127 Remains of a Mural Painting of The

ARAM, 18-19 (2006-2007) 127-151. P.doi: FIGUERAS 10.2143/ARAM.18.0.2020725 127

REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION IN THE SOUTHERN CHURCH OF SOBATA (SHIVTA)

Prof. PAU FIGUERAS1 (Ben Gurion University)

Few sites in the present State of Israel are capable of arousing the emotions of a Christian archaeologist as do the remains of the three Byzantine churches of the ancient town of Shivta, located in the heart of the Negev desert. The town, which was known as Sobata in the Byzantine period,2 has remained practically intact since the time it was finally abandoned towards the end of the eighth century A.D. The first travelers to report on the remains of the mural painting I discuss here were the two distinguished British visitors to the site in 1914, the archaeologist C.L. Woolley and his young voyage companion T.E. Lawrence (later better known as “Lawrence of Arabia”). In their famous book The Wilderness of Zin, which includes a rather detailed description of the southern church, they paid attention to the fact that “the church was singularly barren of stone-carving; on the other hand it showed more traces than remained elsewhere of the tempera painting which must have been the commonest form of church decoration… In the southern apse alone could any coherent design be distinguished, and here the colours had faded under exposure to the light, most of the surface had been scraped away by iconoclasts, and rain-water had brought down lime from the upper ruins and left a thick white deposit over the whole wall-face. Only by wetting the stone were we able to make out and roughly to sketch the original painting. The subject was the Transfigura- tion” (Woolley – Lawrence, 1914-1915, 105). Ninety years have elapsed since the time of Woolley and Lawrence's description, but those traces are still visible, though who knows for how long (Fig. 1). Strangely enough, however, the remains of the mural painting have rarely been mentioned by other archaeologists who later on worked on the site or described this church. Arthur Segal did not report on them in any of his monographs on Shivta (Segal, 1983; 1988; 1988a), neither did the late Joseph Shereshevski, who described the church in some detail in his doctoral dissertation on the Negev settlements (Shereshevski, 1991). A very brief

1 Ben Gurion University of the Negev, Beersheva (Israel). 2 The name was preserved in two documents from Nessana (Kraemer, 1958, 212-214, Pap. 75, and 226-233, Pap. 79), as well as in a variant from a manuscript of the story of Nilus from Sinai (Abel, 1924; Abel, 1935).

06-8819_Aram 18-19_07_Figueras 127 06-26-2007, 17:44 128 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

mention of the painting’s remains was made by the late Fr. Bellarmino Bagatti in his book on the Christian villages of Judaea and the Negev, though he unfortunately mistook their location, attributing them to the central instead of the southern church. Neither did he recognize their subject, though he published three photos of the southern apse (Bagatti, 1983, 201, Tav 28). Only Abraham Negev mentioned the traces of a painting in a general article on Sobata, adding that Woolley and Lawrence recognized in them the Transfiguration scene (Negev, 1993). The photos I used in order to decipher and possibly reconstruct how the original painting could have looked, were taken recently, and from them we can realize how badly the remains are preserved today. The faint traces of painting, mostly in brownish red color, very little in yellow, appear as directly applied upon the pinky original ashlar, which is of local limestone (Fig. 2). On the whole structure of the three apses of this church there are no visible traces of lime or plaster having been applied to the stone, while such a plaster layer is quite evident on the main apse of the central church in the same town. I do not know if the colour of the mural painting had been applied directly to the stones, but this is the way it appears today to anybody visiting the site. A sim- ple explanation for this very unusual phenomenon would be that the remaining traces are not those of the mural painting which once decorated the apse. Rather, these colors could belong to a first and tentative sketch of the whole mural done by the painter before he produced the definitive painting. The latter was normally done on a thin plaster layer, either wet or dry, according to one of two well-known systems, the fresco and the secco. The plaster carrying the final painting might have not survived in our case, and most probably we only have the remains of the temporary sketch. However, if this is the case, it is rather puzzling that that sketch on the stone was done also for all the non- figurative parts of the decoration, including the tooth-pattern around and inside the small recess, today all in red, but once recorded as red and blue by Woolley and Lawrence. On the other hand, these same researchers explicitly stated that they found the whole apse covered with a thick layer of lime. They thought that the lime was deposited by rain-water from the upper ruins. Could not this layer of lime rather be the original lime background of the fresco, which did not retain the colours and has since disappeared altogether? It is also possible that, at a certain moment of history, an iconoclastic scraping away of the paintings occurred, as Woolley and Lawrence claimed, causing the total disappearance of the real painting. This phenomenon seems to be particularly perceptible in the traces of the figure standing to the right of the mandorla or halo around the central figure of Christ (Fig. 3). I shall now try to describe the painting remains from an iconographic point of view. For this purpose comparison with other ancient representations of the

