188 ‘AND SHALL THESE MUTE STONES SPEAK?’ Text as Art Liz James read and understood. What happens, however, when words are considered purely as visual signs and considered in terms of their visuality, rather than what they say? There are several elements to this approach. First, were the words written on objects always meant to be read, and if so by whom? Indeed, did patrons always automatically expect that the texts that they commissioned to be placed on works of art would be read by everyone who saw them? How were the words written on works of art perceived by the illiterate viewer? Second, in what ways might these words be seen when they are considered simply in terms of theit visual appearance and effect, with no thought to their verbal meaning — that is to say, when a formalist reading is applied to them? How do words appear when they are viewed as material objects in a physical context, rather than as texts to be read? How do they function as ornament? To explore these questions, | shall focus on three examples of inscriptions carved onto the facades of churches in Byzantine Constantinople: St. Polyeuktos, the church of Constantine Lips and the pat- ekklesion or funerary chapel added to the church of the Virgin Pammakaristos. The vast church of St. Polyeuktos was buile in the sixth century by the noblewoman Anicia Juliana. Around the walls of that church were carved the words of a seventy-six-line epigram, extolling her lineage and piety.! The text of the epigram was preserved in the Greek Anthology, where, additionally, a marginal manuscript note states that the first forty-one lines of the poem were carved around the nave of the church whilst the rest were in the narthex and courtyard. ‘The actual site of the church was rediscovered in the 1960s, with the uncovering, of a marble block bearing words from the epigram; excavations throughout the 1960s uncovered the foundations of the church and some remains of its fixtures | n discussions of art and text, it is always assumed that words are there to be‘AND SHALL THESE MUTE STONES SPEAK?’ and fittings. Enough of the carved epigram was uncarthed to enable its position within the church itself to be reconstructed.” Inside the church, the epigram had formed an elaborate marble entablarure running around the interior of the building. Surviving fragments of niches, arches and comer pieces showed that sculpted above the letters was a seemingly continuous, sinuous vine-stalk design, with leaves and bunches of grapes, and considerable depth of carving and undercutting, Below the vine, a double border of simple lines with a space between them marked out che epigram, carved in letters eleven centimetres high, contained between unadorned continuous lines (see Fig. 43). Beneath, in the niches, were peacocks, shown frontally, their spreading tails sculpted on the niche surface and their bodies carved in the round. Some traces of bright blue pigment remaining on the background of one of the arches indicated that the frieze was originally coloured.’ Excavation of the epigram in situ enabled Martin Harrison to reconstruct the length of the building in proportion to the length of the epigram. The larger piers and their capitals were some five or six metres in height, thus placing the entab- lature at perhaps six or seven metres above floor level. The text appears to have begun at the east end of the south aisle, next to the apse, to have run down the south wall (lines 1-21), up the north wall (lines 22-41), and to have ended at the north-east corner of the building, next to the apse again. No part was on the west wall. The epigram is effectively a paean of praise to Anicia Juliana; indeed, Mary Whitby has suggested that it can be read within the traditions of the basilikos logos, the speech made specifically in praise of an emperor, cataloguing his virtues.* ‘Thus it opens with praise of Juliana’s family, notably her imperial grandmother, the empress Eudokia. It extols Juliana’s own piety and orthodoxy, both displayed as family traits. It eulogises her Christian charity, commends her building works as immortal and concludes with a prayer for her and her children’s protection by the servants of God, and that her family’s glory may survive for eternity. ‘The epigram inside the church and the remaining thirty-five lines inscribed outside the church have been considered in terms of what they say about Juliana, her imperial ambitions and her challenge to the emperors Justin J and Justinian.’ What I want to do here is to consider the text as a visual document, rather than as a literary text. Was this inscription really designed to be read by visitors to the church? If so, its location within the building is problematic. The opening line, with its reference to the empress Eudokia, is at the cast end of the church. Consequently, the first lines visible to anyone entering from the west end of the church would be lines 20-21 on the south wall, praising Juliana (‘The whole earth cries out that she has made her parents more glorious’) and lines 22-3 on the north wall (‘Where is it not possible to see that Juliana has raised a fine temple to the saints?’). 189LIZ JAMES 190 43, Part of the inscription, church of St. Polyeuktos, Istanbul. Sixth century. (L. James) Viewers would have been able to follow the epigram on the north side of the church as they moved up the nave towards the altar, but this part of the text is, according to the version preserved in the Greek Anthology, its second half. To read the epigram from its perceived opening line, the visitor would have been com- pelled to enter the building and move immediately to the east end. Indeed, entering at the west end and seeing the inscription, visitors to the church would logically follow the text along the north wall first and then down the south wall, ending on lines 20-21. The epigram would still make sense if read that way, but the Greek Anthology’s version suggests that it was not meant to be. Does its location in the church indicate that line 21 was seen as an appro- priate beginning (and, therefore, that this is a text with two lives, one on the wall and an alternative on the page), or does it suggest that the visitor to St. Polyeuktos was expected to enter the church and move straight to the east end to read the epigram? This would contrast sharply with the way another sixth-century writer, Paul the Silentiary, pictured the visitor's experience of Hagia Sophia, for he describes a circumambulation of that church beginning at the west and working up to the east end, the climax of his account.