MASARYKOVA UNIVERZITA Filozofická fakulta Seminář dějin umění

Martin Lešák

The Frescoes of the Tower at Torba: The Intercession and Power of Saints

Magisterská diplomová práce

Vedoucí práce: Ivan Foletti

2016

2

Prohlašuji, že jsem magisterskou diplomovou práci vypracoval samostatně a uvedl jsem všechnu použitou literaturu a prameny.

......

3

4

This thesis would not have been possible without the help, support and guidance of many people. I would like to express my sincere gratitude to Elisabetta Scirocco and Marco

Rossi who opened me the door to libraries in Rome and Milano respectively. I would also like to thank to my family for helping me survive all the stress. To all my friends for their interest, understanding, listening, and questioning my ideas, thank you – especially to the hospital for souls at Via Giorgio Vasari 6, without you all this thesis would not exist. Most of all, I am fully indebted to Ivan Foletti, my advisor, for his wisdom, patience, enthusiasm, and for pushing me farther than I thought I could go.

5

6

Table of Contents

Chapter 1: Introduction ...... 9 Chapter 2: Lost and Found ...... 11 The Reborn Tower at Torba ...... 11 Chapter 3: Status Quaestionis ...... 18 Remarks on the Bibliography and one Hypothesis ...... 28 Chapter 4: Iconographic Description ...... 32 Second Floor ...... 32 First Floor ...... 38 Inscriptions ...... 41 Iconography and its Models ...... 45 Chapter 5: Dating ...... 48 Stylistic Comparisons ...... 49 Chapter 6: Interpretation ...... 54 The “Deësis” and the Tower ...... 54 The Deësis and its Meaning ...... 57 The Last Judgment at Torba? ...... 62 The Votive Chapel...... 66 Reflections and Nuns at Torba ...... 70 Chapter 7: The Chapel in the Tower as a Concept ...... 74 Chapter 8: Conclusion ...... 81 Bibliography ...... 85 Primary Sources ...... 85 Secondary Literature ...... 85 Illustrations ...... 98

7

8

Chapter 1: Introduction

The tower at Torba, located in the Olona Valley, about forty kilometers north of Milano, is without question one of the greatest medieval monuments preserved in the Lombard region.

The interior rooms of this late Roman military tower have been ornamented with the fresco cycle after the Benedictine convent had been established in its vicinity probably during the 8th century. These very frescoes, especially the ones decorating second floor chamber of the tower, are the main topic of this thesis.

Despite the notion of exceptionality which the structure and its inner walls evoke, their importance has not yet been fully recognized by historians and art historians. Considering this fact, we have decided to start the thesis with the presentation of the historical development of its surroundings, which will explain seemingly little attention the tower received and the reason why the first critical studies about the frescoes appeared only in the 70’s of the 20th century. Subsequently, we will discuss the bibliography itself.

Our following questions will be purely art historical. First, we will describe surviving pieces of the frescoes themselves. The stress will also be given to the lost parts of the visual scheme and to their possible reconstruction. Second, we are going to deal with the date and the problems which the phenomenon we call “style” brings to the researchers of medieval cultures. Third, we will be focusing on the second floor chamber and the interpretation of its major iconographic points - we are going to discuss the meaning of Deësis and Maiestas

Domini, concept of the votive image, and also the depiction of the nuns on the west wall of the room. Fourth, following the analyses of these particular parts of the visual scheme, we will build our interpretation of the whole fresco cycle. We will try to explain what was the reason of its existence and how could it be perceived by its audience. Finally fifth, we will be looking for an idea behind the donor’s choice to ornament this tower. The research into these images

9 will therefore lead us to the understanding of the whole monument within its broader context.

We truly believe that an image can lead us on a way of our understanding the history.

In regards to the iconography, we will be working with the classic iconographic studies of Christopher Walter and Yves Christe. Our research into the intended function of the room in the tower, and the way it was perceived by its audience is, however, rooted in the classic studies of Hans Belting and Herbert L. Kessler. Certain parts of the thesis are then based on the more recent art historical methods presented by Aleksej M. Lidov.

10

Chapter 2: Lost and Found

As we will yet fully discover, first studies about the frescoes preserved in the tower appeared only in the second half of the 20th century. The main goal of this initial part of the thesis is to explain, through history of the tower itself, first, why the frescoes have been lost for so long; and second, what train of events led scholars to their discovery. Only then will we discuss the bibliography of the wall paintings, which will hopefully be, thanks to this order, easily intelligible.

The Reborn Tower at Torba

To understand the title of this chapter, referring to the words of Renato Bazzoni who in 1987 celebrated not a survival but a “rinascita”1 of the tower at Torba [fig. 1], the reader needs to be acquainted with the history of this monument, which probably begins as soon as in the 5th century or in the 6th century at the very latest2. Torba is often characterized as a fortified outpost3 of the well-known castrum Castelseprio, to which it used to be physically connected by walls [fig. 2] 4. Therefore, it should not surprise us that also its history is closely linked to this place, originating probably as a Roman military fort, which, according to Pier Giuseppe

Sironi and Gian Piero Bognetti, became a center of a large territorial domain called Seprio5

1 See Bazzoni 1987-1988, p. 154. 2 The most recently Angela Sciillia dated foundations of the tower to the period between the 5th and 6th centuries. Scillia 2013, p. 109. See also Rossi 2011, p. 40 and Colombo 1999, p. 19. Renato Bazzoni places the construction of the tower into a context of the late Roman period, i.e. to the 5th century, when the number of attacks on borders of the empire increased. See Bazzoni 1997, p. 17. For the 5th century dating, see also Chizzoni – Vassalli 1987, p. 96; Panin 1983, p. 93; or Rotondi Secchi Tarugi 1973, p. 88. 3 See Bazzoni 1997, p. 17; Panin 1983, p 93; or Langé 1967, p. 36. 4 These walls broke away from the, probably, contemporary main nucleus of the fortification upon the hill and descended towards the valley. See Scillia 2013, p. 107; Chizzoni – Vassalli 1987, p. 95; or Rotondi Secchi Tarugi 1973. See also Sironi 1973, p. 68 and Villa 1979, p. 304. Villa recalls Sironi’s hypothesis from 1972 of an inclusion of Torba to the fortification of Castelseprio only after the foundation of the convent. 5 The importance of this region increased during first invasions of Germanic tribes in the 3rd and 4th century, when its defense system was strengthened, or rather upgraded. In this period Milano became the capital of the empire (286-402), as well as the center of its defense against the tribes. De Marchi 2013, esp. p. 15. In this context, we should perceive the isolated traces of the fortification in the central zone of the castrum, dated by 11 during the Byzantine and Longobard period respectively6. At this point, we need to remind ourselves the words of Paula Leveto who emphasizes the fact that the first specific reference to Castelseprio as the true center of the region appeared only in the 9th century, thus we cannot be sure of its major role in the previous eras7.

Neither can we be certain of the specific construction reasons of the tower situated in the vicinity of the Olona river, at the foot of the hill where the main nucleus of the castrum unfolds. Bognetti believed that the positioning of the most eastern outpost not only reduced a probability of a successful attack from the valley, but also guaranteed a water supply for the whole complex8. This line of reasoning was not sufficient for Santino Langé who proposed that the motivation of builders was determined rather by the Como-Novaria road unrolling in the valley. He believed that the outpost was built in order to collect tolls and to provide an accommodation to bypassers9. However, also this hypothesis was questioned. Pier Giuseppe

Sironi suggested that the Como-Novaria road passed Castelseprio from the west10; Luisa

Rotondi Secchi Tarugi furthermore added that the route ascended from Olona Valley at

Malnate, a couple of kilometers north of Torba11. Both of the above mentioned scholars thus again emphasized the importance of the water resource (still present) nearby the tower which, in their opinion, the outpost was supposed to serve in protecting12.

scholars to the 4th century. See Scillia 2013, p. 101; Bazzoni 1997, p. 17; or Mirabella Roberti 1973, p. 57. For late Roman fortifications, see Johnson 1983 or Petrikovits 1971. 6 Gian Pier Bognetti believed that Castelseprio originated as a Roman military outpost and only in the Longobard period became the capital of the iudiciaria of Seprio. See Bognetti 1948, pp. 52-56. Whereas Sironi supported the thesis that it became the center of a vast Byzantine military district during the 6th century. See Sironi 1979, p. 16. See also Carver 1987, p. 312, who says, “towers and precinct wall [of the castrum] could be attributed to one of the last urbanizing efforts of imperial Rome, or a precursor of the incastellamento, that ascent of lowland settlements to more easily fortified positions, which is characteristic of early medieval Italy”. 7 For the critical analysis of earlier sources, see Leveto 1985, pp. 2-3. 8 See Bognetti 1958-1959, p. 21. 9 See Langé 1967, p. 36 and p. 39, n. 11. 10 See Sironi 1973, p. 69. 11 See Rotondi Secchi Tarugi 1973, pp. 87-88. 12 See above n. 10 and 11. The combination of these hypotheses can be found in the subsequent bibliography of the Torba Abbey. See for example Bazzoni 1997, p. 17 or Panin 1983, p. 93, n. 3. 12

The first centuries of Torba’s existence are shrouded in mystery, similarly to the case of Castelseprio itself. We can only imagine that during the turbulent period, after the fall of the Western Roman Empire, under the successive rule of the Goths, Byzantines and the

Longobards in this region, the tower retained merely its original defensive function13. It was probably in the course of the 8th century, although the earlier (and even later) dating cannot be excluded, when its primarily military character transformed into the religious one also, and the structure became a part of a Benedictine convent14. Renato Bazzoni suggested that this change was feasible only in the Longobard period during which the military centers turned into places of trade and encounters15. Actually, it seems to also be possible that the nunnery was founded only after the fall of Regnum Langobardorum in 77416.

In fact, the first historical source referring to St. Mary’s monastery at Torba17 dates back to the 22nd of December 1049. 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 St. Maria del Monte in Varese18. Although we have some more recent documents,

13 See De Marchi 2013, p. 15 and Bazzoni 1997, p. 17. 14 See for example Rossi 2011, p. 40; Bazzoni 1987-1988, p. 153 or Chizzoni – Vassalli 1987, p. 95. 15 See Bazzoni 1997, p. 17. 16 In the fall of 772 and the early months of 773, after years of antagonism between the papacy and Desiderius, relations between the two reached a crisis stage. Desiderius attacked Hadrian (772-795) which resulted in Charlemagne’s invasion of Italy in 773, and finally, in the fall of Pavia after several months of siege. On the fall of Regnum Langobardorum and the Lombard king’s politics at the beginning of the 70’s, see for example Indelli 2013, pp. 46-50; Hallenbeck 1982, pp. 137-174; Sefton 1979; and Partner 1972, pp. 29-33. 17 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 incorrectly claimed that the convent was dedicated to St. Blaise. See for example Giulini 1850 (1760), pp. 700, 722 and Bombognini 1856 (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 Rotondi 1922, p. 127, n. 2 and Magistretti – Monneret de Villard 1917. See also Chizzoni – Vassalli 1987, p. 96; Rotondi 1922, p. 127-128; and Langé 1967, p. 36. 18 „secundo campo iacet al. l. ubi dic. Ciscalina: da munte S. Marie de monasterio qui dic. Turba, da alla parte aliquantum in via, et est tab. XXX“. Manaresi 1937, p. 23, item 32. See also Panin 1983, p. 94; Villa 1979, p. 304; Oltrona Visconti 1960, p. 175; and Rotondi 1922, p. 126. For the historical sources regarding Torba, see esp. Villa 1979; Oltrona Visconti 1960; Manaresi 1937; and Rotondi 1922. For the 15th century turn of events, see Panin 1983. 13 starting from 112419, referring usually only indirectly to an abbess or the nunnery itself, these do not reveal much of the convent’s life, let alone of an important decorative campaign or of a substantial historical event20. Nevertheless, the life at Torba must have changed after the famous destruction of Castelseprio in 1287. The castrum was demolished under the command of Archbishop Ottone Visconti command, and only sacred places within the complex were spared of the terror21. That is probably the main reason why also the tower at Torba withstood the tragic fall of Castelseprio. At this point, we should also remind ourselves of the words of

Renato Bazzoni who argues that the tower would probably survive even if it was not protected by its religious armor, since a demolition of a structure so architecturally advanced would be an enormously difficult task22.

The Benedictine nunnery retained some of its possessions even after this event, yet life without the castrum nearby must have been truly complicated23. In fact, with a lease contract signed in 1357, nuns rented all of their estates to Guarnerio and Obizzo Castiglione24.

This is the first sign of economic problems, which probably only deepened during the following years, and which, besides Torba’s isolated and dangerous position within a now depopulated

19 The document from 1124, which records a donation of the land “a meridie Sancte Marie de Turba”, was published in the edition of Luisa Zagni. See Zagni 1992, pp. 77-78, item 48. See also a document from 1198 which was published in the edition of Cesare Manaresi. It reads, “a mane heredum Laurencii Calbi, a meridie abatisse de Turba, a sero S. Marie de Monte”. See Manaresi 1937, p. 264, item 385. 20 Scholars usually refer to a sale contract, regarding directly lands of the convent, from 1204 in which is mentioned “domina aderaxia habatisa eclexie sancta marie de turba”. See Oltrona Visconti 1960, pp. 175-176; Rotondi 1922, p. 126; or Panin 1983, p. 94; for other documents with a reference to Torba from the 13th century (1203, 1233, 1281, etc.), see Oltrona Visconti 1960, pp. 175-176 and Rotondi 1922, p. 127, n. 1. 21 The destruction was ordered after thirty years of quarrel between Sepriese nobility and the Milanese archbishop. See Bazzoni 1997, p. 18; Carver 1987, p. 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 audet vel praesumat in ipso Monte habitare”. See Giulini 1854-1857, pp. 698 or Bognetti 1948, p. 496, n. 977. 22 The tower at Torba was, as argues Bazzoni, more resistant than walls surrounding the whole complex. See Bazzoni 1987-1988, pp. 153-154. 23 See Villa 1979, p. 305. 24 This document was published in Villa 1979, pp. 319-321. See also Villa 1979, p. 308; Oltrona Visconti 1960, p. 176; and Rotondi 1922, p. 128. 14 area, led to the abandonment of the convent in 142625. Under the leadership of Abbess

Beltramina da Velate, the nuns from Torba joined those from the convent of St. Antonino di

Luvinate, which also belonged to the Benedictine order26. It seems that estates of the convent were rented and cultivated by tenant farmers during these years. We cannot be sure though, whether there was also someone looking after the deserted abbey27. 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 Torba28, renovated some of its parts29 only to abandon it again, the last time being in 148130. The second leaving, which truly meant

“death” for the Torba Abbey, was motivated by the same reasons as the first one31. The complex, however, remained in the possession of the Sisters until 1799, when their new convent at Tradate was suppressed in accordance to the policy of Napoleon32. Thanks to

25 Rotondi mentions four documents relating to the decision of nuns dated to 27th, 28th, 29th September, and 1st October. See Rotondi 1922, p. 129, n. 1. In these, we find also a reference to an ongoing local war: “bellorum turbines”. See Villa 1979, p. 308 or Oltrona Visconti 1960, p. 176. 26 See Oltrona Visconti 1960, p. 176 and Rotondi 1922, p. 128. 27 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 is of the opinion that the income from their estates enabled them to live independently even within the different convent. See Villa 1979, p. 308. 28 The return of the nuns was led by the new abbess, Giovannina Sessa. See Panin 1983, p. 95. It also seems possible that to the end of the alliance contributed Francesco Sforza, who informed Pope Callixtus III about a disgraceful behavior of the united groups of nuns, and asked him to restitute the Torba Abbey with a convenient person in its management. A part of the document from 1457, in which we find all of this information, was published in Rotondi 1922, p. 130, n. 19. Other sources regarding these events were published in Panin 1983, p. 97. 29 In a letter addressed to an “illustrissimae reverendissima domina”, who is identified with Bianca Maria Visconti (wife of Francesco Sforza), which was probably written by Abbess Giovannina Sessa after the return of nuns to Torba, was expressed a desire and need to renovate the abbey. Historical as well as visual sources demonstrate that these renovations were indeed carried out. See Panin 1983, pp. 95-96. For the letter, see ibidem, p. 97. 30 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 St. Sepolcro), into which they moved. St. Sepolcro Abbey in Tradate became their new seat. This donation was motivated, again, by the poor overall condition of the convent at Torba. See Villa 1979, p. 309 and Rotondi 1922, pp. 131-132. 31 The reasons are mentioned in a document from 1482. “ex eo quod dictum monasterium vetustate colapsum est et ineptum et in loco solitario et in vale padulosa situm et propterea saluti anime et corporis vestrum omnium redditur contrarium adeo quod pro maiori parte temporis modo una modo alia vestrum plurimum laboratis infirmitatibus”. See Rotondi 1922, p. 132. 32 See Tibiletti 2013, p. 81; Bazzoni 1997, p. 18; Oltrona Visconti 1960, p. 177; and Rotondi 1922, p. 133. 15 several documents, it is possible to deduct that the former abbey was rented to farmers, as soon as it had been abandoned by the nuns33. This rather agricultural character of this place changed only in the 2nd half of the 20th century.

The “rebirth” of the tower, and the whole convent at Torba, was provoked by a discovery of the famous fresco cycle in Santa Maria foris portas at Castelseprio in 1944. Thanks to this finding, for which we must be eternally grateful to Gian Piero Bognetti34, historians and archeologists started to study the complex at Torba. Although their studies were quite superficial they finally also noticed the frescoes preserved under the layers of whitewash in the second floor of the tower. It cannot surprise us that the place was in miserable condition.

In 1967, Santino Langé wrote: “[the complex] presents itself as a farm in a rather poor state at the moment, almost entirely ivy-wounded [fig. 3], with crumbling plasters…”35 The scholars had been aware of the fact that the restoration was necessary [figs. 95-96], yet, it could be initiated only after the farmers left this place (at some point before 1970), and after the property has been purchased by Giulia Maria Mozzoni Crespi in 197636. The new owner, one of the founders of FAI (Fondo Ambente Italiano), officially donated the Torba Abbey to the above mentioned organization in 197737. Consequently, the restoration works38, led by

33 See Bazzoni 1997, p. 18 and Rotondi 1922, p. 134. 34 See Bognetti 1948. 35 “attualmente si presenta infatti come un cascinale piuttosto in cattivo stato, quasi completamente ricopero di edera, con intonaci fatiscenti…“ See Langé 1967, p. 34. 36 See Bazzoni 1997, p. 18. 37 See Magnifico 2013, p. 12. 38 Thanks to the historical development of the tower and its late rediscovery, the frescoes avoided in some cases truly inconsiderate interventions of the later centuries. Actually, there has been only two interventions carried out on the frescoes. The first of them started almost immediately after the monument had been acquired by the FAI, to be more specific, the restorer received the official authorization for carrying out 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 continual degradation caused primarily by an inadequate temperature and humidity. Both of these restoration campaigns have been led by the independent restorer Pinin Brambilla Barcillon. 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 Barcillon, who stated that there are any detailed documentation since, it was not required by the commission 16

Renato Bazzoni (FAI) and Pinin Brambilla Barcillon, started39. At this point, we can conclude this chapter by saying that thanks to these careful interventions the tower rose from the ashes, or rather, it was truly reborn.

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 letting me into their archives and for their great corporation. 39 See Tibiletti 2013, p. 85; Bazzoni 1987-1988, p. 154; and Bertelli 1980, p. 205. 17

Chapter 3: Status Quaestionis

We have already mentioned that historians and art historians had been aware of the existence of the former abbey situated at Torba a long time before they started to write about the frescoes in its tower, as demonstrate references from 60’s of the 20th century40 and even earlier41. However, these references were usually very brief and they only dealt either with convent’s history or with its relation to Castelseprio. We could try to explain this lack of interest in the complex by scholars’ preoccupation with the frescoes in Santa Maria foris portas at Castelseprio discovered in 1944; by its traditional perception as a convent rather than an integral part of the castrum42; but also by the fact that it was almost inaccessible private property in a bad state of preservation. Whatever the reasons for this initial unconcern were, the first mention of the frescoes appeared in 1967 article of Santino Langé Preliminari al complesso monumentale fortificato della Torba43.

In this paper, after he informs the reader about the historical background of the abbey and its church, Langé presents the tower’s decoration. His report is though very limited, since he was able to see only a few second floor fragments of the now revealed frescoes44.

Nevertheless, on the basis of engraved as well as painted inscriptions, the iconography, and the style of visible pieces, he cautiously concludes that the decoration of the upper room was designed for a chapel no sooner than in the 8th and no later than in the 9th century45. He also

40 See E. T. [Eva Tea] 1964, pp. 218-221; Bossi 1961, pp. 192-194; Perogalli – Bascapè 1960, p. 203; or Oltrona Visconti 1960, pp. 175-177. 41 See Bognetti 1958-1959, p. 21; Arslan 1954, p. 413, n. 2; Bognetti 1948, pp. 366 and 396, n. 136; Rotondi 1922, p. 127, n. 2; and Magistretti – Monneret de Villard 1917. 42 See Langé 1973, p. 75. 43 See Langé 1967. Carlo Bertelli ascribed essential role in making the frescoes visible to Adriano Peroni. Cf. Bertelli 1980, p. 205 or Lomartire 1992, p. 215. For the mentioned Peroni’s study, see Peroni 1973a, esp. pp. 385-386. 44 The author mentions the fragment of painted dado on the east wall and the scene of the Mother of God with the Christ child on the south wall. See ibidem, p. 41. 45 See Ibidem, pp. 44, 48. 18 suggests, while describing the first floor arched niches (he does not mention frescoes) and benches, which were then placed around the walls, that this lower room could have been used as a refectory or a “sala di riunioni” very early after the tower’s construction46. We should add that in this study Langé published a rare photo of the scene with [fig. 4], whose face we have lost before the restoration, which, as Langé stressed many times, was necessary to deepen our knowledge of the whole cycle.

However, restoration works started only at the end of 70’s, as we have already noted.

Consequently, it is possible to trace very few references about Torba, let alone about the frescoes, from the period between the publication of Langé’s article and the beginning of this long time awaited operation. Most of them actually appeared in the 32nd volume of Rassegna

Gallaratese di Storia e d’Arte (1973) dedicated to a conference held at Castelseprio in 1972.

Besides studies of Pier Giuseppe Sironi47, Luisa Rotondi Secchi Tarugi48, and Adriano Peroni49, which dealt either with a role of the former abbey in the castrum’s fortification or with a possible importance of its decoration in the debate about Santa Maria foris portas, was also published another paper written by Santino Langé50. The art historian specified the engraved inscriptions on the east wall of the upper room, in which he identified the names of Abbess

Margherita and King Gisberto51 and dated them into the pre-Carolingian age, i.e. into the middle of the 8th century. Then he stated, “Evidently, if the engraving, superimposed on top of the fresco, can be dated into this period, the fresco, which is under it, must originate at

46 See Ibidem, p. 41. 47 See Sironi 1973, esp. p. 68. 48 See Rotondi Secchi Tarugi 1973. Tarugi describes defensive structures which were discovered in the abbey’s vicinity during the archeological excavations in 1968. These findings in her opinion indicates that the castrum of Castelseprio on the hill was connected to Torba already in the 5th century. See ibidem, p. 88. 49 See Peroni 1973, esp. p. 27. 50 See Langé 1973. 51 Cf. Mazza 1978-1979, p. 213, n. 39. 19 least from the same age”52. Six years after his first study about the topic, Langé thus inclined to the earlier dating of the decoration.

As has been already said in the previous chapter, it was only in 1976, when Giulia Maria

Mozzoni Crespi purchased the convent from its owner and donated it to FAI, that the restoration of the complex could finally began. This turn of events led to a greater interest of scholars in the abbey reflected in the growing number of studies mentioning it. The papers of

Maria Teresa Binaghi Olivari53, Renato Bazzoni54, and Sandro Mazza55, presented in 1976 and published two years later, are the tangible evidences of this phenomenon. Yet, because the works on the interior of the tower started only in April 197856, neither of those brought anything new into the debate about the frescoes57. New findings were for the first time presented during the 6th international congress of early medieval studies Longobardi e

Lombardia: aspetti di civiltà longobarda held in Milano in October 1978. Two texts which focused primarily on the wall paintings resulted from this event. They were published in

198058.