06-8819_Aram 18-19_07_Figueras 128 06-26-2007, 17:44 P. FIGUERAS 129

same scene are essential.3 Unfortunately, however, a very small number of representations of the Transfiguration has come down to us from the pre- iconoclastic period in the East. The most ancient and best preserved example is the famous one decorating the central apse of the sixth century church in the Sinai monastery of St. Catherine, donated by the emperor Justinian (Sotiriou, 1953; Forsyth, 1968; Forsyth-Weitzmann, 1971) (Fig. 4). In order to proceed in a systematic way, I will start with the best preserved part, which is the painting of the apostle John (Fig. 5), identified by the inscription of his name. a. The Greek letters…NNIC (Fig. 6) is what is left of the name identifying the prostrated figure in the middle of the lowest part of the painting. It stands for IWANNHC (written IWANNIC), John, the youngest of the three disciples of Jesus who are attending the scene of His Transfiguration. When compared with similar representations, we can also find the abbreviation IW (for IWANNHC) in connection with the prostrated disciple in the Belisirama church (Alti Kilise) of the 12th century, at Göröme in Turkey (Fig. 7). In the Sinai mosaic all the five persons depicted around Jesus are identified by their names written in Greek. In Shivta, however, only the name of John was written, or traces of it are preserved. A second identifying feature of this figure representing John in Shivta would be his lack of a beard, a detail characteristic of all other ancient Transfiguration scenes. However, we only guess that it was originally so here, as just a few colour stains have survived from the young disciple’s face. b. In its physical posture, one could say that the prostrated figure of John is typical of the Byzantine pictorial tradition, as we find it on many mosaics and murals, as well as on icons and miniatures. One example is the figure of peni- tent King David, from an illustrated folio of the well-known early 10th century Psalter from Paris (Lassus 1976, 139, fig. 100), which can be compared to our prostrated figure of John the disciple (Fig. 8). The same posture of the disciple in the middle is found in practically all representations of the Transfiguration scene. In Shivta, it is not clear whether the prostrated disciple is holding his face with his right hand, though traces of the latter are rather well preserved. Interestingly, in Sinai, it is the figure of Peter rather than John who is so depicted, prostrated in the central place and holding his face, which he turns to the viewer rather than to Christ (Fig. 9). This feature alone (but there are oth- ers) repudiates a direct dependence of the Shivta Transfiguration on the one in Sinai, or vice-versa. The reason for the peculiar appearance of Peter in Sinai might be the text of Luke’s narrative, where it is written, “Peter and those who were with him were heavy with sleep but kept awake, and they saw His glory” (Luke 9: 32).

3 I regret not having been able to consult the work by Ch. Ihm, 1960, Die Programme der christlichen Apsismalerei vom vierten Jahrhundert bis zur Mitte des achten Jahrhunderts, Wiesbaden.

06-8819_Aram 18-19_07_Figueras 129 06-26-2007, 17:44 130 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