‘AND SHALL THESE MUTE STONES SPEAK?’ Perhaps, however, the visitor to the church was not expected to read the epigram. Its distance away from the viewer — perhaps some six metres above ground level — suggests that it would not have been easy to see it in order to read it.” There are no word breaks in the epigram, though there is some punctuation and so readers would have been compelled to move around the church, peering well above their heads, spelling out each letter in turn. Although Amy Papalex- androu has argued convincingly for the performance of inscriptions as part of ritual commemorations, either through the reading of an inscription or its memorisation and subsequent recitation, one wonders how well an inscription such as Anicia Juliana’s would work for that purpose.® The epigram is not written in ‘easy’ Greek: it is a long poem, written metrically and in an elaborate, educated language. In other words, any reader would have to be a very competent reader to make much sense out of it, without labouring round the church picking out words and metre slowly and painfully. This immediately raises the questions of how literate a society sixth-century Byzantium was and of how many visitors to the church might have been expected to decipher the inscription. They are not easy questions to answer but it is dear that high-level literacy in the Byzantine world was increasingly restricted to the well-educated elite. Perhaps from the sixth century on, the gap gradually increased between those able to read and write in an Atticising, literary language and those who possessed a functional literacy, operating in a different linguistic and stylistic register chat was closer to the spoken language.” The high style of Anicia Juliana’s epigram suggests that its intended reading audience was the classically educated, highly literate upper class. Others would need it read to them." These same two points about legibility and reading audience can be made with respect to my other examples of inscriptions on churches from Byzantine Constantinople. In both of these buildings, the height of the inscriptions makes the letters difficult to decipher, The church of the Virgin tou Libos, buile by the court official Constantine Lips in 907, and so known as the church of Constantine Lips, has an inscription which runs round the outside of the building. It is carved on a narrow marble cornice on the exterior of the three apses of the north church and above the top of the ground floor windows (see Fig. 44).'' The letters of the text were originally of lead, inlaid into carved grooves only nine centimetres high (see Fig. 45). Two carved inscriptions survive on the parekklesion built on to the south side of the church of the Virgin Pammakaristos shortly after 1310 by Maria Glabas as her husband Michael’s burial place.'? One runs around the outside and one around the inside of the building, Again, they are set at a considerable height. The exterior inscription is at the level of the string course (see Figs. 46 and 47) and runs the length of the building. In addition, on the south fagade of che 191LIZ JAMES 192 parekklesion at roof level, between the two windows, is an inscription in tiles, recording Michael’s name and recording him as founder of the chapel (visible in Fig. 46). The interior inscription, painted in gold letters on a blue background, runs around the building on a marble cornice at two levels, just below the lower windows and at the springing of the vaults (sce Fig. 48). It starts, as was the case with the epigram in St. Polyeuktos, at the south-cast corner on the lower level; lines 1—5 are on the south wall; 6-10 on the west; 11—15 on the north; it ends at the north-east corner at the point where the iconostasis would have begun. The text then moves up a level into the east arm in front of the entrance to the bema, and follows a continuous route around the church on the arms, finishing in the north arm. This ordering also demands a considerable degree of literacy (and enthusiasm for straining one’s eyes upwards) on the part of the reader in order to work out where to start. In both examples, not only the texts are hard to sce, but they are also difficult to read. About half of the inscription from Constantine Lips survives in situ, Its general meaning is that Constantine dedicated a church to the Mother of God and, in return, asked that she secure him citizenship in heaven. However, Cyril Mango and Ernest Hawkins noted that there seem to be three separate parts to the inscription, separated from each other by a cross, and with the line ends marked appropriately. They suggest that because the poetic metre of the inscription changes part way through (parts one and three are written in iambs and part two in hexameters), the complete inscription might have been three distinct poems, each addressed to a different heavenly patron.'? For the Byzantine reader to pick this up implies at least a familiarity with poetic con- ventions and metres, again something for the educated reader, rather than the functionally literate reader. Similarly, at the Pammakaristos, the tile text recording Michael’s role as founder demands that the reader, endeavouring to see it, disentangle contractions and ligatures to make sense of it. Both inscriptions here were almost certainly composed by the court poet Manuel Philes. The exterior text reads as an acclam- atory or ritual lament for the deceased Michael voiced by his widow; the interior one takes the form of an invocation or prayer for the deceased Michael, addressed to Christ. Again, the nature of Philes’s poetry demands a level of familiarity with the style and conventions of Byzantine court poetry and the allusions it contains. As with the epigram as St. Polyeuktos, so here, the readers of such texts would need a considerable literary education. One of the questions raised at the start of this essay asked what happened to the words written on art objects when the illiterate viewer confronted them. Did those who could not read inscriptions perceive themselves as excluded or‘AND SHALL THESE MUTE STONES SPEAK?’ 