Rudolf M. Kloos’s article Die Inschriften im 1. Stock des Turmbaues von Torba is one of them. This study focuses on the painted inscriptions discovered within the recesses in the east and south wall of the first floor59. To be more specific, Kloos analyses the inscription around the cross, or maybe Christogram [fig. 43], and the titulus next to the one of the represented figures (Aliberga) [fig. 51]. To the reading and analyses of both of these we will return later

52 “Evidentemente se il graffito, sovrapposto all’affresco puó avere questa datazione, l’affresco che sta sotto avrá almeno qualche anno di piú in etá.” See Ibidem, p. 77. 53 See Binaghi Olivari 1978-1979. 54 See Bazzoni 1978-1979. 55 See Mazza 1978-1979. 56 See Binaghi Olivari 1978, p. 160. 57 From this period, see also Colombo 1978 and Villa 1979. The latter concentrates on the historical sources related to the convent. 58 The frescoes are also mentioned in the contribution of Pier Giuseppe Sironi. See Sironi 1980, p. 663. 59 See Kloos 1980. 20 on. In this point it needs to be noted that the scholar’s comparisons seems to support the dating of tituli between the 8th and 9th century, and that, according to him, both of the inscriptions originated from the same period, maybe from the same hand60.

Second of the two articles - Relazione preliminare sulle recenti scoperte pittoriche a

Torba - was written by Carlo Bertelli, and it truly laid the foundation of the research into the frescoes61. The art historian, for the very first time, described in detail discovered wall paintings of the first as well as the second floor of the tower. On the basis of his observations, he then arrived to several important conclusions regarding the decoration of the upper chamber. First, its composition, to which dominates enthroned Christ surrounded by two angels on the east wall, was derived from a monumental concept destined to decorate either an apse or a façade62. Second, the obvious overlapping of painted layers resulted from the fresco-secco technique [fig. 5]. Consequently, according to the author, the whole cycle originated from a single decorative campaign63. Third, this campaign antedated the end of the

8th century, as proves Bertelli’s epigraphic analysis of preserved engravings64. It should be also added that, in his opinion, the upper room was embellished only after a first phase of the lower chamber’s decoration had been laid65. Fourth, the fresco cycle was destined for an oratory. It is therefore possible that in the room which it adorned used to be at least one

(maybe even two) moveable wooden altar66. And finally fifth, the frescoes do not have any

60 See ibidem, p. 223. 61 See Bertelli 1980. 62 See ibidem, p. 211. 63 He explains that the first draft was sketched into the wet plaster whereas the rest of the painting was applied to the dry one. See ibidem, pp. 211-213. 64 See ibidem, pp. 215-216. 65 Bertelli correctly observed that there are two layers of painted plaster in the first room. The older one is visible on the east wall (a fragment of painted dado in its lower part; a fragment of a figure above it) and also on the left jamb of the right recess in the same wall. See ibidem, p. 207. Only later Bertelli clearly stated that the second phase of this room’s decoration (Aliberga, cross, etc.) had originated in the same period as the fresco cycle of the upper room. See below, n. 83 or Bertelli 1988, p. 40. 66 Bertelli said, “…da molti indizi appare chiaramente che non siamo qui in una stanza qualunque, ma in un oratorio…” See ibidem, p. 210. 21 stylistic equivalent in a monumental decoration of neither North Italy, which of course included in his opinion Santa Maria foris portas, nor its other parts. In other words the monument seems to be stylistically isolated from almost everything we are familiar with67.

The restorations, excavations, and investigations of the Torba Abbey continued during the 80’s. Most of the new findings, as well as several more or less summarizing articles68, were published in the second part of this decade69. The conference Castelseprio 1287 prima e dopo70, which was held in 1987 to commemorate the 700 years anniversary of Castelseprio’s destruction, played an important role in this sense. Renato Bazzoni, an architect and the co- founder of FAI, described in his contribution architectural transformation of the second floor after the decision had been made to convert it into a sacral space adorned with frescoes71.

Juliette Hanselmann72, while presenting the research of students from the Lausanne

University73 into the fragments of painted plaster found, during the excavations in 1981-1982, in the crypt of the convent’s church, suggested that the first floor of the tower served as a

67 The author asserted that only some generic comparisons with miniatures suggest the North Italian origin of a person responsible for the frescoes’ creation. See ibidem, pp. 216-217 and Bertelli 1988, p. 41. 68 See for example Chizzoni – Vassalli 1987. Worth mentioning is the fact that these authors interlinked the report about Torba’s churches and altars from Liber notitiae sanctorum Mediolani with the second floor of the tower. See ibidem esp. p. 96. For the altars, see also Bertelli 1980, p. 210; and above n. 17. For the mentions about Torba and its frescoes, see also Bazzoni 1986; Leveto 1985 p. 17, n. 36; Sironi 1987; Tamborini 1987; and Mitchell 1985, pp. 132, 158. The last mentioned author pointed out parallels between the decoration of the tower and the convent of San Vincenzo al Volturno. 69 Besides the papers published in 1980, there are two more studies worth mentioning from the first half of the 80’s. One of them deals with historical sources related to the life of the convent in the 15th century, which also means with its architectural transformation during this period, and it is written by Yula Panin. See Panin 1983. The second of those makes only a passing comment about Torba. See Peroni 1984, p. 257. 70 Pier Giuseppe Sironi at this symposium presented a hypothesis that the Benedictine convent at Torba, like the one at Cairate, acted as a xenodochio, i.e. a hospice for pilgrims who crossed the Alps in hope to reach Rome. See Sironi 1987-1988, p. 40. See also Pedroli 1987. 71 See Bazzoni 1987-1988. He stated that the original openings of the east wall were reduced to dim the light entering the room. This decision also provided a workshop responsible for the frescoes with a larger space for rendering the composition on this major wall. See ibidem, p. 156. Cf. Gheroldi 2013, p. 305. The paintings on the jambs of the east windows would suggest either, that the openings were filled up only from the external part of the wall or, that the fillings were created only after the cycle has been finished. 72 See Hanselmann 1987-1988. 73 The research of students from the Lausanne University was conducted under the supervision of Carlo Bertelli in 1986. The article written by Jean-Claude Genoud, in collaboration with the above mentioned art historian, also resulted from these investigations. See Genoud – Bertelli 1987. 22 crypt74. Furthermore, she brought our attention to several technical features of the tower’s decoration75 and to the occurrence of lapis lazuli within the used pigments in the upper room76. This was without any doubt a significant finding since it implied the high status of the

Torba Abbey itself or at least the presence of a rich benefactor of “l’aula di preghiera”77. In addition, Carlo Bertelli in his paper at the symposium suggested that this patron was an unknown noblewoman78.

Carlo Bertelli deepened this hypothesis in the only published monography about the frescoes to this date in 198879. In his opinion, there had been actually two donors, both of them represented on the south wall of the upper chamber, the unknown bishop and the above mentioned noblewoman - comitissa Seprii80. In fact, Bertelli revised, clarified and updated more of his previous propositions in this publication: He newly claimed that nuns had entered the chapel, which had been used, according to the author, until a construction of the convent’s church81, through an entrance in the north wall and not the south one. (An important remark from the liturgical point of view.)82; He clearly described preceding layer of plaster under the

74 The author suggested the possibility that the ground level around the tower could have been higher in the past, so that the first floor would be accessible without external stairs. See ibidem, p. 167 and Bertelli 1988, p. 45. Nevertheless, this hypothesis seems to be unlikely as recently showed Gian Pietro Brogiolo. Cf. Brogiolo 2013, p. 216. 75 Hanselmann pointed out that the face of the nun “Aliberga”, represented in the lower chamber, had been painted on secondary thin layer of plaster applied at the pinpointed location at the end of the decorative process. This praxis, as she said, is, “tipicamente bizantino”. See Hanselmann 1987-1988, p. 167. She also claimed, similarly to Bertelli, that overlaps of paint in the upper floor resulted rather from a change of the original idea during the painting process, than from some restoration. See ibidem, pp. 167-168. 76 See ibidem, p. 168. 77 See ibidem, p. 167. 78 See Bertelli 1987-1988, p. 86. 79 See Bertelli 1988. 80 See ibidem, pp. 28, 47. 81 See ibidem, p. 47. 82 Bertelli 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 were 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 fresco cycle. See ibidem, pp. 12-13, 17. Cf. Bertelli 1980, pp. 208-209. It must be noted that this hypothesis appears already in the article on which Bertelli collaborated with his student Jean-Claude Genoud. See Genoud – Bertelli 1987, p. 70. 23 decoration of both first and second floor (This does not relate to obvious overlapping of painted layers resulting from the fresco-secco technique mentioned above)83; The Author also pointed out the resemblance of the arched alcoves within the walls of the lower room with arcosoliums84. Consequently, he claimed, similarly to Hanselmann, that this chamber had been used as a crypt or rather a mausoleum85; He further asserted that the preserved fragments of the frescoes within the crypt’s recesses (Aliberga, cross etc.) had originated from the same decorative campaign as the fresco cycle of the chapel86; And finally, the author moved the terminus ante quem of the frescoes’ creation to the second decade of the 9th century87. He was however unable to answer, whether we should date them before or after the rule of

Desiderius (756 – 774)88.

Not even this publication stirred up the comparable discussion to the one in the case of Santa Maria foris portas. As the result of that, apart from the suggestion of Giovanni

Valagussa (1997) to date, on the basis of stylistic comparison with a painted saint preserved in Santa Maria presso San Satiro in Milano, the upper room frescoes to the 10th century89, there have not been much information or hypotheses added during the 90’s. It is only worth mentioning that, in 1992, Carlo Bertelli inclined to the dating of the transformation of the military tower to a “luogo di culto”, and thus of the fresco cycle itself, to the period after the

83 The upper room used to be decorated with a simple red band in the lower part of its walls. See ibidem, p. 45. The red framing is also visible under the following layer of plaster in one of the recesses in the lower room. Considering this layer of the first floor Bertelli asserts,”Ció che è ora visibile non è molto, ma presenta notevoli affinitá con la pittura romana del VII-VIII secolo”. In this text the author has mistaken the north wall of the crypt for the east one. “Soltanto sulla parete nord sussistono trace di pittura”. See ibidem, p. 36. In 1980, Bertelli described two layers of painted plaster only in the first floor. See Bertelli 1980, p. 207. 84 See ibidem, p. 45 85 See ibidem, p. 47 86 Bertelli said, “…non vi é dubbio che la decorazione della sala superiore, la croce con la lunga iscrizione nella sala inferiore e, nella stessa, le immagini delle monache in preghiera appartengano tutte ad una stessa epoca“. See ibidem, p. 40 87 To be more specific, he moved the terminus ante quem of the creation of engravings in the upper floor. See ibidem, p. 40. 88 See ibidem, p. 42 89 See Valagussa, 1997, pp. 33-34. 24 fall of the Longobard kingdom, when the region had lost its importance as the border of the realm90. We should also add that, before the end of the millennium were already known ideas summarized for example by Saverio Lomartire91, Renato Bazzoni92, and Silvano Colombo93.

After the turn of the century, on 22nd of September 2001, the conference Giornata di studio Castel Seprio e Vico Seprio took place. Within the edition resulting from this event, we can find, besides papers partly concerning with the tower of Carlo Bertelli94 and Pier Giuseppe

Sironi95, also a text written by Flavia De Rubeis96. In this contribution the author focused on the inscriptions preserved in the lower and the upper floor. From her paleographic analysis she deduced that, though it is impossible to ascribe inscriptions in majuscule (maiuscole longobarde)97 of both levels to one hand - one person, it is possible to assert that they had originated from similar roots98. These roots, she claimed, seems to be closer to the tituli documented in the Longobard fresco cycles of Brescia (San Salvatore) and Cividale (Longobard

Temple), rather than to those of Carolingian area in North Italy99. She dated these inscriptions rendered in fresco, also on the basis of already mentioned engraved tituli in Carolingian minuscule, to the end of the 8th century100.

Second decade of the 21st century has been so far gracious towards the tower studies.

In 2011, Marco Rossi, although acknowledging a completely different context from which had risen the frescoes of the tower and those of Santa Maria foris portas, also pointed out some

90 See Bertelli 1992, p. 4. During the same year, Paula Leveto, within her attempt to reformulate questions which the historiography of North Italian frescoes deals with, mentions our cycle. See Leveto 1992, esp. p. 249. See also Christe 1992a, esp. p. 8. 91 See Lomartire 1992 and Lomartire 1994, p. 49. 92 See Bazzoni 1997, esp. p. 21. 93 See Colombo 1999 [1995], esp. p. 84. 94 See Bertelli 2002. 95 See Sironi, 2002. 96 See De Rubeis 2002, pp. 83-100. 97 See Ibidem, p. 85. 98 See Ibidem, p. 84. 99 See Ibidem, p. 99. 100 See Ibidem, pp. 98-99. 25 of their similarities101. Subsequently, in 2013, a monumental study Castelseprio e Torba: sintesi delle ricerche e aggiornamenti edited by Paola Marina De Marchi was published. We can find many enriching articles concerning the complex within this volume102, yet, from the perspective of the tower’s decoration, the most valuable papers are those written by Vincenzo

Gheroldi103 and John Mitchell, the latter in collaboration with Bea Leal104.

The study of Vincenzo Gheroldi deals with a materiality of the frescoes, and it is more or less technical in its character. On the grounds of the stratigraphic evidence, the geomorphological analysis of the mortars composition, and the recognition of different working techniques with the plaster, the author recognizes, same as Bertelli before him, at least two successive phases of decoration in the first floor of the tower105 and three phases in the second one106. His detailed observations support the Bertelli’s hypothesis of one successive campaign during which the fresco cycle of the upper room was created. Gheroldi explains, in a similar way as Bertelli, that changes and substitutions which are visible on the south107 and west wall108 are in fact, “remakes made during the creation of the work, rather than replacements of a different phase”109. Gheroldi’s analyses on the other hand contradict

101 The author refers to some “accenti iconoci”; the typology of Christ in a clypeus; and the use of velum, huge aureole, as well as classical motives. See Rossi 2011, p. 46. We must add that Rossi published some of his ideas as soon as in 2006. See Rossi 2006, pp. 262-264. 102 See for example Brogiolo 2013; De Marchi 2013; Ibsen 2013; Marinato – Canci 2013; Scillia 2013; and Tibiletti 2013. 103 See Gheroldi 2013. 104 See Mitchell – Leal 2013. 105 The cross on a white background in the left opening of the south wall has no direct stratigraphic relationship with the two previous plasters (A and B). Nevertheless, the mixture and the plaster morphology is closer to the type B. See Gheroldi 2013, p. 296. It should be noted that the author confused the north wall of the lower chamber with the east one in the text. However, the photos of the walls are described correctly. See ibidem, pp.294-295. The same mistake, as has been already said, appeared also in Bertelli’s monography from 1988. See Bertelli 1988, p. 36. 106 During the first phase, the walls were whitewashed and at the bottom of the room was painted a red stripe. This red band was renewed during the second phase. Finally, during the third phase, the whole room was decorated with the now visible fresco cycle. See Gheroldi 2013, pp. 302-303 107 The hand of the donor, who is depicted next to the Madonna was at first higher, also the color of her dress was different. See ibidem, p. 308. 108 The head of the second nun from the left in the row of nuns was originally higher. See ibidem. 109 “…rifacimenti in corso d’opera anziché sostituzioni di altra fase…” See ibidem. 26

Bertelli’s statement, that the paintings in the recesses of the lower chamber (Aliberga, cross, etc.) were created in the same period as the fresco cycle of the upper one110. It is in fact the first phase of the first floor decoration111 that seems to correspond (stylistically as well as technically) to the painted decoration of the upper room112.

In the second of above mentioned studies, John Mitchell adds to the question of the lower chamber’s wall paintings, “…it would appear that all four walls of the room were plastered and painted, in this initial phase, with a complex scheme of figural imagery over a dado of imitation polychrome marble revetment”113. However, the author’s dating is somehow in contrast to what Gheroldi asserts. Mitchell believes that the first floor with all of its decorative phases was completed at the beginning of the 9th century114 and the upper chamber even later in the second quarter of the 9th century115. The main focus of his article is also given to the function of the rooms. The art historian suggests that two upper stories were created for funerary use. The lower room as a place for individual commemoration; the upper chamber as a chapel for memorial masses116. Consequently, he claims, contrary to Bertelli that

“there is no reason to suppose that the chapel served initially as a major liturgical space for the regular use of the community, at a time before a proper church had been constructed”117.

At the end of this part, it should be stated that a research around the area continues even

110 Cf. Bertelli 1988, p. 40. 111 The dado painted in imitation of polychrome opus sectile in the lower section of the east wall; fragmentary remains of probably large narrative composition in the upper part of the same wall; etc. See Gheroldi 2013, pp. 293-295. 112 See ibidem, pp. 293-294. 113 See Mitchell 2013, p. 329. 114 See ibidem, p. 336. 115 See ibidem, p. 344. 116 See ibidem, p. 344. 117 See ibidem. Cf. Bertelli 1988, p. 47. 27 today. The most recent article about the abbey makes, however, only a passing comment about the frescoes118.

Remarks on the Bibliography and one Hypothesis

As we have seen, scholars, led by Carlo Bertelli, asked the basic art-historical questions when dealing with the wall paintings. To answer one the most essential – when had been the decoration created – they found a support in paleography and epigraphy. It is truly only thanks to analyses of frescoed as well as engraved inscriptions on the tower’s inner walls that the majority of them reached very similar conclusion – the date oscillating around 800 – and why can we not find almost any other contra hypothesis or different line of ideas in regards to the dating. In fact, this makes the tower one of the very few monuments, from that period and area, datable on a different basis than those stylistic (in the art-historical sense of the word).

Put differently and perhaps too optimistically, by using words of Paula Leveto, in this case we have a kind of “objective criteria”119 (we will explore later in the text the actual objectiveness of it) on which to base our investigation of the monument. We should truly appreciate this luxury especially when we take into consideration the extensive, and rather never-ending, debate about Santa Maria foris portas120, a monument which, among other reasons, led Hans

118 See Brogiolo – Chavarria Arnau – Giacomello – Marinato 2015, p. 38. 119 Paula Leveto said, “Indisputably, one of the problems in writing the history of medieval painting in Northern Italy is the lack of any objective criteria on which to base a chronology of the monuments”. See Leveto 2000, p. 147. 120 For example Gian Piero Bognetti and Alberto de Capitani d’Arzago suggested the late 7th century dating. They believed that Santa Maria foris portas had been created by Syrian or Palestinian. See Bognetti 1948, pp. 24-28 and De Capitani d’Arzago 1948, pp. 699-700. Kurtz Weitzmann claimed (on the basis of the frescoes’ comparison with the Joshua Roll and the Paris Psalter) that this decoration had originated in the first half of the 10th century. See Weitzmann 1951, p. 95. Ernst Kitzinger supported the date around 800. See Kitzinger 1958, pp. 8-9. Carlo Bertelli then suggested the middle of the 9th century. See Bertelli 1988a, pp. 896-897. Paula Leveto imagined the late 8th or early 9th century as the correct date of its origin. See Leveto 1985, p. 269. John Mitchell’s recent article documents difficulties, which scholars still have when dealing with this monument, in spite of the fact that today we have “the new combined 14C and dendrochronological dating for the roof-beam in the eastern apse in St. Maria, which would seem to locate a terminus post quem and probably a terminus ad quem for the paintings somewhere around 900 or two generations later, in the middle of the 10th century”. See Mitchell 2013, p. 327. 28

Belting in 1967 to write about the problems of art history when dealing with the Italian early medieval artistic production121; a monument of which problematic attribution testifies following proclamation of John Mitchell and Bea Leal: “We are faced with a situation comparable to that in which experts are unable to come to agreement on whether a particular work should be assigned to an artist working in a classicizing narrative mode in 16th century

Europe or to a historicist painter active in the 19th century. The situation is embarrassing”122.

Only the suggestion of Giovanni Valagussa noted above provides a hint of form which would a discussion based only on a formal analysis took in the case of the wall paintings at Torba123.

Besides the question of dating, scholars discussed the function of the rooms which the frescoes adorned. An agreement predominated here also. Actually, it was only Langé who, not being able to see the frescoes of the first floor, suggested a different function of the lower chamber than a crypt - he proposed a refectory or a “sala di riunioni”124. The upper room’s role as a chapel was then generally accepted by all, and only Mitchell specified its funerary use in his recent article125. It must be said though that not one of these art historians truly asked about the reason of existence of the wall paintings themselves. How had they defined the space of the chambers? How had they participated within a liturgy which has been served in the chapel, as proof the commemorative inscriptions under the Christ in front of which stood probably portable altar126, or rather, remembering the substantial term of Aleksej M.

Lidov – “Hierotopy” – which forms the direction of the present-day art historical research, how

121 In his study, Hans Belting dealt with the following question: Was there a survival of domestic artistic traditions in the period between the Longobard invasion in 567 and the coronation of Charles the Great in 800? He also asked, how was the arrival of Longobards, Greeks and Carolingians reflected in the artistic production of the area? See Belting 1967. For the problem which Santa Maria foris portas presents, see for example Leveto 2000; Rossi 2010; and Mitchell – Leal 2013. 122 See Mitchell – Leal 2013, p. 312. 123 See Valagussa 1997, pp. 33-34. 124 See Langé 1967, p. 41. 125 See Mitchell 2013, p. 344. 126 See Bertelli 1980, p. 210. 29 had these paintings drawn their audience into this liturgy127? Considering the rather exceptional integrity of preserved wall paintings of the upper room’s cycle (in comparison with another early medieval monuments), could they even help us to understand liturgical practice in that time and space?

Very important role in the study of the frescoes plays also the technique and stratigraphy. The Bertelli’s idea of the simultaneous creation of frescoes within the recesses of the first floor and the upper room cycle has been only lately contradicted by Gheroldi. From the new hypotheses (Gheroldi, Mitchell) results one logical, yet not articulated, proposition.

If is Gheroldi’s right when suggesting that the first phase of the first floor decoration (The dado painted in imitation of polychrome opus sectile in the lower section of the east wall; fragmentary remains of probably large narrative composition above it; and the fragments of the bordering within the recesses) corresponds stylistically as well as technically to the cycle of the upper room128 - we will refer to it as Layer A from now on, Layer B will be used for frescoes within the first floor’s recesses - and if the Mitchell’s statement that all four walls of the lower chamber have been originally painted129, then we can imagine quite an impressive decorative campaign which determined the appearance of both stories. These decorations could have been related iconographically as well as liturgically to one and other. The question then arises: How could we describe and imagine such interconnections?

Bearing all these information in mind, it becomes very clear that the tower with its wall paintings is truly an exceptional object. All the more striking is thus the little interest which it aroused in the academy world, documented by no mention of this monument during the international conference Studi L’VIII Secolo: Un Secolo Inquieto held at Cividale del Friuli in

127 See Lidov 2004. 128 See ibidem, pp. 293-294. 129 See Mitchell 2013, p. 329. 30

2008130. In 1988, Herbert Kessler stated that “…much of the very best recent scholarly writing has focused on single works or on small groups of related monuments. These monographic studies provide important source material; but the history of medieval art must still be written”131. Now, almost thirty years later, after many brilliant art historical studies and articles contributed to the debate about medieval artistic production, and after many conferences, symposiums, etc. have been held, it still seems that one (and not the only one) truly extraordinary monument has been mostly ignored and overlooked within this dialogue.

Nevertheless, the frescoes of the Torba Tower must be included, without any doubt, into the broader discussion of medievalists and art historians.

130 See Pace 2010. For one of the many examples of this lack of interest, see also Wright 1980. 131 See Kessler 1988, p. 187. 31

Chapter 4: Iconographic Description

Second Floor

We shall start the iconographic analysis of the upper room cycle with the focal east wall [fig.

