-Renaud manuscript' (with an antiphoner): PalMus 16 private collection of Martin Bodmer C. 74 (Old Roman): Lütolf 1987
Editions:
GR; AVIS (texts of six early sources); MMMA 2 (Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319, Old Roman); Sandon 1984, 1986 (Sarum).
(i) Introduction
The term gradual has been used since the early Middle Ages to designate books containing the proper chants of mass. Its derivation is obscure. The earliest surviving
296 ///. Liturgical Books and Plainchant Sources
books of this general type bear the title 'antefonarius' (see the sources edited by Hesbert in his appropriately named Antiphonale missanim sextuplex), as for instance in the title of Brussels 10127-10144 (late eighth century): 'In dei nomen incipit antefonarius ordinatus a sancto Gregorio per circulum anni'. Since the word 'antiphonarius' could also be used of a book containing office chants, gradual is usually preferred for the mass chant-book.
Hucke (1955, 'Graduate') pointed out that in several early occurrences of the word it appears as an adjectivegradalis, referring to the day hours of the liturgy, as opposed to nocturnalis, the night hours. Perhaps there is a connection with the canticum graduum, St Jerome's translation of the mysterious Hebrew rubric for Psalms 120-34, which played an important part in the day hours of the Roman liturgy.
In five of the sources of the Sextuplex the gradual chant is designed 'Resp(onsorium) Grad(ale)', in source R (see the explanation of the siglum below) simply 'Grad(ale)'. This use of 'gradale' is usually explained as 'responsory sung on the altar-steps' (from gradus = step) or 'while the deacon ascends the steps of the ambo to recite the gospel'. An alternative explanation could, however, be 'day responsory', or perhaps here the word has no connection with the previous meaning.
In Ordo Romanus I (c.700) it is stated that the cantor 'cum cantatorio' goes up to sing the responsory (gradual) and alleluia or tract. The statement reappears in Ordines IV, V, and VI. From this a special class of gradual has been identified, containing only these chants. Such manuscripts do indeed survive from the ninth century onward, including the famous St Gall, Stiftsbibliothek 359, which notates graduals, alleluias, and tracts in full but gives only text incipits for the other chants.
Amalarius of Metz (Prologus de ordine antiphonarii, 18, ed. Hanssens, i. 363; PL 105, 1245), writing c\830, says that what the Romans call a eantatorium, the Franks call a gradale. The Romans collect responsories in a responsoriale and antiphons in an antiphonarius. If responsoriale and antiphonarius refer only to the office, eantatorium to the mass, then the antiphons of mass are left unaccounted for. In fact, however, all Amalarius' writings in this context refer only to office chants. Hucke argues therefore that the cantatorium/gradale is once again the book containing chants for the day hours. The Night Office requires far more chants, and the Romans divide them between two books, one for responsories and the other for antiphons. 'I have followed our usage', says Amalarius, 'and place both responsories and antiphons mixed together according to the order of the hours'.
At this early stage, therefore, the nomenclature of chant-books was somewhat fluid, and references to 'antiphonarius' and 'gradale' in early book-catalogues have to be treated with some caution (for example those edited by Gustav Becker 1885: see nos. 7, 8, 23, 24 for 'gradale/gradalis', and nos. 7, 18, 21, 33, 238 for'antiphonarius').
(ii) Graduals without Notation
Although books containing chant texts for mass probably existed as early as any other mass books, they have survived in much smaller numbers. The earliest evidence for
4. Graduals and Cantalotia
297
them that we have is possibly Gennadius' reference, already mentioned (III.3), to the books compiled by Musaeus of Marseilles (d. c.460) at the request of his bishop Venerius (d. 452), though he may mean office responsories, rather than the gradual responsories of mass. At the end of the same century, however, appears the first copy of mass chant texts to have survived, the incipits copied after various lessons in the palimpsest Gallican lectionary Wolfenbtittel, Herzog-August-Bibliothek, Weissenburg 76 (ed. Dold 1936). Dold was with great difficulty able to decipher sixteen of these incipits, some of which correspond with later gradual or offertory texts. They may refer the user across to another book containing the complete chant texts, unless we are to assume that these were known by heart.
The first evidence for the existence of chant-books in Rome itself is by report only and considerably later. At the same time as Bede was writing his history of the English church and people (completed 731), Egbert, a member of the Northumbrian royal family, was ordained deacon in Rome and was then appointed bishop of York (c.732). In a letter of 735 which still survives, Bede advised Egbert to apply for the pallium, which he obtained; he died in 766. His brother Eadberht had become king in 738, and with his support Egbert founded a cathedral school in York, where Alcuin was a pupil and eventually master. In two places in Egbert's Dialogus ecclesiasticus institutionis (PL 89, 377-451) he refers to a chant book ('liber antiphonarius') and sacramentary ('liber missalis') which he had presumably seen in Rome. He shared the popular belief that their author was Gregory the Great and that Augustine had brought copies to England over a century earlier: '. . . beatus Gregorius in suo antiphonari et missali libro per paedagogum nostrum beatum Augustinum transmisit ordinatum et rescnptum . . .' (PL 89, 440-2, cited Gevaert 1890, 80-1 and Ashworth 1958).
Contemporary evidence is to be found in the canons of the Council of Cloveshoe (747):
Ut uno eodemque modo dominicae dispensationis in carne sacrosanctae festivitates, in omnibus ad eas rite competentibus rebus, id est in baptismi officio, in missarum celebratione, in cantilenae modo celebrentur, iuxta exemplar videlicet quod scriptum de romana habemus ecclesia. Itemque ut per gyrum totius anni natalicia sanctorum uno eodem die, iuxta martyrologium eiusdem romanae ecclesiae, cum sua sibi conventienti psalmodia seu cantilena venerentur. (Haddan and Stubbs 1871, 137.)
The reference to a Roman exemplar covers each of the types of book mentioned, for the baptismal rites, for the mass (presumably a sacramentary), and for chant. The feasts of the Sanctorale, governed by the martyrology, are to be sung with the appropriate psalmody and chants. There is no definite implication that either this chant-book or those seen by Egbert contained notation.
The first graduals (or mass antiphoners) which have come down to us all lack notation and were written in north France. They were edited by Hesbert in AMS and their date and provenance have been summarized by Froger (1978, 'Critical'). I list them here with the customary sigla.
298 ///. Liturgical Books and Plainchant Sources
Brussels, Bibliotheque Royale 10127-10144 (B = Blandiniensis), abbey of St Peter on Mont-Blandin, Ghent; late eighth century; the manuscript also contains Ordines Romani (see above, 111.2).
Zurich, Zentralbibliothek, Rheinau 30 (R = Rhenaugiensis), Nivelles; 790s; the parts of chants sung by soloists (gradual verses, etc.) are not given; the manuscript also contains a Frankish-Gelasian sacramentary.
Paris, Bibliotheque Nationale, lat. 12050 (K = Corbiensis), abbey of Corbie; 850s; the manuscript also contains a Gregorian sacramentary with Benedict of Aniane's supplement, copied for Rodradus of Corbie.
Monza, Basilica S. Giovanni CIX (M = Modoetiensis), NE France; second half of ninth century; the manuscript has only soloists' chants.
Paris, Bibliotheque Nationale, lat. 17436 (C = Compendiensis), abbey of Saint-Medard at Soissons (see Froger 1980), later at the abbey of Saint-Corneille at Compiegne; second half of ninth century; the manuscript also contains an office antiphoner, the earliest such to have survived; sometimes known as the 'Antiphoner of Charles the Bald', in the belief that it was presented by Charles (d. 877) to Saint-Corneille.
Paris, Bibliotheque Sainte-Genevieve 111 (S = Silvanectensis), written at the monastery of Saint-Denis for the cathedral of Senlis; 877-82; the manuscript also contains a sacramentary.
A number of fragmentary sources also belong to this early group, of which the most important are perhaps the Lucca fragments of the late eighth century, which have office chants as well (AMS, XXIV-XXVI; Froger 1979, 'Fragment').
The manuscripts do not all record chants in the same way. The Monza source has only graduals, alleluias, and tracts (and of these only the tracts are consistently written out in full) and can therefore be designated as a cantatorium. The Nivelles book (R) by contrast has only texts sung by the choir.
As already mentioned, these manuscripts have no notation, although the Corbie gradual has letters indicating the mode of chants (see below, III. 14). They nevertheless tell us a great deal about the development of the repertory. Particularly interesting are those sections where they differ from one another or lack chants regularly present in later books. There is space here to mention only a few of these peculiarities (see the exhaustive study in AMS).
Most striking is the disunity in the assignment of alleluias to the various masses. In several places the Blandiniensis simply gives the rubric 'alleluia quale volueris', meaning that the singer must at this point dip into a pool of suitable alleluias gathered elsewhere (in this case at the end of the manuscript). This happens in Easter week, for the Sundays after Easter, Whitsun week, and the Sundays after Whitsuntide. There is the greatest disparity between later books in their assignment of alleluias to these occasions. The graduals of the summer Sundays also display peculiarities of ordering, which has given rise to extensive discussion (see Hesbert 1932-3; PalMus 14, 124— 44; Chavasse 1952; Raymond Le Roux 1962; Chavasse 1984).
4. Graduate and Cantatoria
299
The Roman mass antiphoners known to the Franks had a number of features which were altered by them, but which can be found in some later Gregorian sources. Thus M has the Roman offertory for St Michael In conspectu angelorum instead of the Prankish Stetit augelus. Huglo (1954, 'Vieux-romain') drew up a list of Roman characteristics and identified their presence in several later sources, mostly from central Italy.
On the other hand, all Prankish books include a mass for St Gorgonius (9 September), a non-Roman custom initiated when his relics were brought to Metz by Bishop Chrodegang during the reign of Pippin the Short (751-68).
It may be asked how the Franks, seemingly ardently engaged in imitating Roman use as closely as possible, felt at liberty to deviate occasionally from what they found in Roman books. The answer might be along the lines of the apologia presented by Benedict of Aniane, prefacing his supplement to the Gregorian sacramentary: '. . . there are other liturgical materials which Holy Church finds itself obliged to use but which the aforesaid Father [Gregory] omitted because he knew they had already been produced by other people . . .'
(iii) Notated Graduate
The earliest notated graduals to have survived are from the end of the ninth century: Chartres, Bibliotheque Municipale 47 (destroyed in 1944; facsimile PalMus 11), from Brittany.
St Gall, Stiftsbibliothek 359 (PalMus 11/2), from St Gall; the manuscript is a cantatorium, with soloist's chants only.
Laon, Bibliotheque Municipale 239 (PalMus 10), from Laon.
Equally early fragments have also survived from these areas (that is, Brittany, Germany, and north France), but the earliest sources from other areas are somewhat later. England, as might be expected after the devastation wreaked by the Danish invasions of the eighth and ninth centuries, is represented first in the late tenth century. The earliest surviving graduals from Aquitaine are from the early eleventh century, but three tropers of the tenth century survive. Notated Italian sources are practically non-existent before the eleventh century.
By the time of the appearance of the first notated graduals, the assignment of particular chants to particular masses was fairly stable, the main differences between sources being the following:
1. the choice of alleluias for the Easter and Whitsuntide series;
2. the choice of psalm verses for introit and offertory;
3. the choice of chants for saints' days; these could be taken from the common pool, in which case different sources might make a different choice, or (rarely) newly composed.
In the sources of the Sextuplex there is no common pool, except for four masses for a 'pontifex' in K and S only. Nor is there one in many other early sources. There is
300 ///. Liturgical Books and Plainchant Sources
simply a cross-reference from the one saint to another where the appropriate mass is to be found. As the Sanctorale became more heavily populated, however, the frequent cross-referring became inconvenient, and the chants were eventually transferred to a Commune Sanctorum section at the end of the gradual.
Another way in which graduals might differ is in their interweaving of the Temporale and Sanctorale. From later service-books we are used to the placing of all saints' feasts except those immediately following Christmas in a section of their own. The Roman pattern as found in, for example, Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319 (MMMA 2), is given in Table III.4.1.
Table 111.4.1 Interlocking Temporale and Sanctorale in Roman use
Temporale	Sanctorale
Advent Sundays 1-2	
	St Lucy (13 Dec.)
Advent Sunday 3	
Ember Week	
Christmas	
	St Stephen (26 Dec.) to St Silvester (31 Dec.)
Sunday after Christmas	
Epiphany	
Sundays after Epiphany	
	St Felix (14 Jan.) to Annunciation of the BVM (25 Mar.)
Septuagesima to the 4th Sunday	
after Easter	
	SS Tiburtius and Valerianus (14 Apr.) to St Pudentiana
	(19 May)
Ascension	
Sunday after Ascension	
	St Urban (25 May)
Whit Sunday and Whitsun week	
	SS Marcellinus and Peter (2 June) to SS John and Paul
	(26 June)
Sundays 1-4 after Whitsun	
	SS Peter and Paul (29 June) to Octave of SS Peter and Paul
	(6 July)
Sundays 5-9 after Whitsun	
	The Seven Brothers (10 July) to SS Felix and Adauctus
	(30 Aug.)
Sundays 10-15 after Whitsun	
	St Hadrian (8 Sept.) to St Matthew (21 Sept.)
Ember Week	
	SS Cosmas and Damian (27 Sept.) to St Mennas (11 Nov.)
Sundays 16-24 after Whitsun	
	St Cecilia (22 Nov.) to St Andrew (30 Nov.)
4. Graduals and Cantatoria 301
As the sources of the Sextuplex and early notated graduals show, the Franks altered the disposition of the post-Whitsuntide period, as for instance in St Gall 339 (contents listed in Wagner I), displayed in Table 111.4.2.
Thus the saints of the Epiphany period have been amalgamated with the Temporale, not a common procedure, but also not a confusing one, for the Sundays do not differ much in date from year to year. On the other hand, the complete Sanctorale from June to November appears as one block, followed by all the Sundays after Whitsuntide together.
It was a logical step to remove the winter saints from the first part of the gradual as well, so that only those at the end of December remained. The arrangement in the thirteenth-century manuscript London, British Library, Add. 12194 (Graduate Sarisburiense) is given in Table III.4.3.
Of the kyriale and sequentiary more will be said below. Graduals also frequently
Table 111.4.2. Disposition of Temporale and Sanctorale in Frankish use
Temporale Sanctorale
Advent Sundays 1-2
St Lucy (13 Dec.)
Advent Sunday 3 Ember Week Christmas
St Stephen (26 Dec.) to St Silvester (31 Dec.)
Sunday after Christmas Epiphany
1st Sunday after Epiphany
St Felix (14 Jan.)
2nd Sunday after Epiphany
St Marcellus (16 Jan.) to St Agnes (21 Jan.)
3rd Sunday after Epiphanv
St Vincent (22 Jan.) to Annunciation of the BVM (25 Mar.)
Septuagesima to the 4th Sunday after Easter
SS Tiburtius and Valenanus (14 Apr.) to St Pudentiana or Potentiana (19 May)
Ascension
Sunday after Ascension
St Urban (25 May)
Whit Sunday and Whitsun week
SS Marcellinus and Peter (2 June) to St Andrew (30 Nov.)
Trinity Sunday
Sundays 1-24 after the Octave w of Whit Sunday
302 ///. Liturgical Books and Plainchant Sources
Table 111.4.3. Separate Temporale and Sanctorale in later medieval use
Temporale	Sanctorale		Other material
Advent Sundays 1-3			— ■•   -    —- .......t
Ember Week			
Advent Sunday 4			
Christmas			
	St Stephen (26 Dec.	) to	
	St Silvester (31 Dec	•)	
Sunday after Christmas to the			
Sunday before Advent			
Dedication of the Church			
	St Andrew (30 Nov.	.) to	
	St Linus (26 Nov.)		
			Commune sanctorum
			Requiem mass
			Votive masses
			Kyriale and sequentiary
contain processional chants, which, though not part of the mass itself, were often the part of the liturgy which immediately preceded mass (see 1.7). Not all books follow the Sarum pattern exactly, of course. Paris books, for example, put even the Christmas saints' feasts in the Sanctorale. Sometimes the year began with a saint other than St Andrew. Graduals which also include tropes, ordinary of mass melodies, and sequences have different ways of deploying the material: while most place them in separate sections (or have a separate book for them), several Italian books have them among the proper chants for particular masses. The brief descriptions by Emerson ('Sources, MS, II', NG) indicate many of these differences.
Apart from differences of organization or choice of chants, graduals also differ from one another in their musical readings, because of oral transmission and the freedom of local cantors to notate the melodies as they understood them, rather than slavishly following an exemplar. The end result is a host of differences between sources. Some are very obvious differences, such as a quite different melody; the vast majority are trivial as far as the basic shape and character of the melody is concerned.
A little over thirty years ago the monks of Solesmes began publication of material for a new critical edition of the gradual with Gregorian melodies, under the title L<? Graduel rornain. A list of nearly 750 sources appeared first, with brief descriptions. Then the results were published of several comparisons of the sources with one another, carried out by isolating places where the melodic readings tended to show disparity. By counting the number of agreements between sources at these 'lieux variants' it was possible to group the sources into families. (For further work of this kind, see Hiley 1980-1 and 1986, Thurstan'.) One of the most striking results to emerge from the Solesmes investigation was the remarkable unanimity between
5. Antiphoners 303
German sources, compared with the relatively fragmented picture for France and Italy (see below, IX.3).
As can be seen even from the few manuscripts of the Sextuplex, early books of chant texts were often coupled to sacramentaries or lectionaries. Chant incipits were also given in many early sacramentaries. As sources of mass chants, graduals and cantatoria were later joined by notated missals, but persisted as independent types of hook until the end of the Middle Ages.
III.5. ANTIPHONERS (OFFICE ANTIPHONERS)
(i) The Earliest Antiphoners
(ii) Types of Manuscript and Calendric Organization
(iii) Comparison of Sources
Stäblein, 'Antiphonar', MGG; Huglo, 'Antiphoner', NG; Emerson, 'Sources, MS, IF, NG; Andrew Hughes 1982, 161-97, 238-9, 242-4.
Facsimiles:
Berlin, Staatsbibliothek Preußischer Kulturbesitz, Mus. ms. 40047: Möller 1990 Cambridge University Library, Mm.2.9: AS
Durham, Cathedral Chapter Library B. III. 11: Frere 1923 (see also CAO 1) Graz, Universitätsbibliothek 211: Codex Albensis
London, British Library, Add. 30850: Antiphonale Silense (see also CAO 2)
Lucca, Biblioteca Capitolare 601: PalMus 9
Passau, printed antiphoner of 1519: Antiphonale Pataviense
St Gall, Stiftsbibliothek 390-391: PalMus II/l
Vienna, Nationalbibliothek, ser. nova 2700 (with a gradual): Antiphonar von St. Peter Worcester, Cathedral Chapter Library F. 160: PalMus 12 (only the antiphoner, processional, and hymnal)
private collection, the 'Mont-Rcnaud manuscript' (with a gradual): PalMus 16
Editions: AR, AM The text editions in CAO 1-2: CAO 1: manuscripts following the secular or Roman cursus B = Bamberg, Staatsbibliothek, lit. 23 C = Paris, Bibliotheque Nationale, lat. 17436
G = Durham, Cathedral Chapter Library B. III. 11 (see also facs.) E = Ivrea, Biblioteca Capitolare CVI M = Monza, Basilica S. Giovanni C. 12/75 V = Verona, Biblioteca Capitolare XCVIII CAO 2: manuscripts following the monastic cursus H = St Gall, Stiftsbibliothek 390-391 (see also facs.) R = Zürich, Zentralbibliothek, Rheinau 28 D = Paris, Bibliotheque Nationale, lat. 17296 F = Paris, Bibliotheque Nationale, lat. 12584 S = London, British Library, Add. 30850 (see also facs.) L = Benevento, Archivio Capitolare 21
304 ///. Liturgical Books and Plainchant Sources
CAO 3-4: edition of each individual chant text
CAO 5-6: repertorial comparison of 800 sources (see also Ottoscn 1986) (i) The Earliest Antiphoners
The word antiphoner is usually used to mean the book containing the chants of the office hours. This distinction is a relatively recent one, however, for up to the ninth century the word was regularly used to mean any kind of chant-book. If Amalarius be understood correctly (see the passage discussed above, 111.4), the Romans may have gathered both office and mass antiphons together in an 'antiphonarius', and both office and mass responsories together in a 'responsoriale'. As Huglo points out ('Antiphoner', A/G, 482) the Romans may thus have had mass and office chants amalgamated in one book, whereas Frankish practice was always to separate office from mass chants. For these reasons, the 'antiphonarii' which appear in library catalogues of the eighth and ninth centuries (cited by Huglo) cannot be identified definitely with office antiphoners. Whatever the name in the early centuries, 'antiphoner' is used here in its commonest modern sense, to mean a book with all the office chants.
Early books containing office chant texts have survived in very small numbers compared with those containing mass chant texts. The texts copied by Winithar in St Gall, Stiftsbibliothek 1399 in the late eighth century (ed. Dold 1940) and the Lucca fragment of the late eighth century (Lucca, Biblioteca Capitolare 490: ed. Froger 1979, 'Fragment') are the earliest surviving documents. They are followed more than half a century later by the so-called 'Antiphoner of Charles the Bald', which also contains a gradual (ed. Hesbert, CAO 1), and an Aquitanian manuscript of the same type, Albi, Bibliotheque Municipale 44. These sources were not notated.
The earliest antiphoners with notation are:
the 'Mont-Renaud manuscript' (in a private collection), abbey of Saint-Denis, text first half of tenth century, neumes added tenth-eleventh century (PalMus 16).
St Gall, Stiftsbibliothek 390-391, St Gall, copied by the monk Hartker, end of tenth century (text ed. Hesbert, CAO 2; facs. PalMus 11 /1).
Oxford, Bodleian Library, Auct. F. 4. 26, a fragment with Breton notation, tenth century (facs. Nicholson 1913, pi. IX).
The antiphoner which Amalarius of Metz compiled, described in his Liber de online antiphonarii (c.830), is lost. It seems to have been a private initiative without long-lasting consequences, for no surviving antiphoner reflects the ordering of pieces suggested by Amalarius (see Hesbert 1980, 'Amalar'). The actual Metz repertory can nevertheless be reconstructed in its entirety from the ninth-century tonary Metz, Bibliotheque Municipale 351 (ed. Lipphardt 1965), as well as from later Metz manuscripts.
Notated sources from other parts of Europe have survived only from a later period. Two lists of chants survive in tenth-century sources from Limoges (Paris, Bibliotheque Nationale, lat. 1085 and 1240); but notated Aquitanian sources are
5. Antiphoners 305
lacking before the eleventh century. Toledo, Archivo Capitular 44. 1 was written in south France in the eleventh to twelfth century and taken to Spain at a time when Roman use was replacing Mozarabic. Other early Spanish manuscripts are also witness to this change: London, British Museum, Add. 30848 and 30850 were both written at San Domingo de Silos in the late eleventh century (the former is a noted breviary; the text of the latter ed. Hesbert, CAO 2, and facs. Antiphonale Silense). The earliest surviving source from south Italy is likewise a noted breviary, Montecassino, Archivio della Badia 420. North Italian antiphoners of the eleventh century include Monza, Basilica S. Giovanni C. 12/75 (text ed. Hesbert, CAO 1), Monza C. 15/79, and Oxford, Bodleian Library, Misc. lit. 366.
(ii) Types of Manuscript and Calendric Organization
Although the amalgamation of office and mass chants in a gradual-antiphoner is very rare, the presence of both gradual and antiphoner between the same covers of a medieval manuscript is not unknown. The complementing of chants by office lessons and other texts produced the breviary, which is discussed below. At the same time some parts of the office chant repertory were copied separately: in psalters and hymnals, both also discussed below. Another category within the general antiphoner class is the diurnal, where only the chants for the day hours are recorded. The great bulk of the material to be recorded prompted several such subdivisions and subspecies of book. Since such books often included not only chants but also lessons and prayers, two examples are discussed below in the section on breviaries (III. 10).
The arrangement of Temporale and Sanctorale in antiphoners is usually different from that of graduals. The responsories and antiphons for the Sundays after Whitsuntide are usually copied separately from each other, perhaps reflecting the fact that the chants had once been assembled in different books. Benedictus and Magnificat antiphons form another group sometimes copied separately. The tones for psalms and canticles, being so universally familiar, are rarely found, at any rate in earlier antiphoners, and are usually placed in a supplement at the end of the book. They are likely to appear in close proximity to a tonary, and are thus occasionally given in books which otherwise contain no office chants, such as the Saint-Martial tropers Paris, Bibliotheque Nationale, lat. 909 and 1121.
Two different arrangements of the chants may be seen in Table III.5.1. The two sources represented are: Hartker's antiphoner, St Gall, Stiftsbibliothek 390-391, of the closing years of the tenth century (facsimile, PalMus II/2; text edition, CAO 2), and Lucca, Biblioteca Capitolare 601, from the abbey of St Peter at Pozzuoli, a manuscript of the early twelfth century (facsimile, PalMus 9). Both books are monastic, though that is not of relevance to the way the year is organized. (Both editions have tables setting out the order in which the chants appear: PalMus 11/2, pp. 61*—62*; PalMus 9, pp. 16*—20*, which also gives the number of chants for each occasion.) It will be noticed that neither manuscript contains a psalter or a hymnal. Not even incipits for hymns are given.
306 ///. Liturgical Books and Plainchant Sources
Tabic 111.5.1. Deployment of material in two antiphoners
St Gall, Stiftsbibliothek 390-391
Lucca, Biblioteca Capitolare 601
Advent Sundays 1-2
St Lucy (13 Dec.)
Advent Sunday 3 Advent Sunday 4
Christmas Eve to the 5th Sunday after Epiphany
responsories for the weekly office antiphons and short responsories for the
weekly office Trinity
St Sebastian (20 Jan.) to Annunciation of the BVM (25 Mar.)
Septuagesima to Easter week antiphons for Sundays after Easter
SS Philip and James (1 May)
responsories for Sundays after Easter
Commune Sanctorum in Paschaltidc
Finding of the Cross (3 May) Exaltation of the Cross (14 Sept.) SS Alexander, Evcntius, and Theodulus (3 May)
Vigil of Ascension through Whitsun week
St John Baptist (24 June) to St Andrew (30 Nov.)
Commune Sanctorum Office of the Dead
responsories and Magnificat antiphons for the summer Sundays, a group of responsories being followed by a group of Magnificat antiphons
ferial antiphons for the Benedicite, Benedictus, and Magnificat
antiphons for the summer Sundavs
Advent Sundays 1-3
Advent Sunday 4
Christmas Eve to the Octave of the Epiphany 1st Sunday after the Octave of the Epiphany and subsequent ferias: these chants are for Sundays and ferias throughout the year when no feast-day intervenes Sundays after the Octave of the Epiphany
Septuagesima to the 5th Sunday after Easter
Vigil of Ascension through Whitsun week
responsories and Magnificat antiphons for the summer Sundays, a group of responsories being followed by a group of Magnificat antiphons
antiphons for the summer Sundays
St Lucy (13 Dec.) to St Andrew (30 Nov.)
Commune Sanctorum Dedication of the Church Office of the Dead
5. Antiphoners 307
It was customary in antiphoners and breviaries to designate the summer Sundays after the biblical book which provided the lessons of the Night Office. In the Lucca manuscript these Old Testament books, or histories (historia), are assigned as follows:
the Book of Kings, for the first Sunday after Whitsun to first Sunday in August
the Book of Wisdom, for August
the Book of Job, for the first half of September
the Books of Tobit, Judith, and Esther, for the rest of September
the Book of Maccabees, for October
the Books of the Prophets, for November
In any antiphoner the number of proper chants naturally varies greatly from occasion to occasion. The provision for Ascensiontide in the Lucca codex is typical:
\'igil of Ascension
Lauds: Benedictus antiphon (text incipit only: already available elsewhere) Ascension Day
first Vespers: antiphon for psalms; responsory (incipit only); versicle (incipit
only); Magnificat antiphon Night Office: invitatory; six antiphons, versicle, and four responsories for the first
Nocturn; the same for the second Nocturn (versicle incipit only); antiphon,
versicle (incipit only), and four responsories for the third Nocturn Lauds: five antiphons; short responsory; versicle; Benedictus antiphon Little Hours: versicles (incipits only) for Terce, Sext, and None second Vespers: antiphon (incipit only); short responsory; versicle (incipit only);
Magnificat antiphon Friday after Ascension Day
Night Office: Invitatory; one antiphon (incipit only) for each of the first and second
Nocturns
Lauds: antiphon for psalms; Benedictus antiphon (incipit only) Little Hours: antiphon for Prime; versicles (incipits only) for Terce, Sext, and None
Sunday after Ascension Day Lauds: Benedictus antiphon second Vespers: Magnificat antiphon
Understanding what has been omitted and has to be drawn from the common pool, and where to find it, poses considerable problems to the uninitiated.
(iii) Comparison of Sources
As in the case of the gradual, much time and effort has been devoted to studying the relationships between sources, by comparing their selection of chants and (to a lesser extent so far) the variant readings of their texts and melodies. As a glance at the twelve manuscripts whose texts were edited by Hesbert (CAO 1-2) will show, the selection of
308 ///. Liturgical Books and Plainchant Sources
chants for the antiphoner is much less stable than for the gradual, and the same is true of their melodies. The variance in text selection prompted Hesbert to carry out a massive survey of 800' sources (including breviaries as well as antiphoners), comparing their responsories for the four Sundays of Advent (CAO 5 and 6; see also Ottosen 1986). Similar repertory comparisons have been made by Raymond Le Roux for the antiphons of the Night Office and Lauds at Christmas and 1 January (1961), responsories of Epiphanytide (1963), and the responsories of the Triduum Sacrum and Easter (1979). These surveys have proved strikingly capable of isolating families of related sources: those belonging to the same church or diocese, those linked by monastic reform movements, those of the religious orders, and so on (apart from the lengthy discussions in CAO 5 and 6, see also Hesbert 1980, 'Antiphonaire de la Curie', 1980, 'Antiphonaire d'Amalar', 1980, 'Sarum', 1982, 'Antiphonaires monastiques insulaires', 1982, 'Matines de Päques', and Raymond Le Roux 1967). A classic study of a branch of the tradition, with facsimiles of surviving fragments, was made by Gjerlow (1979) for the church of Nidaros. (For extended studies of the contents of individual antiphoners, see Ossing 1966, Ledwon 1986, Möller 1990.)
Musical comparisons have taken place on a much smaller scale, understandably enough, considering the daunting magnitude of the repertory. Udovich (1980) provides parallel transcriptions of the Magnificat antiphons from the ferial office in nineteen sources. Underwood (1982) has compared variant readings in antiphons from a handful of English sources with a view to establishing the pattern of their interrelationships (see also Hiley 1986, 'Thurstan').
III.6. PSALTERS, HYMNALS, COLLECTARS, OFFICE
LECTIONARIES
(i) Psalters
(ii) Hymnals
(iii) Collectars
(iv) Office Lectionaries
The other books beside the antiphoner needed for the celebration of the Divine Office are the psalter and hymnal (essentially for singing), the collectar (which contains prayers), and various books containing lessons: passionaries, homiliaries, and so on. Various different types and combinations of these books are known, assembled to suit the needs of a particular institution or person at a particular time, as convenience dictated. All might be gathered together in the breviary.
(i) Psalters
Steiner, 'Psalter', NG; Andrew Hughes 1982, 224-36; Dyer 1984.
The Latin text of the psalter or Book of Psalms, like the text of the Bible as a whole, was not unified for all times and places. In the early centuries several Latin
6. Psalters, Hymnals, Collectars, Lectionanes 309
translations of the Bible or parts of it were current, made from the Greek Septuagint. There was no special urgency to impose uniformity, for Latin did not replace Greek as the language of the liturgy until the late fourth century. The main rites of Western Europe, Milanese, Mozarabic, etc. had their own translations. As far as the psalter is concerned, the most important types are the following.
(i) The Roman Psalter. Once attributed to St Jerome, this version is still used in St Peter's, Rome, and was used both in other Roman churches and elsewhere in Italy until its replacement by the Gallican Psalter under Pius V (1566-72). Adopted by the Franks from Roman exemplars, it was also the basis of chants sung at mass, in Gregorian as in Roman chant.
(ii) The Gallican Psalter. Made by St Jerome c.392 from the Hexapla Old Testament of Origen, this became very popular in Gaul, perhaps partly under the influence of Gregory of Tours and later of Alcuin. From the Carolingian period this was the translation used for the singing of the office in most churches outside Spain and Italy.
(iii) The Hebrew Psalter. A new translation made by Jerome c.400 as part of his translation of all the books of the Bible from the Hebrew. Jerome's work forms the greater part of the Vulgate Bible ('editio vulgáta' = 'common or popular edition'), but the Gallican Psalter was already too popular to displace and was therefore included in the Vulgate in all lands except Italy (the Roman Psalter) and Spain (the Hebrew).
In order to adapt the biblical psalter for liturgical use, various measures were adopted. The psalms might be left in their usual order, or they might copied in a rearranged order to fit the weekly office cycle. The rearrangement would take a different form according to whether the secular (or Roman) cursus was being followed or the monastic cursus. If the psalms were left in their usual order then notes in the margin might indicate on which day and at which hour they were to be sung. Many psalters contain not just the texts of the psalms but also those of the canticles, sung in a similar fashion at the same services. And the most complete type of liturgical psalter contains the antiphons which accompany the psalms, also invitatories, and perhaps hymn incipits (for which pieces a full hymnal would be needed) and the short lessons and versicles with responses. The whole material necessary to sing the weekly office would then be present.
Such material is not often given in antiphoners and breviaries. It was convenient to have it in a separate book, because the weekly office cycle was performed independently of the festal cycle, breaking off whenever a feast-day intervened and resuming when the feast-day was over. As Steiner remarks, 'the Proper of the Time and of the Saints ... is essentially a more or less lengthy series of interruptions into [the Sunday and ferial] Office' ('Psalter', NG).
No facsimile of a liturgical psalter with notated chants has been published. Leroquais (1940-1) describes a large number. The sections covering the weekly office in the Antiphonale Romanům and the Antiphonale monasticum present the psalter in order as sung but omit the Night Office, while the Liber usualis first sets out the
310 ///. Liturgical Books and Plainchant Sources
Vesper psalms and Magnificat in all eight psalm tones but without antiphons, then proceeds to the hours on Sunday except for the Night Office, which are given in full, then to the Little Hours on weekdays. The new Psalterium monasticum of 1981 restores the Night Office (called Vigils).
(ii) Hymnals
Moberg 1947; MM MA 1; Gneuss 1968; Stäblein, 'Hymnar', MGG\ Steiner, 'Hymn, II. Monophonic Latin', ArG.
Editions with music: Ebel 1930 (with facsimile of Einsiedeln, Stiftsbibliothek 366: see MMMA 1) Waddell 1984 (critical edition of the Cistercian hymnal from manuscript sources) MMMA 1: Stäblein's edition of over 550 melodies, where the following hymnals are given complete:
Milan, Biblioteca Trivulziana 347 (Milan, 14th c.)
Heiligenkreuz, Stiftsbibliothek 20 (Cistercian, 12th-13th c.)
Rome, Biblioteca Apostolica Vaticana, Rossi 205 (Moissac, 10th-11th c.)
Paris, Bibliotheque Nationale, n.a.l. 1235 (Nevers, 12th c.)
Worcester, Cathedral Chapter Library F. 160 (Worcester, 13th c; facs. PalMus 12, 1*—12*)
Klosterneuburg, Stiftsbibliothek 1000 (Klosterneuburg, 1336)
Zürich, Zentralbibliothek, Rheinau 83 (Kempten, early 1 Ith c.)
Einsiedeln, Stiftsbibliothek 366 (Einsiedeln, first half of the 12th c., mid-13th c;
facs. Ebel 1930) Verona, Biblioteca Capitolare CIX (102) (Verona, llth c.) Rome, Biblioteca Casanatense 1574 (Gaeta, 12th c.)
The hymnal was the other chant-book for the office hours more often copied separately from the antiphoner than included in it. Occasionally a hymnal forms part of quite a different book: Paris 1235, for example, is a gradual with a large supplementary section giving tropes, sequences, and ordinary-of-mass chants for the main feasts; in between these come a processional, a collection of short responsories, a tonary, and a hymnal.
The arrangement of hymnals usually resembles that of antiphoners. The hymns for the weekly office usually appear at Epiphanytide. The much-repeated Compline hymns will be given at the beginning of the collection. As an example of the arrangement of a thirteenth-century hymnal, the Sarum manuscript Oxford, Bodleian Library, Laud lat. 95 may be described. There are 100 hymns, distributed as follows (only for the first few are the assignments given in detail):
1 first Sunday of Advent: Vespers
2 Compline except on double feasts and during Lent
3 first Sunday of Advent: Night Office
4 first Sunday of Advent: Lauds
5 Christmas Eve: Vespers
6. Psalters, Hymnals, Collectars, l^ectionaries 311
6		Compline on double feasts except on the Annunciation of the BVM and
		from Easter to Whitsuntide
7		Christmas: Night Office
8		Christmas: Lauds and Vespers
9		St Stephen: Vespers and Night Office
10		Epiphany: Vespers and Night Office
11		Epiphany: Lauds
12-	-19	first Sunday after Epiphany: first Vespers, Night Office, Lauds, Prime
		Terce, Sext, None, second Vespers
20-	-37	Monday to Saturday: Night Office, Lauds, Vespers (no Vespers on
		Saturday)
38-	-65	Quadragesima Sunday to the first Sunday after the Octave of Pentecost
66-	-88	Sanctorale from St Vincent (22 Jan.) to St John the Evangelist (27 Dec.)
		with a supplement
89-	-98	Common of Saints
99-	-100	Dedication of the Church
Stablein's edition of the contents of several hymnals, with other miscellaneous items, in MMMA 1 is the single most important reference tool for the subject, and contains a list of over 360 sources. Many of these are antiphoners, breviaries, or psalters with attached hymnals. The earliest extant hymnals, which have no musical notation, date back to the eighth century; the earliest notated collections are from the eleventh century. While Gneuss (1968) has analyzed a number of early collections and established their interrelationships, Moberg (1947) has carried out almost the only comparable work on later books. The earliest collections, referred to as the Old Hymnal type I, unite in a broad way the monastic uses both of Gaul (St Caesarius of Aries) and Italy (St Benedict) and can be linked to the Ambrosian repertory. Two recastings of the Old Hymnal are then traceable: the Old Hymnal type II is found in Prankish sources of the eighth and ninth centuries; the New Hymnal, which may be connected with the reforms of St Benedict of Aniane (d. 821) is found in Frankish sources from the ninth century onwards and constitutes the basis of most later collections. An exception is the hymnal of the Cistercians, which was a conscious attempt to return to the supposed hymns of St Ambrose: Milanese sources were used for the purpose, though they had to be supplemented by others better known to the French monks (see the discussion in Waddell 1984).
(iii) Collectars
Gy 1960 (with list of sources to r.1400).
The collectar (or orational) is a book containing the collects and often also the short lessons (capitula or chapters) used in the office hours. The same person usually recited both. One each of these prayers and lessons was said at each of the day hours, that is, at all except the Night Office. The lesson followed the psalms, the collect came
312 Liturgical Books and Plain chart t Sources
at the end. Collectars commonly include other material as well, such as the versicle and response and incipits for antiphons, responsories, and hymns. If the book provides the chants in full it has become, in effect, a sort of diurnal.
The earliest surviving collectar is not for the Roman but for the Old Spanish (Mozarabic) rite: the so-called Verona Orational from the early eighth century (Verona LXXXIX; ed. Vives and Claveras 1946).
Two fine collectars from eleventh-century England show what the collectar might encompass.
(i) London, British Library, Harley 2961 ('Leofric Collectar'), Exeter, third quarter of eleventh century (text ed. Dewick and Frere 1914-21):
collectar, including antiphons and responsories, often in full hvmnal eight sequences, with the start of a ninth before the manuscript breaks off incomplete
(ii) Cambridge, Corpus Christi College 391 ('Portiforium of St Wulstan or 'Collectar of St Oswald), Worcester, second half of eleventh century (most of text ed. Anselm Hughes 1958-60):
kalendár
psalter, canticles, litany hymnal
monastic canticles (for the third Nocturn of the Night Office)
collectar: Temporale and Sanctorale (from St Stephen, 26 Dec, to St Andrew, 30 Nov.), including notated chants
supplementary collects for a much fuller Sanctorale (St Silvester, 31 Dec, to St Thomas, 21 Dec.)
collectar:  Commune  Sanctorum and Dedication of the Church (without notation)
blessings and ordeals At this point a new hand has copied private prayers, Latin and Anglo-Saxon. Then a new section begins in yet another hand (or perhaps the first hand in a later phase), giving parts of the office as in a breviary, including the Night Office, with full lessons and notated chants:
Commune Sanctorum
Sunday Vespers and Night Office
summer Sundays
Trinity Sunday
commemorative offices of the Holy Cross and the Blessed Virgin Mary on Saturday; Office of the Dead At the end of the book a later hand has copied O-antiphons.
The thinking behind this particular assemblage of material is not always clear, and the manuscript in its present state may unite books originally prepared apart. At any rate, it allows some insight into the flexibility with which the office texts could be codified.
7. Sequentiaries, Tropers, and Kyriales
313
(iv) Office Ix'ctionaries
While the day hours had only the short chapters for lessons, the Night Office had nine (secular or Roman cursus) or twelve (monastic cursus) lengthy lessons. These were taken from various sources, and books were made to contain each type of lesson. The commonest pattern was: first Nocturn, the Bible; second Nocturn, sermon by one of the Church Fathers; third Nocturn, homily (a short disquisition upon a passage of scripture), usually by one of the four great Latin Church Fathers (Ambrose, Augustine, Jerome, Gregory). For many feasts of the Sanctorale, passages were read from the Vita of the saint in question, which might be gathered together in a legendär, passional, or martyrology.
Basic work on homiliaries has been done by Gregoire (1966, 1980). Saints' lives are published in the massive series Acta sanctorum (a convenient way to find texts is at the entry for the saint in question in Bibliotheca sanctorum). Perhaps the two most important martyrologies are the Hieronymian Martyrology of the fifth century (Acta sanctorum Nov. II, i. 1894; new edition Acta Sanctorum Nov. II, ii. 1931) and Usuard's Martyrology of the ninth century (PL 123, 453-992; PL 124, 9-860; Dubois 1965). (On martyrologies, see Quentin 1908.)
III.7. SEQUENTIARIES, TROPERS, AND KYRIALES
RISM B/Y71; Emerson, 'Sources, MS, II', AG.
Facsimiles of manuscripts containing sequences, tropes, and ordinary-of-mass chants with tropes:
Aachen, Bischöfliche Diözesanbibliothek 13: MMS 2 (omits gradual) Bamberg, Staatsbibliothek lit. 6: see Bamberg 6
Bari, Biblioteca Capitolare 1: MMS 1 (omits ordinary-of-mass chants)
Benevento, Archivio Capitolare 34: PalMus 15
Cambridge, University Library, Add. 710: MMS 4
Chartres, Bibliotheque Municipale 47: PalMus 11
Rome, Biblioteca Angelica 123: PalMus 18
Rome, Biblioteca Casanatense 1741: Vecchi 1955
private collection of Martin Bodmer, C. 74: Liitolf 1987
Editions with music: Paris, Bibliotheque Nationale lat. 1121: Paul Evans 1970 Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319: MM MA 2
Other comprehensive editions or studies: sources from Winchester: Frere 1894, Planchart 1977 Apt, Cathedrale Sainte-Anne 17 and 18: CT 5
Wolfenbüttel, Herzog-August-Bibliothek, Guelf. 79 Gud. lat.: CT 6
There are few written traces of sequences or tropes in the ninth century, though they were certainly sung then. The earliest trope and sequence collections of moderate size
314 ///. Liturgical Books and Plain chant Sources
to have survived are in manuscripts of the beginning of the tenth century onwards. Although this is not so different from the situation of other chants, the textual tradition of the latter goes much further back. The collection of sequence melodies in Chartres, Bibliotheque Municipale 47 (Brittany, late ninth century), the sequences and tropes in Paris, Bibliotheque Nationale, lat. 1240 (Limoges, 930s) and the tropes in Vienna, Nationalbibliothek 1609 (Freising? early tenth century) stand near the beginning of the written tradition.
Sequences, tropes, and ordinary-of-mass chants had their own individual history and their own special function within the liturgy, and hence were often copied in collections separately from other chants. Sometimes they were grouped alongside each other, so that combined volumes of 'special' material were created. These might include other chants as well, such as offertory verses, alleluias, processional chants, liturgical dramas, and so on. The usual name for such a book is 'troper', although tropes in the strict sense are not the only components. Books containing only-sequences, or sequentiaries, are less common; and books containing only ordinary-of-mass chants, known at least since the sixteenth century as kyriales, are even rarer. On the other hand, after the classical tropes for the proper chants of mass had fallen out of use in the twelfth century, it is common to find combined sequentiaries and kyriales. The use of the term 'troper' for these collections is less appropriate, for usually the only tropes present are a few for the ordinary-of-mass chants.
Most of the special material is for soloists to sing. By analogy with the older book containing the soloists' parts of the proper-of-mass chants (gradual and alleluia verses, tracts), the troper is sometimes referred to as a cantatorium; it is better to avoid this usage.
Like sequences and ordinary chants, most tropes are for the mass. Just as frequent as the collection of the special chants in a book apart is their inclusion in the gradual (or later in the missal). Since tropes for the office are restricted to prosulas for the melismas of responsories, the antiphoner was easily able to take these into its fold, and collections of prosulas in tropers are not common. When sequences, tropes, and ordinary-of-mass chants were included in an expanded gradual, they were usually copied in sections of their own. Some manuscripts, however—mainly Italian—give tropes and ordinary chants at the appropriate liturgical place among the proper chants. Sequences can occasionally be found among the proper chants outside Italy as well.
The general trend was away from separate collections of special material and towards its integration into the gradual or missal. After the thirteenth century tropers are rare, partly also because troping itself fell out of fashion.
The manuscripts described by Husmann in RISM B/V/l, Tropen- und Sequenzen-handschriften, do not include graduals (with one or two exceptions) or antiphoners. No representative listing of sequence sources is available, though a provisional one could be compiled by combining Husmann's information with that in Ia> Graduel romain, II: l^es Sources, which covers graduals and missals. For the proper tropes of mass, there are comprehensive lists of sources in Corpus Troporum 1 and 3 (ninety-
7. Sequentiaries, Tropers, and KyHales 315
two in the latter). Ordinary-of-mass chants are equally divided between tropers, graduals, and missals, again usually collected in sections apart, but occasionally amalgamated with the proper chants, or at least cued to their appropriate liturgical place by an incipit. Between them, the catalogues of Landwehr-Melnicki (1955), Bosse (1955), Rcinnau (1967), Thannabaur (1962), and Schildbach (1967), supplemented by Hiley (1986, 'Ordinary'), list over 500 sources of ordinary chants. 496 sources of responsory prosulas are listed by Hofmann-Brandt (1971).
The chief contents of few more or less typical sources may be listed briefly, in order to give an idea of their nature and variety.
(a) Tropers
1. Oxford, Bodleian Library, Bodley 775 (Old Minster, Winchester, mid-eleventh century):
graduals and proper tropes for the whole year, sometimes with alleluias. For each mass the usual order is: trope verses for the introit, gradual verse, alleluia (often only incipit), trope verses for the offertory, trope verses for the communion
ordinary-of-mass chants, troped then untroped
alleluias for the whole year
tracts
sequence melodies
sequence texts, notated At the beginning, after the sequence melodies, and at the end are further sequences and ordinary-of-mass chants, most with tropes, added in the twelfth century.
2. Munich, Bayerische Staatsbibliothek, elm 14322 (abbey of St Emmeram, Regensburg, 1030s):
litany hymns, Exultet
notated sequence texts with melodies in margin graduals and tracts alleluias proper tropes
ordinary-of-mass chants (Kyries, troped then untroped, Glorias, troped then
untroped, etc.) offertory verses (only the incipit of the respond is given) Ite missa est chants
3. Madrid, Biblioteca Nacional 288 (chapel of the Norman rulers of south Italy and Sicily, c.llOO):
tonary processional
Kyries, troped then untroped Glorias, troped then untroped alleluias sequences
offertory verses, some with prosulas
316 III. Liturgical Books and Plainchant Sources
Sanctus, troped then untroped Agnus, troped then untroped responsory prosulas
Benedicamus chants, some with tropes, Benedicamus songs liturgical dramas
offices of SS Julian, Egidius, and Mary Magdalene
4. Paris, Bibliotheque Nationale, lat. 909 (abbey of Saint-Martial at Limoges, c. 1025-30):
proper tropes
chants for the ordinary of mass, troped then untroped (Kyries and start of Glorias
lost) sequences tracts
communion psalm verses
processional
alleluias
offertory verses
tonary
office antiphons for summer Sundays, Trinity office Between the proper tropes and ordinary chants material was added which includes the offices of SS Martial, Valeria, and Austriclinianus of Limoges.
5. Paris, Bibliotheque Nationale, lat. 1118 (Auch?, late 10th c):
combined proper tropes and ordinary-of-mass chants, distributed in liturgical order
feast by feast tonary
prosulas (alleluia and offertory prosulas, also the Fabrice mundi responsory
prosulas, all in liturgical order) sequence melodies notated sequence texts
6. Paris, Bibliotheque Nationale, lat. 10508 (abbey of Saint-Evroult, Normandy, early 12th c):
incipits for introit, offertory, and communion throughout the year Kyries, troped then untroped Glorias, troped then untroped
sequences, often with gradual and alleluia for the mass in question
Sanctus, troped then untroped
Agnus, troped then untroped The rest of the manuscript contains six theory treatises, including Guido of Arezzo's Micrologus.
7. Rome, Biblioteca Casanatense 1741 (abbey of Nonantola, late 11th c): ordinary-of-mass chants (Kyries, troped then untroped, Glorias, troped then
untroped, etc.) fraction antiphons
8. Processionals 317
expanded cantatorium. For each mass the order is usually: trope verses lor the introit, gradual, tract (in Lent, etc.), alleluia prosula, sequence, antiphona ante evangehum, trope verses for the offertory, trope verses for the communion.
(b) Graduals
8. Modena, Biblioteca Capitolare O. I. 7 (Forlimpopoli, near Ravenna, 11 th— 12th C'):
gradual, with tropes, sequences, etc. integrated into each mass in liturgical order. As an example of the most elaborate festal mass, that on Easter Day includes the following: troped introit, troped Kyrie, troped Gloria, gradual, alleluia with prosulas, sequence, troped offertory, troped Sanctus, fraction antiphon, troped Agnus, troped communion.
supplement of festal pieces (ordinary-of-mass chants with tropes, sequences, proper tropes, etc.)
9. Paris, Bibliotheque Nationale, lat. 903 (abbey of Saint-Yrieix, near Limoges, second half of 11th c.)
gradual, with prosulas for alleluias and offertories processional antiphons, preces proper tropes
ordinary-of-mass chants (Kyries, troped then untroped, Glorias, troped then
untroped, etc.) sequences
In 2 and 4 each genre is separate; in 1 only the graduals and proper tropes are integrated. Proper and ordinary tropes are combined in 5. Sequences are to some extent combined with graduals and alleluias in 6. More comprehensive is the integration in 7, before 8 amalgamates the material completely. In 3 and 6 it may be seen that the various chants have been placed in their correct liturgical order (Kyrie, Gloria, sequence, Sanctus, etc.), though the collections are not amalgamated with one another. There are collections of processional chants in 3, 4, and 9; tonaries in 3, 4, and 5 (and theory treatises in 6); and saints' offices in 3 and 4.
III.8. PROCESSIONALS
Gy 1960; Bailey 1971; Huglo, 'Processional', NG. Facsimiles:
Worcester, Cathedral Chapter Library F. 160 (PalMus 12, 232-410); Processionale Sarum
Text editions:
W. G. Henderson 1875 (York); W. G. Henderson 1882 (Sarum); Legg 1899 (Chester); Wordsworth 1901 (Sarum)
Inventories:
Allworth 1970 (Donaueschingen 882—Dominican); Floyd 1990 (English monastic processionals)
318 ///. Liturgical Books and Plainchant Sources
The processional contains processional chants: antiphons (some with verses) and hymns (usually with refrains). The processions where they were sung took place (i) on certain days of special observance, (ii) before mass on feast-days (or at some other time during the day). The repertory was not large, and most early sources copy the chants required into the gradual at the appropriate place. Another place for them was in tropers, as has been seen in the previous chapter. From about the thirteenth century separate books for processional chants become more common, partly because the rubrics for performing the services become much more explicit about this time. The earliest surviving separate processionals date from the twelfth century.
Many of the processions for feast-days borrowed a responsory from Vespers or the Night Office of the day in question; these are not usually copied into the processional but would have been taken from the antiphoner. Conversely, the processional
Table III.8.1. Contents of the Castle Acre processional
Special ceremonies Feast-days
chants for processions after Vespers and the Night Office from the 1st Sunday of Advent to Epiphany
A. O crux benedicta for procession to the rood after Vespers on Sundays from Octave of the Epiphany to Passion Sunday and for the summer Sundays
chants for procession before mass from Advent to Septuagesima Sunday
Ash Wednesday (Blessing of the Ashes and Dismissal of the Penitents)
Palm Sunday (Blessing of the Palms)
Maundy Thursday (Mandatum)
Good Friday (Reproaches, Adoration of the Cross)
Easter Eve
1st Sunday after Easter
Rogation Days
Ascension Day to Corpus Christi
Purification of the BVM (2 Feb.) (Blessing of the Candles)
St Benedict (21 Mar.) to St Nicholas (6 Dec.)
further chants for procession in Advent, Lent, etc.
9. Missals
319
attracted to itself chants that formed part of ceremonies other than mass and the office hours, for example the Maundy antiphons, chants for the Veneration of the Cross on Good Friday and for the Easter Vigil (the Exultet, the hymn Inventor rutili). And because liturgical dramas also contain a processional element these too were sometimes recorded in processionals. As an example of what a well-stocked book might include, the contents of the fifteenth-century processional of the Cluniac priory at Castle Acre, Norfolk (Norwich Castle Museum 158.926.4e) may be summarized. Table III.8.1 lists the occasions in manuscript order in two columns? one for the special ceremonies, one for the feast-days for which a processional chant is recorded. On feast-days the processions took place after Vespers, after the Night Office or after Mass.
No comprehensive survey of the sources has yet been published (but Huglo is preparing one for RISM). Meanwhile Huglo's article 'Processional' (NG) cites several dozen sources, and Bailey (1971) lists the principal chants of many sources in tabular form.
III.9. MISSALS
Leroquais 1924; Le Graduel romain, II: Les Sources; Huglo, 'Missal', A:G; Emerson 'Sources, MS, IF, NG\ Andrew Hughes 1982, 143-59. Facsimiles:
Bencvento, Biblioteca Capitolare VI. 33: PalMus 20
Bratislava, City Archives, EC. Lad. 3, EL. 18 and other fragments: Missale notatum Slrigoniense
Missale Aboense (Dominican): Parvio 1971 Text editions (*signifies noted missal): printed missal of Hereford of 1502: W. G. Henderson 1874, Hetfordensis London, Westminster Abbey, 'Lytlington missal': Legg 1891-7
* Le Havre, Bibliotheque Municipale 330: Turner 1962
* Palermo, Archivio Storico e Diocesano 8: Terrizzi 1970
* Paris, Bibliothcque Nationale, lat. 1105: Anselm Hughes 1963
* Sarum: Legg 1916 (from three early manuscripts) Sarum: Dickinson 1861-83 (from printed editions)
York: W. G. Henderson 1874, Eboracensis (from printed editions)
Sacramentaries with chant incipits in margin:
* Oxford, Bodleian Library, Bodley 579: Warren 1883
Rome, Biblioteca Apostolica Vaticana, Ottob. lat. 313: Wilson 1915
Juxtaposed sacramentary and gradual: Paris, Bibliotheque Nationale, lat. 2291: chants listed Netzer 1910 Paris, Bibliotheque Nationale, lat. 12050: AMS Zurich, Zentralbibliothek, Rheinau 30: AMS
The term 'missal' is best reserved for the book containing all three of the main components of the mass: chants, prayers, and readings. 'Noted missal' can then
320 ///. Liturgical Books and Plainchant Sources
indicate a book where the chants are notated. Noted missals can be found from as early as the tenth century, but they are common only from the twelfth. The oldest is Baltimore, Walters Art Galley M. 6, which contains only festal and votive masses, written probably at St Michael's, Monte Gargano in the tenth century. Other early examples are also Italian, including Benevento, Biblioteca Capitolare 33 (PalMus 20).
Other countries did not take up the idea with any urgency. Many early northern books combine sacramentary, lectionary, and gradual in less integrated fashion. The juxtaposition of the gradual and other books, side by side but not amalgamated, is known as early as the ninth century. Huglo ('Missal', NG) has cited many examples. Numerous early sacramentaries from north France have chant text incipits copied in the margin, sometimes notated. A very few books survive where for each mass first the chants are given and then the prayers. Even rarer are books where the chants and prayers are intermingled in the correct liturgical order. Huglo is able to cite only three examples of the juxtaposition of gradual and lectionary. (See Plate 2 for the remains of one.) A few books are also known where gradual, sacramentary, and lectionary are juxtaposed.
There was at first no great practical benefit to be had from amalgamating gradual, sacramentary, and lectionary in one volume (though the situation whereby sacramentaries began at Christmas, graduals at Advent, could be rationalized, usually in favour of the gradual). But at the end of the eleventh century it became obligatory in Rome, and then customary elsewhere, for the celebrant to recite for himself the texts of all chants (Ludwig Fischer 1916, 80-1). And from the thirteenth century several leading churches, the most important being the papal chapel itself (others include Paris, Salisbury, the Dominicans) revised their liturgies and codified them in comprehensive rubricated manuscripts, which served as models for other uses.
Many missals were not designed to serve the cantor and remained without notation. A large number have notation only for the Canon of the Mass, sung by the celebrant. Another type of missal contains only masses for the highest feasts (sometimes called a 'missale festivum'). Such missals were usually for the use of a bishop or abbot, who might be expected to be present at his 'home' church on such high festivals.
Many missals with notation survive from churches whence no gradual is known, and are crucial witnesses for our knowledge of regional chant traditions (see the descriptions in Graduel rornain II and Emerson). Unnotated missals are also important for the study of chant repertories.
III.10. BREVIARIES
Baumer 1905; Batiffol 1911; Leroquais 1934; Salmon 1959; Salmon 1967; Huglo, 'Breviary', .VG; Andrew Hughes 1982, 197-224, 238-44.
Text editions: Cambridge, Magdalene College F. 4. 11: Collins 1969 printed breviary of Hereford, 1505: FYere and Brown 1904-15
11. Compendia 321
Oxford, Bodleian Library, Rawl. Lit. e. 1* and Gough Lit. 8: Tolhurst 1932-43 printed breviary of Salisbury, 1531: Procter and Wordsworth 1879-86 printed breviary of York, 1476: Lawley 1880-3
The breviary unites all the chants, prayers, and lessons of the office hours in one volume. The reason for assembling such a cumbersome volume—cumbersome both from its sheer size and also because of the frequent cross-referring always necessary in the office—seems to have been the urge to provide standard exemplars of the liturgical use of various churches. The obligation of travelling priests and monks to recite the office for themselves may also have led to the production of such books—the Rule of Chrodegang of Metz (d. 766) already directed that when a cleric could not attend one of the office hours he must say it privately. Most surviving early breviaries are nevertheless for use in monasteries.
A few early sources are juxtapositions of the component parts of the breviary, rather than amalgamations with all the material in correct liturgical order (see Huglo).
Many breviaries contain shortened lessons, a practice which gave the book its name. This practice is known from the papal court from the thirteenth century, where the office was in any case only recited privately. The breviary drawn up under Innocent III (1198-1216) is of this abbreviated type (see van Dijk and Walker 1960).
Noted breviaries are as valuable as antiphoners for the investigation of regional musical traditions. Breviaries without notation are also vital for our understanding of the development of local repertories, as Hesbert (CAO 5-6) has demonstrated.
111.11. COMPENDIA
The division of office from mass was not practised in the Mozarabic or Milanese rites, and their liturgical books amalgamate what in Gregorian usage would have been the antiphoner and gradual. This is rare among Gregorian sources, though a number of early sources juxtapose an unnotated gradual and an antiphoner: Albi, Bibliotheque Municipale 44 and Paris, Bibliotheque Nationale, lat. 17436, both of the ninth century, are two such. The 'Mont-Renaud manuscript' (PalMus 16) is a book of similar sort of the mid-tenth century (probably from Corbie), to which notation was then added about half a century later. Paris, Bibliotheque Mazarine 384 (Saint-Denis, early eleventh century) contains both a notated gradual and a list of chants of the antiphoner. Paris, Bibliotheque Nationale, lat. 12584 is a combined antiphoner and gradual (the antiphoner text edited by Hesbert, CAO 2) of the twelfth century from Saint-Maur-les-Fosses. Again, the two sections are not amalgamated. The book also includes a tonary. It follows Cluniac use, significant in view of the apparent interest' of the Cluniacs in creating compendia to include the entire liturgy.
Such compendia seem to have been attempts to codify the complete use of a particular church for reference purposes. They would have been of particular value to
322 ///. Liturgical Books and Plainchant Sources
the precentor of the church, who normally had charge of the performance of the liturgy.
One of the earliest combined noted breviary-missals is Cluniac: Rome, Biblioteca Casanatense 1907, from San Salvatore on Monte Amiate, of the early eleventh century. Another Cluniac example is the combined noted breviary-missal of Lewes, from the thirteenth century, Cambridge, Fitzwilliam Museum 369. Although a thick book with 517 folios, it uses a minute script, perhaps so that it would be portable, for use as a correctorium when the prior of Lewes visited other English Cluniac houses (see Leroquais, 1935; Holder, 1985). Another Cluniac combined breviary-missal, without notation, has survived from Pontefract or Wenlock: London, British Library, Add. 49363. (For a further noted breviary-missal, not Cluniac, see Salmon 1964.)
The newer religious orders also established master exemplars of their liturgy. Part of the Cistercian exemplar survives in Dijon, Bibliotheque Municipale 114 (see Choisselet and Vernet 1989, pi. 1 after p. 52 for the original list of contents). The master exemplar for the whole Dominican order still survives in Rome, Santa Sabina XIV. lit. 1, the 'Codex of Blessed Humbert of Romans'. This immense book (997 folios measuring 48 X 32 cm.) was compiled c.1260 to codify the liturgy revised by Humbert of Romans, master-general of the order, 1264-77. London, British Library, Add. 23935 contains the same material (minus the ordinal, breviary, and private missal) in small format, presumably for use as a correctorium on visitations in the English province.
The fourteen sections of the Santa Sabina manuscript are titled as follows:
Ordinarium
Martyrologium
Collectarium
Processionarium
Psalterium
Breviarium (for private recitation of the office, with only short lessons, psalm
incipits, and so on) Lectionarium (for the office) Antiphonarium Graduale
Pulpitarium (for one, two, or four friars, depending on the liturgical grading of the services, in the pulpit in mid-choir; it contains invitatories, responsory verses, gradual verses, tracts, and the Litany of the Saints)
Missale conventuale (for High Mass)
Epistolarium
Evangelistarium
Missale minorum altarium (for private masses)
From the same century three other English monastic compendia have survived. Cambridge, University Library Ii. 4. 20 is a combined breviary-missal from Ely (Benedictine). The two others are fully or almost fully noted. The best known is
//. Compendia 323
Worcester, Cathedral Chapter Library F. 160, substantial portions of which were published in facsimile in PalMus 12. The sections of the manuscript are:
antiphoner (Temporale)
Venites
processional
antiphoner (Sanctorale, Commune Sanctorum, Office of the Dead) Magnificat and Benedictus tones
(an added section of the fourteenth century with the office of the Visitation of the
BVM and Corpus Christi) kalendář
psalter with canticles, litany, and collects
hymnal
collectar
Kyries and Glorias gradual
sequentiary (mostly lost) Sanctus and Agnus Laudes regiae mass ordinal
For the performance of the liturgy only the prayers and lessons for mass and the lessons of the office would be required in addition to this.
The principal contents of London, British Library, Add. 35285, from the Augustinián priory at Guisborough, north Yorkshire, are as follows:
tonary
missal (without notation) kalendář
psalter (beginning lost) four ordines from the rituále Venites
antiphoner (Sanctorale and Commune Sanctorum)
office lectionary
tonary
processional
Possibly several originally separate codices were here bound together.
The stimulus to make a complete copy of a church's chant repertory might come from a liturgical or institutional reform, such as the rebuilding of the church in question. This was the case at Piacenza, for example, where after the cathedral had been rebuilt, the compendium Piacenza, Biblioteca Capitolare 65 was copied from 1142 onward. This contains a tonary and theoretical material, sets of tones for office and mass psalmody, a psalter, hymnal, antiphoner, gradual, and collections of tropes and sequences (see Gregoire 1968, 557; RISM B/III/2, 79; Huglo 1971, Ton aires, 174, etc.).
324
///. Liturgical Books and Plainchant Sources
III.12. PONTIFICALS AND RITUALS
(i) Pontificals and Benedictionals (ii) Rituals, Manuals, or Agenda
(i) Pontificals and Benedictionals
[Hiley]: 'Pontifical', NG; Vogel 1986, 225-71.
Pontificals are books containing the texts for services performed by a bishop outside mass and the office: dedication of the church, confirmation, ordination and consecrations, blessings of sacred objects, the sacring and crowning of monarchs, and so on. Chants are sung at these ceremonies—the series of antiphons sung during the lengthy ceremony for the dedication of a church are notable—and these are frequently notated in pontificals. Several of the earliest examples of neumatic notation survive in pontificals, for they were often splendidly decorated and carefully preserved from early times.
The early development of the pontifical has been illuminated by Rasmussen (1978) and its history described by Vogel, with emphasis on the development of the Roman-German type first compiled at St Alban's abbey, Mainz, in the 950s and adopted in Rome within a few years under imperial influence. (See the editions by Vogel and Elze, 1963, and Andrieu 1938-41. For descriptions of pontificals in French libraries see Leroquais 1937; for English ones see Frere 1901; the texts of numerous pontificals have been edited, particularly in the series of the Henry Bradshaw Society.)
One special action of a bishop was to pronounce a blessing at Mass after the Lord's Prayer and before Pax domini semper vobiscum. Collections of these benedictions have been preserved from the Middle Ages, in books known as benedictionals. Some benedictionals for the use of abbots have also survived. (See Baudot, 'Benedictionnaire', DACL; editions by Moeller 1971-9, and several volumes in the series of the Henry Bradshaw Society.)
(ii) Rituals, Manuals, or Agenda
Gy 1960; Vogel 1986, 257-64 (with list of editions).
The counterpart of the pontifical for priests is the ritual or manual, which contains the material for non-episcopal functions outside mass and the office. Prominent among the formularies are benedictions of all sorts and the services for baptism, marriage, and burial. Early rituals are often found in combination with collectars or sacramentaries, which gave the texts to be recited by the priest during the office or at mass respectively. Some rituals reflecting monastic use also survive.
14. Tonaries
325
II 1.13. ORDINALS AND CUSTOM ARIES
Neither medieval nor modern nomenclature is consistent on this point, but a general distinction may be made between (i) ordinals, which regulate the performance of the liturgy, specify which items are to be performed and sometimes the manner of their performance (singers involved, position, etc.), and (ii) customaries, which regulate the administrative organization and way of life of a religious community, secular or monastic. The ordinal may give invaluable information about the repertory of a particular church, and may even contain notated chant incipits, as well as giving details affecting performance. Special items such as liturgical dramas, or the use of polyphony, may be described. The customary may explain the duties and training of the cantor and singers.
Most surviving ordinals date from the thirteenth century or later. This is also the period when service-books were first regularly provided with copious rubrics governing the performance of the liturgy, and it is no doubt symptomatic of a general desire to codify aspects of the liturgy previously left to the practical knowledge of those involved.
No distinction is made between customaries and ordinals in the handlist of sources for over 120 institutions listed in Le Graduel romain II, 189-96. Others are given by Hänggi (1957, xxv-xxxvi) and Jacob (1970). Among modern editions those in the series Corpus consuetudinum monasticarum (CCM) and Henry Bradshaw Society are the most important. See also Angerer (1977) and Fassler (1985).
III.14. TONARIES
(i) Definition and Function
(ii) Type-Melodies for the Eight Modes
(iii) How Psalm Tones were Specified
Huglo 1971, Tonaires; Bailey 1974; Huglo, 'Tonary', NG. Modern editions of medieval tonaries: (i) manuscripts:
Leipzig, Universitätsbibliothek 1492 (15th-c. copy of a Reichenau exemplar of c.1075): Sowa 1935
Metz, Bibliotheque Municipale 351 (Metz, 9th c.): Lipphardt 1965
Naples, Biblioteca Nazionale VIII. D. 14 (Auvergne, 12th c.): DMA A.I
Paris, Bibliotheque Nationale, lat. 1118 (Auch?, late 10th c.) and 1121 (Limoges, early
11th c): C. T. Russell 1966 Paris, Bibliotheque Nationale, lat. 12050 (Corbie, 9th c): reconstructed in AMS, cxxiii-
cxxvi; cf. the reconstruction co-ordinated with the Mont-Renaud manuscript and
Laon, Bibliotheque Municipale 118 in Huglo 1971, Tonaires, 94-101 Paris, Bibliotheque Nationale, lat. 13159 (for Saint-Riquier, late 8th or 9th c): Huglo
1952, 'Tonaire', Huglo 1971, Tonaires, 26-8, Planer 1970
326
///. Liturgical Books and Plainchant Sources
Rome, Biblioteca Apostolica Vaticana, Pal. lat. 1346 (German, 12th— 13th c.): Donato 1978
St Gall, Stiftsbibliothek, 390-391 (St Gall, late 10th c.): facs. PalMus II/l various Sarum sources: Frere 1898-1901
Tonaries compiled from the contents of medieval antiphoners were published in PalMus 9 and 12 and Udovich 1985 (ii) theorists:
pseudo-St Bernard of Clairvaux: Tonale sancti Beniardi: GS ii. 265-77; PL 182, 1154 ff.
Berno of Reichenau: GS ii. 79-91
Frutolf of Michelsberg: Vivell 1919
Jerome of Moravia: CS i. 1-94; Cserba 1935
John ('Cotton' or 'of Afflighem' or of Passau): CSM 1; trans. Babb 1978 Odo of Arezzo: GS i. 248-8 (introduction); CS ii. 81-109 Odorannus of Sens: Bautier et al. 1972 Petrus de Cruce of Amiens: CS i. 262-8; CSM 29
pseudo-Odo in Saint-Die, Bibliotheque Municipale 42 (Franciscan, 14th c.): CS ii. 117-42
Regino of Prüm: CS ii. 3-73
Rudolf of Saint-Trond or Franco of Liege: Quaestiones in musica: Steglich 1911 Walter Odmgton: CS i. 182-250; CSM 14 ~"
(i) Definition and Function
A tonary (tonarius and, sometimes later, tonale, are the medieval terms employed) is a list or series of chants in tonal order. The earliest of them, from the late eighth and ninth centuries, are as old as any liturgical books of chant texts. Being an abstraction from everyday musical practice, though frequently serving a practical need, they were often compiled by theorists who are known also for other theoretical writings.
In tonaries chants in the same mode were listed in series, disturbing the order in which they would have been sung in the church year. It was particularly common to make lists of pieces sung with psalm verses: introits and communions of mass, antiphons of the office. A further subdivision of the series was usually effected, so that those pieces were grouped together which required the same cadence (differentia) for their psalm verse(s). Ex. III. 14.1, an extract from the tonary in Paris, Bibliotheque Nationale, lat. 776, shows how the series were usually presented. The manuscript has listed chants in the first two modes, and has now begun a series for mode 3. A tone for singing the Gloria patri is copied (given at the start of Ex. III. 14.1), followed by the incipits of fifty-nine office antiphons. Ex. III. 14.1 gives the last eight of this series, starting with Tenemus ecce arnia. There follow the first eight examples for the second 'varietas', and all the incipits for the other three sets.
Such lists are found in a great variety of sources. The chief contents of Paris 776, from which Ex. III.14.1 is transcribed, are a gradual; tonaries were often copied in other such liturgical music books, as for example in:
14. Tonaries 327 Ex. III. 14.1. Extract from a tonary, part of the third tone (Paris, Bibl. Nat. lat. 776, fo. 152r)
INCIPIT TERCIUS TONUS
ta*9w**t*9m- •    9   9   9 ^   9^ 0
Glo-ri-a patri et fi-li-o       et spiri-tu-i sancto
0  0»    0    0090ä*      9    0     * __*#**»##_-, 0
0    * V   * *Q
Sic-ut e-rat in princi-pi-o et nunc et semper et in secula seculorum A-men.
> 0 * _    -   B   9    0 0
Tenemus ecce ar - ma.   Sancta legi-o Agau(nensis).   Fidelis seruus et.
lustus germinabit. Tollite portas prin(cipes). Uideo uirum. Leto animo. Sancti qui sperant. VARIETAS .II.
[J -»				
a-a*—*-9 'm m 0	0   ' m -B-	-ie——•—		0 9*9i~t> 0-
p *	-*0-•*-	• u* '-	-•—* —*—*9-	
Glo-ri-a     seculorum A-men.        In e-o-dem namque.   Conuocatis er-go.
—0—00-•—00-0—00-00-
Quando natus est.   Qui de terra est.   Surrexerunt autem.   Hierusalem Iherusalem.
0 00 9 0 t*»t
—•—*f-*~~*9-
Ualerius igitur. Beatus Uincentius. VARIETAS .III.
0 0. 0 _»  m 0 0 0 -M* 0
' 00—0-- i\—0 0      0--m—-a-0-0—0-
I      <r?    '  ' '* %   09 0*0 0'*' ' * 0* / * r ,*
Seculorum A-men.   Ecce concipies. Hec est quern nesciui. Ascendit   Simon Petrus. .IUI._
~       M9 9      9   9—   0   m0   m m
m   0 9 9      00    0     - 0  9   r* m —0-
* *    *0 0   0 0*0 0*09 0 0
Glori-a     seculorum A-men.     Sapientia magnificat. Cum ergo. .V.
9 9M 9   0 m    0   0 a     0   0 M        »0 0 B    §9 9-
0 0"      00Q      0 0 0     0 0      0      0 0      *      0     *^--0-# 0r-
Seculorum A-men.     Fac benigne. Homo quidam. Du-o homines. Cum repente.
0    9  0&*  *    •       0  0  9 0i* •     0    9    0**      0   r*   *'   *    0 ä'   '    '   fO      09 06&Z
Quasi u-nus de. Mari-aer-go. Dornum istam. Pastor bonus. Ego sum pastor. Simon Petrus.
328 ///. Liturgical Books and Plainchant Sources
Monza, Basilica S. Giovanni C. 12/75 (gradual and antiphoner, Monza, early 11th c.) London, British Library, Add. 30850 (monastic antiphoner, Santo Domingo de
Silos, early 12th c; facsimile Antiphonale Silense) Eichstätt, Bischöfliches Ordinariatsarchiv, pontifical of Bishop Gondekar II (1057—
75) (the inclusion of a tonary in a pontifical is altogether unique)
Many tonaries appear in tropers, indicating their f)lace among material for the musician with special skills and interests:
Paris, Bibliotheque Nationale, lat. 1118 (Auch?, late 10th c.) Bamberg, Statsbibliothek, lit. 5 (Reichenau, 1001) Madrid, Biblioteca National 288 (Palermo, c.1100)
Several authors of theoretical treatises also compiled tonaries. Regino, abbot of Prüm and later of St Martin in Trier, wrote a treatise and tonary c.901 for Archbishop Radbod of Trier, in the form of a letter to Radbod. Regino's is the first tonary by a known author. The treatise of Berno, abbot of Reichenau 1008-48, was written for Archbishop Pilgrim of Cologne (1021-36).
It may be pointed out, in view of the confusion present in the older literature, that Odo, abbot of a monastery in Arezzo in the late tenth century, does not come into this category. Odo appears to have been the author of the most widely known of Italian tonaries (CS ii. 78-109, edited from Paris, Bibliotheque Nationale, lat. 10508, from Saint-Evroult, twelfth century, unfortunately a version far removed from the archetype). But he did not compose the well-known treatise Dialogus de rnusica (written by an anonymous monk in north Italy, early eleventh century; GS i. 252-64). The tortuous history of these texts was elucidated by Huglo (1971, Tonaires, 182-224, and 1969). The misattributions so rife in this case result from the common practice of including tonaries in compilations of theoretical writings. Such a compilation is Munich, Bayerische Staatsbibliothek, elm 14272, containing Boethius, Musica enchiriadis, Commemoratio brer is, Alia musica, and a treatise on the monochord, as well as a tonary (St Emmeram at Regensburg, mid eleventh century).
Theorists who discuss the eight-mode system frequently cite examples, which may amount almost to a small tonary in themselves, though rarely giving the actual recitation tones. The celebrated anonymous treatise Musica enchiriadis (north-east France, late ninth century), for example, gives the incipit of just one antiphon per mode. The tenth-century Einsiedeln source of the Enchiriadis, Einsiedeln, Stiftsbibliothek 79, then takes the step of attaching a proper tonary to the treatise (see Huglo 1971, Tonaires, 62)—or one might say that the treatise has been brought in as a preface for the tonary, since many tonaries carry an introductory text explaining the modes. (For another example, where passages of the treatise Dialogus de musica are drafted into the tonary of the fifteenth-century Utrecht antiphoner Utrecht, Bibliotheek der Rijksuniversiteit 408, see Huglo 1971, Tonaires, 183. The twelfth-century tonary in the Utrecht antiphoner Utrecht 406 borrows from both the Dialogus and Berno of Reichenau: see Huglo 1971, Tonaires, 203.) Even closer to the
14. Tonaries
329
tonary in spirit is the Commemoratio brevis (ed. Bailey 1979), 'a veritable treatise on psalmody', as Huglo puts it (1971, Tonaires, 64).
One commonly adduced reason for the compilation of these lists of chants in tonal order is that they were a convenient codification, for the purposes of reference and learning, of the tone and differentia to be used when singing a psalm with each antiphon listed. Given the huge number of antiphons used during the liturgical year (3,000 or more), and the restricted number of tones and endings (around forty), it is not surprising that musicians felt the need to set down in some convenient form the decisions about intonations which they were accustomed to taking. But this simple explanation is not adequate for all types of tonary.
The order in which the compiler of a tonary encountered the chants would have been a liturgical one. Most tonaries, once they have broken down the original order into groups by tone and differentia, disturb the liturgical order no further. Thus within each group there remains a liturgical succession. For example, the ninth-century Metz tonary Metz, Bibliotheque Municipale 351, lists the following twenty-two antiphons under the third differentia (or diffinitio, as the manuscript calls it) of tone 3 (taken from edn. by Lipphardt, 1965, 34-5; * denotes addition). I have appended the liturgical occasion for each chant:
Tu Bethleem 4th Sunday in Advent
Ascendente Ihesu in navim 4th Sunday after Epiphany
Fac benigne in bona 3rd Sunday of Quadragesima
Si in digito Dei 3rd Sunday of Quadragesima
Lignum vite Inventio S. Crucis (3 May)
Tuam crucem adoramus Inventio S. Crucis (3 May)
Claudius quidam St Peter (29 June)
Petrus autem servabatw St Peter (29 June)
Et respicientes viderunt Easter Sunday
Et intravit cum illis Easter Monday
Mult a quidem et 1st Sunday after Easter
Mercennarius est 2nd Sunday after Easter
Ego sum pastor bonus 2nd Sunday after Easter
Si quis diligit me Pentecost Sunday
Si oportuerit me mori Palm Sunday
Beatus vir qui inventus St George (23 Apr.)
Domine spes sanctorum et turns Vigil of All Saints (31 Oct.)
Serve bone et fidelis* Confessors
Iste sanctus pro lege Dei sui a Martyr
Omni tempore de libro Judith
Quidam homo fecit 2nd Sunday after Pentecost
Sancti qui sperant* Martyrs
In fact there are three series here. The first follows the year from Advent to. midsummer, the second goes back to Easter and continues to Pentecost, and the third
330 ///. Liturgical Books and Plainchant Sources
doubles back again, to Palm Sunday. (The two added antiphons are slightly out of place: Seixe bone, for Confessors, should follow antiphons for Martyrs, and the last antiphon should follow Iste sanetus.) The three groups appear to be distinguished by their openings:
G ab b, G a Gb b, etc.
G G Gc a G, etc.
a mixture of the two previous?
There is a certain small disadvantage in retaining a liturgical order within the various groups of chants, for it is not always easy to locate any one piece, at least within very big series like those for tone 1 and tone 8. An alphabetic order would be more useful if a little-known chant were being sought. However, medieval musicians seem not to have felt a need for this, for only in Reichenau sources was it adopted: Bamberg, Staatsbibliothek, lit. 5 (written 1001); the tonary of Berno of Reichenau (second quarter of eleventh century); and the tonary of c. 1075, notated in Hermannus Contractus' interval notation, now surviving only in a fifteenth-century copy (ed. Sowa 1935; see Huglo 1971, Tonaires, 254). (One version of Berno, Vienna, Nationalbibliothek 1836, from Austria, eleventh to twelfth century, actually restores the antiphons to their liturgical order—see Huglo 1971, Tonaires, 272.)
Perhaps, then, the tonary is not so much a reference book for locating the odd troublesome item as a check-list to aid the rehearsal of pieces with similar musical characteristics. Few tonaries attempt the classification of all the antiphons of the year, and most contain only short lists, citing only a few dozen chants from the bigger groups and only a handful of the less common varieties. Thus the majority of tonaries, of the short type, serve neither to specify tones for the whole antiphon repertory, nor to classify melodies into melodic families; their more modest aim seems to be simply to remind musicians of the melodic idioms of the main families. The function of specifying tones was in any case carried out effectively by other means (see (iii) below).
The majority of tonaries contain lists of antiphons. A significant number have introits and communions, and responsories are sometimes included—for they too have a set of elaborate verse tones. But it is clear from even the earliest tonary that more was involved in their compilation than the assignation of psalm tones, for several contain chants that do not have psalm verses. Thus Paris, Bibliotheque Nationale, lat. 13159, from Saint-Riquier in the late eighth (Huglo 1952, 'Tonaire', and 1971, Tonaires, 25-9) or early ninth century (Planer 1970), has a few examples in each mode of the following categories: introit, gradual, alleluia, offertory, and communion. This source seems to be intended as a demonstration of the eight modes, rather than as a catalogue of tones. Paris, Bibliotheque Nationale, lat. 780 (Narbonne, late twelfth century) has alleluias, offertories, invitatories, and processional antiphons beside the more usual chants. The extensive listings in Montecassino, Archivio della Badia 318 (Montecassino, twelfth century) even include sequences (studied by Brunner 1981).
In such sources one may discern a didactic or scholarly intention beyond the
/ 4. Ton arte s
331
specification of psalm tones. An altogether exceptional compilation which shares this character is the famous 'Dijon tonary', possibly in part the work of William of Dijon (cl. 1031), Montpellier, Faculte de Medecine II. 159 (facs. PalMus 8; edn. Hansen 1974). In this source, after a short tonary for antiphons and responsories, a complete repertory of chants for the proper of mass is copied in full. The pieces appear in the following order:
introits and communions: modes 1 to 8 (that is, introits mode 1, communions mode
1, introits mode 2, communions mode 2, etc.) alleluias: modes 1 to 8 tracts: mode 8
graduals: modes D to G (no separation of authentic from plagal modes) offertories: as for graduals tracts: mode 2
For two groups of pieces, the range is specified in the margin, and there is also an indication as to whether the piece uses B or Bb—these are unique to this source. WTithin each modal group, the chants are ordered according to two more criteria: (i) according to their starting note, from lowest upwards; (ii) then, according to their highest note. This remarkable manuscript may be regarded as a master copy of the repertory, of a special kind, organized not according to liturgical principles but abstract musical ones.
The name tonary is also used for a simple set of psalm tones or other tones, without indication of anv of the chants which control their use.
(ii) Type-Melodies for the Eight Modes
The aim of bringing to mind some of the characteristic melodic contours of melodies in a particular mode seems also to lie behind three peculiar sets of chants known almost exclusively from tonaries and theoretical sources. They are placed at the head of each group of antiphons or other chants in a particular mode.
1. The first, particularly intriguing, set of phrases (only about a dozen notes in length) has peculiar pseudo-Greek texts: 'Nonanoeane', 'Noeagis', etc.
2. A second set of longer melodies resembles single-phrase antiphons in melodic style, and they have Latin texts which allude to the number of their tonal group. Thus for tone 1, the chant usually given is Primum quaerite regnum Dei ('First seek ye the kingdom of God', cf. Matt. 6: 36). A different set of such chants is found in St Gall, Stiftsbibliothek 390-391 and related sources, where the first text is Primum mandatum amor Dei est ('The first commandment is the love of God', cf. Deut. 6:5). Another set found in later German sources is given in Huglo 1971, Tonaires, 421.
3. Finally, both the Latin and the pseudo-Greek chants may be suffixed by a melisma or neuma, covering similar melodic ground. (On the various meanings of 'neuma' see below, IV.I.iii.)
332 ///. Liturgical Books and Plainchant Sources
Ex. III.14.2. Type-melodies from a tonary (Pans, Bibl. Nat. lat. 1121, fos. 201v-205v)
AUTENTUS PROTUS   AUCTOR1TAS PRIMA
-' "»—m '
No-an-no - e - a - ne Pri-mum quaerite reg-num De-i
PL AG A PROTI   HOC EST IUNIOR PRIMI
No-   -e - a-gis Se-cundum au-tem simi-le est   hu - ic
AUTENTUS DEUTERUS   HOC EST AUCTORITAS SECUNDA
*—« 0 .-V,.—	—** *       ** 0 0—w—r*-2-
m—»—*--—■—**—--•*-**r*—», —	-*•—*-r—--*-«-.0 -
No-i -  o-e-a-ne Terti-a    di-esest facta sunt
quod haec
PLAGA DEUTERI
No-        -e - a - is Quar-ta ui-gi - li - a   ue-nit  ad e - os
AUTENTUS TRITUS   ID EST AUCTORITAS TERTII
No-i - o - e - a - ne Quinque pru-dentes intrauerunt ad nuptias
PLAGA TRITI
m  m„   *»r '"m      0 .    •*   '  * mm*-
-w0-— 0 0
No-    -e - a-gis Sexta    ho-ra   se-dit super pu-te-um
AUTENTUS TETRARDUS   HOC EST AUCTORITAS .1111.
U           —         —                                  — __                                      -»     . _                                   —__.__._0%_mm---m-mm-—---				
As               *   *      m   m   m*'  '       m, m	*'*m    00ug-=--		'       00 w	'*  0 0%   0 0
w--                             • ** No-i - o- e- a- ne	0 Septem	sunt	spiri-tus an	-te thronum Dei
PLAGA TETRARDI   OC EST IUNIOR	PRIMI			
~*-"TT-Xl* **  m0-00~		*	90       . m0	
fiH       •       0wn«       0,        0                           "0 00     m 00	1**0-. 00 "1,	•	'-«#-	—•-
yzr                                           m -No-e-    -a - gis	Octo sunt	be -	a - ti - tu -	di - nes
Ex. III. 14.2 is a transcription of all three sets of melodies as they appear in the early tenth-century Limoges troper Paris, Bibliotheque Nationale, lat. 1121. The neuma is attached to both the pseudo-Greek and the Latin pieces. After each batch of type-melodies comes a set of antiphon, responsory, introit, and communion incipits.
Bailey, who has studied and made a critical edition of all the three sets of type-melodies from a comprehensive range of sources, dates the introduction of the pseudo-Greek phrases to the early ninth century (see Bailey, 1974, 5-8), and believes (with other scholars) that they are of Byzantine origin. Aurelian of Reome, who is the first to mention them, in the mid ninth century, speaks of them as Byzantine; and melodies of remarkably similar type are known from Byzantine sources (albeit of the twelfth century and later), the echemata (see Raasted 1966, and the parallel copies in Huglo 1971, Tonaires, 384 and Bailey 1974, 12). Bailey points out that the Byzantine
14. Tonaries
333
echemata might be sung by a cantor to bring to mind the characteristic melodic idioms of a mode, before he embarked upon a liturgical item; a similar function may be intended for these Western echemata.
The neumae are probably slightly later in date, while the Latin compositions are not known from earlier than the mid tenth century.
Of all the type-melodies, only the neumae were used hturgically, as special prolongations of the antiphons for the Benedictus, Magnificat, and Quicunque vult, and for the final antiphons of Vespers, Lauds, and the nocturns of Matins (see Frere 1898, ii. 209, for the instruction in Sarum ordinals). In time they seem to have acquired an almost mystical significance, as representations of the 'breath' of the Holy Spirit (see Fluglo 1971, Tonaires, 389). Not surprisingly, perhaps, in the context of extravagant liturgies of the lie de France of the twelfth and thirteenth centuries, they acquired prosula texts and polyphonic upper voices (motets in the Montpellier and Bamberg manuscripts). New melodies of similar type might even be composed. In Laon, Bibliotheque Municipale 263, containing a festal Epiphany liturgy, the kyriale is preceded by a set of tones for the Gloria of the introit. To each tone a melisma is attached, and then there follows a prosula for the melisma. Ex. III. 14.3 gives those for tones 5 and 6 (1-5 are short, 6-8 long).
(iii) How Psalm Tones were Specified
The tonary is only one of several ways in which tones and differentiae were codified. Very many antiphoners actually copy after each antiphon the start or ending to be used, over some abbreviation of the 'Gloria patri' such as 'e u o u a e' ('sEcUlOrUm AmEn'), or simply 'Amen'. Many old sources use code numbers (I to VIII) or letters to indicate the same thing. For example, three graduals of the early tenth century, none originally intended to have notation, Paris, Bibliotheque Nationale lat. 12050 (Corbie), Mont-Renaud (Corbie, notated later in tenth century), and Laon, Bibliotheque Municipale 118 (St Denis), use the following system to indicate the psalm tone to be used for introit and communion (complete integrated lists in Huglo 1971, Tonaires, 94-101):
AP or APR = Authentus Protus PP, PLP, or PLPR = Plagalis Proti
AD = Authentus Deuterus
PD or PLD = Plagalis Deuteri
ATr = Authentus Tritus
PTr = Plagalis Triti
ATD = Authentus Tetrardus
PTD or P1TD = Plagalis Tetrardi
An important group of Swiss sources (studied by Omlin, 1934, with additions by Huglo 1971, Tonaires, 232-51) uses vowels to indicate the tones: a, e, i, o, u, H, y,« co, and consonants for the differentiae: b, c, d, g, h, k, p, q. This system is first seen
334 ///. Liturgical Books and Plainchant Sources
Ex. IIL14.3. From the Gloria tones (Laon, Bibl. Mun. 263, fo. 21v)
QUINTUS TONUS
-■->-•-z--»i 0m-•—--•-m-
-*—•-•—•—-* *t mm—-w-*-*—«—
m      - m0*0m0 m0       %• 0    *    0    m    * * •
Glo-ri-a       se-cu-lorum A-men. Quem glori - f icant ce-!i glo-ri-o-sa
fr—i-	-	•	/*~\ _	00
	r • ' . *—	-BIT	—0 0'***—,—<	- m
uir-tutum ag-mina. Lo - que - bar.       Le-ta  -  re   Ihe - ru-sa - lem.
SEXTUS TONUS_ _
£•   0.   0* m   0   0>m   m . i~0 00b,0**0  S0i»\9 0*0 0 * '*00bj-9 9Z=
Glo - ri - a        se-cu-lorum   A - men.
..ltM ?u*0 rn'0     *s >^ a> * • * • u
Pa-ter e-ternus
-,-.-,-,        0 0—0—-
co-e-ternus       e-ius fi - li- us   e - o-rum compar qui permanet in u - ni-ta-te spi-ri-tus
b#   , *   *    »  t>*   . ** ***** **** J==
w   0 ' 0 -
hi tres u-nus Deus   a quo cuncta salus      perso-nis tribus     est ui-gor u-nus
-0-0-0-'-0-
* 0 J 0
per quem u-ni-uersus    or-bis ue-ge-tatur       su-per omoes Deus quique co-li-tur *00\tm9'0m- Qp    mat   #»»****    000 ~
in se-cu-la be-ne-dictus. In     me - di-o ec-cle-si-e.
in the early eleventh century, in the tonary attached to Hartker's antiphoner (St Gall, Stiftsbibliothek 390-391 copied, fortunately before Hartker's work lost many pages, in St Gall 388), then in many later St Gall sources, and books from Rheinau, Engelberg, Einsiedeln, Disentis, and also from churches in south Germany, Austria, and north Italy. The system survived into the seventeenth century and was actually-revived in the Antiphonale monasticum secundum traditionem Helveticae Congrega-tionis Benedictinae of 1943.
The Breton gradual Chartres, Bibliotheque Municipale 47 (<r.900) uses Roman numerals to specify introit tones, as do many medieval sources. But it also uses a shorthand method of indicating the melodic peculiarities of some melodies. Huglo (1971, Tonaires, 105-8) explains the most common entries as follows:
15. Identifying and Describing Chant-Books 335
the letters z, m, and ul refer to the start (initium'), internal cadence ('medium'), and
final syllable ('ultima') of the chant; the mode is then stated, with the words i ('in') or em i ('eminet in' = 'stands out as').
Thus 'i em i HIT means that the chant starts as a mode 4 piece; 'm i VII' means that it has an internal cadence suggesting mode 7.
There are many medieval antiphon lists whose chief function, besides specifying the selection of chants for the liturgical year, is to codify the psalm tones and endings to be used. These are easily confused with tonary lists, but their true nature is clear when it is realized that the chants are listed in liturgical, not tonal order. (See Huglo 1971, Tonaires, 22-3.)
III.15. IDENTIFYING AND DESCRIBING CHANT-BOOKS
Each different type of liturgical book has its own content and characteristics. Without claiming to cover all important points, I attempt here a brief summary of the features of chant-books which are customarily the object of investigation, and on which one hopes to find information in catalogues, inventories, and handlists.
The three most important pieces of information about the book are: the type of book; its provenance; its date. Manuscript sources almost never carry any title or explicit to make these clear.
The type of book is of course determined by its contents, which after the preceding chapters hardly needs further comment. Experience shows, however, that missals and breviaries should always be described as 'noted' or 'without notation', and the extent of the notation made clear.
The date is partly a matter of palaeography, the accurate chronological identification of the text and music scripts. Partly it is a matter of estimating the significance of the saints represented in the Sanctorale or kalendar, or the date of the latest items in the source.
The biggest problems are often posed by the provenance of the source. A threefold distinction may be observed here, between:
1. the derivation of the source, that is, the liturgical tradition to which it belongs;
2. the place where it was prepared, which may be different from
3. the place where it was used.
Thus Rouen, Bibliotheque Municipale 277, a noted missal of the mid-thirteenth century, can be shown to have been copied in Paris (Branner 1977), but follows the use of Rouen cathedral and was presumably prepared for use there.
Rubricated books may contain information about the personages involved in the performance of, for example, the ceremonial of Holy Week. For example, if a bishop and 'clerici' are involved, the book would have been used in a cathedral rather than a monastery. For graduals and missals the distinction between secular and monastic use is often difficult to make. Kalendar and litanies (if present) and the Sanctorale should
336 ///. Liturgical Books and Plainchant Sources
be inspected to see if particular attention is paid to monastic saints (especially St Benedict and his sister St Scholastica). Office books will naturally follow either secular or Roman cursus or the monastic cursus. Other clues may be gleaned from processional ceremonies, for example those of Palm Sunday, which visit different stations or churches; these may be specified in the source. The date of the service for the dedication of the church may also be marked in the kalendar or be deducible from its position in the Sanctorale, and in rare instances this can lead to an identification of the provenance of the source.
The saints to whom attention is paid are important evidence for the derivation and destination of a source, not so much those who were universally venerated, such as the apostles or early martyrs, as saints of purely local importance. They may be found in three places: (i) in the kalendar; psalters, missals, and breviaries often have a kalendar, whereas graduals, antiphoners, tropers do not; (ii) the litanies of the saints in Holy Week and Ascension week; (iii) the Sanctorale; not only graduals and antiphoners (as well as missals and breviaries) have a Sanctorale, but also collections of sequences and tropes. Although a psalter has no Sanctorale, it may be joined to a hymnal which does.
Repertorial comparisons between sources are a powerful tool for the investigation of derivation and destination. Although it would be possible to list the complete contents of a source and compare it with others, in practice it is necessary to be selective and concentrate on those parts of the repertory which most often differ from source to source. As far as chants are concerned, the following may be picked out.
For mass:
1. Alleluias, including their melodies, which may be identified with the aid of Schlager's catalogue (1965). Text incipits of the series for Easter week and the Sundays thereafter, WThitsun week and the Sundays thereafter should always be noted. Lists for comparison with other sources have been accumulated by various scholars but are unfortunately not available in print in large numbers (for examples see Husmann 1962 'Studien', and Hiley 1980-81).
2. Sequences, including, for rhymed sequences, their melodies (the melodies of older sequences do not vary).
3. Ordinary-of-mass melodies, and tropes, which should be identified with the aid of the catalogues of Landwehr-Melnicki (1955), Bosse (1955), Thannabaur (1962), and Schildbach (1967), supplemented by Hiley (1986, 'Ordinary').
4. Tropes and any other festal material.
For the office:
1. The choice of office chants varies greatly from source to source. In practice it is advisable to restrict one's attention to parts of the repertory already investigated, such as the Advent responsories and their verses listed by Hesbert (CAO 5-6).
2. Hymns, whose melodies should be identified with those published by Stablein (MM MA 1).
15. Identifying and Describing Chant-Books 337
The way in which repertory comparisons can illuminate the derivation of a source may be demonstrated with the aid of the twelve sources edited in CAO. Let us assume for the moment that the provenance of the gradual and antiphoner published in facsimile in PalMus 16, the 'Mont-Renaud manuscript', is unknown. In order to establish the affiliations of the manuscript we should ideally compare it with as many manuscripts as possible, which is what Dom Hesbert did in volumes 5 and 6 of CAO (manuscript 326/728). For present purposes the twelve sources edited in CAO 1 and 2 will suffice.
The responsories of the Night Office for St Andrew (30 November) provide a suitable point of comparison. The twelve sources of CAO 1-2 have between them sixteen different responsories (formulary 120, 117). I have given them in Table III. 15.1 in the order in which they appear in the oldest source, the Compiegne gradual-antiphoner Paris, Bibliotheque Nationale, lat. 17436. Figures indicate the order in which they appear in the other manuscripts. The Mont-Renaud manuscript (fo. lllr-v) appears in the last column.
Table 111.15.1 Responsories in the Night Office of St Andrew: CAO and the Mont-Renaud manuscript
	C	B	E	M	V	H	R	D	F	S	L	MR
Dum de/per-ambulant	1	1	1	1	1	1	1	1	1	1	1	1
Venite post me	2			2	7		2	2			2	2
Mox ut vocem	3	2	2	3	2	2	3	3	2	2	3	3
Oravit sanctus Andreas	4		10		11		9	8	7	3	10	8
Homo dei	5	3	4	6	4	3	4	4	5	5	4	4
0 beata/bona crux	6	7	5	7	5	7	8	7	6	6	9	7
Doctor bonus	7	4	3	4/5	3	4	5	5	4	4	7	5
Expandi manus	8	5	6	8	6	5	10	10	8	7		10
Salve crux	9	6	9			6	7	6	10	10	8	6
Dilexit Andreám	10	8	8	9	8	9	12	11	12	12	11	11
Vir iste	11	9	7	11	9	10	11	12	11	8		12
Cum videret/vidisset	12					8	6		3	11	5	
Dum penderet	13											
Videns/Vidit crucem	14		11	10	10						6	
Beatus Andreas de cruce								9	9	9		9
Cives apostolorum											12	
In most sources the responsories are grouped for the three nocturns in three, fours, or larger groups, sometimes reflecting the secular or Roman cursus, sometimes the monastic, but sometimes with several 'extra' responsories, which could be selected as the precentor saw fit. This is most obviously the case in the Compiegne gradual-antiphoner. It has almost all the responsories found elsewhere. Three of the sources
338 III. Liturgical Books and Plainchant Sources
simply copy all responsories together after the antiphons. (The shelf-marks of the sources are given at the start of III.5; only their provenance is given here; 'G', Durham B. III. 11, has a lacuna at this point.)
C		Compiegne	4 +	3 + 7
B	=	Bamberg	3 +	3 + 3
E	=	Ivrea	3 +	3 + 5
M	=	Monza	4 +	4 + 3
V	-	Verona	3 +	4 + 4
H	-	St Gall (Hartker)	3 +	3 + 4
R	=	Rheinau	4 +	4 + 4
D	=	Saint-Denis	12	
F		Saint-Maur-des-Fosses	4 +	4 + 4
S	-	Silos	4 +	4 + 4
L	=	Benevento	12	
MR	-	the Mont-Renaud manuscript	12	
It may be seen that only one other manuscript has the order of responsories in the Mont-Renaud book, and that is the Saint-Denis antiphoner. The Mont-Renaud source may also come from Saint-Denis, or, slightly more likely, from Corbie, whose liturgy and chant were in most respects identical with that of Saint-Denis. So the little test carried out here accords with other findings about the manuscript. (It may be said that not every formulary finds these two sources in agreement.)
It looks as if the manuscripts all select responsories from a common pool of favourite compositions, practically all of which were already available early in the ninth century (at least in time for the Compiegne source to include them). So the comparison sheds a little light on the historical development of the repertory, which one could follow further by consulting the Old Roman sources.
The importance of comparison of musical variants cannot be overestimated. Not only do they act as a litmus test for the relationships between sources, they constitute valuable evidence as to the mode of transmission of chant. The only extensive work so far carried out is that of the monks of Solesmes, reported in Ixj Graduel romain IV, for proper chants of mass. (For examples of work on other mass chants, office antiphons, and sequences, see Underwood 1982 and Hiley 1980-1 and 1986, 'Thurstan'.) This work is laborious, for its effectiveness depends both on the accumulation of a large number of points of musical variance and their tracking through a large number of sources.
Much of the information elicited by the above methods of investigation has to be reported in articles rather than brief catalogue descriptions of sources, though the latter can summarize the most important points. Catalogue descriptions should, however, include a codicological and palaeographical description of the source and describe the nature and extent of the musical notation. (The latter point is often overlooked: for example, has a missal notation only for the canon of the mass, or for
75. Identifying and Describing Chant-Books 339
all chants?) A photograph of the notation often says more than a paragraph of description.
There are by now a great number of descriptions of chant sources, particularly in the catalogues of individual libraries. Their quality and the space they devote to each source naturally varies greatly. As examples of widely differing concepts of what should be reported, the reader may consult Frere (1894-32), Angles and Subirä (1946), Husmann in RISM B/V/l (1964), Bernard (1965-74), Arnese (1967), Stenzl (1972), Plocek (1973, Fernandez de la Cuesta (1980), and Gottwald's catalogues of music manuscripts in Augsburg, Freiburg im Breisgau, Munich, Nuremberg, and Stuttgart. Fine catalogues of liturgical manuscripts (that is, not just music-books) are now available for several areas of Europe, for example Radö (1973) for Hungary, Gregoire (1968—73) for Italy, and the volumes of Iter Helveticum. Other essential research tools include catalogues of medieval libraries, either original documents (G. H. Becker 1885; the Mittelalterliche Bibliotheks-Kataloge published for Germany, Switzerland, and Austria; Nortier 1957-8-for Norman monasteries) or lists of surviving sources from medieval centres (Ker 1964, 1987 and Krämer 1989).
IV
Notation
IV. 1. INTRODUCTION
(i) Preliminary
(ii) The Signs in Montpellier H. 159
(iii) Neume
Wagner II; PalMus 13; Sunol 1935; Jammers 1965, Tafeln; Stablein 1975; Corbin 1977; Corbin, 'Neumatic Notations, I-IV, NG; Hiley, 'Notation, III, 1', NG; Huglo 1990.
(i) Preliminary
In this chapter first the notational signs used in the earliest chant sources are discussed, the signs generally known as neumes. It is not easy to come to grips with the different styles of notation, their beginnings and historical development, without some basic knowledge of the system or systems involved. First of all, therefore, a brief description of the signs in one particular manuscript is given, a manuscript chosen (a) because it is available for further study in facsimile in a modern transcription, and (b) because it has a letter notation, specifying pitches, as well as the signs commonly known as neumes, which specify mode of delivery. This is the well-known Dijon tonary, Montpellier, Faculte de Medecine H. 159 (PalMus 8; Hansen 1974). It should be clear from the outset, however, that other sources have signs for which there is no counterpart in Montpellier H. 159, and that it has some which they may lack, and that no two manuscripts correspond exactly in the way they employ the signs.
Other styles of music-writing are then illustrated, after which the beginnings of notation are discussed, a topic which has caused much controversy, not least in the last few years. Then the rhythmic nuances indicated in the refined notation of some important early sources are discussed. After that the development of pitch-notation from the eleventh century onward is treated. Some systems found only in theoretical treatises are mentioned, and then the notations of the later Middle Ages and subsequent centuries. Finally, a few thoughts are hazarded about performance and transcription.
/. Introduction
341
Although it has become customary to call the early notational signs 'neumes', the word was more often used in a different sense in the early Middle Ages, to mean a melodic phrase. A neuma could be as short as one note or as long as 101 notes. The meaning of the term is discussed below. Like Cardine (1968, etc.) I have usually spoken simply of 'signs'.
The primary function of musical notation is usually regarded as the indication of pitch, but this was evidently not the case with the earliest notations. It is essential to realize that the music represented was already known by heart as far as its tonality and pitch content were concerned. The notation reminded the singer of details of phrasing, rhythm, and dynamic, together with some refinements of performance associated with the use of the signs known as the oriscus and quilisma, whose significance is not now fully clear. Even after notation began to be used, most performance of chant continued from memory. A notated book was for refreshing the memory, for recordatio, as it was known, in the song-school or before a particular service. We should not, therefore, expect early manuscripts to tell us the same as modern editions such as the Graduale or Antiphonale Romanům, nor regard their notation as some sort of primitive, groping, and flawed start on a long road to ultimate perfection.
(ii) The Signs in Montpellier H. 159
Table IV. 1.1 sets out the signs in Montpellier H. 159, together with corresponding signs in the modern Vatican/Solesmes editions and in the transcriptions given in this book. At the beginning of each line the melodic gesture involved is stated, and the name of the sign in the later medieval sign-tables. (These tables, which appear from the twelfth century onward, have been edited and discussed by Huglo 1954, 'Noms'. Most of the tables are of German origin; for two Italian ones see Ferretti 1929. Only the better-known names are given here.) Montpellier H. 159, but not many other manuscripts, uses the slanting virga (IB) for a note which comes at a low point in a phrase, but is not low in the overall range of the piece. And very few other sources use the descending quilisma. The interpretation of the special signs 11-15 is discussed in IV.3.
Ex. IV. 1.1 shows how some of the signs are used. Each sign is labelled with the number given in the Table. The signs usually give no indication as to the size of the intervals involved. The liquescent sign 15B is used indiscriminately here for a second, a third, and a fourth. The scandicus (4) can just as well be used for Gbd as for Gab. Only the signs for single notes indicate something of relative height, so that 'hic ferť has virga-punctum. The punctum is usually reserved for the lowest of a group of single notes, so that there are more virgae than puncta: thus the 'qui' at 'qui manet' has a virga, not a punctum, likewise 'eť and '-go' at 'et ego'. By analogy one would then expect the scandicus and climacus to consist of but one punctum and two virgae, but the reverse is the case. The slanting virga (IB) is used three times, for a, b, and b respectively; but the punctum can also be found for these notes as well, or even for c
Table I V. 1.1. Notational signs in Montpellier, Faculté de Médecine II. 159
Melodie gesture	Name in medieval	Montpellier	Vatican	Transcrif
(H = higher note	sign-tables		(* = no special	
M = middle note			sign)	
L = lower note)				
1. single notes				
A. H	virga	1		•
B. L (relatively high in range)		/	*	•
C. L	punctum		•	
2. L-II	pes, podatus	J	5	Si
3. H-L	clivis, rlexa	A		
4. L-M-II, etc.	scandicus	i 1	4	• #*
5. H-M-L, etc.	climacus	h K	v.	m • *
6. L-H-M-L	pes subbipunctis		v d *	
7. L-H-L	torculus	A J1 >	«\	rr • •
8. H-L-H	porrectus	ť	Si	•
9. H-L-H-L	porrectus flexus	Al		• 0
10. L-H-L-1I	torculus resupinus	y		
11. repeated notes
A.
B.
C. L-H and H-L + repeated note
12. H-H-L
13. see below, IV.3
A. followed by a punctum
B. joined to the punctum
C. followed by a virga
14. see below, IV.3
15. liquescent neumcs (selection)
A. punctum + ascending liquescence — liquescent pes (L-H)
B. virga + descending liquescence = liquescent clivis (H-L)
C. clivis + descending liquescence = liquescent climacus (H-M-L)
bivirga, trivirga bistropha, tristropha
tngon oriscus
(not found alone)
quilisma
(always between two notes a 3rd apart)
epiphonus cephalicus ancus
(i!
r. n
J
J ft
'7\
344
IV. Notation
Ex. IV.1.1. Communion Ego sum vitis vera (Montpellier, Facultc dc Medecinc II. 159, p. 82)
»3c H 156 13c «56 10 7 J     J     *     J     *      / 4
IA   t     1     *r    1c U   U 1c    7 1c
l  J  J  /  • i  l . J1 .
.jTt   *s   0   q   «'i> >^
go     sum   ui - tis     ue - ra
et   uos  pal - mi - tes qui manet   in me
1A U 1A  I 18 3 USA
U 1c Z 13a 1 1c 7 16 I      J-h.     J    .    i1 /
-/>-
et  e-go in   e - urn
hie    fert       fructum multum
I5e 1a 4    8 18
1A ia 4-   1a13a1a 3
*--»3»
al - le-
-lu
la
al - le-
('maNET'). Clearly there is flexibility in the use of the signs and much depends on melodic context.