c. In Shivta, Peter is the figure appearing behind John, though no identifying name or any letters of it are preserved. The lowest part of his body is hidden behind the prostrated figure, but we must imagine him as kneeling down (Fig. 10). On his head, the hair seems to have been painted short and white, as is common for most representations of Peter in early Christian art as a whole and in the Transfiguration scenes of the Byzantine tradition in particular. In its general appearance the figure can be likened to that of John in the Sinai mosaic and other kneeling figures. He seems to be addressing himself to Christ (like in the Gospel’s account), probably raising one hand towards Him in the original painting. This is the way Peter is depicted in most of the ancient representations of the scene, East and West, for example in the mosaic in the monastery church at Daphni in Greece, from the year 1100 (Fig. 11). In the Shivta mural, Peter is holding a scroll in his right hand, visible only in its upper part. In no other ancient representation of the same scene do we see Peter holding a scroll. However, this should not surprise us. Indeed, in ancient Christian art Peter is often represented holding a scroll, as, for example, on the 10th century ivory tryptich of Haberville (Fig. 12) and a large icon in the An- nunciation Cathedral in the Kremlin, from the late 14th century (Katchalova et al., 1990 pl. 119). The scroll is understood as a sign of the particular authority Jesus had conferred to Peter in Caesarea Philippi (Matthew 16:13-19), thereaf- ter acknowledged to him by the Church. In Shivta and elsewhere, this attribute was obviously used by the painter as an identifying sign. Before turning our attention to the third figure of the scene, I must point out that the two disciples of Jesus commented upon were painted without a nimbus around their heads, thus disregarding the status of saints in which Peter and John were venerated by the Church. But the same trend is found in most of the ancient Transfiguration scenes, not only for the three disciples but also for the two standing figures of Moses and Elijah. This is the case, for instance, in the Sinai mosaic, as well as in a 9th century reliquary cross from Pliska, in Bul- garia (Fig. 13), and a miniature in the 10th century Gospels of Otto III, from the Reichenau school (Fig. 14). In other instances, however, we see a nimbus around the heads of the three disciples and not those of Moses and Elijah, as, for example, in the Transfiguration mosaic in the Roman church of Ss. Nereo and Achilleo from the beginning of the 9th century (Schiller, 1971, fig. 406). Sometimes all the five persons present in the event, beside that of Christ Him- self, are depicted with a nimbus, as in the Daphni mosaic (Fig. 11) and a mosaic icon from the end of the 12th century (Schiller, 1971, fig. 411). But there are cases of the opposite phenomenon, where only the three standing figures have a nimbus. Examples of the latter trend include the relief of a 10th century ivory diptych kept in the Ermitage Museum at St. Petersburg (Kartsonis, 1986, pl. 71a) and a 14th century Greek saccos, traditionally called the dalmatica of Charlemagne (Fig. 15).

06-8819_Aram 18-19_07_Figueras 130 06-26-2007, 17:44 P. FIGUERAS 131

d. We come now to the traces of the figure standing to the left of the two aforementioned ones, which, in contrast, is totally unrecognizable. It must, of course, represent either Moses or Elijah, though the position of these two Bi- blical persons in the scene is not always the same in the traditional iconography. In the Sinai mosaic (Fig. 4) we see Moses to the right of Jesus and Elijah to the left, both identified by their Greek names, and similarly also in the rather symbolic mosaic of the Transfiguration on the apse mosaic of Sant’Apollinare in Classe, in Ravenna (Fig. 16), which dates from the same period (Dinkler, 1964). In the latter occurrence, the two personages are identified by their names in Latin. But in most other instances, their position can be the opposite. When their names do not appear on the representations, which is the normal case, their identification can be ascertained on the basis of other features. In contrast with Elijah, always depicted as an elderly man, with white hair and empty hands, Moses usually appears younger, with dark hair, usually beardless, and in most cases holding a book. Typical representations can be found in the Daphni mosaic (Fig. 11), the mosaic icon (Schiller, 1971, fig. 411), and the Ermitage dyptich (Kartsonis, 1986, pl. 71a) already mentioned. In all three cases Moses stands to the left of Jesus. However, in the 14th century saccos (Fig. 15) Moses stands to Jesus’ right, as in Sinai (Fig. 4) and in Ravenna (Fig. 16). In Shivta, no detail survives of the standing figure (Fig. 3), which was apparently scraped away in ancient times. Nevertheless, its original standing position is certain, and its proportions correspond to those of the two figures we have been considering. With a little imagination, we may even sup- pose that the standing man is pointing to Jesus with his left hand, and perhaps holding a book in his right hand. Or else we might imagine a rather windy scene, with the man’s garments unfolding in the air. e. Let us now look at the central mandorla or elliptical hallo around the whole figure of Jesus, which symbolizes His glory and separates Him from the earthly world (Fig. 1). Attested in Christian art since at least the early fifth century in Santa Maria Maggiore in Rome (Lassus, 1976, 29, Fig. 23), this iconographic convention has a much earlier pagan use. According to André Grabar, its origins must be sought in the East, in Mazdean Persia (Grabar, 1968, 117). I would add that it is possibly related to the ancient use of the zo- diac circle, often sculptured around the sun-god Mithras, such as in the mithreum at Dura Europos, from the third century (Campbell, 1968, Pl. I, Fig. 34) or in a Roman relief from Modena, probably a century earlier (Fig. 17). Ancient Christians, who employed the mandorla to express the transcendent world of God and His saints, applied its use to highlight the divinity of Jesus (Grabar, ibid.) in the momentary vision of His glory enjoyed by His three closest disciples, the same ones who will soon be witnessing His human agony in Gethsemane. The mandorla in the Shivta painting is oval, as in most of other ancient representations of the Transfiguration, such as Sinai (Fig. 4) and

