The Smithsonian Institution Regents of the University of Michigan
The Earliest Islamic Commemorative Structures, Notes and Documents Author(s): Oleg Grabar
Source: Ars Orientalis, Vol. 6 (1966), pp. 7-46
Published by: Freer Gallery of Art, The Smithsonian Institution and Department of the History of Art, University of Michigan Stable URL: http://www.jstor.org/stable/4629220
Accessed: 14/12/2009 06:30
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=si.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
	
9	
STOR	
The Smithsonian Institution and Regents of the University of Michigan are collaborating with JSTOR to digitize, preserve and extend access to Ars Orientalis.
http://www.jstor.org
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES, NOTES AND DOCUMENTS*
By OLEG GRABAR
One of the most characteristic buildings of Islamic architecture is, without doubt, the monumental tomb. The Taj Mahal or the great Mamluk mausoleums in Cairo are visited by thousands of casual tourists, while every traveler in North Africa or the Near East has seen along the roads, on top of hills, in cemeteries of cities and villages, at times even in fields, hundreds of small shrines usually assumed to be the resting place of some saint or hero, or supposed to mark the spot of some celebrated or forgotten action. Many terms are used for these constructions, whether of a rough peasant work or exquisite artistry. They may be called qubbahs and gunbadhs, "domes," after their prevalent form, or turbahs, "tombs," from their most common function, or imnmzudehs, "sons of an im3m," expressing their religious, almost ecclesiastical, connotation. They may be walis or marbats, "places of a holy personage," indicating their relation to some holy hero, maqams, "places," when related to the emplacement of some event, mash-hads, "places of witnessing," true martyria implying a commemorative value. At times, rarer terms such as qasr, "castle," or darguh, "palace," were used for them. All these names—whose precise history and origins are still to be elucidated—illustrate the multiple facets of memorial construction in the minds of the Muslims.
* This study was completed thanks to a grant from the Center for Near Eastern and North African Studies at the University of Michigan.
The memorial building is not a peculiarly Islamic phenomenon. Indian stupas served to honor Buddhist relics; most ancient Near Eastern civilizations developed varying kinds of mausoleums; the nomadic world itself, whether Semitic in Arabia or Indo-European and Turkic in Central Asia, used more or less permanent forms of commemorative architectural symbols; Hellenistic and Roman memorial structures have remained in large numbers; and the martyrium was a central concept as well as form in the growth of Christian architecture.
It is easy to assume that Islamic memorial and funerary construction was but a continuation of the numerous traditions of the pre-Islamic or non-Islamic worlds, and this assumption underlies some of the studies which have been devoted either to precise Islamic examples1 or to the most characteristic forms and functions of such buildings.2 If such was the case, the problem for the historian of Islamic art would be simply to identify various types of commemorative forms used by the Muslims and to study their evolution. There are, however, two objections to the hypothesis that the new civilization simply and directly appropriated the procedures and func-
1 E. Diez, Persien, Berlin, 1923, pp. 51 ft., 73 ff.; K. A. .C. Creswell, Muslim architecture of Egypt, Oxford, 1952, vol. 1, pp. noff.
2 A. Grabar, Martyrium, Paris, 1946, vol. 1, pp. 85-86, 145. E. E. Smith, The dome, Princeton, 1950, passim and esp. pp. 41-44.
8
OLEG GRABAR
tions of old. The first objection is theoretical. It is rare to find any branch of Islamic art merely taking over completed older forms without alteration of shape or meaning. Overwhelmed as they are with Sasa-nian and especially Byzantine and Roman plans and methods of construction and decoration, the monuments of early Islamic art can rarely be considered Byzantine or Sasanian; either in their purpose or in the relationship of their components, these monuments are new and imply new needs or new tastes. It is usually only when a facet of the new Islamic culture developed in a manner which demanded or permitted monumental expression that monuments developed to express it. Therefore, the fact that there were mausoleums in the pre-Islamic world does not by itself explain the existence of Islamic mausoleums; an explanation of their appearance must be given in the cultural terms of the time when they appeared.
The second objection to the notion of a direct passage from pre-Islamic functions and purposes to Islamic times is raised by the early Islamic view on mausoleums. It is clear that early Muslim doctrine condemned any architectural glorification of tombs, and in fact found even most funerary and commemorative ceremonies objectionable. The taswiyah al-qubur, "equalization of tombs (with the surrounding ground)," was felt to be the most appropriate expressions of the equality of all men in death, and veneration of tombs or ceremonies around tombs were considered to derive from improper Christian and Jewish habits.3 From the very beginning this prohibition ran against older tradition of behavior, and stones abound which indicate that certain funerary prac-
tices remained, visits to tombs, lamentations, erection of tents over tombs, setting up of pillars nearby, watering of tombs. All such practices derived from pre-Islamic habits and were maintained in spite of numerous efforts, at least in the first centuries of the Hijrah, to curb them.4 But, the early examples of funerary practices and the opposition to them involved behavior, not constructions. The tomb of the Prophet, whose potential significance as a focal point for worship was self-evident, was carefully kept out of the main direction of prayer in the mosque of Medina; it was unavailable to the faithful and enclosed in an irregular and purposefully ungainly screen; and it was only in the late 13th century that it acquired a cupola.5
If, then, early Islam was so strongly set against the building of mausoleums and against cultic practices in or around tombs, the eventual appearance of thousands of mausoleums and shrines with a memorial significance throughout the Islamic world requires a more precise explanation than the simple statement of a continuation of pre-Islamic habits and practices. An attempt must be made to determine precisely
3 The appropriate texts have been gathered and discussed several times; J. Pedersen, art. mas-djid, part A4, Encycl. of Islam; G.Wiet, MCI A Egypte II, pp. 64 ff.; I. Goldziher, Die Heiligen-verehrung im Islam, Muhammedanische Studien, Halle, 1890, pp. 275 ff.
4 M. Canard, Vie de I'Ustadh Jaudhar, Alger, 1958, pp. 150-151, and esp. note 340; Maqrizi, vol. 1, pp. 312, 313, 327 are but a few examples. For an overall view of early funerary practices, as drawn from hadith literature, see I. Griitter, Arabische Bestattungsbrauche in friihislamischer Zeit, Der Islam, vol.31, 1954, pp. 147ff.
5 J. Sauvaget, La mosquee omeyyade de Me-dine, Paris, 1947, p. 44.
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
9
where, when, how, and why this new and apparently most characteristic Islamic monument appeared. The documentation presented here has been gathered in an attempt to provide an answer to these questions. My original purpose was merely to draw up an annotated list of the earliest remaining sanctuaries, but it soon became apparent that a simple list of standing monuments and of monuments known through inscriptions—i.e. from basically archaeological sources—was not sufficient. On the one hand, among the disused sanctuaries, it is only accidental that some have remained rather than others. And on the other, many of the still popular shrines were so much redone in later centuries that archaeological analysis alone cannot provide the date of foundation of the sanctuary. But, as one begins to cull literary sources for documents on the subject of early shrines, a number of additional problems are raised which require preliminary comments.
The most important of these problems is one of vocabulary. If we define our concern as being an investigation of the origins of a centrally-planned roofed building erected over a tomb or a holy place in order to emphazise and proclaim, as the case may be, the holiness, the glory, the wealth, or the power of an individual or an event, how are we to interpret the numerous instances, among early geographers and historians, of practices which imply an architectural setting of some kind, or to explain the exact significance, in their time, of such words as mashhad or turbahT A few illustrations may help to explain the difficulty.
In his description of the provinces of the Muslim world, the i oth century geographer, al-Muqaddasi, almost always in-
cludes a more or less lengthy paragraph on mashhads7; these vary from specifically mentioned domes over tombs or sanctuaries to simple tombs, caves, or even to ribats, military monasteries which often had acquired a holy significance by their association with religious leaders, but whose architectural forms—at least insofar as we can claim to know them8—were quite different from those of mausoleums, even though their holy character led people to be buried in or near them.9 Nasir-i Khosrow also mentions that many shi'ite mashbads in Tripoli and Tyre looked like ribats.10 In these instances, it seems fairly clear that, even though these institutions played a function similar to that of mausoleums in the religious behavior of their time, they were not shrines in the same sense as the tomb of 'Ali, and their architectural type was different. In order, then, to stay within our definition of the architectural form with which we are involved, these buildings
6 This difficulty with terminology has been recently pointed out by C. Cahen in his re-edition of J.Sauvaget, Introduction a I'histoiredel'Orient musulman, Paris, 1961, p. 14.
7 Al-Muqaddasi, pp. 102-103, I30> M^, 209 to 210, 333-334* 367> 344. etc....
8 The question of the ribat as an architectural form or as a function is still unsolved, at least in the Orient. In North Africa, the works of G. Mar-cais (Note sur les ribats, Melanges R. Basset [Paris, 1923-25], pp. 395 ff.; art. ribat in Encycl. of Islam) and A. Lezine (Le ribat de Sousse, Tunis, 1956) have somewhat clarified the situation, but the degree to which the North African examples are applicable to the Cilician or Central Asian frontiers is still very uncertain.
9 For instance, Ibn al-Jawzi, Muntazam, vol. 8, pp. 214, 303-304, 325, lists one group of examples which can easily be multiplied from historical and geographic sources.
10 Nasir-Khosrow, Safar-nameh, trans. Ch. Schefer, Paris, 1888, pp. 42ft.
OLEG GRABAR
are not included in our list inasmuch as the formal and functional problems they pose are not yet solved and would here lead us astray.
But if the case of the mashhad-ribut is fairly simple, difficulties are greater when one attempts to evaluate the architectural setting suggested by the mashhads to Biblical prophets and events mentioned by Nasir-i Khosrow and Muqaddasi, by the thirteen shi'ite mashhads seen by the Persian traveler in Basrah,11 by the mashhad of the Palm (of the Prophet) and the mashhad of Vows in Baghdad,12 or by the complex sanctuaries of the Seven Sleepers,13 or by numerous other examples found in historians or geographers.14 In most instances there is clear evidence of some kind of construction, but these constructions are not always mausoleums or precisely commemorative buildings.Thus, the mashhad of the Palm in Baghdad was originally a masjid, a private mosque, in which a miracle took place; the first ascertained construction, in the ioth century, around the cave of the Seven Sleepers in Damascus was also a masjid.15 For the Palestinian examples our
11 Ibid., pp. 239-40.
12 G. Makdisi, The topography of eleventh-century Baghdad, Arabica, vol. 6, 1959, p. 289, with all appropriate sources.
13 L.Massignon, Les sept Dormants cPEphere, Revue des Etudes Islamiques, 1955.
14 For instance, in addition to the fairly easily accessible instances in geographical texts, see Ibn al-Jawzi, Muntazam, vol. 8, p. 46, for a mashhad outside Kufah, which was visited by pilgrims to Mekkah; vol. 9, p. 104, mashhad of the cemetery of the Quraysh; p. 139, mashhad of one Muhammad b. Ishaq.
15 S. al-Munajjid, Ahl al-Kahf, Majallah al-Mujm'ah al-'Ilmi al-'Arabi, vol. 31, 1956, pp. 602 ff., esp. 605.
information is less clear, but many of the mashhads were pre-Islamic constructions, simple oratories or even natural features. At the risk of being overly conservative, this list eliminates all instances of mashhad where there is no clear evidence that a specific building was erected in Islamic times for the sole purpose of commemorating a person or an event.
A very similar problem exists in reverse for a group of archaeological documents. In southern Egypt there were until very recently two buildings known as mashhads (one only still remains),16 consisting of a hall covered by nine or six domes, of a minaret, and perhaps of a court. The date of these buildings can be fixed fairly accurately, but not their purpose or use. A passage in Maqrizi may, however, help to explain them and to exclude them from our concern. The Cairene historian described a jami'al-fiyalah, "mosque of the Elephants," from Fatimid times, with nine domes over the sanctuary, whose decoration recalled the backs of elephants in caliphal processions.17 The building is obviously related in type and plan to the southern Egyptian buildings and belongs to a long series of oratories, often built in cemeteries, whose purpose was less precisely commemorative than that of a mausoleum, for they were erected for a different liturgical activity, prayer. While it is true that these constructions served, partly at least, to glorify a man or a family, they were, architecturally and functionally, different both from mausoleums and from congregational mosques. That the practice was not limited to Cairo
16 Creswell, MAE, vol. 1, pp. 144-145 and 149-152.
17 Maqrizi, vol. 2, p. 289.
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
is clearly shown by a passage in Ibn al-Jawzi relating the burial of one 'Ubaydal-lah b. al-riasan in front of his masjid in Baghdad.18
It is in this category of cemetery mosques that I propose to put a construction usually assumed to be a mausoleum, the mashhad of Sharif Tabataba, datable ca. 334/943. K. A. C. Creswell is responsible for the recovery of this curious structure, a square building with nine domes on cross-shaped piers and open on all sides except the central part of the qiblah wall, where there is a mihmb. Nearby, in a more recent building, were found inscriptions with the names of members of the Tabataba family.19 By relating this discovery to a passage in Ibn al-Zayyat describing a mashhad in which eleven members of the family were buried, Creswell suggested that the newly discovered building was that mashhad and dated it at the time of the death of the most celebrated member of the clan. One cannot quarrel with the reconstruction or with its relation with the Tabataba family. But two additional remarks may be made which alter a little its significance. First, there is in the Museum of Islamic Art in Cairo an inscription20 which refers to the "gate (? [uncertain reading]) of the maqbarah" of another member of the family, also mentioned by Ibn al-Zayyat, who died in 348/949. The term maqbarah (see below) simply means a cemetery or a part of a cemetery and this early piece of information (as opposed to the late description of Ibn al-Zayyat) merely suggests some kind of enclosure separating a family plot from other family
18 Ibn al-Jawzi, Muntazam, vol. 6, p. 370.
19 Creswell, MAE, vol. 1, pp. 11 ff.
20 Repertoire, No. 1495.
or single burials. The second point is that the plan of nine domes is to be related to that of the oratories we have just mentioned.21 Later some of these oratories may have come to be called mashhads and the tombs which were near them were covered with domes, but all we can safely suggest is that in the 10th century there were family plots in the larger cemeteries and that oratories were at times added to them for prayer and devotions.
The point of these examples is that the term mashhad, either as it is used today for certain buildings or as it was used in the past, implies different things; whenever it is possible, each instance merits a special analysis and suggests a different explanation, but it seems that, in early times, mashhad was most commonly used for any sacred place and does not always mean a specific construction over it, while, as the centuries went by, many a small building with obvious religious features, such as a mihrub, acquired the name of mashhad.
A third difficulty arises around such words as turbah, maqbarah, and even qabr.22 In the early periods the terms seem to have been interchangeable, and the
21 To the Egyptian buildings should be added the Bib Mardum mosque in Toledo, whose architectural type is the same, whose date is ca. 1000, but whose local function has not yet been explained; G. Marcais, L'architecture musulmane d'Occident, Paris, 1954, pp. 151-152.
22 For the latter term see the constant statement in Ibn al-Banna"s diary that people were buried "in the tomb" (ft qabr or biqabr) of Ahmad b. Hanbal or of other religious men, G. Makdisi, An autograph diary, Bull. School of Or. and Afr. Studies, vols. 18-19 (1956-57), vol. 2, pp. 251, 255; vol. 3, p. 34. The term obviously refers to a whole area and not precisely to the sepulchre proper.