As already mentioned, the extant tables of notational signs date only from the twelfth century onward. The names given to the individual signs are of uncertain derivation, and most are undoubtedly late Latin or pseudo-Greek neologisms. They are probably better known now than they were in the Middle Ages. The literal meanings of those given on Table IV.1.1 are as follows. It will be seen that most are descriptive either of the actual shape of the sign or of the melodic gesture involved. (For further discussion, see Huglo 1954, 'Noms'.)
virga
punctum
pes
podatus
clivis
flexa
scandicus
climacus
pes subbipunctis
torculus
porrectus
porrectus flexus
Lat.: 'staff Lat.: 'point' Lat.: 'foot'
from Gk. pons, pod us: 'foot' from Lat. clivus: 'slope'
possibly an abbreviation of Lat. accent 'circumflexa'
from Lat. scandere: 'ascend'
from Gk. klimax: 'ladder'
a pes with two lower puncta
Lat. 'screw of a wine-press'
Lat. 'stretched out'
a porrectus turned back down
/. Introduction
345
torculus resupinus bivirga, trivirga bistropha, tristropha trigon
onscus
quihsma
epiphonus
cephalicus
ancus
a torculus turned back up
two, three virgae
two, three apostrophes
from Gk. trigonon: 'triangle'
from Gk. horos: 'limit'?
from Gk. kylisma: 'rolling action'
from Gk. epi + phone: 'added sound'
from Gk. kephalis: 'little head'
from Gk. agkon: 'curve'
(iii) Neume
There are two Latin spellings of the word, with meanings really distinct but often confused in the Middle Ages: the Latin word neuma (from the same word in Greek) means a 'gesture'; pneuma (also from the same word in Greek) means 'breath'. The latter form was commonly used to signify the Holy Spirit, and this meaning became applied by transference to neuma, thus linking music to the divine creative force itself.
Although neuma did eventually come to mean a notational sign, that was not its primary musical meaning. More often it was used to refer to a sounding melody, or phrase, in particular one which has no words. Amalarius of Metz, writing about 830, speaks in a well-known passage about a neuma triplex, a set of three melismas, textless melodies, added to certain responsories of the Night Office (ed. Hanssens iii. 54-5). The textless melody of the sequence was also often referred to as the neuma:
Post alleluya, quaedam melodia neumatum cantatur, quod sequentiam quidam appellant (Udalricus of Cluny: PL 149. 655)
Pneuma sequentie quod post alleluia cantatur, laudem eterne glorie significat (Johannes of Avranches: PL 147. 34)
(See Bautier-Regnier 1964, 5; Hiley, 'Neuma', NG). Neuma is also the 'model' melody for each mode found in tonaries or added to antiphons as a festal extension (see III.14.ii).
The use of neuma to mean the written sign seems not to be as old as these. In the controversial eleventh-century treatise Quid est cantus? (which will be referred to again in the discussion of the origins of neumatic notation) it is stated: 'De accentibus toni oritur nota que dicitur neuma' (Wagner 1904, 481-4).
The usual word for written musical signs in the Middle Ages was nota. As a term for alphabetic or comparable signs, this is found as early as the ninth century, in the anonymous treatise Musica enchiriadis and in the treatise of Hucbald of Saint-Amand. The author of Musica enchiriadis has been speaking of the special signs of 'dasian' notation, and ends: 'Sunt et alia plura plurium sonorum signa, inventa antiquitus, quibus evitatis, has faciliores hie inserere curae fuit notas.' ('And there are very many other signs for sounds, found in antiquity, which have been avoided, and
346
IV. Notation
our care was to insert here these easier notes.') Eventually nota signifies the non-alphabetic signs as well—as in the treatise Quid est cantus? mentioned above.
By the eleventh to twelfth century, there had arisen a manuscript tradition of listing neumes in didactic tables. There are two types of table, one short and the other long. One thirteenth-century source of the short table (Huglo 1954, 'Noms', 55, source R) entitles its list 'Nomina notarum', while the usual title for the long table is 'Nomina neumarum'. The last verse of the short table runs: 'Neumarum signis erras qui plura refingis.' ('You err with signs of the neumes when you fashion more.) Thus here 'neume' still means the melody itself, not the sign.
As a final witness to the meaning of neume as a melody, rather than a written sign, Guido of Arezzo's Micrologus, written about 1030, may be cited. In ch. 15 Guido draws an interesting analogy between the construction of metrical verse and that of a melody. The terms he uses are the following. The individual letters of verse (litterae) are comparable to the individual sounds of music (phthongi or soni). As syllables (syllabae) are composed of letters, so musical syllables (syllabae) are composed of one or more sounds. Syllables are made up into parts or feet (partes or pedes); one or more musical syllables make up neumes or parts (neumae or partes). The bringing together of feet makes a verse (versus), which is equivalent to a section (distinctio) of a musical composition, composed of several neumes.
The term neume, meaning a notational sign, has by now become so embedded in the musical literature that it would be pointless to try and restrict its meaning to the other medieval sense of the word. I have nevertheless avoided it when speaking of notation.
IV.2. REGIONAL STYLES
(i) FYcnch and German Notation (see Plates 4, 6-8)
(ii) Palacofrankish, Laon, Breton, and Aquitanian Notations (Plates 1-3)
(iii) Types Related to French-German Notation (Plate 5)
(iv) Other Italian Notations (Plates 11-13)
(v) Examples
PalMus 2-3; Bannister 1913; Suriol 1935.
This chapter gives brief descriptions of some of the more important types of chant notation. It will be recognized, of course, that sources habitually classified together do not always use absolutely identical shapes. The signs given in the tables below are copied from one more or less typical manuscript, rather than being regularized on the basis of the generality of sources. Nor is an attempt made to give all the variant shapes within each source: only the most common are selected.
2. Regional Styles
347
(i) French and German Notation (see Plates 4, 6-8)
Facsimiles:
Bamberg, Staatsbibliothek lit. 6 (German/St Gall): see Bamberg 6
Einsiedeln, Stiftsbibliothek 121 (German/St Gall): PalMus 4
Montpellier, Faculte de Medecine H. 159 (French): PalMus 8
Paris, Bibliotheque Mazarine 384 (French): MMS 5
St Gall, Stiftsbibliothek 339 (German/St Gall): PalMus 1
St Gall, Stiftsbibliothek 359 (German/St Gall): PalMus II/2
St Gall, Stiftsbibliothek 390-391 (German/St Gall): PalMus II/l
private collection, the Mont-Renaud manuscript (French): PalMus 16
Although commonly regarded as distinct, French and German signs are similar in many basic ways. The habit of classifying them apart has come about because whereas up-strokes of the pen in French notation are more or less vertical, those of German sources are usually slanted. Furthermore, the deservedly high degree of attention paid to the most interesting sources of German notation, the St Gall manuscripts, with a wealth of notational detail not found in French sources, has created the impression of a system which is different from that of French notation. If, however, St Gall notation is regarded as exceptional, and a comparison is made between simpler German and French examples, the similarities are obvious.
Table IV.2.1 sets out the usual forms for French and German notation, but places between them the forms found in an intermediate area, as it were. The sources on which the table is based are the Mont-Renaud manuscript (from Corbie, late tenth century), Leipzig, Universitätsbibliothek, Rep. I. 93 (from the Mosel area, tenth century), and London, British Library, Add. 19768 (from Mainz, tenth century). This table may be compared with Table IV. 1.1 in the previous section, for Montpellier H. 159 belongs to this family of notations.
The varieties of French notation found in Cluniac sources have been surveyed by Hourlier (1951, 'Remarques'). The notation of the St Gall sources has long been the subject of intensive study. Because of its especially sophisticated character some of its features are discussed below, IV.3 and IV.5.
(ii) Palaeofrankish, Laon, Breton, and Aquitanian Notations (see Plates 1 —3) Facsimiles:
Düsseldorf, Universitätsbibliothek D. 1 (Palaeofrankish): Jammers 1952 Laon, Bibliotheque Municipale 239 (Laon): PalMus 10 Chartres, Bibliotheque Municipale 47 (Breton): PalMus 1 Paris, Bibliotheque Nationale lat. 903 (Aquitanian): PalMus 13
Unlike the widely established French-German type, Palaeofrankish signs were known in only a small area of north-east France, and survive in but a handful of sources, none of which is a full gradual or antiphoner (for a list of sources, see Hourlier and Huglo 1957). Handschin (1950) gave it the name 'Palaeofrankish',
348
TV. Notation Table I V.2.1. French and German notational signs
			Mont-Renaud	Leipzig Rep. I. 93	London 19768
1.	virga	•	/	/	/
2.	punctum	•	-		-
3.	clivis	•	P	A	
4.	pes	*	J i	j	
5.	porrectus	• t •			
6.	torculus	• »	f		f r
7.	climacus	• • •	(■■	/:	
8.	scandicus		1	/	/ /
9.	quilisma		J	/	y
10.	oriscus	H	s	i	•>
11.	liquescents	o *	,f „	o r J	
believing it to be the oldest type of notation developed in the Frankish empire. Instead of writing pes and clivis with two distinct movements of the pen, this notation uses only one slanting or slightly curved stroke. The porrectus and torculus also tend to form a single arc or angle, rather than a tripartite figure. Quilisma and oriscus do not appear to be present in the earliest examples. Since so many of the examples are brief, often entered in manuscripts not originally designed to contain notation, generalizations about the signs are somewhat hazardous. Those entered in Table IV.2.2 are copied from Düsseldorf, Universitätsbibliothek D. 1 (from PKorvey, tenth century).
The area where Palaeofrankish notation was used is also that where so-called 'Messine' signs were used. The name 'Messine' refers to the supposed origin of the type in Metz, famous in the early ninth century for its chant. No ninth-century notated sources from Metz have survived, and indeed, those from later centuries are also relatively scarce. Corbin (1977, and 'Neumatic Notations', NG) argued that
2. Regional Styles 349
'Lorraine' was a better name for the type, since the area where it was used corresponded to the ancient territory of Lotharingia (from which descends the modern name 'Lorraine'). 'Lorraine' is, however, no more appropriate than the name 'Messine'. Map IV.2.1 shows the centres from which sources with this notation survive. Neither the ninth-century kingdom nor the tenth-century duchy of Lotharingia included the more westerly centres, such as Lille, Noyon, Laon, or Reims; but they did include cities such as Trier and Aachen, and other territory as far as the Rhine where the notation was unknown. The area corresponds better, though still not exactly, with the archdiocese of Reims. Metz is on its periphery. In this book I have called it 'Laon' notation, after its most famous representative, manuscript Laon, Bibliotheque Municipale 239, and because Laon is reasonably central to the area. (For a survey of sources see Hourlier 1951, 'Messine'.)
The most characteristic feature of Laon notation is the small hook often used for the punctum (sometimes called an uncinus). The virga is usually a long flat S shape,
Table IV.2.2. Palaeofrankish, Laon, Breton, and Aquitanian notationalsigns
Dusseldorf D.l      Laon 239      Chartres 47 Paris 1084
1.	virga	•				M
2.	punctum	•	■			
3.	clivis	•	\	1 r		:
4.	pes	•	J		J /	r
5.	porrectus	• • •	Y	f	■V	■ r
6.	torculus	• *		A		4
7.	climacus	• • •	\	i £ H	; - fa)	:: C*0
8.	scandicus	* *		/ / . r i r		_ r
9.	quilisma	uJ	J	/ l	I	t
10.	oriscus	H				
11.	liquescents	ro o »			r <y	1 -/
350 IV. Notation
Map IV.2.1. Centres where Laon notation was used
rather than a straight line; it is hardly ever seen except in combination with the punctum, to form a pes or other sign.
The notation of the manuscript Laon 239 is of a complexity and sophistication comparable only with that of the St Gall manuscripts, and more is said about it below (IV.3 and IV.5).
French-German notation starts its clivis with an upward stroke, but Laon and Breton notation begin instead with a short, almost horizontal movement of the pen, sometimes a shallow arc which is curved in towards the following down-stroke. Breton notation takes its name from the area of north-west Frence, Brittany, whence practically all its sources originate. (For a survey of sources, see Huglo 1963, 'Domaine'.) In many respects it bears a resemblance to Laon notation, though much more compact in overall appearance, and lacking the characteristic Laon hook. It does not have a quilisma. Two signs are used in places where, from other manuscripts, one
2. Regional Styles 351
would expect to see a quilisma or oriscus, but the conventions governing their use in Breton notation are slightly different. Table IV.2.2 shows the Breton signs of Chartres, Bibliotheque Municipale 47 (late ninth century; destroyed in 1944: facsimile in PalMus 11).
Although many notations use combinations of disjunct puncta and virgae as a special variant instead of their more usual conjunct forms for clivis, flexa, etc., few go as far as to make the disjunct forms the rule. This is so, however, with Aquitanian signs (see the survey in PalMus 13, by Ferretti). Again the virga is usually found only 'in composition' with other elements. The signs in Table IV.2.2 are drawn from Paris, Bibliotheque Nationale, lat. 1084 (PAurillac, then Limoges, late tenth century).
It will be noticed that Laon, Breton, and Aquitanian notations all rise diagonally and fall vertically (compare the climacus and scandicus forms).
(iii) Types Related to French-German Notation (see Plate 5) Facsimiles:
London, British Library, Add. 30850 (north Spanish): Antiphonale Silense Rome, Biblioteca Angelica 123 (Bologna): PalMus 18
Studies: Barezzani 1981 (Brescia).
Signs of the FYench-German variety were also used outside their 'home' area. Upright French shapes were copied in England in the late tenth century (summary of sources by Rankin 1987); and also in north Italy, at centres such as Novalesa and Asti, and, with slightly more individual character, at Brescia and Mantua. By contrast, Monza and Bobbio used the slanting German shapes.
Besides this, signs bearing a more general resemblance to the main French-German type are found in both Spain and Italy. Of the two main Spanish types, that used in north Spain had upright shapes, that of Toledo almost horizontal neumes. Both, however, are recognizably related to French signs, in their use of a virga, and an upward/forward movement of the pen at the start of clivis and porrectus. Further research is needed before the identification and use of 'ornamental' signs in the Spanish notations—quilisma, oriscus, and liquescents—is properly understood.
The long, bold strokes used in a well-known source probably from Bologna, Rome, Biblioteca Angelica 123 (eleventh century), are also close relatives of the French-German type. The notation is also found in a source possibly from Modena (flyleaves of Modena, Biblioteca Capitolare O. I. 13), and another seemingly from far to the south, possibly a monastic house in the Abruzzi, Rome, Biblioteca Apostolica Vaticana, Vat. lat. 4770 (closely related by repertory to Benevantan sources).
Sources from the Catalan area of north-east Spain have signs which bear surprisingly little resemblance to those of the Aquitanian area. For example, three vertically aligned puncta signify the ascending scandicus, not a descending climacus as in Aquitanian notation; and there is a preference for conjunct forms. Catalan signs are unlike other types in ascending vertically but descending on a very shallow incline.
352 IV. Notation
Table IV.2.3. Signs related to French-Gennan notation
1.	virga	•	1 y
2.	punctum	•	■ - s
3.	clivis	•	f Un
4.	pes	•	J J v
5.	porrectus	• * *	J ji
6.	torculus		
7.	climacus	•	
8,	scandicus		!//
9.	quilisma		J
10.	oriscus	M	
11.	liquescents	so 0 •	
Leon 8 Toledo 35. 7 Angelica 123      Girona 20
M	n
J	J
*l	
If	
!■■ />	
i	• *
i	
	
Catalan notation has been studied by Mas (1988). The signs in Table IV.2.3 are drawn from Girona, Sant Feliu 20 (eleventh to twelfth century).
(iv) Other Italian Notations (see Plates 11-13) Facsimiles:
Benevento, Archivio Capitolare 33 (Benevcntan): PalMus 20
Benevento, Archivio Capitolare 34 (Benevcntan): PalMus 15
London, British Library, Add. 34209 (Milanese): PalMus 5
Lucca, Biblioteca Capitolare 601 (central Italian): PalMus 9
Rome, Biblioteca Casanatense 1741 (Nonantolan): Vecchi 1955
Rome, Biblioteca Apostolica Vaticana, Vat. lat. 10673 (Benevcntan): PalMus 14
private collection of Martin Bodmer, C. 74 (Roman): Lütolf 1987
Studies: Huglo el al., 1956 (Milanese) Moderini 1970 (Nonantola)
2. Regional Styles 353
Besides French or German shapes, other notations were also adopted in Italy. At Como, Laon signs influenced the notation (see Sesini 1932), Breton ones at Pavia (Huglo 1963, 'Domaine'). Disjunct forms (not at first sight similar to Aquitanian signs, however) were preferred in the very individual style practised at Nonantola. This is but one of many centres whose notation does not reflect the influences discernible in its repertory, for the monastery took over much of the German/Swiss trope and sequence repertory. The notation is related to that found at Torcello, and also appears occasionally in Verona manuscripts. The signs in Table IV.2.4 are drawn from Rome, Biblioteca Casanatense 1741 (twelfth century). A peculiar feature of Nonantolan notation is the way in which letters in the text were provided with long extensions to a height appropriate for the note being sung (which could only be done, of course, for the first note of each syllable).
Sources from many Italian centres have hardly survived at all before the time when staff-notation came into use (eleventh century or later). Generally speaking, the use of a staff often seems to have affected the notational signs, causing the appearance of
Table IV.2.4. Other Italian notations
			Cas.	S.Amb.	Padua	Pistoia	Vat.lat.	Vat. la
			1741	M.29	A.47	C.121	5319	10673
1.	virga	•	1		1	1	1	/
2.	punctum	•	•	•	-	-	-	
3.	cli vis	•> •	I	1	"1 A	I/I		"1 1
4.	pes		1		1	J	J	J
5.	porrectus	•	V	r	A)	>	i	-v
6.	torculus	* •	4		A	A	A	
				[A 1				
7.	climacus	•	r i	\		/-,	i	
8.	scandicus		i	r	} ?		J	J
9.	quilisma	r, w						J
10.	oriscus		-					
11.	liquescents	o 0		5	j	i +	1 V	TT y
354 IV. Notation
more obvious heads and other precisely placed pen-marks. Sources from Milan show this, their signs often being little groups of tiny waved strokes, joined by fine lines. (See examples in Table IV.2.4, from the twelfth-century manuscript Milan, Sant'Ambrogio M. 29.)
Many Italian sources are distinguished by a conjunct scandicus, and the horizontal bar at the start of the clivis is another diagnostic feature. In most respects there is little to choose, as far as the basic shape of the signs is concerned, between the majority of sources written south of a line drawn between Lucca in the west and Ravenna in the east, excepting those types already mentioned. For want of a more accurate term, I have referred to this family as central Italian notation. Of course, there are differences, such as the great variety of liquescent shapes in manuscripts of the Montecassino and Benevento area. Variation in such matters as the thickness of the pen often has a strong effect on the visual aspect of the signs. One would not readily confuse the bold strokes of later Beneventan sources for any others.
Alongside the Nonantola and Milan examples already mentioned, Table IV.2.4 gives some typical shapes from sources from Ravenna (Padua, Biblioteca Capitolare A. 47, twelfth century), Pistoia, Biblioteca Capitolare C. 121 (Pistoia, twelfth century), Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319 (Old Roman, twelfth century), and the region of Bari (Rome, Biblioteca Apostolica Vaticana, Vat. lat. 10673, eleventh century). The variation between sources in the angle of ascent and descent is important (particularly in the climacus; the Milanese source usually descends vertically, but after a torculus the points continue diagonally).
(v) Examples
Ex. IV.2.1 gives another specimen of French notation, like that in the previous chapter, but from a different source: the Mont-Renaud manuscript (Corbie, tenth century: see above, Table IV.2.1). Two other versions of the same antiphon, Be at us Hie servus, in staff notation are copied in parallel with it: Worcester, Cathedral Chapter Library F. 160 (Worcester, thirteenth century) and Paris, Bibliotheque Nationale, lat. 17296 (Saint-Denis, late twelfth century). The notation of modern Solesmes/Vatican books was developed from the signs in manuscripts such as these.
The neumes of the Corbie manuscript include both punctum (dot) and tractulus (dash), always found either for relatively low isolated notes, or as the lower element in a pes. In some sources it seems that the tractulus signifies a longer note than the punctum, or a stressed note (see IV.4). It will be noticed that there are proportionately more virgae than puncta: evidently the scribe preferred to reserve the punctum/tractulus for only the lowest note(s) of a section, rather than reserving the virga for only the highest note(s). The punctum+virga form of the pes is used when the pes begins at a lower pitch than the preceding note. The form with two virgae is used when there is a rise to the start of the pes.
Between the two later versions in staff-notation there are interesting discrepancies. (These are common enough for office chants: see the much more disparate version in
2. Regional Styles
355
Ex. IV.2.1. Antiphon Beatus tile semis (Mont-Renaud manuscript, fo. 119v; Worcester, Cathedral Chapter Library F. 160, p. 427; Paris, Bibl. Nat. lat. 17296, fo. 282v)
\4ont-R. -_//"./". J    I .    I1   ft  J   I I
Wore.
c						a		i		•*	
			■ A		1	T	v .	—r		1	
	\	■	1 "				1		i •		
	1 Be	- a-	tus il	-le seruus	cum	ue-	1 -ne-rit	do	-mimus	e-	ius
p								9			
			■ 4		1	I			^ 3	1	
	—V-		1 "	IV 1 '—1			* ■		—#-=——■-		
quern
		1     l'       1 A	i'	1	1	-	1    ip. Ii		/	
		■    "     1 »	9					n		
w	•		■				m   1 1 ■    1 "	r		
		1 J							1	
	et	pulsa-uerit	ia-	-nu-	-am	et	in-uenerit e-um	ui-	.gi_	lantern.
		■ * • ■	s				■ ■ ■	fh		
&	•			\	■		■   * 1 8		m	
	•				1		! '			
A/V/ 671, based on the St Gall version.) The Worcester source is generally closer to what is indicated by the signs of the Corbie manuscript (liquescent form + punctum for ille, etc.; see also the differences in wording). But it does not usually employ the punctum (square) as opposed to the virga (square with tail). The Saint-Denis source is not so close textually and melodically, but makes use of the punctum for the low notes.
Not all notations use the virga as an isolated note. Ex. IV.2.2 gives the communion Ouis dab it ex Sion from five different manuscripts. At the foot are the pitches given in letter notation by the eleventh-century manuscript from Saint-Benigne at Dijon, Montpellier, Faculte de Medecine H. 159; above it are the neumes of the same manuscript (occasionally differing slightly from the letters). Above these are the neumes of four other manuscripts of the late ninth to the eleventh centuries. There is no isolated virga in either Paris, Bibliotheque Nationale, lat. 776 (Aquitanian) or Laon, Bibliotheque Municipale 239 (Laon). The apostropha (see 3 and 40) is to be found only in Einsiedeln, Stiftsbibliothek 121 (German) and Montpellier H. 159 (French). There is no quilisma (see 41) in Chartres, Bibliotheque Municipale 47 (Breton), and only Montpellier H. 159 uses the oriscus (end of 36, middle of 44).
356
/V. Notation
Ex IV.2.2. Communion 0«« rfa6i7 ex S»o« (P = Paris, Bibl. Nat. lat. 776; C = Chartres, Bibl. Mun. 47; L - Laon, Bibl. Mun. 239; E = Einsiedeln, Stiftsbibl. 121; M = Montpelher, Faculte de Medecine H. 159)
\   . . . / .« \ / * /--'-   ■ y J
c
L
E       /)      /    »>       -      /    //•.     * -    /     //f    A     -     A\    f^'ft   * -    / a (A     |    ui      .     /     | h        .    !     A     A     I    r    /- ///I        •    / J
1x3        4-       5      6 _78       9       fo      11      q    13_^   15 lb
M
Quis da - bit     ex   Si - on sa - lu - ta - re     Is - ra - e] cum a - uer-
p		c	- " 4	A	A .	-	/	
c		J	- - /	-/)			«/	
L		J	/   r 1	J)	r r      r r r	r f	c/	r
E	-  -   / / y -1	«/	i /  / p	1	/    S -	i«> /	f J a.	
»7 <» tj 2o a, 42 23 u 25 2t 17 I) 30 31 *i 21. 2£ ---—-2---^- fO *r#—
— ***** *g *     m'u    m —
-te-rit    do - mi-nus    cap-ti-     -ui-ta-tem   pie - bis    su - e e - xul - ta - bit
S" .
"a
31» 37  «   31   *o      <H 4-a *t*
la-cob et le - ta - bi - tur Is - ra - el.
3. Liquescence, Oriscus, Quilisma
357
IV.3. LIQUESCENCE, ORISCUS, QUILISMA; OTHER
SPECIAL SIGNS
(i) Signs for Liquescence
(ii) Quilisma
(iii) Oriscus
(iv) Virga strata, Pressus, Pes stratus, Pes quassus, Salicus
(i) Signs for Liquescence
Mocquereau in PalMus 2; Freistedt 1929; Göschl 1980; Hiley 1984.
Looking once again at Ex. IV.2.1, one notices that there are no liquescent signs in the Saint-Denis version, and only one in the Worcester copy, at 'ille'. The Corbie manuscript has that one, and others at 'ser(vus)', 'cum', 'ei(us'), and 'et'. Signs for liquescence occur most often where in the text there is a conjunction of a liquid consonant followed by another consonant. The two /'s of 'ille', 'rv' in 'servus', and 'mv' in 'cum venerit' are typical examples. Diphthongs usually call forth a liquescent sign, as at 'eius'. Quite often only one consonant may be present. It is not uncommon also to find liquescents with the syllable 'et', and there are other instances where neither a liquid consonant nor a diphthong seems to be involved. This may, however, indicate that the consonants in question were actually liquescent when the manuscript was being written—such differences of pronunciation are still common today from area to area, so that, for example, in Beding is pronounced like an English _y: 'Wenn die jemsen springen über Berjesjipfel . . .'
The manner of performance of liquescent signs is not entirely clear. A passing-note of some sort seems to be involved, since many manuscripts will record an ordinary neume (say, of two notes) where others have a liquescent (one note + 'passing-note'). The liquescent note itself may in some cases have been sung 'through' the consonant, as it were, vocalized to /, m, n, or r (the most common situations). As Guido of Arezzo said: 'At many points notes liquesce like the liquid letters, so that the interval from one note to another is begun with a smooth glide and does not appear to have a stopping place en route' (trans. Babb 1978, 72).
Comparisons between sources indicate that the use of liquescent signs was inconsistent, as in Ex. IV.2.1. This too is stated by Guido, who continues: 'If, however, you wish to perform [an example] more fully, not liquescing, no harm is done, indeed, it is often more pleasing.' So the signs are there as reminders, though no two scribes would put them in exactly the same places, rather in the way that the rhythmic nuances discussed below (IV.5) are not recorded identically in different manuscripts. Montpellier, Faculte de Medecine H. 159 seems to be a moderately generous source, but is easily surpassed by, for example, the early St Gall sources, and especially by those of southern Italy, which have an unrivalled variety of liquescent shapes.
358
IV. Notation
The written form of the note in square notation of the thirteenth century and later came to be called the plica ('fold').
(ii) Quilisma Wiesli 1966.
The sign-tables depict the quilisma as a wavy shape ending in a longer ascending stroke. This is translated as two notes in the Montpellier letter notation, the first being the special wavy one, the second higher. The name seems to refer to the wavy note in particular.
In most manuscripts the quilisma is nearly always used for the middle note in an ascending-third formation, sometimes only for a minor third. The name seems to derive from the Greek kylisma ('rolling'), kylio ('to roll'). Medieval writers on music do not elucidate its nature very much. For example, in the middle of the ninth century, when notation itself was relatively rare, Aurelian of Reome uses the word tremula to describe a passage usually later notated with a quilisma. Speaking of the gradual Exultabunt sancti V. Cantate domino, at the passage beginning 'can[ticum novum]' he says (Ch. XIII): '. . . flexibilis est modulatio, duplicata, quae inflexione tremula emittitur vox non gravis sonoritas ...'('... the melody is flexible, repeated, where a sound, not low, is given forth with a tremulous inflexion . . .'). Ex. IV.3.1 gives the signs in Montpellier H. 159, Laon, Bibliotheque Municipale 239, and St Gall, Stiftsbibliothek 359.
It has been suggested that the quilisma might involve a portamento delivery, or one with semitone steps filling in the larger intervals, or a gruppetto ornament. In those manuscripts where its use is restricted to the minor-third progression, it constitutes a sign for the semitone step, and may therefore have been used specifically to locate the melody tonally. The idea that the quilisma was primarily of rhythmic significance (Wiesli, Cardine 1968, etc.) does not seem sufficient to explain the special shape of the sign.
Ex. IV.3.1. From gradual Exultabunt sancti V. Cantate Domino
Laon 239 p.137 neumes
suggested transcription
St.Gall 359 p.125 neumes
suggested transcription
Montpellier p.154 neumes
letter notation
/ j
-ti-cum
can-
J. Liquescence, Oriscus, Quilisma
359
(iii) Oriscus
The derivation of the word is obscure: pseudo-Greek oriscos ('a little hill') has been suggested, or horos ('a limit'). The oriscus does not usually stand alone, but, like the wavy component of the quilisma, comes in formation with other elements, typically in a figure such as a—a-G, where the middle note is an oriscus (see three instances in Ex. IV.1.1).
The oriscus is used in the Montpellier codex only as a repeat of the previous pitch: at least, so it seems when one compares other sources. In fact, in the letter notation underneath the musical signs in this manuscript, there is no letter for the oriscus, only a wavy line to show that 'something' is sung. It should also be borne in mind that repeated notes can be shown by other means, for example by placing a virga or punctum close to the preceding sign. It has therefore been suggested that some special vocal delivery was involved. If not, why use the special sign?
Wagner believed that a non-diatonic pitch was involved. This was because of figures like those in Ex. IV.3.2, where the oriscus is now on 6, now on c, as if the 'true' note were actually in between. Once again the semitone step is involved.
Cardine believed its significance was rhythmic, directing rhythmic weight to the succeeding note, which is very often lower. Whether this rhythmic nuance is sufficient justification for the use of a special sign is debatable (see IV.5).
Ex. IV.3.2. Examples of oriscus
from Gradual Eripe me V. Liberator meus
Ji -I
Montpellier p.161
exalta-bis
St.Gall 359 p.85
from Canticum Attende caelum
Laon 239 p. 101 r   r   r        r   r   r        r r
St.Gall 359 p. 104
Paris 776 f.66v
. .y - - . r . -
-•—•—•»
—•—#*
V. Expec-te-   -tur sicut plu-ui-a
360 IV. Notation
In many sources the oriscus can be found in places where the text suggests liquescence, perhaps therefore a sort of liquescence at the unison.
(iv) Virga strata, Pressus, Pes stratus, Pes quassus, Salic us Cardine 1968.
A number of special signs peculiar to the St Gall manuscripts and those closely related to them appear to involve the oriscus. A virga followed by an oriscus was usually run together into one sign, the virga strata. This sign has also been called the franculus, but Huglo (1954, 'Noms', 65) questions the correctness of this identification. The name for this sign on the longer sign-table is gutturalis ('throaty'), which may imply some special vocal effect.
On the shorter sign-table, the oriscus followed by a punctum is called a pressus minor, the virga strata followed by a punctum a pressus maior.
A pes followed by an oriscus is run together into one sign, called a pes stratus, just as happens to the virga alone.
The pes quassus appears to be a pes of which the first note is an oriscus. Sometimes, however, three notes are found in other sources, so that the pes quassus would be a pes with an extra oriscus between the two normal notes. It is often found at the beginning of phrases in mode 3 introits, where other manuscripts have an E—E—F opening. So the semitone step is again involved.
The salicus is a three-note ascending figure where the middle note is an oriscus. Its nature is puzzling, because it can be found in situations where other sources have: (a) two unison notes followed by a higher one; (b) a normal scandicus, that is, three notes in ascending order; (c) a quilisma in three-note ascending formation. Cardine (1968) believed its chief significance was rhythmic, and that the oriscus, as usual in Cardine's interpretation, directed rhythmic weight on to the succeeding note. To distinguish it from the quilisma, which Cardine interpreted similarly, he pointed to its tonal position: the oriscus element in the salicus falls chiefly on / and c, the wavy element of the quilisma by contrast falls chiefly on E and b; furthermore, the note preceding the wavy element in the quilisma is usually long or stressed. (See the comparisons published by Wiesli 1966, ch. 6; studies of the salicus in Laon 239 and St Gall 359 have been published by Picone 1977 and Ponchelet 1973, respectively.)
Without attempting to adjudicate among the different possible interpretations here, the most common characteristics associated with all the neumes involving the oriscus, and also the quilisma, are (i) frequent occurrence at the semitone step, (ii) special vocal delivery, (iii) Cardine's idea of rhythmic drive to the succeeding note. The signs are given in Table IV.3.1, which shows the St Gall form, the usual equivalents in Laon 239, and the form of transcription used in this book.
Wagner called all the liquescent neumes and those involving apostrophe, oriscus, quilisma, with the trigon as well, 'Hakenneumen' ('hook-neumes'), because in the St Gall sources their shapes nearly always include at least one small semicircle or hook
4. The Origins of Chant Notation 361 Table I V.3.1. Special signs in St Gall and Laon notation
	St Gall	Laon	Transcription	
virga strata/gutturalis	r	i ^	->	
pressus minor	r.	1	•	
pressus maior	r-	1 '		
pes stratus		X		
pes quassus		/ / CJ) .«*		• ♦
	/	/ y /	*	
sahcus		•a) / /		
(Wagner II, ch. 8). Thus the apostrophe (used principally St Gall, less frequently in other French-German notations) is used principally for repeated notes on c and other pitches at the semitone, in such chants as graduals. The trigon (three dots set out in triangular formation) was used for a figure such as b-c-a or c—c-a. These signs all derived, Wagner believed, from the apostrophe sign ('Haken' in German). For Wagner their frequent occurrence at the semitone step was crucial, and he believed that chromatic progressions, perhaps even involving quarter-tones, were indicated. While one must certainly reckon with the possibility that the matching of the plainchant repertory to the diatonic scale may have been imperfect, so that many melodies included non-diatonic progressions, the evidence of the plainchant sources themselves has so far been inconclusive, and the statements of medieval writers on music too insubstantial to support Wagner's theory.
Although the neumes discussed in this section have been called 'special' (sometimes 'ornamental'), they are fully integrated into the notation of the earliest sources. It is only from the modern point of view—they had generally fallen out of use by the end of the Middle Ages—that they are 'abnormal', because it is hard to match them to the unequivocal but restricted vocabulary of later notation.
IV.4. THE ORIGINS OF CHANT NOTATION
(i) Introduction
(ii) Early Examples
(iii) Early References to Notation
(iv) Parallel Systems
(a) Prosodic Accents
(6) Punctuation
(c) Ekphonetic Notation
(d) Byzantine Notation
362
IV. Notation
(v) The 'Cheironomic' Theory
(vi) The Early Transmission of Chant
(vii) Some Conclusions
(i) Introduction
The origins of musical notation are obscure. Four main types of evidence may be considered. The first is the evidence of actual examples of musical notation. How old are they? and does anything about them suggest a previous stage of development? The second type of evidence is the information given about notation by medieval writers. The third lies in the existence of parallel or comparable systems of signs to guide vocal performance, which might possibly have suggested or led to the neumatic notation of Western chant: punctuation signs, oratorical accents, Byzantine ekphonetic notation, Byzantine chant notation. The fourth type of evidence is circumstantial. The earliest notation is a record of plainchant melodies. These melodies had previously been performed, learned, and transmitted without the aid of any written record (and thus they continued, to a considerable extent). In these circumstances, one would expect different singers to perform the melodies in different ways. And conversely, when a written transmission became normal, one would expect the differences to have diminished, or even disappeared. If we compare our medieval manuscripts, do we see evidence of the unity attributable to written transmission, or the variety produced by oral tradition?
(ii) Early Examples
The earliest examples of chant notation are the earliest musical notation of any kind in Western Europe. They are generally dated in the ninth century. Only isolated chants or small groups of pieces are notated. The reason for copying them is probably that they were unusual in some way; either (a) they were an unusual addition to the normal repertory, and thus required a special record to be made, or (b) their music was exceptional in comparison with what was normally expected of the singer. At first sight, therefore, it does not look as if it were a general custom to notate the chant repertory as a whole in the ninth century, although Levy (1987, 'Origin', 1987, 'Archetype') has argued that a notated exemplar for mass chants was made as early as around 800. The three earliest surviving books containing a year's cycle of chants date from around 900: Chartres, Bibliothěque Municipale 47 (from Brittany), Laon, Bibliothěque Municipale 239 (from Laon), and St Gall, Stiftsbibliothek 359 (from St Gall). Mirroring the diversity of the earlier examples, these three manuscripts also use different types of neumes.
The dating of early examples to the ninth century usually depends upon palaeographical evidence, and it is not surprising that different scholars have given different estimates. Various lists are available, none containing the same series of sources (see, for example, the twenty-one in Hiley 'Notation, HI, 1', A:G 346, or the
4. The Origins of Chant Notation 363
eleven in Corbin 1977, 30 ff.). Further palaeographical study seems essential. The earliest datable specimen—at least according to one opinion—is a copy of the alleluia prosula Psalle modulamina in Munich, Bayerische Staatsbibliothek elm 9543. The book in question is a copy of St Ambrose's writings, which ends half-way down the final page of the manuscript. At the foot appears a scribal explicit, which tells that the book was copied by one Engyldeo, a cleric at the monastery of St Emmeram in Regensburg from 817 to 834. The blank space in between the main text and the explicit was used to copy Psalle modulamina. (For reproductions see MGG 9, 1625, Smits van Waesberghe 1957/1958, Bischoff 1981.) Bischoff (1974-80, i. 203-4) believed the same hand to have written both main text, prosula text, and neumes, though not all scholars have accepted this (Corbin 1977 omits it from her list of eleven; Stäblein 1975 is notably silent; but see the recent discussion in Möller 1990, 'Prosula'). The date of this one example is not so important, but the difficulties it presents are symptomatic. The general picture remains the same with or without it: sporadic activity through the ninth century, preceding the notation of complete books somewhere towards the end of the century.
The types of neumes represented among the early examples are those known as French, German, and Palaeofrankish. (It is also possible that Spanish examples may be dated to the ninth century: see Huglo 1985, 'Notation Wisigothique'.) Extending the time-limit to the middle of the tenth century, we can add Breton neumes (in Chartres 47), Laon (in Laon 239, and the slightly older flyleaves of Laon 107), and Aquitanian (in Paris, Bibliotheque Nationale, lat. 1154 and 1240, which have both been dated to the early tenth century. It is also noteworthy that Laon and St Gall 359 are not only among the earliest manuscripts with a comprehensive repertory, but also use the most complex and sophisticated systems known. It seems common sense to suggest that they were preceded by less sophisticated codifications of the repertory. Yet the effort does not seem to have been centralized, otherwise the types of neumes would not be so different. Different centres must have had sufficient independence to develop their own styles.
Much has been made of supposed difference of principle underlying the different types of signs. For example, accent-neumes have been held to constitute a different species from point-neumes. Accent-neumes indicate several notes with a single stroke of the pen, as for example the French-German torculus or porrectus. French-German and related types are supposed to belong to this family. Point-neumes tend to indicate each separate note by a separate mark; Aquitanian notation is the obvious example of this type, and Breton and Laon notation have been grouped with it. Yet the distinction is by no means watertight. A glance at Table IV.2.2, with signs of the so-called point-neume notations, shows plenty of signs of the accent-type, for example for the torculus and porrectus again. Palaeofrankish notation, in which a single stroke can represent two notes, has been felt to be qualitatively different to such an extent that Levy, for example (1987, 'Origin'), has seen it as the notation for a first written archetype of chant in the Carolingian period, later superseded by one in another notation. The evidence for this distant period nevertheless remains tantalizingly
364
IV. Notation
scanty, and attempts to construct hiererchies and 'family trees' of notations seem likely to remain hypothetical.
(iii) Early References to Notation
We have many accounts of the pains taken by the Frankish rulers of the eighth and ninth centuries to make Roman liturgy and its chant normative in their rapidly broadening empire. To this end it would obviously have been useful if the Franks could have obtained from Rome written copies of the music which had to be sung in their churches. But were they able to acquire such copies? It has been suggested that they were available as early as 747, for in the records of the Synod of Cloveshoe (somewhere in Britain) reference is made to various aspects of worship, including chant, which are to be performed according to a written Roman exemplar (see above, III.4). From the context it is not, however, certain that the reference is to notated books.
There is a similar difficulty in interpreting the instruction in Charlemagne's Admonitio generalis of 789 that boys should be taught, among other things, nota— 'signs', but of what sort? (MGH Capitularia regum francorum,. i. 60). The word is found qualified by a gloss (121) which connects it with the notarius, the secretary, rather than the writer of music, which implies simply that the boys should be taught how to write. Nor do we find clear references to musical notation in the Institutiones cdnonicorum resulting from the Councils of Aachen of 816 and 817, and these do include sections on the duties of the cantor (MGH Concilia aevi Karolini, i, Pars 2, 414).
From the ninth century come our first Frankish treatises on music, and for the first time there are distinct references to musical notation. Indeed, by the end of the century, when Hucbald of Saint-Amand wrote his De harmonica institutione, different types of neumes were clearly in use (GS i. 117, Babb, 36—7). Back in the middle of the century, Aurelian of Reome's treatise Musica disciplina clearly envisages the presence of written musical signs: 'Plagis proti melodia . . . habet notarum formas . . .' (CSM 21, xix. 34, trans. Ponte 1968, 48) and 'haec consistit figura notarum . . .' (xix. 42, Ponte 1968, 49).
It has been suggested that some of Aurelian's references to acute and circumflex accents also refer to written signs, but this is not explicitly stated: the passages read more easily if we assume that the vocal inflexion, but not a written representation of it, is meant (xix. 10-11, Ponte 1968, 47).
The evidence of documentary references, therefore, seems to indicate beginnings in the first half of the ninth century, which is the same period as suggested by the surviving neumes themselves.
(iv) Parallel Systems: Oratorical Accents, Punctuation, Ekphonetic Notation
Three parallel systems of signs which indicate features of vocal delivery have been compared with neumatic notation and put forward as its predecessor. These are the
4. The Origins of Chant Notation 365
prosodic accents of classical Greek literature, the punctuation signs used in copying texts of the early Middle Ages, and the ekphonetic signs used to indicate the delivery of liturgical lessons.
(a) Prosodic Accents
Thompson 1912, 61-4; Laum 1928; Schwyzer 1939.
'De accentibus toni oritur nota quae dicitur neuma' ('It is from the prosodic accents that there comes a sign called a neume') states the anonymous author of the treatise Quid est cantus? in the eleventh-century Vatican manuscript Pal. lat. 235. The accent-theory of the origins of Western chant notation was put forward by Coussemaker (1852, 154); it gained wide currency through the explanation in Pothier's Les Melodies gregoriennes (1880), and then in Suhol's handbook of chant notation (1935). Not all modern authors have accepted it, however. The main problem has been that it is difficult to detect any continuous use of prosodic accents in Western literary manuscripts up to and including the time when music was first notated in the ninth century.
Prosodic accent-signs were used by the Greeks from early times, but more especially from the third century bc, when Greek gradually became an international language of learning throughout the Near East, and non-natives needed guides on the correct pronunciation of written texts. Aristophanes of Byzantium and his pupil Aristarchos developed the system in Alexandria. In the second century ad Herodian led a resurgence of interest in grammatical studies, including accents, and they are common in second-century papyri, though not always uniform in type. Accents of the type still used today are especially prominent in minuscule hands of the eighth century onward.
The acute accent (oxeia) signified that a syllable should be pronounced with a higher pitch. The grave accent (bareia) signified a lower pitching of the syllable; it is much less common, being generally reserved for syllables which would normally be acute, but for some reason (for example, not ending a clause) should be lower, or at 'normal' pitch. The circumflex accent (perispdmene) signified a higher pitched syllable falling lower.
The signs were divided into groups: the tonoi (pitch-signs), chronoi (duration), pneumata (breathing), and pathe (mutations). Starting a word with a 'rough breathing', that is an H-sound, was signified by the daseia; absence thereof, 'soft breathing', by the psile. Long and short were the makra and bracheia, respectively. The hyphen linked words, denoting elision. The diastole separated them. The apostrophos indicated a short break. The prosodic accents are shown in Table IV.4.1, together with their Latin names.
Such prosodic signs are very rare in Latin sources. Almost the only one found is an acute over an exclamatory monosyllable such as o. On the other hand, many Latin grammarians (in this as in many other ways heavily indebted to Greek models) include brief accounts of prosodic signs. They were certainly known in Carolingian
366
IV. Notation Table IV. 4.1. Prosodic accents
Greek
tonoi
prosodeia oxeia " bareia
chronoi
pneumata
perispomene (oxybareia)
makra bracheia
daseia psile
pathe
apostrophos hyphen
Accent Latin
c
n
^
\J v
\
diastole (hypodiastole) j
torn
accentus acutus " gravis " circumflexus (later r*t)
tempora
spintus
longus brevis
asper (later L c ) lenis (later j 3 )
vanationes apostrophus conjunctio separatio
times, for there exists an eighth-century treatise De accentibus falsely attributed to Priscian (Keil, lii. 517-28); and among writings attributed to Alcuin is a grammar in dialogue form, which, drawing on Priscian, mentions the three toni and two spiritus. The question arises as to whether they reflected or influenced the actual pronunciation practised by Latin speakers at various periods. At least they might have contributed to an emerging practice of notating music.
(b) Punctuation
Early medieval punctuation systems have also been cited as possible catalysts for the new musical notation. Either the signs could have been taken over literally, with modifications of shape or significance, or the general principle of marking syntactic units with a sign might also have been of importance. Chant singers made wide use of well-known melodies which could be adapted to many different texts. Punctuation might therefore guide the application of standard melodic phrases to units of text. (See Bischoff 1986, 224-5.)
Grammarians of antiquity up to Isidore of Seville used a point, placed low, in the middle, or high, to indicate pauses of different length:
comma (short pause) -*--
colon (medium pause) -'-
periodus (end of a sentence) -
4. The Origins of Chant Notation
367
In Carolingian times two systems were prevalent:
usual later Carolingian     Carolingian (Tsidorian')
short pause _*_L_ -•-
long pause .f )   y -!-
A variety of signs for the question mark were known (see Bischoff 1986, 197), some of which bear a distinct resemblance to one of the several quilisma signs:
«S ^     rs   rJ af
Treitler (1982, 'Early'; 1984) has tried hard to see notation from this angle. He has started from the fact that the notation in early chant-books does not record pitches, and must therefore have a different function, one of guiding the articulation and delivery of the text. The signs used to build up a music notation should therefore not be sought among prosodic accents, which were concerned with pitch, but among punctuation signs, which concern the articulation of the texts into its syntactic units. And among other things he has drawn attention to the similarity of the forms of question mark found in late eighth-century manuscripts and forms of the quilisma in the next century. Both indicate a rising vocal gesture.
(c) Ekphonetic Notation
A special type of punctuation was used to guide the intonation of liturgical texts, lessons in particular. The Christian world knew many different systems. The Syrian church had a system employing dots even before the schism of the sixth century. The Byzantine system can be traced back at least to the ninth century, and it then gave rise to the Slavonic, Georgian, and Armenian traditions. Hebrew Bibles also use a system traditionally supposed to have been invented in the ninth century.
That this is something different in degree from simple punctuation may be seen in Byzantine examples (Wellesz 1961, 249 ff., pis. 1-2; Thibaut 1913, 39, 41); see Table IV.4.2. Although the exact meaning of the individual signs is not always clear, the way they function is not in doubt: in general they give an indication of the way each phrase should be delivered, usually by placing one sign at the beginning of the phrase and another at the end. They were usually positioned beneath the appropriate syllables, and might be written in red ink, thus being distinct in both ways from prosodic accents. The signs do not merely mark off the text in syntactic units, they carry a range of pitch signification well beyond that of the punctuation signs. The degree to which they were developed from prosodic accents is debatable, and it is probable that the prosodic accents were themselves known only academically at the time the ekphonetic system was developed, that is, they no longer in practice corresponded to the normal manner of declaiming Greek texts. This would not, however, have prevented their utilization in a newly developing system of lection notation.
368 IV. Notation
Table IV.4.2. Byzantine ekphonetic signs (after Wellesz, 1961, 252)
Oxeia (diplai)		the voice rises to a higher pitch and stays there until the
		second Oxeia
Kremaste	y	rise of the voice with a slight emphasis
Bareia (diplai)	\ w	fall of the voice with a certain emphasis
Apostrophos		lower pitch without emphasis
Apeso exo	j.......y	lower . . . higher
Kentemata	•	rise
Synemba		joining two words smoothly
Hypokrisis	> > >	separation, shorter or longer
Teleia	.......+	pause
Syrmatike		undulating movement
Paraklitike	y	in a beseeching manner
Kathiste	vA.	normal starting tone
There seems little doubt that Byzantine ekphonetic and Byzantine neumatic notation are related to each other, for some of the signs are identical. The oldest stages of Byzantine neumatic notation that can be recovered, however, the so-called Palaeobyzantine types from the tenth century onward, are to be distinguished from ekphonetic notation in that each sign (or group) refers to one syllable, whereas the basis of ekphonetic notation is the phrase. Yet this does not mean that in neumatic notation every syllable has its melodic sign: that stage was not reached until the eleventh century, and in older sources many syllables are left without a sign. Another distinction is that neumatic notation was entered over the text.
It does not seem possible, therefore, to speak of a steady development from accents to ekphonetic notation to neumatic notation, despite the signs held in common. It seems safe to speak of three different systems which shared some ideas and graphic shapes (see Table IV.4.3). The principal reason why they are not more closely related is that they served different purposes. Ekphonetic aids were needed to specify a range of intonation formulas not discernible simply from the normal punctuation of the text. Neumatic notation was needed to suggest much more varied melodic outlines, where each syllable might be delivered with an individual melodic gesture.
The role of ekphonetic notations in the West is still somewhat obscure. Ekphonetic signs of many different types, that is, signs which seem to affect the delivery of complete phrases, rather than separate syllables, may indeed be found in Western books, but they seem to be no earlier than neumes, nor to adhere to any unified tradition or traditions. What is more, no one manuscript seems to use more than eight or nine signs, so that in scope it is something like an extended punctuation system. Indeed, for the experienced singer of a clearly punctuated manuscript, such extra
4. The Origins of Chant Notation 369
Table IV.4.3. Accents, ekphonetic signs, and netimes (Byzantine)
Prosodic accents		Ekphonetic signs		Palaeobyzantine neumes	
Oxeia	/	Oxeia	/	Oxeia	/
		Kremaste	J	Petaste	J
		Kentemata	* ♦	Kentemata*	« •
Bareia	\	Bareia	\	Bareia	\
		Apostrophos	y	Apostrophos	3
Oxybareia	a n				
		Apeso exo	,.....J	Apeso exo*	
Makra	—				
Bracheia	u	Kathiste	</\		
		Syrmatike		Syrma*	
		Paraklitike	V	Paraklitike*	
Daseia	V LC				
Psile	hj3				
Apostrophos	i	Hypokrisis	> > 5 3	Katabasma*	\ 1
		Teleia	+	Stavros*	+
Hyphen		Synemba		Klasma*	
Diastole	■)				
not recognized by all experts
signs were superfluous. There seems little evidence, therefore, to suggest that Western neumatic notation might have used ekphonetic signs as a springboard for the development of its system. (For examples see Sufiol 1935, 181; Jammers 1965, Tafeln, 105; Ekberg, 'Ekphonetic Notation', NG 102; Rankin in Fenlon 1982, 90-4.)
(d) Byzantine notation
From the previous section it will have become clear that a Western adoption or imitation of Byzantine chant notation is very unlikely to have occurred. The idea has nevertheless had its supporters. The Greek names of many neumes, the similarity of some shapes, even if they signified something different, have exercised a considerable fascination. Thibaut, whose studies of Byzantine ekphonetic notation have just been referred to, believed that Western chant notation derived from the Byzantine ekphonetic signs. More recently, Floros (1970) has attempted to reconstruct the earliest Byzantine chant notation and then show that in this early form it was adopted in the West. The evidence on which the theory rests appears shaky, however (see Haas 1975).
370
IV. Notation
(v) The 'Cheironomic' Theory
Before the final body of evidence which might illuminate the early history of chant notation is considered, mention should be made of another theory: that early neumes reflect the manual gestures of the cantor as he directs his choir in performance; they are cheironomic. The idea appears to have originated with Mocquereau (PalMus 11, 96), while discussion of the practicalities of chant-conducting in the Middle Ages began with Kienle (1885; see also Gindele 1951). Nearly all writers on chant notation have at least paid lip-service to the idea (Wagner II, 17; Jammers 1965, Tafeln, 23; Stablein 1975, 28), and it forms part of, for example, the theory of neumatic separation elaborated by Cardine (discussed in the next section). A variety of evidence of cheironomic systems was reviewed by Huglo (1963, 'Chironomie'), and most recently Hucke (1979) has subjected the whole matter to stringent criticism.
Cheironomic systems as commonly understood actually specify the melodic intervals to be sung, and, as Hucke points out, there is no evidence that Western chant notation was linked to any method of this sort. Even literary references to cantors conducting with hand movements are extremely sparse. The theory that chant notation—or at least the St Gall or Laon neumes usually invoked in this context— depicts cantorial hand gestures has therefore no concrete support either from what is known of cheironomy in other contexts (the music of other cultures, modern Western systems such as the Kodaly method) or from contemporary documentation. This does not mean to say that it is erroneous, only that it cannot be proven. Practical experiment suggests that it is possible to reproduce, say, St Gall neumes as hand gestures in conducting a choir; so the reverse is theoretically possible: that conducting gestures were reproduced as written signs. As a historical explanation it lacks all foundation, however, and scholarly caution demands that it be treated, at best, as a picturesque analogy.
(vi) The Early Transmission of Chant
So far I have suggested that (a) the earliest specimens of chant notation date back to the early ninth century; (b) the earliest unequivocal references to notation come from Aurelian of Reome, writing around 850 (the complicated composition and transmission of Aurelian's treatise make caution advisable); (c) some of the materials which might have helped in the development of a basically newfangled notation in the West were available in the Carolingian period: the accents of classical prosody and punctuation signs, with a faint possibility that Byzantine ekphonetic signs or even the beginnings of Byzantine musical notation might have existed to spur the imagination of the Franks.
Before notation was employed to codify chant melodies, and provide an aid to learning and preservation, the music was performed from memory. This state of affairs can actually be deduced from the style of the music itself. A large part of the chant repertory consists of melodies which have been adapted to fit different texts,
4. The Origins of Chant Notation
371
or which use stock ways of beginning and cadencing, and share common ways of delivering texts of a particular type. It might have been expected from this that singers would have had some room for manoeuvre in the performance of such chants. So that for the performance of the tract Jubilate cleo (Ex. II.5.4), for example, different manuscripts—representing the decisions of different cantors about how this text should be performed—would employ different formulas, at least occasionally, so that there would be a certain amount of interchange between Gl, G2, and so on, as the standard phrases of Ex. II.5.4 are labelled. But this is not so. All sources draw upon the same traditional phrases at the same time, follow the same procedure when the length of the text is different, and at first sight, seem much more like copies of the same exemplar than different rememberings and reproductions.
And yet differences, small but persistent, are to be found. They concern liquescent notes, as we might expect, and the use of such signs as the quilisma and oriscus, how to deal with notes at the semitone step (for example: a pes be or two virgae cc?), and so on. In other words, they concern surface detail, not basic structure.
On the other hand, some other types of chant show bigger differences between sources, for example, many antiphons (see those transcribed by Udovich 1980 and Fickett 1983).
This reinforces the impression that we have to do not with a repertory transmitted in writing, but one remembered and later codified differently in different places. For some chants the learning process had been so thorough and exact, the procedures for remembering so diligently rehearsed, that hardly any difference can be seen between sources. For other chants the remembering was not so uniform. (For much valuable comparative material, with discussion, see Van der Werf 1983.)
(vii) Some Conclusions
When trying to decide when the chant repertory was first systematically codified with musical notation, one should bear in mind that early manuscripts were not 'performing scores' in the modern sense, but works for study and reference. There must always have been a greater or lesser discrepancy between the music sung in church and the master exemplar in the song-school, at least in soloists' music.
If our earliest manuscripts are independent of each other—containing different readings, as well as using a different notation—then they do not suggest a date for the beginnings of notation any earlier than their actual date of copying. For their particular churches they might well be the first such books ever made. Whatever the previous uses of notation, it does not seem to have been used for a whole gradual before the late ninth century (and not until a century later for office chants). We might envisage a period when the various notation-types were becoming established, in the ninth century, and were then finally employed for a major codifying project, the whole mass repertory.
If, as Levy believes, a standard exemplar existed earlier, perhaps compiled at Charlemagne's instigation, then later cantors and scribes must have felt free to copy it
372
I\r. Notation
(a) using their own locally developed notations and (b) incorporating their own 'interpretation' of details such as those mentioned above. In other words, they would be working partly from memory, partly from an exemplar. This way of making chant-books might well have been one of the commonest throughout the Middle Ages, for an exactly analogous situation arose in the eleventh and twelfth centuries when churches all over Europe wanted to make new records of their repertory in staff-notation.
The essence of early chant notation must lie, as Treitler has indicated, in its ability to link the appropriate vocal gesture to the text being delivered. This is easiest to envisage in the case of highly standardized melodic units. Treitler uses responsory verses to illustrate the technique (Treitler 1984, 172, reproduced from Rankin 1984). The examples of psalm verses from introits provided by Hucke (1985) can be read as a parallel demonstration. In Ex. IV.4.1 are the openings of some mode 3 introits already cited above (11.11: compare Ex. II.11.1). I transcribe the versions in the Graduale triplex.
Ex. IV.4.1. Opening of mode 3 introits (St Gall notation)
P
P
P
Be-ne - di - ci - te
r»    0 -
-0*t*-w-
Re - pie - a tur
■yst-
-W
E - go      au- -tern
■*fl---a—*-•-
Vo - cem   io-cun-di - ta-tis
We could imagine a scenario something like the following. The cantor knows that a mode 3 melody is traditionally sung for the following texts: Repleatur os meum (GT 246), Ego autem sicut oliva (GT424), Vocetn iocunditatis (GT 229), and Benedicite dominum (GT 607). He will be aiming for c as the reciting note in the first phrase, usually achieved directly from G, or with an ornamental gesture around G and a, and, if the text warrants it, some preliminary movement around D and E. That is, he will know a number of stereotyped openings appropriate for introits in this mode. The neumatic notation reminds him which opening is best for each particular text. Benedicite dominum is the simplest, for only one upward movement is indicated. Repleatur os meum has three ascending gestures, which must be for EEE, DG, and Gc respectively. The ornamental flourish on 'auftemj' in Ego eutem is instantly
4. The Origins of Musical Notation 373
recognizable, and the more leisurely progress of Vocem iucunditatis is also, in its way, quite clear, for it is bound to start with FF, DC, the arrival at c is unmistakable, and in between F, G and a are almost certainly the onlv notes available. The notation therefore guides the controlled delivery of the text towards the first structural crux. (The length of its stay there is also clear in this notation from the apostrophes, associated in this mode principally with c.) It does not specify pitches: the matrix from which they will be drawn is already known. It specifies what words belong to the opening gesture, and how the syllables are disposed within it, its function therefore resembling that of punctuation more than that of modern pitch notation. But it is very highly developed type of punctuation system, one which indicates not simply the main syntactic divisions in the text but the shaping of much detail within those divisions. And, if some signs have been suggested by punctuation signs, others may well have come from oratorical accents, or simply from a sense of the contour of a melody (Duchez 1979, 'Representation').
Once the decision had been taken to record each and every movement of the melody, the latter aspect of chant notation became predominant. The fact that no single point of departure leads to a complete repertory of signs for chant notation, and that different centres employed different styles of writing, suggests a period of vigorous but controlled, new initiative. In this context it seems unnecessary to look outside the Carolingian realm for a pre-existing system.
We might see the early history of notation in three principal stages. Adiastematic notation is best adapted to represent the standard idioms and formulas of the earliest 'Gregorian' layer of music. Its use became most urgent precisely when the old musical types were increasingly being challenged by newer, non-formulaic, compositions. For these, pitch-notation was eventually indispensable. The third stage, the specification of rhythm, was not needed until the music for two or more voices both moving in regular rhythm had to be coordinated; this stage concerns not plainchant but the Parisian polyphony of the twelfth to thirteenth century.
IV.5. THE NOTATION OF RHYTHM
(i) Rhythmic Elements in Early Notations
(ii) Rhythm in Simple Antiphons
(iii) Cardine's 'Gregorian Semiology'
(iv) The Evidence of Theorists
(v) Conclusions
The notation of several early chant manuscripts includes indications of rhythmic differentiation between individual notes or groups of notes. Richest in such indications are the early sources from Switzerland and south Germany (such as St Gall, Stiftsbibliothek 359 and 339, from St Gall, Einsiedeln, Stiftsbibliothek 121, from Einsiedeln, and Bamberg, Staatsbibliothek, lit. 6 from Regensburg) and Laon, Eibliotheque Municipale 239, but very many other manuscripts indicate at least some
374
IV. Notation
degree of rhythmic differentiation. There are several ways in which this was done: (a) the normal shape of a sign was altered in such a way as to suggest a rhythmic alteration, (b) in the Eastern manuscripts, extra elements were added to the normal signs, usually the short bar known as the episema, (c) the way in which several notes were comprehended in one sign was altered, separating what might have been joined and joining what might have been separate, (d) so-called 'significative letters' were placed adjacent to the sign.
The medieval tables of musical signs are later than the early sources just mentioned, and it is not surprising that they provide no explanation of durational significance. Only the Italian tables in Montecassino, Archivio della Badia 318 (Coussemaker 1852, pi. XXXVII; Ferretti 1929, 193, pi. II) and Florence, Biblioteca Magliabecchiana F. 3. 565 (Ferretti, 195) indicate a differentiation: thus in Montecassino 318 we find percussionalis brevis (a dot or punctum) and percussionalis longa (a dash or tractulus).
(i) Rhythmic Elements in Early Notations
The meaning of the significative letters, at least as they were understood at St Gall, is given in a letter attributed to Notker (see PalMus 4, 10 and PI. 5. B-D; also Suhol 1935, 134; NG 13, 132; critical edition by Froger 1962). St Gall tradition ascribed their invention to the Roman cantor 'Romanus', who was supposed to have brought chant-books from Rome to St Gall in the eighth century. Hence the name 'Romanian letters', coined by Schubiger (1858). But we have no contemporary explanation of the other features mentioned above. The only way in which we can approximately understand the notation is by careful comparison of the sources among themselves.
Significative letters can be found in many manuscripts up to the early eleventh century from different areas of Europe, in Breton and Aquitanian sources and the early Winchester manuscripts, for example. But they are commonest in the abovementioned Eastern and Laon manuscripts. Smits van Waesberghe's special study of them (1938-42) reported that Einsiedeln 121 contained the staggering total of 32,378 significative letters. Notker's letter ascribed a meaning to almost every letter of the alphabet, but several cannot be traced in extant manuscripts, or are very rare. Preferences naturally differ from one area to another, and we have no Notker to explain the system in, say, Laon 239. Not all the letters refer to rhythmic properties of performance, some concerning dynamic or melodic features. The commonest in the Eastern sources are the following:
dynamic:   / = cum fragore seu frendore: with harsh attack
k = klenche (Gk.?) or clange: with ringing tone melodic:     a = altius: higher in pitch
e or eq = equaliter: at the same pitch
/, 70, or in = inferius, iosum, or iusum: lower in pitch
iv = inferius valde: much lower
5. The Notation of Rhythm 375
/ = levate: rise to a higher pitch 5 = sursum: ascend to a higher pitch rhythmic:   c = cito or celeriter: quickly
/ = trahete or tenete: drag, hold x = expectate: wait
The preferences of Laon 239 are for the following:
melodic:     eg = equaliter: at the same pitch
h = humihter: at a low pitch
s = sursum = ascend to a higher pitch rhythmic:   a = augete: lengthen
c = cito or celeriter: quickly
n or nl or nt = naturaliter: normal
t — trahete or tenete: drag, hold
In both traditions, m or md = mediocriter: moderately, can qualify either a melodic or a rhythmic indication, or even appear on its own, when its meaning has to be deduced from the context.
(More complete lists of letters may be found in studies of the individual notations, and in Sunol 1935, 134-8, 141, 143-4, etc.)
Among the letters indicating melodic features, e is particularly important, for it specifies the relationship between two disjunct note-groups. The others usually (though by no means exclusively) serve to warn of intervals larger than a tone (or semitone). Ex. IV.5.1. shows a typical use of the letters e, i, m, and s (medieval 5 looks like an elongated r).
Ex. IV.5.1. From offertory Benedictus es Domine (GT 277, notation of Einsiedeln 121)
-6-	1 t     r     ' "	--———-	^, -_!_
/			
W-W7i	..^ «—»**——t—Mj—		* *
v	--0*-		_K_C
in     la - bi - is      me  -  is pronun-ti - a- -ui
Rhythmic differentiation depends almost entirely on c and t. They are the key to understanding the notational signs that have been altered from their usual form or given an episema, for a fairly consistent correspondence (though it is by no means complete) can be seen between the use of, on the one hand, the letter c and the normal shapes for signs, and, on the other hand, the letter / and the altered or supplemented shapes. Typical normal and altered shapes in St Gall notation are given in Ex. IV.5.2, together with two passages horn Alleluia Pascha nostrum, with the signs in St Gall 359 and 339 and in Einsiedeln 121.
The letter c is always used in conjunction with the normal signs, whereas the letter t is usually allied to an episema (in the same or one of the other sources). Striking confirmation of the sense of the episemas comes from the notation of Laon 239. This
376 TV. Notation
Ex. IV.5.2. Normal and modified signs in St Gall notation (selection)
normal
clivis	
pes	J
porrectus	rv
torculus	J)
climacus	A
	/
scandicus	
modified
4 /f n
ji j) j] ju X
A X /. /■- /-
from Alleluia Pascha nostrum (notation of St.Gall 359, 339 and Einsiedeln 121)
359 339      - -E 121   ~e -
--v-r
/*_0-
w*9»
9»9»9 *9
Al-le - lu-
-la.
339     -  /  J a        »/i   /^a A- / «A. /=. <T -
E 121 «•
* *9> *m "m *9 0
immo-la- -tus est
source does not have episemas, but makes much more frequent use of the letters a and /, both meaning a lengthening of some sort. Ex. IV.5.3 gives first the 'normal' and the 'longer' signs in Laon 239, then the same extracts horn Alleluia Pascha nostrum.
One important feature of the second part of Ex. IV.5.2 in particular is that the notes are grouped in exactly the same way—single notes, twos, threes, and so on—in all the sources. The phrasing of the melody was obviously important enough to have been transmitted with a fair degree of uniformity. In a strikingly large number of
5. The Notation of Rhythm Ex. IV.5.3. Normal and 'longer' signs in Laon notation (selection)
377
normal 'longer'
1 'r%\
cnvis
pes J r JX
porrectus *Y
torculus -A r* A -AT
climacus » ^ " r
/ ' /
scandicus • ^ ^
i
from Alleluia Pascha nostrum (notation of Laon 239) re*
Al-le - lu- -ia.
J 'T
f J
ft
•1. ^
f
immo-la-
-tus
est
instances, the group ends with a longer note. The rhythmic significance of the point of separation between the groups has been studied in depth by Cardine (his term for it was 'coupure neumatique').
It is enlightening to observe the consistency with which one of the standard phrases of the repertory is reproduced in these early sources. Ex. IV.5.4 gives a standard phrase from the graduals of mode 5 (phrase Fl in Apel's analysis) with the neumes of Laon 239 and St Gall 359. The notation is given complete for the phrase in Iustus non conturbabitur, whereas for five other graduals only deviations are recorded. (Three of them are not present in Laon 239.) The lack of disagreement is obvious.
The question immediately arises, how slow is slow (tenete or augete) and how fast is fast (cito or celeriter)? There is unfortunately no clear answer. And with how many gradations of duration do we have to reckon (however approximately)?—two (since the signs on their own, without letters, can be roughly divided only into 'normal' and 'longer'), or perhaps more? Do the significative letters simply warn about something
378
IV. Notation
Ex. IV.5.4. A standard phrase from mode 5 graduals (Apel Fl) with Laon and St Gall notation
Iustus conturbabitur GT 476
Christus factus GT 148
* f
s*mm >j
r000O 0r
qui-a do-
-mi - nus
Laon
1 J
tf
St.Gall /V)
fir-
-mat
Exiit sermo* GT 636
St.Gall //
Ecce sacerdos* (no variants) GT 486
Timebunt gentes GT 265
Laon 1n
Laon *
Misit dominus*      (no variants) GT 261
in the notation anyway, or do they mark an extra degree of slowness or fastness? What shades of difference divide the signs given in Ex. IV.5.5?
It is noticeable that in the Eastern sources c is used far more with the clivis than with the pes, whereas in Laon c appears only with the pes.
Ex. IV.5.5. Possible gradations of duration in pes and clivis (Laon and St Gall notations)
<-       longer      shorter -►
Laon
St. Gal!
Laon
i.
St.Gall
t.
c
5. The Notation of Rhythm 379
Some writers have argued that the shorter and longer notes stand in a strict mensural relationship to one another, so that Murray (1963) transcribed the above passage from All. Pascha nostrum as shown in Ex. IV.5.6.
Ex. IV.5.6. Mensural transcription of passages horn Alleluia Pascha nostrum (after Murray 1963)
p i jTf j77m f f j{&-&-tt!jO r dfErTf tuim
Al-le-lu- -ia.
immola-
{7=Fr\	Fff	P   |i    ff--i-7—i	rf ^-				
9 U T t,TJ [J 1	—U^-J	--A-*—i— L_»	r (				' »J J—
-tus est
The Vatican gradual of 1908, produced under Pothier's leadership, did not reflect any of the rhythmic detail of the above-mentioned manuscripts, for Pothier believed it was only of local significance. The editions produced by Solesmes under Mocquereau's direction, however, included dots and horizontal bars, indicating lengthening, which derive from the early sources. In Solesmes practice, these are regarded as rhythmic nuances, not susceptible to rigid measurement. (The Solesmes interpretation under Mocquereau and Gajard, famous from many gramophone recordings, is explained in Mocquereau 1908-27 and Gajard 1951.) At another extreme, it is possible to regard the shorter notes as a type of ornamental figuration, to be performed like the ornaments in Baroque music, or, better, the roulades in some non-European musics. A precursor in some wise of this type of modern performance was Dechevrens, whose fanciful transcription of All. Pascha nostrum is shown in Ex. IV.5.7 (taken from Decheverens 1898, iii/2, 465).
There can be no doubt that rhythmic differentiation was an essential element in the practice of those choirs for whom the St Gall, Laon and other sources were written. The fact that the Laon source is widely separate geographically from the others suggests that this way of singing chant was quite widespread. How long it persisted is unclear.
Several comparative tables of signs from early sources have been published, displaying the range of characters used for each melodic progression. The most detailed tables for St Gall notation have been published by Cardine (1968 etc.), and for Laon 239 in the English translation of Cardine. In PalMus 11 (55-8) and Sunol (1935, 147-52) signs from St Gall, Laon 239, and Chartres, Bibliotheque Municipale 47 are presented.
380 IV. Notation
Ex. IV.5.7. Mensural transcription of passages from Alleluia Pascha nostrum (after Dechcvrens 1898, iii/2, 465)
t^-=- i     h T) 1 1—W f t	i h f .f r		_< ttt *-»—		•—rf-M—*--fh-
gyj.    J j^J' iX J? ' 7 Al-lelu - ia.	i—f * # —fr	t't 7 -\ |i ft	~m f i n		ffl h ~~I
		i j' r i'\	Xi-,-7-^	4-^-1-	1          j  7 tf
..... immo-la-
i1 fTfThH	|-T"*""	f~rf~		,—< -si			=;=^-	~1—v~
		^~	IE Iff 1	— r— -tu	s	«1 7 v.		—^— est .....
(ii) Rhythm in Simple Antiphons
Interesting approaches to the rhythm of office antiphons have been developed by Jammers (1937) and Lipphardt (1950). The most valuable part of their work has consisted in comparing antiphons of the same melodic family in order to see if they had a common rhythmic tradition, using the evidence of the notation in Hartker's ar.tiphoner, St Gall 390-391 (PalMus II/l). This has helped establish conventions for the relationships between the signs for single notes and those for more than one note, at least in predominantly syllabic passages. Jammers provided transcriptions of (among other chants) antiphons from Gevaert's Themes 1-11, Lipphardt from Theme 29. Briefer demonstrations of the same principle can be found in Cardine's w>rk (for example in Semiologie gregotienne, ch. 3). Consider Ex. IV.5.8, which develops one of Lipphardt's examples.
If we assume that the melody proceeds in basically equal notes, we can see that the ar.gular pes at the beginning of phrase two equals two syllables with single notes. The virga with episema for '[sa]cra[ta]' likewise seems equivalent to two simple virgae. So aho the clivis with episema at the end of the third phrase, which in turn suggests that the clivis with c in phrase two should be twice as fast. At the beginning of phrase three, it is particularly interesting to see how the same notes are adapted to two syllables (a 'slow' pes and 'slow' clivis), three syllables ('slow' pes and two virgae), four syllables (four simple notes) or even five syllables ('glori-' with a c, that is quickly). It is not difficult to make a transcription in 2/4 or 4/4 time of these antiphons (Lipphardt opts quixotically for 6/4), arranging stressed syllables to coincide with the beginnings o^ bars. One wonders if the melody had an inherent rhythm, so to speak, for which
Ex. IV.5.8. Antiphons with the same melody (notation of St Gall 390-391, PalMus II/l; pitches derived from Bamberg, Staatsbibl. lit. 25)
r-6-r			/ J	r			/ r r	
Po-			» aa	*—	~fw-	—•-rt-1	0 a	a
W-	• • Be - ne-- /		-dic-ta / J	tu r	-0J.- in	1-~#- mu-li-/ nc	-e - ri-/	-bus
4-								
Kb		—•-•-	0 1	-*%-1	—	^-Tg-	«	•
p1—•— Dig-	-•- -na - re . /	me lau-	-da - re / </	te r	uir-- /	-go sa-/	-cra-/ /	-ta r
3?-								
^5-:-			.	-ft—	~0—*-	• O-	m m	a>
r- n /	0-•- E - xal-- /	/ /	-ta - ta V	es r	san-cta	De-i	ge - ni-/ /	-trix /
3?---			—s m \-					
ft-		—•—0-	--		-	^-	■ •	•
r—•— Gau-	-»-*- -de Ma-/ /	-ri - a / S	uir-	-go f	cun-_ /	-etas /   / -	he - re-/ /	-ses
4-								
ft-		--0-0-			• •-	*   « 0	• *	a>
r- n /	~m—•- Pa - ra-_ /	-dy -si	por-V	-te / /•	per te . /	nobis a-/ -	-per-tae i/ /»	sunt
-Ü—--			-_p-e-					
ft) .	1-.__A-			* «3	>-0—■-		£* *3	* •
Post
partum
-go m-
-ui -o-
-la - ta
per-man - sis-ti
/ /
rr-1										
										
%-		-~0-«-		—-	-—--		—-m-0-			
v- n /	-0-- et V	bene-/ /		—0— -dic-n	*1- -tus	fructus _ /	*ar_ uentris		0- tu-	•- -i.
J-■-										
	—r*-	- •—•—■		———r-1	--		—_—•-			
r—z— da	i—•- mi-- /•	-hi ui r-f /		-*m- -tu-n	- -tern n	^ * cont ra - /	—•- hostes /	/ /	-0- tu-	- -OS.
n										
fa-										
m-	—,——	-•—•-	mm 0*		0^		a •	-		
n /	—0 su-per _ /	choros / /	w w an-ge-	• • -)o-n	9 a -rum / /	^* ad cae - /	w - les-	-ti-a	W reg-	W -na.
J?-■-1										
8) *	*	—0-0-					—-—#—			
**L—— so-p /	—•-1 -la in-- /	-tere-		— -mis-n	a— -ti in / */	^ * u - ni-- /	—♦- -uerso / /		—•- mun-	-- -do. —t
m-■-'	•						*			
yL_=- quae	—•-' ho - di-	* * • -e glori-n		^— -o-/.	*— -sa cum /T	an-ge-. /	-•—-- -lis tri-/ /		-0- -um-- /	~0)- -phas.
0										
ft-									^_	
m-1	1—0-	* *		^*—m—1	-1		-0---1	-—i		i—•-"
De-
geni-
-trix in-ter-ce-de
pro no - bis.
382
IV. Notation
these texts were deliberately composed. Or were there traditional ways of 'rhythmiciz-ing' texts to bring them into line with this and other melodic schemes? It has to be said that this group of antiphons displays a more regular structure than many others, and attempts to find similar rhythmic schemes elsewhere often appear improbable. Furthermore, different performance conventions may have existed for more ornate chants.
Be that as it may, Cardine developed a theory of syllabic equivalence, relating the length of notes to the normal delivery of a syllable with a single note. This time-unit is the 'syllabic beat' ('temps syllabique'). Such a note is normally notated in St Gall sources as an ordinary virga or tractulus. It can be lengthened with an episema or shortened by the letter 'c'. The angular pes and the clivis with episema are then equivalent to two syllabic beats, whereas the normal pes and clivis will represent shorter durations.
(iii) Cardine's 'Gregorian Semiology'
It will by now have become clear how much can be learned from the careful comparison of manuscripts with one another, and of similar musical passages in the same manuscript. Particularly comprehensive and painstaking work of this kind has been accomplished by Cardine and students at the Pontificio Istituto di Musica Sacra in Rome. (There are summaries of these studies by Albarosa 1974, 1977, 1983, and lists in the Cardine Festschrift 1980 and Albarosa 1983. Many have been published in the journals Etudes gregoriennes and Beiträge zur Gregorianik. Cardine summed up his own work in the book Semiologia grego/iana, 1968, which subsequently appeared in French and English versions, 1970 and 1982 respectively).
Cardine dubbed these studies 'semiology', in the general sense of 'study of signs'. (The term is not usually used of the mechanics of writing-systems but of modes of human communication, particularly in the area of linguistics, and also recently in the analysis of music.) Most of the studies have taken a single sign and explored its use throughout one or more of the early manuscripts referred to above. The main emphasis of the work has been on the rhythmic weight of the notes represented by the signs—light or heavy, part of a fluid progression or marking a pause in melodic movement—and their articulation, with special reference to the grouping and separation of notes. It has shown that the early scribes displayed considerable finesse in adapting signs to reflect nuances of performance.
To summarize all the findings to date of this type of research is naturally impossible here, but some idea of the insights gained may be seen through one example, Ponchelet's work (1973) on the use of the salicus in St Gall 359 (the book is a cantatorium; since the salicus is not used in tracts, this means in effect that the investigation concerns graduals and alleluias).
If followed by other notes in a large group, the salicus has an episema on the final virga 104 out of 120 times; but when there is a change of syllable after the salicus the episema is added only six out of 239 times. (All these instances are displayed in tables,
5. The Notation of Rhythm 383
arranged according the melodic context.) This seems to suggest that the last note is always prolonged, for a change of syllable will naturally cause at least a slight delay on the note. For the salicus followed by further notes in a large group, further analysis distinguishes between cases where it is followed by a higher note, a note at the same pitch, or a lower note. Cases are traced where the same melodic phrase is found over one syllable or split between two or more syllables (for example in the cadence formula for the verses of the 'Iustus ut palma' group of mode 2 graduals). Standard phrases of this sort naturally provide prime material for investigation.
The results do not, of course, necessarily bear on other manuscripts. As Ponchelet points out, St Gall 339 is far more sparing of the episema on the last note of the salicus. Does this indicate a change of performance practice at St Gall? How widespread, then, was the type of performance indicated in St Gall 359? St Gall 359 stands near the beginning of the written tradition at St Gall, perhaps near the beginning of the singing tradition; one might surmise that later scribes felt less need to include episemas, the performance tradition being by then more firmly established. The episema in the later manuscripts would by this reasoning carry greater weight. Comparisons with other traditions, those of Laon 239 and Chartres 47, are illuminating. In cases where the salicus in St Gall 359 is followed by other notes, these sources have a salicus on only about one in eight occasions; where the salicus precedes a change of syllable, the proportion is about half. When not notating a salicus, Laon 239 usually has a scandicus with / by the final note, while Chartres 47 has a simple scandicus.
Ponchelet can therefore claim that in the salicus in these early sources rhythmic weight is always directed on to the final note. For Chartres 47 one has to assume that in many cases the weight is felt simply because it is the final note of the three, even in the absence of any graphic emphasis (the oriscus element, or a significative letter). The nature of the sign is investigated purely in rhythmic terms, as is also the case with the quilisma in the studies by Cardine (1968) and Wiesli (1966). This means inevitably that the distinction between all the signs which comprise three notes in ascending order, scandicus, salicus, and quilisma, is a very fine one. If the oriscus element in the salicus (or the 'trembling' element in the quilisma) has any other meaning—say, purely hypothetically, the non-diatonic pitch suggested by Wagner— then this type of study will not reveal it, for its principal tools of investigation are purely rhythmic: episema, significative letters, articulation into note-groups. It has to be said that if the oriscus element really does represent some extraordinary element of performance, its relatively rare occurrence in Laon 239 and Chartres 47, in the contexts investigated, is more difficult to explain.
The early editions published at Solesmes of music-books for the modern liturgy— the Liber usualis is still the best known—included numerous dots, bars, and other signs, inspired by the notation of the early sources. The work of the 'semiologists' has resulted in a new sensitivity to these matters, and it is common for modern performers to use the Graduate triplex, where the notation of Laon 239 and one of the early Eastern sources is copied into the Solesmes edition of the Graduale Romanům of
384
/V. Notation
1974. Manuals of the new type of chant performance have been produced by Agustoni (1963) and by Agustoni and Goschl (1987).
(iv) The Evidence of Theorists
Early medieval writings on chant help but little in understanding how note-lengths were differentiated. One group of related writings, which was discussed at length by Vollaerts (1960), among others, has little to do with the matter, for it refers to the final notes of chants, of their phrases and sub-phrases. As Bower has explained (1989, 'Model'), this is done by analogy with the grammatical structure of the text being sung: the musical delivery should reflect the grammatical structure, by making the hierarchy of text-units clear. This group of writings includes the Scolica enchiriadis (ed. Hans Schmid 1981, pars 1, 86-7), Guido of Arezzo's Micrologus (ch. 15), and the commentaries on the latter by Aribo (CSM 2. 48-50, 65-70) and in the Commentarius in Micrologum (ed. Smits van Waesberghe 1957). Several of Guido's comments refer also to proportional relationships between 'neumes', that is, between phrases of chants; some neumes will be of equal length, while the lengths of others will stand in a proportion of one to two, one to three, two to three, or three to four. This does not refer to the durations of individual notes.
Two other passages have also been adduced in support of a metrical interpretation of the different note-lengths. The Commemoratio brevis of the late ninth century contains a number of references to the different speeds of singing appropriate to different occasions or types of chant, to the maintainance of a steady tempo (except for controlled variations such as a 'rallentando'), and also to long and short notes. Towards the close of the passage it is said that these are in the proportion 2:1:
Breves must not be slower than is fitting for breves; nor may longs be distorted in erratic haste and made faster than is appropriate for longs. But just as all breves arc short so must all longs be uniformly long, except at the divisions, which must be sung with similar care. All notes which arc long must correspond rhythmically with those which are not long through their proper inherent durations, and any chant must be performed entirely, from one end to the other, according to this same rhythmic scheme. In chant which is sung quickly this proportion is maintained even though the melody is slowed towards the end, or occasionally near the beginning (as in chant which is sung slowly and concluded in a quicker manner). For the longer values consist of the shorter, and the shorter subsist in the longer, and in such a fashion that one has always twice the duration of the other, neither more or less. While singing, one choir is always answered by the other in the same tempo, and neither may sing faster or slower. (Ed. Bailey 1979, 103.)
Since the Commemoratio brevis belongs to the same manuscript tradition as the Scolica enchiriadis, one naturally asks if the long notes referred to here are those at the ends of grammatical units, the 'syllables', 'neumes', and 'periods' of chant.
Another relevant text, unrelated to those mentioned so far, is the treatise Quid est cant us? in the Vatican manuscript Pal. lat. 235, discussed by Wagner (1904, also
5. The Notation of Rhythm 385
Wagner II, 355-8) and Baralli (1905). This short treatise, which appears to date from the eleventh century, has already been cited because of its well-known statement that 'the sign known as a neume comes from accents' (see above), and much else in the text relies on grammatical terminology. Just as there are acute, grave, and circumflex accents in the melody (tonus), says the anonymous author, so there are long and short syllables in the feet of the text. He goes on to describe some of the basic shapes of the musical signs, and gives instances from the chant repertory where they are to be found: long and short, the various 'accents', the 'tremula' (quilisma), 'triangulata' (trigon) and so on. On the basis of the analogy with metrical feet, a 2:1 relationship between long and short could be said to exist here.
(v) Conclusions
Whatever the truth behind the matter, some reservations are in order. It seems most unlikely that practice was uniform everywhere. It is true that sources are in general surprisingly uniform in their grouping of notes and even in the placing of such special features as quilismas. And the correspondences between Laon 239 and Eastern sources (and other early sources to varying degress) in matters of rhythmic detail cannot be overlooked. But the agreement is general, not exact. Any claim to have identified an 'authentic' performance tradition should be treated with caution.
Another point has been raised by Hucke (1958, 'Rhythmus'): can we necessarily assume that all types of chant were sung in the same way? We could at least imagine a difference between, say, hymns, antiphons, and responsories, to mention only the office chants. And the author of the Commemoratio brevis repeatedly mentions variety even within the antiphons, for example:
The repetitions of the antiphons which occur between the verses (of psalms) should be at the same speed as the psalms, but when the psalm is finished the antiphon is to be slowed by exactly half to its proper tempo. There is an exception in the case of the Gospel Canticles, which are sung so slowly that their antiphon should follow at the same tempo, and not be further protracted. (Ed. Bailey, 107.)
Finally, our experience of the rhythmic characteristics of music outside the tradition of Western art-music has opened our ears to the possibility of much more flexible patterns than can be recorded easily with conventional Western notation. One has only to look at transcriptions of, say, the chant of the Coptic church (NG 4. 731, col. 1) to become suspicious of simple 'equalist' or 'mensuralist' interpretations. Might not the singing of the ninth century be equally difficult to capture in modern written form? Meanwhile the minute investigation and recording of the ways in which signs were employed provides essential evidence for a better understanding not only of rhythmic matters but also of other aspects of chant notation.
386
IV. Notation
IV.6. PITCH-NOTATION
(i) From the Ninth Century to William of Dijon
(ii) Guido of Arezzo
(iii) Staff-Notation in Different Lands
Smits van Waesberghe 1957 'Origines'; Crocker 1979; Browne 1981. (i) From the Ninth Century to William of Dijon
In Ch. V, where writings on music theory which concern plainchant are discussed, the crucial importance is emphasized of the way in which the plainchant repertory was matched with the pitch-system borrowed through Boethius from classical teaching, the Greek Greater Perfect System. The link seems to have been made in the ninth century: at least, its clearest early formulation is found in the writings of Hucbald of Saint-Amand (d. 930); and the group of treatises using dasian notation—Musica enchiriadis, Scolica enchiriadis, and Commemoratio brevis—possibly earlier than Hucbald, also use a pitch-system. Dasian notation is a pitch-notation, and Hucbald too proposed the use of a pitch-specific notation. His idea was to use a series of letters (borrowed from the Alypian 'vocal' and 'instrumental' letters, again through Boethius), which would be placed by the side of the traditional notation. Hucbald was quite explicit about the reason for wanting both notations: the traditional signs indicated phrasing, longer or shorter notes, trembling notes, and so on, while the letters would make clear the pitch of the notes.
Although the dasian pitch-symbols enjoyed a certain restricted currency in some branches of the theoretical literature, they were never used in regular chant-books. (They are discussed in the next section with other notations used only in the theoretical literature.) Hucbald's set of symbols does not seem to have been taken up at all. But from the latter years of the tenth century there survive a number of sources supplied with pitch-symbols. Instead of the dasian symbols, or Hucbald's even more arcane ones, simple letters of the alphabet were used.
One of the earliest documents of this nature is the famous Winchester manuscript containing organal voices, Cambridge, Corpus Christi College 473. In the fascicle where sequence texts with their music are copied, several compositions are provided with alphabetic letters side by side with the usual signs. Letters are provided only for the first of the double versicles, for the music repeats itself in the second versicle. Conventional cadences are often left without letters. (See Holschneider 1968 and 1978.) The series of letters coincides not with our modern /1-G but with a scale starting on C. Thus the Winchester A-B-C is modern C—D—H, and so on. This arrangement of letters was already described by Boethius, and Hucbald too mentions hydraulic organs whose lowest note would be our G, and whose scale corresponds to our modern major scale. Hucbald does not actually assign letters to this scale, but Holschneider and others have connected the Winchester letter-notation with the
6. Pitch-Notation
387
famous Winchester organ described by Wulfstan, who was also probably the copyist of the relevant portion of Cambridge 473. Coincidentally, this type of letter-notation is again used for a sequence in St Gall, Stiftsbibliothek 380 (p. 176, Carmen suo elitecto). Whether the notation was primarily to aid vocal polyphony, or instrumental performance, remains unclear.
The earliest evidence that some chant-books may have been provided with alphabetic notation corresponding to modern nomenclature comes from Italy. In the first chapter of the anonymous Dialogus de musica (once attributed to Odo of Cluny), written in Italy at the end of the tenth century, the author recommends that the appropriate pitch-letters deduced from the monochord be copied against any antiphon that has to be learned. The teacher will pick out the melody on the monochord, write the pitch-letters into the book, the pupil will then learn by playing the melody on the monochord, reading from the pitch-letters. This replaces the old method of learning by listening to the teacher singing, 'and after a few months' training, they are able to discard the string and sing by sight alone, without hesitation, music they have never heard' (trans. Strunk 1950, 105).
Most surviving books with letter notation are not Italian, however, but French. The most famous is the Dijon tonary, Montpellier, Faculte de Medecine H. 159, written in the first half of the eleventh century, probably at the monastery of Saint-Benigne at Dijon. Its notation has already been used to exemplify French notation at the beginning of this chapter. Its genesis is bound up with the activities of the great abbot William of Saint-Benigne, an Italian from Volpiano in Lombardy. After studying and taking the monk's habit in north Italy he entered the monastery of Cluny, and in 990 was made abbot of Saint-Benigne, with the commission of reforming its observances. In 1001 he was invited by Duke Richard II of Normandy to revive monastic life in the duchy. The results of this reform can be traced clearly in the chant-books of Dijon and Normandy (see below, IX.5), and the alphabetic notation that is found in the Montpellier manuscript can be seen in a number of Norman manuscripts (see Corbin 1955, and, for the fullest list of sources so far, Browne 1981). The alphabet runs from a to p, corresponding to modern convention as follows:
Ex. IV.6.1. Alphabetic notation a-p
abedefghi iklmnop
It is possible that William learnt the idea of using alphabetic notation in Italy. The a-p system is also mentioned by the theorist Odorannus of Sens (985-1046) in a letter of 1032 or 1033 (ed. Bautier et at. 213-14 and cited by Browne 1981, 12).
Practically all the sources using this notation contain relatively little music, a saint's office, a few antiphons or hymns, and so on, many added in blank space in a
388
TV. Notation
manuscript prepared for other purposes. The flyleaves of the Hereford noted breviary Hereford, Cathedral Chapter Library P. 9. vii, however, appear to be the remains of a complete antiphoner notated in this way (see Frere 1894-1932, Bibliotheca i, pi. 2). The famous Montpellier manuscript is by far the most comprehensive source in the group, containing the complete proper chants of mass, arranged in tonal order. It also surpasses the other sources in the sophistication of its alphabetic notation, containing special signs for liquescence, the oriscus, and the quilisma, and five other signs, which have gained some notoriety. They occur where one would expect to find the letters at semitone steps in the scale (see Table IV.6.1). Soon after the discovery of the manuscript Vincent (1854) claimed that they represented quarter-tones, a claim which received comprehensive support from Gmelch (1911). Froger, however, has showed the evidence to be insubstantiable (1978, 'Quarts'). An alternative explanation, that they are represent longer notes, carries rather little conviction. The signs remain mysterious.
Table IV.6.1. Special signs at the semitone step in Montpellier, Faculte de Medecine H. 159
at/3 V
at a (below bb) T at b 1 at e J
Although it does not seem to be related to the latter signs, another notational feature draws attention to the semitone steps in chant. In a number of eleventh- and twelfth-century French and English sources the punctum occurring on the lower note of the semitone step was given a special shape, usually rather like the uncinus of Laon notation. It is found in the staffless notation of some north French (mostly Norman) sources of the eleventh century, and in books with staff-notation (where it was, strictly speaking, redundant) in Norman and English books of the twelfth century (see Plate 10). It is also to be found in eleventh- to twelfth-century Aquitanian manuscripts (see Colette 1990), and passed thence into Spain and Portugal, where it survived until the end of the Middle Ages (Corbin 1952, Musique).
(ii) Guido ofArezzo
The pressure to adopt a pitch-specific notation for chant-books was so great, it seems, that no sooner was the ink dry on the first alphabetically notated books, so to speak, than another type of notation was invented, Guido of Arezzo's staff-notation. As is well known, this became the basis of the musical notation used in Western Europe for centuries to come. Guido tells how he came to invent the notation, and how he was subsequently called to Rome by John XIX (1024-33) to demonstrate it, in a letter to a
6. Pitch-Notation
389
former fellow monk, Michael, in the abbey of Pomposa (north of Ravenna) (GS ii. 43, trans. Strunk 1950, 121 ff.). The notation itself is described in what appears to have been conceived as the foreword to the antiphoner (GS ii. 34; DMA A.Ill; trans. Strunk 1950, 117 ff.). Although no manuscripts survive which can be linked directly with Guido, his intentions are fairlv clear. Stave-lines were scratched onto the parchment with a dry stylus, the F-line was coloured red and the r-line yellow. Guido's choice of F and c as the principle pitches to be identified is understandable, for these mark the upper note of a semitone step. As an alternative way of designating the pitch of the lines, or in addition to colouring the lines, the letters F and c could be set at the start of the stave as clefs. Guido does not specify the number of stave-lines: most often four were employed, for this number accommodates melodies with the range of up to a ninth, sufficient for most purposes, while extra notes can be dealt with by changing the clef or coloured line.
Although Guido does not state it outright, it is possible (and happened in practice) for a coloured line to be drawn through a space: for a chant with high tessitura, if the line for middle C is coloured yellow, then the red /above it will fall in a space. Many early sources with staff-notation use other colours—green is common—and it eventually became general practice to draw four lines in black (sometimes red) ink and use clefs alone to designate pitch. Practice with clefs also varied. Some sources diligently place a letter against every line. F and c are by far the most common, and sometimes high g is found. Much rarer is a gamma-clef for the lowest G. English manuscripts use D and bb clefs more frequently than their continental neighbours, and even fjtj is used as a clef.
Smits van Waesberghe (1951, 'Notation'), following the early history of Guidonian notation in the strict sense through a range of both Italian and foreign sources, could list almost none of the eleventh century. One of the earliest is the Old Roman gradual of 1071, Bodmer C. 74 (facsimile ed. Liitolf 1987). Staff-notation even in a general sense became the norm only in the twelfth century, at least in France, England, and Italy.
In at least one curious instance, the so-called 'Wolffheim Antiphonal' (now University of California at Berkeley 748), the staff was used for a neumatic notation which was not placed accurately upon it; in other words, the staves might as well never have been drawn in (Wagner 1926; Emerson 1958-63).
(iii) Staff-Notation in Different Lands
The shape of the signs deployed on the stave continued to vary from area to area, for the traditional signs could simply be set out in the new way on the page. Only in German lands does the adoption of staff-notation seem to have entailed a radical change. Many German scribes remained faithful to the older staffless notation centuries after it had been adopted in most other centres. The fourteenth-century graduals of Passau, St Florian, and Seckau, for example, are still written without staves (see the convenient list in Le Graduel romain II, 158-71). German notation
390
IV. Notation
was usually strongly sloping in character, and relatively few manuscripts show a vertical ductus. But for staff-notation the vertical alignment became almost universal.
A striking example of the change in appearance of German notations can be seen within a single source, the antiphoner from Petershausen near Konstanz, Karlsruhe, Badische Landesbibliothek, LX Aug., probably dependent on the monastic tradition of Reichenau. The manuscript was from the beginning notated in staff-notation, originally fine little sloping signs of the twelfth century. At the end of the manuscript the Office of the Dead was notated in a different hand, still fine but with practically vertical ductus. Some of this survives, but in the fourteenth century much was erased and replaced by upright 'Gothic' forms. Together with several additions, these scripts provide a fascinating cross-section of hands (studied by Hain 1925).
There appears too to have been a distinct rapprochement between the newer German notation and the later stages of Laon notation, so that in at least some areas it is possible to speak of hybrid forms. The signs shown in Table IV.6.2 can all be found in varying mixtures, a 'Laon' flexa side by side with a 'German' virga, for example.
The 'German' virga, looking somewhat like a horseshoe nail, has given its name to this phase of German notation as a whole: 'Hufnagelschrift'. In later centuries it was commonly written with a broad-nibbed pen (as were other late medieval notations), and in this form is also known as 'Gothic' notation.
Table IV.6.2. Late forms derived from Laon and German notation
single notes
pes
clivis
porrectus
scandicus
climacus
Late Laon    Late German    Bohemian Hungarian
A 1	t <	
J		v*
n	a n	
	ru	
		
		
Manuscripts showing predominantly or even completely Laon forms were written in many German centres: Leipzig (PalMus 3, PI. 174) and Hildesheim (PI. 175), for example. For reasons which are not yet clear, Laon notation was used in a group of sources usually assigned to Klosterneuburg (for example, Graz, Universtatsbibliothek 807: PalMus 19, which nevertheless has German strophici). Further east and north other forms related to Laon and German notations were used. Typical of Bohemian sources is a zigzag form of the pes and scandicus, while Hungarian sources often have a sort of double punctum at the head of a climacus (presumably descended from a right-facing virga). Signs similar to some of these can actually be seen in south and
6. Pitch-Notation
391
central Italian sources, and the possibility of influence from that quarter should perhaps be considered. (See the remarks on East European notation in Szendrei 1985.)
Much work remains to be done on the regional varieties of Eastern notations. In particular, the situation in Germany, Austria, and Switzerland remains to be clarified. On the other hand, Hungarian sources have been magisterially surveyed and classified by Szendrei (1983 and 1988, with over 130 plates) and the main lines of investigation for the East as a whole sketched out (Szendrei 1986). Centres of influence need to be identified, for example among the religious orders, which might explain the adoption of different signs in different churches and the reasons for such phenomena as the 'invasion' of 'German' territory by 'Laon' forms.
In the eleventh century in south France and south Italy many manuscripts were written with exactly heighted notation yet without adopting the Guidonian staff. Aquitanian notation, consisting very largely of separate points, was comparatively easy to set out in such a way that the intervals between notes were accurately indicated. It was common for scribes to rule pages as for a text manuscript, then use each alternate line as a guide for the musical notation. Clefs were not used, so that it was necessary to have an idea of the start of the piece in order to get going. Beneventan notation of the eleventh century followed a similar course; but it developed a bold, angular style, using a broad-nibbed pen, which gives it a quite distinctive surface appearance, however similar to other Italian styles it may be at root (see PI. 12, Beneventan, and PI. 13, Roman).
There was in most cases no great difficulty in adapting traditional 'unheighted' signs to the stave. Even before staff-notation was introduced in north France, for example, the virga and pes were commonly written with a head at the end of the upstroke, and the flexa often had a little foot at the end of the down-stroke (see, for example, PI. 9). Gradually these heads and feet became larger and more prominent, at the expense of the strokes in between, reduced to mere hair-lines. In Parisian manuscripts of the thirteenth century, nearly all the signs are composed of squares joined by these thin lines—forms known as ligatures in the literature on Parisian polyphony of the same period, which naturally took the local notational shapes as its starting-point. The exceptions are the porrectus, with a descending diagonal as its middle element, and the climacus, with rhomboid puncta. Square or quadratic notation, as it is commonly called, naturally existed in many regional varieties (see Plates 14-15, and Sunol 1935, ch. 16). That it became so widespread is due to the cultural pre-eminence enjoyed by Paris in the thirteenth century. Books were produced there for churches in a wide area around Paris (see Branner 1977). The master exemplars of the Dominican liturgical books were produced in Paris and set a pattern for others far and wide, even in Eastern Europe.
Everywhere in Europe in the later Middle Ages there was a move towards larger and thicker notational forms, a general 'Gothicization', it has been remarked. Another general trend was the resolution of all shapes into vertical, horizontal, or diagonal,
392
IV. Notation
curves being eschewed. By the end of the period, nearly all centres were using either a form of square notation or one of the varieties of Eastern notation. One of the last centres to retain its local forms was Milan, whose notation had undergone its own evolution from dainty early medieval signs to impressive squares and lozenges (PI. 16). Here as elsewhere, the adoption of a large format was necessary when it became the practice for choirs to sing from the written page instead of from memory. (The growth of this practice has yet to be comprehensively documented.)
Notations have commonly been divided into adiastematic (or non-diastematic) and diastematic (from the Greek diastema — interval according to whether or not they indicate clearly the intervals between notes. Similar terms are 'unheighted' and 'heighted' (of signs). 'Intervallic' notation is another. Where some intervals remain doubtful the diastematy is said to be 'incomplete', or even (with pejorative overtones best avoided) 'imperfect'. In theory nothing of the flexibility of the older notational phase need have been lost when the pitch-specific mode was adopted. This is clearly borne out by the staff-notation of relatively late German sources, where such features as the quilisma and strophici can be seen (PalMus 3, 132-7). The gradual adoption of staff-notation occurred, however, at a time when many niceties of notation (and presumably of performance) were declining. The salicus is rare outside the earliest manuscripts, the strophici are hardly distinguished notationally outside Eastern sources, the quilisma hardly survived the eleventh century in Western sources.
IV.7. THEORISTS'NOTATIONS
Crocker 1979.
A brief survey is given here of notational systems found only in writings on music theory of the early Middle Ages. They are of as much relevance to chant theory as to the notation of plainchant for service-books. Theorists' notations and notations for practical purposes were naturally related; but theorists felt the need to illustrate theoretical concepts with unambiguous pitch symbols earlier than singing-masters felt the need to give their pupils pitch-specific chant-books to study.
The earliest systems, found in the ninth century, have already been mentioned. Hucbald's notation, which seems to have had no other users, uses the signs shown in Ex. IV.7.1. They were borrowed in the first instance from Boethius (iv. 3-4), who was reproducing the notation of the Greek writer Alypius (3rd-4th c. ad; see Warren Anderson, 'Alypius', NG; Jan 1895, 369; Isobel Henderson 1957, 358; Potiron 1957; Babb 1978, 9, 38). Hucbald applied these signs to the notes of the Greek Greater Perfect System. For the most part, he, and many writers subsequently, referred to the notes of the Greek system by their Greek names, also given in Ex. IV.7.1. In prose discussions this remained the preferred method for well over a century. The real advantage of pitch-symbols was in diagrams and musical examples, where there was often little room for the cumbersome Greek names. It is important to realize that
7. Theorists' Notations Ex. IV.7.1. Ilucbald's letter-notation
393
hb
m
ht
P
*   * \>0 m
O'YHIIE C I MP C  FB T h    U E 0
tetrachords:     hb hyperboleon Greek pitch names:     aa   nete hyperboleon
d   diezeugmenon g     paranete hyperboleon
m  meson f     trite hyperboleon
ht hypaton e     nete diezeugmenon
s   synemmenon d     paranete diezeugmenon
or   nete synemmenon c     trite diezeugmenon
or   paranete synemmenon b paramese
bfc> trite synemmenon
a mese
G    lichanos meson F     parhypate meson E    hypate meson D     lichanos hypaton C     parhypate hypaton B     hypate hypaton A proslambanomenos
Hucbald did not use the A-G nomenclature. In fact, the modern equivalent pitches shown in Ex. IV.7.1 could have started with any note, as long as the sequence of tones and semitones was preserved.
The same is true for the dasian notation employed in the 'Enchiriadis' group of treatises: Musica enchiriadis, Scolica enchiriadis, Commemoratio brevis de tonis et psalmis modulandis, and a few other brief texts (all edited by Hans Schmid 1981). The author of Musica enchiriadis (usually dated to the late ninth century) first describes a series of tetrachords with reference to their intervallic structure, then designates the individual pitches with the dasia signs. If we take low G as the modern equivalent of the lowest note, the scale shown in Ex. IV.7.2 emerges. Nearly all the signs are derived from the Greek aspirant sign, the daseia (modern 'h' sound), which is given extra curls and strokes (like s and c), and turned upside-down and back to front.
Ex. IV.7.2. Dasia signs in the Enchiriadis group of treatises
g__ J_ s_ e_ r_ tetrachords:
i-1   i-1    i-1   i-1 i—i
v   - —
, . . 'EE
, g g^ves
m   *   m - f finales
s supenores e excellentes
r residui
394 IV. Notation
The presence of low Bb, high/#, and cfr, and the absence of middle bb, remain something of a puzzle. Since few of the musical examples touch on the extremes of the range, the outer chromatic notes are not usually a problem in the treatises themselves. Not every note here has a corresponding octave, however, so that when the author discusses octave doublings he simply omits the dasian signs if necessary (Schmid, 27, 32, etc.). Phillips (1984) has argued that the notes were intended to create a more flexible system than the Greek diatonic series for the recording of chants with chromatic notes.
The various demonstrations and diagrams in Boethius' treatise employ a number of alphabetic series. For example, at one point Boethius gives a diagram including all the notes of the Greater and Lesser Perfect System, not only for the diatonic genus but including the chromatic and enharmonic genera as well. To denote all these pitches, a series running from A to Z, AA to LL was required. If the notes of the diatonic scale are extracted from the complete series, the following scale is represented:
Ex. IV.7.3. Boethius' pitch-letters
A     B    C    [     H    I   M    0     X    Y   CC   00   rr   KK LL
This series is actually used in the late ninth century treatise Alia musica.
The A-G alphabetic series with which we are familiar can also be found in Boethius (iv. 14), actually extended beyond G to make a second octave H-0 (omitting J). On another occasion Boethius posits a complete two-octave series A-P (iv. 17), but the argument he propounds does not require the specification of individual pitches and intervals. The earliest medieval adaptation of this appears to be that in Part n of Scolica enchiriadis. (The F-d' series in Musica enchiriadis, Schmid, 30; GS i. 162, is peculiar to Paris, Bibliotheque Nationale, lat. 7202, of the tenth to eleventh century.) Here there is an unexpected and fleeting reference to a two-octave range A-H—P (Schmid, 90-1; GS i. 184-5). The notes in between are not specified here, but the example for which they are used demands that A, H, and P all correspond to modern E, e, and e'. Then in Part m of the same treatise an A-P scale is set out complete (Schmid, 147; GS i. 209), but once again it does not correspond to the modern A-G, a-g, a', but to modern C-b, c-b', c'; it is in fact the series known from the Winchester notation for some sequences, described above, and connected in some way with organs and other instruments (see IV.6).
The series A-P applied to what we would call A-G, a-g, a' finally appears in an anonymous treatise (GS i. 338-342, Anonymous II), apparently written in south Germany about the end of the millenium. Whether there is any connection with the practical use of this series by William of Dijon (see again IV.6) is unknown. Another variant of the series is to be found in Gerbert's Anonymous I (GS i. 330-8).
8. Printed Chant-Books
395
The repeating series A-G, a-g, and so on is first found in the anonymous Dialogus de musica, written in north Italy in the late tenth century. It achieved rapid acceptance, being used also by Guido of Arezzo. The lowest G was given the Greek letter gamma (P). For the third octave double small letters were used, up to a in the Dialogus, up to e in Guido's writings (Ex. IV.7.4).
Ex. IV.7.4. Pitch-letters in the Dialogus de musica and Guido of Arezzo
_  • ~ — —
i
P    ABCDEFGab cdefgl^BcBl)
Although Guido's writings had become well known in south Germany by his time, Hermannus Contractus (1013-54), monk of Reichenau, invented one further notational system, which did not, however, achieve any wider use. This was also a letter-notation, but specified intervals rather than pitches. A dot underneath the letter indicated downward motion. Ex. IV.7.5 lists the letters and transcribes the part of Hermannus' treatise where they are explained. (For a facsimile, see Stablein 1975, 223, where a variant of the system in another south German theory treatise is also reproduced.)
Byzantine notation is also an interval notation, but there is no connection between Hermannus and Byzantine music.
IV.8. PRINTED CHANT-BOOKS
Riemann 1896; Molitor 1901-2 (Italian books), 1904 (German); Tack 1960.
Chant first appeared in printed books in several forms. The creation of musical type (or any other means of printing music) was very time-consuming. Movable type could be prepared either with stave-lines for each character, or printed in a second run on to staves already prepared, both cumbersome procedures. Not surprisingly, many early printed liturgical books simply left space for staves and music to be written in later by hand. Others had printed staves but no notes. Such partially handwritten books can be found as late as the eighteenth century.
The earliest use of movable musical type appears to have been for a gradual printed probably in Constance in 1473. The dating depends upon the fact that the same type was used to print the text of a breviary from Constance, and one copy of this exists with rubrics entered in 1473. (See A. H. King, 1968, with two plates from the only complete exemplar, in the British Library; also Molitor 1904, pi. II, from fragments in Tübingen, Universitätsbibliothek.) The earliest datable book with printed music is theMissale of 1476 printed in Rome by Ulrich Han of Ingolstadt (Molitor 1904, pi. I;
396 IV. Notation
Ex. IV.7.5. Hermannus Contractus' interval notation (Munich, Bayerische Staatsbibl. elm 14965b, fo. IT)
e-    unison                "f   minor 3rd d    perfect 4th ^ minor 6th
S    semitone major 3rd          A   perfect 5th & major 6th
T tone
(,       t  tie- $. tie_e jS^^TTt^Sf^y
£        uoces u-ni-so-nas equat. S      se-mi - to - ni - i dis-tan-ti - am signat.