06-8819_Aram 18-19_07_Figueras 131 06-26-2007, 17:44 132 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Daphni (Fig. 11). Sometimes, however, it is depicted totally round, as in the above mentioned Pliska cross (Fig. 13), the Ermitage dyptich (Kartsonis, 1986, Pl. 71a), as well as in a miniature from the late ninth century manuscript of the Homilies of Gregory of Nazianzus (Schiller, 1971, fig. 409). In very rare cases, such as the Reichenau manuscript, the mandorla does not appear at all (Fig. 14). The colour of the mandorla is usually totally different from the background colour of the whole scene, but sometimes it differs only in intensity and nu- ance. The colour of the mandorla in the Shivta Transfiguration is completely undefined, and I would say non-existent (Fig. 1). Only a little part of its delimiting line has survived, and it differs from the pinky colour of the stones merely in intensity. Woolley and Lawrence described the mandorla and its colour with the words “the vesica of light pink”. My impression is that only the delimiting line was originally painted. f. Finally, we come to the painting remains of the figure of Christ, the central personage of the scene. Unfortunately, however, what is left is so vague (Fig. 18) that we can hardly imagine what Woolley and Lawrence saw and described in 1914. We can distinguish a little part of the nimbus to the left, some yellow (or gilded) lines belonging to the upper part of the garments, particularly to the right, and a few traces of the head and the face in the center. I can also see traces of the right arm and hand. The description made by the English researchers is as follows: “In the centre is Christ, full-face, with hands raised and brought together over the breast. The chiton [here meaning a long tunica] was seemingly of light pink edged with gold, the himation [the cloak or mantle called toga by the Romans] of dark blue; the halo was a plain yellow ring with white centre; the vesica of light pink. The figure was too much damaged to be copied” (Woolley – Lawrence, 1914-15, 105-106). Despite all the details, the description provided by Woolley and Lawrence is of no value for reconstructing the original painting, particularly when it comes to the position of the hands and the colour of the imation, which is the same as the mandorla that surrounds the figure. Both are “light pink”, distinguished only by the gilt edge of the imation. Perhaps the Byzantine painting of the Transfiguration more akin to ours in the original colour of the garments would be that on the Daphni mosaic (Fig. 11), though here we have not a “dark” but a “light” blue for the chiton. About the hands, my suggestion is that they also were treated in a very similar way in both representations. Woolley and Law- rence interpreted them as “the hands raised and brought together over the breast”. This detail is improbable, as we have in Shivta a very visible fold of the imation upon the left side of the breast, as in Daphni. Here, however, that fold is angular, almost square, and not wavy or rounded as in Shivta. If the left hand was “over the breast”, as Woolley and Lawrence wrote, this fold would

06-8819_Aram 18-19_07_Figueras 132 06-26-2007, 17:44 P. FIGUERAS 133

not be visible. One must therefore assume that the left hand was kept down- wards, as in the Daphni, Sinai, and most of the other Byzantine representations of the scene. Regarding the original appearance of Christ’s head in Shivta, we must imagine it with long hair and a black beard, as traditionally depicted in Christian art from the fourth century onwards. It is interesting that in Shivta no light rays seem to have been painted as coming from the glorious body of Christ, as in Sinai and everywhere else. g. Of the other two personages integrating the scene of the Transfiguration, Elijah would be standing on the left of Jesus, as parallel to Moses, the other witness, who was standing to his right. James, the third disciple, would be kneeling, like Peter, his symmetric parallel. Unfortunately, only a few small red stains remain from these two figures. Thus, I have not tried to reproduce them in my sketchy reconstruction (Fig. 19). But I must assume that they were not very different from the models at Daphni, Crete and elsewhere.