12
OLEG GRABAR
word turbah does not mean more than a large plot in a cemetery, which was often bought in advance, in which one or more people were buried, and which could at times be separated from other similar plots by a fence, a wall, or even a portico.23 Here again the amount of construction is not always clear, and in a few instances we have included turbahs because of the probable extent of the constructions erected around the tombs, even though no clear evidence exists about constructions over the tombs. As one reads the geographers of the ioth and later centuries, it becomes quite apparent that an intense life was developing in the cemeteries of the great cities, a life which has recently been illuminated by the late L. Massignon.24 A great deal of attention was given to the places where people were buried; and it was a new form of piety to visit cemeteries and tombs, as, for instance, Abu Bakr Muhammad b. 'Ali, the last of the great Madara'yun family (d. 956-57), used to do every day in Cairo,25 or as the Hanbalites did in Damascus and Baghdad, for "the neighborhood of the saints is preferred in the state of burial, just
23 Ibn al-Jawzi, Muntazam, vol. 7, pp. 207, 240 (buying of a turbah in Kufah), 289 (a dome, which miraculously collapsed, was built over the tomb of a man buried in a turbah); vol. 8, p. 118; vol. 9, p. 10 (two turbahs next to each other); vol. 10, pp. 80 (maqbarah with apparently the same meaning), 134, 142 (turbah used in the plural apparently to indicate plots); Mas'udi, Muruj, vol. 8, p. 234. Other instances are brought out in our list when they occurred in places later to be covered with a dome.
24 L. Massignon, La Cite des Morts au Caire, Bull. Inst. Fr. Arch. Or., vol. 57, 1958.
25 H. Gottschalk, Die Mddara'iyyun, Berlin, 1931, p. 125; Maqrizi, vol. 2, p. 156; Massignon, La Cite, pp. 55-56.
as it is preferred in the state of life."26 These ceremonies, no doubt, all found certain forms of architectural expression, but the latter were far from always involving the tombs themselves.27
As a consequence of these facts, this list eliminates all buildings which cannot be clearly shown to have had, when first erected, a precise memorial connotation and to have been built by Muslims over sacred places and tombs. This rather strict definition was dictated by our primary interest in investigating the origins of the great mausoleums of later times, but it should not overshadow the fact, in itself deserving study, that in many instances a whole apparatus of guesthouses, kitchens, bakeries, enclosures, oratories, gates, even dwellings, sprang up together with the commemorative buildings and, at times, even earlier. The word mashhad referred to this whole phenomenon and a complex but different subject of research from ours would be the development of activities around holy places and the architecture these activities created.
A last preliminary remark concerns our exclusion of the earliest epigraphical document suggesting the existence of the practice of erecting domes over tombs. It occurs in the comparatively modern mauso-
26 G. Makdisi, Diary, vol. 3, p. 34.
27 The examples quoted here do not pretend to be exhaustive; many others can be brought out, including some provided by epigraphy, such as Repertoire, No. 1813, by which a qabr, tomb, is built (band) in 970 for a woman who died in 862. Is this a mere stone or brick cenotaph, such as has been illustrated byD. S.Rice, The oldest illustrated Arabic manuscript, Bulletin of the School of Oriental and African Studies, vol. 22, 1959, pi. VIII?—or should we interpret it as a true mausoleum?
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
13
leum of Dho'l-Nun al-Misri in Cairo and has been published by L. Massignon28 and studied in detail by G. Wiet.29 The inscription is presumed to have been made at the time of his death in 245/859 and includes, in addition to formulas fairly common on epitaphs, the following sentence: "he required in his authentic testament [ft wa-siyyatih al-musnadak 'anhu] that no construction be made at his tomb and no dome [qubbah] be raised over it." The rather neutral first prohibition may refer to any enclosure, while the second one deals very precisely with a dome; this inscription has been taken to be an indication of the prevalence of domes over tombs in the middle of the 9th century, and later evidence exists that such prohibitions were made by holy personages.30 But, with respect to Dhu '1-Nun's inscription, L. Massignon was struck by the expression "in his authentic testament," and especially by the suggestion that the authenticity is shown by an isnad, a chain of authority going back to the mystic himself; the implication would be that this is a later medieval forgery reflecting opinions opposed to the prevalent growth of a funerary architecture. The evidence presented in this study, as well as the fact that all other inscriptions in the building are much later, confirms this interpretation of the text of the inscription. If it is indeed a medieval forgery it is not unique. The celebrated Central Asian tombstone of Abu Zakariya al-Waraghshari, for instance, with the date 230/844 (the actual
28 L. Massignon, Etudes archeologiqnes, Bull. Inst. Fr. Arch. Or., vol.9, I9II> PP- 91~96> Repertoire, No. 440.
29 G. Wiet, MCI A Egypte II, pp. 62 ff.
30 E. Dermenghem, Le culte des saints dans I'Islam maghrebin, Paris, 1954, pp. 39, 118.
date of the death of the personage),31 has been proved to be 13 th century forgery in a recent article by M. E. Masson.32
To these problems presented by the lack of adequate historical dictionaries of Arabic and Persian, one should also add the peculiarities of our information about the early medieval Islamic world. Thanks to al-Maqrizi's description and to the researches of M. van Berchem, G.Wiet, and K. A. C. Creswell, our knowledge of Egypt, especially Cairo, is quite extensive; for Baghdad the texts are more rewarding than the monuments but have not yet been fully exploited; in Central Asia systematic explorations have recently brought to light large quantities of archaeological documents as yet insufficiently related to a rather scanty literary information. But for most of Iran and Syria few authentic literary or archaeological sources remain from before the 12th century. The extent to which these variations in the nature of the information available have unbalanced our results cannot now be said, but the possibility should be kept in mind. As we have attempted to cover the whole extent of the Islamic world, there is no doubt that many examples, especially in texts, have been missed; but it is hoped that this list may serve both to suggest a few conclusions on the origins of a major form of Islamic architecture and to illustrate a method and an approach which may be used in dealing with other types of monuments as well.33
31 Repertoire, No. 310.
32 M. E. Masson, Sredneaziatskie namogilnye kairaki, Epigrafika Vostoka, vol.11, 1956, pp. 8 if.
33 As additional information, archaeological or textual, is acquired, it is hoped that supplements and corrections to our list will be published.
OLEG GRABAR
The monuments are arranged in chronological order with an arbitrary limit set around 1150, at a time when all provinces of the Muslim world had acquired large numbers of mausoleums. Undated buildings are set at an estimated ante quern date, although it is clear that some of the shi'ite sanctuaries, especially in Iraq, preceded the earliest date known for them, and that the chronology of the numerous Central Asian mausoleums has not yet been satisfactorily established.34 A brief description follows each monument and, when it seemed indicated, a discussion of certain problems posed by individual monuments is included. The bibliography appended to each entry is not meant to be complete but simply to refer the reader to the most accessible publications.34a
LIST OF MAUSOLEUMS AND MEMORIAL CONSTRUCTIONS
1 — Jerusalem, Dome of the Rock, dated 691 a.d.This celebrated building consists of a central circular part covered with a dome and surrounded by two octogonal ambulatories. The building has been described by K. A. C. Creswell, Early Muslim architecture, Oxford, 1932, vol. 1, pp. 42-94, and its meaning was discussed by O. Grabar, The Umayyad Dome of the Rock, Ars Orientalis, vol. 3,1959. By its early signifi-
34 There is no single recent work known to me covering the whole of Central Asian architecture; the most comprehensive is by G. A. Puga-chenkova, Puti Razvitiia arhitektury Iuzhnogo Turkmenistana, Moscow, 1958, but the criteria used for the dating of the many anepigraphic monuments do not seem to me to be as yet sufficiently precise to justify all the proposed dates.
cance as a memorial to Muslim domination, by its later development into a shrine to the Ascension of the Prophet, and by its name and shape—the first qubbah of Islam —this building fully belongs to our series, even though its sources and its apparent lack of immediate formal posterity single it out as a unique creation.
2—Samarra, Qubbah al-Sulaybiyah, datable on historical grounds in 862. A central square room covered with a dome
34a Main Abbreviations. Creswell, MAE: K. A. C. Creswell, Muslim architecture of Egypt, Oxford, 1952, vol. 1.
Denike: B. P. Denike, Arhitekturnyi
Ornament Srednei Azij, Moscow 1939.
Ibn al-Jawzi: Ibn al-Jawzi, al Muntazam fi ta' rikh al-muluk wa-l-umam, Hayderabad, 1938 ff.
lutake: Trudy luzhno-Turkmenistanskoi
Arheologicheskoi Kompleksnoi Ekspeditzij, Moscow, 1952ff. (1 r vols, to date).
Le Strange,        G. Le Strange, The lands of the
Lands: Eastern Caliphate, Cambridge,
1930.
Maqrlzi: al-Maqrlzi, K. al-Khitat, Cairo,
1270 H, 2 vols. MCI A: G. Wiet, Materiaux pour un
Egypte II: corpus inscriptionum Arabi-
carum: Egypte deuxieme partie, Cairo, 1929-30 (vol. 52 of the Memoires de I'Institut Frangais d'Archeologie Orientale an Caire).
Pugachenkova:   G. A. Pugachenkova, Puti Razvitiia Arhitektury luzhnogo Turkmenistana, Moscow, 1958 (vol. 6 of lutake).
Repertoire:        E. Combe, J. Sauvaget, G. Wiet, Repertoire chronologique d'epigraphie arabe, Cairo, i93iff.
Survey: A. U. Pope, ed., A survey of
Persian art, Oxford, 1939
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
15
is surrounded by an octogonal ambulatory. The main publication is by K. A. C. Cres-well, Early Muslim architecture, Oxford, 1940, vol. 2, pp. 283-285. The identification of the building as the mausoleum built for the caliph al-Muntasir by his Christian mother has been made by E. Herzfeld, in E. Herzfeld and F. Sarre, Archäologische Reise im Tigris- und Euphratgebiet, Berlin, 1911—20, vol. 1, p. 86; and the eventual discovery there of three tombs seemed to confirm the statements of the texts that the caliphs al-Mu'tazz and al-Muhtadi were buried together with al-Muntasir (Tabari, Ta'rikh, ed M. de Goeje and others, Ley den, 1879ff., vol. 3, pp. 1498-1499, 1711, and 1823; al-Mas'üdi, Mum) al-Dhahab, ed. and tr. Barbier de Meynard and Pavet de Courteille, Paris, 1861-77, v°l- 7> P- 300)-Two problems are still raised by this identification. The first one is that, since there was no contemporary tradition of Byzantine mausoleums for emperors, one may wonder where the Christian mother acquired the idea of erecting a mausoleum over the tomb of her son, unless it be from some obscure provincial tradition; the form of the building may very well have been inspired by the earliest Islamic commemorative construction, the Dome of the Rock, as has been suggested by Creswell. The second problem is that the texts do not mention a building; they simply say that the tomb of the caliph was known and that it was the first caliph's tomb to be known35; Tabari actually even refers to the place of burial as a maqbarah, a term which, as we have
35 The same verb (azhara) is used elsewhere (Ibn al-Jawzi, Muntazam, vol. 6, pp. 45, 199; vol. 9, p. 215) without any suggestion of mausoleum.
seen, is usually difficult to interpret as referring to a built mausoleum. It is possible, of course, that funerary terminology was still quite vague at the time when the two chroniclers wrote, but, while the two points we have made do not necessarily invalidate the identifications made by Herzfeld, they raise the question of the origins of the monument and point to its exceptional character.
3—Qumm, tomb of Fdtimah, second half of the 9th century (?). The history of the sanctuaries of Qumm is quite difficult to disentangle from the legends which have surrounded them and from the fact that the full development of the city as a shi'ite shrine is comparatively late; furthermore, whatever archaeological evidence may exist there has not been collected. For the early period our best source is the Ta'rikh~i Qumm, written in 98 8-89.36 According to it (pp. 213-215), the tomb of the sister of Ali al-Rida (who died in 817-18) first received a covering of mats and then a qubbah. The latter was built two generations later by Zaynab, a granddaughter of Ali. The same person was apparently also responsible for a second qubbah next to the first one. In these two constructions six descendants of Ali al-Rida (mostly in the second generation) were buried. The exact date of the construction of these family mausoleums is difficult to establish. Zaynab outlived Muhammad ibn Musa (see below, no. 11), who had died in 908; hence the qubbahs could have been built either in the
36 The Ta'rlkh-i Qumm by Hasan b. Muhammad Qummi was published in Tehran in 1934; it was used by A. Houtum-Schindler, Eastern Persian Iraq, London, 1897, pp. 63-64, and in a compilation published in 1906 by one Muhammad Husayn Qummi.
i6
OLEG GRABAR
early ioth century or in the latter part of the 9 th. They were considerably enlarged in 961-62 by a local governor, presumably because they had become major places of pilgrimage, even though other geographers of the i oth century do not mention them.
4—Tirmidh, mausoleum of al-Hakim al-Tirmidhi, supposedly of the late 9th century. No plan of the building is known to me and the evidence for the date has been based on its decorative motives: B. V. Veimarn, Iskusstvo Srednei Azij, Moscow, 1940, p. 29, fig. 10; B. P. Denike, Arhitek-turnyi Ornament Srednei Azij, Moscow, 1939, pp. 38 ff., fig. 22. Both authors publish the same decorative fragment which they appropriately relate to the third Sa-marra style. There are, however, definite objections to be made to the date and to the identification of the tomb. Recent work by N. Herr on al-Hakim al-Tirmidhi (among other articles, A Sufi psychological treatise, The Muslim World, vol. 51, 1961, p. 26) shows that the date of his death is quite uncertain; moreover the general evidence about mausoleums over tombs of holy men shows that these were rare before the nth century, and there is, to my knowledge, no outside evidence which would indicate that a cult would have developed early around al-Hakim. As to the ornament it need not be of the 9th century merely because it resembles a Samarra style, for R. Ettinghausen has clearly shown that the "beveled" style persisted for many centuries (The "beveled" style in the post-Samarra Period, Archaeologica Orientalis in Memorial Ernest Herzfeld, Locust Valley, 1952). It is true, of course, that the ornament of the tomb differs from the great ornamental designs known from Tirmidh in the nth and 12th centuries, but there are other in-
stances of survivals of older decorative patterns in religious architecture alongside new ornamental developments.37 Thus either we have here an early mausoleum from around 900 whose present identification is incorrect or a later mausoleum with archaizing designs.
5—Najaf, mausoleum over the tomb of 'Ali. The exact date of the first construction is not certain and for a long time even the precise place of the tomb was controverted (it still is in Mas'udi, Muruf, vol. 4, p. 289). A first qubbah over the spot on which eventual agreement will be made seems to have been erected in 902 (L. Mas-signon, Explication du plan de Kufa, Melanges Maspero, Cairo, 1935, vol. 3, pp. 356-357). Whatever this first construction may have been, it was replaced some time before 317/929 by a new building sponsored by Abu al-Hayja' 'Abdallah b. Ham-däh. Ibn Hawqal's text (ed. J. H. Kramers, Ley den, 1938-39, p. 240) describes the building as a high dome on columns with a door on each side and superb tapestries and rugs. In spite of what has been usually assumed (Le Strange, Lands, p. 76; Creswell, Early Muslim architecture, vol. 2, p. 111), the text does not say that the building was a square and that the dome was held on four columns, an unlikely feature for a building of any size. It is much more likely that it was of circular plan with curtains between columns and that it was open on all sides. The plan has not been preserved as such from early Islamic times, but it can be presumed for certain mosque fountains (for instance in the mosque of Ibn Tülün,
37 M. B. Smith, Imämzäde Karrär at Büzün, Archäologische Mitteilungen aus Iran, vol. 7, 1935, pp. 65 ff.
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
17
Maqrizi, Khitat, vol. 2, pp. 268-269) and in a curious and still unclear feature of the 9th century mosque in Nishapur (al-Mu-qaddasi, p. 319); it can also be recognized in the common tholoi of Gospel manuscripts throughout the early Middle Ages, and especially in Armenia.38 In other words —and this is an important point to which we shall return in conclusion—the form suggested by the texts for the first qubbah at Najaf was not one which was precisely identified with funerary practices. It is apparently from the time of the Hamdanid construction that we may date the development of Najaf into a hallowed cemetery in which princes and simple believers came to be buried. The first mausoleum was replaced or transformed in 366/979-80— 369/979-80 (Ibn al-Athlr, sub anno, 369, unless it is a reference to the 366 reconstruction), burnt in 443/1051, but used again in 479/1086 (Le Strange, Lands, pp. 77ff.).