6]. The enthroned Christ is represented between its two plugged up windows. He has a cruciform nimbus, enormous eyes, beardless cheeks, and a book in his left hand, and he is blessing with his right. Only a fragment remains from the throne, and the same must be said about the lower part of Christ’s body (only his left leg is visible) [figs. 7-8]. However, we can imagine a possible form of the throne on the basis of its comparison with the one from the

Godescalc Evangelistary [fig. 9]132. Christ at Torba is flanked by two angels. The one better preserved, on the Christ’s left side, holds an orb in his left hand [fig. 10]. The visible signs of using a pair of compasses in the process of creating the perfect circle clearly demonstrate, how the sphere was painted. There are preserved only fragments of a wing, head and a right foot from the second angel [fig. 8]. A group of four people is represented on the right side of the wall, beyond the window. There is an agreement between scholars about the identification of the first two figures [fig. 11]. The closest to the main group, as already Carlo

Bertelli stated, is probably John the Baptist, who holds a lamb (or a medallion with the lamb) in his left hand [fig. 12]. The following figure is interpreted, on the grounds of his white hair and beard as St. Peter. Since there are no visible attributes, it is difficult to specify who the remaining figures are. We can only presume they were apostles.

Even more precarious is the situation on the left side of the wall. There are in fact any visible fragments at all. Nevertheless, Bertelli suggested that beyond the left window (to

132 A fragment of the throne is visible on Christ’s left. It is especially this ellipse shape which reminds the cushion on the throne from the Godescalc Evangelistary. There are more parallels between these two representations of enthroned Christ. 32

Christ’s right) a representation of the Virgin Mary (and several more apostles) had been rendered, “following the well-known composition of the Deësis”133. This hypothesis was generally accepted134. The representation of a cross on a velum under the place where the

Virgin has been presumed and the image of the Prodrome on the right side of the composition support, in the opinion of scholars, this statement. Nevertheless, remembering the 7th century text from Constantinople describing a miracle of St. Artemios, important part of which has been published in Hans Belting’s essential study Bild und Kult, a different hypothesis has to be proposed135. 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. Consequently, though the Virgin’s image is the most probable, we cannot exclude the possibility of some other saint, closely linked to the abbey at Torba, being rendered on the left part of the tower’s east wall.

We must add that all figures depicted on the focal east wall are standing above a richly decorated dado painted as a velum [fig. 13]. In addition, we should also emphasize a small painted panel beneath the right window of this wall where two birds, probably peacocks, are delineated, [fig. 14]. As John Mitchell stated, it seems that they are facing a chalice136.

Fragments of another figure standing above a velum, again interpreted as an apostle, are visible on the eastern side of the south wall [fig. 15]137. The composition is then interrupted by a window. Lower parts of three personages, wearing richly decorated long tunics with red clavi, are well preserved on the right side of it [fig. 16]. They are painted over panels of

133 See Bertelli 1988, p. 26. Bertelli claimed, “La presenza di san Giovanni Battista ci autorizza a ricostruire almeno in parte la composizione perduta a sinistra della finestra. Doveva infatti eservi almeno la Vergine, secondo il noto schema della Deesis”. See also Bertelli 1980, p. 210. 134 See for example Mitchell – Leal 2013, p. 337; Rossi 2011, p. 40; or Lomartire 1992, p. 215. 135 Belting 1994 (1990), p. 233. For the text, see ibidem p. 502. 136 See Michell – Leal 2013, p. 339. 137 See Rossi 2011, p. 40; Bertelli 1988, p. 26; and Bertelli 1980, p. 211. 33 marbled dado. Two more figures, one of which is holding a scroll in the right hand, are represented above the (probably later) constructed door [fig. 17]138. It seems that they belong to the above mentioned trio. It is truly difficult to determine who these five individuals are. In fact, in 1980, Bertelli had claimed that all five of them are also apostles139, only to change this hypothesis in 1988, when he identified them as martyrs. The art historian explained this supposition by the fact that the dado imitating a velum changes under these figures to imitate marble, and by their higher placement in the composition. Furthermore, he stated that dresses of five martyrs are more lavishly ornamented than those of the apostles on the east wall140. However, it rather seems that the clothes of martyrs are only better preserved, since also their colors are brighter [figs. 16-18].

The person clad as a bishop – wearing a long tunic, chasuble, and an embroidered stole

[fig. 19] – stands on the right side of the door (from our point of view), outside of the frame and lower than the rest of the represented figures. He holds a book in his right hand. The head of this individual is mostly lost, it seems however that he does not have a nimbus. This, together with his lower position, and the fact that he is being crowned by a martyr, holding his crown in hand, depicted on his left side, led scholars to identify him as a donor or “at least

138 It was especially Carlo Bertelli and Jean-Claude Genoud who dealt with the question of the chamber’s entrance. They suggested that a workshop responsible for the creation of the cycle had had to its disposal “surface pleine” when painting the south wall. In other words, in their opinion, the existing southern door had been walled up before the frescoes were created. This supposition is supported by a surprising reduction of the dado width, and by a use of bricks above today’s door. Furthermore, it would explain why the right opening of the north wall has been extended to the floor, since the entrance would probably be there. See Genoud – Bertelli 1987, p. 70 and Bertelli 1988, pp. 12-13, 17. See also above n. 81. This supposition seems to be not reflected by the more recent studies. John Mitchell and Bea Leal indicated that an only fragmentarily preserved dedicatory inscription in the center of the north wall was composed to be visible after the viewer entered the room through southern entrance. See Mitchell – Leal 2013, p. 340. 139 See Bertelli 1980, pp. 210-211. In fact, 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. Cf. Bertelli 1988, p. 26. In 1988, Carlo Bertelli divided the group of apostles as follows: eight on the east wall; one on the south one; and three on the nord wall. 140 See Rossi 2011, p. 40; Lomartire 1992, p. 215; Bertelli 1988, p. 26. Lomartire as well as Rossi suggest that the figures in white tunics, who stand above the imitation of marble, are “martiri dell’Apocalisse” (Apoc. 7. 9, 13- 14). 34 a ranking ecclesiastic with a particular interest in the community at Torba”141. In addition,

Carlo Bertelli suggested, on the grounds of comparison of this coronation with that of

Angilbert II represented on the famous Golden Altar of Sant'Ambrogio in Milano [fig. 20]142, that the martyr who crowns the bishop (donor) was in fact the saint to which the chapel was dedicated143. A smaller figure, a woman dressed in lay-clothes144, with a square halo145 (this motif could suggest that she was still alive when the decoration was being created146) is depicted further to the right (from our point of view). This woman, who has been identified as a major benefactor of the convent147, offers a candle to the Mother of God holding the child

Christ – Hodigitria [fig. 21]. There are four more saints standing behind the scene of the offering [fig. 15]148.

The west wall is divided into two parts by single axial window [fig. 22]. We will discover eight personages – all of them probably male saints149 – standing above panels of marbled dado on the left side of this opening [figs. 23-25]. Their dresses differs150, and because there are no preserved attributes or even tituli, it seems impossible to specify their identity. We will

141 See for example Mitchell 2013, p. 341; Rossi 2011, pp. 40-44; Bertelli 1988, pp. 28, 47. 142 See Capponi 1996. 143 See Bertelli 1988, p. 28. It is in fact St. Ambrose, saint to which is the altar dedicated, who corronates the archbishop Angilbert II on the altar’s scene. See also Rossi 2011, p. 44. 144 Mitchell describes her clothes as follows: “a yellow over-garment, drawn up over her head, with tight sleeves, worn over a long red tunic”. See Mitchell 2013, p. 341. 145 Bertelli contrarily says that she does not have a nimbus. See Bertelli 1988, p. 28. It is especially the clearly visible line above the woman’s head which suggests presence of the square nimbus. 146 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 for example George Bishop Tatum. This author claimed that “in the ninth century the so-called square nimbus was really a mark of distinction given to persons, real or mythological, living or dead, who by their special qualities or virtues had risen above the common level of mankind but without attaining the position of sainthood”. See Tatum 1944, p. 28. To this topic see also Osbourne 1979, esp. p. 64, who concludes that, although this mark does not give us absolute certainty that the person was alive, this can be deducted from the fact that it was used to reproduce an accurate likeness. See also Fisher 1995. 147 See Mitchell 2013, p. 341; Bertelli 1988, pp. 28 and 47. Rossi stated that she could be the “contemporanea badessa, o della fondatrice del monastero, o di una nobile protetrice della comunitá“. See Rossi 2011, p. 44. 148 See esp. Mitchell 2013, p. 341 and Bertelli 1988, pp. 28-29. 149 All faces preserved in this section have a male physiognomy. 150 For the description of their dresses see Mitchell 2013, p. 342. 35 then notice another line of saints to the right, north of the window [fig. 26]. To be more specific, they are female-saints, as proves the inscription (St. Euphemia), preserved nearby the first person on the left, as well as the physiognomy of surviving faces [fig. 27]. Except for St.

Euphemia, the identification of the female-saints is again impossible.151 Nonetheless, they are depicted higher on the wall than the saints on the left side of the window. The reason for this compositional change is clear. A representation of a nun clad in a dark habit is painted below each of the above mentioned female-saints [fig. 28]. The specific gestures of hands which every single one of these Sisters show are especially noteworthy [fig. 29]. Bertelli observed that “one of their hands is open in a sign of prayer, in second they hold a cross”152. However, the crosses which the author mentioned are not visible anymore. Saverio Lomartire is of the opinion that we may face here the representation of all nuns who lived in the convent at the time the walls were painted153. Consequently, this image has been defined as a kind of

“collective portrait”154. In addition, John Mitchell suggested that “the saints above may represent the name patrons of the nuns below”155. It must be again emphasized, however, that we do not have any information about the Sisters’ life from the period of frescoes’ creation, subsequently, such hypotheses need to be treated with caution.

The north wall of the upper room is without any doubt the one worst preserved [fig.

30]. Some surviving pieces of the fresco are at the western end of it, namely a fragment of lion

(without a nimbus) with a jeweled book in its pawns which supports the lower right part of an aureole [fig. 31]. One of the images embellishing the famous Trier Apocalypse allows us to

151 Some of them are clothed in ornamented courtly garb, others in monastic habits. 152 See Bertelli 1988, p. 29. “Le monache sono otto, tutte in posa frontale, con una mano aperta in segno di preghiera, una crocetta nell’altra”. 153 See Lomartire 1992, p. 215. “…le figure di otto monache, a rappresentare l’intera comunita religiosa di Torba, in atteggiamento orante e un tempo recanti nella mano destra una piccola croce“. Cf. Bertelli 1988, p. 29. Colombo 1999 [1995], pp. 24-25. 154 See Bertelli 1980, p. 213. The author speaks about a “ritratto colletivo di una comunità”. 155 See Mitchell 2013, p. 342. 36 speculate that a depiction of the Lamb of God, or even the triumphant horned lamb of

Revelation, was painted inside of this mandorla [fig. 32]156. However, considering the popularity of a visual scheme called Maiestas Domini157 around 800 [fig. 33], we should support a hypothesis presented already by Carlo Bertelli, who stated that inside of the remaining aureole “must have been an image of the Savior on the throne”158. An example of the similar configuration on Agilbert’s sarcophagus in the north crypt of the abbey church in

Jouarre from the 7th century [fig. 34]159, as well as the one in the apse of the St. Pablo church in Esterri de Cardós from the 12th century, chosen for the analogous position and shape of the lion [fig. 35]160, can help us to imagine how the image at Torba may have looked like. In addition, a fragment of a halo is recognizable under one of the apocalyptic creatures. This suggests the presence of another personages under the above depicted Christ in mandorla.

Lower parts of two figures are then visible between the two windows opposite to the today’s door [fig. 36]. One of them is a high-ranking ecclesiastic to judge from his richly ornamented

156 The Trier Apocalypse, an around 800 created codex in the Stadtbibliothek in Trier (Ms. 31) is one of the most important manuscript of the Apocalypse. It has been suggested that it is in fact a copy of much earlier models, thus it provides us with an important bridge to the Early Christian illustrations. This manuscript contains seventy-four full-page illustrations. On this manuscript, see for example Snyder 1964 or Klein 2006. A mosaic representation of the Lamb on the façade of old St. Peter’s dated to the end of the 7th century must be also mentioned in this context. On the mosaic, see Klein 1992, p. 176. We might even picture, with a certain level of imagination, a presence of Mary in the mandorla, following an iconography of so called Maiesta Mariae. This image, according to Natacha Piano, appeared north of the Alps sometime around 800. However, first surviving examples of this iconography in North Italy are recorded only from the second half of the 10th century, thus it would be rather ill-considered to suggest its presence at Torba. For Maiesta Mariae, see Piano 2003, esp. p. 30. 157 Maiestas Domini belongs between the most frequent and versatile iconographic motifs in Christian art. It is probably based on the description of the Great Vision of the Enthroned in Revelation (New International Version, Apoc. 4. 2-3). Its origin is thus eschatological, even though, as Bianca Kühnel states, it was often used also outside of this apocalyptic context. See Kühnel 2003, pp. 20-21. For this iconography, see for example Christe 1992a; Christe 1996, pp. 123-131; Poilpré 2005 or Cantone 2008. See also Lomartire 1992, p. 215. This last mentioned author 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 says, “the enthroned Christ in majesty”. In this essay I choose to apply this term only when dealing with the north wall image both to avoid a confusion and to follow contemporary art-historical jargon. See Klein 1992, p. 163. 158 See Bertelli 1988, p. 29. 159 For this sarcophagus, see the most recently Delahaye 2007. 160 For the same reason, we could also put to us the apse of Sainte-Eulalie ďEstahón. See Christe 1992a, esp. pp. 6, 10-13; and Christe 1983. 37 stole [fig. 37], second figure is depicted higher161, and wears a long apostolic tunic. John

Mitchell proposed that it is the representation of a material benefactor162, yet, it is equally probable that we are in the presence of another saint or martyr. Between these figures used to be a panel with an inscription – which probably, as Mitchell states, “marked the dedication of the chamber as a sacred space”163. Today, unfortunately, the letters are undecipherable.

We must add two remarks on the iconography. First, another saints, martyrs, as well as ornamental motifs have been rendered on the jambs of the openings. Only rough sketches remains from some of them [figs. 38-39]; the state of preservation of others allows us to recognize some of their facial features [figs. 40-41]. Second, remains of the red decorative band, which had been painted around the whole room before the fresco cycle was created, are clearly visible in the lower part of the walls [figs. 6, 15, 22, and 30].

First Floor

Remains of the first floor frescoes are concentrated on the east wall and within the recesses of the southeastern corner [fig. 42]. Here, we will begin with the description of the second decorative phase of the room – Layer B. To start with, the eastern jamb of the south wall’s alcove used to be decorated with an ochre cross (or Christogram164) on a white background, with the alpha and omega hanging from its two limbs, and a long inscription written around it. Unfortunately, only the lower part of the image survived [fig. 43]. Taking into account the visible letters above the right limb [fig. 44], it seems possible to imagine that the text once surrounded the whole cross. We know similar type of crosses with their massive two-colored

161 According to Bertelli and Mitchell, the first figure is small. Cf. Bertelli 1988, p. 33 and Mitchell 2013, p. 340. 162 See Mitchell – Leal 2013, p. 340. 163 Ibidem. 164 Rudolph M. Kloos suggested that the upper part of the image was comprised of the P-arch shape. See Kloos 1980, p. 220. 38 stems and long, characteristically shaped, bar-terminals, as already Mitchell and Leal stated, from the inner walls of elite tombs preserved in North Italy from the period between the 8th and the 10th centuries [figs. 45-46]165. Nevertheless, tombstones from the same area and time period seem to provide equally suitable comparative material [figs. 47-48]166.

Fragments of decoration are recognizable also on both jambs of the right east wall’s alcove. Two figures are represented on the left reveal [fig. 49]. We can recognize a woman – female ecclesiastic or a nun – depicted in the process of prayer in the lower part of the wall167.

Fragments of another figure, probably of a saint, survived above her. Two praying nuns – or female ecclesiastics – used to be depicted in the lower part of the opposite jamb [fig. 50]. The one on the left, Aliberga, as the titulus next to her reveals, is better preserved [fig. 51]. In fact, only the rest of a hand and upper part of the head give away the presence of the second woman. Two richly dressed saints are represented above them168.

The older, and therefore lower, layer of plaster carrying the painted decoration - Layer

A - is recognizable on both reveals of this east wall’s alcove. However, only the remains of borders defining the space of the recess are visible [figs. 52-54]. This would suggest, as

Mitchell and Leal have already noted, that in the time this room received its first decoration, the walls of the recesses were left bare, with no embellishment169. In 2013, Vincenzo Gheroldi has clearly demonstrated that the plaster of this older layer laps round into the east wall [fig.

165 See Mitchell – Leal, pp. 334-335. In this sense, the painted crosses found during excavations at the basilica of St. Tecla in Milano are closely linked to the one in the lower chamber. For this topic and images of more examples, see esp. Fiorio Tedone 1986. For the crosses from St. Tecla, dated to the period between the 9th and 10th centuries, see ibidem, pp. 413-419. 166 See ibidem, esp. pp. 417-418 and Frati 1978, p. 114. 167 It seems that she holds a stick in her hand. Mitchell and Leal interpreted it as a long candle. See Mitchell – Leal, p. 330. See also Bertelli 1988, p. 39. 168 See Bertelli 1988, p. 39; Mitchell – Leal, p. 331. Bertelli had also suggested that the two upper persons are Empress Helena and her son Constantine. See Bertelli 1987-1988, p. 83. 169 See Mitchell – Leal 2013, p. 329-330. 39

55], where it is possible to observe several fragments of the original composition [fig. 56]170: the painted marble imitating dado in the lower part of the wall – similar to the one which is preserved in the upper room; remains of figures above it – 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 structure171. Mitchell and Leal claimed, that the fragments,

“might be consonant with a representation of the Maries speaking with the Angel at the tomb of the risen Christ, with the guarding soldiers leaning against the Sepulchre in sleep, on the eve of Easter”172. It should be stressed, though, that such hypothesis based on the extremely fragmentarily preserved pieces of the imagery is dubitable at least. Thus, we must satisfy our interest by stating, that the east wall of the lower chamber was once decorated with a large, probably narrative, scene.

Finally, we need to draw our attention to the remains of the painted plaster on the south wall of this lower chamber173. Unfortunately, we cannot determine neither the pictorial motif which was once painted there, nor whether it belongs to the first (Layer A), second

(Layer B), or even later decorative phase. Nevertheless, its own existence is rather in accordance with the statement, that all walls of the lower room were originally decorated with an elaborate scheme (Layer A), than it is in contradiction to this hypothesis.

170 See Gheroldi 2013, p. 293-295. 171 See Mitchell – Leal 2013, p. 329. 172 Mitchell – Leal 2013, p. 329. 173 See Gheroldi 2013, p. 299. We should note that in the description of the image of this fragment is a mistake, since it is described as being on the west wall. Yet, in the text below is the wall defined correctly as the south one. See ibidem, p. 297. 40

Inscriptions

The inscriptions preserved in the first as well as in the second floor of the tower formed an important visual part of the decoration. The second floor tituli have been carried out either in fresco or in engraving. Fragments of the first mentioned type, which formed original part of the cycle, is visible on the south wall, under the scene of Mary’s intercession [fig. 57]. Here we can read following letters: …OE(?)..INE PRECENTEM174. This white inscription on the blue background175, which separated the dado from the figures standing above it, continued around the bigger part of the room. Taking into consideration the chamber’s dimension and the location of the letters, it is obvious that the text was readable for all participants of a liturgy. Sadly, except for the above mentioned words, only a few decipherable letters survived on the east wall [fig. 14]. Consequently, it is impossible to deduce what was the meaning of the whole text.

Frescoed tituli identified also some of the represented saints, perhaps initially even all of them. Today, however, only one inscription may be helpful in our effort to determine who they were. This surviving titulus is visible on the right side of the central opening of the west wall, and it seems to nominate St. Euphemia: SCA EVFEMIA [figs. 58-59]176. In addition, remains of the inscription, also rendered in fresco, are preserved on the north wall. To be more specific, on the already mentioned panel between the figures of ecclesiastic and saint

[fig. 36].

The second type of tituli – the engraved ones – are recognizable only on the east wall, under the depiction of the enthroned Christ [fig. 60]. They have been characterized by Carlo

174 Cf. Rossi 2011, p. 44. The author’s reading is starting with “…OP”. See also Bertelli 1988, p. 35. 175 Cf. Bertelli 1988, p. 17. 176 See Bertelli 1988, p. 29; De Rubeis 2002, p. 83; Rossi 2011, p. 44; and Mitchell – Leal 2013, p. 342. 41

Bertelli as “graffiti commemorative”177, and they are rendered both in majuscule and minuscule. Today we can read only a part of the original four engraved inscriptions, since a segment of the plaster carrying them has been lost. Fortunately though, thanks to the transcription provided by Carlo Bertelli in 1980, we can partly reconstruct their original form178:

† Die.v.m. ivl. [copleta e par] sce margarite …D.M. IVLII. OB. GISBT (Gisbertus). [PBR. IND. XIV. III. KL. N. B … VLVI] PRI[D. KL. FEB OB. AMIRADA ABATI(ssa)] CONSECT(ratio)…

As we have already mentioned, it is especially the script used for the name “Margarite” that is essential for the dating of the whole fresco cycle [fig. 61].

Two intriguing fresco inscriptions survived in the lower chamber of the tower. They both were painted on the later layer of plaster – Layer B, and they both were written in majuscule. The first of them is the titiulus which identifies one of the portrayed nun in the right east alcove [fig. 51]. Scholars have discovered a true palimpsest in this white inscription laid on the ochre background [figs. 62-64]. In other words, they revealed two layers of paint, or rather denomination of the nun. The first older layer has been read mostly as ALIBERGA179 but ALIPERGA appeared also180; the second layer had been initially interpreted by Carlo

Bertelli as CASTANA181, in 1988, he asserted, however, that we should read it as follows: CASTA

ABA(TISSA)182. This reading predominates in the succeeding studies183, even though, as Bertelli

177 See Bertelli 2002, p. 3. See also Bertelli 1988, p. 20. 178 See Bertelli 1980, p. 215. Already in that time the plaster with a big part of the letters had been gone. Carlo Bertelli in his article stated that he had worked with an older photography, which I have not found in the archives of Soprintendenza in Milano. See also Leveto 1985, pp. 17-18, n. 36. 179 See Bertelli 1980, p. 214; Kloos 1980, p. 219; Bertelli 1988, pp. 40-41; Rossi 2011, p. 46; or Lomartire 1992, p. 216. 180 See De Rubeis 2002, pp. 84 and 90. 181 See Bertelli 1980, p. 214; Kloos 1980, p. 219; Chizzoni – Vassalli 1987, p. 96. 182 See Bertelli 1988, p. 40. 183 See Lomartire 1992, p. 216; Rossi 2011, p. 46; 42 emphasized, it is rather difficult to correctly determine what has been written here184. In any case, since it seems that the first layer was covered by the second in a very short period of time185, we face here the phenomenon of “stealing” visual identity by quite simple change in the lettering. Consequently, we can deduce that the representation of the nun was not supposed to depict any specific individual. In other words, it is not a portrait as we are imagining it today186.

The second inscription on the walls of the lower chamber is the most voluminous of them all. Its ochre letters are spread on the white background around the cross in the south wall’s recess [fig. 43]. The Greek redemptive letters Alpha (Α) and Omega (ω) hang on three chains from both limbs of the cross. Four vertically arranged line of three letters once surrounded these chains. Today, some of the letters on the left of the main bar are lost, which makes their reading more difficult than those on the right187:

[H?] [Λ] D V

E E E E

K or X? X H RA

The reading of this text, comprising both Latin and Greek letters, as Hec lex dei vera seems to be the most probable as showed especially Rudolph M. Kloos, who drew our attention to two parallels. First, to manuscript Annales Guelferbytani - Wolfenbüttel 67.5. Aug. 8°, which originated in the 9th century [fig. 65]188. Second, to the baptismal font in SS. Maria e Donato

184 See Bertelli 1988, p. 39. This is demonstrated for example by the diverse interpretation of Pier Giussepe Sironi in 1987. He sees in the second layer of the inscription the name of CASATEIA. See Sironi 1987, p. 145. 185 See Bertelli 1988, p. 39. 186 See also Colombo 1999, p. 22. 187 Kloos as well as Bertelli interpreted the first letter of the first line as “H”, third sign of the same line as “K”, and the first letter of the second line as “Λ”. See Kloos 1980, p. 221; Bertelli 1980, p. 214; and Bertelli 1988, p. 39. Mitchell, in 2013, questioned the reading of these letters and suggested subsitution of “K” for “X”. See Mitchell 2013, p. 334. 188 See Kloos 1980, pp. 221-222. The author reads this inscription as HEC LEX DEI. 43 at Murano, dated by the author to the 8th century [fig. 66]189. The latter even incorporates ligature of the letters R and A, same as in the case of Torba’s inscription.