"p       to-ni dif-feren-ti-am tonat.       S cum T      se-mi-di-tonum sta-tu-it.
Jf- dupli-ca-ta di-tonum ti-tulat. d di-a-tesseron symphoniam de-notat. A A     AsTAei^Ts-c-cds       A A5 e-5     Afe j;ff
. .__a_*»_A_t_K_K_a—_1-•-i--at-*-*-4-rift-->-2-a*-*------
m—m 9 M
Aelta      di-a-pente consonanti-am discri-minat.      ^eltacumS      bi-na cum tritono
dTSxT 4>£ Ix"* " SSt^fTsTTÍ.TTs-txe .  . ;  .    .       f+ A fc   m       .   \   •  ~ •  .   m   -t   ■   ~ •   .  k  'T r-r
■»-*-
limmata docet.     /^el-tacum"f     quaternos cum limmate tonos maximum uide-li-cet. S^TTsrTSTTAde-r&rí: 5      A \>l 4- d_ f T
—•—:-'-!—'■-*-1—*—:- m   * * -
'*'.,■«'',.....»   « ' '
in can-ti-lenis nostris ptongorum interuallum determinat.      Si he no-te cum punctis tst tttsttt cAtJtÍt^T?T^
^     *    ' .
remissas        si-ne punctis in-tensas       uocum differen-ti-as discernunt pretax-a-tas.
A. H. King, pi. III). Whereas the 'Constance' gradual has German notation, the Rome missal uses square forms. Initials in both were entered by hand.
In Han's book, as in many others of the fifteenth century, relatively little music was required, for only the priest's chants are notated. For such short items wood blocks or even metal ones might have been used, but in liturgical books movable type was preferred. The earliest liturgical music-book printed from woodblock appears to be the Obsequiale printed in 1487 for the Augsburg diocese by Erhard Ratdolt (Molitor 1904, pi. VII).
The two books just mentioned are the oldest containing printed music of any sort. By the end of the century over 250 liturgical books with music had appeared. (Molitor gives short accounts of many printers in Italy and Germany.)
8. Printed Chant-Books
397
The notational forms were for the most part the same as in contemporary manuscript books of the corresponding locality. While there is no trace of the quilisma in German prints, strophici can still be found. Both had disappeared from square notation before the fifteenth century. German printers cleverly used the strophicus for liquescent notes as well, simply adding a strophicus after the principal sign. Liquescent notes do not appear in printed square notation; they had in any case practically disappeared from most manuscript books using square notation. On the other hand, numerous mensural rhythmic signs were employed as the occasion demanded: this development served reformed chant, which utilized proportional rhythmic notes.
The basic distinction between music-type using square forms and that using Gothic or German forms has survived until the present century, but the tendency to follow the example of Rome, where square notation was the rule, has led to the general disappearance of Gothic notation. Both forms have been susceptible to wide local variety. For example, several early printers of square notation maintained the distinction between virga, punctum, and rhomb, and some created the diagonal shape necessary for the porrectus. But the virga is comparatively rare in the Medicean edition of the early seventeenth century and its imitators, and the diagonal is absent. The virga was then entirely eliminated in many books (principally French ones) from the later seventeenth century onwards. Signs containing more than one note were then made up simply by placing the required number of square and/or lozenge shapes next to one another. (See Tack 1960, 49-57 for several facsimiles.)
As part of the restoration of the repertory to something like a medieval state, Pothier and the firm of Desclée in Tournai created a new font which was capable of reproducing the basic shapes of fourteenth- to fifteenth-century square notation, as it was known from north French chant books. To these were added a new sign for the quilisma. The first publication with this new music-type was Pothier's book Lxjs melodies grégoriennes of 1880, the first chant-book Pothier's Liber gradualis of 1883; the earliest Solesmes and revised Vatican books then used the same font. It was expanded for the Antiphonale monasticum of 1934 by the addition of signs for the oriscus, apostropha (distropha, tristropha), and liquescent punctum (that is, for liquescence by augmentation of the punctum, rather than diminution of the pes or clivis). The new Antiphonale Romanům has introduced a new batch of signs, designed to reflect even more closely the detail of early notations (principally that of Hartker's antiphoner, St Gall, Stiftsbibliothek 390-391, PalMus II/l; see the table on p. xii of the Liber hymnarius of 1983).
Ex. IV.8.1 gives the same passage from a fourteenth-century manuscript and three modern editions: the Ratisbon Graduale Romanům of 1898, the already partially restored Graduel romain printed in 1872 in Marseilles under the aegis of the Digne Commission, and the restored Vatican Graduale Romanům of 1908. The differences between the notes are of course striking, but also notable is the greater variety of signs in the Desclée font.
398
a'. Notation
Ex. IV.8.1. Communion Pascha nostmm from Graduate (Rome, 1898), Graduel romain (Marseilles, 1872), Graduate (Rome 1908), and Rouen, Bibl. Mun. 250 (Jumieges, 14th c.)
Graduate 1898						
					Mi_ii	
					■  g g^f	
W     1 "		WW	*R   "     1"		■   ■   H *	
ir 1 It Graduel 1872	1 'VP \ \M			II ■		
k               - -						
■     ■ ■	- m  . wm	■ ■		■ a ■	■ ■   ■  f» ■	
			■■■ ■ r"	*♦ ■	■ ■   ■     ▼ ■	
^Graduate 1911			i	▼ ■ 1		
		*		1	■	
■					■ ■ 3*»	
				■   rV W		
Pas - cha /»Rouen 250	-— nos-trum	1 im - mo- _|	-tus         est Christus,		al-le-lu -	ia
--*t-		-«—	m.   hJI L		~z—-—Ai"	
	—n-i/t-	n  ■—L	~tV*¥—*-			-i—
IV.9. MODERN TRANSCRIPTION						
(i) The Liturgical Context
(ii) Transcription
(iii) Transcription from Staffless Notations
From the foregoing chapters it will have become clear that chant notation was from the start quite different in character from modern notation. The absence of pitch in the earliest sources presents special problems of transcription, for to arrive at any pitch-specific interpretation one must have recourse to another manuscript (or manuscripts), often considerably later in date than one's main source. Yet one must preserve as much as possible of the original, possibly to the extent of making a literal copy of its notational signs.
Even the later forms of chant notation contain little of the information about performance practice that we normally expect from a musical score, such as indications of tempo and dynamic. They share this condition with practically all medieval music and a good deal of later music as well. Music sung all day and every day could hardly include extravagant performance practices, for according to its very nature chant was sung in a traditional way, which needed no specification. Customaries and ordinals therefore contain plenty of information about who shall sing what chants, on what occasion, and where in the church. But niceties of performing style were not committed to writing. If such aspects of performance as rhythm, and even pitch, are left out of account in the notation, it is only to be expected that other features should also be lacking.
9. Modern Transcription 399
It has to be admitted, therefore, that we know very little about the performing styles of medieval chant. In fact it is only since the advent of sound-recording techniques—first used for plainchant at the 1904 Gregorian Congress in Rome (Discant Recordings DIS 1-2)—that we can trace the history of chant performance in some of its most essential aspects: rhythm, dynamic, voice production, and so on. This is not the case, of course, with the new types of chant which arose in the seventeenth century and later. On French chant figure, for example, we have both practical manuals and the eyewitness accounts of outsiders (see below, X.6). But there is no unbroken line of tradition back to the Middle Ages.
The method of chant performance is clearly of great importance in the church today. Plainchant has been singled out by several statements from the Holy See in terms such as 'the chief and special sacred chant of the Roman Church' (Sacred Music and Liturgy, §16), and church musicians have felt a special duty to sing the 'right' notes in the 'right' way. The musicologist has the luxury of not having to decide on what is 'right', indeed cannot even see the problem in such terms. This book stresses rather the variety which is to be found in medieval (and later) chant-books. If one seeks to recreate a particular melodic or rhythmic tradition (say that of St Gall) for modern use, that is to select but one tradition out of a multitude.
This book is not a manual of modern plainchant performance, for which one must look to such books as those by Agustoni and Goschl (1987), or at least take note of the directions prefacing the Solesmes/Vatican editions. Even to sketch a historical survey of what is known about chant performance in past centuries would need a book of its own. (For some recent contributions, see Berry 1965-6; Dyer 1978; Brunner 1982.) Instead, a brief consideration is given to three points of particular relevance to performance practice and transcription.
(i) The Liturgical Context
Any chant recorded in any source is an element in the worship of a particular church. Thus the mtroit Puer natus est nobis in the manuscript Paris, Bibliotheque Nationale, lat. 904 has to be understood as part of the liturgy for High Mass on Christmas Day in Rouen cathedral in the thirteenth century. We can, if we like, simply regard the version in that source as typical for the Middle Ages as a whole, but it cannot be assumed a priori that all other medieval manuscripts transmit the melody like Paris 904. We might also assume that the introit was performed in more or less the same way in most churches throughout the year. But experience shows that one should expect special ordinances to apply at the most important mass of Christmas Day, and these may not be the same at Rouen as at other churches. Since Paris 904 has a quantity of rubrics, we are better informed for thirteenth-century Rouen than we are for, say, tenth-century St Gall. For further information we should consult any ordinals that may have survived from Rouen (they may not be thirteenth-century ones, however; in fact there are at least nine, ranging in date from the late thirteenth to the fifteenth centuries: see Le Graduel romain II, 195). These considerations are of
40C
IV. Notation
course especially relevant if one wishes to reconstruct the whole liturgy of a medieval chcrch. Even if only the chant is important for the investigation in hand, it should still be borne in mind that a medieval source records only what was understood to be 'right' at a particular place and time.
(ii) Transcription
Any modern chant transcription should represent the original as faithfully as is compatible with the very notion of transcription. As already stated, performance from memory was for centuries the rule rather than the exception. But where a transcription is needed, either for performance or for study, how should it best be nude?
The answer has seemed to many to lie in the Solesmes/Vatican books of the present century. Especially in the latest volumes of the Antiphonale Romanům there is a variety of signs which can cope with almost all features of early medieval originals: that is, most signs of the medieval source have a counterpart, so that with the modern edition in front of one, one could imagine how the original looked. The biggest advantage of transcription into a sort of square notation is that the grouping of notes in medieval sources can be preserved. This is, however, equally possible with transcription in modern round notes on a five-line stave. The Solesmes/Vatican square notation has by now the aura of tradition and ecclesiastical approbation, and is iniissolubly linked with chant in an almost spiritual way. Despite its aptness, it should nevertheless not be forgotten that it is a late nineteenth-century revival, with modifications, of a notation from the high Middle Ages. Its resuscitation is ccmparable to the revival of Gothic architectural styles (with modifications) as the model for nineteenth-century churches.
Since transcriptions have to be made from a wide variety of medieval notations, a system which is flexible enough to reflect that variety is desirable. In this book I have therefore used round note-heads, with special signs for liquescence, the oriscus, the quilisma, and the strophici. Especially in view of the research of Cardine and his students it seems vital to indicate note-groupings as accurately as possible, and make clear exactly which notes are joined to each other in the same sign. Slurs are used here fcr that purpose. This book contains plenty of examples with both medieval signs and transcription in parallel, which demonstrate what I believe to be a reasonable method. Note-grouping can be reflected simply by placing notes close together, rather than birring them like modern quavers or placing them all under a general phrase mark. The practice of covering all notes sung to a single syllable with a single phrase mark is regrettable, especially in the case of long melismas.
(iii) Transcription from Staffless Notations
For any transcription from a staffless notation on to the modern stave, a source has to be selected to suggest possible pitches. The later source (or sources) should be
9. Modem Transcription
401
acknowledged, even when it is a modesn Solesmes/Vatican book. Ideally one would find a source in staff-notation whose melodic tradition is as close as possible to the earlier manuscript. The monks of Solesmes working on Ix> Graduel romain have had to wrestle with this problem. Whereas the earlier Graduale and Antiphonale cite no manuscript sources to justify the given pitches, the new project has first identified the most important early manuscripts with staffless notation, and then matched each one with a source with staff-notation (see Froger 1978, Edition; Berry 1979).
The copious detail of some of the earliest sources often demands special solutions. The best, but also the most cumbersome, is to copy the signs of the earlier manuscript over the stave for the transcription. As early as 1876 Michael Hermesdorff issued a Graduale with neumes printed over the stave. The invaluable Graduale triplex is an example of this method, with two sets of handwritten neumes over the typeset notation. Otherwise some means has to be devised of representing in the transcription such features as significative letters, episemas, and the distinction between punctum and tractulus (punctum and uncinus). It seems unwise to ignore any detail of the document being transcribed, especially when one is unsure of its precise significance.
PLATES
FACSIMILES OF CHANT BOOKS
INTRODUCTION
The illustrations on the next pages are intended to display both different types of liturgical books containing music and different types of notation. In eighteen photographs it is obviously impossible to give more than an arbitrary selection, but these facsimiles provide at least partial support for Ch. Ill, on liturgical books and plainchant sources, and Ch. IV, on notation.
I have used pages only from sources in British libraries, these being on the whole less well known than those of some other European countries (but see Briggs 1890 and Nicholson 1913: five sources used by Briggs are found here).
The manuscripts in order of presentation:
Plate   1. London, British Library, Harley 1117
Plate  2. Oxford, Bodleian Library, Lat. liturg. d. 3
Plate  3. London, British Library, Egerton 857
Plate  4. London, British Library, Harley 110
Plate  5. London, British Library, Add. 30850
Plate  6. London, British Library, Add. 19768
Plate  7. Oxford, Bodleian Library, Canonici liturg. 350
Plate  8. London, British Library, Arundel 156
Plate  9. London, British Library, Royal 8, C. xiii
Plate 10. Cambridge, University Library, L. 2. 10
Plate 11. London, British Library, Add. 10335
Plate 12. London, British Library, Egerton 3511
Plate 13. London, British Library, Add. 29988
Plate 14. London, British Library, Add. 17302
Plate 15. Oxford, University College 148
Plate 16. Oxford, Bodleian Library, Lat. liturg. a. 4
Plate 17. Edinburgh, University Library 33
Plate 18. London, British Library, printed book IB. 8668
ft-lujviuic dnf.VDrcert «lnf .|tjAA..t«i,-7jj|ica.iiir lU.urn-V'^^vwri.irc illuxcfta-rncrmt-
^"\^> ' betxaim pr*  JulCm circbbe-r-b   mm cjuj ^r.m..i ditto
: j    J -  ^-    „•<"•'! -   , i/    :        Ay j   •       r     ■>  .      .  \ : .j : : . tn   tntintlifui Q^nf -mulnPU '     cmm morbortim ii-tltcu dtnef
y  \ i ^  j   • '   a   .< J ■<     ~' ■ •   *\,<ri    J   -':    y -t»nur prapbetwte- ftu titti yol le-niplunmo T~"»n
fli   titn tp   \i    tif o    - - •   • '   luriHn cc Pr* ' * -
fczurr er prci»ut»r»utiir.VAcl'min*naui cunCrtf opefibnf ecuvrf*bif dl
y   - - /  .  .    .    a <    r    .'i. I J  I        -1    , ' J  •   I | —-•
uma. i*mvtn compcfu    t-r. p7t>pl>mai-.INM .\TVT!ME$LMJO
C~% -   J   J -i ' - A J    1 ' ' J '     r * ~'
rtft» for**f t»*c axbl«*i Ac uefitf Anacbcrrnr-i ni«n<li ccmpfrr na*c m
fimA quo caeli c^per^fumnta A v^m cle-pupe prompfrr .t'jiiam pofr
.    /      • r    y   _ 4 .        .     .     /        *i        y      f   y.   ■  '   .      . -
jfimrjitm ue-nrenf cam "oc UTrumi|ue- citmum fue <r~*rt"> conniltr
^ ' -1 - J   ■ - J '   y, ' " -      -   y   * » ' y  :- ' - ',y   1 r
ctrrl?b<^nTr,p.A-Itiejpifcopa-rn fuc> ta.ni e-vacm btenmo u-rfim w-iciTtT
i= '("( '  V"1 -u 0   '   " "\    -      y    "    1    iy "    -/ •   ' -cleo d i Life to fe-f^aaTr AtTnx>.A.r~>mc «»Su"txir ^Lt*n»rf5«<n corf °T*t/
mcrriTO loeculi lcim| uerx^Jeriio'tti canfcr'titT- outle-xp»-^AA.t>x
y    -     -5 >     J>.-   . -    - -       yn    " r       y * -      I ' *'
p^TCT" fuofUmvcitf' autfafMe-tieftr**™* «mpna.tr wtinamn .Ittier:
„ i   y     .     y    /o   r   f / .   . J   ' •   \ ~j j   -\     .    /    j *
^L^v*t<t|tio7~ ^crcfccnftTuiicf-irtufuc'CiUtr frn<tj\iefiyf*i f.^^f-tirt^ertfyjxctf '.i-j    :n>,   •.•   i- V ^ t"V - -y J
. .    ■ ■/-5   w .   y   /:    J   .    . . —py    ^ .    v r
liierrt f!u«tiiT,cf"fi.iw«- d^pcfflc /IipcrTi^ffcanarr- fc4cC ^ V_/rn-ta,ne preful-
cu-cl»be-|-*,TC f«i -"cpf JTAir tttwert* cat mon Lucrum per enne- aumj>oft thctl.
-x c / • . y • , • -o ■ r \* ■ % J 1 r'-'*r f v , J   -J J :V
«m atuil uctv\"«a^nif"cttumirincin* wrii^uidofflu^wraLtbc hoc fx?^.muf pia pr«ce r« pro'iobif jn-ccrcetiv^-
London, British Library, Harley 1117, fo. 44r. Lives of St Cuthbert, chants for saints' offices, from Canterbury, Christ Church, second half of 10th c. Breton neumes. The plate shows chants for the office of St Cuthbert.
PLATE 1
407
The manuscript contains Bede's prose and verse lives of St Cuthbert, and chants for the Night Office and Lauds of St Cuthbert (fos. 43r-44r), St Benedict (fos. 63r-65r), and St Guthlac (fos. 65r-66v). (The start of the Cuthbert office may be seen in PalMus 2, pi. 81; fo. 63r is in Briggs 1890, no. 6 and part of it in Sunol 1935, pi. 82.) It therefore contained material for both reading and singing, and served as a supplement to the other office-books of the church for which it was written.
At the start of the page are the final responsories of the Night Office. Two are given simply as text cues (Iuravit dominus, Magnificavit ilium) because they would have been available complete in the section for the common of saints in an antiphoner. The final responsory begins on line 2, and is copied complete: R. 0 beatum presulem Cuthberhtum V. Admirandus cunctis operibus.
After the rubric in matutinis laudibus there follow five antiphons for the psalms to be sung at Lauds: Cnsti fortis hie athleta, Qui de rape prompsit aquam, In episcopatu suo, Hinc tanguntur artis sacri, and Mox pater suos affatur. The rubric in evangelia announces the antiphon for the Benedictus canticle, Languor ad crescens in dies. The last antiphon, 0 magne presul Cuthberhte, is for the Magnificat at Vespers or for subsequent commemorations.
The texts of the chants are rhymed (end-stressed), as for example the first antiphon:
Cristi fortis hie athleta ac verus anachorita mundi tempsit haec infima quo caeli caperet summa.
These are classic Breton neumes, fully comparable with those from Brittany itself. The following may be noticed:
(a) both conjunct and disjunct pes and clivis: ^       ^ '
(b) the last note of the climacus often has a long, slender, vertical tail: I
(c) liquescent neumes: ( Q J
(d) the oriscus, either alone, or joined to a subsequent lower note: r -7
(e) the letter / ('levate') for the rise of a fifth in line 2 at 'qui': I
The same office chants appear in the Worcester compendium (PalMus 12, 292 ff.), facilitating the following parallel transcription of the start of the responsory.
T        T T f
O be-a - turn pre-        -sulem   Cuth-berh- -turn
.-A   j . .  A /  \~f . r   .'1   •   J  i   .«i   A, J .
i
~9 "   ~--~ ~"--
qui      gra-ti-a   di-ui-ni mu-neris    ui-   -gens mul-ti - pli- -ci-am
PLATE 2
409
Oxford, Bodleian Library, Lat. liturg. d. 3, fo. 3r. Fragment of a combined lectionary and gradual, from Aquitaine, 10th c. Aquitanian neumes. The plate shows parts of the masses for Monday and Tuesday in Holy Week.
This fragment was originally one page, with text and music set out in two columns. It was then cut down to a suitable size for making guard leaves for a later manuscript. It was folded down the middle, but part of the initial N can be seen on the left of the fold. Text and music are missing up above what is now preserved, so altogether only the lower right-hand quarter (approximately) of a bifolium is preserved.
How does one identify the material on this fragment? The rubrics communio and feria in on the right indicate the communion chant at the end of one mass, and the day on which the next mass is celebrated. The chants, the communion Erubescant and the introit Nos aulem gloriari, can then be located in a modern chant-book, or any other properly indexed edition. The manuscript contains texts as well, so that a missal or the Liber usualis is necessary for identifying everything here visible. It can then be ascertained that on the left is part of the Gospel for the Monday after Psalm Sunday. The last few lines of the lesson and most of the Offertory chant are missing with the top half of the page. Only the closing words of the offertory Eripe me V. Exaudi me are visible: '{intr)es in iudicio cum servo tuo domine', with the cue 'Doce me' for the partial repeat of the respond. After the communion Erubescant the mass for Tuesday of Holy Week begins: the introit Nos autem gloriari and the start of the first lesson, from Jeremiah.
None of the proper prayers is given (they would come after the offertory, communion, and introit) so it looks as if the book combined lessons and chants only. The fragment is listed in Gamber 1968, no. 1381 (under the incorrect shelf-mark Lat. liturg. a. 6), alongside two other manuscripts of this type, one also from Aquitaine and one from Saint-Omer.
In order to achieve a transcription, the Aquitanian neumes may be read off against those in Paris, Bibliotheque Nationale, lat. 903, reproduced in PalMus 13. Paris 776 was used here to guide the transcription of the second half of the communion.
in-du-an-tur pu-do-re   et re-ue - ren-ti - a
-0-
z
qui ma-li-gna locun-tur
aduer -sum me.
«.-dtp. UtíM
J' 1 y. v  fx ff      ľ ^    'íl      / ■ •    -*       * '
H Ciw« to. v A.^.w. ;\uc tu «
T.
é4o»r t m n u f jLc    uf tniŕ
er n bi ned cLmt r m.*
I'll
•   m. 1ť
tiítti    ŕr íitfttr   m nulľkiď »t<^U't-itnj   pniiftit   um líc'h.   Ar !.uniu*i um
'   -*' ,    «i».   «».«.'..ví . v*"'   ' 4
n« ftruro   m<enlp//b»  fU-» tmátť   tir pU  ccat   ábi «{««. non cti «».ifu
- ~? ! V4 ^   ' *        '    * *
tte   «nrtrwrabuf   i«« &»m\ ne W \ nAinA 4tit*«a zuxm AccAct^
ur činia     fuof i
.a
*, Vil! - Ctp ť - i WV« iff -VSC6PÍMVS JUtiCP* 0>ÍWá?f.U öiW ntidn m d*. íu}*m»ř tp-atantr-   -VÍU    1«   u.        , '
/♦ vi*. šL-í-iv'Ut- » v'        *     .j   < . • '
»-*  4   4'i ,«    ^* , /'/        J * ^ ' ' 4Í
iitAtu luattife   ít innitmť t^ru  l'mr cjiu 1pfl-ar   m «■      ) AT III
"í
mm : j'sS <
f ŕ*,
Ír -A1
C OÖ á/ «r .aťuui- „u. « p o*^V l í II - cíp • ^* W\a>1 i vL.
•ismtuuf [íiiVcptui- ríf Aiitnií m«r  au^rec  nuU >iintu-ir me i. i«      m ni i (fcúv
r' .
uertcuvľ tu. a
. _ ŕ.
«       4 I      > «4
t «nomine nu>
c ft- u*) m« ty cuiLtu i aunt u.'ť ť.t
a
/ 7
4 V 4>
d:
j rnttic  <l<-w <
Ü.l«.n'.f «i-.*c mát* cLuvíA " •/'*".*' tú      iíoító f^mte-
PLATE 3
411
London, British Library, Egcrton 857, fo. 52v. Gradual, from Noyon, 11th— 12th c. Laon neumes. The plate shows chants for the proper of mass.
There are no prayers or lessons on this page, for the book is a gradual. Chants are given from three masses of the post-Pentecost season. In the manuscript the masses are numbered, so that when a chant is to be repeated from some previous occasion, a reference to the earlier occurrence is given. The chants on this page are:
(end of mass for the seventh Sunday after Pentecost): gradual Venite fili V. Accedite (incipit only; reference to 'Capitulum lxviii', mass 68); Alleluia Te decet hymn us; offertory Sicut in holocaustum; communion Inclina aurem tuam.
'Dominica VIII'—eighth Sunday after Pentecost—mass 143: introit Suscepimus deus Ps. Magnus deus (incipit only; reference to mass 31); Gradual Esto michi V. Deus in te speravi (incipit only; reference to mass 66); Alleluia Attendite popule tneus; offertory Populum humilem (incipit only; reference to mass 70); communion Gustate et videte.
Ninth Sunday after Pentecost—mass 144 (last letters not visible on photograph): introit Ecce deus adiuvat me Ps. Deus in nomine tuo: gradual Domine dominus noster V. Quoniam elevata est; Alleluia Domine Deus salutis mee.
Facsimiles of other pages will be found in NG xvii. 622 (masses 174—80); Briggs 1890, no. 8 (masses 135-6); PalMus 3, pi. 158 (masses 13-15).
The neumes are not in all instances exactly of the Laon type, hardly surprising when the chief point of reference, manuscript Laon 239, is older by two centuries. The favourite single note is more like a short, sloping, German virga with a head to the left, and the true Laon uncinus is largely absent, except for low notes at the start of a group, where there is little space for a long stem. The Laon virga, like an elongated S, is not used, but the pes looks rather like this. (The pes is somewhat rare in Laon notation, uncinus + virga being preferred.) The flexa is likewise preferred to a pair of vertically aligned uncini; it resembles more the French than the Laon form. The circular quilisma and oriscus like a double o can both be seen in following transcription of the communion Gustate et videte. To arrive at a plausible version, one should if possible compare the neumes with a later manuscript in staff notation from Noyon. Luckily one such has survived: Abbeville, Bibliotheque municipale 7, a notated missal of the 13th—14th century, which therefore guided the choice of pitches in the transcription. (Compare the version in the Liber usualis or Graduate Romanum at 'GusTAte' and 'QUOniam'.)
r S'J'/r r r m . S y> /A   r y   .  A? Jl    , jir A fo   • ' '  '\m »;  . ,«■'■> * ^miSm , ^£^=
Gus-ta-     -te et ui-de-te  quo-ni   -   am   su-a-   -uis  est do-mi - nus
be - a - tus        uir  qui       spe-rat  in e- -o.
f p t ijoirr'""1'" . r.
inc onrrnťá, óf fíunea/,
*a, dominum umucr^a^cVaxitontcur » p,
W     4*^'"V > f a.   n^ŕ W-
tojnšaKulíinc   dominum^ "( jjtit j
, Kb Dííccmc cauíammcam d
------------------------- .lO ny,^
M fry / .      o.y if y p"1 ľl"J'
bbommc miquo a^dolof cnpc- tne- _/,-: tuam        <5ťumxaran mam
t •    * yt % *     •     * » • *
im de
lomirie auôniam rionácrclmcjuif
m jw"- y*"' uí-^'^"1
it^CTaLUre domino quiha-bitar inlion
J ' ., ,/V. / //•/' /   . !>.
bbl^fóranonán páuperum* . ipcrchrrjnum auiiuatc^uéauKa, rxm
'r/   PP.   JI »/'S /   A. ..f.    J J
lit <rcrrccf úpenie impitif íudicare popu immltraa, cVtacctif d t rtw<num patrpc",«
'pxti-parutri rxatidiuic de ,uí-
PLATE 4
413
London, British Library, Harley 110, fo. 56r. Fragment of a gradual, from Winchester, mid 11th c. English ncumes. The plate shows chants for the proper of mass.
As with Plate 2, the use of this leaf for binding a later book has resulted in the loss of valuable notation, in this case on the left-hand side. The surviving labels 'CO' (Commnnio) and 'RG' (Responsorium gradale) show that the fragment is from a gradual, and consultation of a modern edition brings one to the masses for Monday and Tuesday after Passion Sunday. The chants are:
(Monday of Passion Week): end of offertory Domine convertere V. 1. Domine ne in ira tua V. 2. Miserere mihi domine; communion Dominus virtutum Ps. Domini est terra.
[feria] hi statio ad sanctum cyriacum. capitulum: introit Expecta dominum viriliter age Ps. Dominus dluminatio; Gradual Discerne causam me am V. Emitte lucem taum; offertory Sperent in te omnes V. 1. Sedes super thronum V. 2. Cognoscetur dominus.
The rubric 'Station at St Cyriacus' does not, of course, mean that the mass was sung at a non-existent church of St Cyriacus in Winchester. The mass was celebrated at such a church in Rome, and the rubric was simply carried over, even though no longer relevant, when Roman service-books were copied in Francia and elsewhere.
The neumes are exactly like those in one of the famous Winchester tropers, Oxford, Bodleian Library, Bodley 775, perhaps even written by the same hand. If the book had survived complete it would have formed an invaluable complement to the troper, containing in full the proper chants which are either absent or incomplete in the other book. No complete gradual from the Old Minster, Winchester, has survived.
There are numerous features of special interest. The communion has a psalm verse, and both it and the introit are supplied in the margin with the cadence for the psalm tone and with a label indicating the mode: plagalis deuterus (mode 4) and authentus deuterus (a mistake for authentus tetrardus: mode 7), respectively. Such labels are rather uncommon, but are found also in sources from Corbie and Saint-Denis (see Huglo 1971, Tonaires, 90-102). The following transcription of the communion was made with the aid of the Worcester compendium Worcester, Cathedral Chapter Library F. 160 (compare the version in GR or LU).
J    ( L J J    J ^ n
I
Do  - minus
uir-tu-
1 ./• A
-turn   ip - se est
glo-
4-—vv-
-•—mM—
-ri   - e.
::: ;y j*. <f fv L   . ]! . J •  •     - ^ a" ř , - - - " is i ■•. /
.-u.
• i        ;*       i   *   J    pj    i i
ittcobi Ätíulctttŕ" ôrAima utomw ■ cat 'utmäiiô-vctíT uncev&va f Ur um. c
., kyi/A í >.   p     ' />,/^\*,;^.....' •   j • - o ^ a
ľ a S ' ľ-Jt ,;#• /•   w**>f-yV. .Ja f a, i
CP t - f - I . •'; ; ■- 1 ' * r> ■
■        N *    ' í' T*   -    „        í • ~"    f   'j.    -f t
iCiiUMp.vcruíiif a ••■au:- .toft'-'O.stti^fclu^v-ii
i »  ;     ,   ' <* t   /       •        1  /■'•''• í ■
• • • ' .   ,v   • r • •  *      ' '       . • Y
LUL ma •
,\Ť Jl). nf. nputMui • čistia. rauLif. uinívA (7
i. ' ' .   '   f- 1 •  .     , ••  • ,     . . ' «
V- "« /     ^      y   / ^     . . .   . r. •  /      '*  '  ' . '
*: i * -r  J"-. «.
S * I.        ,        ' .,•'!'•- -1       k  -   .  ' .
rí-,»í»<i«««f. j^^^-i^c c4*rw«tf-.-K - ..^.".rv . l .^»«wcniflAr-»»
^/....   . . .' i .*..•''• .-j. •4 .
PLATE 5
415
London, British Library, Add. 30850, fol. 106v. Antiphoner, from Santo Domingo de Silos, 11th c. Spanish neumes (northern type). The plate shows office chants for Easter Day, with the Quern queritis dialogue added at the foot.
The entire text of the manuscript has been edited by Hesbert (CAO 2, manuscript S), and published in facsimile (Antiphonale Silense).
The chants are Gregorian, not Mozarabic, but the old Spanish notation used for Mozarabic chants has been retained, in preference to the Aquitanian neumes found in some comparable Spanish sources (and in the additions to the first pages of this manuscript).
After the final responsory of the Night Office (Dum transisset sabbatum, line 3) there is a text incipit for the Te Deum (line 7). A cross in the margin directs the singer to three lines added at the foot of the page, the Quern queritis dialogue. The rest of the main text is taken up with Lauds antiphons.
Verse and Gloria for the responsory Dum transisset use a version of the standard tone for mode 4. The first half is easily transcribed, but for the rest one would have to locate an antiphoner of the same tradition with diastematic notation.
. /        K      v.....J ■ J . . « J <i [ n
*       »        00 0*0   9U,   00, 0      9P      « #*     0      0 0g 0*     0, 0,9m
Et ual-de ma-ne u - na sab-ba-torum ue-niunt ad mo-nu-men - turn
.   * h    ? f\     , J q J\
or-    -to iam       so-    -le     al-le-lu - ia.
A provisional transcription of the Easter dialogue might be attempted with the aid of contemporary Aquitanian manuscripts (Paul Evans 1970, Early, 155, MM MA 3. 246); but some text variants (such as 'loquutus est' instead of 'praedixerat') are rare, and after 'o celicole' the melody follows an individual line.
Á / * l fkun.crtictf fbctu-m očr&gni ^iydi^':
"S -taíltuľ cfumiA fem per ,
a : * ^ Fri?l4:cpf «TktfterfaruiTi.-- V
^/ť A., ríčtockufatŕemofeoncoWt -t*k '
PLATE 6
^17
London, British Library, Add. 19768, fol. 16v. Tropes and sequences, from the monastery of St Alban at Mainz, mid 10th c. German neumes. The plate shows sequences.
The page shows the end of the sequence for St Andrew, Dens in tua virtute, and the start of the sequence for Apostles (in natale plurimorum apostolorum), Glare sanctorum senatus apostolorum, whose text is attributed to Notker of St Gall. In addition to the neumes entered over each syllable of text, the melody for each phrase is written in the margin. The notation in the margin includes occasional significative letters, for example in line 9. The next line of melody, the first for Clare sanctorum. is underlaid with the vowels of the word 'Alleluia'.
The earliest German copies of sequences in staff-notation are nearly all of the l?th century or later, and one is therefore tempted to look to earlier sources for a guide to transcription. In the example (left-hand side) the start of the melody for Clare sanctorum is transcribed with pitches derived from Rome, Biblioteca Casanatense 1741, fo. 118r (facsimile in Vecchi 1955). The significative letters used in lines 13-15 are:       e ('equaliter'): lines 13 and 14 must start with G
t ('trahere'): more deliberate delivery
c ('cito'): quicker delivery
/ ('iosum'): interval greater than a second at descent
5 ('sursum'): interval greater than a second at ascent Plate 18, which shows the same sequence in the printed Mainz gradual of 1500, gives other pitches, which do not correspond as well with the neumes of London 19768 (right-hand side in the example here). In this case the use of an earlier source, as opposed to a later one from the same region, is justified. London 19768 is nevertheless a monastic source (from the Benedictine abbey of St Alban) and one's first recourse should be to a later manuscript from the same monastery or one in close relation to it.
line 10
11
I
J1 J»
r^
nr
(Plate 18)
■ *
*   * *
A(l)-(l)e-(l)u Si A
Cla-re sanctorum se-na-tus a - posto-lo-rum
•   *   ' .
m m
-a
princeps orbis ter-ra-rum
12
13
14
1
S  ft    /=    A A
rector-que regnorum
15
c c
Ec-cle-si -a-rum mores et uitam mode-ra-re Que per doctrinam tuam fi-de-les sunt u-bi-que
	1-           .»1   »H,--#-		_K_	-0-
\	> *			
Anthi-o-cus et Remus concedunt ti-bi
UrifbitniMhi rcppuür aaufhtit tcnď £ com* 0dtu1biitnt f linuli xtu mwicnb'yfMTifa\js fcmß irrtumaono nofrcfvuc. aimf fallout xxichizrnnff
I OH If CJntzuuiuf cfh iwh' jmer cjuem icrncbxr iVClJJl.lj|<I VldlOB-^A- *
mcffk bílit CT* arrtľ -trní pera V.ircr ^cundem fMmnnjiunt jrneraaut xncitcXrrutt?.
-~  * f  * ŕ-"*''*^ f      ŕ   ,    s r .-    ŕ.       f _/ s s s. _
y   fLutfc/l- tAc^ inter *^o«ínxii»n. m-i^mf |>1*CCDituf áj^nuni iwctfcnttirdicrrc-
wfclf upamUudann-eT cŕTe ^ofTmiuí- V Rtc enrm efc^nem yrrfAynf^reitíhif
/ _     J S J  f v   /* /Z'       /    f    / S--' y s*     J2 ,<t  J   S /
flUuf cLrtuf eil tuj ÍJif caiufuirperuim. fk|> humerttnt
•ŕľ n«r  Jtt ley fotí. jjLy-neléyfon    ýtwXctpC*^ ^ ^
J'«S^* ,sS'-S'--> ysS^ V
J^C   |>cmfferlc-   -yfon JCyvrtflelr      y f*>n- ♦X'ppmftrlŕ  y fan-
j 4/ s . r" j ŕ",    r s* /t J*   '   y yf s* j^sj      „   , s
Lw»<ií,m«j n* b en td* c um «.f t.e PpľárT* vxhx ^fn&r* -mM^riX
jWiaw.tu4m «lamiri& deaf 4«|n»f      fií<u.f |7Jtmf «^utt»UiTf •^•^•f .",V
/ /'^r^ ^/   . ^ / ť -  / - '    /    . J,j&ťj~.
*xi\t*\dr lamfeťe »oEn/! «|umjllif*|»eccetta mamtii /Wciyr 4ŕpn^xtk^ ' ;
G
í   -       í- í!
PLATE 7
419
Oxford, Bodleian Library, Canonici Liturg. 350, fo. 4r. Notated missal, Aquileia?, early 12th c. German neumes. The plate shows proper chants from the third Mass of Christmas Day, with introit trope, Kyrie, and Gloria.
The illuminator never entered the large initial in the middle of the page, but its shape is unmistakable. At the top of the page is the last line of the postcommunion prayer Huius nos domine quesumus sacramenti, then (line 2) another postcommunion, Saciasti domine familiam tuam. These prayers conclude the second Mass on Christmas Day. The rubric for the third Mass follows: 'Ad missam maiorem'. First comes the dialogue trope attributed to Tuotilo of St Gall, Hodie cantandus est nobis, for the introit [PJuer natus est nobis with Ps. Cantate domino canticum novum. Then the Kyrie (Melnicki 1955, no. 74) and Gloria (Bosse 1955 no. 21) are notated.
The fullness of the material is surprising. Notated missals are not common at this date, and the inclusion of tropes and ordinary-of-mass chants in place beside the proper chants in order as they were sung is also unusual. It is, however, sometimes encountered in Italy, a clue to the provenance of the book. The neumes are German, but these were also used in a few North Italian centres: Bobbio, Monza, Moggio. On the basis of the local saints honoured in the Sanctorale, the manuscript has been assigned both to Brescia and Vercelli. If the identification of Kyrie 74 is correct, however, another link emerges, for this rare melody is listed by Melnicki only for sources from Moggio and Aquileia. A transcription may be made with the help of these later sources.
-fl-f    /  S   .~J   /     /__y   J   u   ,S t J /•-.,
^0*5*  m    yjy   ^#0   m0'O    0 •    £*   0 fm  m • '
Ky - ri-e- -ley-son. (3 times)   Christe-le- -yson. (3 times)
/  a   _/ -    -     A.    r»    A J uS1"       .   .   / f: J- /-,     /i ^ A
? *>0 _j» 'in m m sf^o a}       f    &     t t r> **0 _ji u,a 0. jr0 ,7
Kyr- -ri-e -ley-      -son. Kyr- -ri-e- -ley- -son.
(2 times)
^nobis Vjtovňnbmsv ocet mas 1 y<*uí» *pe u «arnoncmmam (Tieawôt í>««s   tftott. nobis - 1 v. qcgtatr
\ ý nt — leifim JM 9 w v V^nTOtmmi-~po mino   qttomam botmsquo     «ww mfteuhtm miftťicoi
V/íuftcuolucrnobisabljoťho Uptbem qucmtrgtrt fanchim cet
turnus k^ma^mm 4n$tr \Z ucm quennf m^wäm. o /?_j> f .. ~» 1  - \ét f ^ t i  *■    t   ň .»/       fi,J    n i**
I/^dfcgfc. mw^rp učup ^r^ijřaar n* nunctaie quia ftmpR:
třtf^^i4w'^«5í07on ^WquernqnamSi fií> «ro eunjrf nun „a*KrMazute ?nts í* ^trjco qma fumrrr $c ä Větutt i máte ittte ttfnref 3 tlbmonutnmtum uminutf ^mm^m^äutn >ommi fcfrtmem «i&imuf ^öteatrem <nna uttwyrc ifrV Cha^Ž
/ / i //«./•/ - * /f /í fc *f. r yW
Lqu©njí' b*—.~ nulrccf <uftc;ötcnitf ^u^tu^^Mí!?!^ tt
aur laptí>cha poftnoiifí quart non feťttáfratír p t t«an ♦.  i  /ľ .      n<j? v    ^ / .
.limit. 4^«pulrxiwi t^fcattT autcefti^ixccm dboíettrno
Uď.
i *^   ríf ^y     f ^ p j t
PLATE 8
421
London, British Library, Arundel 156, fo. 35r. Gradual, sacramentary, and lectionary, Würzburg diocese, early 13th c. German neumes. The plate shows Mass on Easter Eve and the 'Ordo sepulchri' liturgical drama in the Night Office.
Although the book basically contains material for the celebration of mass, an 'Ordo sepulchri' ceremony has been entered (rubric end of line 6), which ends with the Te Deum, showing that it was performed at the end of the Night Office. At the beginning of the page are the concluding chants for the special short form of mass at midnight between Holy Saturday and Easter Sunday: the end of a litany, Kyrie (only the incipit; Melnicki 1955, no. 39), All. Confitemini domino quoniam bonus and tract Laudate domium omnes gentes (incipit only).
The Visitatio is not fully notated, for reasons that are unclear. The text of this ceremony has been edited by Lipphardt (1975-81, no. 371).
The German neumes are thicker than those in the previous plate, and in some cases already have the form known as Gothic or 'Hufnagelschrift' (see Plate 18). The pes is still round, but the scandicus (e.g. line 9, 'crucifiXUM') consists of two lozenges joined lightly to a horseshoe-nail, as in many examples of Gothic notation.
The example transcribes the start of the Alleluia, with pitches derived from Graz, Universitätsbibliothek 807 (PalMus 19). The apostrophe and quilisma are still to be found, but the oriscus has disappeared and looks like avirga (horseshoe-nail), as in the penultimate note of the example.
m
t r vk a ft  a ?i At*
' **     O     »»*    » *» '
A(ll)e-
-0)u-
V. Confi - temi - ni
do-
-mi-no
^ «mnr deaf-'A ^nuf J«- -F  tWjMcn*-uuof
111   ? 1 17 f -   jn1"l j;1 11 h' • A/VJ - 1
alo reduneres    hodie wxrrtt   homo xi^xif ci - tu
/fU - i    -   jflf-.fJi   11t Jt 1 I-- ^ - ni .J-^
f«Jcf »^kexsxi&m yx ^——- cytf.—~ itufc—rere— iw-
fflfr        j t /i j «*i m
Wt fine       &««j>rtti«pw - e«--——      ;> Am
V> ^    n    , .     • \ -      r .7 - 1 1
f   J   f -   f f •   ^     :i    ' 1 1 fi , - 1 <i_> •
rnunc Jtatf rf'wmw tTgitAnd cdomint <ifotwn^&t• -
* ej^wtdnzr «i^wmu*n*    J       !u — «rt>iiur --—--------"
-1 - 11   <        ^  i   V' 1 1     f..7' "    J."-* ♦
London, British Library, Royal 8. C. xiii, fo. 6\ Ordinary-of-mass chants and sequences, from Normandy, llth-12th c. French neumes. The plate shows a Gloria with trope verses.
PLATE 9
423
The notation in Plates 9, 10, and 15 can be seen in a chronological progression, with allowance for the differences of provenance. Many of the neumes on Plate 9 have obvious heads or tails (virga, pes, clivis), which then makes clear their position on the staff in Plate 10.
In lines 11 — 13 there are several examples of the pes stratus, or pes + oriscus. They occur here in the melismas with associated prosulas which occur towards the end of this set of Gloria trope verses. The page begins in the middle of a Gloria (no. 39 in the catalogue of Bosse 1955). On the previous side there have been copied the opening verses of the trope set Laus tua Deus. At the end of line 4 comes the trope verse Ut (or Et) hominem celo redirneres, as far as I know unique to this manuscript. Trope and Gloria verses then alternate until line 10. Here, after the trope verse Conditorgeneris, another trope verse begins, Sapientia deipatris. It is syllabically composed, and each of its phrases is repeated melismatically to the syllable 'E'. The first melisma actually bears the syllable 'Per-', the start of a word completed at the beginning of line 15: '[Pjermanebit in eternum'. The reason for all this is that the 'Permanebit' melisma derives from another context (see above, II.23.hi and vii) and was only subsequently given the 'Sapientia' text.
Although London 8. C. xiii has many close relatives among Norman and Norman-Sicilian tropers, the latter do not contain Sapientia. A version very close to London 8. C. xiii in staff-notation does not seem to exist. The transcription is therefore a compromise version based on both Aquitanian and Italian sources. (Compare the 'O rex gloriae' prosula, Ex. II.23.6.)
J     111-1111.     Jl.i   J 1
*-«-9   _^   *- —**   #        f*-M-i
Con-di-tor  ge-neris   hu-mani       redemptor i-dem-que
m
1 1 1 1 , 1 1 . i f/n. . 1 1_-1 y -3 a n i.. y
---—
♦*r-
P
Sa - pi - en - ti - a   De-i   patris ae-ter-ni   fi - Ii - i Per."*' In - ef - fa - bi-lis  si-ne   fi-ne si-ne prin-ci-pi-o      e. .
J   1  . .   j  n.   ._<r   v. 1..
Sal-ua-re ue-nis-ti   nos. e.
1 \ ■ a ■ n . t 1_i y ft J
'*   -   • »
f-0
i
~r
Nasci dig-na-tus de uir-gi-ne e.
fjji.ii,jqiiiti-ii. •• iftflJJ1
•1 ^ *? i    • 0     ^ *i * • & + • +          ' * & f* & _
Et nunc De-us et homo regnans     O do-mi-ne do-minator       Per-ma-nebit  in e-ternu
m.
ifcrnr iiolnf\nc xp: dmďmrpc. % \yc audi nos.
' 1 , 1-
||igiiifiaw diíe dsnrnifdfmif.ee bot m !| tniiplo nuimme cdiftcmiOTiTfpfatmfvu V duDaii^mfavpdtt.tirqtuoinřdinn. ins Mopnomfopattf tpfc fetnp tuard ficttdtra te toudcnC p. Dqnde fenbifpm infix fíip patimi in utm atmíwta tiht.aíKccdamř qittinn.abaitqii (••í* 'iarfa!aifcf'indtrn'iiiiijiiq[uíu ocndninľič-. die '-*ít í .ittc mmph'fc"^pz™::;:-;:
-yT^j-------■^-1-----U. -.j-^—-t——--ffc----- -
^ Jputiďánictitmn aínidnmioporeír po
Tiftt ^tcrtílttd detnque qnod poliami ďr ti^po ďotntno. funí.'
ífď bna. qniHd. ifcfta. f. fltína. r«n. cca   rcra. toet. cappa. tapíj. im ,s*
n fy1 il e s T K ^ i R a >-c
■■»».■   pli.    o.    pj.   ľO. ft. mu.   gttt.   ft.   jet.   cofij. tflrtg«. cf«. cfjttc.
<v> fradaxiiningiiíotmnirahfnibdrftnníiŕ ulí]! mfnnihfi' ocndctiuT dtar ímnc dtmpí)'.
Jf)«. átU atcfptdr ä ďco gmtn Ocntdt&tomf.<Cimfcdídtn &iyo tfj'ti. (\M$£
a b c d d f g h 111 tu ti o p q r f t; uje j) ť <£. poíhinnarnimrdnitn ccrtic c\ ^ ^
*    dtcar omnmtau banc. tetuoUtuar. S qtn fdti ttioýfm ptmmáifímttbj ifm
PLATE 10
425
Cambridge, University Library, LI. 2. 10, fo. 19r. Pontifical, from Ely, mid 12th c. Early Anglo-Norman square notation. The plate shows chants for the Dedication of a Church.
This pontifical is one of a group of closely related manuscripts written in the century or so after the Norman Conquest of England. The text of one of them was edited by Wilson (1910), and the page shown here, part of the ceremony for the Dedication of a Church, corresponds largely to p. 105 of his edition. The end of a litany is visible at the top of the page. The next three lines (written in red ink, though this is hardly detectable in the photograph) are a rubric saying that the bishop shall rise from praying for the next part of the ceremony, and instructing him on the opening words. The prayer ('Oratio') Magnificare domine follows. On line 9 another rubric begins, for the bishop must now trace the Greek alphabet with his crozier (cambuta) from the left (i.e. north) eastern corner to the right western corner of the church, singing the antiphon Fundamentum aliud. The Greek alphabet is spelt out for him in the middle of the page: not only are the shapes depicted that he must trace, but a guide to pronunciation is also given. The bishop then traces the Latin alphabet from the right eastern to the left western corner of the church, singing the antiphon Hec aula accipiat.
Although many problems of transcription disappear with the introduction of the staff, manuscripts do not lose all their individuality of script and arrangement. The use of a wavy punctum for a note below the semitone step, the mi-neume (see I V.6.1), is characteristic of many Norman and English sources of the twelfth century: there is one in the first antiphon at 'NEmo'. The use of a b-flat clef in both antiphons is typically English. The opening of the first antiphon is given in transcription.
Funda - mentum a - Ii —
ud    ne-mo potest po - ne - re
U4>uútú < teátr* ftrJLufťfč♦ 'títAttrtulocč frtmrďmct L< ttwaty l'A ßf DE F é A he Á ti   K i
\x t l 0 u -* * t O   U At i o it A.
f
u A *   i e u A S   % e n íl ŕ j.    £> j
at ituUtm' muj: fm& or im& fdkw xdtmrx *Gmfi hhvrwr £uuk*fM*®kw ,4^yiu&iám /ťťwmííHw^Uhnumôtu
jf ÖaTcd c f í
; e i o u a c i
j U  a c t   p u a t*
* 4-6    p   a     tf    *   a   4    & h 6 a   c k   c J
• 4tn^u*m rihenänf temprer ttrfmr^nvr rnfö»^
e  ii    c    d   c   a c q
I
a   « íi í a  js  u fl
.luttuqifutf
PLATE 11
427
London, British Library, Add. 10335, fo. 9'. Theoretical writings and tonary, north Italy, early 12th c. Italian notation on dry-point lines, and alphabetic notation. The plate shows part of Guido of Are/zo's Micrologics.
The rather wide vertical spacing of the notes at the top of the page is caused by the use of the lines scratched for text as stave-lines.
Guido of Arezzo's famous treatise Micrologus has been edited by Gerbert (GS ii. 2-24) and Smits van Waesberghe (CSM 4), and translated into English by Babb (1978, 57-83).
The passage explains Guido's peculiar method of composing chants by deriving notes from the vowels of the text to be sung. The text Sancte lohannes has just been set in this manner, whereby the five vowels a, e, /, o, and u correspond to the notes C-G. But Guido intimates that such a small number of notes is rather hampering, and he then sets out a table allowing greater freedom of melodic movement. The whole scale from Tto is set out, and the five vowels assigned twice over, the second time a third higher than the first. An example, Linguam refrenens, follows, in which Guido has stayed within the compass F—f suitable for mode 7. Guido says that he has allowed a deviation from the system at the end ('In sola ultima parte hoc argumentum relinquimus') in order to bring the melody back to the tetrardus mode ('ut melum suo tetrardo conveniens redderemus'); but it is not clear to which note he is referring. (On Guido's 'automatic' method of composition, see Smits van Waesberghe 1951, 'Improvisation'.)
n
eirund fcnca£ pafcfonh
^pnunutn tw^l|um ^* xf£f'
1 J
i m
♦ •»'
U-
VaKcď£%^<rmtir far
f
1 / *
♦ "   "  m. **      i. f
tuf SC&nnntÄ^nttntS v
♦ -\ } - ' s[
*♦ * * *   í       * *    f ♦ '
í: " ■ 1
I . -
PLATE 12
429
London, British Library, Egcrton 35111 (olim Bcnevento, Bibliotcca Capitolare, VI. 29), fo. 174v. Sacramentary with priest's chants notated, from Bcnevento, 12th c. Bcneventan notation on dry-point linies. The plate shows prefaces.
The chief contents of a sacramentary are the proper prayers of mass: collect, secret, and postcommunion. It may also conitain the prefaces proper to various types of feast and seasons of the year, and occasionally these solemn chants are notated. The page shown here contains mostly the prefaice Tn festivitatibus', and towards the end of the right-hand column begins the preface 'In sollemnitatibus'.
Both text and music are written in ithe characteristic Beneventan scripts. The music is not elaborate and the full available range of neumes is therefore not called for. Several of the characteristic Benevenitan liquescent neumes are present:
Pes made liquescent; the end of the tail indicates the pitch of the second note.
Virga plus descending liquescence; the final short turn downward indicates the direction of melodic movement.
Punctum plus descending liquescence; the end of the tail indicates the pitch of the second note.
The start of the preface 'In sollemnitatibus' may be transcribed as given below. I have assumed that the cadences fall on a or G, as is usually the case for these chants. But this also involves assuming that the first custos is a tone too low.
		
A*\_,—m_           .__^ mm ^        mm)            1* MM Ml_                  .1_          AT ^~mm		
	s	
	-w- -	-#-m-m—-™-#-
Per om-ni-a     se - cu - la        se - cu-lo-rum.   A-men.   Domi-nus uo - bis-cum.
Double punctum (?) plus 41 $ liquescence.
Et  cum spi-ri-tu tu-o.      Sur-     -:sum cor-da.
*    t     ' _ » * *   *................. r- -
Vi f ficefce- q§ onttp? t& ur qui ttiUmcttctu^tmice-
——;--v1 \ 1^1» -J*i~^-~-A-—- ---.......- —........■ - '• • ~............
^tnrre{I«^ttfmi& mr? tnortamutuis at lr Ui ia
/ifr-'M ^f t^v ^ j j j t^^A? * A j J
_•    ,J     i, J   v -»
lv1<     1 J\<f T-^-y^-As- J    ^■V*'!
—*~1m1+t ---- —--A--fW-------J-J$   .    t jtf •
awn   ^ fVeg^u^ftfem* f<mr mvcaw
M\>, i^T^-Yvl. -   ITJ J '■ ■■ ' *
J^Ajl-^s-^^.....4J * *.  1 \ *
Ewtc-U>cam iibx-ptfxmSmt tJ^ttTti- attelma $votu\e Jtf. } £a<#ctnpTS;«ramttfUracttmisTmkrfollep
PLATE 13
431
London, British Library, Add. 29988, fo. 74v. Antiphoner, Rome, possibly St Peter's, mid 12th c. Central Italian notation on dry-point lines with red F-line. The plate shows part of papal Vespers on Easter Eve.
The Vespers service celebrated by the pope on Easter Eve, with its magnificent series of Greek and Latin alleluias, constituted one of the chief glories of medieval Roman chant up until the 13th century. The form of the whole ceremony has been set out by Stablein (MMMA 2, 111*—118*), and such of the music as appears in another Roman source, Rome lat. 5319, is edited in the same volume. The service consists of several parts, performed at different stations, hence the multiple singing of the Magnificat. The side shown here corresponds to pp. 114*-116* of Stablein's table (for the previous side of the manuscript see NG i. 488), near the end of the first station. The chants are:
line of music
1:   Magnificat antiphon Cito euntes (Second Station)
3:   processional antiphon In die resurrectionis mee
5:   antiphon Alleluia alleluia
6:   Alleluia V. 1. O kyrios V. 2. Ke gar estereosen 10:   Magnificat antiphon Venite et videte locum
The transcription is of the processional antiphon In die resurrectionis mee (cf. MMMA 2. 526). One of the few problems which arise in this, as in other manuscripts with central Italian and Beneventan neumes, is to differentiate between the different types of liquescence. In the transcription, each liquescent neume is given above the stave and numbered; From the size of the original sign there seem to be at least two 'degrees' of liquescence. The tiny commas used in 2, 4, 5, 6, and 9 seem to demand less emphasis (melodic or dynamic) than the others. It might be thought that they indicate a light liquescence at the unison, but after 4 comes the form of clivis implying a previous lower note, as at 5, so a lower liquescent note is implied.
Q _<yK*
In   di - e    re-sur-rec - ti-o-     -nis me - e      di-cit do - minus al-
JH 3j
z
4
P
-le-     -lu-       -ia con-gre  - ^ga - bo     gen-       -tes et col-li-gam
re^- -na et   ef-    -fun - dam su - per uos      a - quam mun- -dam
äjhtm jäm»B?>*fF^
al-     -le- -lu-
-la.
k íÁp 'fir nr árĺpcrfcinf grnnmu st äeľit>emiP "
• ' mm » - * *"
V"        - •       .' \
hnim  K'**'        íAhtítrtiof hof domine deuf rft&*im
"í
-  ........:   ;>   . * ] *     " "
Rlrti qťtiůum k deúbermď cárům Ufífy mm.iU • nf £jx« &<*;f jki^; unúm vrní CiZiia kammcra
.  * - - -     - *!
A. -     . .       . -    -7 L\f
c  .       .   . - . - :   - i
uenť? jwrrŕ? inuooirr mm dumjj^r eňr^fíi úfyífefe M'/undgWi uenenz ůhiď homm'if fumí in luCA$>
- . »   ; «I
dömitif úciKfur mrwdv m-ueriiec áuofeíhoir
ľ* : .. - "   :  " -    ♦    -   ;* -jMian f/u <pt4 but r! fuémf fibtď h^mmifúci
F..    - :.        . : V   * *ti
f 'm m m        • m
PLATE 14
433
London, British Library, Add. 17302, fo. 8r. Diurnal, monastic, Carthusian, 12th c. Square notation on dry-point lines. The plate shows the end of the series of O-antiphons.
The variety of square (or quadratic) notation known best nowadays is that of books written in Paris in the second half of the 13th and the early 14th century, used not only for plainchant but also, for polyphonic music and the songs of the troubadours and trouveres, and then adapted for the Solesmes/Vatican books of the last hundred years. Other varieties also existed, including that shown here. It is in fact more consistently square than the Parisian type, for even the climacus is written with three square note-heads. Another variety of square notation found in a number of Carthusian books can be seen in London, British Library, Add. 31384 (Briggs 1890, pi. 11), Naples, Biblioteca Nazionale VI. E. 11 (PalMus 2, pi. 47), and Toledo, Archivo Capitular 33. 24 (PaliMus 3, pi.206). English notators, on the other hand, tended to favour the use of lozenges (see Plate 15). Compare the following types:
	Paris	London	other	England
		17302	Carthusian	
virga	1	1	•	f
punctum	•	■	■	
pes	a	2	X	;
clivis		%	1	
climacus	H	s		\
porrectus				
D, F, a, and c clefs may be seen on this page, but none of the dry-point lines has been inked in.
The chants shown here are:
(end of the O-antiphons)
A. O (mens splendor (end only)
A. O Ernmanuhel
A. O rex gentium 'In laudibus' (Lauds of Monday in Ember Week of Advent)
A. Ouerite dominum
A. Cum venerit filius hominis
A. Dies domini
A. Haurietis aquas
A. Gaude . . . (first word only)
The texts of the O-antiphons are of course non-scriptural, but the sources of the other antiphons are indicated in the margin: Isaiah, St Luke, etc.
t/U. ' , ■' */(> ■ I ft
"<>«
n^^WJf^       <i^fott>^wt tatmtrratire tcponte &mwk nvncfotffmt
jCýn*        l^r^m^M occserae^ itt» w ttittnť tiefen. íK ýne
..........
PLATE 15
435
Oxford, University College 148 (kept in Bodleian Library), fo. l()r. Ordinary-of-mass chants and sequences, from Chichester, 13th c. English square notation. This plate shows the start of the Kyries.
As already mentioned (see Plate 14), some English notators of the 13th century favoured the diamond or lozenge shape as a basic notational element. Even the head of the virga tends to slant. The liquescent forms can be seen most easily at 'eLEYson':
'1   11)   .J   *J *
The oriscus as a repetition of the previous note (as in the pes stratus; cf. Plate 9) can still be detected:
The transcription gives the fifth verse of the first Kyrie on the page (line 5):
Qui canunt an-te te      preci-bus an-nu-e       et tu no-bis semper e - ley-son.
. ty> ffi t«:      /«t,a< ..-c *
Christ e
-ley-son.
This is the first page of a collection of Kyries, with and without Latin verses (listed Hiley 1986, 'Ordinary', 39). The rubrics are still faintly discernible at the end of lines 1 and 10:
first Kyrie, Te pater supplices: 'Dominica prima adventus domini cantus sine carmine'
second Kyrie, Clemens rector: 'Dominica secunda cantus sine carmine'
Te pater supplices was frequently sung on Christmas Day, which is no doubt why it appears first here. The rubric indicates, however, that the melody alone ('cantus'), stripped of the Latin verses, was also to be sung on the first Sunday of Advent; similarly with Clemens rector on the second Sunday.
C XXX
I--~4—j----------V—-m-
♦ *
tvcs í nnnm^lp lxcpnngcrc mcnic
♦ * v
»V   ,..--.Aj Al P j~"
its    fic nmr
♦ ♦
rrw^|jiJiPtig
1H M-—*-.-r*   '-|—*
PLATE 16
437
Oxford, Bodleian Library, Lat. Liturg. a. 4, fo. 130r. Chants of the Milanese rite, summer part, 1399. Milanese notation. The plate shows part of the responsory Hi sunt qui secnti for SS Gervase and Prothase.
The scribe of this manuscript, the priest Fazio de Castoldis, wrote at the end that he had finished work on 31 January 1399, at the age of 40.
Hi sunt secuti is a responsory 'cum infantibus', one of those sung after the hymn at Vespers on the most important saints' days (discussed by Moneta-Caglio 1957). The performance of the responsory involved much repetition and the inclusion of two long rnelismas, the melodiae primae and tnelodiae secundae, during the successive repeats of the respond. The overall form seems to have been:
Solo      Respond   Hi sunt qui secuti
(second part) Ad hoc pertingere me me runt ut Christi martyres fierent
Schola   Respond   hi sunt qui secuti (as sung by soloist)
Solo      Verse       Ecce quam bonum
Schola   Respond   (second part) Ad hoc pertingere
with melodiae primae at 'martyres Solo      Respond   Hi sunt qui secuti (first part) Schola   Respond   (second part) Ad hoc pertingere
with melodiae secundae at 'martyres'
Or possibly during the final repeats, the respond was sung complete by the soloist and complete again by the schola, with melodiae.
The schola was constituted by the magister pueri and the boys themselves, hence the nickname of such responsories. Sometimes a boy might have been entrusted also with the soloist's part.
The notation in this massive book (55 X 39 cm.) is large enough for singers to perform from, and presents no difficulties of transcription. The small virga-like note which follows a larger one of the same pitch, as in line 1 at 'per-TIN-GE-re' is a repeated note, not a sign of liquescence. A transcription of the Melodiae primae is given. Lines divide the melody into phrases, which I have labelled a-d; it transpires that the melodiae repeat themselves.
a.
-00-
-r -1     w -*m I *
Mar-ty - res
P
O' r*.
(-res)
-rent.
Di ftirapr tqJmnottf »mm  m ff Ks flu Trrtrm piU
*Ä   irp foujgmtf      trp tou,qfr ifi'foXpnfir    tfi'fô pŕ iŕFtíŔMŕ
10 (on. mtawmůfítt
i C i'- !*'
ft 1    f   n      í1 f        T    I   t"4kt Í
o6(r m tma jmrnomítm* bouř tioblmru^ . a«íHw%
.     p       "Tv:| j.ted i-.jff í Inif »ihm« tt íd t* mu* ffí M* n
PLATE 17
439
Edinburgh, University Library 33, fo. 138r. Gradual, from Holland, 15th c. Gothic notation. The plate shows chants for the ordinary of mass.
Many graduals contain not only the proper chants of mass but also everything else that cantor and schola should sing: tropes (in the earlier Middle Ages), ordinary-of-mass chants, and sequences. The present manuscript concludes with a collection of ordinary-of-mass chants, some of which are provided with indications for dividing the performance between choir and organ (the latter presumably using the chant as a cantus firmus for improvisation). The two Glorias on this page are examples of this. The melodies are well known as nos. I and IX in the Vatican/Solesmes service-books. In between comes a Kyrie, with the Latin rubric 'De domina ii' (a second Kyrie, or Kyrie-Gloria pair, for Our Lady), and, in the margin, the Dutch rubric 'van onse liefue vrouw'. This Kyrie-Gloria pair were frequently sung at Marian masses, and are assigned thus in the Vatican/Solesmes service-books.
The notation is characterized above all by the virga shaped like a horseshoe-nail ('Hufnagel'). A transcription of the opening of the Kyrie is given:
Ky- _ri_e-
-ley - son.
PLATE 18
441
London, British Library, printed book IB. 8668, fos. 113v-ll4r. Printed gradual of Mainz, 1500. Gothic notation. The plate shows sequences.
Early printed service-books with music are no different in arrangement from their manuscript predecessors. The present volume contains the proper chants of mass, then a collection of seventeen sequences (an extremely conservative collection, in which Victime paschali laudes, of the 11th century, is the most recent item), and finally Kyrie-Gloria pairs, Sanctus, and Agnus Dei chants.
The end of the sequence for St Martin, Sacerdotem Christi, occupies the first six lines of the left-hand page. On the last line begins the sequence for apostles, Notker's Clare sanctorum senatus apostolorum, which we have seen in a Mainz version of some 540 vears earlier on Plate 6.
v
Plainchant and Early Music Theo
V.l. INTRODUCTION
Vivell 1912; Ricmann 1920; Smits van Wacsberghe 1969; Gushec 1973; Palisca, 'Theory, Theorists', AG; Schucller 1988; Dyer 1990.
In a general sense a great part of the theoretical writing of the Middle Ages is of relevance to the plainchant repertory. For example, much could be made of the place of 'music' in the history of ideas in the Middle Ages, in the medieval cosmology. But this is not the place for such a demonstration, and no attempt can be made here to summarize all the possible themes. Instead, the following sections will comment on a restricted number of topics which touch directly upon chant composition and practice. The two most important of these are: (i) the way in which the chant repertory was aligned with the pitch system of Greek antiquity, and the concomitant coupling of the modal system with particular pitches; and (ii) the identification of what constitutes the modal character of particular chants. Some of the material treated here borders on the speculative part of music theory (music as the embodiment of celestial harmony), some of it on the pedagogical (how to notate music, how to read it), but no full account of these is attempted.
The growth of a music-theoretical literature from the ninth century onward was stimulated by two factors. The first was the interest in and the desire to emulate the writings of late antiquity that had survived to the time of the Carolingian renaissance. The second was the need to abstract from the practice of music in the liturgy some rules to regulate that practice and make it more uniform from place to place.
The concept of 'musica' inherited by the Middle Ages from classical antiquity was that of musical science rather than practice. The Greeks themselves did not write about their own practical music. So it is not to be expected that 'musica', music theory, should have been of direct relevance to Christian chant. The chant repertory had of course grown up quite independently of any theoretical system, and it is interesting to see how medieval writers wrestled with the heritage of antiquity (mainly as transmitted by Boethius), drew from it what was of practical use in the daily performance of the liturgy, and used some of its simpler doctrines as the foundation for speculative theory, all in a completely new intellectual and musical environment.
The other fascinating process at work through the ninth to twelfth centuries is the
2. The legacy of Antiquity 443
bringing into line of the chant repertory with what had at first been theoretical concepts, which then exerted a retroactive influence on chant itself. It was a short step, though a momentous one, from the recognition that the majority of chants in such and such a mode displayed this or that tonal behaviour, to the deliberate alteration of chants which, as it were, stepped out of line with the majority.
The next section gives a brief account of the musical theory of antiquity which was known to the early Middle Ages. Subsequent sections describe the way in which the pitches sung in chant were assigned places in the Greek Greater and Lesser Perfect System, together with the growth of a literature about the modal quality of chants.
V.2. THE LEGACY OF ANTIQUITY
The principal writings on music of late antiquity known to the Middle Ages were, in chronological order, those of Censorinus (3rd c), Calcidius (4th c), Augustine (354-430), Macrobius (early 5th c), Martianus Capella (early 5th c), Boethius (c.480-c.530), Cassiodorus (<:\485-c.580), and Isidore of Seville (c.599-636). Of these the last four were the most influential, Boethius providing by far the most substantial body of material. Censorinus, in three chapters of his work De die natali, mostly borrowing from Varro (1st c), was concerned with speculative music theory: music and the cosmos, and its ethical influence on man (Censorinus, ed. Rocca-Serra 1980, ed. Sallmann; see also Sallmann 1983). Material of this sort is common to most of the above writers and was repeated by many medieval authors. Also repeated in the Middle Ages, but hardly expanded, were explanations of the harmony of the spheres, such as the one in Macrobius' commentary on the Somnium Scipionis of Cicero (ed. Willis 1970, trans. Stahl 1952). Less commonly encountered is the type of material in Augustine's De tnusica (PL 32. 1081-1194, trans. Taliaferro 1947, Knight 1949). Music is seen here, as elsewhere, as an embodiment of proportion. The concept was generally illustrated with reference to the harmonic intervals between notes—the octave, fifth, fourth, and so on. But in Augustine's treatise the proportions are those displayed in the metres and rhythms of metrical verse (see Crocker 1958, 'Musica'). Music's power over living beings—not only humans but also animals and even plants—is again a principal preoccupation in the musical part of Martianus Capella's De nuptiis Philologiae et Mercurii, which then proceeds to borrow material by Aristides Quintilianus (3rd-4th c.) on the classification of musical knowledge, including its pitches, scale systems, species, and intervals (ed. Willis, trans. Stahl; see also Stahl 1971). The music-theoretical texts of Cassiodorus (ed. Mynors 1937) and Isidore (ed. Lindsay 1911) are likewise mainly concerned with the classification of musical knowledge in groups of topics: of the topics themselves there is relatively little discussion (On Cassiodorus see also Holtz 1984, Strunk 1950; on Isidore Strunk 1950.)
The classic exposition of the relationship between music, proportion, and the universe is in Plato's Timaeus, known to the Middle Ages in the reworking by
444 V. Plainchant and Early Music Theory
Calcidius (ed. Waszink 1962; Waszink 1964; for Plato see Arnoux 1960, Cornford 1937). Here it is explained how the demiurge (easily transmuted by later writers into the Christian God) created the world-soul from a substance which combined that which is indivisible and unchanging (eternal verities) and that which is divisible, material, and subject to change. This substance was divided into sections whose lengths were in strict proportional relationship with each other, 1 : 2, 2 : 3, 3 : 4 and so on. Since the proportions were those of Pythagorean harmonic theory, music could be understood as a reflection of the creation of the world and its divine order.
The divine order was reflected not only in the ordering of the Earth and heavenly bodies, the macrocosm, but also in man, the microcosm, in the balance between his spiritual and physical being. Hence the influence of music upon the state of his soul and his behaviour.
Although Boethius' De institutione musica (ed. Friedlein 1867, trans. Bower 1989) includes chapters about music's ethical power, the harmony of the spheres, Pythagoras' discovery of musical proportions (the story of the hammers of different sizes in the smithy) and other matter, its chief substance is a comprehensive exposition of proportional theory, scales, and species. (On Boethius see also Bower 1967, 1978, 1988, Caldwell 1981, Gibson 1981, White 1981, Chadwick 1981.) Much of Boethius' exposition is set out with reference to the monochord, a method taken up in the Middle Ages, at least by those writers who go so far as to demonstrate the physical basis of the simple proportions. Boethius appears to have wished to attempt a summary of Greek musical learning in the Latin language, at a time when the subjects of the quadrivium (arithmetic, geometry, music, astronomy) were neglected in favour of the rhetorical arts, the trivium (grammar, rhetoric, and dialectic), partly out of fear that Greek learning was in danger of being lost to the world of the late Roman Empire. His plan (which remained uncompleted) appears to have been a reworking of Nicomachus' handbook of harmonics, a lost introduction to music by Nicomachus, and Ptolemy's harmonics (Bower 1978).
Boethius' demonstration of the correct intervals between the notes which make up the Greek Greater and Lesser Perfect System was of great importance. Since he demonstrated not only the intervals of the diatonic scale (the equivalent of, say, our modern A-a', without accidentals), but also those of the chromatic and enharmonic genera, Boethius provided far more information than was useful to medieval musicians: at least, there was no attempt in the Middle Ages to relate the intervals of those two genera to the chant repertory. Boethius' division of the monochord was accomplished in two stages, the first establishing the first, second, and fourth notes of each tetrachord, the second stage filling in the notes which differed from one genus to another. Ex. V.2.1 shows how the first stage proceeded, and then gives the second-stage division of the highest tetrachord. For convenience, the modern equivalent pitches A-a' are shown beneath the line representing the string of the monochord. The complete diatonic scale, its component tetrachords, and the Greek names of each note are given in Table V.2.1, again with hypothetical modern equivalent pitches.
Equally important was Boethius' demonstration of the different species of octave,
2. The legacy of Antiquity Ex. V.2.1. Boethius' division of the monochord
445
First division of the monochord q K     D L,
MM    X E.
—*-1-1—i—
1.	take a string length	AB				
2.	divide AB into 4:	CDE	AB	CB :	DB :	EB
	divide AB by 9:	F	AB	: FB	= 9 :	8
4.	divide AB by 3:	G	AB	: GB	= 3	2
5.	divide CB by 4:	K	CB	: KB	= 4	3
6.	divide DB by 9:	L	DB	: LB	= 9	8
7.	divide DB by 4:	M	DB	: MB	= 4	: 3
8.	divide DB by 3:	N	DB	: NB	= 3	: 2
9.	divide KB by 2:	X	KB	: XB	= 2	: 1
2 : 1
Second division of the monochord (only hyperboleon tetrachord)
A
i_
0
1
6
1. divide AB into 4: let the centre be 0, three-quarters be LL
2. divide LL to B by 8; measure the same length (i.e. one-eighth of LL-B) in front of LL: KK. "    KK-B : LL-B = 9:8
3. divide KK to B by 8; measure the same length in front of KK: FF.
FF-B : KK-B = 9:8
4. divide LL to B by 3; measure the same length in front of LL: DD
DD-B : LL-D = 4:3
5. divide KK to LL by 2; measure the same length in front of KK: HH
HH-B : KK-B = 19 : 18
6. divide DD to FF by 2: EE
DD-B : EE-B : FF-B = 512 : 499 : 486
fifth and fourth which can be found within the system (set out in a different order from that found in extant Greek writings). Each species is distinguished by the arrangement of intervals within it. The interval of the fourth, for example, occurs between five pairs of notes, modern equivalents A—D, B-E, C-F, D-G, E-A. Since, however, the semitone falls at the same point in the tetrachord for A—D and D-G—it
446 V. Plainchant and Early Music Theory
Table V.2.1. Greater and Lesser Perfect System, Diatonic Genus
Tetrachord Pitch-name Medieval
equivalent
	Greater Perfect System		
		"nete hyperboleon	a
hyperboleon		paranete hyperboleon	s
		trite hyperboleon	f
		mete diezeugmenon	e
diezeugmenon_.....		paranete diezeugmenon	d
		trite diezeugmenon	c
		paramese	h
		■mese	a
meson__________		lichanos meson	G
		parhypate meson	F
		:hypate meson	E
hypaton______________		lichanos hypaton	D
		parhypate hypaton	C
		-hypate hypaton	B
		proslambanomenos	A
	Lesser Perfect System		
		"nete synemmenon	d
synemmenon ______		paranete synemmenon	c
		trite svnemmenon	bb
		■ mese	a
meson        _ .		lichanos meson	G
		parhypate meson	E
		: hypate meson	E
hypaton _____		lichanos hypaton	D
		parhypate hypaton	C
		'-hypate hypaton	B
		proslambanomenos	A
forms the middle step of the three—and at the same point in the tetrachord for B-E and E-A, there are only three different species of fourth. In Ex. V.2.2 the species of fourth, fifth, and octave are given with hypothetical equivalent modern pitches, and the semitones are marked with the number of the step at which they occur. To this example have also been added the names of the eight octave species. (Boethius actually sets these out in a different way, imagining them all to occur between the same two pitches when the complete scale is itself transposed to different pitch levels: see iVG xii. 379.) The difference between this nomenclature and the common medieval one is discussed below.
J. A Pilch-System for Plainchant 447 Ex. V.2.2. Species of intervals
species of 4th in diatonic genus:
species of 5th in diatonic genus:
species of octave in diatonic genus:
fe        " *        r •»,     ^ * * ,,     * * *.
Hypodorian Hypophrygian Hypolydian Dorian
_Q---C=,-£-	—-3-	-*-
a"    *   *   a r~~i		r--- 1
	W         M Mi	* m a      ' 1
"V 'a	* m a	*  * . ~1
¥ * Phrygian	* M * • Lydian	9 a Mixolydian
V.3. A PITCH-SYSTEM FOR PLAINCHANT
(i) Introduction
(ii) Hucbald of Saint-Amand
(iii) The Enchiriadis Group of Treatises
(i) Introduction
Ecclesiastical chant plays no part in any of the writings referred to in the last section, and the first evidence that music theory and the practice of chant were on a converging course, so to speak, comes with the Carolingian renaissance of the late eighth and ninth centuries. It is known, for example, that Alcuin, Charlemagne's chief adviser on educational matters, conceived of a liberal arts programme at the imperial court and probably introduced it at his abbey of St Martin's, Tours; there is evidence of it at Orleans, under Theodulf, and especially at Fulda, under Rhabanus Maurus. The
448 V. Plainchant and Early Music Theory
revival of classical learning was exactly contemporary with the establishment of the Roman liturgy and its chant in the empire.
Nevertheless, original new writing about music theory is not found until the next generations of scholars, beginning with Aurelian of Reome, whose treatise Musica disciplina is usually dated r.850. Another group of scholars was connected in various ways with the court of Charles the Bald (823-77, king from 840, emperor from 875), and the schools at Saint-Amand, Laon, Reims, and Auxerre. These are Johannes Scotus Eriugena (c\810-c.877), Remigius of Auxerre (c.841-r.900), and Hucbald of Saint-Amand (c.840-930). Regino of Prüm, writing at the turn of the century, the anonymous treatise known as Alia musica, and the Enchiriadis group of treatises (Musica enchiriadis, Scolica enchiriadis, and the Commemoratio brevis de tonis et psalrnis modulandis) complete what is altogether a very important, even astonishing achievement in creating a body of new music theory, borrowing occasionally from the authorities of the past but often startlingly independent, all datable to the second half of the ninth century.
Johannes Scotus Eriugena (or John Scot) and Remigius of Auxerre left no music treatise in the normal sense. Johannes was born in Ireland about 810 and came to Francia about 845. He is thought to have taught grammar and dialectic at the court of Charles the Bald (though the existence of a formal palace-school is not secure: see McKitterick 1983, 213-14, where Compiegne is proposed as Charles's favourite palace). He was perhaps the only scholar of his age who had a mastery of Greek, and in the early 860s he translated several Greek works for the king. Music is treated in John's work only parenthetically in his masterwork on transcendental philosophy, Petiphyseon or De divisione naturae, and otherwise only in his commentary on Martianus Capella. Martianus' work seems first to have become known again during Charles's reign, after a long period of oblivion (see Huglo 1975, 142if.). Remigius also composed a commentary on it. (On John see Lutz 1939, Bower 1971, Uhlfelder 1976, Contreni 1978; on Remigius see Lutz 1956, 1957, 1962-5.)
(ii) Hucbald of Saint-Amand
Editions: GS i; Babb 1978; Traub 1989.
Literature: Müller 1884; Potiron 1957; Weakland 1956, 1959; Chartier 1973, 1987.
Hucbald's treatise De harmonica institutione (the title is actually Gerbert's) occupies a key position among these writings, not because it was the earliest or the best known (relatively few copies are known), but because it sets out most clearly the way in which classical theory could be explained with the aid of the chant repertory, and could in turn supply an intellectual foundation for a better understanding of chant. Hucbald had a monastic education, at Saint-Amand, Nevers, and Saint-Germain at Auxerre, where he was a fellow student of Remigius of Auxerre. He returned to Saint-Amand as master of the school, founded another school at Saint-Bertin, and in 883 was called with Remigius to revive the cathedral school at Reims. His remaining years were spent mainly at Saint-Amand.
J. A Pitch-System for Plainchant 449
It may well have been among the circle of Remigius and Hucbald that examples of plainchant were first used to illustrate theoretical concepts. Remigius, in his commentary on Martianus Capella, glossing Martianus' definition of the symphoniae (consonances), quotes three introits to illustrate the intervals of the fourth and fifth (Tibi dixit cor meum and Dens in adiutorium meum both begin with a rising fourth, Ocidi mei semper with a rising fifth). Hucbald goes far beyond this, however, citing over sixty chant passages to illustrate intervals and scale-segments. In this the monochord plays no part. The intervals, for example, are not established according to their proportions in terms of length on the string of the monochord; Hucbald simply cites eighteen examples which illustrate nine different intervals in ascending and descending motion. He does not at first name the intervals in question, wishing only to establish the simple notion that there are intervals of differing size. He then shows how the larger intervals consist of combinations of the tone and semitone. The nine intervals which occur in the chant repertory are: semitone, tone, minor and major third, perfect fourth, tritone, perfect fifth, minor and major sixth.
The rudiments which Hucbald is next at pains to demonstrate are: the consonant intervals of the fourth, fifth, octave, octave plus fourth, octave plus fifth, and double octave; and the difference between a tone and a semitone. In the latter connection he cites two examples of chant passages which run through a hexachord, with the semitone in the middle: in modern notation C-D-E-F-G-a and then the reverse, a-G-F-E-D-C. This corresponds, says Hucbald, to the arrangement of strings on the six-stringed cithara, and he suggests that one should practise writing chants on six lines, representing these strings, with the semitone in the right place (see Ex. V.3.1).
Ex. V.3.1. Antiphon Ecce vere Israelite! notated by Hucbald on six 'strings'
ixa_______r     r___e.
—'—9-9-*-*-•-—
Ecce ue-re Is-ra-he-li
ta   in quo do-lus non est. (est)
So far nothing has been said about pitch-names, neither letters nor the Greek names found in Boethius (whose work Hucbald knew and occasionally referred to). This continues in the next part of Hucbald's treatise, where he describes the arrangement of tones and semitones in a two-octave scale. The arrangement actually
450
V. Plainchant and Hady Music Theory
corresponds to that of the diatonic Greater Perfect System, which can be represented by modern A—a', but Hucbald does not immediately state this. His next remark is that water organs and other instruments follow a different arrangement, one which corresponds to our modern major scale (Hucbald again simply gives the order of tones and semitones). Next he explains the classical arrangement of tetrachords within the double-octave scale. This is how Boethius arranges them, he says, and he goes on to cite chant segments where the notes of the tetrachord may be heard, once again in descending and ascending order.
Hucbald's next step is crucially important. He suggests we now imagine the two-octave scale from the bottom up as a series of tetrachords of a different species, the species with semitone in the middle: modern A—B—C-D, D-F-F-G, and so on, and he gives an example from the chant repertory where this scale-segment can be found. At this point Hucbald establishes the place of the synemmenon tetrachord. For Boethius this would have corresponded to modern d-c-db-a, but for Hucbald, since he has introduced the different species of tetrachord, it corresponds to modern G—a-bb-c (Ex. V.3.2; compare Table V.2.1 above). In giving plainchant examples to
Ex. V.3.2. Hucbald's arrangement of ascending tetrachords
illustrate the distinction between diezeugmenon and synemmenon, the disjunct and conjunct tetrachords, he appears to imply that modern bb and fc>tq can occur practically side by side. The version of the introit Statuit ei dominus described by Hucbald would seem to be something like Ex. V.3.3. Only at this point does Hucbald give the Greek names of all the notes in two-octave scale, for which he necessarily has to revert to the Greek tetrachords.
Ex. V.3.3. Introit Statuit ei as described bv Hucbald
	• ■				=^5=		#•# #		rm»a m	-
Sta - tuit	e -i	do-—-—-	-mi	- nus	tes -	ta-men-tum	pa - cis	et	prin -	cipem fe-cit
<m-r'Z m—m-	m -T«			m m m				—»0.		—^=~=-
			~i-				;»* »■			,*>••• •*
e  - um	ut sit	il - li	sa -	cerdo-ti	- i	dig-ni   - tas	in	ae	-	-ter - num.
It is now that Hucbald introduces the idea of pitch-symbols for the individual notes of the scale. As has already been explained (IV.5), the symbols he suggests were borrowed from Alypius via Boethius, but not subsequently taken up by any other writer. Once again, chant examples are given, among them the pseudo-Greek type melody for mode 1 (Ex. V.3.4).
3. A Pitch-System for Plainchant Ex. V.3.4. Hucbald's letter-notation
451
I      M  PMP CF I     M   PC   PM  PC F
Al-le-lu - ia No-ne-no - e - a - ne
As Hucbald himself says, the point has now been reached to which everything looked forward from the beginning, and we are about to see 'the fruit which springs from the seed just sown'. Every aspect of music theory discussed so far, intervals, tetrachords, the complete scale, and the pitch-symbols, has been illustrated by chant examples, and this is tantamount to saying that plainchant 'fits' the scale system just described. This is also the justification for Hucbald's introduction of a different tetrachord species, for the second tetrachord in his arrangement, modern D-E-F-G, corresponds to the four finals of the chant repertory. Hucbald gives the names of the four 'modes or tropes, which we call tones': protus, deuterus, tritus, and tetrardus; and he explains that each of the four notes 'reigns' over an authentic and a plagal trope.
Hucbald's last task is then to make clear the relationship between the starting-note of chants and the finals. A last batch of examples, notated with his pitch-symbols, illustrates a series of starting pitches in chants (nearly all antiphons) which end on each of the four finals. For the tetrardus mode, for example, Ecce sacerdos, starting on modern d, is cited, followed by Beatus venter and Quomodo fiet, both starting on c, Dixit dormnus domino meo, starting on b, Eriimpant monies and Beati quos elegistis, both starting on a, and so on.
It is clear that Hucbald had envisaged at least a large part of the chant repertory as being compatible with the diatonic two-octave scale plus middle bb. If some chants were not capable of being fitted into that framework, and, by implication, of being notated with his pitch-letters, Hucbald does not say so. One wonders, however, whether the Greek system was taken over simply because there was no alternative. The main restriction was that the alternative chromatic step was available only at one point, a few steps up from the finales: D-E-F-G . . . bb-b^-c. If a chant needed the alternative chromatic step at another point relative to the final, it had to be envisaged as occupying a different place in the system. This is the case, for instance, with a number of chants which need the alternative just above the final: they have to be notated with final a: a-bb—b^-c. Nevertheless, once the idea gained a hold that chants stood in this relationship to a clearly definable pitch-system, and once teachers began to pick out chant melodies on monochords where the notes of the Greek system were marked off, pressure must have been felt to eliminate notes which did not fit that system.
Hucbald was probably not the first to make this connection between chant and theory. At any rate, it is also implicit in the anonymous Alia musica and the Enchiriadis group of treatises (which has its own pitch notation, the dasian symbols).
452 V". Plainchant and Early Music Theory
But nowhere else is the connection so painstakingly and clearly explained. Aurelian of Reome, writing about thirty years earlier, has no notion of a pitch-system, and Hucbald's treatise has the aura of introducing something radically new. It is one of the crucial documents of Western music.
Hucbald was undoubtedly one of the leading churchmen and teachers of his time. Luckily a number of compositions by him appear to have survived. His office for St Peter is to be found in many medieval sources (AS 439), and his Gloria trope-set Quern verepia laus occurs in several early manuscripts (Ronnau 1967, Tropen 240, to the melody Bosse 1955, no. 39). (See Weakland 1959.)
(iii) The Enchiriadis Group of Treatises
GS i; Le Holladay 1977; Bailey 1979; Hans Schmid 1981; Phillips 1984.
In the manuscript Paris, Bibliotheque Nationale, lat. 7202 the treatise Musica enchiriadis is attributed to Hucbald, and in Gerbert's edition not only that work but its Scolica and the treatise known as Alia musica were also credited to Hucbald. Miiller (1884) and Muhlmann (1914) were able to disprove the attributions. Meanwhile, no other author has appeared to claim Musica enchiriadis (or the others), and even its date and provenance are unclear (see Phillips 1984). Until decisive evidence turns up, it is probably safe to regard it as a work written in north-east France or west Germany, perhaps even within the orbit of Hucbald's activity, in the second half of the ninth century.
Musica enchiriadis and the Scolica, which subjects some of the topics in Musica enchiriadis to more extended discussion in dialogue form, are most famous for their demonstration of organum of various types. Hucbald also referred to organum by way of illustrating the concept of consonance, and it has been argued that the main point of the discussion of organum in Musica enchiriadis is to illustrate the perfect consonances (Fuller 1981). In the first part of the treatise, on the other hand, the author presents eight chant examples, which are each typical of one of the eight ecclesiastical modes. This is done by means of the dasian notation set out in the opening chapters of the book (see above, IV.7). Each pair of examples is preceded by a brief formula which runs through the notes of one of the four species of fifth (modern equivalents a-d, b-E, c-F, d-G: it should be stressed that, as in Hucbald's work, the alphabetic series with which we are familiar is not connected with the dasian symbols). Like Hucbald's, the examples are set out on lines, though here the pitch of each line is made clear by a dasia sign. The examples for the protus mode are shown in Ex. V.3.5.
The Enchiriadis author, therefore, has an approach diametrically opposed to Hucbald's subtle and gradual method. The pitch-symbols appear at the very start of the work so that the author can immediately illustrate whatever concepts he choose without recourse to memorized chants.
While chant is not the onlv interest of the Enchiriadis author, the Commernoratio brevis is a veritable plainchant primer. Notated in dasian notation, and thus fixed in
3. A Pitch-System for Plainchant 453 Ex. V.3.5. Examples in protus mode from Musica enchiriadis
Modulatio ad principálem protum modum et subiugalem eius
Alcjf le^lu/A/ i&pp. Lau/da/tyte/' Dop mi / num     de IP cae/>lisp.
J._a_
ff_da__num
/ / / Lau_mi_de_Cae.
p_Lki_e._h     J_turn__xe_
b_caelis._cae_Lau_da_Dfium.
Cae/lif  caePlo/ rumP  lauMdaPteFf DepumP,
P
Al-le-lu - ia.        Lau-da-te Do-minum de ce-lis.      Caeli cae-lorum lau-date Deum.
pitch notation over a century before pitch notation was used in chant-books themselves, are nearly seventy musical examples from the standard chant repertory, including the psalm-tone formulas with their multiple cadences, and also the didactic melodies associated with the different modes: Noeanne, Noeais, and so on. The plan is roughly as follows:
(a) pseudo-Greek (Noeanne) melodies and melismas (neumae) and psalm tones (Gloria patri), plus one or two antiphon incipits, for the gospel canticles in each of the eight modes ('used where the performance of a chant is slower, as for instance for the gospel canticles, when time allows');
(b) eight psalm tones 'for the faster chants';
(c) a series of tones which fall outside the regular eight;
(d) different final or median cadences for the psalm tones.
In this work the practical intent is plain. The Commemoratio author even introduces a sign for the quilisma into the dasian notation and a dot to separate note groups, adapting it more closely to the task of notating chant examples (see Bailey 1979, 19). If the given examples are learned, the student will have a thorough grounding in office psalmody and the psalm-tone formulas governed by various typical antiphons. This is not music theory but material for the teaching of chant. Even the additional commentary confines itself to practical matters: the speed of singing, the co-ordination of antiphonal choirs, and so on.
454
V. Plainchant and Early Music Theory
V.4. THE MODES
(i)	General
(Ü)	Aurelian of Reome
(iii)	Regino of Prüm
(iv)	Nomenclature
(v)	Greek Names and Octave Species
(vi)	Italian Theory: The Dialogus de musica
(vii)	Guido of Arezzo
(viii)	South German Writers
(ix)	Modal Theory in South Germany
(x)	Later Syntheses
Powers, 'Mode', NG. (i) General
The definition of 'mode', in the familiar sense of the eight plainchant modes, is not a simple matter. The Latin term modus itself was not universal in the Middle Ages, for it competed with tonus and tropus. There is no space here for a detailed examination of medieval terminology and practice. For the sake of clarity, I have regarded the eight modes as categories for the classification of melodies according to their tonality (implying range and prominent notes, including reciting notes and final notes) and melodic type. It is arguable that the latter feature was originally the deciding factor. It appears that it was the Byzantine system of echoi which inspired Western musicians in the Carolingian period to classify their own chant repertory, and also provided some technical terms for the classification. The echoi are primarily melodic types: 'In general, a hymn composed in an echos will contain a set of melodic turns (motifs, formulae or melody-types) peculiar to that echos; and these structural devices will be found in other hymns composed in the same echos' (Velimirovic, 'Echos', NG). There is plenty of evidence to suggest that the eight-mode system was imposed in the West on a chant repertory not originally so conceived. Western chant did not develop within an eight-mode framework, with melodic material assignable to eight families or types. The eightfold division must therefore have relied primarily on the more abstract criteria mentioned above: range and prominent notes.
That this was so is suggested by the contrast between two medieval works. The tonary attributed to Regino of Prüm (<r.900) classes a number of antiphons in a particular mode because of their opening melodic formula, but with the remark that they close in a different mode. About a century later, the author of the Dialogus de musica states categorically that the final note determines the modality: 'Tonus vel modus est regula, quae de omni cantu in fine diiudicat.' ('A tone or mode is a rule which classifies every melody by its final.') In the next section of the Dialogus, the author prescribes ranges for melodies in the eight modes. In so far as these two writers
4. The Modes
455
are typical, they suggest that theoretical abstractions, rules of thumb—final and range—replaced melodic characteristics as criteria for determining modality.
While such rules of thumb were no doubt useful at an elementary level, musicians continued to look for more comprehensive definitions. It was necessary, for example, to identify the final, which could be done with reference to the intervals on either side of it. Thus D has a tone then a semitone below it, and a tone then a semitone above it, whereas E has two tones below it and a semitone then a tone above, and so on. This focused attention on the scale-segments within which melodies of the various modes moved. Furthermore, more than one note might have the same pattern of intervals around it: D resembles a, E resembles b, F resembles c in this respect. There developed, therefore, a discussion in the theoretical literature about the tonal affinity between notes and hence between the modes they governed.
What was happening, when such a doctrine was developed, was that an abstract scale-segment (the hexachord was the most widely used), with which particular melodies could be compared, replaced a knowledge of the individual melodies or melodic families. We could postulate three (among many) stages: (i) The singer says: 'This melody (like its sisters in the family) belong to this mode . . .'. (ii) The singer says: 'This chant resembles the melody 'Noeagis', therefore belongs to this mode . . .'. (iii) The singer says: 'This chant ends on re of the hexachord, therefore belongs to the protus mode . . .'. In the first stage the singer has to know the modal assignation of all melodies, or at least of groups of similar melodies. In the second stage he has to know the Noeanne melodies and be able to select the right one for comparison. In the third stage he has to match his chant with but a single scale-segment.
The definition of modality was not a purely theoretical exercise in the Middle Ages. I^or chants with psalm verses sung to simple recitation formulas (introits and communions at mass, antiphons during the office) the selection of psalm tone and its cadence depended on the tonal character of the introit/communion/antiphon. It was convenient to be able to classify each chant in one of the eight modes, so that the selection of the corresponding psalm tone could follow as a matter of course. Watertight rules for determining modality therefore facilitated the execution of a large proportion of the chant repertory. In these circumstances we can easily understand how Regino of Prüm, at the end of the ninth century, was inspired to put pen to paper by what he describes as widespread confusion in singing caused by the improper selection of psalm tones ('propter dissonantiam toni').
None of these concerns was directly connected with the heritage of classical music theory. But as soon as chant could be thought of in terms of the pitches of the Greek two-octave system, concepts such as species of octave, fifth, and fourth could be employed in definitions of modality. The two developments went hand in hand, as late ninth-century writings bear witness.
456
V. Plainchant and Early Music Theory
(iij Aurelian ofReome
GS i; Gushee 1963 and CSM 21; Ponte 1961, 1968.
The earliest of the ninth-century writers, Aurelian of Reome, shows very clearly how gross a division existed between the heritage of classical music theory and the needs of practical musicians. His Musica disciplina appears to have been written in the decade before 850 for Bernard, abbot of the monastery of Saint-Jean-de-Reome (not far from Langres in eastern France). From the complicated make-up and transmission it is not clear how much Aurelian himself may have contributed to the treatise in its fullest form. (For example, the start in ch. 8, 'De octo tonis', is found separately elswhere, being attributed in a thirteenth-century source, and thereafter by Gerbert, to Alcuin. While that attribution is false, Aurelian may not have written the passage. Other material is borrowed, sometimes without much adaptation, from Boethius, Cassiodorus, and Bede.) The best-known passages in the treatise are those where Aurelian explains that Charlemagne added four modes to the usual eight because the Greeks (Byzantines) boasted that they had perfected the eight modes; whereupon the Greeks added four more themselves; and how Aurelian asked a Greek the meaning of the Koeanne syllables. (This actually constitutes some of our earliest evidence that Byzantine chant had a modal organization.) More important for the present discussion are the succeeding chapters, where he discusses melodies belonging to the (eight) different modes in purely practical terms.
Aurelian's particular concern is the link between main chant and verse in introits, offertories, and communions at mass and responsories and antiphons of the office. The closest thematic parallel is with the Commemoratio brevis, but the manner of presentation is entirely different, for Aurelian disposes of no pitch-notation. He begins, for example, by describing the three cadences for the psalm tone used with mode 1 introits:
The first authentic mode has several varietates. For the introits it contains three varietates, the first of which is this: Ant. Gaudete in domino semper. The end of the verse flows directly into the beginning (of the introit), and equals it, and is neither pushed upwards nor pressed down.
This corresponds to the modern Solesmes/Vatican reading, where the cadence for the psalm verse falls on D, the starting-note for the introit. The other two joins are described thus:
The second is this: Ant. Iustus es domine, whose psalm verse ending is raised up on high to coincide with the beginning (of the introit). The third is this: Ant. Suscepimus Deus, where the last syllable of the psalm verse dwells a little longer than is the case with the first and second (varietates), and is then raised up on high to fit well with the beginning of the introit.
The third cadence is evidently that which we find in the Solesmes/Vatican books, although the delay on F is just as long for Gaudete in domino. For the second introit,
4. The Modes 457
however, the Solesmes/Vatican books have the same cadence as for Gaudete; Aurelian has a different one in mind (rarely found in the modern books: see Salus autem, or the list of introit psalm-verse tones among the 'Common Tones of Mass' in LU). The three cadences and beginnings of the introits are given in Ex. V.4.1.
Ex. V.4.1. Psalm-verse cadences and openings of mode 1 introits cited by Aurelian of Reome
Gau-de - te      in  do - mi-no    sem  - per
P
E   u      o   u    a     e lus - tus es do-mi-ne
... ^ ^   ft*   '  • ft
-'   "* flV-~p—w-
E  u      o   u   a     e Sus - ce - pimus De - us
It is not always clear to us now why this or that cadence should have been selected, and a good deal of variety exists between medieval sources in this respect. Evidently Aurelian had also considered the matter, though his explanation is far from convincing:
Because of this the mind of the eager reader is troubled why one single tone should have so many sorts of varieties (varietatum modos) . . . He who wishes to know everything else about this science, him we send to 'musica' (that is the science of music), and if he wishes to become versed in the latter, let him turn his eyes to the consonance of proportions and the contemplation of intervals, and also to the certitude of numbers.
Despite this pronouncement, Aurelian then proceeds to offer a more down-to-earth explanation. This is done with reference to the tonus of these introits, which seems to correspond to the chief reciting note, a. In Gaudete, he says, the tonus is reached on the third syllable of the introit, '-te'; in Justus it occurs on the first syllable, in Suscepimus on the second.
It has puzzled modern readers that Aurelian should include offertories in his discussion, for the verses of these chants, at least since the time they were first recorded in notation (late ninth century) have been sung to elaborate melodies. Yet Aurelian says: 'Quod versus offertoriarum per tonos in ipsis intromittantur, cantor nemo (est) qui dubitet.' ('No cantor doubts that the verses of the offertories are allowed to pass into the offertories through the the tones.') This has been interpreted to mean that offertory verses are sung to psalm tones. But Aurelian never describes cadence formulas for the offertory verses, as he does for the introits and communions. His remarks resemble those about the office responsories, and concern the point at which the verse returns to the response: does it fit musically and make sense textually?
458 V. Plainchant and Early Music Theory
By tonus in the above passage, therefore, Aurelian may simply have meant the reciting note, or other important note, which had to be borne in mind when effecting the return to the response.
At the end of his treatise Aurelian includes some edifying remarks about the ethical power of music, and relates, among other stories, one about a monk from Saint-Victor, near Le Mans, who made a pilgrimage to the basilica of St Michael on Monte Gargano on the Adriatic coast of Italy. There he was blessed by hearing angels sing the responsory Gives apostolorurn V. Emitte dornine spirit urn. The monk was able to sing this from memory to the 'clerks of the Roman church' ('Romanae Ecclesiae clericis'), who changed the verse to In ornnern terrarn. The way in which Aurelian can move from music's power over birds and dolphins to a legend about the origins of a particular chant parallels the way in which the rest of his treatise juxtaposes borrowings from the musical knowledge of late antiquity with quite original observations on the practice of plainchant. This is a clear indication of the driving force behind much medieval writing on music: the necessity of defining codes of practice, with or without reference to 'musica' in the classical sense.
(iii) Regino ofPrüm
GS i; Huschen 1962; M. P. Le Roux 1965; Chartier 1965; Bower 1971; Bernhard 1979, Studien.
Something of the same dichotomy is present in the treatise of Regino of Prüm. Since the second half of the eighth century, the monastery of Prüm in the Eifel (western Germany) had been something of a royal foundation. A new monastery church is said to have been dedicated in 799 by Pope Leo III in Charlemagne's presence. Emperor Lothar I died there in 855, having assumed the monk's habit six days previously. As many as 300 monks may have lived there at the peak of its ecclesiastical and cultural achievement. But it suffered grievously at the hands of the Northmen in 882 and 892. Its recovery was due to the leadership of Regino, abbot from 882 to 899. Its wealth made it the object of political intrigue, however, and in 899 Counts Gerhard and Matfred of Hainault forced Regino to leave and had their brother Richarius, a mere boy, made abbot in his place. Regino removed to Trier and placed himself under the protection of Archbishop Rathbod. He accompanied the archbishop on visitations, restored the abbey of St Martin, Trier, and died in the abbey of St Maximin in 915.
Regino's writings fall into the Trier period. The Epistola de harmonica institutione is couched on the form of a letter to Rathbod. It is linked to a tonary with the title 'Octo toni de musicae artis cum suis differentiis'. In about 906 he wrote a book of canon law, De ecclesiasticis disciplinis, for use during visitations, dedicated to the primate of the east Frankish church and imperial chancellor Archbishop Hatto of Mainz. In about 908 Regino completed a chronicle of the world, dedicated to Bishop Adalbero of Augsburg, tutor to the royal children. Regino was one of the foremost churchmen of his time, and therefore a quite different figure from the obscure Aurelian.
4. The Modes
459
An assessment of his Epištola de harmonica institutione and its tonary is again complicated by the transmission. Both have come down in a longer and a shorter form. The two principal sources both date from the tenth century, but neither stems directly from Regino. Brussels, Bibliothěque Royale 2750—65 (connected with the Belgian monastery of Stavelot) appears to transmit Regino's tonary in its original state, but has a short version of the Epištola, one which restricts itself to practical musical matters. Leipzig, Universitätsbibliothek, Rep. I. 93 (probably from Prüm) has a revised version of the tonary, and includes in the Epištola several chapters of speculative music theory, which, as Bower (1971) has shown, depend on the ideas of John Scotus.
Regino's tonary is one of the largest to have survived. This painstaking compilation has already been mentioned because of the numerous chants which Regino reckoned to begin in one mode and end in another. The work is of the highest value for our understanding of the modal assignment, but, despite the Epištola, we have to study the tonary itself to discover the principles involved. Regino, like Aurelian before him, does not define what constitutes modal identity. To see how this was attempted in the Middle Ages we should first inspect the terminology developed from the ninth century onward.
(iv) Nomenclature
Hucbald and the Enchiriadis treatises, and Aurelian of Reome before them, were already familiar with a nomenclature for the eight modes. Following the spelling of the Commemoratio brevis, the terms are:
f authentus    .     ,     „       , . protus . (modern U-mode)
[plagis
. f authentus    .     .      r,      . .
deuterus       .   . (modern L-mode)
[plagis
r authentus ,      „      . ,
tntus . (modern r-mode)
[plagis
.       [authentus    .     .      ~       , . tetrardus       ,   . (modern G-mode)
[plagis
The first witness to this terminology in the West is the (incomplete) Saint-Riquier tonary, Paris, Bibliotheque Nationale, lat. 13159, of the late eighth or ninth century. The terms are derived from Byzantine practice, with the difference that the Greeks did not use the word 'authentic': there were four principle modes (echos protos, echos deuteros, etc.) designated a, b, c, d, and four plagal ones, designated plagios a, plagios b, plagios c, and plagios d. By good fortune evidence of the Byzantine nomenclature survives from about a century before its adoption in the West. There is manuscript witness to its use from possibly as early as the seventh century (see Huglo 1975, 140). The proper chants of the office of Saturday evening and Sunday in the Byzantine rite were organized in an eight-week modal cycle, each week having chants in a different mode. Such chants were copied in the book known as the oktoechos.
460 V. Plainchant and Early Music Theory
Cody (1982, 103) has reported a Melkite Syrian witness of the seventh or eighth century to the modal ordering of Sunday chants. Husmann (1971) has disproved the earlier belief that the hymns of Severus of Antioch, d. 538, were arranged in an oktoechos. Medieval tradition attributed the composition (more likely, organization) of the oktoechos to St John of Damascus (d. r.749). Strunk (1960) accepts the authorship of St Theodore Studites and a date of 794-7 for one particular early cycle of antiphons of the oktoechos, which also argues for the establishment of the system in the eighth century, if not earlier.
Huglo (1978, 138-42) has suggested that the coining of the term authentus was caused by the difficulty of finding an apt translation for echos, since modus, tropus, and tonus were already invested with slightly different meanings. The nearest equivalent to echos would have been tonus, which in the ninth century is generally used of the eight modes of plainchant: but there were only four echoi. A solution can be found in Hucbald, who speaks of the four finales (our D, E, E, and G) which make the appropriate endings for 'quatuor modis vel tropis, quos nunc tonos dicunt', and each in their turn reign over 'geminos . . . tropos'. So four 'tones' command eight 'tropes'. But Hucbald's usage is not that of the majority of early treatises and tonaries, which prefer to speak of eight tones: four principal or 'authentic' ones and four subsidiary or 'plagal'.
Another solution is that found in ch. 8 of Musica enchiriadis, where chant incipits in each of the eight modes are presented under the title: 'Modulatio ad principalem protum modum et subiugalem eius' ('Melodic phrases for the principal protus mode and its subordinate'). That is how 'mode' is commonly used nowadays (and in this book). But this is not the only terminology found in Musica enchiriadis, which, as Atkinson (1989, 'Modi) has recently stressed, illustrates vividly the difficulties of marshalling terms borrowed from classical theory. We still have the same sort of contradictions: 'tone' can mean the interval of a tone, any single note, a psalm tone, or even (a meaning with no ancient etymology) quality of sound (harsh, soft, etc.).
Boethius usually used the word tonus to mean the interval of a whole tone. He was aware, however, that both tonus and tropus were used in classical Greek theory (as they were by later writers such as Martianus and Cassiodorus) to mean the various transpositions of the double-octave scale, which, if one took an octave cross-section at any constant point, would display the different octave species. The different transpositions bore the Greek names Hypodorian, Hypophrygian, and so on. Boethius' preferred term for the transposition scales, however, was modus.
It is not surprising, therefore, that ninth-century writers, inheriting a partially conflicting terminology from late antiquity, and also needing to accommodate the new concept of the eight modes of plainchant, should display some inconsistency. In the later part of Musica enchiriadis, principally concerned with vertical intervals, tonus is restricted to the meaning of an interval of a whole tone. Hence the establishment of modus as the term for the plainchant modes. In the first part of his treatise, however, the Enchiriadis author has himself used tonus for plainchant mode, and is obliged to change course in midstream (actually, in ch. 8).
4. The Modes
461
(v) Greek Names and Octave Species
In ch. 9 of Musica enchiriadis occurs one of the earliest applications of the names of Greek peoples to the modes:
Modes or tropes [the uncertainty of terminology is still evident!] are the species of melody [species tnodulatwmtm] which were discussed above, such as the protos autentus or plagis, deuteros autentus or plagis, or the Dorius, Frigius, Lidius mode, etc.
