The History of the Image of Edessa : The Telling of a Story Author(s): AVERIL CAMERON

Posted by on Jun 28, 2016 in Library | Comments Off on The History of the Image of Edessa : The Telling of a Story Author(s): AVERIL CAMERON

The History of the Image of Edessa : The Telling of a Story AVERIL CAMERON

There is no doubt that in Byzantine society religion had so far extended its domain that it constituted the single most important set of power relations. How and why that situation came about belongs not to Byzantine history, but to the history of late antiquity, where, as always, it is easier to document the change than to explain it. Nevertheless, to see the development over time of a single religious token, as we can with the image of Edessa, an image of Christ “not made by human hands,” will help us to see both where the shifts take place and what the crisis points are. For the thought-world of the Byzantines was not, as is so commonly supposed, entirely static and self-contained : it, too, developed and changed over time, even if the development took the form, as it often did, of a defensive reaffirmation in stronger and more all-embracing terms of the attitudes already adopted. Probably the hardest problem confronting the modern Byzantinist is to understand, with the full degree of seriousness which it deserves, the centrality of religion in the Byzantine world-order, to see how it functioned, and to avoid an anachronistic separation between “purely religious” elements in a given situation and “underlying” or even “real” social or economic factors. In the case of the image of Edessa, and almost uniquely here, we are able to see over time how embedded religion was in Byzantine society and, above all, how the “telling” of the image was progressively extended and developed as new situations arose. It is a long story, but one which can be very suggestive for the problem outlined above.

It very soon becomes clear, for instance, that the Mandylion of Edessa, the miraculous portrait of Christ on a cloth known and loved by generations of Slavs through countless copies in Slavic churches l and venerated by Russian soldiers as recently as during the First World War, cannot be traced back either as a miraculous image or as an impression

1 See A. Grabar, La Sainte Face de Laon: Le Mandylion dans l’art orthodoxe, Seminarium Kondakovianum (Prague, 1931).

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


on cloth, as distinct from a painted picture, beyond the sixth century. The universal understanding of later writers that this image of Christ was somehow miraculously created simply does not occur in the earliest literary sources. This understanding itself goes back to the moment when icons had assumed a special and contemporary significance, and the detailed stories of its origin follow naturally upon that understanding. We can best see how the notion of a picture of Christ at Edessa gathered round itself greater and greater symbolic value by surveying the texts in order, beginning with the earliest.2 The fullest, not surprisingly, is

the latest, namely, the Narratio de Imagine Edessena, composed to commemorate the arrival of the image in Constantinople in A.D. 944; but we can only approach the Narratio after a full consideration of what went before.3

The earliest reference, in fact, to a picture of Christ at Edessa (modern Urfa in Turkey) occurs in a Syriac work known as the Doctrina Addai, which in its present form seems to date from about A.D. 400.4 Eusebius had told the story of King Abgar’s letter to Jesus and Jesus’ reply in the Church History, around 300, 5 claiming to have derived it from city records in Edessa, but not mentioning any picture. While some have claimed that he might have omitted such a reference because of his own distaste for religious images,6 it seems likelier that the motif of the picture entered the complex of the Abgar legend only later, since the

2 For an earlier chronological survey, see the excellent paper by S. Runciman, “Some Remarks on the Image of Edessa,” Cambridge Historical Journal 3 (1931): 238-52; Averil Cameron, “The Sceptic and the Shroud,” Inaugural Lecture, King’s College, London (1980) ( = Continuity and Change in Sixth-Century Byzantium [London, 1981], chap. V), where the subject is treated obliquely. I am glad to have the chance of developing this more straightforwardly. For Edessa in general, J. B. Segal’s Edessa: The “Blessed City” (Oxford, 1970), though impressionistic in parts, is still very useful.

3 Thus it is quite wrong to make it primary, as does Ian Wilson, The Shroud of Turin (New York and London, 1978) (= The Turin Shroud [Harmondsworth, 1979]), e.g., chap. 15. The notion that the Mandylion of Edessa and the Shroud now at Turin are one and the same thing is quite impossible (see fn. 2 above). The Narratio is to be found at PG 113: 425 ff.

4 Ed. and trans. G. Phillips, The Doctrine of Addai the Apostle (London, 1876). A full treatment of the early texts will be found in E. Dobschütz, Christusbilder: Untersuchungen zur christlichen Legende, vol. 1 (Leipzig, 1899), pp. 120ff.; see now also H.J.W. Drijvers, “Edessa,” Theologische Realenzyklopädie (Berlin), 9 (1981): 277-88.