A long titulus is then located on both sides of the cross under these four vertical lines and the letters Alpha and Omega [figs. 67-68]. The reading of this text is not an easy task as demonstrate several variations of its transcription [figs. 69-72]. The major role in its interpretation played the word ALEXANDRIA. At first, Kloos believed that it is a reference to the Egyptian city from which a relic of someone unknown – probably a saint - had been taken to Torba190. Bertelli in the same edition added that the person transferred from Alexandria to

Torba could have been in fact alive. In other words, the mention of the city was, in his opinion, only an important piece of information regarding the place of deceased’s origin191. In 1992, however, Saverio Lomartire asserted that the text mentions Abbess Alessandra rather than the Egyptian city192. It thus became this abbess, and not someone (or rather parts of someone) from Alexandria, for whom the “eternal rest [in the bosom of] Abraham, Isaac, and Jacob” in the inscription is asked193. Lomartire’s hypothesis was accepted by Bertelli194 as well as

Mitchell and Leal195, and it supplied an argument, together with the cross’s typology196 and

189 See Kloos 1977, p. 579 and idem 1980, pp. 221-222. The author reads the titulus as follows: HEC LEX DEI VERA. Cf. Rugo 1975. p. 23, n. 16. Rugo dates this inscription back to the 10th century. His reading is also different from that of Kloos: LEX DEI LEX TEC. 190 See Kloos 1980, pp. 223-224. Rudolph M. Kloos even suggested that it may had been the relics of St. Marco which had reached the Torba Abbey. The author reminded in this context the shipment of St. Marco’s relics to Reichenau in 830. He based this circumspect hypothesis also on the fact that he read in the second line of the text name MARCE. See also Bertelli 1980 pp. 214-215. 191 See Bertelli 1980, pp. 214-215. The author translates the inscription as follows: Nel nome del Signore Salvatore…./ Qui giaccio in questo luogo da Alessandria in un tumulo…/ Signore, concedimi la pace eterna nel grembo di Abramo, Isacco e Giacobbe./ Non osare (di violare qusta tomba?). See also Bertelli 1988, p. 47. Here the author stated that the person coming from Alexandria had been a man. 192 See Lomartire 1992, p. 216. This is the author’s reading of the relevant part: BAD(ISS)A FAMOLA TUA ALEXAN/DRIA. 193 See also Mitchell – Leal, pp. 334. The authors translated the titulus as follows: …that the soul […] of your blessed servant Alexandra may lie in peace in this tomb. Lord [grant] her eternal rest [in the bosom of] Abraham, Isaac and Jacob […]. 194 See Bertelli 2002, pp. 2-3. 195 See Mitchell – Leal, pp. 334. 196 See ibidem pp. 334-335. 44 the resemblance of the recesses with arcosoliums197, for the funerary character of the lower chamber. The questions are whether the funerary function was intended also in a time when the Layer A, contemporaneous with the upper room cycle, has been created; and whether does the later attested funerary use of the lower room automatically leads to the idea of funerary function of the other.

Iconography and its Models

The upper room’s Iconographic scheme has been described by Carlo Bertelli “as an adaptation and a decomposition of a monumental concept, probably destined to decorate either an apse or a façade”198. Of course, we know processions of saints from the naves of the monumental decorations at Ravenna (Sant'Apollinare Nuovo) or Rome (Santa Maria Antiqua). On the other hand, considering that the oldest preserved representations of the enthroned Christ accompanied by Mary, John the Baptist and apostles in the apse are those in Cappadocia from the 10th and 11th centuries199; and that there are not any strong arguments for this scheme to be present on the façade of the cathedral of Turin, which Bertelli seems to meant (the only argument for this statement remains Ernst Kantorowicz’s remark that the cathedral was dedicated by the Lombard King, Agilulf to John the Baptist, the Virgin, and Christ in 602200), it might be useful to search for different models in terms of medium than those monumental.

In fact, there are at least three types of portable objects which we should recall.

Firstly, we could think of panel paintings. In this sense, we should remember the image described by St. Sophronius (Patriarch of Jerusalem 634-638) in his account of the Miracles of

197 See Bertelli 1988, p. 45. 198 “Curiosamente, gli affreschi di questa sala si presentano dunque come la riduzione e la scomposizione d’un concetto monumentale, probabilmente destinato a un’abside o a una facciata”. See Bertelli 1980, p. 211. 199 For the Cappadocian churches and their frescoes, see Thierry 1974. On medieval apse decorations, see Thunø 2015 and Belting-Ihm 1960. 200 See Kantorowicz 1942, p. 70, n. 4 and Bertelli 1980, p. 211. 45

Saints Cyrus and John. In one of his stories, the saint writes about a sick man who dreams about a wonderful church, where he sees a picture showing Christ surrounded by Mary, the

Prodrome, and some of the apostles and martyrs201. Considering then the movability of panel paintings, documented by actions of Benedict the Bishop who brought several panels from

Rome to Wearmouth in the second half of the 7th century202, we can readily imagine the possible importance of this medium for our decorative scheme.

Secondly, we should remember ivory objects. In fact, the ivory triptych from the Museo

Sacro della Biblioteca Apostolica in Vatican, dated between the 10th and 11th centuries, bears the most similar preserved iconographic scheme to the tower’s second floor imagery [fig.

73]203. This triptych, which Ernst Kantorowicz linked to litanies204, shows enthroned Christ flanked by two angels, Mary, the Prodrome, and four saints in the upper register and another saintly figures in the lower one. Of course, the ivory is a precious material and objects made from it appeared mostly in the high social context. Yet, in regards to a noblewoman (comitissa

Seprii) who, according to Carlo Bertelli, was one of the benefactors of the Torba Abbey205, it is feasible to picture that an ivory object played a part in the process of creating the fresco decorative scheme.

Finally, thirdly, we must recall manuscripts. It would be difficult to claim how and which texts circulated in the North Italian area about 800, yet, in considering the iconographic and

201 This text presents several problems. As Ernst Kitzinger critically noted, the part of the narrative may be interpolated. Unfortunately, it is precisely the part which includes the description of an image. However, since the story was known in its present form already to John of Damascus, it cannot be later than the early years of Iconoclasm. See Kitzinger 1954, p. 106, n. 96. Another problem lies in the question of medium itself. The text does not specifically talks about the panel painting, however it is only Bogyay who suggests that it was a fresco. See Kantorowicz 1942, p. 70, n. 4; Bogyay 1967, p. 66; Walter 1968, p. 329 202 On the paintings brought by Benedict the Bishop to Wearmouth, see Meyvaert 1979, esp. p. 74, n. 3. See also Chapter 6 below. 203 On the triptych and its iconography which has come to be known by the term “Great Deësis”, see Kantorowicz 1942, pp. 70-71; Bogyay 1967, pp. 69-70; Kalavrezou 1997, pp. 131-132;.and Gilsdorf 2012, p. 134. 204 See Kantorowicz 1942, pp. 70-71. 205 See Bertelli 1987-1988, p. 86 and idem 1988, p. 47. 46 typological resemblance of the enthroned Christ’s representation in the Godescalc

Evangelistary and that at Torba [figs. 8-9], it is easy to picture that a larger composition from a manuscript served as a model for the second floor’s frescoes. A significant role which scriptoriums played from the end of the 8th century would support this claim206.

In regards to the fact that preserved medieval objects form only a minute fragment of the originally much richer visual culture, it is impossible to say what sort of image and medium stood as a model for the frescoes. Yes, a monumental decoration might have played a major role in this sense, however, by presenting the above mentioned examples, we want to remind the reader that a portable object could have also had a certain significance.

206 See for example Schutz 2004, pp. 135-171 and Mütherich 2004. 47

Chapter 5: Dating

We have already mentioned several important facts about and hypothesis concerning the dating of the fresco cycle at Torba. We have stressed the important role paleography and epigraphy played in this question207. The reader thus already knows that the reasonably certain terminus ante quem for the paintings is provided by the “Margarite” graffito in minuscule engraved into the surface of the painted plaster beneath the enthroned Christ’s representation [fig. 61]. Carlo Bertelli stressed that though writing on a piece of papyrus differs greatly from engraving into a wall, the characteristic form of the letter “r”208 is comparable to the late 8th century North Italian writings, such as Ms. Phillips 1885 from the

Staatsbibliothek in Berlin209. The letters “g’ and “a” then, in his opinion, represent the

Carolingian type, “despite the archaic appearance of the whole inscription”210. These forms, as the author claimed in 1988, would not appear in the second decade of the 9th century211.

Furthermore, the reader is also acquainted with some conclusions of Flavia de Rubeis, who, though extending the terminus ante quem towards the middle of the 9th century212, suggested the dating towards the end of the 8th century, since the frescoed inscriptions in majuscule of both levels correspond to the tituli documented in the Longobard fresco cycles of Brescia (San

Salvatore) and Cividale (Longobard Temple). When comparing these preserved North Italian

207 On paleographic analyses of Torba’s inscriptions, see Bertelli 1980, esp. pp. 215-216; Kloos 1980, and De Rubeis 2002. 208 See Bertelli 1980, p. 215. 209 See Degering 1929, p. 25 and fig. 42. Degering dates this manuscript from Verona between the 8th and 9th centuries. 210 See Bertelli 1980 p. 216. The author states: “Altri elementi significativi sono poi le forme della g e della a, di tipo carolingio nonostante la parvenza più arcaica dell’insieme dell’iscrizione“. On the Carolingian minuscule and its ample bibliography, see Boyle 1984, pp. 140-170. 211 See Bertelli 1988, p. 40. 212 Like Bertelli, De Rubeis points out the form of letters “a” (“con l’asta fortemente inclinata”) and “g” (“con l’occhiello superiore ancora aperto”). Her more careful approach towards this terminus ante quem is probably caused by De Rubeis’ belief that “la scrittura è un fenomeno che può recare attardamenti nelle variazioni delle morfologie delle lettere, fenomeni lenti a sedimentarsi questi mutamenti…” See De Rubeis, pp. 98-99. 48 inscriptions, De Rubeis pointed out the similar morphology of the letters and the vertically developed script system213. To sum up, hers and Bertelli’s comparisons are based on the

“style” of the script214. The question which arises is whether the art historical concept of style and formal analysis support the above mentioned paleographical and epigraphical studies.

Stylistic Comparisons

It was again Carlo Bertelli who as soon as 1980 claimed that the frescoes are not stylistically comparable to any monumental decoration from about 800 preserved either in North Italy or in its other parts. Comparisons with surviving fresco schemes cannot disprove Bertelli’s statement. In this sense, we must draw attention first to the frescoes of the Tempietto

Longobardo at Cividale [fig. 74] and their Hellenistic green-tone character215; secondly, to the fragmentarily preserved painted decoration of San Salvatore in Brescia [fig. 75]216; thirdly, to the abbey of San Vincenzo al Volturno in Benevento [fig. 76]217, close to the border between the Frankish realm (after 774) and the Duchy of Benevento, an area which even after

213 The author describes the letters as follows: Si osservino in particolare le apicature a forcella sulla lettera E,…lo sviluppo verticale della N, la O a mandorla, gli occhielli della R e della P alti, stretti; le apicature sulla T e il tratto della lettera R che tende ad adagiarsi sul rigo di base. See De Rubeis 2001, p. 84-85. 214 The study of Rudolph M. Kloos is in this sense exception, since the author also mentions some typological, or “iconographic”, parallels which would lead to the dating between the 8th and 9th centuries. See Kloos 1980. 215 The Tempietto Longobardo, the oratory of Santa Maria in Valle, was probably commissioned by either Aistulf and Giseltrude (749-756) or by Desiderius and Ansa (756-774). The interior of this small oratory was embellished by using various artistic techniques and materials. Traditionally, the execution of the visual scheme was being related with the presence of artist trained in the Byzantine tradition. See Kiilerich 2010, p. 93. The 8th century frescoes are preserved in the north and west lunettes and their remains are visible also on other walls. See Torp 2006, pp. 16-20. 216 Remains of the original decoration are preserved on the walls of the nave and aisles of the basilica - Christological and hagiographical scenes. The new researches show that the present church was constructed as well as fully decorated in one campaign probably during the 760s. A 9th century date, however, has also been suggested. The basilica dedicated to San Salvatore constituted a part of a female monastery founded at the site by Desiderius and his wife Ansa (756-774) in 753. See for example Stradiotti 2001, pp. 84-103 and Mitchell 2014, p. 169. On the 9th century dating, see Lomartire 1998, pp. 46-47. 217 The frescoes preserved in the monastery of San Vincenzo al Volturno are concentrated within the crypt of one of its churches, although fragments of saints’ heads were found also in other parts of the complex. Thanks to the depiction of Abbot Epiphanius (824-842) who has been portrayed with the square halo, kneeling at the foot of crucified Christ, the frescoes are quite firmly dated into the period of his office, i.e. 824-842. See Mitchell 1985; Hodges 1997, pp. 118-131; and Mitchell 2013 (1993). 49

Charlemagne’s invasion into Italy kept its autonomy218; fourthly, to the main church of the monastery of St. John in Müstair, to the far north of Bolzano, in what is now Switzerland [fig.

77]219; finally fifthly, to the famous Santa Maria Antiqua, the church of the Greeks on the

Forum Romanum, which flourished in the 7th and 8th centuries. Specifically, to the Chapel of

Theodotus from about 750 which belongs among the most important results of these fruitful years [fig. 78]220. The wall paintings from these monuments demonstrate, better than any adjective ever could, the variety of stylistic modes existing from around 750 to 830’s.

Considering again the walls of Santa Maria Antiqua on which are recognizable several stylistic manners originating from the 8th century221, this variability across a larger area, and our subsequent incapacity to relate the wall paintings’ style to any other fresco monumental decoration222, cannot surprise us. However, the small amount of surviving comparative material evokes again the notion of exceptionality of the tower and its frescoes.

218 On the history of this campaign and its results, see for example Indelli 2013, pp. 46-60. 219 The monastery has been founded about 775, however, the paintings in the main church do not date to this initial phase. There was at least one decade delay, as indicates a layer of dirt between the plaster and the intonaco. See Ataoguz 2013, pp. 92-94. Alfred Wyss suggests a date between 800 and 830. See Wyss 2002, pp. 51 and 61, n. 1; Hans Rutishauer supports an earlier dating of the frescoes, he dates them between 785 and 810. See Rutishauser pp. 57-58. The representation of the Last Judgment on the west wall is divided into three registers. 220 The earliest Christian fresco decoration suggests that the central chamber of the structure that would become Santa Maria Antiqua have been transformed to a small chapel as soon as in the first half of the 6th century. The essential fact for our study is however that during the 8th century, Popes John VII (705-707), Paul I (757-767), Hadrian I (772-795), and Leo III (795-816), and a number of lay aristocrats, sponsored some of the decorative campaigns. See Lucey 2007, pp. 139-140. The Chapel of Theodotus, from around 750, belongs to one of those projects. Belting 1994 (1990), pp. 115-126. On the famous church, see for example Osborne – Brandt – Morganti (edd.) 2004 and Andaloro – Bordi – Morganti –Ballardini (edd.) 2016. On the chapel and its wall paintings which, according to Natalia Teteriatnikov, “presents a family offering for the deceased mother and daughter”, see Teteriatnikov 1993, [p. 37]. 221 See above, n. 220. 222 The same must be said about mosaic decorative schemes, created in Rome during the pontificate of Pope Paschal I (817-824). See Goodson 2010. 50

Even though the comparisons of surviving monumental decorations do not allow us to date the frescoes as precisely as the paleographical and epigraphical analyses probably can, comparisons with manuscripts might be still beneficial for our research. In 1988, Carlo Bertelli drew our attention towards two manuscripts223 – the codex of Bishop Eginone (769-799), written in Verona at the end of the 8th century [fig. 79]224, and the manuscript CLXV of the

Chapter Library (Biblioteca Capitolare) in Vercelli, dated to the beginning of the 9th century

[fig. 80]225. Bertelli emphasized that manuscripts’ affinity with the frescoes is “very generic and that it only confirms the North Italian origin of the artist”226. At this point, we must remember one manuscript which might provide more than just this “generic” affinity. Namely, it is the codex of Institutiones of Cassiodorus, (Paris, Bibliothèque Mazarine, 660) [figs. 81-82,

27]. In 2000, Carlo Bertelli suggested that it is the copy of a 6th century manuscript made at the order of Anselm (723-803; Duke of Friuli, then Abbot of Nonantola Abbey), sometime after he had returned to Nonantola from his exile at Monte Cassino, i.e. no later than in the 80’s of the 8th century227.

Following Bertelli’s hypothesis about the Mazarine 660, we might think that the paleographical analyses of Flavia de Rubeis, Rudolf M. Kloos, and Carlo Bertelli are also supported by images. We would thus arrive to the conclusion that the fresco cycle at the second floor of the tower (Layer A) could be truly dated towards the end of the 8th century. In

223 See Bertelli 1988, p. 61. In 1980, the author mentioned also miniatures from the Ivrea’s manuscripts. Cf. Bertelli 1980, p. 217. The Sacramentary of Warmund d’Ivrea, which Bertelli seems to meant, when he was writing about the miniatures, is today dated between 966-1002. See Gatti 2010. 224 The manuscript, also called the Codice di Egino, is a collection of patristic homilies. It is embellished with four illuminations, each depicting one of the fathers of the Church. See Mattke 2000, p. 215 and Castelfranchi Vegas 1993, p.23. 225 It is a collection of canon law texts followed by a series of decretals and one Augustine’s letter, which is illuminated with drawings depicting the history of councils. See Walter 1968a and Hubert – Porcher – Volbach 1980, pp. 142-147. 226 “Ma sono affinità molto generiche, che valgono soltanto a confermare il radicamento dell’artista nell’ambito dell’Italia settentrionale“. See Bertelli 1988, p. 61. 227 See Bertelli 2000, p. 189. On Anselm’s exile, see Braga 2006, p. 514. 51 addition, considering that the name Aliberga written in the first floor (Layer B) has Frankish origin228, and also that there seems to be a kind of relation, typological and iconographic rather than stylistic, between the enthroned Christ at Torba and the one from the Godescalc

Evangelistary [figs. 8-9]229, we might even suggest that the frescoes were rendered only after the fall of regnum langobardorum in 774. From the historical point of view, we could then support the 80’s of the 8th century as a terminus post quem for the frescoes’ creation, since it was only during this decade that, according to Tommaso Indelli, the situation after the invasion of Charlemagne into the Italy settled and stabilized230.

However, the question of the date of Mazarine 660 which stood at the beginning of our ideas is much more complicated231. Since, as Bernhard Bischoff stated, “we know nothing about its history”, we have to date and place it on the bases of stylistic analyses of words and images. The most recently, Giulia Orofino stated that its style is in comparison to other manuscripts attributed to Nonantola Abbey “differently developed, explainable by the later date232”. Accordingly, she ascribed it to the end of the 9th century233. Of course, Orofino’s explanation might seem insufficient, especially if we think of two generations and several

228 See Bertelli 1988, p. 40 and Bertelli 1992, p. 4. 229 The Godescalc Evangelistary, one of the first Carolingian manuscripts, is dated between 781 and 783. Thanks to the dedicatory verses, it is possible to connect the commission of this luxurious object with the baptism of Charles’s son, Pepin the Hunchback, at Rome in 781. Six whole-page illustrations (Four Evangelists, enthroned Christ, and the Fountain of Life) precede the text of the codex. See for example Bullough 1965, p. 99-101 and 134; Schutz 2004, p. 379; and Crivello – Denoël – Orth 2013. 230 See Indelli 2013, pp. 54-59. 231 Mazarine 660 is comprised of two parts. One of those belongs to the 15th century. Logically, it is the second one in which we are interested. The manuscript is without any doubt a mysterious object. To use the words of Bernhard Bischoff: “nulla si conosce della sua storia”. This author claims that the paleographical analysis reveals two generations of scribes and several different hands. One of those is, in the author’s opinion, of the Nonantolan character; 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 and 118; Cavallo 2000, p. 89; Albarello 2000, pp. 158-159. For the early 10th century date, see Molinier 1885, pp. 301-302. For doubts about its North Italian origin, see Palma 1983, p. 147, n. 23, cf. Orofino 2006, p. 560. 232 “…il Cassioodro mostra uno stile altrimenti maturo, spiegabile con una cronologia più avanzata, verso la fine del secolo,…” See Orofino 2006, p. 561. 233 See ibidem. 52 hands which may have been responsible for the manuscript’s rendering234, but it is still more words than Bertelli had used to support his supposition in 2000. Subsequently, if we want to insist on a kind of similarity between the frescoes and Mazarine 660 (a similarity which must be perceived with cation due to the difference between the mediums), we would have to move one of these monuments through time. Should we do so with the wall paintings, i.e. date them to the second part of the 9th century, we would have to explain the late 8th century character of the frescoed inscriptions presented by Flavia de Rubeis. On the other hand, if we move the manuscript’s date towards 800, we would have to question Orofino’s position. From our point of view, after reading all the paleographic arguments of Rubeis, Kloos, and Bertelli, it would be difficult to renounce their hypothesis, and thus we still support the late 8th century dating for the Layer A. Layer B is probably only several years younger considering the similarity of faces and scripts preserved in the both floors. In this sense, the fresco decoration preserved at Torba could be used as an argument for the dating of the Mazarine 660.

However, after these few pages, we should be conscious of difficulties which the phenomenon we call “style” brings to the researchers of early medieval cultures, limited by the massive amount of lost objects and monuments. Accordingly, since we base our conclusion, solely on the stylistic analyses of both letters and images, we must be critical to our own hypothesis and to be open to different suggestions, maybe even the 9th century dating.

234 See Bischoff 1983, pp. 108 and 118. 53

Chapter 6: Interpretation

As we have seen, it is especially the second floor of the tower which raises intriguing questions regarding the iconography. There are in fact three major iconographic points, besides the row of apostles, saints, martyrs, and nuns. First, the depiction of Maiestas Domini on the north wall. Second, the votive image on the south wall. And third, without any doubt, the enthroned

Christ surrounded by angels, John the Baptist, and probably Mary on the east wall. According to John Mitchell and Bea Leal, the latter mentioned group, for which the Greek term “Deësis”

(δέησις) was frequently used by art historians who dealt with this east wall imagery235, has been depicted “in generic reference to the Last Judgement”236. We will start our interpretation with reflecting on this statement.

The “Deësis” and the Tower

To start with, we need to ask whether the term “Deësis” is actually appropriate in regards to the second floor decoration at Torba. Accordingly, we must focus our attention to the word

“Deësis” itself.

This Greek term, originating from the verb deomai (δέομαι - to plead), was used for a description of petitions presented to the late Roman and Byzantine emperors (or rather officials)237. The Christian usage, parallel with the juridical one, employed the word to characterize “an entreaty, supplication or petition addressed to God or to holy men, as well as intercessory prayers, whether general or liturgical”238.