It is the treatise known as Alia musica, however, which gives the first comprehensive exposition of this terminology. The author, or editor, appears to have combined material from two previous writers with his own observations (see Muhlmann 1914, Chailley 1965, and Heard 1966), as the title perhaps indicates: 'Alia musica' may be not Latin for 'another treatise on music' but Greek (halia) for 'a compilation about music' (Heard 1966). As far as modal terminology is concerned, the author simply assigns the Greek names to the eight plainchant modes without regard to their former nature. In Boethius the names denoted transposition scales, through which the seven species of octave could be displayed; no internal hierarchy of notes was implied. The plainchant modes, on the other hand, cannot be equated simply with scales: a number of notes are more prominent than others, and melodies in a particular mode may be related to each other by melodic charateristics.
Table V.4.1 shows how the Greek names were reordered to provide the medieval modal names. The eighth mode, the Hypermixolydian (literally 'beyond the Mixolydian', rather than Hypomixolydian, 'below the Mixolydian') is derived from Ptolemy, as reported by Boethius. Ptolemy did indeed advocate an eighth transposition scale an octave above the first, hence higher than all the others ('Hyper-'). Writers on the medieval plainchant modes simply changed the name to conform with the other plagal modes. Referring to the pitch-letters derived from Boethius' divisions of the monochord, the author of Alia musica justified the distinction between Dorian and Hypermixolydian by invoking the different 'median note' of each mode {mediarn chordam) which is 'guardian of its quality' (qualitatis custodem). For the Hypermixolydian this is M (modern G), for the Dorian O (modern a).
Not until the humanist rediscovery of classical texts in the sixteenth century were attempts made to restore the Greek names to their original octave species.
In view of the borrowing of the Greek names for octave species in Alia musica, it is not surprising that one of the authors should also have taken the radical step of specifying octave ranges for the plainchant modes. One of his successors in the same treatise specified the same ranges split into a fourth plus a fifth, positing the idea of a median tone where the division is to be found (hence the distinction between mode 1 and mode 8). The plagal modes have the modal final as their median note, that is a fourth above the start of the range, whereas the authentic modes have the fifth as their median. The nomenclature modus is retained for octave species, while the plainchant mode is tropus.
Range is indeed a major preoccupation in the treatise. Even the ranges of the
462 V. Plainchant and Early Music Theory
Table V.4.1. Greek and medieval modal names
I......'
Boethius
Phrygian
.4//o musica
Dorian
Hypodorian Ilypodorian
Tonaries, Hucbald, Aurelian, etc.
Protus
Authentus
Plagalis
1...«
Dorian Phrygian Mixolydian Hypophrygian
f
Deuterus
I
Authentus
Plagalis
i......
Hypolydian Lydian
Lydia n
Hypolydian
Hypophrygian Mixolydian
Uypermixolydian
Tritus
Authentus
Plagalis
\
Tetrardus
I
Authentus
Plagalis
I = final of plainchant mode
Noeanne formulas for each mode are stated, although these by no means correspond with the scale-segments defined elsewhere. One senses here a certain tension between melodic pattern (the Noeanne melodies) and abstract concept (scale-segments) in the definition mode.
There is a trace of the same dichotomy in the way in which the chant examples in Musica enchiriadis are presented. Each is prefaced, as we have seen, with a phrase which descends through the fifth above the four finals in turn. Although the author
4. The Modes
463
does not say so, these are in fact the four possible species of fifth (cf. Ex. V.2.2 and Ex. V.3.5).
(vi) Italian Theory: The Dialogus de musica
GS i; Strunk 1950; Oesch 1954; Huglo 1969, 1971 'Prolog'.
The first efflorescence of music theory in the Carolingian century took place primarily in north-east France and the Moselle area, in the late ninth century. From the next few decades less material has come down to us, until at the end of the tenth century and in the early eleventh two new groups of writings appear, from Italian and south German writers respectively.
The Italian texts are characterized by a strongly practical purpose, and are evidently motivated by the need to train the ability to relate heard and sung melodies to written pitches. In some respects they are thus the successors to Hucbald's writings.
The earliest of these texts is an anonymous treatise written in dialogue form probably in Lombardy in the closing years of the tenth century. It was for a long time attributed to the famous abbot Odo of Cluny (d. 942), former pupil of Remigius of Auxerre, who is known to have composed a number of office chants, but who cannot be shown to have left any writings on music theory. The grounds for the attribution are simply that the Dialogus mentions the modal corrections made to the antiphon 0 beatum pontificem by a certain abbot Odo. This is naturally of no consequence for the authorship of the Dialogus itself, but medieval copies, and later Gerbert, placed Odo's name at the head of the treatise and assumed Odo of Cluny to be the abbot in question. The interrelationship of the writings involved in this complex of material has been clarified by Huglo (1969, 1971, 'Prolog'; see also 1971, Tonaires, 183-224 and 'Odo', NG):
(i) Prologue 'Formulas quas vobis' (GS i. 248) and tonary (CS ii. 117, from a source rather distant from the archetype); the prologue was written by a Benedictine abbot, the tonary agrees with the liturgical use of Arezzo, so Huglo attributes them to 'abbot Odo of Arezzo', about whom, however, nothing more is known.
(ii) Prologue 'Petistis obnixe' (GS i. 251, Huglo 1971, 'Prolog'); probably conceived as prologue for an antiphoner in alphabetic notation, later used as an introduction to (iii); composed by a monk, possibly Camaldolese or Vallombrosan.
(iii) The dialogue treatise (GS i. 252), in which the Odo of (i) is mentioned.
The first chapters of the Dialogus describes the monochord. The crucial remark is that one may mark up an antiphon in a chant-book with the same alphabetic letters that one uses for finding the notes on the monochord. The author's division of the monochord is given in Ex. V.4.2 (cf. that of Boethius in Ex. V.2.1 above).
Subsequent chapters emphasize the difference between whole tone and semitone, and describe the perfect consonances fourth, fifth, and octave. The intervals which
464
V. Plain chant and Early Music Theory
Ex. V.4.2. Division of the monochord in Dialogus de mnsica
i
			
	T   A  B   C   D E	F	G   a   b   b c
	1    2   3    4   5 6	7	8   9   17 10 11 1
1.	Take a length of string TX		
2.	divide TX by 9:	A	TX:AX - 9:8
3.	divide AX by 9:	B	AX :BX = 9:8
4.	divide LX by 4:	C	LX:CX = 4:3
5.	divide AX by 4:	D	AX:CX = 4:3
6.	divide BX by 4:	E	BX:EX = 4:3
7.	divide CX by 4:	F	CX:FX = 4:3
8.	divide TX by 2:	G	TX:GX = 2:1
9.	divide AX by 2:	a	AX:aX - 2:1
10.	divide BX by 2:	b	BX:bX = 2:1
11.	divide CX by 2:	c	CX:cX = 2:1
12.	divide DX by 2:	d	DX:dX = 2:1
13.	divide EX by 2:	e	EX:eX = 2:1
14.	divide FX by 2:	f	FX:fX = 2:1
15.	divide GX by 2:	g	GX:gX = 2:1
16.	divide aX by 2:	a a	aX:aX = 2:1
17.	divide FX by 4:	b	FX:bX = 4:3
f g
occur in chant melodies are then illustrated with chant examples, very much as Hucbald had done, though neither the tritone nor any sixth is admitted here.
It is at this point (ch. 6) that the author discusses two problematic antiphons. O beatum pontificem (AS 586) lies mostly in the middle of the D-mode range and resembles mode 2 antiphons like the O-antiphons of Advent; because of two high-lying phrases it must, however, be assigned to mode 1 (cf. prologue to Odo's tonary, GS i. 249, also CS ii. 121). The problem is relatively minor, and is perhaps mentioned only because of the danger of attaching undue importance to the opening alone in deciding modality. The other antiphon, Do/nine qui operati, is more difficult, for it seems to call for both bt\ and bb. It begins like many mode 6 antiphons (cf. O admirabile commercium), and we might therefore expect it to be notated with final on F. But at a later point, Eb would then be required, an impossibility in the notation of the tenth to eleventh century. An answer to the problem would be to notate the antiphon a fifth higher, ending on c, so that both b\ and bb could be used, as is done, for instance, in the Worcester antiphoner (PalMus 12). Another solution would be to rewrite the passages with Eblbb, as is done in the Lucca antiphoner (PalMus 9). The author of the Dialogus recommends a different procedure: the opening should be altered to resemble mode 8 antiphons such as Amen [amen] dico vobis (AM 489). This would eliminate the semitone step under the final. The antiphon would then
4. The Modes
465
presumably be notated in G. Strunk (1950, 112) gives the various openings involved (Odo's preferred opening should be emended, however). The scale-segments involved are given in Ex. V.4.3.
Ex. V.4.3. Notes required to notate antiphon Domine qui operati, various versions
Worcester, final c
the same with final F
altered opening, final G
Lucca, final c
Chapter 8 begins with the well-known definition of 'mode' quoted at the beginning of this section: the final determines the mode. Not only this, it also governs starting-note and range. The rest of the chapter amplifies this—one recollects the examples at the end of Hucbald's treatise—by explaining that the relationship of starting-note to final should be that of the six intervals enumerated previously (minor and major second, minor and major third, fourth, fifth) or the unison. (The only exception is that mode 3 melodies often begin on c, a minor sixth above the final E.) Individual phrases within the chant (distinctiones) had best end on the final, which leads to the rule: 'A chant belongs most to that mode towards which its distinctions most frequently lead.' Chapter 9 defines the range of chants in the various modes (Ex. V.4.4). Eight chapters, one for each mode, provide examples to illustrate the above prescriptions.
Ex. V.4.4. Ranges of chants in the eight modes, Dialogus de music a Mode 1 Mode 2
—Tiy-
■v
Mode 3 Mode 4
466
\. Plainchant and Early Music Theory
It is clear from the remarks in ch. 6 that the author is perfectly ready to 'improve' chants whose modality is uncertain, and it is not surprising to read in the last chapter of the dialogue that chants which do not obey the rules have been composed by 'presumptuous and corrupt cantors' ('furtivam singularitatem a presumptoribus vitiatisque cantoribus factam esse non dubito'.) It is in precisely this climate that the revisions of the chant repertory undertaken by the Cistercians should be understood.
The last part of the dialogue develops the interesting idea that by virtue of the intervals on either side of it, each note has a certain modal quality. This follows from the concept that the four finals, D, E, F, and G, define their respective modes in various ways. Notes which occupy similar scale-segments to one of the four finals therefore share their modal character. For example below A there is a tone, above it there is a tone, a semitone and two whole tones; that is also true of D. So A 'preserves the rule of the first mode: therefore not without reason is it called first'. The 'similitudes' between notes and modes which the author enumerates are given in Ex. V.4.5.
Ex. V.4.5. 'Similitudes' of notes to the various modes, Dialogus de musua
(vii) Guido of Arezzo
GS ii; CSM 4; Babb 1978; DMA A.III, IV; Strunk 1950; Ocsch 1954; Smits van Waesberghe 1951, 'Guido', 1953, 1957, Kxpositiones, 1974; Waeltner and Bernhard 1976.
Guido of Arezzo was intimately acquainted with the Dialogus and even recommends its use on some occasions. But his own teaching goes well beyond it in scope and clarity. Guido was born probably around 990, and was educated in the Benedictine monastery at Pomposa (in the Po delta north of Ravenna), but became unpopular, apparently as a direct result of his music teaching. It was at Pomposa that he invented his system of staff-notation (see above, IV.6). His Aliae regulae, conceived as prologue to an antiphoner with the new notation, describes the system. Guido moved to Arezzo around 1025; his most important treatise, Micrologus, completed around 1030, was dedicated to Bishop Theodald of Arezzo. Pope John XIX (1024-33) called Guido to Rome to show him the new antiphoner, as Guido reports in his Epištola ad Michaelem or Epištola de ignoto cantu. His whereabouts in later years is unknown.
Guido's new notation has already been described. As so often, it is the lack of agreement between singers that has caused him to invent the system. Each cantor makes his own antiphoner, he complains, so that instead of the 'Antiphonarium Gregorii' the talk is of Leo's or Albert's antiphoner. Guido sees his system not simply
4. The Modes
467
as a better method than the alphabetic way of notating chant. It has implications for the whole way in which chant can be grasped as a tonal construct. This is made clear above all in the Epistola.
The Epistola introduces the hymn Ut queant laxis. The melody Guido uses may well have been composed by him for the purpose in hand, for it is unknown elsewhere at this time. Its well-known peculiarity is that each line begins with a successively higher note of a hexachordal series C-D-E-F-G—a (Ex. V.4.6). The reason why the learner should fix the hexachord and its six syllables in the mind is that he will have a point of reference for any chant. Guido says that learning melodies by rote from a master (the traditional method) or by picking them out on the monochord (by means of alphabetic notation, the method of the Dialogus) is adequate for beginners but bad for more advanced students. His new learning method reduces the time for learning the chant repertory to two years (previously ten were required). A boy can learn a new chant in three days. This is because the new system helps one to read music previously unknown; and it helps one to notate a melody one knows. The point of the exercise is summed up as follows: 'Ergo ut inauditos cantus, mox ut descriptos videris, competenter enunties, aut indescriptos audiens cito describendos bene possis discernere, optime te iuvabit haec regula.' ('Therefore this rule will be of greatest help to you so that you may sing competently an unknown chant as soon as you see it written, or so that you may be able to perceive straightaway how to write down an unwritten chant that you hear.') The singer is instructed to compare the ends of phrases, in whatever chant is being dealt with, with the notes of the hexachord. This is so that he will be able to gain a proper tonal orientation. The advantages of using a single hexachord as point of reference are many. The hexachord encompasses the cadential formulas of most chants, certainly in the simpler antiphons which were presumably taught first in Guido's song-school; one hexachord is better than four species of fifth or seven species of octave.
Ex. V.4.6. Hymn Ut queant /axis, Guido of Arczzo
P
-*-"-9*-•-*-*   ^   *-'-*-*"-»    •    V   *- - *• »
UT queant  la - xis       RE-so-na-re fi-bris       Ml-ra ges-to-rum     FA-muli tu-o-rum
'*»     M,      M      * '     •     '     M ^
- *- -m-
SOL-ue pol-lu-ti        LA-bi-i   re-a-tum     sancte Io-an-nes.
Perhaps we could imagine a choirboy learning the antiphon shown in Ex. V.4.7. He can see the red F-line, knows that it marks the semitone step, mi-fa on his hexachord. He can quickly locate the final of the chant in his hexachord, which means he can hear it in his inner ear, or sing it out loud, with the correct constellation of tones and semitones on either side. He can then go on to locate the finals of each phrase, indeed all the notes of the piece, with relative ease. This antiphon is of course a very simple
468
V. Plainchant and Early Music Theory
Ex.V.4.7. Antiphon Diligite Dominum (Lucca, Bibl. Cap. 601, PalMus 9, p. 456)
Di - li - gi - te do-minum    omnes sancti     e - ius +      quo - ni -am ue - ri - ta-tem
re - qui ret    do - mi - nus.
example, but shortly the student will be used to making the connection between clefs, coloured lines, and particular pitches, and be able to read at sight from the stave, using the hexachord only occasionally for reference. This would have been far quicker than learning by rote from a teacher, or from notes picked out on the monochord. (Although he might have attained fluency in reading alphabetic notation at sight, that would not have taken a shorter time than reading staff-notation.) The reverse process, notating Diligite dominum when it was already known from memory, could proceed equally quickly through the intermediary of the hexachord. The first phrase would be matched to the syllables re-re--ut—re--fa-mi—re, which would then be translated into D-D-C-D-F-E-D in staff-notation.
Guido does not mention the steps described above in every detail. I have tried to imagine how it was done. Nor does he say how this learning method was applied to chants with a wide range. Soon after Guido hexachords were imagined on other notes of the scale, on F (with bb, or round b, hence 'hexachordum molle') and on G (with bfc), or square b, hence 'hexachordum durum'); the hexachord on C was the 'hexachordum naturale'. One other pedagogical device subsequently attributed to Guido is not mentioned anywhere in his writings: the 'Guidonian hand', where each note of the scale has a position on the fingers of the (left) hand, with its correspoding hexachord syllable. The earliest source so far known, Montecassino, Archivio della Badia 318, of the late eleventh century, displays the range T-a, without hexachord syllables, but with the letters T and S indicating the intervals between pitches. The earliest hands with hexachord syllables seem to be somewhat later. Fig. V.4.1 shows the hand in a Bavarian manuscript of the twelfth century, Munich, Bayerische Staatsbibliothek elm 14965b, and the series of hexachords indicated on it. The hand also indicates whole tone and semitone steps, and against certain of the letters are given the modes in which those pitches are the dominant (reciting) note (circled on my copy for ease of identification). The direction of reading is given in a smaller accompanying diagram. (See Smits van Waesberghe 1969, 129 for facsimile, 120-43 for hands of all sorts; also Russell 1981.)
Both the Regulae and the Micrologus assign modal qualities to the notes of the scale, as the Dialogus had done (Guido's term for the correspondences between notes, the similitudo of the Dialogus, is affinitas). The longer discussion in the Micrologus (ch. 7) introduces the concept of four modi vocum (literally, 'the modes of the notes',
4. The Modes 469
Fig. V.4.1. Guidonian hand after Munich, Bayerische Staatsbibliothek, elm 14965b
perhaps better, 'the modal qualities of the notes'). This describes quite simply the constellation of tones and semitones around each note of the scales. Significantly, each note except G is placed in a hexachord, a clear link with the hexachord and its ut—re-mi syllables in the (later) Epistola. Guido enumerates four patterns (see Ex. V.4.8). The intention is obvious: if we can locate a note in one of these scale-segments, then we can notate it, or sing it, as the case requires.
The Micrologus also treats other topics covered by the Dialogus: the division of the monochord, the intervals found between notes in plainchant (the same as those enumerated in the Dialogus), the names of the modes, their range, the importance of
470 V. Plainchant and Early Music Theory
Ex. V.4.8. The modi vocum, Guido of Arezzo
the final, and the 'correction' of wrong melodies (ch. 10: 'De modis, et falsi meli agnitione et correctione'). Two chapters concern organum, and two chapters, on the ethos of the modes and on Pythagoras' discovery of music (!), constitute rare references back to classical theory (there were none in the Dialogus). Guido's comparison of the elements of music with the parts of language has already been mentioned (IV. 1). Although its roots lie in a tradition of conceiving music in grammatical terms (see Phillips and Huglo 1985, Bower 1989), these chapters (15-16) develop in an original way, with numerous observations on the relationships between musical phrases. Chapter 17 describes the way in which a melody may be composed mechanically by assigning notes according to the vowel in each syllable of the text: let the vowel 'a' be given the note C, 'e' be given D, 'i' E, and so on; several different arrangements are suggested. Guido nevertheless departs from the method at the end of the chant in order to make a satisfactory cadence (see plate 11).
As far as the modality of chant is concerned, Guido's main concern is not an abstract definition. His two primary aims seem to be: (i) to enhance our understanding of chant in ways that will facilitate the reading of notation and, conversely, the notating of chants; and (ii) to provide guidelines for the composition of new chants and the correction of poor ones.
(viii) South German Writers
In bulk the Italian theoretical literature of the tenth to eleventh centuries is easily overshadowed by that written in south Germany. From France rather little new writing is known, save for the treatise by Odorannus of Sens (985-1046: see Bautier et al. 1972, also Villetard 1912, 1956). Mention should also be made of two letters commenting on Boethius' De inslitutione musica, and a treatise on the measurement of organ-pipes and the division of the monochord, by Gerbert of Aurillac (c.940-1003), later Pope Sylvester II (see Sachs 1972 and Huglo, 'Gerbert d'Aurillac', NG). The generally meagre quantity of material does not mean, however, that French musicians no longer studied music theory. Proof that they did can be found in manuscripts like Munich, Bayerische Staatsbibliothek, elm 14272. This was mostly copied at Chartres by the monk Hartvic of the Benedictine abbey of St Emmeram in Regensburg, and contains, among other things, Boethius' De institutione musica,
4. The Modes
471
glossed partly by Fulbert of Chartres (bishop 1007-28), a tonary, writings on the trivium, the three main treatises of the Enchiriadis group, a monochord tract, and the Alia musica (see RISM B/III/3, 110). In Regensburg it contributed significantly to the studies in music theory which flourished there. Yet, on the whole, works of the originality of those mentioned here do not seem to have been written in France, or have been lost. Smits van Waesberghe repeatedly asserted the existence of a school of music theory in Liege and neighbouring areas, related to the south German tradition, though the true picture remains somewhat unclear. The importance of the south German area can be judged from Table V.4.2, a list of the more important writers and works of the eleventh century.
There is clearly no space here to discuss all members of this colourful group. It contains several very eminent churchmen: Berno, monk of Prüm, called in 1008 by the Emperor Henry II to be abbot of the famous abbey on the Reichenau in Lake Constance; William of Hirsau, monk of St Emmeram in Regensburg, from 1069 abbot of Hirsau in the Black Forest, fountainhead of a great monastic reforming movement, the 'German Cluny'; Theogerus, monk of Hirsau, appointed in 1088 by William to be abbot of St Georgen in the Black Forest, elected bishop of Metz in 1117, though unable to occupy the see; Honorius of Augsburg, perhaps the most widely read popular theologian of his time. Other figures remain somewhat mysterious. Johannes has been identified implausibly as an English monk, or as abbot of the abbey of Afflighem near Brussels; he was almost certainly Bavarian, and possibly wrote his treatise for Bishop Engelbrecht of Passau (1045-60: see Hohler 1980, 57-8). Of Aribo nothing certain is known.
The contributions naturally vary widely in scope and interest. The material connected with St Emmeram in Regensburg exemplifies this (it has been elucidated principally by Bischoff). Otker is known only as the author of a diagram called the Quadripartita figura (see Bronarski 1926, also Smits van Waesberghe 1969, 90), apparently for identifying on the monochord the species of fourth and fifth, and the 'chordae constitutivae' of each mode, that is, the notes whose modal character identifies them with each mode, a concept perhaps borrowed by Aribo in a diagram of his own, nicknamed the 'Caprea'. Otloh's musical activities are overshadowed by his many other interests, but he appears as William of Hirsau's partner in two dialogue treatises, on music and on astronomy; he may be author of a Kyrie trope; and he copied two music theory manuscripts, Munich, Bayerische Staatsbibliothek, elm 14523 (which includes all Guido's writings) and elm 18937. William of Hirsau's connection with St Emmeram has already been mentioned. Henricus/Honorius of Augsburg also spent several years as monk in Regensburg. The presence of the Enchiňadis treatises at St Emmeram, in a source copied at Chartres, and Otloh's copy of Guido's writings, show how cosmopolitan might be the cultural links of an important Benedictine monastery in its heyday.
472 V. Plainchant and Early Music Theory
Table V.4.2. Writers on music theory in south Germany in the eleventh century
Reichenau:
Pseudo-Bernelinus (before Berno)
Musica (GS i) Berno (d. 1048) (see Oesch 1961)
Prologus in ton avium (GS ii)
Tonatius
De consona tonorum diversitate (DMA A.VI6) De mensura monochordi (?) (DMA A.VI«) Hermannus Contractus (1012-54) (see Oesch 1961) De musica (GS ii, Ellinwood 1936)
Explicatio litterarum et signorum (GS ii, Ellinwood 1936) Versus ad discemendum cantuurn (GS ii, Ellinwood 1936)
Bavarian and related centres: Otker of St Emmeram
Mensura quadripartitae figurae (c. 1050?) (Mettenleiter 1865; Bronarski 1926) Otloh of St Emmeram (1032-70, older contemporary of William of Hirsau) (Bischoff 1955) William of Hirsau (d. 1091)
Musica (1070-80) (GS ii; Müller 1883; CSM 23) Johannes 'Cotto' or 'of Afflighem' (from Passau? c. 1060?)
De musica cum tonatio (GS ii; CSM 1; Babb 1978) Aribo (of Freising?)
De musica (1068-78) (GS ii; CSM 2) Anon.
Commentarius in Micrologum (c. 1075?) (Smits van Waesberghe 1957, Expositiones) Frutolf of Michelsberg (Bamberg) (d. 1103)
Breviariurn de musica et tonatius (Vivell 1919) Henricus/Honorius of Augsburg ('Augustodunensis', c. 1100)
Dialogus de musica (DMA A.VII; Huglo 1967, 'Henri'; Flint 1982) Theogerus of Metz (c. 1050-1120, pupil of William of Hirsau at Hirsau)
Musica (GS ii)
Liege?:
'Wolf-Anonymous' (c. 1060)
De musica (Wolf 1893) Franciscus of Liege (1047-83) or Rudolf of Saint-Trond (c. 1070-1132)
Ouaestiones in musica (Steglich 1911)
(ix) Modal Theory in South Gertnany
Guido's influence eventually became very strong in south Germany, as elsewhere, but Berno was writing before it had effect. Nor does Berno use any pitch-notation, where necessary referring instead to the Greek pitch-names, proslambanomenos and so on. There is no explicit connection with sight-singing or notating music. (All passages
4. The Modes
473
referring to letter-notation in Gerbert's edition are interpolations, principally from Gerbert's Anonymous I—GS i. 330: see Oesch 1961, 84-90; on the other hand, Smits van Waesberghe in DMA A.VI referred to this author as Berno I.) The use of the monochord is envisaged, as embodiment of the classical scale system, for example when Berno recommends that certain antiphons be transposed up a fifth, for otherwise they will be impossible to locate on the monochord ('Nisi enim hae antiphonae, Alias oves habeo, Domine qui operati sunt [mentioned in the Dialogus de musical, quae sexti toni sunt, in quintum transponantur locum, hoc est, a parypate meson in trite diezeugmenon, nequaquam in regulari monochordo servare poterunt ordinem suum.')
Berno's principal treatise, conceived as prologue to a tonary, is, according to the author himself, largely a summary of older rules, as is made clear by frequent citations or paraphrases of Hucbald and, to a lesser extent, Pseudo-Bernelinus (GS i. 313). Apart from ch. 4, a speculation on the significance of the numbers 1, 2, 3, 4, and their sum 10, it stays within the confines of basic rudiments, as a brief account of his discussion of modality will make clear. From Pseudo-Bernelinus comes a description of the modes in terms of scale-segments, in fact as species of fourth plus species of fifth, just as in Guido's exposition (see Ex. V.4.8).
There are obvious similarities with the Alia musica writings. Berno (GS ii. 69), in the words of Pseudo-Bernelius, says that the modes 'consist of these tetra- and pentachords ('Protus constat ex prima specie diapente . . .'). His ninth chapter nevertheless sets out guide-lines for range which are more flexible than the octaves just outlined. We should therefore understand the scale-segments as loci within which the chant will typically move. After dealing with chants whose modality is for one reason or another unclear, Berno, like his Italian contemporaries, discusses the relationship between the final and the beginnings and endings of the component phrases of a chant. He also speaks of the socialitatis foedus which exists between the finals and notes a fifth higher, and sometimes a fourth higher. A host of examples are adduced to show how some chants are best imagined as being placed in a different part of the two-octave system, to take advantage of the sociales, rather than ending on the normal finals. The communion De fructu operum tuonim is an example of this (discussed exhaustively by Jacobsthal 1897, 52-62, 136-78, and by Bomm 1929, 60-2, 97-103). Berno regards it as a mode 8 chant, but says that if one starts on the final (our G, his lichanos meson) a problem involving the semitone arises in the middle of the chant; for this reason the chant should be assigned to the position a fourth higher, our c.
The communion as it is found in the modern Solesmes/Vatican books has our modern major tonality. If one wanted to notate such a piece in the Middle Ages ending with one of the four regular finals, it would have to be placed on F and use bb, in other words like our modern F major. The chant lies in the lower register, and is thus assigned to mode 6 (F plagal) in modern books. The reason for Berno's remarks can be seen if the version in the Dijon tonary, Montpellier, Faculte de Medecine H. 159, is inspected, for here the chant is indeed notated with final on c, and makes
474 V*. Plainchant and Early Music Theory
use of both b\\ and bb in alternating phrases. The Montpellier manuscript actually assigns the communion to mode 6, whereas for Berno it is a mode 8 piece. Whether one regards the piece as F-mode (with hypothetical Ft] and Eb) or G-mode (hypothetical Ft] and F#) appears to be a matter of taste, normally to be decided with reference to the ending: is the note under the final in the last phrase a whole tone (G-mode) or a semitone (F-mode)? (Although by this reasoning Montpellier should have chosen mode 8, it nevertheless assigns the piece to mode 6: is this because in Guido's theory of affinities G has no co-final?) The scale-segments used in the Vatican/ Solesmes version and in Montpellier H. 159 are given in Ex. V.4.9. (The Cistercians had a different solution. De fructu has at least one D cadence, and its opening resembles D-mode communions such as Posuemnt mortalia. The Cistercian revision eliminated the passages with the Eb/bb step, and strengthened the D-mode element, ending the piece on D and creating a mode 1 communion.)
Ex. V.4.9. Notes required to notate communion De fructu, two versions
A Vatican/Solesmes_
$       • " ' ' ^
Montpellier H.159 m
Between B erno and the younger Reichenau monk Hermannus Contractus ('Herman the Cripple'—see the moving biographical sketch by his pupil Berthold, MGH (Scriptores) Annales et chronica aevi Salici, 267, PL 143, 25, Ger. trans. Oesch 1961, 117) came the impact of Guido's writings. That is not to say that Hermannus' treatise is in any way like, say, the Micrologus in character. (For a German work of that nature one must turn to Johannes 'Cotton'.) But Hermannus uses pitch-letters, and some of his topics reflect ideas in Guido's writings. Of prime interest is Hermannus' response to Guido's hexachordal modi vocum, those scale-segments which are decisive for understanding the intervallic relation of the final to the surrounding notes. Hermannus' equivalent are the sedes troporum, or 'seats of the modes'. These are determined by taking each of the four tetrachords which make up the diatonic scale (to which William of Htrsau added the synemmenon tetrachord), and adding a tone on either side (see Ex. V.4.10).
Another difference from Guido is Hermannus' insistence on a dual modal function for D (and d). For the author of the Dialogus and Guido D was plainly protus, and only A had the same disposition of whole tones and semitones on either side. But Hermannus assigns modal quality on the basis of the position of the note in the four tetrachords. Thus .1, D, a, and d have protus quality, P>, F, b, and e have deuterus quality, and so on. D, which appears as first note of the finales and fourth note of the graves, therefore has dual quality.
4. The Modes
475
Ex. V.4.10. Modal qualities of notes, Ilermannus Contractus
P
modal quality
protus   deuterus   tritus tetrardus
graves
tinales
'*                           ^uno no tpic				■
		a		
	W			
				
excellentes
-Q-3yncmmenon-				-0-
		bo	a	
t>T\	a			
v \i m				
It is therefore not surprising to find Hermannus asserting that the tetrachords A-D and D-G are different species of fourth. The A-D tetrachord starts with the protus note of the graves and ends with the protus note of the finales, whereas the D-G tetrachord starts with the tetrardus note of the graves and ends with the tetrardus note of the finales. The tetrachords have quite different modal quality and thus constitute the first and fourth species of fourth respectively. By a similar process of matching a protus step with another protus step, Hermannus defines four species of fifth and four species of octave. The fifths run between the finales tetrachord and the superiores (D-a, F-b, F-c, G-d), the octaves between finales and excellentes (D-d, K-e, F-f, G-g).
Although this reasoning has a certain logic, it is clearly at variance with classical theory. And while previous deviations from classical theory—such as, for example, Hucbald's redefinition of the tetrachords—usually had a practical purpose, Hermannus' ideas are not obviously the product of experience in learning and teaching chant. The examples of chants in the different modes which he casually cites near the end of his treatise have no essential connection with what has preceded: 'It also helps greatly in recognizing the quality of the modes (troporum qualitatem) if you meditate assiduously upon the customary cadences and diffinitiones, many versicles in their modes, and extended chants most of all . . .'. The balance struck between abstract theory and practical application which characterizes the writings of, say, Hucbald and Berno does not obtain in Hermannus' treatise. His theorizing is nevertheless of an original kind, not a reversion to the abundant theoretical knowledge to be found in Boethius.
476
v. Plainchant and Early Music Theory
(x) Later Syntheses
During the thirteenth century polyphonic music assumed an importance in the work of progressive theorists at least equal to that of plainchant. It is true that some writers continued to interest themselves in plainchant only, at least judging by the writings that have come down to us: Engelbert of Admont (d. 1331; GS ii. 287-369) is one such. On the other hand we have many short texts, and some complete treatises, which are almost entirely preoccupied with polyphony, for example, the well-known treatise of Anonymous IV (CS i. 327-65; Reckow 1967). But for those writers who attempted to cover theoretical aspects of all branches of practical music, plainchant and polyphony are of equal interest. The transition is unmistakable. Polyphony occupies two of the twenty chapters in Guido's Micrologus (r.1030). In the twelfth century, polyphony is discussed only in the last two of twenty-nine chapters of the Schneider/London/Naples anonymous edited by Pannain (1920), and in only two of the eighteen by the so-called 'St Martial' or 'De La Fage' anonymous (ed. Seay 1957; see Sarah Fuller 1977, who shows that it is dependent on Cistercian plainchant teaching, for example in its using the term maneriae for the four modes D, E, F, and G). Against this we can set the twin treatises by Johannes de Garlandia of about 1240, De plana musica and De mensurabili tnusica (see Baltzer, 'Johannes de Garlandia', NG). From the early fourteenth century we have the Lucidarium (concerning plainchant; ed. Herlinger 1985) and the Pomenum (concerning polyphony) of Marchettus of Padua; and of the seven books of Jacobus of Liege's massive Speculum musicae, Book 6 concerns plainchant, Book 7 polyphony (CSM 3). (The above-mentioned writers naturally form a mere selection. Much work still remains to be done even on some of the better-known named authors of chant theory; for anonymous ones see the bibliographical information in Palisca, 'Anonymous theoretical writings', NG.)
Whatever the intellectual attractions of Hermannus' 'seats of the modes', the practical advantages of Guido's hexachords ensured them universal popularity. By the thirteenth century it was customary to identify pitches not only by their letter but also by their solmization syllables: thus middle c, which falls in three hexachords, is c sol fa ut: no other pitch has this combination of syllables. The choice between syllables made possible a change from one hexachord to another, a process known as mutation (mutatio). If a chant went beyond the range of a single hexachord, it would have to be assigned syllables ('solmized') from an adjacent hexachord. As Johannes de Garlandia explains:
One undertakes [a mutation] because of ascent or descent, as for example in Cfa ut, because, if for this note one were to take fa, one could ascend as far as the third note. If one wished to reach the fourth note, however, one would have to take ut in this same Cfa ut, which is a mutation of fa into ut.
The practical application of Guido's modi vocum and of Hermannus' 'seats of the modes' (although Hermannus does not explain it) is that chants with chromatic notes
4. The Modes
All
can be envisaged and notated at a pitch complementary to the usual one, ending with one of the alternative finals (the 'affinals' in Guido's terminology, usually now 'co-' or 'confinals'). But since both the 'modi vocum' and the 'sedes troporum' were conceived as hexachords, it is clear that the solmization syllables will fit them exactly. Whether the chant closes on a co-final or not, it will be solmized in the same way. So Jacobus of Liege can quite simply say: 'chants which are ended in re are of the first and second mode . . . chants which are ended in mi are of the third and fourth mode . . . chants ending in fa are of the fifth and sixth mode . . . chants which are ended in sol are of the seventh and eighth mode . . .' (6. 217). The next chapter (Book 6, ch. 76) deals with chants ending on ut and la. Those ending on ut are to be assigned to mode 5 or 6 if they end on C, F, or c, because of the coincidence and affinity with fa at those pitches. (There is a coincidence with sol on the higher c, but no affinity, because of the whole tone step beneath sol.) If ut as final falls on T, G, or g, then the mode is 7 or 8. If a chant ends on la, it may be assigned to modes 1 to 4: if it ends on a or I, there is both coincidence and affinity with both re and mi, so modes 1 to 4 are all possible; if it ends on E then mi is the only alternative solmization, and the chant should be assigned to mode 3 or 4.
Jacobus can be seen here working out with great thoroughness a technique of modal identification which was implicit in Guido's teaching from the beginning, even if not explained in Guido's own writings. The essence of the technique is the use of an invariable scale-segment—the hexachord—for tonal orientation. It is not basically different from the technique of using the Noeanne formulas to fix in the mind the modality of a chant. The fact that it was an unvarying, abstract pattern made it applicable, through transposition, to any situation likely to arise in the chant repertory, and it was keyed securely into the system of staff-notation. Its success is shown by the fact that even in the much more complicated tonal designs of fourteenth- and fifteenth-century polyphonic music it continued to provide a means of tonal orientation.
VI
Plainchant up to the Eighth Century
VI.1. INTRODUCTION
The rest of this book is a survey which aims to provide a historical background to the previous parts. It is necessarily sketchy, both because of the nature of the evidence and because of the compression of material involved. Particularly for the early centuries, before music was notated, great care is required in considering the nature of whatever plainchant might have been sung. The issues are unusually complicated, much often depending on the interpretation of very brief references, or even single words. There is no space here for a detailed examination, and consequently no room for more than a rough indication of the directions in which recent scholarship has been moving.
The evidence from the centuries before musical notation was developed is perforce more useful for the history of liturgy than of music. While we can learn much about the liturgical function of particular chants, we must continue to speculate about their musical character. This period will be treated only lightly. A special section is nevertheless given to the role of Gregory the Great (d. 604), whose name, more than any other, is indissolubly associated with plainchant.
An important way of filling gaps in our knowledge of early forms of liturgy in the various churches of Christendom has been to compare them with each other and note po:nts of similarity. What is common to them is often the result of a common ancestry. (The classic study of this type is Baumstark 1958.) In a similar fashion— though to a far lesser degree—the veil covering music before the ninth century can be pierced by comparative study of parallel chant repertories. In theory such a study could include all known Christian chant repertories. No attempt is made in this direction, however, beyond a listing of those repertories in VI.2. Later in the book (VIII) the chant which we know as 'Gregorian', the subject of Chapter II above, is placed side by side with the other principal Western chant repertories which have come down to us: the old Italian chant of Milan, Rome, and Benevento, and Old Spanish (or Mozarabic) chant. It is generally agreed, as a very broad hypothesis, that where these repertories show any sort of formal or musical similarity, and yet have developed more or less independently of each other, the similarities may indicate a state of the repertory earlier than the one that has come down to us. If we have three or four similar versions of chant, the common basis of those versions may go back
/. Introduction
479
earlier than the ninth century, beyond the time when the first notated books appeared. Yet the fact that these west European repertories diverge as much as they do suggests that extreme caution must be exercised in assessing the possible interrelationships between Western and Eastern chant repertories. And in the East, too, the absence of notation before the ninth century (or later—some repertories were not notated before the present century) raises a formidable barrier.
The first codification in a form of musical notation of the chant we know as 'Gregorian' is the result of the Prankish endeavours to establish Roman liturgy and its chant as normative in their land. There is nevertheless plenty of evidence that the Franks did not simply reproduce the Roman liturgy and its music exactly. Extensive liturgical modifications were necessary to accommodate local requirements. At the same time, the Franks expanded the chant repertory in radically new ways: the repertories of sequences and tropes are the best-known examples of this. Two other achievements of the ninth and tenth centuries are placed in context immediately after the chapter on compositional activity: the codification of the chant repertory with musical notation, and the development of a body of writings on music theory. These have to be understood in the light of both the liturgical renewal and the general intellectual revival of the ninth century, the so-called 'Carolingian renaissance' under Charlemagne and also the 'second Carolingian renaissance' under Charles the Bald. The significance of the 'Carolingian century' is discussed in Chapter VII.
For many centuries both before and after the time when the Franks were learning and codifying the Roman repertory, the Byzantine Empire and church offered a political and liturgical example which might from time to time have been emulated in the West. At a very basic level, Byzantine chant also has roots in the music of the early church which it shared with the Western chant repertories, perhaps discernible in certain musical features common to both Western and Eastern chant. On a more detailed level, we know of Byzantine chants which were translated and sung in the West, and it is probable that the eight-mode system of Gregorian chant is derived from Byzantium. The possible influence of Byzantine chant in the West is considered in VIII.2.
The principal features of the Old Italian chant repertories are discussed in VIII.3— 5. As already remarked, the 'Gregorian' repertory was first codified in Francia. Unfortunately, it seems thereby to have displaced local repertories, commonly referred to as 'Gallican', before they achieved written form. The scant remains of Gallican chant are discussed after the Old Italian repertories, in VI11.6. Old Spanish chant was indeed codified, but at a time when diastematic notation (indicating pitch precisely) was not yet used. Only a handful of Old Spanish chants can be transcribed today. Nevertheless, the shape of the melodies codified in adiastematic notation can be recognized sufficiently well to permit a fair degree of certainty about some aspects of Old Spanish chant (VIII.7).
Subsequent chapters pursue the history of Western chant after the Carolingian settlement.
480
VI. Plainchant up to the Eighth Century
VI.2. THE CHURCHES OF CHRISTENDOM Baumstark 1958.
The subject of this book is plainchant in the West, meaning the chant sung in the European Latin liturgies used as far east as the areas which correspond roughly to modern Poland, Czechoslovakia, and Hungary. To understand the origins of Western plainchant, a knowledge of the early history of the liturgy is important. This implies consideration not only of the Roman liturgy but also of others, Eastern as well as Western. Ultimately it is also possible to compare the chant repertories of these other rites with Western ones, though in case of many rites the written tradition is very late. Studies of this type have, however, hardly begun, and even questions of basic methodology remain as yet unanswered. It must also be admitted that contact between the different Christian communities was interrupted at an early date by theological differences, for example the separation of communion caused by the Monophysite and other disputes, and by political divisions after the fall of the Roman empire, so that the prospect of meaningful musical comparisons is faint.
The earliest Christian community, in Jerusalem, was presided over by St James after the other Apostles spread abroad, and was the place of the first church council of about the year 49. But after the Jewish revolt of 66 and consequent Roman destruction of the city in 70 Jerusalem had little importance. Then in about 326 Helena, mother of the emperor Constantine, came to venerate the holy places of the city, which began an important fashion of pilgrimage and seems ultimately to have contributed to making the liturgy more topical, commemorative, bound up with the remembrance of persons, places, and events. Uniquely valuable and detailed descriptions of some of its ceremonies have survived, written by the pilgrim Egeria or Etheria, thought to have been a Spanish nun or abbess, who visited many places in Eastern Christendom at the end of the fourth century (Corpus Chnstianonirn Series Latina, 175; Eng. trans. Gingras 1970, Wilkinson 1971). At the Council of Chalcedon in 451 the patriarchate of Jerusalem was created. After the Islamic conquest of the Holy Land in the 630s the patriarch resided mostly in Constantinople. Although a fair amount is known about the rite of Jerusalem (not least from Armenian and Georgian borrowings), there are of course no records of its music (see Baldovin 1987, Leeb 1979).
The two most important Middle Eastern Christian centres in the early centuries were not Jerusalem but Antioch and Alexandria.
The church at Antioch in Syria, traditionally founded by St Peter, was by the fourth century third in rank of the patriarchal sees (after Rome and Alexandria). In the next century it was split by christological controversies. The Greek church of Constantinople and the Latin church of Rome have held to the doctrine that Christ has two natures, human and divine, within a single person, affirmed by the Council of Chalcedon. The belief that Christ has two separate persons as well as natures,
2. The Churches of Christendom 481
Diphysitism, condemned at the Council of Ephesus in 431, was supported by part of the Antiochene church, usually called 'Nestorian' ('East Syrian', 'Church of the East', or 'Assyrian'), which by the end of the Middle Ages had actually spread as far as south India and China. Monophysitism, the doctrine that Christ has one nature and person, condemned at Chalcedon, was supported by the 'Syrian Orthodox' ('Jacobite' or 'West Syrian') party of the church at Antioch as well by the Alexandrine and Armenian churches. The liturgy of the Jacobite church, known as the 'Liturgy of St James' (reflecting its supposed derivation from Jerusalem), adopted Syriac as its language. The Antiochene party which rejected both heresies was called 'Melkite' and retained Greek in its liturgy.
Books of the Syrian rites were rarely notated, though some notated ones, principally Melkite, have survived. Research on the music of the various Syrian liturgies has therefore relied on transcriptions made from modern performance. (See Husmann, 'Syrian Church Music', NG; by far the greater part of recent research was accomplished by Husmann.)
The patriarchate of Alexandria, second in rank to Rome before the rise of Constantinople, was traditionally held to have been founded by St Mark, hence the name 'Liturgy of St Mark' for its rite. The church opted for Monophysitism. It became known as the Coptic church after the use of the vernacular (Coptic, descended from ancient Egyptian) in the liturgy. The church became relatively isolated after the conquest of Egypt by Islam in 641. Although a form of ekphonetic notation may have been used in the tenth and eleventh centuries, Coptic chant has been transmitted to the present only orally. (See Borsai, 'Coptic Rite, Music of the', \G\)
The Ethiopian church is a daughter of the Syrian Jacobite and Coptic churches, and was subject to the latter until this century. The oldest notated sources of its chant date from the sixteenth century. (See Hannick, 'Ethiopian Rite, Music of the', NG.)
The Armenian church was recognized by the country's ruler, Tiridates III, in 301, even before the Roman emperor Constantine's toleration of Christianity. It became schismatic half a century after Chalcedon. Elements of its liturgy are a faithful reproduction of the Jerusalem liturgy of the fifth century (Renoux 1961-2, 1969-71). Both an ekphonetic and an indigenous chant notation are known from Armenian liturgical books, said to date from as early as the ninth century. (See Hannick, 'Armenian Rite, Music of the', NG.)
The church of Georgia dates from the fourth century. Monophysite from the Council of Chalcedon onwards, it returned to orthodoxy at the end of the sixth century. Since the development of the Georgian rite was much influenced by the Georgian monasteries in the Holy Land (St Sabas near Jerusalem and on Mount Sinai), it may incorporate some elements of the chant of Jerusalem (see Tarchnischvili 1959-60). Georgian lectionaries with Byzantine ekphonetic notation survive from as early as the tenth century, and chant-books in an indigenous notation are apparently equally ancient. (See Hannick, 'Georgian Rite, Music of the', NG.)
In 330 Constantine refounded Constantinople and made it his capital, on the site of
482 VI. Plainchant up to the Eighth Cent my
Greek Byzantium, which appears to have had a Christian community from the second century. At a council in the city in 381 the bishop was given precedence second only to Rome, and became patriarch in 451. Much of its early liturgy appears to have derived from Antioch, whence St John Chrysostom came in 398 to be its bishop. The liturgy in general use still bears his name. That reserved for the Sundays of Lent (except Palm Sunday), Maundy Thursday, the Eves of Easter, Christmas, and Epiphany, and the Feast of St Basil (1 January) is known as the Liturgy of St Basil (d. 379).
The political supremacy of Byzantium eventually lent great influence to its liturgy as well. It was translated into Old Slavonic by SS Cyril and Methodius upon the conversion of the Slavs of Moravia in the mid-ninth century, and when Vladimir, prince of Kiev, was baptized in 988, he also introduced the Byzantine liturgy in Old Slavonic. While the Greek expansion into Moravia was short-lived—it clashed with German political interests—the Slavonic liturgy was permanently established in Bulgaria (from the mid-ninth centry) and Serbia (from the tenth). We have musical documentation of Byzantine chant from the ninth century (ekphonetic notation possibly from even earlier), of Russian chant from the late eleventh, of Bulgarian chant only from the thirteenth, and Serbian from the fifteenth. (Byzantine chant is discussed briefly below, VIII.2; on the Slavonic families of chant see Vehmirovic, 'Russian and Slavonic Church Music', NG.)
The patriarchate of Rome coincided more or less with the western half of the Roman empire, comprising Italia, Illyricum, Africa, Hispania, Gallia, and Britannia.
The bishops of Rome in the first two centuries were all Greek-speaking. Greek remained the language of the Roman liturgy until the fourth century (Klauser 1946). We should in any case expect a strong similarity to have existed between the rites of the various Christian churches at this early period. This means that, with due caution, evidence about one church can be considered for its relevance to another. The later the period, the less safe does this become, for it is by then a matter of assessing influences upon established liturgies, rather than the development of similar usages from common origins. The general picture appears to be one of great conservatism at Rome in face of new liturgical developments. Rome's political importance dwindled after the establishment of Constantinople; it was replaced at times by Milan and Ravenna as imperial residence, and it suffered from barbarian invasion and from wars between the Byzantine empire and Goths and Lombards. Yet this did not weaken its constancy to liturgical tradition. On the other hand, the considerable independence of other Western uses from those of Rome may be attributed to its political weakness. Since it is by no means easy to trace the development of Roman liturgical use through the vicissitudes of the early centuries until the time from which we first have comprehensive liturgical books (the eighth and ninth centuries), attempts to understand its chant in anv detail are extremely difficult.
Aquileia in north Italy assumed the rank of patriarchate in the sixth century, from the seventh century being split into two seats, one at Grado, the other at Cividale. Despite speculation about an independent rite and even an independent Old Aquileian chant repertory, practically nothing is known about its early liturgy.
2. The Churches of Christendom 483
Ravenna and Milan both developed local rites. Ravenna was seat of the imperial residence in the fifth century, and capital of the Byzantine Exarchate from 540 to 751. Again, very little is known of an independent musical repertory. Milan, on the other hand, has retained its liturgical independence until the present day, and its chant repertory, first codified in the twelfth century, constitutes prime evidence of the degree to which local uses varied in the early Middle Ages. It has been argued that the political link between Milan and Benevento in south Italy, first under the Ostrogoths, then the Lombards, provided the basis of a similar chant repertory in the two cities. At any rate, by the time of the writing down of chant in the Beneventan region, another local repertory different from the Roman had obviously developed. (On the local Italian chant repertories see below, VIII.3-5.)
The church in North Africa, centred on Carthage, used a Latin Bible and liturgy even before the church in Rome, but little is known in detail of its rite. Weakened by the Vandal conquest (429)—the Vandals were Arian Christians—restored by the reconquest under Justinian (534), the church faded after the Arab conquest at the end of the seventh century. Nothing of its musical repertory is known.
The term 'Gallican' has been used to indicate all the various north Italian rites independent of Rome and those of Gaul and Spain as well. Alternatively, the liturgies of Spain, Gaul, and the Celtic lands (primarily Ireland) have been grouped together. The term is usually now restricted to the liturgies of Gaul alone. The evidence for these is incomplete and there was doubtless considerable variation from area to area and from century to century. (We are best informed about the church in the Rhone valley.) No written record of Gallican chant was made before the establishment of the Frankish-Roman liturgy in the eighth and ninth centuries. A certain amount of Gallican material survives, however, in 'Gregorian' books. The nature of Gallican chant and the way it differed from Roman is nevertheless not easy to define, not least because of the differences between the two versions of the Roman chant repertory which have come down to us: the 'Gregorian', first codified by the Franks in lands where the Gallican rite had been used, and 'Old Roman', the chant found in books from Rome itself. It is possible to argue that 'Gregorian' chant is at least in some respect a 'Gallicanized' version of the Roman repertory (see VIII.6).
By the expression 'Celtic church' is usually meant the church in Britain and Ireland until the time when Augustine of Canterbury (d. 604 or 605) and later churchmen brought Roman usages, which gradually supplanted Celtic ones. For most of England this happened during the sixth and seventh centuries, while Ireland remained unaffected until considerably later. The process of conversion to Roman customs forms a parallel, some two centuries earlier, with the change in Francia during the eighth and ninth centuries. It should be emphasized that, even less than in Gaul, there is little evidence of a centrally organized church with a 'use' prescriptive within a diocese. Almost no chant which can be considered authentically Celtic has survived: the change to Roman use in England occurred long before musical notaton was used, and no notated books survive from Ireland until those of the thirteenth century, based on English (Sarum) use.
484 VI. Plain chant up to the Eighth Century
The other branch of the Roman church for whose rite not only the texts but also the music achieved codification was that of Spain. Various terms have been used to designate this rite and its chant: 'Visigothic' (Spain was ruled by the Visigoths from the fifth century until the Islamic invasion of the eighth), 'Old Spanish', and 'Mozarabic' (a term properly used to designate the Christian population living under Arab dominion). Although much can be learnt about its music, practically all surviving chants are recorded in adiastematic notation. The coming of pitch-specific notation coincided with the reconquest of Spain from the Moors and the importation of the Frankish-Roman rite and Gregorian chant. We have almost no manuscripts with Old Spanish chant after this event, and consequently almost no records in pitch-specific notation (see VIII.7).
It will be clear from this brief outline that reconstruction of the chant of the early church is extremely hazardous. By the time written records of the chant repertory of any of the branches of the Christian church were made, they had had ample time to develop along independent lines, through centuries of often radically changing circumstances which cannot fail to have had a considerable impact on the liturgy and its music.
In the next two sections the history of the most important musical forms of the Roman chant are outlined and parallels are drawn with chants of other rites where appropriate. The emphasis is upon the other Western rites. Since almost nothing of the chant of the Celtic lands survives I have largely disregarded Celtic practices (see Warren 1881/1987; Curran 1984). Despite the lean evidence regarding Gallican practices, I have drawn upon them more regularly to point up similarities with other rites.
VI.3. THE EARLY CHURCH
Stäblein, 'Frühchristliche Musik', MGG; Hannick, 'Christian Church, Music of the Early', NG; Smith 1984; McKinnon 1986, 1987, Music.
Until the so-called Edict of Milan of 313, by which Constantine and Licinius sanctioned the freedom of worship of Christians (and adherents of any other religions), Christians were often persecuted savagely, religious meetings took place in private, often in secret, and public worship was impossible. In these circumstances the order of worship remained simple and was in many respects likely to have been unformalized.
Until recently it had been assumed that the Jewish Book of Psalms constituted a chief source of texts for singing in the early church, and that continuity must have existed between Jewish and early Christian worship. This view is no longer accepted. Hannick ('Christian Church, Music of the Early', NG) has summarized the large
3. The Early Church
485
number of witnesses to the singing of non-psalmodic hymns, and both Smith (1984) and McKinnon (1986) have provided ample evidence that Jewish public worship could have given to Christian worship neither its form nor its content. Two types of Jewish worship enter the matter, that of the Temple in Jerusalem, and that of the synagogues (synagoge is the Greek translation of Hebrew beth ha-knesset: 'place of assembly). In the grand setting of the Temple a different psalm for each day of the week was sung, with instrumental accompaniment, at the daily sacrifice; the Hallel psalms, Pss. 113-18, which have the refrain 'Alleluia', were also sung on important days of the Jewish year. Both because of their special associations and their special manner of performance, it seems unlikely that these psalms would have been adopted bv the small, private groups of Christians. The Temple was destroyed by the Romans in the year 70. The evidence from the synagogue is equally unpromising, for a different reason. There is no evidence that in New Testament times meetings in the synagogue were for worship in the normal sense of the word. The synagogue was a secular rather than an exclusively religious meeting place. While passages of scripture (including passages from the psalms) were undoubtedly read there and expounded, it was not normally a place of prayer or psalm-singing. In the decades after the destruction of the Temple, which removed at a stroke the whole focus of Israel's religious life, something like an ordered service of worship became established in the synagogues, a partial substitute for what had been lost. But psalm-singing, or more specifically the singing of the daily psalms once used in the Temple, was one of the last elements of Temple worship to be taken up in the synagogue, to judge by its absence from documents of Jewish religious teaching before the sixth century.
More promising is the evidence linking Christian religious meetings with private Jewish religious exercises. The prayers shema and tefillah, both part of the Temple liturgy, were to be recited privately, and the Hallel psalms were sung at the ceremonial meal, the Seder, at Passover. Indeed, if the Last Supper of Christ and the disciples took place on the first night of Passover, the likelihood is that the 'hymn' they sang (Matt. 26: 30, Mark 14: 26) was the Hallel.
Most early references to singing during Christian worship are too vaguely expressed to be very enlightening. The words 'psalm' and 'hymn' usually have only the general meaning of 'song of praise', and cannot unconditionally be connected with the Old Testament psalms, or with any particular poetic or musical form.
It has been argued that some early Christian songs are quoted in passages such as 1 Tim. 3: 16, or Rev. 4: 8, but neither constitutes conclusive evidence (McKinnon 1987, Music, 16-17). Few newly composed texts have come down to us. The forty-two 'Odes of Solomon', perhaps written towards the end of the first century by a Syrian Christian, are compositions like the Old Testament psalms, and may be representative of new Christian composition (McKinnon 1987, Music, 23-4; Charlesworth 1977). Much has been lost, not least the heretical hymns of Bardesanes (Bar-Daisan) of Antioch (154-222) and Paul of Samosata, bishop of Antioch in the third century. Their popularity and consequent danger to orthodox Christians seems to have aroused suspicion against all non-biblical chants, which were then banned in
486 VI. Plainchant up to the Eighth Century
272 at the second Synod of Antioch and in the fourth-century Canons of Laodicea (the reference is to idiotikous psalrnous: McKinnon 1987, Music, 119).
Something of the nature of Bardesanes' songs can perhaps be gleaned from the songs of Ephrem of Edessa (d. 373), who is supposed to have composed orthodox contrafacta to the melodies which Harmonios, son of Bardesanes, had written for his father's heretical poems. Prominent among them are the madrasha, a strophic solo song with choral refrain, and the sogitha, a strophic song in dialogue form. Their nearest equivalents in the medieval Western liturgy are perhaps the processional hymns of Passiontide: Gloria laus et honor, Audi iudex mortuorum, and Patige lingua.
Much of the musical component of early Christian worship must have been improvised or at least performed as the occasion demanded, according to no fixed order of service. Something of the character of the services may be gleaned from 1 Cor. 14: 26-7, where Paul comments on what seems to be a largely improvised type of worship, probably prevalent in other places as well as Corinth: 'When ye come together, every one of you hath a psalm, hath a doctrine, hath a tongue, hath a revelation, hath an interpretation.' It has also been supposed that early Christian worship might occasionally have resembled that of other sects, for example the pre-Christian monastic Therapeutae from near Alexandria, whose worship is described by Philo in the early first century (ed. Conybeare 1895, ch. 11). Here the believers rise after their Sabbath meal, form a choir of men and a choir of women, with a musically experienced leader. They sing hymns of varied types, some together, some alternatim (antiphonois). On the other hand, in one important description of the eucharist, celebrated on Sunday morning, written by Justin Martyr (rlOO-c. 165) in Rome, there is no reference to singing of any sort (McKinnon 1987, Music, 20).
The distinction between eucharist and other forms of worship is important. In the first three centuries the eucharist was commonly the culmination of a common religious meal, the agape or love-feast. Singing clearly took place during the agape. Tertullian, writing at Carthage in the closing years of the second century, mentions both singing 'from the sacred scriptures' and improvised singing (McKinnon 1987, Music, 43, Taft 1985, 26-7). A document known as the Apostolic Tradition, attributed to the Roman priest Hippolytus, written in the early third century, describes an agape where the Hallel psalms are sung by various persons, with 'Alleluia' as the refrain (McKinnon, 47). If not by the agape (which gradually fell out of use), then the eucharist might be preceded by other customary observances. Thus in commenting on a charismatic sister of the Montanist movement, Tertullian says she is inspired by the various components of the synaxis which precedes the eucharist: scriptural readings, psalms, a homily, and prayers.
The usual time for the agape and/or eucharist was the evening. Of the possible musical component, if any, of religious observances at other times of the day we know little. The Apostolic Tradition describes a number of occasions for private prayer during the day, and a morning communal assembly for instruction as well as the common evening meal.
Although we can speculate, on the basis of later evidence, as to which texts might
4. Office Chants before the Eighth Century 487
have been sung during these religious observances, we have almost no precise evidence. The same is naturally true of their music. The single piece which has survived (fragmentarily), the hymn recorded on the Oxyrhynchus papyrus of the late third century, in Egypt, raises as many questions as it answers (facsimile and various transcriptions in Stäblein, 'Frühchristliche Musik', MGG). The Greek vocal notation with rhythmic signs allows a relatively secure transcription. It appears to be a solo song in the authentic G-mode (or, less anachronistically stated, using the diatonic Hypolydian scale), with texts in anapaestic dimeter. Although the text is sacred, it is impossible to say whether the music is typical of its time and place, either in its rhythm or its melodic style. The piece was recorded on the back of an (earlier) cereals account, thus for private purposes. Its function (if any) in contemporary Christian worship remains unclear.
VI.4. OFFICE CHANTS BEFORE THE EIGHTH CENTURY
Stablein, 'Fruhchristliche Musik', MGG; Hannick, 'Christian Church, Music of the Early', NG; Baumer 1905; Hucke 1953, 'Entwicklung', 1973; Salmon 1959, 1967; Jungmann 1960; Heiming 1961; Leeb 1967; Winkler 1974; Cheslyn Jones et al. 1978, ch. V; Bradshaw 1981; Taft 1985; McKinnon 1987, Music; Dyer 1989, 'Monastic'.
The official toleration of Christianity and the end of persecution must have had wide-ranging consequences for the form and content of worship. It is no accident that information about the liturgy is much more plentiful from the fourth century onwards. It is often no easier to interpret than that from the previous period because of the differences of observance between different areas. But from the fourth century it is possible to see clearly many of the chief characteristics of the liturgy familiar to us from the Middle Ages. Two developments are particularly important.
First, public worship became a matter of more organized ceremonial conducted by trained personnel in churches built for the purpose. By the end of the fourth century we have ample evidence of formal liturgies of various kinds. This means, for example, that songs were performed by cantors specially detailed to carry out this duty, rather than by any person present inspired to sing (see Hucke 1953, 'Entwicklung', 177-85, with information on the formation of the trained choir; also Foley 1982, Fassler 1985, McKinnon 1987, Music, 109). The ceremonial aspects of worship were elaborated in order to add mystery and inspire awe. This can be seen, for example, not only by the liturgy described by Theodore, bishop of Mopsuestia (392-428), north-west of Antioch, but also by his commentary on it (Dix 1945, 282-8). It has been supposed that more extended liturgical actions may have provided the occasion for new accompanying chants.
Secondly, there was a dramatic movement—already under way in Egypt towards the end of the third century—by many religious persons to cut themselves off from the world and lead a solitary life, alone as hermits, as hermits answering occasional
488 VI. Plainchant up to the Eighth Century
calls to meetings, or as communities of monks and nuns. The monastic communities developed their own forms of communal worship, which occasionally interacted with their 'secular' counterparts but remained essentially distinct. Many such communities engaged in more or less continuous worship, usually organized in a system of office hours different from the prayer hours which were observed in the previous centuries. The chief purpose of the religious exercise was continual prayer and praise, particularly through the complete recitation of the Old Testament psalms. This ultimately led, among others, to the type of medieval office known to us in the Roman and Benedictine uses. The secular office, by contrast, was naturally selective in its texts. In recent times it has therefore become usual to distinguish a 'cathedral' or 'clerical' office, the non-monastic use, from a 'monastic' office. We are better informed about monastic practices than cathedral or secular liturgies, because of the survival of the 'rules' of several founding fathers of monastic orders.
Thus from the churches of the western half of the Roman empire there is only fragmentary evidence from before about 600 to inform us about cathedral liturgies, while from the same area we have the Ordo Monasterii (part of the so-called Ride of St Augustine), probably written in North Africa by a follower of St Augustine (ed. Verheijen 1967), the rules for the monks and nuns of St Caesarius of Aries and his successor Aurelian, bishop of Aries 546-53 (ed. Morin 1942, McCarthy 1960), the late sixth-century rule of St Columbanus for the Celtic church (ed. Walker 1957), and then, for early sixth-century Italy, the Regula Magistri (ed. de Vogue 1964, trans. Eberle 1977) and the Rule of St Benedict (ed. Hanslik 1960, de Vogue and Neufville 1971-7). With the aid of the last two it has been possible to reconstruct much of the contemporary monasticism in the convents which were attached to the basilical churches in Rome. These are by no means the only uses which can be reconstructed, or about which something is known, as a glance through the accounts of Bradshaw (1981) and Taft (1985) will show.
Common to the cathedral or secular uses was a basic system of daily morning and evening prayer and Sunday eucharist, already established by the end of the fourth century. Many components of morning and evening prayer were characteristic of the time of celebration: Ps. 62 (Deus Deus mens ad te de luce vigilo) for morning prayer and Ps. 140 (Domine clamavi ad te . . . elevatio manuum mearum sacrificium vespertinum) in the evening. Morning prayer also nearly always included Pss. 148— 50, and evening prayer often began with the ceremony of lamplighting. In addition to all these, an all-night vigil service was occasionally performed on the eve of important feast-days. (Those known to Gregory of Tours are listed by Taft 1985, 182.) Here too psalmody took its place alongside readings and prayers, but generally speaking there is little evidence of the recitation of the psalms in cycles.
In these morning and evening services the bulk of the music consists of Old Testament psalms. That is something of a contrast from the early centuries. Other items certainly had a place: the Gloria in excelsis in the morning, and the hymns composed by Hilary of Poitiers and St Ambrose of Milan, for instance. But the psalms are ubiquitous.
4. Office Chants before the Eighth Century 489
In the monastic office the psalter was recited continuously, at first informally and then in groups of psalms distributed among the fixed hours, so arranged that the whole psalter could be performed within a fixed period, usually a week (see the summaries in Bradshaw 1981 and Taft 1985). Cathedral uses with especially numerous psalms, often to be found in the West, seem likely to have been influenced by monastic practice. For example, the Second Council of Tours in 567 specifies what is practically a monastic round: twelve psalms at Lauds, six with 'Alleluia' at Sext, twelve at 'duodecima' (an evening service distinct from the lamplighting ceremony), and a number during the Night Office which varied according to the month—as many as thirty in December (Taft 1985, 149). Ambrosian hymns are permitted, and also hymns by other authors, whose names should be written in the margin (implying a written collection of hymn texts).
We can only speculate on the basis of later evidence what the music during these services was like (for example, four of Ambrose's hymns are thought to have survived with something like their original melodies). Even in important matters of performance practice—the division between soloist and choir, the role of refrain verses, and the use of alternating groups of singers or choirs—there is considerable difficulty in interpreting the early evidence, no doubt because of the considerable variety that existed. Important distinctions which have to be borne in mind are those between monastic and secular practice, and between the performance of the Old Testament psalms and the singing of newly composed verses or songs. The greatest difficulties arise in understanding what is meant by 'response' and 'antiphon' and the concepts 'responsorial' and 'antiphonal psalmody'.
In one sense 'response' and 'antiphon' might seem to be identical, with the meaning of 'refrain', a verse or verses sung after each section of another chant, for example after each verse of a psalm. A possible distinction here would be that the response takes its text from the psalm or other piece being sung (usually its first verse, sung first by the leader and repeated by the chorus), whereas the antiphon is an independently composed song. If that seems a trifling difference, we should remember the sensitivity displayed particularly in the third and fourth centuries in the face of non-biblical songs: those of Bardesanes in Antioch, or those of the Arians in Constantinople described by Socrates and Sozomen (McKinnon 1987, Music, 101-4). General conservatism with regard to non-scriptural lyric compositions is one of the things which distinguishes the medieval Roman liturgy most strongly from the Byzantine, where hymn genres such as the kontakion, kanon, and sticheron flourished in great abundance. (The evolution of these is revealing: see Wellesz 1961, chs. VI and VIII-IX.)
The concept of the antiphon and the antiphonal manner of performance has, however, an ancillary meaning of 'alternating' between groups. It has been argued that 'responsorial' relates primarily to the alternation between soloist and chorus, 'antiphonal' to that between two choirs, say the people and the trained choir, or the two halves of a choir. We should also consider the possibility that the two meanings of 'antiphon' may not be mutually exclusive: the antiphon to be both newly composed,
490 17. Plainchant up to the Eighth Cent toy
non-scriptural, and performed by alternating choirs. Isidore of Seville (c.560-633), for example, says that antiphons are chanted by two choirs alternately (Etymologies 6. xix, De ecclesiasticis officiis 1. vii and ix). He may have been thinking of alternatim performance of the psalm verses to which the antiphons are a counterpart. But in view of the usual solo singing of the psalms themselves at this date, it seems more likely that the antiphon itself is to be chanted first by one side of the choir, then by the other. That is what we find, much later, described by Amalarius of Metz in his Liber officialis of r.823 (Hanssens, ii. 433, quoted by Dyer 1989, 'Monastic', 68). First the antiphon is begun by a soloist from one choir, and seems to be completed (or repeated) by both choirs together. The performance apparently continues:
Psalm verse: soloist Antiphon: first choir
Psalm verse: soloist
Antiphon: second choir Psalm verse: soloist Antiphon: first choir
Psalm verse: soloist
Antiphon: second choir
(and so on)
The great attraction of 'antiphons'—whatever they may have been—is difficult to understand when so little is known about their character and content. There is no space here to examine each reference in order to distinguish between possible meanings (Hucke 1953, 'Entwicklung' provides the most comprehensive and judicious review). I shall extract only a few salient points which are particularly relevant to the medieval usage in the West.
Early reports connect the first singing of antiphons with Antioch, and, lest their non-biblical nature and their use by heretical groups like the Arians should taint their image, a legend grew up that Bishop Ignatius of Antioch (martyred c. 107) had seen a vision of angels singing 'antiphonal hymns', and introduced the same sort of songs into his own church (McKinnon 1987, Music, 102). The fourth century is, as with so many things, the period when the fashion is first widely recorded, and support for its Eastern origin comes from both St Augustine and the biographer of St Ambrose. Augustine says that during the time when Ambrose's church in Milan was under threat from the Arians, in 386-7, 'hymns and psalms' were first sung 'after the manner of the oriental regions' to raise the spirits of those under siege (Confessions 9.7.15; McKinnon 1987, Music, 154). This has been connected with a remark by Paulinus, former secretary of St Ambrose, writing in 422, but describing the same events as Augustine, that at this time 'antiphons, hymns, and vigils' were first performed in Milan. As so often, there is no precise evidence here of what 'antiphon' means, though its connection with inspiriting congregational singing is clear. (On the other hand, Leeb 1967 pointed out that Ambrose himself only mentions solo psalms with congregational responses in his sermons: see the discussion of the gradual below, VI.5.)
4. Office Chants before the Eighth Century 491
The recent study of Dyer has made it clear that for early monastic psalmody solo performance was the rule. Each monk in turn played his part in singing while his brethren meditated on what was being sung. We do not know what the music for this singing was like. Something akin to the familiar tones may have been used, though solo delivery naturally allowed greater freedom of musical expression.
The meditative role of the monks not taking their turn to sing a psalm was, however, occasionally varied by their adding responses or antiphons to the solo psalm. In John Cassian's Institutes, written about 415 for the bishop of Apt, drawing on Cassian's experience of Egyptian monasticism, he says that monks in Palestine first stand to sing three 'antiphona' (usually taken to mean psalms with antiphon refrains), then sit while three monks in turn sing a psalm solo, the other brethren 'responding' (presumably singing short responses). Various arrangements of psalms with antiphons and psalms with responses are to be found among early monastic uses. In some, the balance is weighted heavily in favour of psalms with antiphons. This is the case in the Rule of the Master, for example (ed. de Vogue 1964; see also Bradshaw 1981, 140, Taft 1985, 122-3), where the proportion is over three to one.
Several items in the Rule of the Master are labelled responsoria, and the question arises as to whether these are in fact the same as the old psalms with responses. (See Hucke 1953, 'Entwicklung', 172-3, with reference to the Rule of Benedict, slightly later than the Rule of the Master and drawing heavily upon it.) They appear only as single items, not in the sets typical for the recitation of the whole psalter. In fact, they stand outside the cycles of psalms. We may surmise that pieces like the short responsories known to us from medieval sources may have featured among them, but a more ornate manner of delivery cannot be ruled out.
It is evidently necessary to make a clear distinction between the singing of short responses (such as 'Alleluia') during the cyclic singing of psalms, and the singing of responsories independently of the psalm cycle. It is the latter which we see in the Rule of the Master, and also the Rule of St Benedict, where responsories are performed during the Night Office in alternation with lessons, a new arrangement with which we are familiar from the Middle Ages: the texts are specially selected for their relevance to the lesson. No doubt this could have been done ad libitum. If the performance resembled that of medieval short responsories, the singer of the responsory would have selected a passage and sung an opening verse, which would be repeated by the others present (either choir or congregation is conceivable). The soloist would have continued with further verses, the others responding after each verse with the same verse as they had sung at the start. In Benedict's Rule we see this systematically organized for the first time, and may assume that the medieval cycles of lessons and complementary responsories had their birth here.
When perusing old orders of service (such as those gathered together by Bradshaw 1981 or Taft 1985), it is of course not always possible to know whether the responsorial chant is part of a psalm cycle or the complement to a lesson. The sixth-century 'Psalter of St Germain-des-Pres' (Paris, Bibliotheque Nationale, lat. 11947; see Huglo 1982, 'Repons-Graduel'), a psalter with responsory texts, seems to be
492 17. Plainchant up to the Eighth Century
designed for cyclic performance of the psalter. On the other hand, two lectionary fragments have survived with psalm verses apparently intended for responsorial performance: Wolfenbuttel, Herzog-August-Bibliothek, Weissenburg 76 from the fifth to sixth century (ed. Dold 1936) and Mount Sinai, St Catherine's monastery Gr. 567, possibly a Spanish manuscript of the seventh century (Lowe 1964).
We have two reports of the selection of responsory texts as early as the fifth century (Vogel 1986, 302-3). Gennadius relates that the priest Musaeus of Marseilles made a selection of readings for feast-days and passages for responsories from the psalms (responsoria psalmomrn capitula) to go with them (McKinnon 1987, Music, 170). Sidonius Apollinaris credits Claudianus of Vienne with something similar (cited by Jeffery 1984, 161). With the Wolfenbuttel and Mount Sinai fragments, and with the reports about Musaeus and Claudianus, it is difficult to know if we are dealing with the office responsory or the gradual of mass (the Wolfenbuttel fragment also has what may be offertory texts), but the overall nature and purpose of the extracts is clear.
The early history of the chief forms of office music, the antiphon and the responsory, is therefore by no means clear in detail. Without knowing more about the precise texts sung and their manner of performance, we cannot say very much about their music. The same two forms, or something related to them, were used in the mass liturgy, the subject of the next section, which may help our understanding of the office chants.
Hymns are distinct from the forms discussed so far by reason of their poetic text and their independence from psalmody or readings (although we cannot be sure that all early references to 'hymns' would fit this definition). Being non-scriptural, they were occasionally regarded with the same suspicion as antiphons. They were not part of the early Roman monastic office, nor of the Rule of the Master, but Benedict included them. On the other hand, they are recorded for several other early Western liturgies, particularly secular ones. It has been suggested that a statement by Jerome may refer to the singing of hymns at morning prayer in Roman secular use (Taft 1985, 143). They are known from Milan in St Ambrose's time and north Africa in St Augustine's. For Gaul we have the hymns of St Hilary of Poitiers (c.315-67), and they are mentioned at several Gallican synods. The full set of daily hymns of the monastic rule of Aries of the 530s is known (Taft 1985, 103-4). Huglo has reconstructed the whole Gallican hymnal ('Gallican Rite, Music of the', NG). Canon 13 of the Fourth Council of Toledo (633), presided over by St Isidore of Seville, allowed the singing during the office of the Gloria in excelsis and other non-biblical hymns such as those of Hilary and Ambrose. That hymns, like antiphons, were valued for the opportunity they gave for popular participation is indicated by the following passage from the Life of St Caesarius of Aries (c.470-542):
He added also and provided that the lay inhabitants memorize psalms and hymns, and chant, some in Greek, others in Latin, proses (prosas) and antiphons with a high and melodious voice, like the clergy, so as not to have any time to waste on gossip while in church. (MGH, Passiones vitaeque sanctorum, 463-4; Taft 1985, 151.)
4. Office Chants before the Eighth Century
493
The nature of the 'proses' is uncertain.
The texts of numerous Celtic ecclesiastical hymns have survived, though their liturgical function and position are not often clear (Warren ed. Stevenson 1987, pp. lxxxiii—lxxxix).
The sense of contrasting monastic and non-monastic offices has been dulled somewhat since this early period because of the prevalence of a monastic type since the Carolingian renaissance. The differences between secular and monastic use in the Middle Ages are relatively minor when set beside the adoption by secular uses of the monastic daily Night Office and Little Hours. The distinction between the two types of office was already blurred in Rome from the fifth century onward, when monastic communities became attached to the basilical churches such as the Lateran and St Peter's. Benedict's Rule appears to have been influenced by the practice of these communities. After the reforms of the eighth century in Francia, clerics were increasingly expected to perform a full cycle of office hours after the Roman or a quasi-monastic pattern.
The prevalance of the full monastic-style office is not restricted to Roman/ Benedictine use. We have both secular and monastic Old Spanish sources of office chants; it is clear that the secular services were simply built out into a monastic cycle (see W. C. Bishop 1924). Rather than outline the whole of the Old Spanish monastic liturgy (see Taft 1985, 115-20), I summarize the chief forms of chant, which may be compared with the Roman/Benedictine categories of antiphon, Venite antiphon, great responsory, little responsory, and hymn.
Antiphons for psalms and canticles. In the Night Office and Lauds, two 'normal' antiphons and an alleluiatic antiphon (that is, one with 'alleluia' exclamations in the text) were grouped with a responsory to form the unit known as a 'missa'. (The 'alleluiatici' often served as introit antiphons at mass.) Another type of antiphon distinguished only by the topicality of its text was the 'matutinarium' for Lauds. The 'Benedictiones' were antiphons to be sung with the Benedictus canticle (Dan. 3: 52) of the Night Office. Antiphons were also sung during the other hours.
'Laudes', like alleluiatic antiphons without a psalm, were sung in the little hours and Vespers.
Great responsories were sung in each 'missa' of the Night Office, and singly in the other hours.
The Eucernarium' or 'Vespertinum' is a responsorial chant (varying numbers of verses are found, and some pieces lack a verse) which begins Vespers.
The 'Sonus' of Lauds appears to resemble the Sacrificium (offertory chant) of mass in form and musical style, and some pieces serve in both functions. They occasionally display long sequence-like melismas with repeat structures.
The Milanese office, like the Roman, betrays a heavy infiltration of monastic forms (see again Bishop 1924). In the time of St Ambrose (bishop of Milan 339-79) a daily morning and evening service and occasional vigil services are documented, though their detailed content is unclear. The medieval order has been edited by Magistretti
494 17. Plainchant up to the Eighth Century
(1904-5) and summarized by Lejay ('Ambrosien (rit)' DACE) and Borella (1964, eh. 11); it can also be followed through the introduction and transcription in PalMus 6.
The repertory of Milanese office chants consists not only of the expected simpler antiphons to accompany psalms and canticles, and responsories for the readings, but also a number of other items, usually as a part of ceremonial lacking in the Roman/ Benedictine office. The following may be mentioned.
For the morning service (a form of Night Office and Lauds following one another with little division) another responsory was sung after the opening hymn. After the main sequence of Night Office items, concluded outside Advent and Lent by the Te Deum, a special 'antiphona ante crucem', with doxology, was sung at the start of Lauds. The Gloria in excelsis deo was sung after the psalmody. Towards the end of the service a 'psallenda', a processional antiphon with doxology, was sung, as the choir proceeded to the women's baptistery. There a responsory 'in baptisterio' was sung, then another psallenda on the way to the men's baptistery.
The same stational ceremony concluded Vespers. Vespers began with the 'Lucernarium', an elaborate antiphon with verse (cf. the Old Spanish lucernarium). There followed the 'Antiphona in choro' without psalm verse, and the 'Responsorium in choro' sung by a deacon. On important feasts the last two chants were replaced by the 'Responsorium cum infantibus', with its lengthy sequence-like melismas (cf. the Old Spanish sonus).
Thus both Old Spanish and Milanese offices include a number of substantial items not found in Roman use, some of them chants which do not fall into the categories of either responsory or antiphon for the psalms and canticles. Bailey and Merkley (1989, 16) have pointed out a considerable overlapping of use between the very rich Milanese repertory of processional antiphons and office antiphons (usually under the name 'psallenda'). We cannot always know how much of this material was present in the non-Roman rites before the earliest chant-books, written after the eighth century. But we should be aware of the alternative lines of liturgical development which took place in parallel with the Roman one.
The next section pursues some of the same themes as this with regard to the chants of mass.
VI.5. MASS CHANTS BEFORETHE EIGHTH CENTURY
(i) The Gradual
(ii) The Introit and Other Chants at the Start of Mass
iii) The Chants at Communion
iv) The Offertory
(v) The Chants beside the Lessons
Stablein, 'Fruhchristliche Musik', MGG; Hannick, 'Christian Church, Music of the Early', NG; Duchesne 1919; Dix 1945; Jungmann 1948 etc. (1962 edn. cited here);
5. Mass Chants before the Eighth Century
495
Cheslyn Joneset at. 1978, ch. Ill; Dyer 1982; Jeffery 1984; McKinnon 1987, 'Gradual'; Bailey 1983, Alleluias, 1987; McKinnon 1987, Music.
(i) The Gradual
The presence of responsories in the office, performed singly in response, as it were, to a reading, suggests a parallel with the gradual of mass. McKinnon (1987, 'Gradual') has shown that the evidence for the presence of a distinct chant among the lessons of mass first accumulates towards the end of the fourth century. While in the early centuries the eucharist had been preceded by readings, ending with a homily, there is no definite information about a chant at this point in the service. One arrangement is documented by St Augustine, who refers to the sequence epistle-psalm-gospel as a set of three readings (McKinnon 1987, Music, 161). It appears that the psalm may have changed from reading to chant, the gradual in fact (but see section v below for the possibility that the ancestor of the tract was involved in this development).
McKinnon sees the establishment of the gradual as part of the general enthusiasm for the Old Testament psalms which took hold of Christianity in the later fourth century. Whereas an Old Testament psalm might occasionally have been declaimed among the lessons, it now became the norm, and a special manner of performing it arose, whereby the congregation or choir punctuated its performance with a refrain, the respond, a verse taken from the psalm itself. The 'Apostolic Constitutions', which describe the mass liturgy of Antioch in the later fourth century, says that after readings from books of the Old Testament someone should sing the psalms, and the people should respond with 'akrostichia' (McKinnon 1987, Music, 109). It was then customary for a homily to be preached on the same respond verse. St Ambrose (d. 397), extolling the virtues of psalm-singing generally, remarks that congregations are noisy when a lesson is being read, but that singing the responses to a psalm involves them in the service (McKinnon 1987, Music, 127); while he may not necessary have had the gradual of mass in mind, what he says seems applicable to that chant.
Quite how the chant evolved from what was presumably a fairly simple chant with congregational participation to the elaborate chants we know from many churches of Christendom (Byzantine prokeimenon, Milanese psalmellus, Mozarabic psalmo) is not clear. As early as the late fourth century, however, Augustine refers both to a plainer type of performance of the psalm (note the singular, possibly implying the gradual) which Bishop Athanasius of Alexandria (d. 373) had preferred, and to the entrancingly melodious psalmody of Milan in St Ambrose's day (Confessions, 10. 33. 49-50; McKinnon 1987, Music, 155).
Jeffery (1984) has provided confirmation of McKinnon's thesis in an unexpected way (his article actually preceded McKinnon's but deals primarily with the next stage in the history of the gradual). A passage in the Liberpontificalis which can probably be dated to the early sixth century states that Pope Celestine I (422-32) decreed the 150 psalms should be sung 'before the sacrifice', whereas previously only the epistle
496 VI. Plainchant up to the Eighth Century
and gospel had been recited. In the early ninth century Amalarius of Met'/ interpreted this to mean that Celestine had introduced the singing of the introit at the start of the Roman mass, an interpretation followed by practically all subsequent writers. But, as Jeffery explains, the passage makes much better sense if Celestine had introduced the singing of a responsorial psalm between epistle and gospel, the gradual in fact. The explanation is especially convincing because Celestine seems to have had experience of the liturgy and its music in Milan at exactly the same time as St Augustine, and might well have wished to introduce the same customs that Ambrose had received from the East and which Augustine said were being taken up all over.
A slightly later interpolation into the Liber pontifical is says that the psalms introduced by Celestine were 'antephanatim ex omnibus', which Jeffery explains by the responsory texts having been extracted 'as excerpts' from the parent psalms. The passage reminds one of the later attribution to Gregory the Great of an 'antiphonarius cento', a patchwork of antiphons. Reports, and actual sources, of such excerpts exist from as early as the fifth century (see above, VI.4); some of these may refer to the office responsory, whose history must have run partly parallel to the gradual.
We shall return to the gradual in a discussion of the chants between the lessons as a whole—gradual, alleluia, and tract—at the end of this chapter (section v below).
(ii) The Introit and Other Chants at the Start of Mass
Where does Jeffery's new interpretation of Celestine's contribution to the Roman mass leave the introit? The earliest direct evidence for Rome is the Ordo Romanus I, from the early eighth century, where it is described as a psalm with antiphon: as many psalm verses are sung as required for the entrance of pontiff and curia (Andrieu 1931 — 61, ii. 83). The Gallican antiphona ad praelegendum had psalm verses, as also the Old Spanish praelegendum, and some of the latter were sung both as entrance chants for mass and office antiphons (the 'alleluiatici'), with the same psalm tones. On the other hand, the Milanese entrance song, the ingressa, has no psalm and is therefore by nature more like the large non-Roman processional antiphons. In all these cases, it would seem that ceremonial considerations stimulated the development of the repertory, but evidence for their origins and development is sparse.
Gregorian sources usually indicate a single psalm verse for the introit, but some early books (particularly Paris, Bibliotheque Nationale, lat. 17436) also have a 'Versus ad repetendum'. This corresponds to the instruction in Ordo Romanus XV (third quarter of the eighth century; although it was compiled by an admirer of the Roman liturgy, he is thought to have worked in Burgundy or Austrasia). The following order of performance is indicated (Andrieu 1931-61, iii. 120; cf. Jungmann 1962, i. 417, 420; Dyer 1982, 15):
introit antiphon psalm verse 1 introit antiphon
[possibly further psalm verses each followed by the introit antiphon, as time allows]
5. Mass Chants before the Eighth Century 497
Gloria patri part 1 introit antiphon Gloria patri part 2 introit antiphon Versus ad repetendum introit antiphon
It has been suggested that the versus ad repetendum is a relic of Gallican practice (Huglo, 'Gallican Rite, Music of the', NG\ cf. Froger 1948, 'Introit'), but it is found in Old Roman sources.
The introit forms a group with Kyrie and Gloria at the start of the Roman mass which is partly matched by the opening chants in other Western rites. Before proceeding to a brief discussion of the communion chant, which has much in common with the introit, a few words about these other chants are in order.
The first chants in the Gallican, Old Spanish, and Milanese liturgies, the antiphona ad praelegendum and the Milanese ingressa, have already been mentioned. The main chants up to the first lesson are listed in Table VI.5.1.
Table VI.5.1. Chants at the start of mass in four rites
Roman    Gallican Old Spanish Milanese
Introit     ad praelegendum ad praelegendum ingressa
Gloria Gloria
+
Kyrie Kyrie Trisagion Trisagion Kyrie ?
Gloria
Bencdictus/Sanctus Deus
The supplicatory verses Kyrie eleison were used on numerous occasions in various rites in various parts of the liturgy, both office and mass, in litanies and processions. The time when litanies with Kyrie eleison became a part of the opening ceremonies of mass is not clear. Possibly Pope Gelasius (492-6) was responsible as far as Rome as concerned (Capelle 1934). Gregory the Great (d. 604) appears to have extended their use to non-festal days. In Milan three simple Kyrie invocations form a mere pendant to the Gloria in excelsis. Porter (1958, 22-3) doubted whether the brief Kyrie chant indicated in the 'Expositio' of the Gallican liturgy was firmly entrenched. Both the Gallican and Old Spanish liturgies had an extended litany chant, the preces, after the lessons.
The Gloria was sung in the morning service as early as the fourth century. Pope Symmachus (d. 514) is said to have introduced or at least extended its use in the Roman mass.
498 17. Plainchant up to the Eighth Century
The related Gallican and Old Spanish rites both included the Trisagion Agios o theos (sung both in Latin and Greek), like the Gloria a 'song of the angels'. This corresponds to the use of Byzantium and the Liturgy of St James.
Like the Gloria in excelsis, the Benedictus or Canticle of Zacharias (Luke 1: 68-79), has links with morning prayer. The Expositio antiquae liturgiae Gallicanae says that the Benedictus was performed 'alternis vocibus' (Ratcliff 1971, 5). In Lent the antiphon Sanctus Deus angelorum was sung instead.
The author of the note on Celestine I mentioned above, concerning the gradual, remarked that previous to his introduction of psalmody only the epistle and gospel had been performed before the sacrifice. Possibly he has in mind the cluster of items epistle-gradual-gospel, and his remark has no relevance to what was performed before the epistle. It is also possible that the singing of introit, Kyrie, and Gloria at the start of Roman mass all postdate Celestine (422-32).
(iii) The Chants at Communion
Versus ad repetendum are also to be found in the same early sources for the communion chant, and for its Gallican equivalent, the trecanum. In the early eighth-century Expositio antiquae liturgiae Gallicanae the 'threefoldness' of the trecanum is described as a reflection of the Holy Trinity (Ratcliff 1971, 16), in a way which Jungmann (1962, ii. 490) interprets as follows:
Father      Son Holy Spirit    Son Father
antiphon    psalm verse   Gloria psalm verse antiphon
The word 'trecanum' has also been explained as a transliteration of the Greek trikanon, the name for Ps. 33. No text of a Gallican trecanum actually survives to support either of these explanations.
Ordo Romanus I (Andrieu 1931-61, ii. 105) states that the psalm verses are to be continued until the pontiff gives a sign, whereupon the Gloria and the 'verse' are to be sung. By 'verse' either the communion antiphon (previously called 'antiphona') or a versus ad repetendum might be meant. Once again Ordo Romanus XV (Andrieu 1931-61, iii. 124) is more explicit, and indicates a performance exactly like that of the introit (see above). One branch of the manuscript tradition of Ordo Romanus I indicates that the schola and the subdeacons alternated in the performance, though what portions they sang is not clear.
As might be expected, the singing of a chant at the fraction of the bread, or the distribution of the communion, is attested at an early date. The singing of Ps. 33 during the distribution is recorded both by Cyril, bishop of Jerusalem 349-87, and in the 'Apostolic Constitutions' (liturgy of Antioch, later fourth century; McKinnon 1987, Music, 77 and 109). Verse 8 of the psalm, 'O taste and see that the Lord is good', no doubt inspired this choice. From both accounts it appears that a soloist performs the psalm, but there is also early evidence of a choral response (see McKinnon 1987, Music, 82, 120, 144.) Indeed when Cyril encourages the people to
5. Mass Chants before the Eighth Century 499
listen to the cantor singing 'O taste and see', it may partly be because that is the very verse they will themselves take up as a choral response.
Early evidence for Rome is lacking. And the fact that the Western rites differ from one another in the placing and character of their chants in the eucharist as a whole suggests that the tradition of these chants is not as old as that of the gradual. Although Gregory the Great placed the Paternoster before the fraction (he defends the practice in a letter to Bishop John of Syracuse), there is no mention of it in most early ordines (it is absent from Ordo Romanus I, for example, but appears in V, X, and XV). In Ordo Romanus I the Agnus Dei (introduced by Sergius I, 687-701) is sung during the fraction, and the communion chant during the communion of those present.
Gallican usage had the fraction, and an accompanying fraction antiphon (without psalm verse), before the Lord's Prayer. The Agnus Dei was not sung. The trecanum, equivalent in function to the Roman communion chant, followed. This order is also that of the Old Spanish rite: a chant 'ad confractionem' (without psalm verse), then the Paternoster, then a chant 'ad accedentes' (with psalm verses). Only the Old Spanish rite had an antiphon a little earlier in the eucharist, the chant 'ad pacem' for the kiss of peace: this was an antiphon with psalm verse. Both antiphon 'ad accedentes' and 'ad pacem' have the same psalm tones as office antiphons, which is the case with the Roman communion as well.
The Milanese chants follow the same order again: confractorium—Paternoster— transitorium. Neither the fraction nor the communion chant has psalm verses. Oddly, such musical concordances as exist link the Roman communion not with the Milanese transitorium but with the confractorium, while the Milanese transitoria show links with Gallican or even Eastern fraction chants.
(iv) The Offertory
The origins of the offertory chant as it appears in the Gregorian tradition of the late eighth century onward are obscure. It was long believed to have been an antiphon with psalm verses chanted in a way similar to the introit and the communion, an 'action chant' like them. Dyer (1982) has shown conclusively that no firm evidence exists for the offertory as an antiphon with psalm verses. Hucke's observations on the responsorial nature of many texts are significant (Hucke 1970). Ordo Romanus I says that it should be terminated when the pontiff gives the choir the nod; but nothing is said about the nature of the chant.
The earliest reference to a chant appears to be in St Augustine's writings, where he reports he was criticized for introducing singing from the psalms both 'ante oblationem' and during communion (Retractationes ii. 11 or alternative numbering 37; McKinnon 1987, Music, 166). 'Ante oblationem', before the sacrifice, may refer to the gradual rather than the offertory. In any case there is no clear indication as to the manner of performance of these psalms (which has not prevented scholars from assuming that both were done with antiphons). The date of the introduction of an offertory chant into Roman use is unknown.
500 17. Plainchant up to the Eighth Century
Chants with an equivalent function in other Western rites are the Old Spanish sacrificium and the Milanese offerenda. Both of these resemble the Roman chant in having one or a few verses, sung in a predominantly (often highly) melismatic manner throughout. The Old Spanish sacrificia are sometimes found as office chants, the soni of the Night Office and Vespers.
The Gallican chant sung during the offertory procession was the sonus. The identity of terminology with the Old Spanish sonus is presumably not accidental; the chant was apparently somewhat different, however. In the description which turns up in Ordo Romanus XV (Andrieu, iii. 122-3 and 74 ff.; Dyer 1982, 15), Laudate dominion de celis (called an antiphon) is sung three times during an extended procession. In the Expositio antiquae liturgiae Gallicanae (Ratcliff 1971, 10-13) the sonus concludes with a threefold alleluia, the laudes. (The Old Spanish laudes, on the other hand, are the equivalent of the Roman alleluias of mass.) Details of the procession and the concluding threefold alleluia correspond to one of the most splendid parts of the Byzantine mass liturgy, the 'Great Entrance', where the Cheroubikon hymn is sung with a triple alleluia.
(v) The Chants beside the Lessons
The early history of the gradual has been touched upon above in a brief discussion of 'antiphonal' and 'responsorial' psalmody. It remains to place it in context in the set of chants sung in close proximity to the lessons of mass. The various Western rites have the arrangements set out in Table VI.5.2.
Table VI.5.2. Chants beside the lessons in four rites
Roman	Gallican	Old Spanish	Milanese
	Prophecy	Prophecy	Prophecy
	Benedictiones	Benedictiones	
		Psalmo: Clamor/	Psalmellus
		Threnos	
		Preces	
Epistle	Epistle	Epistle	Epistle
Gradual	Responsorium		
Alleluia/			Alleluia/Cantus
Tract			
	Antiphona ante evangelium		Antiphona ante evangelium
Gospel	Gospel	Gospel	Gospel
	Sanctus post evangelium	Laudes	Antiphona post evangelium
	Prcces		
5. Mass Chants before the Eighth Century 501
The Milanese antiphons before the gospel appear to have been sung only at Christmas, Epiphany, and Easter, but those after the gospel were sung all the year. These are processional antiphons without verse, to accompany the procession of the deacon to and from the ambo. The Gallican rite had a similar custom, but the return after the gospel was accompanied by singing the Sanctus (as well as in its normal position during the consecration).
The benedictiones in the Gallican rite as described in the 'Expositio' followed one of the readings, presumably the first. It is not clear how much of the canticle, the 'Hymn of the Three Children' (Dan. 3: 52-90) was sung. In the Old Spanish use a variety of related chants were sung, with texts drawn from the canticle, only on the highest feasts.
The Gallican responsorium was sung by boys; its musical character is unknown. The equivalent chants, the Old Spanish psalmo and Milanese psalmellus, like the Roman gradual, are of course ornate. These three are all responsorial in text selection and musical form. The Old Spanish psalmo (psabni pulpitales in the Leon antiphoner) was on some important feasts followed by a 'clamor', after which the repeat section of the psalmo was sung once again.
It is not clear whether the Gallican rite knew an alleluia approaching the dimensions found in Rome, Milan, and Spain. The threefold alleluia which concluded the offertory chant, the sonus, may have been such a piece. The Old Spanish alleluia chant, the laudes, follows the gospel.
A special chant can be found for the Lenten season in all except Gallican use: the Roman tract replacing the alleluia, the Milanese cantus with a similar function, and the Old Spanish threnos replacing the psalmo.
A number of Gallican preces have survived with music in south French manuscripts. Two can be reckoned concordances with Old Spanish preces. The latter were sung on some Lenten Sundays, but also in the office.
The three most extended chants among these are the gradual, alleluia, and tract, and their equivalents in the non-Roman rites.
Bailey (1987, 35-9) has cautioned against the assumption that all early references to a chant between the lessons concern the gradual, for a congregational response is not always specified. Solo chants without refrain, such as the medieval tract, might also have been sung. If we accept the idea that the chant between epistle and gospel had previously been read as a psalm, then the tract, whose text is psalmic in the medieval sources, is as likely a candidate as the gradual.
The introduction into the Roman mass of the alleluia as an ecstatic melismatic chant between the lessons seems to have gone through progressive stages. The early documentation of its use is difficult to interpret, not least because of possible confusion with simple congregational responses of 'alleluia' and the addition of alleluia extension phrases to many chants when sung during Eastertide. One possible sequence of events is as follows, but other interpretations have been proposed. (See Froger 1948, 'Alleluia', Blanchard 1949, Wellesz 1947 and 1954, Martimort 1970, Jammers 1973, to cite only recent studies.)
502 VI. Plainchant up to the Eighth Century
In the fourth century Rome may have sung a special, independent alleluia chant on Easter Day only. By the time of John the Deacon (later John I, d. 526) it was sung on other Sundays during Eastertide (John's letter to Senarius; Isidore, De ecclesiasticis officiis, speaking of African customs). The singing of the alleluia may have continued to spread through the church year, for Gregory the Great, corresponding with Bishop John of Syracuse in 598, implies that when he allows its use beyond Eastertide this actually represents a cutting back of an even more widespread practice 'brought here by the Greeks' (Gregory's letter). I have here chosen to believe that it was the extent of the alleluia's use that Gregory cut back (the word is amputavimus), though other explanations have been offered: that Gregory amputated the melodies we know as sequences, or an extended jubilus for the repeat like the Milanese or Old Roman melodiae (Jammers), or introduced verses instead of the jubilus (Wagner II, 92, Apel 1958, 377). All of this is beside the point, however, if Gregory was speaking of alleluia refrains and end-phrases, as is possible. The lack of unanimity in the medieval tradition certainly speaks against any early establishment of the repertory as we know it from the later chant-books.
Special to Rome was the performance of Greek and Latin alleluias during papal vespers on Easter Eve. This and other features of the Old Roman alleluia repertory has led to speculation about possible Byzantine influence. Although the alleluia was sung before the gospel at mass in the Byzantine rite, the musical setting of the word alleluia itself remained brief, in contrast to the Western traditions.
Despite the evidence of the early Wiirzburg lectionary (see above, III. 3. ii), two lessons, rather than three, are to be found in the Roman liturgy (as in the Byzantine). The Old Spanish, Gallican, and Milanese rites have three. Martimort (1984) has pointed out that the fifth-century Jerusalem liturgy, as preserved in the Armenian lectionary, had a varying number of lessons, and he concludes that the number of lessons was fixed in the different rites relatively recently.
The following possible developments in the Western rites may therefore be suggested. If Bailey is right, the tract in Roman use is the descendant of a reading between the epistle and gospel. The gradual was added as a post-epistle responsory. The singing of the alleluia gradually expanded through the church year, replacing the tract in all seasons but Lent. The theory that a chant might once have been a reading appears to require that there were originally four lessons in the Old Spanish and Milanese rites, an arrangement likewise with ancient testimony (Baumstark 1958, 44). At Milan one lesson became the cantus, eventually replaced outside Lent by the alleluia, while the psalmellus entered the liturgy after the prophecy. There is no trace of a tract-like chant in the (admittedly imperfectly known) Gallican rite, but in the Old Spanish rite one might suppose that the threnos suffered the fate of the Roman tract and Milanese cantus. The threnos would have been replaced by the psalmo. The alleluia chant, thelaudes, was placed after the gospel; but it too was replaced in Lent by another type of responsorial chant, still called laudes but resembling rather the psalmo.
Explanations of a similar sort can be devised if we follow McKinnon and treat the gradual/psalmo/psalmellus as the successor to a reading. To follow a lesson by a
6. Gregory the Great 503
responsorial chant with related text is an ordering well known from the office. But, as Martimort (1970) has pointed out, it is by no means clearly demonstrated in the mass (see also Hucke 1973, 163-71). While the Milanese psalmelli do bear this relationship, other chants do not; the alleluia is linked rather to the gospel, which in most rites follows it. Randel ('Mozarabic Rite, Music of the', NG) has argued that the arrangement of psalmo and clamor in the Old Spanish liturgy cannot be the remnant of the singing of a complete psalm.
Gregory the Great (PL 77. 956 ff.) once remarked that mass could last anywhere up to three hours. We should not underestimate the amount of ceremonial and solemn repetition of chants which may regularly have been practised.
VI.6. GREGORY THE GREAT
(i)	Introduction
(«)	Gregory and the Deacons
(ill)	Gregory and the Alleluia
(IV)	Isidore of Seville; the Liber pontificalis; the List of Chant 'Editors' in Ordo
	Romanus XIX
(v)	The Anglo-Saxon Tradition
(vi)	The Biographies by Paul Warnefrid and John Hymmonides
(vii)	The 'Gregorian' Sacramentary
(viii)	The Prologue Gregorius praesul
(IX)	The Reception of the Gregory Legend in the Ninth Century and Later; the
	Dove; the Modes
(x)	Gregory and the Lenten Communions
(XI)	Conclusions
I lucke	1955, 'Entstehung',  1958, 'Problemen'; Stäblcin, 'Gregor I.', MGG, 1968,
MMMA 2; Steiner, 'Gregorian Chant', NG; Hucke, 'Gregory the Great', NG. (i) Introduction
The 'Gregorian' chant repertory bears the name of St Gregory the Great, pope from 590 to 604. Gregory's actual share in its composition, or the regulation of its liturgical use, has been much debated. Since there is a gap of nearly three centuries between Gregory's life and the appearance of the first completely notated chant-books, it is unlikely that what finally entered the written musical record is what Gregory knew, even if we subtract the chants for those days whose liturgies were added to the calendar after Gregory's time. It is nevertheless possible to argue that an ancient core of the repertory, in something very much like the state in which we find it in the late ninth century, might date back to Gregory. The arguments are circumstantial, for we have no reliable contemporary witness to any musical activity on Gregory's part, and must rely on inferences drawn from the date of the texts and the style of the music. More difficulty arises from the fact that the very idea of Gregory as a composer cannot
504 VI. Plainchant up to the Eighth Century
be traced back earlier than the eighth century, although there are pieces of evidence that some chants may very well be as old as Gregory's time.
This section therefore examines those of Gregory's writings which touch on musical matters, then the biographical writings of the seventh to the ninth century, by which time the Gregorian legend was fully established. Some parts of the chant repertory which Gregory is alleged to have composed are briefly inspected, and a summary is attempted.
Gregory's own authentic writings do not allude to the composition of music on his part, but nor, in the main, are they of the type that would do so, consisting largely of homilies and commentaries on sacred texts. He was an energetic leader in both civil and ecclesiastical affairs, and a prolific writer, the last of the four great doctors of the medieval church (after Augustine, Jerome, and Ambrose), so that activity at least in the sphere of liturgical organization has never been denied to him: it would be normal on the part of such a church leader. But the only authentic references by him to matters affecting music concern (i) the singing of chant by deacons; and (ii) the occasions when the alleluia and Kyrie might be sung: that is, they are liturgical prescriptions rather than purely musical ones, at least at first sight.
(ii) Gregory and the Deacons
At a Council held in 595 (Mansi 1757-98, x. 434), Gregory proposed a number of reforms touching upon the lives of ecclesiastics (though nothing is said of any body which might be called a 'schola cantorum'). One of these arose from the fact that clerics were apparently being promoted to the rank of deacon on account of the beauty of their singing (see Duchesne 1919, 169-70 for epitaphs praising their singing skill). But such an officer should be occupied more in preaching and the care of souls than in 'charming the people with his singing'. Deacons shall therefore henceforth not assume the functions of cantor, but only read the gospel on solemn feasts. 'As to the singing of the psalms and the other lessons, I decree they shall be executed by subdeacons, and, in case of necessity, by clerks in lower orders.'
(iii) Gregory and the Alleluia
Part of a letter written in October 598 by Gregory to John, bishop of Syracuse, about Roman usage concerning the singing of the alleluia, has occasioned much dispute (PL 77. 956). The passage on question was brought together with a wealth of other early evidence about the alleluia by Froger (1948, 'Alleluia'). I have expressed reservations above (VI.5.v), as did Martimort (1970), about the usual assumption that Gregory was discussing the alleluia chant of mass.
The same letter touches upon Gregory's provision for singing the Kyrie sometimes with 'other things' (presumably other Latin verses) and on weekdays simply with 'Kyrie eleison' and 'Christe eleison' (see above, 11.17).
6. Gregory the Great 505
In another passage Gregory defends placing the Lord's Prayer straight after the Canon of mass, where it is recited by the priest alone.
(iv) Isidore of Seville; the Liber pontificalis; the List of Chant 'Editors' in Ordo Rom anus XIX
As Solange Corbin pointed out (1960, Eglise, 173), the oldest 'life' of Gregory is that of Isidore of Seville (De illusttibus sctiptoribus ecclesiasticis, ch. 27, PL 75. 490), who has nothing to say of liturgical work on Gregory's part. Isidore (c.560-636) knew of Gregory from his brother Leander, and was also very familiar with his writings. He was presumably the source of some of the information in the life by his pupil Ildefonsus of Toledo (c.607-67) (De viris illusttibus, ch. 2, PL 96), likewise unhelpful in the present case.
The Liber pontificalis (editions by Duchesne and in MGH) is a composite collection of biographies of the popes, which began in the sixth century and was gradually built up, with contributions of increasing length, down to the fifteenth century. The information on Gregory, very scanty, was probably recorded in 638. It says nothing of the composition of chants, and is in fact silent about liturgical activity altogether. Nor does it mention the existence of a schola cantorum, with an interest in which Gregory was also to be credited. Not until the biography of Sergius I (687-701), who is said to have been educated under the care of a 'prior cantorum', does the existence of a school seem to be implied.
An interesting document entitled 'De convivio sive prandio atque cenis monachorum' has survived, which can be dated to the late seventh century (the last pope on it is Martin I, 649-55). It ends with a remarkable list of popes and Roman abbots who, its author claims, made important contributions to liturgy, chant, and other matters. At one time it was claimed to be the composition of John, abbot of the monastery of St Martin and Archicantor of St Peter's, who came to England in Pope Agatho's time (678-81) to teach chant in Northumbria (as recounted by Bede). This attribution has been rejected since Andrieu's edition of the text in no. XIX of his Ordines Romani (Andrieu 1931—61, iii. 223-4), but that the document may stem from late seventh-century Rome (albeit in the ninth-century manuscript St Gall, Stiftsbibliothek 349), is not now seriously questioned. The evaluation of its contents, however, is not easy, particularly as regards the reasons for its selection of personages.
Eight popes, and the three abbots, are mentioned as having 'edited' or 'ordered' (edidit; ordinavit) the cycle of chants for the year (omnem annalem cantum; anni circoli cantum). Thus of Boniface II (530-2) it is said that 'inspired by the Holy Spirit, he assembled the rule and arranged the cycle of chants for the year' ('regolam conscripsit et cantilena anni circoli ordinavit'). The remarks are quite formal, and no one personage stands out as more important for music than any other. Knowledge of the liturgy and competence in the celebration of the divine service were of course among the prime requirements of any high-ranking Roman prelate in these times. The progression from monk in one of the Roman monasteries to abbot and at the same
506 v7. Plainchant up to the Eighth Cent toy
time leader of the schola cantorum was normal. The list is nevertheless selective in that not every pope is mentioned. Next after Boniface comes Gregory. Relatively much is made of him, and his commentaries and homilies are all cited in reverent terms; then comes the formal close 'et cantum anni circoli nobili edidit'.
(v) The Anglo-Saxon Tradition
An important impulse towards the creation of a Gregorian legend must have come from England. The first reason for this was Gregory's action in sending Augustine to England as a missionary in 597, the first step in the reclamation of England for the Roman church. The second reason was that English monasticism in the later seventh century entered a period of splendid achievement, so that it was eventually able to influence Frankish thinking on liturgy and learning, not least the Frank's notion of Gregory, held in such high esteem in England.
There are two important English sources of information about Gregory. The first is an anonymous life written by a monk of Whitby some time between 704 and 714 (ed. Colgrave 1968); the other is Bede's history of the English church and people, completed in 731 (ed. Plummer, trans. Sherley-Price). The achievements of Gregory here recounted do not include musical ones, and hardly even touch upon liturgy. Of the many charming episodes later to be recounted by John Hymmonides—Gregory's attention to the schola, his compilation of an antiphoner, and so on—the Whitby monk says nothing. WTe do find for the first time the story of the dove which descends upon Gregory; but he is writing not neumes, as depicted in 'Hartker's Antiphoner', but a commentary on Ezekiel. The author had access to the information about Gregory in the Liber pontificalis, and in return his life was known later in Rome, to John Hymmonides. Its only surviving source is an early ninth-century copy, made probably at St Gall and now part of St Gall, Stiftsbibliothek 567.
Bede's history is of special interest as much for what it says about the various cantors who came from Rome in the seventh century as for its information on Gregory. Gregory's answers to questions Augustine sent back regarding problems he was facing are justly regarded as crucial to an understanding of the significance of the mission. Gregory recommends (i. 27) that Augustine make use of whatever liturgical customs seem best for the new English church. (The passage was quoted in the ninth century bv Amalarius (ed. Hanssens, i. 363) in justification of his revision of the antiphoner.) This is consistent with the attitude displayed in the abovementioned letter to John of Syracuse, to whom he wrote:
If this church [i.e. that of Constantinople] or anv other has some good thing, then, just as I reprimand my inferiors when thev commit a crime, so am I ready to imitate them when they have something of worth. It would be follv so to set up the primacy as to disdain the learning of what is better.