5 Historia Ecclesiastica I.13. For the latest view on the date of Eusebius’s Church History, see T. D. Barnes, “The Editions of Eusebius’s Ecclesiastical History” Greek, Roman, and Byzantine Studies 21 (1980): 191-201, who has Eusebius beginning it in the mid 290s.
6 Runciman, “Some Remarks,” p. 241 f. For a valiant attempt to argue against this general attitude in Eusebius, see Sister C. Murray, “Art and the Early Church,” Journal of Theological Studies 28 (1977): 303-345.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


pilgrim Aetheria, who visited Edessa in about A.D. 380, saw no picture, but merely the letter of Jesus.7 No doubt the Doctrina Addai derives from the same set of material known to Eusebius, but that the picture was an element added to the story at a late stage also seems likely because two sixth- or seventh-century papyri which appear to preserve variants of the whole complex earlier than and distinct from that of the Doctrina are equally without mention of a picture.8 We must now ask, therefore, what the story of the picture in the Doctrina amounts to and why it should have entered an otherwise self-contained complex of myth at precisely this stage.

A close look at the Doctrina reveals that already there has been some conflation of personages. Abgar sent to Jesus among other envoys one Hannan (Greek Ananias), described as the keeper of the archives, who is said to have written down for the king everything he saw Jesus doing and all that he heard him say. The next stage is the sending of a letter, again via Hannan, to whom Jesus also dictated his reply. We then learn that Hannan was the king’s painter, and that he painted a picture of Christ for Abgar. But the conversion of Abgar is not achieved here (as it is in later texts) by the sight of the picture: the conversion of Edessa is apparently laid upon Thomas after the ascension of Christ, and Thomas in turn sends Addai (Greek Thaddaeus), who is at the centre of this text.

It is Addai who cures Abgar of his disease, and it is through Addai that Abgar becomes a Christian. Clearly this text is concerned with the apostolic origins of the church of Edessa, that is, with affirming its authority.9 Here the picture is of minimum importance. Yet there are already several different elements uneasily coexisting: the letter, the picture, Hannan, Addai, Thomas. There was already an established connection between Thomas and Edessa, and the pilgrim Aetheria came to see Thomas’s shrine.10 The Doctrina, however, promotes the claim that Addai brought Christianity to Edessa, with a rather crude attempt to harmonize this story with the existing ones about the letter and with the Thomas connection. By the sixth or seventh century, as the Greek

7 Peregrinano Aetheriae, 19, ed. H. Petré, in Sources chrétiennes 21 (1948): 162-71.
8 R. Peppermüller, “Griechische Papyrusfragmente der Doctrina Addai,” Vigiliae Chris tianae 25 (1971): 289-301.
9 It is likely, in fact, that Christianity came to Edessa in the second century.
10 For Thomas and Edessa, see Drijvers, “Edessa,” pp. 282f.; A.F.J. Klijn, Edessa, de Stad van de Apostel Thomas (Baarn, 1962). The Acts of Thomas are of the third century and have been thought to have originated in Edessa; already, however, Thomas is

presented as the apostle of Parthia and ïndia.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


Acts of Thaddaeus (Addai) show,11 the emphasis has changed; though Addai is still the nominal hero of the Acts, the agent of Abgar’s conversion is the picture, which is now a miraculous picture on cloth; furthermore, the cure takes place even before Addai arrives at Edessa. Thomas can now be discarded altogether, because now everyone knows that what really mattered in this complex was the image of Christ. Further, Hannan’s role has had to be modified : he could not have painted the picture if it was a miraculous image. So we read that indeed he was unable to do so, whereupon Christ took a cloth, asked for water with which to wet his face, and then impressed his image on the cloth. From now on, this or a version of it becomes the standard explanation for the origin of the image.12 By comparison, the reference to a picture in the Doctrina Addai is unemphatic and low-key; the thrust of the text is clearly concentrated elsewhere. Thus the entry of the picture into the tradition is likely to be for unspectacular reasons.13 While the letter was venerated not only in Edessa but over a wide area, it probably had its critics:14 an extra proof would be welcome to the Edessan church authorities, and would by this time be likely to take pictorial form. An authentic picture of Jesus would be more immediate and persuasive evidence of the unimpeachable origins of Edessan Orthodox Christianity than even a supposed letter, which after all was only dictated. The transfer from letter to picture – or, rather, the juxtaposition of letter and picture – is part of the move towards the codification of the religious discourse. By telling about the religious tokens, and by increasing its complexity, the subject’s potential is increased. Thus the “addition” of the picture is not an unfortunate accretion to be swept away by the historian,1 5 but a deliberate move in the Christianization process needing full appreciation. And the recording of this move in literary texts (and,

we must suppose, spoken intercourse, too) is a crucial part of the process. The move necessitated an actual picture, and called forth explanations of the picture’s origins; but at the same time the explanations gave the picture (and later the miraculous image) its real power in society.