235 See Bertelli 1988, pp. 26, 29, 34, and 42 and Bertelli 1980, pp. 210-211. 236 See Mitchell – Leal, p. 342. 237 See Gilsdorf, p. 134. 238 See Walter 1968, p. 317. 54

It was the art historian Christopher Walter, whose articles formed, and still forms, the core of the research into the Deësis iconography239, who, in 1968, tried to unravel the meaning of δέησις used in association with several ikons listed in Byzantine monastic inventories from the 12th century240. The author realized that whatever may have been the exact nuance of this term in the vocabulary of his contemporary art historians, they would all agree to the sort of picture to which they would apply it – the image of Christ between the Virgin and the

Prodrome, the panel from the Princeton University Art Museum represents the ideal example in this sense [fig. 83]241. However, this, according to him, had not been the case for Byzantines living in the 12th century, who would not use the same term when describing the particular artistic scheme. This specific meaning of the word was introduced only at the end of the 19th century by the Russian art historian Kirpičnikov242. To use the words of Walter, there is in fact

“no positive evidence that the subject which we call nowadays a Deësis was given the same name by the Byzantines. The only case where we find the word δέη or δέησις associated with the picture is when a petition is actually being presented to Christ in the name of the donor of the picture”243. Consequently, the art historian stated that there are in fact three sorts of images which could be reasonably called a Δέησις: first, an image of a saint or the Virgin praying; second, an ex-voto where the donor presents his petition; third the Paraklesis where the Virgin presents a petition244. From his perspective, we should thus use the word “Deësis”

239 See Walter 1968; idem 1970; idem 1977; and idem 1980. For the introduction into this iconography’s studies, see also Bogyay 1966. 240 He worked with documents stored in the archives of the Russian monastery of Saint Pantaleimon on Mount Athos. Among the papers there is for example an inventory dating from the year 1143, which had been taken in the monastery of the Theotokos of Xylourgos. See Walter 1968, p. 311. 241 On this panel, see Mouriki 1968. 242 See Walter 1968, p. 312. Walter claimed: “If nevertheless in our time we have come to call by the name of Deësis the iconographic type which sows the most powerful advocates for humanity bent before Christ in prayer, it is thanks to the Russians, who so called the ikon which crowned the iconostasis, and thanks to the Russian scholar who called attention to this fact”. See ibidem, p. 324. See also Kirpičnikov 1893. 243 See Walter 1968, p. 317. 244 See Walter 1968, p. 323 55 for the image of Mary receiving the candle from the donor on the south wall [fig. 19] rather than for the east wall imagery [fig. 6].

However, before we jump to this rather hasty conclusion, we must also remember the art historian Anthony Cutler, who has challenged Walter’s idea in his more recent study. By introducing the passage of the late 11th century text Life of Lazarus the Galesiote by Gregory, his disciple, he proved that the term δέησις had also been applied by Byzantines in association to another images245. Consequently, Cutler “opened up” the definition of a Deësis and declared that, in the period between the 9th and 12th centuries, the form of an iconography linkable to the word was more expandable as well as flexible246.

Considering the use of the term from Anthony Cutler’s perspective, the earlier mentioned possibility that some other saint, closely linked to the abbey at Torba, and not the

Virgin has been rendered on the left part of the tower’s east wall, would appear as an unimportant remark247. As he claimed, “it mattered little whether there were two, three, or five figures; whether Mark ‘replaced’ the Prodromos, Martha was ‘substituted’ for Mary, or all human forms gave way to angels”248. We must add, however, that he was aware of the fact that the lack (or the presence) of specific iconographic element, should exclude the possibility of interpreting the image with a triadic composition as a Deësis249. He mentioned, for example,

245 See Cutler 1987, p. 147. This passage describes death of an old monk named Nikon “…who lay down on his straw mat in the place in which there are holy images of the Theotokos and of the archangel Michael stretching out [their arms] in supplication (δέησιν) to the Saviour, and quietly surrendered his soul to God through the hands of the angels”. 246 See ibidem, p. 151, n. 57. See also Gilsdorf 2012, p. 134. 247 See above Chapter 4. 248 See ibidem, p. 153. 249 Cutler mentioned the reliefs on the Ivory Scepter of Leo VI. See Cutler, p. 151, n. 57. For this object, with a relief of Christ flanked by Peter and Paul, as well as one with the Mother of God surrounded by Archangel Gabriel and the emperor, see Corrigan 1978. 56 the absence of the gesture of entreaty250 or adoration251, which is absent at Torba where John the Baptist stands frontally, holding the lamb, not pleading, not gesturing. In this respect, he is comparable to the Prodrome represented in the illuminated copy of the Christiane

Topography in the Vatican Library [fig. 84]252. John the Baptist is the subject of the recollected composition as reveals the text emphasizing him as the last of the prophets and his central position between five figures. The group of three to the left (Mary, Christ, and the Prodrome), which has, according to Walter, “a definite homogeneity”, is the subject of our attention253.

We can observe that, contrary to rather passive, Torba-like, role of John the Baptist, Mary stretches her arms out in the gesture of adoration towards Christ254. This could have been, and probably was, the case at Torba also. Thus, we can conclude that despite the absence of the iconographic motif mentioned by Cutler and the destroyed left part of the image, it is still possible to apply the term “Deësis” in regards to the upper chamber’s east wall imagery.

The Deësis and its Meaning

The question now arises as to how this iconography can be explained. What does it signify, or rather, what meaning does it conceal? Most of the scholars, starting from the end of the 19th century, explained for them rather narrowly specified iconographic motif as a derivation of the eschatological vision255. Consequently, they believed that this imagery bears an eschatological connotation. Studies of Nikodim Pavlovich Kondakov256, Jean Fournée257, or

250 See Cutler, p. 151, n. 57. 251 See Walter 1967, pp. 326-327 Christopher Walter points out that “The same ritual gestures, extending the arms, were used for adoration or for supplication, whether in imperial ceremonies, the liturgy or even in everyday life”. Therefore, there always will be a kind of equivocality in those. 252 This manuscript has been identified as the 9th century copy of a 6th century original. See Walter 1968, p. 328. 253 See ibidem. 254 See Ibidem. 255 For summary of interpretations and ideas concerning the Deësis, see esp. Bogyay 1967, pp. 61-64. 256 See Kondakov 1892, p. 274. 257 Fournée claimed: “Elle appartient dès le début à l’iconographie du Jugement Dernier, qu’elle se contente souvent de résumer”. See Fournée 1964, p. 104. 57

Louis Bréhier258, are truly great examples of the phenomenon. The last mentioned author, for example, claimed that “the Deësis is only an episode of the Last Judgment”259. Despite the fact that Kirpičnikov, already in 1893, claimed that originally there were no eschatological connotation linked to the image260, it was only in the second part of the 60’s that the traditional perception radically changed thanks to the studies of Thomas von Bogyay and

Christopher Walter261.

While trying to deal with the previous hypotheses concerning the Deësis, Christopher

Walter stated, basing his idea on the thorough study of images and inscriptions connected to them that “the advocacy of the Virgin and the Prodrome at the Last Judgment is simply a development of an already existing faith in their intercessory powers… 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 by reason of their intimate relationship with

Christ and by reason of their acknowledgment of his divinity during his life on earth”262.

According to Walter and Bogyay, the Deësis has been added to the Last Judgment theme, and thus acquired a new significance, only during the 11th century263. Bearing their ideas in mind, it would seem rather impossible to support the John Mitchell’s and Bea Leal’s proclamation that the Deësis on the east wall of the tower has been depicted in generic reference to the

Last Judgement264. In contrary, on the basis of their opinions, we should interpret it as a logical part of heavenly court’s vision, in which John and perhaps also Mary figure as the witnesses of Christ’s divinity.

258 See Bréhier 1928, p. 147. 259 Bréhier stated“…la Deësis n’est qu’un épisode du Jugemet Dernier”. See Bréhier 1928, p. 147. 260 See Kirpičnikov 1893, esp. pp. 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. 261 See Bogyay 1967 and Walter 1968. 262 See Walter 1968, p. 336. See also Kantorowicz 1942, p. 71. 263 See Walter 1968, p. 335; Walter 1971, p. 265; Bogyay 1967, pp. 62, 71-72. 264 See Mitchell – Leal, p. 342. 58

We must however pursue one more lead in our examination of the famous iconographic motif. This lead in a sense affirms the paradox of art history, since the more we are moving away from a certain phenomenon in the past the more we are getting to recognize its peculiarities. In 2016, Ivan Foletti and Valentine Giesser published an article about Rome in the 9th century265, in which they interpreted the decoration of the famous San Zeno chapel at

Santa Prassede [fig. 85]266. This funerary chapel, or, to be more precise its embellishment, ordered by Pope Paschal I (817-824) for his mother Theodora, provides another point of view on the tower’s decoration. The iconographic concept of this space, without any doubt specific in Rome’s context, contains besides other things the vision of an empty throne – so called

Etimasia, Transfiguration, Anastasis, and the most essentially the depiction of the Prodrome and Mary surrounding God – who is represented by the light which intrudes into the chapel through the east window267. All these iconographic elements, as the art historians pointed out, are constituents of later Last Judgment’s representations – such as the one preserved on the reverse west façade at Torcello originating in the 11th century [fig. 86] or the one in the parecclesion of the Church of the Holy Saviour in Chora from 14th century. Ivan Foletti and

Valentine Giesser suggested, also on the basis of the usual night vigil – Pannyichis – which has

265 See Foletti – Giesser 2016. 266 From rather extensive bibliography of the San Zenon chapel in Santa Prassede see for example Nilgen 1974; Brenk 1975; Wirenfeldt Asmussen 1986; Mackie 1989; Wisskirchen 1992; Mackie 1995; Goodson 2010; or Pirochtová 2012. 267 Foletti and Giesser claim that “In St. Zeno, la scena con Maria e Giovanni Battista che circonda il lucernario deve essere letta come un’autentica Deesis, ove la finestra è figura del Cristo, simboleggiato dalla luce”. The Authors base their “God-Light” hypothesis mainly on a liturgical chant, a vigil to be more precise, which was likely composed by Patriarch of Constantinople Germanus I (715-730). This text celebrates Lord - “giudice di (noi) tutti”, the light, and the salvation. The waiting for a dawn and the idea of the return of the Judge are overlapping in this poem. The hypthesis of Foletti and Giesser is also grounded in the way windows have been used in the eastern architectural praxis, since, as the art historians claim, “in Oriente esse sono considerate quale canale attraverso cui filtra la luce divina” See Foletti – Giesser 2016, p. 228-230. For the different, rather pragmatic use of windows in the West and its “metaphysical” utilization in the East see Ivanovici 2013. Ivanovici is skeptical towards the use of widows as representations of the Divine Light in the West, the San Zeno chapel included. However, the presence of Greek monks at Santa Prassede in Rome would allow to suppose the eastern mode of the use of windows. See Ivanovici 2013, p. 45; cf Foletti 2016, p. 229. 59 been sung in the period between an individual’s death and his or her funeral, that we should perceive San Zenon’s decoration as a kind of the Last Judgement representation. This carefully laid hypothesis challenged, on several levels, the older ideas about the Deësis iconography. It suggested that this image has formed a part of the Last Judgement iconography much sooner than in the 11th century. In fact, since it were Greek monks, documented by Liber Pontficalis268, with whom we can link the conception of the chamber’s decoration, it seems possible to imagine that the connection between the two had existed even before the first phase of iconoclasm begun. It thus disproves Walter’s and Bogyay’s argument considering the Last

Judgment’s non-existing relation to the Deësis and their following interpretation of this image when dated before the 11th century. The Prodrome and Mary in the San Zeno chapel, would not figure only as mere witnesses of Christ’s divinity, although, we must emphasize that this role is here also essential, since God – the Light – is represented only by an architectural feature – a window. They would be in fact pleading with God – the Judge, similarly to the monks who were singing during the whole night, for the soul of deceased. This aid 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, but also it may help to a soul which has been in the interim, i.e. a place between these two judgments269.

268 “Qui dum sanctissimi atque coangelici praesulis haec intima cordis vigilantia gererentur ut reconditorum ibidem sanctorum corporum Deo indesinenter super astra placentium precibus apud omnipotentem Dominum iuverentur, construxit in eodem loco a fundamentis cenobium, quod et nomine sanctae Praxedis virginis titulavit; in quo et sanctam Grecorum congregationem adgregans, quae die noctuque grece modulationis psalmodie laudes omnipotenti Deo sanctisque illius ibidem quiescentibus sedule persolverent introduxit”. See Duchesne 1892, p. 54. 269 The distinction between two judgments should be taken into consideration when we are dealing with a funerary chapel, judgment’s depiction, or an intercessory role of some figures. The early Christians assumed that the end of time and the following resurrection would happen in the near future. With time it started to be clear that the apocalyptic vision was not going to happen this soon. Consequently, the explanation of the two judgments’ role must have been provided. Furthermore, it must have been established what happens to an embodied soul (anima separata) in the so called interim – the intervening time between the death and the final judgment. Trying to explain these questions thus also meant to explain the relationship between two essential 60

If we then accept the above described reading of the San Zeno decorative scheme, the generic Last Judgment reference of the Deësis, suggested by John Mitchell and Bea Leal in

2013 becomes again relevant. At this point however, we need to recall the words of Ernst

Kantorowicz, who explained as soon as in 1942, that even though the Deësis is closely linked to the Last Judgment, it “is at the same time an independent subject” which doesn’t have to be related to it270. Now, having in mind the Last Judgment iconography at the San Zeno chapel, we are going to look for another possible images preserved in the second floor of the tower at Torba and try to decide whether we should perceive Torba’s Deësis as this Last Judgment – eschatological – reference or rather as an independent subject which in content has nothing to do with it.

substances comprising men – body and soul. Unsurprisingly, thinkers such as Tertullian (d. after 220), Augustine (354-430), or Thomas Aquinas (1225-1274), commented on the questions which these two separated judgments stirred up. See for example Angenendt 1997 and Mecklenburg – Mertens 2013, esp. p. 19-21. To understand what the interim and the two judgments meant for an 8th century individual, we have to refer to a miracle described by Bede (672/673-735). This splendid example is to be found in the fifth book (twelfth chapter) of the Bede’s well known Ecclesiastical History of the English Nation, where the author describes a vision of “a certain person, who had been some time dead, rose again to life, and related many remarkable things he had seen” [241]. The resurrected man called Dryhthelm (Drithelm), one of the Northumbrians, portrays the structure of the place where the souls waited during the interim: First, a fiery pit from which there is no way to all eternity; Second, cold and flames consuming vale where are the souls of punished ones, who “at length have recourse to repentance at the point of death, and so depart this life” [244]. These souls, as the man states, will be received into the kingdom of heaven “at the day of judgment” [245], since they confessed and repented at the deathbed. However, the prayers, alms, and fasting of the living can bring them some relief even sooner; Third, the flowery place in which are received good souls, who cannot be yet accepted into the kingdom of heaven; Finally fourth, there Is already kingdom of heaven into which the perfect souls enter immediately after they depart their bodies. This miracle proofs that prayers and intercession for one’s soul were not sought only during the Last Judgment. See Bede 1939, 5. 12, pp. 241-246. In fact, it was already Augustine of Hippo who divided the souls in the interim according to their need of help from the living. For the perception of interim within the medieval context, see also Foxhall Forbes 2010. For the relevant Augustine text, see ibidem, p. 63. On the intercession for the souls in the interim, see Oakes 2008, p. 21. 270 See Kantorowicz 1942, pp. 70-71. 61

The Last Judgment at Torba?

We have already mentioned several iconographic elements which led Ivan Foletti and

Valentine Giesser to presuppose allusion to the Last Judgment in the San Zeno chapel

(Etimasia, Transfiguration, Anastasis, and Deësis)271. There is actually none of these iconographic motifs, besides many time mentioned Deësis, visible on the walls of the second floor chamber at Torba. In fact, only one image, rather one fragment of it, is from a certain perspective linkable to the Last Judgment. Namely, it is the fragment of a lion supporting the rest of mandorla in the upper part of the room’s northwestern corner [fig. 31]. It has been suggested, on the grounds of the remaining pieces, that it used to comprise a part of an iconographic motif known in today’s art historical literature as Maiestas Domini 272. The link between this scheme and the Last Judgment is however truly problematic.

In the core of this rather difficult connection lies the question of how should we interpret this visual scheme. The Maiestas Domini as 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 the series of visions of the End, or in eschatology. It is due to this background, as well as due to the revision of Apocalypse undertaken by exegetes in the 12th century273, that an image of Christ standing or seated in a mandorla framed by the four living creatures appeared in the 12th century Last Judgment iconography – the baptismal font from St. Boniface in Freckenhorst, dated to 1129 [fig. 87] 274,

271 See Foletti – Giesser 2016, pp. 228-230. 272 See above n. 157. 273 See Thunø 2015, p. 74. See also Christe 1992, p. 235. Latter mentioned art historian explains that during the 12th and 13th centuries the interpretation of the Apocalypse underwent “…a vast program of revision in Paris undertaken by exegetes…” such as Joachim of Fiore (d. 1202) and various Franciscan. 274 Depiction of Christ showing his wounds in a mandorla flanked by the four animals is rendered on this baptismal font. The resurrected are represented on his right side. Haloes around their heads are helping us to distinguish them from the condemned on the Christ’s left. The image is accompanied by the text from Matthew 25:34 and 41. On this image see Christe 1999, p. 180. 62 or the relief of the main portal at Church of St. Trophime in Arles, finished probably in 1178,

[fig. 88] should be remembered in this context275. Even more important though, is that on the basis of this origin, as well as on the basis of the above mentioned images from the 12th century, the modern authors linked the Maiestas Domini (as well as another apocalyptical motifs), when depicted on its own before the 12th century, to a triumphal theophany removed to the end of time. Consequently, they interpreted it as the Last Judgement’s reduction276.

Following the logic of those art historians, we must mention a text of Bede (673-735) who, while presenting the life of his teacher Benedict the Bishop, describes a decoration of St.

Peter’s church founded by his master:

[Benedict] brought back many holy pictures of the saints to adorn the church of St. Peter he had built: a painting of the Mother of God, the Blessed Mary ever-Virgin and one of each of the 12 apostles which he fixed round the central arch on a wooden entablature reaching from wall to wall; pictures of incidents in the gospels with which he decorated the south wall and scenes from St. John‘s vision of the apocalypse [imagines visionum Apocalipsis beati Iohannis] for the north wall. Thus all who entered the church, even those who could not read, were able, whichever way they looked, to contemplate the dear face of Christ and His saints, even if only in a picture, to put themselves more firmly in mind of the Lord’s Incarnation and, as they saw the decisive moment of the last judgement [vel extremi discrimen examinis] before their very eyes be brought to examine their conscience with all due severity277.

275 On the elaborated iconography of the portal which includes processions of both the damned and the saved souls, see Christe 1999, pp. 199-201. 276 See Christe 1999, p. 150. The author states, “…ont été interprétées par l’histoire de l’art comme des apparitions judiciaires, comme des Jugements derniers ‘réduits’”. On this question see also Christe 1992, esp. p. 236. Apocalyptic motifs in early Christian art has been, in fact, usually understood similarly in these eschatological terms. See Klein 1992, p. 161. Very important role in this sense played especially the apse mosaics of St. Apollinare in Classe in Ravenna or that of St. Pudenziana in Rome. For interpretation of these apses in eschatological terms see Dinkler 1964, esp. pp. 77-100, Matthiae 1967, pp. 59-63. Cf. for example Filipová. 277 “Picturas imaginum sanctarum quas ad ornandam aecclesiam beati Petri apostoli, quam construxerat, detulit; imaginem videlicet beatae Dei genetricis semperque virginis Mariae, simul et duodecim apostolorum, quibus mediam eiusdem aecclesiae testudinem, ducto a pariete ad parietem tabulato praecingeret; imagines evangelicae historiae quibus Australem aecclesiae parietem decoraret; imagines visionum Apocalipsis beati Iohannis, quibus septentrionalem aeque parietem ornaret, quatinus intrantes aecclesiam omnes etiam litterarum ignari, quaquaversum intenderent, vel semper amabilem Christi sanctorumque eius, quamvis in 63

The north wall of St. Peter’s church, as we read, has been decorated with images – painted panels of considerable number – from St. John’s vision of the apocalypse278, which seem to be in this case somehow linked to the concept of the Last Judgement [vel extremi discrimen examinis]279. Thus, comparing the Torba Tower with St. Peter’s church through this perspective, we could arrive to the conclusion that the viewer standing in front of north walls of both the church and the tower would be in presence of the iconography based on the apocalyptic vison, linkable even to the Last Judgment.

However, this line of reasoning seems to be indefensible, not only because we have just made a parallel between a single motif based on the St. John’s vision and quite a few unknown panel paintings once ornamenting the St. Peter's church at Wearmouth, but mostly because, as Yves Christe convincingly asserts, the significance of the Maiestas Domini is

“present ecclesiological, rather than the future eschatological”280. This idea is based on the ecclesiological reading of John’s Apocalypse introduced by Tyconius – a North African lay theologian who lived at the end of the 4th century281. Throughout his exegesis, to use accurate and clear words of Eric Thunø, “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

imagine, contemplarentur aspectum; vel dominicae incarnationis gratiam vigilantiore mente recolerent; vel extremi discrimen examinis, quasi coram oculis habentes, districtius se ipsi examinare meminissent“. See Darby 2010, pp. 1-2, n. 2 or Bede 1954, pp. 405-406. 278 On the paintings brought by Benedict the Bishop to Wearmouth, see Meyvaert 1979, esp. p. 74, n. 3. 279 There is a certain amount of ambiguity in this reference. It seems that apocalypse motifs in its first part are actually linked to the Last Judgment mentioned in the second one. It would thereby be truly tempting to suggest that these concepts are overlapping in the mind of the 8th century thinker. However, this idea, which might deserve more profound analyses, should be treated carefully. Bede’s description is truly general, hence we do not have any idea what have been depicted on the painted panels, or whether the Last Judgment, which Bede mentions, should be truly linked to the north wall of the church. Furthermore, Bede is one of the commentators on the Apocalypse who draws inspiration from Tyconius’s late fourth-century commentary which, in fact, de-emphasizes the eschatological value of the Apocalypse, and thus removes it also from a concept of the Last Judgment. See Thunø 2015, p. 240, n. 29 and Christe 1996, pp. 19-51. 280 For this topic, see Christe 1992, [p. 255]. See also Christe 1996, pp. 7-17, which summarizes art historian’s previous scholarship on this subject, or Thunø 2015, pp. 74-81. 281 On his commentary, see for example Steinhauser 1987. 64 earth and in heaven”282. The theologian’s commentary, which de-emphasizes the eschatological value of John’s vision, was in fact essential for shaping the nature of major Latin commentaries on the Apocalypse from the 4th to the 12th century283. Subsequently, during this period, the Apocalypse was not perceived eschatologically, as “a revelation about the end of time”284, but as “a vison of the glory of God already realized, a present eschatological reality”285. Following this reading of the Apocalypse, we must conclude by stating that the

Maiestas Domini on the north wall of the second floor chamber at the Torba Tower, a motif drawn from John’s vision, does not allude to the end of time, the Last Judgment, or the second coming of Christ. Instead, it represents His glory in the already existing, present, celestial

Church-Kingdom286.

Of course, the frescoes ornamenting the north wall of the room are, as we have already seen, particularly ill-preserved. Thereby, we cannot exclude the possibility, however small it is, that other signs would suggest different interpretation287. Yet, to dispel our doubts, we should draw our attention towards the oldest representation of the Last Judgment in the

West, datable around 800. We discover this image, rendered in fresco, in the main church of the monastery of St. John in Müstair, to be more precise, on the church’s west wall [fig. 89]288.

This representation is divided into three registers. The Parousia, the rolling up the heavens, and the raising of the dead are iconographic motifs depicted in the upper part. The Christ-

Judge in a round mandorla flanked by several angels and twelve seated apostles are then

282 See Thunø 2015, p. 74. 283 See Ibidem. See also Christe 1996, pp. 19-51 and idem 1992, p. 235. 284 See Christe 1992, p. 237. 285 See ibidem, p. 238. 286 For the idea of the Church now, as the kingdom of Christ and the kingdom of heaven, for which understanding is essential Augustine’s book City of God, see Thunø 2015, pp. 74-75. 287 Yves Christe, for instance, claims that Maiestas Domini “rarely introduced the Last Judgment”. See Christe 1992, p. 257. 288 See above, n. 219. 65 recognizable in the middle register. The redeemed and the doomed are represented in the lower part of the composition289. There is no uncertainty from the iconographic point of view about the depicted subject. Subsequently, also the iconographic differences between the fragment from Torba and the Last Judgment at St. John’s monastery – a substitution of the four animals for the angels surrounding the mandorla – support the idea that around the

Maiestas Domini in the tower has not been depicted any visual reference to the Last

Judgment’s iconography.

