5
Prayers and Protection:TheTower atTorba Reconsidered
MARTIN F. LEŠÁK
The tower at Torba (fig. 1) is the most eastern fortified outpost
of the well-known castrum Castelseprio. It is situated about forty
kilometers north of Milan in the valley of the Olona river at the
foot of the hill to where the main nucleus of the castrum extends
(fig. 2). The origins of this military building date back to the 5th
or 6th century at the very latest1; however, the wall paintings decorating
two floors of the structure were created soon after the
Benedictine convent was founded nearby. This most likely occurred
during the 8th century2. Though the first reference to the
tower appeared as early as in the first half of the 20th century3,
the wall paintings were only mentioned in 1967 by Santino
Langé4, and did not awaken much scholarly interest in the following
years5. This fact is especially striking in comparison to
the bibliography regarding famous frescoes, which is preserved a
few hundred meters from the tower in Santa Maria foris portas,
one of the most studied monuments in Lombardy6. To this day,
the only remaining monograph about the murals at Torba is
Carlo Bertelli’s piece (of fifty-nine pages) from 19887 and therefore
1 Most recently Brogiolo dated foundations of the tower to the 5th century: G.
P. BROGIOLO, La torre tardo-antica e il monastero alto-medievale di Torba, in
Archeologia classica e post-classica tra Italia e Mediterraneo, eds. S. Lusuardi Siena,
C. Perassi, F. Sacchi and M. Sannazaro, Milano 2016, pp. 225-232. Cf. A.
SCILLIA, Analisi stratigrafica degli alzati, in Castelseprio e Torba. Sintesi delle
ricerche e aggiornamenti, ed. P. M. De Marchi, Mantova 2013, pp. 93-123, esp.
109; G. P. BROGIOLO, Per una storia religiosa di Castelseprio: il complesso di Torba
e la chiesa di S. Maria foris portas, in Castelseprio e Torba…, 2013, pp. 213-222;
M. ROSSI, Castelseprio nell’alto Medioevo, in Storia dell’Arte a Varese e nel suo territorio,
ed. M. L. Gatti Perer, Varese 2011, pp. 8-49, esp. 40; and S. COLOMBO,
Conoscere Castelseprio, Torba e dintorni, Varese 1999 (I ed. 1995), p. 19. Renato
Bazzoni places the construction of the tower into the context of the late Roman
period, i. e. to the 5th century, when the number of attacks on borders of the
empire increased. See R. BAZZONI, Il Monastero di Torba e i suoi affreschi, in
Conservazione e valorizzazione degli affreschi nella provincia di Varese, ed. P. C.
Marani, Varese 1997, pp. 17-22, esp. 17. For the 5th century dating, see also
E. CHIZZONI - G. VASSALLI, Il Monastero di Torba, in «Tracce», VIII (1987/2),
pp. 95-102, esp. 96; Y. PANIN, Documenti inediti quatrocenteschi per Torba, in
«Arte Lombarda», 64 (1983/1), pp. 93-98, esp. 93; or L. ROTONDI SECCHI
TARUGI, Il baluardo difensivo di Castelseprio verso Torba, in «Rassegna Gallaratese
di Storia e d’Arte», XXXII, 119 (1973), pp. 81-88, esp. 88.
2 The dedication of the abbey to St. Mary is documented by historical sources
from 1049 and 1204. However, on the grounds of references to Torba in Liber
notitiae sanctorum Mediolani, originating probably at the end of the 13th century,
some historians inaccurately claimed that the convent was dedicated to
St. Blaise. See for example G. GIULINI, Memorie spettanti alla storia, al governo
ed alla descrizione della città e campagna di Milano, ne’ secoli bassi, IV, Milano
1855 (I ed. 1760), pp. 700, 722 and F. BOMBOGNINI, Antiquario della Diocesi
di Milano, Milano 1856 (I ed. 1790), p. 100. For the references, from which
it would be possible to deduce that, at the time, there was a church of St. Mary
with an altar dedicated to St. Calocerus and a monastery church of St. Blaise
with two altars, first dedicated to St. Peter, second to St. Quriaqos; see G. ROTONDI,
Un passo di Galvano Flamma e il monastero di Torba, in «Archivio Storico
Lombardo», XLIX (1922), pp. 119-134, esp. 127, n. 2 and Liber notitiae
sanctorum Mediolani: manoscritto della Biblioteca capitolare di Milano, eds. M.
Magistretti and U. Monneret de Villard, Milano 1917. See also CHIZZONI VASSALLI,
1987, p. 96 and S. LANGÉ, Preliminari al complesso monumentale fortificato
della Torba (Castelseprio), in «Castellum», V (1967), pp. 33-48, esp. 36.
3 See G. P. BOGNETTI, Aggiornamenti su Castelseprio III, in «Sibrium», IV
(1958-1959), pp. 19-79, esp. 21; E. ARSLAN, L’architettura romanica milanese,
in Storia di Milano, III, 1954, pp. 395-521, esp. 413, n. 2; G. P. BOGNETTI, S.
Maria foris portas di Castelseprio e la storia religiosa dei Longobardi, in Santa
Maria di Castelseprio, eds. G. P. Bognetti, G. Chierici and A. De Capitani
d’Arzago, Milano 1948, pp. 15-511, esp. 366, 396; ROTONDI, 1922, p. 127;
and Liber notitiae sanctorum…, 1917.
4 See LANGÉ, 1967. Carlo Bertelli ascribed essential role in making the murals
visible to Adriano Peroni. See C. BERTELLI, Relazione preliminare sulle recenti
scoperte pittoriche a Torba, in Longobardi e Lombardia: aspetti di civiltà longobarda.
Atti del 6° congresso internazionale di studi sull’alto medioevo (Milano, 21-
25 ottobre 1978), Spoleto 1980, I, pp. 205-218, esp. 205 or S. LOMARTIRE,
Torba, Monastero di Santa Maria, Torre, in Pittura tra Ticino e Olona, Varese e
la Lombardia nord-occidentale, ed. M. Gregori, Milano 1992, pp. 215-216. For
the mentioned Peroni’s study, see A. PERONI, O svjazi fresok cerkvi Santa Marija
v Kastel’seprio i baziliki San Sal’vatore v Brešii (in Russian), in Vizantija južnye
Slavjane i drevnjaja Rus, Zapadnaja Evropa, ed. V. N. Graščenkov, Moskva
1973, pp. 375-387, esp. 385-386.
5 See for example P. G. SIRONI, Le difese di Castelseprio nel XIII secolo, in «Rassegna
Gallaratese di Storia e d’Arte», XXXII, 119 (1973), pp. 67-73, esp. 68; ROTONDI
SECCHI TARUGI, 1973; A. PERONI, Spunti per un aggiornamento delle discussioni
sugli affreschi di S. Maria di Castelseprio, in «Rassegna Gallaratese di
Storia e d’Arte», XXXII, 119 (1973), pp. 19-27, esp. 27; S. MAZZA, Il complesso
fortificato di Torba, in «Sibrium», XIV (1978-1979), pp. 187-215, esp. 213; R.
BAZZONI, Il nucleo altomedioevale di Torba, in «Sibrium», XIV (1978-1979), pp.
217-218; R. M. KLOOS, Die Inschriften im 1. Stock des Turmbaues von Torba, in
Longobardi e Lombardia…, 1980, pp. 219-224; or BERTELLI, 1980.
6 See for example BOGNETTI, 1948; K. WEITZMANN, The Fresco Cycle of S.
Maria di Castelseprio, Princeton 1951; E. KITZINGER, Byzantine art in the period
between Justinian and Iconoclasm, in Berichte zum XI. Internationalen Byzantinisten-Kongress,
IV, 1, München 1958, pp. 1-50; P. D. LEVETO, Castel Seprio:
Architecture and Painting, Dissertation thesis, Indiana University, 1985; or J.
MITCHELL - B. LEAL, Wall pantings in S. Maria foris portas (Castelseprio) and
the tower at Torba. Reflections and reappraisal, in Castelseprio e Torba…, 2013,
pp. 311-345.
7 See C. BERTELLI, Gli affreschi nella torre di Torba, Milano 1988.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 5
Arte Lombarda | MARTIN F. LEŠÁK
6
his ideas and interpretations have primarily shaped our understanding
of the wall paintings from then on8. In fact, the second,
and so far, the last in-depth study concentrating specifically on
the tower’s decoration, is an article written by John Mitchell and
Bea Leal from 2013. Mitchell and Leal published their study in
the monumental volume Castelseprio e Torba: sintesi delle ricerche
e aggiornamenti, edited by Paola Marina De Marchi9 – largely
following Bertelli’s suggestions. The opinions of authors diverge
on the function of the upper floor chapel. While Bertelli suggested
it was used as a main liturgical space of the convent before
a proper church had been constructed10, authors of the
most recent article challenged his view. Mitchell and Leal believed
that it was decorated and redesigned «expressly for funerary
use […] as a chapel for memorial masses, remembering and
praying for the salvation of deceased sisters and benefactors»11.
The scarce historiography of the tower provides answers to
art-historical questions; however, some of them are based on imprecise
arguments and observations, and therefore require our
vigilance and a reconsidering. Firstly, after focusing on the little
scholarly attention the tower at Torba has attracted in the past,
the present article will attempt to introduce an in-depth reconsideration.
We will discuss the relation between the decoration
of the two floors, a proclaimed stylistic isolation of the murals
and consequently, we will touch on the topic of their dating. Finally,
reflecting on the iconography of the upper room’s visual
scheme, we will propose a new layer of understanding of the wall
paintings’ raison d’être and their function. In other words, in the
last section of the article, we will be interpreting the choice to
reuse the military structure in a new sacral context.
The Reborn Tower at Torba
The murals of the tower at Torba are one of the most coherently
preserved decorations from the 8th and 9th centuries created in
the West. Here, we specifically address the iconography of the
upper chamber’s visual scheme, not the condition of its preservation.
To understand the lack of attention given to the study of
this object12, we should not forget Renato Bazzoni’s claim from
1987, celebrating not a survival but a «rebirth»13 of the tower at
Torba, referring to its history full of struggle and affliction.