44, East end of the north church of Constantine Lips, Istanbul. Tenth century. (L. James) alienated? How far, indeed, were such inscriptions deliberate devices to exclude others? Did spectators, literate and illiterate, ignore the epigrams or did they pethaps simply devise their own readings? The epigrams recorded in the Greek Anthology seem to us to imply a concern with inscriptions as literature, to be copied down and imitated by men of learning. In contrast, however, the eighth-century Parastaseis syntomoi chronikai, a text characterised by its translators as a sort of guidebook to the monuments of Constantinople, suggests the creation of new meanings, irrespective of any pos- sible dedications carved on the statue.'* Chapter 64, for example, describes statue of a man and donkey and offers the interpretation of ‘philosophers’ that this represents the inversion of the natural order.'? Niketas Choniates, however, records the statue as the one of Nikon and Nikander, set up at Actium by 16 Elsewhere, the Parastaseis talks of a statue of Augustus after a battle-field vision. the empress Verina, which it says is more correctly identified as one of Athena.’ Ina similar fashion, Eusbius described a statue of Bellerophon as one of Joshua, an ascription echoed almost a thousand years later by Niketas Choniates.'® Should we take it that these statues were not identified through inscriptions or that Eusebius and Choniates could not or did not read the captions or that they simply saw in accordance with their own times and their own beliefs? In the Parastaseis, 193LIZ JAMES 45. Detail of the inscription from the east end of the north church of Constantine Lips, Istanbul. Tenth century. (L. James) 194 seven philosophers debate the meanings of statues in the Hippodrome.'? This debate, which makes little sense to the modern reader, because what the philo- sophers say about the identity of these statues bears no relevance to anything we might expect to find written on a statue base, suggests again that if there were inscriptions, people did not necessarily read them. In a culture where the meaning of images was relatively fluid and the inscriptions on statues were not always read, what makes us believe that the inscriptions on buildings were always read and, when read, were read ‘correctly’‘AND SHALL THESE MUTE STONES SPEAK?’ 46. Exterior view of the parekklesion, church of the Virgin Pammakaristos, Istanbul. Thirteenth century. (L. James) and ‘accurately’ in our terms, that the memories invoked were memories relating to the person commemorated in the statue or inscription? As Dale Kinney has pointed out, if someone does not know what a statue is of or what date it is, his or her interpretation of it is unfettered.” Viewers seek to explain what they see in the context of what they know or are familiar with. It seems to me, therefore, that it is a legitimate question to ask whether the only, or indeed the primary, function of inscriptions was their legibility. Were these stones designed simply to speak? Irene Bierman has treated public texts as definers of space and boundaries, marking territory as belonging to a particular ruler, faith, language group.” Amy Papalexandrou has shown how the inscrip- tions on the exterior of the church at Skripou potentially served as memorial rexts, to be read and re-read to commemorate and pray for the dead.** Both of these approaches stress the ‘speaking’ role of inscriptions. What, however, if they were simply seen and could not be or were not read? Elsewhere, Rosalind Thomas has hinted that in Classical Greece, certain inscriptions might, through their style, underline the non-written significance of memorials.” John Scheid has suggested that the Roman Arval dedications were meant for display and memorial, rather than for reading. Similarly, Jaé Elsner has noted the monumentality and 195LIZ JAMES 47. Detail of the inscription on the exterior wall of the parekklesion, church of the Virgin Pammakaristos, Istanbul. Thirteenth century. (L. James) inscriptional nature of surviving carvings of the text of the Res Gestae of Augustus, discussing how the monumental nature of writing can turn writing itself into a sign, a visible sign of a permanent empire.* In a Byzantine context, Margaret Mullett has proposed that letter forms themselves were meaningful even when they could not be read, and that the simple existence of the written word was perceived as significant at every level of society.”° In this context, the visibility of written words was significant in itself, without the presence of a literate person to read those words aloud. The power of texts was particularly pronounced in the Byzantine world where the spoken language was not the same as the written.”° As Marc Lauxtermann has pointed out, because most Byzantines were illiterate, they did not read inscriptions but rather gazed at them in literate cultures, the signified meaning is far more important than the sign; in illiterate cultures, the sign matters more.’” Oleg Grabar illustrates a carpet page from a fifteenth-century Islamic manuscript which, to the uninitiated or less literate simply appears decorative, but to those who know the script is a repetition of the name of Ali, the son-in-law of the Prophet.”* Grabar also cites examples of Islamic buildings where, he suggests, the point of the writing is no longer the concrete message of the words but something"AND SHALL THESE MUTE STONES SPEAK?’ 