Since, there seems to be missing the reference to the Last Judgment or eschatology in

Maiestas Domini and the rest of the frescoes as well, we can probably perceive the presence of the Virgin and John the Baptist outside of this eschatological context. Accordingly, we must conclude our research into these iconographic elements by stating that no direct allusion to the Last Judgment or eschatology is to be found, neither in one of the chamber’s decorative parts nor in the scheme as a whole. Logically, if we follow the idea that the Last Judgment and eschatological themes have been connected to the funerary function, as in the case of the funerary chapel of San Zeno in Rome, should we now reconsider also the hypothesis of

Mitchell and Leal, who claimed that the tower with its two stories was restored “expressly for funerary use”290? Before answering the question, let us first present our interpretation of the upper room’s wall paintings.

The Votive Chapel

Our interpretation of the chamber’s decorative scheme is based on the understanding of the south wall scene, to be more precise, on the image of an offering to Mary with child. To comprehend this visual motif properly, we need to return to the end of the 4th century when

289 On the iconography, see Christe 1999, pp. 152-171. For photos, see Goll - Exner - Hirsch 2007, pp. 212-225. 290 See Mitchell – Leal 2013, p. 344. 66 the cult of Mary has been still developing291. The fierce reaction of Bishop Epiphanius of

Salamis292, recorded in a letter from around 367293, to a praxis of women294 who “in the name of the Ever-Virgin make an offering (epitelein) of small loaves”295, plays an important role in this sense. The bishop condemned this cult of Mary as “wicked and criminal, a perversion of the Holy Spirit’s preaching“296, or, to use clear words of Hans Belting, as “a relapse into heathenism”297. The bishop’s response to the practice offers two truly important information.

First, 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 betterment to the supplicants, already in the 4th century298. Second, these practices were at first condemned by theologians, because, in fact, they were based on pagan origins, to be more specific, on many cults of goddess-mothers (Cybele, Diana of Ephesus), still popular during the 3rd century299. However, this initial opposition of theologians changed after the council of Ephesus in 431, during which was Mary recognized 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 God300. It is exactly this tradition and belief on which has been partly developed the type of picture visible in the tower.

291 See Belting 1994 (1990), pp. 32-36; Oakes 2008, pp. 7-23; Turchi 1951; 292 Also known as Epiphanios of Cyprus. Cf. Esbroeck 2005, p. 63. 293 Epiphanius quoted this letter in seventy-eighth chapter of his later work Panarion. See Coyle 1993, p. 145. Cf. Belting 1994 (1990), p. 34. 294 These groups of women, who had emigrated from Thrace to Arabia, were called Kollyridians or Philomarians. See Belting 1994 (1990), p. 558, n. 13; Coyle 1993 [1978]; and Delius 1963, p. 100. 295 For the English translation of the letter, see Coyle 1993, p. 145. 296 See ibidem. 297 See Belting 1994 (1990), p. 34. 298 The special role of Mary as intercessor is recognized also in a papyrus fragment of a Greek prayer directly addressed to her, which is known as the Sub Tuum. This papyrus is dated to the 3rd or the 4th centuries. See Oakes 2008, p. 20. 299 See Belting 1994 [1990], p. 33. 300 See Ibidem or Turchi 1951, p. 47. 67

The other part is then connected to even older belief, namely, to the assumption that angels and saints intercede for Christians in heaven. This idea had been presented already in the first half of the third century by Origen301, and its institutionalization initiated in the East and the West from the late 3rd century and the late 5th century respectively302. The tradition was truly essential in the process of successive visualization of Saints’ and Mary’s intercessory powers, which also means in the forming of a cult image. In fact, first visual schemes resulting from this belief – image of a local saint with prostrating figures at his feet - appeared already around 400303 [fig. 90]. The development culminated in St. Demetrius’s church in Salonika where the local saint’s representation multiplied all over the church [fig. 91]304. However, at

St. Demetrius, the mosaic images which depicted the intercessory power of the saint appeared in a new votive, or ex-voto, function. This means, according to Hans Belting, that they were

“meant to present private individuals or public figures as the saint’s client”305. By funding of an image depicting the intercessory power of the saint, the private “client”, besides deepening his or her personal relationship with the saint306, asked or thanked for advantages of the actual

301 In the sixth chapter of his De oratione Origen wrote, “But these pray along with those who genuinely pray— not only the high priest but also the angels who ‘rejoice in heaven over one repenting sinner more than over ninety-nine righteous that need not repentance,’ and also the souls of the saints already at rest”. For English translation, see Origen 2001, p. 17. On this treatise, see Quasten 1962a, pp. 66-69. On Origen, see ibidem, pp. 37-101. 302 See Oakes 2008, p. 20. 303 The saint’s image situated beneath the church of SS. Giovanni e Paolo is exemplary in this sense. See Belting 1994 (1990), pp. 80-82. 304 The Dating of the mosaics of St. Demetrius’s church, discovered in 1907, is limited by a fire in 1917 which destructed many scenes. Fortunately, in 1909, Walter M. George created several aquarelles of the later annihilated parts of decorations, thereby these have not been completely lost. See for example Cormack 1969, pp. 17-19. The mosaics are dated into two executive phases. First, before around 620 when the original church suffered damages resulting from another fire. Second, after the year of this disaster. Ernst Kitzinger stated that the gap between the images (ex-votos) made before the fire (these were concentrated mainly in the inner north aisle of the basilica) and those made after (votive images on the piers flanking the entrance to the chancel, etc.) is not great. The art historian dated the first group to the late 6th and early 7th centuries; the second one to the middle of the 7th century. See Kitzinger 1958, pp. 20-27. Of course, other hypotheses exist. Jeffrey C. Anderson for example supports the date around 680 for the second group of images. See Anderson 1999, p. 55. On the dating of the first basilica to the beginning of the 6th century, see Spieser 1984, pp. 165-214, esp. p. 212. See also Bakirtzis 2012. 305 See Belting 1994 (1990), p. 82. 306 See Brown 1981, pp. 56-58. 68 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. The variation of this scheme, in which Mary substituted a saint, was quite usual motif of the 8th century visual culture307. Subsequently, its presence in the second floor of the tower is not really surprising. Nevertheless, its interpretation leads us towards intriguing hypothesis.

We would like to suggest that it is from this intercessory perspective that we have to perceive the whole decoration of the upper room at Torba. The fresco cycle with its accentuation of apostles, martyrs, saints, John the Baptist, and Mary, as well as with its strikingly high number of portrait images of contemporary individuals, would thereby be interpreted as a votive scene which has expanded on the four walls of the room, i.e. a votive chapel. Thus, the people entering this small room would found themselves in the presence of powerful advocates, the nearest figures to Christ himself. In fact, a person, probably a nun, would be surrounded by the vision of the heavenly kingdom. Of course, this kingdom might be perceived by its viewer in the eschatological sense, as the promised eternal kingdom, a future reality somewhere at the end of time. Accordingly, we just cannot rule out the possibility of the funerary use of this votive chapel and the whole tower, since the intercession of the saints would be asked with this eschatological vision in mind. However, this heavenly place could also be interpreted either as an paradisiac place of joy in the vision of the interim described by Bede308, or in the ecclesiological terms of Tyconius reading of John’s Apocalypse, as the “now” existing celestial kingdom309. In these cases, the intercession would be asked

307 We will find examples in Santa Maria Antiqua or Santa Sabina in Rome. See Teteriatnikov 1993 and Foletti - Gianandrea 2015. 308 See above, n. 269. 309 See above “The Last Judgment at Torba?”. 69 either for the souls in the place between the two judgments or for people living in the present time. Even though the ambiguity between these modes of intercession is the most likely explanation, we can also imagine, remembering Ernst Kantorowicz’s interpretation of the mentioned ivory triptych from Vatican [fig. 73]310 and the fact that the funerary images in the lower chamber were painted only after the both rooms have been embellished with fresco decoration311, that the pleadings with the saints were primarily related to the present terrestrial existence. In that instance, we would have to abandon the connection of the tower with dead, and replace it with the concept of the “tower of life”.

Reflections and Nuns at Torba

Following this line of ideas, we must return to the last still unexplained part of the cycle, i.e. the depiction of praying nuns in the lower part of the west wall [figs. 28 and 92]. Interpretation of this scene in the intercessory context has been already suggested by several authors312. We can thus borrow the words of John Mitchell and Bea Leal, who wrote that these nuns are

“calling on their celestial patrons to intercede with Christ for their own elevation and preservation and for the health and salvation of the benefactors of the monastery”313. Their role within the cycle seems to be thus clear, however, we would like to add several further remarks concerning this image.

310 Kantorowicz stated that this triptych, iconography of which is the most similar preserved iconographic scheme to the tower’s second floor imagery, “represents in lavish style a litany or a suffrage for the Basileus which in content has nothing to do with either a traditio legis or a Last Judgment”. See Kantorowicz, p. 71 and above “Iconography and its Models”. 311 We could imagine that this desire to be buried in the lower room has been caused by the presence of some relics, which had been put there for different reasons. See below Chapter 7. In this context, we must recall the abbey at Farfa, to be more precise, we must remember Abbot Sichardus (830-842), who built there, as documents Constructio farfensis (from the 9th century), the oratory “joined to the church of the Virgin with a crypt below where he honorifically interred the bodies of the holy martyrs Valentine and Hilarius, translated from Tuscany, together with the body of St. Alexander the son of Felicitas”. It has been suggested that the base of this oratory correlate with the later bell tower. As always, there also are different hypotheses. See McClendon 1987, pp. 7 and 76. 312 See Bertelli 1988, p. 29 or Mitchell – Leal 2013, p. 343. 313 See Mitchell – Leal 2013, p. 343. 70

To start with, we have to emphasize the nuns’ position in the lower part of the wall, which is, on such a scale, exceptional within this decorative scheme. We must explain this deviation in terms of hierarchy, or rather hierarchies. Firstly, the arrangement visualizes the fact that the nuns were not an integral part of the celestial vision. In other words, they have not been worthy of a higher position, since their distance from God has been too remote. And secondly, it also reflects a need to divide the minimalistic, square, originally only military chamber into liturgically functional sections314. The eight nuns in the lower part of the wall have to be perceived as an important expression of this necessity, to which in fact belongs also the alternation of velum and marble imitations315. Logically, since the liturgical practice, or rather the arrangement of persons present at the liturgy, is transferred onto the walls, we could label this image as a kind of “ritual mirror”. It truly seems that the actual nuns entering the chamber would be conducted by the scheme itself. Put differently, the image would lead their way.

This reflection and ritual mirroring into the visual scheme leads us towards one intriguing hypothesis. It is based, firstly, on the visual resemblance of the nuns’ representation with an actual mirror, to be more precise, the Sisters seems to be depicted as if they were in a mirror316. The idea is supported by the fact that the nuns who entered the room, led by this image towards their liturgical position, would be in certain point standing face-to-face to their pictorial counterparts. The impression of looking in a mirror would thus be without any doubt strong. Secondly, this hypothesis is founded on Augustine’s idea of the mirror of human mind

314 Liturgy has been served in the chamber as proof the commemorative inscriptions under the Christ in front of which stood probably portable altar. See Bertelli 1980, p. 210. See above “Remarks on the Bibliography and one Hypothesis”. 315 On this alternation of the imitation of hanging vela and the polished marble opus sectile, see especially Mitchell – Leal 2013, pp.337-339. These authors claim that the dado is used “to articulate and introduce hierarchical structure”. This practice is to be found also at St. Vincenzo al Volturno. See ibidem, p. 337. See also Bertelli 1988, p. 26. 316 On the mirror and mirroring, see for example Pentcheva 2014. 71 described in his De trinitate317. In fact, this treatise, though written at the beginning of the 5th century, has been well-known and circulating in Europe at the end of the 8th century, as proofs the fact that Alcuin (735-804) wrote theological work summarizing this particular Augustine’s writing318 and that Bede’s treatises (673-735) also demonstrate the awareness of it319.

Augustine talks about a mirror already in the Soliloquia320 when he deals with the question of falsity321. He describes mirrors as material objects made by men, however, it is the

Nature itself that creates, “by representation”, “the images reflected from them”322.

Nevertheless, it was only in already mentioned De trinitate, of which, as Sabine MacCormack claims, “the Soliloquia are a kind of preludes”323, that Augustine put into use a metaphorical concept of mirror, essential for our hypothesis. While trying to explain the doctrine of the

Trinity in light of Scripture and seeking traces of this “Trinity which is God” in the inner man324,

Augustine referred, many times, to Apostle Paul’s words, “for we see now through a mirror obscurely [in enigma], and then face to face; now I know in part, and then I shall fully know,

317 For original, see Augustine 1968. For English translation of De trinitate, see Augustine 2002. On the Augustine’s idea of human mind as a mirror, see for example; Melchior-Bonnet 2001 [1994], pp. 110-111; TeSelle 2002 [1970], pp. 307-308; Pollack 2008, pp. 286-288; or MacCormack 2012, pp. 405-407. 318 See Cavadini, 1981. 319 See Fitzgerald 1999, p. 95. 320 The Soliloquia were written at Cassiciacum (near Milano) shortly after Augustine resigned his chair of rhetoric in 386 and decided to join the church. See Fredriksen 2012, p. 89. They are composed in form of a dialogue between Augustine and his own reason (ratio), see Fuhrer 2012, p. 274, and their central theme is “to know God and the soul” (Sol. 1.2.7). See also MacCormack 2012, pp. 398-399. For original text, see Augustine 1986. For English translation, see Augustine 1910. 321 (Sol. 2.6.10) See Augustine 1986, pp. 57-58 and Augustine 1910, pp. 66-67. 322 Augustine, in the person of Ratio states, “Moreover those objects of the sense of sight which appear to us as if really the things they look like, are produced and fashioned, some by nature, some by living beings. Nature produces these inferior similitudes either by reproduction [gignendo similitudines] or representation [resultando similitudines]: by reproduction, as when children are born resembling their parents, by representation, as in the case of every sort of reflector; for, although men make nearly all mirrors, they do not make the images reflected from them [eas quae redduntur imagines]” (Sol. 2.6.11). See Augustine 1910, p. 68 and Augustine 1986, pp. 59-60. 323 See MacCormack 2012, p. 399. 324 “I wished, therefore, to ascend as it were by steps, and to seek in the inner man a trinity of its own kind, both in knowledge and wisdom, as we previously sought it in the outer man, in order that we might come with a mind more developed by exercise in these lower things to the contemplation of that Trinity which is God [trinitatem quae deus est], according to our own modest capacity, if we can do even this, at least in an obscure manner and through a mirror” (Trin. 13.20.26). See Augustine 2002, p. 134 and Augustine 1968, p. 418. See also MacCormack 2012, p. 399. 72 as also I was known” (Young's Literal Translation, 1 Corinthians 13:12)325. Trying to understand this mirror Augustine says:

If we inquire what this mirror is, and of what sort it is, the first thing that naturally comes to mind is that nothing else is seen in a mirror except an image. We have, therefore, tried to do this in order that through this image which we are, we might see Him by whom we have been made in some manner or other, as through a mirror (Trin. 15.8.14)326.

Augustine believed, that the unity of God imaged, in however distant a sense, in the human mind327. Subsequently, if “nothing else is seen in a mirror except an image”, the Paul’s mirror has to bear a reference, according to Augustine, to the man himself, to his or her mind. In other words, we are the image of God328, and “through this image which we are, we might see

Him… as through a mirror”. However, we might see Him only “in enigma”, which means, according to the theologian, likeness, “yet a likeness that is obscure and difficult to perceive”

(Trin. 15.9.16)329. In fact, as Augustine states, seeing through this mirror and this enigma is possible only for those, “…who look upon their mind as an image, so that they are able, in some way or other, to refer what they see to Him, whose image it is, and also to see by conjecturing that which they now see through the image by beholding, since they cannot yet see face to face” (Trin. 15.23.44)330.

325 “Videmus nunc, inquit, per speculum in aenigmate, tunc autem facie ad faciem” (Trin. 15.8.14), see Augustine 1968, p. 479. 326 See Augustine 2002, pp. 181-182. “Quale sit et quod sit hoc speculum si quaeramus, profecto illud occurrit, quod in speculo nisi imago non cernitur. Hoc ergo facere conati sumus ut per hanc imaginem quod nos sumus uideremus utcumque a quo facti sumus tamquam per speculum”. See Augustine 1968, p. 479. 327 See MacCormack 2012, p. 403. See also Ayres 2000, p. 74. 328 TeSelle 2002 [1970], p. 308. 329 See Augustine 2002, p. 184. “Vnaest enim cum tota sic dicitur: Videmus nunc per speculum in aenigmate. Proinde quantum mihi uidetur, sicut nomine speculi imaginem uoluit intellegi; ita nomine aenigmatis quamuis similitudinem tamen obscuram et ad perspiciendum difficilem”. See Augustine 1968, p. 482. 330 See Augustine 2002, p. 214. “…illi qui eam [mente] tamquam imaginem uident, ut possint ad eum cuius imago est quomodocumque referre quod uident, et per imaginem quam conspiciendo uident etiam illud uidere coniciendo quoniam nondum possunt facie ad faciem”. See Augustine 1968, p. 522. 73

Following Augustine’s thoughts as well as the strong impression of reflecting in Torba’s scene, we could cautiously interpret the nuns at Torba, represented as if they were in a mirror, as a reference to or a reminder of the theologian’s idea that it is actually only through the mirror of mind, which is in His image, that we can contemplate God331. To conclude, a Sister entering the room, possibly from the north door332, led by the nuns’ image towards her liturgical position, would be in certain point standing face-to-face to this “community’s mind” which would remind her of the foundations of the Christianity itself, and of the proper way to contemplate God. Returning then from this image towards the east wall and the vision of heaven, the person would be ready for the liturgical ritual itself.

We must stress one more time that Augustine’s De trinitate circulated in Europe in the

8th century as the Bede’s texts and especially Alcuin’s summarizing version of this treatise demonstrate. It is thus within the realms of possibility that the person (persons) responsible for the scheme, as well as nuns for whom this scheme has been intended, was (were) acquainted with these Augustine’s ideas. Therefore, our reading of this part of the imagery seems to be thinkable333.

Chapter 7: The Chapel in the Tower as a Concept

While interpreting the frescoes in the second floor of the tower, we have emphasized their intercessory role and suggested that they were designed for a votive chapel334. Now, since it

331 Pollack 2008, p. 286. 332 See Genoud – Bertelli 1987, p. 70 and Bertelli 1988, pp. 12-13, 17. See also above n. 81 and 161. 333 On the dissemination of Augustine’s works in the Early Middle Ages, see also Weidmann 2012, pp. 434-436. The author points out that to form an idea of how the theologian’s texts circulated in the Middle Ages, we need to consider two surviving manuscripts with works by Augustine and the evidence of writers who cited him. It seems that the entire collection of his works may have been in Rome as soon as around 440, which means that the texts may have been accessible in Italy from then on. Augustine has also been read in the 8th century England. During the so called Carolingian Renaissance his treatises were widely read, cited, and disseminated by monasteries and their scriptoria. See also Saak 2012, pp. 465-467. 334 We have thus rejected Bertelli’s hypotheses that the tower was the only religious space of the convent when it has been painted and that its function was later replaced by the church which stands in its close 74 is certain that the tower had stood on the eastern edge of the castrum centuries before its two inner rooms were painted, the question arises whether it was necessity – for example, lack of resources – that inspired the person responsible for the visual scheme to use until then only military tower in a new sacral context. Should we not rather look for a diverse and perhaps even intellectual idea from which this choice resulted?

We will start our search with the Shepherd of Hermas, which is, although numbered among the Apostolic Fathers, one of the apocryphal apocalypses335. This book of revelations as revealed to Hermas in Rome by the agency of two heavenly figures (an old woman and an angel in the form of a shepherd), has been dated between the 1st and 2nd centuries336. From its two main parts and conclusion337, it is the first section consisting of four visions that is of particular interest to us.

The dreamer Hermas is accused of unrighteous desire in the first vision. While considering and doubting his own heart Ecclesia – the Church – comes before him in disguise of an old woman (vis. 1, 2)338 and urges him to do penance for his sins (vis. 2, 3)339. The same ancient lady appears to the man in the third vision and shows him a great tower under construction (vis. 3, 2)340. Subsequently, she explains: “The tower which you see being built is

vicinity. See Bertelli 1988, p. 47. In this context, we must emphasize that a part of today’s church was dated by Alessandro Dejana between the 8th and 9th centuries already in 1981. See Dejana 1981, p. 319. See also Brogiolo 2013, pp. 215-216 and Mitchell 2013, p. 344. The archeological research which is still in progress under the guidance of Gian Pietro Brogiolo will hopefully provide more information in this regard. 335 See Quasten 1962, pp. 92-105; Bogdanos 1977. For an original text as well as English translation, see Lake 1959 (1913). 336 As Johannes Quasten points out, Hermas refers in his second vision to Pope Clement who held office at the end of the 1st century. The Muratorian Fragment informs us, however, that the author was brother of the Roman bishop Pius (140-155). Quasten explains this contradiction as follows: “The two dates are accounted for by the way in which the book was compiled”. See Quasten 1962, p. 92 and Bogdanos 1977, pp. 33-34. 337 The first section contains four visions; the second one comprises the twelve commands and parables one to nine. See Quasten 1962, pp. 93-94. 338 See Lake 1959 (1913), pp. 10-13. On the ancient lady as the Church (vis. 2, 4), see ibidem pp. 24-25. 339 See ibidem, pp. 12-15. 340 See ibidem, pp. 28-33. 75 myself, the Church, who have appeared to you both now and formerly” (vis. 3, 3)341. Ecclesia also points out to Hermas that whenever the building “has been finished, the end comes (vis.

3, 8)”342. This end, which is shown in the fourth vision, she claims, will come soon, because the tower “will quickly be up (vis. 3, 8)”343. To understand this tower, we must add that the stones of diverse type from which it is being built are men, fitting or unfitting into the structure.

Accordingly, the construction itself gives to the reader great amount of information about the many facets of each man’s soul344. Furthermore, as Theodore Bogdanos explains, it also informs him or her about “God’s universal order on its cosmological, sacramental, social, and moral levels”345. Besides being the Church and a reminder of the end, it is thus, more than anything else, “a model for an inner reconciliation and reordering… a powerful symbol of universal and personal integration”346.

Considering the survival of several manuscripts – in Latin translation as well as in the original Greek – in the West from the 2nd to the 6th century and from the 9th to the 15th century, it is clear that the Shepherd of Hermas was well-known and in circulation347. In fact, Ernst R.

Curtius labeled this text as “the most important document of early Christian vision literature”348. Question now arises whether we should connect the tower at Torba with this

Hermas’s vision349.

341 See ibidem, pp. 34-35. 342 See ibidem, p. 49. 343 See ibidem. 344 See Bogdanos 1977, p. 39. 345 See ibidem, p. 40. 346 See ibidem. 347 See ibidem, p. 34. Cf. Quasten 1962, p. 103. 348 See Curtius 1953 (1948), p. 103. 349 In this context, we must remember the miniature from illuminated copy of the Commentary on the Apocalypse written around 776 by the monk Beatus of Liébana. This illumination, dated to 970, shows an illuminator Emetrius sitting with another scribe in a room beside the bell-tower at the monastery of Tavara. The connection between the scribes making the copy of the Apocalypse and the tower evokes a deep interest. However, only a research into possible link between Beatus’s commentary, or the monastery of Tavara, and the Shepherd of Hermas might proof some relation between those two towers. For the illumination, see Alexander 1992, p. 9; Williams 1992, pp. 43-49 and pl. 257. 76

Unfortunately, it is impossible to find a definite answer to this question. In fact, following the above indicated method, we could recall many different references to a tower and relate them to the choice of rendering frescoes at Torba. For example, we may remember words from Song of Solomon: “Thy neck as a tower of the ivory” (Song 7:4), which have been interpreted by some early Christian fathers (Cassiodorus, Philo of Carpasia) - “tower” as the

Church and “ivory” as its pure priests350. In addition, we might recall words from the second book of Samuel where it is God himself who is described as the high tower351. Wanting to be truly speculative, we might once more recollect Augustine, who, while trying to comprehend the text written by Apostle Paul (2 Corinthians 3:18)352, explains that the word speculantes, used in the Latin translation of Paul’s letters, means “beholding through a mirror [speculum], not looking out from a watchtower [specula]” (Trin. 15.8.14)353.