The first centuries of the tower’s existence are shrouded in
mystery. We can only assume that during the turbulent period
after the fall of the Western Roman Empire, the tower retained
its original defensive function14. Based on the architectural layers
of the structure itself15, the wall paintings preserved within
its walls and the historical development of the region, we can
presume, it was during the 8th century the tower’s primarily military
character transformed into a religious one also, becoming
a part of a Benedictine convent16.
8 See LOMARTIRE, 1992; S. LOMARTIRE, La pittura medievale in Lombardia, in
La pittura in Italia, L’Altomedioevo, ed. C. Bertelli, Milano 1994, pp. 47-89;
F. DE RUBEIS, Le iscrizioni della Torre del monastero di Santa Maria di Torba:
un problema paleografico, in Castel Seprio e Vico Seprio. Aggiornamenti, atti del
convegno (Castelseprio - Torba, 22 settembre 2001), ed. C. Benati, Castelseprio
2002, pp. 83-100; M. ROSSI, La tradizione dell’antico nella pittura altomedievale
del territorio del Seprio, in Medioevo: il tempo degli antichi, ed. A. C.
Quintavalle, Milano 2006, pp. 257-266; M. ROSSI, Il problema Castelseprio,
in L’ VIII secolo, un secolo inquieto, atti del convegno internazionale di studi
(Cividale del Friuli, 4-7 dicembre 2010), ed. V. Pace, Cividale del Friuli 2010,
pp. 131-137; ROSSI, 2011.
9 See MITCHELL - LEAL, 2013.
10 See BERTELLI, 1988, p. 47.
11 See MITCHELL - LEAL, 2013, p. 344.
12 For example, the tower was not mentioned at the international conference
L’VIII secolo…, 2010.
13 Renato Bazzoni writes about «rinascita», see R. BAZZONI, Torba non più dimenticata,
in «Sibrium», XIX (1987-1988), pp. 153-164, esp. 154.
14 See P. M. DE MARCHI, Castelseprio e il suo territorio in età longobarda e carolingia,
in Castelseprio e Torba..., 2013, pp. 15-44, esp. 15, and BAZZONI, 1997,
p. 17. It has been suggested that the tower’s position was supposed to reduce a
probability of a successful attack from the valley and guarantee a water supply
for the whole castrum. See BOGNETTI, 1958-1959, p. 21; SIRONI, 1973, p. 69;
and ROTONDI SECCHI TARUGI, 1973, pp. 87-88. Cf. LANGÉ, 1967, pp. 36, 39.
15 See SCILLIA, 2013, p. 109. Cf. BROGIOLO, 2016, p. 228.
16 Renato Bazzoni suggested this change was feasible only in the Longobard
period during which the military centers turned into places of trade and encounters.
See BAZZONI, 1997, p. 17. Yet, it seems to also be possible that the
nunnery was founded only after the fall of Regnum Langobardorum in 774. On
the fall of Regnum Langobardorum and the Lombard king’s politics at the
1. The tower at Torba (Castelseprio), exterior. 2. The castrum Castelseprio, general plan.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 6
Prayers and Protection:TheTower atTorba Reconsidered
7
The first historical source referring to St. Mary’s monastery
at Torba dates to the 22nd of December 104917. Neither this document
nor more recent sources18 reveal much of the convent’s
life, let alone of an important decorative campaign. Nevertheless,
life at Torba must have undergone a change after the famous
destruction of Castelseprio in 1287. The castrum was demolished
under the command of Milanese Archbishop Ottone
Visconti and only sacred places within the complex were spared
of the terror19. That is most likely the reason why the tower at
Torba withstood the tragic fall of Castelseprio20. The Benedictine
nunnery retained some of its possessions even after this
event, yet the life without the castrum was complicated21. Economic
troubles22, together with the convent’s isolated and dangerous
position within a now depopulated area, led to its second
and final23 abandonment in 148124. The departure of the nuns
truly meant “death” for the tower and the murals within it.
Thanks to several documents25, it is possible to deduct that the
former abbey was rented to farmers, as soon as it had been abandoned
by the nuns26. It was used in the everyday life of its new
residents until the second half of the 20th century.
It is no surprise that scholars found the former abbey in a miserable
condition. In 1967, Santino Langé confirms the poor condition
of the abbey by saying: «attualmente si presenta infatti
come un cascinale piuttosto in cattivo stato, quasi completamente
ricopero di edera, con intonaci fatiscenti e finestre casualmente
ricavate, prive di sottolineature architettoniche»27. Langé knew the
restoration of the murals hidden under the layers of whitewash
was necessary (fig. 3). He also knew the process could not be initiated
until after the farmers left (at some point before 1970), and
after the property has been purchased by Giulia Maria Mozzoni
Crespi in 197628. The new owner, one of the founders of FAI
(Fondo Ambiente Italiano), officially donated the Torba Abbey to
beginning of the 70’s, see for example T. INDELLI, Langobardia, I Longobardi
in Italia, Padova 2013, pp. 46-50; J. T. HALLENBECK, Pavia and Rome: The
Lombard Monarchy and the Papacy in the Eight Century, Philadelphia 1982, pp.
137-174; D. S. SEFTON, Pope Hadrian I and the Fall of the Kingdom of the Lombards,
in «The Catholic Historical Review», LXV (1979/2), pp. 206-220; P.
PARTNER, The Lands of St. Peter, London 1972, pp. 29-33.
17 This document records a donation of two estates, the second of which is located
near Torba, by the Archbishop of Milan, Guido da Velate, to the nuns of
S. Maria del Monte in Varese: «secundo campo iacet ad l. ubi dic. Ciscalina:
da munte S. Marie de monesterio qui dic. Turba, da alia parte aliquantum in
via, et est tab. XXX». See Regesto di S. Maria di Monte Velate, sino all’anno
1200, ed. C. Manaresi, Roma 1937, p. 23, item 32. For this and other historical
sources regarding Torba, see also PANIN, 1983, p. 94; E. T. VILLA, Le
monache di Torba e i Castiglioni di Casciago, in «Archivio Storico Lombardo»,
CIII (1979), pp. 303-330, esp. 304; G. D. OLTRONA VISCONTI, Il monastero
di Torba (note per una cronologia), in «Rassegna Gallaratese di Storia e d’Arte»,
XIX, 75 (1960), pp. 175-177; and ROTONDI, 1922, esp. 126.
18 We refer to a document from 1124, which records a donation of the land «a
meridie Sancte Marie de Turba», published in Le pergamene della Basilica di S.
Vittore di Varese (899-1202), ed. L. Zagni, Milano 1992, pp. 77-78, item 48.
See also Regesto di Santa Maria..., 1937, p. 264, item 385 for a document from
1198 which reads: «a mane heredum Laurencii Calbi, a meridie abatisse de
Turba, a sero S. Marie de Monte».
19 The destruction was ordered after thirty years of quarrel between Sepriese
nobility and the Milanese archbishop. See BAZZONI, 1997, p. 18; M. O. H.
CARVER, S. Maria Foris Portas at Castel Seprio: A Famous Church in a New Context,
in «World Archaeology», XVIII (1987/3), pp. 312-329, esp. 312; LEVETO,
1985, p. 4; or BOGNETTI, 1948, pp. 357-362. The order of the archbishop was
clear: «Castrum Seprium destruatur, et destructum perpetuo teneatur, et nullus
audeat vel praesumat in ipso Monte habitare». See GIULINI, 1855, p. 698 or
BOGNETTI, 1948, p. 496.
20 Renato Bazzoni argues that the tower would probably survive even if it was
not protected by the presence of nuns, since a demolition of a structure so architecturally
advanced would be an enormously difficult task. The tower at
Torba was, as argues Bazzoni, more resistant than walls surrounding the whole
complex. See BAZZONI, 1987-1988, pp. 153-154.
21 On the convent and life of the nuns, see especially VILLA, 1979 and PANIN, 1983.
22 With a lease contract signed in 1357, nuns rented all their estates to Guarnerio
and Obizzo Castiglione. This is the first sign of economic problems, which
probably only deepened during the following years. See VILLA, 1979, pp. 308
and 319-321; OLTRONA VISCONTI, 1960, p. 176; and ROTONDI, 1922, p. 128.
23 The nuns of the abbey left the convent for the first time in 1426. They
joined nuns from the convent of S. Antonino di Luvinate, which also belonged
to the Benedictine order. Giuseppe Rotondi mentions four documents relating
to the decision of nuns, in which we find also a reference to an ongoing local
war: «bellorum turbines». See ROTONDI, 1922, p. 129. See also VILLA, 1979,
p. 308 or OLTRONA VISCONTI, 1960, p. 176. The fusion between the two convents
lasted for approximately thirty years. At some point during the period
between 1457 and 1460, the nuns returned to their home at Torba, renovated
some of its parts only to abandon it again, the last time being in 1481. See ROTONDI,
1922, p. 130 and PANIN, 1983, pp. 95-97.
24 In 1481, Ubertetto Pusterla, the uncle of the then abbess, Margherita
Pusterla, donated to the nuns a house in Tradate (close to the church of S. Sepolcro),
into which they moved. S. Sepolcro Abbey in Tradate became their
new seat. See VILLA, 1979, p. 309 and ROTONDI, 1922, pp. 131-132.
25 Yula Panin argues that document from 1436, which records an anxious effort
of the nuns to ensure the right to draw water from an irrigation ditch, so their
lawns could be watered, suggests a presence of a tenant farmer. See PANIN, 1983,
p. 94. Ettore Tito Villa believes that the income from their estates enabled them
to live independently even within the different convent. See VILLA, 1979, p. 308.
26 See BAZZONI, 1997, p. 18; ROTONDI, 1922, pp. 133-134; and PANIN, 1983,
p. 94. In fact, the complex remained in the possession of the Sisters until 1799,
when their new convent at Tradate was suppressed in accordance to the policy
of Napoleon. See also T. TIBILETTI, Testimonianze letterarie e indagini archeologiche
a Castelseprio. Interpretazioni, problemi, spunti di riflessione, in Castelseprio
e Torba…, 2013, pp. 45-86, esp. 81 and OLTRONA VISCONTI, 1960, p. 177.