48. Interior view of the parekklesion, church of the Virgin Pammakaristos, Istanbul. Thirteenth century. (L. James) mply the fact of writing itself: the exact meaning of the words can else, pethaps s 9 become secondary to the composition of their parts.” What this suggests is that in the case of these epigrams on church walls, at times, depending on the audience, the words have meaning and their presence is a sign of class and status. At other times, with other viewers, the words on the inscriptions both inside and outside the churches under discussion may stop functioning as words and start functioning as signs. As signs, the words might be perceived in various ways; here, I want to consider two of these possibilities: words as magical signs and words as ornamental signs. It is very clear that in Byzantium, both reading and writing were crucial elements of magical practices and that words and letters carried considerable magical and talismanic powers.*’ Byzantine magic worked above all through words and the manipulation of words and letters, as written sources as far apart as Ammianus Marcellinus’s account of the persecution of magic under Constantius Tl, Valentinian and Valens and fourteenth-century accounts of trials for sorcery make clear.*! Repeating letters in magical formulae and patterns seemingly increased their potency. The spell in a fourth-century papyrus in Leiden repeats the vowels in three columns in different but seemingly senseless combinations.** 197LIZ JAMES 198 Eulogia, clay blessing tokens given out by holy men and often carrying a medicinal significance, were often written on, and that writing was not necessarily for reading but to complement the healing power of the sacred dust.” Letter forms seem to contribute a further level of iconic power. Ie is perhaps for this reason that amulets often include incantations or words with a quasi-scriptural formulation or even arrangements of letters that appear as gibberish and meaningless to us.>* A ring in the Kelsey Museum of Archaeology, Ann Arbor, depicts a Holy Rider with an inscription that runs round the edge of the ring and reads ‘ogyrdkmapeotufzezouzerbe’ (cf. also Fig. 30 in Chap. 7).?° This must have had significance for the wearer, but whether that was a meaning involving reading or simply the presence of letters in a particular order is another matter. The coupling of the letters with the Holy Rider probably provided a formula powerful in its own right.*° To increase the ability to read and write, magical names were written on paper, mixed with consecrated wine and given a slow reader to drink.?” A seventh-century story told about St. Symeon the Holy Fool describes how he rendered a sorceress ineffective by giving her an amulet on which was inscribed a Christian prayer that she, being illiterate, was unable to read, but which she recognised as magical through the presence of the letters.** Only the wise could interpret the magic of letters. In the Old Testament, only the prophet Daniel could read the mysterious writing on the wall of the palace of Nebuchadnezzar. In the cighth-century Parastaseis, ‘philosophers’, or ‘wise men’, regard ancient inscriptions (that it is clear they cannot or do not read) as pre- dicting the future.*? In the Parastaseis, it is often as a result of reading inscriptions or prophecies about statues that the spectator is made aware of the malevolent forces inherent in statues. In one story, as soon as the inscription on a particular statue is read, the statue falls on the head of the reader, killing him; in another, an observer of a statue comments, ‘I would have been better off if I had not read the inscription’.*° Although the Parastaseis has been seen as a text that describes the malevolent power of images, it also treats words and inscriptions as potentially powerful and dangerous; when statues fall on people’s heads in its stories, the reason for their fall is often not as a result of looking at the statue but of reading the inscription. Words possessed power and magic qualities, for initiate and non- initiate alike. They could be positive or negative, depending on their context. There is scope for suggesting that the same magical powers applied to texts written around church walls. If words and lever forms contained power simply in their existence, it is conceivable that such inscriptions served as a visual, rather than read, sign of the efficacy of the image.“ Where words, especially words that included the name of God and his saints, appeared on buildings, this perhaps served to enlist their magical powers in the protection of the building. On city walls, gates and even aqueducts, crosses, inscriptions (and even what we define as‘AND SHALL THESE MUTE STONES SPEAK?! ‘masons’ marks’), often in the form of monograms, are placed at key points, areas of the greatest structural weakness such as keystones of arches, in corners and above gates.“? When words are perceived as possessing a protective function, their appearance on exposed parts of buildings has a purpose in addition to their commemorative role. At St. Polyeuktos, Constantine Lips and the Pammakar- istos, the texts are written at the springing of the first storey above ground level, a structurally vulnerable point of any building. At St. Polyeuktos and at the Pammakaristos, they run the length of the walls inside the church, as if wrapping them in protection; at Lips, the inscription runs the length of the apse, for perhaps the same reason. The presence of words as (magical/powerful) signs, even if their sense is hard to read, is visually very apparent in all three cases. That inscriptions such as those at St. Polyeuktos, Constantine Lips and the Pammakaristos are located in positions on the building that made them difficult to read from floor level suggests another role. That pictures in churches were there not only for the human worshipper but also for God himself seems very clear from Byzantine attitudes to and uses of images. Why should the same not be true of words? Many images in churches bear texts that could not have been legible from the floor of the church; the inscription held by Christ in the narthex panel above the