However, we would like to present a less hypothetical interpretation built on different line of ideas. To be more specific, the following hypothesis will not be based on the perception of the tower as an iconographic unit to which we could relate almost any kind of textual reference, but, it will be anchored on one hand in the perception of the tower’s materiality and historicity, and, on the other hand, in the already presented reading of the frescoes within its walls.

Firstly, we should thus return to the beginning of our research into the tower at Torba.

In the opening chapters of the thesis, it has been already stressed that during the turbulent

350 Jeffrey 1992, p. 773. 351 “My God [is] my rock — I take refuge in Him; My shield, and the horn of my salvation, My high tower, and my refuge! My Saviour, from violence Thou savest me!” (2 Samuel 22:3). 352 “Nos autem reuelata facie gloriam domini speculantes in eandem imaginem transformamur de gloria in gloriam tamquam a domini spiritu”. See Augustine 1968, p. 479. 353 See Augustine 2002, p. 182. “Speculantes dixit, per speculum uidentes, non de specula prospicientes”. See Augustine 1968, p. 479. By connecting this text with the tower, we would thus suggest that the person responsible for the choice based his or her idea on an intellectual play with words and, in a broader sense, on the idea of contemplating God – a claim which would be supported only by our reading of the nuns represented as if they were in a mirror. See above “Reflections and Nuns at Torba”. 77 period, after the fall of the Western Roman Empire, the tower retained its original defensive function354. This has probably changed in the course of the 8th century, when the late antique military outpost has been already turned into the castrum, a place of trade and encounters, and a convent has been established in the tower’s vicinity355. While working on the assumption of this transformation, Renato Bazzoni stated that at Torba, “the clash of arms has been substituted for whispering prayers”356. By using these words, he thus suggested that the protective function of the tower had lost its importance – it was, after all, replaced, substituted. This, however, seems to be misleading choice of words, since it is truly hard to imagine that the castrum’s fortification would be perceived by inhabitants of the place as unnecessary during the 8th century.

If we then return to our intercessory reading of the frescoes and to the interpretation of the second floor chamber as a votive chapel one of which benefactors might have been a high-ranking noblewoman living in the castrum, we might see the choice of ornamenting the tower in a different light. We believe that this choice should not be perceived as a

“substitution” of a military function for a religious one, but rather as an upgrade of its protective purpose by the addition of a new religious aspect. We would like to suggest that a person who donated funds to create frescoes within the tower, might have do so not only for his or her personal betterment, whether this would relate to now, then, or even afterwards, but for the betterment of life and protection of all citizens of Castelseprio. During the liturgy, the nuns, and sometimes perhaps also the donor, while seeing the celestial vision, would then be pleading with the saints not only for themselves, but also for the health, well-being, and the endurance of walls of the castrum and the people living in it.

354 See De Marchi 2013, p. 15 and Bazzoni 1997, p. 17. 355 See Bazzoni 1997, p. 17. 356 “Torba aveva visto sostituire al fragore delle armi, il sussurrare delle preghiere…“ Bazzoni 1987-1988, p. 153. 78

In this sense, we should recall the protective role of saints during the siege of

Constantinople by the Avars in 626, when the icons of the Virgin357, or perhaps her relics358, were placed above the gates and carried in solemn processions around the walls359. Even more intriguing might be a recollection of the frescoes preserved in the “rotonda” tower of Ansperto in Milano360. This tower in the garden of the Archaeological Museum (the ex-Monastero

Maggiore, or Monastero di S. Maurizio) has a Roman origin, and it constituted a part of

Milano’s fortification. According to some, it has been restored by Archbishop Anspert (861-

881) during the 9th century361. The frescoes within have been rendered only in the 14th century. Of course, chronological gap separating these wall-paintings and those surviving at

Torba is considerable, however, similarities between the two are clear. Firstly, the context of both is the same since they are located on the inner walls of reused military towers in the vicinity of the female Benedictine convent362. And secondly, there is an iconographic analogy between them. In fact, the paintings of the tower in Milano 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 surrounding, in this case, crucified Christ [figs. 93-94]. These arguments seem to be more than enough to create a link between the two monuments. The frescoes from the 14th century, which might even follow some older iconography (maybe from

Anspelt’s time), could thus be interpreted in the same way as those in the Torba Tower.

357 Traditional scholarship has maintained that icons of the Virgin were carried in procession during this siege of Constantinople in 626. See Belting 1994 (1990), 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 idem, pp. 497-498. See also Clemena 2010, p. 68. Cf. Pentcheva 2006. 358 Bissera V. Pentcheva is of the opinion that it was rather relics than icons that were used at this time to ward off the enemy. See Pentcheva 2006. 359 See Belting 1994 (1990), pp. 36-37. 360 On the frescoes, see Bisogni 1986. 361 For different hypotheses concerning Anspert tower’s origin, see Rossi 1914, p. 152-180. 362 See Bisogni 1986, p. 3. 79

On the bases of the protective role of saintly images during the early medieval sieges; analogies between the “rotonda” tower of Ansperto and the tower at Torba; and the most importantly, on the grounds of the presented interpretation of the frescoes, we conclude with the following statement: The wall paintings in the tower at Torba, while reflecting a deep belief in the intercessory power of saints, Mary, and John the Baptist, has been created to improve the protection of Castelseprio through this tower, since it was not the tower which protected the saints, but the prayers of the nuns which might through the saints’ intercession with God protect the tower and the whole castrum.

80

Chapter 8: Conclusion

This foregoing thesis has dealt with the tower at Torba and its fresco decoration. It has started with the presentation of the historical development of its surroundings. We have thus emphasized tower’s connection to Castelseprio and its original military function to which has been in the course of the 8th century added the religious one when the structure became a part of a Benedictine convent. We have also explained effects of the destruction of

Castelseprio in 1287 on the life of nuns, who left the place in 1481. The discovery of the frescoes in the second part of the 20th century and the subsequent “rebirth” of the tower in the 70’s, have also been dealt with. Only then we have discussed the bibliography and the significant role of Carlo Bertelli in the tower’s studies.

This thesis continued with a description of the frescoes situated in the two floors of the monument. Besides images themselves, we have also concentrated on the surviving tituli.

The iconographic description has been concluded with a reflection on the models of the upper room’s composition. Next sequence has opened the question about the frescoes’ date. We have pointed out the stylistic similarities between Mazarine 660 and the wall paintings at

Torba. Furthermore, we have also stressed the importance of the inscriptions for a dating of the chapel. Following the paleographic studies of Flavia de Rubeis, Rudolph M. Kloos, and

Carlo Bertelli, we have supported the late 8th century date, however, we have urged caution in regards to the objects which art historians or paleographers attribute only on the grounds of the style.

The interpretation of the upper room cycle has been the subject of the next part of this thesis. We have started with reflecting on John Mitchell’s and Bea Leal’s statement that the

Deësis has been painted in general reference to the Last Judgment. Consequently, we have dealt with the term “Deësis“ itself and its meaning. By doing so, we have stressed the words

81 of Ernst Kantorowicz, who claimed that even though the Deësis is closely linked to the Last

Judgment, it doesn’t have to be related to it. Subsequently, we have tried to reconsider whether we should perceive imagery at Torba as this Last Judgment – eschatological – reference or rather as an independent subject which in content has nothing to do with it.

Having in mind the Last Judgment iconography in the San Zeno chapel, where Deësis appears, we have attempted to find some other possible references to the eschatology on the walls of the second floor of the tower. Accordingly, we have interpreted Maiestas Domini, the iconography, which had been in the past connected to the Last Judgment. In this question, we have followed the ideas of Yves Christe who pointed out that due to the ecclesiological reading of John’s Apocalypse introduced by Tyconius, this image was not perceived eschatologically but as a vision of the glory of God already realized. Therefore, we have not found any eschatological reference represented in the chamber. These observations have led us to question the recent hypothesis of Mitchell and Leal, who claimed that the tower with its two stories was restored expressly for funerary use.

In the next part of the thesis we have tried to search for an alternative explanation.

With help of the votive image on the south wall of the room, the votive character of the chamber has been suggested. We have also pointed out that by this commission the person, perhaps comitissa Seprii, besides deepening his or her personal relationship with the saint, asked or thanked for advantages of the actual saint’s intercession. It has been stressed that we have to perceive the whole decoration of the upper room at Torba from this intercessory perspective. A person, probably a nun, who would climbed the stairs to plead with the saints and martyrs for the intercession, would thus be surrounded by the vision of the heavenly kingdom. We have claimed that this kingdom might be perceived by its 8th century viewer in its eschatological, ecclesiological, or “interim” sense. Accordingly, we have demonstrated

82 existence of diverse types of intercession – for death, for life, for souls in the interim, etc. We have also outlined, acknowledging that the nuns were the most probably pleading for all of these modes of help, the way of interpretation in which we would abandon the connection of the tower with dead, and replace it with the concept of the “tower of life”. Before we have reached the last part of the thesis, we had presented the reading of the nuns’ scene on the west wall of the room. Our interpretation of this particular image was based on the

Augustine’s idea of the mirror of human mind.

The final part of the thesis has dealt with the reasons which inspired the person responsible for the visual scheme to use until then only military tower in a new sacral context.

First, we have worked with the tower as with an iconographic unit to which we have related several textual references, known to the 8th century man. However, our final hypothesis has been anchored in a different line of ideas. On one hand, in the perception of the tower’s materiality and historicity, and, on the other hand, in the already presented reading of the frescoes within its walls. By emphasizing the original function of the structure, analogy between the “rotonda” tower of Ansperto and the tower at Torba, and the protective role of saintly images during the early medieval sieges, we have supported the idea that the choice to create a votive chapel has to be perceived as an upgrade of tower’s protective purpose by the addition of a new religious aspect. Subsequently, we have asserted that a person who had donated funds to create frescoes within the tower, had done so not only for his or her personal betterment, but for the betterment of life and protection of all Castelseprio’s citizens.

Though we have arrived to the conclusion of the thesis, we believe that we have not used up all the scholarly potential of the monument. Many intriguing questions still remain open. What was the tower’s relation to the remaining structures of the Benedictine convent?

How can we imagine life of the nuns in this particular place? These and other questions can

83 only be answered with further research of historians, art historians, paleographers, archeologists, etc. Fortunately, archeological research is today still in progress. It is deeply hoped that a newly discovered evidence, which might shed some light on the convent’s history, will encourage future investigations of the structure and that the frescoes will be included into the broader discussion of scholars. After all, the tower truly deserves our fully attention.

84

Bibliography

Primary Sources

Augustine, The Soliloquies, Rose Elizabeth Cleveland (transl.), Boston 1910.

Augustine, De trinitate, Libri XIII – XV, in: W. J. Mountain (ed.) Corpus christianorum, Series Latina, vol. 50a, Turnhout 1968.

Augustine, Soliloquiorum libri duo, in: Wolfgang Hörmann (ed.), Corpus scriptorum ecclesiasticorum Latinorum, vol. 89, Wien 1986, pp. 1-98.

Augustine, On the Trinity, Gareth B. Matthews (ed.), New York, 2002.

Bede Venerabilis, The Ecclesiastical History of the English Nation, Stevens John (transl.), London – New York 1939 [1910].

Bede Venerabilis, Lives of the Abbots, in: John E. King, (transl.), Baedae Opera historica, vol. 2, 1954, pp. 391-445.

Origen, Origen on Prayer, in: William A. Curtis (trans.), Grand Rapids, MI: Christian Classics Ethereal Library, New York 2001.

The Shepherd of Hermas, in: Kirsopp Lake (transl.), The Apostolic Fathers. vol. 2, The Shepherd of Hermas, The Martyrdom of Polycarp, The Epistle to Diognetus, London 1959 (1913).

Secondary Literature

Albarello Carlo, Cassiodoro Institutiones [Entry in Catalogue], in: Carlo Bertelli – Gian Pietro Brogiolo (edd.), Il futuro dei Longobardi, L’Italia e la costruzione dell’Europa di Carlo Magno, Milano 2000, pp. 158-159.

Alexander Jonathan J. G., Medieval illuminators and Their Methods of Work, New Haven – London 1992.

Andaloro Maria – Bordi Giulia – Morganti Giuseppe – Ballardini Antonella (edd.), Santa Maria Antiqua tra Roma e Bisanzio, Milano 2016.

Anderson Jeffrey C., A Note on the Sanctuary Mosaics of St. Demetrius, Thessalonike, Cahiers Archéologiques, vol. 47, 1999, pp. 55-65.

Angenendt Arnold, Geschichte der Religiosität im Mittelalter, Darmstadt 1997.

Arslan Edoardo, L’architettura romanica milanese, in: Storia di Milano, vol. 3, 1954, pp. 395- 521.

85

Ataoguz Kirsten, The Apostolic Ideal at the Monastery of Saint John in Müstair, Switzerland, Gesta, vol. 52, n. 2, 2013, pp. 91-112.

Ayres Lewis, “Remember That You Are Catholic” (serm. 52.2): Augustine on the Unity of the Triune God, Journal of Early Christian Studies, vol. 8, n. 1, 2000, pp. 39-82.

Bakirtzis Charalambos, Saint Demetrios, in: Charalambos Bakirtzis (ed.), Mosaics of Thessaloniki, Athens 2012, pp. 128-179.

Bazzoni Renato, Il nuclo altomedioevale di Torba, Sibrium, vol. 14, 1978-1979, pp. 217-218.

Bazzoni Renato, Torba, otto anni di lavoro, Cá de Sass, n. 95, 1986, pp. 19-22.

Bazzoni Renato, Torba non più dmenticata, Sibrium, vol. 19, 1987-1988, pp. 153-164.

Bazzoni Renato, Il Monastero di Torba e i suoi affreschi, in: Pietro C. Maran (ed.), Conservazione e valorizzazione degli affreschi nella provincia di Varese, Varese 1997, pp. 17- 22.

Belting Hans, Probleme der Kunstgeschichte Italiens im Frühmittelalter, Frühmittelalterliche Studien, vol. 1, 1967, pp. 94-143.

Belting Hans, Likeness and Presence, A History of the Image before the Era of Art, Chicago 1994 [1990].

Belting-Ihm Christa, Die Programme der christlichen Apsismalerei vom vierten Jahrhundert bis zur Mitte des achten Jahrhunderts, Wiesbaden 1960.

Bertelli Carlo, Relazione preliminare sulle recenti scoperte pittoriche a Torba, in: Atti del 6. Congresso internazionale di studi sull'alto Medioevo, Milano, 21-25 ottobre 1978, vol. 1, Spoleto 1980, pp. 205-218.

Bertelli Carlo, Pittura di età carolingia nell’Italia settentrionale e a Castelseprio, Sibrium, vol. 19, 1987-1988, pp. 81-86.

Bertelli Carlo, Gli affreschi nella torre di Torba, Milano 1988.

Bertelli Carlo, Castelseprio e Milano, in: Carlo Bertelli (ed.), Bisanzio, Roma e l’Italia nell’Alto Medioevo, vol. 2, Spoleto 1988a, pp. 869-914.

Bertelli Carlo, L’Alto Medioevo, in: Massimo Lanza (ed.), Pittura tra Ticino e Olona, Milano 1992, pp. 3-9.

Bertelli Carlo, Aspetti dell’arte promossa dai Longobardi in Italia nell’VIII secolo, in: Carlo Bertelli – Gian Pietro Brogiolo (edd.), Il futuro dei Longobardi, L’Italia e la costruzione dell’Europa di Carlo Magno, Saggi, Milano 2000, pp. 189-195.

Bertelli Carlo, Ultimi studi sulle pitture di Castelseprio e Torba, in: Carmen Benati (ed.), Castel Seprio e Vico Seprio, Castelseprio 2002, pp. 1-8.

Binaghi Olivari Maria Teresa, Castelseprio: Santa Maria Foris Portas e Torba. Stato attuale di conservazione degli affreschi, Sibrium, vol. 14, 1978-1979, pp. 223-224.

86

Binaghi Olivari Maria Teresa, Il monastero di Torba, in: I Longobardi e la Lombardia, Milano 1978, pp. 159-160.

Bischoff Bernhard, Manoscritti Nonantolani dispersi dell’epoca Carolingia, La bibliofilia, vol. 85, 1983, pp. 99-124.

Bisogni Fabio, Gli affreschi della "Torre di Ansperto" a Milano, Arte Cristiana, n. 74, pp. 3-14.

Bogdanos Theodore, The Shepherd of Hermas and the Development of Medieval Visionary Allegory, Viator, vol. 8, 1977, pp. 33-46.

Bognetti Gian Piero, S. Maria foris portas di Castelseprio e la storia religiosa dei Longobardi, in: Gian Piero Bognetti – Gino Chierici - Alberto De Capitani d'Arzago, Santa Maria di Castelseprio, Milano 1948. pp. 15-511.

Bognetti Gian Piero, Aggiornamenti su Castelseprio III, Sibrium, vol. 4, 1958-1959, pp. 19-79.

Bogyay Thomas von, Deēsis, in: Klaus Wessel (ed.), Reallexikon zur Byzantinischen Kunst, vol. 1, Stuttgart 1966, pp. 1178-1186.

Bogyay Thomas von, Deesis und Eschatologie, Byzantinische Forschungen, vol. 2, 1967, pp. 59-72.

Bombognini Francesco, Antiquario della Diocesi di Milano, Milano 1856 (1790).

Bossi Maria Antonietta, La chiesa del monastero di Torba (ora cascinale), Rassegna Gallaratese di Storia e d’Arte, vol. 20, n. 79, December 1961, p. 192-194.

Boyle Leonard E., Medieval Latin Palaeography, A Bibliographical Introduction, Toronto – Buffalo – London 1984.

Braga Gabriella, Testimonianze di vita monastica italiana fra nord e sud nell’VIII secolo, in: Giovanni Spinelli Osb (ed.), Il monachesimo italiano dall’età longobarda all’età ottoniana (secc. VIII – X), Cesena 2006, pp. 509-534.

Bréhier Louis, L´art chrétien, Paris 1928.

Brenk Beat, Zur Bildprogramm der Zenokapelle in Rom, in: Homenaje al Prof. Helmut Schlunk, Madrid 1975, pp. 213-222.

Brogiolo Gian Pietro, Per una storia religiosa di Castelseprio: il complesso di Torba e la chiesa di S. Maria foris portas, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: Sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 213-222.

Brogiolo Gian Pietro – Chavarria Arnau Alexandra – Giacomello Federico, Marinato Maurizio, Scavi e Ricerche 2013-2014 nel complesso di Torba (VA), in: Paul Arthur, Marco Leo Imperiale (edd.), VII congresso nazionale di archeologia medievale, vol. 2, Firenze 2015, pp. 34-37.

Bullough Donald, The Age of Charlemagne, Bergamo 1965.

Cantone Valentina, Ars Monastica, Iconografia teofanica e tradizione mistica nel mediterraneo altomedievale (V-XI secolo), Padova 2008.

87

Capponi Carlo (ed.), L’ altare d’oro di Sant’Ambrogio, Milano 1996.

Carver Martin Oswald Hugh, S. Maria Foris Portas at Castel Seprio: A Famous Church in a New Context, World Archaeology, vol. 18, n. 3, 1987, pp. 312-329.

Castelfranchi Vegas Liana, L’ arte medioevale in Italia e nell’Occidente europeo, Milano 1993.

Cavadini John C., Alcuin and St Augustine’s “De Trinitate”, Augustinian Studies, vol. 12, 1981, pp. 11-18.

Cavallo Guglielmo, Libri e cultura nelle due Italie longobarde, in: Carlo Bertelli – Gian Pietro Brogiolo (edd.), Il futuro dei Longobardi, L’Italia e la costruzione dell’Europa di Carlo Magno, Milano 2000, pp. 85-103.

Chizzoni Enzo - Vassalli Giorgio, Il Monastero di Torba, Tracce, vol. 8, n. 2, aprile 1987, pp. 95-102.

Clemena Antonova, Space, Time, and Presence in the Icon, Farnham 2010.

Christe Yves, A propos des peintures d’Esterri de Cardós et de Sainte Eulalie d’Estahón, Les Cahiers de Saint-Michel de Cuxa, vol. 14, 1983, pp. 59-67.

Christe Yves, The Apocalypse in the Monumental Art of the Eleventh Through Thirteenth Centuries, in: Richard Kenneth Emmerson (ed.), The Apocalypse in the Middle Ages, Ithaca 1992, pp. 234-258.

Christe Yves, Une 'Maiestas Domini' de type Lombard, Arte Lombarda, n. 102-103, 1992a, pp. 5-13.

Christe Yves, L’ Apocalypse de Jean: sens et développements de ses visions synthétiques, Paris 1996.

Christe Yves, Jugements derniers, Saint-Léger-Vauban 1999.

Colombo Silvano, Finalmente Torba, Lombardia nord/ovest, vol. 50, n. 4, 1978, pp. 23-25.

Colombo Silvano, Conoscere Castelseprio, Torba e dintorni, Varese 1999 [1995].

Cormack Robin S., The Mosaic Decoration of S. Demetrios, Thessaloniki: A Re-Examination in the Light of the Drawings of W. S. George, The Annual of the British School at Athens, vol. 64, 1969, pp. 17-52.

Corrigan Kathleen, The Ivory Scepter of Leo VI: A Statement of Post-Iconoclastic Imperial Ideology, The Art Bulletin, vol. 60, 1978, pp. 407-416.

Coyle J. Kevin, The Fathers on Women and Women’s Ordination, in: David M. Scholer (ed.), Women in Early Christianity, New York – London 1993 [1978] pp. 117-168.

Crivello Fabrizio - Denoël Charlotte - Orth Peter, Das Godescalc-Evangelistar, Darmstadt 2013.

Curtius Ernst Robert, European Literature and the Latin Middle Ages, New York 1953 [1948].

88

Cutler Anthony, Under the Sign of the Deēsis: On the Question of Representativeness in Medieval Art and Literature, Dumbarton Oaks Papers, Studies on Art and Archeology in Honor of Ernst Kitzinger on His Seventy-Fifth Birthday, vol. 41, 1987, pp. 145-154.

Darby Peter, Bede and the End of TIme, Farnham – Burlington 2012.

De Capitani d’Arzago Alberto, Gli Affreschi di S. Maria di Castelseprio, in: Gian Piero Bognetti – Gino Chierici - Alberto De Capitani d'Arzago, Santa Maria di Castelseprio, Milano 1948, pp. 537-711.

Degering Hermann, Die Schrift, Atlas der Schriftformen des Abendlandes vom Altertum bis zum Ausgang des 18. Jahrhunderts, Berlin 1929.

Dejana Alessandro, Gli scavi nella chiesa di Torba, in: Atti del 1° Convegno Archeologico Regionale, Brescia 1981, pp. 317-326.

Delahaye Gilbert-Robert, Le décor du sarcophage d’Agilbert dans les cryptes de Jouarre, Paris et Ile-de-France, vol. 58, 2007, pp. 27-60.

Delius Walter, Geschichte der Marienverehrung, München – Basel 1963.

De Marchi Paola Marina, Castelseprio e il suo territorio in etá longobarda e carolingia, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: Sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 15-44.

De Rubeis Flavia, Le iscrizioni della Torre del monastero di Santa Maria di Torba: in problema paleografico, in: Carmen Benati (ed.), Castel Seprio e Vico Seprio, Castelseprio 2002, pp. 83- 100.

Dinkler Erich, Das Apsismosaik von S. Apollinare in Classe, Köln 1964.

Duchesne Louis (ed.), Le liber pontificalis, vol. 2, Paris 1892.