27 See LANGÉ, 1967, p. 34.
28 See BAZZONI, 1997, p. 18.
3. The tower at Torba, second floor, east wall, before 1977.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 7
Arte Lombarda | MARTIN F. LEŠÁK
8
the above-mentioned organization in 197729. Consequently, the
restoration works30 led by Renato Bazzoni (FAI) and Pinin Brambilla
Barcilon started31. Thanks to these careful interventions the
tower rose from the ashes, or rather, returning to the words of Bazzoni,
it was truly «reborn».
The tower was under private ownership up until 1976, making
it almost inaccessible and in a poor state of preservation. This fact
is the major cause for its relatively scarce historiography. Another
valid reason for this, as Langé pointed out in 1973, might also be
the traditional perception of the complex as a separate convent
rather than an integral part of the castrum32. The castrum itself attracted
scholarly attention thanks to the discovery of the Santa
Maria foris portas’ frescoes in 194433. The latter wall paintings with
their Hellenistic character aroused a great and widespread interest34,
understandable in the context of the forties of the 20th century.
During this time scholars such as Erwin Panofsky, Ernst
Kitzinger and Kurt Weitzmann intensely dealt with the question
of medieval renaissances35. It seems the Hellenistic forms and highquality
frescoes in Santa Maria foris portas, grasped the attention of
scholars away from the highly stylized and worse preserved paintings
from the very beginning. However, it must be emphasized that
the first discovery in the Santa Maria foris portas in 1944 and the
increased interest in the area linked to it, eventually led to the uncovering
of the painted walls at Torba.
Two Chambers and Layers of Painted Plaster
We have already pointed out the rare integrity of the preserved
visual scheme of the chapel in the second floor of the tower. In
the following paragraphs, we will focus on its iconography and
its relation to the layers of the remaining painted plaster in the
chamber below.
The focal point of the upper cycle, which unfolds on the walls
of a small, nearly square room, is the enthroned beardless Christ
with a cruciform nimbus represented between two secured windows
of the east wall (fig. 5). He is flanked by two angels. The one better
preserved on Christ’s left side, holds an orb in his left hand, painted
by using a pair of compasses (fig. 4). A group of four people is represented
on the right side of the wall, beyond the window.The closest
to the main group is John the Baptist holding a lamb, or rather
a medallion with a lamb. The following saint may be interpreted as
St. Peter, based on his white hair and beard. It is difficult to specify
whom the other two figures represent, though; it is possible they are
apostles36. There are no visible fragments of saints to be found on
the left side of the wall. Nevertheless, following the well-known
iconography of Deësis, we can presume the iconic presence of Virgin
Mary37, though different possibilities cannot be excluded38. The
probable position of her depiction, as John Mitchell and Bea Leal
observed, could be accentuated by as well as deduced from a large
cross on a dado painted as a velum in the lower part of the wall39.
The opposite west wall is divided by a single axial window
(fig. 6). Eight figures – all of them most likely male saints – are
depicted above a dado painted as a marble on the left side of the
29 See M. MAGNIFICO, Presentazione, in Castelseprio e Torba…, 2013, p. 12.
30 Thanks to the historical development and late rediscovery of the tower, the murals
avoided in some cases inconsiderate interventions of the later centuries. In fact,
only two interventions have been carried out. The first started almost immediately
after the monument had been acquired by the FAI.To be more specific, the restorer
received the official authorization for proceeding to the works from Soprintendenza
on the 25th of August 1977. The second restoration campaign started only
thirty years later in 2007. It was motivated by a continual degradation caused primarily
by an inadequate temperature and humidity. The independent restorer
Pinin Brambilla Barcilon led both restoration campaigns. Sadly though, a documentation
has been preserved only to the later and much more delicate intervention,
i. e. to those from 2007. We can explain this lack of earlier documents by laconic
commentary of Barcilon, who stated that there is any detailed documentation
since it was not required by the commission at the end of the 70’s. I would
like to thank Soprintendenza Archeologia, Belle Arti e Paesaggio per la Città
Metropolitana di Milano as well as FAI for their cooperation. For the documents
concerning the restoration as well as the tower’s history from 1960’s, see Milano,
Archivio Monumenti della Soprintendenza, folders o/2 4580 and o/2 4580 II. For
the early photos of the tower and the murals see Milano, Archivio Fotografico della
Soprintendenza, entry «(VA) Torba». See also Il notiziario del FAI, from 1980 on.
31 See TIBILETTI, 2013, p. 85; BAZZONI, 1987-1988, p. 154; and BERTELLI
1980, p. 205.
32 S. LANGÉ, L’inserimento del monastero fortificato di Torba nel dibattito di
Castelseprio, in «Rassegna Gallaratese di Storia e d’Arte», XXXII, 119 (1973),
pp. 75-79, esp. 75.
33 See BOGNETTI, 1948.
34 See above note 6.
35 See K. WEITZMANN, The Joshua Roll: a Work of the Macedonian Renaissance,
Princeton 1948. E. KITZINGER, Mosaic Pavements in the Greek East, and the
Question of a «Renaissance» under Justinian, in Actes du VIe Congrès International
d’Études Byzantines, Paris, 27th July - 2nd August 1948, II, Paris 1951, pp. 209-
223; and E. PANOFSKY, Renaissance and Renascences, in «The Kenyon Review»,
VI (1944/2), pp. 201-236.
36 At first, Bertelli imagined that nine apostles (three on the east wall and six
on the south one) are depicted on Christ’s left side. Three more, according to
the author, were represented on the east wall, following the possible representation
of Mary. In 1988, Carlo Bertelli divided the group of apostles as follows:
eight on the east wall; one on the south wall; and three on the north wall. See
BERTELLI, 1980, pp. 210-211. Cf. BERTELLI, 1988, p. 26.
37 See BERTELLI, 1980, p. 210 and BERTELLI, 1988, p. 26. For the iconic presence,
see H. BELTING, Iconic Presence. Images in Religious Traditions, in «Material
Religion» (published online), XII (2016/2), 235-237.
38 We must recall the 7th century text from Constantinople describing a miracle
of St. Artemios. This miracle describes a chancel screen on which has been
depicted Christ between John the Baptist and the church patron St. Artemios
in large icons. See H. BELTING, Likeness and Presence, A History of the Image before
the Era of Art, Chicago 1994 (I ed. 1990), pp. 233, 502.
39 See MITCHELL - LEAL, 2013, p. 337.
4. The tower at Torba, second floor, east wall, detail of an angel.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 8
Prayers and Protection:TheTower atTorba Reconsidered
9
5. The tower at Torba, second floor, east wall.
6. The tower at Torba, second floor, west wall.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 9
Arte Lombarda | MARTIN F. LEŠÁK
10
central opening. Another eight saints are represented to the right,
north of the window. Thanks to the inscription (SCA EVFEMIA,
St. Euphemia) and the physiognomy of preserved faces, we can
assume that they are all female. A praying nun in a dark habit is
painted below each of these eight saints40. The nuns’ position in
the lower part of the wall can be explained. Firstly, they were not
an integral part of the heavenly realm and secondly, there was a
need to divide the minimalistic, square, originally only military
chamber into functional sections41. It truly seems, the actual
nuns entering the chamber were conducted by the scheme itself.
Put differently, the image seems to lead their movement towards
their designated places within the chapel. Due to the absence of
attributes or tituli, apart from St. Euphemia, a precise identification
of the figures depicted on this wall is impossible.
The same could be said about fragments of six figures in the
left part of the south wall42. In fact, the only clearly recognizable
person on this wall is the Mother of God holding the child
Christ – Hodigitria (fig. 7, tav. 2) – represented on the right side
of the current entrance43. A smaller figure, a woman dressed in
lay-clothes with a square halo (this motif could suggest that she
was still alive when the decoration was being created) offers a
candle to the Virgin44. Carlo Bertelli identified her as an unknown
noblewoman, «comitissa Seprii», one of the benefactors
of the murals45. There are four more saints standing on the right
side of the offering scene and one martyr depicted on the left,
crowning a cleric dressed as a bishop46. Due to his lower position
outside of the frame, the absence of a nimbus, and the coronation
act itself, the latter has been identified as another donor, or
«at least a ranking ecclesiastic with a particular interest in the
community at Torba»47.
40 Saverio Lomartire believes that we may be facing the representation of all
nuns who lived in the convent at the time the walls were painted. Consequently,
this image has been defined as a «collective portrait». See LOMARTIRE, 1992,
p. 215 and BERTELLI, 1980, p. 213. In addition, John Mitchell with Bea Leal
suggested «the saints above may represent the name patrons of the nuns below».
See MITCHELL - LEAL, 2013, p. 342. Since we do not have any information
about the sisters’ life from the period of murals’ creation, such hypotheses
need to be treated with caution.
41 The alternation of velum and marble imitations also belongs to this necessity.
See MITCHELL - LEAL, 2013, pp. 337-339. These authors claim that the dado
is used «to articulate and introduce hierarchical structure». See also BERTELLI,
1988, p. 26.
42 The most eastern figure on the south wall has been generally interpreted as
an apostle. See for example ROSSI, 2011, p. 40 and BERTELLI, 1988, p. 26. The
other five figures were read by Bertelli in 1980 as apostles also. See BERTELLI,
1980, pp. 210-211. However, in 1988, he claimed they are martyrs. Cf.
BERTELLI, 1988, p. 26.
43 It was especially Carlo Bertelli and Jean-Claude Genoud who dealt with the
question of the chamber’s entrance. They described three stages of the upper
floor existence: First, the room was accessible through an inner stairway; it had
only windows, no doors; the walls were bare. Second, the door on the south
wall was created; the walls have been whitewashed (the simple red band was
painted in the lower part of the walls). Third, the southern entrance was walled
up; the door in the north wall emerged; the walls were decorated with the mural
cycle. See J.-C. GENOUD - C. BERTELLI, Torba, in «Etudes de lettres, Revue
dela Faculté des lettres de l’Université de Lausanne», CCXIII (1987/2-3), pp.
67-83, esp. 70 and BERTELLI, 1988, pp. 12-13, 17. Cf. BERTELLI, 1980, pp.
208-209. Ideas of Genoud and Bertelli were not reflected in the more recent
studies. See MITCHELL - LEAL, 2013, p. 340.
