This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Embellishing the Steps: Elements of Presentation and Style in The Heavenly Ladder of John Climacus JOHN DUFFY
he Sinai Peninsula, a bleak and barren wilderness jutting into the northern end of the Red Sea, acted like a magnet from Early Christian times, attracting to its solitude men and women earnestly engaged in the struggle to save their eternal souls. In the religious sphere the special mark of the place was its association with Moses and his meetings with God. It was here, near the elevation known as Mt. Sinai, that the future prophet came face to face with the divine and received the charge to lead the Israelites out of Egyptian bondage;1 it was on the summit of Mt. Sinai that Moses later accepted into his hands the tablets of God’s law.2 The primordial contacts between heaven and earth were to dominate the image of the location—a holy ground to the Jew, Christian, and Muslim—for the rest of time. In the second half of the sixth century, when the Byzantine Empire reached its greatest extent, Justinian I ordered to be constructed, near the foot of Mt. Sinai and on the traditional site of the burning bush, one of his great and lasting monuments, the monastery of St. Catherine, originally intended to double as a fortress at this strategic point in the region. Over the course of its long history the spiritual foundation, in addition to housing an active community of religious men, became a rich repository for works of Christian art, principally in the form of icons, books, and sacred vessels. The building complex and other glories of the monastery were for the first time explored and systematically examined in this century, during the joint Princeton-Michigan-Alexandrian expeditions to Sinai in 1958, 1960, 1963, and 1965.3 Among the book treasures housed in the library, and represented by numerous copies
T
I am grateful to Kathleen Corrigan and Ioli Kalavrezou for valuable help provided during the preparation of the lecture version of this article, and to Eunice and Henry Maguire for interesting observations after its delivery. I would also like to thank the anonymous reader whose comments contributed in a most helpful way. 1 Exod. 3:1–12. 2 Exod. 20:1–17, 31:18. 3 For a brief summary of the history of the Sinai expeditions, see the preface to G. H. Forsyth et al., The Monastery of Saint Catherine at Mount Sinai: The Church and Fortress of Justinian (Ann Arbor, Mich., 1973), 1–4. An accessible and informative account of the expeditions was also published by two of the leaders, George H. Forsyth and Kurt Weitzmann, in National Geographic 125 (1964): 83–127.
2
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
in the collection, is an indigenous Sinai product, variously known in English as The Heavenly Ladder or The Ladder of Divine Ascent, a recognized classic of human spirituality that has had a wide influence throughout the Orthodox world.4 What is known for sure about the author of the work is very little indeed. The Heavenly Ladder, as far as the record shows, is the only book he wrote. His name was John, and he was destined to be given, for an obvious reason, the surname Climax or Climacus. He was the hegoumenos of St. Catherine’s, most likely in the first half of the seventh century when he was already advanced in age.5 Even if the designation scholastikos, assigned to him in the title of a vita of uncertain date, is genuine, it probably means no more than that he was a learned individual.6 The Ladder itself had a modest enough beginning, coming into existence in response to a request from the abbot of the nearby monastery of Raithou, who asked Climacus to put together a spiritual guide for the members of that community. Ostensibly intended for a local audience in the Sinai Peninsula, the work of Climacus made its way well beyond those borders, reaching monasteries and private homes in numerous lands and many languages throughout the Middle Ages and into modern times. The fate of the Ladder in several different cultures would constitute separate chapters in the history of the work’s travels and influence.7 Within the Greek manuscript tradition, additional material was picked up along the way: an index of the thirty chapters with a short introduction and epilogue; the vita of Climacus by Daniel of Raithou, already mentioned; anecdotes about his life; marginal scholia; and illustrations ranging from one or two miniatures to full cycles of pictures. Without questioning in any way the interest and importance of the various elements attached to the later Climacus, especially the illustrations8 and the marginal comments,9 Among the best general introductions to this work, with relevant bibliographies, are G. Couilleau, s.v. “Jean Climaque,” DSp 8:369–89; K. Ware, introduction to John Climacus, The Ladder of Divine Ascent, trans. C. Luibheid and N. Russell (New York, 1982), 1–70; J. Chryssavgis, Ascent to Heaven: The Theology of the Human Person according to Saint John of the Ladder (Brookline, Mass., 1989); and M. Heppell, introduction to St. John Climacus, The Ladder of Divine Ascent, trans. L. Moore (London, 1959), 13–33. Of fundamental importance is D. Bogdanovic´, Jovan Lestvichnik u vizantijskoj i staroj srpskoj knjizhevnosti (Jean Climaque dans la litte´rature byzantine et la litte´rature serbe ancienne) (Belgrade, 1968), in Serbo-Croatian with French summary. 5 The question of Climacus’s approximate dates has not been—and, due to the lack of hard evidence, is not likely to be—finally settled. Most scholars nowadays, however, would probably be willing to agree that his death occurred at a time not far distant from the year 650. One of the more intriguing theories about the course of his life was developed by L. Petit, s.v. “Jean Climaque,” in DTC 8.1:690–93, who argued that Climacus, before he became a monk at a late stage, had been a married man with a professional career. The hypothesis, though reasonably supported by several concrete arguments, clashes head-on with the import of a passage in the Ladder itself (S. Giovanni Climaco: Scala paradisi, ed. P. Trevisan, 2 vols. [Turin, 1941], 2:157–59 [hereafter Trevisan]) where the wording seems clearly to imply that he had been a monk already in his early years (e“ti ne´ o" w“n). In one of the stories attributed to Anastasius of Sinai, in F. Nau, “Le texte grec des re´cits du moine Anastase sur les saints pe`res du Sinaı¨,” OC 2 (1902): 58–89 (the account is at pp. 63–64, no. 6), it is claimed that John was tonsured at the age of twenty. 6 The vita is written by the monk Daniel from the monastery of Raithou, and its historical value has been seriously questioned by scholars; the text is included in all editions of the Ladder. 7 The main lines of the work’s spread and influence are succinctly laid out by Couilleau, “Jean Climaque,” 381–88. 8 Comprehensively studied in the book of J. R. Martin, The Illustration of the Heavenly Ladder of John Climacus (Princeton, N.J., 1954). For a detailed and instructive investigation into the production of one of the most prominent illuminated copies the reader is referred to K. Corrigan, “Constantine’s Problems: The Making of the Heavenly Ladder of John Climacus, Vat. gr. 394,” Word and Image 12.1 (1996): 1–33. 9 Given the absence of a critical edition of the main text, it is not surprising that the mass of scholia and marginal comments has never been thoroughly sifted. 4
JOHN DUFFY
3
for the purposes of this study of presentation and style I concentrate largely on the original documents that passed hands, so to speak, between the two monastic superiors, John of Raithou and John Climacus. There are four documents in question:10 (1) the letter of the abbot of Raithou entreating his friend to write a book of spiritual guidance; (2) the letter of reply by Climacus; (3) the Ladder itself; and (4) the short treatise at the end of the Ladder known as the “Homily to the Pastor” (Pro` " to` n poime´ na lo´ go"), the guidance written specifically for the superior of a monastic community and addressed directly to the abbot of Raithou. It is entirely understandable that the third document—the thirty chapters of the Ladder—is usually, and often exclusively, the focus of attention in discussions of Climacus’s work. It needs to be stressed, however, that from the point of view of literary production the other three pieces should be treated as smaller but not insignificant parts of the same enterprise. Therefore, before moving to the Ladder proper in the second part of the article, I spend some time dealing with the front and back material, that is, the exchange of letters at the beginning and the “Homily to the Pastor” at the end. In writing to Climacus the abbot of Raithou remarks that he was encouraged by the words of Moses, “Ask thy father and he will tell thee, thy elders and they will inform thee.” 11 He is now humbly approaching Climacus, he explains, as father, elder, and exceptional teacher, asking him to become a second Moses; he wants him, as one who has been to the mountain and seen the vision, to prepare a set of “God-inspired tablets” for the instruction of the new Israelites who have left the world and taken up the monastic life. He is not trying to flatter but is simply repeating what is well known. He therefore hopes to receive soon those words of guidance inscribed on tablets that will point the way unerringly and become a ladder leading those who have chosen the angelic state up to the gates of heaven. If Jacob, a herder of mere sheep, he goes on, was able to experience an awesome vision of the ladder, surely the head of a spiritual flock can be expected to provide not just a vision, but a secure path up to God.12 This is the gist of the letter from Raithou, and it not only provides the first impetus for the book’s composition, but also introduces two ideas that will be important from the point of view of its presentation, namely, the general overarching notion of Moses as the medium for the delivery of God’s law, and the particular image, itself from the Mosaic Book of Genesis, of the ladder, seen by Jacob, set up between heaven and earth.13 The letter of reply by Climacus is a good rhetorical match for the abbot’s epistle.14 John of Raithou, with typical monastic humility, had approached as a suppliant, presenting himself in the most lowly guise: a sinner before an angelic spiritual father, an ignorant person before a talented, inspired teacher and paragon of virtue. Climacus, in response, deftly tosses the ball back with the depictions reversed: himself ignorant, poor in virtue, and a mere learner, his correspondent an exemplar of dispassion, purity of heart, and humility, the best of teachers whose mind is illuminated by divine light. However, afraid of offending the mother of all virtues, holy obedience, he has decided to accept the com10 They are transmitted together in most of the main manuscripts and printed versions. Throughout this article we cite the text of these documents according to the two-volume Italian edition of Trevisan, S. Giovanni Climaco: Scala paradisi. 11 Deut. 32:7. 12 Trevisan, 1:31–33. 13 Gen. 28:12–17. 14 Trevisan, 1:35–39.
4
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
mand of a moral superior. But, he cautions, the treatise he is sending for the community at Raithou is only a poor sketch which his better, the true artist, must bring to completion and embellish. It should be pointed out that Climacus makes no mention of the ladder in his letter; on the other hand, he does accept, however uneagerly, the role of Moses. And that reluctance itself is fully in character, since the original Moses was also very slow to accept the mission of leading the Israelites and mounted a series of objections to the Lord. One of them was prompted by a conviction of his own inadequacy: “Who am I that I should go to Pharaoh?” 15 Another stemmed from his alleged lack of eloquence, and on that basis he succeeded in having his brother Aaron, a more talented speaker, made his spokesman. With regard to the “Homily to the Pastor,” 16 the first observation to be made about a document that is rarely given equal time with the regular chapters of the Ladder is that Climacus himself treats it as part of the larger work; that much is clear from the opening words addressed to the abbot of Raithou: “In this earthly book I have put you in the very last place.” Set up along the same lines as the preceding thirty chapters, it is intended specifically for the abbot in his capacity as leader of a community, since, as Climacus had already indicated in the prefatory letter, it would be most inappropriate to send to a master a work intended for learners. However, even taking this special approach Climacus is uncomfortable, as he openly confesses. “But, father of fathers,” he says, “even as I send this to you I am afraid of hearing those words, ‘You who instruct others, do you not instruct yourself?’ So when I have said one last thing, I will bring this discourse to a close.” 17 The “one last thing,” in the event, turns out to be a complex and occasionally eloquent climax in which the ever reluctant Moses, now using the author’s advantage and seizing the last word, takes off his mantle of the prophet, so to speak, and places it in its full glory on his fellow hegoumenos instead. John of Raithou is depicted as the abbot whose soul has already been illuminated and has achieved unity with God. He is colorfully portrayed as the “guide of guides” (oJdhgo` " oJdhgw'n)18 of his brother monks and, in an elaborate metaphor, is put through each of the major episodes of the Mosaic epic, including the ascent, the vision of God, and the descent to Horeb, whence, glorified in soul and body, he carries back to his people “the tablets of knowledge and of the ascent” (ta` " pla´ ka" th'" gnw´ sew" kai` th'" ajnaba´ sew").19 In this conception/conceit, John of Raithou is the true Moses and the paragon of monastic virtue who needs no spiritual guidance from another. Even as author of the Ladder Climacus is content to cast himself as the great prophet’s mere mouthpiece, a second Aaron, and, he insists, a very poor one at that. Then, in a closing compliment, Climacus reverts to the image of the ladder of virtues and here too credits the other, as a skilled architect, with providing “the foundation, or rather the completion” for the book.20 The two words used, qeme´ lion and plh´ rwma, are without doubt multivalent and suggestive, and they are not meant to indicate precise Exod. 3:11. Trevisan, 2:322–77. 17 Ibid., 363. 18 Ibid., 367. 19 Ibid., 369. 20 Ibid., 375. 15 16
JOHN DUFFY
5
contributions. The field of ideas includes the abbot’s original request to compose the guide, the winning over of a reluctant author, and the provision of an exemplary life to serve as a model. Moreover, it may not be stretching things too far to see in the two words a reference to the literary “beginning” and “end” as supplied by the letter from John of Raithou and the “Homily to the Pastor” addressed to him. In another sense, the plh´ rwma is also what comes after the last step has been climbed: it is the summit of virtue where the soul is made one with love, which is God. And this is precisely how John of Raithou is depicted in the final paragraph of the “Homily.” If, therefore, while recognizing the Ladder of thirty steps as the central work of the author, we take into account the three surrounding documents as well, we will then be restoring the full context of the composition and in that way looking at a rhetorical whole. From this perspective it becomes clear that John Climacus had two main metaphors in his mind’s eye at the time of writing, Jacob’s ladder on the one hand, and Moses on the other; and while the figure of Moses is especially prominent in the front and back material, it also, in a real sense, hovers over the entire enterprise. The importance of the image for the original scheme is confirmed incidentally in two other ways worth at least a passing mention. In the vita written by the monk Daniel, transmitted in even the earliest manuscripts of the Ladder, Climacus is portrayed as a “New Moses” who after a vision on Mt. Sinai produced ta` " qeogra´ fou" aujtou' pla´ ka".21 Secondly, among the Narratives attributed to Anastasius of Sinai there is a charming anecdote that relates how, on the day on which John was installed as hegoumenos at Sinai, a group of six hundred guests was entertained to a meal at the monastery. As they ate, Climacus noticed a certain individual with short hair and dressed in Jewish fashion busily going around and giving orders to the cooks and various servants. Afterwards, when the servants themselves sat down to eat and the individual could not be located to join them, the narrative informs us: “Our holy father John said to us, ‘Let him be. There is nothing strange in the fact that the lord Moses came to serve at his own place.’”22 Furthermore, the prominence of the Mosaic image in the original configuration may not be unconnected with the issue of the work’s name, which has never been fully clarified. We have no critical edition of Climacus, all current versions depending on the 1633 text prepared by the Jesuit Matthew Rader, and there is not likely to be one anytime soon, given the enormous number of manuscripts involved. A very preliminary look at some of the oldest surviving copies from Sinai and Istanbul (of the tenth and eleventh centuries)23 reveals three possible contenders for the original Greek title: (1) Kli´max qei´a" ajno´ dou, or “Ladder of Divine Ascent”; (2) Pla´ ke" pneumatikai´, or “Spiritual Tablets”; and (3) Lo´ go" ajskhtiko´ ", or “Ascetic Discourse.” It may well be that, of the three, Pla´ ke" Ibid., 1:17–19. Included, as part of the transmitted material, in Trevisan, 1:19–21; see also Nau, “Le texte grec des re´cits,” 64, no. 7. In addition, two other Anastasian stories (Nau, ibid., nos. 32 and 34) refer specifically to the idea of the new or second Moses. 23 I have examined microfilms of Sinai gr. 421 (9th/10th century), Sinai gr. 417 (10th century), and Istanbul, Ecumen. Patr. 126 [134] (11th century). Descriptions of the Sinai manuscripts will be found in the catalogue of V. Gardthausen, Catalogus codicum graecorum Sinaiticorum (Oxford, 1886), 100–101; for Sinai gr. 417, there is also a full account in Martin, Illustration of the Heavenly Ladder, 186–87. For the Istanbul copy, see A. Tsakopoulos, Perigrafiko` " kata´ logo" tw'n ceirogra´ fwn th'" biblioqh´ kh" tou' Oijkoumenikou' Patriarcei´ou, 2, Tmh'ma ceirogra´ fwn JI. Monh'" Ag. J Tria´ do" Ca´ lkh" (Istanbul, 1956), 153–54. 21 22
6
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
pneumatikai´ is the strongest candidate. For one thing, it is prominently displayed, in decorated frames, in a number of the earliest illustrated copies. For another, it is the only one of the three aspirants in whose company the word title is used. In the oldest extant manuscripts, among the front matter we read: Pro´ logo" tou' lo´ gou ou» ejpwnumi´a pla´ ke" pneumatikai´ (“Prologue to the work entitled ‘Spiritual Tablets’”).24 The idea of Moses and the tablets is uniquely appropriate for the place of composition, and it served Climacus for other purposes as well. On a superficial level it allowed him considerable scope for the rhetorical game, which he played to the hilt, in both prologue and epilogue, with his friend from Raithou. And deeper down it gave the treatise itself an aura of inspired teaching and a sense of authority, which in no way interfered with the author’s pious wish to be considered a mere mouthpiece of a higher power. For all of its importance, however, to Climacus’s original conception and presentation, the Mosaic metaphor was destined to be dominated by its companion, the ladder metaphor. The ladder image, more visually compelling for a start, was in any case used for a substantially different purpose. Though not the only structural principle in operation in the work, this device, with its thirty steps, supplies a definite, if somewhat lightly attached, framework. It is true that the text of Climacus, as laid out, does not show anything like a strict hierarchical progression from one spiritual step to the next; however, it is not quite fair to conclude, as is sometimes done, that the presentation of vices and virtues is unsystematic. In fact, as Guerric Couilleau has demonstrated,25 there is a surprisingly high degree of pattern to be detected in groups of steps and some subtle thematic correspondences between groups and individual topics within them. One might call this logical or even theological order, because it is based on doctrinal content. Since this article concerns almost exclusively the literary side of the work, there is no need to pursue this line further and it will have been enough to mention in passing the convincing analysis of the French scholar. Instead I briefly consider a different aspect of system and order, highlighting the idea that Climacus was very conscious, for literary as well as doctrinal purposes, of drawing attention at the beginning or end of topics to the sequence between them. Order (ta´ xi"), sequence (ajkolouqi´a), and other closely related terms figure prominently in the border areas between steps. A few specific examples illustrate this point. At the end of the fourth step, the very long chapter on Obedience, and just before the fifth step, on Repentance, 24 As printed in Trevisan, 1:41. The decorated framing device appears in, among other copies, Princeton Garrett 16, fol. 8v (for a detailed description of the manuscript, see Martin, Illustration of the Heavenly Ladder, 175–77); Sinai gr. 417, fol. 4r (ibid., 186–87); and Paris Coislin gr. 263, fol. 7r (ibid., 172–74). With regard to the image of Moses, it should not be forgotten that ever since the 6th century monks at Sinai, whenever they looked up at the apse mosaic in their basilica, saw depicted two pivotal episodes from the prophet’s career, namely, the epiphany at the burning bush and the reception of the tablets. The question of the “original” title is a complicated one, and the remarks made here do not pretend to offer any kind of final solution. The issue might have been resolved by the find in St. Catherine’s, less than a quarter of a century ago, of the earliest witness to the text of Climacus—six folios in late biblical uncial script dating from the 7th or 8th century. From the little information so far made available it would appear that the title section is not among the surviving fragments; see L. Politis, “Nouveaux manuscrits grecs de Mont Sinai,” Scriptorium 34 (1980): 5–17, esp. 9. 25 “Jean Climaque,” 373–74.
JOHN DUFFY
7
begins, the author supplies these closing words: oJ ajqlhth` " sth'ke tre´ cwn ajfo´ bw". proe´ drame´ pote Pe´ trou Ij wa´ nnh"⭈ prote´ taktai de` nu'n uJpakoh` metanoi´a"⭈ oJ me` n ga` r prolabw` n uJpakoh'", oJ de` e”tero" metanoi´a" tu´ pon kaqe´ sthken.26 Here Climacus uses one of his favorite images—the monk as an athlete involved in a dromos—to lead into a Gospel reference (John 20:4) where John the Evangelist is reported to have outrun Peter to the tomb of Christ; this, in turn, is offered as the reason why the presentation of Obedience precedes the chapter on Repentance, John being the symbol of the one and Peter the symbol of the other. The opening words of the sixth step, on Remembrance of Death, provide a similar, if less elaborate, type of rationale for its particular position in relation to the chapter on Mourning that follows: panto` " lo´ gou prohgei'tai e“nnoia. mnh´ mh de` qana´ tou kai` ptaisma´ twn prohgei'tai klauqmou' kai` pe´ nqou"⭈ dio` kata` th` n oijkei´an ta´ xin kai` ejn tv' lo´ gv te´ qeitai.27 And the placing in turn of the eighth step, on Placidity and Meekness, after Mourning is explained in this way: w”sper u”dato" ejn flogi` kata` mikro` n prostiqeme´ nou telei´w" hJ flo` x ajposbe´ nnutai, ou”tw kai` tou' ajlhqinou' pe´ nqou" to` da´ kruon pa'san th` n flo´ ga tou' qumou' kai` ojxucoli´a" ajpoktei´nein pe´ fuke⭈ dio` aujto` kai` ajkolou´ qw" teta´ camen.28 Those instances will have shown that one is dealing with a style of ordering that is not internally generated and based on doctrine, but is rather externally imposed and inspired by creative imagination. It is a kind of sequencing that one would be justified in labelling rhetorical order, in the sense that it has more to do with embellishment than with logic; it could be regarded as a type of decorated bordering added to prettify the presentation. This feature is by no means unique to Climacus. It is also found in at least one of the works of a contemporary writer, Sophronius of Jerusalem, who will be brought into the discussion later on as well. One of his more substantial productions is The Miracles of Cyrus and John, an account of seventy cures performed by the two saints in their shrine at Menouthis in Egypt.29 In this text Sophronius is extremely attentive to matters of presentation, whether it be the careful arrangement of groups of miracles along geographic or other lines, or the smoothing out of the crossing from one miracle account to the next. The method of transition is reminiscent of Climacus, and a few examples will suffice to indicate that the same type of imposed order and sequence is at work. The subject of miracle 14, Qeo´ pempto", is said at the outset to appear in an appropriate se26 Trevisan, 1:203 ⫽ Luibheid and Russell (as above, note 4), 120–21: “Keep running, athlete, and do not be afraid. Once John outran Peter, and now obedience is placed before repentance. For the one who arrived first represents obedience, the other repentance.” In giving translations for passages of the Ladder proper, I generally follow the version of Luibheid and Russell, with occasional modifications. 27 Trevisan, 1:243 ⫽ Luibheid and Russell, 132: “As thought comes before speech, so the remembrance of death and of sin comes before weeping and mourning. It is therefore appropriate to deal now with this theme.” 28 Trevisan, 1:287 ⫽ Luibheid and Russell, 146: “As the gradual pouring of water on a fire puts out the flame completely, so the tears of genuine mourning can extinguish every flame of anger and irascibility. Hence this comes next in our sequence.” There is an interesting discussion of the bridging between chapters involving both the image and the text in Corrigan, “Constantine’s Problems,” 20 and 21; the artist made a conscious effort to improvise connections even in cases where the layout of the text presented difficulties. 29 For the critical edition, following an analysis and study of the miracles, see N. Ferna´ndez Marcos, Los Thaumata de Sofronio: Contribucio´n al estudio de la incubatio cristiana (Madrid, 1975).