6—Bukhara, Mausoleum of the SumU-nids, datable before 943. This celebrated construction is generally called the tomb of Isma'il (Creswell, Early Muslim architecture, vol. 2, pp. 367ff.; L. Rempel in Bull. Amer. Inst, for Persian Art and Arch., vol. 4> 1935, PP- 199ff-)- The cleaning of the building in the thirties and discoveries in Bukhara libraries brought to light three new documents about it: a waqf rescript copied in 1568-69 relating that Isma'il had given land for the tomb of his father Ahmad and that several princes were buried there; several bodies in the tomb itself; a fragment of a Kufic inscription on a wooden plaque at the eastern entrance with the name of Nasr ibn Ahmad ibn Isma'il
38 C. Nordenfalk, Die Spatantiken Kanon-tafeln, Goterborg, 1938, pis. 24, 39, and passim.
(d. 943).To my knowledge these documents have not been published as such; their earliest mention known to me is in Denike, Arhitekturnyi Ornament, p. 8; they have been recalled in various other works, in particular the important study of V.L.Vo-ronina, Kharakteristiki arhitektury Sred-nei Azij epohi Samanidov, Trudy Akad. Nauk Tajik SSSR, vol. 27, 1954, pp. 41 ff-The exact interpretation of these documents for the dating of the monument is difficult to make, inasmuch as the waqf statement may have referred to a turbah (on which see introduction) rather than to a qubbah. While an early date before the death of Isma'il in 903 is not excluded,39 it may seem preferable, from a strictly methodological point of view, to use the earliest available archaeological document and to date the monument from the reign of Nasr (913-943). In plan the monument is a simple square with four openings and a large central dome; there are four small cupolas in the corners of the building over a gallery. The extraordinary brickwork has often been analyzed.
7—Salamiyah, mashhad. An undated inscription, assumed on epigraphical grounds to be of the first part of the 1 oth century, refers to a mashhad presumably built (the verb has disappeared) by one Abu al-Faraj 'Abd al-Wahhab b. ... 'Abbas b. 'Abd al-Samad (Repertoire, No. 949). Salamiyah is known to have been a major shi'ite center, but the inscription
39 This is the date admitted by Madame Pu-gachenkova in her latest study, Mazar Arab-ata v Time, Sovetskaia Arheologia, 19614, p. 203, whereas earlier she had been less definitive, as in the important survey she wrote with L. Rempel, Vydaiuchiesia Pamiatniki Arhitektury Uzbeki-stana, Tashkent, 1958, p. 65.
i8
OLEG GRABAR
does not say whether we are dealing with an actual construction over a holy place or, as is perhaps more likely, with some feature associated with a hallowed area.
8— Baghdad, tombs of '' Abbcisid caliphs. As in the cases of Fatimid and Bu-wayid tombs (nos. 16 and 29), the lack of remaining tombs makes an interpretation of the texts rather uncertain. With the exception of the three princes buried in Sa-marra (see above, no. 2), the early 'Abbasid caliphs and members of their families do not appear to have been buried with any particular formalities. By the 4th century of the hijrah, however, clear evidence exists of one or several turbahs reserved for them in Rusafah (Ibn al-Jawzi, al-Muntazam, vol. 6, p. 324). Some texts dealing with al-Radi, al Ta'i, his mother and son, al-Mutf, seem to imply individual turbahs (Ibn al-Jawzi, vol.7, pp.79, 139; al-Suli, Akhbar, tr. M. Canard, Algiers, 1946, vol. 1, p. 23 8), while others suggest some sort of collective necropolis (Ibn al-Jawzi, vol. 8, pp. 113, 217; vol. 10, p. 128). Some constructions existed in or around these turbahs, which were endowed, but with the single exception of al-Radi (d. 329/940-41), there is no evidence of a monument over the tomb itself. As far as al-Radi is concerned, Yaqut (Buldan, sub Rusafah) mentions the existence of a qubbah. The date of its construction is not known, to my knowledge, but even though al-Khatlb, for instance, does not mention it, it may have been built at the time of his death. The precise architectural characteristics of the dome are not known.
9— Damascus, tomb of Mu'awiyah, before 332/943-44. Mas'udi (Muruj, vol. 5, p. 14) relates that the first Umayyad caliph "was buried in Damascus by the Bab
al-Saghir; his tomb is visited to this day.. over it a bayt was erected, which is open on Mondays and Thursdays." From a further statement by the same author (vol. 7, p. 90) it can be assumed that no such construction existed in 212/827-28. It has been asserted that in 270/883-84, Ahmad Ibn Tulun built a qubbah over the tomb of Mu'awiyah.40 While this is not impossible in the light of Ibn Tulun's activities and beliefs,41 the major text supporting it (Ibn Taghri-birdi, al-Nujum al-Zahirah, Cairo, 1929, vol. 3, p. 47) simply says that the Tulunid prince erected four riwtiqs around it (lit. "over it," 'alayhi, which does not make sense architecturally) and assigned people to read the Koran by it and to keep candles lit. This first construction seems, therefore, to have been an enclosure rather than a construction over the tomb and it may even be wondered whether the bayt of Mas'udi implies much more than what Ibn Tulun had accomplished. An explanation for the choice of Mu'awiyah's tomb can easily be provided by the significance of the caliph as an anti-shi'ite hero.42
1 o—Aleppo, mashhad of Sahykh Mu-hasin, dated 3 51/962. The present building is of the 13th century (J.Sauvaget, "Deux sanctuaires shi'ites d'Alep, Syria, vol. 9, 1928), but an inscription referring to it as a mashhad has been preserved by Ibn Shaddad (Repertoire, No. 1557; in the edition by D. Sourdel, Damascus, 1953, p. 48;
40 'Abd al-Qadir al-Rihawi, Qubur al-Uza-ma' ft Dimishq, Majallah al-Majma' al-'Ilmi al-'Arabi, vol. 34, 1954, pp. 648-649.
41 O. Grabar, The coinage of the Tiilunids, New York, 19 5 7, pp. 3 3 ff.
42 Ch. Pellat, Le culte de Mu'dwiya an Hleme siecle de I'Hegire, Studia Islamica, vol. 6, 1956, pp. 53 ff.
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
19
E. Herzfeld, Materiaux pour un Corpus Inscr. Arab.: Alep, Cairo, 1956, vol. 1, pp. 193 ff.), and a legend has been preserved of the miraculous discovery by Sayf al-Daw-lah of the tomb of this obscure son of Hu-sayn. The term masbbad, of course, does not guarantee that there was an actual construction over the tomb.
11—Qumm, tomb of Muhammad b. Musa, }66/<)-/6-jj. This sanctuary, like that of Fatimah in the same city (see above, no. 3), seems to have begun as a family maqbarah, eventually being transformed into a mashhad (Ta'rikh-i Qumm, pp. 215 ff.). Muhammad himself died in 908, but it is apparently only in 366/976-77 that a dome was built over his tomb (modern Ta'rikh-i Qumm, p. 131; Houtum-Schind-ler, pp. 63-64). As in the instance of the tomb of Fatimah, this sanctuary seems at this time to have been of local and parochial significance only, since it is not mentioned by the geographers.
12—Tim, mausoleum known as "Arab-ata," dated in 367/977-78. This recently discovered mausoleum is of considerable architectural significance (G. A. Pugachen-kova, Mazar Arab-ata v Time, Sovetskaia Arheologia, 19614, pp. 198 ff., where all other studies are mentioned, the most important one being the account by the discoverer, N. I.Leonov, in Sov. Arh., i9604, pp. 186ff.). It is in the mountainous regions around the Zerafshan valley but it has not yet been possible to identify the medieval city—apparently from the ruins near the mausoleum a minor one—with any one city known from geographical descriptions. According to local lore, the mausoleum was erected over the tomb of an early Arab conqueror. The inscription has not yet been published in its entirety. The mausoleum is
a square building with a single richly decorated facade higher than the side walls and its dome is carried on the earliest known instance of a muqarnas squinch. Thus the building is an important transitional one between the Bukhara mausoleum and later "Seljuq" ones.
13— Mashhad, mausoleum of 'Ali al-Rida, before 375/985. This celebrated imam was buried by al-Ma'mun near the tomb of Harun al-Rashld (Tabari, vol. 3, p. 1030); both tombs were well known in the 9th century, but the earliest reference to major memorial construction is by al-Muqaddasi (p. 333), who wrote that the amir Amid al-Dawlah Fa'iq built a fort with houses and bazars and a splendid oratory (masjid) near the tomb. Although the text does not precisely say so, it is probable that some construction was made over the tomb. The amir is known as early as 354/ 965, but it is likely that he did not embark on any major program of construction until 372/982-83, when his situation seemed more secure. Since al-Muqaddasi wrote around 375/985, we may date the buildings around these years. The buildings were destroyed and rebuilt several times in the course of the 12th century. The first archaeological documentation from the shrine consists in an inscription dated ^512/1118 (Repertoire, No. 2978). The Mongols sacked the city, but propably not the shrine, for there are inscriptions there from 612/1215 (Repertoire, Nos. 3783-84, with an important bibliography).
14— Mosul, sanctuary of Jonah, second half of the 10th century. The history of this sanctuary is still unknown and so is the nature of its earliest constructions. But al-Muqaddasi mentions that the daughter of Nasir al-Dawlah built it up and endowed
20
OLEG GRABAR
it (p. 146). A recent archaeological study-failed, however, to discover any certain constructions earlier that the 6th century of the hijrab (S. al-Daywahji, Jami' al-Nabi Yunis, Sumer, vol. 10, 1954, pp. 26off.), but could not do more than survey the present buildings rather superficially.
15—Kerbela, mausoleum of Husayn, before 369/979-80. Our earliest definite evidence about the existence of major constructions at Kerbela certainly appears considerably after the fact. Already under Härün al-Rashid financial support was given by the caliph's wife to people taking care of the tomb (Tabari, vol. 3, p. 752) and in 236/850-51, al-Mutawakkil had the tomb leveled and the buildings around it destroyed (Tabari, vol. 3, p. 1407); these texts do not, however, indicate the existence of a mausoleum. By the time of Ibn Hawqal (978) and from Ibn al-Athir's account of the year 369/979-80, it is evident that a large masbhad existed to which cAdud al-Dawlah had made repairs or additions (Ibn Hawqal, p. 243). The archaeological evidence (A.Nöldeke, Das Heiligtum al-Husayns zu Kerbela, Türkische Bibliothek, vol. 11,1909; M. Streck, Kerbela, Festschrift E. Sachau, Berlin, 1915, pp. 393 ff., with interesting comments on the possible pre-Islamic origins of the cult) is practically nil, but it may be suggested that no major construction existed in 951, since al-Istakhri does not mention any. It is quite likely that Adud al-Dawlah was responsible for its recreation, as is suggested by later sources (Mustawfi, Nuzhat al-Qulüb, London, 1919, p. 39). There is no textual proof that an actual mausoleum was built over the tomb, but, if we recall that this is the time when many other shi'ite tombs acquired one, it can be assumed.
16— Buwayhid tombs, second half of the 10th century. The major evidence for the existence of Buwayhid mausoleums comes from al-Muqaddasi, who relates that, at Rayy, the Daylamite princes built high and solid domes over their tombs and lower ranking princes erected smaller ones (Muqaddasi, p. 210). On the other hand, under the year 388/998, Ibn al-Athir tells that Samsam al-Dawlah was eventually buried in the turbah bani Buwayb, whereas Adud al-Dawlah, Baha al-Dawlah, and several other members of the family were buried in the Najaf-Kufah cemetery, near the shrine of cAli; Ibn al-Jawzi refers to at least cAdud al-Dawlah's sepulchre as a turbah (Ibn al-Jawzi, Muntazam, vol. 7, p. 149). No definite archaeological evidence exists concerning these mausoleums. Herzfeld claims to have seen the copy of a wooden door presumably from the mausoleum of Adud al-Dawlah (MCIA: Alep, p. 157, n. 1), but his information is not clear; it is also possible that the wooden panels with unusual inscriptions from the year 363/973-74 (Repertoire, Nos. 1831-32) may have come from Buwayhid cenotaphs, but they do not imply the existence of large constructions. Since no archaeological evidence exists and since the texts do not fully agree, the only safe conclusions to draw would be to assume that there were Daylamite princely mausoleums which may indeed have reflected by their size the importance of the princes buried in them, that these mausoleums were probably grouped together, but also that they are more clearly ascertainable for Iran than for Iraq, where many of the principal Bu-wayhids were buried.
17— Hebron, sanctuary of Abraham, before 985. The early history and develop-
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
21
merit of the greatest Muslim sanctuary dedicated to Old Testament prophets and to their wives has been greatly complicated by the Crusades and the Muslim recon-quest (H. Vincent and H. MacKay, Hebron, Paris, 1921, p. 159ff.). From the most trustworthy early description, by al-Muqaddasi, we can safely assume that in the latter part of the 10th century there was a stone dome built by Muslims (al-Muqaddasi insists on this point) over the tomb of Abraham, but presumably not over that of other prophets. The latter were included in the mashhad which contained also hostels, bakeries, and various other institutions necessary for the upkeep of pilgrims (al-Muqaddasi, p. 172; G. Le Strange, Palestine under the Moslems, London, 1890, pp. 309ff.).
18— Sarakhs, mausoleum (?) of an uncle of 'Ali al-Ridu, before 985. Al-Muqaddasi writes (p. 333) that there was a tomb at Sarakhs on which a mashhad was built. It is likely that it included a dome.43
19— Central Asia, mausoleum of Ahmad, late 10th century. This building, a simple square with apparently two entrances facing each other, is only known through an unmarked photograph remaining in Leningrad (Pugachenkova, pp. 178-179, ill. p. 179). On its facade there is an inscription in brick, on which the word ahmad can be read, hence the name given to the building. It is built in typical "Kho-rasani" brickwork. Its date is impossible to determine with certainty, but a very late 10th century one is not improbable.
43 For reasons that do not appear clearly, B. Spuler, Iran in Friihislamischer Zeit, Wiesbaden, 1952, p. 181, n. 11, suggests that the Sarakhs in question was near Qazvin. The point, however, is not important for our purposes.
20— Chehur Juy (anc. Amul on the Oxus), anonymous mausoleum, late 10th century (?). This small (6.20 by 6.20 meters) construction with a heavy low dome is remarkable in several ways, even though of very mediocre execution. It has three doors, the qiblah side being provided with a mihrab; two large pilasters transform one of the sides into a sort of embryonnic facade (Pugachenkova, pp. 177-178). The proposed date in the late 10th century has to be accepted with caution.