1 1 See below, fn. 32.
12 For example, St. John Damascene, De fide orthodoxa IV. 16; De imaginibus 1.27.
13 Rather than as a counterpart to pictures of Mani (Driivers, “Edessa,” p. 280).
14 It was officially declared apocryphal in 494 (E. von Dobschütz, Das Deeretum Gelasianum, vol. 8 [Leipzig, 1912]). Procopius seems to reflect some doubt – see below

and fn. 24.
15 Runciman, “Some Remarks,” p. 239: “a parasitic growth to the story which in a few centuries entirely covered it.” But the story was not a “given” at a particular point in time, but a growing and changing phenomenon.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


These explanations were not slow in forthcoming, once the further idea of icons not made by human hands had spread. For a time, the simple notion of Hannan’s painting was sufficient. But by the late sixth century the first of the icons “not made by human hands” had appeared in the Syriac-speaking milieu to which Edessa belonged.16 Chief among them was the Camuliana icon, first attested in Syria in 558 and taken to Constantinople by 574. 17 The official hierarchy, therefore, recognized the potential of these powerful religious tokens; the Camuliana Christ-icon was paraded round the battlefield and on the walls of Constantinople, and both events were fully recorded.18 Thus the icon acquired a multifaceted role, both in private devotion and as a symbol of unity. It was only to be expected that the image at Edessa should also serve to fulfil these new functions. Perhaps the only too human painting by Hannan (or what passed for such) had been lost. At any rate, we hear for the first time at the end of the sixth century of just such a miraculous image at Edessa, “found” in the city gate by the bishop just at the time of the greatest Persian threat to the city, in A.D. 544.19 The problems surrounding the date of the first appearance of the real icon (to which I shall return) are of less interest in themselves than the setting in which the “finding” was located in the later texts. For Edessa in A.D. 544 was deeply divided; not only were there some who looked for ways of avoiding the fate that had befallen Antioch in 540 by making a rapprochement with Chosroes, but the church was at odds internally. Jacob Bar’adai had very recently been made Monophysite bishop of Edessa – in name only, it is true, yet with a rapidly growing

constituency.20 The Orthodox needed a counterweight to his magnetic pull, just as they needed an explanation for why Edessa was actually spared, if at high cost. We shall see the rival religious groups wrangling over possession of the image in the seventh century; and a similar context, together with the tension of extreme danger from the Persian army, was the stage for the image’s “finding.”

Evagrius, writing at the end of the sixth century, had no doubt that Edessa had been saved by the miraculous image. Yet Procopius, who

16 See E. Kitzinger, “The Cult of Images before Iconoclasm,” Dumbarton Oaks Papers (hereafter DOP), 8 (1954): 85-150.
17 See Averil Cameron, “Images of Authority: Elites, Icons, and Cultural Change in late sixth-century Byzantium,” Past and Present 84 (1979): 18ff.

18 A. Grabar, L’Iconoclasme byzantin (Paris, 1957), pp. 3 Iff.
19 Evagrius, HE IV.27.
20 E. Honigmann, Évêques et évêchés monophysites d’Asie antérieure au VIe siècle, CSCO Subsidia, 2 (Louvain, 1951); W.H.C. Frend, The Rise of the Monophysite Movement

(Cambridge, 1972), pp. 283 ff.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


wrote much nearer to the time of the actual events and had a detailed

knowledge of Edessa, did not even mention it.21 It has been thought that the “rational historian like Eusebius” would have left it out.22 But

Procopius was no more rational than Evagrius. He did not leave out the miracle which saved Apamea from Chosroes on the same occasion, nor the efforts of Bishop Megas of Beroea, nor the sign sent from God before the sack of Antioch.23 Furthermore, he knew of the letter of Jesus at Edessa and was clearly intrigued by it; his “rationalist” doubts about the authenticity of its addendum promising that Edessa would never be captured did not prevent him from concluding that the letter and its addition served a useful social purpose for the Edessenes.24 It is very hard to believe that Procopius would have left the image out of his ac- count, had he known of it. Much simpler to suppose that the miraculous image was “found” only after the event, and that its immediate function was to justify a course of action taken by a certain sector of the Edessenes. Evagrius, whose bias was deeply Orthodox, very naturally emphasized the “miracle” which saved the town from the Persians, for it was an Orthodox miracle, and the image was housed in the newly rebuilt Orthodox cathedral. Procopius, on the other hand, did not mention it, not merely because he did not know, it, but, because it did not yet exist. He was, indeed, writing very soon after 544, certainly before 550. 25 Thus the image was born during the tense questioning which would have followed Edessa’s costly escape from Chosroes. By comparison, when Evagrius wrote, in the 590s, such icons in Byzantine-Persian warfare had become almost a commonplace,26 and the connection of Edessa’s escape with a miraculous icon would have been fitting.