8
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
quence (ajkolouqi´an euJra´ meno" eu“kairon) because in following H j li´a" of number 13 he preserves the alphabetical order,30 and Ij wa´ nnh" of miracle 15 receives similar praise for the same reason.31 Miracle 27, the story of one Theodore, is said to follow nicely on number 26, which concerns a woman called Theodora.32 On a more playful level, Paul the pauper, the subject of miracle 18, is credited with knowing where to take his place in the sequence: he comes after the wealthy John of number 17, just as a beggar goes after a rich man in the marketplace.33 From the foregoing it should be clear that the attention paid to order and sequence by the two writers is informed by a similar spirit and is in the nature of artistic enhancement meant to contribute to the neatness of the presentation. It is time now to direct our attention to individual steps of the Ladder and to focus on aspects of its literary style. The separate chapters, like the work as a whole, do not reveal an immediately obvious structure, but nevertheless allow the careful reader to discern a general pattern. Various scholars (primarily Dimitrije Bogdanovic´, Couilleau, and Kallistos Ware)34 have proposed a number of slightly different solutions. I myself would present the following pattern. In many of the steps one finds these four elements: (1) a brief introductory statement; (2) concise definitions of a virtue or a vice; (3) a general discussion of the theme of the step combining exegesis, admonition, illustrative stories, and personal observations; and (4) a short formulaic closing statement. If this grid does not fit snugly onto every chapter taken at random, the explanation is that Climacus will not allow himself to be held to either a large overall scheme or anything like a fixed pattern in the individual steps. And much the same could be said of his writing style, one of the special attractions of the work: it is unpredictable and can swing, in the turning of a page, from finely crafted definitions that have a hint of poetry to the pedestrian prose of instruction that reads like school lecture notes, from striking metaphors and images that have all the simplicity and earthiness of a Homeric simile to passages of enigma and obscure allusion. The multiple facets of the work’s style have been admirably identified and discussed by Bogdanovic´ in a separate chapter of his book on Climacus,35 and there is no need to repeat his findings, but it may be useful to examine more closely a few of the larger elements of style and to advance the consideration of the topic one or two paces further than has been done in the past. Some of the more remarkable features of Climacus’s prose style, it can be argued, resemble closely the qualities that one associates with certain types of Byzantine liturgical writing, in particular the sermon. The sermon here should be taken in a broad sense, to include the metrical homily of Eastern origin, the kontakion, that was very popular in the sixth and seventh centuries. The three elements isolated for examination here are poetic quality, litany structure, and drama. A series of texts serve to illustrate various aspects of these features. Our first extract is an example of one of the two main kinds of definition to be found Ibid., 271. Ibid., 272. 32 Ibid., 292. 33 Ibid., 277. 34 See Bogdanovic´, Jovan Lestvichnik, 221; Couilleau, “Jean Climaque,” 374; and Ware, introduction to John Climacus, 13. 35 Jovan Lestvichnik, chap. 5, esp. pp. 127–45. 30 31
JOHN DUFFY
9
in a typical step of the Ladder, that is to say, a single long period or sentence consisting of a chain of descriptive phrases, which have been rearranged here in order to show the structure as built by the successive members, or cola, of the sentence. ponhri´a ejsti`n eujqu´ thto" ejnallagh` peplanhme´ nh e“nnoia oijkonomi´a yeudome´ nh (5) kekolasme´ noi o”rkoi sumpeplegme´ noi lo´ goi buqo` " kardi´a" a“busso" do´ lou pepoiwme´ non yeu'do" (10) fusikh` loipo` n oi“hsi" tapeinw´ sew" ajnti´palo" metanoi´a" uJpo´ krisi" pe´ nqou" mikrusmo` " ejxomologh´ sew" e“cqra (15) ijdiognwmo´ ruqmo" ptwma´ twn pro´ xeno" ajnasta´ sew" ajnti´qeto" u”brewn meidiasmo` " memwrame´ nh kath´ feia (20) ejpi´plasto" eujla´ beia daimoniw´ dh" bi´o"
This is Climacus’s definition of ponhri´a (wickedness or depravity) from the twenty-fourth step.36 One can see that every phrase, with one main exception, is composed of either two nouns, or a noun and an adjective, in juxtaposition. Each colon is independent and makes a complete statement with two words, one of which carries the chief burden of description. We may call this poetic compression and power, with everything but the essentials stripped away. For example, the nouns lo´ goi (6) and yeu'do" (9) are reasonably simple and straightforward. It is left to the loaded adjectives to convey the special meaning and nuance intended by the author: sumpeplegme´ noi lo´ goi suggests words that are interwoven, complicated, subtle, and therefore deliberately misleading; pepoiwme´ non yeu'do" is falseness that has become a quality, a poio´ n or poio´ th", in other words, falseness or deception that has developed into a habit. In general, the vocabulary is chosen with deliberate care for its suggestiveness and figurative power, and attention is paid to the sound created. In at least three instances a conscious effort is made to create parallel and balanced phrases. kekolasme´ noi o”rkoi (5) and sumpeplegme´ noi lo´ goi (6) match each other in a number of ways: the type of words and their order (participle ⫹ noun); the gender, number and, case endings; the number of syllables in each phrase; and, last but not least, the exact correspondence in the pattern of unaccented and accented syllables (∼ ∼ ∼ x ∼ x ∼).37 The following lines, buqo` " kardi´a" /a“busso" do´ lou (7–8), are also artfully composed, with the use of buqo´ " and a“busso" opposite each other, the match in syllable count, and Trevisan, 2:85. The sign x is used here to indicate a stressed syllable; and ∼, to denote an unstressed syllable. In this extract, the same rhythm appears also in pepoiwme´ non yeu'do" (9) and daimoniw´ dh" bi´o" (21). 36 37
10
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
the same rhythm (x ∼) at the end. Similarly, in the case of the pair ptwma´ twn pro´ xeno"/ ajnasta´ sew" ajnti´qeto" (16–17) one may point to certain artistic elements, such as alliteration or the opposing of the words beginning with pro´ - and ajnti´- and used in combination with two other concepts that are diametrically opposed (sin and resurrection), as well as the balance in the end rhythm (x ∼ ∼). The few elements of phrase-end rhythm and balance in the extract just examined are a far cry from the full patterns of accented and unaccented syllables that characterize poetry like the kontakion, and it would be easy to claim that such traces are purely accidental in Climacus. However, there are numerous sections of the Ladder, and by no means confined to definitions, in which features of this kind are in evidence.38 Here are four sample pieces from three different chapters: (x ∼) x ∼ ∼ x ∼ ∼ (x ∼) x ∼ ∼ x ∼ ∼ x∼∼x∼∼ x∼∼x∼∼
hJsucasth´ " ejstin oJ boh´ sa" ejnargw'" “ EJ toi´mh hJ kardi´a mou oJ qeo´ ".” hJsucasth´ " ejstin ejkei'no" oJ eijpw´ n “ Ej gw` kaqeu´ dw kai` hJ kardi´a mou ajgrupnei'.” klei'e me` n qu´ ran ke´ llh" sw´ mati kai` qu´ ran glw´ ssh" fqe´ gmati kai` e“ndon pu´ lhn pneu´ mati40
C.
oJ me` n ajqumw'n nh´ cetai u”dasin, oJ de` ajkhdiw'n fu´ retai plh´ qesi41
x∼∼∼x x∼∼∼∼x x∼∼∼x x∼∼∼∼x (∼ x ∼) x ∼ x ∼ ∼ (∼ x ∼) x ∼ x ∼ ∼ (∼ x ∼) x ∼ x ∼ ∼
(∼ ∼ ∼ ∼ x) x ∼ ∼ x ∼ ∼ (∼ ∼ ∼ ∼ x) x ∼ ∼ x ∼ ∼
D. qu´ ra ga` r tou' prote´ rou ta` plh´ qh tw'n trauma´ twn, tou' de` deute´ rou oJ plou'to" tw'n kama´ twn42
(∼ ∼ ∼ x ∼ ∼) x ∼ ∼ ∼ x ∼ (∼ ∼ ∼ x ∼ ∼) x ∼ ∼ ∼ x ∼
If one considers the line endings and the area marked off by, and including, the last two accented syllables, a lot of parallelism is evident. The pattern in texts A and C (the socalled double dactyl, x ∼ ∼ x ∼ ∼), it should be remarked, is a very common end rhythm both in kontakia and in highly rhetorical Byzantine prose texts, and examples are indicated in the hymn extracts cited below.43 Further, in passages A, B, and D there are occurrences of complete cola that are fully balanced in rhythm; this is almost the case in C as well, and one slight piece of “metrical” license would take care of the extra syllable in the second line. But apart from the pattern of accented and unaccented syllables, one 38 Bogdanovic´ has already, using other parts of the Ladder, stressed its rhythmic and poetic qualities; see Jovan Lestvichnik, 142–45. Cf. French summary: “Les figures sonores donnent un effet puissant de rythme et la prose de Jean Climaque se transforme souvent en prose rythme´e voire en vrai poe`me” (p. 222). 39 Trevisan, 1:75. 40 Ibid., 2:235. 41 Ibid. 42 Ibid., 41. 43 For rhythm in Byzantine prose texts, which is normally confined to the phrase ending, or clausula, the ¨randner, Der Prosarhythmus in der rhetorischen Literatur der reader is referred to the classic study by W. Ho Byzantiner (Vienna, 1981). On the high frequency of the double-dactyl form, see ibid., 27, 42.
JOHN DUFFY
11
also notes in C and D another element of correspondence, the clear signs of assonance and rhyme, while in A and B one finds instances of homoeoteleuton. Once again one is reminded of the kontakion. Consistent rhyme and assonance are not a usual feature of Byzantine poetry until much later in the Middle Ages; even in Romanos only one part of one hymn has, as we might say, serious signs of these characteristics, namely, the third strophe (vv. 6–13) of the second kontakion on Joseph: h«n me` n ajgaqh` tou' despo´ tou hJ eu“noia, a“crhsto" de` li´an hJ tau´ th" dia´ noia⭈ e“sterge dia` semno´ thta oJ ajnh` r to` n Ij wsh´ f, e“qelge dia` faulo´ thta hJ gunh` to` n eujgenh'⭈ e“terpe me` n ejkei'non hJ ojrqo´ th" tou' tro´ pou, e“trwse de` ejkei´nhn wJraio´ th" prosw´ pou⭈ ou»to" aujtv' to` n oi«kon pare´ dwken, au”th aijscrw'" to` sw'ma proe´ dwken⭈44
(x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼)
(x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼)
In particular, the two lines starting with e“terpe me` n ejkei'non show not only the complete matching in accent patterns (as required), but also much correspondence in sounds at the beginning and the end of words. Still, for the feature in full—and almost unique— bloom, one must go to the anonymous Akathistos Hymn, which most scholars would assign to either the sixth or the seventh century: Cai're, di∆ h»" hJ cara` ejkla´ myei⭈ cai're, di∆ h»" hJ ajra` ejklei´yei⭈ cai're, tou' peso´ nto" jAda` m hJ ajna´ klhsi"⭈ cai're, tw'n dakru´ wn th'" Eu“a" hJ lu´ trwsi"⭈ cai're, u”yo" dusana´ baton ajnqrwpi´noi" logismoi'"⭈ cai're, ba´ qo" dusqew´ rhton kai` ajgge´ lwn ojfqalmoi'"⭈ cai're, o”ti uJpa´ rcei" basile´ w" kaqe´ dra⭈ cai're, o”ti basta´ zei" to` n basta´ zonta pa´ nta⭈ cai're, ajsth` r ejmfai´nwn to` n h”lion⭈ cai're, gasth` r ejnqe´ ou sarkw´ sew"⭈ cai're, di∆ h»" neourgei'tai hJ kti´si"⭈ cai're, di∆ h»" proskunei'tai oJ pla´ sth" cai're, nu´ mfh ajnu´ mfeute.45
(x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼)
(x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼)
The sample before us is the first of the twelve structurally identical thirteen-line units in the poem, each displaying a high degree of rhyme and sound correspondence. One need go no further than the first two lines of this particular unit to see a good illustration of these elements, and there is a rich display of matching and contrasting in the third couplet as well. Two further points are worth mentioning in passing: the deliberate positioning of u”yo" and ba´ qo" opposite each other in that couplet is similar to, though more effective than, Climacus’s buqo` " kardi´a"/a“busso" do´ lou, which we have already discussed;46 and the type of two-word paradox in the refrain of the last line (nu´ mfh ajnu´ mfeute) is a device that Climacus likes as well. Two good examples would be ajnh´ dono" P. Maas and C. A. Trypanis, eds., Sancti Romani Melodi cantica (Oxford, 1963), no. 44, 356–57. C. A. Trypanis, Fourteen Early Byzantine Cantica (Vienna, 1968), no. 1, 30. 46 Above, p. 9. 44 45
12
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
hJdonh´ , “pleasureless pleasure,” 47 and mi'so" a“mison, “hateless hatred,” 48 the latter used to describe the conflicting feelings a monk will have for his earthly parents. Still another feature that immediately catches one’s attention in the Akathistos is the repeated greeting of the archangel Gabriel to the Virgin Mary at the Annunciation, the cairetismo´ ". It forms a litany in each of the twelve longer strophes of the kontakion, all closing with the same refrain of the last line. In some parts of Climacus too one finds a similar type of litany or chain structure, though without a refrain. It makes its appearance especially, but not exclusively, in the other common form of definition found in the Ladder, as in the first of the two samples that follow: 1. meta´ noia´ ejsti ajna´ klhsi" bapti´smato"⭈ meta´ noia´ ejsti sunqh´ kh pro` " qeo` n deute´ rou bi´ou⭈ meta´ noia´ ejsti tapeinw´ sew" ajgorasth´ "⭈ meta´ noia´ ejsti swmatikh'" kataklh´ sew" dihnekh` " ajnelpisti´a⭈ meta´ noia´ ejsti aujtomato´ krito" logismo` " kai` ajme´ rimno" aujtome´ rimno"⭈ meta´ noia´ ejsti quga´ thr ejlpi´do" kai` a“rnhsi" ajnelpisti´a"⭈ metanow'n ejsti kata´ diko" ajkatai´scunto"⭈ meta´ noia´ ejsti diallagh` kuri´ou dia` th` n th'" tw'n ejnanti´wn toi'" ptai´smasin ajgaqw'n ejrgasi´a"⭈ meta´ noia´ ejsti suneido´ to" kaqarismo´ "⭈ meta´ noia´ ejsti eJkou´ sio" pa´ ntwn tw'n qliberw'n uJpomonh´ ⭈ metanow'n ejsti dhmiourgo` " oijkei´wn kola´ sewn⭈ metanow'n ejsti qli´yi" gastro` " ijscura` kai` yuch'" plh'xi" ejn aijsqh´ sei krataia'. 2. oJ ejn ajlhqei´a to` n ku´ rion ajgaph´ sa", oJ ejn ajlhqei´a th'" mellou´ sh" basilei´a" ejpizhth´ sa", oJ ejn ajlhqei´a po´ non peri` tw'n eJautou' ptaisma´ twn ejschkw´ ", oJ ejn ajlhqei´a mnh´ mhn kola´ sew" kthsa´ meno" kai` kri´sew" aijwni´ou, ´ ´ ´ ´ oJ ejn ajlhqeia fobon th'" eJautou' ejxodou ajnalabwn, oujk e“ti ajgaph´ sei, oujk e“ti fronti´sei h‘ merimnh´ sei, ouj crhma´ twn, ouj kthma´ twn, ouj gone´ wn, ouj do´ xh" tou' bi´ou, ouj fi´lou, oujk ajdelfw'n, oujdeno` " ejpigei´ou to` para´ pan.
In the first example, from the opening of the fifth step,49 there is a repeating series defining or describing meta´ noia, repentance; the language is highly figurative, and the chain of definitions seems to be an especially effective way to introduce one of the more moving and famous chapters of the work, in which the author describes a prisonlike monastery outside Alexandria, to be discussed later. The second passage, from the second step, on Detachment,50 is once more an opening paragraph of a chapter. Here the Trevisan, 1:273. Ibid., 2:135. 49 Ibid., 1:205. 50 Ibid., 69. 47 48
JOHN DUFFY
13
repetition of the phrase oJ ejn ajlhqei´a helps to build up the positive side of the notion of monastic detachment, to counterbalance the whole series of material and emotional attachments from which the aspiring monk is called to cut himself loose. Sophronius of Jerusalem, a contemporary of Climacus who spent ten years in one of the monasteries of the Sinai Peninsula, is another writer who is very fond of this rhetorical arrangement in his prose writings. The extracts below, taken from two of his sermons, are, like those of Climacus, presented here according to the colon structure, which is easy to detect in the case of Sophronius, since one of the identifying marks of his style is that every unit ends in a double-dactyl rhythm. For present purposes, the device of repetition speaks for itself, and a couple of words of additional comment will suffice. 1. cai´roi", w« cara'" th'" uJperourani´ou gennh´ tria⭈ cai´roi", w« cara'" th'" uJperta´ tou maieu´ tria⭈ cai´roi", w« cara'" th'" swthri´ou mhtro´ poli"⭈ cai´roi", w« cara'" th'" ajqana´ tou parai´tie⭈ cai´roi", w« cara'" th'" ajle´ ktou mustiko` n katagw´ gion⭈ k.t.l. ti´" sou fra´ sai th` n ajglai?an dunh´ setai; ti´" sou fa´ nai to` qau'ma tolmh´ seie; ti´" sou khru´ xai qarsh´ sei to` me´ geqo";51 k.t.l.
(x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼)
2. eij ou«n to` patriko` n aujtou' qe´ lhma pra´ xoimen pi´stin ajlhqh' kai` ojrqo´ doxon e“conte", kai` th` n Ij smahlitikh` n rJomfai´an ajmblu´ noimen kai` th` n Sarakhnikh` n ajpostre´ yoimen ma´ cairan kai` to´ xon to` jAgariko` n katea´ xoimen kai` th` n iJera` n Bhqlee` m oujk eij" makra` n qeasoi´meqa kai` ta` ejn aujth' katopteu´ soimen qau´ mata kai` to` n qaumatourgo` n aujto` n Cristo` n eijsaqrh´ saimen kai` su` n ajgge´ loi" aujtv' th` n uJmnvdi´an boh´ saimen. . . . 52
(x ∼ ∼ x ∼ ∼) (x ∼ ∼ x ∼ ∼)
The first piece, from a homily for the Feast of the Annunciation, is close in spirit and function to the cairetismo´ " of the Akathistos, enhancing the sense of excitement at the announcement and celebrating the amazing role of Mary in the divine plan. The second piece, though less imposing in form and more mundane in spirit, will nevertheless support the point under discussion. It comes from a sermon on the Nativity, delivered in Jerusalem at Christmas of the year 634, a few years before the defiant city fell to the Arab conquerors. Here the patriarch Sophronius lists a series of happy events that will result if the Christian citizens follow God’s will and keep the orthodox faith. The polysyndetic string can be said to add a tone of solemnity and confidence to his pronouncements. The last element of Climacus’s style chosen for comment is drama, and once more it is worth bearing in mind both the kontakion genre and the writings of Sophronius of Jerusalem. It is generally recognized that vivid dialogue, lively questions, and dramatization are prominent among the stylistic marks of sixth- and seventh-century hymns.53 Sophronius too very often enlivens his sermons with hymnodic touches—litany-like repPG 87.3:3237. RhM 41.3 (1886): 508.28–509.3. 53 See ODB 2:960; and Maas and Trypanis, Sancti Romani Melodi cantica, xxii. 51 52
14
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
etitions (as seen above), refrains, strings of pointed questions, and re-creations of dramatic scenes complete with dialogue, such as the first encounter between Christ and John the Baptist (in his sermon for the Feast of the Epiphany)54 or the appearance of the archangel Gabriel to Mary (in the sermon for the Feast of the Annunciation).55 In the Ladder also there are a number of passages of dramatic intensity, none more emotional than the fifth chapter, on Repentance, whose major scene is set in a monastery not far from Alexandria. Climacus had gone to pass some time there and during his stay learned of a special section, known as “The Prison” (Fulakh´ ), about a mile distant from the main community, where erring monks were sent for penitence. Granted permission to see it, the author spent a month with the inmates, and he includes an account of his experiences in the fifth step. It is certainly the strangest part of the work, and easily the most moving. Climacus is well aware of that; like a Moses addressing the Israelites, he says to the audience of his book, “Come gather around, listen here and I will speak to all of you who have angered the Lord. Crowd around me and see what he has revealed to my soul for your edification.” 56 What follows is a veritable visit to the underworld, with a catalog, in gruesome detail, of self-inflicted misery, deprivation, and punishment. With the visitor we see the harrowing sights and hear the groans and anguished questions of the tormented. Some sense of the account is conveyed by a few brief citations: “I saw some of those accused yet innocent men stand all night until dawn in the open air, their feet never moving, pitifully pounded by the natural urge to sleep, giving themselves no rest, reproaching themselves, driving sleep away with abuse and insults”;57 and later: “Others sat in sackcloth and ashes on the ground, hiding their faces between their knees, striking the earth with their foreheads. Others constantly beat their breasts, recalling their past lives and the conditions of their souls. Some shed their tears on the ground, while others, unable to weep, struck themselves. Some raised over their own souls a lament for the dead, since the strength to bear their heart’s grief had left them. Others moaned inwardly, stifling the sounds of their wailing until, unable to bear it any longer, they would suddenly cry out.” 58 Or again: “You could see the tongues on some of them dry and hanging from their mouths in the manner of dogs. Some punished themselves in the blazing sun, others tortured themselves in the cold, while others, again, drank only as much water as would keep them from dying of thirst.” 59 So the catalog goes on relentlessly, page after page, and ends appropriately on a supercharged note—the scene played out whenever an inmate was about to die. The others gathered around their brother in his final hour, while his mind was still working, and plied him with eager queries: How was he feeling? What were his hopes and expectations? Had he achieved what he had worked so hard for, or was his struggle a failure? Had he been given any kind of assurance or was he still uncertain in his hopes? And finally: “Can you say anything to us, 54 A. Papadopoulos-Kerameus, ed., Ana j ´ lekta JIerosolumitikh'" stacuologi´a" (St. Petersburg, 1898; repr. Brussels, 1963), 5:151–68, esp. 154. 55 PG 87.3:3217–88, esp. 3237. 56 Trevisan, 1:207. 57 Ibid., 209. 58 Ibid., 209–11. 59 Ibid., 215.
JOHN DUFFY
15
brother? Please tell us, so that we may know how it will be for us. Your time is over and you will never have another chance.” 60 Climacus’s considerable flair for the dramatic manifests itself in another connection that is of double interest, since it has a bearing on the question of structure within the work as well. With the eighth chapter starts a series of steps that begin to pay particular attention to the passions or vices. In the eighth step, Anger comes to the forefront. At the beginning of the ninth step, on Malice, the author makes an important statement, which speaks directly to his effort to maintain a certain order in presentation: “The holy virtues are like the ladder of Jacob and the unholy vices are like the chain that fell from the chief apostle Peter. The virtues lead from one to another and carry heavenward the man who chooses them. Vices, on the other hand, beget and stifle one another.” 61 The chain mentioned here refers to the shackles that fell off Peter when he was released dramatically from Herod’s prison by an angel,62 and it is introduced now as a device to link the series of vices, in the special sense of the family link. What this amounts to is, in effect, that for the whole middle portion of the thirty steps—meaning chapters 8 to 23 where the passions are highlighted—there are two structural metaphors at work, namely, the ladder of virtues and the chain or family of vices; and of the two, it is the family notion that controls, at least rhetorically, the sequence of chapters in this central section. For some of these steps Climacus goes a degree further. In short scenes at the end of five chapters, he presents several of the major passions in the dramatic role of the tyrant (tu´ ranno"); he produces the characterization for Anger, Despondency, Gluttony, Lust, and Insensitivity. The main purpose of the highly figurative exercise, apart from its usefulness for the structure, is to teach the causes and cures of the passions, and do so in a vivid and memorable way. Adhering more or less to the same general pattern in each instance, the staging is as follows. The tyrant in question, for example, Anger (Qumo´ ") in the eighth step, is hauled in chains before a tribunal and forced, under rough interrogation, not only to provide the names of his parents (i.e., causes) and relatives, but also to reveal, against himself, the identity of his enemies (i.e., cures). In the case of Anger, the subject is sufficiently intimidated to supply the information without much of a struggle. Appearing in the fourteenth step, the female tyrant Gluttony (Gastrimargi´a) is, by comparison, more recalcitrant and defiant; she does catalog her relatives, but warns at the end, “The thought of death is my enemy always, but there is nothing human that can really wipe me out.” 63 Lust is the passion highlighted in the fifteenth step, and it provides the occasion for a dramatic episode of a special kind. The role of the demon of Lust, for the purposes of the tribunal showdown, is played by the tyrant of the body, and the casting leads to a very interesting confrontation, taking the form of a schizoid struggle within the narrator/ interrogator himself. Throughout the fifteenth step Climacus is depicting vividly the never-ending war waged over chastity—the war between the two natures, the angelic and the material—and the efforts of the spirit to tame and conquer the tyrant of the Ibid., 223–25. Ibid., 303. 62 Acts 12:7. 63 Trevisan, 1:355. 60 61
16
PRESENTATION AND STYLE IN THE HEAVENLY LADDER
body. At the same time there is a clear recognition that it is a war in part between friends, since body and soul are inseparable, if uneasy, partners. By the end of the step the conflicting feelings have reached a high pitch of emotional intensity. Before the interrogator can directly confront the tyrant to extract the secret of how he might be conquered, before the dialogue of soul and body can occur, the soul, or mind, must face certain questions within itself. How can it tie up this body, this friend, and treat it like the other tyrants? “Before I can bind him he is let loose, before I can condemn him I am reconciled to him, before I can punish him I bow down to him and feel sorry for him. How can I hate him, when my nature disposes me to love him? How can I break away from him, when I am bound to him forever? How can I do away with him when he is going to be resurrected with me? How can I make him incorrupt when he has received a corruptible nature?” 64 These are the contradictions and divisions, and the split in emotions is further heightened by the conciseness of the phrases that follow, until at the end the struggle subsides, for the moment, into three questions that may produce quiescence but not an answer: “What is this mystery surrounding me? What is the meaning of this mixture that I am? How can I be to myself both an enemy and a friend?” 65 Here is the central portion in the original: pw'" mish´ sw o’n fu´ sei ajgapa'n pe´ fuka; pw'" ejleuqerwqw' v» eij" aijw'na" sunde´ demai; pw'" katargh´ sw to` kai` su` n ejmoi` ajnista´ menon; pw'" dei´xw a“fqarton to` fqarth` n eijlhfo` " fu´ sin; ti´ eu“logon ei“pw tv' ta` eu“loga kekthme´ nv dia` th'" fu´ sew"; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . kai` ga` r kai` su´ nergo´ " ejsti kai` pole´ mio" kai` bohqo` " kai` ajnti´diko" kai` ajntilh´ ptwr kai` ejpi´boulo"⭈ qerapeuo´ meno" polemei' kai` thko´ meno" ajtonei', ajnapauo´ meno" ajtaktei' kai` saino´ meno" ouj fe´ rei. a‘n luph´ sw o”lw" kinduneu´ sw, a‘n plh´ xw oujk e“cw dia` ti´no" ta` " ajreta` " kth´ somai⭈ to` n aujto` n kai` ajpostre´ fomai kai` periptu´ ssomai. ti´ to` peri` ejme` musth´ rion; ti´" oJ lo´ go" th'" ejmh'" sugkra´ sew"; pw'" eJautv' ejcqro` " kai` fi´lo" kaqe´ sthka;
The section of the Ladder examined above is appropriate for bringing to a close the discussion of selected stylistic features in Climacus, because it exhibits all of the elements focused on here: it has bits of rhythm and snatches of poetic phrasing, there are parts in litany-like arrangement, and the dramatic quality is pervasive and undeniable. The following can be said by way of general conclusion: There were essentially two things that I set out to accomplish in this article. The first was to restore some clarity to 64 65
Ibid., 401. Ibid.