21— Near Merv, mausoleum known as Kizbibi (now disappeared), late 10th century. This simple construction (7.50 by 7.50 meters) has a single entrance; inside there is a domed room with four deep recesses giving it an almost cruciform shape. A date in the 10th century is apparently suggested by the brick technique (Pugachenkova, p. 175).
22— Merv, mausoleum in the Imam bůb cemetery, late 10th century (?). This building also has now been destroyed and is only known through photographs (Pugachenkova, pp. 179-180). It is archaic in that it has four openings, and its brick technique (all layers seem to be one instead of two bricks thick) differentiates it somewhat from other buildings assigned to this period.
23— Mestorinn, mazar ShirKabir, late 10th century (?). Madame Pugachenkova, who has studied the building (Pugachenkova, pp. i68ff.; Iutake, vol. 2, pp. i94ff.), proposes an earlier date, perhaps even in the late 9th century, but its comparatively complex character would rather suggest a somewhat later date, following in that Madame Krachkovskaia's analysis of the epigraphy (V. A. Krachkovskaia, Evoluzia kuficheskogo pisma, Epigrafika Vostoka, vol. 3, 1949, p. 17). As it is now the mazar
22
OLEG GRABAR
consists of several parts, of which the most ancient is a single domed hall with a cupola carried on several recessed squinch arches; the walls were originally covered with stucco, of which only a few fragments have remained; these can properly be related to ioth century stucco in Iran, as in Nayin (Survey, pis. 265 ff.); a single entrance was not on the axis of the building, which is otherwise provided with a large mihrab. Whether this hall was the original mausoleum over some unidentified spot or a prayer hall attached to a tomb is not very clear from the available evidence, but the later history of the site with several additions to the original construction and its character as a center for pilgrimage until the 19th century strongly suggest that it had commemorative connotations from the very beginning. The city in which it is found has been identified as the Dihistan of the Hudud al-Alam (tr. V. Minorski, London, 1937, p. 133) and of al-Muqad-dasi (al-Muqaddasi, pp. 358-359).
24— Mizdakhuneh (Kirghizia), ioth or early 1 ith centuries. A small mausoleum is mentioned there by various authors (for instance, E. A. Davidovich and B. A. Lit-vinskij, Ocherki areologij. raiona Isfara, Trudy Akad. Nauk Tajik SSR, vol. 35, 1955, p. 187), but the original publication has not been available to me.
25— Biskra, Mausoleum of Sidi Oq-bah. Neither the date of the building nor the original shape of the mausoleum to the great conqueror of the Maghrib seem to be clearly ascertained; the earliest remains, a wooden gate datable ca. 1000, may be of a mausoleum but also of an enclosure without major construction over the tomb itself (G. Marcais, Le tombeau de Sidi Oq-ba, Annales Inst., Et. Or., Alger, vol. 6,
1939-41, reproduced in Melanges d'his-toire et d'archeologie, Alger, 1957, vol. 1, pp.i5iff.).
26— Cairo, mausoleum of Abu al-Fadl J a'far, 391/1001. The evidence gathered by G. Wiet (MC IA Egypt II, p. 101) permitted him to conclude that a mausoleum had been built over the tomb of this member of the remarkable family of viziers, the Banu al-Furat. The texts usually mention the word turbah and it is possible that we have here another instance of a family plot with surrounding constructions rather than a true mausoleum over the tomb.
27— Gunbadh-i Qäbüs, dated 397/ 1007. This earliest and magnificent example of tower-tomb is called a qäsr in the inscription (Repertoire, no. 2118). Its significance as a work of art and as a peculiar monument, at the same time mausoleum and royal symbol, has been fully analyzed by E. Diez and Max van Berchem (Chura-saniscbe Baudenkmäler, Berlin, 1918, pp. 39ff. and iooff.) and later by A.Godard (Survey, pp. 967 ff.).
28— Baghdad, mausoleum of Abu Dä-'üd b. Siyämard, presumably from the time of his death in 399/1008-09. Ibn al-Athlr relates that he was buried near the qabr al-nudhilr (see above, p. 10) and that its qub-bah was well known.
29— Resget (Mazanderan), anonymous mausoleum, 400/1009-10 (?). This building, whose inscription has mostly disappeared, belongs to the category of tower-tombs on a circular plan (A. Godard, Les tours de Lajim et de Resget, Ahtär-e Iran, vol. 1, 1936, pp. io9ff.).
30— Cairo, saba' banät, ca. 1010. Four square buildings covered with domes still stand and two others have been excavated. They are all alike and almost of the same
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
23
size with a door on each side; the largest is provided with two small mihrabs, one on each side of one of the doors. All the mausoleums are surrounded with an enclosure with a single door, whose sill has been preserved (Creswell, MAE, vol. 1, pp. xoyiL). The buildings have been identified with the seven mausoleums said by Maqrlzi to have been built over the graves of members of the Maghrebi family executed by al-Ha-kim (Maqrlzi, vol. 2, p. 459). The event took place in 1010 and such is the date suggested by Creswell for the mausoleums. Architecturally, of course, these simple constructions with their squinches and octagonal drums with windows can well be assigned to the first half of the 11 th century. Whether, on the other hand, these mausoleums were likely to have been erected immediately after the execution of the Maghrebis is perhaps less certain and they might have to be dated after al-Ha-kim's death in 1021.
31—Cairo, mausoleums of the Fatimid caliphs. From the point of view of our investigation of monumental memorial constructions, the exact position of the tombs of the Fatimid caliphs is difficult to ascertain. The main passages in Maqrlzi which deal with the subject (Maqrlzi, vol. 1, pp. 407-408; vol. 2, pp. 49 and 442-443) are clear enough in indicating that the caliphs and members of their families were buried in special areas, the turbah al-Za'farUn, which was part of the palace complex, the extreme south of the Qarafah cemetery (Wiet, MCI A Egypte II, p. 132), or, later, the area to the north of the Bab al-Nasr. It is also clear that mosques were built nearby, as the magnificent mosque built in 366/ 976-77 by the mother of the caliph al-Azlz (Maqrlzi, vol. 2, p. 318), and that a
considerable amount of money was put into the servicing with candles and chanting of the tombs of the imams in the palace itself (Maqrlzi, vol. 1, p. 408). But most of these texts are not clear on what precise type of structure, if any, was built over the tomb proper. Ibn al-Jawzi (Muntazam, vol. 7, p. 156) does mention that in 380/ 990-91 the great vizier Ya'qub b. Yusuf, was buried in the qubbah inside the palace which had been prepared for the caliph al-Azlz, but, since I do not know of other texts describing the palace resting places as qubbahs, one may wonder whether the work qubbah is not meant here in its later sense of "mausoleum" rather than in its precise sense of "domed chamber," or whether Ibn al-Jawzi's information is altogether correct. One may, then, conclude that greater importance was given by the Fatimids to the tombs of their caliphs than by the Abbasids. The existence of funerary chambers in the palace, while obviously related to the common practice of burials in houses, had an added significance in a regime which emphasized lineage so much. But there is no fully documented description of these rooms and their actual shape is not clear. It is quite possible that the Fatimids followed to a degree, intentionally or not, the Byzantine practice of one or two major mausoleums fitted into some other construction (the church of the Holy Apostles in Byzantium, the palace in Cairo), in which the tombs of the caliphs and of members of their families were put. As in Byzantium, the rule was probably not absolute (Ph. Grierson, Tombs and obits of Byzantine emperors, Dumbarton Oaks Papers, vol. 16, 1962). The subject deserves a more complete textual study than has been accomplished so far.
24
OLEG GRABAR
32— Mil-i Radkan, mausoleum of the ispahbad Muhammad b. Wandarin Bä-wand, 407-11 /i 016-21. This remarkable example of the tower-tomb, quite different from the slightly earlier Gunbadh-i Qäbüs, is called by three different names in its inscriptions: mashhad (Repertoire, no. 2312), qasr (Repertoire, no. 2313),44 and gunbadh (in the pahlevi inscription read by Herz-feld). Its architectural and symbolic meanings have been discussed by Diez and van Berchem (Chursanische Baudenkmäler, pp. 36ff. and 87ff.).
33— Lajlm, imäm-zadeh 'Abdallah, dated 413/1022. It is in fact the tomb (qabr in the inscription, Repertoire, no. 2331) of one Shahriyär b. al-eAbbäs and another instance of the circular tower mausoleum (A. Godard in Athür-e Iran, vol. 1, pp. io9ff.).
34— Damghan, Pir-i 'Alamdär, 417 or 419/1026-27 or 1029. Called both a qub-bah and a qasr, this mausoleum to a häjib, Muhammad b. Ibrahim, is also a circular tower-tomb (Repertoire, no. 2352; Survey, pi. 339 B; cf. also Bull. Amer. Inst, for Persian Art and Arch., vol. 4, 1936, pp. i39ff.).
35— Ghazni, tomb of Mahmud, in or shortly after 421/1030. The exact shape of the mausoleum built over the tomb of Mahmud by his son Mas'ud is not known precisely, but it was most probably a cupola over a square plan (M. Nazim, The life and times of Sultan Mahmiid of Ghazna, Cambridge, 1931, pp. 124,167). The epigraphi-cal evidence derives from the sarcophagus and from wooden doors later taken to India (Repertoire, nos. 2377, 2379, 2380).
44 See the modified reading by E. Herzfeld, Postsasanidiscbe Inschriflen, Archäologische Mitteilungen aus Iran, vol. 4, 1933-34, pp. 140ff.
36— Sangbast, mausoleum of Arslün Jadhib, dated 419/1028. So far as I have been able to gather there is no absolute certainty about the identity of the personage for whom this mausoleum was built, but its simple square with a dome on squinches and a door on each side makes the date plausible (Survey, pp. 923, 986-988,1275, and pi. 260B, C; Diez, Churasanische Baudenkmäler, pp. 52-55; Schroeder in BAI-PAA, vol. 4,1936, pp. 136ff.).
37— Yazd, Duväzdeh Imam, dated 429/1037. This square mausoleum with a single door and a mihräb is of considerable importance for being the earliest known building in Central Iran with a muqarnas squinch (Survey, pp. 1001-1005, pi. 274). Its inscriptions have not yet been published, and no information exists with regard to the personage in whose honor the mausoleum was built.
38— Baghdad, Käzimayn, mausoleums of Müsa al-Käzim (d. 800) and of Muhammad al-Taqi (d. 219/834). The earliest information known to me about this celebrated sanctuary is that, during the riots of 443/1051, it was plundered of its rich treasures and that, after the plunder, the two wooden cupolas and the sarcophagi over the tombs of the imams were burned, one of the sarcophagi having been opened. At the same time most of the surrounding tombs were destroyed, including tombs of Buwayhid amirs and of members of the 'Abbäsid family (Ibn al-Athir, sub anno 443; Ibn al-Jawzi, Muntazam, vol. 8, p. 150). The extent of the establishment thus destroyed suggests that the sanctuary, including the domes, had been established for some time. The most likely period is the end of the 10th and the beginning of the 1 ith centuries, in all probability before
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
25
the growth of the shrine to Abu Hanifah (see next item).
39— Baghdad, mausoleum of Abu Hanifah, 437-38/1045-47. The earliest reference to some construction by the tomb is in al-Muqaddasi (p. 130), who relates that one Abu Ja'far al-Zammam erected a suf-fah apparently to the side of the tomb; it was probably some kind of platform for prayer or gatherings. The next information comes from Ibn al-Jawzi who relates (Muntazam, vol. 8, p. 245) that in 437 or 438/1045-47 new constructions were made which include a roof (saqaf) over the tomb of the holy man; it is interesting to note that the work was sponsored by Turks. In 459/1067, a complete reconstruction took place which involved strong foundations, a brick mausoleum, a madrasah, and other institutions which required considerable acquisitions of land and the transfer of many tombs. Whatever the later history of the sanctuary may have been,45 it would seem, then, that the tomb of Abu Hanifah had acquired some kind of architectural recognition as early as in the late 1 oth century, but it is only in the early decades of the 1 ith that we have specific documentation for a construction over the tomb. In 1067 a complete institution was created there, one of the earliest of its kind in Islam, as has been shown by G. Makdisi.
40— Kermin, mausoleum of Mir Say-yid Bahr, datable in the first half of the 11 th century. The city is probably the Kar-miniyah of medieval texts, between Buk-
45 G. Le Strange, Baghdad under the 'Abbasid Caliphate, Oxford, 1900, pp. 159ff.; M. Jawad, al-Nizamiyah, Sumer, vol. 9, 1953, p. 324; G. Makdisi, Muslim institutions of learning in eleventh-century Baghdad, Bull. School of Or. and Afr. Studies, vol. 24, 1961, pp. 19 ff.
hara and Samarkand (Le Strange, Lands, p. 468). The building is a simple square (4.50 by 4.50 meters inside) with a single entrance, a developed facade, and no gallery. The original publication is by V. Nil-sen, Mavzolei. . ., Materiály po istorij i teoriy arhitektury Uzbekistana, Moscow, 1950; it was discussed by V. L.Voronina, p. 47 and L. Rempel, Arhitekturnyi ornament Uzbekistana, Tashkent, i96i,p. 152, both of whom tend to date the building in the late 10th century (cf. Pugachenkova, p. 272). The building is provided with an inscription of which, apparently, only the basmalah is readable (V. A. Nilsen, Monu-mentalnaia Arhitektura Bukharskova Oa-zisa, Tashkent, 1956, pp. 37ff.). It is difficult to assign a precise date to this construction. However, because of the existence of a rather developed facade extending beyond the limits of the walls of the building, a later date than the one proposed by the discoverers seems preferable.
41—Aswan, anonymous mausoleums, datable in the first half or the middle of the nth century. A full description of these 49 buildings and of the incidents which led to the disappearance of their inscriptions and tombstones can be found in Creswell's summary of and additions to Monneret de Villard's original publication (U. Monneret de Villard, La necropoli musulmane di Aswan, Cairo, 1931; Creswell, MAE, vol. 1, pp. 131 ff.). There is considerable variety in the detail of the mausoleums, but a unity of basic forms, squares covered with a dome on squinches and on a drum; some buildings have four doors, others only one and a mihrab. The construction material is usually cheap mud brick for the walls, baked brick for arches and vaults. The main
26
OLEG GRABAR
problem posed by the Aswan mausoleums is the evaluation of their significance. The rather primitive construction and the large number of buildings suggest that we are not dealing here with the sanctuaries of single saints and heroes, as in the cAlid shrines of Iraq. For some reason, considerable numbers of people of more modest origins buried in Aswan thought it appropriate to have mausoleums erected over their tombs or had memorials built by family or followers. Is it possible to identify some peculiarity of Aswan which would explain this development? Or is it merely accidental that so many qubbahs have survived from there, whose existence we should assume for other cities as well? To the last question the evidence presented so far and the rather detailed account which concern at least the cities of Baghdad and Cairo seem to indicate that there was no general practice of mausoleum building. It is, therefore, something precisely related to the city of Aswan which must explain its memorial constructions. The peculiarity of Aswan was that it was on one of the pilgrimage routes to Mekkah and that it was the last major Muslim city before Nubia. This latter quality made it a thughr, a frontier area (Maqrizi, vol. i, pp. 197-199) where holy men and warriors gathered in order to guard the frontiers of Islam, to partake of an intensive if little known spiritual life, and perhaps to die as martyrs. There were other institutions which developed around the frontier areas, in particular the ribat, a sort of military monastery which had been analyzed in detail in the Maghrib. In fact Aswan was known as a ribat:*6 Because of the holiness attached to them, ribats were often places in or near which people were buried and they became,
at times, mashhads, as we have seen before (see above, p. 9; numerous examples of the relationship between the ribut and burials can be found in the necrologies of Ibn al-Jawzi and Ibn al-Athir). For Aswan, we may then suggest that the mausoleums express one aspect of the "frontier" spirit, the identification either of those who had died for the faith or of those who guided warriors for the faith. Since many of them were of modest origins, since many of the ribats and other related institutions were private foundations, and since no cults developed around most of them, the simple character of the buildings, which were mere memorials, can easily be explained. It remains to be seen whether similar phenomena occurred in other frontier areas. To be true, the Cilician and Central Asian frontiers have been the scene of so many invasions that many earlier monuments have been destroyed, but the very remarkable number of recently discovered mausoleums from Central Asia could, partly at least, be attributed to the same causes.