Another, less easily dated text bears on this question of date : a hymn in Syriac celebrating the rebuilt Orthodox cathedral at Edessa, which mentions the image.27 We have a terminus post quern for the church, and thus for the image, of A.D. 553, when Amazonius became bishop.28

21 BP 11.26-27 (and cf. 12.6-13.11).
22 Runciman, “Some Remarks” (fn. 4), p. 244.
23 Apamea: BP 11.11. 14-30; ibid., 6.17f.; Antioch: ibid., 10.1. 24 £/> 11.12.30.

25 Books 1 to 7 of the Wars were finished in 550 (B. Rubin, Prokopios von Kaisareia [Stuttgart, 1954], pp. 80-81).
26 Cameron, “Images of Authority,” p. 23.
27 A. Grabar, “La témoignage d’une hymne syriaque sur l’architecture de la cathédrale d’Edesse au VIe siècle et sur la symbolique de l’édifice chrétien,” Cahiers archéologique 2 (1947): 41 ff.; translation: C. Mango, The Art of the Byzantine Empire, 312-1453 (Englewood Cliffs, N.J., 1972), pp. 57 ff.

28 According to Jacob of Edessa, ed. and trans. E.W. Brooks, CSCO Script. Syri III. iv (Paris, 1905), p. 243.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


There are problems about the interpretation of the reference to the image, and it is oddly unemphasized for something that was to become so celebrated so soon.29 Yet the hymn does refer clearly enough to an image not made by human hands, at a time very soon after the Persian attack, though later than Procopius’s Persian Wars. By contrast, the author of the Syriac Chronicle of Edessa, written soon after 544, already used the theme of divine intervention in the escape of Edessa from Chosroes, but with no mention of a miraculous image.30 All this points to an emergence, by some means or other, of a miraculous image at Edessa in the middle of the sixth century. The discourse had thickened, yet at its heart was a physical object, which we must now consider.

For there seems to have been a change in the object – or, if not that, then a change in the way in which the object was seen. From now on, the texts refer to a miraculous impression on cloth. Paint was not enough; the image must have been transferred by Christ himself as he pressed a napkin to his moistened face : an alternative has it bedewed with sweat during the agony in the garden.31 This development is the product both

of an increasingly persistent call for the most immediate memorials of Christ – for whom there was, after all, nothing to equal the Virgin’s robe and girdle at Constantinople or her house at Ephesus – and of the growing complexity of the discourse about the image. More and more detail was being added about the origins of the image, and an account of its exact miraculous coming-into-being was now a prerequisite for a satisfactory understanding. Thus the Greek Acts of Thaddaeus32 know the image as an impression on a cloth, as we have seen. Hannan is on his way down to second-rank, but has retained a place in the story. The image, however, is central, even over the part played by Thaddaeus himself. As yet, though, we do not find the embroidery that Hannan was prevented from painting Christ’s picture by the radiance of the divine presence, or the conflation of the Hannan and Thaddaeus stories that make Thaddaeus-Addai the bearer of the image to Edessa. The total story grows only gradually. Another text seems to point to the notion of

a cloth already in the seventh century, namely, the odd passage in the

29 See Cameron, “The Sceptic and the Shroud,” pp. 9-10.
30 Ed. I. Guidi, CSCO Script. Syri III.xiv.1 (Paris, 1903), trans., p. 11.
31 Narrai io, PG 113:432Df.
32 R.A. Lipsius and M. Bonnet, Acta Apostolorum Apocrypha, vol. 1 (Leipzig, 1891), pp. 273-78.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