JOHN DUFFY
17
the original rhetorical context of The Heavenly Ladder’s composition and presentation, with the following result: if fuller account is taken of the front and back material—the exchange of letters and the “Homily to the Pastor”—the figure of Moses emerges as an element of some importance in the overall scheme of things. And if it were someday shown that the original title of the Ladder was in fact Pla´ ke" pneumatikai´, that would be but a fitting capstone, since there is already enough real Mosaic presence in the Ladder and in the transmitted documents surrounding it to give to the treatise the tone of a set of inspired guidelines. The image of Moses also helps, from the reader’s point of view, to take pressure off the ladder concept as the organizing device; in conjunction with the concept of the chain of vices, it makes us realize that the ladder idea was never meant to carry the full burden of the work’s structure by itself. For just as the appearance of multiple, sometimes protean, metaphors is a notable feature of individual passages within the work, so too is it a vital element of the Climacus’s general aesthetics of presentation.66 The second objective was to focus on the author’s manner of writing. Here three features were singled out for examination, and an effort was made to suggest a literary affinity based on those elements of style, or, in other words, to identify a style context. The genre of the kontakion and the prose sermons of Sophronius of Jerusalem were brought forward as exhibiting a kindred spirit—as compositions in which similar types of artistic strategy were employed to embellish the writing and heighten the experience of the reader or auditor. If this is an accurate association, it should be made clear at the same time that no claims are being made about any direct influence of these writings on each other. Rather, I am suggesting that the rhetorical character of Climacus’s work, while retaining its own individual stamp, would seem to place it squarely within a particular tradition of Greek Christian writing.67 We know too little about the facts of John Climacus’s career to say anything concrete about his rhetorical training, but wherever he received it, whether inside or outside the monastery, he surely put it to good use, since the single book he wrote is at once a classic of spiritual wisdom and a fascinating, albeit challenging, work of literature. Harvard University 66 From another point of view, as Alice-Mary Talbot has reminded me, the story of Moses and the Ladder share the theme of ascent. 67 That tradition in turn might well represent a regional style within the Greek-speaking part of the Byzantine Empire. At any rate, its origins are commonly held to lie in the East and it is said to reflect specifically ¨nborn, Sophrone de Je´rusalem: Vie monastique et confession dogSyriac modes of composition. See C. von Scho matique (Paris, 1972), 102–3; Maas and Trypanis, Sancti Romani Melodi cantica, xii-xiii, for kontakia. Petit too (“Jean Climaque,” 692), commenting briefly on Climacus’s style, mentions the “school of Antioch” in this connection. The question would merit further exploration by someone competent in both Greek and Syriac literature.
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
The Byzantine Letter of Consolation in the Macedonian and Komnenian Periods A. R. LITTLEWOOD
et the foreign critic beware how he meddles with style, that impalpable essence which surrounds thought as with an atmosphere, giving to it its life and peculiar tone of colour, differing in different nations, like the atmospheres which envelop the different planets of our system, and which require to be comprehended that we may interpret the character of the objects seen through their medium. None but a native can pronounce with any confidence upon style, affected as it is by so many casual and local associations that determine its propriety and its elegance.” 1 Thus wrote William H. Prescott, the historian of Mexico, in his discussion of the work of his predecessor Antonio de Solı´s. One must reject such a wary, indeed pessimistic, approach to the literature of an alien culture since it would, of course, preclude any critical literary comment on Byzantine literature today; but at the same time one may also regret that Prescott’s scruples have rarely troubled the Byzantine scholar in the past. In his blind adherence to modern literary aesthetics, he has too often assassinated—with a bludgeoning crudity or, yet more effectively, with a deftness more silken than even a Thug’s rumal—the character of Byzantine literature for his generation. But now, fortunately, the telling of this tale is supererogatory at last, for modern scholarship, led by figures such as Jakov Ljubarskij and Alexander Kazhdan, is at last choosing the briar-entangled path of disinterested assessment. One genre, however, is frequently omitted in this recent trend and is, together with the progymnasma, the last area of Byzantine literature to attract sympathetic understanding. It is the genre of the letter. As long ago as 1932, at the Third International Congress of Byzantine Studies in Athens, Joannes Sykutris called upon his colleagues to divest themselves of their modern prejudices and take into account Byzantine criteria in their
“L
I wish to express my thanks to M. Mullett of the Queen’s University, Belfast, for generously sending me portions of her Ph.D. thesis and of her book Theophylact of Ochrid: Reading the Letters of a Byzantine Archbishop, Birmingham Byzantine and Ottoman Monographs 2 (Birmingham, 1997), before its publication; to M. P. Vinson of Indiana University for helpful suggestions; and especially to my colleague M. R. Cole for not only introducing me to some of the recent psychological literature on the subject but also instigating and discussing with me this article’s final paragraph. 1 W. H. Prescott, The Conquest of Mexico, bk. 2, chap. 8.
20
THE BYZANTINE LETTER OF CONSOLATION
judgment of Byzantine letters;2 but his clarion sounded a lonely note and few have taken their stand beneath his banner. Frustrated in their desire to extract precisely detailed information and irritated by yet another variation of a beloved cliche´,3 most scholars have tended to castigate the Byzantine letter as either a tortuous Chinese puzzle or little more than the Byzantine equivalent of the modern commercial greeting card.4 Valentin Smetanin has attempted to construct a methodology for approaching these letters (creating his own field of Byzantine “epistolology”) for both theoretical and practical purposes,5 thereby “unlock[ing] hidden meanings and decipher[ing] elaborate codes which would otherwise be inaccessible,” 6 but his efforts do not appear to have achieved much success, at least on the practical side, and he has had few disciples. Other scholars, most notably Apostolos Karpozilos,7 have conscientiously scrutinized letters of different periods to extract factual information, while Ljubarskij has subjected the letters of Michael Psellos to a thorough and penetrating analysis largely to reveal the personality of the author.8 It is, however, Gustav Karlsson who has, by his meticulous cataloguing of epistolographic formulae,9 laid the groundwork for the more literary examination of the letter. In recent years this direction has been most successfully followed by Margaret Mullett,
´ tudes J. Sykutris, “Probleme der byzantinischen Epistolographie,” in Actes du IIIe Congre`s international d’E ´ pistolographie,” in RE, suppl. byzantines (Athens, 1932), 295–310, following similar comments in his article “E 5. I should also note the slightly earlier pioneering work of Sister Agnes Clare Way in The Language and Style of the Letters of St. Basil (Washington, D.C., 1927) (for later work on Basil’s letters, see the summary in P. Hatlie, “Redeeming Byzantine Epistolography,” BMGS 20 [1996]: 235–37). 3 There appears to be a double standard employed here: when Shakespeare makes Gertrude say to her son Hamlet “All that lives must die,” we have a literary gem fit for a dictionary of quotations; but when, centuries earlier, a Byzantine says the very same thing, we have a hackneyed commonplace. It may be well to heed the words of S. Medcalf commenting on the Western medieval attitude, which is analogous to the Byzantine: “All literature is necessarily involved with commonplaces—situations, relations, places and other things which are familiar to the audience; but the modern writer tends to be embarrassed if this use of commonplaces is too obvious, whereas the medieval took it for granted as a technique. Geoffrey of Vinsauf (ca. 1200) . . . thought that one of the writer’s principal choices was whether to amplify his subject matter or to abbreviate it. On the whole, medieval writers tended to amplify and to rely on traditional commonplaces” (in The Cambridge Guide to the Arts in Britain, ed. B. Ford [Cambridge, 1988], 2:98). 4 This has been most memorably expressed by G. T. Dennis in The Letters of Manuel II Palaeologus, CFHB 8 (Washington, D.C., 1977), xix, but notice his change of attitude in “The Byzantines as Revealed in Their Letters,” in Gonimos: Neoplatonic and Byzantine Studies Presented to Leendert G. Westerink at 75, ed. J. Duffy and J. Peradotto (Buffalo, N.Y., 1988), 155–65. 5 See V. A. Smetanin, “Epistolologija pozdnej Vizantii: Postanovka problemy i obzor istoriografii,” ADSV 14 (1977): 60–76; idem, “Epistolologija pozdnej Vizantii: Proelevsis (konkretno-istoricheskaja chast’),” ibid., 15 (1978): 60–82; idem, “Teoreticheskaja chast’ epistolologii i konkretno-istoricheskij efarmosis pozdnej Vizantii,” ibid., 16 (1979): 58–93; idem, “Idejnoe nasledie Vizantii i ‘dekonkretizatsija’ (na primere epistolografii),” ibid., 21 (1984): 95–108; idem, Vizantijskoe obshchestvo XIII–XIV vekov po dannym epistolografii (Sverdlovsk, 1987). 6 Hatlie, “Redeeming Byzantine Epistolography,” 214; see ibid., 213–16, for a brief but useful summary of Smetanin’s theories. 7 See A. D. Karpozilos, “Realia in Byzantine Epistolography, X–XII c.,” BZ 77 (1984): 20–37; idem, “Realia in Byzantine Epistolography, XIII–XV c.,” ibid., 88 (1995): 68–84. 8 J. N. Ljubarskij, Mikhail Psell: Lichnost’ i tvorchestvo. K istorii vizantijskogo predgumanizma (Moscow, 1978). For a summary of other modern work on Psellos’s letters, see Hatlie, “Redeeming Byzantine Epistolography,” 241–43. 9 See G. Karlsson, Ide´ologie et ce´re´monial dans l’e´pistolographie byzantine, 2d ed. (Uppsala, 1962). 2
A. R. LITTLEWOOD
21
who has struck an exemplary balance between contemporary and modern criteria.10 Much, nevertheless, remains to be done, as has recently been pointed out by Peter Hatlie in his excellent survey of the trends, over the years, in the modern scholarship on Byzantine epistolography.11 A possibly fruitful approach is to consider letters written on the same subject both by the same writer and by different writers.12 The letter of consolation is chosen here for two reasons. It ought, because of its importance to the recipient, and in some cases to the writer as well, to demand of the latter his best efforts.13 At the same time it provides the author with a severe challenge—to show his literary artistry despite the difficulty of avoiding a mere unadorned and unintegrated parade of hackneyed quotations and paramythetic topoi, such as “death is common to all,” “death brings an end to earthly sufferings,” “the deceased is with God,” “it is impious to criticize God’s dispensation,” “excessive lamentation shames the deceased,” or “self-control is an example to others.” 14 This challenge could be forcibly shown by a comparison with the texts of cards of sympathy in our modern, much vaunted society. Religious cards rarely claim more than that God will comfort the bereaved, or assert the theologically unsound conviction that the 10 Of Mullett’s many articles on letters, those most concerned with literary aspects are “The Classical Tradition in the Byzantine Letter,” in Byzantium and the Classical Tradition, ed. M. E. Mullett and R. D. Scott (Birmingham, 1981), 75–93; “Writing in Early Medieval Byzantium,” in The Uses of Literacy in Early Medieval Europe, ed. R. McKitterick (Cambridge, 1990), 156–85, esp. 172–85; “The Language of Diplomacy,” in Byzantine Diplomacy: Papers from the Twenty-fourth Spring Symposium of Byzantine Studies, ed. J. Shepard and S. Franklin (Aldershot, 1992), 203–16; “Originality in the Byzantine Letter: The Case of Exile,” in Originality in Byzantine Literature, Art and Music, ed. A. R. Littlewood (Oxford, 1995), 39–58; and Theophylact of Ochrid. It is sad that even in the recent edition of The Letters of Ioannes Mauropous, Metropolitan of Euchaita, CFHB 34 (Thessalonike, 1990), Apostolos Karpozilos’s commentary barely touches on matters literary. 11 “Redeeming Byzantine Epistolography,” 213–48. As Hatlie points out (ibid., 230–31), new directions may be suggested by looking into the work of John White and his Ancient Epistolography Group, which itself considers the very early Byzantine centuries in its examination of the Semitic, Persian, Greek, and Roman epistolographic traditions. Hatlie also encourages (ibid., 231–34) the greater use of “computer-aided text analysis.” For bibliographies of primary sources and modern scholarship, see N. B. Tomadakes, Buzantinh` ejpistolografi´a: Eijsagwgh´ , kei´mena, kata´ logo" ejpistolografw'n, 3d ed. (Athens, 1969); V. A. Smetanin, Epistolografija: Metodicheskaja razrabotka k spetsial’nomu seminaru dlja studentov-zaochnikov Istoricheskogo fakul’teta (Sverdlovsk, 1970); T. V. Popova, “Vizantijskaja epistolografija,” in Vizantijskaja literatura, ed. C. C. Averintsev (Moscow, 1974); H. Hunger, Die hochsprachliche profane Literatur der Byzantiner (Munich, 1978), 1:234–39. 12 This was done for letters of exile by Mullett, “Originality,” while I briefly considered letters of consolation and letters to the emperor written by men awaiting trial, in “An ‘Ikon of the Soul’: The Byzantine Letter,” Visible Language 10 (1976): 217–19. Limited to a single author, some types of letter, including the consolatory, were long ago usefully compared by Sister M. Monica Wagner in “A Chapter in Byzantine Epistolography: The Letters of Theodoret of Cyrus,” DOP 4 (1948): 119–81. 13 See, however, below, p. 22. 14 Byzantine writers themselves were, of course, aware of this difficulty. An interesting early example is Julian’s letter to the recently widowed Himerios (Ep. 201 in L’empereur Julien: Oeuvres comple`tes, ed. J. Bidez [Paris, 1924], 1.2:229–31), in which the emperor admits that it would not be fitting to offer the usual condolences to an orator so adept himself in using them to instruct the bereaved who through ignorance lack the necessary self-control. He consequently relates a fable, which, he claims, may be unfamiliar to Himerios: the grief of Darios for his dead wife was so excessive that Demokritos saw fit to attempt to shame the king out of it by asking for the names of three persons who had never mourned for anyone. Even here, however, Julian cannot refrain from alluding to a famous Homeric passage (Odyssey 4.220–21; see below, note 79), while his tale is unlikely to have been unknown to Himerios since it had earlier appeared in Lucian, with the names of Herodes Attikos and Demonax (Demonactis vita 25).
22
THE BYZANTINE LETTER OF CONSOLATION
deceased is with God, while secular efforts rarely do more than either encourage happy memories or declare that although the sender cannot think of any adequate words, his or her thoughts are with the bereaved.15 As an indication of the sort of material that, I think, would be unearthed by a thorough examination of all surviving Byzantine letters of consolation, in this preliminary survey I deal only with the letters of the Macedonian and Komnenian periods that I have found adequately edited for my purpose;16 and I have chosen to examine their style primarily in terms of content rather than in terms of verbal expression. My selection of nineteen letters is, therefore, to some extent random and not in any way specifically designed to exhibit diversity of treatment. However, it is impossible to know how typical this sample is of the letters of consolation from the literati, since such letters are surprisingly uncommon in the epistolary collections (it is hard to believe that patriarchs in particular wrote so few).17 By my estimate only 1 to 2 percent of surviving Byzantine letters are consolatory. This could be an anomaly of survival, but it may also suggest one of two possibilities: either that consolation was conventionally offered orally, letters being a “second best” dictated by absence,18 or that many were written hastily, to meet the immediate need, and were then not considered of sufficient literary merit to be copied into collections by either writer or recipient. With only two exceptions,19 all the letters that I have selected are addressed to men; but the relationships of the deceased to the recipients are very varied—wife (in 2 cases), father (2), mother (2), brother (2), sister (2), friend (2), husband, son, daughter, brother-in-law, unidentified male relation by marriage, abbatial predecessor, and exiled priest. The Byzantines had many literary treatments of death, such as the ejpikh´ deion, the 15 The following are typical modern examples (ellipses indicate not my omissions but the actual punctuation on the cards): “May He walk with you during this difficult time”; “May God be with you in this time of sorrow . . . comforting you, guiding you, sending you warm reminders of His love”; “God knows the sadness you feel at losing one you loved so much, so let Him help to ease your burden with His tender, loving touch”; “May it comfort you to know that your loved one is safe in God’s care and that friends are praying for you in your time of sorrow”; “Even though our faith tells us that we should rejoice that your loved one is with God, it is still difficult for those left behind not to feel sad and lonely. May the peace and the comfort of friends see you through this difficult time”; “May the joys you knew help comfort you in this, your time of sorrow”; “Your loved one will always be remembered”; “Your loss is my sorrow too”; “Sharing in your sorrow and wanting you to know that this brings far more sympathy than words could ever show”; “May you somehow know the sympathy that words cannot express”; “I wish I knew the perfect thing to say right now. I can only tell you that you’re special, and that you’re in my thoughts now more than ever”; “So many thoughts and feelings fill our hearts right now, but the words are difficult to find. We want you to know that we care . . . and we’re so very sorry”; “At times like this, ‘I’m sorry’ sounds so empty, compared to what you’re going through. And yet, it’s all I know to say.” The above are a random selection, the complete message being given in each instance. Byzantine confession to aphasia (e.g., Photios Ep. 201) is ever but a temporary aberration. 16 Thus, unfortunately, I have not considered, inter alia, the corpus of letters by Michael Psellos. 17 For instance, among the 115 published letters of Athanasios I of Constantinople there is not a single letter of consolation. They seem to have been more common among the early church fathers: for instance, Basil wrote fourteen, and Theodoretos twelve; but there are only four consolation letters surviving of Gregory of Nazianzos (including one to Gregory of Nyssa on the death of the latter’s brother, Basil) and only three of John Chrysostomos (figures are taken from S. K. Stowers, Letter Writing in Greco-Roman Antiquity [Philadelphia, 1986], 148–49, 152). 18 I suspect that some letters (e.g., Epp. 47 and 156 of Nikolaos Mystikos, discussed below, pp. 27–28) may have been written to court officials, and especially members of the imperial family, in addition to oral consolation in the self-preservatory desire to exercise tact and diplomacy in dealing with rivals or superiors. 19 Ep. 245 of Photios (below, pp. 25–26) is addressed to an abbess, and Ep. 5 of Gregory of Oxeia (below, p. 32), to a daughter of John II Komnenos.
A. R. LITTLEWOOD
23
ejpita´ fio" lo´ go", the qrh'no", the monvdi´a, and the paramuqhtiko` " lo´ go", all of which were interrelated, overlapped with each other, and sometimes replaced each other.20 Their subject matter was a combination of Christian and pagan elements. Although the latter have a very long history, with ancient authors tracing them back to Achilles’ words of comfort addressed to Priam,21 the Byzantines chiefly employed the rhetorical formulations crystallized in the late third century A.D. by Menander of Laodikeia on the Lykos.22 The letter of consolation, however, stands somewhat apart from such literary treatments: they were public, it generally private; they were addressed to a group, it usually to an individual;23 they commonly celebrated an important public figure, it usually consoled a man24 on the loss of a parent, wife, child, or sibling; they were principally encomia of the deceased delivered at funerals and memorial services, it dealt principally with the feelings of the bereaved. The Byzantine letter of consolation shares much of its Christian content with the other funerary genres, but places a greater emphasis, especially in the early centuries, on 1 Thessalonians 4:13–18. The first two verses of this passage were much quoted: “But I would not have you to be ignorant, brethren, concerning them which are asleep, that ye sorrow not, even as others which have no hope. For if we believe that Jesus died and rose again, even so them also which sleep in Jesus will God bring with him.” The final words, “Wherefore comfort one another with these words,” were naturally taken as a divine directive to use the passage.25 For its non-Christian content26 the Byzantine letter of consolation, unlike the other genres, depends not so much on the Menandrian tradition as on Stoic (especially Chrysippan) and Cynic theories.27 These ultimately derive from the now lost work On Grief (Peri` Pe´ nqou") of the early Academic philosopher Crantor of Soli, which the Stoic Panaitios recommended be learned by heart. Philosophers, and during the Second Sophistic rhetoricians as well, considered it part of their practical duty to give oral or, if that were not possible, written consolation to their less fortunate fellow citizens. Dio Chrysostomos indeed claims that many men of their own accord invited philosophers “to come and speak See, further, M. Alexiou, The Ritual Lament in Greek Tradition (Cambridge, 1974). Homer Iliad 24.507–51. In general, see K. Buresch, Consolationum a graecis romanisque scriptarum historia critica (Leipzig, 1886); and R. Kassel, Untersuchungen zur griechischen und ro¨mischen Konsolationsliteratur (Munich, 1958). 22 Menander Rhetor, ed. D. A. Russell and N. G. Wilson (Oxford, 1981), 2.9, pp. 160–65; 2.11, pp. 170–79; 2.16, pp. 200–207. 23 Doubtless, the letters of consolation were often read to other relatives at the time and subsequently perhaps to friends at a home theatron, but in the first instance they were intended for a specific individual, as the contents often make very clear (e.g., “you knew when you married her that she was mortal,” from a letter [Ep. 156] of Nikolaos Mystikos, on which see below, pp. 27–28). 24 Far less common are letters of consolation addressed to women, but see Ep. 245 of Photios (below, pp. 25–26), Ep. 5 of Gregory of Oxeia (below, p. 32), Plutarch’s letter to his own wife (below, note 70), and Basil’s letter to the wife of Nektarios (below, note 89). 25 As A. J. Malherbe observes, “To the Church Fathers, who knew the classical genres, this section of the letter appeared close to a consolatory epistle” (“Exhortation in I Thess.,” NT 25 [1983]: 254). It is one of the few passages to be found in two letters of this selection (see below, note 79). On Christian aspects of consolation in general, see R. C. Gregg, Consolation Philosophy (Cambridge, Mass., 1975). 26 The two sides, Christian and non-Christian, do, of course, have certain things in common, of which Malherbe gives a brief summary (“Exhortation,” 255). 27 See, in particular, A. J. Malherbe, The Cynic Epistles (Missoula, Mont., 1977). 20 21
24
THE BYZANTINE LETTER OF CONSOLATION
comforting words” upon the death of a relative.28 The professed experts thus built up a number of arguments,29 which may be found employed by classical authors such as Cicero and Plutarch30 and later by the Byzantines. It is, however, in a short treatise of the second or first century B.C. on the different types of letter, ascribed to a certain Demetrios, that we find the basic format of the letter of consolation, which, like many of the later Byzantine letters, begins with the sumpa´ qeia and proceeds to the parai´nesi". The author’s model letter, dealing not just with bereavement but with any calamity, is as follows: “When I heard of the terrible things that you met at the hands of thankless Fate, I felt the deepest grief, considering that what had happened had not happened to you more than to me. When I saw all the things that assail life, all that day long I cried over them. But then I considered that such things are the common lot of all, with nature establishing neither a particular time nor age in which one must suffer anything, but often confronting us secretly, awkwardly and undeservedly. Since I did not happen to be present to comfort you, I decided to do so by letter. Bear, then, what has happened as lightly as you can, and exhort yourself just as you would exhort someone else. For you know that reason will make it easier for you to be relieved of your grief with the passage of time.” 31 We may now look at our selection of Byzantine letters.32 To consider their writers in roughly chronological order, we begin with five letters by the scholarly Photios that were perhaps all written in the period between his two patriarchates yet exhibit huge differences in both content and tone.33 The first letter (Ep. 234) is a lengthy consolation that Photios sent his brother Tarasios who had lost his daughter in the bloom of her youth and ripe for marriage. Photios emits a cri de coeur for Elijah, Elisha, Peter, and Paul—all of whom, the Bible says, brought the dead to life—to open a lengthy and dolorous exordium in which he gruesomely describes the once comely girl’s awful appearance in death, her lips “contracted for dissolution,” her eyes “having poured out their vital stream, cover[ing] over what remains with inani28 Dio is complaining that men treat philosophy like medicine, the rich not seeing fit to employ a philosopher until they have encountered misfortune, which alone renders them “tolerant of philosophers’ words and admitting that they need comfort” (Or. 27.8–9). See also Julian’s letter to Himerios (above, note 14). 29 Stowers finds (Letter Writing, 142) the most common arguments to be as follows: “1. Death is inevitable. 2. Death is the fate of all, kings and beggars, rich and poor. 3. The person’s memory and honor will live on in spite of death. 4. Death releases one from the evils of life. 5. The funeral and the tomb are a great honor to the deceased. 6. Either death is nonexistence and does not occur to the dead or it leads to some happier state.” In respect to this last argument, it should be pointed out that some pagan writers (e.g., Seneca in Consolatio ad Marciam 25.1, 26.3) come closer to Christian, and therefore Byzantine, belief in claiming that the deceased is now in the company of the blessed souls. Stowers warningly goes on (Letter Writing, 144) to note Ep. 164 of Gregory of Nazianzos, which advises that philosophical arguments are to be reserved for the educated, others being offered instead sympathy, exhortation, and often a rebuke for their despair. 30 E.g., Cicero Tusculanae disputationes; Plutarch Moralia 101F–22A, 608B–12B. 31 Translated by Stowers, Letter Writing, 144. Stowers gives further translations of six real letters of consolation from the 1st (?) to the 6th century A.D., including three from papyri. 32 All summaries and translations of letters in the selection are my own. 33 Photii Patriarchae Constantinopolitani Epistulae et Amphilochia, ed. B. Laourdas and L. G. Westerink, 6 vols. (Leipzig, 1983–88), 2. The five consolatory letters have been rendered into English by D. S. White, Patriarch Photios of Constantinople: His Life, Scholarly Contributions, and Correspondence Together with a Translation of Fifty-two of His Letters (Brookline, Mass., 1981), nos. 1–5, pp. 115–35. Her translations, with brief notes, of Epp. 234 and 245 had earlier appeared in GOTR 18 (1973): 47–58, and Classical Folia 29 (1975): 31–43, respectively.