west door of Hagia Sophia is only one such example. It is conceivable that the eternal prayer conveyed in epigrams written on buildings, and indeed other art forms such as icons, was there to be read by God the eternal reader. Texts and prayers could be activated by God without a human orator, just as an image could be without a human viewer. At St. Polyeuktos, the praise offered to Anicia Juliana is also an eternal evocation of her piety, lest we — or God — forget. The inscription at Constantine Lips describes the church as a lavish gift, emphasising the donor's generosity, but stressing his prayer for spiritual salvation. In the Pammakaristos parekklesion, the inscription inside the building prays for Michael’s soul. Another inscription in mosaic runs around the apse, commemorating the founding of the building as a thank-offering to God on the part of Martha the nun (Maria Glabas’s name in religion) and commemorating the deceased Michael, another reminder or con- stant prayer to God, but also a placing of the names of the donors in the most sacred, and safe, part of the church.*? Writing prayers on a church wall was a magical act, for it made those prayers everlasting, messages for God, to remind him eternally of whose monument this was, invoking his readership forever. In the discussion so fat, words and letters have been considered as words and letters — as text. | want now to move to consider them in terms of their visual appearance — as images. The story of St. Symeon and the sorceress underlines the nature of words not as communicators of information but as incomprehensible magical signs whose potency resided in their appearance as, in Lauxtermann’s 199LIZ JAMES 200 formulation, things to be gazed at. The visual appearance or pattern of letters ordered in three columns in the Leiden papyrus, or arranged circling a Holy Rider, or snaking along the walls of St. Polyeuktos may be as significant and powerful as their verbal quality, the more so because more accessible to any viewer. One might draw a comparison between the babble of letters on the ring depicting a Holy Rider in the Kelsey Museum with the fragment from St. Polycuktos shown in Figure 43. This reads ‘oudautededaekas’, less gibberish, perhaps, but the sense of letters blurring together remains the same, Letters and inscriptions do not only engage with language; they are also visual patterns whose power resides in their visual nature in an oral society, as well as in their role as texts.“ Indeed, after c. 1000, Byzantine epigraphy became more ornamental, more calligraphic and less legible than it had been earlier.’> Another example is provided by monograms of people's names on buildings or on seals, for example, where letters are used to create visual patterns, again mystical and only comprehensible to the initiate.“ In addition, seals and monograms translate the Byzantines’ pleasure in puns and word-play to the visual sphere. Lauxtermann cites the example of an Iconoclast inscription put up extolling the cross as the symbol of life. The viewer did not need to be literate to get the message, for the cross-shaped acrostic was picked out in gold letters forming the shape of a cross.*” The message to the general viewer was in the sign, not the words. Inscriptions on buildings may achieve something similar. The power of the written word was present in the visible form of the inscription itself, whose size, materials and colours might create feelings of awe. The letters formed a part of the work of art possessing, visual and ornamental qualities possibly as significant as their verbal ones, and perhaps more accessible to their viewers. If inscriptions such as those at St. Polyeuktos, Constantine Lips and the Pammakaristos church are defined as ‘ornamental’, then the significance of ornament itself needs some consideration.*® Within medieval art history, there is a long tradition of discussing and describing the ornamentation of Western art, leading to an understanding that there is a sense in which ornament completes the object to which it belongs, serving a function that might be practical or aesthetic or both.*? Changes in ornament can indicate the relative importance of different parts of a building, highlighting the focal point of a building, for example, or marking out doors and windows. The type of ornament used may also make points about wealth or political identity on the part of the patron of the building: the use of Italianate or traditional forms in Tudor buildings, for example.” Although the development of rules of ornament was very much a nineteenth- century preoccupation, this is not to say that systems, traditions and hierarchies of ornament did not hold a place within Byzantine art. Rather, these are not‘AND SHALL THESE MUTE STONES SPRAK?! concepts that have been explored within Byzantine art history, where art historians have often been concerned with the figural aspect of Byzantine art, perhaps led by the Byzantines’ own concern to justify figural imagery. Nevertheless, within Byzantine art, there is a great deal of ornament.”! Ornament is not simple decoration. Ernest Gombrich argued that ornament fulfilled three roles: framing the object; filling in gaps; and linking different elements together, all highly practical aspects. Oleg Grabar has further suggested that ornament can be seen as the subject of the design: it is not simply a filler but, rather, can transform the purpose of its carrier.” It is these aspects that are now considered in the three cases used here, At both Constantine Lips and the Pammakaristos, the ornament, including the inscriptions, used on the facades of these buildings serves a very practical end of linking parts of an object together so that the ornament seems to be the natural conclusion of the process of making. The inscriptions are made of thin strips or blocks of white marble that run around the apses of the churches of Constantine Lips and along the long south wall of the parekklesion, serving almost as a visual girdle around these buildings. These cornices serve to link the building