Exactly what Augustine managed to introduce in the way of liturgical customs— Roman, Gallican—is debatable (see Deanesly 1964). But contacts with Rome were
6. Gregory the Great 507
maintained. The importance of proper liturgical observance to the life of the church is underlined by the fact that in 680 John, archcantor of St Peter's and abbot of the monastery of St Martin, came from Rome to teach in Northumbria. At Monkwear-mouth, 'in accordance with the instructions of the Pope [Agatho], Abbot John taught the cantors of the monastery the theory and practice of singing and reading aloud . . .' (Bede, iv. 18). Liturgical books were produced for reference, cantors from elsewhere came to learn, and John himself travelled about. Bede also says that both in the seventh century and in his own time there came north singers 'trained in vocal music by the successors of blessed Pope Gregory's disciples in Kent'—in this instance Maban, r.710 (v. 21); previously, and the first to do so, came Eddi, or Stephen, r.670 (iv. 22). The proud link with Gregory was still present in the minds of eighth-century English monks; indeed it was to remain so for many centuries (for the example of eleventh-century Glastonbury see Hiley 1986, 'Thurstan').
A member of the Northumbrian royal house at the time Bede was writing, Egbert, provides the link with the Carolingian era. Ordained deacon at Rome, he was appointed bishop of York about 732. In a letter of 735 which still survives, Bede advised him to apply for the pallium, which he obtained. His brother Eadberht became king in 738, and with his support Egbert founded a cathedral school in Y'ork where, among others, Alcuin was a pupil and eventually master. The vitality of the Gregorian tradition was maintained. Indeed, whatever state of codification the liturgy had reached at York in Egbert's time (he died in 766), the composition of the books was attributed to Gregory himself. What is more, these were not books of prayers, which one might expect, but chant-books. In two places in his Dialogus ecclesiasticae institutionis (PL. 89, 377-451), a treatise on discipline, reference is made to
beatus Gregorius in suo antiphonario et missali libro per paedagogum nostrum beatum Augustinum transmisit ordinatum et rescriptum . . .
beatus Gregorius per praefatum lcgatum, in antiphonario suo et missali . . . non solum nostra testantur antiphonaria, scd et ipsa quae cum missalibus conspcximus ad apostolorum Petri et Pauli limina. (PL 89, 440-2, cited Gevaert 1890, 80-1 and Ashworth 1958.)
Although Gevaert thought the text unauthentic, modern historians seem not to dispute Egbert's authorship (see, for example, Haddan and Stubbs 1869-71, iii. 413— 16, and Ashworth 1959, despite the argument that the books cannot have been compiled originally by Gregory). It seems almost certain that these were books with texts only, not notation. Egbert's testimony antedates by only half a century such Italian documents as the Roman antiphoner of Hadrian I (772-95) seen by Amalarius at Corbie, the late eighth-century antiphoner fragments from Lucca (Lucca, Biblioteca Capitolare 490), with the 'Gregorius praesul mentis' prologue (Froger 1979, 'Fragment'), and the eighth-century Italian missal fragments (chant texts in smaller hand) Rome, Biblioteca Apostolica Vaticana, Vat. lat. 10644—see Gamber 1968, no. 1401, also facs. Gamber 1962, 336). The Corbie antiphoner is lost; the others are unnotated.
508
VI. Plainchant up to the Eighth Century
(vi) The Biographies by Paul Warnefrid and John Hymmonides
Two biographies of Gregory, widely separated in character and date, illustrate the enormous growth of interest in Gregory during the eighth to ninth centuries, that is, at the time during which Frankish rulers made such efforts to 'Romanize' the liturgy in their lands.
Paul Warnefrid ('the Deacon') wrote a brief biography in the second half of the eighth century. It exists in two versions, the original one (ed. Grisar 1887) being shorter and the other (in PL 75. 1-59) interpolated. Even in the PL edition it takes less than twenty columns. John Hymmonides, writing in 872-3, left a text of nearly 200 columns in the modern edition (PL 75. 58-242).
Grisar's edition is based principally on three Montecassino manuscripts of the eleventh century (145, 146, and 110) and sixteen other Italian sources. Other, non-Italian sources are variously interpolated with extra stories and other material. Paul Warnefrid gives the 'Lombard' (south Italian) version of Gregory's work. John Hymmonides (also 'the Deacon') knew of two lives, a Lombard one (Paul's) and an English one (the monk of Whitby's), and at the behest of Pope John VIII (872-82) now provided a Roman vita. Since even his ardent collector's zeal missed some of the stories in the interpolated Paul text, it may be assumed he knew only the original.
Paul does not give the story of the dove which descends upon Gregory, although the homilies are mentioned. Nor does he say anything of liturgical, let alone musical, activity on Gregory's part. The interpolated versions (PL 75. 57-8) add the dove, but during Gregory's work on Ezekiel, not on an antiphoner: a familiar of Gregory's notices that Gregory, dictating in the next room, is making pauses, and, spying through the keyhole, sees that this happens whenever a wondrous dove on his head speaks into his ear. The servant is forbidden to tell of the events, but does so when he hears ill being spoken of Gregory after his death. Also among the interpolations is the information that Gregory arranged the entire population of Rome in seven groups, to sing the Letania septiformis after a visitation of the plague (PL 75. 59).
The great litany, sung as the penitential procession goes from church to church, is described in greater detail by John (PL 75. 80). He knows and describes pictures of Gregory, with a dove, but music does not enter into his explanation (PL 75. 222). After Gregory's death, Peter the Deacon, Gregory's familiar, is himself on his deathbed: seeing that some of Gregory's books are about to be burnt, he forbids the sacrilegious act, citing the dove's miraculous appearance as evidence of the books' holy inspiration.
The main interest, however, of John's book lies in sections ii. 6-10 (PL 75. 90-2), where Gregory's liturgical and musical activity is described in colourful prose (printed as verse by Stablein, MM MA 2. 143*). In ch. 6 we are told that Gregory 'antiphonarium centonem cantorum studiosissimus nimis utiliter compilavit', 'constituted' the song-school with two dwellings, by St Peter's and the Lateran palace respectively, where his couch, cat-o'-nine-tails for helping the choirboys learn their chant, and the 'authentic antiphoner' are still on display for the reverent visitor.
6. Gregory the Great 509
Although it is easy to dismiss this as pious fiction, John cannot be presumed to have invented wilfully. It is likely that he recounted what was believed in his day, and one must explain how these beliefs arose. The 'authentic antiphoner' may be one of those seen by Egbert, or a successor. The decision to place Gregory's name in writing at its head—this happened to the sacramentary sent by Hadrian to Charlemagne in the late 780s and the only slightly later 'Blandiniensis'—is difficult to pin down chronologically. It may have been contemporaneous with its appearance on the sacramentary. It may have happened in France rather than Rome. But even before the 780s both Rome and its clients evidently believed in Gregory's authorship. Upon the book could easily have followed the couch, the whip, and a belief that Gregory was the originator of arrangements in the field of musical administration as well as composition. Let it be stressed again that we need not mistrust John's veracity, even though we doubt the antiquity of the things he described.
The other chapters in John's narrative speak of the reception of Gregory's chant in other lands. The Germans and Gauls cannot sing it properly (ch. 7); John the Cantor takes it to Britain (ch.8); Charlemagne tries to ascertain why the Franks are not singing as the Romans do, and Metz is established as a chant centre (chs.9-10). Intriguingly, John says that when the Roman singers were challenged, 'they probably showed the authentic antiphoner' to the Franks. Does this mean that it had musical notation, in John's day at least? But how old was the book he knew?
As in the case of the sacramentary, whose ancestry scholars do indeed trace back to a (no longer extant) redaction by Gregory I, it is likely that Frankish demands for 'authentic' chant books stimulated their redaction and production in Rome. It is also likely that the Anglo-Saxon tradition was a decisive factor: witness their interest both in Gregory personally and in acquiring authentic books (for example, the booty brought back from six journeys to Rome by Benedict Biscop, abbot of SS Peter and Paul's, Canterbury and founder of the double monastery of Monkwearmouth and J arrow (d. 689)—related by Bede in his História Abbatum).
(vii) The 'Gregorian'Sacramentary
As already recounted (see above, III.3), Gregory also lent his name to the 'Gregorian sacramentary', a prayer-book of the Roman church which in one form at least—the so-called Hadrianum—was sent to Charlemagne by Pope Hadrian at some time between 784 and 791 at the request of the Frankish ruler. (Charlemagne's request had been conveyed to Hadrian by Paul Warnefrid.) In the letter which accompanied it, and in the earliest surviving copy, Cambrai, Bibliothéque Municipale 164, dated 811-12, it is attributed to Gregory. It seems possible that manuscripts of chant texts with a similar attribution also circulated, and that is the gist of the verse prologue (in hexameters) Gregorius praesul (see below, section viii). A Saint-Martial writer of the eleventh century actually said that the Gregorius praesul prologue itself was composed by Hadrian.
The Hadrianum represents not a modern book of the late eighth century designed
510 VI. Plainchant up to the Eighth Century
to meet Frankish requirements, but a book of masses as normally celebrated by a Roman pope, the stational liturgy when he visited the various churches of the Eternal City, in a recension probably little different from the state in which it was left by Gregory II, who added the formularies for the Thursdays of Lent. About eighty prayers in the 'Gregorian' sacramentary are reckoned to be by Gregory the Great (Ashworth 1959, 1960), their identification depending on parallels with his authenticated homilies and other writings. But the title of the sacramentary, 'In nomine domini incipit sacramentorium, de circulo anni expositum, a sancto Gregorio papa Romano editum, ex authentico libro bibliothecae cubiculi scriptum', may well refer as much to Gregory II as to Gregory I.
(viii) 77?^ Prologue Gregorius praesul
Perhaps by analogy with the sacramentary sent to Charlemagne, perhaps because Roman antiphoners already had such a title, the earliest extant 'chant' book bears a title ascribing its contents to Gregory. Brussels, Bibliotheque Royale 10127-10144, the so-called 'Blandiniensis', written somewhere in north France in the last years of the eighth century, begins with the following:
in dei nomen incipit antefonarius
ordinatus a sancto gregorio per circulum anni
Perhaps just as old, however, is the Lucca source of the Liberpontificalis, Lucca, Biblioteca Capitolare 490, possibly once part of a chant-book, which already has one of the longest versions of a complete poem beginning 'Gregorius praesul mentis' (facs. PalMus 2, pi. 3). It is not clear whether its Italian rather than northern provenance is significant. This and other versions of the prologue have been discussed by Stablein (1968), who also reported an even older source, now lost, from the diocese of Vercelli. Two of the sources edited by Hesbert in AMS (see 2-3) may be cited here: Monza, Basilica S. Giovanni CIX (written in north-east France in the middle of the ninth century), and Paris, Bibliotheque Nationale, lat. 17436 (Soissons, slightly later):
Monza CIX Paris 17436
gregorius praesul gregorius praesul
meritis et nomine dignus meritis et nomine dignus unde genus ducit
summum conscendit honorem      summum condescendens honorem
qui renovans monumenta renovavit monimenta
patrumque priorum patrum priorum
tum conposuit hunc et composuit hunc
libellum musicae artis libellum musicae artis
scolae cantorum scolae cantorum
in nomine dei per anni circulum
Stablein also reports the remarkable fact that the Metz patterning of its liturgy and music on a Roman model went so far as the adaptation of the Gregorius text to
6. Gregory the Great 511
'Anchilramnus presul dum summum conscendit honorem . . .', though without mention of a 'book of musical art' (Stablein 1968, 543 n. 23, after Andrieu 1930, 350).
(ix) The Reception of the Gregory Legend in the Ninth Century and Later; the Dove; the Modes
The dove is not seen in connection with the dictation of music until the frontispiece of the late tenth-century antiphoner of Hartker of St Gall (PalMus II/l), where neumes are actually being written. A chant-book and a representation of Gregory are brought together for the first time in the Monza cantatorium of the mid-ninth century, Monza, Basilica S. Giovanni CIX (north-east French; ed. Hesbert in AJVIS). Here an ivory diptych of the late antiquity depicting two consuls was altered to show Gregory and King David (Stablein, 'Gregory I.', MGG, 773; for fuller iconographical information see Stablein 1968, n. 39).
Also legendary, it seems, is the attribution to Gregory of the creation of the four plagal modes in response to Ambrose's four authentic ones, a story first put about, as far as I am aware, by Aribo in the eleventh century (GS i. 210, CSM 2).
Gregory's authorship of sacramentary, antiphoner, or missal was variously accepted in the ninth century and later. Amalarius of Metz recorded attributions to Hadrian at the head and tail of the Roman antiphoner (Hanssens, iii. 14; cited Hucke 1955, 'Entstehung', 261 n. 21), but it is clear that he and others of that time had doubts over its authenticity (Hucke, 261-2).
Elsewhere the name of Gregory was plainly used as a means of establishing the authority of the chant. Pope Leo IV (847-55), writing to Abbot Honoratus of (?) Farfa, uses the words 'Gregoriana carmina', perhaps for the first time, while ordering the chant to be sung and the Roman liturgy observed on pain of excommunication (Hucke, 264).
In a short life of Hadrian II which Adhemar of Chabannes appended to a summary of the Liber pontificalis (Paris, Bibliotheque Nationale, lat. 2400, Limoges, eleventh century; ed. Duchesne; pp. clxxxii-iv), the verse prologue is attributed to Hadrian I (772-95). This is the remarkable document which attributes the composition of tropes to Gregory I, sequences and more tropes to Hadrian I, their introduction in France to Hadrian I and Charlemagne, and their revivification after a period of negligence to Hadrian II (Gautier 1886, 38). All that was best in medieval chant therefore went back to Gregory.
(x) Gregory and the I^enten Communions
Some peculiarities in the choice and the transmission of communion antiphons in the Lenten period have suggested that developments as old as Gregory are at issue. (See Cagin in PalMus 3, Hesbert 1934, 198 and /IMS, p. xlviii, several articles by Callewaert reprinted 1940, especially Callewaert 1939, Hesbert in PalMus 14, p. 225, and finally Hucke and Huglo, 'Communion', NG). The matter is not susceptible to
512 VI. Plain chant up to the Eighth Century
incontrovertible proof and is outlined briefly here only in order to demonstrate the sort of arguments which have been adduced in order to reconstruct an early chapter in chant history.
The masses for the Thursdays of Lent were arranged during the pontificate of Gregory II (715-31). We may assume that the chants for the other twenty-six weekdays were already in existence before then. Nearly all the twenty-six communions for those days have texts taken from Pss. 1 to 26, in numerical order, whereas the added Thursday communions fall outside this scheme. Yet the numerical arrangement is itself probably not ancient, since the other chants for those days are not ordered numerically. The numerically ordered communions nevertheless display a stable musical transmission when finally codified in the late ninth century.
In among the numerically ordered pieces there are, however, five non-psalmodic communions, replacements, as it seems, for communions drawn from Pss. 12, 16-17, 20-1. These are Opor'tet te, Qui biberit, Nemo te condemnavit, Lutum fecit and Videns dominus.
Three of the five 'odd' communions quote from the gospel of the mass. Morin thought that the three masses in question were originally masses for the scrutiny of catechumens. Callewaert argued that one of the three, Videns dominus, bore the stylistic hallmarks of an authentic text by Gregory, and that Lutum fecit and Qui biberit might also be his. On this basis, Callewaert suggested that it was Gregory the Great who might well have been responsible for the reorganization of the scrutiny masses, which were shifted from Sundays to weekdays because of their diminishing importance in a completely Christian society (Chavasse 1948). Musicologists are clearly not the only scholars eager to see Gregory as an active reformer of the liturgy.
The five 'odd' communions appear in musical sources with considerable inconsistency, amounting to different melodies. What are thought to be the original melodies, the melodies which appear in the earliest sources, are simple and syllabic, distinctly plainer in musical style than most communions. The other melodies for the five which appear in divergent sources seem to be attempts to provide more 'communion-like' substitutes. But why should the five communions have such a special musical character? Does it mean that the sixth- or seventh-century redactor who was responsible for putting them into place borrowed them from another context, one where their musical style was the appropriate one? That would imply that both sets of melodies were already in existence when the scrutiny masses were reorganized. The sequence of events might have been as follows:
1. The ancient scrutiny masses have old and simple communion antiphon melodies.
2. The communions for the weekdays of Lent have more ornate melodies, of the type more recently in fashion, when the masses for those days (always omitting the Thursdays, however) were compiled.
3. The old scrutiny masses are placed (by Gregory?) into new weekday positions, still retaining their ancient melodies, displacing what are actually newer antiphons.
6. Gregory the Great 513
All this assumes, of course, that we still have something very like the original melodies in our ninth-century musical sources. As I see it, Gregory could not have been responsible both for the liturgical change and for composing text and melody of the 'odd' communions.
Callewaert also suggested, again on grounds of the style and content of their texts, that Gregory was the author of antiphons of Sexagesima and Quinquagesima Sundays. Their music has not, so far as I know, been investigated with this possibility in mind.
(xi) Conclusions
It is no doubt spiritually and morally reassuring, for us as for the Carolingians, to be able to attach the name of a canonized author to the body of chant we know as Gregorian. But the available evidence does not permit this. Apart from the insufficient documentary tradition, liturgical and musical analysis has found no strong evidence that the melodies first recorded in the ninth century could be as old as Gregory. No doubt many of the same texts were sung with the same liturgical assignment in Gregory's time as later, but we must reckon with both more subtle and more radical changes in singing practice in the interim, changes which call into question any notion of 'sameness' between the sixth and the ninth century. In this context, the very idea of a 'composer' of chant is anachronistic, and we should do better to uncouple Gregory as a person from the chant repertory.
His name retains its usefulness, in the sense that 'Gregorian' chant is neither of one specific time, nor wholly Roman, nor wholly anything else. A legendary label is as good as any.
VII
The Carolingian Century
VII.1. INTRODUCTION
The events which led to the establishment of the chant repertory we know as 'Gregorian' are inextricably bound up with political developments of the eighth century, principally the establishment of a strong and extensive Frankish kingdom, which by the end of the reign of Charlemagne (768-814) covered an area roughly corresponding to that of modern France, West Germany, and north and central Italy. This is the 'Francia', the empire of the Franks, referred to in this book. Charlemagne was deeply concerned that the church in his domain should follow the liturgy according to Roman practice, as far as possible, which naturally included the plainchant of the liturgy.
The next section recounts some of the stages in the establishment by the Franks of a Roman chant repertory in their empire. Since the needs of the Frankish church did not correspond in every detail with Roman practice, adjustments were necessary. But the Franks also enriched the music of the liturgy in ways which had no precedent in Roman practice (VII.3). A special part in the establishment of a corpus of liturgical chant in Francia was played by notated music-books, which as far as can be seen at the moment became increasingly important at the end of the ninth century (though some scholars would place this development as much as a century earlier). VI 1.4 comments on the cultural and historical role of the earliest notated books. A final section discusses the place of music in the general cultural awakening popularly known as the 'Carolingian renaissance'.
VII.2. THE ESTABLISHMENT OF ROMAN CHANT IN
FRANCIA
Klauser 1933; Hucke 1954, 'Einführung'; Vogel I960, 1965 'Cultuclle', 1965 'Liturgique'; Stäblein in MM MA 2; Wallace-Hadrill 1983.
Rome in the early centuries had neither the desire nor the power to regulate details of liturgical practice in all parts of Western Europe. The example of Gregory the Great shows this clearly: his recommendation that Augustine use whatever liturgical materials he found good, and his correspondence in liturgical matters with other
2. Establishment of Roman Chant in Francia 515
prelates, where he is at pains to explain and defend what he has done but does not seek to impose it on others (see above, VI.6).
Nor did the other churches of Western Europe expect Rome to be able to lay down a pattern of worship, or try to copy Roman use. The import of the various councils which laid down liturgical prescriptions is not that Roman practice shall be followed but that the liturgy shall be performed in a way that is seemly and proper. The prescriptions naturally took into account local traditions and conditions. This is what we find at the Council of Agde (506), where bishops from the whole of south-west France and Spain gathered, or the Fourth Council of Toledo (633), held under the presidency of St Isidore, which was attended by no less than sixty-two bishops from Spain and Septimania (their liturgical prescriptions are summarized by Taft 1985, 147-8 and 159-60 respectively).
The change in outlook which led to wholesale adoption of Roman practice may well be due to the influence of the Anglo-Saxon missionaries who in the late seventh and early eighth centuries brought new German territories into the Christian fold: Willibrord in Frisia from 690, and WTynfrith, who took the name of Boniface, in central Germany and Bavaria in the first half of the eighth century. After organizing the German church, Boniface turned his attention to the Frankish congregation, reformed through a series of synods in the 740s.
Boniface had been called to Rome in 722, where Gregory II (715-31) gave his blessing to his work. To Anglo-Saxons such as Boniface and his disciples, several of whom became bishops in Germany, the traditional connection of their church with Rome was vital. Their work was accomplished in a spirit of obedience to and in frequent consultation with Rome. Newly founded monasteries in Germany followed the Benedictine rule.
The Frankish church under the newly founded Carolingian dynasty of Pippin III (751-68) was consequently far more Rome-oriented than previously. In 750 Pippin had sent the Anglo-Saxon Burghard of Eichstätt and the Frank Fulrad of Saint-Denis to Rome to gain approval, in effect, for the deposition of the Merovingian king and his own succession. In 754 Pope Stephen II (752-7) visited Pippin to seek military aid against the Lombards, and Pippin was consecrated king at Saint-Denis.
There is no need here to recapitulate subsequent political interchange between Francia and Rome. It constitutes, however, the essential background to the adoption of the Roman liturgy in Francia. The decisions of the numerous Frankish church councils of the mid-eighth century are concerned not with liturgy but with fundamental church discipline, so we cannot trace the details of liturgical interchange very clearly, but its pace unquestionably began to quicken. In 760 Pippin's brother, Bishop Remigius (or Remedius) of Rouen, went to Rome to ask that a Roman teacher of chant be allowed to come north, while monks from Rouen learnt the chant in Rome under George, the primus scholae. Simeon, secundus of the Roman schola cantorum, was sent by Paul I (757-68) to teach Remigius' clerics. After a while George died, and Paul had to recall Simeon to succeed him, while assuring Remigius that the Rouen singers in Rome would be brought to perfection under Simeon's instruction.
516 VII. The Carolingian Century
Already the records of the Council of Cloveshoe in England (747) speak of a chant-book sent from Rome (VI.6.v). Paul I sent not only Simeon but also books to Rouen: an 'antiphonale', a 'responsale' and books on orthography and the seven liberal arts, all in Greek. (Vogel 1960, 242, MM MA 2, 148*). This evidence that books with chant texts occasionally found their way north from Rome is very important. (I have assumed that at this time they would not be notated.) But on the whole we are much better informed about the Ordines Romani and sacramentaries which were prepared from or under the influence of Roman models than about Roman or Romanized books with chant texts, for we have no chant-books of the eighth century which would reveal the state of the repertory before and during the wave of Roman influence in the second half of the century.
It was Metz, rather than Rouen, which became renowned as a centre of Roman chant in Francia. Chrodegang of Metz (d. 766), the leader of the Frankish church after Boniface was killed on a mission to Frisia in 754, composed a rule for the canons of his city which borrows heavily from Benedictine practice and refers frequently to Roman customs (Mansi xiv. 313; PL 89. 1059-1120). It marks a crucial stage in the 'monasticization' of the office for the secular clergy. The life of Chrodegang by Paul Warnefrid (d. 799) attributes to him the introduction of the 'cantilena Romana' in Metz (MGH (Scriptores) Scriptures rerutn Sangallensium, etc., 260-70, PL 95. 720; see Buchner 1927). Even a stational liturgy of the papal type, whereby the bishop celebrated mass in different churches on different days, was introduced in Metz (Klauser 1930, Andrieu 1930).
Pepin as the prime mover is then repeatedly mentioned in the numerous documents issued under Charlemagne which concern the singing of chant (Vogel 1960, 230-1, 265-6). What his father had begun became a prime element of Charlemagne's governance. The injunction to sing according to Roman use is repeated time and time again in the numerous capitularies issued under his instructions. The examination of priests included the question as to whether they sang the 'cantus Romanus' in the Night Office and at mass (MGH (Leges) Capitularia re gum Franconim, i. 235; see also 110 for the required adherence to Roman liturgy).
Charlemagne's promulgation of a uniform Roman liturgy in Francia has been attributed to political motives, as a means of buttressing the political unity imposed in many regions by force of Frankish arms. This view is too cynical. At the beginning of the famous 'Admonitio generalis' of 789, Charlemagne's extensive proposals for ecclesiastical and educational reform, he recollects the work of King Josiah in reforming the worship of Israel (2 Kgs. 22-3; MGH Capitularia re gum Franconim, i. 54, cited by McKitterick 1977, 2). The good king is careful of the religious observances of his people, suppresses idolatry, and observes the (written) law. Charlemagne's interest could also take a personal turn. If Notker's report is to be believed, he himself listened carefully in order to hear if the singing was correct. His biographer Einhard says he attended church four times a day, and could have intoned the psalms and lessons publicly had he wished to do so. Nor was he without experience of the Roman liturgy, being in Rome four times in all, including the Easter
2. Establishment of Roman Chant in Francia 517
of 787, which he spent with Hadrian I, and Christmas 800, when he was crowned emperor.
If the second half of the eighth century saw the most vigorous contacts between Roman and Frankish singers, there seems to have been a decline in the ninth. Around 820 Helisachar, the chancellor of Louis the Pious (814-40) and abbot of Saint-Riquier and of Saint-Aubin at Angers, wrote to Bishop Nidibrius of Narbonne in terms which suggest that the Roman way of singing responsories was no longer known (see Huglo 1979, 'Remaniements'). At first Metz provided an alternative. The Capitulary of the missi dominici (Charlemagne's emissaries who carried out civil and ecclesiastical visitations) at Thionville in 805 says that for a knowledge of chant one must go to Metz (MGH Capitularia re gum Erancorum, i. 121). Archbishop Leidrad of Lyon (799-814) reformed the liturgy of Lyon according to the use of the imperial palace itself, with the help of a cleric of Metz (MGH (Epistolae) Epistolae Karolini aevi, ii. 542, Hucke 1954, 'Einführung', 185, MMA 2, 149*).
The famous story of the Roman and Frankish cantors also gives Metz a place in chant history. The affair, seen through Roman eyes, is told by John Hymmonides (PL 75. 91, Hucke 1954, 'Einführung', 180, MMMA 2 144*). He says that Charlemagne noticed the lack of harmony between Roman and Frankish singing and sent two clerics to Hadrian to learn Roman chant; they came back to Metz and through them the singing in all Gaul was corrected. After they died the situation deteriorated, so Hadrian sent two of his singers to Gaul. The singing in Metz had deteriorated much less than elsewhere.
The life of Charlemagne written by Notker Balbulus of St Gall gives the Frankish side of the story (Hucke 1954, 'Einführung', 178-9). Here the astonishing accusation is made that twelve Roman singers, having been sent to Francia by Pope Stephen II (according to Notker; it would actually have been Stephen III) at Charlemagne's request, set out to confuse the Franks by deliberately singing wrongly, being envious of the glory of the Franks. To circumvent this animosity, an apologetic Pope Leo, Stephen's successor (it would actually have been Hadrian I), suggested that singers be sent disguised to listen to the services sung properly in Rome itself. They came back to Metz and the royal palace at Aachen respectively. Thus the 'ecclesiastica cantilena' is called 'Messine', of Metz.
Even if deliberate misinformation be ruled out, the establishment of Roman chant in Francia was a hazardous undertaking, relying principally on human memory for the learning and promulgation of a vast repertory of music. The process seems to have lasted roughly half a century, coinciding with the reigns of Pepin and Charlemagne. The celebration of mass was altered from whatever Gallican forms were used in Francia to coincide with the Roman liturgy, and the Roman chants of mass had to be learned. We could imagine that the Anglo-Saxon churchmen would have contributed their knowledge of Roman chant, but direct contact with Rome itself was essential. A first generation of singers learned in Rome or from Romans, then taught their art to the next generation. The celebration of the office hours presumably followed a similar course; perhaps Roman/Benedictine practice was established earlier and more firmly, at least in German lands, through the work of Boniface and his fellow Anglo-Saxons.
518 VII. The Carolingian Century
In the absence of Anglo-Saxon chant-books this remains speculation, but the prestige of Anglo-Saxon learning and liturgical practice cannot have been diminished by the presence at the imperial palace of Alcuin of York as Charlemagne's chief liturgical adviser.
VI I.3. THE FRANKISH EXPANSION OF THE CHANT
REPERTORY
No book with chant texts sent from Rome to Francia has survived, nor anything that can be called a direct copy of one. The result of the process of establishment of Roman chant in Francia can nevertheless be discerned in the earliest surviving Frankish chant-books, the unnotated graduals edited by Hesbert in AMS. And an idea of how they may have differed from Roman use can be gained if we compare them with the earliest surviving books from Rome itself, the Old Roman chant-books of the eleventh century. We can compare, that is, a Frankish version of the Roman liturgy with a Roman one. A list of the principal differences has been drawn up by Huglo (1954, 'Vieux-romain', 108—9). Some concern details of the text: on some occasions the two versions have quite different texts. Quite often this is a case of selecting a different set of mass chants from the common of saints for a particular individual saint. Other differences touch on the selection of the alleluia, a notoriously unstable genre. Whatever the date when these differences of practice arose, it is clear from the history of the various Roman and Frankish sacramentary types that a certain amount of accommodation between different versions of the chant repertory must have gone on. We can get an idea of what was happening also through the work of Amalarius of Metz (see IX.2).
The ninth century saw, however, not just a consolidation of the Roman chant repertory in Francia but also vigorous enrichment. The chronology of this expansion is not always clear, for notated sources appear first in the tenth century. We cannot always be sure whether we are dealing with relatively new items, or compositions of, say, the second Carolingian renaissance of the time of Charles the Bald (king 840-77, emperor 875-7), or of Charlemagne's time, or perhaps even relics of older non-Roman practice which the Franks were unwilling to relinquish.
The main areas of expansion were the following. I give a reference to the chapter earlier in this book where each genre is discussed in more detail.
1. Whereas the repertory of introits, graduals, and communions appears to have remained relatively stable after the Carolingian settlement, the number of alleluias increased dramatically (11.14). The number of frequently sung alleluia melodies in the Old Roman sources (only three common ones) and Milan (one!) may indicate how many the Franks are likely to have known from their Roman teachers. The repertory of melodies by the end of the ninth century was already about sixty (many used for several different texts) and continued to expand rapidly.
Just as dramatic in its way—although the opportunities for singing it were more
3. Frankish Expansion of the Chant Repertory 519
limited—was the increase in the number of tracts: from twenty-one in the earliest sources to twice that number in the average of the twelfth to thirteenth centuries. Some possible instances of Frankish expansion can actually be seen in the earliest sources. The Old Roman chant-books have Gregorian melodies for the tracts of the Easter Vigil, which should mean that they were adopted from Frankish practice; anciently no tracts were sung here. The tract Eripe me for Good Friday is in neither the Old Roman nor all of the earliest Frankish sources, and was still referred to as 'new' in the tenth century (see Hucke in Hermann Schmidt 1956-7, 932). Domine non secundum, the tract for Monday, Wednesday, and Friday in Lent, is also a latecomer to the repertory.
The choice of offertory verses was also not fixed at the time of the Frankish settlement. Not only this, recent studies have revealed the existence of non-Roman offertories in the Gregorian repertory which may have come from Gallican use (Baroffio and Steiner, 'Offertory', NG, and especially Levy 1984).
2. The origin of the sequences which were sung after the alleluia on feast-days in the Middle Ages is controversial (see the fuller discussion II.22.vi). Whatever the truth of the matter, the repertory at the end of the ninth century probably consisted of only about thirty widely known melodies. Most new compositions circulated only locally. But, as with the alleluia, new compositions were constantly being produced. The total number of melodies known from the Middle Ages is very large, the number of texts vast.
3. The repertory of ordinary-of-mass chants known in Rome was small, and only a few were widely known in Francia. Rapid expansion took place, different pieces being composed in different places. Particularly notable was the composition of trope verses of various kinds. (On the Kyrie repertory and its tropes, see 11.17 and II.23.viii, on the Gloria 11.18 and II.23.iii, vii, and xi, the Sanctus 11.18 and II.23.xi, and on the Agnus 11.20 and II.23.xi.) The Nicene Creed was not sung at mass in Rome in Carolingian times, and the Franks sang it here in continuation of older Gallican practice (11.21.).
4. The provision of complementary trope verses, especially for the proper chants of mass, is one of the most striking developments of the new period. Again, the time and place of origin of the practice are obscure (11.23).
5. The famous Quern queritis dialogue may well have been written in the ninth century, for it was known all over Europe by the end of the millenium. Its composition has nevertheless been placed in the early tenth century by some. The chief compositional activity in the area of liturgical drama took place, however, after the end of the millenium (11.25).
6. It is still not possible to gain an idea easily of the amount of new composition for the office hours accomplished by the Franks. The sources edited by Hesbert in CAO often show striking lack of unanimity in choice of chant, and this variety is matched in the transmission of many musical items. It is presumably the result of conflicting, if already 'Romanized', practices during the eighth century, and the lack of a Roman example against which to control variants. There was naturally a lot of writing for the
520 VII. The Carolingian Century
offices of local patron saints. The chief musical interest of the latter lies in the frequent adoption of a modal order for the chants, a purely artificial arrangement which reflects the Frankish interest in speculative music theory. (See Crocker 1986 for a musical study of such 'numerical' offices.) The writing of verse texts for such offices is another Carolingian and post-Carolingian development (11.26).
VI1.4. THE CODIFICATION OF PLAINCHANT
Treitler 1984; Levy 1987 'Archetype', 1987 'Origin'; David Hughes 1987.
The existence of books containing chant texts has been mentioned repeatedly in previous chapters. The acquiring and copying of service-books and the compilation of new ones was undoubtedly one of the chief concerns of Charlemagne and his advisers. It was part of the great programme of ecclesiastical and educational reform which went to make up the 'Carolingian renaissance' of literacy and learning. The high priest Hilkiah had produced a book of law from which King Josiah learned the ways of righteousness. The written word as a means of ensuring the right performance of the liturgy was obviously of great importance to the Franks. Did their chant-books contain not just chant texts but music-writing? Did the various types of Caroline minuscule script, pioneered chiefly at Corbie in the second half of the eighth century and perfected in Charlemagne's palace school, have a musical parallel or parallels? Where and when was the attempt first made to notate the chant repertory in total (At least for the mass: all the indications are that the office chants were not codified until later.)
I have discussed above (IV.4) the evidence for and against the notating of chant-books in Charlemagne's time. Neither surviving documentary evidence nor contemporary report allows us to state unequivocally that complete chant-books were prepared for Charlemagne. WTe could wish for more manuscripts of chant texts from the ninth century, but those that have survived are all unnotated. Argument a ex silentio are never wholly satisfactory, but the silence is in this case almost overwhelming. The graphic materials out of which a notation could emerge were already present: punctuation and accent-signs familiar to all who had received an education in grammar. From the dates of surviving fragments one might even suggest that a system of music-writing was invented at the palace school in Aachen, or Metz, or Tours in Alcuin's time. The ultimate proof is lacking. And this still does not constitute evidence that the newly learned chant repertory was transmitted as a whole in musical notation.
The chant repertory, even after musical notation became usual in the tenth century, had to be learned. The notation controls (in rehearsal, though presumably not during the service itself) the detail of melodies already known by heart. This implies, of course, that when the melody is well enough known, notation is unnecessary for
5. Music in the Carolingian Renaissance 521
performance. Necessity being the mother of invention, the question is: when did notation become necessary?
The obvious answer is: when the repertory had expanded to a point where it was no longer possible to memorize it all successfully. And this point, I would suggest, was reached later rather than earlier in the ninth century, close to the date of the earliest surviving notated graduals. The problem lay not so much with the 'Gregorian' repertory of proper-of-mass chants as with the sequences, tropes, and ordinary-of-mass melodies being produced in ever-increasing numbers and in musical styles quite different from the formulaic chants learned from Rome. This would also explain why there is no unequivocal sign of a notational archetype from which those manuscripts were descended. Different notational styles had already evolved in different areas. They were now applied to the notating of a repertory still well known but in danger of expanding out of control.
There is a parallel with the situation in the eleventh century, when staff-notation first became widely used for learning (not performing) the repertory (see IV.6 and V.3). The earliest notated chant-books, of about 900, are for the mass. If the date of the Hartker Antiphoner is representative, office-books were not notated for another century. Is it coincidence that the author of the Dialogus tie music a and Guido of Arezzo speak of pitch-notation primarily in conjunction with the antiphoner? Once again it is possible to envisage the expansion of a repertory to the point of unmanageability, necessitating the invention of radically new pedagogical methods.
I therefore hold to the view that musical notation, while it may possibly have been invented as early as Charlemagne's time, was not used to facilitate learning and teaching during the first heroic phase of establishing the Roman chant repertory in Francia.
VII.5. THE PLACE OF MUSIC IN THE CAROLINGIAN
RENAISSANCE
Smits van Wacsberghe 1952, 1970; Manfred Schuler 1970; Iluglo 1975.
While I have expressed doubt that musical notation was used by the Franks as a principal means of mastering the Roman chant repertory, they were certainly concerned with a musical-technical matter of a different nature, namely, the modal system. Aurelian of Reome, writing in the decade before 850, says that Charlemagne added four modes to the usual eight. Byzantine musicians, who had prided themselves on the invention of the original eight, thereupon added another four themselves. While Charlemagne's role is doubtless a pious fiction, the earliest surviving tonary— Pans, Bibliotheque Nationale, lat. 13159, from Saint-Riquier, possibly written as early as late eighth century—shows that the eight-mode system was already understood in Charlemagne's time (see III. 14). It may indeed have come to the West from Byzantium (see VI11.2).
522 V77. The Carolingian Century
As I have argued in V.4, the reasons the Franks were interested in the modes were partly practical and partly intellectual. If one understood the mode of an antiphon, an introit, or a communion, one could choose the appropriate tone for the psalm verse(s) which these chants accompanied. The same is true for office responsories. But the presence of other mass chants—graduals, alleluias, and offertories—in the Saint-Riquier tonary (and some later tonaries) suggests that the classification was important for its own sake, a link with the corpus of classical music theory which was being unearthed by the Franks along with so much of the heritage of antiquity, even though it was by no means easy to reconcile with the Roman chant repertory they were trying to master.
Despite the place of music among the seven liberal arts, the 'seven pillars or steps' of Alcuin's educational programme (PL 101. 853), there is little evidence of original thought about music theory at Charlemagne's court. Beside 'David' (Charlemagne), 'Flaccus' (Alcuin), 'Timotheus' (Paulinus of Aquileia), and 'Homer' (Angilbert, later abbot of Saint-Riquier), 'Idithun' had his place as the instructor of the boys in both practical and theoretical music:
Instituit pueros Idithun modulamine sacro, Utque sonos dulces decantent voce sonora. Quot pedibus, numeris, rithmo stat musica discant.
(MGH (Antiquitates) Poetae Latini aevi Carolini, i. 246; cited by Manfred Schüler 1970 and Huglo 1975). Idithun (Jeduthun) was a musician in the temple of King David (1 Chr. 16: 41-2 and 25: 1; also Pss. 39, 62, 77), and he makes a reappearance in one of those modern psalms, Notker's sequence Sumrni triumphum regis (von den Steinen 1948, i. 240). His identity at Charlemagne's court is unknown. The 'musica' he taught is evidently the knowledge of proportions as exemplified in metrical poetry, as treated in St Augustine's De musica. But he must have been a personage of consequence. Several singers of the court chapel are known to have gone to to high ecclesiastical office: Anstrannus, cantor at Charlemagne's court, became bishop of Verdun in 800; Hucbert, praecantor under Louis the Pious, became bishop of Meaux in 823; Johannes, cantor at the court of Lothar II, became bishop of Cambrai in 766 (Manfred Schüler 1970, 26-7).
Charlemagne's own teacher in the quadrivium was Alcuin. A treatise on music which Alcuin is reported to have composed (Manfred Schüler 1970, 34) appears to have been lost, unless it survives in the short text De octo tonis in musica (GS i. 26-7), taken up in Aurelian of Reome's Musica disciplina (Gushee 1963 and CSM 21).
Whatever the beginnings under Charlemagne may have been, it remains true that the first flowering of medieval music theory belongs to the second half of the ninth century. Aurelian of Reome is a somewhat isolated figure, but the writings of Johannes Scotus Eriugena (d. r.877), Remigius of Auxerre (d. c.900), Hucbald of Saint-Amand (d. 930), and Regino of Prüm are all directly linked to one another and, directly or indirectly, with the entourage of Charles the Bald. Yet the trend which
5. Music in the Carolingian Renaissance 523
seems to have set in under Louis the Pious, a transference of the main intellectual activity away from the court to the great monasteries and episcopal schools of north France and Germany, continued. The place of origin of some important writings— the h'nchiriaciis group of treatises foremost among them—is still unknown. But it is no accident that Hucbald, Regino, and Berno of Reichenau should all have been abbots, distinguished for much more than their contributions to music theory.
Despite the apparent lack of new writing on music theory in the time of Charlemagne himself, it is there that the foundations were laid for a new understanding of what a 'musician' should be. For Boethius there was an unbridgeable gulf between the connoisseur of 'musica' (the science of music theory), the composer, whose understanding of music is by instinct rather than by the power of reason, and the performer, the mechanic (Bk. 1, ch. 34). Although this doctrine was reproduced in modified form by Aurelian, for example (ch. 7), and in Guido of Arezzo's catchy rhyme 'Musicorum et cantorum', numerous musicians combined all three types in themselves with honour. Perhaps the single most important motive for writing about music among medieval theorists was the desire to improve the performance of the liturgy. (Chapter V above therefore concentrates on writing of a practical intent.) Aurelian and the substantial figure of Regino are early examples of practising church musicians who contributed to the theoretical literature. Hucbald was simultaneously singer in the performance of the liturgy, composer, and theorist. These have their forerunner in the unknown 'Idithun', the instructor in both chant and music theory, the paradigm of the new type of medieval church musician.
VIII
Gregorian Chant and Other Chant
Repertories
VIII.1. INTRODUCTION
In this chapter I shall attempt to summarize the relationship of Gregorian chant to other types of early chant in the West, so that Gregorian chant itself may be understood in a rather broader context than would otherwise be possible. Something of the similarities and differences in liturgy between Roman, Milanese, Old Spanish (Mozarabic), and Gallican uses has been indicated in previous chapters (VI.4-5). Now the musical aspects of this relationship are sketched in.
One point of the chapter is to help define the nature of what the Franks learned from Rome and what they recorded in the first music manuscripts, that is, to answer the question, what is Gregorian chant? As is well known, the earliest music manuscripts from Rome itself contain quite different music for the same texts as Gregorian chant, music which is commonly called Old Roman chant. If this is Roman, but differs from Gregorian, in what sense is Gregorian chant 'Roman'?
Old Roman chant is not the only old chant repertory of Italy, and it bears similarities of musical style to the two other extant repertories, namely Milanese and Old Beneventan chant. These provide the musical context for understanding Old Roman chant. The three 'Old Italian' repertories are discussed in VIII.3-5. To understand Gregorian chant, on the other hand, we should at least consider the possibility that the Franks were influenced by the traditional chant of Gaul, that is, Gallican chant. There is an obvious obstacle to this, in that the Gallican rite and its chant were suppressed in favour of Roman use, and practically nothing of the chant repertory survives. But a sister repertory, that of the Old Spanish rite, does survive, albeit in staffless neumatic notation. On this basis we can make at least a tentative guess at the nature of what the Franks suppressed and what may have influenced their understanding of the Roman repertory.
The reader may justifiably have expected greater space to have been devoted to these non-Gregorian chant repertories. There is no doubt that, despite a number of fine studies, they are still not generally well known. The reason for this is the preoccupation of early studies with the music still sung in the Roman church, Gregorian chant. But not only has historical circumstance contributed to their
2. The Influence of Byzantium 525
neglect, some of the non-Gregorian repertories have been denigrated as aesthetically inferior to the Gregorian, an unfortunate and baseless justification of something which had quite different causes. The situation at the moment is that while there are excellent studies of individual genres of Old Spanish, Old Roman, and Milanese chant (such as those of Bailey), which are in many respects superior to anything yet available for Gregorian chant genres, and while the whole corpus of Old Beneventan chant has been discussed in exemplary fashion by Kelly, a balanced survey of all of these is not yet possible in the way that it is possible for Gregorian chant, certainly not in a book of this length. In particular, there is a need to return to Gregorian chant in the light of the new knowledge about the other repertories. This is work that will occupy scholars for many years to come.
During much of the period from the fourth to the ninth century, the eastern Roman empire and its capital Byzantium enjoyed greater material wealth, political power, and ecclesiastical prestige than the kingdoms of the West, and often influenced Western affairs directly, most obviously in Italy, parts of which were long under Byzantine rule. The possibility has often been discussed that the Byzantine liturgy and its chant could have influenced Roman use. A preliminary section (VIII.2) is devoted to this question.
VIII.2. THE INFLUENCE OF BYZANTIUM
(i) Introduction
(ii) The System of Eight Modes
(iii) Antiphons for the Adoration of the Cross and Other Chants in Old Italian Repertories
(iv) The Trisagion
(v) The Frankish 'Missa graeca'
(vi) The Communion Omnes qui in Christo
(vii) The Byzantine Alleluias
(viii) Textual Concordances without Musical Similarity
(ix) The Veterem hominem Antiphons for the Octave of Epiphany
PalMus 5; PalMus 14; Brou 1938-9, 1948-52; Wellesz 1947; Baumstark 1958; Jammers 1962, Musik; Levy 1958-63; Huglo 1966; Jammers 1969; Levy 1970; Strunk 1977, 297-330; Kelly 1989, 203-18; Levy, 'Byzantine Rite, Music of the', NG.
(i) Introduction
At first sight it might seem that a good case could be made out for the likelihood of Byzantine influence on Roman chant. After the fall of the Western Roman empire, Italy was for long periods a territory disputed between Byzantium and the invading Germanic peoples. The reconquest of Italy by Justinian from 533, temporary though it proved, introduced what has been called the 'Byzantine period' in the history of the papacy, lasting over two centuries. Eleven out of thirteen popes between 678 and 752
526 VIII. Gregorian Chant and Other Repertories
were Greek or Syrian by birth. The outbreak of the iconoclastic persecution in Byzantium in 726 resulted in the movement of many Greek churchmen to the West. If there were evidence of widespread Byzantine influence in Roman chant it would be easy to explain. On the other hand, the recurrent political and theological differences which separated Rome and Byzantium may as easily have contributed to the separate development of their chant repertories. Some fundamental differences cannot be overlooked. The Western rites never followed Byzantium in the composition of the vast number of kontakia and kanones which constitute one of the glories of Byzantine music. On the other hand, Byzantium remained rigidly economical where Rome developed a cycle of proper-of-mass chants (Strunk 1977, 316): basically only one entrance chant corresponding to the introit, two offertories, two dozen koinonika (communion chants).
The history of the Roman liturgy—and in the absence of musical documents this is all we can argue from—is one of general independence from the Byzantine rite. The influence of Byzantium is seen rather in individual details rather than overall form and content. It may be that some aspects of Roman and Byzantine chant resemble each other because they derive from a common source—though rarely has a serious attempt been made to demonstrate what this might mean in terms of actual pieces of music (see the articles by Levy). Yet the available musical evidence, like the liturgical, relates to isolated items. Most of these are to be found, moreover, not in the Gregorian repertory but in the Old Italian traditions: Old Roman (rarely), Milanese, and Old Beneventan.
Jammers (1962, 183), in a study that influenced the thinking of several other scholars (principally Stablein: cf. Stablein, MMMA 2, 58*) argued that Gregorian chant was the result of Old Roman chant carried out at papal behest (specifically, that of Vitalian, 657-72, but on the basis of sources no older than the thirteenth century!) to make it more compatible with the Byzantine practice of ison singing, a type of chanting with vocal drone unfortunately not documented before the fifteenth century. (The obvious deficiencies in Jammers's arguments were stated succinctly by Nowacki 1985, 260-1.) Even if there were better documentary evidence for the 'ison' theory, it seems unwise to propose such all-embracing hypotheses about the style of Byzantine and Roman music in the seventh century on the basis of musical sources of the ninth to tenth centuries (Gregorian), the eleventh (Old Roman), and later (Byzantine).
More significant, it appears to me, is Jammers's observation that chant is bound to a text and often shapes itself according to the needs of text declamation. Differences of language have far-reaching musical consequences.
I have already discussed the possible influence of Byzantine on Western notation (IV.4).
Needless to say, in what follows I have not attempted to list each instance of possible Byzantine influence or Western borrowing. (For further references, see the bibliography in Levy, 'Byzantine Rite, Music of the', NG; Brou lists over forty possible borrowings, many of which have not yet been subjected to musical analysis; Jammers 1969 povides another list.) It seems possible to distinguish the following
2. The Influence of Byzantium 527
categories: (a) musical similarities which may go back to the common roots of Christian music in the early church; that this is by no means unlikely might be argued from the analogy of Greek readings in the Roman mass, which persisted into the ninth century (Vogel 1986, 296-7); (b) Western borrowing of texts where no corresponding musical sharing has been established; (c) Western versions of Byzantine chants; (d) new Western compositions with Greek texts for which no Byzantine model appears to have existed.
Byzantine musical influence can be seen to reduce itself largely to a number of individual instances. The overwhelming impression is that Roman chant developed largely independently of Greek models. There is a somewhat larger number of borrowings in the non-Roman Old Italian liturgies, the Milanese and Old Beneventan.
(ii) The System of Eight Modes
Raasted 1966; Bailey 1974; Hucke 1975.
The eight-mode system may have been taken up by the Franks after the Byzantine example. It does not seem to have been part of what they learned from Rome, for it plays no part in the organization of the Old Roman chant repertory. The 'noeanne' formulas of Frankish tonaries, which helped remind singers of the pitch-constellations of various groups of chants, appear to have been borrowed from the intonation formulas of Byzantine chant (see III. 14). It is true that the modal system appears to have exerted influence on the shape of certain chants which otherwise had peculiarities consorting uneasily with it. Yet the Gregorian repertory cannot be said to have been formed from the beginning in conformity with the system. The Byzantine influence is therefore indirect and secondary.
(iii) Antiphons for the Adoration of the Cross and Other Chants in Old Italian Repertories
The antiphons O quando in cruce, Adoramus crucem tuam, Crucem tuam adoremus for the Adoration of the Cross, and some other Holy Week items, were sung both in Greek and Latin in Benevento and some other Italian centres (Wellesz 1947, 21-2, 68-77; PalMus 14, 305-8; Drumbl 1976; Kelly 1989, 207-18). Some have Byzantine forebears. For example, O quando in cruce, found in Beneventan and Ravenna sources, is a version of a Byzantine troparion which can be followed back to the rite of Jerusalem in the seventh century. Its presence in Ravenna should mean that it was already used in the liturgy there before the fall of the Byzantine Exarchate of Ravenna to the Lombards in 752.
These are examples of chants whose sources are principally Italian. Others are the Ambrosian ingressae Coenae tuae mirabili and Videsne Elisabeth, which both appear to be modelled on Byzantine pieces (PalMus 5, 9-13; Levy 1958-63).
528
VIII. Gregorian Chant and Other Repertories
(iv) The Trisagion
Another chant from the ceremony of the Adoration of the Cross, the Trisagion Agios o theos, is sung in the Roman rite in Greek and Latin as part of the improperia. In the Byzantine rite the Trisagion was sung several times every day in various services. Levy (1972) has argued that both Eastern and Western versions rely on a common modal tradition, and work within the same set of notes, without the one being derived directly from the other.
(v) The Prankish 'Missagraeca'
Levy 1958-63; Huglo 1966; Atkinson 1981, 1982'Missa Graeca', 1989'Do.™'.
A set of mass chants in Greek appear somewhat sporadically in Western manuscripts of the ninth century onward. Atkinson's careful investigation of the origin and transmission of those for the ordinary of mass (the others are much less widely disseminated) has led him to propose that they were put together between 827 and 835, after the Eastern emperor Michael II had sent a copy of the works of the Pseudo-Dionysius the Areopagite to Louis the Pious. These writings were translated under the direction of Abbot Hilduin of Saint-Denis. Atkinson believes the texts of the 'missa graeca' could have originated at the same time, which coincided with the special concern for the unity of Christendom detectable in the works of such writers as Amalarius of Metz and Agobard of Lyons.
Atkinson argues persuasively that the melody for the Greek Agnus Dei (O amnos tu then), was composed expressly to complete the set. On the other hand, the melodies of the Gloria (Doxa en ipsistis), Credo (Pisteuo eis en a theon), and Sanctus (Agios) may well be earlier borrowings from the East. In fact Levy believes that the Sanctus represents an ancient recitation formula which underlies the whole Anaphora of mass, the Lord's Prayer, and the Te Deum as well (Levy 1958-63; see 11.18), comparable in this respect with the Trisagion just mentioned.
The rest of the 'missa graeca'—introit, Kyrie, offertory, and communion—appears to be adapted from Latin chants. The mass was usually sung at Pentecost, when two alleluias would normally have been performed, but no special provision for them appears to have been made. Some other alleluias with Greek texts are known, however, from a small number of mostly north French and English manuscripts: Dies sanctificatuslYmera agiasmeni (Brou 1938-9) and a set in Cambrai, Bibliotheque Municipale 75 (Saint-Vaast at Arras, early eleventh century; Brou 1948[-52], 172-6). None of these appears to be based on Eastern originals, but they are certainly evidence of a lively interest in Greek learning, typical particularly of northern French centres in the second part of the ninth century. The 'missa graeca' was later rounded out at Saint-Denis to make a complete set of prayers for the octave of the feast of the abbey's patron saint.
2. The Influence of Byzantium
529
(vi) The Communion Omnes qui in Christo Levy 1970.
The mass of Easter Eve where the neophytes were baptized has very few chants in the Roman rite and consequently in Gregorian chant-books. Levy has uncovered a number of proper chants for the mass in Old Italian sources. One of these, the offertory chant Omnes qui in Christo, can be found in Roman sources as the communion on the Saturday of Easter week. What is more, it appears to be a translation of the Byzantine baptismal chant Hosoi eis Christon, with essentially similar music. Other aspects of this set of chants, particularly their modal unity, lead Levy to suspect roots in the early Christian period before the division of the empire and the cleavage between the main Christian churches.
(vii) The Byzantine Alleluias Thodberg 1966.
Like the Roman and other Western repertories, Byzantine chant has a cycle of proper alleluias. They have a standard set of short 'alleluia' openings, one for each of six modes (the two F modes are not used), with florid verses. Thodberg (1966, 168-95) believes that melodic similarities can be demonstrated between three of these melodies and three of the Greek alleluias sung at Roman Easter Vespers and in Easter Week, whose chants are found in the Old Roman chant manuscripts. Since Latin versions of these alleluias are also known, Thodberg's analysis embraces the Gregorian and Milanese traditions as well (O kyrios — Dominus regnavit decorem, Oty theos — Quoniam deus magnus, Epy si kyrie — In te domine speravi; Snow had already noticed the melodic strangeness of these alleluias in the Old Roman repertory: Apel 1958, 499).
(viii) Textual Concordances without Musical Similarity
Texts common to Byzantine and Western rites have occasionally been identified, without the melodies being recognizably similar. An example is the text of the Holy Week chant Vadit propitiator, which has been identified as part of a kontakion by Romanos the Melode (first half of sixth century). It appears in various liturgical functions with various melodies in the three Old Italian chant repertories: Milanese, Old Roman, and Beneventan. (PalMus 5, 6-9, Hesbert 1938-47: 1945, 73-8, Kelly 1989, 206, 295-6).
530
VIII. Gregorian Chant and Other Repertories
(ix) The Veterem hominem Antiphons for the Octave of Epiphany
Ilandschin 1954; Lemane 1958; Strunk 1964 = 1977, 208-19.
Notker Balbulus' life of Charlemagne, written for Charles the Fat in 883-4, relates that the emperor heard members of a Byzantine legation—according to Handschin the embassy sent to Aachen in 802 by the empress Irene—singing in a service on the Octave of Epiphany, and ordered a Latin version of the chants to be made, matching the original syllable for syllable: this was the series of antiphons starting with Veterem hominem for Lauds on the Octave Epiphany. The Byzantine originals have been identified, although, as Strunk has stated, without Notker's story one would not have guessed that the pieces were the same. At least the mode is the same in both traditions. If such borrowings are to be recognized on musical grounds alone, closer stylistic analysis of the Western antiphon repertory is clearly required.
VIII.3. OLD ITALIAN CHANT I : ROME
0)	Introduction
(Ü)	Sources and Studies of Old Roman Chant
(iii)	Examples: Communions
(iv)	Graduals
(v)	Antiphons
(vi)	Offertories
(vii)	Alleluias
(viii)	Oral Tradition
(i) Introduction
Three Italian chant repertories have survived, from Rome, Milan, and Benevento respectively, which in many respects are quite different from Gregorian chant. There are faint traces of other repertories as well, but their remains are so slight that I have to pass them by. (See Levy, 'Ravenna Rite, Music of the', NG, and Bryant, 'Aquileia', NG.)
The Roman chant is for the Roman liturgy, just as is Gregorian chant, the great majority of their texts being the same. The relationship between these two sorts of chant for the one liturgy is the subject of a special section in this chapter. Since the Roman chant was superseded in Rome by Gregorian chant in the thirteenth century, I shall follow the practice of some other writers and call it 'Old Roman'.
The Milanese liturgy is different from the Roman, as I have already indicated above (VI.4—5). Its chant has survived in use until this day.
Although the sources for the Beneventan liturgy before its replacement by the Roman are few, enough is known to allow us to give it a place equivalent to the Milanese. The chant created for this old liturgy may be called 'Old Benevantan', for it was superseded by the Gregorian.
3. Old Italian Chant 1: Rome
531
The stylistic similarities in the music of these three chant repertories allow us, with due caution, to group them under the designation 'Old Italian', whose idiom differs in certain fundamental ways from Gregorian chant.
(ii) Sources and Studies of Old Roman Chant
Old Roman chant, as already stated, serves the same liturgy as Gregorian chant, but without the liturgical modifications made by the Franks in the eighth and ninth centuries. That means that for the majority of chants of the Roman liturgy, both an Old Roman and a Gregorian version exist. Apart from the musical differences, there are a number of cases where the Old Roman and Gregorian books have a different chant for a particular liturgical occasion: several cases for the mass chants, and many more for the office, where variety between uses is always much greater. Other differences concern the wording of chant texts. Some of the liturgical and textual details peculiar to Old Roman books have occasionally been found in manuscripts of Gregorian chant (Huglo 1954, 'Vieux-romain', Frenaud 1959).
The principal sources of the Old Roman melodies are three graduals (Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319, San Pietro F. 22, and the Martin Bodmer private collection Bodmer C. 74, formerly Phillipps 16069) and two antiphoners (Rome, Biblioteca Apostolica Vaticana, San Pietro B. 79 and London, British Library, Add. 29988). A facsimile of Bodmer C. 74 has been published (Lutolf 1987) and a transcription of Vat. lat. 5319 (by Landwehr-Melnicki in MMMA 2). Cutter 1979 has edited the texts of Vat. lat. 5319 and San Pietro F.22, together with Georgii's (inaccurate) copy of the texts of Bodmer C. 74.
The earliest of these manuscripts is Bodmer C. 74, dated 1071; the latest is San Pietro F. 22, from the thirteenth century.
The first survey of each of the major Old Roman chant genres was made by Snow (in Apel 1958, 484-505). The best recent survey of this type is Hucke, 'Gregorian and Old Roman Chant', A'G. Extended studies of most categories have been completed:
introits: Connolly 1972, 'Introits and Communions'; see also Connolly 1972, 'Archetypes', Connolly 1975, Connolly 1980 graduals: Van Deusen 1972; see also Hucke 1955, 'Gregorianischer', 1956 tracts: Hans Schmidt 1955, 1957, 1958 alleluias: Stablein in MMMA 2, 119*-140*
offertories: Dyer 1971, Kahmer 1971; see also Hucke 1980, 'Aufzeichnung' communions: Murphy 1977
office antiphons: Nowacki 1980; see also Nowacki 1985
great responsories of mode 2: Cutter 1969; see also Cutter 1967, 'Oral', 1970, 1976 psalm tones: Dyer 1989, 'Singing' ordinary-of-mass chants: Boe 1982
532
VIII. Gregorian Chant and Other Repertories
(iii) Examples: Communions
Stated very broadly, and allowing for numerous exceptions, Old Roman chant displays many of the same formal characteristics as Gregorian chant; the text is treated in the same way. The two settings agree on where phrases begin and end, and the settings of those phrases usually have the same shape and tonal character. One finds the same structural elements in both types of chant: intonation figures, recitation passages, cadence figures. The main difference between the Gregorian and Old Roman chant concerns surface detail. Old Roman chant is more ornate. A few examples will illustrate some of the similarities and differences.
Ex. VIII.3.1 is the communion for the mass on Christmas Eve in its Old Roman and Gregorian versions. Most of the Gregorian version can be seen as a recitation around F, with most cadences on D, and the more ornate Old Roman version seems to have the same centre of gravity. In both versions the second phrase rises to a, the third starts on D and ends on F. The last phrase is more adventurous, rising to high c and falling back to D, then up to a again before the final cadence. Some of this is
Ex. VI11.3.1. Old Roman and Gregorian versions of communion Revelahitur gloria Domini (Rome, Bibl. Ap. Vat. Vat. lat. 5319, fo. 10v; Montpellier, Faculte de Medecme H. 159, p. 26)
Old Roman
É
V
Re-ue - la - bi - tur glo - ri - a     do - mi-ni
Gregorian
m9 '*
I
v
Re-ue-la-     -bi - tur glo - ri - a   do - mi-ni
et   ui-de-bit  om- -nis ca-
íp     m   * *    *** 0 *0*0
et ui-de-bit   om- -nis   ca- -ro
-2H
sa - lu - ta- -re   De- -i no- -stri.
I
rgz
0*9    '0*     '»m m 0 «J
-000-m~
sa-lu - ta- -re       De- -i nos  - tri.
J. Old Italian Chant I: Rome
533
entirely typical for a communion in this mode, but other features, particularly the handling of the last line, show a more than casual resemblance between the two versions.
The Old Roman version has many more notes than the Gregorian. It does not have as many D—F or F-D leaps, and they tend to occur in the longer melismas. The whole is less 'gapped' than the Gregorian: it is typical that the Old Roman version descends from the high c by step. The musician who made this version understands the phrase ends to be in the same place, on the same notes, but marks them by melismas, standard for Old Roman communions. (For a parallel example, see Hucke, 'Gregorian and Old Roman Chant', NG, Ex. 1, whose end-phrase 'salutare dei nostri' is identical to Ex. VI11.3.1 in both versions.)
In many places the two versions are almost identical, and there is evidently a close relationship between them. Despite the greater plainness of the one, the verbosity of the other, one would hesitate to say the Old Roman was a decoration of the Gregorian, or the Gregorian a simplification of the other. They are rather two realizations of the same basic idea, one in a rather restrained, the other in a more florid idiom.
(iv) Graduals
The standard phrases seen in Gregorian graduals, tracts, and responsories have counterparts in the Old Roman versions. Rather than cite what would be a lengthy example in both versions, I give the Old Roman versions of two verses from graduals of the so-called 'Iustus ut palma' type. In fact the group would have to have another name in the Old Roman repertory, because Iustus ut palma was not sung as a gradual in Rome. In Ex. VIII.3.2 I give the verses (V// enarrant and In manibus for the graduals A summo celo and Angelis suis respectively. These are three-phrase verses. A comparison can be made with the five-phrase verse in Ex. II.5.1: the three phrases in Ex. VIII.3.2 correspond to the first, second, and fifth phrases in Ex. II.5.1, which end d, ay a, as here.
(v) Antiphons
The standard Gregorian melodies of the office antiphon repertory have their Old Roman counterparts, and the versions are often closely similar—not surprisingly, in view of the simplicity of the melodies in question. Here Nowacki has made the important observation that the two repertories do not always have the same standard melody (see Nowacki 1985, 246-7, Exx. 4-6). Ex. VIII.3.3 is an example of this: Valde honorandus est and Si vere fratres are two members of a large family of antiphons using the melody which Nowacki has labelled D.2.i. The equivalent melody in the Gregorian repertory is la in Frere's discussion (AS 65), Gevaert's theme 6. In the Gregorian tradition, however, Si vere fratres has a different melody, a popular G-mode melody, VIIc in Frere, Gevaert's theme 23. This too is well known
534
VIII. Gregorian Chant and Other Repertories
Ex. VIII.3.2. Gradual verses Celt enatrant and /// manihus portabunt te (Rome, Bibl. Ap. Vat. Vat. lat. 5319, fos. 7r, 41v)
f.7r
a    a   a —r
•   «i »   í-u      ;.Mj itMf. ™ SEE
In  ma-ni-bus   por - ta-
-bunt te
Í--#-'•tQm.mn            ttsT^ m-	-9-31-W0-—-7"3I-^9-
fk-*    * *m—*-*-a—*— *»%—**m—	-••a,   '   '-''a'a'a*   a *--
r---:- glo-      -ri-am    De- -i	et     o - pe - ra manum
-M--'-*>,»#«a-*-**-U*9A *»m m*T\  miM~\-Í9m *m\-m~-	--
&r-s-a—a *ij aaaj—r *9r0-m*m-1	
ne un-   -quam of - fen
-das
ad
;»* p« .t^.9 ' •
-1US
annun-ti-at fir-
-ma-
* w a    ě a?a+ am* a * t*» 9*mam*m-
la-
-pi-dem
pe-
-dem
P
-men - turn.
P
•«H# Cm,
tu - um.
in the Old Roman repertory, Type G. 1 in Nowacki's analysis. Specie tua is an antiphon with the G-mode melody in both traditions.
(vi) Offertories
The Old Roman offertories use two standard types of material (Dyer 'Formula' A and B, Kahmer 'Singweise' 1 and 2) in a way which is not matched in the Gregorian repertory. Formula A is a monotone or decorated recitation which hovers around the semitone step, found in twenty-seven pieces, usually more than once in the same piece. Formula B is a more mobile melodic phrase, and can likewise be used more
3. Old Italian Chant I: Rome
535
Ex. VIII.3.3. Antiphons in Old Roman and Gregorian versions (Rome, Bibl. Ap. Vat. San Pietro B. 79, fos. 34\ 69r, 139v; AS, 62, 142, 664)
Old Roman
m    0    s    '**    * •    *"    *    *    ****   »m 0   0    0    4**   4i ^
-f-^- - * *'t       m ' m
Valde   ho-no-randus est       be - a - tus lo - an - nes       qui su-pra pectus do-mi-ni
P
in ce - na re - cu-bu - it. Sarum
~9~
' 1
Valde   ho-norandus est be - a-tus lo-han-nes        qui supra pectus do-mi-ni
r   *     9     0   0   0 J~~
in   ce-na re-cu-bu-it.
Old Roman
im' *
-4—0-
**0 0,m        S0 ' 0j   +   s   •   1} .
--*0-v-' 04 -4—Hz—0—
Si ue-re fra-tres Sarum
di-ui-tes es-se cu-pi-tis         ueras di-ui - ti-as a - ma-te.
»"?   0   *—0—0-
100
t    *   9   * 0^
-0—»—0-
Si   ue-re fratres
di-uites es-se cu-pitis ueras di-ui -ti-as a-ma-te.
Old Roman
V-^—T	•										
X           'm *		—0—	-#-	4    0    0   0\ m					• •	9    0    0           - »I	
				---- mf	V	if				*    •     • •	
						w	—				
V _ Spe-ci	- e	tu	-a	et pulchritu-di -	ne	tu	- a	in-	-tende	prospe-re proce-de et	reg-na.
Sarum											
v .	4			•   *    a ~							
		—W~	—^—	•          •    J> *0				•	* •	_      ^      M___y_ m__	
FA\ *					#	_				•        0    9    tm   * 0	0 0-
						—^	W			w	
Spe - ci	i-e	tu	-a	et pulchritu - di -	■ne	tu	-a	in-	-tende	prospere   pro-ce-de et	reg-na.
than once in the same offertory; in fact, some offertories consist of little more than repetitions of this same basic phrase. Both formulas appear in pieces in any tonality, though one notices that twelve of the twenty-nine offertories with Formula B have a final on F, two more on C. This is because Formula B itself is strongly F-orientated. Not surprisingly, some of the Gregorian versions of Old Roman offertories with Formula B have an F final.
Ex. VIII.3.4 is the verse Quipropitiatur from the offertory Benedic anima mea in its Old Roman and Gregorian versions. The offertory respond ends on D in the Old
536
VIII. Gregorian Chant and Other Repertories
Ex. VIII.3.4. Old Roman and Gregorian versions of offertory verse Qui pwpitiatur (Rome Bibl. Ap. Vat. Vat. lat. 5319, fo. 47v; Montpellier, Facultc de Medccine II. 159, p. 261)
Old Roman
»•» í*r>   *? éjr*
*a/í*
—wk-
Qui pro-pi-ti - a - tur   om-ni - bus
in-i-   -qui-ta - ti-bus
tu - is
I
et re-
-di - met
de in-te - ri - tu ui - ta
tu
73
•1
qui co - ro-
-nat te
in mi-se-ra - ti - o-
			
--i-^-		v	--Z*<-^-■-S*mw    0 S ímm-=ř-**mY'2           1 ~š^*l
Q) « * -—í*-	O*		-(i   f»    *    £•*•*** »g    w*0M *ufl*-
-ne
et mi-se-ri-
-cor-di
Gregorian___
f*"' »m *tu   mm    »Mi at*     m    •* ^ -
a  a  m  a   '            '"a       "* "      *     "a   a m  m  s M   aM* m'*a —ar
Qui propiti - a- -tur om-ni-bus in - i-qui-ta - ti - bus tu - is
		
		Ii»   a'   a   a   a /a
Im          *• •		M   a*                          a   L #m, a
a #•		W - -w -W-w-
et   re - di-met de in  -  te - ri-tu uitam tu  - am
a^   a*   a   * **m a*a^u í*a
qui  co- ro-nat te
a   (**• f#f  .    _.. iCacJl #H ^   _  «   a9*..*}*} nraM
*s   •   a   a   a a V   * 'a   f, 9*4aB0Z
in   mi - se - ra-ti - o-     -ne et mi-se - ri - cor-        -di - a.
Roman, on F in the Gregorian version, while both versions of the verse end with an imperfect cadence, as it were, on G, after being strongly F-orientated. The formulaic make-up of the Old Roman version is quite clear (marked in the example). The two phrases which usually make up the formula appear in the order ab-ab-a-ab. As usual, the accentuation of the text governs the way the formula is used. In this case it makes no sense to align the two versions. The Gregorian version is set out with the same line divisions as the Old Roman, but there is no corresponding repetition of formulas. The melody pushes constantly up to c\ repetition of small motifs is there, but the vocabulary as well as the form is different from the Old Roman.
j
3. Old Italian Chant I: Rome
537
Ex. VI11.3.5 is the verse Ouoniam angelis suis from the offertory Scapulis suis (endingK in the Old Roman, C in the Gregorian version). In this case the Old Roman version ends on a, the Gregorian on G, just as in Ex. VIII.3.4. The ornate recitation of the Old Roman version is to some extent matched in the Gregorian, in the repeated cs or c-a oscillations.
The Old Roman offertories, like the Gregorian, occasionally have long melismas. Here the two versions again diverge strongly.
Despite the difference in the text of these versions, and the frequent lack of unanimity in choice of verse(s) (which is characteristic of the Gregorian tradition as a whole) the text tradition is in one important respect largely unanimous: the unusual feature of text repetition is common to both Old Roman and Gregorian offertories
Ex. VIII.3.5. Old Roman and Gregorian versions of verse Ouoniam angelis from offertory Scapulis suis (Rome Bibl. Ap. Vat. Vat. lat. 5319, fo. 43r; Montcpellier, Faculte de Medecine II. 159, p. 295)
i
Old Roman
? P. ft ,T. gfr-y^ mu*9m .=
Quo-ni
am     an - ge - lis su-
-1S
P
5£
man
da - uit
de
te
ut cus-
-to - di-
-am
te
■fi^ Zf"
9 n    9 M   9 i
■  r m >
*****
in
om - ni - bus
ui - is
tu - is.
Gregorian
Quo - ni - am     an-    -ge-lis su-
-is manda-
-uit   de te
ut custo-     -di - ant te       ne urn-       -quam offen- -das
'7> jr..
?• u *,^=
ad
-pi   - dem   pe - dem tu-
-um.
538 VIII. Gregorian Chant and Other Repertories
(though not found in the above examples), and occurs nearly always in the same places.
(vii) Alleluias
The ornamented recitation figure of the offertories ('Formula A', Ex. VIII.3.5) is also to be found in some Old Roman alleluias. There are very few Old Roman alleluia melodies—only four are used with any frequency: one in D, one in E, and two in G, of which one is the melody of the alleluias sung at Vespers in Easter Week. This grand series of Easter alleluias, some with Greek texts and some with Latin, has attracted considerable comment, both from Amalanus of Metz in the ninth century and from modern writers (most recently Smits van Waesberghe 1966, Stáblein in MM MA 2, 84*-140*, van Dijk 1969-70). Some of the verses are divided into two parts. In the first part, a recitation in sung on c with frequent torculi cdc. Then the second half is 'announced' before being sung to a different recitation with torculi bca, the torculus we have just seen in the offertory (Ex. VIII.3.5). Ex. VIII.3.6 gives the first verse of Alleluia Paratům cor meum.
Ex. VIII.3.6. First verse oíAlleluia Paratům cormeum (Rome, Bibl. Ap. Vat. Vat. lat. 5319, fo. 89v)
"P-—			oPl- ri—~-	-			r_m^m>_ /*m%.  -*cj._—_
X                     ■ •	■ ě'ě			M*M		»•*   f ě	
				*m			
^ \J-W----■--m							
Pa - ratum	cor me	- um	De-	-us	pa -	ra-tum cor	me- -um.
3*£
•mm*9 *mm  •% **é *mB   w     *     9 '•'mtm9»a fsmď^
Can-ta - bo.
Can-   -ta - bo   et   psalmum di-   -cam do - mino.
There are meagre performance rubrics in the only two musical sources, Vatican 5319 and London 29988 (see the facsimiles in NG i. 488 and this book, Plate 13). These indicate that the primicerii (leaders of the choir) sang the 'announcements' and the schola took up the rest of the verse, presumably singing the first part. The lengthy description of the ceremony in Ordo Romanus XXVII, in particular of the first alleluia, Alleluia Dominus regnavit decorem (Andrieu 1931-61, iii. 363), allowed Smits van Waesberghe to reconstruct a more imposing type of performance, as might have been witnessed in late eighth-century Rome. Alleluia Dominus regnavit decorem shares one more peculiarity with other Old Roman alleluias: an extended repeat of the alleluia after the verse(s), called in Ordo Romanus XXVII the 'alleluia secunda', equivalent to the melodiae of the Milanese alleluia repertory but only rarely found in the Gregorian. The full performance appears to have been something like this:
Alleluia (+ jubilus?) primus scholae cum parafonistis infantibus
Alleluia (+ jubilus?) parafonistae viriles
Alleluia + jubilus subdiaconus cum infantibus
J. Old Italian Chant I: Rome
539
Verse 1*
Alleluia (+ jubilus?)
Anouncement Verse 2, part 1 Announcement Verse 2, part 2
Alleluia (+ jubilus?)
Announcement Verse 3, part 1 Announcement Verse 3, part 2
Request to archdeacon** Alleluia secunda Alleluia + jubilus
subdiaconus cum infantibus
parafonistae viriles
parafonistac viriles parafonistae infantes parafonistae viriles parafonistae infantes
parafonistae viriles
parafonistae viriles parafonistae infantes parafonistae viriles parafonistae infantes
primus scholae
primus scholae cum pafafonistis infantibus parafonistae viriles
*not sung with the recitation formulas and announcements of verses 2 and 3
'same music as announcement
(viii) Oral Tradition
Several writers have pointed out that the Old Roman sources are often in disagreement over melodic details. In particular Cutter (1967, 'Oral') has shown numerous cases of divergence. Nowacki's demonstration (1985) of how changes may have come about in the assignment of typical melodies to particular antiphon texts is another aspect of the continuing oral tradition in Rome. Hucke (1955, 'Gregorian-ischer') had already realized that in some graduals—a generally very systematically organized group—the Gregorian sources transmitted a frozen, archaic state while the Old Roman ones showed evidence of further stylization.
More surprising is the wide disagreement which Dyer (1989, 'Singing') has shown to have existed between the two Old Roman antiphoners in their repertory and assignment of psalm tones. They have over 100 between them—a prodigality which Dyer associates with the survival of solo psalmody in Rome—but only half are found in both sources. (There is no obvious eight-tone, eight-mode system.)
All this disagreement suggests that the sources were notated in the midst of a continuing tradition of oral transmission and performance. In many respects the Old Roman tradition, even on the eve of its suppression in the thirteenth century, never reached the degree of written fixity indicated in sources of Gregorian chant. This is important for our assessment of the relationship between the Old Roman music as we first see it in the eleventh century, Gregorian chant as we first find it in the sources from around 900, and Roman chant as it might have been known to the Franks in the later eighth century.
Much more could (and has been) said about the interrelationship between particular items and between parts of the repertory in their Old Roman and Gregorian versions
540 VIII. Gregorian Chant and Other Repertories
(the reader is referred especially to the penetrating essays by Hucke). Before a summing up of the matter is attempted, the other Old Italian repertories will briefly be discussed.
VI11.4. OLD ITALIAN CHANT II: MILAN
PalMus 5-6; Gatard, 'Ambrosien (chant)' (DACL); Cattaneo 1950; Hucke 1956; Jesson in Apel 1958, 465-83; Weakland, 'Milanese Rite, Chants of, MY:; Baroffio in Fellerer 1972, 191-204; Baroffio 'Ambrosian Rite, Music of the', AG; Heiming 1978.
The Milanese (or Ambrosian) rite is related to, but in many respects fundamentally different from, the Roman. The chief differences in the number and nature of important chants have already been noted (VI.4-5). Despite the frequent incompat-ability of the two rites, some Roman chants were adopted in Milan (see Hucke 1956), and this allows a certain amount of comparative musical analysis. It hardly needs saying that the liturgy and its chant in our medieval sources cannot be exactly that known to St Ambrose (d. 397), whose name has been given to both rite and chant. It seems possible, however, that four hymns by him have survived (see 11.15; also Stablein, 'Ambrosius', A1GG, Cattaneo 1950).
Our knowledge of the Milanese liturgy in its full medieval splendour depends largely on two sources edited by Magistretti: a manual of the eleventh century containing the texts of the complete liturgy (Magistretti 1904-5), and an ordinal compiled about 1125 by the clerk Beroldus (Magistretti 1894). There are summaries by Lejay ('Ambrosien (rit)', DACL), King (1957, Primatial, 287-456, for the mass), Paredi ('Milanese rite', NCE), and Borella (1964).
The chant of the Milanese rite is the only Latin non-Roman medieval repertory to survive in reasonably complete state in transcribable notation. Reports of attempts to suppress it at various times in the Middle Ages appear to be legendary. For example, in the eleventh century the Milanese chronicler Landulphus (De ritibus ecclesiasticis 2. x-xii, PL 147. 855) said that Charlemagne had had an Ambrosian and a Gregorian sacramentary placed side by side on the altar. By the will of God, both opened simultaneously, proving that both were equally authoritative.
Milanese chant survives in a large number of manuscripts (described by Huglo et al. 1956), of which the earliest date from the eleventh century. The earliest comprehensive sources are of the twelfth century:
1. three chant-books for the winter season (Milanese books combine mass and office and divide winter from summer services):
London, British Library, Add. 34209 (facsimile and transcription PalMus 5-6);
Milan, Biblioteca capitolare F. 2. 2;
the 'Varese—Eredi Bianchi' manuscript, privately owned;
2. one for the summer services:
Bedero di Val Travaglia, San Vittore B.
4. Old Italian Chant II: Milan
541
The written musical tradition of Milanese chant, like that of Old Roman chant, therefore begins considerably later than that of Gregorian. An interesting guide to the Milanese chant of the seventeenth century, reflecting the reforms of St Charles Borromeo, was written by Perego (1622).
Two modern service-books edited by Sunol contain editions of the music for mass and vespers respectively, the Antiphon ale miss arum (1935) and Liber vesperalis (1939). These do not always reflect medieval practice. An equivalent volume for the morning service did not appear. Critical editions of the alleluias and cantus (tracts) have been made by Bailey (1983, Alleluias, 1987) and the complete antiphon repertory has been indexed, edited, and classified according to melodic type by Bailey and Merkley (1989, 1990). A survey of the mass chants as a whole was made by Jesson
(1955) , and summary accounts of the complete repertory have been made by Jesson (in Apel 1958, 465-83) and Baroffio ('Ambrosian Rite, Music of the', NG). Studies of the psalm tones have been made by Bailey (1977, 1978), of the offertories by Baroffio (1964), and Hucke has compared Gregorian, Old Roman, and Milanese graduals
(1956) . There are studies of the 'responsoria cum infantibus' by Moneta Caglio (1957) and Bailey (1988). Multiple versions of the antiphons of the ferial office were published in parallel by Claire (1975). (For further bibliography see Baroffio, 'Ambrosian Rite, Music of the', A^G; Heiming 1978 surveys recent liturgical research.)
The simple psalmody of the Milanese office was not organized according to an eight-mode/eight-tone system. The complicated system in the modern Liber vesperalis, with over 150 different cadences, is not authentic. Bailey found only sixty-nine in the early sources, of which only twenty-seven were commonly used, and these corresponded with relatively few exceptions to the thirty chief antiphon melodies used. This suggests a possible model for the Roman system before it was organized on modal lines by the Franks.
Until reliable editions from the early sources and analyses of the different genres have been carried out, only general remarks about much of the office repertory are possible. None of the larger antiphon types has been investigated, such as the long morning 'antiphona ad crucem', the evening 'antiphona in choro', and the numerous simpler psallendae. Discussion of the responsory repertory has concentrated on the elaborate melismas in repeat form of the 'responsoria cum infantibus' or 'cum pueris'. Much fundamental work remains to be done. It is clear, for example, that the various types of responsory ('post hymnum' and 'ad lectionem' in the morning, 'in choro' and 'in baptisterio' in the evening) use typical melodies and verse formulas, within each genre and as a larger group, though less unvaryingly than the Gregorian office responsories, and with a number of irregularities in the matching of a particular verse tone to a particular final in the responsory. But no systematic analyses have been published. Jesson has noted instances of recurrent formulas in the psalmelli (graduals). Bailey has edited and analysed all the cantus (tracts) and shown that all are
542
VIII. Gregorian Chant and Other Repertories
elaborations of the same type-melody, related to the Old Roman and Old Beneventan melodies, which indicates a common parentage.
A formal peculiarity of many Milanese chants—we have seen something like it in the Old Roman offertories—is the repetition of musical phrases or formulas to generate the music for longer verses. This is to be found, as Jesson points out, in numerous non-responsorial chants of mass: ingressae (introits), the 'post evangelium' antiphons, and especially the transitoria (communions: see table in Jesson 1955, 104). The transitorium Te laudamus, for example, uses the same music six times over (Jesson 1958, 478). There is great variety in the degree of ornateness in these pieces. The transitorium Gaude et letare (PalMus 5, 63, PalMus 6, 72), for example—in the simple form AABAABC—is largely syllabic, with a coating of the little curlicue ornaments (often a pes subbipunctus) beloved of the Milanese singers. The splendid transitorium for the Septuagesima Sunday Convertimini omnes, on the other hand, is heavily decorated and includes a lengthy melisma in the final phrase of text. In Ex. VIII.4.1 it is set out in a way which makes clear the repeat structure. As usual, ornamental figures are omnipresent: the torculus subpunctis (pes subbipunctis) appears three times in the A-phrase, twice in C, twice in D, and so on.
Melismas with repeat structure are to be found in many genres of Milanese chant: the psalmellus (gradual), alleluia, and offerenda at mass, the office responsory. Perhaps the best known are those to be found in some office responsories, usually the responsories 'ad lectionem, and 'in baptisterio' of the morning service. These elaborate chants, of which Moneta Caglio (1957) listed forty, often carry the rubric 'cum pueris' or 'cum infantibus'. Some have one long melisma, or melodiae, some have two, the melodiae primae and melodiae secundae. The performance scheme appears to be that usual in responsorial practice in the Roman rite (see II.4), where the responsory is sung twice at the beginning and again complete at the end, but it is conceivable that the repetitions were more numerous on the highest feasts. A melisma may appear within the verse (usually on one of the last syllables). Its primary purpose was to enhance the frequent repetitions of the responsory. The usual place for the melodiae primae was the partial repeat of the responsory. The second melisma is for the repeat of the whole responsory. It is not always clear who should sing the various sections, for the early manuals do not specify the arrangement in full. By analogy with other such pieces one might suggest the following (the practice no doubt varied from feast to feast):
responsory responsory responsory
soloist
magister and boys
verse
schola soloist
responsory second part with melodiae primae responsory
responsory with melodiae secundae
magister and boys schola
magister and boys
4. Old Italian Chant II: Milan 543 Ex. VIII.4.1. Transitorium Convertimini omnes (London, Brit. Lib. Add. 34209, p. 136)
i
i=s-0*+
* m
0-0 *      *T *0 fi*
i
Con-uer - ti c
-nt
om-nes si - mul ad   De - um
*    m*-^
mun-
-do
cor- -de e
et
a-
-ni - mo
1
in o - ra-ti-C
-ne
le - iu-
-ni - is   et  ui - gi - li-is mul-tis
7t
fun - di -
-te
preces uestras      cum la-
-cn-mts
90g09.
-£•—
ut  de - le-
-a-
-tis
cy   - ro-
-gra-pha     pec-ca - torum ues - tro - rum.
-«-#-
0' •
-0-^-f£s 1» M-^-
0*   smi' £;~
Pri-
-us- -quam uo - bis re-pen-ti-
C
,o m^
su-per-ue-ni - at   in - te - ri   - tus.
An-
-te quam uos profun- -dum
-a—zr<-   iTi          /^*m o	—^,  —i-j "3 «a ■'Tfi—=-	
	—-0m* *—*■--—*' 0*0>	
mortis    ab - sor-be - at et  cum cre-a-tor nos-     -ter        ad-     -ue-     -ne - rit
3 i. .__^ ^ ^ K
^ .^/t^^; ft. fm™. I.*, a.*      '< f...
Pa - ra-
1
V->--»—tr-
-tos nos    in - ue - ni   - at.
544 VIII. Gregorian Chant and Other Repertories
If this seems excessively elaborate, it is relatively modest in comparison with the singing of the alleluia at mass on Quadragesima Sunday, the day of the 'farewell to the alleluia' before Lent (reconstructed by Bailey 1983, Alleluias, 25).
alleluia (twice) four boys (in ambo)
alleluia (twice) lectors (in choir)
verse lectors
alleluia four boys (in ambo)
alleluia with melodiae primae lectors
alleluia with melodiae secundae four boys (altar steps)
alleluia with francigenae four boys (in choir)
Moneta Caglio listed forty 'responsoria cum infantibus', of which four are reassignments of mass alleluias. Among these he identified some fifteen sets of melodiae. Bailey's study (1988) makes it clear that they were frequently re-employed for chants other than their original 'mother' chant. The putative origin can be rediscovered by analysing the melodiae, which are usually built up through a process of elaborated repetition of the section of mother chant which they replace and enhance. When melodiae secundae are present, they will be a further elaboration of the germ cell. Six such sets of melodiae can be discerned. Ex. VIII.4.2 shows one of the shorter examples, the responsory 'in baptisterio' at the morning service of the fourth Sunday in Advent, Sperent in te (for a different way of setting out the melisma see Bailey 1988). The second phrase of the respond has the same music as the second phrase of the verse, so that a repeat of the respond from the third phrase seems logical. It is in this phrase that the melodiae are sung. The melodiae begin with 'Querentes te' (A) and end with 'domine' (XYZ), as in the respond. In between come four phrases. I have marked them with letters to show their relationship to each other and to the mother phrase 'Querentes in te':
A QR A-OPQR-W A-PQR-Q A-PQR A XYZ
The A phrases are ways of getting up to c, the tonal centre of the melisma. The repeats within the melisma are clear enough. W is a pendant to the body of those repeated phrases. If Bailey is right, OPQR should therefore derive from XY, Z being a final cadential flourish. And indeed this seems to be the case, except that O is a preliminary phrase, Q an internal extension. Of course, there is no one right way to analyse such a freely flowing melody, only several plausible ones.
The alleluias show improvisatory melodic generation of a different kind, where, as Bailey has shown, the longer repeated sections can themselves be broken down into small cells often following a repeat scheme: aabbcc, and so on. The Milanese alleluia repertory is small. There are ten alleluia melodies and ten verse types. But since two of the ten melodies are musically related to others, there may once have been only eight. Most of the verse types are related to the alleluia melodies.
Each of the alleluias has at least one melodiae for the extended repeat after the verse. Three of them have melodiae primae and melodiae secundae. Melodiae tertiae exist in the shape of the '[melodiae] Francigenae', sometimes supposed to be of
4. Old Italian Chant II: Milan 545
Ex. VIII.4.2. Responsorium cum infantibus Sperent in te (London, Brit. Lib. Add. 34209, p. 22)
jj> c* . .     ft "'.j** p.. p.. &
Sperent in   te     o- -mnes qui no-ue-       -runt   no - men   tu  - urn
quo-     -ni- -am non    de- -re- -lin - ques
_A__ *_v_z<_
L% pi** • **. *m *^t ^,;*. ,r* ,
que - ren-   -tes  te do- -mi-ne.
—-z. *Ts*m -     *******      f***m~*m      M~> ^   £*»        M     tl *^* £* m *^r*
ft     m      N*  #     ** 3*0* *   * '  '*      * 0* ' "* *     9*  * * *
V.   Ti-bi   do-     -mi-ne de
-ctus est pau-
v - - ■
-per    pu-pil-                            -lo tu e-
l-1
-ris       ad- -iu - tor.
Ay_«-__
Que-ren  -  tes te
_O P__6L__R._W
_A_P_6L _^_W
		••••                          »*~^ 'T. '7*         j- i*i
ft «* '-	0 **	*   f *                         •    #•                   #    *f    * 0*                          «J ^0**0
do- -mi-ne.
546 VIII. Gregorian Chant and Other Repertories
Gallican origin, though there is no proof of this. It appears to be a nickname like those of sequence melodies: 'Occidentana', 'Metensis maior', 'Bavverisca'. All the melodiae can be derived from the mother alleluia, the 'Francigenae' from Alleluia VI in Bailey's edition.
The 'Francigenae' are about 240 notes long; Alleluia V is about 110 notes long, its melodiae primae about 240, its melodiae secundae over 320 notes long. But there is space here to show only a much more modest example. Ex. VIII.4.3 gives the melodiae of Alleluia VI (originally an Advent alleluia with verse Venite). The notes derived from the mother alleluia are marked with an asterisk; brackets between the asterisks show where phrases from the other alleluia are left out.
Ex. VIII.4.3. Melodiae of Ambrosian Alleluia VI (London, Brit. Lib. Add. 34209 p. 269)
** + [ ] ******* [
Al-le-      -lu- -ia.
m'\-"T>—s~ -• *m m ——m *m m -        ti 9 - m-wm w • »i—n--m
We do not know how old the Milanese melodiae in their present form may be, just as we do not know the age of the Old Roman melodiae. Is the habit of singing extended repeats after the alleluia verse in some way Old Italian? But it is found in Old Spanish chant as well. Is it old enough to be a possible model for the Frankish sequence? Did the Frankish sequence derive from (now lost) Gallican alleluias of this sort?
There is common melodic ground between the Milanese alleluias, their Old Italian counterparts in Rome and Benevento, and, presumably through Rome, the Gregorian melodies. Bailey believes all ten Milanese alleluias are related to Old Roman and Gregorian ones, and that the two completely surviving Old Beneventan melodies can be matched as well.
Milanese chant, as will have become clear, is often extremely ornate. This often means that possible cases of identity between Milanese and other chants are difficult to assess, especially when comparing the other ornate Old Italian repertories. Is one comparing 'essential' notes or picking out details of little real significance? A basis for
4. Old Italian Chant II: Milan
547
comparison does exist in a number of chants which Milan appears to have borrowed from Rome and which can sometimes be found in as many as four versions: Old Roman and Gregorian, Old Beneventan and Milanese. (Quite often the same piece will appear with different liturgical functions in the different traditions. Well over 100 mass chants with at least partial melodic correspondence between Gregorian and Milanese traditions are listed by Jesson 1955.)
Peculiarities in the handling of formulas, and some cases of very closely similar melodic readings, have led Hucke (1956) to believe that for those Milanese graduals of mode 2 that have Gregorian and Old Roman parallels the Milanese version derives from the Gregorian, not the Old Roman version. It seems likely that this is the case with other borrowings.
Ex. VII1.4.4 gives parallel versions of the introit/ingressa Invocavit me for Quadragesima Sunday, from the Gregorian, Old Roman, and Milanese tradition. The Milanese version has an 'alleluia' phrase at the end not reproduced here. Despite the varied ornamental clothing applied to it, the same melody appears to lie under all three versions. Not that one can identify any precise 'Urmelodie', but a common understanding of tonality and melodic shape is present. We can easily understand some of the surface differences as different conventional ways of making identical gestures. The same conventional cadence appears in the Gregorian version at the end of phrases C, E, and F. The Old Roman version has a different way of making a cadence, used in its fullest form at the end of phrases C and E, in a shorter form for phrase B. The Milanese version has yet another way of doing it, seen for phrases C and F. Some other figures seem to be common ornaments: the figure cbabc in Milan (3, 10, 16), for example, where the Gregorian version has cbc, ccc, cc, respectively. The opposite figure aGFGa appears in the Milanese version at 12, 15, 16, and 18. The Old Roman uses the figure G(a)bcb, a(b)cdc, or b(c)ded six times (7, 10, 12, 19, 25, 35), though this appears in Milan as well (7, 23, 25). If we can see past such conventional figures, the common elements and the differences appear more clearly. The Old Roman version differs from the other two by ending the A phrase on a, descending to G at 9, and starting the D and F phrases on c. The Milanese version stands alone at 12 when it goes for F instead of c. To this extent the Milanese and Gregorian versions form the closer pair.
It hardly needs saying that such melodic analysis must be carried out on a wide and systematic basis. Arguing from single pieces is of little value. But common sense tells us in this and many other cases that we are dealing with simple and ornate versions of the same melody. Since it seems likely that Milan took the introit over from the Gregorian rather than the Roman tradition, we have here a case of a relatively simple melody being transformed when sung in a more decorative idiom. If we believe that the Franks heard the piece in something like the Old Roman version given here, we might consider their Gregorian version to be a less accomplished, 'ersatz' version of something they found rather difficult to manage in its full splendour. There is, however, another possibility. Both the Milanese and Old Roman versions are recorded in sources no older than the eleventh century. Should we reckon with a
548
VIII. Gregorian Chant and Other Repertories
Ex. VII1.4.4. Gregorian, Old Roman, and Ambrosian versions of introit/ingrcssa Invocavit me (Montpellicr, Faculte de Medecine II. 159, p. 86; Rome, Bibl. Ap. Vat. Vat. lat. 5319, fo. 41v; London, Brit. Lib. Add. 34209, p. 147)
G regorian
-fir
r.t
ztsz
A
In - uo-ca-1     2 3 Old Roman
-bit
me
5
8
et
-go ex
8 9
au-
10
-di - am
11 12
um
1',
*'l    ,     l7*'*B   *s,   ?'* m'*,,
0 a
0Z.
-bis
A mbtosian
m      i      *      *'  ' jtj
^i*~*  *   £m*B *»i»*  ;* r.'Cs »7»^-
0 0* 9        1*       a0J0*    *0*  0 0
1--7,-,-	-f---S-0---
fe -O---*—----<^	p> , y* *—« -——
S^i-» -1- -.- /-    e-        -ri-      -pi - am          e- -um ^     15          16         1?     18           19 20 Old Roman	et  glo-ri   -   fi - ca- -bo £j21   22   23       2<t     25 26
lb-*^-»-iii   />a   a***'*^ —i	a   •      •   in t
JA)-»-2-m*0m * °-9-*—•—*—** n Ambtosian	
^——^1«* »» ^i.-j—-^-	--fi—'"m-a£*i ****** <as--^S*-- -(fm**0   9**t  " **-** *£*-W***a-
Efl}-—a*—*-*-**-r»M-*-*—*--*•-1	—»-0-0*.-»-*-«-■-
Gregorian
& Wi . 1	——'**"', r«*>-	,i*   *o--jw-
5?--—— e- -um 27 28 Old Roman	1---w Ion   -   gi-tu-di - ne      di - e- -rum ^ 29       ?0  31   32    33      3*.    35 36	—--*0-a - ad - im - pie- -bo p 37      38     39 <<0
«L--**m	-«--*-•«-.   »t«   M0t'm0iS\ i»,a-	-4-"<  a*7*, -a '"a--
til Ambrosian	_2___> .-k—0-M-- -nem	-a      • *j-r-
	-a*~*t.    *    ** -,-£7*-*r^	if*      *      i*r'-T~£+.-
&-           *' *	_*      w m_                                                       •_a_*--	-*-r-■-■-«0-f0 * 00- 1--«-
Gregorian
35=
e- -um.
M i. Z
Old Roman
Ambrosian_
$   i?™ •*
5. Old Italiem Chant III: Benevento
549
process of accretion of decorative detail between the ninth and eleventh centuries? Did the Romans sing something like the Gregorian version to the Franks, then through further embellishment develop the ornate singing tradition represented by the Old Roman copy of the eleventh century? How old is the ornate style of singing recorded in the Old Roman and Milanese chant books?
Comparison of different Milanese manuscript sources does occasionally reveal small differences of detail, extra passing notes, an extra flourish, the extra repetition of a phrase in a melisma. But the overall level of decoration is hardly altered thereby. Of course, the creation of a written tradition in Milan may well have stopped further embellishment (in the written sources themselves: what was sung may have been another story). But the impression one gains is that the elaboration is not particularly recent. The same is true of Old Roman chant. The differences between the sources are different versions within the same musical style.
Consideration of the third surviving Old Italian chant repertory, the Old Beneventan, also suggests that the high level of decorative detail is not a modern accretion but part and parcel of a general conception of how chant in general should be sung, a conception prevalent in Italy but seemingly foreign to the Franks.
VI11. 5. OLD ITALIAN CHANT III: BENEVENTO
Andoycr 1911-21; Hesbcrt 1938-47; Bailey 1983; Iluglo 1985, 'Beneventain'; Kelly 1985,"'Montecassino\ 1987, 1989; PalMus 14, 15, 20.
Despite the modest amount of music surviving from the Old Beneventan repertory, it has been well served by scholarship, principally by Hesbert (especially 1938-47 and in PalMus 14) and recently by Kelly, who has now provided a comprehensive survey of all aspects of the repertory (1989). A set of facsimiles will shortly appear as PalMus 21.
Of the more than eighty sources containing pieces of Old Beneventan chant, none is a complete chant-book for this repertory, and very many have but a handful of items or duplicate each other. The repertory therefore survives in a very incomplete state. It suffered the same fate as all others in Western Europe save the Milanese, giving way to Gregorian chant by the twelfth century, though not before it was recorded in diastematic notation. Much of it survives in the form of items of special local interest or veneration in the midst of Gregorian chant-books. The two principal sources are Benevento, Archivio Capitolare 38 and 40, in both of which a number of Old Beneventan masses (eight and thirteen respectively) are simply copied after the corresponding Gregorian ones.
Much about the Beneventan liturgy remains unknown, but in some respects—chiefly affecting the arrangement of prayers and lessons of mass—it differs from the Roman, often agreeing with Milanese use instead. For example, the introit chant is called the 'ingressa' and has no psalm verse, as at Milan. There are points of musical
550 VIII. Gregorian Chant and Other Repertories
correspondence with Milan as well, and Beneventan musicians referred to their chant as 'Ambrosian'. It has been suggested that these reflect a period of closer cultural contact between Milan and Benevento, when both were under Lombard domination, although the evidence is not strong enough to speak of a common 'Lombard' liturgical use uniting Milan and south Italy. The Lombard invaders of Italy of the sixth century, gradually converted to Catholicism during the seventh, did not at first conquer Ravenna (Byzantine) or Rome itself. Soon after the Lombard seizure of Ravenna in 751, Charlemagne added the whole of north and central Italy to the Frankish kingdom in 773-4, leaving Benevento as an independent duchy. This survived in one form or another, in the teeth of both Frankish and Byzantine incursions, until its extinction in the mid-eleventh century at the hands of the Normans. As well as Benevento itself, especially the ducal church of Santa Sofia, Montecassino was an important centre of Old Beneventan practice.
The contrast, even conflict, between 'Ambrosian' (Beneventan) and Gregorian chant is illustrated in a poem preserved in Montecassino, Archivio della Badia 318 (Amelli 1913, Cattaneo 1950, 23-6). Charlemagne has ordered that 'Roman' chant be sung in every church. A trial is arranged between two choirboys, singing Roman and Ambrosian chant respectively. After a while the boy singing Ambrosian falls while the other sings on.
Altogether Kelly (1988, 164-5) was able to list sixty-four texts found at least partially in both the Old Beneventan and at least one of the Gregorian, Old Roman, and Milanese repertories. The Milanese repertory, with forty-four concordances, stands the closest, while thirty-nine texts are shared with the Gregorian and thirty-two with the Old Roman. Convincing musical concordances with Old Beneventan chants are fewer: twenty-seven Milanese but only six Gregorian and the same six Old Roman. There is strikingly little uniformity of liturgical assignment, only six items.
Some aspects of Old Beneventan practice stand in contradiction to Roman and Milanese; for example, the Old Beneventan offertories are short, simple pieces without verses (there is some crossing of liturgical boundaries with communions and office antiphons). Then again, in some instances we appear to be dealing with a general Old Italian practice which contradicts Gregorian; for example, the same melody is used for practically all Old Beneventan alleluias.
Like some lengthy Milanese antiphons (an example of a transitorium was given above, Ex. VIII.4.1), some Old Beneventan pieces are composed by the simple repetition of a melodic idea. Kelly gives as examples the especially long Easter ingressa and communion (112-14, 128-9), also three ingressae which share melodic material (117—18). There are seven melodic phrases (A-G) in all, employed in the three pieces as follows:
ABCDEFG   ABCDEFG   ABCDEFG   ABCDEFG ABCDEFG Petrus dormiebat      *   *    ***    *   **   **    *   ** **
Gaudearnus omnes   * *      *#*##        #      # * The last phrase ('G') is the most consistently used.
5. Old Italian Chant III: Benevento
551
Kelly has discussed in satisfying detail numerous items where the Old Italian and Gregorian repertories have the same basic melody in different stylistic manifestations. If in need of a blunt generalization to counterpoint Kelly's fine discrimination, one might fairly say that the Old Beneventan versions are at least as ornate as any other, and often more ornate. They support the broad hypothesis that Old Italian chant favoured a more elaborate style than is found in the Gregorian tradition. What is more, the Beneventan melodies have an older manuscript tradition than the others. Rome, Biblioteca Apostolica Vaticana, Vat. lat. 10673 (PalMus 14) and Benevento, Archivio Capitolare 33 (PalMus 20), which both contain chants of the Old Beneventan Holy Week liturgy in significant quantities, both date from the early eleventh century.
One example must suffice to show two Old Italian ways of embellishing what is basically a simple recitation around the semitone step. The communion chant Hymnutn canite is common to the Milanese and the Old Beneventan traditions but is not found in the Roman. In Ex. VIII.5.1 it is transposed up a fifth from the edition in
Ex. VIII.5.1. Ambrosian and Old Beneventan versions of communionHymnum canite (after Antiphonale . . . Mediolanensis, 225; Benevento, Arch. Cap. 38, fo. 47r)
Ambrosian
f£ p • —*•—«•»—f** '—«p-*t—	-T**-*-f*rt-
Hymnum ca - ni   -  te         ag - ni      mun- -di Old Beneventan _Q-^- -	la - va - cro
$5—p <i  *            —ft,, ,	
-te
ag-   -ni mun-
-di
ua-
-cro
Ambrosian
	__-                                            .             f^-<_                  , . .
&—-»i-	-£**—•-m-*m-i**--"-•—
rv\_9 _92-m--k_€-92-	-• -k-M-,-m-M—m\M-»-
V ( ton-          -tis      re   -   na - ti Old Beneventan -a-——--	sa - ti-a   -   ti         corpo-    -re Chri-sti
kc—é——•—s—	-3   i*o . !Ji^mm—.             **■—f* •
£Ei___—Jt_a-■-0_£-	_92._m.-92._■_0-92.-92.---
-tis   re-      -na - ti
sa - ti - a - ti
cor-po-      -re Chris-ti
Ambrosian
0-#-
Hal - le - lu-jah hal-le- -lu-jah. Old Beneventan
Al - le-
-lu - ía
Al -le - lu - ia
Old Beneventan__
552 VIII. Gregorian Chant and Other Repertories
Sunol's Milanese Antiph on ale; the source of the Beneventan version, Benevento 38, has no clefs, so its pitches are also chosen to match it with the common Beneventan recitation pattern (see Kelly 1989, 107 for an extract and other examples). The alleluias at the end of the Old Beneventan piece are not part of the recitation pattern but a typical ending attached to several chants in the Easter season (for other examples see Kelly 1989, 129 and the table on 115).
VIII.6. GALLICAN CHANT
Leclercq, 'Gallicane (liturgie)', DACL; Gastoue 1939 (1937-9); Porter 1958; Stablein, 'Gallikanische Liturgie', MGG; Huglo in Fellerer 1972 219-33, 'Gallican Rite, Music of the', NG.
The development of liturgies in Gaul was obviously affected by the invasion of Germanic tribes in the fifth century, that is, at a time when forms of public worship all over Christendom were taking on their varied shapes. Visigothic hegemony in the early fifth century was gradually replaced by that of the Franks, who extended their control as far as the Pyrenees in the early sixth century. Their king Clovis was baptized in Reims on Christmas Day 497 or 498. By the mid-sixth century they already ruled an area equivalent to modern France and much of West Germany. Throughout this period liturgical practices developed largely, though not completely, in isolation from Rome, to a certain extent in parallel with those of the Visigothic kingdom in Spain. The principal similarities and differences between the various rites, as far as they affect chants, have been summarized above (VI.4-5). The subsequent adoption of Roman use under Pippin and Charlemagne has also already been outlined (VII.2).
Although we are used to the concept of 'a rite', uniform in form and content wherever it was observed, it is inappropriate for the church in Gaul before the adoption of the Roman rite in the Frankish kingdom. The surviving service-books, rules, synodal prescriptions, descriptions such as the Expositio antiquae liturgiae gallicanae (ed. Ratcliff 1971), and passing comments on the liturgy such as those in the Historia Francorum by Bishop Gregory of Tours (d. 594) make it clear that there were many variations within a general pattern.
A handful of early Gallican service-books have survived (see Vogel 1986, 108, 275-7, 320-6) and these reflect the diversity of usage. Thus the lectionary in Paris, Bibliotheque Nationale, lat. 10863, sixth to seventh century, and the 'Lectionary of Luxeuil', Paris, Bibliotheque Nationale, lat. 9427, seventh to eighth century, have the same system of three readings at mass, but almost never the same choice of lessons. The sets of prayers for mass first published by F. J. Mone (1850), then edited by Mohlberg et al. (1958), differ considerably from those of the so-called 'Missale Gothicum' (ed. Mohlberg 1961), though both follow the same order of service.
By exceptional good fortune some of the surviving books include chant texts:
6. Galilean Chant
553
sixteen for chants after the lessons in Wolfenbuttel, Herzog-August-Bibliothek 4160 (Weissenburg 76), a lectionary of around 500 from southern Gaul (ed. Dold 1936; see also Morin 1937), and the respond texts marked in the so-called 'Psalter of Saint-Germain-des-Pres', Paris, Bibliotheque Nationale, lat. 11947, of the sixth century (Huglo 1982, 'Repons-Graduel'). What may well be Gallican chant texts have also survived alongside Celtic material, for example in the so-called 'Bangor Antiphonary' of the late seventh century (ed. Warren 1893-5; see Curran 1984). The fragments edited by Morin (1905) may be Celtic or Gallican. Several items, distinctive by virtue of their enthusiastic literary style, appear to have survived in Old Spanish or Milanese books. The greater part of what survives with music is, however, to be found in Gregorian chant-books from France. For one reason or another, items from the older indigenous use were preserved alongside the newer Roman repertory.
The remains of the chant repertory of the Gallican rite are so meagre that it is practically impossible to generalize about its musical style. After the pioneering work of Gastoue (1937-9), comprehensive summaries of our present knowledge have been made by Stablein and Huglo.