A. R. LITTLEWOOD
25
mate lids.” He then admits that he has been carried away by grief, in itself a thing of destructive nature and one with which they would insult their ancestors’ patient endurance of even more cruel deaths. He proceeds to find and give comfort by reflecting that death is common to all and that there are worse deaths than that of this girl who expired in her parents’ arms and left no children to grieve for her. There follow a philosophical argument that longevity or brevity of life is irrelevant to happiness, and a theological argument that her youthful death not only is God’s will but also enables her to pass into immortal life with fewer bodily defilements. In a touching section Photios then visualizes the girl herself comforting her parents: “Why do you strike yourself in grief, my father, why do you bewail me as if I have departed to evils? Paradise has been allotted to me as my abode, a sweet sight to gaze upon with my eyes, and a sweeter thing to enjoy, and my experience of it exceeds all belief. . . . We are all wise [there] in divine and heavenly wisdom . . . and our life is all feast and festival. . . . Some day you will go there too together with my dear mother. . . .”34 Photios continues by urging his brother “not to give way to lamentation, for men must set a good example to women,35 or where will they get their comfort? . . . We must not act like women.” He ends with numerous scriptural arguments and a prayer that they may be courageous.36 Ep. 245 was written for an abbess named Eusebia, and is Photios’s sole surviving letter addressed to a woman.37 It is an extraordinary missive and was presumably well merited since otherwise it would have been highly insulting to a recipient in her position. To the complete exclusion of all other topics commonly found in letters to the bereaved, its 138 lines are devoted entirely to deploring excessive lamentation. It may be briefly paraphrased thus: “Since we are not the first human beings, we cannot be upset by an unexpected occurrence. You complain that your sister, your ‘sole consolation after God, the comfort of [your] sorrows, the dissolution of [your] sufferings, the prime cause of [your] joys, has abandoned [you],’ but so have your parents and all your ancestors back to Adam, and you yourself will abandon others. You should not be upset that she has now been released from corruption. She has not disappeared into non-existence but is at peace with her Maker. Why should we be angry against the law of nature? Plants, animals, and even the sun, which traverses the heavens ‘like a happy giant,’ are mortal. Will you risk divine wrath in lamenting God’s gift of immortality to his handiwork? By sullenly bewailing the fact that your sister’s soul is now being rejoicingly received by the angels, you are disgracing your marriage with Christ, your immortal bridegroom, for whom you 34 The passage even includes a verbal reminiscence of a line in the opening lament, surely meant to console the writer who had bewailed the fact that the crooked serpent had crept into Paradise and who is now assured that it is no longer there. This reminiscence is also improved by variatio, for, while Photios asks pw'" eij" to` n para´ deison ei”rpusen oJ ponhro` " ejkei'no" kai` skolio` " o“fi"; (ll. 13–14), the deceased says ajlla` nu'n me` n oujde` oJ skolio` " ejkei'no" kai` ponhro` " o“fi" cw´ ran e“cei uJferpu´ sai (ll. 134–35). 35 The need to set a good example, be it to a specific individual, to women in general, or even to everyone, had become a topos long before Photios’s time. An early example specifically in a work of consolation occurs in Seneca Consolatio ad Polybium 5.4 f. 36 This too is an ancient topos. The most famous examples are probably the words that Propertius puts into the mouth of the dead Cornelia when she bids her surviving husband weep for her in private but “deceive [their children’s] kisses with dry cheeks” (4.11.79–84), and the Younger Pliny’s eulogy of Arria (Ep. 3.16.3–6) who cheerfully reported to her seriously sick husband Caecina Paetus the imaginary convalescence of their son, whose funeral she was at that time having to arrange on her own. 37 Out of a total of 299 letters.
26
THE BYZANTINE LETTER OF CONSOLATION
abandoned all family ties. Mourning would have been legitimate if your sister had been a sinner, but she was a pious virgin and you are insulting her. Cease your lamentation and immerse yourself in ‘the blessed and indestructible love of Christ, your pure and immaculate bridegroom’!” In Ep. 201 Photios claims that he wished to write in consolation to George, metropolitan of Nikomedia, as soon as he heard of the death of their mutual friend, a priest who had died in exile and had clearly been a member of the persecuted Photian party; initially, however, he found that he was unable to do so owing to his own sorrow.38 He then reflects that though the death is like the ancient offering of first-fruits to God, yet the root remains to produce further fruit. They should not take it amiss that in dying young their friend beat them in the race to attain virtue. That he died amid hardships and persecution is consoling, for an athlete should be tested “amid struggles . . . and the bloodthirstiness of his persecutors” and “appear before the judge still dripping in the sweat of the contest.” He was the first offering, chosen for his outspokenness, and his release from toils must not be mourned since his virtues now have their reward. He died before being able to enjoy the restoration of the church (i.e., the Photian party), but the attainment of earthly happiness precludes the recompense of heavenly blessedness; nevertheless, by his closeness to God he may now perhaps be able to expedite that restoration. Photios ends with the hope that George may take additional consolation in the knowledge that their friend’s successes depended upon George’s “instruction and zeal.” Another George39 must have been startled to read the opening words of his letter from the patriarch (Ep. 105): “I forbid you, best of friends, to be called anymore my friend.” Photios’s explanation is that a mutual friend has deserted his earthly friends for eternal ones, and that, since he himself has thus yet again been grieved by the death of a friend who meant much to him, he should imitate Timon: as Timon, perhaps after meeting with unusual human wickedness, called himself a misanthrope, so Photios should now become a “misophile.” He then rails against the source of his sorrow (o”sti" pote´ ejstin ejkei'no"), a plaint that appears at first to verge on blasphemy until we learn that this same source may quite soon put an end to him and drive him away to God for whom he yearns. The letter ends with a plea to George, who, Photios claims, is the sole remaining comfort of his life, to attend to the rites in his place and protect and lighten the sorrow of the relatives. The letter is unusual for this survey both in that it alone appears to break the news of the death to the recipient and in that the sole consolation contained therein is that to be extracted by the writer from the recipient. The final letter of consolation (if it can be so classified) by Photios (Ep. 131) is by far the shortest, a mere three lines addressed to his brother Tarasios on the death of a mutual friend. In full it runs thus: “Our friend has died in the body; but not even you would deny that he left behind his goodness (ajreth´ ) throughout his life as an imperishable monument. Shall we, then, call anybody happy at all if we take thought to bewail him?”
38 ¨ther (PhoWesterink (Laourdas and Westerink, Photii, 2:98) argues rightly, I think, against J. Hergenro tius, Patriarch von Constantinopel: Sein Leben, seine Schriften und das griechische Schisma, vol. 2 [Regensburg, 1867], 267–71) in maintaining that this letter was written before rather than after Photios’s recall from exile. 39 Identified in the lemma simply as “deacon and koubikoularios,” he was possibly, as Westerink (Laourdas and Westerink, Photii, 1:144) suggests, the same as the “deacon and orphanotrophos” of Ep. 136.
A. R. LITTLEWOOD
27
The letters of another well-educated patriarch, Photios’s friend Nikolaos Mystikos,40 are of especial interest in that we know something of their recipients. The first (Ep. 47)41 is addressed to Constantine, a Paphlagonian eunuch who had been castrated in his infancy by his father in the hope of a successful career in the civil service, which was triumphantly fulfilled by his appointment in 908 as Grand Chamberlain (parakoimomenos), a position he held until 919 except during the brief interlude of Alexander’s reign. Since Constantine was devoted to Nikolaos’s bitter foe Zoe Karbonopsina, and inherited much of Nikolaos’s power when Zoe seized control from the council of regency in 914, he was certainly the patriarch’s most important opponent after the empress herself. There exist two versions of a consolatio sent by Nikolaos to the chamberlain on the death of the latter’s sister, perhaps the wife of Leo Phokas. The first version, which was left unfinished,42 is principally a lengthy lamentation over the bitter event, which temporarily robbed the patriarch of voice, hearing, and even reasoning powers. Nikolaos finally goes on to say that he himself took comfort from reflecting on the dispensation of the Creator, who orders all things for the best, expresses his hope that Constantine’s sister will reach Paradise, and exhorts the chamberlain to terminate his own natural grief. The second version is far less personal: lamentation and what can almost be described as self-pity have all but vanished, and in their place are found further arguments and an elaboration, with many examples, on the theme of the dispensation of the Creator. Why are there two versions? Since Nikolaos’s relations with Constantine must have been strained, to say the least, he probably found this letter rather hard to write, despite the vast experience with this sort of thing that was surely a concomitant of his lofty ecclesiastical office. It is, therefore, not unreasonable to suppose that in an attempt to carry out what he held as his humane and Christian duty he emphasized his own personal involvement in the sad event, but before sending his letter had the good sense to realize that he had gone too far and that his protestations would smack of insincerity—perhaps even that they were a little unbecoming for a patriarch. Accordingly, he composed and sent a more restrained and dignified letter of condolence that was less likely to jar on the sensibilities of the chamberlain, but by some chance kept his earlier attempt, which found its way into a collection of his letters and has been thus fortuitously preserved as indication of Nikolaos’s personal difficulties in writing the letter.43 The recipient of the second letter (Ep. 156) is Romanos I Lekapenos, who lost his wife Theodora in 922. The opening remarks may be thus summarized: “God has been good even in this, for, if such a thing had to happen, it was far better that you should For a brief summary of recent work on this letter collection, see Hatlie, “Redeeming Byzantine Epistolography,” 239–40. 41 Nicholas I, Patriarch of Constantinople: Letters, ed. R. J. H. Jenkins and L. G. Westerink, CFHB 6 (Washington, D.C., 1973). 42 It concludes, in the only manuscript that contains it (Cod. Patmiacus 178), with the last few sentences of the second version, itself preserved only in Cod. Vind. Phil. gr. 342. 43 Jenkins believed that the second version was written first (“A ‘Consolatio’ of the Patriarch Nicholas Mysticus,” Byzantion 35 [1965]: 159–66), but J. F. Mitchell showed convincingly that this was not so, on the grounds that the juncture of the first version with the conclusion of the second is both ungrammatical and nonsensical and that it involves repetition of ideas (“A Consolatio of the Patriarch Nicholas Mysticus: Further Remarks,” ibid., 37 [1967]: 136–42). Westerink believed that the first version “was probably left unfinished because of the negative tone of its first paragraph” (Jenkins and Westerink, Nicholas I, 548). 40
28
THE BYZANTINE LETTER OF CONSOLATION
have to bear the grief rather than she, who in her womanly weakness would have given way to blasphemy. Let us thank the Lord who cares for those who trust in him, amongst whose number you are! Let us cast aside our grief lest God condemn us for our mutterings of complaint (goggu´ zonte").” Mention of the resignation of David is followed by two classical examples, which lead the writer back to Romanos: “Therefore, do nothing unworthy of yourself, but, since you knew when you married her that she was mortal and you have enjoyed her love, do not be downcast and sullen (kathfh` " kai` skuqrwpa´ zwn) as if you have suffered an injustice from God. Is it not a fearful thing to imagine ourselves punished by God when he has raised you to your sacred position? Let us not grieve any longer, I beg you, but let us rather thank and glorify him in this as in other matters and trust that he will increase his benefits if we strive to serve him with all our strength.” From this hard-hitting exhortation it is not unnatural to assume that Romanos, who appears to have been devoted to his wife, momentarily gave way to a grief that was considered hardly consonant with his eminence and which prompted Nikolaos to express himself with such bluntness, and perhaps gave him the opportunity vainly to attempt to establish a position of psychological ascendancy over a remarkable emperor.44 The third consolatory letter of Nikolaos (Ep. 46) was sent to a man whom the patriarch did not know personally, Gurgen II of Abasgia, whose father, Constantine III, died in 915/6. Here the condolence is perfunctory: in the very first sentence, despite its only moderate length, Nikolaos manages to express not only his own grief at the death of Constantine, but also sympathy toward his son and successor and praise of the widely known virtues of the deceased. He adds at greater length that he himself is consoled by the fact that Gurgen is the image of his father whom, the patriarch prays, he may even outshine. Before closing the letter with an item of business, Nikolaos mentions that he is sending the king a new robe (as token of Byzantine recognition of his accession). This letter is simply a piece of diplomacy; Nikolaos makes no attempt to pretend to a personal relationship where none existed. It is interesting to compare the letters by the two patriarchs, both of whom had enjoyed secular careers prior to their elevation, with one by a contemporary and, again, scholarly archbishop. As can well be imagined, Arethas has little time for softness and sentiment. In a letter to Kosmas Magistros45 he dwells at length upon the superior and blessed state of death, a belief for which he has both classical and scriptural authority, and ends with a demand to Kosmas to refrain from lamentation as an example to the other relations of the deceased. He is far more distant than Nikolaos (even when allowance has been made for his contorted style), and his tone suggests that he would have little symThe tone of this letter may be compared with that of a much milder imperial consolation, which was probably sent by Gabriel, metropolitan of Thessalonike, most likely to John VII Palaiologos on the death of his young son sometime during the opening years of the 15th century (for the text, summary, and discussion, ¨ BG 16 see G. T. Dennis, “An Unknown Byzantine Emperor, Andronicus V Palaeologus [1400–1407?],” JO [1967]: 175–87). The writer grieves at the death of the son who he had prayed would one day succeed his father, expresses sorrow for his own absence, claims that it is his duty to write a consoling letter to the emperor despite the fact that the latter is not so small-minded as to require it, envisions the son by the throne of God, and adduces theologically comforting biblical passages and tales of the saints, from whom paradeigmatic lessons may be learned in just the way that artists use their models. For further letters of consolation sent to members of the imperial family, see below, pp. 31, 32. 45 Arethae archiepiscopi Caesariensis Scripta minora, ed. L. G. Westerink (Leipzig, 1968), 1: no. 22. 44
A. R. LITTLEWOOD
29
pathy for any weakness; we must hope that he knew his man, but in any case his letter allows us to know him a little better. A world away in tone is a brief letter of consolation by an anonymous protonotarios, possibly Symeon Metaphrastes,46 to a friend who has lost his mother. He begins by assuring the mourner that the blow he has suffered is enough to vanquish a soul of adamant, and “thus it is no wonder that you are in this state since you have lost your mother who was your pride and comfort” and who made “our life of pain and peril seem to you all sweet and kind.” Having thus gently excused his friend’s grief and allowed his mind to linger on his mother’s virtues, the writer reminds him firmly that they (the sympathetic use of the first person plural should be noted here, as so often elsewhere) are not the first to suffer such a loss, since it is of human nature to die. He then directs the attention of the bereaved to his mother’s placid, dignified, and almost happy death, which should be an inspiration to him. Our anonymous author ends thus: “And be yourself an example even to others of fortitude and endurance in the face of your sorrows, and let all marvel at your courage and congratulate you on your stoutheartedness and nobility, and your mother on being blessed in offspring and on her dignified mien.” Thus, in only twentyeight lines, his friend has been brought on a gentle but firm rein from helpless despair, with tears almost openly encouraged, through bitter-sweet memories to a manly resolve that is forged by his mother’s noble end and is perhaps reached also for her sake. This little letter was not written in haste or thoughtlessly 47 and does great credit to its author’s tact and sensibilities. There survives from the tenth century another letter composed to console a man sorrowing at the death of his mother.48 It was written by Theodore, bishop of Nicaea, to the protospatharios Leo, and may be summarized as follows: “You grieve much at the death of your mother, but not unreasonably, for how could you not since you lived with her for so long? She had borne her widowhood of nearly forty years with exemplary austerity and piety. Now you have nobody to whom to turn, neither father, wife, nor children. Even a man of adamant would be overcome by grief in this situation, yet your mother would not want you to be downcast because of her. So put off your excessive sadness, show yourself worthy of her, and console your brother who is not as strong in the face of afflictions, for you will thus please her, be an instructive example to others, and not act contrary to God’s will.” Despite similarities with the previous letter, this one exhibits some notable differences, for the good bishop is at pains to consider the specific situation of the bereaved: he indicates his sympathy for what was clearly excessive grief by expatiating not only on the mother’s virtues but also on the comforts that would in normal circumstances have been available from other members of the family, before he concludes with a demand for a more manly attitude. The last of our consolatory letters from the Macedonian period is the most overtly rhetorical, and approaches nearest to what in an author of the Greco-Roman world 46 Ed. S. P. Lampros, “ jEpistolai` ejk tou' Biennai´ou kw´ diko" Phil. Gr. 342,” Ne´ o" JEll. 21 (1927): 27–28. On the authorship, see J. Darrouze`s, “Inventaire des ´epistoliers byzantins du Xe sie`cle,” REB 18 (1960): 128. 47 A good example of the care taken in the letter is the repetition of the word ajpokrou´ w, not the commonest of words, which occurs both when the writer reminds the son of how his mother comforted him during her life and when he urges the son to take inspiration from his mother after her death. 48 ´ pistoliers byzantins du Xe sie`cle, ed. J. Darrouze`s, AOC 6 (Paris, 1960), 276–77. Ep. 6 in E
30
THE BYZANTINE LETTER OF CONSOLATION
would be termed the Grand Style. It was written by Philetos of Synada49 to Nikephoros Balanites, the strategos of Melitene in Cappadocia, on the death of the latter’s son. Philetos assures his friend that he can understand his grief, since he is a son of a father if not a father himself, but that death is the common lot of man. Nikephoros must make no ignoble50 utterance, must not say that his son was snatched away “in the very bloom of youth just when the down was appearing on his cheeks”; instead he must accept the unfathomable decisions of God, must take David, the mother of the Maccabees, and Job as his examples, must be happy that he has sent his son ahead to God, who desires such a soul released from mortal toils. For another example Philetos suggests himself who lost his father at birth and his mother in his infancy, and then (in language reminiscent of Hellenistic epigrams) “my only brother, in whom I had placed all my hopes, I suddenly saw by a gravestone hidden” (ejp∆ ajdelfv' de` mo´ nv ta` " o”la" ejlpi´da" e“cwn, a“fnw kai` aujto` n ei«don . . . li´qv kai` ta´ fv krupto´ menon). He concludes with a reminder that no good will accrue from grief. In this letter one notices Philetos’s insistent involvement in his friend’s sorrow,51 which is marked by the anaphora in the first three sentences, each beginning with a dominant oi«da (“I know”); in a nice touch the very last word, oijo´ meqa (“we believe”), is clearly intended, although a different verb, to recall the first word—but now, after the description of the losses that both have experienced, the sympathetic singular has been changed into a united plural.52 Before we turn to the Komnenian period, notice should be taken of another letter in which the author does not offer consolation to the bereaved but instead both expresses his delight that the report he has had of the recipient’s death has proved false and recalls his own earlier thoughts and emotions. It was written by an anonymous schoolteacher (the so-called Anonymus Londiniensis) to a metropolitan of Neokaisareia.53 A summary demonstrates well the Byzantines’ ambivalent attitude toward death: “The news of your death was not unexpected, but, thank God, proved to be untrue. I did not know whether to be sad or joyful. I was sad when I thought of your good character, but joyful when I thought of the rewards that await a life well lived and of your unassailable reputation. When we thought you dead, such were the considerations that moved us. Now that we know you to be alive, they move us no less. May I see you soon.” 54 We may now consider examples from the Komnenian period. The three surviving Ep. 4 in ibid., 251–53. It must be hoped that Philetos intended no pun, and that his friend saw none, in using the word ajgenh´ ", for which “childless” is also an attested meaning. 51 Compare with the letter of the anonymous protonotarios (above, p. 29). Wagner noted Theodoretos’s “personal share in the sorrow of his friends” in his condolatory letters (“A Chapter,” 160 and n. 133). 52 In any study of this genre in the Macedonian period, one should also note a second letter by Philetos ´ pistoliers byzantins, 249–50), wherein a boy briefly laments the death of his mother and (Ep. 1 in Darrouze`s, E requests consolation. Darrouze`s, arguing from the fact that this contradicts Ep. 4 where Philetos mentions his mother’s much earlier death, concludes that this lament is “soit d’un enfant en ge´ne´ral soit d’un particulier au nom duquel parle Phile´tos: c’est une lettre fictive ou un exercice de rhe´torique” (ibid., 252 n. 2). The letter opens with apostrophes of the mother and the grave and closes with a specific message that gives it at least the appearance of authenticity (even if it was written by Philetos on behalf of another); but, whether genuine or fictitious, it nevertheless represents contemporary Byzantine attitudes toward bereavement. 53 Ep. 73 in “The Correspondence of a Tenth-Century Byzantine Scholar,” ed. R. Browning, Byzantion 24 (1954): 447. 54 Ibid., 416. 49 50
A. R. LITTLEWOOD
31
consolatory letters of Theophylact of Ochrid 55 exhibit great dissimilarities, but, in view of Mullett’s study,56 the discussion here will be brief. His most regular consolatio is that to Michael Pantechnes on the death of his father (Ep. 39). The writer opens with the usual lamentation, but then insists that though the earth has lost a star, yet God has gained a sun: since the merits of the deceased are worthy of heaven, “why do you grieve that they have gone to their proper abode?” It would be selfish, he claims, for Michael to begrudge his father the blessedness of heaven simply because he enjoyed the latter’s presence on earth. In the last dozen lines Theophylact encourages his friend in his personal knowledge that the father’s virtues are continued in the son. Still following the normal forms of such letters, but much briefer (only thirteen lines long in Paul Gautier’s edition), is a consolation on the death of Michael Psellos that was sent to the latter’s brother (Ep. 132). Theophylact’s words of sorrow shared sound more stark than is usual: “I am not unaware that you are in pain, . . . but I too grieve, smitten by the pangs of friendship.” His sorrow is doubled because through distance he can give only epistolary comfort by reminding that Michael is not dead but has gone to live with God. His last words, “you are not the only one to know his life, but we all know what sort of a man he was,” may be, as Mullett opines, a curt ending because Psellos was too great a man to be mourned “in a more obviously eloquent—or personal—manner,” but they could equally be a disguised criticism of the recipient’s unwillingness to share his brother with the world, especially if taken in conjunction with the opening words.57 Very different from both these letters is one addressed to the caesar Nikephoros Melissenos (Ep. 73). This is a fairly lengthy piece of diplomatic tact. The consolation itself is brief, for Theophylact may have deemed it improper to make any remarks that the caesar could possibly have found patronizing (the tone is quite the opposite of that exhibited by Nikolaos Mystikos toward Romanos I). The opening has a unique twist: the archbishop hopes that as he himself takes consolation in letters from Melissenos, so will Melissenos, in his sorrow at the death of his brother-in-law, the sebastokrator Isaac Komnenos, take consolation in God. Then, as if abashed at making so plain the relationship between writer and recipient of the letter, Theophylact proceeds to pose the question of Melissenos’s preparations for his own death, but this is done with extreme delicacy. The letter ends, rather incongruously, with the hope that the accompanying gift of fish will be good for both the caesar’s appetite and his spiritual health.58 A fourth letter of Theophylact (Ep. 37) reacts to the news of a death, but it can hardly be classed as a letter of consolation since the archbishop makes little attempt to console.59 Addressing Symeon, his spiritual father and the new abbot of a monastery at Anaplous on the European shore of the Bosporos,60 Theophylact expresses his pleasure at the death of The´ophylacte d’Achrida: Lettres, ed. P. Gautier, CFHB 16.2 (Thessalonike, 1986). Theophylact, 138–44. 57 Ibid., 143. Psellos is not otherwise known to have had a brother; for a discussion of the problem, see Gautier, The´ophylacte, 113–15. 58 Theophylact sent either fifty or five hundred fish. On an earlier occasion he had sent Melissenos two hundred salted fish (Ep. 13). 59 Mullett indeed calls it “a celebration rather than a consolation” (Theophylact, 138). 60 This may be the monastery of Philotheos; see Gautier, The´ophylacte, 252 n. 1, and R. Janin, La ge´ographie eccle´siastique de l’Empire byzantin, vol. 1, Le sie`ge de Constantinople et le patriarcat œcume´nique, pt. 3, Les ´eglises et les monaste`res, 2d ed. (Paris, 1969), 494. 55 56
32
THE BYZANTINE LETTER OF CONSOLATION
his correspondent’s spiritual father and predecessor, whose desire to become a hermit he found troubling. Here we can therefore see a critical type of honesty that rarely appears in such letters. About half a century later Gregory, abbot of Oxeia, writes a letter of condolence to Alexios Komnenos Euphorbenos upon the death of his wife (Ep. 2).61 Taking advantage of the letter to send both greeting and consolation, he quotes the Thirty-fifth Psalm to the effect that Alexios must put the fear of God before his eyes in accepting what has happened, which he finds terrible to say but which is necessary for his Majesty to hear. He reminds Alexios of his wife’s sufferings and asks how he could wish her to suffer yet more on earth. There follows a rare element in Byzantine letters, a description, albeit brief, of the funeral, after which Gregory assures Alexios that his wife’s soul is now in heaven and concludes with the hope that they may both attain the crown eternal. Despite his deference he is far blunter than Theophylact in the letter to the caesar Nikephoros Melissenos without approaching the peremptory tone of Nikolaos Mystikos’s hard-hitting advice to Romanos I Lekapenos.62 There is a second letter of condolence by Gregory to a member of the imperial family (Ep. 5), in this case Theodora, the third daughter of John II Komnenos, who had lost her husband, the gloriously styled Panhyperprotosebastohypertatos Manuel Anemas. Every sinner who genuinely repents on his deathbed, Gregory begins, will be saved, and one who dies “in the angelic profession” of monasticism will be assuredly restored by God to the image of the Creator; but he adds delicately that he knows that letters of consolation can be as irritating to grieving spirits as “the softest cloth to an inflamed eye.” 63 He proceeds to expatiate on man’s tendency to think in human terms, and on the uncertainty of human life, which flows continually but not always calmly. Quoting the apostle Paul amid an avalanche of quotations, he reminds Theodora that suffering brings patience, and patience testing: “He who runs from the suffering runs from the testing.” 64 She should not be distressed at what has happened, for Abraham, the prophets, and even Christ himself have all died. Our abbot ends with a lengthy apostrophe to the dead husband who now enjoys the numerous blessings of heaven and is himself blessed in the eyes of Gregory and his fellow monks for all his benefactions to them. To conclude our selection, John Tzetzes provides a remarkable letter of condolence65 addressed to Leo Charsianites upon the death of the latter’s brother. In all types of Byzantine letters may biblical and classical allusions be found, appearing on their own, mixed together, or in separate sections (although letter-writers had the good taste and courtesy to indulge in classical allusions only in correspondence with those who had the learning to appreciate them).66 In a truly Christian society, however, as Byzantium was, comfort 61 “Les lettres de Gre´goire, higoume`ne d’Oxia,” ed. P. Gautier, REB 31 (1973): 203–27. On the identification of this Alexios Komnenos, see ibid., 206–8. 62 Above, pp. 27–28. 63 This is a quotation from the opening sentence of Basil’s letter of condolence to the wife of Nektarios (Ep. 6; see below, note 89). 64 Rom. 5:4. 65 Ep. 38 in Ioannis Tzetzae Epistulae, ed. P. A. M. Leone (Leipzig, 1972). 66 This can be seen best in the epistolary collections of churchmen, many of whose correspondents had little or no classical training; see further Mullett, “The Classical Tradition,” 92, and A. R. Littlewood, “A Statistical Survey of the Incidence of Repeated Quotations in Selected Byzantine Letter-Writers,” in Duffy
A. R. LITTLEWOOD
33
for the bereaved must be drawn largely from Christian convictions—yet this letter of Tzetzes is completely devoid of biblical allusions while at the same time drawing extensively on pagan writers, such as Homer, Aeschylus, Sophocles, Euripides, Plutarch, and Plato. Another remarkable feature of this letter is Tzetzes’ justification of the immoderate grief with which, in traditional manner, he opens his letter. He tells the tale of Crassus who, when taunted by his political opponent L. Domitius Ahenobarbus for burying and mourning a pet moray eel, retorted that, though he had indeed done that, yet Domitius had buried three wives without a tear: if Crassus had thus bewailed a single eel, how could he, Tzetzes, not mourn a much greater loss? While there appears to be no clear difference between the Macedonian and the Komnenian letters, our examples are sufficient to show that in both periods, despite a general compositional tripartite form of lamentation, consolation, and encouragement, there is yet a surprisingly wide range of subjects and tone that demonstrates remarkable variatio in a highly circumscribed subgenre.67 Our surviving letters, of course, are not necessarily typical. They were written by, and largely for, members of the intelligentsia who delighted in their precious artistry. They were considered worthy of preservation, not in a desk drawer to be read over by some poor aged widow or mother hoping to recall the sympathy offered by her friends as she continued to grieve, but for their own inherent merit. They were sent mainly to men and kept, at least in part, as pieces of literary art, and perhaps read aloud to friends who admired them and desired copies—or how else should there be so many, and found so often in more than one manuscript? How effective, however, were these letters in their primary purpose, which is to console? Unless we believe that Byzantine epistolographers were all Thespians indulging themselves in an enormous charade completely divorced from reality (and the only inference to be drawn from the remarks of some scholars is that they think that they were), we must accept the supposition that recipients delighted in receiving letters—there is assuredly an abundance, indeed a superabundance, not only of thanks for them but also of expressions of joy at them. One example must suffice. Symeon Metaphrastes writes to a friend: “When your letter reached me, these worries were dissipated like the shadows of dreams after awakening. When I got it into my hands, I loosed its fastenings and immediately looked at its length, just as the thirsty gaze at the size of the cup before drinking; then, slowly, dwelling on every syllable, I read it, prolonging for myself the pleasure and desiring not to stop the cause of my pleasure until I was satisfied. . . .”68 Over the whole chronological range of Byzantine literature a letter was valued as a consolation for the absence of the writer;69 how much more so, one must assume, was the letter that attempted to share the grief of the bereaved. And the fact that they were genuine and Peradotto, Gonimos (as above, note 4), 149. The Byzantines similarly adapted their imagery, style, and, ˇevon occasion, even level of language to suit their correspondents; see Ljubarskij, Mikhail Psell, 72–74; I. S ¨ B 31.1 (1981): 307; and R. Anastasi, “Michele Psello al Metroˇcenko, “Levels of Style in Byzantine Prose,” JO polita di Euchaita (‘Epist’: 34 pp. 53–56 K.-D.),” Studi di filologia bizantina 4 (1988): 108–15. 67 The mid-14th century can extend the range remarkably through a letter in which Nikephoros Gregoras, with tongue in cheek, consoles a friend over his beautiful, young wife, just recently wed but already unfaithful (Ep. 129 in Correspondance de Nice´phore Gre´goras, ed. R. Guilland [Paris, 1927], 221–25 ⫽ Ep. 123 in Nicephori Gregorae Epistulae, ed. P. A. M. Leone [Matino, 1982], 2:318–20). 68 ´ pistoliers byzantins, 150. Darrouze`s believed that the friend was Niketas of Smyrna. Ep. 89 in Darrouze`s, E 69 See in particular Karlsson, Ide´ologie, 45–47.