44 See MITCHELL - LEAL, 2013, p. 341; BERTELLI, 1988, pp. 28 and 47. Regarding
this figure, Marco Rossi stated: «potrebbe forse trattarsi della contemporanea
badessa, o della fondatrice del monastero, o di una nobile protettrice della
comunità». See ROSSI, 2011, p. 44. The belief that a square nimbus was the
marker of a living person is based on the 9th century description of the portrait
of Gregory I in SS. Andrea e Gregorio at Rome. It must be noted, however, that
this bizarre piece of iconography does not always have to identify someone who
is alive, as reminded George Bishop Tatum. See G. B. TATUM, The Paliotto of
Sant’Ambrogio at Milan, in «The Art Bulletin», XXVI (1944/1), pp. 25-45, esp.
28. See also J. OSBORNE, The Portrait of Pope Leo IV in San Clemente, Rome: a
Re-Examination of the So-Called ‘Square’ Nimbus, in Medieval Art, in «Papers of
the British School at Rome», XLVII (1979), pp. 58-65, esp. 64. See also S.
FISHER, The Square Halo and Other Mysteries of Western Art, New York 1995.
45 See C. BERTELLI, Pittura di età carolingia nell’Italia settentrionale e a Castelseprio,
in «Sibrium», XIX (1987-1988), pp. 81-86, esp. 86 and BERTELLI,
1988, pp. 28, 47.
46 The cleric wears a long tunic, chasuble, and an embroidered stole. Carlo
Bertelli suggested, on the bases of comparison of this coronation with that of
Angilbert II depicted on the famous Golden Altar of S. Ambrogio in Milano,
that the martyr who crowns the bishop was in fact the saint to which the
chapel was dedicated. See BERTELLI, 1988, p. 28 and ROSSI, 2011, p. 44.
47 See MITCHELL - LEAL, 2013, p. 341. See also ROSSI, 2011, pp. 40-44 and
BERTELLI, 1988, pp. 28, 47.
7. The tower at Torba, second floor, south wall, detail with Hodigitria.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 10
Prayers and Protection:TheTower atTorba Reconsidered
11
Only few fragments of the decorative scheme are preserved
on the north wall of the room. The lower parts of two figures
are recognizable between two openings, opposite today’s door.
The one, next to the right opening, seems to depict a different
ecclesiastic authority, judging from a richly ornamented stole.
The other was identified by Bertelli as a saint or a martyr.
Mitchell and Leal, however, believed that it represents a material
benefactor48. Furthermore, there used to be a panel between
these two figures with an inscription, but unfortunately,
today it is undecipherable49. An essential piece of the cycle
survived at the western end of the same wall, namely a lion
with a jeweled book in its paws, supporting the lower right
part of the mandorla (fig. 8, tav. 1). Considering the popularity
of the visual scheme known as Maiestas Domini50 around
800, we follow a hypothesis already presented by Bertelli, who
stated that the mandorla «doveva contenere un’immagine del
Redentore in trono»51.
The upper room’s iconography is rich in art-historical significance;
therefore, two more observations must be made regarding
it. Firstly, saints, martyrs, and ornamental motifs are painted
on the jambs of openings. Secondly, remains of the red decorative
band, which had been painted around the whole room before
the creation of the above-described cycle are clearly visible
in the lower part of the walls. In fact, Vincenzo Gheroldi recognized,
on the grounds of stratigraphic evidence, geomorphological
analysis and the recognition of different working techniques,
three layers of painted or whitewashed plaster in this
room52. The first and second of these layers are related to the
red band. The murals are painted on the third one, which is superimposed
on the other two and separated into pontatas. Here
we must stress that, subtle changes and substitutions were made
on this third painted plaster, traceable on the south and west
walls (fig. 9). As Gheroldi observed, they are «rifacimenti in
corso d’opera anziché sostituzioni di altra fase»53. In other
words, as Bertelli already presupposed54 and Gheroldi supported
via his research, the whole mural cycle of the upper chamber
originated from a single decorative campaign.
48 Cf. BERTELLI, 1988, p. 33 and MITCHELL - LEAL, 2013, p. 340.
49 This panel, according to Mitchell and Leal, «marked the dedication of the
chamber as a sacred space». See MITCHELL - LEAL, 2013, p. 340.
50 For this iconography, see Y. CHRISTE, Une ‘Maiestas Domini’ de type Lombard,
in «Arte Lombarda», 102-103 (1992/3-4), pp. 5-13; Y. CHRISTE, L’ Apocalypse de
Jean: sens et développements de ses visions synthétiques, Paris 1996, pp. 123-131; B.
KÜHNEL, The End of Time in the Order of Things, Regensburg 2003; A.-O. POILPRÉ,
Maiestas Domini. Une image de l’Église en Occident, Paris 2005; or V. CANTONE
VALENTINA, Ars monastica. Iconografia teofanica e tradizione mistica nel
Mediterraneo altomedievale (V-XI secolo), Padova 2008. See also LOMARTIRE, 1992,
p. 215. The last-mentioned uses the term also for the enthroned Christ on the
east wall.This is understandable, since this denomination means in general, as Peter
K. Klein states, «the enthroned Christ in majesty». I have chosen to apply this
term only to the north wall image both to avoid confusion and to follow contemporary
art-historical jargon. P. K. KLEIN, Introduction: The Apocalypse in Medieval
Art, in The Apocalypse in the Middle Ages, eds. R. K. Emmerson and B. McGinn,
Ithaca-London 1992, pp. 159-199, esp. 163. Cf. É. PALAZZO, L’invention chrétienne
des cinq sens dans la liturgie et l’art au Moyen Âge, Paris 2014.
51 See BERTELLI, 1988, p. 29. Illuminations of the famous Trier Apocalypse allow
us to speculate that a depiction of the Lamb of God, or even the triumphant
horned lamb of Revelation, was painted inside of this mandorla. On
the manuscript, see for example J. SNYDER, The Reconstruction of an Early
Christian Cycle of Illustrations for the Book of Revelation - the Trier Apocalypse,
in «Vigiliae Christianae», XVIIII (1964), pp. 146-162 or P. L. KLEIN, L’Apocalypse
de Trèves, in Le Jugement dernier entre Orient et Occident, ed. V. Pace, Milano
2006, pp. 37-39.
52 See V. GHEROLDI, I rivestimenti aniconici e i dipinti murali della Torre del
Monastero femminile benedettino di Torba, in Castelseprio e Torba…, 2013, pp.
293-310, esp. 300-303. Cf. BERTELLI, 1988, pp. 12-13, 17.
53 The hand of the donor who is depicted next to the Madonna on the south
wall was at first higher, also the color of her dress was different. On the west
wall, the head of the second sister in the row of portrayed women (from the
left) was originally higher. See GHEROLDI, 2013, p. 308.
54 See BERTELLI, 1980, pp. 211-213. See also BERTELLI, 1988, pp. 22, 29.
8. The tower at Torba, second floor, north wall, detail.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 11
Arte Lombarda | MARTIN F. LEŠÁK
12
Multiple layers of painted plaster are also preserved in the
lower room. The oldest one includes a dado painted in imitation
of polychrome opus sectile in the lower section of the east
wall. It is identical to the same motif on the west wall of the
upper chamber55. Secondly, there are remaining fragments of a
large narrative composition above the painted dado56. Mitchell
and Leal suggested an alluring, yet untenable hypothesis considering
the murals’ state of preservation. According to the
scholars, fragments «might be consonant with a representation
of the Maries speaking with the Angel at the tomb of the risen
Christ»57. Thirdly, a border defining the space of the east recess
is recognizable on its reveals. This delimitation would suggest
that when this room received its first decoration, the walls of
the recesses were left bare, with no embellishments. Remains
of an unspecified, painted plaster are also preserved on the
south wall.
Several figures are displayed in the alcoves of the southeastern
corner. We can still recognize three saints and three praying
women – most likely nuns. One is called Aliberga, as a
palimpsest titulus next to her reveals58. Furthermore, an ochre
cross (or Christogram59) is depicted on a white background
with a long inscription around it60. These murals were painted
on a subsequent layer of plaster several years after the first layer
of the lower room had been finished61. Carlo Bertelli believed
they originated in the same period as the cycle of the
upper room62. John Mitchell with Bea Leal suggested the first
floor, including all its decorative phases, was completed even
before the work on the murals above began63. However,
Gheroldi disproved both latter hypotheses by technological
research. The scholar demonstrated, it is the first layer of plaster
of the lower chamber which corresponds to the third layer
above; the layer on which the murals were created64. The typology
of the imitated opus sectile visible on the walls of both
the lower and upper chamber strongly supports his conclusion.
Consequently, we can imagine an impressive decorative
campaign resulting in murals covering four walls of both
chambers; most likely related both in its iconography as well
as its functional use65.
55 See GHEROLDI, 2013, p. 295, fig. 4a.
56 We refer to a fragment of an inclined head and the lower part of a standing
figure, both turning left; and almost unidentifiable painted elements in
the upper left section of the central field – it has been suggested that these
used to form a part of an architectural structure. See MITCHELL - LEAL,
2013, pp. 328-329.
57 See MITCHELL - LEAL, 2013, p. 329.
58 For reading this palimpsest inscription, see BERTELLI, 1980, p. 214; KLOOS,
1980, p. 219; BERTELLI, 1988, pp. 40-41; ROSSI, 2011, p. 46; or LOMARTIRE,
1992, p. 216.
59 Rudolph M. Kloos suggested the upper part of the image was comprised of
the “P” shape. See KLOOS, 1980, p. 220.
60 The inscription in ochre letters spread on the white background around the
cross comprises of the Greek redemptive letters Alpha (Α) and Omega (ω)
hanged on three chains from both limbs of the cross; letters which can be read
as Hec lex dei vera around the chains; and a long titulus located on both sides
of the cross under the letters Alpha and Omega. The reading of this text is not
an easy task. The major role in its interpretation played the word ALEXANDRIA.
In 1980, Kloos suggested that it is a reference to the Egyptian city from which
a relic of an unknown saint had been taken to Torba (KLOOS, 1980, pp. 223-
224). Yet, Bertelli proposed in the same publication that the person transferred
from Alexandria to Torba could have been alive. In other words, the mention
of the city was, in his opinion, only an important piece of information regarding
the place of a deceased’s origin (BERTELLI, 1980, pp. 214-215). Then, in
1992, Saverio Lomartire asserted that the text refers to an Abbess Alessandra.