A further difficulty exists in that the Gallican chants which did survive in late copies may have been edited to make them conform to the 'Gregorian' style. Abbot Hilduin of Saint-Denis, writing to the emperor Louis the Pious (MGH (Epistolae) Epistolae Karolini aevi, iii. 330), said that the office of St Denis included Gallican chants and therefore had to be 'Gregorianized'. W'hen we inspect the music of those chants which seem to be of Gallican origin, it is often difficult, if not impossible, to say what is specifically Gallican about it. That it did indeed sound different in some way is stated by Walahfrid Strabo in De rebus ecclesiasticis, ch. 22, written in the 820s (PL 114. 946). He says it can be distinguished by both its texts and its music ('verbis et sono'). He mentions Rome's 'more perfect science of chant' and the 'inelegance' and 'lack of correctness' in Gallican use. The stylistic contrast between the language of Roman and Gallican prayer texts has often been discussed: the concision of the Roman as against the flowery and colourful Gallican. It is naturally more difficult to translate Walahfrid's remarks into concrete musical terms. In practice one has to look for texts with Gallican hallmarks, preferably with a concordance in another non-Roman use, and musical form or melodic style which differs from Gregorian practice. Analytical techniques are not yet well enough developed for distinctions always to be possible. The so-called 'Gallican cadence' (CDD, CEE, or DEE, EGG, Gaa, etc., often notated with a pes stratus) is not a sure indicator, for although it is not found in the layer of chant taken from Roman use, that is Gregorian chant, it was used very widely in items composed by the Franks from the eighth century onward.
An example of a possible Gallican item is the Palm Sunday processional antiphon Cum audisset. This includes a sentence, 'Quantus est iste', which recurs in an antiphon, Curvati sunt, known from the Old Spanish and Milanese liturgies. Since Cum audisset is not found in Roman books, the probability exists that it is a Gallican
554 VIII. Gregorian Chant and Other Repertories
piece, like several other processional antiphons (cf. 11.9). The recurrence of a number of openings and cadential figures is notable (Ex. VIII.6.1: the scribe has given an alternative text on lines 5—6, presumably not for singing).
The majority of surviving Gallican chants are to be found for occasions when Roman use provided no model. Chants for the votive masses which were added by the Franks to the Roman mass-books may provide the occasion, as in the introit Prosperum iterfaciat (ed. Stablein in MGG), found in the Aquitanian gradual Paris, Bibliotheque Nationale, lat. 776 (11th c.) for the mass for those setting out on a journey. Another group is constituted by the 'preces', originally a litany at the end of the first part of the Gallican mass, reassigned in Gregorian books to the Minor Litanies of the days before Ascension. This custom of singing litanies in procession was taken over by the Roman church from the Gallican. Huglo (A7G) has listed twenty-four possibly Gallican items; examples have been edited by Stablein (MGG) and Huglo (1955, 'Preces'). The relationship between these litanies and the Kyries with Latin verses has already been discussed (11.17).
On the other hand, not all additions to the imported Roman repertory can be assumed to be Gallican. Gastoue (1938, 10-11), Stablein (MGG, 1307) and Wright (1989, 47-8) thought that the E-mode melody for the Gradual ihigelis suis must be Gallican. While the Franks may well have added this gradual to the mass on Quadragesima Sunday (see AMS, L-LII), they simply used the usual Gregorian formulas for E-mode graduals.
Some Gallican items appear to have been preserved as an enrichment of Roman use. The Exultet and hymn Inventor rutili for the Easter Eve liturgy are examples of this. Others turn up in the liturgies for local French patron saints: St Denis, St Maurice, St Remigius. For example, the mass for St Dagobert, Merovingian king of the Franks, preserved in books from Saint-Denis, includes an antiphon 'ante evangelium', an item of the Gallican mass but not the Roman mass (facsimile and transcription by Stablein in MGG). The presumption is that the music is Gallican, though the precise ways in which it differs from, say, many longer office antiphons in Gregorian books have not yet been distinguished sufficiently clearly.
An important group of possibly Gallican pieces is being uncovered in the offertory repertory. The Gallican sonus accompanied a procession of some splendour, apparently modelled on Byzantine use. Chiefly on grounds of musical style, Gastoue had already believed the offertory Memor sd dominus to be Gallican (ed. Stablein in MGG). This piece appears without verse in the Old Roman tradition. The non-Roman offertory Elegerunt apostoli survived in Gregorian books for St Stephen's Day in the face of competition from the Roman In virtute. On grounds of textual and in part musical concordances with Old Spanish and Milanese chants, Levy (1984) has added more offertories to this group. In most of these cases there are versions in the Old Roman repertory. Acceptance of a non-Roman origin appears to mean that their musical style was 'Romanized' to make the Old Roman versions we know.
Ex. VIII.6.2 gives part of the offertory Sanctificavit Moyses (see also the discussion by Baroffio 1967) in Gregorian, Old Spanish (with staffless notation only),
Ex. VIII.6.1. Processional antiphon Cum auclissel populus (Paris, Bibl. Nat. lat 776, fo. 54v)
zee
•2    0 m*9U»m •=
Cum au - dis-     -set  po- -pu-
»* 0
-I us
qui-a Ihesus ue - nit Ihero-so- -li-mam
• •**0't
ac-ceperunt ramos pal-ma-rum
et   e - xi-e-runt   e-i
ob-ui - am
i*—•—i'-*--* 0*—•-
0*      *0 0
*-•-*      0      '-00*    ' 0*0
et clamabant pa - ri - ter di-centes
hie
est   qui uenturus est
*      * \*.»9*t0*-
-0-0-
est    salus nos-ti
ad sa-lu-    -tern genti - um
hie
t*l**0  0t   0-r
-**-'-•   0mt* «j
0 *
et redempti
Is - ra - hel
quan - tus est is - te cu-i tro-ni qubn - txs fst ks - tf     cx-k trp-nk
—0-* —0-9-9---r-"-
et do-m:-na-ci - o - nes oc-currunt ft dp-mk-nbek - p - nfs pc-cxrrxnt
0—0
hie   est    rex  Is - ra - hel
*p *' % v,f''^
He-ma-nuhel
-0*0-1
et
ge - nus     Da - uid
-0—0*
ex mundi    sal-ua - tor
No-li ti - mere   fi - li - a Si
9ittm fo-
se
m0*~0 £0      00 00
0   0*--*0 **0
ec-ce rex tu - us  ue - nit   ti - bi
-dens
-»—■—-—e-
*9—0
35
9-9-
super pullum a - si - ne
sicut    scrip-turn est
sal-
-ue
rex
fabri-cator mun-di
0        • 'v , 0 ,*0=
qui ue-  -nis-ti     re-di-     -me-re nos.
556 VIII. Gregorian Chant and Other Repertories
Ambrosian, and Old Roman versions. The text is a free compilation of parts of the book of Exodus. The Gregorian and Old Spanish versions have the respond Sanctificavit; after the verses there is a repeat from 'Fecit sacrificium', introduced by a special new phrase 'Tunc Moyses'; there follow the (very long) verses Locutus est and Oravit Moyses (Ott 1935, 114; Montpellier, Faculte de Medecine H. 159, ed. Hansen, no. 918). The Old Roman version in Rome, Biblioteca Apostolica Vaticana, Vat. lat. 5319 (MMMA 2, 350) splits the verses into four verses each, eight in all. Only the last phrase of the respond 'In conspectu' is given for the repeat. In Milan the first part of the verse Oravit is the respond, the verse being formed by the rest of Oravit plus what was the repeat in the Gregorian version, 'Tunc Moyses fecit sacrificium'.
If the offertory had come to Rome from elsewhere, one would have expected the Roman musicians to apply their usual formulaic method for providing music (see VI11.3). To some extent in the verses they did so, and for these parts of the verses there is no relation with the Gregorian version. The Milanese version is generally not close to the Gregorian (Baroffio 1967 interprets it as using a basic melody several times over in varied fashion). In so far as the Old Spanish neumes can be properly understood, they appear to indicate a version much closer to the Gregorian, at least in places. Ex. VI11.6.2 shows the close of the second verse, where both Gregorian and Old Spanish versions have a melisma in AAB form (the Old Spanish actually AABB'BB'CCD). The Old Roman version uses the end of a recitation formula for 'Quia ego sum Deus' and the full formula for the last phrase, perhaps evidence of Roman moderation as opposed to Gallican rhetorical display.
Ex. VI 11.6.2. Part of offertory Sanctificavit Moyses in Old Spanish, Gregorian, Ambrosian, and Old Roman versions (Antifonano visigotico, fo. 305'; Montpellier, Faculte de Medecine II. 159, p. 256; Antiphonale . . . Mediolanensis, 333; Rome, Bibl. Ap. Vat. Vat. lat. 5319, fo. 13()v)
OR
A			-i	a y	X	in	.4	J J   n   JJ j .
								S~m       ^P; , ,_---._
/                 ' *								
			*		* «2		•	
	•#	m		—w-		—w-		a a    * *m m
V fa -	- ci	- es	au-	-tern	me-	-a	non	ui - de - bi - tur    ti - bi
								
	*.							
rm *	—Z.			i*	'*» * *m.		#	*   *         m m~
\¥-—	# #	'»	•		W W	—#-		9         *m #
-8-TT*								
/L *	w	•	V	1	» «m		M* *	
								
								
-em non   ui-   -de - bis
7. Old Spanish (Mozarabic) Chant 557
a
os  j - j a J f   r       j j j    . 3 a v
Gr
P
qui - a    e - go   sum   De-      -us os - ten-dens mi-ra-bi- -li-
AM
•   •  *   a &
ORp
-•-
m*h  l,     —    .. .-. tan   b« ~
Gr ffT)      j~»      > * y *»   V« j*l   tau.^ • m  LV#  -0f #>_-
in      ter- «*»
OS
Gr
-ra.
-ra.
in ter-
(-ra.)
in   ter- -ra.
VIII.7. OLD SPANISH (MOZARABIC) CHANT
Liturgy. W. C. Bishop 1924; Prado 1927, 1928; Cabrol, 'Mozarabe (La liturgie)', DACL; King 1957, Pnmatial; Rivera Recio 1965.
Chant: Wagner 1928, 1930; Rojo and Prado 1929; Brou 1947, 'Psallendum' and 'Etudes', 1948, 1950, 1951, 'Alleluia', 1951, 'Sequences'; Husmann 1958-61; Huglo
558 VIII. Gregorian Chant and Other Repertories
1964; Pinell 1965; Brockett 1968; Randel 1969, 'Psalmody', 1969, Psalm Tones, 1973, 1977, 1985, 'Mozarabic Rite, Music of the', AG; Levy 1985.
Facsimile: Antifonario visigótico (León, Archivo Catedral 8); Antiphonale Hispaniae (Saragossa).
I have called 'Old Spanish' the chant sung in what is now central and northern Spain and Portugal up until (and in some instances beyond) the imposition of the Roman liturgy and concomitant Gregorian chant in the eleventh century. It is often called 'Visigothic' or 'Hispanic', but mostly 'Mozarabic'. The Visigoths, who ruled most of south-west France in the fifth century before being ousted by the Franks, brought all of Spain under their control in the sixth. Our chief source of knowledge of the liturgy in the Visigothic kingdom is the De ecclesiasticis officiis of Isidore of Seville (d. 636), with additional information in his Etymologiae and the records of the Fourth Council of Toledo (633) held under his leadership. In 711 the Visigothic kingdom, with the exception of the extreme north-west, was conquered by the Omayyad Arabs. Christian worship nevertheless continued; the word 'Mozarabic' refers to Christians living under Muslim rule. Although a 'Spanish mark' was established under Charlemagne, the Christian reconquest of the peninsula did not get under way until the eleventh century, a major landmark being the conquest of Toledo in 1085. Northern liturgical practices followed northern arms into Spain, not least through Cluniac influence over the monastic houses of northern Spain. Pope Alexander II (1061-73) took a strong interest in the reconquest, and his agent Hugo Candidus was energetic in the replacement of the Old Spanish (Mozarabic) rite. Alexander's policy was continued by Gregory VII (1073-85), prime advocate of papal supremacy. Bernard, appointed archbishop of the newly regained Toledo, was French: he had been abbot of Cluniac Sahagun. Some churches in Toledo were nevertheless permitted to continue with the old rite. (As usual, reports of various legendary 'trials' of the rival rites—in this case by duel and by fire—have come down to us: see King 1957, Primatial, 510.)
How much of the chant for this rite survived thenceforth in its early medieval form is not clear. In 1500 and 1502 respectively Cardinal Jimenez de Cisneros published a missal and breviary of a revived Old Spanish rite. The chants differ considerably from any known early sources. There was a further revival, with new chant-books, at the turn of the eighth century under Cardinal Lorenzana, and the Old Spanish rite is still celebrated in one of the chapels of Toledo Cathedral.
Over twenty major sources of Old Spanish chant exist (see Pinell 1965; forty-six sources altogether are listed by Randel 1973, though some contain only one or two pieces). Yet barely two dozen melodies were copied in the transcribable Aquitanian notation imported from the north (written over erasures in Madrid, Academia de la História, Aemil. 56). The significance of the various types of Spanish neumes is generally clear, but the pitches of the chants are irrecoverably lost. Despite these obstacles, a surprising amount is known about the repertory, chiefly through the work
7. Old Spanish (Mozarabic) Chant 559
of Brou and of Randel, whose catalogue of 1973 is a complete list of all known chants, indexed according to liturgical tradition and genre. Randel has also completed penetrating studies of the office responsories and psalm tones. This brief account is chiefly based on his writings.
Like the Gallican rite (in so far as it can be known in any detail), the Old Spanish rite was host to a variety of liturgical traditions, of which four are generally distinguished. They observed the same structure of office and mass but often choose different chants or, in the case of concordances, have different musical variants. One tradition centres on Leon, one on the upper Ebro valley, the area known as the Rioja, and northern Castile, including the abbey of Santo Domingo de Silos. The Rioja tradition, albeit with variants, was also known in Toledo, where it is known as Toledo 'tradition A'.. The Toledan 'tradition B' is restricted to only three sources, but it is this that provided the basis of Cardinal Cisneros's revision in the fifteenth to sixteenth century.
As well as the liturgical differences, two different styles of notation were used, the one upright, used in Leon and Castile, the other heavily slanted, used in Toledo and perhaps some other centres (see IV.2). The Toledan notation can itself be divided into two traditions. One employs more rounded forms, and was used for the liturgical tradition 'A'. More jagged shapes were used in conjunction with tradition 'B'.
Apart from the documents of St Isidore's time, two early service-books of the period before 711 have survived, a 'Liber ordinum' containing the order of service (ed. Ferotin 1904), and the collection of office prayers known as the Verona 'Orationale' (Verona, Biblioteca Capitolare LXXXIX, ed. Vives 1946), particularly important because chant texts are cued in the margin. Since most of these chants appear in later sources, it is assumed that many of the texts of the liturgy were already established before the Moslem conquest. The earliest books with notation are usually dated in the tenth century, but most come from the eleventh. The early books combine office and mass formularies, like the Milanese books. One of the earliest, from Leon, has been published in facsimile (Antifonario visigotico). It contains two prologues, the second of which describe the practice of responsorial and antiphonal psalmody (see Randel 1969, 'Psalmody').
In two important respects Old Spanish chant can be shown to ally itself with the other non-Gregorian repertories and against Gregorian practice. Randel (1977, 1985) has shown that for the antiphonal psalmody of mass and office (the two seem indistinguishable) four tones account for the great majority of surviving examples (admittedly meagre in number). His study of responsory tones in the various traditions also demonstrates that the Gregorian system of eight modes and eight tones did not exist. The Leon tradition has seven, with four being particularly common, and the Rioja tradition also knew these. Toledo tradition A, although apparently transmitting the same responsory melodies, had a different system of assigning verse tones, and the tones themselves are different, two being particularly common. Toledo tradition B has four tones, one common.
560 VIII. Gregorian Chant and Other Repertories
Randel has also discerned the presence of standard phrases in responsorial chants, in the vespertini at the start of Vespers. Type-melodies, on the other hand, underlie the repertory of threni in Lent.
Several different types of chant sometimes include long melismas with repeat structure: the soni of the morning and evening services, the clamor, laudes (alleluia), and sacrificium (offertory) of mass (facsimiles and copies in Brou 1951, 'Alleluia'; see also Husmann 1958-61).
It is frustrating that the two chant repertories which were presumably closest to what the Franks knew before the adoption of Roman practice, their own Gallican chant and that of the sister liturgy in Spain, should both be inaccessible to us. Possible direct links between both of them and the Gregorian repertory were mentioned in the last section, where Gallican offertories were discussed. There is one other way, however, in which the Old Spanish chant repertory might shed light on the origins of Gregorian chant. I have drawn attention to the generally ornate style of the Old Italian chant traditions, which sets them apart from the Gregorian. In this respect Old Spanish chant is more akin to Gregorian. Perhaps that is another small indication that Gregorian chant is the product of non-Italian musicians.
VIII.8. CONCLUSIONS
van Dijk 1961, 1963, 'Papal, 1963, 'Gregory', 1966; Stablein in MM MA 2; Hucke 1955, 'Gregorianischer', 1956, 'Gregorian and Old Roman Chant', NG, 1980, 'View'; Nowacki 1985.
In the last section it was again remarked that a system of eight modes and eight tones which complement them for the singing of office psalms and responsory verses was unique in the West to Gregorian chant. The eight-tone arrangement, and the smaller number of psalm-tone endings by comparison with Old Roman chant, have suggested to Dyer a connection with the change from solo to choral psalm-singing, a novelty of Carolingian practice. This is one among many radical changes brought about by the Prankish church. Others are the obligation of all clerics to celebrate the full round of office hours, the filling out of the Temporale and Sanctorale, and the enlargement of the chant repertory with sequences and tropes. The need for written records to support the huge edifice thus constructed led to the compilation of tonaries and the use of musical notation. All this is Frankish innovation.
It still remains, however, to try and sum up the relationship between the two chants for the Roman liturgy, that is Gregorian and Old Roman chant. On the one hand, the repertory codified by the Franks at the end of the ninth century (earlier according to some) is the Gregorian. On the other hand, the only chant-books written in Rome itself, from the eleventh to the thirteenth century, have Old Roman chant.
Several scholars (notably Smits van Waesberghe, Jammers, and Stablein) have argued that Gregorian chant was created in Rome itself and coexisted there with Old
8. Conclusions
561
Roman. The liturgist Stephan van Dijk argued in a series of articles that Gregorian chant was that sung by the papal schola for the special liturgical use of the papal chapel. Old Roman chant was that of the city churches. But, while the special practice of the papal chapel is well documented, it is by no means clear that a different way of singing should have been created to accompany it. While some chants would be unique to one liturgical use or another, the great majority of chants required would be common to all churches of Rome. We are speaking of variants upon the one Roman liturgy, not differences of the sort that divide Roman from Milanese or Old Spanish use. Nor does it seem likely, on the face of it, that the Gregorian repertory, which is rather sober and simple when compared with Old Roman, should have been created as part of an attempt to imitate 'the majestic court ceremonial of Byzantium', as van Dijk put it (1966, 302). The example of Benevento is important here. Old Beneventan chant and Gregorian did indeed exist side by side, but as representatives of two different rites, the aboriginal Beneventan and the invading Roman respectively. Given the eventual dominance of the Roman rite, Old Beneventan chant was bound to disappear.
Much work has still to be done to understand the musical relationships between Gregorian and Old Roman chant. This has to proceed at two levels. Where there is no demonstrable musical relationship, as for example in many offertories, one needs to ask why this should be so, which involves the history of the liturgical practice of that genre of chant and the particular service of which it is part. Where a musical similarity exists, what is common and what is dissimilar has to be explained, involving minute analysis of idiom and form. In VIII.3 above a number of instances were cited or exemplified, where the same musical idea was realized in different ways, each with its own characteristic turns of phrase. Hucke (NG), speaking of graduals, and seeing the Gregorian versions as 'consistent adaptations of the Old Roman ones', has called this 'a translation from a foreign musical language'. Behind this lies the conviction that two different peoples were concerned, the Franks and the Romans, not two groups of musicians within Rome itself. By extension, one should then ask into which language was Roman translated? (To which the only answer is a type of Gallican chant). The evidence to prove or disprove this thesis is, however, lacking. While analysis can establish what constitutes a Gregorian idiom, parallel Gallican examples are lacking, and it seems unlikely that Old Spanish chant will provide an adequate substitute.
Since the codification of the Gregorian repertory preceded that of the Old Roman by some 200 years, it was possible for the Gregorian tradition to influence the Old Roman at the time of its writing down. There are signs that this happened. Not only were Gregorian pieces taken up in Rome itself (for example, the cantica of Holy Saturday, numerous alleluias, not to mention Frankish sequences, alleluia prosulas, ordinary-of-mass chants and the like). There are peculiarities in the readings of some Old Roman chants which suggest 'contamination' by Gregorian versions (Hucke 1980, 'Offertorien'), as if the Old Roman notator had a Gregorian chant-book to hand. (As Hucke points out, notation must have first become known in Rome through the medium of Gregorian books.)
562 Mil. Gregorian Chant and Other Repertories
Given the evidence of a continuing oral tradition in Rome even past the time of the earliest notated books, it seems to me impossible that they should represent in all details the melodies sung to the Franks in the eighth century. Since Beneventan and Milanese chant is generally as ornate in idiom as Old Roman, I think it highly likely that the Roman chant of the eighth century already had an ornate idiom, in other words, that Old Roman chant preserves the spirit, if not always the letter, of the eighth-century state. To this extent, the oft-quoted remarks of Johannes Hymmonides (John the Deacon: PL 75. 90, Stäblein in MMMA 2 143*, van Dören 1925, 53) carry conviction: the clumsy Frankish singers were too inexpert to manage the fine detail of the Roman melodies. The theme was later taken up by Adhemar of Chabannes (d. 1034) who refers to the 'tremulas vel vinnolas sive collisibiles vel secabiles voces' of the Romans, in so far as one can understand these terms (Chro?iicon, II. 8, ed. Chavanon 1897, 81; cited van Dören 1925, 53, Hucke 1954, 'Einführung', 181). One would not call the Gregorian melodies clumsy, but it seems conceivable that the Franks sang the Roman chants in a more straightforward idiom than the Romans themselves, something perhaps more akin to that of their native traditions. That is a matter of surface detail. As regards learning the standard formulas, type-melodies and the methods of using them, the Frankish singers were evidently very successful.
IX
Persons and Places
IX.1. INTRODUCTION
This chapter will inevitably seem incomplete to those seeking a full and balanced account of the history of plainchant in all the various churches of Western Europe during the Middle Ages. At present, however, it is not possible to write such an account, for investigation of the various chant traditions has so far proceeded very unevenly, and full coverage would in any case exceed the proper limits of this book. I have therefore ventured a few remarks only on those persons and places, those musicians, churchmen, and churches, that have attracted particular attention so far, or that I myself have found particularly interesting. The purpose of the chapter is partly to point readers in the direction of information they may be seeking, but partly also to give an idea of the sort of research which has been carried out, its aims and methods.
Individual creation, original composition, calls for individual appreciation. But before one can decide what is individual in a particular manuscript, a great deal of patient comparison of sources has to be carried out. Scholars have developed a number of short cuts, as it were, to facilitate comparisons and highlight similarities and differences between sources, and some of these are repeatedly mentioned in what follows. Some are repertorial: the choice of alleluias at mass for the Sundays after Pentecost and of responsories for the Sundays of Advent; others concern musical differences, variants in the detail of melodies (these methods are outlined briefly above, III.15). Such comparisons are naturally only a first step in the evaluation of sources, a means of finding one's bearings in an ocean of evidence, on a veritable mountain of manuscripts. Yet, for many sources little progress beyond these first steps has been made. I hope, therefore, that the reader will forgive the piecemeal approach and gain some benefit from these brief visits to some of the more interesting corners of chant history in the West.
From time to time readers may well find themselves asking, what was chant like at St Gall, at Cluny, at Montecassino? The answer is that in most respects it was like chant anywhere else. Rarely does the repertory of a church stand out by reason of items which are unique in style, or even items which are unique but similar to others. And the small differences in detail which help distinguish one manuscript from another are of bibliographical rather than purely musical interest. No doubt, if one
564
IX. Persons and Places
had been able to visit the places in question, one would have been able to distinguish the singing of one medieval schola from another. But it would probably be by virtue of aspects of performance such as voice production, tempo, and dynamic, none of which is reflected in the manuscript sources that have come down to us.
The first section in this part of the historical survey is about Amalarius and Metz in the second quarter of the ninth century, which continues the story from Ch. VII and illuminates some of the consequences of the Carolingian settlement. St Gall (IX.3), Benevento and Montecassino (IX.8), and Limoges (IX. 10) are mentioned as three examples of places which made particularly significant contributions to the chant repertory.
These sections concern important historical developments in the chant repertory. But William of Dijon's bringing of chant to a previously devastated Normandy (IX.5), and the chapters on chant in England, Spain, and north and east European countries (IX.6, 11-12), concern the implantation of a previously formed body of music in new territory. Only in the case of Normandy and England have I entered into detail, in order to explain a type of development which was repeated, with local variations, in many other areas. For the last two sections I have limited myself largely to mentioning recent research.
The bringing of a chant repertory to any of these new lands mirrors the Frankish importation of Roman chant. There was a political dimension to almost all these enterprises, in that the local ruler practically always took the initiative in deciding who should come and initiate, restore, or reform worship in his domain. The initiative of Pippin and Charlemagne is echoed later by Richard of Normandy, who brought William north from Dijon, by Edgar of England, and so on.
It may surprise some readers that some obviously important centres such as Chartres and Paris, or Weingarten and Regensburg, have not been singled out for special mention. The liturgy and chant of Paris in the twelfth to thirteenth centuries, for example, has been the subject of much intensive study (most recently Wright 1989), largely because of the polyphony sung there. Such studies could be and in some cases have been made for dozens of churches across Europe. There is, alas, simply no space to report them all here, though I have indicated the more important editions of liturgical books or particular repertories where possible in previous chapters.
Map IX. 1 shows the archdioceses of Latin Christianity towards the end of the Middle Ages. Because of numerous reorganizations of church administration it is strictly accurate for no particular date. To some extent the boundaries reflect the national divisions of the early Middle Ages. Thus the archdiocese of Lund included Denmark and what is now southern Sweden, like the medieval kingdom of Denmark. The diocese of Riga is a product of the settlements of the German orders of knights. The Holy Roman or German Empire included in the west the dioceses of Cologne, Trier, Besangon, Vienne, and Aries, in the east Magdeburg, Prague (for the kingdom of Bohemia), and Salzburg. Further east lay the Polish and Hungarian archdioceses. During the Middle Ages Latin Christianity was still expanding eastwards, and Spain
/. Introduction Map IX. 1.1. The archdioceses of late medieval Europe
565
566
IX. Persons and Places
was being reconquered from the Moors, so that the Portuguese diocese of Lisbon, and the Spanish ones of Santiago and Toledo, moved correspondingly south. The archdiocese of Seville was only established in 1248, that of Granada only in 1492.
Maps IX.2, 3, and 4 indicate some of the more important churches from which chant sources have survived, in Italy, Germany, and France, respectively. The maps are intended to provide some orientation during the rapid course of this chapter.
Map IX. 1.2. Ecclesiastical map of Italy
A -Trent
^-- co",(,i+m J? ilauileia
\ tMonztl // /—^-A
T     f K^yA^   A____„        ) \
bologna t    7^ % SX
/'   ^\ Forlimpopoir   * Ravenna V^'S^v
I % Arc::!)
\ t Cubbw
X
tN
ore in
Ragusct >ubn
1.
a Rome
'\      t   % Bencvenlp
Rnri
{ Archbishopric J       Palermo /'
* Bish°Pr'c V.^ Crtlrt.ii« i/
t Monastery ~~\ 1
/. Introduction Map IX. 1.3. Ecclesiastical map of Germany
567
J Archbishopric i Bishopric 1" Monastery
t Quedlinburg
Liege tf ^Aachen ( C fsiavelot
)
1~Pni»i t\
\ngelheun
\ Main:
C\ Bamberg
t Pragu
\Ec(üen,acly-Yr.er     t l^' tJMd: ^
Heidenheim
Cone ~f/
Regensburg (Rdlisbon
t Hi)
Augsburg 1
Murbach t     ;S( B/fls>-
Otlobeuren\    / j"/ f      t     J   J5eeo" [Salzburg^ Kremsmiinsler
1 A/imf\
Klosterneaburg Melk Mienn
1" Einsiedeln Engelbert) "f _____
\
t S«/mm
568 IX. Persons arid Places
Map IX. 1.4. Ecclesiastical map of France
J Archbishopric t Bishopric -f Monastery
( Sl-Omcr
I Ť^T/iwxiífmu'
i + Sl-Amand
St-Riquier 1
t   Arms   Í Crtwfirrti
Bnypm
Moiii-Sf-MkfiW
Í Rfiim's
^0
AiH/Viist" ,
;^^{Jk-* Rí""'l!   í     A   —-
liim^cS ,        Bemtvtiis /í^t ~ÍT .      i Reim T Bi'f
t SI-Éiroit/l t Lo Mííiis
Prtris
i C Soissoiis •pm/ís
t C/wrlrrs
Ton/ ^
Angcn if FoiildTiifdl y
Tours
X-
--y+Ficiiry .
\
\ t D''<"'      Í Bcsmiíon
'■ 1 <-
t Clumj
f Sl-Lrónmrí
Liinořřps Í
t Sl-Yrim t Amjouldmc
t Au rillen
Lye
i Lc Phi;
' Sl'öl!
I.íí GíYÍ/lrff
. J CCirtrlíVuse
Nmví/esrtŤ
" \      + MfllSSílí
\ K i
t Al ■(    N A'ř"' "' '     )Í Toulouse
Avignon ň +
/ * * A/>(
Narbonne í
Vid) t      i / Geronať
2. Am a (an us of Met
569
IX.2. AMALARIUS OF METZ
Hanssens 1948-50, Cabaniss 1954, Lipphardt 1965, Roger Evans 1977, Huglo 1979, 'Remaniements', Ilesbert 1980, 'Amalar'.
Despite the efforts of Pippin and Charlemagne to achieve uniformity of practice, much of the detail of liturgical observance was never standardized. The Roman mass replaced the Gallican, the Benedictine and Roman forms of the office were established. But the texts and chants performed during these services could not be made the same everywhere. Roman exemplars were lacking, inadequate, or self-contradictory, and had to be supplemented for Frankish use. This can be documented for the numerous sacramentaries (collections of prayers for mass), and there is no reason to suppose that things were much different for chant texts. The earliest sources of chant texts for the proper of mass (those edited by Hesbert in /\MS) indicate that the basic corpus was established in Charlemagne's time, if not earlier. The way differences could arise in the choice and arrangement of office texts is illustrated by the activities of Amalarius of Metz (r.775-c.850). As to the music for these texts, while we can marvel at the relative uniformity of the earliest preserved copies, there are numerous differences of detail between them.
Amalarius may have been educated under Alcuin at Tours or Aachen, and he is thought to have been archbishop of Trier between 809 and 814. He knew the Byzantine liturgy from an imperial mission to Constantinople in 813-14. He returned via Rome, and had another opportunity to learn Roman practice at first hand on a mission in 830-1. He may have been director of the palace school at Aachen in the 830s (Hanssens, i. 68 ff.). He is famous chiefly for two works on the liturgy. The Liber officialis (Hanssens, ii) explains the theological and spiritual significance of each part of the liturgy, texts through the church year, and how it is performed. Though its content is chiefly hermeneutical, it conveys a lot of detail about early ninth-century practice. The Prologus de ordine antiphonarii (c.831) and Liber de ordine antiphonarii (r.840) (Hanssens, i. 361-3, iii. 9-109 respectively; trans. Roger Evans 1977) refer to an antiphoner that Amalarius proposed to introduce, though its destination is unclear. Its contents are noted in some detail. Hanssens, in a magnificent appendix (iii. 110-224), sets out its contents, as far as they can be reconstructed, in parallel with the Old Roman antiphoner Rome, Biblioteca Apostolica Vaticana, San Pietro B. 79, one of the earliest Gregorian antiphoners, Paris, Bibliotheque Nationale, lat. 17436 (texts only), and one of the earliest notated antiphoners (St Gall, Stiftsbibliothek 390-391).
The differences between Amalarius and the other sources are very interesting and important, but it is the spirit behind them which interests us here. Amalarius was concerned, like Helisachar before him (see VII.2) and Aurelian after him (V.4), with the literal sense of responsories and their repeat structure. But he tells us that many series of pieces in Rome were simply given as sets from which a choice could be made
570
IX. Persons and Places
(this is still the case in the ninth-century antiphoner Paris 17436). Amalarius would prefer a logical order, such as the order of verses as they appear in the Bible. And there are indeed many examples of such numerical ordering in the Gregorian gradual, though not necessarily due to Amalarius' prompting; it may be that he wished to apply to office chants a principle already well established at mass. He tells us that in his antiphoner he marks chants according to the usage he prefers: 'R' for Rome, 'M' for Metz and 'IC' for the 'indulgentia' and 'caritas' which he requests to be extended to his own individual choice.
Amalarius was appointed to the see of Lyon in 835 after its previous archbishop Agobard (814-35/838-40) had been exiled. But it was not difficult for Agobard and his archdeacon Florus to stir up trouble about Amalarius' writings, which were condemned at a synod in Quierzy in 838, whereupon Agobard regained his see. A justificatory letter of Agobard of 838, presented as a 'Praefatio antiphonarii ecclesiae Lugdunensis' in one source, generally called the 'Liber de correctione antiphonarii' (MGH Epistolae Karolini aevi, iii. 238; PL 104. 329-50), explains the reasons for Agobard's own revision of the antiphoner, the removal of what was 'superfluous, mendacious, or blasphemous'. Florus' writings (PL 119) also condemn Amalarius' innovations. Agobard's revision chiefly consisted in removing non-biblical texts. Huglo (1979, 'Remaniements') gives numerous examples of Agobard's revisions and others carried out in like spirit in the books of Lyon and neighbouring dioceses. Hymns were not sung in Lyon before the twelfth century, and Agobard may have inspired distrust of troping in some churches (Gy 1983). The Carthusians, too, subsequently eliminated non-scriptural texts from their liturgy. (In Plate 14, a Carthusian manuscript, the scriptural sources of chants are noted in the margin.)
In Rome, Amalarius was unable to obtain a Roman antiphoner for the Emperor Louis, for they had been given to Abbot Wala of Corbie (in 825). Travelling to Corbie, therefore, he was finally able to consult a 'responsoriale' bearing an attribution to Hadrian I (772-95) (Hanssens, i. 361, iii. 14).
Whatever the destination of Amalarius' antiphoner, it does not seem to have been taken up in Metz, in or near which he lived in the latter part of his life. This is clear from a comparison with the ninth-century tonary of Metz (Metz, Bibliothěque Municipale 351, ed. Lipphardt 1965; see 202-11, also Hesbert 1980, 'Amalar'). But the tonary does not agree with what Amalarius himself says was Metz practice ('M'). These complications are witness to a period of fluidity in liturgical repertories. We may surmise that the liturgical reforms of Benedict of Aniane (r.750-821), officially promulgated at the Council of Aachen in 817, occasioned both revision and codification, at least in monastic repertories. But their exact workings out in extant liturgical books is unclear.
Careful comparison of their repertories has enabled Lipphardt and Huglo to trace relatives of the Metz tonary in a number of later sources: from Reichenau (Bamberg, Staatsbibliothek lit. 5, of the year 1001), from Quedlinburg (Berlin, Staatsbibliothek, Mus. 40047, r.1020), and from Nonantola (Rome, Biblioteca Casanatense 54, early eleventh century). Lipphardt connects this with the effects of the monastic reform
J. St Call
571
movement of Gorze, which lies a few miles west of Metz. Gorze was founded by Chrodegang of Metz, and revived after a period of decline by Bishop Adalbero of Metz (919-62). It and its associated house St Maximin at Trier became widely influential in the second half of the tenth century, introducing reforms of monastic observance into numerous houses at the behest of ecclesiastical or lay patrons. The reform came to Reichenau in the time of Abbot Ruodmann (972-85) in the person of Sandrat of St Maximin in Trier. The effect of such reforms as that of Gorze (see Hallinger 1950-1) and later Hirsau in the Black Forest (Jakobs 1961) upon monastic discipline is well documented, but it is fair to say that their consequences for the chant repertory have not been investigated in sufficient detail. The example of Cluny is discussed below.
IX.3. ST GALL
Schubiger 1858; Marxer 1908; van Doren 1925; Clark 1926; Scheiwiler 1938; von den Steinen 1948; Labhardt 1959-63.
It is unusual for a chant manuscript as old as Metz, Bibliotheque Municipale 351 to survive from a cathedral church. Most of the history of chant in the ninth and tenth centuries has to be reconstructed from monastic books. The monasteries were vital for securing the foundations of the Carolingian renaissance. The earliest surviving tonary, Paris, Bibliotheque Nationale, lat. 13159, is from Saint-Riquier; one of Saint-Riquier's abbots, Angilbert, was Charlemagne's son-in-law. A valuable description of the monastery, its library, and some of its liturgical ceremonial during Angilbert's time survives (ed. Lot 1894), which stops just short of providing the detail about performance practice useful for the musicologist (see Heitz 1963). Wala of Corbie, just mentioned as having carried off an antiphoner during a mission to Rome, was also related to the royal family. Corbie played a leading role in the development of Caroline minuscule script (the ancestor of the type used in this book). Saint-Denis, where Stephen II and his chapel stayed while visiting Pepin in 752, has also been mentioned (VIII.2.iv). Its abbot, Hilduin, responsible for the 'missa graeca', was Louis the Pious's archchancellor. Curiously, as well as the tonary from Saint-Riquier, we have graduals with tonal indications in early sources from both Corbie and Saint-Denis: Paris, Bibliotheque Nationale, lat. 12050, Laon, Bibliotheque Municipale 118, and the 'Mont-Renaud' manuscript, all of the late ninth or early tenth century (III. 14.iii). None were originally notated, though the last was provided with notation around the end of the tenth century. The chant tradition of Corbie and Saint-Denis, as revealed both in its selection of chants and the melodic detail of those chants, was closely similar, and one would like to know if this reflects the period when both monasteries were so intimately linked to the Carolingian liturgical reforms.
One of the first and one of the only monasteries before the end of the millennium from which notated manuscripts have been preserved in any numbers is St Gall, south-east of Lake Constance. The earliest, the cantatorium St Gall, Stiftsbibliothek
572
IX. Persons and Places
359 (PalMus II/2), is usually dated around 900. Later in the century appear the gradual St Gall 339 (PalMus 1) and two sources of sequences and tropes, St Gall 484 and 381 (facsimile in preparation by Arlt and Rankin). If the dating of St Gall 359 is correct, the manuscript must have been known by Notker Balbulus ('the Stammerer', d. 912), the famous author of a life of Charlemagne and numerous sequence texts, and Tuotilo (d. 915), who is reputed to have composed tropes while playing on his rotta (a type of triangular psaltery).
Notker's account of the difficulties of Frankish singers learning Roman chant says that of the two Franks who learned in Rome itself, one went to Aachen, the other to Metz. Later musicians were evidently concerned to establish St Gall's claim to a share in the authentic Roman tradition, for Ekkehard IV (c.980-r. 1060), in his Casus Sancti Galli (probably of the 1040s), says that while one singer did indeed go to Aachen, the other fell ill on the way and stayed at St Gall, teaching the monks from the Roman antiphoner he had with him, and instructing Notker in the meaning of the significative letters used in its notation (MGH Scriptores rerum Sangallensium, 102, cited van Doren 1925, 128). There survive several copies of a letter attributed to Notker, written to an otherwise unknown monk Lantpertus, which explains these letters (critical edition by FYoger 1962.)
Although Ekkehard IV's chronology has been shown to be shaky, and a certain amount of legend has crept into his narrative, he paints a fascinating and charming picture of life and learning in the monastery, of the activities of Notker and Tuotilo, Ratpert and Hartmann, and Notker's teachers, Iso and the Irishman Moengal or Marcellus (MGH Scriptores rerum Sangellensium; the parts relating to music are reproduced by Gautier 1886; see also Schubiger, etc. above). Of the composers of the rich trope repertory in the St Gall manuscripts (the early St Gall 484 and 381, then 376, 378, 380, and 382 from the eleventh century), we know nothing, except for Ekkehard's ascription of two pieces to Tuotilo. This is regrettable, because the early tropers in particular contain many pieces not known elsewhere, including the collection of melismatic extensions to introits, often with prosula texts. On the other hand, the sequence texts which can be more or less reliably attributed to Notker, the famous Liber hymnorum, have been identified on stylistic grounds (von den Steinen 1948). The difficulty of matching the neumatic notation of the early sources with appropriate later copies in staff-notation means that no critical musical edition of Notker's book exists, though transcriptions made with reference to early French sources have been published bv Crocker (1977). Schubiger edited many of the sequences from later German sources, including St Gall 546, a massive compilation of sequences, ordinary-of-mass chants, and some other items copied by the St Gall cantor Joachim Cuontz in preparation for the canonization of Notker in 1512, 600 years after his death. By this time several of Notker's pieces had fallen out of the repertory, and Cuontz did not know their melodies. (On the ordinary-of-mass chants see Marxer 1908, on the sequences see Labhardt 1959-63.)
No less important than its trope and sequence manuscripts are the St Gall books with proper of mass and office chants, in particular St Gall 359, 339, and the antiphoner compiled around 1000 by Hartker (d. 1011), St Gall 390-1 (PalMus II/l).
J. St Gall
573
The latter is one of the earliest of all notated antiphoners, St Gall 359 one of earliest of all fully notated chant-books. Their importance was recognized by the chant reformers of the nineteenth century, and it was believed for a time that St Gall 359 was the very manuscript brought from Rome by Romanus, hence the publication of a facsimile of it by Lambillotte in 1855. The fine detail of their notation has made the early St Gall books particularly important for the investigation of chant performance. As to the liturgical and melodic tradition they represent, while neither they nor any other books can be shown to transmit any particularly authoritative archetype, Carolingian or other, they belong to a particularly closely knit group of German sources. This is clear from the Solesmes comparison of melodic variants in proper-of-mass chants (Graduel romain IV) and also Hesbert's work on Advent responsory series (C40 5—6). It has yet to be shown, however, that St Gall was the fons et origo of these traditions in eastern lands.
In Table IX.3.1 I have summarized some of the results of the Solesmes survey. The method of investigation involved identifying 100 'lieux variants' in the proper chants of mass where sources tended to differ from each other. Agreements and disagreements between all the sources were then counted, and groups of sources in total or near agreement were identified. Within a limit of twenty-five disagreements out of 100 (75% agreement), the sources divided into two main blocs, eastern and western, with sources from Vercelli and Le Mans, respectively, forming two eccentric blocs. Within the eastern bloc, however, the great majority of sources were in agreement to within six differences (94% agreement), whereas at that level of agreement the western sources could only form mostly very small groups, representing a single church or religious order. The list omits all single sources which can also be assigned to the main blocs, and those for which insufficient readings were available.
The great majority of the readings on which manuscripts differ are hardly perceptible to modern ears. But there is a quite audible distinction in the treatment of the notes at the semitone steps in the diatonic scale. Where many southern and western sources (Aquitanian and Beneventan ones seem to be most consistent) have a reading such as ded, or aba at the top of a phrase (or aba, if the scribe uses the flat sign at all), northern and especially eastern manuscripts prefer dfd and aca. Intimately related to this phenomenon is the question of which reciting note, b or c, is appropriate for the third psalm tone and recitation passages in compositions in modes three and four. The eastern propensity to choose the higher note is often referred to as constituting the 'German chant dialect', though it would seem more appropriate to call it a regional accent, since nothing in the basic vocabulary or grammar of the chants is affected. Numerous examples were given by Wagner (1930-2; see also Wagner 1925), and Heisler (1985, 1987) has recently considerably broadened the scope of the inquiry. (See also Stäblein, 'Deutschland. B. Mittelalter. I. Der römische Choral im Norden', MGG, and Agustom 1987.) Gajard (1954) and Cardine (1954) demonstrated cogently that early St Gall manuscripts used b, not c, as the reciting note, and this would seem to imply that the preference for the higher note at the semitone step is a relatively later development.
574 IX. Persons and Places
Table IX.3.1. Sources grouped according to their musical variants
showing groups of sources with no more than 6 out of 100 disagreements (number of sources in each group given in parentheses)
Western bloc
25 groups, 83 sources
1. Saint-Denis (earlier), Corbie, Worcester, Downpatrick (5)
2. Saint-Denis (later), Saint-Corneille at Compiegne (3)
3. Saint-Maur-des-Fosses (2)
4. Paris, Jerusalem (7)
5. Paris (another group) (2)
6. Cambrai (2)
7. Arras (2)
8. Anchin, Lille, Saint-Amand (3)
9. Andenne(2)
10. Sens(3)
11. Chartres(2)
12. Lyon (2)
13. Autun(2)
14. Saint-Benigne at Dijon (2)
15. Bee(2)
16. Sarum (3)
17. Toulouse, Albi, San Millan de la Cogolla (5)
18. Valence (2)
19. Ivrea or Pavia (2)
20. Forlimpopoli, Ravenna (2)
21. Piacenza (2)
22. Montecassino, Benevcnto (11)
23. Cistercian, Dominican (7)
24. Carthusians (5)
25. Franciscans (3)
Eastern bloc
5 groups, 98 sources
1. St Gall, Einsiedeln, Pfävers, Rheinau, Schaffhausen, Augsburg, St Georgen, St Blasien, Weingarten, Zwiefalten, St Peter at Bantz, Minden, Bamberg,
St Emmeram at Regensburg, Prüfening, Passau, Salzburg, Admont, Seckau, St Lambrecht, Moggio, Kremsmünster, Garsten, St Florian, Seittenstetten, Melk, Blaubeuern, Stockerau, Klosterneuburg (90)
2. Seeon, Schäftlarn (2)
3. Trier (2)
4. Monza(2)
5. Aquileia (2)
IX.4. CLUNY
General: Sackur 1892-4; Joan Evans 1931; Congress Cluny 1949; Congress Todi 1958; Hunt 1967; Wilmart 1970; Valous 1970; Cowdrey 1970. Customs: PL 149-. 633-778; CM 2; CCM 10.
Chant: Leroquais 1935; Hourlier 1951, 'Clunisienne', 1959; Huglo 1957; Renaudin 1972; Garand 1976; Huglo, 'Cluniac Monks', A'G; Davril 1983; Huglo 1983; Steiner 1984, 'Music', 1987; Holder 1985; Lamothe and Constantine 1986; Hiley 1990.
4. Čluny
575
Perhaps the most famous of all Benedictine monasteries, Cluny, near Macon in Burgundy, was founded in 909, long after such Merovingian foundations as Corbie, or Irish ones such as St Gall. It owed obedience to no diocesan authority but only to the pope. This arrangement, and the quality of its monastic observance, inspired other Benedictine houses to reform themselves after the Cluniac model. Others were obliged to do so by their lay or ecclesiastical patrons, or were founded as priories over which Cluny had authority. Under a series of outstanding abbots—Odo (927-42), Mayeul (948-94), Odilo (994-1049), Hugh (1049-1109), and Peter the Venerable (1122-57)—Cluny rose to a position of unprecedented ecclesiastical and temporal power, commanding the obedience of over 1,000 houses, mostly in Burgundy and the rest of France, but also in northern Spain, Italy, and England.
After the twelfth century its influence declined in the face of competition from the newly fashionable ascetic orders, especially the Cistercians, who were responsible for much intemperate polemic against what they saw as a betrayal of the simple ideals of early monasticism. Among other things, the complexity and length of the liturgy was criticized. From the extant evidence it does not, however, seem that Cluny's liturgy was exceptional in its own day. It is now seen not as boundless prolixity for the sake of ritualism alone, but as the expression of a vision of the endless hymn of praise of heaven itself (Waddell 1982, Leclercq 1974, 308).
The surviving chant sources with music from Cluny itself are pitifully few. The earliest are two from about 1075, that is, from the time of Abbot Hugh: a gradual, Paris, Bibliotheque Nationale, lat. 1087, and a noted breviary, Paris 12601 (Hourlier 1959). Unfortunately, later musical manuscripts are almost totally lacking. Only the noted breviary of 1317 of Saint-Victor-sur-Rhins, near Roanne, can be cited, a manuscript which was prepared at Cluny (Davril 1983, Huglo 1983). Recourse is therefore necessary to those from affiliated monastic houses. For manuscripts in staff-notation one may consult, for example, for the mass, a gradual of the early twelfth century from Souvigny (Bourbonnais, compiled for Sauxillanges, Auvergne), Brussels, Bibliotheque Royale II. 3823 (Huglo 1957); for the office, a twelfth-century antiphoner from Saint-Maur-des-Fosses, Paris 12044 (Renaudin 1972); and for mass and office combined, the breviary-missal of Lewes of the late thirteenth century, Cambridge, Fitzwilliam Museum 369 (Leroquais 1935, Holder 1985). (The text of the twelfth-century Cluniac antiphoner of Saint-Maur-des-Fosses, Paris 12584, has been edited by Hesbert in CAO 2.)
Presumably because its extant sources, and indeed its time of foundation, were too late in date to interest the Benedictine restorers of plainchant, partly perhaps also because some of the Cistercian criticism has coloured modern attitudes, Cluny's liturgy and chant have not received detailed attention in proportion to their fame. The tendency has been rather to assume Cluniac influence on liturgical music simply because it 'must have been influential'. Thus Abbot Odo, known to have composed three hymns and twelve antiphons for the monastic office of St Martin, was also credited with the compilation of a number of theoretical works, all of which can now be assigned to other authors: the treatise known as the Dialogus de mustca, which
576
IX. Persons and Places
originated in northern Italy in the late tenth century; a tonary and associated treatise likewise of the late tenth century, compiled by Abbot Odo of Arezzo; a tonary of the fourteenth century which is clearly of Franciscan origin (see Huglo, 'Odo', NG).
More bizarre is the case of the famous 'Codex Calixtinus' or 'Liber Sancti Jacobi', renowned for the unique collection of conductus (rhymed Latin songs) and polyphonic music which it contains (ed. Whitehill el al., with facsimile of the offices; edition of music also by Wagner 1931). The manuscript was copied in the 1160s, for the great pilgrimage church of Santiago de Compostela, where it still resides. That was not, however, the original destination of the material. The manuscript was written in central France, possibly at Vezelay. It contains texts and music for the festal liturgy of St James, an account of St James's translation to Galicia, his miracles, a book about the pilgrimage routes to Santiago, and a book of the legendary exploits of Archbishop Turpinus, one of the Twelve Peers of Charlemagne. At the beginning of the book its author states that a large proportion of the material was compiled at Cluny, and attempts have therefore been made to understand the book as an instrument of the Cluniac ecclesiastical reconquest of Spain, and the replacement of the Mozarabic rite by the Roman (see Hamel 1950, Stablein 1963, 'Modale Rhythmen'). This thesis unfortunately takes the nature of the book too seriously. Hohler (1972) has shown that the Pseudo-Turpinus part of the 'Liber Sancti Jacobi' was a text designed to teach Latin to schoolboys, full of deliberate grammatical howlers and malapropisms, not to mention such humorous stories as the one where the Emperor Charlemagne orders his cavalry to charge across rough country with their horses blindfolded. And the St James liturgy cannot be a Cluniac compilation. Among other things, the form of the office hours is not always monastic. And the special Latin songs and polyphony are a type of music which Cluny is unlikely to have cultivated with much vigour.
From the exhaustive description of Cluny's liturgy by Ulrich of Zell (PL 149. 633— 778) and the contents of the contemporary gradual Paris 1087, it seems clear that Cluny did not have tropes sung in its liturgy, with the sole exception of a troped Agnus Dei on Easter Day. While the argumentuni ex silentio is never reliable, Ulrich's habit of dwelling on each unusual detail of the liturgy, not least its more festal music, makes his mention of but one trope telling. Furthermore, sequences were sung with texts at Cluny only on the highest feasts; on other feast-days only the melodies were sung, as wordless melismas. Does this denote a deliberately reserved attitude towards 'non-canonical' material, a later manifestation of the ideas of Agobard of Lyons? (Cf. Gy 1983.) At any rate, with respect to these pieces at least, Cluny appears to have steered a moderate course. Here is none of the inflated liturgical splendour attacked by Cluny's opponents. (See Hiley 1990.)
Cluny's influence upon monastic observance and discipline and relations with ecclesiastical authority has been studied more than its liturgy. When a monastery submitted to Cluniac authority, did it necessarily observe the same liturgy as Cluny? In the case of many houses no service-books survive to give us an answer. The standard tests on the chant repertory provide at least some answers, however. Map IX.4.1 shows the provenance of sources which agree with Cluny in their choice of
4. Čluny
577
Map IX.4.1. Monasteries following the use of Cluny, Saint-Benigne at Dijon, or Bee
Coldingham
a 2, w-x
Durham a
■ Cluny
• Saint-Bénigne, Dijon
* Bee
)
Ponlefract
\
Worcester'* ,
)  • Evesham
Gloucester.^ »\\linciuombe si Al&ons
Abingdon S .Westminister
\j __
Canterbury*
Muchelney
Lpwcs
Moní-Blfiřid/i!
Fecamp '^z^.—^ IltHlIříjeS
Aiirftiii , . Sí-Ai«fltid ■
Marchiemies u Corbie
1)
____ x ■ Reims
Caen a . ~\ Meulau
I Troarn Eiwhx.
„  l      Sl-Évroii/I . , A *Concím St-Maur-les-Fossés
Mont-St-Michel v-x \„
■ Nogenl-le-Rolrou \
Mendome i Angers m       Tours .
Sl-Fbrenl-les-Saumur
Vézclay i
Montier-la-Celle
Dijon
Souvigny ■
Cluny J
■ Limoges
w
Moissac
578
IX. Persons and Places
post-Pentecost alleluias and Advent responsories (cf. CAO 5-6), and in the melodic readings of their proper-of-mass chants, the Solesmes 'lieux variants' (Graduel rotnain IV). Similar information for two other monastic families is registered on the map, which will be referred to in two subsequent sections.
Although work on many sources remains to be done, it is clear that when a monastery adopted the order of service and selection of chants performed at Cluny, it might continue to sing those chants in the old way. This is the case at Saint-Martial of Limoges (Paris, Bibliotheque Nationale, lat. 1132), Marchlennes (Douai, Bibliotheque Municipale 114), Anchin (Douai 90) and Saint-Amand (Valenciennes, Bibliotheque Municipale 121), all of which have the Cluniac alleluia series but non-Cluniac melodic variants.
Ulrich of Zell (1029-93), the author of the customary just referred to, grew up in Regensburg, where he was educated alongside the great William, future abbot of Hirsau in the Black Forest. Later he entered Cluny, where he wrote the Consuetudines to inform William about Cluniac observance. But William did not introduce the Cluny liturgy in the houses which were reformed at first or second hand from Hirsau. Although practically none of Hirsau's liturgical books has survived, we have a good idea of what they contained from manuscripts of monasteries which were reforrhed from Hirsau. In Hesbert's survey of Advent responsory series, a group of mostly south German monasteries emerged which evidently reflected their common link with Hirsau: Prüfening, Zwiefalten, and St Georgen, Schwarzwald, were founded or reformed from Hirsau itself; SS Ulrich and Afra, Augsburg, and Gengenbach from St Georgen; St Emmeram, Regensburg, from Admont; Rheinau from Petershausen. Assessing such relationships with the aid of the Solesmes 'lieux variants' in proper-of-mass chants is, however, much more problematical, for the observed difference between eastern sources is so slight. It appears that musicologists will have to develop other means of distinguishing between such books.
Within Benedictine monasticism as a whole many monasteries became grouped under the aegis of a founding or reforming centre whose discipline or liturgy they adopted. Cluny, Gorze, and Hirsau, Grandmont, Camaldoli, and Vallombrosa are just a few of the better-known names. Much work remains to be done to discover the significance of these groupings for the chant repertory. The sort of chant propagated in all of them was traditional. Only the Cistercians reformed the actual melodies to be sung (see below, X.2).
IX.5. WILLIAM OF DIJON
Chomton 1900; PalMus 8; Muglo 1956, 'Tonaire', 1971, Tonaires; R. Le Roux 1967; Bulst 1973, 1974; Hansen 1974.
W7illiam of Dijon was born of a noble Piedmontese family in Volpiano in Lombardy in 961. When already a monk he met Mayeul of Cluny and entered that house, eventually attaining to such esteem that he was entrusted with the reform of Saint-Benigne at Dijon in 989. He made this the springboard for the reform of other
5. William of Dijon 579
monasteries, particularly in Piedmont. (The customary of one of these survives, from San Benigno at Fruttuaria, close to his birthplace: ed. in CCM 12 and by Albers in CM 2 and 4). In 1001 he was invited by Duke Richard II of Normandy to revive the monastic houses of his duchy, only then beginning to recover from years of devastation and neglect at the hands of Richard's forebears. From Fecamp on the Channel coast he and his successors, in particular his nephew John of Ravenna, who succeeded him as abbot of Fecamp in 1028, and Thierry, abbot of Jumieges and then Mont-Saint-Michel, transformed monastic life in Normandy. On his death in 1031 he was head over more than forty houses in France and Italy. A pious picture of the great man was written by his disciple Rodulphus Glaber, who described his wide-ranging interests, including medical and musical ones. The latter are, however, mentioned in purely general terms—he 'emended' the singing of antiphons, responsories, and hymns so that they were sung better than anywhere else (PL 142. 715)—from which nothing specific can be learned.
While much is known about his reforming work, and the customary he composed for Fruttuaria is extant, it is impossible to know just how much musical expertise William had. The question is not a purely academic one because of the existence of the famous tonary with dual notation, in neumes and letters, Montpellier, Faculte de Medecine H. 159, which may have been composed by him or under his direction (facs. PalMus 8, edn. Hansen 1974).
That the tonary comes from Saint-Benigne or one of its sister houses in Normandy is beyond dispute. The offertory Laudate dominum quoniam was given the prosula In Hierusaletn in honour of St Benignus (PalMus 8, 217-18, Hansen no. 880—1). And by choice of pieces and by its melodic variants Montpellier H. 159 agrees with other books from Saint-Benigne, from Jumieges, and Mont-Saint-Michel. By all the usual tests these and other Norman houses form a particularly cohesive group: Fecamp and Saint-Evroult are two other notable members of it, together with some English houses. (As well as CAO 5-6 and Graduel romain IV, see Le Roux 1967, Michel Robert 1967, Hiley 1980-1 and 1986, 'Thurstan'). These monasteries are represented on Map IX.4.1 above, in company with those following Cluniac use and that of Bee, whose influence is discussed in the next section.
The main part of Montpellier H. 159 is a copy of all the proper chants of mass in tonal, rather than liturgical order. But on its flyleaves is a fragmentary tonary of the conventional kind, that is with the incipits of office antiphons only, which matches a tonary of the same type in the Fecamp antiphoner Rouen, Bibliotheque Municipale 245 (A. 190) (see Huglo, 1956 'Tonaire', 1971, Tonaires, 328-33).
The main tonary in Montpellier H. 159 has been described briefly above (III.14), and also its alphabetic notation (IV.6). Given the Italian extraction of William and John of Ravenna, it is perhaps not surprising to find alphabetic notation disseminated among the Norman monasteries, for the anonymous Dialogus de musica, written in north Italy in the late tenth century, describes just such a system (albeit without the special signs for liquescence, quilisma, and oriscus in the Montpellier manuscript).
The liturgical ordering of chants which unites books from Saint-Benigne and those
580
IX. Persons and Places
from Norman monasteries is not that of Cluny. The 'Cluniac' nature of William's reforms lies elsewhere. Perhaps it is significant that we do not find tropes for the proper of mass in books from the revived monasteries—a sign of Cluniac 'abstinence'? They do, however, have full complements of sequences, and ordinary-of-mass chants: the twelfth-century troper from Saint-Evroult, for example, Paris, Bibliothéque Nationale, lat. 10508, has one of the biggest collections of ordinary-of-mass chants and tropes of its time, surpassed only by the collections in Norman-Sicilian manuscripts, to which it is in any case related (listed in Hiley 1986, 'Ordinary'). And there is ample evidence of compositional activity in honour of the local patron saints, for all of whom new offices would have been written as soon as musical practice was firmly established. Huglo has speculated that the office of St Benignus, also celebrated in Normandy, might have been written by William himself, though we have no proof of this. At first, as one might expect, outside help was occasionally needed: Ordericus Vitalis, the chronicler of Saint-Evroult, reports that the office of the monastery's patron saint Evroult (Ebrulphus) was first composed according to the secular (Roman) cursus at the request of Abbot Robert de Grantmesnil (1059-61) by Arnulf, the precentor of Chartres cathedral and a pupil of the famous Fulbert. Two young monks, Hubert and Ralph, went to Chartres to hear it sung (in the days before staff-notation was used at Saint-Evroult). The office was given its final monastic form by the addition of nine antiphons and three responsories by Guitmund, who had come to Saint-Evroult with the next abbot, Osbern (1061-6), from the monastery of Sainte-Catherine-du-Mont at Rouen (Chibnall 1969-80, ii. 108-9). Such a pattern of events must have been repeated very often—and of course not in Normandy alone. On numerous occasions the pious composer's name is recorded for posterity by an equally pious chronicler, bypassing for us the anonymity of the liturgical books themselves. The results can easily be seen by turning the pages of, for example, Hesbert's catalogue of manuscripts from Jumiéges (MMS 2): Rouen, Bibliothéque Municipale 248 (A. 339), an antiphoner of the thirteenth century, contains the complete proper offices for Philibertus and Aycadrius, the abbey's patron saints, Audoenus (Ouen) of Rouen, and Benignus (Bénigne). Plates LXXVI-LXXVII in MMS 2 give the start of the Philibert office, with the antiphons of the first nocturn of the Night Office in modal order 1-6. The office of St Audoenus, patron of the diocese, was more adventurous, containing numerous prosulas (see Kelly 1977).
IX.6. ENGLAND BEFORE AND AFTER THE NORMAN
CONQUEST
Graduale Sarisburiense; AS; PalMus 12; Frere 1894, 1898-1901; Eeles 1916; David and Handschin 1935-8; Knowles 1963; Holschneider 1968; Hartzell 1975, 1989; Planchart 1977; Gjerl0w 1979; Hiley 1980-1, 1986, 'Thurstan', 1987; Hesbert 1982, 'Antiphonaires'; Underwood 1982; Droste 1983; Rankin 1984, 1987.
6. England before and after the Conquest 581
After the heroic age of Anglo-Saxon monasticism in the time of Bede and Benedict Biscop, Boniface, and Alcuin, the church in England was practically destroyed by the Scandinavian invaders of the ninth century. It is not at all certain that even a single monastery succeeded in continuing observance of the office. Since, not surprisingly, no service-books have survived this period, the contribution of Anglo-Saxon practice to the introduction of Roman chant under Pippin and Charlemagne is very difficult to assess. If anything did survive into the earliest English chant-books of the late tenth century, we cannot yet recognize it. The evidence seems to tell against such a possibility, for what we see in the later manuscripts is clearly based on continental traditions, principally north French ones.
The English revival, like the Carolingian, depended upon the co-operation of civil and ecclesiastical authority, in the English case between King Edgar (959-75) and three great bishops: Dunstan, Oswald, and Ethelwold. Dunstan (c.909-88) was abbot of Glastonbury from about 940, then had to leave England during a period of anti-monastic government in the 950s, spending a few years at the monastery of St Peter, Ghent. In 960 Edgar made him archbishop of Canterbury. Although he counts as an inspiring force behind the monastic revival, it is difficult to see tangible results in the history of liturgical music. He played the harp and made instruments: he gave an organ to Malmesbury, bells to Abingdon and Canterbury. The stories that he composed the Kyrie Rex splendens and the antiphon Gaudent in celis are, however, false. (For these anecdotes see Stubbs, 1874). Oswald (d. 992) was a monk at Canterbury, then at Fleury (Saint-Benoit-sur-Loire); he returned to England in 959 and was ordained bishop of Worcester by Dunstan in 962. Ten years later he was made archbishop of York. He had followed the usual practice in this period of monastic revival by replacing the clerics of Worcester Cathedral by monks, and he, like his colleagues, was responsible for the restoration and foundation of other monasteries, notably Ramsey, where he installed as first head Germanus, who had studied with him at Fleury. As far as we are concerned, Ethelwold was undoubtedly the most important of these men. He was a monk at Glastonbury, abbot of Abingdon, and from 963 bishop of Winchester, the chief city of Wessex and the main seat of the royal court, where he expelled the clerics from the Old Minster and replaced them with monks. It is from Winchester that most of the earliest English notated liturgical books come. About 972 Ethelwold and Edgar called together an ecclesiastical council to lay down guide-lines for the conduct of the English church. Monks from Ghent and Fleury were there. The document which resulted, the Regulatis concordia (ed. Symons 1953), describes many interesting details of liturgical practice, including the performance of the Visitatio Sepulchri on Easter morning.
Ghent and Fleury were not the only foreign sources of influence. Already at Abingdon Ethelwold had been concerned for unanimity of chant practice and invited monks from Corbie to provide an example in reading and singing (Stevenson 1858, i. 129, cited by Knowles 1963, 552).
After the Norman conquest William I (1066-87) replaced practically the complete Anglo-Saxon church hierarchy, abbots as well as bishops, by Normans. Norman
582
IX. Persons and Places
influence had already been felt during the reign of Edward the Confessor. Two more Italians who, like William of Dijon, had been important in Norman monasticism led the reorganized Anglo-Norman church. These were Lanfranc, monk of Bee (not part of William of Dijon's family) from 1042, first prior of St Stephen's, Caen, founded by Duke William, and first archbishop of Canterbury after the Conquest, consecrated by one of his former pupils at Bee, Pope Alexander II. He was succeded by another Italian, Anselm, prior of Bee from 1063, who in turn became archbishop of Canterbury (1093-1109).
These are the figureheads. We do not know how many English cantors, politically a less important company, were replaced by Normans. But when the English chant-books from before and after the Conquest are compared with their continental counterparts, the various lines of influence stand clearly revealed, first from Corbie, then from monasteries of the Dijon family and from Bee (see for example the demonstrations in Hartzell 1975, Hiley 1980-1, Hesbert 1982, 'Antiphonaires', Hiley 1986, 'Thurstan'). WTe must of course use these labels with caution, for in practice it is not usually possible to pinpoint the specific source of influence in question. The uses of Corbie and Saint-Denis, for example, are practically indistinguishable according to the rough tests so far developed. Similarly for Fecamp, Jumieges, Mont-Saint-Michel, and so on. But the general pattern is clear. From Winchester we have only pre-Conquest books, which clearly ally themselves with Corbie practice. After the Conquest we can see, for example, the order of service of Bee (judged by the alleluia and responsory test) taken up at Christ Church, Canterbury, Worcester, and Durham, that of Dijon at Westminster, Evesham, and Winchcombe. And study of the melodic variants in their chants shows that Bee versions were taken up at St Albans, Dijon ones at Gloucester. But interestingly, the books of most churches do not show the melodic variants typical of Norman sources. In other words, even when the order of service was revised, the chants were sung as before, something we have already noted in the case of books from Cluniac houses. So we find the Corbie/Winchester melodic tradition perpetuated after the Conquest at Christ Church Canterbury, Crowland, Peterborough, Exeter, Worcester, the Augustinian priory of Guisborough, even the secular use of York, and Downpatrick in northern Ireland. (See Map IX.4.1 above.)
Of the thirty or so liturgical books or fragments with musical notation that have survived from England up to the time of the Conquest (see Rankin 1987) about a third come from Winchester (Rankin 1984 describes the slightly later activity at Exeter). Pride of place is held by the two famous tropers, Cambridge, Corpus Christi College 473, probably written in the closing years of the tenth century, and Oxford, Bodleian Library Bodley 775, from about the middle of the eleventh. At least parts of the Cambridge manuscript may have been written by the cantor of the Old Minster, Winchester, Wulfstan, author of a lost music treatise and of a long poem in honour of St Swithun which contains much invaluable information about the Anglo-Saxon cathedral of Winchester and a celebrated description of its organ. He is generally thought to have been the composer of the organa in the Cambridge manuscript, over
6. England before and after the Conquest
583
150 second voices to be sung with chants of the mass and office to make two-part polyphony.
Both the tropes and the polyphony in the Winchester manuscripts have been studied in detail (see Planchart 1977, Holschneider 1968). Many of the tropes were known abroad, most in France, some in eastern sources—possible north French intermediaries, such as tropers from Corbie or Saint-Denis, unfortunately do not survive, nor is any troper from Fleury extant. It is a pity that no later English troper of this type exists, one transmitting the pieces in staff-notation, for all the compositions in honour of local English saints, known only from the Winchester sources, remain untranscribable.
It would perhaps be overbold to say that the Norman Conquest, and the new men who took command of the church in England, were directly responsible for a 'purge' of tropes from English books. The evidence is insufficient. But, while the proper tropes are no longer recorded, the advent of the new Norman repertories of ordinary-of-mass chants and their tropes (such as those in the Saint-Evroult troper mentioned above) is clearly visible. The Winchester tropers also contain individual repertories of sequences which do not appear to have survived the Conquest (listed in Frere 1894). About half the seventy-odd texts are unique to Winchester, a strikingly large proportion. Most of the melodies for these texts are known elsewhere. The process must have resembled somewhat that at St Gall, where a century earlier Notker (at least in some cases) provided new texts to replace previous ones which he thought unsatisfactory. Alas, some of the unique Winchester pieces are written to melodies not known elsewhere, and therefore remain lost to us.
The gravity of the changes led to unrest. At Glastonbury the new Norman abbot Thurstan, previously monk at Caen, is reported to have tried to make the monks sing the 'chant of William of Fecamp' (presumably William of Dijon/Volpiano), which, among other things, led to a bloody fight between the recalcitrant monks and Thurstan's retainers. One would like to know more about the 'chant' which caused the trouble: did the monks object to learning new pieces, or to singing old ones in a new way? (On this question, see David and Handschin 1935-8 and Hiley 1986, 'Thurstan').
While the derivation of the monastic uses in England is reasonably clear, the use of the secular cathedrals survives in a frustratingly incomplete state. We have chant-books, both for office and mass, from only three: Salisbury, Hereford, and York.
The foundations of the Salisbury liturgy as we know it appear to have been laid by Bishop (St) Osmund in the late eleventh century. The twelfth-century unnotated gradual Salisbury, Cathedral Chapter Library 149 shows the mass liturgy established before the first notated chant-books appear. These seem to have been due to the initiative of Bishop Richard Poore, who in 1218 changed the site of the cathedral, and who may be the author of the customary and ordinal: these survive in several successive 'editions' (see Frere 1898-1901). Among the earliest chant-books are three notated in very similar hands, of about the second quarter of the thirteenth century:
584
IX. Persons and Places
the graduals London, British Library, Add. 12194 (part facs. in Graduate Sarisburiense) and Oxford, Bodleian Library, Rawl. lit. d. 3, and the antiphoner Cambridge, University Library, Mm. 2. 9 (from Barnwell near Cambridge, facs. in AS), followed by the noted missal Manchester, John Rylands Library, lat. 24, made in Salisbury around 1260 for Henry of Chichester, canon of Exeter Cathedral, to present to the cathedral there (Hollaender 1942-4).
Salisbury ('Sarum') use became a sort of national rite by the later Middle Ages. The reasons for this are not entirely clear, but among the crucial factors were the following: (i) the seat of the archbishopric, Christ Church Canterbury, was a monastic house whose liturgy could not be automatically used in secular cathedrals and parish churches; (ii) the royal chapel seems to have used the Salisbury liturgy, though from what time is not known; indeed a note in a fourteenth-century Sarum missal states that the use of the royal chapel is actually normative for the rite elsewhere (Hohler 1978, 37); (iii) from the thirteenth century onward, professional workshops in Oxford, London, and later Cambridge appear to have been able to supply books of Salisbury use 'on demand', so to speak, making it easy for churches to acquire books of this type rather than those of their diocesan cathedral (Droste 1983). A good illustration of this is the missal Munich, Bayerische Staatsbibliothek, elm 705, which bears a colophon indicating that it was copied in Oxford, and whose calendar has the dedication feast of the parish church of Lydd (diocese Canterbury).
Early sources from Wales, Ireland, and Scotland are lacking. It was under English influence or in the train of English arms that the chant traditions represented in the earliest extant sources (of the thirteenth century) were formed, with the usual modicum of items in honour of local saints.
Very rarely can we detect 'local flavour' in the chants that were produced locally. There is no fundamental stylistic difference between chant written in England and that written in France. Of the sequences for St Patrick in Irish sources, for example, Letabundus decantet is a contrafactum of the popular Christmas sequence (see MMS 4 for a facsimile of Cambridge Add. 710); and while Hesbert comments on the 'Irishness' of Laeta lux est hodierna (in the same manuscript), its 'major mode' melody seems to the present author typical of many contemporary compositions. The office of St David in Aberystwyth, National Library 20541 E (see O. T. Edwards 1987) is a contrafactum of the office of St Thomas of Canterbury. (For Scottish remains see McRoberts 1952 etc. and Woods 1987.) The aim is to provide fitting music for the praise of the saint, and what is unusual or radically new will not as a rule be fitting.
7. North Italy
585
IX. 7. NORTH ITALIAN TRADITIONS
Planchart 1985.
Apart from the separate rite of Milan, and the scanty remains of other non-Roman uses, chant-books from Italy north of Rome are Gregorian. But they display great variety of content and appearance, some of it the result of establishing local traditions, some due to the importation of foreign practices. A few examples of these overlapping uses may be given here.
The different styles of notation in Italian manuscripts have already been mentioned, and the occurrence of German notation at Monza and Bobbio, Breton notation at Pavia, and Laon notation at Como (IV.2). This accords to some extent with the affiliations revealed by the melodic variants in chants for the proper of mass: Monza (Monza, Basilica S. Giovanni C. 12/75 and C. 13/76) allies itself with the eastern 'bloc'. The nearest relative of Vercelli, Biblioteca Capitolare 186, from Balerna near Lake Como, is the Laon gradual Laon, Bibliotheque Municipale 239. The sources from Pavia (the ascription is provisional), Ivrea, Biblioteca Capitolare 60 and Vercelli, Biblioteca Capitolare 56, are related, though not so closely, to Breton sources. A majority of the tropes in Ivrea 60 appear to have been imported from France, yet its sequence repertory is dominated by German pieces. The latter feature is not unusual in north Italy. The monastery of Nonantola, near Bologna, for example, had a sequence repertory largely based on Notker's collection; one wonders if this fact is connected with the early appearance of a tonary like the Reichenau one in a Nonantola manuscript (Rome, Biblioteca Casanatense 54—see III. 14). (Sequences in all Italian sources up to r.1200 have been catalogued by Brunner 1985, who is also preparing an edition of the Nonantola sequences.)
Since numerous German clerics were appointed to important positions in the Italian church throughout the early Middle Ages, it is not surprising that the influence of German liturgical practice was strong. It seems certain, for example, that the early appearance of the German or 'Augsburg' form of the Visitatio Sepulchri, 'Quern queritis o tremule mulieres', in Aquileia (Udine, Biblioteca Capitolare 234, eleventh to twelfth century) is due to the appointment of Augsburg canons to the patriarchate in the eleventh century: Eberhard in 1042-8 and (more likely) Heinrich in 1077-84. This would also be why the melodic variants in the surviving Aquileian graduals Rome, Biblioteca Apostolica Vaticana, Rossi 76 and Udine 8. 2 (both thirteenth century) agree with sources in the eastern 'bloc'. (On Aquileia see also Huglo 1976).
Apparently ancient non-Roman chants have often turned up in relatively late Italian sources (see Wilmart 1929, Huglo 1952, 'Antiennes', 1954, 'Vestiges', Levy 1970, 1971, Baroffio 1978, Boe 1987). We should not, therefore, automatically assume that items in Italian manuscripts which are also known in France or Germany were necessarily imported into Italy. As a simple illustration of this point one might take
586
IX. Persons and Places
the tropes for the introit of mass on Easter Day as they appear in Piacenza, Biblioteca Capitolare 65, the massive compendium of the entire chant repertory of Piacenza cathedral, begun in 1142 not long after the commencement of the new cathedral church. The edition and lists of tropes in CT 3 make comparisons with other repertories easy.
Piacenza 65 begins with a version of the Quern queritis dialogue, one where, after the sentences Quern queritis—Iesurn Nazarenurn—Non est hie . . .he nunaate quia surrexit dicentes, the piece concludes with the verses Alleluia resurrexit dominus. Eia carisshni verba canite Christi, presumably composed to lead into the Easter introit Resurrexit. The verse FJia carisshni is found otherwise only in a few other Italian sources and in Cambrai, Bibliotheque Municipale 75, from Saint-Vaast at Arras. Where did it originate? That dialogue has been discussed frequently (see Rankin 1985, 'Quern queritis1). But for the introit itself Piacenza 65 has more trope verses (Ex. IX.7.1).
The trope verses Qui dicit patri and Mirabile laudat appear elsewhere only in a small number of Italian sources, from Novalesa near Mont Cenis, Pistoia, and Benevento. Hodie exultent, however, turns up in a slightly different range of Italian sources and in the two earliest St Gall tropers, St Gall, Stiftsbibliothek 484 and 381. The text was also copied into Munich, Bayerische Staatsbibliothek, elm 14843, which some scholars have dated as early as the ninth century. The second part of the verse, 'resurrexit leo fortis deo gracias', is what usually follows 'Alleluia resurrexit dominus' in the dialogue at the sepulchre. What has happened? Has that sentence been lifted from the dialogue as part of a revision, and if so, was it done at St Gall or in Italy? At St Gall Hodie exultent iusti starts a series of verses, rather than finishing one off as in Piacenza and other Italian sources.
In a recent discussion of some other verses beginning 'Hodie . . .' Reier pointed out (a) that they had wide distribution in Italian sources, and (b) that they all had the same melody (Reier 1981, i. 157). Although Reier's examples were introductory tropes, unlike Hodie exultent iusti in Piacenza, Hodie exultent has exactly the same melody as the others. That is the same as Hodie resurrexit leo fortis in other versions of the Easter dialogue. So we should at least consider the possibility that Hodie exultent iusti is an old Italian introductory trope, perhaps displaced by the popular Quern queritis dialogue, which in turn in some versions took a verse from Hodie exultent iusti. This sort of reasoning carries no conviction if the music is not taken into account. It is the use of the same melody for several different trope verses which is unusual and alerts one to the possibility of an 'original' layer of Italian material.
After these trope verses Piacenza 65 has the Easter Alleluia Pascha nostrum with a prosula for the 'Alleluia'. There follows the sequence Clara gaudia festa paschalia. The melody of this sequence was widely known, being called 'Angelica' and 'Nobilissima' at Winchester, 'Musa' in Angers, Bibliotheque Municipale 144, and 'Romana' in St Gall (which may not be entirely fanciful). Notker composed two texts for it, Johannes jfesu Chris to and Lauren ti David magni martyr. Clara gaudia was known in France, as was another Easter text for this melody, Die nobis, and another
7. North Italy 587
Ex. IX.7.1. Trope verses Qui cheat patri, etc. for introit Resurrexi (Piaccnza, Bibl. Cap. 65, to. 235v)
RESUR-RE-XI ET ADHUC TECUM SUM        AL - LE - LU - IA.
Tl
12
Qui di - cat pa-        -tri   pro-phe-ti-ca uo-ce.
T2
T3
fig   p   9*    »'09    9J9    ,t4»    0    49   £99    i*Mg   40       9»  ,4*    • M*»9*
PO-SU-IS-TI       SU-PER   ME MANUM TU - AM    AL-LE -LU-IA.
0    m     "»0»0^jf4*     ' '        '*»4    9     & 0
Mi-ra-bi-le lau-dat fi-     -li-   -us patrem.
^Ct9   4    9    ČT"  gjyp   ^   0fr 9 •     •   ^   *   0Pf ?U f0
MI-RA - BILIS  FAC-TA     EST  SCI-EN-TI-A      TU - A   ALLE-LUIA AL-LE-LU - IA.
PS   j)     <j >    '   i   " 9   •    0    9    *    9aj9 0-i00±
DO-MI-NE PROBASTI ME (reS5      . ET RESURRECTIO-NEM MEAM.
omitted)
* 9s
9 *• '*  0Cp  0    0       • é.
Ho-di -e   ex-ultant ius - ti       re-sur - re-xit   le - o     for-tis    De-o gra-ti-as
di-   -ci-te e  -  ia.     (followed by cues for RESURREXI and GLORIA PATRI)
text for martyrs, Candida contio. In the earliest substantial Italian collection of sequences, Monza, Basilica S. Giovanni 13/76, Clara gaudia and Notker's two texts all appear. Previous discussion of the sequences has concentrated on Notker's texts and their possible models. Crocker (1977, 146-59) treated Clara gaudia and Die nobis as 'West Frankish' but thought neither a wholly convincing starting-point for Notker. He recognized that the melody stood rather outside the mainstream of the sequences he was discussing, but did not pursue the matter further. In fact the melody bears all the hallmarks of an Italian composition (see above, II.22.v): repetitive melody and consequent restricted range, with occasional groups of two or three notes to a syllable (in Italian sources; they are often 'tidied up' in foreign ones). And its prominent place on Easter Day alerts one to the possibility that in Piacenza 65
588
IX. Persons and Places
(as in many other Italian sources: see Brunner 1985, 220) the piece stands in its original place. The text is dramatically colourful, an account of the harrowing of hell, as Crocker observes, with demons howling in line 5, saints acclaiming the Lord in line 6 (Ex. IX.7.2).
Once again it is musical features which point in the direction of what may be the 'original'. In cases like this, Italian chant does not sound like chant everywhere else.
Ex. IX.7.2. Sequence Clara gaudia festa paschalia (Piacenza, Bibl. Cap. 65, fo. 236r)
<k      m ±00       *        m        ifi       *0^ 0
6
Cla-ra gau-di - a      fes - ta pas-cha-li - a
Mi* • •   * •   #  i*   *)   * m   »   *  0   *   0 m
Congaudent cetus per om - ni - a Dul-ce de-cantant al - le - lu-ia
In qua Christus per crucem redemit a - nimas in-fer-no de-di-tas A proto - plas-to quotquot in hoc secu-lo  pro-ge-ni - te fu-e-rant
Pa-tri-ar-charum om-niumque simul prophe-tarum regum ponti - fi - cum De - ti - nebantur claustra tartare-a mortis cru-o-re retru-...... -sa
-•-«—*—*—a-*-s-TX-•-*"-a-
00     m    0 0     * ' 0*0*-*&- -*-
Donec uic-tor mortis do-minus omni-um at-que sanctus san-cto-rum Cum crucis trophe-o  in-fernum pe-netrans ab-e - git claustra   se - ua
00   0   0   0   *   *   *  *   *   *  * *   *   .   *   •   0   *b .   *-6 *  * 0
Quis  ... est de-mones u - lulant cruci - fer    qui nostra ut  de-us soluit uin-cu-la cuncta Fu-ga-tis te-nebris fulgu-rat the-a-1ra - lis  orror ru-ti-lat   lu-mi-ne perlustra-tus
-2-a—r—■—---—•—•—s-nrr
-0-*-•-«-*-•-0*0W
0 9
Clamabant sancti adue-nis-ti   O iam domi-ne   re - gum rex a-ue Quern o-lim ua-tes pre-ci-ne-re iam nos habet re-demptor rex Christe
*     a     *    *     '     •     *     a *
•*     0 0
Tunc  ... Ie-sus cum leta sanctorum ag-mi-ne     proces-sit   mor-te uin-cto Cu - i psal-le - re laudes cum omni cardi-ne      non cessat   om-nis e-tas
Decantant al-le-lu-ia.
8. Benevento and Montecassino
589
IX.8. BENEVENTO AND MONTECASSINO
The existence of a non-Roman rite and corresponding non-Roman chant in the Lombard duchy of Benevento has already been discussed (VIII.5). The principal sources of Old Beneventan chant are in fact Gregorian books with relics of the displaced repertory still copied alongside the imported one. Of course, these books are also important as sources of Gregorian chant, with the local peculiarities of repertory and melodic readings we should expect of any group of Gregorian sources. Most of the surviving chant manuscripts are from Benevento itself, which has slightly obscured the fact that the great Benedictine monastery of Montecassino lay in the same area and shares many details of chant practice with Benevento (see Planchart 1990). Kelly (1985, 'Montecassino', 1989) has made it clear that Old Beneventan chant was sung at Montecassino as well. Unfortunately, Cassinese sources are fewer and later, and it is therefore difficult to assess its contribution to the establishment of Gregorian chant and the burgeoning repertory of sequences and tropes in the Beneventan region. (The sequences are catalogued by Brunner 1985, the trope repertory is being edited by Boe and Planchart in the series Beneventanum Troporum Corpus.)
The principal sources from the two centres, which date from the early eleventh to the thirteenth century, are the following (in Benevento, Archivio Capitolare, Montecassino, Archivio della Badia, and Rome, Biblioteca Apostolica Vaticana):
from Benevento:
noted missals: Benevento 33 (PalMus 20) and 30 (partially notated) graduals: Benevento 38, 39, 40, 34 (PalMus 15), and 35 combined missal and breviary: Benevento 19—20 noted breviary (winter part): Benevento 22 antiphoner: Benevento 21 (text edn. by Hesbert in (AO 2) from Montecassino:
noted missals: Rome, Vat. lat. 6082 (possibly for San Vincenzo al Volturno) and
Montecassino 540 (used at Plombariola) gradual: Montecassino 546 (winter part) antiphoner: Montecassino 542 (winter part)
troper: Urb. lat. 602 (inventory and partial reconstruction Boe 1985)
theory compendium and tonaries: Montecassino 318 (RISM B/III/2, Huglo 1969,
Brunner 1981)
related sources:
graduals: Rome, Vat. lat. 10673 (only a large fragment remains; PalMus 14) from Apulia or possibly Benevento, and Paris, Bibliotheque Nationale, n. a. 1. 1669 from Gubbio
noted missals: Rome, Barb. 603 from Caiazzo, Baltimore, Walters Art Gallery 6 from
Canosa (text ed. Rehle 1972-3), and Rome, Ottob. lat. 576 (partially notated) troper: Naples, Biblioteca Nazionale VI G. 34 from Troia
590
IX. Persons and Places
(Assignment of these sources to particular institutions and dates is occasionally revised, but the general area is not in doubt. See Boe 1985, Kelly 1989, and the catalogue (in progress) of sources in the Archivio Capitolare at Benevento by Mallet and Thibaut.)
This short list does not mention other types of liturgical books or the numerous fragmentary sources, some from Dalmatia (the Adriatic coast of modern Yugoslavia). In particular, a number of splendid rolls bearing the Exultet chant for the lighting of the paschal candle on Easter Eve have been preserved (Avery 1936, Cavallo 1973). The distinctive script and notation of manuscripts from the area has stimulated scholars to hunt down a great number of sources, and we are relatively well informed about the extent and nature of the surviving corpus (see Loew and Brown 1980).
The story of the supersession of the Old Beneventan rite and chant by the Roman rite and Gregorian chant is a complicated one but very interesting. A great deal of it has been reconstructed by Hesbert (1938-47, PalMus 14, etc.), and Kelly (1989, 18-25) provides a recent summary. The version of the Roman rite as it appears in its earliest Beneventan sources (Benevento 33 from around 1000) has many archaic features which had long been overhauled in the Frankish empire, but which persisted in Rome itself, as may be seen in the Old Roman chant-books of the eleventh century. That the Beneventan books with Roman order of service have Gregorian chant has been taken to denote Frankish, rather than specifically Roman influence, and it has been argued that the Gregorian chant must have arrived in Benevento at the time when the liturgical archaisms were still current in the Frankish empire, that is in the eighth century. It is clear from the music manuscripts themselves that Gregorian and Old Beneventan chant lived side by side, so to speak, for over a century. Could they have coexisted since as far back as the late eighth century? Kelly sees Paul Warnefrid (the Deacon), biographer of Gregory the Great, as a possible agent for Frankish-Gregorian chant in Benevento (Kelly 1989, 22-5). The age of the introduction of Gregorian chant seems to be confirmed by the fact that the Feast of the Twelve Brothers, introduced at Benevento in 760, has Old Beneventan chant, whereas for the Feast of St Mercurius none survives, nor for two saints whose use was taken up at Benevento in the ninth century: Januarius and Bartholomew. Yet for Barbatus, whose cult also developed in the ninth century, some Old Beneventan antiphons do survive.
The role of Montecassino, St Benedict's own monastery, is unclear. Founded in the early sixth century, it was destroyed by the Lombards in 577 and had to be refounded by Petronax of Brescia in 717. In the interim monks from Fleury had carried off the mortal remains of Benedict and his sister Scholastica. So the Frankish 'acquisition' of Roman chant had a tangible monastic counterpart. What the chant of Montecassino may have been like in the eighth century we do not know, which is unfortunate because Boniface sent his disciple Sturmi, future abbot of Fulda, to Rome and Montecassino for two years to study the Benedictine rule. Charlemagne sent there in 787 for a copy of the rule. In 883 it was burned by the Saracens and remained deserted for sixty years. There follows the period for which Old Beneventan chant can be
8. Benevento and Montecassino
591
documented at Montecassino, after which we have the following valuable record of moves to introduce Gregorian.
What the non-Italian popes who occupied the Holy See from the end of the tenth century and again from 1046 onward thought of the Old Roman chant they heard in Rome, so different from the Gregorian chant of their native lands, we do not know. It was by the German Gregory V (996-9) that a German type of pontifical was introduced in Rome (see III. 12), but we have no specific record of changes in chant practice in Rome until much later. In the mid-eleventh century Montecassino had two German abbots in succession, Richerius and Frederick, the latter becoming Pope Stephen IX in 1057. In 1058, on a visit to Montecassino, he forbade the singing of 'Ambrosianus cantus', by which is meant Old Beneventan chant (MGH (Scriptores) Die Chronik von Montecassino, 352—3). The task of suppressing it must have been that of the great abbot Desiderius, the future Victor III, whose abbacy marks a cultural high point in Montecassino's history (Bloch 1985, Cowdrey 1983). It is disappointing that practically no musical source of this time has survived, with the exception of the huge theoretical compendium, Montecassino 318.
The Gregorian tradition established at Benevento and Montecassino and the repertories of tropes and sequences, though naturally containing distinctive local details, were by no means parochial. Many of the tropes and sequences known there appear to have travelled long distances (see Brunner 1981), and the melodic variants in the proper chants of mass, collated by the monks of Solesmes (Graduel romain IV), are surprisingly close to those of Aquitanian sources. The reasons for this are not clear.
In the late eleventh century and twelfth, Norman invaders conquered South Italy and Sicily. With them came churchmen and the chant traditions of their native land (Hiley 1981, 1983, 'Quanto', 1988). But this seems to have made little impact upon Beneventan and Cassinese practice. Only one manuscript shows a merging of the different traditions. This is the troper from Troia, Naples VI. G. 34, compiled late in the twelfth century (RISM B/V/l, 175-6; one page facs. Arnese 1967, 146-9 and pi. X, Stablein 1975, 145).
The manuscript is written in Beneventan script and notation. Even as late as this, there are Old Beneventan remains: the processional antiphon Deprecamur te domine in an Old Beneventan version (Levy 1982, 95) and the Old Beneventan Peccavimus domine (Kelly 1989, 284), hardly more than a threefold variation of the same simple phrase (Ex. IX.8.1).
But the winds of change have blown strongly. The manuscript has almost nothing in common with the great Beneventan graduals: two of its sequences only are taken from the traditional Beneventan repertory (Quipurgas animas and Precursor Christi). The Laudes regiae, the acclamatory chant for the crown-wearing of the ruler, sung after the Gloria on the highest feasts of the year, have in this manuscript a version that is peculiar to Norman sources (Kantorowicz 1946). Practically all the Kyrie and Gloria melodies and tropes are taken from the Norman tradition (inventory in Hiley 1986, 'Ordinary'; the Sanctus and Agnus are lost from the end of the manuscript).
592
IX. Persons and Places
Ex. IX.8.1. Rogation antiphon Peccavhnus Domine (Naples, Bibl. Naz. VI. G. 34, fo. 6r)
/7
i      '  ' ' '      r» ——
i
_._- *—i—r
Pec-ca - ui - mus do-mi-ne pec - ca - ui-mus
j J__
4-6   *>   0    0     '   *    #3 .-
et   pec-ca-ta   nos-tra cog-nos-ci - mus
•     * m
clemen - tis-si - me   De - us   do - na   no - bis ue-ni-am.
Next come farsed lessons, one of the most extensive collections known (see II.23.xii and Stablein, 'Epistel', MGG), with several unica. Many of these were also known in Sicily and north France, though, since their south Italian sources are equally early, it is not clear where some of these pieces originated. Were they brought by the Normans, or taken back north by them? Ex. IX.8.2 gives the start of the lesson Lectio libri Sapientiae . . . Qui timet Deum faciet bona for St John the Evangelist with the farsing verses Ad laudem regis glorie unique to this manuscript. The melody of the lesson is largely freely composed; its prose contrasts strongly with the regular, rhyming, farsing verses.
Most of the incomplete collection of alleluias found in Naples VI. G. 34 was not known in Benevento, and the sequences include some new rhymed compositions, typical of north France, particularly Paris, in the later twelfth century: Potestate non natura, Exultemus in hac die, and so on. The sequence for Troia's patron saints Pontianus and Anastasius is unfortunately incomplete, so that an example of local composition is spoiled for us. St Secundinus has his name inserted in a common sequence borrowed probably from Norman use.
The type of manuscript itself is not one we normally associate with Benevento, where ordinary-of-mass chants, tropes for the proper of mass, and sequences were usually entered in place within each mass, as is often the case in Italian sources. The nearest to it is Vatican Urb. lat. 602, which, however, seems to have had no sequences or farsed lessons. To sum up, Naples VI. G. 34 is a manuscript where the type of notation, traditional Beneventan, gives no clue to the origin of the chants, which range from Old Beneventan through Norman and beyond. The study of overlapping palaeographical and repertorial traditions in manuscripts like this is one of the most fascinating areas of chant research.
8. Benevento and Montecassino 593
Ex. IX.8.2. Start of farsed lesson with verses Ad laudcm regis glorie, etc. (Naples, Bibl. Naz. VI. G. 34, fo. 46v)
Tip
L2
Ad    lau-dem   re - gis     glo - ri - e uox in - to - net ec-cle-si-e
prop-ter   Io - han-nis   me-ri-ta hec  re - ci-tans pre - co-ni - a.
li fft   /.« tt*
LEC - TI-0 LI - BRI     SA - PI -EN-TI - A.
y
T2 ftp      *     ft    0    '    "    99-*   * ft   0   .   p    #    *g   . 4
Pro-cla-mat sa - lu - ber-rime spi-ri-tus sancti   car - mi-ne
quam fi - de - les     per-pen-di -te
f'»^    0-*'-*•-9-**-*-0--*-
CM'I   TI - MET DE-UM FA - CI - ET BO-NA.
T3 fff\ _      ,    »   f 3fc
-»-*v-Q   t-«2j-■-•—'** _^ 0 0
Ut per-ci-pi-at gau-di - a con - di - to-ris perhempni-a. ^_2____£_
L3 jfo       .    0   A     _ ^    f-^   0 0*
ET QUI CONTINENS EST IUSTI-TI - E APPREHENDAT IL-LAM
-gy 0 , 0 00 00 #g   f< ^      «   * # 0 * ^=
ET OB-UI-A-UIT IL-LI   QUA-SI MATER HONORI-FICATA.
594
IX. Persons and Places
XI.9. ROME AND THE FRANCISCANS
Hiischen, 'Franziskaner', MGG; Abate 1960; van Dijk and Walker 1960; van Dijk 1961, 1963, Sources', Salmon 1967; Hesbert 1980, 'Curie'.
I have adopted the view above (see VIII.8) that Old Roman chant was the only chant sung in Rome until about the eleventh century, and that when we meet it in its earliest notated form in 1071 (the date of manuscript Bodmer C. 74) it is a somewhat altered descendent of the chant which the Franks learned from Roman cantors. We do not know how often non-Romans noticed differences from the chant of the Franks. We may assume—though we have no record of it—that the first foreign popes, installed when the Emperor Otto III was trying to reform the papacy, would have been aware of the contrast: the German Gregory V (996-9), and the Frenchman Gerbert of Aurillac, Sylvester II (999-1003). However, as already mentioned (IX.8), it is in the wake of the series of German popes in the mid-eleventh century that we find Stephen IX objecting to 'cantus Ambrosianus' at Montecassino. That is, of course, not Old Roman chant; the point is, it is not Gregorian either, and Gregorian chant is what was sung thereafter at Montecassino. And it may be significant that the earliest surviving Old Roman chant-book dates from a time when the Old Roman repertory may have been coming under pressure, when it may have been felt necessary to take steps for its preservation.
The type of pontifical codified at Mainz in the mid-tenth century, the so-called 'Roman-German' pontifical, was being used in Rome shortly thereafter, brought, it is thought, by Emperor Otto I on one of his many visits. This copious and well-organized collection of pontifical ceremonies is a different matter from a gradual or antiphoner, of course. But it is indicative of the way a non-Roman document could alter Roman liturgical practice. On the other hand, the pontifical compiled for Innocent III (1198-1216) still includes three Old Roman antiphons (Huglo 1954, 'Vieux-romain', 102—4).
Gregorian chant could also have been sung close at hand to the papal chapel from the time when Alexander II (1061-73), a former canon of St Fridian's at Lucca, reformed the monastery attached to the Lateran, whose personnel performed the services in the Lateran basilica (the pope and his chaplains celebrated in the adjacent chapel of St Lawrence). The reform was confirmed in 1105 when the Lateran Congregation of Canons Regular was founded as a descendant of St Fridian's (van Dijk 1961, 424). The ordinal of the canons as codified in the mid-twelfth century is known (ed. Ludwig Fischer 1916), but it contains no chant. (See Gy 1984.)
The first clear evidence that mass and office were performed with Gregorian chant in the papal chapel comes from the thirteenth century, when a series of revisions brought all liturgical practice in Roman churches, papal or civic, into harmony, with which the Franciscan order of friars concorded. Of Innocent Ill's revision of mass and office of the papal chapel in 1213-16 only the ordinal survives (ed. van Dijk and
9. Rome and the Franciscans
595
Walker 1975). Honorius III (1216-27) made another revision of the breviary for outside use, which the Franciscans adopted in 1230. They adopted a revised missal shortly afterwards. Both breviary and missal were revised by Haymo of Faversham, General of the Franciscan Order from 1240 until his death in 1244, at the request of Innocent IV (1243-54) (ed. van Dijk 1963, Sources). From this point papal and Franciscan use were to all intents and purposes the same. The Roman city churches appear to have come into line somewhat gradually. In the 1250s Cardinal John Cajetan Orsini devised a compromise liturgy to combine the use of the Vatican (urban) and the Lateran palace (papal), but on becoming Pope Nicholas III (1277— 80) he ordered the destruction of all non-conforming books (van Dijk 1956, 1960).
We can therefore tell what chant was sung in the papal chapel by inspecting notated Franciscan books (papal ones do not survive). The noted missal Naples, Biblioteca Nazionale VI. G. 38 (facs. Arnese 1967, pi. XVI, PalMus 2, pi. 30) has the liturgy of Honorius Ill's revision (van Dijk 1969-70). The earliest books have central Italian notation (Naples VI. G. 38; see also Hiischen, 'Franziskaner', MGG; and NG iii. 269 for a breviary of 1224). The preface to the gradual of the 1250s specifically states that square notation should be used (ed. van Dijk 1963, Sources, ii. 359; see also i. 110), which by that time was rapidly becoming universally popular.
The Solesmes survey of variant readings on proper-of-mass chants places Naples VI. G. 38 with other Franciscan books in a group showing no strong resemblance to any other. Hesbert's investigation of the responsory series (1980, 'Curie') showed that, once evident innovations have been set aside, an affinity with other central Italian uses became clear—Perugia, Carinola, Florence, and Piacenza. Possibly some such process led to the establishment of the musical readings. The Roman-Franciscan gradual appears to have been adopted by the Celestines, the order of Benedictine monks founded by Celestine V in 1250, to judge from the single source investigated by the monks of Solesmes. On the other hand, Celestine office-books follow the Montecassino order of service. The Olivetans, founded in 1319 by Giovanni Tolomei on Monte Oliveto near Siena, adapted the Roman-Franciscan breviary for monastic use. The same happened in monasteries descended from the reform at Santa Giustina in Padua in the early fifteenth century. Subiaco was reformed from Santa Giustina, and German monks took the reform to the newly founded Melk on the Danube, whence the Roman-Franciscan use in a revised monastic form spread to other Austrian and south German monasteries.
Tonaries survive which were composed by two clerics attached to the curia (Huglo 1971, Tonaires, 225-9), though neither has the status of an official document. The treatise Practica artis musicae written in 1271 by the English priest Alfred (or Aluredus or Amerus), in the entourage of Cardinal Ottobono Fieschi (the future Hadrian V) contains two tonaries, one compiled according to French and English use, the other according to Roman (ed. Ruini in CSM 25). By contrast, Elias Salomon's Doctrina scientiae musicae of 1274 (GS iii), though compiled at the curia, does not reflect its usage.
Churches outside Rome did not stand under any obligation to adopt the Roman-
596
IX. Persons and Places
Franciscan reforms. Even the Lateran basilica celebrated its own form of the office until the late fourteenth century.
The attitude of St Francis and the Franciscans to liturgical music is discussed by D'Angers (1975) and also by Huschen (MGG). The rhymed office of St Francis was edited by Felder (1901).
IX. 10. AQUITAINE AND SAINT-MARTIAL AT LIMOGES
PalMus 13; MM MA 3; Crocker 1957, 1958; Chailley 1957, 1960, Ecole; Emerson 1962, 1965; Herzo 1967; Treitler 1967; Steiner 1969; Paul Evans 1970, Repertory, 1970, 'Elements'; Brockett 1972; Roederer 1974; Planchart 1977; Planchart and Fuller, 'St Martial' NG.
A number of important chant manuscripts have survived from south-west France, that is roughly south of the Loire and as far east as the Saone-Rhone valley, corresponding more or less to the Aquitaine, Gascony, and Septimania of Charlemagne's empire and generally known for convenience as 'Aquitaine'. The manuscripts are usually immediately recognizable by their distinctive notation, one of the earliest whose pitches can be read accurately. They have attracted a great deal of attention, partly because of their legibility, but also because they contain large repertories of tropes and sequences, and because a late sub-group among them contains numbers of the rhythmic, rhyming songs of the twelfth century, some in polyphony. A good number of them passed through the hands of the librarian of the abbey of Saint-Martial at Limoges, Bernard Itier, in the years around 1200, and the label 'Saint-Martial' has sometimes been applied to more or less the whole group. Only a handful can be shown definitely to have been written for Saint-Martial, a few more possibly for other churches in Limoges, while the majority are not from Limoges. The provenance of several is unfortunately not at all clear.
While the sources do indeed share common traditions, to a greater or lesser extent, they are naturally fortuitous survivals, which leave some awkward gaps in our knowledge. The earliest important liturgical source, the unnotated combined gradual and antiphoner Albi, Bibliotheque Municipale 44, dates from the second half of the ninth century. But then there is a complete gap for both types of book for over a century. The earliest collection of tropes and sequences is dated around 930, followed by two particularly large and important collections of a similar nature from the closing decades of the century. We have no notated graduals until well into the next century, and no complete notated office-book until the fourteenth century. (A possible exception is the twelfth-century antiphoner Toledo, Archivo Capitular 44. 2. Its notation is Aquitanian and it is usually said to come from Aquitaine, but its closest 'relatives' in Hesbert's survey—CAO 5—are all Spanish.)
The principal manuscripts, in roughly chronological order up to about 1200, are listed in Table IX. 10.1. It goes without saying that the label 'troper' covers a great
/0. Aquitaine and Saint-Martial Table IX. 10.1. Aquitanian sources to cA200
597
Manuscript	Type		Provenance	Date
Albi 44	gradual-antiphoner			9th c, second half
Pans 1240	troper (p o s t)		Saint-Martial	c.930
Pans 1084	troper (p s t)			late 10th c.
Pans 1118	troper (p s t)		Auch ?	late 10th c.
Paris 1085	antiphoner		Saint-Martial	late 10th c.
Pans 1120	troper (p o s f)		Saint-Martial ?	r.1000
Paris 1121	troper (p s f t)		Saint-Martial	r.lOOO
Pans 887	troper (p o s)		?	early 11th c.
Pans 1138-1338	sequentiary (s)		Limoges	early 11th c.
Paris 909	troper (p s f t)		Saint-Martial	early 11th c.
Paris 1119	troper (p s)		Saint-Martial	c.1030
Pans 1137	sequentiary (o s	f)	Limoges	f.1030
Paris 903	gradual (p o s f)		Saint-Yrieix	11th c, first half
Paris 1133	sequentiary (o s	f)	Limoges	mid-11th c.
Paris 1135	sequentiary (o s	0	Limoges	mid-11th c.
Pans 1136	sequentiary (o s	0	Limoges	mid-11th c.
London 4951	gradual (f)		Toulouse	mid-11th c.
Paris 776	gradual (f t)		Gaillac near Albi	late 11th c.
Paris 1132	gradual (o s f)		Saint-Martial	late 11th c.
Paris 1134	sequentiary (o s	f)	Saint-Martial	late 11th c.
Pans 779	troper (p o)		Limoges?	late 11th c.
Pans 1871	troper (p o s)		p	late 11th c.
Pans 1177	sequentiary (o s	0	?	late 11th c.
Paris 780	gradual (f t)		Narbonne	late 11th c.
Paris 1139	song-book		?	early 12th c.
Paris 3549	song-book		?	12th c.
Paris 3719	song-book		?	12th c.
Pans 778	sequentiary (o s)	1	Narbonne	late 12th c.
Paris 1086	sequentiary (o s)		Saint-Leonard, Noblat	late 12th c.
p = tropes for proper-of-mass chants
o = ordinary-of-mass chants
s = sequences
f = offertory verses
t = tonary
variety of contents, and I have used it here basically to separate books without the full range of mass-proper chants from graduals. 'Sequentiary' means that the book does not contain tropes for proper-of-mass chants. Two other important sets of chants frequently found are ordinary-of-mass chants and offertory verses. Several of the manuscripts contain tonaries (studied by Huglo 1971, Tonaires). But many of them
598
IX. Persons and Places
contain much more besides: tracts, alleluias, gradual verses, processional chants, special offices, and so on. (For summaries of the contents of most of these manuscripts, see Emerson, 'Sources, MS, §11', NG; Graduel romain II; RISM B/V7 1.) The later books with rhymed versus and Benedicamus songs I have simply called song-books. The dating of many sources is also only approximate; in many cases the manuscripts are composite, containing sections of different sorts written at different periods, brought together for convenience at binding more than for any musical reason.
Two manuscripts now in the chapter library of the cathedral at Apt are also usually considered in conjunction with the Aquitanian sources. Apt 17 is written in Aquitanian notation, while Apt 18 rarely uses it and mostly employs French notation. (This is not in itself surprising, for French notation was used even over the Alps at La Novalesa.) The provenance of Apt 18, the earlier of the two, from c\1000, is unclear, while Apt 17, of the mid eleventh century, is from Apt itself. (See Bjorkvall in CT 5.)
Benedict of Aniane (c.750-821), the great monastic reformer and liturgical organizer of Louis the Pious's reign, came of an Aquitanian family and as early as 779 established in his homeland a monastery whose practice was widely imitated. (Late in his life he had the monastery of Inde, or Cornelimunster, near Aachen, built for him by Louis, which was supposed to be a model for other houses.) But it seems unlikely that the later Aquitanian chant-books contain anything that can be identified as springing directly from his work in that region. On the other hand, a number of survivals of Gallican chant have been traced, especially in the graduals Paris 776 and 903 (see VI11.6).
It is, however, for their repertories of sequences and tropes that the majority of the manuscripts are chiefly prized. The literature on their tropes is by now particularly extensive, and only a few points need be made here. Although by repertory the trope collections are related closely to one another, they are by no means uniform, and the interplay of traditions and recasting of material is evident to a considerable degree. This sort of variation between sources characterizes the transmission of tropes right across Europe, of course, and often the Aquitanian sources stand together against others. But within Aquitaine the variations are often great. In Weiss's edition will be found many instances of strongly variant melodies, particularly involving Apt 17. Planchart and others have pointed out that the earliest troper, Paris 1240, often shares variants with north French manuscripts, rather than with later Aquitanian sources. In these cases it may represent an imported early tradition that was subsequently reworked by Aquitanian musicians. (There are outlines of the relationships between sources in Weiss 1964, 'Problem', David Hughes 1966, Hiley 1983, 'Observations'; the easiest way to see the varied selection and arrangement of trope verses is through the volumes of CT and especially Planchart 1977; for further literature see 11.23.)
The sequences, too, are found in many and varied traditions (inventories by Crocker 1957, many editions in Crocker 1977). There is a striking contrast between the moderate collection in Paris 1240 and the vast series in Paris 1084 and 1118. The later sources opt for sizes in between those extremes. In Paris 1086 and 778, and in an
10. Aquitaine and Saint-Martial 599
added section of Paris 1139, the new sequences with regular rhythm and rhyme, associated with Adam of Saint-Victor in Paris, enter the repertory (Husmann 1964).
Two historical events impinged upon the monastery of Saint-Martial and left traces in its musical manuscripts. In the 1020s a movement was set afoot, whose chief apologist was Adhemar of Chabannes, to gain for Martial the rank of apostle. In 1031 the Council of Limoges declared the apostolicity authentic, although the decision had little effect elsewhere (except possibly to stimulate other churches to promote the cause of their own saints). But at Saint-Martial a new mass (beginning with the introit Probavit, replacing chants from the common of saints beginning with the introit Statuit ei) and a special office were composed. Other saints associated with Martial and with Limoges, Austriclinianus and Valeria, had offices composed at the same time (Emerson 1965). It is likely that Adhemar composed them all. The prime source for these is the manuscript Paris 909, whose physical make-up was extensively rearranged to take in the new material. Ex. IX. 10.1 shows the last of the responsories of the Night Office from the office of St Martial. It has two verses, rather than the usual one, both of which, with the Gloria, have the same melody. Two new melismas, both with form AABBC, are provided at the end, to replace the music for 'omnium' when the last part of the responsory is repeated after the verses. The probable order of performance would then have been:
Respond
VI, repetenda with melisma 1 V2, repetenda with melisma 2 Gloria, repetenda
Ex. IX.10.1. Responsory O sancte Dei apostole (Pans, Bibl. Nat. lat. 909, fo. 68r)
i
^» ™  t 9- '- -*—*-
R-      O sancte    De - i    a - pos- -to-le
P
O Mar - ti - a-
-lis   pnnceps e - gre-gi-e
pas - tor      et  pro-to-dux     A- -qui-ta - no - rum
*-"-*-*0m  0m 0
0, 0, «„» * ■ * s .™#^ t
au-di       prae- -ces seruu-lo-
-rum
if—•-i*—"—"—*—*—1-:—
Rep.   et in-ter-ce  - de pro sa-lu-te
3zr
0 v^** 0§z
om-
-ni-um po-pu-lo
rum
600
IX. Persons and Places
(Ex. IX. 10.1 cont.)