34
THE BYZANTINE LETTER OF CONSOLATION
letters, written with the practical purpose of giving consolation to a specific individual and not simply as artistic variants on a set theme, is shown, I believe, both by the specific personal details that they contain and by the wide range of tone that they exhibit, including, as has been seen in the letter of Theophylact to his spiritual father, even pleasure. Byzantine letters of consolation are, then, genuine letters actually sent to and, at least initially, intended for the bereaved.70 Those that have survived, however, are all carefully written and to some extent depend upon the artifices of rhetoric. Rhetoric today has an opprobrious connotation for most people, but one must never forget that rhetorical devices are simply devices that speakers have found effective over the years and that frequently spring to the lips of even the unlettered when they are emotionally aroused. To a modern reader a highly literary letter of consolation would smack of insincerity; to a Byzantine a hastily bought sympathy card would demonstrate lack of true friendship. For a modern reader “it is the thought that counts”; for a Byzantine reader it is both the thought and the time and effort spent that count. To begin, here is an example of the confidence that a writer has in his own rhetorical ability. In the late twelfth century Constantine Manasses opens a consolation to the recently widowed sebastos John Kontostephanos71 in the following way: “I have come again72 to mix another bowl for you, most noble lord. I have come to you as an orator, . . . not to tear open the wounds of your grief . . . but to restrain the anger and subdue the tumult of your soul . . . and soon this draught will seem to you to banish pain and allay your anger, more potent as it is than the drugs with which Helen sweetened the bowl of Telemachos.” But does the reception of such compositions justify such confidence? Of help here are three letters of Niketas Magistros,73 which he wrote in the second quarter of the tenth century. With the first (Ep. 1) he informs the protasekretis Gregory of the death of an unknown magistros. To modern ears it sounds as little more than a rhetorically composed lament, lacking personal applicability to the recipient. In his reply Gregory may have offered commiseration, but in any case he indubitably eulogized his correspondent’s style, for Niketas’s second letter (Ep. 3), which is itself replete with allusions to Homer, expresses his delight at compliments received: “Thrilled by my letter more than the young son of Philip was by Lysippos’s painting, you exalt me by your praises and write that those who read it are more eager to approach my guiding rudder than those who listened to the lyre of Arion and were charmed by his songs.” He proceeds to aver that his letter was not a literary exercise written for display but a genuine expression of his feelings, on which he expatiates for some lines. Although it could be argued that Niketas was not emotionally traumatized by this death, since the deceased was not Not only do all these letters easily fall within the Byzantine metron of a letter, but their comparative brevity indicates that they could have been written fairly soon after receipt of news of the death (although, of course, that alone cannot prove that they actually were). They are thus very different from Plutarch’s letter of condolence to Apollonios (Moralia 101F–22A), which is really a treatise that must have taken many days to compose. His letter of condolence to his wife on hearing of the death of their daughter Timoxena (Moralia 608B–12B) is perhaps just short enough to be a genuine letter, although it is hard to believe that his wife could have found any solace in it at all. 71 “Dva proizvedenija Konstantina Manassi, otnosjashchiesja k smerti Theodori Kontostefanini,” ed. E. Kurtz, VizVrem 7 (1900): 636–45. 72 Constantine had previously composed a more public monody (ibid., 630–35). 73 Nice´tas Magistros: Lettres d’un exile´ (928–946), ed. L. G. Westerink (Paris, 1973). 70
A. R. LITTLEWOOD
35
a relative, he does claim that the magistros was a “second father” to him; and our third letter (Ep. 12) concerns the death of his own son. This is addressed to his friend John the Patrician, who had heard of the death by means of a letter from a third party, a certain Rhodophylles, whose excessive rhetoric John had clearly found distasteful. “You have criticized the letter of Rhodophylles,” writes Niketas, “and not only ridiculed him for having played the tragedian with his laments of Hekuba, but reproached me too, the old man wasted away by his great grief. . . .” His subsequent long and rhetorical threnody, which abounds in classical allusions, serves both to emphasize his point made in the letter to Gregory and to remind us that for a Byzantine rhetoric and emotion were not incompatible.74 Further support may be found in Michael Italikos’s claim75 that, despite the fact that he is unconsolable at the death of his friend Constantine Hagiotheodorites, he is still grateful to his brother for a letter of consolation whose composition, rhythms, and beauty of diction have made him more cheerful. The bereaved today may take pleasure in the picture on the card of sympathy and still grieve; the Byzantine could clearly take pleasure in the artistry of the letter and still grieve. Allusions and quotations themselves are not necessarily mere padding and decoration. It is extremely rare for them not to be entirely apposite,76 and it is surprising, given the narrow scope of paramythetic quotations, how many different ones authors manage to find—even references to Job are usually to different passages. In the twenty letters of consolation mentioned above,77 their respective editors have identified 143 different quotations and allusions;78 of these only four are used by two writers, two by three, and none by more than three,79 although Photios does twice use the same passage.80 We may thus infer that the writers depended upon their own memory rather than delved into a handbook of quotations (that is rather a modern vice; if every Byzantine writer had his 74 Nevertheless, one may wonder how often a Byzantine played the precious game simply in order not to be excluded from the small circle of literati. Symeon Metaphrastes revealingly condemns rhetoric on the grounds not of emotion but of intelligibility when he informs a bishop that if he had intended his letter to ´ pistoliers byzantins, 154–55). be obscure he had assuredly succeeded (Ep. 94 in Darrouze`s, E 75 Op. 4 in Michel Italikos: Lettres et Discours, ed. P. Gautier (Paris, 1972), 89–91. 76 Nikolaos Mystikos criticizes the Bulgar khan Symeon on two different occasions for what he believes are inapposite quotations (Ep. 21, ll. 105–12; Ep. 25, ll. 67–72). 77 To the nineteen letters summarized above (excluding the letters of Anonymus Londiniensis and Niketas Magistros) I have added the lengthy consolation of Constantine Manasses (above, p. 34). 78 I have added four quotations from the consolation of the anonymous protonotarios, since the edition is without apparatus fontium, and one that was missed by Gautier from the letter of Gregory of Oxeia (Ep. 5, ll. 12–13), which is interesting in that, of all the quotations, it alone is taken from an earlier letter of consolation (see above, p. 32 and note 63). 79 Homer Odyssey 4.220–21 is quoted by Arethas, the anonymous protonotarios, and Manasses; Phil. 1:23 is quoted by Theophylact, Gregory of Oxeia, and Manasses; 2 Kings 12:15–23 is summarized by Photios and alluded to by Nikolaos Mystikos; Isa. 35:10 (⫽ 51:11) is quoted by Photios and Gregory; 2 Cor. 12:2 is quoted by Manasses and alluded to by Photios; 1 Thess. 4:13 is quoted by Photios and, with the subsequent four verses, referred to by Arethas. Of the 143 passages, 57 are from seventeen books of the Old Testament, 39 from fourteen books of the New Testament, 45 from twenty-three pagan authors, and just 2 from church fathers (Gregory quotes Basil, and Manasses perhaps alludes to a passage in the funeral oration by Gregory of Nazianzos for his brother Kaisarios). 80 In one instance, he accurately quotes the famous words of John 11:25 (oJ pisteu´ wn eij" ejme´ , ka‘n ajpoqa´ nh, zh´ setai) in two different letters (Ep. 234, l. 157, and Ep. 245, l. 116); the other instance should perhaps not be counted since it is simply a combination of a quotation and an allusion to Matt. 9:15 in the same letter (Ep. 245, ll. 119–20 and 102).
36
THE BYZANTINE LETTER OF CONSOLATION
book of commonplaces at hand, we should expect an abundance of such manuscripts, but they are in fact rare).81 Photios opens his letter to Tarasios with a call for Elijah and his other resurrectionary colleagues from the Bible before proceeding to a plethora of quotations. He was on good terms with his brother; is one to believe that in a letter of consolation he wishes to distress his brother with superfluous scholarship and insincere rhetoric?82 Finally, let us briefly examine Byzantine letters of consolation in terms of the needs of the bereaved as indicated by modern psychology and psychiatry.83 Here we must bear in mind four things. First, the Byzantines were probably better able to cope with loss than are citizens of the so-called developed world today. Death, because of its irremediable finality, is always a shock, even if it has been expected throughout a lingering illness and even when it has perhaps been hoped for to terminate pain or a vegetal state. Never81 Although there are about forty manuscripts of tags from Menander the Comedian, the compilation of Stobaios and collections of riddles were not commonly employed by epistolographers. The best-known individual collection of extracts is probably that in Cod. Heidelb. Palat. gr. 129, long attributed to Maximos Planoudes but ascribed to Nikephoros Gregoras by A. Biedl (“Der Heidelberger Cod. Pal. Gr. 129—eine ¨rzburger Jahrbu ¨cher fu ¨r die Altertumswissenchaft 3 [1948]: Notizensammlung eines byzantinischen Gelehrten,” Wu ˇevcˇenko’s claim that the hand is that of Gregoras himself (“Some 100–106), an attribution corroborated by I. S Autographs of Nicephorus Gregoras,” ZRVI 8.2 [1964]: 447–50). For a more detailed discussion of the use of quotations, see Littlewood, “A Statistical Survey,” 137–54. 82 Moreover, in the Zeitgeist of Byzantium a desire for the skill of the prophets and the apostles was perhaps not as far-fetched as it would be in the Western world today. One only has to recollect the rumor that through the necromancy of Photios’s friend Theodore of Euchaı¨ta the deranged emperor Basil was able to embrace the phantom of his dead son Constantine—and that in the Orthodox world of the 20th century, ` sto` n Gre´ ko, chap. 19), the poet Angelos Sikelianos expended if one can believe Nikos Kazantzakis ( Anafora j his strength in a valiant but anguishingly frustrated attempt to raise a corpse from the dead by stretching his own body over it all night and breathing into its mouth, just as Elisha had done. 83 Although the modern study of grief may be said to have its beginning with Robert Burton’s Anatomy of Melancholy, which was first published in 1621 (with the final edition in 1651/2), proper scientific examination of the matter really began with Josef Breuer’s investigations into the mental health of Anna O., a young woman disturbed by the terminal illness and subsequent death of her father. These researches, which began in 1881, were eventually made public in 1893 in the short but founding study of modern psychoanalysis, ¨ ber den psychischen Mechanismus hysterischer Pha¨nomene: Vorla¨ufige Breuer’s and Sigmund Freud’s “U Mitteilung,” Wiener medizinische Presse 3 (1893): 121–26, 165–67 (more accessible in the collected Studienausgabe [Frankfurt am Main, 1971], 6:13–24). Although subsequently Freud published in 1917 his brief “Trauer ¨r ¨arztliche Psychoanalyse 4:288–301 (more accessible in Studienausund Melancholie,” Internationale Zeitschrift fu gabe [Frankfurt am Main, 1975], 3:197–212), and M. Klein in 1940 her “Mourning and Its Relation to Manic Depressive States,” The International Journal of Psycho-analysis 21:125–53, almost no interest was taken in the normal person’s reaction to death until an address by Pope Pius XII to the World Union of Family Organizations at Castel Gandolfo in 1957. Welcoming greater care for widows, he claimed that this was a “sujet auquel ` se trouvent ces foyers jusqu’ici on n’a pas preˆte´ assez d’attention, en partie `a cause de l’impuissance meˆme ou sur le plan de l’action sociale” (Acta Apostolicae Sedis 49, ser. 2.24 [1957]: 899). The last twenty-five years have seen a substantial body of work on the subject of bereavement. To the layman the most sound and useful treatments seem to be Y. Spiegel, Der Prozess des Trauerns: Analyse und Beratung (Munich, 1973); J. Schneider, Stress, Loss, and Grief: Understanding Their Origins and Growth Potential (Baltimore, Md., 1984); D. Klass, Parental Grief: Solace and Resolution (New York, 1988); C. M. Sanders, Grief: The Mourning After. Dealing with Adult Bereavement (New York, 1989); G. H. Pollock, The Mourning-Liberation Process, 2 vols. (Madison, Wisc., 1989); and especially B. Raphael, The Anatomy of Bereavement (New York, 1983), and C. M. Parkes, Bereavement: Studies of Grief in Adult Life, 2d ed. (London–New York, 1986). I am grateful to the late Alexander Kazhdan for pointing out to me that one of the most perceptive psychological studies of mourning appears in Thomas ¨der (Frankfurt am Main, 1962), esp. chap. 7, “Der Zerrissene” of Der Junge Mann’s novel Joseph und seine Bru Joseph.
A. R. LITTLEWOOD
37
theless, in his everyday life a Byzantine was usually far more aware of the possibility of death than is modern man with his trust in modern medicine.84 This is especially true in the case of children: Byzantine parents, due to the high mortality rate of children, were, if not resigned to, at least unsurprised by losing half or even more of their children; parents of today rarely contemplate the chance of losing a single child.85 Second, although most people even today can cope with their grief aided only by family and friends, psychiatric textbooks are primarily concerned with the small percentage of those who cannot; and most of the case studies are of women rather than men, since the former more frequently survive their spouses. Third, grief is a process, not a state. We do not know exactly how soon after a death any particular letter reached its recipient, and therefore we do not know at what stage of his grief he was and what, consequently, were his needs. Moreover, letters are not winged words but permanent records of sympathy designed to give comfort long after their initial reading, and they should therefore be suitable not only for early but also for later stages of grieving, and indeed for the whole long period that should follow its resolution. Fourth, the emphasis of modern psychiatric counseling is on listening. Although at least since the time of Artemon, the editor of Aristotle’s letters, it had been considered that “a letter should be written like a conversation since it is like one half of a conversation,” 86 it is still only half, and, despite the possibility of a reply that not only expresses thanks but also details the feelings of the bereaved, the letter must necessarily be at a great disadvantage in comparison with a genuine dialogue. All modern authorities emphasize that there are stages in the grieving and healing process,87 though they disagree on certain particulars and even on the number of stages The still scanty evidence for the life expectancy of the Byzantines is summarized by A.-M. Talbot in “Old Age in Byzantium,” BZ 77 (1984): 267–78. She concludes that “for most Byzantines life was cut short, in youth or middle age” (ibid., 269). There is as yet little evidence from skeletal remains, but the samples that have been published, from different periods and different areas of the empire, suggest that the mean age of death for adults (reports define adults as those surviving to either 15 or 17.5 years) was around 35 for men and some five or six years fewer for women. Epigraphers concur in calculating “that only half the adult Byzantine population reached the age of 35” (ibid., 268). In the capital itself, Talbot quotes two surveys of skeletal remains, one giving an average age at death of 28 to 29 years from a sample of seventy-six individuals in a 12th-century cemetery at Sarac¸hane (the church of St. Polyeuktos), and the other, surprisingly different, of 46.2 years for men and 37.3 for women from Kalenderhane (ibid., 267). See further E. Patlagean, Pauvrete´ ´economique et pauvrete´ sociale `a Byzance, 4e–7e sie`cles (Paris–The Hague, 1977), 73–112. As may be expected, however, the situation is not the same with the educated upper class, to which all the writers and recipients of our letters of condolence belong. Talbot cites Kazhdan’s figures of 62 years as the average age at death for a group of thirteen authors (Greek and Latin) from the 6th century, and 71 years for another group of fifteen from the late 11th and 12th centuries (A. P. Kazhdan, “Two Notes on Byzantine Demography of the Eleventh and Twelfth Centuries,” ByzF 8 [1982]: 117), as well as her own finding of 67.3 years for the literati of the Palaiologan period (ibid., 269). 85 Modern North American cards of condolence, like other types of cards from that continent, aim to save the purchaser from the toil of writing anything but a signature and in consequence specify the relationship of the deceased to the bereaved. It is, therefore, worthy of note that although cards are frequently designed to give solace for a lost father, mother, grandfather, or grandmother (or even pet), I have never seen one for a son, daughter, child, or baby (but one well-known maker did announce in early 1998 that it was now commercially viable to produce cards of sympathy for a death by suicide). 86 Demetrius De elocutione 223. See further Karlsson, Ide´ologie, 34–45. 87 This is often presented as if it were a modern discovery, but it was understood, even if not closely analyzed, by ancient writers (e.g., Seneca Consolatio ad Marciam 1.8; Pliny Ep. 5.16.10–11; Plutarch Moralia 102A). 84
38
THE BYZANTINE LETTER OF CONSOLATION
(from three to six; for the following comments I have arbitrarily concluded that there are four distinguishable stages). The first stage is that of shock, characterized by disbelief, numbness, confusion, alarm, and tears, and in the normal person this tends to end soon after the funeral. A letter would arrive, therefore, sometime toward the end of this phase or, probably more frequently, a little after; and, as already concluded, the actual shock to a Byzantine was probably less than it is nowadays since he was far more accustomed to the ever imminent possibility of death. For both of these reasons a Byzantine letter had less need to deal with many of the reactions mentioned by psychologists and psychiatrists. One element, however, is outstanding. Almost all letter writers realized the psychological benefit of giving way to tears88 and opened their letters with either an encouragement to cry or sympathetic laments of their own. This is particularly notable in that most of the recipients were male rather than female,89 for—despite the tradition of heroic crying first found in Homer90 whose warriors cry with as much gusto as they fight—tears were still often considered somewhat more womanly than manly in both ancient Greece and Byzantium.91 Today, especially in northern Europe, valiant attempts to stifle tears are frequent, but these attempts, especially if they are successful, tend merely to postpone, sometimes indefinitely, the process of healing. Modern scientific theory here corroborates the intuitive understanding of Byzantine (and classical) letter writers.92 88 There existed even the separate genres qrh'no" and monvdi´a that were devoted to lamentation (see Alexiou, Ritual Lament). 89 Outside the chronological bounds of this survey are two letters of St. Basil (Epp. 5 and 6) that, even with allowance made for the different personalities of the recipients, reveal something of what Basil considered the needs of men and women to be, since they are addressed respectively to Nektarios (probably the future Constantinopolitan patriarch) and his wife on the death of their son. 90 See H. Monsacre´, Les larmes d’Achille (Paris, 1984), esp. 137–57. Edward Gibbon makes a sardonic comment about the Crusaders’ propensity to give way to tears: “A reader of Villehardouin must observe the frequent tears of the marshall and his brother knights. . . . They weep on every occasion of grief, joy, or devotion.” (The History of the Decline and Fall of the Roman Empire, ed. J. B. Bury [London, 1912], 6:400 n. 49.) 91 On Byzantine display of grief, see J. Kyriakakis, “Byzantine Burial Customs: Care of the Deceased from Death to the Prothesis,” GOTR 19 (1974): 37–72; D. Abrahamse, “Rituals of Death in the Middle Byzantine Period,” GOTR 29 (1984): 125–34; and H. Maguire, “The Depiction of Sorrow in Middle Byzantine Art,” DOP 31 (1977): 123–74. Christian teaching seems gradually to have curbed, if not eradicated, the excessive wailing of antiquity, against which the early church fathers frequently inveigh. St. Basil thus decrees in his homily De Gratiarum Actione: “Therefore neither men nor women should be permitted too much lamentation and mourning. They should show moderate distress in their affliction, with only a few tears, shed quietly and without moaning, wailing, tearing of clothes and grovelling in the dust, or committing any other indecency commonly practised by the ungodly” (PG 31:229C, trans. in Alexiou, Ritual Lament, 28; see further her second chapter, “From Paganism to Christianity,” ibid., 24–35). 92 This intuitive understanding is, of course, widespread among many disparate cultures. An outstanding example is that of the Huli in the Southern Highlands of Papua, whose women have, for therapeutic purposes, a duguanda or “crying-house” in which those close to the deceased even sleep for several weeks after the funeral (S. Frankel and D. Smith, “Conjugal Bereavement amongst the Huli People of Papua New Guinea,” British Journal of Psychiatry 141 [1982]: 303). Parkes claims that this need to cry was recognized by Pius XII: “The encyclical [sic] of Pope Pius XII delivered in September 1957 to the World Federation of Family Organizations points out that there is nothing ignoble in tears and that the widow should withdraw from the activities of the world for a ‘reasonable period of mourning’” (Bereavement, 180, n. 1). Unfortunately, the pope says no such thing either in his address to that organization or in any encyclical, although in the former he does compassionately mention the widow’s “douleur ineffac¸able” and “l’angoisse qui l’enserre comme d’une infranchissable muraille” (Acta Apostolicae Sedis 49:900). It is worth noting that Menander Rhetor, in his formulation for funeral orations, says that those delivered soon after the death should open with lamentations that should be “fully worked out,” whereas those delivered much later should be merely enco-
A. R. LITTLEWOOD
39
The second phase of grieving is one of awareness of the loss and is generally marked by yearning, anger, and frustration. It is at this phase, we may perhaps assume, that most letters arrive, especially if they had to travel a long distance.93 The love and yearning of the bereaved for the deceased are frequently mentioned; sometimes, as in the case of Manasses’ letter to John Kontostephanos,94 they are ennobled by being presented as a final gift to the deceased wife who has not herself been left to mourn a dead husband. The second feeling, anger,95 is today generally directed at doctors and nurses96 who, it is believed, could have saved the beloved if they had been more competent. Lacking the modern delusion in the paramountcy of medicine, the Byzantine tended to blame God.97 If one can rely on our letters, this trait was especially feminine, Nikolaos Mystikos even believing98 that her own prior death saved Romanos’s wife from the possible blasphemy in both words and deeds that she might have committed had she been the one widowed. Nevertheless, men too were resentful, and our writers argue that this is improper because it is contrary to the wishes of God, and that the deceased has now escaped the pain of illness here, or, if there has been no sickness, simply the troubles of earthly life and that he or she must consequently be better off in heaven. The bereaved should, therefore, accept the will of God, be happy for the deceased, and, despite the pain, not be brokenhearted.99 The third diagnostic feeling at this stage, which is the frustration that many bereaved feel today at their inability to cope with daily life, plays only a small role in Byzantine letters, perhaps partly because it may have been considered a little insulting, especially if addressed to men. As already noted, psychiatrists always emphasize that anyone counseling the bereaved must primarily be a good listener, especially at the first and the second stages of mournmia except when the speaker is closely related to the deceased, for “memory does not grant him relief from sorrow even after a year” (Russell and Wilson, Menander Rhetor, 2.11, pp. 172–74). Wagner, not comprehending psychological needs, finds an “inconsistency” in “the first part of the funeral speech which sought to heighten the grief and the second part which endeavored to allay it” (“A Chapter,” 160 n. 134). 93 See above (p. 22) where I suggest that these letters may usually have been sent only when absence made oral consolation impossible. The time taken for the news of a death to reach the writer may also have been frequently greater than the time taken for the letter to reach its destination. 94 Above, note 71. 95 As with lamentation, anger’s repression in grief is considered harmful; see D. C. Maddison and W. L. Walker, “Factors Affecting the Outcome of Conjugal Bereavement,” British Journal of Psychiatry 113 (1967): 1057–67. However, no Byzantine letter writer, to my knowledge, encourages the expression of anger. For a general study of anger and its practical outlet in aggression, see K. Lorenz, Das sogenannte Bo¨se: Zur Naturgeschichte der Aggression (Vienna, 1963). Lorenz demonstrates how deep-rooted this emotion is in humans in comparison with other animals, a connection first seriously studied, in a much larger context, by Charles Darwin in The Expression of the Emotions in Man and Animals (London, 1872). 96 And, in cases of widows and widowers, anger can even be directed at the deceased for abandoning the spouse. 97 This is not uncommon even today; see Parkes, Bereavement, 100. Thomas Mann’s Jacob indulges in a particularly complex expostulation against his God who “nicht Schritt gehalten hat in der Heiligung” since ¨ bermut umbringt die Frommen und Bo ¨sen” ( Joseph und seine Bru ¨der, 1:644). “er im U 98 Ep. 156. 99 The complaint of a broken heart is not just romantic fiction but is medically verifiable: three-quarters of the increase in mortality in the first six months of bereavement are attributable to coronary thrombosis and arteriosclerotic cardiac disease; see Parkes, Bereavement, 34–49, 214–16 with extensive bibliography. Parkes notes (ibid., 34) that in Doctor Heberden’s Bill classifying the 1,357 deaths in London in 1657 ten are attributable to “Griefe.”