42—Near Tolas (modern Janbul), mausoleum of Babaji-khutun, first half of the nth century (?). This rather unusual building is a square (6.80 by 6.80 meters) with a curious ribbed low dome in brick, an inscription on the facade which is said to contain the name Babaji-khatun, and a simple decoration of arches and medallions (A. Margulan and others, Arbitektura Ka-zakhstana, Alma-Ata, 1959, pp. 97 ff.).The date is hypothetical and especially difficult to establish on stylistic grounds for so remote a region. Its justification resides
46 J. Maspero and G. Wiet, Materiaux pour ...la geograpbie de I'Egypte, Cairo, 1914-19, p. 15.
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
^7
mainly in the fact that the second half of the century witnessed elsewhere in Central Asia considerable modifications in decorative techniques and esthetic values which are not apparent here. But, of course, arguments of that order can only be used with caution in dealing with provincial centers.
43— Rayy, circular tower-tomb, now destroyed. The inscription known through Coste's transcription (P. Coste, Monuments modernes de la Perse, Paris, 1867, pi. LXIV) and a photograph by Curzon (G. N. Curzon, Persia, London, 1892, vol. 1, p. 350) has been published in the Repertoire (no. 3153) as being of 546. In reality the first digit is missing and, if one considers the extreme simplicity of the construction, 446/1054-55 may seem preferable. The tomb is called a qubbah and the builder was one cAbd al-Jalll b. Faris, the treasurer.
44— Damghfin, so called Chehel Dukh-taran, dated 446/1054-55. The exact date of this well-known circular mausoleum is not very clear (Repertoire, nos. 2572-73; E. Herzfeld, Khorassan, Der Islam, vol. 11, 1918, p. 168; Survey, pi. 340a), just as the name of the personage buried in it is uncertain. From the inscription and from the popular tradition attached to it, it seems, however, likely that it was built for a whole family.
45— Abarquh, Gunbadh-i 'Ali, dated 448/1056-57. It is an octagonal building built for a descendant of the Flruzanid dynasty, Hazarasp b. Nasr, by his son (Repertoire, no. 2582). The building (Survey, p. 1023, pi.335-336; A.Godard, Abarquh, Athar-e Iran, vol. 1, 1936, pp. 49ff.) is remarkable for being in rubble rather than in the more common brick technique of the time.
46— Baghdad, mausoleum of the shaykh Ma'ruf al-Karkhi, before 459/ 1067. The earliest information available to me on this mausoleum is that it burned down accidentally in 1067, being mostly in wood (Muntazam, vol. 8, p. 105; G.Mak-disi, The topography of nth century Baghdad, Arabica, vol. 6,1959, p. 286; esp. M. Jawad, al-Imarah ..., Sumer, vol. 3, x947> PP- 54As early as 345/956-57 (Muntazam, vol. 6, p. 382, describing the burial of one Muhammad b. 'Abd al-Wahid al-Zahid in the suffah which faced the qabr of the shaykh) some sort of construction existed in the area of the tomb, but it does not seem to have consisted in a mausoleum.
47— Imam Dur, mausoleum of Muslim b. Quraysh, ca. 478/1086. Whether the shrine was actually built for the prince whose name appears on the inscription (Repertoire, no. 2756) or for Muhammad b. Musa, a son of the fifth imam (E. Herzfeld, Damascus I, Ars Islamica, vol. 9, 1942, pp. 18 ff.), or for both, is not clearly established. The building itself is a square covered with the earliest instance of the spectacular muqarnas dome which will be so characteristic for Iraqi and Syrian architecture in the following century.
48— Cairo, so-called mosque al-Juyu-shi, dated 478/1085. The recent study of this well-known monument by K. C. A. Creswell has posed anew the question of its purpose. Max van Berchem, who was the first to have brought attention to it, had considered it to be a mausoleum over a tomb with an attached small sanctuary consisting of a court, a hall for prayers with a dome in front of the mihrab, in other words a typical martyrium (Max van Berchem, Une mosquee d'epoque fatimite,
28
OLEG GRABAR
Bulletin de lTnstitut Egyptien, vol. 2,1888; Notes d'archeologie arabe, Journal Asiati-que, 8th series, vol. 17, 1891, pp. 479-481; K. A. C. Creswell, MAE, vol. 1, pp. 15 6ff.). It is quite true that the inscription calls the building a mashhad (Repertoire, no. 2753; Creswell mistakenly calls it a zawiyah), but the mausoleum, which is found near it, has been shown by Creswell (p. 157) to be a very late addition. It is, therefore, the mosque itself which is a memorial structure; in other words an architectural form created for prayer is used with a commemorative function emphasized by the founding inscription. The problem is to discover what was being commemorated. An interesting peculiarity of the shrine is the unusual choice of Koranic quotations found in its inscriptions, as Max van Berchem has already noticed. The two passages on the dedicatory inscription on the facade (LXXII, 18, and IX, 109) are standard passages found in mosques and refer to the masajid Allah. But the passages found in the domed room in front of the mihrcib are, with one exception (XXIV, 36-37), much rarer in monumental inscriptions: XI, 36: "My counsel, if I want to counsel you, will not profit you, if God wants to keep you astray; He is your Lord and you will be returned to Him"; XXIV, 11: "Those who spread the slander are a small group among you; do not think it a bad thing for you; in fact it is good for you; to every man who carried it what he has earned of the sin and to the one who had the greatest share an awful doom"; X, 24: "Yet when He had delivered them [people in a storm] behold they rebeled on the earth with untruth; O men, your rebellion will only turn against you; a [brief] enjoyment of the life of the world, then unto Us is your return
and We shall proclaim to you what you used to do"; XLVIII, 1-5: "Lo We have given thee a signal victory, that God may forgive thee of thy sin that which is past and that which is to come, and may perfect His favour unto thee and may guide thee on a right path; and that God may help thee with strong help; He it is Who sent down peace of reassurance into the hearts of the believers that they might add faith unto their faith; God's are the hosts of the heavens and the earth, and God is ever Knower, Wise; that He may bring the believing men and the believing women into Gardens underneath which rivers flow, wherein they will abide, and may remit from their evil deeds; that in the signs of God is the supreme triumph"; and finally, XXXV, 39: "He it is who made you regents [khalz'if] on the earth; so he who disbelieveth, his disbelief be on his own head; their disbelief increases for the disbelievers, in their Lord's sight, anything but abhorrence; their disbelief increases for the disbeliever naught but loss." In the center of the cupola the names of Muhammad and 'Ali are intertwined. The quotations from the Koran all emphasize two points: the doom prepared for slanderers, disbelievers, dissenters, and rebels, and the victory given to the Prophet, to the caliphs, and to the pure men for whom the building was erected. If one recalls the remarkable position of the mashhad on the edge of a cliff overlooking the whole city, it may be suggested that its purpose was to symbolize the victory achieved only a few years earlier by Badr al- Jamali in the name of the caliph al-Mustansir over the rebellions and disorders which for a long time plagued the Fatimid empire. Like the Dome of the Rock, then, the "mosque al-
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
29
Juyushi," as an expiatory chapel and a symbol of victory, was a building commemorative of a precise historical event. As the event receded into time and lost its pungency, the memory of the purpose of the building faded away and, in order to fit with the more common commemorative constructions, it acquired a funerary sense and a small mausoleum with a tomb was even added to it.
49— Aleppo, maqam Ibrahim, 479/ 1086. Only an inscription with the names of Malikshah and Nizam al-Mulk (Repertoire, no. 2760; E. Herzfeld, MCI A Alep, p. 177) reflects this early sunnite memorial construction in Aleppo (J. Sauvaget, Alep, Paris, 1941, p. 107).
50— Salamiyah, mashhad of 'Ali b. Jarir (?), 481/1088. This mashhad, presumably built over the tomb of an obscure officer, is only known through an inscription (Repertoire, no. 2772).
51— Tunis, mausoleum called "Sidi Bou Khrissan," 486/1093. This small square building was apparently open on all four sides and is dated by an inscription (S. M. Zbiss, Documents d'architecture fatimite d'Occident, Ars Orientalis, vol. 3,
i959>P-3°)-_
52— Tunis, cupola known as Msid al-
Qubbah, undated, but typologically related to the preceding (Zbiss, Ars Orientalis, vol. 3, p. 30).
53— Cairo, mausoleum of Sayidah Nafisah, 482/1089-90. This celebrated Cairene mausoleum no longer remains in its original shape, which was a very simple square construction with a mihrcib and one or three doors (D. Russell, A note on the cemetery of the ''Abbusid caliphs, Ars Is-lamica, vol. 62, 1939, pp. 168 ff.). Maqrlzi (vol. 2, p. 442) reports that the 'Abbasid
governor at the time of Nafisah's death (206/821-22), 'Ubaydallah b. al-Sari, had already built a cenotaph around the tomb with marble plaques, whose inscriptions have been assumed to have been copied on a later refection (Repertoire, no. 162, with important comments by G. Wiet in MCI A: Egypte II, pp. 33 ff.). It is doubtful, however, whether a real construction was erected at that time, not only because the general evidence collected here would contradict it, but also because Maqrlzi transmits the story strictly as a rumor without documentation. Furthermore, the remaining inscription, presumably an Ayyubid copy, is suspicious by its use of the word mashhad and by its curiously antiquarian character in that, as G. "Wiet has pointed out, the Koranic quotation stops precisely on the spot where, according to the legend, the holy woman had stopped when she died. There is a literary flavor to it which makes one question its archaeological value. The second inscription from the shrine also comes from Maqrlzi (vol. 2, p. 442; Repertoire, no. 2776); it refers to the construction of a door under al-Mus-tansir and gives a precise date. Its archaeological index is much greater than that of the first inscription and, while it is possible, even likely, that there were earlier constructions over the tomb of the holy woman, they cannot be dated more precisely.
54—Ascalon, mashhad of the head of Husayn, 584/1091-92. Nothing is known of the shape of this great Fatimid mausoleum and much in its purpose is still not clear. Even its exact date is a matter of controversy, although the very clear epi-graphical indications of the minbar made for the shrine and eventually brought to
3°
OLEG GRABAR
Hebron make it certain that the building existed in 1091-92 (Repertoire, nos. 2790-91; M.V. Berchem, La chaire de la mosquee d'Hebron, Festschrift E. Sachau, Berlin, 1915, pp. 298 ff.; G.Wiet, Notes d'epigra-phie, Syria, vol. 5,1924, pp. 2i7ff.).
55— Mehmandust (near Damghan); a ten-sided tower-tomb was found there with a date which was read as 49 x by Schroe-der (E. Schroeder, BAIPAA, vol. 4, 1936, p. 135, fig. 4) and published as 490 by Godard (A. Godard, Athar-e Iran, vol. 4, 1949, p. 259, fig. 210). The building is thus to be dated ca. 1096.
56— Iarti-Gunbadh (near Sarakhs), dated 491/1098. This comparatively large (12 by 12 meters) mausoleum is remarkable in many ways. First it is one of the few early Central Asian mausoleums to provide us with a dated inscription. Second, its zone of transition from square to dome has a rather subtle variation on the muqar-nas type first seen at Tim and Yazd (see above, nos. 12 and 37) and later throughout Central Iran. Third, it has elaborate decorative compositions on all four sides, even though there v/as only one entrance; the type of composition (corner columns, pilasters framing the central part, etc.) relate the monument to the 10 th century Samanid monuments rather than to the tradition of a single facade typical of 11 th century monuments in the area. Fourth, for unexplained reasons, its eastern facade was more elaborate than the southern one on which the entrance is found (G. A. Pu-gachenkova, Iarty-Gumbez, Epigrafika Vostoka, vol. 14, 1961, pp. 12 ff.).
57— Aysha-bibi mausoleum, near Ta-las. This heavily decorated brick mausoleum is comparatively small (8 by 8 meters) but remarkable for two features: a facade
wall which may have been higher than the dome, and an extraordinary decoration of terracotta fragments in which ornamental principles of Samanid times appear to have been executed in a new decorative technique (A. Margulan and others, Arhitektura Kazakhstana, pp.99ff.; Denike, pp.98ff.). Like the group of buildings which follows, this mausoleum is difficult to date and could in fact be put anywhere in the nth century.
58— Say at (on the right bank of the lower part of the Kafirnigan valley in Tajikistan). Two mausoleums (ca. 10.50 meters to the side) are preserved there and are connected with each other by a vaulted passageway. The interest of the building resides in its brickwork, especially in the squinches, and in a 16-sided zone above the squinches. Parts of its facade were provided with a decoration which is certainly later. The mausoleums are not dated (A. M. Be-lenitzkij, Mavzolei u seleniia Saiat, Trudy Sogdiisko-TadjikskoiArheologicheskoi Ek-speditzij, vol. 1, Moscow, 1950, pp.207ff.; also in KSIIMK, no. 33, 1950, unavailable to me).
59— Sarakhs, mausoleum of Abu Fadl. This very monumental construction, described in detail by Madame Pribytkova (A. M. Pribytkova, Pamiatniki Arhitek-tury XI veka v Turkmenij, Moscow, 1955, pp. 6ff.), consists of a square of heavy brick masonry carrying a gallery and a superb dome; while its sides are decorated with blind arcades, its facade is of a new type with projecting towers making a sort of entrance eywdn. It has been dated by Madame Pribytkova and others who have followed her in the second quarter of the nth century. The personage for whom it is supposed to have been built (see refer-
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
31
ences in review by O. Grabar, Ars Orienta-lis, vol. 2, 1957, p. 545) died in 1023, but, so long as there is no epigraphical or textual justification for its attribution, the identification, which is of popular origin, is hardly more than a terminus post quern. On the other hand, the tremendous development of the facade relates this building to definitely dated 12th century examples. It is, therefore, mostly because of the sobriety of the ornamentation and of the decoration of arcades on the sides of the building that a late nth century date is tentatively proposed.
60— Mihnah (near mod. Mean; Le Strange, Lands, p. 394), mausoleum of Abu Sa'id. In plan this other remarkable mausoleum is similar to the preceding one (Příbytková, pp. 20 ff.; Pugachenkova, pp. 272 ff.). Some of its proportions are different and its facade and side decorations are quite different. Its most remarkable characteristic, however, is that it has a double brick dome. As in the preceding example and for somewhat the same reasons, a late nth century date seems preferable to the earlier date proposed by some of the Russian scholars who have written about the building.