Esbroeck Michel van, The Virgin as the true Ark of the Convent, in: Maria Vassilaki (ed.), Images of the Mother of God, Aldershot – Burlington 2005, pp. 63-68.

E. T. [Eva Tea], Operazione salvataggio, l’ attività dell’“Oasi“ di Varese, Arte Cristiana, vol. 52, 1964, pp. 211-222.

Filipová Alžběta, Santo, vescovo e confessore. L'immagine di Apollinare nei mosaici di Classe, in: Nicolas Bock - Ivan Foletti - Michele Tomasi (edd.), L’éveque, l’image et la mort. Identité et mémoire au Moyen Âge, Roma 2014, pp. 431-444.

Fiorio Tedone Cinzia, Dati e riflessioni sulle tombe altomedievali internamente intonacate e dipinte rinvenute a Milano e in Italia settentrionale, Atti del 10 Congresso Internazionale di Studi sull’Alto Medioevo, Spoleto 1986, pp. 403-428.

Fisher Sally, The Square Halo and other mysteries of Western art, New York 1995.

Fitzgerald Allan D., Augustine Through the Ages: An Encyclopedia, Grand Rapids 1999.

Foletti Ivan – Gianandrea Manuela, Zona liminare, Il nartece di Santa Sabina a Roma, la sua porta e l’iniziazione cristiana, Brno 2015.

89

Foletti Ivan - Giesser Valentine, Il IX secolo: da Pasquale I (817-824) a Stefano V (885-891), in: Mario D’Onofrio (ed.), Le committenze pontificie a Roma nel Medioevo, Roma 2016 (in press), pp. 219-237.

Foxhall Forbes Helen, Diuiduntur in Quattuor: The Interim and Judgement in Anglo-Saxon England, The Journal of Theological Studies, vol. 61, n. 2, 2010, pp. 659-684.

Fournée Jean, Le jugement dernier: essai d’exégèse d’une oeuvre d’art: Le vitrail de la Cathédrale de Coutances, Paris 1964.

Frati Vasco et al., San Salvatore di Brescia, Materiali per un museo, vol. 1, Brescia 1978.

Fredriksen Paula, The Confessions as Autobiography, in: Mark Vessey (ed.), A Companion to Augustine, Chichester 2012, pp. 87-98.

Fuhrer Therese, Conversationalist and Consultant: Augustine in Dialogue, in: Mark Vessey (ed.), A Companion to Augustine, Chichester 2012, pp. 270-283.

Gatti Evan A. In a Space Between: Warmund of Ivrea and the Problem of (Italian) Ottonian Art, Peregrination, vol. 3, n. 1, 2010, pp. 8-48.

Genoud Jean-Claude – Bertelli Carlo, Torba, Etudes de lettres, Revue dela Faculté des lettres de l’Université de Lausanne, vol. 2-3, n. 213, 1987, pp. 67-83.

Gheroldi Vincenzo, I rivestimenti aniconici e i dipinti murali della Torre del Monastero femminile benedettino di Torba, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 293-310.

Gilsdorf Sean, Deēsis Deconstructed: Imagining Intercession in the Medieval West, Viator, vol. 43, n. 1, 2012, pp. 131-173.

Giulini Giorgio, Memorie spettanti alla storia, al governo ed alla descrizione della città e campagna di Milano, ne'secoli bassi, vol. 4, Milano 1855 (1760).

Goll Jürg - Exner Matthias - Hirsch Susanne, Müstair, die mittelalterlichen Wandbilder in der Klosterkirche, München 2007.

Goodson Caroline J., The Rome of Pope Paschal I. Papal Power, Urban Renovation, Church Rebuilding and Relic Translation, 817-824, Cambridge 2010.

Hallenbeck Jan T., Pavia and Rome: The Lombard Monarchy and the Papacy in the Eight Century, Philadelphia 1982.

Hanselmann Juliette, I frammenti di affreschi di Torba: ricomposizione e interpretazione, Sibrium, vol. 19, 1987-1988, pp. 165-172.

Hodges Richard, Light in the Dark Ages, The Rise and Fall of San Vincenzo al Volturno, London 1997.

Hubert Jean – Porcher Jean - Volbach Wolfgang Fritz, L’Europa delle invasioni barbariche, Milano 1980.

90

Ibsen Monica, Arredo liturgico da Castelseprio e dipinti murali da S. Maria di Torba. Scavi 2009, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 423-431.

Indelli Tomasso, Langobardia, I Longobardi in Italia, Padova 2013.

Ivanovici Vladimir, Windows and Church Space in Early Medieval Byzantium and West, Opuscula historiae artium, vol. 62, 2013, Supplementum, pp. 38-47.

Jeffrey David Lyle (ed.), A Dictionary of Biblical Tradition in English Literature, Grand Rapids 1992.

Johnson Stephen, Late Roman Fortifications, Totowa 1983.

Kalavrezou Ioli, Triptych Icon with the Deesis and Saints, in: Helen C. Evans – William D. Wixom (edd.), The Glory of Byzantium, New York 1997.

Kantorowicz Ernst, Ivories and Litanies, Journal of the Warburg and Courtauld Institutes, vol. 5, 1942, pp. 56-81.

Kessler Herbert L., On the State of Medieval Art History, The Art Bulletin, vol. 70, n. 2, 1988, pp. 166-187.

Kiilerich Bente, The rhetoric of materials in the Tempietto Longobardo at Cividale, Valentino Pace (ed.), L’ VIII secolo, un secolo inquieto; atti del convegno internazionale di studi, Cividale del Friuli 2010, pp. 93-102.

Kitzinger Ernst, The Cult of Images in the Age before Iconoclasm, Dumbarton Oaks Papers, vol. 8, 1954, pp. 83-150.

Kitzinger Ernst, Byzantine art in the period between Justinian and iconoclasm, in: Berichte zum XI. Internationalen Byzantinisten-Kongress, München 1958, pp. 1-50 (IV, 1).

Klein Peter K., Introduction: The Apocalypse in Medieval Art, in: Richard K. Emmerson – Bernard McGinn, The Apocalypse in the Middle Ages, Ithaca – London 1992, pp. 159-199.

Klein Peter K., L’Apocalypse de Trèves, in: Valentino Pace (ed.), Le Jugement dernier entre Orient et Occident, Milan 2006, pp. 37-39.

Kloos Rudolf M. Neue Inschriftenausgaben und -untersuchungen, Deutsches Archiv für Erforschung des Mittelalters, vol. 33, 1977, pp. 570-588.

Kloos Rudolf M., Die Inschriften im 1. Stock des Turmbaues von Torba, in: Atti del 6. Congresso internazionale di studi sull'alto Medioevo, Milano, 21-25 ottobre 1978, vol. 1, Spoleto 1980, pp. 219-224.

Kondakov Nikodim Pavlovich, Geschichte der Denkmäler des byzantinischen Emails, Frankfurt am Main 1892.

Kühnel Bianca, The End of Time in the Order of Things, Regensburg 2003.

Langé Santino, Preliminari al complesso monumentale fortificato della Torba (Castelseprio), Castellum, vol. 5, 1967, pp. 33-48.

91

Langé Santino, L’inserimento del monastero fortificato di Torba nel dibattito di Castelseprio, Rassegna Gallaratese di Storia e d’Arte, vol. 32, n. 119, 1973, pp. 75-79.

Leveto Paula D., Castel Seprio: Architecture and Painting (Dissertation thesis), Indiana University, 1985.

Leveto Paula D., The Historiography of North Italian Frescoes Dating from 800 to 1000 A.D., with a Special Focus on Castelseprio: Chaos, Continuity, or Crisis?, in: Sandra Barberi (ed.), Medioevo aostano, 1. Atti del convegno internazionale (Aosta, 15-16 maggio 1992), Torino 2000, pp. 247-256.

Lidov Aleksej M. (ed.), Hierotopy - studies in the making of sacred spaces, Moskva 2004.

Lomartire Saverio, Torba, Monastero di Santa Maria, Torre, in: Mina Gregori (ed.), Pittura tra Tiino e Olona, Varese e la Lombardia nord-occidentale, Milano 1992, pp. 215-216.

Lomartire Saverio, La pittura medievale in Lombardia, in: Carlo Bertelli (ed.) La pittura in Italia, L’Altomedioevo, Milano 1994, pp. 47-89.

Lomartire Saverio, Riflessioni sulla decorazione del San Salvatore di Brescia alla luce delle nuove indagini archeologiche, in: Matthias Exner (ed.), Wandmalerei des frühen Mittelalters, München 1998, pp. 40-48.

Lucey Stephen J., Art and Socio-Cultural Identity in Early Medieval Rome: The Patrons of Santa Maria Antiqua, in: Éamonn ó Carragáin - Carol Neuman de Vegvar (edd.), Roma Felix, Aldershot 2007, pp. 139-158.

MacCormack Sabine, Augustine on Scripture and the Trinity, in: Mark Vessey (ed.), A Companion to Augustine, Chichester 2012, pp. 398-415.

Mackie Gillian Vallance, The San Zeno chapel, a prayer for Salvation, Papers of the British School at Rome, vol. 57, 1989, pp. 172-199.

Mackie Gillian Vallance, Abstract and vegetal design in the San Zeno chapel, Rome: the ornamental setting of an early medieval funeraly programme, Papers of the British School at Rome, vol. 63, 1995, pp. 159-182.

Magistretti Marco - Monneret de Villard Ugo (edd.), Liber notitiae sanctorum Mediolani: manoscritto della Biblioteca capitolare di Milano, Milano 1917.

Magnifico Marco, Presentazione, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: sintesi delle ricerche e aggiornamenti, Mantova 2013, p. 12.

Manaresi Cesare (ed.), Regesto di S. Maria di Monte Velate, sino all’anno 1200, Roma 1937.

Marinato Maurizio - Canci Alessandro, I cimiteri di S. Maria di Torba e della chiesa pievana di S. Giovanni a Castelseprio: i risultati delle analisi antropologiche, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 173- 181.

Matthiae Guglielmo, Mosaici medioevali delle chiese di Roma, vol. 1, Rome 1967.

92

Mattke Christiane, Mémoire et devoirs d’un évêque, l’ iconographie du ’codex’ d’Éginon, in: Léon Pressouyre (ed.), Utilis est lapis in structura, Paris 2000, pp.215-230.

Mazza Sandro, Il complesso fortificato di Torba, Sibrium, vol. 14, 1978-1979, pp. 187-215.

McClendon Charles B., The Imperial Abbey of Farfa, New Haven – London 1987.

Mecklenburg Michael – Mertens Thom, Introduction: The Last Judgement in Medieval Sermons, in: Mertens Thom, Sherwood-Smirth Maria, Mecklenburg Michael, Schiewer Hans- Jochen, The Last Judgement in Medieval Preaching, Turnhout 2013, pp. IX-XXIV.

Melchior-Bonnet Sabine, The Mirror, A History, New York 2001 [1994].

Meyvaert Paul, Bede and the church paintings at Wearmouth-Jarrow, Anglo-Saxon England, vol. 8, 1979, pp. 63 – 77.

Mirabella Roberti Mario, Le mura di Castelseprio, Rassegna Gallaratese di Storia e d’Arte, vol. 32, n. 119, 1973, pp. 57-64.

Mitchell John, The Painted Decoration of the Early Medieval monastery, in: Richard Hodges – John Mitchell (edd.) San Vincenzo al Volturno, The Archaeology, Art and Territory of an Early Medieval monastery, Oxford 1985, pp. 125-176.

Mitchell John – Leal Bea, Wall pantings in S. Maria foris portas (Castelseprio) and the tower at Torba. Reflections and reappraisal, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 311-345.

Mitchell John, The Crypt reappraised, in: Federico Marazzi (ed.), La cripta dell’abate Epifanio a San Vincenzo al Volturno, Cerro al Volturno 2013 (1993), pp. 409-467.

Mitchell John, The Painted Decoration of San Salvatore di Brescia in Context, in: Gian Pietro Brogiolo - Francesca Morandini (edd.), Dalla corte regia al monastero di San Salvatore - Santa Giulia di Brescia, Mantova 2014, pp. 169-201.

Molinier Auguste, Catalogue des manuscrits de la Bibliothèque Mazarine, Paris 1885.

Mouriki Doula, A Deësis Icon in the Art Museum, Record of the Art Museum, Princeton University, v. 27, n. 1, 1968, pp. 13-28.

Mütherich Florentine, Die Erneuerung der Buchmalerei am Hof Karls des Großen, in: Florentine Mütherich (ed.), Studies in Carolingian Manuscript Illumination, London 2004, pp. 1-61.

Nilgen Ursula, Die grosse Reliquieninschrift von Santa Prassede: eine Quellenkritische Untersuchung zur Zeno-Kapelle, Römische Quartalschrift für christliche Altertumskunde und Kirchengeschichte, vol. 69, 1974, pp. 7-29.

Oakes Catherine, Ora Pro Nobis, The Virgin as Intercessor in Medieval Art and Devotion, London 2008.

Oltrona Visconti Gian Domenico, Il monastero di Torba (note per una cronologia), Rassegna Gallaratese di Storia e d’Arte, vol. 19, n. 75, December 1960, pp. 175-177.

93

Orofino Giulia, Da Montecassino a Nonantola, la tradizione illustrativa delle Institutiones di Cassiodoro, in: Giovanni Spinelli Osb, Il monachesimo italiano dall’età longobarda all’età ottoniana (secc. VIII - X), Cesena 2006, pp. 553-607.

Osborne John, The Portrait of Pope Leo IV in San Clemente, Rome: A Re-Examination of the So-Called'Square' Nimbus, in Medieval Art, Papers of the British School at Rome, vol. 47, 1979, pp. 58-65.

Osborne John – Brandt J. Rasmus – Morganti Giuseppe (edd.), Santa Maria Antiqua al Foro Romano, Roma 2004.

Pace Valentino (ed.), L’ VIII secolo, un secolo inquieto; atti del convegno internazionale di studi, Cividale del Friuli 2010.

Palma Marco, Alle origini del “tipo di Nonantola”: nuove testimonianze meridionali, Scrittura e civiltà, vol. 7, 1983, pp. 141-149.

Panin Yula, Documenti inediti quatrocenteschi per Torba, Arte Lombarda, vol. 64, 1983, pp. 93-98.

Partner Peter, The Lands of St. Peter, London 1972.

Pedroli Alberto, Castelseprio settecento anni dopo: un convegno ed una nuova guida, Tracce, vol. 8, n. 4, October 1987, pp. 289-290.

Pentcheva Bissera V., Icons and Power, the Mother of God in Byzantium, University Park, Penn. 2006.

Pentcheva Bissera V., Mirror, Inspiration, and the Making of Art in Byzantium, Convivium, vol. 1, n. 2, 2014, pp. 10-39.

Perogalli Carlo - Bascapè Giacomo C., Castelli della pianura lombarda, Milano 1960.

Peroni Adriano, Spunti per un aggiornamento delle discussion sugli affreschi di S. Maria di Castelseprio, Rassegna Gallaratese di Storia e d’Arte, vol. 32, n. 119, 1973, pp. 19-27.

Peroni Adriano, O svjazi fresok cerkvi Santa Marija v Kastel’seprio i baziliki San Sal’vatore v Brešii, in: Viktor Nikolaevič Graščenkov (ed.), Vizantija južnye Slavjane i drevnjaja Rus, Zapadnaja Evropa, Moskva 1973a, pp. 375-387.

Peroni Adriano, L’arte nell’etá longobarda, in: Giovanni Pugliese Carratelli (ed.), Magistra Barbaritas, I Barbari in Italia, Milano 1984, pp. 229-297.

Petrikovits Harald von, Fortifications in the North-Western Roman Empire from the Third to the Fifth Centuries A.D., The Journal of Roman Studies, vol. 61, 1971, pp. 178-218.

Piano Natacha La Maiestas Mariae, Diffusione e sviluppo di un tema iconografico nell’Occidente medievale (IX-XIII secolo), Arte Cristiana, vol. 91, 2003, pp. 29-42.

Pirochtová Antonína, Kaple San Zenone v bazilice Santa Prassede v Římě, MA thesis, Masaryk University, Brno 2012.

Poilpré Anne-Orange, Maiestas Domini, Une image de l’Église en Occident, Paris 2005.

94

Pollack Tamara, Light and Mirror in Dante’s Paradiso, Ann Arbor 2008.

Quasten Johannes, Patrology, vol. 1, The Beginnings of Patristic Literature, Utrecht – Antwerp 1962.

Quasten Johannes, Patrology, vol. 2, The Ante-Nicene literature after Irenaeus, Utrecht – Antwerp 1962a.

Rossi Ismaele, La chiesa di San Maurizio in Milano, Milano 1914.

Rossi Marco, La tradizione dell’antico nella pittura altomedievale del territorio del Seprio, in: Arturo Carlo Quintavalle (ed.), Medioevo: il tempo degli antichi, Milano 2006, pp. 257-266.

Rossi Marco, Il problema Castelseprio, in: Valentino Pace (ed.), L’ VIII secolo, un secolo inquieto; atti del convegno internazionale di studi, Cividale del Friuli 2010, pp. 131-137.

Rossi Marco, Castelseprio nell’alto Medioevo, in: Maria Luisa Gatti Peter (ed.), Storia dell’Arte a Varese e nel suo territorio, Varese 2011, pp. 8-49.

Rotondi Giuseppe, Un passo di Galvano Flamma e il monaster di Torba, Archivio storico lombardo, vol. 49, 1922, pp. 119-134.

Rotondi Secchi Tarugi Luisa, Il baluardo difensivo di Castelseprio verso Torba, Rassegna Gallaratese di Storia e d’Arte, vol. 32, n. 119, 1973, pp. 81-88.

Rugo Pietro, Le iscrizioni dei secoli VI, VII, VIII esistenti in Italia, vol. 2, Venezia e Istria, Padova 1975.

Rutishauser Hans, Das Benediktinerkloster St. Johann in Müstair als Zeuge karolingischer Kulturpolitik, in: Karl der Grosse und Europa, Frankfurt am Main 2004, pp. 55-66.

Saak Eric L., Augustine in the Western Middle Ages to the Reformation, in: Mark Vessey (ed.), A Companion to Augustine, Chichester 2012, pp. 465-477.

Schutz Herbert, The Carolingians in Central Europe, their History, Arts, and Architecture, Leiden – Boston 2004.

Scillia Angela, Analisi stratigrafica degli alzati, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 93-123.

Sefton David S., Pope Hadrian I and the Fall of the Kingdom of the Lombards, The Catholic Historical Review, vol. 65, n. 2, 1979, pp. 206-220.

Sironi Pier Giuseppe, Le difese di Castelseprio nel XIII secolo, Rassegna Gallaratese di Storia e d’Arte, vol. 32, n. 119, 1973, pp. 67-73.

Sironi Pier Giuseppe (ed.), Nuova guida di Castel Seprio¸ Tradate 1979.

Sironi Pier Giuseppe, Nuove osservazioni e considerazioni su alcuni monumenti di Castel Seprio, in: Atti del 6. Congresso internazionale di studi sull'alto Medioevo, Milano, 21-25 ottobre 1978, vol. 2, Spoleto 1980, pp. 659-664.

95

Sironi Pier Giuseppe, Le pitture di Torba, in: Pier Giuseppe Sironi (ed.), Castel Seprio, Storia e Monumenti, Colombo – Tradate 1987, pp. 143-151.

Sironi Pier Giuseppe, La situazione viaria nel Seprio fra tardoantico e medioevo, Sibrium, vol. 19, 1987-1988, pp. 35-52.

Sironi Pier Giuseppe, Vico Seprio, Castel Seprio,Torba: situazione attuale e problemi aperti, in: Carmen Benati (ed.), Castel Seprio e Vico Seprio, Castelseprio 2002, pp. 9-16.

Snyder James, The Reconstruction of an Early Christian Cycle of Illustrations for the Book of Revelation - the Trier Apocalypse, Vigiliae Christianae, vol. 18, 1964, pp. 146-162.

Spieser Jean-Michel, Thessalonique et ses monuments du IVe au Vie siècle, Paris 1984.

Steinhauser Kenneth B., The Apocalypse Commentary of Tyconius, A History of Its Reception and Influence, Frankfurt am Main 1987.

Stradiotti Renata (ed.), San Salvatore - Santa Giulia a Brescia, Ginevra 2001.

Tamborini Marco, Iconografia ottocentesca del complesso di Torba e descrizioni coeve di Castelseprio, Libri & documenti, vol. 13, n. 1, 1987, pp. 37-42.

Tatum George Bishop, The Paliotto of Sant' Ambrogio at Milan, The Art Bulletin, vol. 26, n. 1, 1944, pp. 25-45.

TeSelle Eugene, Augustine the Theologian, Eugene 2002 [1970].

Teteriatnikov Natalia, An interpretation of the program of the private chapel in S. Maria Antiqua, Cahiers archéologiques, vol. 41, 1993, pp. 37-46.

Thierry Nicole, A propos des peintures d’Ayvali köy (Cappadoce), Zograf, vol. 5, 1974, pp. 5- 22.

Thunø Erik, The apse mosaic in early medieval Rome, time, network, and repetition, New York 2015.

Tibiletti Thea, Testimonianze letterarie e indagini archeologiche a Castelseprio. Inerpretazioni, problemi, spunti di riflessione, in: Paola Marina De Marchi (ed.), Castelseprio e Torba: Sintesi delle ricerche e aggiornamenti, Mantova 2013, pp. 45-86.

Torp Hjalmar, Il tempietto longobardo, Cividale del Friuli 2006.

Turchi Nicola, Il gesto d’intercessione nell’arte religiosa bizantina, Colloqui del Sodalizio, vol. 1, 1951, pp. 46-50.

Valagussa Giovanni, Dagli inizi della pittura al Duecento, in: Mina Gregori (ed.), Pittura a Milano dall’Alto Medioevo al Tardogotico, Milano 1997, pp. 31-42.

Villa Ettore Tito, Le monache di Torba e I Castiglioni di Casciago, Archivio storico lombardo, vol. 103, 1979, pp. 303-330.

Walter Christopher, Two Notes on the Deēsis, Revue des études byzantines, vol. 26, 1968, pp. 311-336.

96

Walter Christopher, Les dessins carolingiens dans un manuscrit de Verceil, Cahiers archéologiques, vol. 18, 1968a, pp. 99-107.

Walter Christopher, Further Notes on the Deēsis, Revue des études byzantines, vol. 28, 1970, pp. 161-187.

Walter Christopher, The Origins of the Iconostasis, Eastern Churches Review, vol. 3, n. 3, 1971, pp. 251-267.

Walter Christopher, Bulletin on the Deësis and the Paraclesis, Revue des études byzantines, vol. 38, 1980, pp. 261-269.

Weidmann Clemens, Augustine’s Works in Circulation, in: Mark Vessey (ed.), A Companion to Augustine, Chichester 2012, pp. 431-449.

Weitzmann Kurt, The Fresco Cycle of S. Maria di Castelseprio, Princeton 1951.

Williams John, The Illustrated Beatus, vol. 2, The Ninth and Tenth Centuries, London 1994.

Wirenfeldt Asmussen Marianne, The chapel of S. Zeno in S. Prassede in Rome: New aspects on the iconography, Analecta Romana Instituti Danici, vol. 25, 1986, pp. 67-87.

Wisskirchen Rotraut, Die Mosaiken der Kirche Santa Prassede, in Rom, Mainz am Rhein 1992.

Wright David H., 1980 Sources of Longobard Wall Painting: Facts and Possibilities, in: Atti del 6. Congresso internazionale di studi sull'alto Medioevo, Milano, 21-25 ottobre 1978, vol. 2, Spoleto 1980, pp. 727-739.

Wyss Alfred, Restaurierungsgeschichte der Wandmalereien in der Klosterkirche bis 1960, in: Alfred Wyss - Hans Rutishauser – Marc Antoni Nay, Die mittelalterlichen Wandmalereien im Kloster Müstair, Zürich 2002, pp. 51-61.

Zagni Luisa (ed.), Le pergamene della Basilica di S. Vittore di Varese (899-1202), Milano 1992.

97

Illustrations

98