Chronicle of John of Nikiu, a distorted text written in Greek but surviving only in Ethiopie, where a Jew of Alexandria called Aubaruns is said to have possessed the mandil and towel with which Christ girded himself when he washed the feet of the disciples; these objects are said to have passed to the church of the Tabenniosites in Alexandria in the early sixth century (patriarchate of Timothy, 51 7-35). 33 Surely this is a garbled version of the Abgar legend, and it links Abgar with a cloth. It also seems to show that the term mandylion (meaning “small cloth, kerchief, napkin”) was beginning to be applied to the image on the cloth, or rather to the cloth itself, almost as soon as the idea of a cloth took hold. The Arabic mandil, later also applied to the image, is obviously a Greek loan-word,34 and means much the same as the Greek (in turn derived from the Latitf manielium or mantele). Another important element has entered the complex : the image of Edessa is to be thought of as an impression on a small cloth, the sort that a man might use to wipe his face, and this is how it is glossed in the later Greek writers.35

A natural question is what was the image of Edessa actually like. Was there at some stage an actual substitution of a piece of cloth for a painted icon? Or did the image always look much as any other ancient icon might have looked, even if the local people told themselves that it was on a piece of cloth? The tenth-century Narrado tells of its having been nailed onto a board, and fixed in a gold frame.36 Probably it was indistinguishable from any other precious Christ-icon; what changed was the discourse, not the object. That, at any rate, is what is suggested by references to painted copies being made well enough to be mistaken for the original.37 The language for the description of works of art is notoriously elusive and ambiguous. The word eikon, for instance, continued to be commonly used for the image by writers who were convinced that it was an impression on a cloth, while the language of painting could be applied to it with an equal freedom. Most telling, however, is probably the account by Michael the Syrian of copies made from the image in the early Arab period; the artist carefully used dim colours so that the copy would look old, which suggests to me (for we

33 R.H. Charles, The Chronicle of John, Bishop of Nikiu, translated from Zotenbergs Ethiopie text (London, 1916), pp. 144-45.

See below on Eutychius, fn. 48.
35 E.g., by cheiromaktron (Narratio, PG 113:429D) or soundarion (e.g., Germanus of Constantinople; see fn. 41 below).
36 PG113:437A.

37 Michael the Syrian, Chronicle XI. 16, trans. J.-B. Chabot, II. hi (Paris, 1904), pp. 475 ff.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


are also told that this copy fooled people into accepting it as the original) that the image never actually looked like a cloth at all. For this reason, we must conclude that the artist of the Sinai icon of Abgar and the Mandylion, in which the depiction is of a free-flowing cloth, was using his imagination;38 the reality must have looked quite different.

The early Arab period in Edessa, and the years immediately preceding it, were in fact crucial in the development of stories about the image. Just as the miraculous image had been “found” in a context of tension and division, so now it became a talisman fought over by Monophysites and Orthodox. The reign of Maurice (582-602) saw severe struggles between the two groups, during which four hundred Monophysite monks were said to have been killed. Later, however, the Monophysites (or “Jacobites”) gained the upper hand, and had gained control of the cathedral of Edessa under Heraclius, since that emperor had handed it back to the Orthodox.39 It was at this point that a copy was made by the Jacobites arid (so the story goes) passed off to the Orthodox as the genuine article.40 In the last years of Byzantine rule in Edessa, the image’s political potential, which had been there from the beginning, was greatly intensified, and, we must suppose, the detailed story of its origins as a face-cloth was worked out, giving it an even greater claim to authoritative and immediate contact with Christ. Certainly by the early eighth century the image had achieved a wider fame, though Edessa itself

swas cut off from the Byzantine world by Arab rule. Two factors worked to increase its prestige and indeed to preserve it – the Iconoclastic controversy and the control of Edessa by the Arabs. The former called forth a yet more developed codification of the meaning of icons, in which the image of Edessa now held a central place; the latter protected the image from the destruction which was the lot of most of the great icons in the Byzantine world, while allowing its reputation to grow unhindered, for the Arabs made no attempt to weaken or destroy the church in Syria. A heightening of the political role of the image, therefore, was followed by a set of circumstances which peculiarly favoured its rise to become the major surviving early icon during the Iconoclastic period.

In this process, a crucial factor was the amount of attention given to it in contemporary iconophile writings, even though as a miraculous

38 See K. Weitzmann, “The Mandylion and Constantine Porphyrogennetos,” Cahiers archéologiques 11 (1960): 163-84.
39 Michael the Syrian, Chron. X.23, XI. 3; see R. Devréesse, Le patriarcat d’Antioche depuis la paix de l’église jusqu’à la conquête arabe (Paris, 1945), p. 293.