61— Astan-baba near Karkhi (Le Strange, Lands, p. 404, so-called mausoleum of Alamberdar. This is another instance of an anonymous mausoleum of heavy proportions with a slightly developed facade, a low and wide dome, and most original brick squinches (Pribytkova, pp. 77ff.; Pugachenkova, pp. 268ff.).
62— Tirmidh, mausoleum in the cemetery known as Sultan-Sa'dah. The only publication available to me that deals with the religious buildings of Tirmidh is the poorly printed second volume of the ac-
count of the Tirmidh archaeological expedition (in Trudy Akad. Nauk Uzbek SSR, ser. 1, Tashkent, 1945, pp. i96ff.). From this account it would appear that several mausoleums may, on architectural grounds, be dated fairly early, but the only one which has been discussed more fully (Denike, pp. 12-14) shows close similarities to the preceding monuments and should be dated in the late nth century.
63— Mausoleum at Baba-Gamber (small town on the river Mug). Madame Pugachenkova (pp. 273 ff.) has discovered a photograph of a mausoleum now disappeared; it is a rather simple construction with a high facade which can be dated in the nth century.
64— Uzgend, so-called middle mausoleum. This simple square building is squeezed between two later and dated buildings. Cohn-Wiener (E. Cohn-Wiener, 7uran, Berlin, 1930, pp. 18, 35) had thought that it might be of the early 11 th century, perhaps the tomb of one of the first Karakha-nids (cf. also Denike, pp. 14ff.). Its large gate and extensive decoration make it more likely to be a somewhat later monument.
65— Vekil-Bazar, in Turkmenistan, mausoleum of 'Abdallah ibn Burayda, late 11 th century. This small mausoleum with two Kufic inscriptions, one in brick, the other in stucco, is not dated and is only known to me through a reference in Epigrafika Vostoka, vol. 16, 1963, p. 149.
66— So-called Shaburgan-ata (in the area of Bukhara). This very curious building is octagonal with a large protecting facade (Nilsen, pp. 5 5 ff.). It is dated around 1100 because of the simplicity of its decoration and of its methods of construction. Nilsen mentions also, p. 61, two other similar buildings but without any
32
OLEG GRABAR
reference. The plan is curiously close to that of the Jabal-i Sang near Kerman (Survey, pi. 281).
67— Asterabad (Mazanderan), mausoleum. This building known from an illustration in Diez, Persien, Islamische Bau-kunst in Churasan, Berlin, 1923, pi. 8, is a low octagonal construction and has been related by E. Schroeder to our no. 55, but it is possible that it is later.
68— So-called Khoja Chisht (near Herat). Two square mausoleums are found there, apparently with two entrances on one axis, a double dome, and a very good decorative design of bricks. The fragments of inscriptions which have been published are not easily legible and do not seem to be of a historical character (O. V. Nieder-mayer, Afghanistan, Leipzig, 1924, pp. 62-63, pis-182-184).
69— Mausoleum of Muhammad Bash-shar, near Kolohozhion (in Tadjikistan). This complex building is in fact a mosque-mausoleum. It is divided into three parts: a large central domical room with a later portal and a mihrub to the right of the entrance and two side areas which are partly covered with domes, one of which contained two tombs. The zone of transition has a sort of pendentive in brick which is comparatively rare in Central Asia (L. S. Bretanitski, Ob odnom maloizvestnom pamiatnike, Materiály i Issledovaniia po arheologij SSR, vol. 66, 1953, pp. 325 ff.). The date is not given but around 1100 is suggested as plausible on comparative grounds.
70— Dihlstnn, mashhad. The name mashhad is given to a large cemetery near Dihistan, where seven mausoleums are still standing, of which five are datable around 1100 on architectural grounds (Pugachen-
kova, pp. 292 ff.). These mausoleums, which may have fulfilled a function similar to that of the Aswan examples, have rather curious forms. In plan they are circular or octagonal, with, in a few instances, thin half-towers or star-like points. In elevation they are very squat and curiously tapered, somewhere between the tower and the polygon covered with a dome.
71— Ferav (southern Turkemistan). Two small mausoleums were found, one a square with projecting entrance, the other also a square with large facade but provided inside with a curious series of recesses on four different axes. (Pugachenkova, pp. 299 ff.).
72— Kufan (Le Strange, Lands, p. 394). The ruins remain of a small square mausoleum with a long entrance IwUn (Pugachenkova, pp. 301-303). The mausoleum was redone in the second half of the 13 th century, as is attested by inscriptions—to my knowledge unpublished—on its stuccoes, but the original construction is supposed to be earlier.
73— Tolas, so-called mausoleum of Karakhcin. This mausoleum, popularly attributed to the founder of the Karakhanid dynasty (cf. above, no. 64), was destroyed in the early part of this century. On the basis of remaining photographs it is rather difficult to decide on its date, but its decorative designs (Denike, p. 15) are certainly later than the beginning of the dynasty, and, while an early core to the building is conceivable, the visible parts can hardly be dated before 1100.
74— Astan-baba (cf. above, no. 61). A shrine stands there consisting of four domed rooms of different sizes fitted within an irregular rectangle (Pugachenkova, pp. 286ff.). Various added constructions
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
33
suggest that this was a mashhad of some significance. The various legends which exist with respect to the identity of the people buried there do not help in explaining its original function or even shape. One of the domes is set on a very developed type of brick squinch. The mausoleum has been dated in the ioth century, but a date around iioo seems preferable because of the extent of the whole complex (B. A.Lit-vinskij, Arhitekturnyi kompleks, Trudy Akad. Nauk Tajik SSR, vol. 17, 1953, pp. i3iff.).
75— Irayadh, a village in Kuhistan. According to Yaqut (sub voco), a khcinqcih for sufis was there and by it a mashhad with a qubbah over the tomb of shaykh Abu Nasr al-Zahid al-Irayadhi, who died after 500/1006-07 (cf. E. Herzfeld, Kho-rasan, Der Islam, vol. 11, 1918, p. 168). Yaqut does not give us the date of the building and it is only a suggestion that it was erected in the first decade of the ioth century. It is interesting in illustrating a phenomenon for which there probably are other examples in texts: the honoring of a local holy man through a mausoleum and through some philanthropic or pious organization. Conversely many of the anonymous buildings we have listed may be explained in similar terms.
76— Paahtabad (inTajikistan). Of two mausoleums discovered there, one, a simple square construction with an embryonic portal, has been dated in the first decade of the 12th century (B. A. Litvinskij in Trudy Akad. Nauk Tajik SSR, vol. 17, 1953)-
77— So-called Khoja Roshnay, in the area of Isfara (V. Barthold, Turkestan-down to the Mongol invasions, 2nd. ed., London, 1958, pp. 160-161). A simple
square mausoleum of mediocre construction was found whose system of squinch arches recalls that of the later mausoleum of Sanjar in Merv, hence the suggested date at the beginning of the 12th century (E. A. Davidovich and B. A. Litvinskij, Ocherki arheologij raiona Isfara, Trudy Akad Mauk Tajik SSSR, vol. 35, 1955, pp. i85ff.).
78— Mausoleum also known as Khoja Rushnai, in the southern part of the Sur-khan valley, not far from Tirmidh. It is a square building with a projecting facade and two side entrances; the dome is on squinches consisting of a series of recessed arches. On comparative grounds the discoverer of the building has suggested the late nth century as a probable date (V. A. Nilsen, Nekotorye syrtzovye... postroiri, Istoria Materalnoi Kultury Uzbekistana, no. 3, 19625pp. i02ff.).
79— Turtas gunbadh, in the same area and of a similar type (ibid., pp. i07ff.).
80— Uiuk-gunbadh, also in the same area, but different from the preceding ones in that it has no axial entrance but a side one; on the other hand, its squinches are of a rather peculiar type, as though some local craftsman was trying to copy the newly developed Iranian muqarnas without quite understanding its significance (ibid.,pp. 109 ff.). It may be that this building is a little earlier than thepreceding two.
81— Alasher-khcLneh, in the Kengir valley of Kazakhstan. Built in baked bricks, it is a square building with heavy "pylons" on the main facade. Its walls are heavily decorated in "brick-style," and its exact date is uncertain (A. N. Margulan, Arhi-tekturnye pamiatniki raiona r. Kengir, Krat. Soob. Inst. 1st. i Mat. Kult., vol. 28, 1949, pp. 45-46).
34
OLEG GRABAR
82— Cairo, Khadra Sharif ah, 501/ 1107. This building with an entrance complex, a court without porticoes, and a sanctuary with three halls, the central one being covered with a dome, poses a problem. When it was cleared, tombs were discovered in the rooms to the side of the dome (Creswell, MAE, vol. 1, pp. 224-226); the whole construction should then be interpreted as a funerary compound. On the other hand, Maqrlzi's text used to identify the building and to date it (Maqrizi, vol. 2, p. 452) refers to this and other similar structures as masjids. If the tombs are indeed comtemporary with the building, the use of the word masjid would indicate an interesting extension of the meaning of the word or a change in the understanding of the building between the time of its foundation and the 15th century. If the tombs are not contemporary, the building should be related to the category of oratories built in cemeteries but not primarily for memorial purposes, which I have discussed in my introduction. By Maqrlzi's time, of course, many of these distinctions had become somewhat blurred and it is possible indeed that it is in the later Fa-timid period that the tremendous growth of an architecture in cemeteries (Creswell, p. 226) led to the convergence of two originally separate architectural and functional traditions, the funerary oratory and the mausoleum. The Khadra Sharif ah would then be one of the earliest examples of a new type of building.
83— Busra, mausoleum^) of shaykh Safi al-Dln, before 503/1109-10. An inscription on a sarcophagus refers to the waqf of a madrasah and to the renovation of a tur hah (Repertoire, no. 2932). There is no absolute certainty that a mausoleum
was involved, but the combination mausoleum- madrasah becomes so common during the 12th century that it is possible to interpret the Busra structure as an early example.
84—Aswan, mashhad, ca. n 10. This remarkable structure bears a close resemblance in plan to the Juyushi mosque (see above, no. 48). Since it has preserved a cenotaph, it must be assumed to have been a funerary building, although the identity of the personage involved is not known and there is no absolute certainty that the cenotaph is contemporary with the building (Creswell, MAE, vol. 1, pp. 222-224). The problem here is in reality the same as for no. 78.
8 5 —Damascus, mausoleums of Safwah al-Mulk and of Duqaq, 504/1110-11. The mausoleum of Safwah al-Mulk, now disappeared, consisted in a central dome supported by two half-domes (J. Sauvaget, Les monuments ayyoubides de Damas, vol. 1, Paris, 1938, pp. iff.). Its inscription is of great interest for the history of funerary and commemorative terminology; it begins as follows: "Has ordered the construction of his mashhad and of the turbah which is in it..." (Repertoire, no. 2942, with the correction already made by Sauvaget, p. 8, n. 8). As Sauvaget has reconstructed it on the basis of literary and topographical evidence, the construction of Safwah al-Mulk, made in her life-time (she died in 1119), was fitted into an existing complex of buildings which included the mausoleum of her son Duqaq (who died in 1104), an oratory, and perhaps a khanqah for sufis; all of them made up the mashhad and it is likely that the introduction into it of a new mausoleum required certain refections (hence the word 'imarah in the inscription,
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
35
which can refer to a new building as well as to a reconstruction). The word turbah appears to refer to the mausoleum Safwah al-Mulk built for herself; if so, it would be one of the earliest instances of the clearly documented use of the word for a domed building over a tomb. It is possible, on the other hand, that the masbhad refers to the mausoleum and the turbah to a cenotaph.47 86—Merv, mausoleum of Muhammad b. Zayd b. 'Ali Zayn al-'Abidin, 506/1112 to 13. It is only recently that the inscription of this long known mausoleum has been entirely deciphered, but, to my knowledge, M. E. Masson's reading has never been published (it is paraphrased with a discussion of the building in Pugachen-kova, pp. 304ff.). The mausoleum was built by the local governor Sharaf al-Dln Abu Tahir b. Sa'id and is referred to in the inscription as a mashhad. Like the Astan-baba complex (see above, no. 74) it is a couriously composed construction of four domical rooms of varying sizes—the largest of which was over the tomb of the cAlid— arranged in an irregular rectangle. One of
47 There may be yet another possibility. The text of the inscription (now gone), as it was published in the Repertoire from van Berchem's notes, begins: amara bi-imarah hadha al-mashhad wa al-turbah qubbah(?).... This is obviously impossible and Sauvaget had already proposed the obvious correction of qubbah into fihi, following in this a note left by van Berchem. If the inscription had been perfectly clear, as experienced an epigraphist as Max van Berchem could not have been misled into the inconsistency shown in his original reading. It may be wondered whether al-qubbah should not be read instead of al-turbah, a possible suggestion if one simply assumes that the beginning of the noun was not clear. Were the inscription to read ... al-mashhad wa al-qubbah fihi, the difficulty in understanding turbah as referring to a domed mausoleum would be gone.
the halls was an oratory. Its walls were painted and the remains are accepted as being of the early 12th century. The facade, as reconstructed by Madame Pugachen-kova, had a composition based on three arches, a comparatively uncommon motive in Iranian architecture of that time.
87— Aleppo, mausoleum of the qadi Abu al-Hasan, 508/1114-15. Known only through Ibn Shaddad (ed. D. Sourdel, Damascus, 1953, p. 35), it is called a turbah and the text insists on the remarkable quality of its stones. There is no absolute certainty that it was domed.
88— Damascus, cemetery Dahduh, 514/1120. An inscription mentions a qubbah built by the mother of Fakhr ak-Dln Buri (Repertoire, no. 2981).
89— Cairo, mausoleum of Sayidah 'Atikah, ca. 1120. This small square mausoleum is to an aunt of the Prophet (Cres-well, MAE, vol. 1, pp. 229 ff.).
90— Cairo, mashhad of Umm Kult-hum, before 516/1122. Only a mihrab remains of this building, and the date as well as the identification is hypothetical but plausible (Creswell, MAE, vol. 1, pp.239 to 240). The textual reference which permitted the identification (Ibn Doukmak, Description de I'Egypte, ed. K. Vollers, Cairo, 1893, vol. 4, p. 121; Maqrizi, vol. 2, p. 442) relates that seven mausoleums were redone at that time; the other six have probably disappeared.
91— Qus, mausoleum, ca. 1125. The date of this simple qubbah is based on its architectural characteristics (Creswell, MAE, vol. 1, pp. 236-238).
92— Cairo, mausoleum known as ikh-wtit Yusuf, ca. 112 5. Similar to the preceding and dated at the same time (Creswell, MAE, vol. 1, 234-236).
36
OLEG GRABAR
93— Cairo, anonymous mausoleum, first third of the 12th century. This small square construction in front of the khun-qcih of Baybars (Creswell, MAE, vol. 1, pp. 227-228) belongs to the same group.
94— Cairo, mausoleum of Muhammad al-Ja'fari, before 1125. Another instance of the same type (Cresswell, MAE, vol. 1, pp. 232-234).
95— Cairo, mausoleum known as shaykh Yunis, before 1125. This mausoleum, typologically related to the preceding ones, had, until Creswell's studies, been thought to be the mausoleum of Badr al-Jamali (Creswell, MAE, vol. 1, pp. 232 to
234)-
96— Huday-Hazar, some 28 kms. from Bayram 'Ali in Turkmenistan, near the ruins of a medieval village. This simple square building is remarkable for its large facade and for its sober decoration (Puga-chenkova, pp. 3ioff.).