visually. They represent solid, tangible architectural features. They do not articulate sur- faces or forms, but rather join the whole form of apse or exterior wall together, establishing a sense of linearity. At Lips, the white marble entablature is a very distinctive visual feature on the brick exterior of the building, visually binding the tripartite apses together, a binding echoed by the horizontal line of dog-toothed bricks immediately below the roof level. The cornice also separates off parts of the structure. At the Pammakaristos, the visual effect is of a defined horizontal band of white forming the sills of the first floor windows, echoed by the placing of large white blocks in the same position below the second floor windows, thus separating the three stories of the building. Inside this church, the same effect is apparent in the positioning of the marble bands of the inscriptions below the windows at first and second storey levels. Ornament is also used to bring life to the blank spaces of buildings. The frieze from St. Polyeuktos, which is the most decorative of the three, combined the twists of the vine with the curves of the peacock tails and the bodies of the peacocks bursting into a three-dimensional plane. The use of varying depths of carving and of colour and possibly gilding would have enhanced the surface, creating patterns of light and shadow. These are features also apparent in the external friezes at Constantine Lips and the Pammakaristos. Decorative brick working, dog-toothing and recessed bricks, increasingly popular in Byzantium, are all apparent at Constantine Lips, sug- gesting a desire to enhance and elaborate external architectural surfaces with patterns of high and low relief, light and shadow.”* The marble inscription, 201LIZ JAMES 202 sharply angled outwards, with its letter-shapes, contributes also to this visual effect. The south wall of the parckklesion at the Pammakaristos appears strikingly symmetrical at first sight, with windows and blind niches balanced all the way up. One break in this apparent symmetry comes on the thitd level, where the repeating pattern of three round-headed arches grouped together is broken by a single round-headed arch, a pointed arch and a circle of decorative brickwork, which does, in fact, balance the circle of brickwork to the right of the adjoining window. The use of blocks of white marble and bands of brickwork throughout the wall surface serves to decorate the surface still further. Just below the level of the roof is the striking brick inscription recording Michael as founder of the building. The letter forms become a part of the ornamental fagade in their own right, in a way similar to the use of Pseudo-Kufic scripts at Hosios Loukas in Phocis.** In both Constantine Lips and the Pammakaristos, the bands of inscription serve to create a sense of order and definition. They are used to articulate levels within the buildings, in contrast to the vertical thrust of the windows and niches and the movement of the circular patterns in the brickwork. Here, and at St. Polyeuktos, they break away and out from the planes of the wall. The letter forms themselves within each inscription form a part of the ornament. The scripts used in each example differ and create differing visual effects. At St. Polyeuktos, the plain lines of the entablatures enclose letters formed of upright and cross- strokes, all verticals and horizontals as far as is possible, with the minimum of curvature. The inscription itself is marked out, framed and set apart, but the angular letters have a decorative quality of their own, with many straight and incisive lines, which contrast with the vine plants above and peacocks below. ‘This differs from the carved entablature from the later sixth-century church of Saints Sergios and Bacchos in Constantinople. Here, very rounded, flowing letters sit in the middle of elaborate ornamentation, egg and dart, deep-drilled acanthus scrolling, beaded rolls and leaf and arrow. At Lips, the effect is harder to be sure of, since the original letters were of metal sunk into the marble, but the letter forms seem more expansive and rounder and contrast with the more upright, tight and elongated letters, with their moments of deep drilling, at the Pammakaristos. One effect of this ornamentation is to enhance the words. Another is to add visual qualities to the words and possibly even to divert the viewer's attention from the words themselves. At St. Polyeuktos, the most dramatic of these examples, spectators would perhaps find their eyes pulled in two directions away from the lettering, towards the elaborate twisting vine scrolls above it and the derailed and, presumably, dramatic peacocks below it. The entablature would bedazzle the eye, with the interlace of the vine scrolling, the undercutting and the contrasts of'AND SHALL THESE MUTE STONES SPEAK?" colour, light and shade. In all three cases, however, architectural space is demarcated by carved ornament; the carved inscriptions serve to create a visual sense of rhythm, balance and order on the wall in areas of the church — the apse, the main wall — where such things mattered. ‘The inscribed word in Byzantium was a powerful medium of communication and commemoration, not just for what it said but for how it looked. It seems to me that there is a case for understanding these letters on walls as silent, unread inscriptions, performing a function in their silence and potential illegibility. In this context, they also fulfil an active ornamental function. The inscriptions on the three churches I have discussed here serve practical functions, filling and linking the architecture of the buildings, emphasising and underlining various aspects of the buildings’ appearance. Their presence also transforms the purpose of the walls that bear them, The texts represent the magic of letters to the non- initiate and convey eternal silent prayers that give the buildings they adorn meaning, These stones need not speak; their expressive powers lie equally in their mute form as in their verbal content. NOTES My grateful thanks to Michelle O'Malley and Amy Papalexandrou for thoughts, comments and discussion (especially about the visibility of the inscriptions considered here!) 1 For the text of che epigram, see the Greet Anthology 1,10. For the identification of the marble blocks with the poem in the Greek Anthology, sec C. Mango and 1. Sevienko, ‘Remains of the church of St. Polyeuktos at Constantinople’, DOP 15 (1961), 243-7. 