40
THE BYZANTINE LETTER OF CONSOLATION
ing. Although this is clearly impossible for a letter-writer, it is worth noting that he may be trying to compensate for this lack by writing about what he assumes the bereaved is feeling, and therefore emphasizing the sense of loss and sympathetically sharing it when he too knew the deceased. The bereaved is thus in a sense himself telling the writer his own grief. Eulogy comes in here too,100 although modern psychiatrists, worried as they are by excessive attachment to the deceased in abnormal cases, tend not to consider its comforting aspects to the healthy. The third phase is often one of withdrawal,101 and here the Byzantine epistolographer helps the bereaved to move into the fourth phase, that of healing, by giving him a purpose in life—he must be worthy of the deceased. Further help in this direction would be given in subsequent letters that did not allude to the death but took the recipient’s mind away from it by dealing with other subjects. Rarely, however, can we date letters so precisely as to know which were sent within the few weeks following the letter of sympathy. Nonetheless, there is one brief letter from the tenth century that is of significance in this context: Leo of Synada102 thanks his correspondent for his letter of sympathy but now asks for something “bright and gay (caropo´ n ti kai` carmo´ sunon) so that I may know that you are cheering me up with your letters.” The single major difference between the underlying ideas of Byzantine epistolary consolation and the standard modern therapeutic approach is the latter’s affective, rather than cognitive, emphasis. Effective psychotherapy, according to Carl Rogers, demands on the part of the healer empathy, genuineness, and an “unconditional positive regard” for a client.103 “Rogerian” psychotherapy, though a general methodology not specifically developed for bereavement, is employed today by an overwhelming majority of clinicians. The uncritically sympathetic understanding and sharing of emotional suffering, which are the staple ingredients of the wording of modern cards of condolence, are thus in line with medical practice. This affective element is also, of course, found frequently in Byzantine letters, but there it is usually balanced, overshadowed, and occasionally even obliterated by a cognitive emphasis, which is realized in the demand for changes in thinking: the Byzantine writer attempts to wrest the bereaved from an absorption in self-pity by stressing theological and other implications. Herein the Byzantines foreshadow the 100 The role of eulogy is, however, generally not great in letters, since it more properly belongs to other funerary genres, such as the ejpita´ fio" lo´ go". 101 This is exacerbated today by a tendency to shun the widow, sadly mentioned by Pius XII (Acta Apostolicae Sedis 49:899) and constantly addressed in modern treatments of bereavement. It has, however, little part in Byzantine letters, probably because widows, if they did not enter a monastery, still had an important role to play within their own families. 102 Ep. 38 in The Correspondence of Leo, Metropolitan of Synada and Syncellus, ed. M. P. Vinson, CFHB 23 (Washington, D.C., 1985), 62. 103 “[Optimal therapy] would mean that the therapist feels this client to be a person of unconditional selfworth: of value no matter what his condition, his behavior, or his feelings. It would mean that the therapist is genuine, hiding behind no defensive fac¸ade, but meeting the client with the feelings which organically he is experiencing. It would mean that the therapist is able to let himself go in understanding this client; that no inner barriers keep him from sensing what it feels like to be the client at each moment of the relationship; and that he can convey something of his empathic understanding to the client. It means that the therapist has been comfortable in entering this relationship fully, without knowing cognitively where it will lead, satisfied with providing a climate which will permit the client utmost freedom to become himself.” (C. R. Rogers, On Becoming a Person: A Therapist’s View of Psychotherapy [Boston, 1961], 185; see also esp. 47–48, 283–84.)
A. R. LITTLEWOOD
41
psychotherapeutical theory of Albert Ellis and Aaron Beck, which is still not widely accepted and has not yet been applied, at least to any appreciable extent, in “griefcounseling.” This theory holds that, since distress is the result of disordered thinking, the inappropriate emotional responses engendered by such thinking can be countered by diverting the patient into new lines of thought.104 Did the Byzantines instinctively feel that their insistence on challenging the bereaved to direct their minds to what they believed were appropriate thoughts would the more swiftly allay feelings of grief? Another controversial modern theory, that of Martin Seligman, teaches that a major reason for depression, which is a condition akin to that of grief, is the patient’s belief that no actions taken by him can have any effect upon his life, a misapprehension that can be remedied only by encouraging him to bestir himself and thus disprove it by becoming again “an effective human being.” 105 According to this theory, the Byzantine letter writers, by urging not only new mental activity but also new actions in the form of exemplary behavior, would again be materially aiding the bereaved to overcome their grief. University of Western Ontario 104 “The depressed . . . individual has certain idiosyncratic cognitive patterns . . . which may become activated either by specific stresses impinging on specific vulnerabilities or by overwhelming, nonspecific stresses. . . . Cognitive psychotherapy may be used symptomatically during depressions to help the patient gain objectivity toward his automatic reactions and counteract them.” (A. T. Beck, Depression: Causes and Treatment [Philadelphia, 1970], 318; see further esp. 228–40, 253–73; and also A. Ellis, Reason and Emotion in Psychotherapy [New York, 1962], 105–6.) 105 M. E. P. Seligman, Helplessness: On Depression, Development, and Death (San Francisco, 1975), 100 (cf. p. 105: “The central goal of therapy for depression . . . is the patient’s regaining his belief that he can control events important to him”).
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Photius as a Reader of Hagiography: Selection and Criticism TOMAS HA¨GG
yzantine hagiography is a rich and rewarding field of study. Still, modern scholars working in the field sometimes give voice to frustration over their inability to understand, in a more profound sense, their own object of study, let alone like it. Certain Lives of saints, or certain types of such Lives, are no doubt even more distant from modern mentality and taste than most other Byzantine works of literature—at least, from the mentality of modern Western scholars. Furthermore, modern critical tools are clearly inadequate to explain the success and abundant survival of some of these works. It may therefore be of interest to see how a scholar and literary critic among the Byzantines themselves—and one with the high stature and independent mind of Photius at that— approached this form of literature. In his Bibliotheca,1 Photius devotes eight codices to literary works that can be classified as saints’ Lives: the Life of John Chrysostom (cod. 96), the Life of Gregory the Great (cod. 252), the Martyrdom of the Seven Child Saints (cod. 253), the Martyrdom of Timothy the Apostle (cod. 254), the Martyrdom of St. Demetrius (cod. 255), the Life of Sts. Metrophanes and Alexander (cod. 256), the Life of Paul the Confessor of Constantinople (cod. 257), and the Life of Athanasius the Great (cod. 258). All but one are presented by Photius as anonymous. The exception is the Life of John Chrysostom, which he says was written by George, bishop of Alexandria, admitting, however, that he possesses no external information about the author. Several codices are quite substantial: the Life of Athanasius is summarized in twenty-three (printed) pages, that of John Chrysostom in fifteen, and that of Paul the Confessor in ten. Following Warren Treadgold in his useful tabulated dissection of the Bibliotheca,2 we may classify some other codices as hagiography as well: the Apocryphal Acts of the Apos-
B
My quotations of the Greek text of the Bibliotheca follow Photius, Bibliothe`que, ed. R. Henry, 8 vols. (Paris, 1959–77) and vol. 9, index by J. Schamp (Paris, 1991). My translations are adapted from those of Nigel Wilson, whenever the codices in question are contained in Photius, The Bibliotheca: A Selection Translated with Notes, ed. N. G. Wilson (London, 1994). I am grateful for constructive criticism of an earlier version of the article from participants in the Dumbarton Oaks Symposium of 1996, “Aesthetics and Presentation in Byzantine Literature, Art, and Music,” where it was first presented. I would also like to thank two anonymous readers and Christian Høgel for their comments. 2 W. T. Treadgold, The Nature of the Bibliotheca of Photius, DOS 18 (Washington, D.C., 1980), 174. 1
44
PHOTIUS AS A READER OF HAGIOGRAPHY
tles (cod. 114),3 the various writings ascribed by Photius to Clement of Rome (codd. 112– 13), John Cassian’s Conferences (cod. 197), and John Moschus’s Pratum spirituale (cod. 199). Even so, hagiography is not among Photius’s most favored theological genres; other categories, such as apologetics, homilies, church history, and dogmatic and exegetical works, occupy a greater number of codices each. Still, some of his hagiographical reports are detailed enough—and seem independent enough—to make it worthwhile to observe Photius as a reader and critic of this kind of literature. My emphasis throughout is on Photius’s own attitude to hagiography proper, as implicit in his choices or explicitly stated in his criticism. This means that the notoriously difficult problems of precisely what manuscripts, or what recension of a certain text, he had before his eyes is secondary in my treatment. I do not thus attempt to probe further into the surviving hagiographical manuscripts than experts like Franc¸ois Halkin have already done. In some cases the results arrived at by other scholars seem convincing to me, while in other, more doubtful, cases such an ambition would have meant research of a kind and dimension beyond the scope of this article. I also take for granted that Photius actually read the works he says he read, and did not use someone else’s summary, unless he says so, or take over someone else’s literary criticism mechanically. I have argued elsewhere for this approach to the Bibliotheca, taking Photius at his word with regard to his bibliographical information; and the recent criticism of Jacques Schamp4 has not persuaded me to think otherwise. Photius sometimes states that he has not been able to find a certain book and has to rely on secondary sources, or that he has read only part of a book; that he would positively mislead his readers in other cases does not make sense. It is another matter that some of the terminology Photius uses in his bibliographical introductions is open to different interpretations—or, rather, has been interpreted differently by different scholars, though mostly, in my opinion, without good reasons. For if one cares to accustom oneself to Photius’s sometimes idiosyncratic use of words like ejklogh´ and e“kdosi", by reading a number of his introductions and conclusions and comparing the terminology to the actual contents of the codices in question, the problem tends to evaporate. It is, however, arguable that the codices of the second part of the Bibliotheca (codd. 234–80)—where most of this peculiar terminology appears—have a different genesis from the earlier ones,5 and that this might affect the question of Photius’s personal reading. Since seven of our eight hagiography codices proper belong to this second half, the issue cannot be ignored here; but such problems are not allowed any prominent place in my considerations. Instead, I hope to be able to bring to the topic some fresh perspectives and comparative material not used previously in this context, by illustrating how a Byzantine scholar 3 Photius’s criticism of these Acts is analyzed by E. Junod, “Actes apocryphes et he´re´sie: Le jugement de Photius,” in Les Actes apocryphes des apoˆtres: Christianisme et monde paı¨en, ed. F. Bovon et al., Publications de la Faculte´ de the´ologie de l’Universite´ de Gene`ve 4 (Geneva, 1981), 11–24. 4 J. Schamp, Photios historien des lettres: La Bibliothe`que et ses notices biographiques, Bibliothe`que de la Faculte´ de philosophie et lettres de l’Universite´ de Lie`ge 248 (Paris, 1987), 95–99. Cf. T. Ha¨gg, Photios als Vermittler antiker Literatur: Untersuchungen zur Technik des Referierens und Exzerpierens in der Bibliotheke, Acta Universitatis Upsaliensis, Studia Graeca Upsaliensia 8 (Uppsala, 1975), 131–37, and Treadgold, Nature, 41–48. 5 ¨ B 32.4 See, in particular, Treadgold, Nature, 37–51. Cf. also J. Schamp, “Flavius Jose`phe et Photios,” JO (1982): 185–96; his conclusions concerning cod. 238 do not admit of generalization.
¨ GG TOMAS HA
45
of the middle Byzantine period looked at Christian narrative texts of late antiquity and the early Byzantine period. Since Photius read many works of pagan history, biography, and fiction, it appears natural to bring in his attitude to these three pagan genres for a comparison. Did he, for example, like some modern scholars, judge saints’ Lives solely according to their degree of historicity, or is there any indication that he could have also read hagiographical texts for their narrative qualities or other artistic merits, as he clearly did with ancient fiction? Did he merely seek factual information on the earthly careers of specific saints, or was he also prepared to see the genre’s raison d’eˆtre in a broader Christian perspective? In other words, could he have contemplated hagiography as intended for spiritual edification rather than satisfying the kind of intellectual curiosity so typical of Photius himself? The first part of the article is devoted to an attempt to determine what most interested Photius himself in the saints’ Lives, that is, why he read them. The basis for a study of his selection principles is, of course, provided by the cases in which the original text has survived and been identified, so as to permit the direct comparison between two texts. This applies to codex 96, the Life of John Chrysostom, which is my principal object of study. To bring out proportions, principles of selection, and tendencies, I use the same method as I did for the study of Photius’s profane reading in my book Photios als Vermittler antiker Literatur (1975), thus facilitating comparisons. In the second part, the emphasis is on Photius as a literary critic, with special reference to his reading of hagiography, that is, how he read and judged the Lives of saints. Emil Orth’s listing and discussion of Photius’s critical vocabulary 6 as well as George Kustas’s study of Photius’s literary criticism7 provide some of the background here. THE SELECTION A study of Photius’s principles of selection must begin with the books themselves, before we turn to the actual comparison of texts. One may take for granted that Photius had huge quantities of hagiographical texts to choose among; the menologia and other collections of saints’ Lives compiled in the ninth through the eleventh century witness to the abundance of texts of this genre available in Constantinople. Many of these, of course, were contemporary compositions, and such works were obviously not on Photius’s agenda for the Bibliotheca: the most recent hagiographical book he discusses is the Life of Gregory the Great (cod. 252), perhaps belonging to the middle of the eighth century;8 6 E. Orth, Photiana, Rhetorische Forschungen 1 (Leipzig, 1928), and idem, Die Stilkritik des Photios, Rhetorische Forschungen 2 (Leipzig, 1929); cf. the warning in P. Lemerle, Le premier humanisme byzantin (Paris, 1971), 195 n. 56: “doit ˆetre consulte´ avec prudence.” 7 G. L. Kustas, “The Literary Criticism of Photius: A Christian Definition of Style,” Hellenika (Thessalonike) 17 (1962): 132–69; cf. also idem, “History and Theology in Photius,” GOTR 10 (1964): 37–74. 8 The identity of the text Photius had before him is disputed. F. Halkin, “La date de composition de la ‘Bibliothe`que’ de Photius remise en question,” AB 81 (1963): 414–17, argued that Photius used a short (surviving) Greek Life (BHG 721) which, according to H. Delehaye, “S. Gre´goire le Grand dans l’hagiographie grecque,” AB 23 (1904): 452, built on the huge Latin Vita S. Gregorii (BHL 3641), written in the 870s by Johannes Diaconus (Hymmonides); on this text, see W. Berschin, Biographie und Epochenstil im lateinischen Mittelalter, vol. 3, Karolingische Biographie, 750–920 n. Chr., Quellen und Untersuchungen zur lateinischen Philologie des Mittelalters 10 (Stuttgart, 1991), 372–87. This would place the composition of the Bibliotheca some decades later than is usually thought, and Halkin’s argument has been accordingly scrutinized and
46
PHOTIUS AS A READER OF HAGIOGRAPHY
all the others are from the seventh century or earlier. But even if he thus restricted himself to texts already possessing some historical patina, he no doubt had plenty to choose from. From this abundance, Photius included in the Bibliotheca just eight hagiographical works proper. One fact is immediately conspicuous: the emphasis—with regard to the number of works chosen as well as the comprehensiveness of their treatment—clearly lies on politically important figures of church history. There are substantial codices devoted to two well-known bishops of Constantinople, John Chrysostom and Paul the Confessor, and to the famous bishop of Alexandria, Athanasius the Great; in addition, a shorter codex concerns the life of a pope, Gregory the Great. Two more obscure early bishops of Constantinople, Metrophanes and Alexander, are treated in one large codex. On the other hand, the legendary lives of early popular saints play a rather modest role in Photius’s selection: only six pages altogether are spent on the Martyrdoms of Timothy the Apostle, St. Demetrius of Thessalonike, and the Seven Child Saints of Ephesus. These crude figures give us a first indication of what kind of hagiographical texts Photius mainly wanted to communicate to the readers of the Bibliotheca—primarily perhaps to his brother Tarasius as the formal addressee of the whole work, but at the same time also to the wider circle of intellectually curious contemporaries whose instruction and enjoyment he certainly had in mind when he undertook the arduous task of composing it. What he wanted to bring to their attention was evidently not, in the first place, legends of the most celebrated early martyrs and saints. We may presume that most such Lives were already excluded by being easily available, in one version or other, in other places; after all, the most important general criterion for detailed treatment in the Bibliotheca appears to have been the difficulty of acquiring the book in question. But there must also have been rare books surviving on saints of a less public career than the famous bishops whom Photius favored, which means that his emphasis on politically important figures remains significant. This interest is no doubt directly connected with his own ideal of the kind of role that a church leader should play in society: the future patriarch of Constantinople reveals his political instinct.9 To what extent does this first impression agree with Photius’s actual choice of material from the books he read? In investigating his criteria of selection, one may start with his treatment of George of Alexandria’s On Chrysostom (cod. 96, BHG 873, CPG 7979) and by quoting Photius’s own recommendation to other readers at the end of this codex (83b19– 21): oJ de` suggrafeu` " ou»to" oujk ojli´ga fai´netai paristorw'n⭈ ajll∆ oujde` n kwlu´ ei tou` " ajnaginw´ skonta" ejklegome´ nou" ta` crh´ sima ta` loipa` parora'n. In Nigel Wilson’s translation, “This writer is obviously much given to inaccuracy; but there is nothing to prevent readers from selecting what is valuable and overlooking the remainder” (p. 120). It is questionable whether paristorw'n here really means “given to inaccuracy”; I return to that below. Now, what Photius expects his readers to look for is “what is valuable,” or “useful” (ta` rejected by several scholars; see B. Hemmerdinger, “Le ‘codex’ 252 de la Bibliothe`que de Photius,” BZ 58 (1965): 1–2; Lemerle, Humanisme, 190 n. 48; Treadgold, Nature, 30–31; Schamp, Historien des lettres, 70–75. Treadgold (Nature, 30–31) instead suggests the date of 741–752 for the Greek Life that Photius read. 9 One of the anonymous readers of the article has emphasized this point (cf., e.g., the prominence given to the bishop’s conflicts with the imperial power in the Lives of John, Paul, and Athanasius, and partly in the Life of Metrophanes and Alexander as well) and rightly pointed out that such nonliterary factors deserve closer scrutiny than has been possible in the present study.
¨ GG TOMAS HA
47
crh´ sima). We may presume that he has chosen for his summary precisely that—what is useful in his own eyes—when boiling down a text of some 325 pages to fifteen.10 What, then, has he “selected,” and (no less interesting) what has he “overlooked,” in this Life? Some work has already been done on this topic, notably by Rene´ Henry in the notes to his translation.11 By close comparison with the original text, he was able to show that Photius often read several pages of the text before writing down (or dictating) his summary, since in the summary some facts appear at an earlier place than in the original. But there are also literal borrowings inserted in the summary that cannot readily have been made from memory. Thus, the codex well fits the type that I have called “das analytische Referat”;12 in other words, as Treadgold explains, “descriptions probably composed by referring back to the original text (or possibly to notes on it).” 13 It seems that Photius read George of Alexandria’s work in about the same way as he read Procopius, De Bellis 1–2.19 (cod. 63), which was my chief example for this kind of epitome. It is interesting that the proportion between the epitome and the original text turns out to be very similar for the two authors: approximately 1 to 20 in Procopius’s case, and about 1 to 21.5 in the case of George of Alexandria.14 At this point, a short digression on Photius’s methods of reading and his composition of the various codices of the Bibliotheca seems appropriate. There is still much idle speculation on these matters, but also too much defeatism as to the possibility of knowing anything at all about them with certainty. However, on the basis of the surviving originals of the works that Photius read, it is possible to distinguish a few basic types of treatment, and for each of these Photius’s reading method and use of memory can be deduced with a fair degree of confidence.15 There is no mystery about Photius’s memory capacity. For the short “synthetical summaries” of books his memory serves him well, but seemingly without any exceptional qualities; he summarizes after he has finished reading, there is a topical rather than chronological ordering of the facts, and his interests decide the selection. For the longer “analytical epitomes,” such as those of Procopius and George of Alexandria, he makes notes (or dictates summaries) progressively; changes in order occur only within the two, or five or ten, pages he happens to read in each portion, and instances of literal borrowing show his immediate access to the text of his model. The “excerpts” proper, finally, are copied directly from the model, sometimes even including obvious corruptions, spelling mistakes, and so on; there is no superhuman memory at work there, but probably a secretary. Problems remain, of course, such as the use of reading notes, the use of dictation versus copying, the role played by secretaries, and the division of work between the years of reading and the actual composition of the Bibliotheca. Furthermore, the three basic types of treatment are sometimes mixed; and in some of the cases where the original I have used the size of Henry’s (Bude´) pages as the norm; the 215 pages of F. Halkin’s edition of George of Alexandria (Douze re´cits sur saint Jean Chrysostome, SubsHag 60 [Brussels, 1977]) roughly correspond to 325 pages of Henry’s edition. 11 Henry, Bibliothe`que, 2:48–63, 207–11. 12 Ha¨gg, Vermittler, 184–94. 13 Treadgold, Nature, 118. 14 Photius: ca. 18,600 characters; George of Alexandria: ca. 402,480 characters. 15 There is ample concrete demonstration of this in the notes to the translation in Henry, Bibliothe`que, as well as in T. Ha¨gg, “Photius at Work: Evidence from the Text of the Bibliotheca,” GRBS 14 (1973): 213–22, and idem, Vermittler. 10
48
PHOTIUS AS A READER OF HAGIOGRAPHY
work is lost it is not possible to be sure of the proper classification.16 But it is, in my opinion, inadmissible to continue speaking and writing as if nothing at all could be demonstrated about these matters, or to continue indiscriminately calling all the codices “excerpts,” “epitomes,” or “re´sume´s,” when there are clearly defined categories, each being the result of a different genesis and displaying different characteristics. My principal concern, however, is not Photius’s methods of reading but his principles of selection and the attitudes to the hagiographical genre that they mirror. Henry makes some general remarks on that issue with regard to codex 96, to the effect that Photius has retained only the essential biographical facts while leaving out a number of long and monotonous edifying passages as well as some absolutely useless repetitions.17 This verdict is echoed by Wilson, who goes even further by stating, “Photius’ re´sume´ is superior to the original because it omits long moralising passages that give no information.” 18 Using less subjective tools of description, we may now see what kind of material was bypassed in Photius’s summary. Since no critical edition proper of George of Alexandria’s biography yet exists, I use for my purposes Halkin’s text of 1977,19 which is based on two manuscripts. Two other manuscripts formed the basis for the old text printed in volume 8 of Sir Henry Savile’s Chrysostomi Opera Omnia of 1612; Halkin’s text was not available to Henry in 1960, and unknown to Wilson in 1994. The first observation is that the flowery and verbose language characterizing great parts of George’s narrative is boiled down by Photius to a very prosaic statement of the basic facts. For instance, if George in his prologue solemnly pronounces that he will “twine together his different sources into one harmonious narrative, as if bound together in a golden chain” (chap. 1, 72.25: eij" mi´an tina` eujarmo´ nion dih´ ghsin sumple´ xai, w”sper ajlu´ sei crush' sundedeme´ nh⬍n⬎), Photius simply says eij" e’n sunaqroi´sasqai (78b35), “combine in one.” When Photius himself has occasional recourse to metaphorical language, it is indicative that he uses metaphors of his own instead of the ones employed by George. Photius says, for instance, that John gi´netai aujto` " tu´ po" kai` kanw` n tw'n monazo´ ntwn (79b8); the closest equivalent in George’s very detailed description of John’s exemplary life is the phrase pare´ dwken de` aujtoi'" o”ron ejgkratei´a" kai` proseuch'" kai` ajskh´ sew" (chap. 6, 92.17).20 There is thus no question of mirroring George’s style, of which Photius is in fact highly critical (see below, p. 57). Nor does he 16 The most detailed classification, as well as a tentative attribution of each codex to a certain type, is to be found in Treadgold, Nature. 17 Henry, Bibliothe`que, 2:211: “Il n’a vraiment retenu que l’essentiel des faits d’ordre biographique et il a ˆte´ nombre de longs de´veloppements ´edifiants et monotones, de redites absolument inutiles.” Cf. laisse´ de co ˆte´, ici comme ailleurs, les longs de´veloppements ´edifiants de Jean et les ibid., 209: “Photius laisse de co discussions, toujours les meˆmes, qu’il a avec ceux qui s’adressent `a lui.” 18 Wilson, Bibliotheca, 113. 19 Halkin, Douze re´cits, 69–285. More than twenty manuscripts of the work are known (see ibid., 69 n. 3; and a list in C. Baur, “‘Georgius Alexandrinus,’” BZ 27 [1927]: 1–2). On the commonly accepted identification of this George, bishop of Alexandria, with the bishop whose time of office was from ca. 620 to 630, see P. R. Norton, “The Vita S. Chrysostomi by Georgius Alexandrinus,” CPh 20.1 (1925): 69–72; however, Baur, “Georgius,” 5–7, rejects the identification. 20 Instead of being inspired by his immediate source, Photius’s tu´ po" kai` kanw´ n appears to echo elements of monastic hymnography, a context to which ejgkra´ teia belongs as well (cf., e.g., H. Follieri, Initia hymnorum ecclesiae graecae, MR I 27, MV XI 64; and J. Schiro, Analecta hymnica graeca e codicibus eruta Italiae inferioris, 13 vols. [Rome, 1966–83], 8:76). I thank Christian Hannick for these references.