97— Aleppo, mausoleum built by Amr b. abi al-Fadl, 522/1128. The curiosity of this mausoleum, known only through an inscription (Repertoire, no. 3027), is that it is called an eywun.
98— Cairo, mausoleum of the head of Zayn al-'Abldln, 525/1130-31. A rather complex set of problems is connected with this mausoleum (Maqrizi, vol.2, pp.436, 440; G. Wiet, MCIA: Egypte II, pp. 2i3ff.). First, as Maqrizi already pointed out, the mausoleum was built in fact for Zayn's son, Zayd. Second, the date 525 is the date of the invention of the head and the earliest reference to a construction is in a bad Ottoman copy of a Fatimid inscription giving the date 549/1154 (Repertoire, no. 3163). Third, the historical circumstances of the invention, as they are told by Maqrizi, contain a number of pe-
culiarities and obviously legendary elements which have been discussed in detail by G. Wiet. If we have preferred the date of 525 to 549 it is merely that the miraculous invention of the head of an imam is likely to have been followed by the building of a shrine and that a waiting period of almost a quarter of a century seems rather meaningless. The monument contained an oratory with court, an eywan, and a domed mausoleum, but most of the present structure is modern.
99— Cairo, mashhad of Sayidah Ruq-qayah, 527/1133. This mausoleum with a central domed hall, two side halls, and a porticoed ante-hall, has often been studied (Creswell, MAE, vol. 1, pp. 247ff.; and esp. G. Wiet, MCIA: Egypte II, pp. 195 ff.), and G. Wiet, in particular, has brought out all the complex features of Fatimid piety which are involved in it.
100— Sohag, 529/1134. An inscription published in the Repertoire (no. 3071) refers to some construction ('imUrah) which served to commemorate a victory of the army commanded by the son of the caliph al-Hafiz. Although the events which are so commemorated are not recorded and the nature of the construction unknown, we have here another instance of a rare type of memorial construction, the victory monument.
101— Madinah, mausoleum of 'Abbas b. 'Abd al-Muttalib and of al-Hasan b. 'Ali, before 5 29/1134-3 5. Our major source for the monuments of Madinah is Samhudi, who relates that there is some argument as to the time when this mausoleum was built but opts for 529 and the caliphate of al-Murtashid (al-Samhudi, Kitab wafa' al~ wafti, 2 vols., Cairo, 1326 H., vol.2, p. 101). The building had a high dome and
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
37
two doors, one of which was open all day.48 So far as I have been able to discover, this is the earliest certain instance of a mausoleum in Madman, but it is possible that a number of Alid mausoleums were destroyed in earlier times.
102— Mazar-i Sharif, mausoleum to 'Ali, 530/1136. Although no part of this celebrated Timurid building seems to be as early as the 12th century, a sanctuary was apparently created there as early as 1136 (O. von Niedermeyer, Afghanistan, pp. 65 ff).
103— Shahristan (near Isfahan), ImcLm-zudeh Sehzndeh Husayn and Ibrahim, ca. 532/1137-38. This mausoleum has been assumed by A. Godard to be that of the caliph al-Rashid, who was indeed killed near Isfahan (A. Godard, Les anciennes mosquees, Arts Asiatiques, vol. 3, 1956, p. 55), but the evidence of contemporary chronicles does not justify this conclusion, although, of course, this simple octagon (Survey, p. 1023 and pi. 334 A) could well be dated at this time.
104— Rayy, so-called tomb of Tughril, 534/1139-40. This much reconstructed tomb is still pretty much of a mystery. The evidence for the date is based on an inscription, now in the Museum of Art at the University of Michigan (formerly in the possession of E. Herzfeld), which is said to have come from the tomb (E. Herzfeld, imcimzadeh Kurrcir, Arch. Mitt, aus Iran, vol. 7, 1935, p. 80); cf. Appendix.
105— MarcLghah, so-called red mausoleum, 542/1148. This square mausoleum,
48 It is conceivable that a few other Medinese mausoleums, for which no date is given in Sam-hudi, are as early as the first part of the 12th century (in addition to our no. 107); only a thorough study of the personages involved could provide a solution.
with a dome on muqarnas and a superb facade, is empty, but its inscription has preserved the names of the patron and of the builder (Repertoire, nos. 3135 and 3136; A. Godard, Les monuments de Ma-mgha, Paris, 1937^.4).
106— Turan-pusht (between Abarquh and Yazd), Gunbadh-i shaykh Junayd, 543/1148-49. An octagonal tomb. The plan and photographs have been published by A. Godard (Les coupoles iraniennes, Athar-e Iran, vol. 4, 1949, fig. 213 and 258).
107— Madinah, mashhad of Ismahl b. Ja'far al-Sadiq, 546/1151-52. As described by Samhndi (p. 104), this was a complex building with a court and a well in addition to the qubbah itself. The date is given by an inscription.
108— Cairo, mashhad of Sayidna al Husayn, 549/1154-55. In 1154-55, the head of Husayn was moved from Ascalon to Cairo, where a new sanctuary, included in the palace area, was built (Creswell, MAE, vol. 1, pp. 271-273; Maqrizi, vol. 1, pp. 427 ff.). Most of the present building is modern.
109— Cairo, mausoleum of Muhammad al-Hasawati, ca. 1125-55. The main curiosity of this mausoleum is that it seems to have reverted to earlier patterns by being open on three sides (Creswell, MAE, vol. 1, pp. 259-260).
110— Cairo, mausoleum of Yahya al-Shabih, ca. 1150. This is a more complex structure with a court and an oratory, in addition to the domed room over the tomb (Creswell, vol. 1, pp. 264 ff.).
111— Cairo, mausoleum of Qasim abu Tayyib, ca. 1150. Here also we are probably dealing with a complex building (Creswell, MAE, vol. 1, pp. 269-270).
38
OLEG GRAB AR
ii2—Tils, anonymous mausoleum. This well-known large mausoleum with a superb facade has not yet been satisfactorily dated (E.Diez, Khurasan, pp. 57-62; Survey, pp. 1072 ff.), but the evidence of other Central Asian mausoleums makes it quite likely that the middle of the 12th century is a possible date (Pribytkova, pp. 19 ff.).
CONCLUSIONS
This list of 112 items corresponding to over 160 mausoleums erected before the middle of the 12th century suggests a num-of conclusions and raises a series of problems.
The first point to be noted is that the two earliest monuments in the list—the Dome of the Rock in Jerusalem and the Qubbah al-Sulaybiyah in Samarra—are extraordinary monuments, which, in their function and in the circumstances which surround their construction, stand out as exceptional creations explicable only through precise events of their time. It is, therefore, in the late 9th, but especially in the 10th and nth centuries that we can ascertain the growth of commemorative buildings. Their function was almost always funerary; there are only a few instances (nos. 48, 100) in which other aims are clearly indicated.49 This result coincides
49 We have entirely left out of the discussion a whole group of Iranian monuments whose commemorative significance is still a matter of uncertainty, the manar. There is little doubt that the tall "minarets" which are so numerous in cities, the countryside, or remote valleys, like the extraordinary example at Jam (A. Maricq et G. Wiet, Le minaret de Djam, Paris, 1959), did not serve merely for the call to prayer. They have been called
with evidence provided by the geographers of the time. The writers of the 3rd century of the hijrah (9th century), like Ibn Rostah or Ya'qubi, not only do not mention monuments built over tombs, but also rarely list names of personages buried in given places; on the other hand, the writers of the 4th century (10th century a.d.), like al-Muqaddasi or Ibn Hawqal, almost systematically provide the reader with such lists. By the time of Ibn Jubayr, in the 6th century (12th a.d.), places of burial and of pilgrimage and mausoleums are standard fare, and guide-books appear for pilgrims.50 The comparison between monuments and texts leads to a further remark: the interest in places of burial of holy men, their more or less systematic cataloguing, and cultic and social functions around tombs and in cemeteries—as they appear in the studies of L. Massignon and G. Makdisi on Cairo and Baghdad—seem to have preceded the actual appearance of an architectural form to single out and to honor tombs and holy places. It is difficult, if not impossible, to determine the time which separated the
towers of victory (E. Diez, Persien, pp. 73 ff.), but they obviously could not all fulfill that purpose. Some of them certainly had a commemorative aim, which explains their use as tombs. But, ty-pologically and genetically, they belong to a different group and their understanding must await more complete archaeological, epigraphical, and literary documentation than has so far been available. The importance of such precise studies is fully shown by the many surprising conclusions reached by Madame Sourdel-Thomine in her reading of the inscriptions on two such manars (Deux minarets seldjoukides d'Afghanistan, Syria, vol. 30, 1953).
50 For instance, al-Harawi, Guide des lieux de pelerinage, tr. J. Sourdel-Thomine, Damascus, 1957-
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
39
growth of the two phenomena, but it should be pointed out that, both during the period with which we have dealt and in later centuries, several important branches of Islam, banbalism being a conspicuous example, consistently avoided the use of architectural symbols for their heroes and great events.
As far as the purposes of the mausoleums are concerned, the fact emerges that the overwhelming majority of early mausoleums served either to emphasize shi'ite holy places or to glorify princes from smaller dynasties, usually heterodox. This is not surprising, for the very basic shi'ite emphasis on descent from the Prophet and the mystical significance of the succession of imams might naturally result in the desire to transform into places of veneration the real or alleged places where the members of the holy family were buried or lived. Princely mausoleums are, in earlier instances, particularly characteristic of the new dynasties of Iran and it is interesting to contrast the funerary practices of these princes with the comparative simplicity of the burials of the Abbasid and even Fati-mid caliphs. Many more princely mausoleums probably existed than have been identified, since so few local chronicles are known and since the meaning of the building faded away with the disappearance of the dynasties which built them. Some of the remaining anonymous mausoleums of Central Asia, for instance, could originally very well have been princely mausoleums to which at a later date a holy man or a holy event was attached. Many of the early sunni shrines (nos. 9, 49) were probably built in answer to the growth of shi'ite places of veneration; it is known, for instance, that in Aleppo one of the first acts
of the conquering Turks was to create a sanctuary to Abraham in order to counterbalance the earlier shi'ite shrine.51 In their search for personages around whom cults and ceremonies were to be developed, the sunnis tended either to use scholars, Companions of the Prophet and early conquerors, or Old Testament Prophets, especially Abraham and Solomon, whose Islamic associations are particularly strong.52 It is, of course, unlikely that all sunnite mausoleums and mashhads were built as a reaction to shi'ism, and the many early constructions of Egypt and Central Asia in particular, if properly dated, pose a major problem. In some instances (no. 17), older sanctuaries were taken over. Elsewhere (no. 48) attention has been brought to one aspect of sunnism which may have affected the growth of shrines, even though it is still comparatively little known: the frontier spirit of the ghazw, which often involved allegiance to a man or to the place from which the jihad originated. To perpetuate the memory of the founder or leader of such a group through a mausoleum is a logical enough procedure, and it could well explain the monuments at Aswan as well as many Central Asian ones. Altogether, however, the evidence of the buildings seems to indicate that it is not before the ubiquitous impact of shi'ism in what Mas-signon had called the "Qarmatian century" that either branch of Islam began systematically to transform its tombs and historically meaningful places into sanctu-
51 J. Sauvaget, Alep, Paris, 1941, pp. i24ff., with a discussion of the further development of these shrines in the second half of the 12 th century.
52 J. Sourdel-Thomine, Les anciens lieux de pelerinage, Bulletin d'Etudes Orientales, vol. 14, 1951-54, pp. 65 ff.
40
OLEG GRABAR
aries and to build shrines over them. It might also be noted that a significant number of early mausoleums served for several personages, one of which, usually, was of greater holiness or quality than the others. Whether this feature is merely accidental or whether it reflects the fairly early identification of certain parts of cemeteries for specific individuals or families (cf. introduction) is still unresolved.
The third conclusion to emerge from our list is that while early shrines existed throughout the territories of Islam and while some of the earliest sanctuaries appeared in Iraq, where cAli and Husayn had died, the two areas which show by far the greatest concentration of early mausoleums are the northeastern Iranian provinces and Egypt. North Africa and Spain, from which considerable literary and archaeological remains are available, seem to have been least affected at this time by the development of mausoleums.
A last conclusion concerns the architectural types which are involved. On this subject our list did not lead to any major new conclusion, at least insofar as the mausoleum itself is concerned. The tower-tomb is peculiar to the mountains and plains around the Caspian sea and to the northern edge of the Iranian plateau; it affected a few monuments in Central Iran (no. 45 for instance). The square or polygonal "canopy" type of tomb covered with a dome existed everywhere. In the earliest instances it was open on all sides and then, little by little, was transformed into a hall with a single door—or sometimes two doors— and a mibrtib. Unusual plans, accessory buildings, crypts, combinations or several square halls into one ensemble, and mosque-mausoleums appear only rarely, if at all,
before the middle of the 12th century, but they exist and pave the way for the great monumental ensembles consisting of a tomb and of pious or philanthropic institutions which will appear in the second half of the century. Yet the single simple "canopy" is never abandoned. The growth of the pishtUq, the high gate almost separated from the mausoleum itself, in northeastern Iran is more difficult to date; and I have preferred here a late nth century date to the early nth century date proposed by most Russian archaeologists, even though the apparently clearly dated example at Tim (no. 12) has complicated the problem considerably, just as it has put the question of the muqarnas on a new basis. The interest of the many instances of the pishtUq which have been listed is in showing many variants, often embryos, of a feature which was destined to become one of the most spectacular characteristics of later Iranian architecture, and whose purpose and evolution require a separate study. The point which may be of greater significance, as one deals with the physical appearance of the mausoleums, is that, even though there is clear evidence that quite early there existed constructions in cemeteries (cf. the introduction), it is only slowly that a composite building appears which united in a single architectural ensemble the manifold functions of a shrine.
While these conclusions appear to follow the evidence provided by the documents, they also raise problems, which can, at this stage, only be defined, for their solution demands investigations which go much beyond my original aim.
The first of these problems concerns certain aspects of the purpose of the mausoleums. It has been noted that two areas
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
41
of the Islamic world—Egypt and Central Asia—seem to have developed mausoleums in the 10th and nth centuries to a much greater degree than other provinces, and that many of these cannot readily be explained as obviously religious shrines or princely memorials. The problem is to evaluate the significance of these facts and to find out whether there were certain peculiarities common to the history and civilization of these two areas which could explain the phenomenon, or whether entirely different causes led to the same results in both areas. The difficulty of the problem is complicated by the fact that much of our view of early Islamic history is based on chronicles whose concern was narrow geographically and limited in content. Furthermore, the concern of many historians, until very recently, has been in the secure-ment of a fundamental chronology of events and in problems of doctrine rather than in the establishment of the cultural, social, and economic factors which formed the civilization of the different provinces of the Islamic world.