2 For the excavations, see R. M. Hartison, Excavations at Saraghane in Istanbul (Princeton, NJ: Princeton University Press, 1986), 2 vols, and also M. Harrison, A zemple for Byzantium (London: Harvey Miller, 1989). For the epigram, see C. L. Connor, “The epigram in the church of Hagios Polyeuktos in Constantinople and its Byzantine response’, Byz 69 (1999), 479-527. ‘The arch with line 32. See Plate 95 in Harrison, Temple. Mary Whitby, “The St. Polyeuktos epigram (AP 1.10): a literary perspective’ in S.P. Johnson (cd.), Greek literasure in Late Antiquity: Dynamism, didacticiom, clasicism (Aldershor: Ashgate, 2006), 159-87. For the basiikos logos in the context of empresses, see L. James, Empress and power in early Byzantium (London: Leicester University Press, 2001), Chap. 2. 5 See, for example, C. Milner, “The image of the rightful ruler: Anicia Juliana’s Constantine mosaic in the church of Hagios Polyeuktos’ in P. Magdalino (ed.), New Constantines: Rhythms of imperial renewal in Byzantium, 4th-13th centuries (Aldershot: Ashgate, 1994), 73-81; B. Kiilerich, “The image of Anicia Juliana in the Vienna Dioscurides: Flatery or appropriation of imperial imagery?” Symbolae Osloensis 76 (2001), 169-90. RO 203LIZ JAMES. 13, 14 15 16 18 19 20 21 See R. Macrides and P. Magdalino, ‘The architecture of ekphrasis: Construction and context of Paul the Silentiary’s poem on Hagia Sophia’, BMGS 12 (1988), 47-82. I owe the suggestion of a text with swo lives to Amy Papalexandrou. For thoughts about problems the Byzantines might have had in actually seeing art, see R. $. Nelson, “To say and to see: Ekphrasis and vision in Byzantium’ in Nelson (ed.), Visuality before and beyond she Renaisance (Cambridge: Cambridge University Press, 2000), 143-68, esp. 148-9. A. Papalexandrou, ‘Text in context: Eloquent monuments and the Byzantine beholder’, Word and Image 17 (2001), 259-83. On levels of Byzantine literacy, sce R. Browning, ‘Literacy in the Byzantine world’, BMGS 4 (1978), 39-54, and M. E, Mullett, ‘Writing in early mediaeval Byzantium’ in R. McKitterick (ed.), The uses of literacy in early mediaeval Europe (Cambridge: Cambridge University Press, 1990), 156-85. On this theme, see Papalexandrou, “Text in context’, which makes it clear that archaising qualities in performance are less of a concern than we might think. The same is true for homilies: see, for example, M. Cunningham, ‘Dramatic device or didactic tool? ‘The function of dialogue in Byzantine preaching’ in E. Jeffreys (ed.), Rhetoric in Byzantium (Aldershot: Ashgate: 2003), 101-16. Now known more generally as Fenart Isa Cami. ‘T. Macridy with A. H. S. Megaw, C. Mango and E. J. W. Hawkins, ‘The monastery of Lips (Fenart Isa Camii) at Istanbul’, DOP 18 (1964), 249-318. For the inscription, see the section by Mango and Hawkins, esp. 300-301. Known now as the Fethiye Camii, A. H. S. Megaw, ‘Notes on recent work of the Byzantine Institute in Istanbul’, DOP 17 (1963), 368-71, gives the inscription’s text and location within the church. See H. Belting, C. Mango and D. Mouriki, The mosaics and frescoes of St. Mary Pammakaristos (Fethiye Camii) at Istanbul (Washington, DC: Dumbarton Oaks, 1978), 16, for comments on the inscriptions. Sec Papalexandrou, “Text in context’, 269-71 and 276-7, for discussion of the text and placement of the poem. Mango and Hawkins, ‘The monastery of Lips’, 301 and Plate 1. ‘A. Cameron and J. Herrin (eds. and trans.), Constantinople in the early eighth century: The Parastaseis syntomoi chronikai (Leiden: Brill, 1984). Cameron and Hertin (eds.), Parastaseis, Chap. 64, 140-47 and commentary on 258. De Signis 650. The original of this story is in Plutarch’s Life of Anthony, Chap. 65, though Choniates himself does not mention Plutarch. Cameron and Herrin (eds.), Parastaseis, Chap. 61, 138-9. De Signis 857.17-858.1. Text in Niketas Choniates, Historia, J. L. van Dieten (ed.), Corpus Fontium Historia Byzantinae 11 (Berlin and New York: de Gruyter, 1975). For a discussion of the nature of De Signis as a text, see A. Cutler, ‘The De Signis of Nicetas Choniates: A reappraisal’, American Journal of Archaeology 72 (1968), 113-18. Cutler sees Choniates distinguishing between the people's faith in monuments as wonder-working signs and his own aesthetic appreciation of them, deriding the irrational fear of signs and inanimate objects. Cameron and Herrin (eds.), Parastaseis, Chap. 64, 140-47. D, Kinney, “The horse, the king and the cuckoo: Medieval narrations of the statue of Marcus Aurelius’, Word and Image 18 (2004), 372-98. 1A. Bierman, Writing signs: The Fatimid public text (Berkeley: University of California Press, 198).‘AND SHALL THESE MUTE STONES SPEAK?! 22. Papalexandrou, “Text in context’. 23 R. Thomas, Literacy and orality in Ancient Greece (Cambridge: Cambridge University Press, 1992), 51-6. My thanks to Amy Papalexandrou for this reference. 24 J. Elsner, ‘Inventing imperium: Texts and the propaganda of monuments in Auguscan Rome’ in J. Elsner (ed.), Art and text in Roman culture (Cambridge: Cambridge University Press, 1996), 32-53. 25 Mullett, ‘Writing’, 163, where she discusses Byzantium as a ‘residually oral’ society. 26 As M. McCommick, “Textes, images et iconoclasme dans le cadre des relations entre Byzance et 'Occident Carolingien’ in Testo e immagine nell'alto medievo 41 (1993), 100-101, points out. Also, see the important analysis by Simon Franklin, Writing, society and culture in Early Rus, c. 950-1300 (Cambridge: Cambridge University Press, 2002), Chap. 6. 27 M. D. Lauxtermann, Byzansine poetry from Pisides to Geometres. Texts and contexts, I (Vienna: Osterteichische Akademie der Wissenschaften, 2003), 272-3. 