¨ GG TOMAS HA
49
aspire to produce an artistic short account of his own, giving scenic life to the various episodes. It is plain factual summary all along; and when the narrative is occasionally twisted into clumsy constructions, those are of Photius’s own making and not taken over from his model. When he does borrow half a phrase, which happens now and then,21 it is for convenience, not for mimesis. The summarizing thus prunes George’s narrative of most of its descriptive details. Some further examples may be considered. While the description of John’s mother’s grief at the death of his father is elaborated by George for half a page (chap. 3, 76.29 ff), Photius simply refers to it by the plain word tou' pe´ nqou" (79a10). Correspondingly, the long debate on paganism and Christianity between John and Anthemius, the priest of Athena, is condensed from six pages of direct speech (pp. 80–86) into two lines (79a19– 21) that state the fact and explain what qualities in John’s speech made him the winner. Close to the end of the biography, George meticulously specifies what happened to each of John’s friends and supporters after his final deposition from the bishop’s throne, whereas Photius picks out just two names, adding that there were nearly twenty other bishops removed. This is the procedure all along. A strict selection of the basic facts and a plain narrative style thus combine to explain Photius’s brevity in those parts of the Life that he does choose to include; and that brevity, in fact, accounts for most of the shortening of the text into the above-mentioned proportion of 1 to 21.5. It would not be correct to infer from Henry’s and Wilson’s sweeping statements that this proportion is to any great extent the result of the wholesale omission of entire parts of the text, such as miracles or “moralising passages.” Such passages do disappear, it is true, but together with other types of sermons, conversations, comments, or descriptions. In fact, up to chapter 47, i.e., for two-thirds of the biography, there is hardly a single chapter22 that is wholly omitted; the principal topic of each receives its brief attention, in the chronological order of events. From chapter 47 on, Photius is obviously beginning to lose patience: some chapters are totally omitted, the summarizing becomes more radical still, and for the final fifteen chapters George’s narrative sequence is no longer respected. I see no reason to ascribe this phenomenon to any particular aversion to or disinterest in the kind of material narrated in the last third of the Life. The contents of the omitted chapters are not, in principle, different from those of the previous chapters that Photius did include in his summary. The explanation seems simple enough—a natural fatigue after reading and summarizing some two hundred pages of the author’s rather prolix narrative. Thus, the degree of compression, rather than disinterest in certain kinds of material, accounts for the so-called omissions. It is also beside the point to assert, as Henry does, that Photius omits the “repetitions” in George’s account, since the kind of material that George may be said to repeat is not part of Photius’s summary even the first time it appears. It is more interesting to try to define what Photius is really looking for. Among the various forms of treatment he uses, for this biography he has chosen the analytical epitome. Accordingly, it is not his intention to excerpt whole passages of particular mateE.g., 79b28–29; chap. 11, 102.15–16. A more special case is the almost literal quotation of the first sentences of Theophilus’s letter to John (82a31–34; chap. 39, 186.4–6). 22 The division into chapters is, of course, modern (Savile); but, with their very different lengths, they constitute topical units and are thus usable for our present purpose. 21
50
PHOTIUS AS A READER OF HAGIOGRAPHY
rial or stylistic interest, as in his excerpts proper; nor is he content with a very general re´sume´, as in his brief synthetical summaries; instead, he wishes to follow the course of events, from John’s birth to his death, in their chronological order. Predictably, he mentions the places to which John’s activities take him and the most prominent persons he meets, the various types of books he writes, the theological or political controversies within the Church that he takes part in, and so on; but with the same conscientiousness Photius records the miracles John performs, insofar as George describes them in some detail (cf. 79b19–34 and chaps. 8–9, 11–13; 80a17–25 and chaps. 18–19). It is, on the whole (at least for the first two thirds of the work), a very professional job, once the premises are accepted, and one Photius has performed with considerable loyalty to the text before him. What, then, does Photius mean by his concluding words that George oujk ojli´ga fai´netai paristorw'n, while nothing prevents the reader from choosing ta` crh´ sima for himself and overlooking the rest? Wilson’s translation of paristorw'n, “given to inaccuracy,” 23 and ˆte´ de la ve´rite´” are hardly correct; this is already indicated by the implied Henry’s “a` co contrast with ta` crh´ sima. Furthermore, in the summary itself there occurs no criticism of facts that would confirm Photius’s dissatisfaction with George’s factual accuracy or truthfulness (in contrast to his style). Kustas, in his article “History and Theology in Photius,” 24 interprets Photius’s words in a somewhat different way: George of Alexandria “appears to relate much that is contrary to history” (my italics). “Deviation from the truth of history,” Kustas explains, “prejudices the crh´ simon.” 25 This seems to be a far-fetched explanation, again lacking the support of any actual criticism of facts in Photius’s epitome. Since this appears to be the only place where Photius uses the verb paristorei'n,26 we cannot get any immediate external help.27 However, there may be a clue at the end of codices 256 and 258, where the simple verb ijstore´ w occurs: eij de´ tina para` tou` " a“llou" iJstorei' . . . (474a19), “If he narrates something differently from other writers . . .”; and ejn polloi'" de` kefalai´oi" kai` kaino´ tera para` tou` " a“llou" iJstorei' (485b6), “In many chapters the account is novel in comparison with other writers” (Wilson, p. 243). The sense in which the verb iJstorei' appears here, “he narrates,” seems applicable to our context (while the sense of the preposition para´ is not). The most reasonable interpretation, then, of paristorw'n is that George, according to Photius, “narrates things that are beside the main point.” And “the main point”—what is “useful” in reading a biography like this— is the factual account of the historical events of the saint’s life, the way Photius distills it in his own summary. When summarizing Procopius in codex 63, he did not admit the mythographical, geographical, strategical, or anecdotal material into his summary, but
23 The Italian translation by C. Bevegni, in Fozio, Biblioteca, ed. N. Wilson (Milan, 1992), 227, which preceded the English edition, similarly has: “Il nostro autore risulta spesso poco affidabile.” 24 Kustas, “History,” 53. 25 Kustas may have been influenced by the earlier (partial) English translation of the Bibliotheca by J. H. Freese, The Library of Photius (New York, 1920), 1:187: “The writer appears to relate much that is contrary to the truth of history.” 26 Cf. Orth, Stilkritik, 108. 27 LSJ, s.v., gives such alternatives as “inquire by the way” (corrected in the supplement of 1996 to “learn by the way”) and “narrate or notice incidentally,” neither of which suits our passage.
¨ GG TOMAS HA
51
concentrated on the historical events; similarly, his epitome of John’s Life is largely free from the edifying elaboration and scenic impersonation of his model. This does not necessarily mean that Photius would not be interested in that kind of material—no less so than in geography or pagan mythology, which he does excerpt in other codices of the Bibliotheca, provided the material seems exotic enough. It is, of course, quite possible, although he omits to say so, that he found George’s edifying contributions insignificant or unoriginal (and that this is why he chose the analytical epitome as his method in the first place). But, more importantly, Photius’s procedure may tell us something fundamental about his attitude to hagiography: it is similar to his attitude to historiography, i.e., he reads hagiography to learn the basic facts of the saint’s life rather than to receive edification or aesthetic pleasure. All this may vary, of course, according to the kind of hagiographical work that actually lay before Photius; therefore, I proceed to some other Lives of saints, which he included in the Bibliotheca, as a check. Photius’s epitome of a Life of Athanasius (cod. 258, BHG 184) seems to confirm our conclusions in some important respects, while differing in others. It should be noted, however, that we do not possess as clear-cut an object of comparison in this case as in the former one. First, some have assumed that Photius read only an abridgment of the work in question, basing their argument on the phraseology at the beginning and the end of the codex: ajnegnw´ sqh ejk tou' lo´ gou . . . (477b21) and o”ti hJ suggrafh` ejx h»" hJ prokeime´ nh proh'lqen ejklogh´ . . . (485b3). However, an analysis of Photius’s terminology in the later codices shows that the word ejklogh´ , “selection,” is used there to refer to his own written epitome, not to his model being an abridgment, exactly as the word e“kdosi" in similar contexts refers to Photius’s own “publication” of the text within the framework of the Bibliotheca; nor must the introductory phrase ajnegnw´ sqh ejk be interpreted as indicating an abridged model or a partial reading.28 The second objection is more serious: while Photius’s actual model was no doubt close to the extant “premetaphrastic” biography of Athanasius (BHG 185 ⫽ PG 25:CCXXIII–CCXLVI),29 the exact relationship remains obscure. Either both had the same (lost) model, or Photius’s model was a (lost) later redaction of BHG 185.30 If we assume, however, that Photius’s model was not very different from what we have 28 This is obvious already in the present codex if one only cares to compare the introduction with the conclusion. Henry translates them as “Lu en parti . . .” and “L’ouvrage dont est tire´ le pre´sent abre´ge´ . . .” (pp. 18, 40); Wilson, as “Read an abridgement of a work . . .” and “The work from which the present summary derives . . .” (pp. 231, 243); both translators thus render the conclusion correctly but go astray (in different ways) in the introduction (cf. Wilson, Bibliotheca, 243 n. 1: “Theoretically one could also translate ‘I have read part of . . .’,” which adopts Henry’s equally mistaken interpretation). For further demonstration of Photius’s usage in this part of the Bibliotheca, cf. Ha¨gg, Vermittler, 131–37, and Treadgold, Nature, 41–48. Treadgold suggests that the standard phrase ajnegnw´ sqh was added mechanically (by the secretary?) to epitomes that already had the heading ejk tou' lo´ gou . . . A correct translation should thus disregard the ejk (or imitate the anacoluthon resulting from the secretary’s mechanical procedure: “Read. [A selection] from . . .”). 29 J.-P. Migne, following the Maurist edition of Athanasius, refers to this text (PG 25:CCXXIII–CCXLVI) as “ex Metaphraste,” and to BHG 183 (PG 25:CLXXXV–CCXI) as “incerto auctore.” The reverse relationship was argued by B. Beck, “Die griechischen Lebensbeschreibungen des Athanasius auf ihr gegenseitiges Verha¨ltnis ¨ berund ihre Quellen untersucht” (diss., University of Jena, 1912), 79–80, and confirmed by A. Ehrhard, U lieferung und Bestand der hagiographischen und homiletischen Literatur der griechischen Kirche, TU 51 (BerlinLeipzig, 1937), 2:591 n. 10. I thank Christian Høgel for providing this reference. 30 The latter alternative is argued by Beck, “Lebensbeschreibungen,” 80–81.
52
PHOTIUS AS A READER OF HAGIOGRAPHY
in BHG 185, the following conclusions may be drawn with regard to Photius’s method. The degree of compression here is much less than in the codex devoted to the vita of John Chrysostom; the ratio may have been almost as little as 1 to 2.31 As a result, Photius’s summary is more vivid and readable; there are even some small portions of dialogue directly quoted (480b20 ff and 480b40 ff). When confronting a short Life, he understandably did not feel the need to reduce the size as drastically as when reading a work of hundreds of pages; he even allowed his summary to be longer this time than in the case of the Life of John Chrysostom. But in other respects, his method remains the same. He mainly keeps to the historical facts of the narrative, at the same time remaining loyal to his author; for instance, when an anecdote is the only thing told about Athanasius’s childhood and parents, this is what Photius summarizes, in quite some detail. Several letters from Constantine and others, directly quoted in the Life, are consistently replaced with a few words about their main contents, as are the sermons and discussions in the Life of John Chrysostom. The resultant epitome is thus not very different in our two cases: there are narrative and factual information throughout, and one would probably not have guessed that the two models were so different in narrative style and size. Little new light can be thrown on Photius’s method by his summary of a Life of Gregory the Great (cod. 252): according to Halkin and others,32 his model does not survive.33 The fact that his epitome consists of only two edifying episodes from Gregory’s life, rather than containing a narrative of his career, is obviously due to Photius’s model’s concentration on the same two episodes. All we have to judge the epitome’s relation in quantity to the original is his initial statement, ajnegnw´ sqh Grhgori´ou tou' Dialo´ gou oJ bi´o" ou» hJ e“kdosi" ejklogh´ n tina ajnagra´ fei (466b26), “Read the Life of St. Gregory Dialogus, of which the [present] publication records a selection.” 34 To all appearances, this is an epitome more similar in type to codex 258 (Athanasius) than to codex 96 (John Chrysostom): a short Life has again led to a less drastic abridgment. It may be registered as a novel trait, however, that even a Life without much historical information could interest Photius; and, in fact, in his concluding remarks he speaks in a positive tone about Gregory’s own biographical works, saying that he “included in them stories of edifying nature” (dihgh´ mata swthri´an ejkpaideu´ onta) (467b3). In the epitome proper, the telling of the first anecdote is excused by the remark that the episode narrated “may be specially indicative of (a‘n ma´ lista carakthri´seie) his wonderful humanity and charity” (466b38). This is, of course, the classical motivation among biographers, known especially from Plutarch (Alex. 1.2–3), for concentrating on typical or “characterisMy calculations have shown ca. 31,500 characters in Photius’s summary and ca. 66,000 in BHG 185. See note 8 above. Unfortunately, Berschin, Biographie, 387, follows F. Halkin, “Une courte Vie latine ine´dite de Saint Gre´goire le Grand retraduite du grec,” Me´langes Euge`ne Tisserant (Vatican City, 1964), 4:379– 87, obviously ignorant of the fact that Halkin’s hypothesis has been commonly rejected; the anonymous Greek text (published by R. Abicht and H. Schmidt, “Quellennachweise zum Codex Suprasliensis,” ASP 18 [1896]: 152–55) does not belong to John’s Nachleben, but possibly derives from one of his sources. 33 Treadgold, Nature, 30, observes that “codex 252 has no verbal parallels with Halkin’s text.” 34 Cf. ibid. Wilson, Bibliotheca, 228, in his translation (“Read selections from . . .”) is misled by Henry’s text, which prints the version of the manuscript M, in spite of acknowledging (Henry, Bibliothe`que, 7:207 n. 1) that the A version is the original one (cf. Treadgold, Nature, 30 n. 41; Schamp, Historien des lettres, 70 n. 5). Schamp, ibid., 70, translates correctly: “Lu de Gre´goire ‘le Dialogue’ la Vie dont cet publication transcrit un choix,” but then confuses the matter again in his n. 5. 31 32
¨ GG TOMAS HA
53
tic” episodes rather than the historically well-known events in a person’s life. It would no doubt be rash simply to ascribe the remark to Photius himself, for it is quite possible that he repeats it from his source—which had, after all, made the actual choice of preferring the anecdotal to the historically important. But a repetition, too, would be indicative. One may compare this approach with Photius’s own practice in his series of excerpts from the Parallel Lives of Plutarch in codex 245, where he in fact omits the historical matter, both biographical and political, in favor of a number of anecdotal and moralizing passages.35 His interest in the short and anecdotal Life of Gregory the Great is thus not without parallels, if one widens the perspective to include ancient biographical literature as well. I shall not go into similar detail with regard to the remaining Lives. The anonymous Politei´a of Metrophanes and Alexander, bishops of Constantinople (or rather Byzantium), to which Photius devotes codex 256 (BHG 1279), is in reality an account of church politics in the time of Constantine, with the two bishops as shadowy figures in the background.36 Photius summarizes this account in considerable detail—the ratio is approximately 1 to 3—and seems quite happy with reading not an actual saint’s Life but a piece of church history. Codex 257 (BHG 1472), the Life of Paul the Confessor, is again an analytical epitome of what appears to be a rather similar ratio (1:2 to 1:3?), with instances of literal borrowing within a rather free summary.37 Together with the Life of Athanasius (cod. 258) studied above, these two obviously form a subgroup among Photius’s analytical epitomes,38 characterized by an unusually low degree of compression. This may be what Photius himself refers to at the beginning of codices 257 and 258: ejf∆ oJmoi´a sustello´ meno" ejkdo´ sei (474a24), “shortened to be published in a similar form,” and to` n i“son diatupou´ meno" tro´ pon (477b23), “fashioned in the same way.” 39 Similar expressions may in fact be followed back continually as far as codex 253 (except for codex 254). At the beginning of codex 253, Photius states: . . . ejx ou» kefalaiw´ dh" dietupw´ qh crei´a (467b17), “from which a useful summary has been fashioned,” or (adapting Wilson’s paraphrase) “from which a concise narrative of the essential points [has] been prepared.” 40 This, then, would be Photius’s way of describing his procedure in these five (or six, if we regard the omission in codex 254 as accidental) hagiographical codices. Of the three Cf. Ha¨gg, Vermittler, 139–41. The text is printed by M. Gedeon, “Mnhmei'a th'" ejkklhsiastikh'" iJstori´a" tou' d⬘ aijw'no",” jEkk. Alh j ´ q. 4 (1884): 285–91, 296–300, 306–10, 321–26; Photius’s treatment is discussed by P. Heseler, “Hagiographica III,” BNJ 13 (1936–37): 81–92, as well as by Henry, Bibliothe`que, vol. 7, in his notes. On the character of BHG ¨fe Metrophanes und Alexandros von Byzanz,” BZ 59 (1966): 1279, see also F. Winkelmann, “Die Bischo ¨ berlieferung der Vita Metrophanis et Alexandri (BHG 1279),” in 47–71, and idem, “Die handschriftliche U Studia Patristica, ed. F. L. Cross, TU 92 (Berlin, 1966), 7.1:106–14. 37 See Henry, Bibliothe`que, 8:215 (and successive notes), and the unpublished edition of the original life (BHG 1272a) in E. D’Haene, “Vita Pauli Episcopi Constantinopolitani: Tekstuitgave en commentaar,” 2 vols. (diss., Rijksuniversiteit Gent, 1971–72), vol. 2 (the handwritten format of the Greek text makes an exact numerical comparison with Photius laborious), with comments on Photius’s version in 1:114–17, 133–34. 38 This was already pointed out by D’Haene, “Vita Pauli,” 1:133–34. 39 Wilson, Bibliotheca, 243 n. 2, explains, “i.e. abridged,” mistakenly attributing this quality to the work that Photius read (cf. also ibid., n. 1); but it is Photius’s own version that is an abridgment, not his model. 40 Wilson, Bibliotheca, 230 (I substitute “has” for “had,” since again the reference is not to the model but to Photius’s version). 35 36
54
PHOTIUS AS A READER OF HAGIOGRAPHY
that have not yet been discussed, two deal with hagiographical legends of early popular saints, the Seven Child Saints (or Sleepers) of Ephesus (cod. 253)41 and St. Demetrius (cod. 255).42 Since these stories were widely spread and many versions survive, it is difficult to identify models and thus to say anything specific about the ratio of compression. It may be that they are so short simply because the models were short, rather than because Photius had less interest in their contents than in the historical Lives, as I suggested at the start. As low a ratio as 1:2 to 1:3 would be the logical inference from Photius’s statement about using the same method; but I am not sure that one is allowed to take his statement as a very technical declaration. Rather, I have the impression that the summarizing is considerably more radical in these codices,43 mirroring Photius’s personal priorities. Only a thorough examination of the manuscript tradition of each of the two hagiographical legends might give us an answer. There remains only the Martyrdom of Timothy the Apostle in codex 254. Photius starts with a short summary of the Martyrdom (27 lines) and then adds two items, the second of which is an excerpt describing the Ephesian Katagogia feast to Dionysius (468b35–469a2). Thanks to that excerpt, it is possible to state that Photius read a version of the Martyrdom that was fairly close to the one surviving in Greek.44 This identification would not have been possible on account of the preceding summary only, in which Photius retells in his own words and sequence some of the contents of the Martyrdom—in other words, a typical synthetical summary of his. He is loyal to his text in that he retains (without comments) its tendency to concern St. John’s doings in Ephesus more than St. Timothy’s own.45 Photius ends his summary with the words tau'ta dh` kai` toiau'q∆ e”tera dihgei'tai to` martu´ rion aJplouste´ ra fra´ sei tou' aJgi´ou Timoqe´ ou (468b29), “These and similar things are told in the Martyrdom of St. Timothy in a rather simple style.” He has obviously got tired of recounting the whole story (all the more so since its style did not correspond to his own ideals; cf. below, p. 57 and note 59). His impatience in this case Cf. BHG 1593–99; on the many known versions (in various languages) of the legend of the Seven Sleepers of Ephesus, see the monograph of P. M. Huber, Die Wanderlegende von den Siebenschla¨fern: Eine literargeschichtliche Untersuchung (Leipzig, 1910). For further references, see Henry, Bibliothe`que, 7:209 n. 2. According to Huber, Wanderlegende, 40–43, Photius’s model must have been close to a couple of the extant Greek versions, but identical with none, since it contained original elements absent in these. 42 Cf. BHG 496–98. Henry, Bibliothe`que, 7:212 n. 4, notes that the two Lives of St. Demetrius printed in PG 116 have little in common with Photius’s summary. The unedited Athos manuscript with the same incipit as in Photius, to which Henry (ibid.) refers (⫽ BHG 496b), is in fact a transcription of Photius’s summary (according to Treadgold, Nature, 164). 43 With regard to cod. 253, this is Huber’s impression too: “Es scheint, dass Photius, der, wie er selbst am Anfang sagt, diese Legende gelesen und nun kurz im Auszug geben will, seine Vorlage in sehr freier Weise benu ¨ tzt hat . . .” (Huber, Wanderlegende, 41–42), i.e., it is a synthetical summary rather than an analytical epitome. 44 Edited in H. Usener, Acta S. Timothei (Bonn, 1877), 7–13. According to Henry, Bibliothe`que, 7:212 n. 1, 213 n. 3 (and Usener, Acta, 30), the literalness of the excerpt is not marked enough to warrant that this is the very version of the Martyrdom that Photius read; in the summary itself, the verbal coincidences are few and insignificant. If Photius’s model was close to Usener’s version in compass, the ratio between the summary and the original would be approximately 1 to 4.5 (ca. 1,350 against ca. 6,000 characters); but such a figure has less significance in the case of a synthetical summary than when relating to an analytical epitome. 45 For the earlier discussion of this apparent disproportion in the Martyrdom and an attempt at explanation, see H. Delehaye, “Les Actes de Saint Timothe´e,” in Anatolian Studies Presented to W. H. Buckler (Manchester, 1939), 77–84. Delehaye opposes Usener’s (and others’) high opinion of the authenticity of the sources for the Martyrdom; he finds it largely fictitious and places its composition in the 5th century. 41
¨ GG TOMAS HA
55
emerges after his reading only four pages and writing less than one (of printed text); with the historical Life of John Chrysostom, the similar reaction does not happen until he has read more than two hundred pages and written ten. It is not difficult to see where his main interests lie. THE CRITICISM Photius is, of course, the prime object for the study of literary criticism in the middle Byzantine period. In several of the codices of the Bibliotheca he embarks on detailed stylistic judgments, and there are interesting passages of the same kind in his Letters and the Amphilochia as well. There seems to be a fairly general consensus that these stylistic judgments are Photius’s own, in contrast to his biographical sketches, which he has mostly taken over from others.46 For his literary criticism, Photius mainly uses the tools provided by the rhetorical tradition from antiquity, especially the terminology codified by Hermogenes in the second century A.D. But he also borrows from earlier writers, such as Demetrius and Dionysius of Halicarnassus. These sources of Photius’s critical vocabulary were carefully explored in the late 1920s in works by Orth and Gu ¨ nther Hartmann;47 but for an inspired view of what it is that Photius is really doing with this borrowed terminology, one has to turn to an article by Kustas of 1962, with a telling subtitle, “The Literary Criticism of Photius: A Christian Definition of Style.” 48 This is the kind of article in which many of the most important insights are provided in the footnotes, while the main argument is not always quite convincing. Still, on balance, I think that Kustas has managed to capture some of the spirit, if not always the letter, of Photius’s critical project. His key to an understanding of Photius is the parallelism in vocabulary between ethics and literary criticism. Now, the correlation between stylistic and ethical values was of course part of the ancient critical tradition as well; but what happens if ethics change, while the terms remain the same? It is true, Kustas admits to Orth and Hartmann, that Photius’s terminology of literary criticism mainly draws on the classical tradition, but it “is adjusted so that the adjectives descriptive of literary standards become identical with those expressing ideals of Christian behaviour.” 49 Style and character are made to coincide. This is demonstrated in detail and explained as the result of Photius’s effort to come to grips with the style of St. Luke, St. Paul,50 and other Early Christian writers, which obviously does not conform to the classical norms of style. To condense a long argument into one concrete example, the ca´ ri", “(divine) grace,” of St. Paul’s words is inseparable from the ca´ ri", “charm,” of his style. In contrast, we may add, the Apocryphal Acts of the Apostles, which Photius treats in codex 114, are found to be lacking, in perfect conformity with their heretic contents, the “inborn charm” (th'" . . . ejmfu´ tou ca´ rito", 90b27) that characterizes the Gospels.51 Cf. Schamp, Historien des lettres, passim. Orth, Photiana; idem, Stilkritik; G. Hartmann, “Photios’ Literara¨sthetik” (diss., Rostock, 1929). Cf. also R. Henry, Essai sur le vocabulaire technique de la rhe´torique du patriarche Photius (Lie`ge, 1931) (non vidi), and idem, “Proclos et le vocabulaire technique de Photius,” RBPH 13 (1934): 615–27. 48 See above, note 7. 49 I quote from Kustas’s own summary of his earlier article in “History,” 65 n. 88. 50 In this respect, Kustas could partly base himself on the study of B. Wyss, “Photios u ¨ ber den Stil des Paulus,” MusHelv 12 (1955): 236–51. 51 Cf. Junod, “Actes apocryphes,” 18–19. 46 47
56
PHOTIUS AS A READER OF HAGIOGRAPHY
Thus, though using the old terms, Photius creates his own theory of style, to which Christian ethical values are central. Kustas arrives at a very positive judgment of Photius’s originality and subtleness as a literary critic, in some ways more “modern” than his Byzantine successors. In a recent article, Dmitrij Afinogenov has returned to these “aspects of innovation” in Photius’s literary criticism,52 he agrees with Kustas’s high appreciation of Photius’s achievement in this area53 but, by widening the perspective, reaches partly different conclusions. He shows that Photius applies terms like e“mfuto" ca´ ri", “inborn charm,” not only to the Gospels or St. Paul but also to the style of pagan authors; the introduction of such terms as “inborn beauty” or “naturalness” was, he argues, “a conscious and purposeful modification” of Hermogenes (p. 341). They become stylistic terms, applicable to pagan and Christian authors alike. To Photius, according to Afinogenov, St. Paul and the apostles follow “universal rules of the art of persuasion, not some particular techniques invented by pagan scholars” (p. 344). On the whole, Afinogenov concludes, Photius’s attitude to literature “looks much closer to the modern perception of art than anyone could imagine” (p. 345). So much for the general background. Returning to our eight hagiographical codices in the Bibliotheca, we are in for some disappointment. Partly, this is no doubt due to the fact that seven of the eight codices belong to the second part of the Bibliotheca, in which excerpts and epitomes dominate and personal comments are few.54 Most of Photius’s detailed literary criticism is in the first part, in which codex 96 is the only hagiographical codex proper. But it also, I would suggest, tells us something about Photius’s attitude. He does not seem to have paid any particular attention to hagiography as an important and typically Byzantine genre. One may compare it with epistolography, a form of literature that Photius himself practiced all his life and on which he often makes interesting remarks when collections of letters are included in the Bibliotheca. He repeatedly refers to what he sees as “the rules of the epistolographic style”—[oJ] ejpistolimai'o" carakth´ r (cod. 143, 98b31) or oJ tw'n ejpistolw'n carakth´ r (cod. 138, 98a19). From his scattered remarks, it has been observed, “it is possible to extract a definite theory of epistolography.” 55 One may also compare Photius’s approach to hagiography with his reference to something he calls the “ecclesiastical” style: aJplou'" de` kata` th` n fra´ sin kai` safh´ " ejsti kai` ejggu` " tou' ejkklhsiastikou' kai` ajperie´ rgou carakth'ro" (cod. 126, 95a22), . . . meta` safhnei´a" kai` aJplo´ thto" kata` to` n ejkklhsiastiko` n th'" eJrmhnei´a" tu´ pon (95a42).56 These remarks concern Clement’s letters to the Corinthians and Polycarp’s letter to the Philippians, respectively, and the terminology does not seem to appear in connection with the works of church history that Photius actually treats in the Bibliotheca.57 The typical or ideal “ecclesiastical” style is evidently supposed to be clear and simple. With regard to hagiography, nothing similar exists to indicate that Photius was interD. E. Afinogenov, “Patriarch Photius as Literary Theorist: Aspects of Innovation,” BSl 56 (1995): 339–45. ˇevcˇenko, “Levels of Style in Byzantine Prose,” JO ¨ B 31.1 (1981): 299. Partly modifying the views of I. S Afinogenov (“Theorist,” 339 n. 2) also bases himself on a study by S. S. Averintsev, which I have not been able to consult. 54 For the characteristics of the second part and an attempt at explanation, see Treadgold, Nature, 37–51. 55 Kustas, “Criticism,” 152; cf. also Hartmann, “Literara¨sthetik,” 50–51. 56 ˇevcˇenko, “Levels,” 296; Afinogenov, “Theorist,” 342. Cf. S 57 But cf. also cod. 67 (34a1) on Sergius the Confessor’s style as suited to ecclesiastical history. 52 53
¨ GG TOMAS HA
57
ested in the rules of this genre, as he was in the theory of epistolography, or that he associated it with some particular level or ideal of style, as he did for church history. Perhaps this is only a natural reaction to the fact that hagiography confronted him in greater abundance and diversity of style and ethos than any other form of Byzantine literature: he could not see the forest for the trees. He has more interesting things to say about pagan biography as a genre, in particular Damascius’s Life of Isidore (cod. 181),58 than about its Christian continuation. But his failure to create a generic theory or define a special style for hagiography does not mean that he refrains from any kind of stylistic comment on the specimens he has read. On George of Alexandria he comments: e“sti me´ ntoi th` n fra´ sin aJplou'" kai` eij" pollh` n cudaio´ thta katenhnegme´ no", mhde` tou'to dh` to` para` toi'" grammatikoi'" kata` cei'ra", th` n tw'n ojnoma´ twn kai` rJhma´ twn su´ ntaxin hjkribwme´ no" (78b27–31); or, in Wilson’s translation, “However, in his style he is simple and falls into great vulgarity, because he is not even precise in his use of nouns and verbs, which is well within reach of the educated” (p. 113). Here, then, is one of the instances where aJplo´ th", “simplicity,” is not a virtue of style to a Byzantine mind.59 The term for “vulgarity,” cudaio´ th" with the adjective cudai'o", is to be found five more times in the passages of literary criticism in the Bibliotheca, always with reference to Christian works.60 For instance, commenting on the authenticity of a work attributed to John Chrysostom, Photius notes that it is “vulgar” in style and thus cannot be authentic (cod. 274, 510b13). Concerning Ephraem Syrus he notes that it is a wonder (qaumasto´ n) that such salvation and benefit can issue forth by means of such a vulgar style (cod. 196, 160b7)—an inversion of what he says about some Greek novels, namely, that indecent and useless contents are conveyed in an accomplished style (codd. 87, 94). I look at two further stylistic comments in the hagiographical codices. They are rather similar in structure, and I need to quote only one of them, occurring at the end of the Life of Metrophanes and Alexander (cod. 256): . . . hJ suggrafh` . . . ou“te pantelw'" eij" to` dihkribwme´ non kai` sofo` n th'" fra´ sew" kai` th'" dianoi´a" ejkmemo´ rfwtai, ou“te pro` " to` cudai'on kai` hjmelhme´ non diape´ ptwken (474a15–18), or “The work . . . is neither wholly structured so as to achieve precision and skill in style and thought, nor does it surrender to vulgarity and negligence.” At the end of the Life of Athanasius (cod. 258, 485b4–6), there is the same contrast between “negligence” and “precision,” with the same terms used, and there is also a parallelism implied between style and content. This characterization of the extremes between which the work under review is said to be situated is something very typical of Photius’s critical method. What is missing here, however, is some word of praise for the aurea mediocritas—for the natural and the mean (to` me´ tron) are his avowed stylistic ideal. This is no “golden” mean, we may infer, but just ordinary dull prose. Evidently, none of the hagiographical works Photius reviewed had a particular appeal to him aesthetically.