40 Seefn.36.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


image not made by human hands it was strictly irrelevant in the technical arguments about the circumscribability of Christ. Nevertheless, iconophiles appealed to the prestige and tradition surrounding the image and to its authority as, they believed, contemporary witness to the appearance of Christ. Germanus, the patriarch of Constantinople deposed in 730 by Leo III for his support of icons, cited the example of the image of Edessa in a speech made before the emperor.41 Above all, St. John Damascene, the greatest defender and apologist of icons in the eighth century, who himself wrote under Moslem rule, more

than once used the same example, clearly drawing on the Greek Acts of Thaddaeus.*2 These passages in the works of Germanus and John Damascene became famous; the patriarch Nicephorus, himself deposed by the Iconoclast emperor Leo V in 815, repeated Germanus’s arguments about the Edessan image almost verbatim.4’3 And so, through Iconoclasm and the debates to which it gave rise, the image of Edessa was transformed from a local possession fought over by local groups into

a major religious token in the Christian east as a whole. The iconophile

writers of the eighth and ninth centuries brought it out into the full glare

of publicity and made the image the centre of a far wider debate than had touched it in Edessa itself.

We can still, however, capture a glimpse of the local honour paid to the image in the eighth and ninth centuries from the work of Theodore Abu Qurrah, bishop of Harrãn in the late eighth to ninth century, who wrote a treatise on images in Arabic in which the preeminence of the image of Edessa is clear.44 For Theodore, the most notable of all images is that “in our city” (i.e., Edessa). To bring that home to any who did not appreciate how sanctified the image was, he resorted to some strange examples: supposing, he said, that there was a picture of someone’s father on the door of the church and everyone spat on it as they went in – how would the son feel?45 Or – and here we see the kind of

41 Quoted by George Monachus, Chron., p. 740 de ‘Boor, and cf. p. 321.16f. 42 See fn. 12.

43 PG 100:461A, with 260.
44 Theodori Abu Kurra, De Cultu Imaginum, ed. J. Arendzen (Bonn, 1897), chap. 23, pp. 46-47; translation: G. Graf, Die arabischen Schriften des Theodor Abu Qurra (Paderborn, 1910), pp. 328-30. A happy meeting with Dr. Sydney Griffith, who is preparing a new edition of Theodore’s work, drew my attention to its importance, as to that of Eutychius (see below).
45 Theodore, chap. 23. As Dr. Griffith points out, Theodore’s immediate concern was a pastoral one, to defend images in general, and especially the image of Edessa from critics near at home who were evidently protesting at the honours paid to it.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


argument that was offered by the opponents of images – suppose there was a king whose mother had committed adultery, and that a painter had painted a picture of her in the act and then showed it all round the city – would not the king grow furiously angry and tear the painter limb from limb? It would be no kind of defence, Theodore continued, for the wretched painter to cry ‘Tve done nothing to your mother; it’s only paint.” It was obvious, then, that the attention paid to an image, good or bad, was attention paid to the person represented. Honouring the image of Christ at Edessa with processions and feast days was the same as honouring Christ,46 and conversely, insults to the image were insults to Christ. So Theodore spoke of the homage paid to the image in Edessa

just after 800.47 Clearly, iconoclastic arguments had reached Edessa, probably from Moslem as well as Byzantine quarters, and provoked in men like Theodore an intensified devotion to images and a detailed defence. The image’s holiness had been taken for granted in its early days; now an iconophile dialogue had to be elaborated. It is harder, however, to know whether to press Theodore’s language – again the language of colours and paint – or the implication that the image was placed on the very door of the church. Perhaps neither should be taken literally. But in view of the next Arabic allusion to the image, this is an important point.

We come next, in fact, to the Book of the Demonstration of Eutychius, the tenth-century patriarch of Alexandria.48 In this work Eutychius tells of the “signs” which God sent to prove the truth of Christianity (as Jesus himself is said in the Qu’ran to be a sign); they include the holy places, relics, the oil of the chrism, and the Eucharist. But the only relic actually mentioned is the image of Edessa. Eutychius says: “the most wonderful of His relics which Christ has bequeathed to us is a napkin in the Church of ar-Ruhã [i.e., Edessa] in the region of Jezireh. With this Christ wiped His face and there was fixed on it a clear image, not made by painting or drawing or engraving and not changing.”49 The terminus ante for this passage is 944, when the image went to Constantinople. But note that Eutychius not only uses the term mandil, the Arabic equivalent of mandylion, but also expressly denies that the image was a painted icon – denying it the very word (as-surãh) which Theodore had un-