Thanks to Maqrizi's Kbitat, to the existence of good epigraphical surveys, and to the still comparatively few published Geniza fragments, the Egyptian situation can be better understood than the Eastern Iranian one. Even a rapid perusal of Maqrizi's description of Cairene monuments53 clearly shows that the 10th and nth centuries are characterized by the building of many religious monuments by private individuals, at times members of the ruling families acting as private believers (this is particularly true of women), more often officers, administrators, dignitaries, and
53 Maqrlzi, mainly vol. 2, esp. pp. 244 if.
even merchants.54 The social basis for the sponsorship of a monumental architecture widened remarkably. The result is evident in the nature of the buildings: from the few large early mosques (al-Azhar, al-Hakim) and palaces, we move to many small oratories (of which only a handful remain)55 and to many pavilions, private houses and apartments throughout the city.56. Within this context, if the conclusion is acceptable that the main original impetus for mausoleums derived from religious shi'ite and dynastic royal needs, it may be suggested that quite rapidly, at least in Cairo, a sort of "democratization" of the mausoleum took place to include all who could afford and desired to have one built. In addition to its religious and princely connotations, the mausoleum became a symbol of conspicuous consumption. In many instances such mausoleums were to be related to philanthropic (schools, hospitals, guesthouses) or devotional (monasteries, ribats, oratories) institutions. Whether this occurred commonly before 1150 in Egypt is still difficult to say, except in the one instance of the oratory (masjid), which may well have preceded the mausoleum in cemeteries (see above p. 10). Outside of Cairo, the only instance of mausoleums on any large scale is provided by the Aswan constructions. The explanation for them is suggested by their location in a frontier city, with
54 See the very remarkable inscription of a merchant who built a mosque and planted a palm tree for the benefit of all Muslims (Repertoire, no. 2173), dated 1011-12.
55 Mosques al-Aqmar and al-Salih Tala'i, Creswell, MAE, vol. 1, pp. 242 ff. and 275 ff.
56 The admirative description of Nasir-i Khosrow is, in this respect, particularly remarkable.
42
OLEG GRABAR
a special veneration of martyrs and the leaders of the various ribclts organized for the defense of Islam.
But the historical circumstances of Egypt were not duplicated in toto in northeastern Iran. The extraordinary variety of cultures and influences which created Islamic Central Asia is only now beginning to emerge from obscurity, and the process of Islamization of Iranian and Turkish groups there is still far from clear. Of the four causes which can be proposed to explain the growth and development of Egyptian mausoleums—shi'ism, princely symbols, wider social patronage, the frontier—two are clearly present in Central Asia. It is there that we find most of the earlier instances of purely secular tombs (nos.6, 16, 27, 32), and there is little doubt that all the symbols, institutions, and allegiances which developed on the frontiers of the Muslim world were particularly acute in Central Asia.57 Many ributs existed there; many were sponsored privately, and in or near some of them founders or their descendants were buried.53 The social history of the 10th and nth centuries has not yet been entirely worked out, but the evidence provided by archaeology59
57 In summarized form the point comes out in V. V. Barthold, Four studies on the history of Central Asia, vol. 1, Leiden, 1956, pp. 18ff. and 70; and it appears quite clearly as one glances through al-Narshaki's, History of Bukhara, tr. R. N. Frye, Cambridge, 1954.
5S See above p. 9.
59 The best, even though not always fully documented and at times schematic, description for the southern part of Central Asia is in Puga-chenkova, pp. 118 ff., especially the introductions and conclusions of chapters. For the northern areas one should consult the syntheses of S. P. Tolstov on Khorezm.
seems to indicate that there was a shift or power from landowning dihqUns to the cities with their military rulers and their numerous social and professional organizations. The passage from one to the other probably took place around the time of the fall of the Samanids (i.e. late 10th and early nth centuries). It may be suggested that, in northeastern Iran, the original impetus for mausoleums derived from princely constructions, but that, just as in Egypt, the widening of the patronage and changes in religious and cultic habits (here related more precisely to the importance of semi-religious orders guarding the frontier and of social organizations with mystical overtones) led to a wider use of the monumental tomb. This explanation, however, cannot be more than a hypothesis so long as our information is so limited on the individuals who were buried in the mausoleums or the events they commemorated; and the uncertain chronology of the tombs adds to the difficulty.
There is, however, a further point which emerges from these remarks. Just as we have seen that funerary cults preceded the building of mausoleums, so it appears that the mausoleum is closely related in another way to a whole group of social and pious institutions that appeared before it, especially the ribat and the private oratory. The central feature of the latter is that it was heavily endowed and sponsored by a wide variety of people. As such the oratories reflected more than piety; they were also forms of investment and of self-glorification. These secular ideals were carried over to the mausoleums. They explain why, in later centuries, with only few exceptions (for instance the mausoleums to al-Shafi'i in Cairo), next to popular, almost pagan,
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
43
memorials to local saints, the characteristic domical memorial will be sponsored by and built for princes, members of their families, dignitaries, and officers. The Taj Mahal is the supreme illustration of the fact that by its origins, its sponsorship, and its growth, the Islamic mausoleum reflected secular conceptions. Next to these memorials to rulers are the great shi'ite sanctuaries to which constant attention was given through the centuries. Sunnite shrines, however, remained few in numbers. Most of them were created in the historically and culturally tumultuous period which extended from the time of breakdown of the' Abbasid caliphate in the early ioth century to the Mongol conquest in the middle of the 13th. Few of them received, in the following centuries, the care and attention which was given either to cAlid shrines or to the new secular ones. The two factors, thus, which first caused the growth of mausoleums—sh'ism and secular glorification—remained throughout as the main sources of memorial constructions.60
60 On a lower social and artistic level, the appearance of local shrines, along roads or on elevations of the ground, dedicated to local holy men and often with pagan overtones, cannot be dated precisely. None of the ones known to me are earlier than the 12th century. Even though many reflect cults of pre-Islamic origin, the evidence seems to indicate that the construction of popular qubbahs followed the growth of official shrines rather than continued a pre-Islamic tradition of small sanctuaries and mausoleums of popular origin. The facts that Islam lacked the ecclesiastical organization of Christianity and that its major developments were urban rather than agricultural also explain why these symbols of popular and peasant beliefs were not incorporated into the mainstream of the faith, as they were in Christianity. Descriptions of few of these buildings
A problem of comparable complexity confronts the art historian when he tries to explain the origins of the forms used for the mausoleums. The square covered with a dome and the simple canopy are, of course, not new architectural forms: they were used in the funerary architecture of antiquity. The problem is that there is little evidence that these forms had been used for these purposes in the centuries which preceded or accompanied their appearance in the Islamic world. In the Christian empire of Byzantium, imperial sarcophagi were put in Constantinopolitan churches, but, with a few exceptions in Armenia, specific funerary architecture for individuals does not seem to have developed. In early Christian times many instances exist of mausoleums for saints and heroes which are clearly related to pagan types, but these antique funerary cults and forms were soon sublimated in the martyrium church, which by the 6th century had developed a large repertory of shapes and forms based on the central plan and had almost abandoned the simple types which Islam used in the ioth century; it is interesting to note, however, that some of the Egyptian examples preserved the early types much longer.61 In Iran the ancient tombs of the Achaemenids and of the Par-thians were not continued in Sasanian times.62 Towers of silence and ossuaries
are published; some indications exist in Sarre-Herzfeld, Arcbaeologische Reise, im Eupharat- und Tigris-Gebiet, Berlin, 1911-20, and in T. Canaan, Mohammedan saints and sanctuaries in Palestine, Jerusalem, 1927.
61 A. Grabar, Martyrium, vol. 1, pp. 47 ff., 144, and passim.
62 See, however, some curious paintings in Panjikent, Zhivopis Drevnego Pianjikenta, Moscow, 1954, pi. XX.
44
OLEG GRABAR
were presumably the only monumental forms associated with funerary practices.63 And one cannot explain all types of monumental tombs from the impact of the Indian stupa or of nomadic mounds form Central Asia, while an influence of pre-Islamic Arabian practices could hardly have affected eastern Iran or even Egypt to the exclusion of closer provinces. Under these circumstances the assumption by the Muslims around the beginning of the i oth century of an ancient funerary form which had not been used for several centuries for these purposes poses an important problem.
Several answers might be suggested. It may be that in parts of the Islamic world, older martyria of simpler form were still used, as for instance in the biblical sanctuaries of Syria and Palestine, and in Egypt, and that these modest symbols served as models for the Muslims. Or perhaps the Islamic world revived for its own purposes architectural formulas which had existed for these very purposes several centuries earlier. It may be asked, however, to what extent the medieval Islamic world was conscious of the significance of the early monuments which were standing in the areas it had conquered several centuries after the conquest. Moreover, examples taken from the former Christian Near East cannot be used to explain the Central Asian buildings.
A more likely explanation of the use of the dome over a square or polygon is not its association with funerary architecture in the past, but its significance in other aspects of Islamic architecture as a sign of honor and veneration. It was common in
63 R. Ghirshman, Etudes Iraniennes II, Arti-bus Asiae, vol. 2, 1948, pp. 292 ff., with an important bibliography.
mosques in front of the mihrub from the moment the mibrub was introduced in Medina64; it was a common feature of throne rooms (Mshatta, Baghdad, Samarra), gates (Khirbat Minyah, Baghdad), and pavilions (better known through descriptions as in Egypt or Iran) in the architecture of palaces; it was found over fountains in palaces (Khirbat al-Maf jar) or in mosques (mosque of Ibn Tulun in Cairo). The domical structure was adopted as the most common form to be used over holy spots and tombs not because of its precise funerary attribute but because of its general meaning as a sign of veneration.
This last point may also serve to explain the cloudy origins of the northern Iranian tower-tomb. There a different type of structure, whose symbolic meaning as a memorial, a tower of victory, or a beacon has often been discussed,65 was used for funerary architecture because of its abstract significance as a symbol of power or holiness rather than precisely for any funerary reason. In the case of both forms, then, it would appear, in this hypothesis, that, in its search for a monumental expression of certain spiritual and cultural needs, the Islamic world did not always invent new forms or borrow them directly from other cultural and artistic traditions, but at times adapted an existing architectural vocabulary of its own to new needs. If it happened that the final result bears resemblance to earlier, pre-Islamic monuments erected for similar purposes, it is not so much that they were borrowed directly one from the other,
64 J. Sauvaget, La mosquee omeyyade de Medine, fig. 5 on p. 91.
65 See the references to the works of Diez under nos. 27 and 66; cf. also note 49.
THE EARLIEST ISLAMIC COMMEMORATIVE STRUCTURES
45
but that both derived their architectural forms from the wide repertory of more or less abstract formulas which the Mediterranean and the Near East had in common since antiquity.66
These considerations on the sponsorship and purpose of the Islamic commemorative building and on the forms it took go much beyond our original aim of gathering together a series of documents on a certain type of building. They serve, however, to illustrate some of the problems which are posed in any investigation of the manner in which Islamic civilization created the artistic forms which eventually became identified with it. It is only when detailed investigations of other architectural entities—the masjid, the ribat, the khanqfih, the khan—and of the cultural and social phenomena which led to their growth have been made that it will fully be possible to understand the way in which Islamic architecture was formed and to explain the forms it took.
APPENDIX BY C. G. MILES
In 1940 the late Ernst Herzfeld, then at the Institute for Advanced Study in Princeton, showed me an iron plaque in his possession bearing an inscription dated 534 H. After his death this inscription became the property of his sister, Mrs. Charlotte M. Bradford, who entrusted it to me for safekeeping. It is now in the Museum of Art at the University of Michigan in Ann Arbor. When Herzfeld first brought
66 Cf. the analyses of the "basilicalhall" made by Sauvaget, op. cit., pp. 158 ff.
the inscription to my attention I took some brief notes on what he told me of the circumstances in which he acquired the plaque. It appears that during his residence in Iran he found and purchased the inscription in one of the Teheran bazaars. Some time later while examining a notebook of a Persian antiquarian friend of his he saw a drawing of this same inscription in situ over the door of the then unrestored "Tower of Toghrul" at Rayy. Herzfeld later verified this provenance to his own satisfaction by measurements and an examination of the restored brickwork where the tablet had formerly been fixed in the tower. Unfortunately I do not now recall what he told me of its exact position in the doorway.1
The inscription in question is cast on a single sheet of iron (about 2 mm. in thickness), measuring 80.5 cm. in length and from 12 to 13 cm. in height. Notches at the top and bottom of the righthand end give it a tabula ansata form,2 at least at this end. The plaque is pierced by eleven circular holes, approximately 9 mm. in diameter, through which nails or bolts were driven to attach the sheet to the architectural member on which it was mounted. There are five of these holes above the upper line
1 The existence of this dated inscription has been noted several times, but never with any explanatory details. In Arch. Mitt. Iran, vol. 7, 1935, p. 80, Herzfeld's list of early Kufic and Naskhi inscriptions in Iran included the entry "534, Rayy, sog. Turm des Toghrul, Eisenthiir." This is the basis of the bare mention of the date of the tower in A. V. Pope, Survey of Persian Art, London, 1938, vol.2, p. 1022, and vol.4, plate 346; and of my entry, 534, Rayy, tower, in Ars Islamica, vol. 8, 1941, p. 108.
2 See Ernst Herzfeld, Die Tabula ansata in der islamischen Epigraphik und Ornamentik, Der Islam, vol. 6, 1916, pp. 189-199.
46
OLEG GRABAR
of inscription and six below the lower line. The text reads as follows:
fecit 'Abd al-Wahhab al-QazvIni b. Fakhra-var at the end of Rajab of the year 534 [March,
1140 a. d.]
The rather naive, quasi-Kufic characters are completely unadorned. Several peculiarities suggest that the craftsman who designed the letters was not really versed in the Kufic alphabet but that he attempted to give the inscription a Kufic appearance. For example, the lam of 'amala does not descend below the base line, nor does the nun of thalathin or the khti of salkh; the dul of eabd is not Kufic, the sin of khams, being in the medial form, should join to the following mim, but it does not; the final ya's of Qazvini and fi are curiously abbreviated. Despite these errors and the imperfect balance of the two lines, some thought and planning in the laying out of the inscription is evident in the spacing of letters in the upper line that permits the tops of several tall characters in the lower line to extend into the upper register.
With regard to the reading and content of the inscription there can be no doubt about the date and the name 'Abd al-Wahhab, and very little doubt about the correctness of the nisbah al-Qazvmi.The only problem is that of the correct reading of the proper name immediately preceding the date. Fakhravar, "Bringer of Glory," although evidently previously unrecorded, is a quite possible Persian name. I am in-
debted to Professor V. Minorsky for drawing my attention to analogous names compounded with -avar, such as Numavar? Bakhtavar* and of course Dinavar.5 The order of the names is curious: one would expect al-Oazvini to follow the father's name. The unsophisticated character of the inscription suggests the possibility that the addition of the father's name,6 who may have been a Daylamite, with whom the inhabitants of Qazvin had frequent, not always hostile, intercourse, may have been an afterthought.
Just what it was that 'Abd al-Wahhab "made" (camala) cannot now be determined. It is unlikely that he was the architect or the builder of the tower itself; such an individual would doubtless have recorded his achievement in a more impressive inscription. Herzfeld's implication is that the plaque was associated with an iron door. In any case there can be little doubt that the inscription provides us with a specific date for at least a part of "Togh-rul's Tower" at Rayy, a date shortly after the death of Toghrul I, sultan of the 'Iraq branch of the Seljoqs.
3 E.g., Hasan b. Namavar, one of the last survivors of the Buyids (cf. Juvaini, The history of the world-conqueror, tr. John A. Boyle, Manchester, 1958, vol. 2, p. 697).
4 Cf. F. Justi, Iranisches Namenbuch, Marburg, 1895, p. 61.
5 Ibid., p. 85.
6 The first word in the second line is undoubtedly bin and not [a~\bu; all the waw's in the inscription have clear open heads. Also a close examination of the plaque itself shows that no letter is missing at the beginning of the second line.