CONCLUSIONS

Summarizing what has been said till now, and trying to understand the meaning of the existence of these remains, I would like to underline three spe- cific points: 1. In spite of its state of deterioration, we recognize that the Transfiguration scene decorating the southern apse of the south church in Shivta is one of the oldest representations of that scene that has survived, and certainly the oldest on a mural painting. We must assign it to the beginning of the sixth century, the period in which the two side apses were probably built in that church, re- placing earlier square rooms (Negev, 1993). From about fifty or sixty years later we have the mosaic representations of the same scene on the central apse of the churches of St. Catherine on Mt. Sinai and Sant’Apollinare in Classe in Ravenna, and possibly in the mosaic of the eastern façade of the Eufrasiana basilica in Porec (Croatia), of which unfortunately almost nothing remains. 2. A comparative study of the Shivta remains with the Sinai mosaic has led us to observe that, despite the striking similarities in the distribution of the six personages comprising the scene (with the standing figures Elijah, Moses, and Christ inside mandorla, and the three kneeling disciples, the middle one almost prostrated), there is a significant difference of detail. Indeed, the prostrated disciple in the middle is Peter in Sinai and John in Shivta. The artists of the Byzantine tradition after the iconoclastic period will adhere to the Shivta model: John is always represented prostrated, sleeping or falling, head down- wards, while Peter does not appear to be so surprised. He is usually depicted looking up at Christ, obviously telling Him about building three tents or huts

06-8819_Aram 18-19_07_Figueras 133 06-26-2007, 17:44 134 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

(a detail that in one case is also expressed in painting). In Sinai, John and James raise their arms looking up to Christ, while Peter, on the floor, is holding his head. Thus we can conclude that the Sinai mosaist and the Shivta painter worked independently from one another. The Sinai model was never adopted, not only in the detail of the placing of Peter and John but also in the symmetric similarity of Moses and Elijah’s figures, with no apparent difference of age and no special attributes, and actually looking more like ancient philosophers than biblical figures. 3. The existence of a painting of the Transfiguration (the oldest such painting known to date), in the Negev desert, in a side chapel of a Byzantine church in the remote town of Sobata, deserves a comment. We understand that in the Sinai church, the display of that event is central, probably because of the close relationship between the theophany to Moses and the Children of Israel, thought to have occurred on that place (Exodus 19), and the theophany on the high mountain to the three disciples, with the presence of Moses and Elijah (Matthew 17:1-13), both of which are related to Mount Sinai. In the church of Sant’ Apollinare in Classe in Ravenna, the same scene has also been represented on the central apse, but in a rather symbolic way, as described previ- ously. The scene fills also the central place in the Pliska cross, so that the mandorla is totally circular and includes not only Christ but also Moses and Elijah, a feature which first appears in the eighth century Chapel of S. Zeno in the Roman church of S. Prassede (Grabar, 1968, 103, n. 22). We do not know how the scene was represented in the mosaic which once existed inside the gable of the façade in the Eufrasiana of Porec. In any case it was given a prominent place also there. There might be a historical reason for the sudden interest in the event of the Transfiguration of Christ in front of his closest disciples, a Gospel event that had never been represented in early Christian art, neither in the Catacombs nor on sarcophagi. Schiller points out that “it was primarily… during the religious controversies of the fourth century that the Transfiguration was looked upon as proof of Christ’s divine nature” (Schiller, 1971, 147), a dogma that was denied by the Arians, for whom Christ was only a human being. In the fifth century East, Chalcedonians and anti-Chalcedonians alike recognized the deep theo- logical contents of the episode, that includes more than one element of a real theophany (the hand of God the Father expressing in some representations, such as in Fig. 14, what according to the Gospel was made perceptible by the heavenly voice). The Transfiguration was seen by the fourth century theolo- gians as a revelation of Christ’s divinity to the three disciples. Following the comments on the glorious event by the author of the second epistle of Peter (1:16-18), John Chrysostom interpreted the Transfiguration as prefiguring the second coming of Christ, when “the Son of Man shall come in the glory of the