It thus became this abbess, and not relics from Alexandria, for whom the «eternal
rest [in the bosom of] Abraham, Isaac, and Jacob» in the inscription is
asked (LOMARTIRE, 1992, p. 216). This reading together with the cross’s typology
and the resemblance of the recesses with arcosoliums provided Mitchell and
Leal arguments to state that the lower chamber was used in a funerary and
memorial sense. See MITCHELL - LEAL, 2013, p. 326.
61 See GHEROLDI, 2013, pp. 293-299.
62 See BERTELLI, 1988, p. 40.
63 See MITCHELL - LEAL, 2013, pp. 336 and 344.
64 See GHEROLDI, 2013, p. 294.
65 Remains of a painted plaster are also visible on the south wall of the lower
room. Unfortunately, we cannot determine neither the fragments’ iconography,
nor whether they belong to the first, second, or even later decorative phase of
this chamber. See GHEROLDI, 2013, p. 299. Nevertheless, its existence is rather
in accordance with the statement, that all walls of the lower room were originally
decorated with an elaborate scheme, than it is in contradiction to this hypothesis.
See also MITCHELL - LEAL, 2013, p. 329 and BROGIOLO, 2016, p. 227.
9. The tower at Torba, second floor, west wall, detail.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 12
Prayers and Protection:TheTower atTorba Reconsidered
13
The Date and the Stylistic Isolation
Regarding the question of dating, we first need to focus on the
so far omitted construction history of the tower. We have already
established that its origins extend to the 5th or 6th century at the
very latest66. Its upper floors, however, do not seem to belong to
the initial phase of construction. Angela Scillia dated both levels
in which the murals are preserved before the 8th century, yet,
Gian Pietro Brogiolo attributed the upper chamber to a later period.
The latest 14C (Carbon-14) analyses of the mortar are taken
from the external masonry in the area around the entrance
into the chapel. These analyses led the scholar to suggest a generic
dating of this upper elevation between the 9th and 10th centuries67.
This result, which would give us a terminus post quem
for the murals, should be however, as the author itself emphasizes,
taken with caution, considering difficulties brought by
analyses of a mortar. In addition, the question of entering the
room, as Bertelli and Jean-Claude Genoud pointed out, seems
to be a problematic one, due to structural changes. These
changes, especially regarding the openings, occurred after the
wall paintings were already finished68. Consequently, it is still
necessary to search for different methods of examination to
bring us closer to a correct solution.
Paleography and epigraphy played a significant role in dating
the murals69. The reasonably certain terminus ante quem was
provided by the «Margarite» graffito, in the form of a minuscule
engraving into the surface of the painted plaster beneath Christ’s
representation on the east wall (fig. 10)70. Carlo Bertelli stressed
that writing on a piece of papyrus differs greatly from an engraving
into a wall, nonetheless, he compared the two. According to
him, the characteristic form of the letter «r» is comparable to the
late 8th century North Italian writings71. Furthermore, the letters
«g» and «a», in his opinion, represent the Carolingian type72.
These forms, as the author claimed in 1988, would not appear
after the second decade of the 9th century73.
Painted inscriptions preserved in majuscule located in both
chambers helped scholars determine the date. Flavia De Rubeis,
for example, suggested the murals originated towards the end of
the 8th century, because the painted inscriptions correspond to
the tituli documented in the Longobard mural cycles of Brescia
(San Salvatore) and Cividale (Longobard Temple)74. Scholars
have reached similar conclusions on dating, thanks to analyses
of both painted and engraved inscriptions on the tower’s inner
walls. The dates oscillate between fifty years around the year
80075. A question that arises is whether the art historical concept
of style supports those conclusions76.
Before attempting to resolve it, however, we must address
the technique of the murals and the condition of their preservation,
inherently linked to it. The wall paintings were not
rendered in buon fresco but fresco-secco technique. The first
draft of the scheme was sketched into a still wet plaster (separated
into pontatas), whereas most of the details (facial features,
folds of draperies, etc.) were applied on the already
dried surface77. The murals’ resistance to harmful aspects of
time, was in result, weakened from the very beginning. Thus,
after the mentioned restorative works at the end of the seventies;
including the removal of layers of whitewash covering the
paintings, an implementation of the stylistic analysis was, and
still is, problematic.
66 Cf. BROGIOLO, 2016, pp. 227-228 and SCILLIA, 2013, p. 109.
67 See BROGIOLO, 2016, pp. 227-228; BROGIOLO, 2013, pp. 216-219; and
SCILLIA, 2013.
68 See above note 43.
69 For paleographic and epigraphic analyses of Torba’s inscriptions, see BERTELLI,
1980, pp. 215-216; KLOOS, 1980; and DE RUBEIS, 2002.
70 Today, we can read only a part of original four engraved inscriptions, because
a segment of the plaster carrying them has been lost. Bertelli characterized
these inscriptions as «graffiti commemorative». See C. BERTELLI, Ultimi
studi sulle pitture di Castelseprio e Torba, in Castel Seprio e Vico Seprio, 2002,
pp. 1-8, esp. 3 and BERTELLI, 1988, p. 20. Fortunately, thanks to a transcription
provided by the scholar in 1980, we can partly reconstruct their original
form. See BERTELLI, 1980, p. 215. See also LEVETO, 1985, pp. 17-18, n. 36.
71 Bertelli compared the inscription with Ms. Phillips 1885 from the Staatsbibliothek
in Berlin. See BERTELLI, 1980, pp. 215-216. For the manuscript,
see H. DEGERING, Die Schrift, Atlas der Schriftformen des Abendlandes vom Altertum
bis zum Ausgang des 18. Jahrhunderts, Berlin 1929, p. 25 and fig. 42.
72 In the words of Bertelli himself, «Altri elementi significativi sono poi le
forme della g e della a, di tipo carolingio nonostante la parvenza più arcaica
dell’insieme dell’iscrizione». See BERTELLI, 1980, p. 216. On the Carolingian
minuscule and its ample bibliography, see L. E. BOYLE, Medieval Latin
Palaeography, A Bibliographical Introduction, Toronto-Buffalo-London 1984,
pp. 140-170.
73 See BERTELLI, 1988, p. 40. Cf. DE RUBEIS, 2002, pp. 98-99.
74 De Rubeis pointed out the similar morphology of the letters and the vertically
developed script system. See DE RUBEIS, 2002, pp. 84-88 and 98-99.
75 See for example LANGÉ, 1967, pp. 44 and 48; LANGÉ, 1973, p. 77; BERTELLI,
1980, pp. 215-216; BERTELLI, 1988, p. 40; DE RUBEIS, 2002, pp. 98-99;
MITCHELL - LEAL, 2013, pp. 336 and 344.
76 See for example J. ELSNER, Style, in Critical Terms for Art History, eds. R. S.
Nelson and R. Shiff, Chicago-London 2003 (I ed. 1996), pp. 98-109; A.
PINOTTI, Formalism and the history of style, in Art History and Visual studies in
Europe, eds. M. Rampley et al., Leiden 2012, pp. 75-90; H. WÖLFFLIN, Principles
of art history. The problem of the development of style in early modern art, eds.
J. Blower (transl.), E. A. Levy and T. Weddigen, Los Angeles 2015 (I ed. 1915).
77 See BERTELLI, 1980, pp. 211-213.
10. The tower at Torba, second floor, east wall, «Margarite» graffito.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 13
Arte Lombarda | MARTIN F. LEŠÁK
14
Furthermore, the fact that preserved medieval objects form
only a minute fragment of the originally much richer visual culture,
led Bertelli to claim the wall paintings do not have any
stylistic equivalent in a monumental decoration from the 8th and
9th centuries – neither in North Italy nor in its other parts78.
Comparisons with surviving murals cannot disprove the formal
isolation of Torba (Tempietto Longobardo at Cividale, San Salvatore
in Brescia, San Vincenzo al Volturno in Benevento, main
church of the monastery of St. John in Müstair, Sant’Ambrogio
alla Rienna at Montecorvino Rovella, Chapel of Theodotus in
Santa Maria Antiqua, etc.). In fact, there has only been one attempt
to link the murals to another wall painting. Based on the
stylistic comparison with a saint preserved in Santa Maria presso
San Satiro in Milano, Giovanni Valagussa cautiously suggested
dating the upper room scheme to the 10th century79. Because of
the thicker contours of the Milanese saint’s face, his overall starker
and more rigid character, and the fact Valagussa neglected the
already discussed analyses of inscriptions, it would be rather difficult
to support his position.
Having in mind the problematic state of preservation and the
differences between two media, we might draw our attention to
manuscripts also. Illuminations of the codex of Bishop Eginone
(769-799), written in Verona at the end of the 8th century, and
the manuscript CLXV of the Chapter Library (Biblioteca Capitolare)
in Vercelli, dated to the beginning of the 9th century have
already been linked with the wall paintings by Bertelli. The art
historian emphasized the analogies between the two being very
generic and confirming the North Italian origin of the artist80.
78 See BERTELLI, 1980, pp. 216-217 and BERTELLI, 1988, pp. 40-41.
79 See G. VALAGUSSA, Dagli inizi della pittura al Duecento, in Pittura a Milano
dall’Alto Medioevo al Tardogotico, ed. M. Gregori, Milano 1997, pp. 31-42,
esp. 33-34.
80 See BERTELLI, 1988, p. 41. For the Codex of Bishop Eginone, also called
the Codice di Egino, with four illuminations, each depicting one of the Church
fathers, see Ch. MATTKE, Mémoire et devoirs d’un évêque, l’iconographie du
‘codex’ d’Éginon, in Utilis est lapis in structura, ed. L. Pressouyre, Paris 2000, pp.
215-230 and L. CASTELFRANCHI VEGAS, L’ arte medioevale in Italia e nell’Occidente
europeo, Milano 1993, p. 23. On the manuscript CLXV of the Chapter
11. Paris, Bibliothèque Mazarine, ms. 660, f. 133r.
12. The tower at Torba, second floor, west wall, two lost heads of nuns. 13. The tower at Torba, second floor, west wall, face of a saint.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 14
Prayers and Protection:TheTower atTorba Reconsidered
15
The illumination of a musician from the codex of Institutiones
of Cassiodorus (Paris, Bibliothèque Mazarine, ms. 660)
still provides us with more than a «generic» affinity (fig. 11)81.