VI-      Ad- -nun-ci - as - ti     o-pe-ra do - mi - ni
—a-*-~-a   '* a'i—*^ -»   ^ »9-
et   facta   e - ius    in-tel-le- -xis - ti. Etin-ter...
\ U  «*f«   f.           m an* '*'' «•    *   a    » a\=& a w »t»  ,'r-- * —
V2.      Prae- -fulgens glo-ri - o-sus   in conspectu do- -mi-ni
m   a    a   a    .    n'   *■    *   '    *»#»,H  a.      *-- ^
a   * ' -'     a      »•»-jr;-' &
tu - o-rum susci-pe   preces ser-uu-lo- -rum. Et in-ter...
HE
U t»*0      ... »   a.     a   a   .   a    *   •    t ip *«,.#H
-J  m 9na—j*--w- -
Gl-      Glo- -ri-a     patri et fi-li-o   et spi-ri - tu-i sancto
et nunc et  per e - on. Et in-ter...
Ml       Om- -ni-um (populorum)
^    »'« uaa       a>rM «fl   -IU   "^t^^i   -II .....«'3««.,;
M2       Om- -ni-um
(populorum)
In 1062 Saint Martial was reformed from Cluny. The change can be seen in the order of chants, for example the post-Pentecost alleluia series in Paris 1132 and 1134. Table IX. 10.2 sets out three different series to be found in Limoges books: (a) that of the cathedral of St Stephen, (b) Saint-Martial before the Cluniac reform, (c) the Cluniac series. The Cluniac series of Advent responsories appears in several Limoges books from the same time onward, including the fourteenth-century noted breviaries Paris 783 and 785.
//. The Hispanic Peninsula Tabic IX. 10.2. Post-Pentecost alleluia series in Limoges
Limoges cathedral Saint-Martial     Cluniac Saint-Martial
before 1062
Paris 9438    Paris 1137    Paris 909* Pans 1132    Paris 1134
7: 12 7: 12 7: 12 7: 12 7: 12
17 17 17 17
20 20 20 20 20
30 30 30 30
46 46 46 46 46
64 64 64 64
70
77 77 77 77 77
80 80 80 80 80
87 87 87 87 87
89 89 89 89 89
92
94: 1 94: 1 94: 1 94: 1 94: 1
94: 3 94: 3 94: 3 94: 3 94: 3
97
99
104 104 104 104 104
107 107 107 107 107
110 110
112
113: 1 113: 1 113: 1
113B: 11      113B: 11      113B: 11 113B: 11      113B: 11
116: 1 116: 1
117 117 121 121
124
129 129 129 129 129
137: 1 137: 1 137: 1 137: 1 137: 1
145 145 145 145
146 146 146 146 146
147: 12 147: 12 147: 12
147: 14        147: 14        147: 14 147: 14 147: 14
Highest number of concordances (among these five sources): Pans 1137    Pans 9438    Paris 1134        Pans 1134    Paris 1132 (21) (21) (19) (22) (22)
*Paris 1121 has the same series as Paris 909, but there is a lacuna after 104.
602
IX. Persons and Places
Soldo Table IX.102:
The alleluias are designated bv the psalm from which their verse is taken. Where other verses from the psalm were used for medieval alleluias, the one used here is stated.
7:	12 Deus iudex iustus	104		Confitemini domino et invocate
17	Diligam te domine	107		Paratům cor meuni
20	Domine in virtute tua	110		Redemptionem misit
30	In te domine speravi . . . accelera	112		Laudate pueri
46	Omnes gentes	113:	1	In exitu Israel
64	Tc decet hymnus	1 13B	: 11	Qui timent dominum
70	In te domine speravi . . . eripe me	116:	1	Laudate dominum omnes gentes
77	Attendite populc meus	117		Dextera Dei
80	Lxultate Deo	121		Lctatus sum
87	Domine Deus salutis mec	124		Qui confklunt
89	Domine refugium	129		De profundis
92	Dominus regnavit decorem	137:	1	Contitebor tibi
94:	1  Venite exultemus	145		Lauda anima mca dominum
94:	3 Quoniam Deus magnus dominus	146		Qui sanat
97	Cantate domino	147:	12	Lauda Iherusalcm dominum
99	lubilate Deo	147:	14	Qui posuit
XI.11. THE HISPANIC PENINSULA AFTER THE
RECONQUEST
Sablayrolles 1911-12; Angles 1922, 1931, 1935, 1938, 1970; Corbin 1952, Musique; Donovan 1958; Fernandez de la Cucsta 1980, 1985; Huglo 1985, 'Penetration'; Castro 1990.
Whatever the importance of the Cluniac monks in the 'reconquest' of Spain for Christianity in the latter part of the eleventh century (see VIII.7), little tangible evidence of their liturgical chant has so far been identified in Spain, unless it be negative features such as the failure of the Quern queritis and Easter dramatic ceremonies to take root in the western half of the peninsula (Donovan 1958, ch. 6). Studies so far accomplished show that the Gregorian chant-books of north Spain depend heavily, as we should expect, on Aquitanian traditions, just as Aquitanian notation is found in Spanish books of this time (for example, Paris, Bibliotheque Nationale, lat. 742 is an antiphoner of <r.1200 from Ripoll with Aquitanian notation: Ripoll was reformed from Saint-Victoire at Marseille; see Lemarie 1965). The closest 'relatives' of the following books, as detailed in Graduel romain IV, are all Aquitanian: Madrid, Academia de la Historia 18 and 45 (both graduals from San Millan de la Cogolla, early 12th c), Madrid, Palacio 429 (noted missal, Castilian, late twelfth century), Toledo, Archivo Capitular 35. 10 (gradual, Toledo, thirteenth century). Hesbert's survey of responsory series (CAO 5-6) found the Cluniac order in books from Moissac, which was an important source of influence upon Spain, and also in a twelfth-century antiphoner from Santo Domingo de Silos. The eleventh-century Silos manuscripts London,  British  Library, Add.  30848 and 30850 (facs. in
II. The Hispanic Peninsula 603
Antiphonale Silense, text ed. by Hesbert in CAO 2, by contrast, share an order unknown elsewhere (the latter, like many other books, includes some survivals of the Old Spanish rite). If the sources now in Toledo were used there, then they reflect a variety of imported uses. The important twelfth-century antiphoner Toledo, Archivo Capitular 44.2, as already mentioned, may be Aquitanian, though all related sources are Spanish; the thirteenth-century antiphoner Toledo 44.3 is Carthusian; the twelfth-century breviary Toledo 48.14 is Camaldolese. Much work remains to be done on the musical traditions of these sources.
A number of important manuscripts containing tropes and sequences have survived, which, while displaying obvious contacts with south French traditions, have numerous individual contributions to the repertory. Among the most important are the Vich tropers, Vich, Museo Episcopal 105 (eleventh to twelfth century) and 106 (twelfth to thirteenth century), and the twelfth-century Huesca troper, which include tropes for the proper chants of mass as well a ordinary-of-mass chants and sequences. Gros (1983) has described and edited the text of the fourteenth-century processional Vich 117 and compared it with its sister manuscript in Erlangen, Institut fur Musikwissenschaft (Stablein 1975, 160-1). There are a number of manuscripts, from Tarragona, Burgos, Gerona, and Tortosa, with ordinary-of-mass chants and sequences (sometimes polyphonic). Of these (brief descriptions and bibliography by Husmann in RISM B/V/l), Barcelona, Biblioteca de Cataluna M. 911, from Gerona in the fifteenth century, may be singled out because of its large provision of responsory prosulas, more numerous than in any other known source (Hofmann-Brandt, 1971, Bonastre 1982)-. All these sources come from Aragon except for the one from Burgos in Castile.
The famous 'Codex Calixtinus', probably copied in central France at a centre such as Vezelay around 1160, was destined for the famous pilgrimage shrine of Santiago da Compostela (ed. Whitehill et al. 1944; music alone ed. Wagner 1931; on the nature of the source see Hohler 1972). A copy of it was made at Santiago in 1173 by Arnalt de Munt, a monk of Ripoll (Barcelona, Corona de Aragon 99).
Corbin (1952, Musique, with further information in 1960 Deposition) has surveyed Portuguese liturgical books and their chant. (See also Braganga 1975, 1985.) It is said that when Lisbon was captured from the Moors in 1147 by a joint Portuguese-English expedition, an Englishman named Gilbert was appointed bishop of Lisbon and introduced the Salisbury missal into use there (Almeida 1910-22, i. 574 n. 1). Queen Philippa of Lancaster is reputed to have heard the canonical hours daily according to Salisbury use (Almeida, i. 627), and may have inspired her sons, Saint Ferdinand (1402-43) and King Duarte (Edward, 1433-8), to do the same. Extant service-books to support these claims have not yet been discovered. Corbin's extensive survey—she was able to list over eighty sources—reveals the expected influence of Aquitanian practice, easily visible in the notation, with other French traditions, including the Cluniac, making a contribution. Numerous Cistercian books survive. Corbin describes a number of local liturgical customs and their music.
604
IX. Persons and Places
IX.12. NORTHERN AND EASTERN EUROPE
(i) Scandinavia and Iceland
(ii) Poland
(iii) Czechoslovakia
(iv) Hungary
The pattern which we have seen when church life was revitalized in Normandy at the beginning of the tenth century was repeated wherever the Roman rite and its plainsong moved into new regions of the West. Much of chant research in these circumstances consists in identifying the starting-point(s) of the 'colonization' and isolating those parts of the repertory composed in the newly established church. The starting-point can often be revealed by palaeographical as well as repertorial studies. The newly composed material was usually called forth by the cult of a local patron saint, which most often entailed a new sequence for the mass and a new office, very possibly modally ordered and with rhyming, rhythmic text. A few remarks have been made about the introduction of Gregorian chant in the Hispanic peninsula; this chapter looks briefly at countries at the opposite end of Europe: Scandinavia, Poland, Czechoslovakia, Hungary, and Dalmatia. I have not attempted to summarize all the results of what in many cases is very distinguished and extensive research, and have simply cited some of the more important studies.
(i) Scandinavia and Iceland
The three archdioceses of medieval Scandinavia coincided geographically with medieval, not modern political divisions. Nidaros (Trondheim) covered an area somewhat larger than modern Norway. Lund included Denmark and the south of modern Sweden, where the cathedral was in fact situated. Uppsala took in the rest of modern Sweden and also Finland. The church in Denmark was organized chiefly from north Germany, while that in Norway owed much to England. Liturgical practice in both lands would have been established in the late tenth and eleventh century. At the Reformation liturgical books were systematically destroyed and reused to bind and cover other books: vast numbers of fragments remain.
Numerous fragments from English sources of the eleventh and twelfth centuries have survived in Norwegian collections, many with notation, and these are no doubt indicative of an important source for Norwegian liturgies and their chant (see Toni Schmid 1944, Gjerl0w 1961, 1970, 1979). But as Gjerl0w has shown, other sources and influences were present: the Nidaros ordinary draws upon both Lanfranc's Deereta and the German commentators Bernold of Constance and Honorius of Augsburg (Gjerl0w 1968); while much of the order of services can be traced to English uses, no one point of departure emerges. Gjerl0w's intensive exploration of the sources of the Nidaros antiphoner (1979) weighs the evidence of all the
12. Northern and Eastern Europe 605
fragmentary remains, then the choice of chants, and shows once again a strong relationship with the various English traditions as well as some points of contact with German ones. The Nidaros breviary of 1519 has been published in facsimile (Breviatium Nidrosiense). The Nidaros sequentiary has been published in facsimile with transcriptions by Eggen (1968).
Remarkable evidence of traditions in Iceland, similar to the Norwegian, and deriving ultimately from the same sources, has been discovered and presented in exemplary fashion once more by Gjerl0w (1980). A proper mass and office for Iceland's patron, St Thorlac, have survived (ed. Ottosson 1959).
Apart from around 15,000 fragments with musical notation (Haapanen 1922-32, 1924, Andersen and Raasted 1983), only about a dozen notated Danish liturgical books survive, none of which are graduals or antiphoners (Abrahamsen 1923, Asketorp 1984). Stromberg showed that a pontifical from Reims must have been brought to Lund, interacting with the widespread Roman-German pontifical (Stromberg 1955), and Asketorp was able to support this conclusion to some extent with a study of the melodies (1984). Husmann's comparison of post-Pentecost alleluia series in Danish missals (1962 and 1964-5, 'Studien') shows the expected influence of north German uses. The Lund missal of 1514 has been published in facsimile (Missale Lundense).
For Sweden, the fragmentary remains of the printed gradual of Vasteras (c.1513) have been published in facsimile by Toni Schmid (Graduate Arosiense), with sequences for Sweden's patron saints. The Linkoping ordinal has been edited by Helander (1957). Kroon (1953) studied Swedish repertories of ordinary-of-mass chants, Moberg (1927, 1947) those of sequences and hymns. Special offices have been studied by Undhagen (1960), Onnerfors (1966), Milveden (1964, 1972), and Lunden (1976). St Bridget (Birgitta: c. 1303-73) founded the Order of the Brigittines at Vadstena in 1346. A breviary of their house at Syon near London has been edited by Collins (1969).
The thousands of fragments which survive in Finland have been surveyed by Haapanen (1922-32, 1924).
(ii) Poland
Morawski 1988.
The unbroken history of Roman Catholicism in Poland since the end of the tenth century has meant that a large number of medieval sources are preserved. Research in the last thirty years has been particularly energetic. Numerous individual sources have been analysed (results published in Musica medii aevi and Muzyka religijna w
606
IX. Persons and Places
Polsce: Materiály i studia). Most indicators point in the direction of Bavaria, Switzerland, and Austria as the ultimate sources of Poland's chant traditions, apart from books of the religious orders.
We are probably better informed about Polish sequence repertories than those of any other country, thanks to the work of Kowalewicz (1964), Morawski (1973), and Pikulik (1973-6, 1974—101 sources—and 1976). Pikulik (1978, 1990) has catalogued and analysed the ordinary-of-mass repertories (ninety sources), and also Marian alleluias (Pikulik 1984). Polish pontificals have been studied by Miazga (1981).
(iii) Czechoslovakia
Konrád 1881; Mutter 1930, 1931; Nejedlý 1904/rev. 1954-6; Schoenbaum 1960; Vamcký 1960.
The bishoprics of Prague (Bohemia) and Olomouc (Moravia) had contrasting histories. Until its destruction at the beginning of the ninth century by the Magyars, the church in Moravia was influenced as much by the Greek as by the Roman church. Thereafter it resembled Bohemia in coming under strong German influence. In the tenth century Benedictine monks, from the Reichenau, from Korvey in Saxony, and especially from St Emmeram at Regensburg, played a leading part in the Bohemian church. The results of this in surviving chant manuscripts have yet to be investigated. From the eleventh century collegiate chapters were increasingly important, and the later religious orders made their usual contribution.
Notations in medieval Bohemia have been discussed by Hutter (1930, 1931). Whereas in the twelfth century German notation signs were used, in the thirteenth the Laon type of Gothic notation was taken up. Prague already knew staff-notation in the twelfth century.
Plocek's splendidly detailed catalogue of sources now in Prague (1973), with numerous melodic incipits, gives a good idea of the range of surviving material (see also Podlaha 1910). Among the earliest and most interesting sources is Prague, Archív Metropolitní Kapituly, cim. 4, a collection of the twelfth to thirteenth century from Prague Cathedral of offertory verses and ordinary-of-mass chants (Spunar 1957), many of the latter being 'up-to-date' north French compositions. Numerous manuscripts from Czechoslovakian libraries were filmed by Stáblein and thus available for his work on hymns and the studies of Melnicki, Bosse, Thannabaur, Schildbach, and Hofmann-Brandt. The link with south German and Austrian traditions is clear, while for the repertories of sequences and ordinary-of-mass chants contact is evident with the newer traditions of the twelfth to thirteenth centuries coming from centres such as Paris. The Latin songs usually called cantiones were also written in large numbers in Bohemia and Moravia (Orel 1922). The offices of St Procopius and St Ludmila have been studied and edited by Patier (1970, 1986).
12. Northern and Eastern Europe
607
(iv) Hungary
Szigeti 1963; Dobszay 1985, 1990'Plainchant'.
The Hungarian church was formally constituted under King (St) Stephen I in 1001, when ten bishoprics under Esztergom (Gran) were established. The Hungarian bishops are reported to have decided in 1100 to normalize their liturgy according to the order of service in Bernold of Constance's Micrologus de ecclesiasticis officiis (Kennedy 1956, 234). Repertorial studies have not yet confirmed this.
The earliest surviving sources with notation are from the late eleventh century, the first of many manuscripts taken to Zagreb after its diocese had been founded under Hungarian rule in 1094 (Vidakovic 1960). A fully notated antiphoner of the early twelfth century, the so-called 'Codex Albensis' from Szekesfehervar (Stuhlweissenburg, Alba Regia), has been published in facsimile (Codex Albensis), as has an early fourteenth-century noted missal from Esztergom (Missale notatum Strigoniense).
Chant research has been pursued in Hungary in recent years with exceptional energy and authority. In virtually all areas substantial and systematic studies have been made which often surpass those available for other countries. Rado's catalogue (1973) of liturgical books in Hungarian and neighbouring libraries has provided the foundation for further basic work, in particular the catalogue, analysis, and classification by Szendrei (1981) of over 850 complete and fragmentary sources, with over 100 facsimiles. Equally impressive is Szendrei's account of Hungarian chant notations (Szendrei 1983, with copious facsimiles; revised German version 1988, 'Geschichte'). The dominant type was that developed at Esztergom when staff-notation was taken up at the beginning of the twelfth century, resembling south German staff-notation in many respects, but seemingly a consciously evolved new variety.
Offices for Hungarian saints have been edited by Falvy (1968). The corpus of sequences in Hungarian sources has been edited by Rajecky and Rado (1956), and an edition of the complete repertory of office antiphons is in preparation.
Repertorial studies have concentrated on office-books, a number of which have been completely inventoried and compared with numerous extra-Hungarian traditions as well (see Ullmann 1985, Dobszay 1988, Dobszay and Proszeky 1988).
Hungarian experts distinguish a Zagreb liturgical tradition from the dominant Esztergom one. Sources have been studied by Hofler (1967) and Hudovsky (1967, 1971).
The Dalmatian coast area, part of Croatia in the eleventh century and variously under Hungarian, Bosnian, and Serbian control, received chant-books from south Italy and made its own with Beneventan notation (Birkner 1963, Hudovsky 1965, Bujic 1968; see also the Dalmatian sources in Loew-Brown 1980 and the items studied by Gyug 1990).
X
Reformations of Gregorian Chant
X.l. INTRODUCTION
Fellerer, 'Choralreform', MGG.
The Roman-Franciscan chant repertory does not mark a radically new beginning in the history of Western chant, except in so far as it meant the end of Old Roman chant in the Eternal City itself. Whatever musical reforms it may have involved, they embody a concept of Gregorian chant not in any substantial way different from that found in most churches in most lands. The Cistercian order of monks, however, revised its chant repertory according to fixed principles which entailed quite drastic alterations in the melodies they had inherited. Therefore their chant comes first in this chapter on reforms of the chant repertory.
Since no early medieval chant-books have identical musical readings throughout, they must all, to a greater or lesser extent, have been made by, or according to the directions of, cantors with their own individual ideas about how the melodies should be sung. Medieval writings on chant are liberally strewn with references to cantors who 'emended' the melodies, and books that are 'bene emendati'. The more specific the notation, the more obvious are the differences between the books, as for example in the case of the 'rhythmic' and other signs in the early St Gall and Laon manuscripts. The necessity of specifying pitch relationships added another area of variance between sources. Sometimes it is clear that tonal preconceptions entered into the matter, so that chants were variously notated according to different notions of tonality: less straightforward passages involving semitone steps were made more obvious, tonally eccentric passages were recast in the interests of modal unity, and so on. Some manuscripts stand out as relatively heavily 'edited' in these respects, for example the Saint-Yrieix gradual, Paris, Bibliotheque Nationale, lat. 903 (PalMus 13; see Stuart 1979).
The Cistercians went well beyond this degree of editorial interference, and, while previous versions can all to some extent be called little 'reforms' (but 'editions' would be better), theirs was a major recasting of the repertory. Their revision was adopted in large measure by the Dominicans. After these two I discuss the chant of other religious orders, although these do not encompass major musical departures comparable with that of the Cistercians.
2. The Cistercians
609
The next major reform of the chant repertory took place in the aftermath of the Council of Trent and can be regarded as part of the Counter-Reformation. This radical overhaul of the Roman chant repertory was but the beginning of a series of recastings of the repertory in other lands, particularly in France, where so-called 'Neo-Gallican' chant was written as a substitute for the medieval or the newer Roman melodies.
X.2. THE CISTERCIANS
Delalande 1949; Marosszeki 1952; Cocheril 1956, 1962; Gumpel 1959; Canal 1960; Huschen, 'Zisterzienser', MGG; Berry, 'Cistercian Monks', NG; Huglo 1971, Tonaires, 357-67; Waddell 1970, 'Prologue', 1970 'Origin', 1971, 1976, 1980, 1984, 1985; Choisselet and Vernet 1989.
The Cistercian order takes its name from the monastery of Citeaux, near Dijon, founded in 1098 by the Benedictine monk Robert of Molesme and like-minded brethren, who wished to recover a simpler and stricter form of Benedictine monasticism than that generally practised in their time. It rose to Europe-wide fame and influence through the example and activity of St Bernard of Clairvaux (1090-1153). Daughter houses observed the same liturgy as Citeaux. Such uniformity was not entirely new, for although Benedictine houses were bound to no one liturgy, identity of liturgical practice between groups of houses can occasionally be found, as when they stand in a mother-daughter relationship. Thus the Norman monasteries refounded from Saint-Benigne of Dijon all reflect, more or less exactly, the liturgy of Saint-Benigne. The importance attached to this uniformity and the scale on which it was established in the Cistercian order was, however, unprecedented.
The Cistercian musicians took the performance of the liturgy in its 'pure' and 'original' form very seriously. We know little of the earliest musical practice at Citeaux, though the office was much shortened in an effort to recover the state prescribed in the Rule of St Benedict, often interpreted in a very literal way. It appears, however, that the ancient reputation of Metz as guardian of the best chant tradition was known at Citeaux, and monks were sent to Metz to make a copy of the 'authentic' antiphoner there. The thirteenth-century Metz antiphoners Metz, Bibliotheque Municipale 83 and 461 may well represent the tradition which the monks would have encountered (Waddell 1970, 'Origin'). The Cistercian copies are lost. The Messine chant was in any case eventually pronounced unsatisfactory and St Bernard was authorized to oversee a new revision. By about 1190 a complete copy of the Cistercian liturgy had been compiled, to be used as standard for all other manuscripts, now Dijon, Bibliotheque Municipale 114 (facs. Huschen MGG, Choisselet and Vernet 1989); the last five parts of this massive book, including hymnal, gradual, and antiphoner, have unfortunately been missing since the sixteenth century.
610 A'. Reformations of Gregorian Chant
The modal considerations which exercised the Cistercians, like other editors, were mentioned in the previous section. To these were added quite new guide-lines. In notating the chants B-flats were to be avoided, by transposition if necessary. Chants were edited to lie predominantly within a single octave, ten notes being the limit. Long melismas were cut back. The last prescription is a clear reflection of a more austere aesthetic, as is the one that repetition of words should be avoided.
The history leading up to these changes and the main idea behind them are described in a short prologue 'Bernardus humilis abbas Clarevallis' written by Bernard and commonly attached to Cistercian chant-books (PL 182, 1122-3, CSM 24. 21-2). The principles of the musical reform are described in a longer tract, regarded as a preface to the chant-books, beginning 'Cantum quern Cisterciensis ordinis ecclesiae' (PL 182. 1122-32, CSM 24. 23-41). This is thought to have been put together by Guy de Cherlieu, monk of Clairvaux and from 1131 to 1157 abbot of Cherlieu. The theoretical part of the tract, which is supposed to justify the emendations, is in fact taken from an earlier treatise, the 'Regula de arte musica' (CS ii. 150-91) of another Guy—Guy d'Eu (Guido Augensis; not Guy de Chalis, as Coussemaker had it). This Guy was also a monk at Clairvaux, apparently from 1132 at Longpont, and later abbot at an unknown monastery. His treatise, beginning 'Premunitos autem esse volumus', was written for his pupil William, first abbot of Rievaulx in Yorkshire, in 1132. Furthermore, the so-called 'Tonale sancti Bernardi' at the end of early Cistercian antiphoners is also from Guy d'Eu's treatise (GS ii. 265-77, PL 182. 1153-66). (This complicated situation would be considerably simplified if these were the same Guys, but the required proof has not yet turned up.)
Marosszeki carried out a number of tests to establish the parentage of the Cistercian chant tradition, and these have to some extent been refined in Graduel romain IV and by Hesbert in CAO 5. The order of Advent office responsories does indeed agree with the series in the thirteenth-century antiphoner of Saint-Arnoul at Metz, Metz 83. The melodic variants in the gradual point more in the direction of Dijon, situated only 20 km. north of Citeaux (whereas Metz is over 250 km. further), but no gradual from Metz survives for comparison.
In view of what has been said, it may seem otiose to compare Cistercian books with traditional ones. But the alterations mentioned in the famous preface are not encountered very often. One has only to look at the pages from Cistercian graduals reproduced in PalMus 2-3 and Marosszeki to see how resolutely they keep to the usual path, B-flats, melismas, and all. Marosszeki and Delalande (1949) certainly provide many examples of alterations, but they do not, by and large, leap to the eye in medieval manuscripts.
The notation in Dijon 114 is a variety of French notation on lines with a number of Laon (Messine) features, the type which eventually developed into what is usually called 'Gothic' notation, and in various manifestations this was the notation used in very many subsequent Cistercian books. It has been suggested that Cistercian books were the prime agent for the propagation of this notation in Germany and lands east, in other words of the change from staffless to staff-notation and from German to
J. The Dominicans
611
Gothic signs (Hourlier 1951, 'Messine', 153-7). Szendrei (1985) has pointed to the lack of exactitude in use of the terms 'Messine' and 'Gothic' and shown that the Cistercians' notation was only one of several varieties involved in the new developments in south Germany and Austria.
On four occasions the Benedictine Rule calls a hymn an 'Ambrosianum'. The Cistercians therefore tried to find the hymns of St Ambrose for their revised office (or rather, office restored to its 'authentic' form and content), travelling to Milan for the purpose (at a time when the Milanese were probably not yet using notated hymnals). A hymnal with thirty-four texts, to nineteen melodies, was compiled, many of them heretofore quite unknown outside Milan. This was used among other places at the monastery of the Paraclete founded by Abelard for Heloi'se, and two interesting letters by Abelard survive, commenting on the general strangeness and literary peculiarities of the hymnal (Waddell 1976), one of them Abelard's only letter to St Bernard. The hymnal was indeed eventually found generally unsatisfactory, and a new one prepared with sixty texts to thirty-seven melodies (seven of the previous melodies were dropped). The non-'Ambrosian' hymns were assigned to hours other than Vespers, the Night Office, and Lauds, when the Rule of St Benedict does not use the specific term 'Ambrosianum'. Both the first and second hymnal have been edited by Waddell (1984) with definitive commentary.
X.3. THE DOMINICANS
Bonniwell 1944; Delalande 1949; Huschen, 'Dominikaner', MGG; Gleeson 1972; Berry, 'Dominican Friars', NG.
The Dominican order of friars (Ordo Praedicatorum, Order of Friars Preachers), founded by St Dominic, was formally constituted in 1220. Its organization was strongly centralized, like that of the Cistercians, and its chant practice also followed the Cistercian in many respects. The office is, however, celebrated according to the Roman or secular cursus, not the monastic.
The first moves to formulate a standard liturgy appear to have been taken by the second Master-General, Jordan of Saxony (1221-37). A new revision was set in motion in 1244 by John the Teuton (1241-52). At the General Chapter of 1245 a commission of four brothers was appointed, one from each of the four most important provinces of the order, France, England, Lombardy, and Germany, who were to pool their knowledge and materials. (Hesbert 1980, 'Sarum' shows how the Sarum office may have played a part in the final recension.) Working sessions in Angers produced no generally acceptable result, and in 1250 the brothers were commissioned to visit Metz, as the Cistercians had done. After further disagreement the matter was placed in the hands of the newly elected Master-General Humbert of Romans (1254-63), previously Provincial of France and probably already influential in the work of revision. His proposals were approved at the General Chapters of 1255 and 1256. His
612 X. Reformations of Gregorian Chant
master copy of the entire liturgy and its music still exists, the 'Correctorium Humberti', or 'Correctorium Sancti Jacobi' (after the Dominican headquarters at Saint-Jacques in Paris where it was made), or 'Le Gros Livre' (now Rome, Santa Sabina XIV. lit. 1). A portable copy for the Master-General's use also survives (London, British Library Add. 23935), and another complete copy of the fourteenth century in Salamanca. A set of rules for the copying of chant-books, based on the Franciscan one, was composed (Van Dijk 1963, Sources, 118, Huglo 1967, 'Reglement').
The Cistercian and Dominican series of alleluias for the Sundays after Pentecost are identical, and, as Delalande has shown, the differences during the rest of year are very few. Many of the bowdlerizations of the chant repertory carried out by the Cistercians were perpetuated in the chant of the Dominicans, though in many instances there was a reversion to the norm. Delalande (1949) has provided numerous examples of complete or partial agreement of Cistercian and Dominican books against the general tradition. Occasionally the Dominicans operated where the Cistercians had declined to do so.
Ex. X.3.1 (adapted from Delalande 1949, Tab. IXbis) shows part of the gradual Exsurge domine fer opem nobis V. Deus auribus nostris (discussed also by Bomm
Ex. X.3.1. From gradual Exurge Domine in Gregorian and Cistercian-Dominican versions (Piacenza, Bibl. Cap. 65; St Petersburg, Saltykov-Shchedrin Public Lib.; Delalande 1949)
„ Piacenza
E-xur - ge do- -mi-ne Leningrad
^ b      f'm   *    t< ff »j   0*0   *****    9    *~t i* *» 0~j-
Cistercian-Dominican_
P
Piacenza
* 'Is
p
... in  di - e- -bus       an-        -ti- -quis.
Leningrad *
Cistercian-Domini can
et in di - e- -bus     an-        -ti- -quis.
4. Chant in other Religious Orders 613
1929, 116-20). This piece is tonally problematic, and several different ways of notating it are to be found. The Cistercians, followed exactly by the Dominicans, devised a unique way which entails a quite drastic revision.
The differences between notated versions concern the relationship between the main reciting note—bb or c, even d near the end of the verse—and the final, either E or D. The situation is complicated by the fact that the reciting note appears to change during the piece and the verse may begin at a level higher than the respond did. All possible combinations of these are to be found:
Respond   reciting note: c or bb ... c
cadence: E or D
Verse       reciting note: c or c . . . d or b b . . . c
cadence: E or D
To this knot the Cistercians took a sharp sword indeed, staying firmly with c-recitation throughout and then, presumably feeling that the usual descent to the cadence was not adequately prepared, eliminating it altogether by excising a short phrase and ending on a instead of E or D. Ex. X.3.1 gives the beginning of the gradual and the end of the verse in two of the more traditional versions, followed by the Cistercian/Dominican one.
X.4. CHANT IN OTHER RELIGIOUS ORDERS
(i) The Carthusians
(ii) The Premonstratensians
Unlike the Cistercians, other monastic orders did not make a radical revision of the chant repertory but adapted what lay to hand in a fairly straightforward way, as monasteries had traditionally done. While each order was concerned for correctness and uniformity of practice throughout the order, their traditions do not contain many surprises.
(i) The Carthusians
Huschen, 'Kartäuser', MGG; Lambres 1970, 1973; Becker 1971, 1975; Huglo 1977, 'Livres'.
The Carthusian order was founded by St Bruno of Cologne at the Grande Chartreuse (Charterhouse) in the Alps a little north of Grenoble in 1084. It was conceived as a contemplative order with a strongly eremitical character. A maximum of twelve monks for each monastery was fixed, and rarely relaxed. The monks lived alone in their cells, coming together only for worship and for meals on feast-days. The Little Hours were said privately. The order had no codified rule until the fifth prior of the Grande Chartreuse, Guigo (Guigues de Chätel), drew up a customary, completed in
614 A". Reformations of Gregorian Chant
1127 (PL 153, 635-760). This included regulations not just for the monastic discipline and way of life but also for the liturgy and its chant, with a tonary (ed. Becker 1975). While the office until then had followed the Roman or secular cursus, Guigo's liturgy had the monastic cursus. Some simplifications of liturgical and musical practice seem to have been introduced: non-biblical texts were avoided, as in the liturgy of Lyon (for that reason hymns were sparingly used); melismas in office responsories were avoided; all those parts of the repertory where expansion can often be found in medieval sources—for example, alleluias, ordinary-of-mass melodies— were kept small. The order had no interest in musical splendour.
Huglo has shown that the post-Pentecost alleluias of Carthusian books correspond with those of Clermont, the Advent responsories with those of Clermont and Aurillac. The melodic variants in proper-of-mass chants agree most often with those of Valence, one of the Aquitanian group of sources. Early Carthusian books have Aquitanian notation, but a very straightforward type of square notation was adopted at the end of the twelfth century which lacks an upward stroke at the start of the clivis (see Plate 14).
(ii) The Premonstratensians
Hiischen, 'Augustiner', MGG, 'Pramonstratenser', MGG; Lefevre 1933, etc.; Weyns 1967, 1973.
The rule of St Augustine was a document probably drawn up by one of St Augustine's followers and partly on the basis of his writings, which specified a number of particular observances and offered more general comment about the nature of a common religious life. It was little used in the early Middle Ages, but was taken up by the communities of clerics who increasingly from the mid-eleventh century and the period of the 'Gregorian reform'—so called after the reforming pope Gregory VII, 1073-85—formed religious groups devoted to a life of poverty and celibacy in imitation of the first Christians. They became known also as 'Canons Regular' because they followed a Rule (Regula). They used the liturgy of whatever diocese they inhabited, except in the case of congregations grouped under the leadership of a particularly influential house. Such were the Augustinian canons of Saint-Victor, Paris, whose version of the Paris liturgy was used by other Victorine Augustinian canons (for example at Bristol in England). The most influential were the Premonstratensians, founded by St Norbert at Premontre, a little west of Laon, in 1120. They were particularly active in east Germany and Hungary.
The liturgy of the Premonstratensians has been described in various works by Lefevre, who has edited several ordinals. The text of the mass has been studied by Weyns (1967, 1973). It is not clear to what extent the chant was unified among Premonstratensian houses. The monks of Solemes tested the melodic variants of one source from Premontre itself, and other Premonstratensian manuscripts from the Low Countries, Germany, and Austria. The gradual from Premontre (Soissons,
5. The 'Medicean' Gradual
615
Bibliotheque Municipale 85) and two of the German sources were closely similar, and similar to books from Liege and Andenne (not Laon). Others were closer to other north French uses, and one was grouped within the eastern European 'bloc'. It is possible that some houses entered the Premonstratensian congregation after their chant tradition was established.
The various congregations of Augustinián Canons Regular are quite distinct from those of the Augustinián hermits or friars. This order was formed by Pope Alexander IV in 1256 from various eremitical communities. They became particularly numerous in German lands. They had no centralized chant tradition.
Little attention has so far been paid to the chant of later religious orders. The Carmelite Order, founded in Palestine in the mid-twelfth century and transferred to Italy a century later, observed the liturgy of an ordinal by Sibert de Beka from 1312, but the melodies were not prescribed. Office sources have been surveyed by Boyce (1990), who has also edited two special offices (1988, 1989). A number of sources with music have survived from the Brigittine Order, founded by St Bridget of Sweden about 1346 at Vadstena. This was originally a double order of both monks and nuns; while the brothers followed the liturgical use of the local diocese, the sisters celebrated an individual Marian office liturgy, referred to as the 'cantus sororum', attributed to Petrus Olavi of Skanninge. Its text has been edited by Collins (1969) and Lundén (1976); Servatius (1990) has made a musical edition of the antiphons. Of the chant of the German reforms of Bursfeld, Kastl, and Melk, relatively little is known, except through the work of Angerer (1974, 1979) on Melk. The Melk reform, in essence yet another return to a 'purer' Benedictine monasticism, was promoted by Albert V of Austria, who founded the famous monastery overlooking the Danube in 1412.
X.5. THE'MEDICEAN' GRADUAL (1614-15)
Molitor 1901-2; Weinmann 1919.
The threat to the Roman church from Protestantism led to repeated calls for a universal council, which eventually took place in Trent in northern Italy intermittently between 1545 and 1563. One of the minor matters discussed was the music of the church, decrees affecting which were issued at the twenty-second to twenth-fourth sessions. As far as plainchant was concerned, these were couched in general terms. In all music, intelligibility of the text was important. An important provision was made that provincial synods should establish musical practice with respect to local traditions and circumstances.
It was in the Roman chant-books compiled in the aftermath of the Council that concrete shape was given to the broad instructions of the Tridentine decrees. A reformed breviary had been published in 1568, a reformed missal in 1570. In 1577 Giovanni Pierluigi da Palestrina and Annibale Zoilo were commissioned to prepare
616 A'. Reformations of Gregorian Chant
the music for them, but did not complete the task. Not until 1614 and 1615, in fact, did the two volumes of a new gradual appear, its editors being Felice Anerio and Francesco Soriano, remnants of a six-man committee which had begun work in 1608. The work was published by the Medici Press in Rome, hence its usual appellation of the 'Medicaea' or 'Medicean edition'.
An antiphoner along the same lines was never planned. But already between 1582 and 1588 Giovanni Guidetti, the papal chaplain, singer in the papal chapel and pupil of Palestrina, had published a series of handbooks in small format with music not in the gradual, the most important of which was the Directorium chori of 1582, containing recitation tones for prayers, lessons, and psalms, with hymns, versicles, and responses, and short responsories. An important innovation in Guidetti's Directorium was the use of proportional note values:
12       3 4 ♦     ■    ■ ■
Since the third value only occurred when followed by the first, any piece was exactly measured in units of the second value.
It is important to remember that the use of these books was never officially binding upon the church. It was the example they set that was most telling. The way towards further rewriting of the ancient melodies was open for anyone with the will and ability to follow it.
The drastic nature of the Anerio-Soriano revision cannot be understated. It was a complete recasting of the repertory according to the aesthetic principles of the humanist age, carried out by composers whose musical sensibilities were formed by the polyphonic church music of Palestrina and his contemporaries (see Meier 1969). (Exactly what Palestrina may himself have contributed to the revision is unclear. He was occupied with it in 1578, and his work and ideas would have been known to Soriano, who was his and Zoilo's pupil, and to Anerio, who was Soriano's pupil and Palestrina's successor as composer to the papal chapel.) The corner-stones of the revision were modernization of the tonality and rearrangement of the notes to reflect correct declamatory principles, so that unaccented syllables should not have more notes than accented ones. Every type of chant was affected, the more ornate ones most of all.
Although the Medicaea has not been published in facsimile, it was accurately reproduced in the Graduale Romanům published by the firm of Pustet in Regensburg (Ratisbon) in 1871, edited by Franz Xaver Haberl. For thirty years this edition enjoyed papal authority as the approved form of chants for mass, and its contents are widely available for study. Ex. X.5.1 shows the type of revision carried out in the Medicaea. Omnes de Saba is the mode 5 gradual for Epiphany. In its Gregorian version, when the melody hovers around c, b \ is sung. In the verse the final F exerts a stronger tonal attraction and 6bs are frequent. The Medicaea avoids b\ altogether. The fluttering repetitions of c and bb are eliminated; notes are not repeated at all
5. The 'Medicean' Gradual
617
Ex. X.5.1. Medieval and Medicean versions of gradual Oinnes de Saba (Piacenza, Bibl. Cap. 65; Graduale, Rome, 1898)
P
Piacenza
>6   »t    i-f> -r
P
Om-Pustet
-nes de  Sab - ba
ue-ni - ent     au - rum  et tus de-
jr*jr
*     Cm •
P
Om-Piacenza
-nes   de Sa-
-ba      ve - ni - ent,      au - rum   et thus de-
-
-rr
f f
-fe - ren - tes Pustet
et  laudem do-mino    an - nun - ti - an - tes.
r* * *—r
m   m m-
-fe - ren-Piacenza
-tes,    et laudem D6mi-no   an-nun-   -ti - än - tes.
1      rm «I«' mmrrm -#    0       wm      a m
*       J      ■     <     * "I'**
Sur -ge Pustet
et il - lu-mi -na-
-re
•   ■   *—y
V. Sur - ge, Piacenza
et il - lu-mi - nä-
-re
'i ft_
'* m *m
m   m   'rn^'m   m m
Hie-ru - sa - lern Pustet
qui -a   glo-   -ri - a do-
^5,
5=
P
Je - rü-Piacenza
-sa-lem:
qui-a   glö - ri - a
Do
mi-ni
-9+mm-
=#*^
su - per te Pustet
or - ta est.
bm m
3^
su - per te
or   -  ta est.
618 A'. Reformations of Gregorian Chant
without a change of syllable (see the final cadences of both parts). Each melisma on the final syllable of a phrase is cut, and the balance between accented and unaccented syllables is altered: for example, deferentes is changed from 1-2-2-5 notes to 1 — 1—6-1, Saba venient from 3-13-1-3-1 to 5-2-2-1-2. The accented syllables are marked in the Pustet edition, though not in the Medicaea.
As a piece of early seventeenth-century plainchant the new version is admirably moderate, mellifluous, and consonant. But none of the ecstatic, other-worldly quality of the medieval version remains.
X.6. NEO-GALLICAN CHANT
Baumer 1905; Henri Leclereq, 'Liturgies neo-gallicanes', DACL; [Hiley], 'Neo-Gallican Chant', NG; Fontaine 1980; Brovelli 1982.
New Roman liturgical books with texts revised on humanist lines were briefly fashionable, of which the Breviary of the Holy Cross by Cardinal Quihonez (1535) is an example. But the Council of Trent (1545-63) did not perpetuate such initiatives. The new Roman breviary and missal of Pius V (1568 and 1570 respectively) were by no means drastic revisions of the traditional texts, and would have been found quite normal by the average late-medieval cantor. Those cantors obliged to provide music for services where they were used would have had no difficulty in continuing to employ the traditional chant-books, perhaps one of the many printed editions which appeared in the sixteenth century. They would in any case have had to do precisely this for the office. Popular Italian books included those printed by Junta in Venice: the Graduate Romanum of 1606 and 1611 and the Antiphonarium of 1603. Another is the Antiphonarium published in Antwerp in 1611 for the diocese of Mechelen (Malines), which was used by Haberl for the Pustet antiphoner of 1878. (See Tack 1960 for numerous facsimiles.)
But the seeds of literary and musical revolt against the medieval heritage had been sown, and the next two centuries were to see large numbers of quite new books where texts and melodies were revised or completely replaced. To the extent that they mark a complete break with the medieval chant which is the main subject of this book, they could be omitted from further consideration here. And research on what is in fact a vast heritage of such service-books has hardly begun. A brief statement of the general situation must therefore suffice.
The making of new chant-books was particularly energetic in France, where the church's relationship with Rome was a matter of political and doctrinal controversy. The decisions of the Council of Trent regarding the constitutional structure of the Roman church were not accepted in France, where a strong school of thought supporting freedom from papal authority had existed ever since the thirteenth century. These beliefs were enshrined in the Four Gallican Articles of 1682, drawn up by Bossuet, and although they were formally withdrawn a decade later, their essence
6. Neo-Gallican Chant
619
was propagated throughout the next century. The tide of opinion did not begin to turn back in an ultramontanist direction until after the restoration of the French monarchy in 1814. The later seventeenth and eighteenth centuries therefore mark the period of composition of what is commonly known as 'Neo-Gallican' chant.
Of the many liturgical books produced in many French dioceses during this period, those for the office are more radical revisions than those for mass. Particularly numerous were the new hymns, by such authors as Santeuil and Coffin (see Pocknee 1954). The most important new liturgies were those of Paris, where Archbishop Frangois de Harlay issued a new breviary in 1680 and Archbishop Charles de Vintimille both a new breviary and a new missal, in 1636 and 1638 respectively. The breviary of Cluny of 1686 was also influential. It has been estimated that ninety out of 139 French dioceses had non-Roman books by the end of the eighteenth century (Fontaine 1980), the Paris books being adopted in over fifty of them.
In the 'Neo-Gallican' chant-books both revision of older chants and composition of quite new ones is to be found. The revision is generally more radical than that of the Medicaea, particularly from the tonal point of view. Accidentals were introduced and whole sections shifted to make more obvious tonal sense (from the Baroque standpoint). Ex. X.6.1 shows two chants for mass at Epiphany from the Graduel de Paris of 1754.
Ex. X.6.1. Extracts from Graduel de Paris, 1754 (pp. 110, 112)
In the sequence Ad Jesum accurite the note values are usually pairs 1 ♦ , sometimes »j ■ . I have assumed that triple time was intended throughout.
In the offertory Reges Tharsis I have transcribed note values as follows:   ^   = •
■ = • ♦ = •
PROSE. Du 1. (Ton)
* a
Ad Jesum ac - cu-ri-te, Stel - la fo - ris prae-di-cat,
\* * 6   J  1 *
Cordaves-tra sub-di-te In-tusfi-des in-di-cat
Re-gi no - vo Genti-um. Redemptorem om-ni-um.
1 1			1				—j-	—i—	■ '    ■    — -' ■
t>   }    I'1 i	0    1 0	—-1			-C	-i-b-.—	—tt-■---		Ljk-U
Hue af-fer-re mu-ne-ra Haec e-rit    gra - tis-si - ma
Volun-ta - te li-be-ra, Sal-va-to - ri vic-ti-ma,
Sed mune-ra Mentis sa-cri
cor -di-um. fi - ci -um.
m
mm
-6—_—1 v ~J7j.
Et thus de-si - de-ri-um. Thu-re De-us Gen-tium.
-*—G
et myrrham aus-te-ri-tas, Ho-mo myrrha, co - li-tur
Of - f ert au - rum ca-ri-tas, Au - ro Rex ag - nos - ci-tur,
JbJ   J I J J J £
--
Ju-dae-a, gau - den - ti-bus Post cus-to-des    o - vi-um,
Non in-vi - de Gen-ti-bus Se Ma-gi fi - de - li-um
Re-tectum mys-te-ri-um. Jungunt in con -sor-ti-um.
620
(Ex.X.6.1 cont.)
X. Reformations of Gregorian Chant
v7 * s
In unum tu - gu - ri-um. Nascentis ex - or - di-um.
Qui Ju-dae - os ad-vocat Christus, Gentes convo-cat Beth-le-em fit     ho-di-e        To-ti-us Ec-cle-si-ae
m
Profe - rat im - pe - ri-um.
Regnet Christus     cor-dibus,       Et victis   re - bel-li-bus
9  _J s y
-men.
OFFERTOIRE. Du 8. (Ton)
0m     » 0
Re-ges Tharsis
et  in-    -su-lae   mu-   -ne-ra     of-      -ferent:   Reges A - ra-bum
P
et   Sa - ba
do   -  na ad-du  - cent;
et a - do - ra - bunt e - urn om
nes
*4* 9*0*
**T»  , 00* i**099m0*00m
Re-ges    ter - rae:       omnes   gen-       -tes      ser - vi - ent
Apart from this type of chant, two others were developed which have as much in common with Baroque song as with plainchant. These were (i) chant figure or plainchant, plainchant musical, and (ii) chant sur le livre or fleuretis. In the first (Pineau 1955, Fuller, 'Plainchant musical', NG), a measured and ornamented type of chant was sung by a soloist, choir sections being accompanied by organ harmonies or by a serpent. It is known from treatises by Nivers, Poisson, Lebeuf, and La Feillee, and from the Cinq messes en plain-chant (1669) by Henri Du Mont, of which the Messe royale was especially popular. In the second (Prim 1961) a soloist improvised florid counterpoint over a chant performed by a choir in rhythm and reinforced again by the serpent.
Neo-Gallican chant in France was not the only repertory of non-Roman and non-medieval chant. Numerous other books of the eighteenth and early nineteenth centuries, especially those published in Germany, are likewise witness to attempts to reform plainchant in accordance with contemporary taste. Those produced by Reiner Kirchrat in Bonn (Theatrum musicae choralis, Cologne 1782) and Caspar Ett in Munich (Cantica sacra, 1827, with organ accompaniment) are examples. (Tack 1960
6. Neo-Gallican Chant
621
gives several facsimiles from German publications.) The French plainchant musical is clearly derived from Italian canto fratto (as opposed to canto fermo puro), exemplified in Viadana's '24 Credo a canto fermo' of 1619. The accompaniment of chant, increasingly common in the seventeenth and eighteenth centuries, was itself an inducement to recompose chants in a style easier to harmonize and more in the style of contemporary solo and concertante motets. (The history of plainchant accompaniment is a subject unfortunately too large for discussion here: see Sohner 1931 and Wagener 1964.)
Contemporary works which offer discussion of or instruction in plainchant are essential for an understanding of the attitudes current in those times. Those written in the seventeenth and eighteenth centuries have been listed and discussed by Fellerer, both briefly (1972, 'Choralpflege') and more comprehensively (1985).
XI
The Restoration of Medieval Chant
XI. 1. THE RETURN TO THE SOURCE
Fellerer, 'Choralreform', MGG, 1974-5, 1985.
After the period of perpetual revision and new composition it was perhaps inevitable that the idea should have gained ascendancy of returning to the plainchant as it had 'originally' been. Several factors determined the course that events would take.
First, liberalism in the church as a whole was discredited, not least by the French Revolution, which, it was felt, had not been hindered by the Jansenist and Gallican movements in the French church. There was a similar reaction to the ecclesiastical reforms of Holy Roman Emperor Joseph 11 (1765-90), the general tenor of which had been to increase religious tolerance and limitation of papal authority to spiritual matters. The reaction implied a return to Rome in liturgical as in other matters. The call to action was sounded as early as 1811 by Alexandre-Etienne Choron in his Considerations surla necessite de retablirle chant de VEglise de Rome dans toutes les eglises de I'Empire francais. (Choron had already distinguished himself as an editor of Josquin, Goudimel, Palestrina, and Baroque masters.)
Secondly, the perennial tendency of the church to renew itself by returning to an earlier and supposedly 'purer' state coincided with the wider romantic tendency to idealize the past. When the Age of Reason challenged the old certainties, disaster had ensued. Churchmen yearned for the Age of Faith. To the 'Gothic revival' in church architecture, a comparable revival in church music would be joined. Its musical ideals—held by a remarkably wide body of opinion in Italy, France, and Germany— were embodied in traditional plainchant and the polyphonic music of Palestrina (see Fellerer, 'Caecilianismus', MGG).
Thirdly, the techniques of manuscript study which were essential for revealing the secrets of medieval sources had already been developed, particularly by the Benedictine monks of the Congregation of Saint-Maur during the late seventeenth and eighteenth centuries. It would take until the end of the nineteenth century for similar expertise to accumulate in the handling of the first notations, but the whereabouts of many important sources was already known from the Maurists' work, for example, through Mabillon's study of the Ordines Romani, published as early as 1689.
7. The Return to the Source
623
The particular part played by the Benedictine abbey of Solesmes near Le Mans after its restoration by Dom Prosper Gueranger in the 1830s will be discussed in the next section. Solesmes was of course not alone in studying ancient sources as a means of renewing the quality of the church's music.
Some of those who favoured a return to the Roman liturgy were not interested in the medieval melodies, and wished to reintroduce the Medicean and other editions of that period. The Medicaea plus the Antiphonale of Peter Liechtenstein, Venice, 1580, formed the basis of the several books produced in the 1840s and 1850s by Duval, de Vogt, and Bogaerts under the auspices of Cardinal Sterckx in Mechelen (Malines). As long as Rome itself remained inactive, this was also the position of the Cecilian movement in Germany, founded in 1868 in Regensburg by Franz Xaver Witt (see Lickleder 1988), and the justification for the editorial line taken by Haberl in the publications by the firm of Pustet in the same city.
Some of the landmarks in recovering the medieval melodies are as follows. In 1846 J.-L.-F. Danjou discovered the eleventh-century manuscript Montpellier, Faculte de Medecine H. 159, with alphabetic letters and neumes in parallel. In 1851 Louis Lambillotte published a hand-drawn facsimile of St Gall, Stiftsbibliothek 359, unfortunately not accurately done, believing that this was the manuscript which, according to St Gall legend, was sent from Rome itself by Pope Hadrian I (IX.3). Lambillotte's Graduate Romanum and an Antiphonarium were published in 1855 in Paris after his death by Dufour, but did not follow St Gall 359—not surprisingly, since the necessary work of comparing it with other sources to decide on pitches to interpret its neumes would only be accomplished later.
Several editions with partially restored melodies were prominent in the supersession of Neo-Gallican books in France, of which the gradual promoted by the bishops of Reims and Cambrai (1851) is the most significant. It was reputedly based, at least in part, on the Montpellier manuscript and other medieval codices, but the restoration was still half-hearted. The other books took a good deal of account of the Reims— Cambrai edition, but were still more timid, especially in melismatic chants.
Discussion of the merits and demerits of these and other French editions forms a lively chapter in that country's scholarly music criticism, and very numerous were the memoirs published, the committees and societies formed (see Fellerer 1974-5, esp. 143-5 on the Reims-Cambrai commission). In Germany a pioneering role was played by Michael Hermesdorff (1833-85). In 1869 he founded a branch of the Allgemeiner Cacilienverein for the Trier diocese and from 1872 to 1878 edited the journal Cdcilia (founded in 1862, later the Kirchenmusikalisches Jahrbuch), which from time to time included facsimiles of medieval chant sources. In 1872 he founded a Verein zur Erforschung alter Choral-Handschriften (Society for the Investigation of Old Chant Manuscripts), and he was Peter Wagner's choirmaster and teacher. He published several chant editions, including a Graduate . . . Trevirensis (1863), an Antiphonale (1864), Kyriale (1869), and a second Graduate ad normam cantus S. Gregorii (Trier, 1876—82). The latter was a remarkable publication, which not only restored practically the entire medieval shape of the melodies, but also included small
624 AY. The Restoration of Medieval Chant
printed neumes, imitating the shapes of later medieval Trier manuscripts, over the music on the staff.
As a means of gauging the relative degree of restoration effected in some of the above-mentioned editions, Ex. XI. 1.1 gives the opening of the gradual Exurge domine non prevaleat homo. For ease of comparison I have transcribed all the versions on to the modern stave, but reproduced the note-shapes of all the nineteenth-century editions.
The editions of the Graduate (1871 and 1873) and Antiphonale (1878) published in Regensburg by the firm of Pustet were edited by Franz Xaver Haberl. The gradual was a faithful revival of the Medicaea, the antiphoner was based on the printed editions of Venice 1585 and Antwerp 1611. Chants for new feasts or anything else required by liturgical changes since the seventeenth century were provided by Haberl and Witt. In 1871 Pius IX declared the Regensburg editions to be the only versions officially recognized by the Roman Church, an astonishing monopoly enjoyed by Pustet for thirty years, and the first of its kind, for the Medicean gradual had never had this status.
XI.2. SOLESMES AND THE VATICAN EDITION
Rousseau 1945; Combe 1969; Bescond 1972.
The Benedictine monastery of Saint-Pierre de Solesmes dates back to the early eleventh century, but after the ruin of the French Revolution it had to be founded anew. Dom Prosper Gueranger (1805-75) bought the property in 1832, and settled there with five priests a year later. In 1837 it was constituted an abbey, with Gueranger as its first abbot, and head of the French Benedictine Congregation, by Gregory XVI. (From 1880 to 1895 and again from 1901 to 1922 the community lived in exile, for some years in the latter period at Quarr Abbey on the Isle of Wight.) Gueranger was passionately committed to the restoration of Roman use in France and of authentic melodies in the chant, ideas set out at length in his Institutions liturgiques (1840-51).
It was apparently in 1856 that the first work with a medieval chant-book was carried out, when Dom Paul Jausions transcribed the 'Rollington Processional', from Wilton Abbey (now lost: see Combe 1969, 31-2, Benoit-Castelli 1961). In 1860 Dom Joseph Pothier (1835-1923) joined Jausions in the work of transcription. Two years later the transcription on to lines of the first manuscripts in staffless neumes were tackled, Angers, Bibliotheque Municipale 91 by Jausions, and St Gall Stiftsbibliothek 359 by Pothier. In 1864 a Directorium choti was compiled. Already by 1867 Pothier felt able to write the first draft of a 'method', of the type which abounded at the time. And by 1868 a new gradual had also been prepared. These eventually became the Melodies gregoriennes of 1880 and the Liber gradualis of 1883. Meanwhile a lithographed Processionale had been published in 1873.
Ex. X1.1.1. Opening of gradual Exurge Domine non prevaleat homo: Montpellier, Faculté de Mcdecine H. 159, p. 160, Graduale (Rome, 1898), p. 79 (Pustet edition), Graduel romain (Paris, 1874), p. 205 (Reims-Cambrai edition), Graduel romain (Marseilles, 1872), p. 92 (Digne Commission edition), Graduale . . . Trevirensis (Trier, 1863), p. 124 (Hermesdorff 's first edition), Graduale ad normám cantus S. Gregorii (Trier, 1876) (Hermesdorff's second edition)
Montpellier
Ex-
-ur- -ge
do - mi - ne
nonpre-ua-     -le- -at
ho-
-mo
Pustet
Ex-súr - ge Dó - mi-ne,    non praevá - le-at ho-mo
p
Reims-Cambrai
A ■'■'1 I ''1 ^
E - xur - ge, Do
mi-ne.
non praeva-
-le-at
ho-Digne
-mo
■■1   > ■^■M^r
'   1 Hr^i
Ex-ur- -ge Do- -mi-ne, Hermesdorff I
non prae-va-   -le-at ho-
-mo
E - xurge,       Do- -mi-ne,
non pre-va-   -le- -at
—■ • ■
ho-
-mo
Hermesdorff 11
E-
-xur - ge,
Do- -mi-ne,
non pre-va-
-le- -at
ho- -mo
626 AY. The Restoration of Medieval Chant
Pothier was but one star in a brilliant constellation of scholars of liturgy and church history at Solesmes: Pitra (later cardinal, whose aid was later to be invaluable in gaining permission for manuscripts to be brought to Solesmes and photographed), Cagin, Cabrol, and later Quentin and Wilmart, to mention only a few. In 1875 André Mocquereau (1849-1930) joined the community. It was he who conceived the idea of the series of facsimiles Paléographie musicale, whose indisputable evidence should support Pothier's publications. The first volume (St Gall 339) appeared in 1889. Particularly significant was the presentation in volumes 2 and 3 of nearly 200 medieval manuscript versions of the same piece, the gradual Iustus ut palma. At the international congress on Gregorian chant in New York in 1920 Mocquereau described the Justus ut palma volumes as a 'war machine' ('engin de guerre') in the struggle against the Pustet privilege, 'une sorte de "tank" scientifique, puissant, invulnerable, capable ďenfoncer tous les raisonnements ennemis'. (Mocquereau 1920-1, 9, cited by Combe 1969, 126.)
In the aftermath of the Franco-Prussian War (and the above was pronounced after the First World War) it was perhaps not only for musicological reasons that French scholars disputed the privilege enjoyed by the Pustet edition. The argument was often acrimonious. The considered views of Haberl, editor of the Regensburg books, laid stress on the papal approbation the Medicaea enjoyed, the necessity of obedience to Rome, the musical value of an edition prepared by such great musicians as Palestrina, Anerio, and Soriano, and the practical merits of the melodies as against the much more difficult versions restored by Solesmes (Haberl 1902).
An international congress held at Arezzo in 1882 vindicated the work of Solesmes, only to have its resolutions rejected by Rome. In 1884 Pitra presented the Liber gradualis to Leo XIII, who praised its scientific worth but made it clear that it would not be adopted by the Vatican as normative. An important point was gained when the Jesuit Angelo de Santi, editor of the journal Civiltá cattolica in Rome, was won round by Mocquereau, and the students of the French seminary in Rome also adopted the Solesmes melodies. Other Roman choirs, including the Sistine choir itself under Antonio Rella, gradually followed suit. The decree of 1883 confirming support for the Pustet edition was withdrawn by Leo XIII in 1899, and when Pustet's monopoly expired in 1901 it was not renewed. Leo XIII died in 1903 and his successor Pius X acted quickly to sanction the restored chant. The famous motu propno 'Tra le sollecitudini', dated St Cecilia's Day (22 November) 1903, appears to have been drafted by De Santi. A triumphant congress held in Rome in 1904 to mark the thirteenth centenary of St Gregory gave rapturous acclaim to the restorers and the choirs singing the medieval melodies. (Many of the speeches and performances were actually recorded by the Gramophone Company: see Berry 1979.)
Meanwhile a string of important publications had appeared from Solesmes, mostly Pothier's work. A selection of sequences, votive antiphons, and other pieces, the Variae preces, was published in 1888. The first edition of the Liber antiphonarius was published in 1891. Pothier's old Processionale was issued in 1893 as a Processionale monasticurn, and in 1895 there appeared a Liber responsonalis. (These two contain
2. Solesmes and die Vatican edition
627
almost the only editions of the great responsories of the Night Office to have been issued, for the variousAntiphonale^ which have appeared are for the day hours only.) A revised Liber Gradualis was issued in 1895, and the first edition of the compendium Liber usualis.
There remained the task of preparing official Vatican editions of the Graduate and Antiphonale. In 1904 a commission under Pothier was appointed by the Vatican to supervise the edition, which was to be prepared by the monks of Solesmes. Pothier was no longer resident at Solesmes, having been appointed prior of Liguge in 1893 and abbot of Saint-Wandrille in 1898. Perhaps inevitably in a project as complex as this, differences between members of the overseeing commission and the executive arose. For example, Pothier favoured c rather than b as the reciting note in mode 3 and mode 8 chants and recitation formulas. He and others (Wagner, Gastoue) were readier than Mocquereau to admit the validity of later medieval sources, representatives of 'the living tradition'. Bescond (1972, pi. XXIX) gives a facsimile of a page from the preparatory drafts of the Kyriale. Here we see the Solesmes proposal for the Kyrie Lux et origo and Gloria T, with the commission's amendments to Solesmes proposals. The beginnings run thus:
Solesmes:        Gab a ah cbbG
Ky- ri-e e-Commission:   Gac a accbG
Ky- ri-e
Solesmes:        b    aGGGabb* abaGG
* = liquescent
The amendments were chiefly those of Pothier, Wagner, Gastoué, and Grospellier. It is clear that no sensible amendments could be made without recourse to manuscript sources, available to very few of the commission. The Solesmes draft was naturally presented without critical apparatus.
Disagreement reached such a pitch that after 1905 Solesmes did not take further part in the editorial work. The Graduate Romanům of 1908 was in effect a new edition of Pothier's Liber gradualis, re-edited in the light of the Solesmes Liber usualis of 1903. The Antiphonale Romanům followed in 1912, based on Pothier's Liber antiphonarius.
In the same year as the publication of the Vatican Kyriale (1905), an alternative Solesmes version was issued by Desclée with rhythmic indications: bars and dots for longer notes, accents to indicate the 'ictus', for that note in a group which the singer should understand to be the most important, without stressing it dynamically. These remained a characteristic of Solesmes editions of the official books, and of course of their own house productions, notably the Liber usualis.
Commission:
Glo-ri- a in ex-cel-sis de-o b    aG G G ab b* abcb aG G Glo-ri- a in ex-cel-sis de-o
628
A7. The Restoration of Medieval Chant
XI.3. PRACTICAL EDITIONS AND SCHOLARLY RESEARCH
Stablein, 'Gregorianik', MGG; Hucke 1988, 'Choralforschung'.
After the bruising experience of the Vatican edition, Solesmes remained charged with the preparation of chant editions for the Roman church, a task it continues to perform. During the present century, however, circumstances have changed, in some ways dramatically. Mocquereau worked in the belief that an 'original' form of the earliest melodies could be established by comparison of the oldest sources, in the way that Gueranger had asserted years before: 'When manuscripts of different periods and countries agree upon a particular reading, we can safely assert that we have discovered the Gregorian phrase.' As this book has tried to make clear, it is not at all certain that an 'original' form of this type ever existed. From the Old Roman manuscripts, for example, we can see some of the Roman ways of singing melodies in the eleventh to thirteenth centuries (more than one, for the manuscripts do not agree). We can see several Frankish ways of singing the same melodies in the ninth century, the tenth century, and so on. But the manuscript tradition is too variable for a single 'authentic' reading to be deduced even from a small group of the earliest sources. In Le Graduel Romain IV the monks of Solesmes showed how a number of early traditions could be identified, and proposed a comparative edition where the divergent readings would be recorded (see also Froger 1978, 'Edition'). The work is not complete, and a corresponding edition of office chants is also only in the preparatory stages.
It is not only the magnitude of the task which has delayed its completion. The changes in the liturgy announced at the Second Vatican Council (1962-5) have necessitated the preparation of other new service-books, of which the Psalterium monasticum (1981) and the Liber hymnarius (1983) have so far appeared. These, like all the previous Solesmes editions, are service-books for practical use.
In the history of plainchant research, the practical and the abstract have been inextricably entwined. Scholarship devoted to finding out and understanding the music of the past and present has sometimes been almost inseparable from a religious belief in a 'pure' tradition and the necessity of providing the 'best' attainable form of the melodies for contemporary use. And contemporary use does not stand still. The goal of more recent liturgical reform is no longer the 'Age of Faith' which inspired the efforts of the chant restorers, but rather still earlier centuries from which no liturgical music will ever be recoverable. Christians have not usually found it difficult to worship in churches containing architectural elements of widely differing age, and the liturgy itself is an amalgam of compositions and ceremonial of diverse origins. The plainchant repertory, just like the repertories of polyphonic music which developed alongside it, contains a multiplicity of forms and styles, again of widely differing origin—another point which I hope has become clear in the course of this book. As far as its use in modern worship is concerned there is much to be said for the view of St Gregory, giving advice to his missionary Augustine:
3. Practical Editions and Scholarly Research
629
My brother, you are familiar with the usage of the Roman Church, in which you were brought up. But if you have found customs, whether in the Church of Rome or of Gaul or any other that may be more aceptable to God, I wish you to make a careful selection of them, and teach the Church of the English, which is still young in the Faith, whatever you have been able to learn with profit from the various Churches. For things should not be loved for the sake of places, but places for the sake of good things. Therefore select from each of the Churches whatever things are devout, religious, and right; and when you have bound them, as it were, into a Sheaf, let the minds of the English grow accustomed to it. (Bede, A History! of the English Church and People, i. 27, trans. Sherley-Price, 73.)
That would mean that Carolingian chant of the ninth century could take its place beside rhymed songs of the twelfth or a hymn of the seventeenth century, and, since the 'authenticity' of the chant is no longer an issue, the selection may be made on grounds of religious quality alone.
These are, however, problems of liturgical practice, not musicology, which is surely mature enough as a discipline to distinguish between scholarly and practical concerns in the recovery of old music. The problems facing chant scholarship are great enough without the added complications of liturgical controversy. They concern above all the enormous quantity of potential evidence. What are the right questions to ask when confronted by the huge numbers of liturgical music-books from the ninth century to the present day? How is the material to be controlled and investigated? Exactly what is there in these books, only a tiny fraction of which have been inventoried? What does it tell us about man as a religious being and as a creator of music? Compared with these problems, the tasks which the restorers of the nineteenth century set themselves were essentially simple, concerned with but a part of the chant repertory, mighty though the labours were that were needed to carry out those tasks. It is my hope that this book will have conveyed a sense not only of past achievements but also of the problems that scholarship still faces. But above all, I hope it has helped create an awareness of the inexhaustible musical riches of Western plainchant.
INDEX OF TEXT AND MUSIC INCIPITS
Citations in bold indicate a musical example.
A solis or/ns cardhw, hymn 142, 284
A summo celo, gradual 533-4
Adccuam agm, hvmn 234
Ad dextrampains, Agnus trope 233
Adjesum accurate, sequence 619-20
Ad laude/n regis glorie, farsed lesson 592-3
Ad te Domine levavi, offertory 123
Adam novus velerem, song in Fleury Peregrinus play 27 Adesl dies lelicie, antiphon 276-7 Adiulorium nostrum, antiphon 91-2 Adoramus crticem liiain, antiphon 527 Adoma thalamum sunm, processional antiphon 31, 102, 104
Adoru te devote, hymn 25
Aeterne rerun) conditor, hymn 140, 281
Affluens delictis, sequence 195
Agios (Greek Sanctus) 162-4, (163), 528
Arioso Theos, Trisagion 36-7, 498, 528
Agnoscat omne saecultim, hymn 142
Agnus, Old Roman 166-7
Agnus 15 232
Agnus 34 (Vatican XVII) 166-7 Agnus 81 a 233 Agnus 98 166
Agnus 101 (Vatican XVIII) 149, 166-7
Agnus 114 (Vatican IX) 167
Agnus 136 (Vatican IV) 167
Agnus 144 167-8
Agnus 164 (Vatican XVI) 167-8
Agnus 209 (Vatican XV) 167
Agnus 220 (Vatican XI) 166-7
Agnus 226 (Vatican II) 166-8, 167
Ales diet nuntius, hymn 142
Alias ores habeo, antiphon 473
Alleluia, antiphon 39
Alleluia V (Milanese) 546
Alleluia VI (Milanese) 546
Alleluia Ante thronum trinitatis 137-8
Alleluia Attendiie papule mens 133-4
Alleluia Beatus sanctus Martinas 134
Alleluia Chnstus resurgens 201
Alleluia Concussion est mare 203-4
Alleluia Coufitemini Domino quoniani bonus 38, 420-1
. Miel uia C 'um esset Stephan us 13 4
Alleluia Dies sanctijicatus 131
Alleluia Dies sanclificaluslYmera agiasnieni 528
Alleluia Dominus dixit ad me 131-3
Alleluia Dominus regnavit decorem 529, 538
Alleluia Ego sum pastor bonus 134
Alleluia Epi si kyrie 529
Alleluia Kxcila Domine 131
Alleluia In te Domine speravi 529
Alleluia Levita Laurentius 5
Alleluia Suptie facte sunt in Chana 134-5
Alleluia Okynos 430-1, 529
Alleluia O Maria rubens rosa 138-9
Alleluia Ora voce pia 138
Alleluia Ostende nobis 132, 181-3, 182
Alleluia Oty theos 529
Alleluia Paratum cormeum 538
Alleluia Pascha nostrum 375-7, 379-80, 586
Alleluia Quoniani Dens magnus 529
Alleluia Surrexil Dominus et occuriens 134
Alleluia \ eni sancte Spiritus 43
Alleluia \ eni sponsa Christi 135-6
Alleluia Yirga lesse floruit 137
Alma fulgens, sequence 183
Alma redemptoris mater, Marian antiphon 104-7
Alto consitio, Latin liturgical song 248
Anten amen dico vobis, antiphon 464
Angeli eorum, antiphon 234
'Angelica', sequence melodv 586
Angelis suis, gradual 533-4, 554
Angelorum ordo sacer, sequence 181
Angelus Michael, alleluia prosula 203-4
Anima nostra, gradual 80
Annua sancte dei, processional hvmn 146
At/ua sapienliae, introit 221
Ardua spes inundi, processional hymn 146—7
Ascendenle lesu m navim, antiphon 329
Ascendil Dens in iubilalione, offertory 126
Ascendil Simon Petrus, antiphon 327
.\sperges me, antiphon 22, 148
Altende celuni, canticum 359
Audi iudex mortuorum, processional hymn 486
'Aurea', sequence melody 200
Aurea luce, hymn 234
Aurea lux terra, processional hymn 146
Aurea virga, sequence 153
Aurei Hominis Maria, Marian antiphon 107
Ave beatigenninis, processional hymn 146
Ave gratia plena, processional antiphon 102
Ave Maria, prayer 25
Ave mater salvatoris, Latin liturgical song 247 Ave regina caelorum, Marian antiphon 104-6 Ave i'irgo speciosa, verse 43
'Bavverisca', sequence melody 546 Be den hoi mats, song 248 Beatiquos elegistis, antiphon 451 Beatus Andreas de cruce, rcsponsory 337 Beatus ille senus, antiphon 355
632
Index of Text and Music Incipits
Beatus Nicholaus, responsory 206 Beatus reenter, antiphon 4SI Beatus Vincentiiis, antiphon 327 Beatus vir, antiphon 234 Beatus vir qui inventus, antiphon 329 Beaus peres, Sanctus trope 225 Benedic annua mea, offertory 535-6 Benedicam Dominum, offertory 126, 129 Benedicamus flon orlo, Benedicamus 214 Benedicamus . . . Verbi iungando cannina,
Benedicamus 214-15 Benedicite Dominum, introit 372 Benedicta tu in mulieribus, antiphon 381 Benedictas es Do/nine, offertory 375 Benedict us es in ftrmamento, canticle 23 Benedictas es . . . in labiis, offertory 130 Benedixisti, offertory 201-2
Caeli caelorum laudate Deum, antiphon 453
Candida concio, sequence 587
'Caput', melisma 36
('aritas Dei, introit 115
Cannen suo diiecto, sequence 387
Ce que Ysaies nos escrit, farscd lesson 237
'Chorus', sequence melody 178
Christ ist erstanden 225, 261
Christe qui lux es, hymn 142-3
Christe redemplor, Kyrie trope 213
Christe redemptoromnium, hymn 234
Christifortis hie athleta, antiphon 406-7
Christus fact us est, gradual 80
Christus manens, verse 43
Christus resurgens V. Dicant nunc ludei, processional
antiphon 104 Cist aigneaus, Agnus trope 225 Cito euntes, antiphon 254 Cives apostolorum, rcsponsory 337, 458 Clamaverunt iusti, gradual 81 Clangant cigni, sequence 178 Claragaudia festa paschalia, sequence 586-7, 588 Clare sanctorum, sequence 200, 416-17, 440-1 Claris vocibus, sequence 173-5, 174 Claudius quidam, antiphon 329 Clemens rector, Kyrie trope 434-5 Coenae tuae mirabili, Milanese ingressa 527 Collegerunt pontifices, processional antiphon 34, 102 Commovisti, tract 83, 85 Concelebremus sacram, sequence 239 'Concordia', sequence melody 178 Confessio et pulchritudo, introit 5 Confirmation est cor, responsory 43 ('ongaudentes exultemus, sequence 189-91 Congaudentes iubilemus, Jube song 249 Convertimini omnes, Milanese transitorium 542-3 Convoeatis ergo, antiphon 327 Cordepatrisgenitus, Benedicamus song 43, 224 Credidi propter bait us sum, antiphon 87 Credimus in unum Deum 169-70 Credo Vatican 1 169 Credo Vatican II 169 Credo Vatican III 171
Credo Vatican IV ('Credo cardmalis') 171
Credo Vatican V 169
Credo Vatican VI 169
Credo Vatican VII 171
Cntcem tuam adoremus, antiphon 37, 527
Crux benedicta nitet, hymn 142
Cnix fidelis, hymn (cf fange lingua) 37, 145-6
Cultor agri, antiphon 276, 278
Cum appropiuquaret Dominus, processional antiphon 34, 102
Cum audisset populus, processional antiphon 34, 102,
553-4, 555 Cum ergo, antiphon 327
Cum natus esset Dominus, processional hymn 146
Cum pervenisset bcatus Andreas, antiphon 97-8
Cum repente, antiphon 327
Cum venerimus, processional antiphon 104
Cum videretlvidisset, responsory 337
Cunctis diebus, antiphon 93-4
Currebant duo simul, antiphon 261
Cuivali sunt, antiphon 553
Cnstodi nos, versiele with response 30
Cuslodiebant testimonia, antiphon 61
Da laudis homo, Latin liturgical song 246
Da nobis famuli's, offertory prosula 202
Da nobis potenti, offertory prosula 202
Data est michi, antiphon 258
De fructii openim tuonim, communion 473-4
De necessitatibus, tract 82
De sancte lohanne, sequence 178
De stipends affero nuntio, Latin liturgical song 248
Dei sapientia, Benedicamus song 242
Deprecamur te, rogation antiphon 102, 591
Descendit de celis, responsory 200
Desiderio desideravi, antiphon 93
De us creatoromnium, hymn 140
DensDeusmens (Ps. 62) 488
Dens Deus mens, offertory 126
Dens Deus mens, tract 34, 82-3
Deus enimfinnavit, offertory 129
Deus in adiutorium meum, versiele with response 25-6,
28, 35, 43, 248 Deus in adiutorium meum, introit 449 Deus paterfilium suum, introit trope 221-2 Die nobis, sequence 586-7 Dicant nunc ludei, see Christus resurgens Dicit Dominus implete, communion 119-20 Dies est leticie, song 225 Diffusa est gracia, rcsponsory 120 Diffusa est gracia, communion 120 Dignare me, antiphon 381 Dilectus isle domini, introit trope 197—9 Dilexit Andream, responsorv 337 Diligite Dominum, antiphon 468 Dixit Dominus Domino meo, antiphon 451 Doctor bonus, rcsponsory 337 Documentis allium, antiphon 276, 278 Domine qui operati, antiphon 464, 473 Domine spes sanctorum, antiphon 329 Domine audivi, tract 83
Index of Text and Music Incipits
633
Domine clainavi ad te (Ps. 140) 488 Domine Dens mens, offertory 130 Domine cxaiidi, tract 83
Domine labia meet apenes, vcrsicle and response 25 Domine non secundum, tract 44, 519 Dominies! terra (Ps. 23) 45 Dominus legifer noster, antiphon 114 Dominus rirtiilum, communion 412-13 Dominus vobiscum, versicle with response 47-8 Dornum is tam, antiphon 327
Doxa en ipsistis theo (Greek Gloria in excelsis Deo) 158, 528
Dulce carmen, versus with double cursus 240
Dum delperambulanl, responsory 337
Dum penderei, responsory 337
Dum transissel Sabbatum, responsory 39, 252, 261,
414-15 Duo homines, antiphon 327
ľ, voces unisonas equal (Hcrmannus Contractus) 396
Ecce annunlio vobis, antiphon 43
Ecce carissimi, processional antiphon 104
Ecce concipies, antiphon 327
Ecce iam Christus, sequence 221
Ecce iam Johannes, introit trope 217-19
Ecce iam noctis, hymn 143-4
Ecce lignum cruris, antiphon 36
Ecce oculi, introit 113, 115
Ecce pulchra, sequence 153
Ecce sacerdos, antiphon 451
Ecce sacerdos, gradual 80
Ecce vere Israelita, antiphon 449
Ego autem sicul oliva, introit 372
Ego sum alpha et w, processional antiphon 103-4
P'.go sum Dens, processional antiphon 101
Ego sum pastor bonus, antiphon 327, 329
P.'go sum resurreclio, antiphon 45
Ego sum vitis vera, antiphon 87, 344
Eia nuncpueri', Benedicamus 214
Eia recolamus, sequence 173-5, 234
Elegeriiul aposloli, offertory 554
Etnendemus in melius, antiphon 32
I'.mendemus in melius, responsory 271
h'.n rex vetul, processional hvmn 34
Epiphaniam domino, sequence 178
Enpe me, gradual 359
Enpe me, tract 82-3, 519
Erubescant, communion 408-9
Erumpant monies, antiphon 451
Et intravit cum Ulis, antiphon 329
Et respicientes viderunt, antiphon 329
Elerne virgo memone, responsory prosula 207
Ex eins t umbo, responsory 205
Exaltabo te, gradual 81
Exaltata es, antiphon 381
Exaudi nos Deus, antiphon 31
'Excita domine', sequence melody 181
Exiil senno, gradual 80
Expandimanus, responsory 337
Exsultabunl sancli, gradual 79, 358
Ex surge Domine feropem, gradual 612
Exsurge Dotnine uou prevaleat, gradual 624-5 Exultemus in hac die, sequence 592
'Fabricae mundi', mclisma 200, 316
Eac benigne, antiphon 327, 329
Eidehs senits, antiphon 93-4, 327
Eidetn auge. Paternoster farse 238
Einnetur manus tua, processional antiphon 44
Flete fideles animae, Lament of Mary 266
'Flos filius', mclisma 214
'Francigenae', Milanesemelodiae 545-6
'Frigdola', sequence melody 178
Eulgens preclara, sequence 181
Fundamentum aliud, antiphon 424-5
Gaude eia unica, sequence 178
Gaude et letare, Milanese transitorium 542
Gaude felix parens Yspama, antiphon 276-7
Gaude Maria, responsory 207
Gaude Maria virgo, antiphon 381
Gaudeamus omnes, antiphon 94-6,95
Gaudeamus omues, Old Beneventan ingressa 550
Gaudenl in celis, antiphon 581
Gaude te in Domino semper, introit 456-7
Gloria 2 158-9
Gloria 5 160
Gloria 11 (Vatican XIV) 224 Gloria 13 (Vatican ad lib. II) 158 Gloria 22 (Vatican ad lib. Ill) 158 Gloria 23 (trope Spintus el alme) 161 Gloria 24 (Vatican ad lib. I) 159 Gloria 25 (Vatican V) 158-9 Gloria 38 (Vatican VIII) 158
Gloria 39 ('Gloria A') 157-8, 200, 209, 224, 228-30 Gloria 39 trope Laus tua Dens . . . Sapienlia Dei
patris 422-3 Gloria 43 (Vatican XV) 158 Gloria 51 (Vatican XI) 158 Gloria 56 (Vatican IV) 158, 208-9, 224 Gloria A, see Gloria 39 Gloria in excelsis Deo, antiphon 263 Gloria laus el honor, processional hymn 34, 146, 486 Gloria . . . Pater elernus, doxology prosula 334 Gloria . . . Quernglorificant cell, doxology prosula 334 Gloriosa sanclissimi, office 275 Gregonus praesul 223, 503, 507, 509-11 Gregorius vigdiis, antiphon 276 Gustate el videte, communion 116, 410-11
I lac est clara dies, verse 43 llaecdies, gradual 29-30, 77 Heine concordifamulatu, sequence 178 llec aula accipial, antiphon 424-5 Elec est quem nescivi, antiphon 327 Hemdes enim, antiphon 93-4
lieu heu hen quo casu sortis, lament of Daniel in Ludus
Danielis 272 /// sunt secuti, responsory 436-7 Hie est discipulus, antiphon 93—4 Hie vents Dei famulus, song in Ludus Danielis 270-1 'Hicronima', sequence melody 178
634
Index of Text and Music Incipits
lIodic can fundus est, introit trope 222-3, 418-19
/Iodic dominus testis ('hristus, sequence 183
/Iodic exultcnt tusti, introit trope 586-7
/Iodic intacta vugo, antiphon 43
'Hodie Maria virgo', sequence melody 153
I Iodic resurrexil Ico foiiis, introit trope 586
I Iodic surrexit leofortis, Benedieamus 214-15
Ilodierne lux did, sequence 191,192, 193
Homo Dei, rcsponsory 337
Homo quidain, antiphon 327
J Josanna filio David', antiphon 34
Ilosoieis Christon, Byzantine baptismal chant 529
/ lost is I/erodes impie, hymn 142
Humanuni genus, Agnus trope 233
Humilipiece, processional hymn 146-7
Hyinnum canite, communion 551
lam lucis oi In sulere, hvmn 8
Jam siugit hora tertia, hymn 140
Immulemur habitu, antiphon 31
In conspectu angelomm, offertory 299
In die qiuvido, processional antiphon 104
/// die resuireetionis mee, processional antiphon 430-1
/// die sollempnitatis, offertory 123-4
/;/ eodem namque, antiphon 327
In exitu Israel (Ps. 113) 64
In Hicmsalcm, offertory prosula 579
In medio ecclesie, introit 197-9, 216, 218-19
In medio ecclesie, rcsponsory 200
//; omncm terrain, gradual 79
In paradisian, antiphon 45
In virtu le, offertory 554
Incomparabiliter cum iocunditate, song 283
Ingrediente Domino, processional antiphon 34
Inlliens in celum, rcsponsory 72
Inventor rutili, processional hymn 38, 142, 319, 554
Inviolata Integra, rcsponsory prosula 207
Invocavit me, introit/ingressa 547-8
Isle confessor, hymn 144
Isle sanelits pro lege Dei, antiphon 329-30
Jubilate Deo omnis terra, offertory 125
Jubilate Deo universa terra, offertory 124-6, (125),
128-30,(129) Iubilatc Domino, tract 83-4, 371 lubilcmtis cordis voce, song 270 lubilemits regi nostiv, song in Liulus Daniclis 270 luravit Dominus, gradual 81 Instils es Domine, introit 456-7 lustusgenninabit, antiphon 327 Justus non coiiturbabitur, gradual 377-8 lustus ul palma, gradual 77, 533, 626 'lustus ut palma', sequence melody 173
Jacobe virgiiici frater, rcsponsory 278 Jerusalem Jerusalem, antiphon 327 Johannes Jesu Christo, sequence 286, 586 Jube domne bencdiccrc, versiele 249-50
Kyrie 7 155
Kyrie 39 (Lux et origo, Vatican I) 28, 152, 155, 627 Kyrie 47 (Vatican VI) 152
Kyrie 48 (Kyrie Jons bouitatis, Vatican II) 155, 211-12 Kyrie 52 155
Kyrie 55 ('I'e Christe/palersupplices, Vatican ad hb.
VI) 152-3, 155,210,211,213,434-5 Kvrie68 (Vatican XIV) 152-4,154 Kyrie 74 418-19 Kvrie 77 155 Kyrie 84 155
Kvrie 95 (Vatican VIII, 'de angclis') 155, 158, 171 Kyrie 97 156
Kyrie 102 (Clemens rector, Vatican ad lib. I) 152, 155, 434
Kvrie 124 152, 155
Kvrie 139 156
Kyrie 142 155
Kyrie 144 152, 155
Kvrie 151 (Vatican XVI11) 152,155
Kvrie 155 (Vatican XV) 152-5,154
Kyrie 171 (Vatican IX) 438-9
Kyrie Auetor celorum 155
Kyrie Devote canentes 151, 155
Kyrie 'de angelis', see Kyrie 95
JLaeta lux est hodierna, sequence 584 l^apidabant Stephanuni, rcsponsory 72 JauuIu Sion salvatorcin, sequence 192 'Laudate Deum', sequence melody 181 Laudate Dominum, tract/canticum 38 Ixiudate Dominum quoniam, offertory 579 Ixiudate Dominum de eclis, antiphon 453, 500 Laudatepueri (Ps. 112) 61 Landes enteis attolamus, sequence 192, 194 Landes Deo, sequence 178
Landes Deo dieam per secuta, epistle farse 236-7, 249, 283
IMildes omnipotens, processional hymn 146 iMudum caniiina, sequence 178 ImuiciiIi Davul magni martyr, sequence 586 Laus tua, Gloria trope 200
lA'labiiiuli iltbilemus, Latin liturgical song 243—4
Ix'tabuiuliis decanlct, sequence 584
la'tabundits cxidtet jidelis chorus, sequence 43, 192, 234
I atemur "audits, offertory prosula 43, 204
I'A'to animo, antiphon 327
Liber carnis vinculo, antiphon 277
Libera me Domine, rcsponsory 45
Lignum vile, antiphon 329
Linguam refienens temperet, didactic song 426—7 Lumen ad revelatwnem gentium, antiphon 31 Lumen eternum, Sccptruni prosula 228-30 Lutum fecit, communion 512 Lux de luce, sequence 183 Lux hodie, verse 43
Lux omni festa poplilo, Benedieamus song 242-3 'Lvra', sequence melody 153
Magi veniitnl, processional antiphon 265 Magister elicit, antiphon 93 Mandátům novum, antiphon 34, 35 Mane pnma sabbati, sequence 192 Maria ergo, antiphon 327
index of Text and Music Incipits
635
Mana Magdalena, antiphon 261
Martyr Christi gratia, Hosanna prosula 225, 230-1
Mentor sit Dominus, offertory 554
Metitibus intenlis, introit trope 285
Mercenarius est, antiphon 329
Merito her patimur, responsorv 271
'Metensis niaior', sequence nielodv 546
Misi digitum, antiphon 258
Mitte maitiim tuain, communion 118-19
Modicum el non videbitis, antiphon 94-5
Max ul vocem, responsorv 337
Mnlta quidem, antiphon 329
'Musa', sequence nielodv 586
Musirorum et canloruni, didactic rhyme 523
Xato caniwt omnia, sequence 234
Xatttsest, trains est, Latin liturgical song 245
Xatns ante secuta, sequence 234
-W avert as, gradual 79
Xenw te candemnavit, communion 512
Xesciens mater, antiphon 43, 94
Xigra sum, antiphon 93-4
'Nobilissima', sequence melody 586
Xon vos demergat, processional antiphon 101
Xos anient gloriari, introit 408-9
Xos pueri iubilo, introit trope 285
Xos c/id viximus, antiphon 63-4
Xostrifestigaudiuni, Latin liturgical song 249
Xotuin fecit Dominus, antiphon 63, 87
Xitnc elerieorum eoneio, Latin liturgical song 248-9
Xunc fratres enesis, introit trope 285
Xilliquid Dan obseivan, from Ludits Damelis 272
O admirabde commercium, antiphon 95-6, 464
OAdonay, antiphon 98
O amiios tit then, Greek Agnus Dei 528
O heatalbona crux, responsorv 337
O bealum ponlijicem, antiphon 463-4
() bealum president, responsorv 406-7
O rrucifer bone, hviiin 142
() crux henediela, processional antiphon 318
()gloria sanctorum, Gloria trope 208-9, 224
O laudabilis rex, Gloria trope 224-7, 228-30
O mater nostra, responsorv 207
O quando in cruce, antiphon 527
O Radix /esse, antiphon 98
0 redemptor sume cannen, hymn 35
O rex glorie, Rcgnum prosula 208-9
O sancle Dei apostate, responsorv 599-600
O Sapienlia, antiphon 98
O virgo virgin urn, Marian antiphon 105
'Occidentana', sequence mclodv 546
Otto sunt beatitudines 332
Oculi rnei, introit 221, 449
unities de Saba, gradual 616-18, (617)
Ottilies qui in Christo, offertorv/comrnunion 525, 529
Ottilies tuagratia, Hosanna prosula 225,230-1
Omni tempore, antiphon 329
Omnipotent tern semper adorant, canticle 23
Oportet devot a mettle, responsory prosula 206
Oportet te, communion 512 Oravit sanctiis Andreas, responsorv 337 Otganicis canamits moduhs, sequence 173, 175, 176, J 78
Orientis pat tibus, Latin liturgical song 43 'Ostende (minor)', sequence nielodv 182, 221 'Ostende (maior)', sequence melody 200
Pangat simul eia, sequence 178-9
Range lingua gloriosi corporis mystérium, hymn 145
Range lingua gloriosi proelium certatninis, processional
hymn 25,35,37, 142, 144-6, (145), 239, 282, 284, 486 Raradisiporte, antiphon 381
Raranymphus salutat virginem, sequence 192, 193
Pascha nostntm, communion 118-19, 398
Past is visce nbus, h y m n 142
Pastor bonus, antiphon 327
Ratefacte sunt ianue celi, responsory 72
Pax in tetris nunciatur, Latin liturgical song 263
Pax vobis, antiphon 259
Peccavimus Domine, rogation antiphon 591-2 Perpetua manině, Sanctus trope 234 Petrus autem servabatur, antiphon 329 Petrus donniebat, Old Beneventan ingressa 550 Pisteuo eis ena theon, Greek Credo 528 Plaudits ante nescia, Lament of Mary 266 'Planctus cigni', sequence mclodv 178, 188 Rlangant cigni, sequence 178 Plebs Domini hoc die, Latin liturgical song 246-7 Ropule mens, respond 36-7 Populus gentium, chapter 43 Portio mea, antiphon 87
Populus Sion convertiinini, processional antiphon 101
Post pa tin in virgo, antiphon 381
Post sex annos, responsory 278-9
Rostquain celoritin Dominus, Benedicamus song 270
Roteslate non natura, sequence 592
Precamur nostras, sequence 181-3, (182)
Rreco novuset celicus, antiphon 277
Precursor ( Vmslt, sequence 591
Rrimo dierum omnium, hvmn 143
Primutn mandátům amor Dei est 331
Primům que rite regit wn Dei 331-2
Pro cundis dedudus, Agnus trope 232
Pro eo quod non credidisti, antiphon 94—5
Probavit, introit 599
Prospěním iterfacial, introit 554
Rsallemodidamina, alleluia prosula 201
Psallite regt magno, introit trope 217
Piter natus nobis est, introit 221-3, 399, 418-19
Pueri IJebreoritm portantes, processional antiphon 34
PueriIlebreorittn vestinienla, processional antiphon 34
Rura de urn, sequence 178-9
Quando natus est, antiphon 327
Quarta vigilia venit ad eos 332
Quasi umis, antiphon 327
Que mutter habeas, antiphon 95-6
Quem creditis super astra ascendisse 255
Quem ethera et terra, responsory prosula 207
Quern queritis in presepe 254-5, 264
636
Index of Text and Music Incipits
Quem queritis in sepulchre) 223, 250-5, (253), 263, 267,
414-15,519,585-6,602 Quern ten a pmitus aethera, hymn 142 Quem vere pia laus, Gloria trope 452 Qui bibeiit, communion 512 Qui de terra est, antiphon 43, 327 Qui dial patri, introit trope 586-7 Qui habitat, tract 82-3 Quipurgas aninias, sequence 591 Qui regis sceptra, sequence 181 Quia vidistime, antiphon 258-9 Quid tu virgo mater, sequence 265 Quidam homo fecit, antiphon 329 Quinque pruclentes intraverunt 332 Quis dabit ex Sion, communion 355-6 Quomodo fiet, antiphon 451 Qitoniam Deusmagnus, invitatory antiphon 99
lieges Tharsis, offertory 123, 619-20
Regina caeli, Marian antiphon 104, 106
Regnuin tiiitrn soliclum, Gloria trope verse 199-200
Repleaturos meum, introit 372
Requiem eternam, versicle 45
Resonel internet, Latin liturgical song 249
Respire Domine, introit 221
Resurrexi, introit 197, 216-17, 252-3
Resurrexil Dominus, short responsory 87
Revelabititrglona Domini, communion 532
Rex benedicte, processional hymn 146
Rex caeli domiiie, versus with double cursus 240
Rex omnipotens, sequence 234
Rex sanctorum angelorum, processional hymn 38
Rex splendens, Kyrie trope 581
'Romana', sequence melody 586
Sacerdotes tin, introit 115 Sacrata libri dogmata, processional hymn 146 Salus ciutem, introit 457 Salus etenia, sequence 200 Salvator initncli satva nos, antiphon 37 Sähe crux, responsorv 337 Salve fesla dies, hymn 43, 142, 146,269 Salve lacteolo decorat um, processional hymn 146 Salve regina, Marian antiphon 104, 106-7 Salve sancta parens, introit 281 Salve virgo virginion, Regnuin prosula 209 Salve virgo singularis, Latin liturgical song 264 Sancta legio :\gaunensis, antiphon 327 Sancte cruris celebremus, sequence 183-4 Sancte Johannes meritoruni tuorum, didactic chant 426-7
Sancte Paule pastor bone, versus with double cursus 239-41
Sancti baptisle, sequence 173, 175, 177, 178 Scinrti qui sperant, antiphon 327, 329 Saudi Spiritus, sequence 234 Sanctificavit Moyses, offertory 554, 556-7 Sanctorum mentis, hymn 145 Sanctum diem celebremus, sequence 184 Sanctus 32 (Vatican XVII) 162 Sanctus41 (Vatican Will) 161-2
Sanctus 49 (Vatican IV) 162 Sanctus 56 (Vatican III) 230-1 Sanctus 66 163 Sanctus 68 164-5 Sanctus 89 163 Sanctus 98 164-5 Sanctus 111 163 Sanctus 116 (Vatican VIII) 162 Sanctus 154 (Vatican i) 163 Sanctus 155 163 Sanctus 177 (Vatican XII) 162 Sanctus 202 (Vatican XI) 162 Sanctus 203 (Vatican II) 162 Sanctus 216 163-4 Sanctus 223 (Vatican XV) 162-3 Sanctus Deus angelorum, antiphon 498 Sapientia magnificat, antiphon 327 Scapulis suis, offertory 537 Scitote quia prope est, antiphon 61 Secundum anient simile est hide 332 Secundum mcignam misericordiam, antiphon 91-2 Seclit angelus, sec Stetit angelus Septem sunt Spiritus 332 Serve bone et fidelis, antiphon 329-30 Sexta hora sedit superputeuin 332 Si in digito Dei, antiphon 329 Si oportuerit me mon, antiphon 329 Si quis diligit me, antiphon 329 Si vere frat res, antiphon 533, 535 Sicut ceivus, tract 82 Simon Petrus, antiphon 327 Sinite parvulos, antiphon 264 Sitirit in le, antiphon 91-2 Solve iitbente Deo, antiphon 274 Somno refectis artubus, hymn 142—3 Sospes nunc efficitur, responsorv prosula 205 Sospitali cledit egros, responsorv prosula 205-6 Specie tua, antiphon 534-5 Speciosa facta es, Marian antiphon 105-6 Sperent in te, Milanese responsorium cum infantibus 544-5
Spiniiis et cdme, Gloria trope 161, 223
Staus a huge, sequence 234
Staus beata Agatha, antiphon 97-8
Staus beata Agues, antiphon 97—8
Statuit ei Dominus, introit 450, 599
Stetit angelus V. Crucifixion in came, processional
antiphon 104 Stetit angelus, offertory 299 Stiips lesse, responsory 214 Stola iocunditatis, sequence 192 Sub altare Dei, responsory 265 Sub tiiunipresidium, Marian antiphon 105 Subvenite sancti Dei, responsorv 45 Summi triumphum regis, sequence 522 Superfluinina, offertory 123, 129 Super te leruscdem, short responsory 86 Surgepropera, Old Beneventan ingressa 550 Sitrrexerunt autem, antiphon 327 Surrexit Christus hodie, song 225 Surrexit Dominus, communion 118-19
Index of Text and Music Incipits
Surrexil Dominus de sepulchro, antiphon 254, 258-60 Sur/v.xit oiim sicul dixit, antiphon 254 Suseepimus Dens, introit 456-7 Suscipe elemental/ ptebs, processional hymn 146
lantiim ergo sactamentum, hymn 25-30
Te decet laus, hymn 27
Te laiidamtts, Milanese transitorium 542
Te pater supplices, Kyrie trope 434-5
Tecum prineipium, gradual 77-9, (78)
Tellus ac ethra, processional hymn 239
Tempora florigem, poem 224
Tenemus ecce anna, antiphon 327
Terribilis est, introit 45
Terribilis rex ahne, sequence 181
Tertia dies est quod haec facta sunt 332
Thesaurus novegracie, Benedicamus song 246
Tibi dixit cor meum, introit 449
Tibi laus perennis auetor, hymn 142
Timete Dominum, introit 114
Tolle anna tua, responsory 99
Tolliteportas, antiphon 45, 327
Tolliteportas, gradual 77
Tollite portas, offertory 126-8
Tria sunt munera, processional antiphon 265
Trinitas deitas, Sanctus prosula 209
Trinitatis thalamum, Marian antiphon 108
Tu autem Donime, versicle 248
Tu Bethleem, antiphon 329
Tuam criteem adoramus, antiphon 329
t'nxerunt Salomonen! Sadoch sacerdos, antiphon 44 l rbs beala Iemsalem, hymn 234 l't quean! laxis, didactic hymn 467
Vadit propitiator 529 Valde honorandus est, antiphon 533, 535 Valerius igitur, antiphon 327 Veni doctorprevie, organum 43 Veni redemptorgentium, hymn 140 Vemt ad Pelruin, antiphon 36 Venite et videte, antiphon 254-5 Yenite post me, responsory 337 Venite venite, antiphon 35 Verbum cam factum est, responsory 207 Verbum patris hodie, Bencdicarnus song 264 Verbum superman, hymn 234 Veteran hominem, antiphon 525, 530 Yexilla regis, hymn 142, 144, 284 Victimae paschali laudes, sequence 189-90, 260-1, 284
Videbant oinnes Stephanum, responsory 72
VidenslVidit crucem, responsory 337
Videns Dominus, communion 512
Videns Rome vir beatus, responsory 275
Video virum, antiphon 327
Yiderunt otnnes, gradual 81
Videsne FAisabeth, Milanese ingressa 527
Vidi aquam, antiphon 22, 148
I 'ir iste, responsory 337
Virginisproles, hymn 144
Virgo hodie fidelis, antiphon 43, 94
Virgo verbo concepit, antiphon 43
Vocem iocunditatis, introit 372-3
Vocivita, Sanctus prosula 209
Zachee festinans descende, antiphon 45
INDEX OF MANUSCRIPT AND PRINTED SOURCES
Citations in bold indicate facsimile and partial transcription.