06-8819_Aram 18-19_07_Figueras 134 06-26-2007, 17:44 P. FIGUERAS 135

Father with his angels”.4 Similarly, Basil of Cappadocia perceives the Trans- figuration as an anticipation of His glorious Parousia.5 And in the words of Leo I, “The Incarnation of the Logos effected by the Holy Ghost through Mary and the Transfiguration which took place on Tabor proclaim that Christ is the Son of God and that he was sent from heaven”.6 Even for a later monophysite writer such as the Syrian Jacob of Serugh (451-521), who commented on the Transfiguration in one of his metric homilies, that event was of course clear evidence of Jesus’divine nature.7 In my view, the interest in the Transfiguration narrative as a Scriptural proof of the divine nature of Christ could have been highly influential for the plastic representation of the scene in the southern church of Shivta. Palestine had been the scene of extreme conflict between the two Church factions, the Chalcedonians and the Monophysites, in the fifth century. Not far from Shivta, in the Gaza region, Peter the Iberian, his clergy and his monastic community, might have initiated the interest in the Gospel text of the Transfiguration. It is therefore plausible that the painting of the Transfiguration in Shivta is some- how related to that controversy. On the occasion of the transformation of the two former pastophoria or sacristies into apsidal chapels in the early sixth century, the local clergy found it fitting to introduce that event, so much talked about, as a decorative painting. A similar reason may have motivated the decorative mosaic of the Transfiguration on the central apse of the new church that emperor Justianin later erected in Sinai. In any case, it is obvious that, from the sixth century on, the Transfiguration is ubiquitous in Christian art, particularly in the East. Also later, after the long break in Christian art resulting from the iconoclastic crises, it makes its glorious reappearance in the Daphni mosaic, in the frescos of the cave churches of Göröme in Cappadocia, in the illumination of manuscripts, and the tradition has continued to this day on the innumerable icons spread all over Russia and neighboring countries. In the West, however, it has not been so common, though there are also numerous representations. The Roman Church did not officially celebrate a special feast of the Transfiguration until the fifteenth century (A.D. 1457), following the victory of the Christian army against the Mus- lim Turks in Belgrade on August the 6th, 1456 (Réau, 1957, 576). In the Greek Church, on the other hand, it was one of the traditional twelve feasts of the Christian year since before the year 1000. In conclusion, the representation of the Transfiguration in Shivta seems to correspond to that trend in the icono-

4 John Chrysostom, Comm. Matth. (Migne, PG, 58, 554). 5 Basil, Homily on Psalm 44:5 (Migne, PG, 29, 400). 6 Leo the Great, Homily 51 (Migne, PL, 54, 312). 7 See Homily 49, “On Transfiguration”, in P. Bedjan, Homiliae Selectae Mar-Jacob Sarugensis, II, Paris-Leipzig, 1906, p. 347-375. Thanks to Prof. Sebastian Brook for calling my attention to this reference.

06-8819_Aram 18-19_07_Figueras 135 06-26-2007, 17:44 136 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

graphic program of the Eastern Church, that employed it to reaffirm its faith- fulness to the dogma of the dual nature in the person of Jesus Christ.

REFERENCES

F.M. Abel, “Une mention Byzantine de Sbaïta”, Byzantion, 1924, pp. 51-58. F.M. Abel, “Note sur Sbaïta”, Journal of the Palestine Oriental Society, 1935, pp. 7- 11. B. Bagatti, 1983, Antichi villagi cristiani di Giudea e Neghev, Jerusalem. L.A. Campbell, 1968, Mithraic Iconography and Ideology, Leiden. A. Cutler – J.M. Spieser, 1996, Byzance Médievale, 700-1204, Paris. E. Dinkler, 1964, Das Apsismosaik v. S. Apollinare in Classe, Köln-Opladen. G.H. Forsyth, 1968, “The Monastery of S. Catherine at Mount Sinai, the Church and Fortress of Justinian”, Dumbarton Oaks Papers, 22, 3-19. G.H. Forsyth – K. Weitzmann, 1971, The Monastery of S. Catherine Mount Sinai, Ann Arbor. A. Grabar, 1968, Christian Iconography: A Study of Its Origins, Princeton. A.D. Kartsonis, 1986, Anastasis, The Making of an Image, Princeton. I. Ya. Katchalova, N.A. Mayasova, L.A. Shchennikova, 1990, The Annuntiation Ca- thedral of the Moscow Kremlin, Moscow. C.J. Kraemer, ed., 1958, Excavations at Nessana, vol. III, Non-literary Papyri, Lon- don. J. Lassus, 1976, The Early Christian Art and Byzantine World, London-N.Y.-Sidney- Toronto. A. Negev, Sobata, 1993, in E. Stern, ed., The New Encyclopedia of Archaeological Excavations in the Holy Land, Jerusalem, pp. 1406-1407. L. Réau, 1957, Iconographie de l’art chrétien, vol. II, Iconographie de la Bible, II, Nouveau Testament, Paris. M. Restle, 1969, Wall Painting in Asia Minor, vol. III, Shannon, Ireland. G. Schiller, 1971, Iconography of Christian Art, vol. I, London. A. Segal, 1983, The Byzantine City of Shivta (Esbeita), Negev Desert, Israel, Oxford. A. Segal, 1988, Shivta – Portrait of a Byzantine City in the Negev Desert, Haifa (He- brew). A. Segal, 1988a, Architectural Decoration in Byzantine Shivta, Oxford. J. Shereshevski, 1991, Byzantine Urban Settlements in the Negev Desert, Beersheva. G. Sotiriou, 1953, “Das Verklärungmosaik in der Kirche des Sinaiklosters”, in Atti del 7o Congresso Bizantino, Palermo 1951, Roma, pp. 246-252. C.L. Woolley and T.E. Lawrence, 1914-1915, The Wilderness of Zin, (PEF Annual), London.