Its comparison with one of the nuns’ faces in the chapel, preserved
unfortunately only in the early photographs (fig. 12), is
especially striking. The physiognomy of both – the elongated
shape of their noses and faces in general, their almond shaped
eyes created by the thick contours, as well as their emphasized
cheeks – reveals a significant resemblance. The same features are
visible also on other surviving faces at Torba (fig. 13, tav. 3). In
2000, Carlo Bertelli suggested that Mazarine 660 is the copy of
a 6th century manuscript made at the order of Anselm (723-803;
Duke of Friuli, then Abbot of Nonantola Abbey). This occurred
sometime after he had returned to Nonantola from his exile at
Monte Cassino, i. e. no later than in the eighties of the 8th century82.
Following his hypothesis, we may believe, the paleographical
and epigraphical investigations of Bertelli, De Rubeis
and Rudolf M. Kloos are supported by formal aspects.
The question of dating Mazarine 660 is, however, much
more delicate83. Bernhard Bischoff, the main authority on western
medieval manuscripts, stated: «nulla si conosce della sua storia»84,
and thus, similarly to the murals’ case, we must rely only
on the style of images and words when questioning its origin.
Most recently, Giulia Orofino stated its style is in comparison to
other Nonantola Abbey’s manuscripts attributed to the first half
of the 9th century «altrimenti maturo, spiegabile con una
cronologia più avanzata, verso la fine del secolo»85. If we consider
the two generations involved and several hands which may
have been responsible for the manuscript’s rendering86, Orofino’s
reasoning seems insufficient, yet it is a more detailed explanation
than the one Bertelli had used to buttress his supposition in
200087. Which claim should be then supported?
We suggest Bertelli’s hypothesis in question of dating the
Mazarine 660. This would be the logical conclusion, considering
the Mazarine 660’s visual link to the murals, the analyses of
the inscription in the tower mentioned above and the iconographic
similarities between the enthroned Christ at Torba and
that of Godescalc manuscript from the eighties of the 8th century
(fig. 14). We use the interconnection between the murals
and the manuscript, firstly, to date the manuscript closer to the
Library in Vercelli, which is illuminated with drawings depicting the history
of councils, see Ch. WALTER, Les dessins carolingiens dans un manuscrit de Verceil,
in «Cahiers archéologiques», XVIII (1968), pp. 99-107 and J. HUBERT J.
PORCHER - W. F. VOLBACH, L’Europa delle invasioni barbariche, Milano
1980, pp. 142-147. In 1980, Bertelli suggested a comparison with miniatures
from the Ivrea’s manuscripts. Cf. BERTELLI, 1980, p. 217. The Sacramentary
of Warmund d’Ivrea, which Bertelli most likely meant, is today dated between
966-1002. See E. A. GATTI, In a Space Between: Warmund of Ivrea and the
Problem of (Italian) Ottonian Art, in «Peregrination», III (2010/1), pp. 8-48.
81 For the musician, see especially B. BISCHOFF, Manoscritti nonantolani dispersi
dell’epoca carolingia, in «La bibliofilia», LXXXV (1983), pp. 99-124, esp. 118.
82 See C. BERTELLI, Aspetti dell’arte promossa dai Longobardi in Italia nell’VIII secolo,
in Il futuro dei Longobardi. L’Italia e la costruzione dell’Europa di Carlo Magno.
Saggi, eds. C. Bertelli and G. P. Brogiolo, Milano 2000, pp. 189-195, esp.
189. On Anselm’s exile, see G. BRAGA, Testimonianze di vita monastica italiana
fra nord e sud nell’VIII secolo, in Il monachesimo italiano dall’età longobarda all’età
ottoniana (secc. VIII-X), ed. G. Spinelli, Cesena 2006, pp. 509-534, esp. 514.
83 Mazarine 660 is comprised of two parts. One of these belongs to the 15th
century. Logically, it is the second one which we are interested in. According
to Bernhard Bischoff, the paleographical analysis reveals two generations of
scribes and several different hands. One of the hands writes in the manner
known at Nonantola; two or three others are comparable to the script of the
Sessoriano 95, dated probably to the first half of the 9th century. See BISCHOFF,
1983, pp. 108, 118; G. CAVALLO, Libri e cultura nelle due Italie longobarde, in
Il futuro dei Longobardi… Saggi, 2000, pp. 85-103, esp. 89; and C. ALBARELLO,
Cassiodoro, Institutiones, in Il futuro dei Longobardi. L’Italia e la costruzione
dell’Europa di Carlo Magno, exhibition catalogue, eds. C. Bertelli and G. P.
Brogiolo, Milano 2000, pp. 158-159. For the early 10th century dating, see A.
MOLINIER, Catalogue des manuscrits de la Bibliothèque Mazarine, Paris 1885,
pp. 301-302. For doubts about its North Italian origin, see M. PALMA, Alle
origini del «tipo di Nonantola»: nuove testimonianze meridionali, in «Scrittura e
civiltà», VII (1983), pp. 141-149, esp. 147. Cf. G. OROFINO, Da Montecassino
a Nonantola, la tradizione illustrativa delle Institutiones di Cassiodoro, in Il
monachesimo italiano…, 2006, pp. 553-607, esp. 560.
84 See BISCHOFF, 1983, p. 118.
85 See OROFINO, 2006, p. 561.
86 See BISCHOFF, 1983, p. 118.
87 See BERTELLI, 2000, p. 189.
14. Paris, Bibliothèque Nationale, Nouv.Acq.Lat. 1203, Godescalc Evangelistary
(781-783), f. 3a, Enthroned Christ.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 15
Arte Lombarda | MARTIN F. LEŠÁK
16
year 800 – rather than to the second half of the 9th century. Secondly,
we consider the opposite view to linger on the dating of
the murals, once decorating both upper levels of the tower
around 800. Subsequent layers of painted plaster documented
in the southeastern corner of the lower chamber are probably
only several years younger considering the similarities in the
physiognomy of faces, technique and scripts88.
The Intercession and Power of Saints at Torba
We have seen the wall paintings of the upper chamber raise intriguing
questions regarding the iconography and function of
the room, hence, the tower itself. Beside the row of apostles,
saints, martyrs, and nuns, there are three iconographic points
which come into our focus. It begins with the depiction of
Maiestas Domini on the north wall, the votive image on the
south wall, and the enthroned Christ surrounded by angels,
John the Baptist, and (most likely) Mary on the east wall. According
to John Mitchell and Bea Leal, the latter mentioned
group, for which the Greek term Deësis (δέησις) has been frequently
used89, was depicted «in generic reference to the Last
Judgement»90. We will start our interpretation of the visual
scheme by reflecting on this statement.
Starting from the end of the 19th century, most scholars
(Nikodim P. Kondakov, Jean Fournée, or Louis Bréhier) explained
the rather narrowly specified iconographic motif Deësis
as a derivation of the eschatological vision91. They believed this
imagery bears an eschatological connotation; an understanding
which Mitchell and Leal follow in their statements. This traditional
interpretation radically changed thanks to studies of
Thomas von Bogyay and Christopher Walter in the second part
of the sixties92. In 1968, Walter stated, «the advocacy of the Virgin
and the Prodrome at the Last Judgment is simply a development
of an already existing faith in their intercessory power [...].
This faith in their intercessory powers in its turn was a development
of an already existing belief that they were exceptionally
powerful in the heavenly court»93. According to Walter and Bogyay,
Deësis had been added to the Last Judgment theme later
during the 11th century. Only then did it acquire new significance94.
Following their hypotheses, and the suggested dating of
the wall paintings at Torba, it would be difficult to support
Mitchell and Leal’s proclamation. Bearing in mind ideas of Walter
and Bogyay, we would likely adopt the interpretation of John
and Mary as the logical part of heavenly court’s vision, in which
they figure as the witnesses of Christ’s divinity.
However, before reaching such a conclusion, it is necessary
to pursue one more lead in our examination of Deësis. Namely,
it is the funerary chapel dedicated to San Zenone at Santa
Prassede, most likely ordered by Pope Paschal I (817-824) for
his mother Theodora95. The iconography of this space is without
a doubt specific to Rome’s context. It includes an «Empty
Chair», Transfiguration, Anastasis, and most essentially the depiction
of the Prodrome and Mary surrounding God – the light
invading the chapel through the east window96. All these iconographic
elements, as Ivan Foletti and Valentine Giesser recently
pointed out, appear in later representations of the Last Judgment.
Consequently, art historians interpreted San Zenone’s
decorative scheme also as a version of the Last Judgment iconography97.
In their view, John the Baptist and Mary, do not figure
88 See J. HANSELMANN, I frammenti di affreschi di Torba: ricomposizione e interpretazione,
in «Sibrium», XIX (1987-1988), pp. 165-172, esp. 167 and DE
RUBEIS, 2002, pp. 84-85.
89 On the usage of this term and the introduction into this iconography, see
for example Ch. WALTER, Two Notes on the Deēsis, in «Revue des études byzantines»,
XXVI (1968), pp. 311-336 and A. CUTLER, Under the Sign of the Deēsis:
On the Question of Representativeness in Medieval Art and Literature, in Studies
on Art and Archeology in Honor of Ernst Kitzinger on His Seventy-Fifth Birthday,
«Dumbarton Oaks Papers», XLI (1987), pp. 145-154. For using the word in
the murals’ historiography, see BERTELLI, 1988, pp. 26, 29, 34, 42 and BERTELLI,
1980, pp. 210-211.
90 See MITCHELL - LEAL, 2013, p. 342.
91 See N. P. KONDAKOV, Geschichte der Denkmäler des byzantinischen Emails,
Frankfurt am Main 1892, p. 274. J. FOURNÉE, Le jugement dernier. Essai d’exégèse
d’une oeuvre d’art: le vitrail de la Cathédrale de Coutances, Paris 1964, p. 104; L.
BRÉHIER, L’art chrétien, Paris 1928, p. 147. For a summary of interpretations and
ideas concerning the Deësis, see for example T. BOGYAY, Deesis und Eschatologie,
in «Byzantinische Forschungen», II (1967), pp. 59-72, esp. 61-64.