28 Istanbul, Topkapi Serai Museum, Hazine 2152, fol. 9v; O. Grabar, The mediation of ornament (Princeton, NJ: Princeton University Press, 1992), 18 and Plate 3. 29 Grabar, Ornament, esp. 80, 99-100. Grabar’s argument is that in the Muslim world, this is related to the concept of words holding communities together, acting as their mediation between lives and levels of living ~ words of prayer, authority, pleasure. 30 For a useful definition of the term ‘magic’, see R. P. H. Greenfield, ‘A contribution to the study of Palacologan magic’ in H. Maguire (ed.), Byzantine magic (Washington, DC: Dumbarton Oaks, 1995), 118. For a discussion of magic and writing, see Franklin, Writing, society and culture, Chap. 7, 255-74. 31 Sce M. T. Fogen, ‘Balsamon on magic: From Roman secular law to Byzantine canon law’ in Maguire (ed.), Byzantine magic, 99-117, esp. 111-14. 32M. W. Meyer and R. Smith (eds.), Ancient Christian magic: Coptic texts of ritual power (Princeton, NJ: Princeton University Press, 1999), translates a variety of magical spells which share a formulaic use of words, with repetitions of names and phrases, repeated lecters decteasing in number and considered patternings on the written surface, often interwoven with symbols and images. For the Leiden papyrus example, see H. Maguire, The icons of their bodies: Saints and their images in Byzantium (Princeton, NJ: Princeton University Press, 1996), 119-20. 33. Sec J. Russell, The archaeological context of magic in the early Byzantine period’ in Maguire (ed.), Byzantine magic, 41; G. Vikan, ‘Art, medicine and magic in early Byzantium’, DOP 38 (1984), 69-70; Vikan, ‘Byzantine pilgrims’ art’ in L. Safran (ed.), Heaven on earth: Art and the church in Byzantium (University Park: Pennsylvania State University Press, 1998), 229-66, esp. 243 on. 34 The trisagion and ‘Lord, save!” are particularly popular incantations. See J. Russell, “The archaeological context of magic in the early Byzantine period’ in Maguire (ed.), Byzantine magic, 35-50, esp. 37-42. Greenfield, ‘Palacologan magic’, discusses these in a fourceenth- century context, indicating the continued power of such spells. 35. This is described and pictured in Maguire, The icons of their bodies, 123 and Fig. 107. 36 See Maguire’s discussion in Chapter 7 and Fig. 30 in this book. 37 Lauxtermann, Byzantine poetry, 197. 38 Leonsios de Neapolis, Vie de Syméon le Fou, A. J. Festugitre (ed.), (Paris: P, Geuthner, 1974), 96-7, and H. Maguire, ‘Magic and the Christian image’ in Maguire (ed.), Byzantine magic, 62-3. 39 Cameron and Herrin (eds.), Parastascis, e.g., Chap. 24, 84-7; Chap. 61, 138-9; Chap. 64, 140-7, for the debate between philosophers over the meanings of statues in the Hippodrome. 205LIZ JAMES 206 40 4l 42 43. 45 46 47 48 49 50 51 52 53 54 See Cameron and Herrin (eds.), Parastaseis, e.g. Chap. 14, 76-9; Chap. 28, 90-91; Chap. 65, 146-7. On this theme, see Franklin, Wrising, society and culture, 241-8. See, for example, the cases in Amy Papalexandrou’s piece in this volume. Space docs not allow me to develop this idea further in the context of secular buildings, but I am grateful to Jim Crow for discussions about it in che context of city walls and aqueducts. Also see S. Curtié, ‘Design and structural innovation in Byzantine architecture before Hagia Sophia’ in R. Mark and A. $. Gakmak, Hagia Sophia from the age of Justinian to the present (Cambridge: Cambridge University Press, 1992), 16-38, esp. 17-22 on the use of crosses. Lam grateful to Amy Papalexandrou for this reference. For this text, see P. A, Underwood, ‘Notes on the work of the Byzantine Institute in Istanbul: 1954’, DOP 9-10 (1955-6), 298. Papalexandrou, “Text in context’, looks at Byzantine monumental inscriptions in the context of the ‘non-rational’ uses of writing. See A. Kazhdan et al. (eds.), Oxford dictionary of Byzantium (Oxford: Oxford University Press, 1991), vol. 1, ‘Epigraphy’. HL. Maguire, ‘Magic and geometry in Early Christian floor mosaics and textiles’ in W. Hérandner, J. Koder and O, Kresten (eds.), Andrias: Herbert Hunger zum 80. Geburtstag, JOB 44 (1994), 265-74, where he illustrates this with examples from papyti laid out in various ‘ornamental’ ways. Mullect, ‘Writing’, 183, on seals as visual signs Lauxtermann, Byzantine poetry, 162 and 178. For the ornamental atticulation of fagades on Byzantine churches, see the work of Robert Ousterhout, including ‘Observations on the recessed brick technique during the Palacologan period’, Archaiologikon Deltion 39 (1990), 163-70; ‘Some notes on the construction of Christos ho Pantepoptes (Eski Imaret Cami) in Istanbul’ Deltion tes Christianikes Archailogikes Etairias 16 (1991-2), 47-56; Master builders of Byzantium (Princeton, NJ: Princeton University Press, 1999), esp. 173-80, 194-200 and 224-6, On the significance of ornament more generally, see Grabar, Ornament, esp. Chap. 13 M. Snodin and M. Howard, Omament: A social history since 1450 (New Haven and London: Yale University Press, 1996). ‘As a basic introduction, see N. Coldstream, Medieval architecture (Oxford: Oxford University Press, 2002). See T. Mowl, Elizabethan and Jacobean style (London: Phaidon, 1994), for example. One issue here is what words the Byzantines used to describe ornament, E. Gombrich, The sense of order (London: Phaidon, 1979); Grabar, Ornament. See Ousterhout, Master builders, 194-200. On the use of Kufic and Pseudo-Kufic, see A. Walker, ‘Conflating the others: Pseudo-Kufic in Middle Byzantine art’, paper presented at the Medieval Academy of America Annual Meeting, Seattle, April 2004; ‘Magic and meaning in Byzantine ar: Pseudo-Arabic ‘omament and the San Marco cup’, paper presented at the College Art Association Annual Conference, Adlanta, February 2005. | am grateful to Alicia Walker for chese references. Papalexandrou, ‘Memory tattered and torn’, 69 and 72, has suggested that the use of texts, especially foreign texts on buildings, helped to create a shared sense of identity with the past or of conquest ~ the placing of ‘their’ language on ‘our’ building as a sign of triumph.