06-8819_Aram 18-19_07_Figueras 136 06-26-2007, 17:44 P. FIGUERAS 137

Fig. 1. Shivta, southern apse today, general view.

06-8819_Aram 18-19_07_Figueras 137 06-26-2007, 17:44 138 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 2. Shivta, the painting remains.

06-8819_Aram 18-19_07_Figueras 138 06-26-2007, 17:44 P. FIGUERAS 139

Fig. 3. Shivta, the scratched figure.

06-8819_Aram 18-19_07_Figueras 139 06-26-2007, 17:44 140 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 4. The Sinai mosaic, 6th century (after Lassus, 1976, p. 62, fig. 46).

Fig. 5. Shivta, the prostrated John’s figure.

06-8819_Aram 18-19_07_Figueras 140 06-26-2007, 17:44 P. FIGUERAS 141

Fig. 6. Shivta, John’s name.

Fig. 7. Transfiguration fresco in Belisirama church, Göröme, Turkey, 12th century (after Restle, 1969, p. 514).

06-8819_Aram 18-19_07_Figueras 141 06-26-2007, 17:44 142 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 8. David in Paris Psalter, 10th century (after Lassus, 1976, p. 139, fig. 100, detail).

Fig. 9. Peter’s figure in the Sinai mosaic (after Lassus, 1976, p. 62, fig. 46, detail).

06-8819_Aram 18-19_07_Figueras 142 06-26-2007, 17:44 P. FIGUERAS 143

Fig. 10. Peter’s figure in Shivta.

06-8819_Aram 18-19_07_Figueras 143 06-26-2007, 17:44 144 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 11. Daphni mosaic, 12th century (after Schiller, 1971, fig. 408).

06-8819_Aram 18-19_07_Figueras 144 06-26-2007, 17:44 P. FIGUERAS 145

Fig. 12. Haberville ivory, 10th century (after Lassus, 1976, p. 143, fig. 104).

06-8819_Aram 18-19_07_Figueras 145 06-26-2007, 17:44 146 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 13. Pliska cross, Bulgaria, 9th century (after Cutler–Spieser, 1996, 26).

06-8819_Aram 18-19_07_Figueras 146 06-26-2007, 17:44 P. FIGUERAS 147

Fig. 14. Otto III Gospels, Reichenau manuscript, 10th century (after Schiller, 1971, fig. 416).

06-8819_Aram 18-19_07_Figueras 147 06-26-2007, 17:44 148 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 15. Greek saccos or “Charlemagne’s dalmatica” (after Schiller, 1971, fig. 415).

06-8819_Aram 18-19_07_Figueras 148 06-26-2007, 17:44 P. FIGUERAS 149

Fig. 16. Symbolic Transfiguration, S. Apollinare in Classe, Ravenna, 6th century (after Dinkler, 1964).

Fig. 17. Mithras in mandorla, Modena relief, 2nd-3rd century (after Campbell, 1968, Pl. XVII, fig. 695).

06-8819_Aram 18-19_07_Figueras 149 06-26-2007, 17:44 150 REMAINS OF A MURAL PAINTING OF THE TRANSFIGURATION

Fig. 18. Shivta, remains of Christ figure.

06-8819_Aram 18-19_07_Figueras 150 06-26-2007, 17:45 P. FIGUERAS 151

Fig. 19. Sketchy reconstruction of the Shivta painting.

06-8819_Aram 18-19_07_Figueras 151 06-26-2007, 17:45