92 See BOGYAY, 1967 and WALTER, 1968. In fact, it was already Aleksandr I.
Kirpičnikov who in 1893 claimed that, originally, there was no eschatological
connotation linked to the image. See A. I. KIRPIČNIKOV, Deisus na Vostoke i na
Zapade i ego literaturnye paralleli (in Russian), in «Žurnal Ministerstva narodnogo
prosvesšěnija», CCXC (1893), pp. 1-26, esp. 7-8. Thomas von Bogyay
remarked that even though the western authors cited Kirpičnikov’s study frequently,
they read it rarely. See BOGYAY, 1967, p. 63.
93 See WALTER, 1968, p. 336. See also E. KANTOROWICZ, Ivories and Litanies,
in «Journal of the Warburg and Courtauld Institutes», V (1942), pp. 56-81,
esp. 71.
94 See WALTER, 1968, p. 335; Ch. WALTER, The Origins of the Iconostasis, «Eastern
Churches Review», III (1971/3), pp. 251-267, esp. 265; and BOGYAY,
1967, pp. 62, 71-72.
95 From the ample bibliography about the San Zenone chapel in Santa
Prassede, see for example U. NILGEN, Die grosse Reliquieninschrift von Santa
Prassede: eine Quellenkritische Untersuchung zur Zeno-Kapelle, in «Römische
Quartalschrift für christliche Altertumskunde und Kirchengeschichte», LVIX
(1974), pp. 7-29; B. BRENK, Zur Bildprogramm der Zenokapelle in Rom, in
«Archivo español de arqueología», XLV-XLVII (1972-1974), pp. 213-222; G.
V. MACKIE, The San Zeno chapel, a prayer for Salvation, «Papers of the British
School at Rome», LVII (1989), pp. 172-199; R. WISSKIRCHEN, Die Mosaiken
der Kirche Santa Prassede, in Rom, Mainz am Rhein 1992; C. J. GOODSON, The
Rome of Pope Paschal I. Papal Power, Urban Renovation, Church Rebuilding and
Relic Translation, 817-824, Cambridge 2010; A. PIROCHTOVÁ, Kaple San
Zenone v bazilice Santa Prassede v Římě, Dissertation thesis, Masaryk University,
Brno 2012; and I. FOLETTI - V. GIESSER, Il IX secolo: da Pasquale I (817-
824) a Stefano V (885-891), in La committenza pontificia dei papi a Roma nel
Medioevo, ed. M. D’Onofrio, Roma 2016, pp. 219-237.
96 See FOLETTI - GIESSER, 2016, pp. 228-230. Cf. V. IVANOVICI, Windows and
Church Space in Early Medieval Byzantium and West, in «Opuscula historiae artium»,
LXII (2013), pp. 38-47, esp. 45.
97 They based their interpretation also on the reading of night vigil – Pannyichis
– which has been sung in the period between an individual’s death and
his or her funeral. FOLETTI - GIESSER, 2016, pp. 228-230.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 16
Prayers and Protection:TheTower atTorba Reconsidered
17
as mere passive witnesses of Christ’s divinity, quite the contrary.
In this funerary chapel, they are actively pleading with God –
the Judge – for the souls of the deceased98. Their carefully laid
hypothesis challenged Walter’s and Bogyay’s argument on several
levels, considering the Last Judgment’s non-existing relation to
Deësis before the 11th century. Furthermore, this idea made the
statement made by Mitchell and Leal relevant once again. According
to them, the Deësis in the tower bears a generic reference
to the Last Judgment. Yet several iconographic elements («Empty
Chair», Transfiguration, Anastasis) which form a Last Judgment
representation at the San Zenone chapel are missing at
Torba. This seems to demonstrate, paradoxically, that John and
Mary on the tower’s walls should not be perceived exclusively in
this Last Judgment – eschatological – context. We believe that
these figures should be understood in Walter’s perspective, as a
logical part of the heavenly court’s vision which is timeless in its
essence, and thus cannot be linked to the end of time only.
This assumption is strengthened by our understanding of
Maiestas Domini on the north wall of the upper chamber. One
of the most frequent and versatile iconographic motifs in Christian
art is based on the description of the Great Vision of the
Enthroned in Revelation (Apoc. 4). Its origin thus lies in a series
of visions of the End times – eschatology. Due to the background,
modern authors link it to a triumphal theophany removed
to the end of time. They even interpret it, similarly to
Deësis’ case, as the Last Judgement’s reduction99. However, as
Yves Christe convincingly asserts in accordance with ideas of
Peter K. Klein, from Tyconius’s ecclesiological reading of John’s
Apocalypse, the significance of the Maiestas Domini in early
middle ages is «present ecclesiological, rather than the future eschatological»100.
It denotes His glory in the already existing, present,
celestial Church-Kingdom more than representing the end
of time, the Last Judgment, or the second coming of Christ.
This ecclesiological approach is essential for shaping the nature
of major Latin commentaries on the Apocalypse from the 4th to
the 12th century. Throughout Tyconius’ exegesis, Eric Thunø
claimed: «John’s vision of the end of time was transported from
the edge of history back into the present to describe the current
Church existing simultaneously on earth and in heaven»101. Subsequently,
early medieval representations based on the Apocalypse
can be understood as visions of the celestial and terrestrial
Church now.
The presented understanding of Deësis and Maiestas Domini
suggests that a direct reference to the Last Judgment or eschatology
in the chamber’s decorative scheme is unlikely. If we follow
the idea that the iconographical themes bearing eschatological
connotations have been connected to the funerary function,
as in the case of the San Zenone chapel in Rome, should
we not reconsider the hypothesis of Mitchell and Leal? They
claimed that the tower was restored «expressly for funerary
use»102. This question is indeed essential. After all, it was the
representations of the southeastern corner of the lower room,
described above, which led scholars to presuppose the originally
intended funerary character of the whole tower103. However,
these depictions were only later additions to the scheme. In
other words, they should not be used to determine the upper
chamber’s raison d’être.
Our answer to the question starts with the image of an offering
to Mary with Child represented on the south wall. This
scene rises from the tradition of the belief in the intercessory
powers of the Virgin104 (or saints in general) and of votive images.
Hans Belting explained that votive images, appearing for
example in St. Demetrius’s church in Salonika, were «meant to
present private individuals or public figures as the saint’s
client»105. By funding of an image which in words of Georges
Didi-Huberman «can offer itself equally in propitiation, in a time
of expectation and unfulfilled desire»106, the private «client», besides
deepening his or her personal relationship with a saint107,
98 The intercession of Mary and John has been, of course, desired in this funerary
context since it may help during the general Last Judgment at the end
of the time and during the particular judgment of an individual upon death.
Furthermore, it also may help to a soul which has been in the interim, i. e. a
place between these two judgments. See for example Bede Venerabilis, The Ecclesiastical
History of the English Nation, transl. J. Stevens John, London - New
York 1939 (I ed. 1910) and H. FOXHALL FORBES, Diuiduntur in Quattuor: The
Interim and Judgement in Anglo-Saxon England, in «The Journal of Theological
Studies», LXI (2010/2), pp. 659-684.
99 See Y. CHRISTE, Jugements derniers, Saint-Léger-Vauban 1999, p. 150 and Y.
CHRISTE, The Apocalypse in the Monumental Art of the EleventhThroughThirteenth
Centuries, in The Apocalypse in the Middle Ages, 1992, pp. 234-258, esp. 236.
100 For this topic, see CHRISTE, 1992, p. 255; KLEIN, 1992; CHRISTE, 1996,
pp. 7-17; or E. THUNØ, The apse mosaic in early medieval Rome, time, network,
and repetition, New York 2015, pp. 74-81. On Tyconius’s commentary, see for
example S. B. STEINHAUSER, The Apocalypse Commentary of Tyconius, A History
of Its Reception and Influence, Frankfurt am Main 1987.
101 See THUNØ, 2015, pp. 74-75; CHRISTE, 1996, pp. 19-51; and CHRISTE,
1992, p. 235.
102 See MITCHELL - LEAL, 2013, p. 344.
103 Here we refer, first, to Saverio Lomartire’s reading of the inscription around
the cross. See LOMARTIRE, 1992, p. 216. It must be pointed out that Rudolph
M. Kloos suggested a different interpretation of this text. Cf. KLOOS, 1980,
pp. 223-224. Second, to the typology of the same cross. See MITCHELL - LEAL,
2013, pp. 334-335. Third, to the resemblance of the recesses themselves with
arcosoliums. See BERTELLI, 1988, p. 45. And fourth, to the technique used for
creating the faces within the recesses which was labeled as «typical funereal».
MITCHELL - LEAL, 2013, p. 336.
104 An offering to Mary was perceived, at least within some groups, as a desirable
act, since Mary’s intercession could bring a spiritual as well as terrestrial
advantages to the supplicants, already in the 4th century. Theologians
opposed this practice at first. It changed after the council of Ephesus in 431
which recognized Mary as the Mother of God (Theotokos). This decision
led to general veneration of the Virgin as well as to the birth of new texts
emphasizing her role as an intercessor with God. See BELTING, 1994, pp. 32-
36; C. OAKES, Ora Pro Nobis, The Virgin as Intercessor in Medieval Art and
Devotion, London 2008, pp. 7-23; and N. TURCHI, Il gesto d’intercessione
nell’arte religiosa bizantina, in «Colloqui del Sodalizio», I (1951), pp. 46-50.
105 See BELTING, 1994, p. 82.
106 See G. DIDI-HUBERMAN, Ex-Voto: Image, Organ, Time, in «L’Esprit Créateur»,
XLVII (2007/3), pp. 7-16, esp. 9.
107 See P. BROWN, The Cult of the Saints, Chicago 1981, pp. 56-58.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 17
Arte Lombarda | MARTIN F. LEŠÁK
18
asked or thanked for advantages of the actual saint’s intercession.
Thereby, this scheme is from the visual point of view, an imagery
of the active intercessory role of a saint, and from the view of its
own existence, an entreaty or an acknowledgement of a private
individual.
It is from this intercessory-votive perspective that we
should perceive the whole decoration of the upper room at
Torba. The murals with its accentuation of apostles, martyrs,
John the Baptist, and Mary, as well as its strikingly high number
of depicted contemporary individuals, should be interpreted
as a votive scene which has expanded on the four walls of
the room, i. e. a votive chapel. People who would climb the
stairs into this cramped space would be surrounded by a representation
of powerful advocates, or rather the Court of Heaven.