46 Theodore, chap. 23.
47 He was bishop in Harran from 800 to 812.

48 P. Cachia and W. Watt, Eutychius of Alexandria, The Book of the Demonstration, CSCO, 192 (Louvain, 1960), p. 384.
49 Cachia and Watt, Eutychius, p. 384.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


hesitatingly applied to it. This alone is indicative of how unwise it is to press exact terminology in specific passages. More importantly, though, Eutychius’s reference tells us of the centrality of the image, standing alone as a direct token of Christ, and with a preeminence far beyond anything known in its earlier history. It is matched, for Eutychius, only by the coffin and shroud of the Virgin, which had been taken to Constantinople and which “proved” her Assumption. No such relic of Christ existed to rival the claim of the Edessan image, and certainly no such physical token as could be claimed for the Virgin.50

Given this multiplication of attention paid to the image, the Byzantine government was bound to want it removed from the obscurity of Edessa to a proper home in Constantinople. But that did not happen until 944, in the context of renewed Byzantine-Arab warfare. And now the Byzantines were in the stance vis-à-vis Edessa that Chosroes had been in in the sixth century: Edessa was an enemy city which the Byzantines wanted to capture. The sources, not unnaturally, claim that the trading of the image for 200 Moslem prisoners was reluctantly agreed to by the caliph and that the people of Edessa resisted surrendering the image with force. More probably they bought their safety this way as they had paid Chosroes to leave them alone in 544. 51 On both occasions the reasons for Edessa’s escape were disguised in miraculous stories. Again we hear of copies of the image, and the attempt to pass one off as the genuine article.52 One such was in a Nestorian church in Edessa; another, kept with the image itself, was associated with the cure of a daughter of Chosroes I during the sixth-century invasion. When the image went to Constantinople it was accompanied by a copy on a tile, subsequently kept with it in the Pharos chapel. This tile, too, had to be given a miraculous origin. The Narrations author knows it as having originated when Hannan-Ananias was taking the original cloth back to Edessa: he hid it for safety in a heap of tiles, but there was a fire, for which he was unjustly blamed. He was saved from the charge when his accusers discovered the

50 It is striking that no author of this period mentions a surviving burial cloth or shroud of Christ which might have had equal claims with the Mandylion, or indeed with the Virgin’s relics. Eutychius is a case in point; more noteworthy, perhaps, is the fact that Photius, for all that his homilies of Holy Saturday dwelt on the events of the burial of Christ, never suggests any such surviving shroud – nor do the remaining Byzantine homilists, for whom Holy Saturday was naturally a standard theme.

51 So A. Toynbee, Constant ine Porphyrogenitus and His World (Oxford, 1973), p. 319; Narrado, PG 113:444f. ; Georg. Mon. Cont., pp. 918-19; Leo Gramm., pp. 325-26;

Theoph. Cont., p. 432; ps. Symeon, pp. 748-49. 52 See Runciman, “Some Remarks,1′ p. 249.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


cloth itself and a tile beside it with the imprint transferred onto it. From that time both relics were preserved together at Edessa.53 Much of the Narratio recording the transfer to Constantinople in 944 is concerned with the details of the journey and the wonders on the way, and with the liturgical reception of the image in the city. For the first time we encounter the notion of the image being placed with the letter in a casket, from which it was removed by the emperors.54 It was certainly also fixed on a wooden backing and framed in gold with an inscription (all attributed to Abgar himself)55 But the Narratio also preserves an

alternative version of its origin – “a different story,”56 in which Hannan has no part and in which Christ used the cloth to wipe his face during the agony in the garden, after which he gave it to Thomas, telling him to give it to Abgar through Thaddaeus-Addai after the Ascension. We recognise here in a more precise organisation the same conflation observable in the Acts ofThaddaeus. Curiously, the Narratio claims to be an archaiologia, an “archeology” of the image. And it is interested in proof: the story of the “finding” of the image in the sixth century (all connection with the picture in the Doctrina Addai has been lost) is adorned not only with the name of Evagrius, who recorded it in his Church History, but also with those of three patriarchs who had written a letter to the emperor Theophilus in support of images, citing the image of Edessa and the story of its finding.57 In a sense, the Narratio marks the end of this story ; the discourse about the image has been fully developed, and the image itself has come home to be placed beside the crown of thorns, the lance, and the tunic of Christ, where it can be quietly taken for granted until disturbed by the arrival of the Crusaders. It was seen in the Pharos

chapel and reported shortly before 1204, when it was listed by Robert of Clari.58 But during these years there was no need for further elaboration of its story ; that had been done in the context of Edessa, and brought

53 PG 113:432Af.
54 PG 113:449B.
55 PG 113:437Af.
56 PG 113:432Df.
57 PG 113:441 A.
58 1200-1201: Anthony of Novgorod, ed. B. de Khitrowo, Itinéraires russes en Orient

(Geneva, 1889), pp. 97-98; Nicolaus Mesantes, in A. Heisenberg, Nikolaos Mesantes, Die