Aachen, Bischöfliche Diözcsanbibliothek 13 313 Aachen, Dombibliothek 20 87, 91-3 Abbeville, Bibliotheque Municipalc 7 411 Aberystwyth, National Library of Wales 20541 E 584 Albi.Bibliothčque Municipale 44 303, 321, 596-7 Angers, Bibliotheque Municipale 91 624 Angers, Bibliotheque Municipale 97 179 Angers, Bibliotheque Municipale 144 586 Annecy, privately owned manuscript 275-6 Antiphonale missaruin (Milan) 541, 556-7 Antiphonale monaslicum 62, 257, 289, 309, 397 Antiphonale Romanům 62, 289, 309, 341, 401, 627 Antiphonale Romanům (Ratisbon, 1878) 624 Antiphonale Sarisburiense (Cambridge, University
Library Mm.2.9) 48, 66, 93, 106, 114, 120, 206-7,
303,535
'Antiphoner of Charles the Bald', see Paris, Bibliotheque
Nationale, lat. 17436 Aosta, Seminario Maggiore 9.E.19 250 Apt, Cathédralc Saintc-Anne 17 313, 598 Apt, Cathédrale Sainte-Anne 18 313, 598 Assisi, Biblioteca Comunale 695 57-8, 192-3
Baltimore, Walters Art Gallery M.6 320, 589 Bamberg, Staatsbibliothek lit.'5 328, 330, 570 Bamberg, Staatsbibliothek lit. 6 295, 313, 347, 373 Bamberg, Staatsbibliothek lit. 23 303, 337-8 Bamberg, Staatsbibliothek lit. 25 99 'Bangor Antiphonary' 157, 553 Barcelona, Biblioteca de Catalunya M.662 205, 603 Barcelona, Corona de Aragon 99 603 Bari, Biblioteca Capitolare 1 313 Bayeux, Bibliotheque Municipale 121 42 Bedero di Val Travaglia, San Vittorc B 540 Benevento, Archivio Capitolare 19-20 589 Bencvcnto, Archivio Capitolare 21 303, 337-8, 589 Benevento, Archivio Capitolare 22 589 Benevento, Archivio Capitolare 29 428-9 Benevento, Archivio Capitolare 33 295, 319-20, 352, 551,589-90
Benevento, Archivio Capitolare 34 122, 126-8, 150,
233,241,295,313,352,589 Benevento, Archivio Capitolare 35 233 Benevento, Archivio Capitolare 38 161, 549, 551-2, 589 Benevento, Archivio Capitolare 39 589 Benevento, Archivio Capitolare 40 549, 589 Berkeley, University Library 748 ('Wolffhcim
Antiphonal') 389 Berlin, Staatsbibliothek Preußischer Kulturbesitz,
Mus.ms.40047 303, 570
'Blandinicnsis', see Brussels, Bibliotheque Royale
10127-10144 Bodmer, M., private manuscript collection of, see
Cologny
'Bohn Codex', see Trier, Stadtbibliothek 2254 Bratislava, City Archives, EC.Lad.3, EL. 18 and other
fragments (Missale notation Strigoniense) 295, 319,
607
Brussels, Bibliotheque Royale 2750-65 459 Brussels, Bibliotheque Royale 10127-10144
('Blandiniensis') 185, 290, 296, 298, 509-10 Brussels, Bibliotheque Royale 11.3823 575 Burgos, Monasterio de Las Huelgas 284
Cambrai, Bibliotheque Municipale 61 151, 153-4, 189 Cambrai, Bibliotheque Municipale 75 221, 528, 586 Cambrai, Bibliotheque Municipale 78 180 Cambrai, Bibliotheque Municipale 164 292, 509 'Cambridge Songbook', see Cambridge, University
Library Gg.5.35 Cambridge, Corpus Christi College 391 ('Portiforium of
St Wulstan') 13-16, 312 Cambridge, Corpus Christi College 473 163, 216, 254,
386-7,582
Cambridge, Fitzwilliam Museum 369 54, 322, 575 Cambridge, Magdalene College, F.4.11 320 Cambridge, University Library F7f.l.l7 41, 249, 269 Cambridge, University Library (ig.5.35 ('Cambridge
Songbook') 239 Cambridge, University Library Ii.4.20 322 Cambridge, University Library LI.2.10 405, 424-5 Cambridge, University Library Mm.2.9 584; see also
Antiphonale Sarisburiense Cambridge, University Library, Add. 710 ('Dublin
Tropcr') 313, 584 " Cant us selecti 105, 211 'Carmina Burana', see Munich, Bayerische
Staatsbibliothek, elm 4660a Chartres, Bibliotheque Municipale 47 180, 185-6, 295,
299, 313-14, 334, 347, 349, 351, 355-6, 362-3,
379,383
Cividale, Museo Archelogico Nazionale CI 266 'Codex Albensis', see Graz, Universitätsbibliothek 211 'Codex Calixtinus', see Santiago de Compostela Colognv, Biblioteca Bodmeriana C.74 (olim Phillipps
16069) 150, 166-7, 295, 313, 352, 389, 531, 594 'Compcndiensis', see Paris, Bibliotheque Nationale,
lat. 17436
'Corbiensis',see Paris, Bibliotheque Nationale, lat. 12050
Index of Manuscript and Printed Sources
639
Darmstadt, Hessische Landes- und Hochschul
bibliothek 1946 295 'Dijon Tonarv', see Montpellier, Faculté de Mcdecinc
H.159
Dijon, Bibliothěque Municipal 114 322, 609-10 Donaucschingen, Fürstlich Fürstenbergische Hofbibliothek 882 317 Douai, Bibliothěque Municipale 90 578 Douai, Bibliothěque Municipale 114 578 'Dublin Troper', see Cambridge, University Library, Add.710
Durham, Cathedral Chapter Library B.III. 1 1 303, 337-8
Düsseldorf, Universitätsbibliothek D.l 293, 347-9
Edinburgh, University Library 33 405, 438-9
Edinburgh, University Library 123 278-9
Eichstätt, Bischöfliches Ordinariatsarchiv, pontifical of
Bishop Gondekar II 328 Einsicdcln, Stiftsbibliothek 79 328 Einsiedeln, Stiftsbibliothek 121 198, 295, 347, 355-6,
373-6
Einsiedeln, Stiftsbibliothek 366 310 Erlangen, Universitätsbibliothek 464 225 Erlangen, Universität, Institut für Musikwissenschaft s.n. 603
'Fleurv Plavbook', see Orleans, Bibliothěque Municipale 20 f
Florence, Bibliotcca Magliabccchiana F.3.565 374
Girona, Sant Feliu 20 352 Gniezno, Archiwum Archidiecezjalnc 149 295 (hadttale Romanům 121, 148, 289, 341, 383, 397-8, 401, 411, 627
(iraduale Romanům (Ratisbon, 1898) 397-8,616-17, 624-5
(iraduale Sarisbiiriense (London, British Library,
Add. 12194) 52, 150, 211 Graduate triplex 383, 401 Cradiielde Pans (1754) 619 Graduel mmain (Marseilles, 1872) 171, 397-8 Graduel mmain (Paris, 1874—Reims-Cambrai
edition) 623, 625 Graz, Universitätsbibliothek 211 ('Codex
Albcnsis') 303,607 Graz, Universitätsbibliothek 756 ('Seckau Cantionarium')
250
Graz, Universitätsbibliothek 807 122, 126, 198, 222-3, 295, 390, 421
'Hartkcr Antiphoner', see St Gall, Stiftsbibliothek 390-391
Heiligenkreuz, Stiftsbibliothek 20 310 Hereford, Cathedral Chapter Library P.9.vii 388 Hereford, printed breviary of 1505 320 Hereford, printed missal of 1502 319 Huesca, Catedral 4 603
Ivrea, Biblioteca Capitolare 60 217, 585 Ivrea, Biblioteca Capitolare CVI 303, 337-8
'Jacobus', see Santiago de Compostela
Karlsruhe, Badische Landesbibliothek, Aug.pcrg.60 61,
63,65,390 Klosterneuburg, Stiftsbibliothek 1000 310 Klostcrncuburg, Stiftsbibliothek 1013 71-3 Kyriale sen ordinarium missae 148, 627
Laon, Bibliotheque Municipale 107 363
Laon, Bibliotheque Municipale 1 18 325, 333, 571
Laon, Bibliotheque Municipale 239 295, 299, 347, 349-
50, 355-6, 358-9, 362-3, 373-9, 383, 385, 411, 585 Laon, Bibliotheque Municipale 263 41, 180,210-12,
234-6, 238, 333-4 'Las Huelgas manuscript', see Burgos, Monasterio de
Las Huelgas Le Havre, Bibliotheque Municipale 330 319 'Lectionary of Luxeuil', see Paris, Bibliotheque
Nationale, lat. 9427 Leipzig, Universitätsbibliothek Rep.1.93 347-8, 459 Leipzig, Universitätsbibliothek 391 295 Leipzig, Universitätsbibliothek 1492 325 'Leofric Collectar', see London, British Library, Harlev
2961
'Leofric Missal', see Oxford, Bodleian Library, Bodley 579
Leon, Catedral 8 352, 501, 556-7 'Liber ordinum', see Madrid, Real Academia de la Historia 56
'Liber Sancti Jacobi', see Santiago de Compostela,
unnumbered manuscript Liber antiphottarius 626 Liber gradualis 624, 626-7 Liberhymnarius 397, 628 Liber responsorialis 67, 70, 104, 626 Liber usualis 2, 49, 62, 70, 121, 148, 289, 309, 383, 409,
411, 627 Liber vesperalis (Milan) 541 Limoges, Bibliotheque Municipale 2 225, 237-8 London, British Library, Add. 10335 405, 426-7 London, British Library, Add. 12194 292, 301-2, 584;
see also (Iraduale Sarisburiense London, British Library, Add. 17302 405,432-3 London, British Library, Add. 19768 347-8, 405, 416-
17, 441
London, British Library, Add.22514 283 London, British Library, Add.23935 322, 612 London, British Library, Add.29988 405, 430-1, 531, 538
London, British Library, Add.30848 305, 602 London, British Library, Add.30850 303, 305, 328,
337-8, 351, 405, 414-15, 602-3 London, British Library, Add.31384 433 London, British Library, Add.34209 352, 540, 543,
545-6, 548 London, British Library, Add.35285 323 London, British Library, Add.36881 41, 247 London, British Library, Add.49363 322 London, British Library, Arundel 156 405, 420-1 London, British Library, Cotton Caligula A.xiv 217 London, British Library, Egerton 857 405, 410-11
640
Index of Manuscript and Printed Sources
London, British Library, Egerton 2615 41, 270-2 London, British Library, Egerton 3511 405, 428-9 London, British Library, Egerton 3759 174 London, British Library, Harley 110 405, 412-13 London, British Library, Harley 1117 405, 406-7 London, British Library, Harlev 2961 ('Leofric
Collectar') 312 London, British Library, Harley 4951 597 London, British Library, printed book IB.8668 405,
417,440-1
London, British Library, Royal 8.C.xiii 405, 422-3 'Lucca Antiphoner', see Lucca, Biblioteca Capitolare 601 Lucca, Biblioteca Capitolare 490 298, 304, 507, 510 Lucca, Biblioteca Capitolare 601 ('Lucca
Antiphoncr') 73,86,94-5,97, 105-6,303,305-7,
352, 464, 468 'Lytlington missal', see Westminster Abbey-Madrid, Real Academia de la História, Aemil. 18 602 Madrid, Real Academia de la História, Aemil. 45 602 Madrid, Real .Academia de la História, Aemil. 56 ('Liber
ordinum') 558-9 Madrid, Biblioteca Nacionále 288 41, 201-2, 243, 315—
16, 328
Madrid, Biblioteca Nacionále 289 41, 242-3, 245-6, 249 Madrid, Biblioteca Nacionále 19421 41, 191, 208-9,
223,225,230-1,243 Madrid, Biblioteca Nacionále V.20-4 255-60 Madrid, Palacio Real 429 602 Mainz, printed gradual of 1500 417, 440-1 Manchester, John Rylands Library lat. 24 584 Metz, Bibliotheque Municipale 83 609-10 Metz, Bibliotheque Municipale 351 304, 325, 329, 570 Metz, Bibliotheque Municipale 461 609 Michaelbeuern Cart. 1 284 Milan, Biblioteca Capitolare F.2.2 540 Milan, Biblioteca Trivulziana 347 310 Milan, Sant'Ambrogio M.29 353-4 'Missale Aboensc' 319
Missale notaluni Stiigoniense, see Bratislava, City Archives, EC.Lad.3, ĽL.18 and other fragments
Modena, Biblioteca Capitolare O.I.7 163, 184,203,217, 317
Modena, Biblioteca Capitolare O.I.13 351 'Modoetiensis', see Monza, Basilica S. Giovanni CIX Mont-Renaud manuscript (in private collection) 75,
295, 303-4, 321, 325, 333, 337-8, 347-8, 354-5,
571
Montecassino, Archivio della Badia 110 509 Montecassino, Archivio della Badia 145 509 Montecassino, Archivio della Badia 146 509 Montecassino, Archivio della Badia 318 330, 374, 468, 550, 589
Montecassino, Archivio della Badia 420 305 Montecassino, Archivio della Badia 540 589 Montecassino, Archivio della Badia 542 589 Montecassino, Archivio della Badia 546 589 Montpellier, Faculté de Médecine H. 159 ('Dijon Ternary') 47, 110-14, 118-29, 133,201-2,295, 331, 340-4, 347, 355-9, 387-8, 473-4, 532, 536-7, 548, 556-7, 579, 623, 625
Monza, Basilica S. Giovanni CIX ('Modoetiensis') 298, 510-11
Monza, Basilica S. Giovanni C. 12/75 303, 305, 328, 337-8, 585
Monza, Basilica S. Giovanni C. 13/76 585, 587 Monza, Basilica S. Giovanni C. 15/79 305 'Moosburg Gradual', see Munich, Universitätsbibliothek 2° 156
Mount Sinai, monastery of St Catherine, Gr.567 492 Munich, Bayerische Staatsbibliothek, elm 705 584 Munich, Bayerische Staatsbibliothek, elm 4660a
('Carmina Burana') 257, 267, 269 Munich, Bayerische Staatsbibliothek, elm 9543 201, 363 Munich, Bayerische Staatsbibliothek, elm 14272 328,
470
Munich, Bayerische Staatsbibliothek, elm 14322 315 Munich, Bayerische Staatsbibliothek, elm 14523 471 Munich, Bayerische Staatsbibliothek, elm 14843 586 Munich, Bayerische Staatsbibliothek, elm 14965b 396, 468-9
Munich, Bayerische Staatsbibliothek, elm 18937 471 Munich, Universitätsbibliothek 2° 156 ('Moosburg Gradual') 55, 139, 160, 195, 250
Naples, Biblioteca Nazionalc YLE.ll 433 Naples, Biblioteca Nazionale VI.G.34 589, 591-3 Naples, Biblioteca Nazionale VI.G.38 595 Naples, Biblioteca Nazionale VIII.D.14 325 Norwich, Castle Museum 158.926.4e 319
Offertoriale triplex 121
Orleans, Bibliotheque Municipale 129 176
Orleans, Bibliotheque Municipale 201 ('Flcurv
Playbook') 251,257,263,265-6,269-71 Oxford, Bodleian Library, Auct. F.4.26 304 Oxford, Bodleian Library, Bodlev 579 ('Leofric Missal')
293,319
Oxford, Bodleian Library, Bodlev 775 77, 174-5, 216,
218-19, 224, 226-7, 254, 315,413, 582 Oxford, Bodleian Library, Canonici liturg.350 405,
418-19
Oxford, Bodleian Library, c Mus. 126 50, 53 Oxford, Bodleian Library, Gough Lit.8 321 Oxford, Bodleian Library, Lat.liturg.d.3 405, 408-9 Oxford, Bodleian Library, Lat.liturg.a.4 405, 436-7 Oxford, Bodleian Library, Laud lat. 95 141, 143-5, 310-11
Oxford, Bodleian Library, Misc.lit.366 305 Oxford, Bodleian Library, Rawl.lit.e. 1* 321 Oxford, Bodleian Library, Rawl.lit.d.3 584 Oxford, University College 148 405, 434-5
Padua, Biblioteca Comunale A.47 217, 353-4
Palermo, Archivio Storico e Diocesano 8 319
Paris, Bibliotheque Mazarine 384 321, 347
Paris, Bibliotheque Nationale, lat. 742 602
Paris, Bibliotheque Nationale, lat. 776 77-8, 80-1, 83-
4, 132-3, 151, 170, 182, 201-2, 326-7, 355-6, 359-
61, 409, 554-5, 597-8 Paris, Bibliotheque Nationale, lat. 778 597-8 Paris, Bibliotheque Nationale, lat. 779 597
Index of Manuscript and Printed Sources
641
Paris, Bibliotheque Nationale, lat. 780 330, 597 Paris, Bibliotheque Nationale, lat. 783 600 Paris, Bibliotheque Nationale, lat. 785 600 Paris, Bibliotheque Nationale, lat. 887 169, 214-15, 597
Paris, Bibliotheque Nationale, lat. 903 103, 122, 133-6, 151, 213, 295, 317, 347, 409, 597-8, 608
Paris, Bibliotheque Nationale, lat. 904 147, 263, 295, 399
Paris, Bibliotheque Nationale, lat. 905 137-8
Paris, Bibliotheque Nationale, lat. 909 305, 316, 597,
599- 601
Paris, Bibliotheque Nationale, lat. 1084 181-3, 204-5,
217, 219, 349, 351, 597-8 Paris, Bibliotheque Nationale, lat. 1085 304, 597 Paris, Bibliotheque Nationale, lat. 1086 597-8 Paris, Bibliotheque Nationale, lat. 1087 575-6 Paris, Bibliotheque Nationale, lat. 1105 319 Paris, Bibliotheque Nationale, lat. 1112 52 Paris, Bibliotheque Nationale, lat. 1118 63, 163, 204-5,
216, 316, 325, 328, 597-8 Paris, Bibliotheque Nationale, lat. 1119 167, 222, 597 Paris, Bibliotheque Nationale, lat. 1120 597 Paris, Bibliotheque Nationale, lat. 1121 224, 226-7,
305, 313, 332,597,601 Paris, Bibliotheque Nationale, lat. 1132 222, 578, 597,
600- 1
Paris, Bibliotheque Nationale, lat. 1133 597 Paris, Bibliotheque Nationale, lat. 1134 232, 597, 600-1
Paris, Bibliotheque Nationale, lat. 1135 597 Paris, Bibliotheque Nationale, lat. 1136 597 Paris, Bibliotheque Nationale, lat. 1137 597, 601 Paris, Bibliotheque Nationale, lat. 1138-1338 597 Paris, Bibliotheque Nationale, lat. 1139 41, 105-8, 243-
4, 248-9, 265-7, 270, 597, 599 Paris, Bibliotheque Nationale, lat. 1154 239-41, 363 Paris, Bibliotheque Nationale, lat. 1240 180, 186, 200,
215, 224, 226-7, 304, 314, 363, 597-8 Paris, Bibliotheque Nationale, lat. 2291 319 Paris, Bibliotheque Nationale, lat. 2400 511 Paris, Bibliotheque Nationale, lat. 3549 41, 246, 248,
597
Paris, Bibliotheque Nationale, lat. 3719 41, 248, 597
Paris, Bibliotheque Nationale, lat. 7193 292
Paris, Bibliotheque Nationale, lat. 7202 394
Paris, Bibliotheque Nationale, lat. 9427 ('Lectionary of
Lruxcui!') 552 Paris, Bibliotheque Nationale, lat. 9432 166 Paris, Bibliotheque Nationale, lat. 9438 600 Paris, Bibliotheque Nationale, lat. 10508 159, 164-5,
224-30,316, 328,580 Paris, Bibliotheque Nationale, lat. 10863 552 Paris, Bibliotheque Nationale, lat. 11947 ('Psalter of
Saint-Germain-des-Pres) 91, 491, 553 Paris, Bibliotheque Nationale, lat. 12044 205, 575 Paris, Bibliotheque Nationale, lat. 12050
('Corbiensis') 298, 319, 325, 333, 571 Paris, Bibliotheque Nationale, lat. 12584 205, 303, 321,
337-8, 575
Paris, Bibliotheque Nationale, lat. 12601 575
Paris, Bibliotheque Nationale, lat. 13159 141, 325, 330,
459, 521, 571 Paris, Bibliotheque Nationale, lat. 16819 264 Paris, Bibliotheque Nationale, lat. 17296 67, 303, 337-
8, 354-5, 357 Paris, Bibliotheque Nationale, lat. 17436
('Compendicnsis', 'Antiphoner of Charles the Bald')
298, 303-4, 321, 337-8, 496, 510, 569 Paris, Bibliotheque Nationale, nouv.acq.lat. 1064 257,
260
Paris, Bibliotheque Nationale, nouv.acq.lat. 1177 597 Paris, Bibliotheque Nationale, nouv.acq.lat. 1235 85, 217,310
Paris, Bibliotheque Nationale, nouv.acq.lat. 1669 589 Paris, Bibliotheque Nationale, nouv.acq.lat. 1871 597 Paris, Bibliotheque Sainte-Genevieve 111 ('Silvanectensis') 298
Parma, Biblioteca Palatina 98 56 Passau, printed antiphoner of 1519 262, 303 Passau, printed gradual of 1511 150, 295 Piacenza, Biblioteca Capitolare 65 59, 61-2, 323, 586-8, 612, 617
Pistoia, Biblioteca Capitolare C. 121 198, 253, 353-4 'Portiforium of St Wulstan', see MS Cambridge, Corpus
Christi College 391 Prague, Archív Metropolitní Kapituly cim.4 606 Prague, University Library VI.C.20 225 Processionale monasticum 32,66,70, 105,626 Provins, Bibliotheque Municipale 12 64, 101-2, 217,
219
'Psalter of Saint-Germain-des-Prěs', see Paris,
Bibliotheque Nationale, lat. 11947 Psalterium monasticum 310, 628
Reims, Bibliotheque Municipalc 224 57 'Rhcnaugiensis', see Zürich, Zentralbibliothek, Rheinau 30
'Rollington Processional' 624 Rome, Biblioteca Angelica 123 180, 203, 295, 313, 351-2
Rome, Biblioteca Apostolica Vaticana, Barb.lat. 603 589 Rome, Biblioteca Apostolica Vaticana, Ottob.lat. 313 319
Rome, Biblioteca Apostolica Vaticana, Ottob.lat. 576 589
Rome, Biblioteca Apostolica Vaticana, Pal.lat. 235 365, 384
Rome, Biblioteca Apostolica Vaticana, Pal.lat. 1346 326 Rome, Biblioteca Apostolica Vaticana, Reg.lat. 316 292 Rome, Biblioteca Apostolica Vaticana, Rossi 76 585 Rome, Biblioteca Apostolica Vaticana, Rossi 205 310 Rome, Biblioteca Apostolica Vaticana, San Pictro
B.79 166,531,535,569 Rome, Biblioteca Apostolica Vaticana, San Pietro
F.22 531
Rome, Biblioteca Apostolica Vaticana, Urb.lat. 602 589, 592
Rome, Biblioteca Apostolica Vaticana, Vat.lat. 4770 351 Rome, Biblioteca Apostolica Vaticana, Vat.lat. 5319
150-1, 166, 295, 300, 313, 353-4, 531-2, 534, 536-8,
548, 556-7
642
Index of Manuscript and Printed Sources
Rome, Biblioteca Apostolica Vaticana, Vat.lat. 6082 589 Rome, Biblioteca Apostolica Vaticana, Vat.lat. 10644 507
Rome, Biblioteca Apostolica Vaticana, Vat.lat.
10673 295, 352-4, 551, 589 Rome, Biblioteca Casanatense 54 570, 585 Rome, Biblioteca Casanatense 1574 310 Rome, Biblioteca Casanatense 1741 175, 219, 313, 316-
17, 352-3, 417 Rome, Biblioteca Casanatense 1907 322 Rome, Santa Sabina XIV.lit. 1 277-8,612 Rouen, Bibliotheque Municipals 245 (A. 190) 579 Rouen, Bibliotheque Municipalc 248 (A.339) 580 Rouen, Bibliotheque Municipalc 250 (A.233) 398 Rouen, Bibliotheque Municipalc 277 (Y.50) 335 Rouen, Bibliotheque Municipalc 310 (A.293) 56
Saint-Die, Bibliotheque Municipalc 42 326
St Gall, Stiftsbibliothek 339 295, 301, 347, 373, 375-6,
383, 572, 626 St Gall, Stiftsbibliothek 349 505 St Gall, Stiftsbibliothek 359 295-6, 299, 347, 358-63,
373, 375-7, 382-3, 571-3, 623-4 St Gall, Stiftsbibliothek 360 146 St Gall, Stiftsbibliothek 376 216, 572 St Gall, Stiftsbibliothek 378 572 St Gall, Stiftsbibliothek 380 387, 572 St Gall, Stiftsbibliothek 381 146, 163, 175, 197, 572,
586
St Gall, Stiftsbibliothek 382 146, 572 St Gall, Stiftsbibliothek 388 334
St Gall, Stiftsbibliothek 390-391 ('Hartker Antiphoner') 27, 70, 75, 303-6, 326, 331, 334, 337-8, 347, 380-1, 397, 506, 521, 569, 572 St Gall, Stiftsbibliothek 484 180, 198, 572, 586 St Gall, Stiftsbibliothek 546 139, 150, 156, 164-5, 168, 177,572
St Gall, Stiftsbibliothek 567 506
St Gall, Stiftsbibhothek 1399 304
St Petersburg, Saltvkov-Shchedrin Public Library
O.y.1.6 122, 127, 295, 612 Saint-Quentin, Bibliotheque Municipalc 86 267 Saint-Victor-sur-Rhins, breviary 575 Salisbury, Cathedral Chapter Library 149 583 Salisbury, printed breviary of 1531 321 Santiago de Compostcla, unnumbered manuscript
('Jacobus', 'Liber Sancti Jacobi' or 'Codex
Calixtinus') 39-41, 249, 276, 576, 603 'Scckau Cantionarium', see Graz, Universitätsbibliothek
756
Sens, Bibliotheque Municipalc 6 40 Sens, Bibliotheque Municipalc 46 41 'Silvanectensis', see Paris, Bibliotheque
Sainte-Genevieve 111 Soissons, Bibliotheque Municipalc 85 614-1 5
Toledo, Archivo Capitular 33.24 433
Toledo, Archivo Capitular 35.7 352
Toledo, Archivo Capitular 35.10 602
Toledo, Archivo Capitular 44.1 305
Toledo, Archivo Capitular 44.2 596, 603
Toledo, Archivo Capitular 44.3 603
Toledo, Archivo Capitular 48.14 603
Tours, Bibliotheque Municipalc 927 257, 267-9, 271
Trier, Stadtbibliothek 2254 ('Bohn Codex') 122
Turin, Biblioteca Nationale Lniversitaria F.I.4 250
Udinc, Biblioteca Arcivcscovile 234 261, 585 Udine, Biblioteca Arcivcscovile 8.2 585 Utrecht, Bibliotheek der Rijksunivcrsiteit 406 328 Utrecht, Bibliotheek der Rijksunivcrsiteit 408 328
Valenciennes, Bibliotheque Municipalc 121 578
' V arese-Ercdi Bianchi manuscript', privately owned 540
Yariaepieces 105, 211, 626
Yästcräs, printed gradual of c. 1513 295, 605
Vercelli, Biblioteca Capitolare 56 585
Vercelli, Biblioteca Capitolare 186 585
Verona, Biblioteca Capitolare LXXXV (80) 291
Verona, Biblioteca Capitolare LXXXIX (84) ('Verona
Orational') 312, 559 Verona, Biblioteca Capitolare XCVIII 303, 337-8 Verona, Biblioteca Capitolare CIX (102) 310 'Verona Orational', see Verona, Biblioteca Capitolare 89 Vich, Museo Episcopal 105 603 Vich, Museo Episcopal 106 603 Vich, Museo Episcopal 117 603
Vienna, Österreichische Nationalbibliothck 1609 314 Vienna, Österreichische Nationalbibliothek 1836 330 Vienna, Österreichische Nationalbibliothckser.nov.2700 295, 303
Westminster Abbey, 'Lytlington missal' 319 Wolfcnbiittel, I Icrzog-August-Bibliothek, Weissenburg
76 294, 297, 492, 553 Wolfcnbiittel, 1 lerzog-August-Bibliothek, 79 Gud. hit. 313
'Wolffheim Antiphonal', see Berkeley, University Library 748
Worcester, Chapter Library F.160 ('Worcester Antiphoncr') 68, 85-7, 107, 145-6, 201, 218-19, 303, 310, 317, 323, 354-5, 357, 407, 413, 464
York, printed breviary of 1476 321
Zürich, Zentralbibliothek, Rheinau 28 303, 337-8 Zürich, Zentralbibliothek, Rheinau 30 ('Rhenaugiensis') 298,319
Zürich, Zcntralbibliothek, Rheinau 83 310
INDEX OF NAMES AND TERMS
Citations in bold indicate a musical example.
Aachen 87, 292, 349, 517, 520, 530, 569, 572; Council of (798) 169; Council of (816) 364; Council of (817) 364,570
Abate, G. 594
Abelard, Peter 194, 241, 61 1
Abingdon 581
Abrahamsen, E. 605
Ada sanctorum 313
Adalbero, bishop of Augsburg 458
Adalbero, bishop of Metz 571
Adam de la Halle 247
Adam of Saint-Victor 194, 599
Adhémarof Chabannes 511, 562, 599
.\dniomttogeneralis 364, 516
Adniont 578
Adoration of the Cross 36-7, 265, 525, 527-8
Advent 7, 11-12, 70, 86, 92
affinity, modal 455, 466 (similitude), 468, 473-5
Afflighem 471
agape (love-feast) 486
Agatho, pope 505, 507
Agdc, Council of (506) 55
agenda 324
Agnus Dei 8, 19,24,38, 148-50, 163, 165-8, 183,499;
invocation in litanv 51; tropes 196, 213, 220, 223, 225,
231-3,519 Agobard of Lyon 528, 570, 576 Agustoni, L. '384, 399, 573 Albarosa, N. 382 Albers, B. 579
Albert Y of Austria, duke 615 Albi 201
Alcuin 292, 297, 309, 366, 447, 456, 507, 518, 520, 522, 569, 581
Alexander II, pope 558, 582, 594 Alexander IV, pope 615
Alexandria 365, 480; see also Coptic Church Alfonzo, P. 88
Alfred (Aluredus, Anierus) 595
Aliamusica (Anon.) 328, 394, 448, 451-2, 461-2, 471, 473
Aliae regulae (Guido of Arezzo) 466, 468
alleluia 6,23,29,46, 127-8, 130-9, 172, 181, 185-9, 273, 294, 518-19, 525, 528-9, 561, 563, 578, 582, 600-2; early history 496, 500-5; Old Roman 531, 538-9; Milanese 541-2, 544, 546; Old Beneventan 550; Old Spanish 560; Marian 136-9; prosulas 201-4, 296, 298-9, 314-17, 330-1, 336, 363, 382, 561
allcluiatici (antiphons, Old Spanish) 493, 496
Allworth, C. 317
Almeida, F. de 603
Alvpius 386, 392, 450
Amalariusof Metz 5-6, 70, 73-4, 77, 82, 185, 200, 251, 290, 296, 304, 345, 490, 496, 506-7, 511, 518, 528,
538, 564, 569-70
Ambrose, St 140, 142, 311,313, 363, 488-90, 492-3,
495- 6,504,511,540,611 Ambrosian chant, see Milanese chant Ambrosiani (hvmns) 142-3, 281, 284, 489, 611 Amelli, A. M.'550
Amiens 166 Anchin 578 ancus 343, 345 Andersen, M. G. 605 Anderson, W. 392 Anderson, G. A. 284 Andover, R. 549
Andrieu, M. 185, 289-91, 324, 496, 498, 500, 505, 511,
516, 538 Ancno, F. 616, 626 Angerer, J. F. 325, 615 .Angers 586; abbey of Saint-Aubin 517 Angilbert, abbot of Saint-Riquier 522, 571 Angilramnus of Metz 5 11 Angles, II. 205, 284, 339, 602 Anonymous I (GS i) 394 Anonymous II (GS i) 394 Anonymous IV (CS i) 476 Anselin, St 582
Anstrannus, bishop of Verdun 522
Antioeh 480-2, 489-90, 495, 498
Antioch, Second Synod of (272) 486
antiphon 25-30, 35, 37, 39, 46, 87-99, 102, 109, 114-15, 117, 273-9, 305-7, 309, 324, 326, 329-35, 380-2, 455-6; early history 489-94, 496, 513; fraction 316-17, 498-9; Old Roman 53 1, 533-5, 539; Milanese 541; Marian 30, 46, 100, 104-8; processional 34, 46, 100-4, 318-19, 330, 496, 553-5 ; votive 30, 105
antiphona: ad accedentes (Old Spanish) 499; ad cruccm confractioncm (Old Spanish) 499; ad cruccm (Milanese) 494, 541; ad pacem (Old Spanish) 499; ad praelcgcndum (Gallican, Old Spanish)
496- 7; ante evangelium 98, 317, 554, (Gallican, Milanese) 500-1; in choro (Milanese) 494,
541; post evangelium (Milanese) 500-1, 542
antiphoner 61, 288-9, 303-1 1, 314, 321-3, 328, 336-7, 388-9, 414-15, 430-1, 506-9, 511, 516, 521, 531,
539, 571-2, 596-7, 616, 618
Apel, W. 69, 75, 77, 80-83, 89, 123, 131-3, 194, 286, 377, 502, 529, 531, 540
644
Index of Names and Terms
Apostles' Creed, see Credo in Deum patrem
omnipotentem Apostolic Constitutions 495, 498 Apostolic Tradition (Hippolytus) 486 apostrophe 361; see also strophici Apt 491,598
Aquileia 261, 263, 419, 482, 530, 585 Archer, P. 9 Arezzo 328, 466, 626 Aribo 384, 471-2, 511 Aristarchus 365 Aristides Quintilianus 443 Aristophanes of Byzantium 354 Arit, W. 39-43, 213-14, 221, 234, 237-8, 242, 246, 248-50, 572
Armenian church 481; ekphonetic notation 367; rite
and chant 480, 502 Arnalt de Munt 603 Arnese, R. 339, 591, 595 Arno, bishop of Salzburg 292 Arnoux, G. 444
Arnulf, precentor of Chartres cathedral 580
Arras, and monastery of Saint-Vaast 211, 221, 528, 586
Ascension Dav 9-11, 13
Ash Wednesday 9-11, 13, 31-2, 44
Ashworth, H. 297, 507, 510
Askctorp, B. 605
Asti 351
Atchley, E. G. C. F. 290
Athanasian Creed, see Quicunque vult
Athanasius, bishop of Alexandria 495
Atkinson, C. M. 5, 62-3, 149, 162, 165-6, 168, 223,
460, 528 Aubry, P. 172, 194 Auch' 316, 325, 328, 597 Auda, A. 273
Augsburg 261, 263, 396, 585; monastery of SS Ulrich and Afra 578
Augustine of Hippo, St 26, 140, 282, 313, 443, 488, 490,
492, 495-6, 499, 504, 522 Augustine of Canterbury 102, 297, 483, 506, 514, 628 Augustinián Canons Regular 614-15; Hermits or
Friars 615 Aurelian of Aries 141, 488
Aurelian of Reome 63, 332, 358, 364, 370, 448, 452,
454, 456-9, 462, 521-3, 569 Aurillac 351
Auxerre, and abbev of Saint-Germain 448 Avery, M. 590
Babb, W. 326, 357, 364, 427, 448, 466, 472 baculum 40
Bailey, T. 30,32,58,62-3,90, 100-1, 104, 130, 188-9, 317, 319, 325, 329, 332, 384-5, 452-3, 495, 501-2, 525, 527, 541, 544, 546, 549
Baldovin, J. F. 480
bal lata 250
Baltzcr, R. 476
Bamberg 338-9
Bannister, H. M. 346
baptism 33, 38, 42, 44, 51, 297, 324
Baralli, R. 385
Bardcsanes (Bar-Daisan) of Antioch 485-6, 489 Barezzani, M. T. R. 351 Bari 190,354 Barnwell 584
Baroffio, B. 121, 123, 125, 519, 540-1, 554, 556, 585
Batiffol, P. 320
Baudot, J. 324
Bäumer, S. 320, 487, 618
Baumstark, A. 478, 480, 502, 525
Bautier, R.-H. 326, 387, 470
Bautier-Regnier, A.-M. 345
Bayart, P. 273
Baveux 42
Beare, W. 279, 281, 285
Beauvais 39-43, 169, 233-5, 238, 243, 248-50, 257, 259
Bee 577, 579, 582
Becker, G. H. 296, 339, 613-14
Bede 10, 102, 290, 297, 407, 456, 505-7, 581, 629
Beleth.J. 40
Bcnedicamus Domino 24,27-30, 148, 150, 196,213-15, 238, 242
Bcnedicamus songs 41, 108, 193,238-9,242-3,316
Benedicite, canticle 64
Benedict Biscop 509, 581
Benedict of Aniane 292, 298-9, 311, 570, 598
Benedict of Nursia, St 178, 311, 336, 407, 590
Benedictine monasticism and liturgical practice 27, 488,
491-4, 517, 569, 578, 590, 615; see also Rule of St
Benedict Bcnedictio cerei 38
benediction, by bishop after Paternoster 324 Benediction with the Blessed Sacrament 24 benedictional 324
benedictiones (Gallican, Old Spanish) 493, 497-8, 500-2
Benedictus Dominus Deus Israel, canticle 22, 28, 34, 45; antiphons for 88, 96, 98, 105, 305-7, 333; in Gallican and Old Spanish mass, see Benedictiones; tones for 58, 64, 323
Benedictus es Domine, canticle/hymn 23, 58, 68
Benedictus, part of Sanctus 24
Benevento, Beneventan chant (inch Old
Beneventan) 150, 181, 204, 217, 249, 338-9, 354, 391, 429, 478, 483, 524-7, 529-30, 542, 546-7, 549-52, 561-2, 564, 573, 586, 589-93
Bcnoit-Castelli, G. 624
Benton, J. F. 194
Bernard, abbot of Saint-Jean-dc-Reöine 456 Bernard, archbishop of Toledo 558 Bernard of Clairvaux, St 609-11 Bernard, M. 264-5, 273, 275-6, 339 Bernclinus (Pseudo-) 472-3 Bernhard, M. 458, 466
Bernoof Reichenau 326, 328, 330, 471-5, 523 Bernold of Constance 604, 607 Beroldus 540
Berry, M. 399, 401, 609, 611, 626 Berthold, monk of Reichenau 474 Bcscond, A. 624, 627 Bewerunge, H. 286
Index of Names and Terms
645
Bmford-Walsh, H. 47 Birkncr, G. 607
Bischoff, B. 269, 363, 366-7, 471-2
Bishop, E. 34
Bishop, W. C. 493, 557
bistropha 343, 345
bivirga 343, 345
Bjork, 1). 149-50, 152, 211-12, 252 Björkvall, G. 201, 204, 598 Blanchard, P. 501
Blessing of the Candle 38,51, 100, 102; of the Hoiv
Oils 35; of the New Fire 38 Bloch, II. 591 Blume, C. 181 Bobbio 250, 351, 419, 585
Boc, J. 150-1, 156-8, 162, 166, 196, 212, 531, 585, 589-90
Boethius 239, 328, 386, 392, 394, 442-6, 449-50, 456,
460-3, 470, 475, 523 Bogaerts, C. C. 623 Bologna 351
Bomm, U. 117, 124, 127, 473 Bonastre, F. 205, 603
Boniface, St (Wynfrith) 294, 515-17, 581, 590
Boniface II, pope 505-6
Bonniwell, W. R. 611
Boor, H. de 252, 261
Borella, P. 494,540
Borsai, I. 481
Bosse, D. 148, 156-61, 209, 315, 336, 419, 452, 606 Bossuet, J. B. 618 Botte, B. 7 Bouver, L. 5
Bower, C. M. 187, 438, 448, 458-9, 470
Boy Bishop 40
Boyce, J.J. 615
Bradshaw, P. 487-9, 491
Braganca, J. 603
Bragard, R. 201
Branner, R. 335, 391
Brescia 351,419
breviary 287-9, 305, 307-9, 311, 320-2, 336, 595, 615, 618
Bridget, St 605, 615 Briggs, II. B. 405,407,41 1,433 Brigittinc Order of nuns 605, 615 Bristol 614
Brockctt, C. W. 257, 558, 596
Bronarski, L. 471-2
Brou, L. 188, 525-6, 528, 557, 559-60
Brovclli, F. 618
Brown, V. 590, 607
Brown, L. E. G. 320
Browne, A. C. 386-7
Brunhölzl, F. 106
Brunner, L. 183, 185, 399, 585, 588-9, 591 Bruno, bishop of Toul (Pope Leo IX) 275 Bruno of Cologne, St 613 Bryant, D. 530 Buchner, M. 516 BujicB. 607
Bulst, N. 578
Burghard of Eichstätt 515
Burgos 603
Burgundv 188
burial 42, 44, 324
Bursfeld 615
Buzga, J. 250
Byzantium: Byzantine rite and chant 31,58, 102, 149, ' 169, 188, 220, 332, 454, 456, 459, 479, 482-3, 495, 498, 500, 502, 521, 525-30, 554, 561, 569; notation 361-2, 367-70, 395; sec also Constantinople
Cabaniss, A. 569
Cabrol, F. 289, 557, 626
Caen, priory of St Stephen 582
Caesarius of Aries 141, 294, 311, 488
Cagin, P. 511, 626
Caiazzo 589
Caleidius 443-4
Caldwell, j. A. 444
Callcwaert, C. 511-13
Camaldoli, Camaldolese 578, 603
Cambridge 584
Campbell, t. P. 251
Canal, J. M. 609
canon (of mass) 24,287,291,320
Canosa 589
cantatorium 77, 295-303, 314, 382
Canterbury 102, 581-2; abbey of St Augustine 239;
cathedral priory of Christ Church 406, 582, 584;
abbey of SS Peter and Paul 509 cantica of Holy Saturday 36, 38, 82-3, 561; sec also
tract canticle 21,28-9
cantio 108, 238-50, 606; see Latin liturgical song
cantionale 250
canto fermo, canto fratto 621
cantus (Milanese) 500-2, 541-2
Cantus sororum (Brigittinc office) 615
( \intns quem ("istercicusis ordinis ecclesiae 610
Capelle, B. 497
capitulary 293-4
capitulum, see chapter
Capua 294
Gardine, E. 109, 341, 358, 370, 373, 379-80, 382-3,
400,573 Carinola 595 Carmelite Order 615 Carthage 483, 486
Carthusian Order 433, 570, 603, 613-14 Cassiodorus 443, 456, 460 Castle Acre, Cluniac priory 318-19 Castro, E. 602
Casus Saudi Calli (Ekkehard 1V) 199, 572
Catania 209, 223-5
catechumens 23, 33
Cattanco, E. 540, 550
Cavallo, G. 590
Cccilian movement 623
Celestine I, pope 495-6, 498
Celestine V, pope 595
646
Index of Names and Terms
Cclcstinc Order 595
Celtic church and chant 483-4, 488, 493, 553; see also
Irish church Censořinus 443
cento, ccntonization 74-5, 224, 496 cephalicus 343, 345 Chadwick, H. 444 Chaillev.J. 40,207,461,596 Chalcedon, Council of (451) 480-1 Chambers, E. K. 43-4 chant figure 399,620 chant sur le livre 620
chapter (short lesson) 27-30, 39, 54, 85, 309, 311, 313
Charlemagne, emperor 17, 292-3, 364, 371, 447, 456, 458, 479, 509, 511, 514, 516-18, 520-3, 530, 540, 550, 552, 558, 564, 569, 571-2, 576, 581, 590, 596
Charles the Bald, emperor 298, 448, 479, 518, 522
Charles the Fat, emperor 185, 530
Charles Borromeo, St 541
Charlesworth, K. 485
Chartier, Y. 448,458
Chartrcs 64, 150, 217, 470, 564, 580
Chartreuse, La Grande (Charterhouse) 613
Chavanon, J. 562
Chavasse, A. 293-4, 298, 512
Chclles 292
Cheroubikon 500
Chevalier, U. 39, 41, 43, 249
Chibnall, M. 580
Chichester 435
Choissclet, D. 322, 609
Chomton, L. 578
Choron, A.-E. 622
Christmas Day 8, 11-12, 27; Christmas Eve 12, 56;
Christmas season 8,39-42, 191 Chrodegang, bishop of Metz 299, 516, 571 chromatic notes 68, 92, 123-4, 145, 361, 394, 451, 464,
474
Chmnicon (Adhémar of Chabannes) 562
Cinq messes en plait/chant (Henry l)u Mont) 171
Circumcision, feast and office 12, 39-43, 96, 171, 233-
6, 243, 249, 270 Cisneros, Jimenez de, cardinal 558-9 Cistercian Order, Citeaux 310-11, 322, 466, 474, 476,
575, 578, 603, 608-13 Cividale 266, 482 Claire, J. 85,87-8,90-1,541 clamor (Old Spanish) 500-3, 560 Clark, J. M. 571 Claudianus of Viennc 492 climacus 341-2, 344, 348-9, 352-3, 390 clivis 342, 344, 348-9, 352-3, 390 Cloveshoc, Council of (747) 297, 364, 516 Clovis, king of the Franks 552 Clunv 19-20, 54, 67, 321-2, 347, 387, 558, 563, 571,
574-80, 582, 600-3, 619 Cochcril, M. 609 Codv, A. 460 Coffin, C. 619 Colette, M.-N. 388 Colgrave, B. 506
collect 6, 23, 27, 30, 36, 49, 291, 311-12 eolleetar 288, 308, 310-12, 322-4 Collins, A. J. 320, 605, 615 Columbanus, St 488 Combe, P. 624, 626
comes (liber eomitis, liber commicus) 293 Comes of Murbach 294
Comes (Capitulary, Leetionary) of YVurzburg 294, 502 Connnenioiatio brevis de tonis el psalmis nuuiulandis
(Anon.) 62-3, 328-9, 384-6, 393, 448, 452-3, 456,
459
Commenlarius in Micmlogum (Anon.) 384, 472 communion (chant) 19, 24, 38, 46, 109, 116-23, 326,
330-1, 456, 518; Old Beneventan 550-2; Old
Roman 531-3; early history 497-9,503,51 1-13;
tones for psalm verses 58, 64-5, 330, 455, 457;
tropes 196, 215-16; see also transitorium, Milanese Como 353, 585
Compicgne 448; abbey of Saint-Corneille 264-5,298, 337
Compline 16, 19-21,29-30,35-7,39,85, 104-5, 142,
249, 310 computus (compotus) 9
conductus 39-43, 108, 238-50, 273, 576; see also song,
Latin liturgical Confessions (St Augustine) 490, 495 confirmation 42, 44, 324 confractoriurn (Milanese) 499 Connolly, T. 58, 62, 66, 109, 531 Conrad II, emperor 189 Constance 395-6
Constantine I, emperor 480-1, 484 Constantine, C. G. 574
Constantinople 480-2, 489, 506, 569; see also
Byzantium Contreni, J. 448 Conversion of St Paul, play 269 Conybeare, F. C. 486 Coptic church and chant 385, 481 Corbie 298, 321, 325, 333, 338, 347, 354-5, 413, 507,
520, 570-1, 575, 581-3 Corbin, S. 269, 340, 348, 363, 387-8, 505, 602-3 Corinth 486 Cornelimunster 598 Cornford, F. M. 444 coronation 42, 44, 324 Corpus Troporum 196 Corpus Christi 25, 192 coupure ncumatique 377
Coussemaker, E. de 251, 260, 263-7, 269, 365, 374, 610
Coutances 269
Cowdrey, H. E.J. 574, 591
Credo in Deum patrem omnipotentem (Apostles'
Creed) 25,30, 168; farse 234-5 Credo in unum Deum (Nicenc Creed) 8,23, 148-50,
168-71, 519; farse 168-9, 171, 196, 233-4, 235-6,
621
Crichton, J. D. 5
Crocker, R. 88, 150, 152, 156, 161, 165, 168, 172, 178, 180, 185-7, 196, 211, 238, 279, 282, 284, 286, 386, 392, 443, 572, 587-8, 596, 598
Index of Names and Terms
647
Cross, Exaltation of the Holv 11 — 12; Invention of the
Holy 192 Crovvland 582 Cserba, S. 326 Cuontz, J. 156, 572 Curran, M. 484, 553
cursus (secular or monastic form of office) 21, 25-30,
38-9, 201, 309, 313, 336-7, 580, 614 cursus (prose rhythm) 285-6 customary 325, 398 Cutter, P. 66, 69-70, 531, 539 Cyril, St 482
Cyril, bishop of Jerusalem 498 D'Angcrs, (). 596
Dagobert, king of the Franks, St 554
Dalton, J. N. 7
Damilano, P. 250
dance, liturgical 40, 207, 225
Daniel, Play of, see Lucius Danielis
Danjou, J.-L.-F. 623
Darragon, B. 91
dasian notation 386, 393-4, 451-3
David, king of Israel 511, 522
David (Charlemagne) 522
David, L. 580,583
Davidson, C. 251
Davril, A. 252, 574-5
De accentibus (Anon.) 366
De computo (Hrabanus Maurus) 10
De eomivio sizeprandio atque cenis monachonim 505
De die natali (Censorinus) 443
De divisione naturae (Johannes Scotus Eriugena) 448 De ecclesiasticis disciplinis (Regino of Prüm) 458 De ecclesiasticis officiis (Amalarius of Metz) 5 De ecclesiasticis officiis (Isidore of Seville) 502, 558 De harmonica institutione (Hucbald of Saint-Amand) 364, 448
/)e illustns sctiptoribus ecclesiasticis (Isidore of
Seville) 505 De institutione musica (Bocthius) 444, 470 De La Fage Anonymous 476
De mensurabili musica ( Johannes dc Garlandia) 476
De modis musicis (Anon.) 63
De musica (Augustine) 443, 522
De nuptiis Philologiae el Mercurii (Martianus
Capclla) 443 De octo tonis (Anon.) 456, 522 De plana musica (Johannes de Garlandia) 476 De rebus ecclesiasticis (Walahfrid Strabo) 553 De ritibus ecclesiasticis (Landulphus) 540 De litis illusttibus (Ildefonsus of Toledo) 505 Deanesly, M. 506 Dechevrens, A. 379-80 Decrela (Lanfranc) 604
Dedication (Consecration) of a Church 42, 45, 51, 324,
424-5 Delalande, D. 609-12 Delaporte, Y. 274 Deleglise, F. 273
Deposition of the Cross 35, 37, 251
Desclec, publishing firm 397, 627 Deshusses,J. 291-2
Desiderius, abbot of Montecassino, Pope Victor III 591 Dewick, E. S. 13, 312 dialect, German chant 573
Dialogus de musica (Anon.) 328, 387, 395, 454, 463-7,
470,473-4, 521,575,579 Dialogus ecclesiasticus itistitutionis (Egbert of
York) 297, 507 Dickinson, F. H. 172, 319
differentia (diffinitio) 58, 61, 89, 92, 105, 108, 326-9, 333,335
Digne, ecclesiastical commission 171, 397, 625
Dijk, S. van 321, 538, 560-1, 594, 612
Dijon, monastery of Saint-Benigne 201, 331, 355, 387,
577-9, 582, 609-10 Dirige 19 Disentis 334 diurnal 305,312,432-3 Dix, G. 487,494 Doble, G. 7 Dobson, E.J. 266 Dobszay, L. 88,90,607
Doctiina scientiae musicae (Elias Salomon) 595
Dold, A. 294, 297, 304, 492, 553
Dominic, St 274, 276-8
Dominican Order 320, 322, 391, 608, 611-13
Donato, G. 326
Donovan, R. B. 251, 602
Dören, R. van 562, 571-2
Downpatrick 582
Doxology, see Gloria
drama, liturgical 46, 250-73
Drinkwelder, (). 172
Dronke, P. 204,240
Droste, D. 580, 584
DrumblJ. 264,527
Du Mont, 11. 171, 620
Ouarte (Edward), king of Portugal 603
Dubois, J. 313
Duchesne, L. 494, 504-5
Duchez, E. 373
Dufav, G. 36
Du four, J. 623
Dumas, A. 292
Dunstan, St 581
Durham 582
Duval, E. 623
Dyer, J. 63, 121,308,399,442,487,490-1,495-6, 499-500, 531, 534, 539, 560
Eadberht, king of Northumbria 297, 507
Easter Sunday 8-10, 13, 19, 26, 32-3, 39; Easter Eve, Easter Vigil, see Holy Saturday; Easter season 8-9, 12-13,32, 148; calculation of date 9-10
Easter Play 267-9
Ebel, B. 310
Eberhard, canon of Augsburg 585 Ebrulphus (Evroult), St 580 echemata 332-3 cchos, echoi 454, 459-60
648
Index of Names and Tenns
Eddi (Stephen) 507
Edgar, king of England 564, 581
Edward the Confessor, king of England 582
Edwards, K. 40
Edwards, O. T. 273, 584
Eeles, F. C. 580
Egbert, bishop of York 297, 507, 509 Egena (Etheria) 150, 480 Eggen, E. 172, 605 Ehrensberger, H. 289 Eichstätt 205, 328
Einhard, biographer of Charlemagne 516 Einsiedcln 310, 328, 334, 373 Eisenhofer, L. 1 Eizenhöfer, L. 281 Ekberg, G. 369 Ekenbcrg, A. 1
Ekkehard IV of St Gall 199, 215, 572
ekphonetic notation 364-5, 367-9, 481
elevation of the sacrament 24
Elias Salomon 595
Ellinwood, L. 472
Ely 322, 425
Elze, R. 324
Ember Davs 11-13, 23, 68, 77
Emerson, ). 238, 295, 302-3, 313, 319-20, 389, 596,
598-9 Engelberg 334 Engelbert of Admont 476 Engelbrccht, bishop of Passau 471 Engyldeo 363
Ephesus, Council of (431) 481 Ephrem of Edessa 486
Epiphany 10-12, 27, 40-41, 56, 70, 104, 235, 265, 269-
70, 310, 333 epiphonus 343, 345 episema 374-8, 380-3
epistle 6, 23, 54, 247, 293-4, 495-6, 498, 500-2; farse
234, 236-8, 249 Epistola ad Michaelem (Epistola de iqnolo canlii) (Guido
ofArezzo) 466-7,469 Epistola de haimonica institutiane (Rcgino of
Prüm) 458-9 epistolary 287, 293-4, 322 Epstein, M. J. 273 Erbacher, R. 62 Escorial, El 5 cstampic 188 Esztergom (Gran) 607 Ethelwold, bishop of Winchester 581 Ett, C. 620
F.tymolagiae (Isidore of Seville) 490, 558 cuouae (seculorum amen) 333 cvangeliary 287, 293, 322 Evans, j. 574
Evans, P. 196-7, 215, 220, 223, 313, 415, 596 Evans, R. 569 Evesham 582 Exeter 312, 582, 584
Expositio antiquae liturgiaegallicanae 188, 294, 497-8, 500-1,552
Expositio lotius missae (Amalanus of Metz) 5-6 Exultet 38, 49, 51-2, 315, 319, 554
Falconer, K. 223 Fallows, D. 239 Falvy, Z. 273, 607 Farmer, D. H. 7
farse 41, 168-9, 196, 225, 233-7, 249, 592-3; see also trope
Fassler, M. 194, 279, 325, 487
Fazio de Castoldis 437
Feast of Fools 39-43
Fecamp 579, 582
Felder, H. 273, 596
Feilerer, G. 540, 552, 608, 621-3
Fcnlon, I. 369
Ferdinand of Portugal, St 603 feria 12
Fernandez de la Cuesta, I. 339,602 Fcrotin, M. 559
Fcrrctti, P. 69-70, 77, 87, 90, 93, 98-9, 286, 341, 351,
374 Fiala, V. 289 Fickctt, M. van Z. 371
Fieschi, Ottobono, cardinal, Pope Hadrian V 595
Fischer, L. 320, 594
Flaccus (Alcuin) 522
Flavigny 292
Fleming, K. 117
flcuretis 620
Fleury, and monastery of Saint-Benoit-sur-Loirc 269,
581", 583,590 flexa 342,344 Flint, V. 472 Florence 595 Floros, C. 369 Florus of Lvon 570 Floyd, M. 317 Foley, E. 487 Fontaine, G. 618-19 Fontevrault 225, 238 Forlimpopoli 317 Fortescue, A. 1 Franca, U. 88
francigcnae, Milanese mclodiac 188-9, 544, 546 Francis, St 274, 596
Franciscan Order 104, 326, 594-6, 608, 612
Franciscus/Franco of Liege 326, 472
Franks 70, 73-4, 81, 86, 101, 109, 152, 166, 187-8, 289-92, 296, 299, 301, 304, 309, 311, 364, 370, 479, 483, 506, 509-10, 514-25, 527, 531, 541, 546, 549, 552-4, 560-2, 564, 569, 572, 590, 594, 628
Frederick, abbot of Montccassino, Pope Stephen IX 591
Frcising 314
Frenaud, G. 53 1
Frere, W. H. 1,7, 13, 48, 66, 69, 71, 73, 75, 88-93, 115, 180, 216, 293, 303, 312-13, 320, 324, 326, 333, 339, 388, 533, 580, 583
Friedlcin, G. 444
Froger, J. 109, 297-8, 304, 401, 497, 501, 504, 507, 572, 628
Index of Names and Terms
649
Frutolf of Miehclsberg 127, 326, 472 Fruttuaria, abbey of San Benigno 579 Fulbert, bishop of Chartres 471, 580 Fulda 447 Fuller, I). 620
Fuller, S. 238, 248, 452, 476, 596 Fulrad of Saint-Denis 515
Gaeta 310 Caillac 597 Gajard, J. 379, 573
Galhcan liturgy and chant 32, 98, 101-2, 104, 122, 134, 140, 150, 153, 178, 188-91, 220, 275-6, 281, 285, 292, 294, 297, 479, 483, 506, 518-19, 524-5, 546, 559-62, 569, 598; early history of office chants 492; of mass chants 497-502; music, and comparison with Gregorian 552-7
Gambcr, K. 287-8, 291-3, 409, 507
Garand, M. 574
Gastoué, a. 552-4, 627
Gatard.A. 540
Gauticr, L. 146, 173, 196-7,216,220,248,511,572 Gavel, M. H. 279 Gelasius, pope 497 Gelineau, J. 1
Genealogy of Christ 27, 56-8 Gengenbach 578 Gennadius 291, 297, 492 Gennrich, F. 266
George, primus of the Roman schola cantorum 515 Georgian ekphonetic notation 367; rite and chant 480 Gerbert of Aurillac, Pope Sylvester II 470, 594 Gerbert, M. 394, 427, 448,"452, 456, 463, 473 Germanus, abbot of Ramsey 581 Germanus of Paris 188 Gerona 603
Gcvaert, F. 63, 87-94, 297, 380, 507, 533
Ghent, St Peter's abbey on Mont-Blandin 290, 298
Gibson, M. 444
Gillingham, B. 238
Gindcle, C. 370
Gingras, G. 480
Gjerl0w, I,. 308, 580, 604-5
Glastonbury 507, 581, 583
Glecson, P. 611
Gloria in excelsis Deo (Greater Doxology) 6, 8, 19, 22, 24, 35-6, 38, 68, 148-50, 156-62, 166; added melismas 196-7, 199-200, 208-9; earlv historv 488, 492, 497-8; f arse 234; prosulas 208-9; tropes' 161, 196, 220,223-30,519
Gloria patri et ťälio (Lesser Doxologv) 25-30, 33, 70, 73-4,85-7, 109, 141,200-1,333
Gloucester 582
GmclchJ. 388
Gneuss, H. 140-1, 310-11
Gocde, N. de 172, 240-1
Göllncr, T. 56
Gondekar, bishop of Eichstätt 328
Good Friday 11, 13, 32-3, 35-7, 39, 251, 265, 319
Gorzc 571,578
Goschl.J.B. 384,399
gospel 23, 27, 54-8, 88, 247, 293-4, 495-6, 498, 500-4;
farse 234 Gottschalk of Orbais 239 Gottwald, C. 339 Grado 482
gradual (book) 287, 289-90, 293-303, 305, 307, 310, 314-15, 318, 320-3, 326, 328, 335-7, 408-13, 420-1, 438-9, 440-1, 521, 531, 570, 596-7, 616, 618
gradual (gradual rcsponsorv) (chant) 6, 23, 46, 69, 76-82, 122-4, 132, 294, 296-8, 314-17, 330-1, 377-8, 382, 518, 547, 561; early history 494-6, 499-503; Old Roman 531, 533-4, 539, 547; prosula 209; see also psalmellus, Milanese
Gradual Psalms (Pss. 119-33) 19
Gran (Esztcrgom) 607
Grandmont 578
Great Entrance (Byzantine) 500
Greater/Lesser Perfect System (svs/ema teleion) 443-4,
446, 450 Greek Sanctus, see.Agios Greek Creed 168 Greek Gloria, seeDoxa Gregoire, R. 313, 323, 339 Gregorian Congress, Rome 1904 399, 626 Gregory I 'the Great', St, pope 4, 26, 74-5, 151-2, 274,
292, 294, 296-7, 313, 478, 496-7, 502-14, 590, 626,
628-9
Gregory II, pope 510, 512, 515
Gregory V, pope 591, 594
Gregory VII, pope 558, 614
Gregory XVI, pope 624
Gregory, bishop of Tours 309, 488, 552
Grisar.'H. 508
Gros, M. 603
Grospellier, A. 627
Grotefend, H. 7
Gubbio 589
Gueranger, P. 623-4, 628 Guidetti, G. D. 55, 616
Guido of Arezzo 316, 346, 357, 384, 386, 388-9, 395,
426-7, 454, 466-74, 476-7, 521, 523 Guigo (Guigues de Chatel) 613-14 Guisborough 323, 582 Guitmund, monk of Saint-Evroult 580 Gumpcl, K.-W. 609 Gushee, L. A. 442, 456, 522 gutturalis 360 Guy de Chalis 610 Guy de Cherlieu 610 Guvd'Eu 610
Gv, P.-M. 196, 311, 317, 324, 570, 576, 594 Gyug, R. 607
Haapanen, T. 605 Haas, M. 369
Habcrl, F. X. 616, 618, 623-4, 626 Haddan, A. W. 297, 507 Hadrian of Nisida 294
Hadrian I, pope 292, 507, 509, 511,517, 570, 623 Hadrian II, pope 511 Hadrian V, pope 595
650
Index of Names and Tenns
Hain, K. 390
Hallcl psalms (Pss. 113-18) 485-6 Hallinger, K. 571 Halperin, D. 47 Hämel, A. 576 Han, U. 395-6
Handschin, J. 178, 188, 197, 200, 240, 347, 530, 580, 583
Hänggi, A. 325
Hannick, C. 481,484,487,494
Hansen, F. E. 47, 110, 121-2, 126,340,556,578-9
Hanslik, R. 488
Hanssens, J.-M. 73, 77, 185, 200, 251, 296, 345, 490,
506, 511,569-70 Hardison, O. B. 5-6, 33, 251 Harlay, F. de, archbishop of Paris 619 Harmonios, son of Bardcsanes of Antioch 486 harp 581
Harrison, F. LI. 18, 40, 104, 150, 238, 250, 266 Hartker of St Gall 572; see also St Gall, Stiftsbibliothek
390-391 Hartmann of St Gall 146,572 Hartvic, monk of St Emmeram, Regensburg 470 Hartzell, K. D. 580, 582 Hatto, archibishop of Mainz 458 Haug, A. 180 Havmo of Faversham 595 Heard, E. B. 461 Hebrew ekphonetic notation 367 Heiming, (). 487, 540-1 Heinrich, patriarch of Aquileia 263, 585 Heisler, E. 573 Heitz, C. 571 Heiander, S. 605
Helena, mother of Emperor Constantine 480
I Ielisachar, abbot of Saint-Riquicr and Saint-Aubin at
Angers 517,569 Ileloisc 611 1 lendcrson, 1. 392 Henderson, W. 317, 319
I lenricus (I lonorius) of Augsburg ('Augustoduncnsis')
471-2, 604 Henry II, emperor 169, 189, 471 Hcnrv of Chichester, canon of Exeter 584 Hereford 34, 209, 388, 583 Herlinger, J. 476
Hermaniuis Contractus 108, 330, 395-6, 472, 474-6 Hermesdorff, M. 401, 623, 625 Herodian 365 Herzo, A. M. 596
Hesbert, R.-J. 89, 131, 185, 296-8, 304-5, 307-8, 321, 336-7, 415, 510-11, 518, 529, 549, 569-70, 573, 575, 578, 580, 582, 584, 589-90, 594-6, 602-3, 610-11
hexachord 449, 455, 467-9, 474-7
Hilary of Poitiers 488,492
Hildesheim 283, 390
Hildoard, bishop of Cambrai 292
Ililduin, abbot of Saint-Denis 528, 553, 571
Hilev, D. 148, 302, 308, 315, 324, 336, 338, 340, 345, 363, 435, 574, 576, 579-80, 582-3, 591, 598, 618
Hilkiah, high priest of Israel 520
Ilippolytus, Romanpriest 486
Hirsau 471, 571, 578
Hispanic chant, sec Old Spanish chant
historia 273, 307
Historia abbat um (Bede) 509
Historia Fiancorum (Gregory of Tours) 552
I löfler, J. 607
Hofmaim-Brandt, H. 200, 204-5, 315, 603, 606 Hohler, E. C. 39, 41, 274, 283, 471, 576, 584, 603 Holder, S. 322, 574-5 Ilolman, H.-J. 200-1 Holmes, S. 9
Ilolschneider, A. 178, 180, 386, 580, 583 Holtz, L. 443
Holy Saturday (also Easter Eve, Easter Vigil, Paschal Vigil, etc.)' 10-11, 13, 32-3, 36, 38-9, 44, 51, 157, 162, 166, 319, 430-1, 502, 529
Holy Week 11, 32-9, 56, 100, 290, 335-6
Homer (Angilbert) 522
homiliarv 288, 308, 313
homily 26, 313, 495
Honoratus, abbot of Farfa 5 11
I lonorius I, pope 292
I lonorius III, pope 595
I lonorius of Augsburg 471-2, 604
Hope, D. 291
Hoppin, R. II. 273
Hornof, V. 250
Hosanna in cxcelsis (part of Sanctus) 24; prosulas 209,
225, 230-1 Hourlicr, J. 88, 347, 349, 574-5, 611 Hrabanus (Rhabanus) Maurus 10, 447 Hucbald of Saint-Amand 178, 223, 274, 345, 364, 386,
392-3, 447-52, 459-60, 462-4, 473, 475, 522-3 I Iucbcrt, bishop of Meaux 522
Hucke, II. 66, 69-70, 73, 76, 79, 81, 85-6, 88, 90, 93, 109, 116, 121-2, 296, 370, 372, 385, 487, 490-1, 499, 503, 51 1,514, 517, 519, 527, 531, 533, 539-41, 547, 560-2, 628
Iiudovskv, Z. 607
Huesca 603
Hufnagelschrift 390, 421, 439 Hugh, abbot of Cluny 575 Hugh of Saint-Victor 194
Hughes, Andrew 7-8, 16, 21-2, 24-8, 33, 56, 273-4,
287-9, 295, 303, 308, 319-20 Hughes, Anselm 13, 172, 178, 181, 312, 319 Hughes, David 201, 520, 598
Huglo, M. 47,49,55,88,90-1, 104, 116, 140, 150, 152, 158, 168-9, 196-7, 199, 213, 234, 238, 240-1, 269, 287-8, 295, 299, 303-4, 317, 319-21, 323, 325, 328-35, 340-1, 346-7, 350, 352-3, 360, 363, 370, 413, 448, 459-60, 463, 470, 472, 491-2, 497, 511, 517-18, 521-2, 525, 528, 531, 540, 549, 552-4, 557, 569-70, 574-6, 578-80, 585, 589, 594-5, 597, 602, 609, 612-14
Hugo Candidus 558
Humbert of Romans 322, 611-12
Hunt, N. 20, 574
Huschen, II. 458, 594-6, 609, 611, 613-14
Index of Names and 'Venns
llusmann, II. 173, 180-1, 186-7, 192, 194,220, 314,
336, 339, 460, 481, SS7, 560, 599, 603, 605 I hitter, J. 606
hvrrm 22, 25-30, 39, 41, 46, 131, 140-8, 172, 197, 216, 239, 305, 309, 336, 488-90, 492-3, 570, 611; processional 34, 146-8,239,318-19,486
hymnal 288, 305, 308-11, 323, 336, 611
ictus 280-2, 627 Idithun 522-3
Ignatius, bishop of Antioch 490
Ildcfonsus of Toledo 505
Improperia (Reproaches) 36-7, 265, 528
Indc 598
Ingolstadt 395
ingressa (Milanese) 496, 542, 547-9, (Old Bcneventan)
549-50 Innocent III, pope 321, 594 Innocent IV, pope 595
Innocents, Holy 23, 40, 146; liturgical drama 265, 269 Institutes ( John Cassian) 491 Institutiones canonicorum 364
introit 6, 19, 22, 36, 38, 44, 46, 109-16, 111-13, 121-3, 151, 166, 253, 299, 326, 330-1, 372-3, 456, 518, 526, 547-9; added melismas 196-9, 572; earlv history 494, 496-8; Old Roman 531, 547-8; "tones for psalm verses 58, 64-5, 286, 330, 333-4, 455-7; tropes 196, 215- 23, 586-7; see also ingressa, Milanese, Old Bcneventan
invitatory antiphon 25, 46, 88, 99-100, 306-7, 309, 330
invitatory psalm (Venitc) 25, 99; tones for 46, 58, 62, 66-8, 67, 323
Irene, empress 530
Irish church, chant and sources 157, 188,483-4
Irtenkauf, W. 207, 250, 273, 289
Isidore of Seville 366, 443, 490, 492, 502-3, 505, 515,
558-9 Iso of St Gall 186,572 ison singing 526 he missaest 24, 35, 39, 148, 150 ltier, Bernard 596
Justus ut palma group of graduals 77-81, 383, 533-4 hersen, G. 225 Ivrea 338-9
Jacob, A. 325
Jacobite (West Syrian) Church 481 Jacobsthal, G. 124, 473 Jacobus of Liege 201,476 Jakobs, II. 571 James, St 480, 576
Jammers, E. 46, 130, 250, 273-4, 276, 293, 340, 347,
369-70, 380, 501-2, 525-6, 560 Janson,T. 285 J arrow 509 Jausions, P. 624 Jeanneteau, J. 90 Jeffery, P. 492, 495-6 Jerome, St 296, 309, 313, 492, 504 Jerome of Moravia 326 Jerusalem 150,480-1,502,527
Jcsson, R. 540-2, 547 Jewish religious worship 484-5 Johannes de Garlandia 476 Johannes of Avranchcs 345
Johannes Scotus Eriugena ( John Scot) 448, 459, 522
Johannes, bishop of Cambrai 522
John VIII, pope 508
John XIX, pope 388,466
John of Ravenna, abbot of Fecamp 579
John Cassian 491
John Chrysostom, St 482
John 'Cotton' or 'of Afflighem' or of Passau 326, 472-3 John Hymmonides 'the Deacon' (Pope John I) 74, 502
3, 506, 508-9, 517, 562 John, abbot of St Martin's and archicantor of St Peter's
Rome 505, 507, 509 John, bishop of Syracuse 151, 499, 502, 504, 506 John of Damascus, St 460 John the Teuton 611 Johner, D. 46
Jones, Charles 205, 274, 283
Jones, Cheslyn 1-2, 5, 291, 487, 495
Jonsson, R. 183, 196, 274
Jordan of Saxony 611
Joseph II, emperor 622
Josiah, king of Israel 516, 520
jubilus (part of alleluia) 130-7, 181,502,538-9
Julius Caesar 9
Jumicges 56, 186, 579-80, 582
Jungmann, J. A. 5-6, 22, 487, 494, 496, 498
Justin Martyr 486
Justinian, emperor 144, 483, 525
Kahmer, I. 121, 531, 534 Kainzbaucr, X. 76 kalendář 9, 13-16,323,335-6 kauon 526
Kantorowicz, E. H. 591 Kastl 615 Keil, II. 366
Kellv, T. F. 181, 200-1, 204-7, 225, 525, 527, 529,
549-52, 580, 589-91 Kemptcn 141, 310 Kennedy, V. L. 607 Ker, N."339 Kicnle, A. 370 King, A. A. 540, 557-8 King, A. H. 395-6 Kirchrat, R. 620 Klauser.T. 293,482,514,516 Klostcrncuburg 223, 310, 390 Knight, W. F. J. 443 Knowles, D. 19, 580-1 Kohrs, K. H. 181, 185 koinonikon 526 Konrád, K. 606 kontakion 102, 528 Korvcy 348,606 Kowalewicz, H. 606 Kramer, S. 339 Krieg, E. 267
652
Index of Names and Temis
Kroon, S. 605 kyriale 302, 313-17
Kvrie eleison 6, 8, 19, 22, 28, 36, 38, 148-56, 162-3, 166-8, 183, 214; early history 497-8, 504; invocations during Office 27, 30, 35; with Latin text, tropes, and prosulas 196, 210-13, 223, 434-5, 519, 554; in litanies 51
La Feillec, F. de 620 Labhardt, F. 195, 571-2 lai 188, 195, 248, 266 Lambilotte, L. 573, 623 Lambres, B. 613
Lament of Rachel (Lamentatio Rachelis, Ordo Rachelis)
251, 265-7, (266) Lament of Marv (Planctus Manae, Maricnklage) 251,
265-6
Lamentations of Jeremiah 35-6, 38, 55 Lamothe, D. R. 574 Landulphus 540
Landwehr-Melnicki, M. 28, 148, 150-6, 167, 211, 295,
315, 336, 419, 531, 606 Lanfranc 19,582,604 Lantpertus 572 Laodicea, Canons of 486
Laon 39-41, 169, 233-7, 243, 265, 270, 299, 347, 349,
373, 448, 585, 608 lauda 250
laudes, Gallican 188-9, 500; in Old Spanish mass 188-
9, 500-2, 560; in Old Spanish office 493 Laudes regiae 323,591 Lauds 18, 20-2, 28-9, 34-6, 38, 85, 88, 249 Laum, B. 365 Lawley, S.W. 321 Lazarus, plav of 269 Lc Holladav", R. 452 Le Mans 573 LcPuv 39-41,243,249 Lc Roux, M. P. 458 Le Roux, R. 298, 308, 578-9 Leander, brother of Isidore of Seville 505 Lebeuf, J. 620 Lechnor, J. 1 Lcclereq, FL 552, 618 Leclercq, J. 575
lectionarv 288, 291, 293-4, 297, 308, 313, 320, 322-3,
408-9, 420-1, 492, 552-3 Lectionarv (Comes, Capitulary) of Würzburg 294, 502 Ledwon, J. 308 Leeb, H. 480, 487, 490 Lcfevre, P. 614 legenda 26 legendär 313 Legg, J. W. 289, 317, 319 Leidrad, archbishop of Lvon 5 17 Leipzig 390 Lejav, P. 494, 540 Lemarie, J. 530, 602
Lent 8, 10-13,23-4,31-3,44,70, 104, 117 Leo III, pope 458, 517 Leo IV, pope 511
Leo IX, pope 275-6 Leo XIII, pope 626 Leon 559 Leoninus 285
Leroquais, V. 309, 319-20, 322, 324, 574-5
lessons 23, 25-30, 35-6, 38-9, 82, 305, 307, 491, 495-
6, 500-3, 552-3; chants to accompany procession
246-50, 500-1; farsed 196, 225, 233-8; recitation
formulas for 46-9, 54-8 Lew, K. 1, 102, 121-2, 149, 161-2, 183, 362-3, 371,
519-20, 525-30, 554, 558, 585, 591 Lewes 54, 322, 575
Liber tic correct tone antiphonarii (Agohanl of Lvon) 570
Liber de online antiphotiarii (Amalarius of Metz) 569
Liber de riris illustribus (Gcnnadius) 291
Liberhymnorum (Notker) 572
Liberoffiaalis (Amalarius of Metz) 490, 569
Liber pontificalis 156-7, 165, 495-6, 503, 505-6,
510-11 Licinius, emperor 484 Lickleder, C. 623 Liechtenstein, P., publishers 623 Liege 25, 471 Lille 349
Limoges, and monastery of Saint-Martial 224, 238, 265, 304, 305, 314, 316, 325, 332, 351, 509, 511, 564, 578, 596-601
Limoges, Council of (1031) 599
Lincoln 18
Lindisfarne 294
Lindisfarne Gospels 294
Lindsay, W. M. 443
Linköping 605
Lipphardt, W. 250-1, 264, 284, 304, 325, 329, 380, 421, 569-70
liqucseence 341, 343, 348-9, 352-3, 357-8, 360, 371,
388, 397, 400, 429, 431 Lisbon 603
litany 20,27,44-5,51,53, 100, 149-52,(151), 183, 212,
Little Hours (Day Hours) 18, 29-30, 34, 38, 85, 88,
310-11 Littlchalcs, H. 287
liturgical drama 46, 250-73, 314-17, 325
Liturgy of St Basil 482; of St James 481, 498; of St
John Chrysostom 482; of St Mark 481 Liutward, Bishop of Vercelli 185-6 Loew (Lowe), K. A. 492, 590, 607 London 584 Longpont 610
Lord's Prayer, see Paternoster Lorenzana, cardinal 558 Lot, F. 571 Lothar I, emperor 458 Lothar II, emperor 522
Louis'the Pious', emperor 292, 517, 522, 528, 553, 570, 598
Low Sundav 13
Lowe (Loew), F. A. 492, 590, 607
Lucca 354; St Fridian at 594
lucernarium (Old Spanish) 493; (Milanese) 494
Index of Names and Terms
653
Lucidarium (Marehettus of Padua) 476 Ludmila, St 606
Lud us Danit'lis 41, 251, 266, 270-3
LullyJ.-B. 171
Lumen Christi, intonation 38
Lund 604-5
Lunden, T. 605, 615
Lütolf, M. 150, 295, 313, 352, 389, 531
Lutz, C. E. 448
Lvdd 584
Lyon 87, 570, 614
Maban 507 Mabillon.J. 622 McArthur, A. A. 7 McCarthy, M. C. 488 McGee, T. 252-3
McKinnon, J. 291, 294, 484-7, 489-90, 492, 495, 498-9, 502
McKitterick, R. 5, 448, 516 McRobcrts, D. 584 Macrobius 443 madrasha 486 Madrignac, A. 131 Magistretti, M. 493-4, 540
Magnificat, canticle 21, 29, 35, 39, 43, 99, 431; antiphons for 88, 96-9, 105, 305-8, 333; tones for 58, 64, 310, 323
Mainz 347, 441, 594; abbcv of St Alban 324,417
Mallet, J. 590
Malmesbury 581
Mandatum ceremony 35, 251
maneriae 476
Mansi, G. D. 504, 516
Mantua 351
manual 324
Marcellus (Moengal) of St Gall 186, 572
Marehettus of Padua 476
Marchiennes 578
Marcusson, O. 196, 204
Marian office (Brigittine) 615
Mark, St 481
Marosszcki, S. R. 609-10
marriage 42, 44, 324
Marseilles, abbey of Saint-Victoire 602
Martial, St 599'
Martianus Capella 443, 448-9, 460 Martimort, A.-G. 1,294,501-4 Martin I, pope 505
martyrologv 297, 313, 322; Hieronymian 313; Usuard's 313
Marxer, 0. 150,571-2
Mary, Blessed Virgin, Mass of 18, 439; Hours of 18;
Purification of 12, 31, 100, 102, 105 Mas, J. 352
Mass 3-7, 16, 18-19, 22-5, 31, 34-7, 45, 290; early history 494-503, Mass for the Dead 44; Mass of the Paschal Vigil 38; Mass of the Presanctified 33; votive 18, 44; see also Holy Saturday
Matins, see Night Office
matutinarium (Old Spanish) 493
Maundy Thursday 11, 13, 31-6, 44, 319 Maveul, abbot of Cluny 575, 578 Mcchclen (Malines) 618, 623 Mediccan gradual 397, 615-19, 623-6 Meier, B. 616 Melk 595, 615
Melkite church 481; Melkite Syrian chant 460
Melnicki, see Landwehr-Melnicki
melodiae (melismas in Old Roman and Milanese chant)
187-9, 436-7, 502, 538-9, 542, 544-6 mensural notation 156, 165, 167-8, 397 Merccnicr, F. 105 Mcrkley, P. 90, 541 Methodius, St 482 Mettenlciter, D. 472
Metz 87, 299, 304, 325, 329, 348-9, 471, 509-10, 516-
17,520, 564, 569-72, 609-11 Miazga, T. 149, 168-9, 171,606 Michael II, emperor 528 Michael, monk of Pomposa 389 Micmlogus (Guido of Arezzo) 346, 384, 426-7, 466,
468-9, 474, 476 Micmlogus de ecclesias/icis ufficiis (Bernold of Constance)
607
Milan, Edict of (313) 484
Milan and Milanese (Ambrosian) chant 87, 90, 101, 105, 130-1, 140-1, 169, 187-9, 200, 309-11, 321, 352, 354, 392, 436-7, 478, 482-3, 518, 524-7, 529-30, 538, 550-3, 556-7, 559, 561-2, 585, 611; early history of office chants 490, 492-4; of mass chants 497,499-503; music, and comparison with Gregorian 540-9
Miller, J. H. 1
Milveden, I. 605
Missa maior 20
Missa Chrismalis 35
Missa graeca 525, 528, 571
Missa matutinalis 20
missal 287, 289, 293-4, 303, 314-15, 319-20, 322-3,
335-6, 418-19, 507, 595, 615, 618 Missalc Bobbiense 292 Missale Francorum 292 Missale Gallicanum vetus 292 Missale Gothicum 292, 552 Misset, E. 172, 194 Moberg, C. A. 140, 172, 310-11, 605 Mocqucreau, A. 77, 285, 370, 379, 626-8 mode 47, 58, 61, 66, 88, 91, 99, 127, 167, 273-4, 298,
326, 328-35, 451, 454-77, 479, 520-2, 525, 527, 539,
541,559-60 Moderini, A. 352 Moeller, E. 324
Moengal (Marcellus) of St Gall 186, 572 Moggio 419
Mohlberg, L. C. 281,291-2,552 Moissac 310, 602 Molitor, R. 395-6, 615 Möller, H. 303, 308, 363 Mone, F. J. 281, 552 Mone Masses 281, 552 Moneta Caglio, E. 437, 541-2 Monkwearmouth 507, 509
654
Index of Names and Terms
monochord 328, 387, 451, 469-71, 473; Bocthius' division of 444-5; division in Dialogus de musica 463-4
Mont-Blandin, see Ghent
Mont-Saint-Michel 579, 582
Monte Gargano, basilica of St Michael 319, 458
Montccassino 354, 508, 550, 563-4, 589-91, 594-5
Monza 328, 338-9, 351, 419, 585
Moosburg 55
Morawski, j. 605-6
Morin, G. 291,294,488,512,553
Mount Sinai, monastery 481
Mozarabic chant, see Old Spanish chant
Mühlmann, W. 452, 461
Müller, H. 448, 452, 472
Muller, H. F. 220
Murphy, J. M. 116-17, 120, 531
Murray, G. 379
Musaeusof Marseilles 291, 293, 297, 492 Musica discipline! (Aurelian of Reome) 364, 448, 456-8, 522
Musica enchinadis (Anon.) 240, 328, 345, 386, 393-4,
448, 452-3, 460, 462 Mynors, R. A. B. 443
Naples 294
Narbonne 330, 597
Nejcdly, Z. 250,606
Neo-Gallican chant 609, 618-21, 623
Nestorian (East Syrian) Church 481
Netzer, H. 319
Neufville.J. 488
neuma, melisma for each mode 331-3, 345, 453; musical phrase sung on one syllable 341, 345-6; neuma triplex 200, 345
ncume 340-1, 345-6
Ncvcrs 85, 146, 217, 310, 448
New Year's Day, see Circumcision
Nicaea, Council of (325) 10, 169
Nicene Creed, jw Credo in unuin Deum
Nicholas, St, plavs of 269
Nicholas III, pope 595
Nicholas of Clairvaux 194
Nicholson, E. W. B. 304, 405
Nicolai., M. G. 285
Nicomachus 444
Nidaros (Trondhcim) 308, 604-5
Nidibrius, bishop of Narbonne 517
Night Office 18-22, 25-7, 32, 34-9, 66, 70, 83, 85, 88-9, 99, 104, 290, 307, 309-10
Nivelles 298
Nivers, G. 620
Noblat, monastery of Saint-Leonard 597 nocturn 25-7, 34-5, 39 Nocturns, see Night Office
Noeanne melodies 331-3, 451, 453, 455-6, 477, 527 Nonantola 150, 219, 224, 316, 352-4, 570, 585 None 18, 20-1, 29-30, 32, 35-6, 85, 249 Norbcrg, D. 279,281-3 Norbert, St 614 Norden, E. 285
Nortier, G. 339 Norton, M. L. 251, 261
Notker'Balbulus'('the Stammerer') of St Gall 146, 173, 178, 185-6, 200, 265, 286, 374, 516-17, 522, 530, 572, 583, 585-7
Novalesa, La 351, 586, 598
Nowacki, E. 47, 76, 89-90, 526, 531, 533-4, 539, 560 Noyon 349, 411
Nunc Dimittis, canticle 22, 30-1
O-antiphons 29, 88, 91, 98-9, 105, 432-3
Odo totii de musicae cutis cum suis differentiis (Regino
of Prüm) 458 Odelman, E. 196 Odes of Solomon 485 Odilo, abbot of Cluny 575 Odington, W. 326 Odo, abbot of Cluny 387, 463, 575 Odo of Arezzo 326," 328, 463, 576 Odorannus of Sens 326, 387, 470 Oesch, H. 463,466,472-4 offerenda (Milanese) 500, 541-2 offertory 19, 23, 38, 46, 102, 121-30, 132, 297, 299,
314-17, 330-1, 456-7, 519, 526, 554, 556-7, 561;
tropes 215-16; prosulas 196, 201-4; early historv
494, 499-500; Old Roman 531, 534-8; Old Bcneventan
550; see also offerenda, Milanese, sonus, Gallican,
and sacrificium, Old Spanish office (Divine Office) 19, 21-2, 25-30, 45, 520, 580;
early history 487-94; offices with verse texts 46,
273-9
Office of All Saints 19; of the Dead 18,44 Officium stellae 264-5 oktoechos 459-60
Old Roman chant and sources 63, 70, 87, 90, 109, 120, 130-1, 157, 162, 166-7, 185, 187-9, 223, 338, 389, 430-1, 478, 483, 497, 502, 518-19, 524-6, 529, 541-2, 546-50, 554, 556-7, 560-2, 569, 594, 608, 628; music, and comparison with Gregorian 530-40
Old Spanish (Hispanic, Mozarabic, Visigothic) rite and chant 122, 169, 285, 305, 309, 312, 321, 415, 478-9, 484, 524-5, 546, 553-4, 556-7, 576, 603; early history of office chants 493; of mass chants 498—503; music, and comparison with Gregorian 556-61
Olivetan Order 595
Olomouc 606
Omlin, E. 333
Onncrfors, A. 605
orational 311; see also collectar
Ordericus Vitalis 580
ordinal 288, 323, 325, 398, 594
ordinary (chants etc.) 8, 22-5, 46, 148-50, 302, 310, 336, 438-9, 519, 521, 561, 580, 603, 606; Old Roman 531
ordination 42, 44, 291, 324
ordines 288; Ordincs Romani 289-91,298,516,622
Ordo monasterii 488
Ordo prophetarum, play 267
Ordo Romanus I 151, 166, 290, 296, 496, 498-9
Ordo Romanus 11 185
Ordo Romanus III 166,290
Index of Names and Terms
655
Ordo Romanus IV 152,296 Ordo Romanus V 296, 499 Ordo Romanus VI 74, 296 Ordo Romanus X 499 Ordo Romanus XII 290 Ordo Romanus XIII 290 Ordo Romanus XIV 290 Ordo Romanus XV 496, 498-500 Ordo Romanus XIX 503, 505-6 Ordo Romanus XXVI 290 Ordo Romanus XXVII 538 Ordo Romanus XXXB 290 Ordo Romanus L 291 Orel, D. 250,606
organ 386-7, 394, 439, 450, 470, 581-2, 620 organum, see polyphony Origen 309
Origny-Sainte-Benoítc 267
oriscus 341, 343, 345, 348-9, 352-3, 359-61, 371, 383,
388, 400 Orleans 447
Orsini, John Cajetan, cardinal, Pope Nicholas III 595 Osbern, abbot of Saiiit-Evroult 580 Osmund, St, bishop of Salisbury 583 Ossing, H. 308
Oswald, bishop of Worcester 581
Otker of St Emmeram, Regensburg 471-2
Otloh of St Emmeram, Regensburg 471-2
Ott, K. 121-2, 556
Otto I, emperor 594
Otto III, emperor 594
Ottosen, K. 304, 308
Ottósson, R. A. 273, 605
Oxford 584
Oxyrhynchus hymn 487
Padua, monastery of Santa Giustina 595 Palermo 201, 328 Palestrina, G. P. da 615-16, 626 Pahsea, C. V. 442, 476
Palm Sunday 9, 11-13, 32-4, 100, 102, 251, 336
Pannain, G. 476
Paraclete, monastery 611
paraphonistac 538-9
parapter 58, 62-4
Paredi.A. 540
Paris 40-1, 51, 57, 73, 77, 161, 190, 194, 285, 302, 320, 335, 391, 433, 564; Notre-Dame 194; Saint-Vietor 194, 614; Saint-Jacques 612
Parvio, M. 319
Paschal Candle 38
Paschal Vigil, see Holy Saturday
Paschal Kyrie (Kyrie 39, Vatican I) 28
Paschal time, see Easter season
Pascher, J. 7
Passau 389
Passion 34, 36, 56
Passion Play 266
Passion Sunday 13, 33
passionary 288, 308, 313
Passiontiďe 75, 92, 144
Paternoster 24-7, 30, 54, 162, 499, 505, 528; farse 196,
234, 237 Patier, D. 273,606 Paul, St 486 Paul I, pope 515-16
Paul Warncfrid ('the Deacon') 503, 508-9, 516, 590 Paul of Samosata, bishop of Antioch 485 Pauhnus of Aquileia 169, 239, 522 Paulinus, secretary of St Ambrose 490 Pavia 353, 585 penance 44
Penitential Psalms 31, 35 Pentecost, season 10-11,13 Pcrego, C. 541
Pcregrinus play 257, 260, 269-70, 273 pericopc 293
Periphyseon ( Johannes Scotus Eriugena) 448 Perugia 595
pes 342, 344, 348-9, 352-3, 390; pes quassus 360-1;
pes stratus 360-1; pes subbipunctus 342,344 Peter, St 480 Peter the Deacon 508 Peter the Venerable, abbot of Cluny 575 Peterborough 582 Petershausen 61,63,65,390,578 Petronax of Brescia 590 Petrus Olavi of Skanninge 615 Petrus de Cruce of Amiens 326 Pfaff, R. 1, 16 Philip II, king of Spain 4 Philippa of Lancaster, queen of Portugal 603 Phillips, N. 240, 394, 452, 470 Philo 486
Piacenza 252-3, 323, 586, 595 Piconc, C. 360
Pierre de Corbeil, archbishop of Sens 40 Pikuhk.J. 606
Pilgrim, archbishop of Cologne 328 Pineau, C. 620 Pinell, J. M. 558
Pippin III 'the Short', king of the Franks 292, 299, 515,
552, 564, 569, 581 Pistoia 216, 252, 254, 354, 586 Pitman, G. 121 Pitra, J. B. F., cardinal 626 PiusV, pope 309, 618 Pius IX, pope 624 Pius X, pope 626 Placebo 19
plainchant musical 620-1
Planchart, A. E. 196, 215-17, 220, 223, 238, 313, 580,
583, 585, 589, 596, 598 planctus 239, 265-6 Planer, J. H. 325,329 Plato 443 plica 358
Plocek, V. 339, 606 Plombariola 589 Plummer, C. 506 Pocknce, C. 619 podatus 342, 344
656
Index of Names and Terms
Podlaha, A. P.-A. 606 Poisson, L. 620 Poitiers 144 Poland 149, 171 Polheim, K. 285-6
polyphony 32, 36, 41, 43, 73-4, 77, 325, 333, 387, 452,
470, 476-7, 564, 576, 582-3, 603 Pomerium (Marchcttus of Padua) 476 Pomposa 389, 466 Ponchelet, R. 360, 382-3 Ponte, J. P. 364, 456 Pontefract 322
pontifical 42, 287-8, 291, 324, 328, 424-5, 591, 594, 606
Poore, Richard, bishop of Salisbury 583
porrcctus 342, 344, 348-9, 352-3,'390, 397; porrectus
flexus 342, 344 Porter, W. S. 285, 497, 552 postcommunion (pravcr) 24, 35, 39, 49, 291 Pothier, J. 365, 379, 397, 624, 626-7 Potiron, H. 392, 448 Pozzuoli, abbey of St Peter 305 Practica artis musicae (Alfred) 595 Prado, G. 55, 169, 557 Praeconium paschalc, see Exultet Prague 225,606
prayers, recitation formulas for 46-54, 162 precentor 39—40
preces: at Compline 30; litany in Gallican and Old
Spanish rite 151, 317, 497, 500-1, 554 Preface (of mass) 23, 49-50, 162, 291 Premonstratensian Order 613-15 Premontre 614
Premunitos an tent esse volumus 610 pressus 360-1 Prim, J. 620
Prime 18-21, 29-30, 85, 249 Priscian 366
procession, liturgical 30-3, 44-5, 51, 82, 100, 102, 104-5, 221, 239, 251, 253, 264-5, 269, 290, 302, 318-19, 336; see also antiphon, processional and hvmn, processional
processional (book) 310, 315-19, 322-3, 603 Procter, F. 289, 321 prokeimenon 495
Prologus de ordine antiphonarii (Amalarius of Metz) 296, 569
proper (chants etc.) 8, 22-5, 302, 314, 526 prophecy (Gallican, Old Spanish, Milanese) 500-2 prosa 492
prose (sequence text) 173 ; see also sequence
prosula 32, 43, 183, 196-214, 225, 230-1, 286, 314-17,
333, 572 Pröszeky, G. 607 Prudentius 142 Prüfening 578 Prüm 328, 458-9, 471 psallcnda (Milanese) 494, 541
psalm tone 46, 58-64, (59-60), 66, 89, 92, 99, 158, 286, 326-35, 453, 560; Milanese 541; Old Roman 531, 539; Old Spanish 559
psalmellus (Milanese) 495, 500-3, 541-2, 547
Psalmi Familiäres 19-20
psalmo (Old Spanish) 495, 500-3
psalmodv: antiphonal 70, 122, 489, 559; direct 77;
rcspoiisorial 70,91, 122,489,559 psalms 25-30, 39, 45, 58, 308-10, 484-6, 488-95 psalter 288, 305, 308-11, 322-3, 336 Psalter: Gallican 66, 83; Hebrew 309; Roman 66, 309 Pseudo-Alcuin 185 Pseudo-Bernard of Clairvaux 326 Pscudo-Bernelinus 472-3 Pseudo-Dionysius the Arcopagitc 528 Pscudo-Odo 326 Ptolemy 444,461
punctum 341-2, 344, 348-9, 352-3 Pustet, F., publishers 616, 618, 623-6 Pythagoras 444, 470
Quadragesima (Sunday) 10-11, 13 Quadripat titafigura (Otkcr) 471-2 Ouaestiones in musica (Rudolf of Saint-Trond or
Franciscus/Franco of Liege) 326, 472 Quarr Abbey 624 quarter-tones 361,388 Quedlinburg 570 Quentin, H. 313, 626
Quicumque vult (Athanasian Creed) 30, 168, 333 Quid est cantus? (Anon.) 345-6, 365, 384 Quierzv, Svnod of (838) 570
quilisma 341, 343, 345, 348-9, 352-3, 358, 371, 383,
385, 388, 392, 397, 400, 453 Quinönez, F. dc, cardinal 618 Quinquagesima Sunday 10, 13,83
Raasted, J. 332,527,605 Raby, F. J. E. 279
Radbod (Rathbod), archbishop of Trier 328, 458
Radegunda, Queen 144
Radö, P. 339,607
Rajecky, B. 607
Ramsey 581
Randel", D. 503, 558-60
Ranke, E. 294
Rankin, S. k. 251-2, 257, 260, 351, 369, 372, 572, 580,
582, 586 Rasmussen, N. K. 324 Ratcliff, E. C. 188,498,500,552 Ratdolt, E. 396
Rathbod, archbishop of Trier 458
Ratisbon, see Regensburg
Ratpertof St Gall 146-7, 572
Ravenna 217, 317, 354, 482-3, 527, 530, 550
recitation formulas 46-58, 162,616
Rcckow, F. 476
Reconciliation of Penitents 31, 33, 35, 44 Regcnsburg, and monastery of St Emmeram 217, 3 15, 328, 363, 373, 470-1, 564, 578, 606, 616, 623-4, 626 Regino of Prüm 89, 326, 328, 448, 454-5, 458-9, 522-3 Reginold, bishop of Eichstätt 205, 274 Regida Magistn, see Rule of the Master Regidaiis ( oncordia 19, 254, 581
Index of Name s and Terms
657
Rühle, S. 292,589
Reichenau 285, 325, 328, 330, 390, 395, 570-1, 585, 606
Reichert, G. 173
Reier, Ii. J. 586
Reims 57, 349, 448, 552, 605
Kella, A. 626
Remigius of Auxerre 448-9, 463, 522 Remigius (Rcmedius), bishop of Rouen 515 Renaudin, A. 574-5 Renoux, A. 481
Réóme, abbe v of Saint Jean 456 Reproaches (Impropcria) 36-7, 265 Requiem 44
reservation of the sacrament 24
rcsponsorium (Gallican, at mass) 500-2; ad lectionem (Milanese) 541-2; cum infantibus/pueris (Milanese) 494, 541, 544-5 ; in baptistériu (Milanese) 494, 541-2, 544; in choro (Milanese) 494, 541; post hymnum (Milanese) 541
responsorv 25-30, 35, 37, 39, 273-9, 291, 296, 305-7; early history 489-96, 517
responsorv, great (rcsponsorium prolixum) 21-2, 25-7, 29, 32, 46, 69-76, 79, 82, 93, 99, 104, 109, 114-17, 120, 122, 305-7, 330-2, 336-8, 456-7, 563, 569, 578, 582, 600, 610; melismas 76, 196, 200-1, 204-7; Old Roman 531, 533; tones for verses 46, 48, 58, 62, 65-6, 158,286,330,560
responsorv, short (rcsponsorium breve) 22, 28-9, 39, 46, 85-8,91,305-7,310
liclracttones (St Augustine) 499
Rhabanus (Hrabanus) Maurus 10, 447
Rheinau 334, 338-9, 578
Ribav, B. 77
Richard II, Duke of Normandy 387, 564, 579
Richard of Saint-Victor 194
Richerius, abbot of Montecassino 591
Riemann, H. 395, 442
Righetti, M. 1
Ripoll 602-3
ritual 324
Rivera Recio, J. F. 557
Robert de Grantmcsnil, abbot of Saint-Evroult 580
Robert of Molesme 609
Robert, L. 47, 54
Robert, M. 579
Robertson, A. W". 150
Roeca-Serra, G. 443
Rodradus, abbot of Corbie 298
Rodulphus Glaber 578
Roederer, C. 104, 596
Rogation Days, Rogation Sundav 11, 32, 51, 100, 147, 290
Rojo, C. 169, 557
Romanian letters, see significative letters
Romanos the Melode 529
Roinanus, Roman cantor 374, 572-3
Rome, Roman liturgy and chant 70, 73-4, 81, 100-2, 104, 150-2, 157, 165-6, 169, 289-93, 296-7, 299-300, 304-5, 309, 312, 320-1, 324, 352, 364, 374, 388, 391, 395-6, 413, 448, 458, 466, 479-80, 482-4, 486, 488, 492-4, 496-521, 524-7, 530-1, 540-2, 547, 549,
552-4, 556, 560-2, 564, 569-73, 576, 590-1, 594-6, 604, 608, 615-16, 618, 622-3, 626, 628-9; see also Old Roman chant
Rome: Lateran palace, basilica of St John 290, 292, 493, 508, 594, 596; chapel of St Lawrence 594; Vatican, basilica of St Peter 290, 292, 309, 493, 508
rondeau 247-8
Rönnau, K. 149, 199-200, 223-4, 452 rotta 199, 572
Rouen 34, 137, 147, 263-5, 269, 335, 399, 515-16;
monastery of Saintc-Catherinc-du-Mont 580 Rousseau, O. 624 Rudolf of Saint-Trond 326,472 Ruini, C. 595
Rule:of St Augustine 488, 614; of St Benedict 70,141, 488, 491-3, 609, 611; of St Cacsarius 488, 492; of Chrodegang of Metz 321, 516; of St Columbanus 488; of the Master {Regula Magistrí) 488, 491-2
Ruodmann, abbot of Reichenau 571
Russell, T. A. 468
Sablavrolles, M. 602 Sachs, K.-J. 470 Sackur, E. 574
sacramcntary 287-9, 291-4, 297-8, 303, 320, 324, 420-
1, 428-9, 516, 518, 540, 569 Sacramentarv, Gelasian 10, 292, 297; of Gellone 292;
Gregorian 292-3, 298-9, 503, 509-10; Hadrian 17,
292, 509; Leonine 291; of Trent or Salzburg 292; of
Verona 291 sacrmcium (Old Spanish) 122, 493, 500, 560 Sahagun 558 St Albans 582
Saint-Amand 152, 292, 448, 578
Saint-Benoit-sur-Loire, monastery at Fleury, see Fleury Saint-Bertin 448
Saint-Denis 67, 298, 304, 321, 333, 338-9, 354-5, 413,
515, 528, 554, 571, 582-3 St Emmeram, jw Regensburg Saint-Évroult 224, 316, 328, 579-80 St Florian 389
St Gall 146-7, 163, 186, 197, 199,215-17,220,223, 299, 304, 334, 338-9, 347, 350, 357, 360-1, 373-4, 382, 399, 563-4, 571-5, 583, 586, 608, 623
St Georgen 471, 578
Saint-Martial, see Limoges
St Martial/De La Fage Anonvmous 476
Saint-Maur-des-Fossés 205,"321, 338-9, 575
Saint-Quentin 4
Saint-Riquier 141, 325, 330, 459, 517, 521, 571
St Sabas, monastery near Jerusalem 481
Saint-Vaast, see Arras
Saint-Victor, monastery near Le Mans 458
Saint-Victor, Augustinián abbey in Paris 194
Saint-Victor-sur-Rhins 575
Saint-Yricix 317, 597, 608
salicus 360-1, 382-3, 392
Salisbury 34, 51, 583-4; see also Sarum
Sallmann, N. 443
Salmon, P. 320, 322, 487, 594
San Millán de la Cogolla 602
658
Index of Names and Terms
San Salvátore on Monte Aniiate 322
Sanctorale 300-2, 305-7, 310-11, 335-6
Sanetus 8, 19, 23-4, 68, 148-50, 161-7; prosulas 209,
225, 230-1; tropes 196, 220, 223, 225, 230-1, 519 Sanetus post evangelium (Gallican) 500-1 Sandon, N. 150, 295 Sandrat of St Maximin, Trier 571 Santeuil, J.-B. de 618 Santi, A. de 626
Santiago de Compostela 39-41, 234, 576, 603 Sarum use and sources 32, 36, 38, 47-8, 56, 71, 90,
105-6, 141, 143, 150, 172, 301-2, 310, 320, 326, 333,
483, 583-4, 603, 611 Sauxillanges 575
scandicus 341-2, 344, 348-9, 352-3, 383, 390 Sceptrum prosula (cf. Regmtm mum solidům) 228-30 Scheiwilcr, A. 571
Schildbach, M. 148, 165-8, 223, 315, 336, 606 Schlager, K. 6, 68, 130-5, 137-8, 173, 187, 204, 209, 336
Schmid, H. 384, 393-4, 452 Schmid, T. 604-5 Schmidt, Mans 76, 531 Schmidt, Hermann 7,33,519 Schmitz, A. 238
Schneidcr/London/Naplcs Anonymous 476
Schoenbaum, C. 606
Schubiger, A. 374, 571-2
Schneller, H. M. 442
Schuler, E. A. 251
Schuler, M. 521-2
Schwvzer, E. 365
Scolica enchiriadis (Anon.) 384, 386, 393-4, 448, 452 scrutiny of catechumens 23, 33, 512 Seav, Ä. 476 Scckau 389
secret (prayer) 23, 49, 291
scculorum amen (close of Gloria patri) 61
Sedulius 142,284
Senarius 502
Sen I is 298
Sens 39-41, 169, 171, 207, 233-5, 238, 243, 249-50 Septuagesima Sunday 13 sequela 173
sequence 23, 41, 46, 104, 131, 134, 153, 172-95, 207, 212, 221, 225, 239-40, 248, 266, 273, 279, 286, 310, 313-17, 323, 330, 336, 386-7, 394, 417, 440-1, 479, 494, 511, 519, 521, 546, 560, 572, 576, 580, 583, 585-8, 591-2, 596-9, 603, 606; (versus) with double cursus 179-80, 239-41; tropes 196,223
sequentiarv 302, 313-17, 323, 597
Sergius I, pope 102, 165, 499, 505
sermon 26, 313
serpent 620
Servatius, V. 615
Scsini, U. 353
Severus of Antioch 460
Sevcstre, N. 197, 220
Sexagesima Sunday 10,13
Sext 18, 20-1, 29-31, 85, 249
Sherlcy-Pnee, L. 10, 102, 506, 629
Sibert de Beka 615 Sidler, H. 121, 123, 127 Sidonius Apollinaris 492 Siffrin, P. 281 Sigl, M. 150
significative letters (Romanian letters) 56, 374-8, 572 Silos, abbey of Santo Domingo 305, 328, 338-9, 415, 559, 602
Simeon, secundus of the Roman sehola cantorum 515-16
Skeat,W.Vv. 294
Slavonic liturgies and chant 482; ekphonetic notation 367
Smith, J. A. 484-5
Smits van Waesberghe, J. 180, 201, 209, 363, 374, 384,
386, 389, 427, 442, 466, 468, 471-3, 521, 538, 560 Snow, R. 529, 531 Socrates 489 sogitha 486
Soissons, abbey of Saint-Medard 298
Solesmes, abbey of Saint-Pierre 148-9, 302-3, 379,
383, 397, 400-1, 433, 573, 578, 591, 595, 614, 623-8 Somnium Scipionis (Cicero, commented by
Macrobius) 443 song, Latin liturgical (conductus, versus, cantio) 39-43,
46, 193-4, 238-50, 266-7, 269, 273, 279, 284, 576,
598, 606
Song of Simeon, see Nunc Dimittis
Song of Zacharia, see Benedictus
sonus, Gallican 500, 554, 556-7; Old Spanish 493-4,
500, 560 Soriano, F. 616, 626 Souvignv 575 Sovva, IL 330 Sozomcn 489
Spanke, II. 188, 239-40, 242, 248, 250 Speculum musicac (Jacobus of Liege) 476 Sponsus, play 194, 267, 271 Spunar, P. 606 Srawlev, J. II. 4
Stäblein, B. 47, 51, 53-8, 62, 66, 74, 76, 88, 92, 102, 104, 109, 116, 130, 140-4, 146-7, 150, 155-7, 165-6, 168-9, 172, 178, 180-1, 183, 185, 188, 195-7, 199-201, 205-6, 209, 220-1, 223-4, 234, 236, 238, 240-1, 248-50, 278, 293, 295, 303, 310-11, 336, 340, 363, 370, 395, 431, 484, 487, 494, 503, 508, 510-11,514, 526, 531, 538, 540, 552-4, 560, 562, 573, 576, 591-2, 603, 606, 628
Stahl, W. II. 443
station, stational liturgy 31, 252, 264, 336, 413, 431,
510,516 Stauffacher, M. 238 Stavelot (Stablo) 459 Steer, G. 267 Steglich, R. 326, 472 Stein, F. A. 250
Steinen, YY. von den 181, 185-6, 522, 571-2 Steiner, R. 1,41,47,64,66-8, 109, 121-3, 125, 140,
196, 200-1, 204, 220, 225, 279, 295, 308-9, 310,
503, 519, 574, 596 Stcnzl.J. 339
Index of Names and Tenns
659
Stephan, R. 225
Stephen II, pope 515, 517, 571
Stephen III, pope 517
Stephen IX, pope 591,594
Stephen I, St, king of I lungarv 607
Stephen, bishop of Liege 274
Stevens, J. 251, 265-6, 269
Stevenson,J. 493, 581
Stotz, P. 146, 216
Stratnian, C. J. 251
Strecker, K. 279
Strehl, R. 221
Strbmberg, B. 605
strophici 390, 392, 397, 400; see also apostrophe,
bistropha, tristropha Strunk, 0. 220, 387, 389, 443, 460, 463, 465-6, 525-6,
530
Stubbs, W. 297, 507, 581 Sturmi, abbot of Fulda 590 Subiaco 595 Subita, J. 338 Suchier, W. 284 suffrages 29
Sunol, G. 77, 340, 346, 365, 369, 375, 379, 391, 407,
541, 552 Super populum (prayer) 24 Sylvester II, pope (Gerbert of Aurillac) 470, 594 Synimachus, pope 157, 497 Svmons, T. 254, 581 Syon 605
Syrian ekphonetic notation 367
Syrian churches, see Jacobite, Melkite, Nestorian church Szekesfehervär (Stuhlweissenburg) 607 Szendrei, J. 391,607,611 Szigeti, K. 607
Szöverify, J. 108, 140, 189, 273
Tack, F. 171, 395, 397, 618, 620 Taft, R. 486-9, 491-3, 515 Taliaferro, R. C. 443 Tarchnischvili, M. 481 Tarragona 603
Te Deumlaudamus 26-7,44,57-8,68-9, 156, 162,
220, 252-4, 261, 269, 273, 420-1, 494, 528 Temporale 300-2, 305-7, 310-11 Tenebrae 34-8
Terce 18-21, 29-30, 85, 249, 253 Terrizzi, F. 319 Tertullian 486
Thannabaur, P. 148, 161-4, 167,209,223,225,315,
336, 606 Theodald, bishop of Arezzo 466 'Theodore Studites, St 460 Theodore, archbishop of Canterbury 294 Theodore, bishop of Mopsuestia 487 Theodulf of Orleans 146, 447 Theogerus of Metz 471 -2 Therapeutae 486 Thibaut.A. 590 Thibaut, J. B. 122,295, 367
Thierry, abbot of Jumieges 579 Thionville, Capitulary of (805) 517 Thodberg, C. 529 Thomas Aquinas 42, 145, 192 Thompson, E. \V. 365 Thorlac, St 605
threnos (Old Spanish) 500-2, 560
Thurstan of Caen, abbot of Glastonbury 583
Tiinaeus (Plato) 443
Timotheus (Paulinus of Aquileia) 522
Tiridates III of Armenia 481
Toledo 351,558-9, 602-3; Council of (589) 169;
Fourth Council of (633) 492, 515, 558 Tolhurst, J. B. L. 18, 25, 321
1'onale sancli llernardi (pseudo-Bernard of Clairvaux) 326, 610
tonary (tonarius, tonale) 61, 63, 89, 141, 305, 310, 315-17, 321, 323, 325-35, 345, 454, 458-9, 462-3, 471, 473, 521, 527, 560, 571, 579, 585, 595, 597, 614
tonus peregrinus 58, 62-4
Toreello 353
torculus 342, 344, 348-9, 352-3; torculus resupinus
342, 345 Tortosa 603 Toulouse 597 Tournai 397
Tours 520, 569; abbey of St Martin 447; Second
Council of (567) 489 tract 23, 36, 38, 46, 69, 76-7, 82-5, 294, 296, 298,
314-17, 382, 519; early history 495-6, 500-3;
Old Roman 531, 533; prosula 209; see also cantus,
Milanese tractulus 382
transitorium (Milanese) 499, 542-3, 551-2
Traub, A. 448
trecanum (Gallican) 498-9
Treitler, L. 74, 82, 130, 220-1, 238, 242, 367, 372, 520, 596
Trent, Council of (1545-63) 609, 615, 618 Tres elerier, play of St Nicholas 269 Triduum 8, 19", 32, 34-9, 55
Trier 349, 569, 623-4; abbey of St Martin 328,458;
abbey of St Maximin 458, 571 trigon 343, 345, 361, 385 Trina Oratio 19-20
Trinity, Holy 274; Trinity Sunday 11,13 Trisagion 36-7, 497-8, 525, 528 * tristropha 343, 345 trivirga 343, 345 Troia 589, 591-3
trope 22,41,43,46, 138-9, 148, 152, 161, 168-9, 183, 196-238, 251, 289, 302, 310, 313-17, 323, 336, 479, 511, 519, 521, 560, 570, 572, 576, 580, 583, 586-7, 591-2, 596-8, 603
tropcr 289, 313-18, 328, 336, 591-2, 596-8
tropus, meaning'mode' 451, 454
Tuotilo of St Gall 199, 215, 223, 419, 572
Turco, A. 88,90
Turner, D. H. 319
Turpinus 576
Tyrcr.J.W. 33
660
Index of Names and Terms
Udalricus of Clunv (Ulrich of Zcll) 345, 576, 578
Udovich, J. 88, 308, 326, 371
Uhlfelder, M. L. 448
Ullmann, P. 607
uncinus 349, 388
Underwood, P. 308, 338, 580
Undhagen, C.-G. 605
Uppsala 604
Urban IV, pope 25
Utrecht 328
Vadstena 605, 615 Vagaggini, C. 5 Valerian, emperor 4 Vallombrosa 578 Valous, G. de 574 Van der Werf, II. 371 Van Deusen, N. 248, 531 Vanicky.J. 606 Varro 443
Vatican Council, Second (1962-5) 628 Vccchi, G. 100, 150, 172, 196, 204, 209, 266, 313, 352, 417
Velimirovic, M. 454, 482
Venantius Fortunatus 142, 144, 146, 224, 282, 284
Venerius, bishop of Marseilles 297
Venite exultemus Dominum (Ps. 94), see invitatory
psalm Vercelli 419, 573 Verheijcn, L. 488 Vcrnet, P. 322, 609 Verona 310, 338-9, 353
vcrsiclcsand responses 22, 25-30, 39, 45, 309, 312;
recitation of 47-8 versus 39-43, 108, 238-50; at St Gall 146-7; versus
(sequence) with double cursus 179-80, 239-41; versus
ad rcpetendum 109, 496-8; versus ante officium 221;
see also Latin liturgical song Vespers 16, 20-2, 28-9, 32, 35, 37-8, 85, 88, 104-5,
249
vespertinum (Old Spanish) 493, 560 Vezclay 576 Viadana, L. 621 Vich 603
Victor III, pope 591 Victor of Capua 294 Vidakovic, A. 607 Vigils, see Night Office Villctard, H. 39, 41, 234-5, 238, 249, 470 Vincent, A.-J.-H. 388 Vintimille, C. de, archbishop of Paris 619 virga 341-2, 344, 348-9, 352-3, 397; virga strata 360-1
Visigothic chant, see Old Spanish chant
Visitatio sepulchri 39, 256-7, 262-3, 420-1, 581, 585
vita 26, 89, 313
Vitalian, pope 526
Vivell, C. 326, 442, 472
Vives, J. 312, 559
Vladimir, prince of Kiev 482
Vogel, C. 7, 287-9, 291, 293, 324, 492, 514, 516, 527, 552
Vogue, A. de 488, 491 Volberts, J. VV. A. 384 Volpiano 387, 578
Volturno, abbev of San Yincenzo 589 Vosyka, V. 250
Waddell, C. 194, 310-11, 575, 609, 611 Waeltner, E. L. 466
Wagner, P. 39, 41, 69, 85, 117, 146-7, 149, 166, 234, 249, 295, 340, 345, 359-61, 370, 383-5, 389, 502, 557, 573, 576, 603, 623, 627
Wala, abbot of Corbie 570-1
Walahfrid Strabo 553
Waldramm of St Gall 146
Walker, G. S. M. 488
Walker, J. H. 321, 594-5
Wallace-Hadrill, J. M. 514
Warren, F. E. 293, 319, 484, 493, 553
WaszinkJ. 444
Weakland, R. 197, 199, 274, 448, 452, 540
Weingarten 564
Weinmann, K. 615
Weinnch, L. 241
Wcisbein, N. 192
Weiss, G. 199, 220-1, 598
Wellesz, E. 31, 367-8, 489, 501, 525, 527
Wenlock 322
Westminster 582
Wevns, N. I. 614
Whit Sunday (Pentecost Sundav) 8-11, 13,32, 169,
528; Whitsun Eve 51 Whitby, monk of, author of life of St Gregory 506 Whitby, Svnod of (664) 10 White,'A.'444
Whitehill, W. M. 39, 41, 576, 603 Wiesli, W. 358, 360, 383 Wilkinson, J. 480 William I, king of England 581-2 William, abbot of 1 lirsau 471-2, 474, 578 William, abbot of Rievaulx 610 William of Volpiano (or Saint-Benigne, Dijon, or Fecamp) 331, 386-7, 394, 564, 578-80, 582-3 Willibald, bishop of Eichstätt 205 Willibrord 515 Willis, G.G. 7 Willis, J. 443
Wilmart, A. 294, 574, 585, 626 Wilson, U.A. 319 Wilton Abbey 624 Winchcombe 582
Winchester 74, 163, 181,211,216-17,219-20,254,
315, 374, 386, 394, 413, 581-3, 586 Winithar 304 Winkler, G. 487 Winterfeld P. von 179, 240 Wipo 189 Witt, F. X. 623-4 Wolf, J. 472 Woods, I. 584
Index of Names and Terms
661
Worcester 14-16, 310, 312, 354-5, 581-2 Wordsworth, C. 287, 289, 317, 321 Wormald, F. 13 Wright, C. 554, 564
Wulfstan, cantor at the Old Minster, Winchester 387, 582
Wynfrith, see Boniface
Yearlev.J. 265
York 53, 292, 297, 507, 518, 581-3
Young, K. 33, 37, 39, 251-3, 260, 263, 265-7, 269, 283
Zagreb 607 Zoilo, A. 615-16 Zwiefaltcn 578
Osburg |