This court is outside of the category of time and therefore
we can interpret it, for example, in its eschatological sense, as
a future reality somewhere at the end of time. It can also represent
a paradisiac place of joy in the vision of the interim described
by Bede108; and in the ecclesiological terms of Tyconius’
reading of John’s Apocalypse, as the «now» existing celestial
kingdom109. It is more than clear that the nuns, and sometimes
perhaps also the donors, while seeing this timeless celestial
vision would plead with the saints for diverse types of intercession.
They could have plead for both well-being of the
living and for souls of the deceased, having in mind either Last
Judgement or the place between the particular judgment and
the last one – interim. In this sense, we just cannot rule out the
possibility of the funerary use of this votive chapel suggested
by Mitchell and Leal. Yet, we would like to put emphasis on
one specific plea concerning life (the present terrestrial existence)
rather than death, which will hopefully shed light on
the choice of people responsible for the visual scheme to use
until then only military tower in a new sacral context.
The following hypothesis regarding the donors’ decision will be
anchored on one hand in the already presented reading of the murals,
and, on the other, in the tower’s materiality and historicity.
It has already been stressed that during the turbulent period,
after the fall of the Western Roman Empire, the tower retained
its original defensive function110. This most likely changed during
the 8th century, when the convent was established in the tower’s
vicinity111. While trying to understand this transformation,
Renato Bazzoni stated that «Torba aveva visto sostituire al
fragore delle armi, il sussurrare delle preghiere»112. These words
suggest that the protective function of the tower had lost its importance
at some time during that century – arms were, after all,
replaced and substituted by prayers.
The political situation after the fall of the Lombard empire
in 774, and the suppressing of Rotgaudo’s riot in Friuli in 776,
likely calmed the north Italian area. In 787, as Tommaso Indelli
wrote, that stability had spread to the whole Peninsula113 and
was reinforced by Charlemagne’s coronation as Holy Roman
Emperor in 800. Despite the historical background, it is difficult
to imagine that the castrum’s fortification was perceived by its
inhabitants as unnecessary from the military point of view at the
time of the murals’ creation. Furthermore, it was not only military
powers threatening the people living within the walls. Aerial,
intangible forces linked to the notion of the biblical wilderness
– a place associated with death, chaos, disobedience, temptation
and testing by demons – (what will be later called selva
oscura by Dante) were as real as a potential military or, more
broadly, corporeal threat and perilous to people living in the castrum114.
In this regard, we should recall, for example, that it was
early medieval hermits who were entering in the wilderness to
fight against these powers115 and the sound of bells was believed
to have capacity to disperse demons, as Bishop Gerard of ArrasCambrai
described in his text around 1025116.
If we then return to our intercessory reading of the wallpaintings
and to the interpretation of the second-floor chamber
as a votive chapel (one of the benefactors of which might have
been a high-ranking noblewoman living in the castrum), we
might see the choice to embellish the tower in a new perspective.
We believe this decision, risen from the high intellectual circles
– based on the painted inscriptions and comparison of the murals
with Mazarine 660 – should not be perceived as a «substitution»
of a military function for a religious one, but rather as an
upgrade of its protective purpose through the addition of a new
religious aspect. We would like to suggest that a person (people)
who donated funds to create murals within the tower, might
have done so not only for his or her personal betterment
(whether this would relate to now, then, or even afterwards), but
mainly for the betterment of the safety of all citizens of Castelseprio.
Therefore, during the prayers, nuns, and perhaps also
donors, while viewing the heavenly court, pleaded with saints
especially for the «well-being», and the endurance of both castrum’s
walls and the people living in it. The prayers activated the
power of represented saints to fight, protect, and by divine help
fortify Castelseprio against both intangible and tangible military
threats, thus making the saints truly present.
108 See above note 98.
109 See above note 100.
110 See DE MARCHI, 2013, p. 15 and BAZZONI, 1997, p. 17.
111 See BAZZONI, 1997, p. 17.
112 See BAZZONI, 1987-1988, p. 153.
113 See INDELLI, 2013, pp. 46-60, esp. 58.
114 See for example V. DELLA DORA, Landscape, Nature, and the Sacred in
Byzantium, Cambridge 2016, esp. 118-144; and J. HOWE, Creating Symbolic
Landscapes, Medieval Development of Sacred Space, in Inventing Medieval Landscapes,
Senses of Place in Western Europe, eds. J. Howe and M. Wolfe, Gainesville
2002, pp. 208-223.
115 See DELLA DORA, 2016, esp. 133-137 and HOWE, 2002, esp. 212-214.
116 See J. H. ARNOLD - C. GOODSON, Resounding Community: The History and
Meaning of Medieval Church Bells, in «Viator», XLIII (2012/1), pp. 99-130,
esp. 120-121. See also E. LOON-VAN DE MOOSDIJK, Pilgrim Badges and Bells,
in Art and Symbolism in Medieval Europe, eds. F. Verhaeghe and G. de Boe,
Zellik 1997, pp. 149-154.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 18
Prayers and Protection:TheTower atTorba Reconsidered
19
With regard to the protection against military danger by a
represented saint, we recall the siege of Constantinople by the
Avars in 626. During this time, the icons of the Virgin were
placed above the gates and carried in solemn processions around
the walls117. Considering aerial (demonic) threats, as understood
by Herbert L. Kessler, in the Anagni’s oratory of Saint Thomas
Becket, representations on the jambs of windows dated to the
12th century refer to this spiritual struggle between good and evil,
and as such possess an apotropaic function118. To support our
reading of Torba’s murals, wall-paintings preserved in the «rotunda»
tower of Ansperto in Milano dated to around 1300, must also
be mentioned119. This tower in the garden of the Archaeological
Museum (the ex-Monastero Maggiore, or Monastero di San
Maurizio) is of Roman origin, and it constituted a part of Milano’s
early medieval fortification. According to some, it had been
restored by Archbishop Ansperto (861-881) during the 9th century120.
The chronological gap separating these wall-paintings
and those surviving at Torba is considerable, however, similarities
between the two are clear. Firstly, the historical context of both is
similar. Both are located on the inner walls of reused military
towers near the female Benedictine convents121. Secondly, there
is an iconographic analogy between the two, which allows us to
presuppose that the murals from around 1300 follow an older
iconography (maybe from Ansperto’s time). The paintings represent,
similarly to those at Torba, a procession of saints in the upper
register, praying nuns in the lower one, as well as Mary and
John the Baptist, in this case, surrounding the crucified Christ
(fig. 15). These arguments encourage us to suggest a link between
the two monuments and interpret the murals preserved in the
tower of Ansperto in an equivalent way as those at Torba. This,
consequently, would further our understanding because the latter
cannot be considered as an isolated case.
On the grounds of the protective role of saintly images in the
early middle ages, historical and iconographical analogies between
the «rotunda» tower of Ansperto and the tower at Torba,
and most importantly, on the grounds of the historical circumstances
of the latter and the presented interpretation of its upper
floor visual scheme, we will conclude with the following statements:
The wall paintings of the tower at Torba, while reflecting
a deep belief in the intercessory power of saints, Mary, and John
the Baptist, were created around 800 to enhance the protection
of Castelseprio. It was not the tower which protected the saints,
but the prayers of the nuns, through the saints’ intercession with
God to protect the tower, the whole castrum, and the people living
in it. This protection was sought against a potential corporeal
danger (which includes a military threads), but also against
demons, a spiritual danger spreading around the castrum in the
wilderness that surrounded it, threatening its inhabitants on day
to day basis.
15. The interior of Ansperto tower, around 1300.
117 Traditional scholarship has maintained that icons of the Virgin were carried
in procession during this siege of Constantinople in 626. See BELTING, 1994,
pp. 36-37, 62. For the text of George the Pisidian from 626, in which the author
praises the patriarch who «demanded to speak, hastened to the city wall,
and steadfastly held up to the [Avars] the awe-inspiring image of the unpainted
painting», see BELTING, 1994, pp. 497-498. See also A. CLEMENA, Space, Time,
and Presence in the Icon, Farnham 2010, p. 68. Cf. B. V. PENTCHEVA, Icons and
Power, the Mother of God in Byzantium, University Park (PA) 2006.
118 H. L. KESSLER presented these ideas at the international conference Liminality
and the Medieval Art, which took place in Brno at the Masaryk University
between the 11th and the 12th December 2017. The paper was called at At
Hell’s Upper Edge.
119 On the murals, see for example F. BISOGNI, Gli affreschi della «Torre di Ansperto»
a Milano, in «Arte Cristiana», LXXIV (1986), pp. 3-14; G. B. SANNAZZARO,
Gli affreschi nella torre poligonale: proposta di studio per il riquadro con i Tre santi in
carcere, in Le torri romane del Monastero Maggiore, ed. D. Caporusso, Milano
2017, pp. 59-62; and P. M. DE MARCHI, Il monastero dei Santi Maria, Maurizio e
Sigismondo, detto il Maggiore. Il Medioevo, in Le torri romane..., 2017, pp. 15-30.
120 For different hypotheses concerning Ansperto tower’s origin, see I. ROSSI,
La chiesa di San Maurizio in Milano, Milano 1914, pp. 152-180 and DE
MARCHI, 2017.
121 See BISOGNI, 1986, p. 3.
This article grew out of my master’s thesis (2016). First and foremost, I would like
to thank Ivan Foletti and Marco Rossi for reading various drafts, providing feedback
that greatly improved my arguments. I would also like to express my sincere gratitude
to Elisabetta Scirocco for providing me the access to libraries in Rome.
Referenze fotografiche
1, 4-6, 8, 10, 13, tavv. 1, 3: photo by the author; 2: after DE MARCHI, 2013,
p. 52; 3: after M. LEŠÁK, The Frescoes of the Tower at Torba: The Intercession and
Power of Saints, Masarykova Univerzita, 2016, fig. 95; 7, 9, tav. 2: after
BERTELLI, 1988, pp. 53, 57; 11: © Bibliothèque Mazarine, Paris; 12: after Il
Notiziario del FAI; 14: after LAUDAGE - HAGENEIER - LEIVERKUS, 2006, p. 160;
15: after BISOGNI, 1986, p. 4.
182-183_01Lesak_p5-19_IVb.qxp_Layout 1 25/10/18 09:20 Pagina 19