Palastrevolution des Johannes Comnenos (Würzburg, 1907), pp. 29 ff. ; 1204: Robert of

Clari, La Conquête de Constantinople, ed. P. Lauer (Paris, 1924), chap. 83. The relation of

chapters 82 and 83 to chap. 92 on Blachernae is complex, but not relevant to the present

discussion. However, the discussion in A. Nada Patrone, ed., Roberto di Cari, La

Conquista di Costantinopoli (1198-1216) (Geneva, 1972), pp. 18 ff., is entirely confused and should be avoided.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


to a conclusion in the Narrano, itself as much a formal celebration of the image as the ceremonies it describes with which the image was received in the capital – a progress by boat round the city from Blachernae to the Golden Gate, then another through the city to the

Augusteum, St. Sophia and the imperial palace, where it was placed on the throne in the Chrysotriklinos before being finally deposited with the letter in the Pharos chapel.59 As for the appearance of the image by then, we may guess that it was faint and dim: Constantine Porphyrogenitus was miraculously aided to see in it what was indistinguishable to his brothers-in-law and rivals.60 So the same story shows that the image had not lost its political potential. Its removal to Constantinople boosted the claim to the throne of Constantine Porphyrogenitus; on the way, to use the phrase of Steven Runciman, the image itself “declared its political opinions”61 by inspiring a man possessed by demons to cry out that the kingdom belonged to Constantine. The kudos for Constantine was as great as that won by Heraclius with the recovery of the True Cross, and the event was publicised in art and ceremonial. But there was no more “telling” to be done.

When the Crusaders entered Constantinople in 1204, the image was still in its place in the Pharos chapel. But from that time on its history is dark. Most probably it was the toella which passed with most of the other Pharos relics in 1247 from Baldwin II to Louis of France.62 Subsequently its history goes undocumented. The Sainte Chapelle, where the Eastern relics were placed, was sacked in the French Revolution, and with this, one supposes, the image of Edessa disappears. If it was not transferred to France, its history is equally silent. No one speaks for the image of Edessa after 1204, only the copies in Western or Slav churches which preserved its memory, out of context, for many centuries.63 The original

had lost its point and never attained in the West the commanding position it had enjoyed in the East.

The image of Edessa belongs to the years of the ascendency of images. Its history, therefore, despite the early signs in the Doctrina Addai, is properly compact. Protected from official Byzantine Iconoclasm by its

59 PG 113:449B-452D.

60 Theophanes Cont., p. 432 Bonn.

61 Runciman, “Some Remarks,” p. 249.

62 So Runciman, “Some Remarks,” pp. 251-52. For the text, see de Riant, Exuviae

sacrae Constantinopolitanae, vol. 2 (Geneva, 1878), pp. 134-35.

63 For the former: Runciman, “Some Remarks,” p. 251; for the latter: Grabar, La Sainte Face.

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to


remote situation in Arab Syria, it was uniquely able to ride the storm and reemerge with an unrivalled claim to antiquity. Through all of this we can see its place at the centre of social and political consciousness, and we can see the growing detail and exactitude with which the story of the image was told, together with the steady reinforcement of its authority by a closer and closer tracking down of its bodily contact with the physical being of Jesus. Indeed, the affirmation of body imagery, which also took the form of emphasis on objects that had been in bodily contact with holy personages or holy places, was an important aspect of the establishment of early Christianity.64 Icons “not made by human hands” naturally played a special part in this development. Thus the image of Edessa is both a sign of the authorities’ wish for control and a prime example of the way religious tokens could, in this society, become tension points in a network, both synchronie and diachronic, of shifting power relations. The image of Edessa may not help very much with the why, since, as we have seen, its story really begins only when the preeminence of such tokens in the social organisation of the Byzantine world has already been established. But it is an illuminating example of the how, which is perhaps where historians should stop. Why Orthodox Christianity became for Byzantine society the touchstone of all else has to be explained in relation to the Christianisation of late antiquity, and then especially with reference to the sixth and seventh centuries. But this

was when the story of the Edessan image was just beginning. I offer the “telling” of this image in admiration to Ihor Sevcenko, who has done so much by his detailed and exact scholarship to open up the ensuing centuries, when, for very good reasons, the reputation of the image of Edessa was at its most potent.

King’s College London

64 As can be seen, I owe much to M. Foucault, The History of Sexuality, vol. 1: An Introduction (Eng. trans.; London, 1978) (here pp. 65, 126ÍT.).

This content downloaded from on Thu, 23 Jun 2016 20:51:41 UTC All use subject to