Esp. 126b33–38. Cf. Kustas, “History,” 57–58. Cf. also cod. 254, 468b30 (quoted above); but contrast, e.g., cod. 126, 95a22 (also quoted above). Cf., on ˇevcˇenko, “Levels,” 303, and G. L. Kustas, Studies in the parallel issue of obscurity as a potential virtue, S Byzantine Rhetoric, Analekta Blatadon 17 (Thessalonike, 1973), 91–93. 60 Codd. 89 (66b34), 107 (88a37), 196 (160b7), 256 (474a18), 274 (510b13). 58 59
58
PHOTIUS AS A READER OF HAGIOGRAPHY CONCLUSION
Hagiography, then, is not a favored genre in the Bibliotheca, nor does it seem to have been a form of literature in which Photius took a special theoretical interest. In spite of his concern for ancient biography, he does not make any specific comparisons between that genre and the Lives of saints he reviews. Rather, he appears to have read those Lives of saints to which he devotes most energy and space as historical narratives, valuable for the information they give on the career of prominent ecclesiastical figures and on their times. Edification does not seem to be what he is seeking in hagiography, nor what he is recommending others to read it for. Still, he is remarkably loyal to the works he reviews, and it is more by his choice of works than by arbitrary epitomizing that he reveals his real preferences. His taste, by implication, seems to have been not too dissimilar from that of many modern scholars: what they deprecate in Byzantine hagiography, Photius simply omits; and his emphasis on the historical facts, almost to the repression of the edifying and aesthetic aspects, prefigures what has long been the dominant trend in modern Western scholarship. University of Bergen
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
The Aesthetics of Sacred Space: Narrative, Metaphor, and Motion in Ekphraseis of Church Buildings RUTH WEBB
INTRODUCTION: THE PROBLEMS OF DESCRIPTION ttempting to describe the interior of Hagia Sophia in Constantinople in the twelfth century, Michael the Deacon proclaimed the difficulty of the task: it is barely possible for anyone to encompass such a vast and varied subject in a discourse.1 His complaint is, of course, a familiar rhetorical topos, part of the speaker’s ritual captatio benevolentiae. But, like many topoi, it expresses a significant truth: description, the representation of the visual through the medium of the word, is a problematic enterprise that raises many questions. To what extent is it possible, for example, to represent a material object in an immaterial, intelligible medium such as language? How does one represent any static three-dimensional object in a medium that unfolds in time? How can one represent in words the totality of visual experience—the infinite varieties of color, space, depth, texture, light, and shade—offered by even the simplest object?2 The more one considers these questions, the more justified Michael’s complaint seems, as do the words of Photios, who in his tenth homily claims that the church of the Virgin of the Pharos surpasses the canons (no´ moi) of ekphrasis, or the second-century rhetor Aelius Aristeides, who points out that the city of Rome is too vast to see, let alone to describe.3 Clearly a verbal description can never be entirely adequate to its object. This observation, self-evident as it may be, has important implications: the composition of
A
The original version of this article was written with the support of the British Academy Post-doctoral Fellowship. I would like to thank Margaret Mullett, Liz James, Asen Kirin, and the anonymous reviewers for their comments. 1 C. Mango and J. Parker, “A Twelfth-Century Description of St. Sophia,” DOP 14 (1960): 233–45 (text by Michael the Deacon, referred to hereafter as Michael the Deacon, Ekphrasis), 1.22–26: ouj cei'ron de` kai` lo´ gv perihgh´ sasqai, o”son e“ni, mh` ga` r to` pa'n, ejpei` kai` mo´ gi" tou't∆ aujto´ ti" lo´ gou pro´ fasin ejnsthsa´ meno" tv' panti` ejpexi´oi, poiki´lw" e“conti kai` makrv'. 2 For the discussion of these questions, see M. Beaujour, “Some Paradoxes of Description,” Yale French Studies 61 (1981): 27–59. I would not seek to suggest that individual rhetoricians necessarily felt themselves at a loss when tackling such subjects, simply that the conventions of the genre within which they were working allowed for the expression of the real difficulties involved in describing. 3 Photios, Homiliai, ed. B. Laourdas (Thessalonike, 1959), 10.7: to` ka´ llisto´ n te ei«nai to` n nao` n kai` wJraio´ taton kai` nikw'nta no´ mou" ejkfra´ sew" parasth'sai proh´ rhmai; Aelius Aristeides, To Rome, in Opera, ed. B. Keil (Berlin, 1898), 2:39.1–2: peri` h»" [i.e., Rome] mh` o”ti eijpei'n kata` th` n ajxi´an e“stin, ajll∆ oujd∆ ijdei'n ajxi´w" aujth´ n.
60
THE AESTHETICS OF SACRED SPACE
any description involves the selection of the details to be included.4 Moreover, the linear unfolding of a text—still more of an orally delivered discourse—demands the imposition of a temporal order onto material that is in reality perceived simultaneously by a viewer. For although a work of art or architecture may be experienced sequentially, as one moves through the building or lets one’s eyes travel across the scene, any single glance takes in more than could be expressed in a single statement.5 These and other related questions have been highlighted by modern criticism, but they are relevant to the reading of descriptive writing from any period. As the inadequacy topos suggests, the classical and Byzantine rhetorical traditions may be seen to anticipate, albeit in less explicit terms, the observations of modern criticism. What is most interesting in the present context, however, is the particular response to the problem that one finds in Byzantine ekphrasis. In this article I consider the literary problems involved in describing a subject as complex in appearance and function as a church building, and identify some of the strategies employed by authors of ekphraseis to tackle these difficulties. To some extent these strategies had their roots in the classical rhetorical tradition of ekphrasis, which I discuss briefly. But the description of the interior of a building as replete with meaning as was a church was a new task, for which there existed no direct precedent in the classical tradition, and thus it is particularly interesting to see how Byzantine authors approached this challenge. The present article is therefore first and foremost a study of the aesthetics of the literary and rhetorical representations of sacred space, concentrating precisely on those passages or aspects of the works that tell us little about the concrete fabric of the buildings, but much about how they could be written about and, perhaps, how they were perceived. I concentrate in my survey on the descriptions of two major Constantinopolitan monuments, Hagia Sophia and the church of the Holy Apostles. In the sixth century, Justinian’s first rebuilding of Hagia Sophia was celebrated by Prokopios of Caesarea, and his second by Paul the Silentiary.6 Six centuries later, Michael the Deacon celebrated the “eternal novelty” of the building in a speech probably composed for the encaenia.7 Constantine Rhodios’s verse ekphrasis of the church of the Holy Apostles and the Wonders of Constantinople was dedicated to Constantine VII in the tenth century.8 The occasion for the composition of Nikolaos Mesarites’ prose ekphrasis of the Holy Apostles is unclear.9 Other examples of ekphrasis provide useful comparisons with these works. The ekphrasis embedded within Eusebios’s panegyric on Paulinos’s new church at Tyre stands at the beginning of the tradition of ekphraseis of church buildings and for this reason deserves 4 See, for example, R. Barthes, “L’effet de re´el,” in Litte´rature et re´alite´, ed. R. Barthes et al. (Paris, 1982), 86: “Si elle n’e´tait pas soumise `a un choix esthe´tique ou rhe´torique toute ‘vue’ serait ine´puisable par le discours.” 5 Beaujour, “Some Paradoxes.” 6 Prokopios, Buildings 1.1; Paul the Silentiary, Ekphrasis of Hagia Sophia, in P. Friedla¨nder, Johannes von Gaza und Paulus Silentiarius (Leipzig, 1912), 227–65 (hereafter Paul the Silentiary, Ekphrasis). 7 Mango and Parker, “Twelfth-Century Description.” 8 Constantine Rhodios, Description des oeuvres d’art et de l’e´glise des Saints Apoˆtres, ed. E. Legrand (Paris, 1896) (hereafter Constantine Rhodios, Ekphrasis). 9 Nikolaos Mesarites, Description of the Church of the Holy Apostles, ed. and trans. G. Downey, TAPS, n.s., 47 (1957): 855–924 (hereafter Mesarites, Ekphrasis). Downey suggests that it may have been composed in order to raise the prestige of the church after the imperial tombs were plundered by Alexios III in 1197.
RUTH WEBB
61
particular attention. Chorikios’s encomia of Bishop Marcian include some useful comments about the problems of describing, as do the homilies by Leo the Wise and Photios that celebrate newly restored or newly built churches.10 These ekphraseis of church buildings have traditionally been used as sources of information for reconstruction and mirrors of the realities they describe.11 In the case of the Holy Apostles the complete destruction of the actual monument has made the need for this type of reading particularly acute. But the problems involved in such referential readings are notorious;12 when measured against the yardstick of “objective description,” the ekphraseis are usually found wanting. However, as the questions raised at the beginning of this article suggest, “objective description,” although a useful working notion, is itself neither self-evident nor unproblematic. Even the most starkly informative guidebook description involves a drastic selection of details to be expressed in words.13 Such an account may contain verifiable statements about material elements of the construction, specifying dimensions and orientation and using technical terminology, but much is still omitted. Any description necessitates the selection and ordering of details according to the rules of language. In this sense one can argue that all descriptions, and not just those by Byzantine authors, are mirrors that must in some way “distort” their subjects. At the same time, by the very act of selecting, ordering, and presenting material they can act as a sort of commentary on their subjects. In order to see how these ekphraseis can be best used in relation to the monuments they describe, it is surely important to treat them first and foremost as texts with their own logic and organizing principles.14 If we read them in this light, the distinction between those words and phrases that refer to the material appearance of the buildings and those (often dismissed as “rhetorical ornamentation”) that refer to other, unseen, aspects becomes less important. A literary approach to ekphrasis can contribute to a study of aesthetics from two points of view. Most simply, it can tell us about 10 Eusebios, Historia ecclesiastica 10.4; Chorikios, Laudatio Marciani 1 and 2, in Choricii Gazaei opera, ed. R. Foerster and E. Richtsteig (Leipzig, 1929); Photios, Homiliai 10; Leo the Wise, Panygerikoi logoi, ed. Akakios (Athens, 1868), 28, 34. Akakios’s edition is difficult to find; partial translations are given by C. Mango, The Art of the Byzantine Empire (Toronto, 1986), 202–5, and A. Frolow, “Deux ´eglises byzantines d’apre`s des sermons peu connus de Le´on VI le Sage,” EtByz 3 (1945): 43–91. 11 ˜ oz, “Alcune fonte letterarie per la storia dell’arte bizantina,” NBACr 10 (1904): 221–32, and idem, A. Mun “Le ejkfra´ sei" nella letteratura bizantina e i loro rapporti con l’arte figurata,” in Recueil d’e´tudes de´die´es `a la me´moire de N. P. Kondakov (Prague, 1926), 139–42, was one of the first to suggest the use of Byzantine ekphraseis as a source for art history. Legrand (Description des oeuvres) and Friedla¨nder (Johannes von Gaza) include diagrammatic reconstructions of the monuments described by Constantine and Paul, respectively. 12 On the problems raised by the ekphraseis of the Holy Apostles, see particularly A. Epstein, “The Rebuilding and Redecoration of the Holy Apostles in Constantinople: A Reconsideration,” GRBS 23 (1982): 79–92. 13 See, for example, the account of the dome of Hagia Sophia in J. Freely, Blue Guide: Istanbul (London, 1997), 48: “The main support for the dome is provided by four enormous and irregularly shaped piers standing in a square approximately 31m on a side. From these piers rise four great arches, between which four pendentives make the transition from the square to the circular base of the dome. Upon the cornice of the circular base rests the slightly elliptical dome, of which the east-west diameter is about 31m and the north-south approximately 33m, with the crown soaring 56m above the floor . . .” R. Krautheimer, “Introduction to an ‘Iconography’ of Medieval Architecture,” JWarb 5 (1942): 1–33, contrasts this modern conception of architecture with the interests of medieval (mainly Western) architects and patrons. 14 Epstein, “Rebuilding and Redecoration,” demonstrates the importance of an understanding of literary form to the assessment of the historical value of ekphraseis.
62
THE AESTHETICS OF SACRED SPACE
the textual aesthetics of description, the rhetorical strategies favored by authors in presenting their subject matter. But there is also a more complex dimension: these particular texts bear a certain relation to a material object and often to a particular occasion;15 understanding their literary form can help to clarify that relationship and maybe point toward the aesthetics of the buildings themselves and their qualities, perhaps not always material or visible ones, that may have been perceived and appreciated. It is for this reason that I have chosen to treat these texts, which span a period of six centuries, in a single survey: all address the same basic problems of describing and share strategies of verbal representation. TEXTS AND THEIR CONTEXTS The historical relationship between the texts in question and their subjects is often fairly easy to identify, although each occupies a slightly different position within a complex set of factors such as patronage, occasion, and genre. Most were composed for a particular occasion and were originally recited in or close to the monuments they describe. Eusebios pronounced his panegyric within the church itself, while Paul the Silentiary’s ekphrasis appears from the manuscript headings to have been recited in two different locations, the Imperial and Patriarchal palaces.16 The audience were thus either familiar with the buildings or actually able to see them as they listened to the speaker, a paradox to which several authors refer.17 At the same time the authors also had to bear in mind possible later audiences. Chorikios distinguishes the audience in the church from the future readers who have never seen it but who may gain a vivid impression from his writings.18 In the fourth century, Libanios’s autobiography gives evidence for the circumstances in which an orator might give a repeat performance of his own work. The story concerns the unfortunate Bemarchios who presented an old speech of his celebrating Constantius’s Great Church at Antioch. According to Libanios, the members of the audience were left glancing at each other in utter confusion as Bemarchios “rambled on about pillars, trellised courts, and intercrossing paths which came out heaven knows where.” 19 Whether the failure was due to Bemarchios’s poor rhetorical skills or can be ascribed to sheer ill will on the part of Libanios we shall never know; but the anecdote serves to remind us that these speeches in their original contexts had functions that went far beyond the merely informative or representational. This later performance was part of an ongoing professional rivalry between Bemarchios and Libanios. In Bemarchios’s case, the nature of the audience and the occasion made all the difference to the reception of the speech—for, as Libanios tells us, Bemarchios had earlier made his fortune with the very same speech.20 For a rich exploration of the relation between Paul the Silentiary’s ekphrasis of Hagia Sophia and its historical context, 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. 16 M. Whitby, “The Occasion of Paul the Silentiary’s Ekphrasis of S. Sophia,” CQ 35 (1985): 216–17. 17 Chorikios, Laudatio Marciani 1.16. 18 Ibid. 19 Libanios, Autobiography, ed. and trans. A. F. Norman (Oxford, 1965), 41: diexio´ nto" aujtou' ki´ona" dh´ tina" kai` kigkli´da" oJdou´ " te uJpallh´ lwn temnome´ na" ejmpiptou´ sa" oujk oi«d∆ o”poi. . . . 20 Ibid., sections 39–41. 15
RUTH WEBB
63
Such detailed, albeit partial, information about the context of a performance and the audience response is rare. For the most part we are limited to the authors’ own accounts of the aims and problems of their enterprise. In several passages Chorikios discusses the limitations of verbal representation—the aspect of Bemarchios’s speech that Libanios had singled out for ridicule. Speaking on the church of St. Sergios, Chorikios notes that his description can encompass only those elements of its subject that are suitable for rhetorical presentation, going on to remind his audience that a mere imitation (mi´mhsi") will always fall short of accuracy.21 Leo the Wise, in his speech on the church of Stylianos Zaoutzas, discusses the challenge of describing a sight that his audience can see. In an elegant proemium he characterizes his enterprise as a reward to the artist, granted in exchange for the pleasure of the spectacle. The speech is therefore in some sense an equivalent to its subject, a monument in words.22 Leo implies that one should consider his speech as a verbal artifact whose equivalence to its subject is not based solely on its representational function. The same idea is present, in a more developed form, in Nikolaos Mesarites’ ekphrasis. In the invocation to the apostles, which precedes his account of the interior of the Holy Apostles, the author makes explicit the metaphor of verbal monument. With a prayer for divine inspiration, he refers to his composition as a “building (oi«kon) which I have undertaken to construct (oijkodomh´ sai) with the material of words and the skill of my intellect.” But then he goes on to define his work as much more than a simple verbal copy of the church: the speech, and the very process of composition, will allow both the author and every “lover of the apostles” to perceive the beauty of the church more acutely (trano´ teron) and purely (kaqarw´ teron).23 Mesarites thus proposes a dual function for his speech. It is a verbal equivalent, a carefully wrought work of art in its own right, but at the same time it can complement the material building, which is both a referent and a parallel. Indeed, it is far more than a static literary monument—rather, its composition and recitation are processes that involve and affect both the speaker and the audience. Taken together, all these remarks suggest that the rhetor is able to construct in words an equivalent to the monument he is describing. The word has its own ways of working that are very different from those of the visual arts, and it may fall far short of its model in some respects, as Chorikios states, but it can also make its own contribution, as Mesarites suggests. Our authors are imprecise about the particular nature of their medium and its demands; but in many ways the tradition of ekphrasis within which they were working suggested ways of creating a verbal equivalent to a subject that fully exploited the characteristics of the verbal medium, such as its temporal dimension and its ability to evoke that which is absent as vividly as that which is present. The theory and practice of ekphrasis 21 Chorikios, Laudatio Marciani 1.9, 1.16: pa'sa de` mi´mhsi" h»tton pw" fe´ retai th'" ajkribei´a". These words are repeated verbatim by Ioannes Phokas in the introduction to his ekphrasis of the Holy Land; PG 133:928B. 22 Leo the Wise, Homily 34, ed. Akakios (as above, note 10), 274–75. 23 Mesarites, Ekphrasis 12.4: eij ga` r mh` ku´ rio" di∆ uJmw'n oijkodomh´ sei moi oi«kon tou'ton, o’n tai'" ejk lo´ gou u”lai" kai` toi'" ejk noo` " tecnourgh´ masin oijkodomh´ sai proujqe´ mhn, i”n∆ e“coimi di∆ aujtou' kajgw` kai` pa'" filapo´ stolo" pro` " to` tou' uJmete´ rou oi“kou ka´ llo" trano´ tero´ n te kai` kaqarw´ teron ejnora'n, eij" ma´ thn oiJ oijkodomou'nte" ajnqrw´ pinoi´ moi logismoi` kai` lo´ goi kekopia´ kasi.
64
THE AESTHETICS OF SACRED SPACE
had always allowed for the possibilities and limitations of the verbal medium in ways that were developed by Byzantine authors. ANCIENT AND BYZANTINE THEORIES OF DESCRIPTION The aim of ekphrasis in rhetoric had always been less to give a complete and accurate account of a particular object than to convey the effect that the perception of that object worked upon the viewer. The speaker aimed first and foremost to appeal to the listener’s imagination, often through the use of generalized images that were more likely to correspond to the prior experience of the audience, rather than to provide an accurate verbal transcription of a particular object. Vivid language seems to have been thought to evoke the effect of perception upon the listener, to make him or her feel “as if ” in the presence of the scene.24 Implicit in the classical conception of ekphrasis is, therefore, an emphasis on the use of language to represent not merely the sensible appearance of the subject, but its more general, intelligible characteristics. This is an aspect of ekphrasis that Byzantine authors exploited in their descriptions of church buildings, among other things. The strict modern distinction between description (understood as the representation of static objects) and narration (the representation of actions or events) was not observed. In ancient rhetoric, ekphrasis fell somewhere between these two categories, and was often a vivid and detailed narration of events. An ekphrasis is by definition an account of anything—from a battle to a person to a season—that has the quality of vividness (ejna´ rgeia) necessary to make an audience “see.” Unlike description in modern theories of narration, ekphrasis was not defined as an interruption to the temporal flow of a surrounding narrative, but usually incorporated some temporal progression itself. This is apparent from the instructions for ekphraseis of battles in the Progymnasmata, which recommend a progression from the events that preceded the battle to the battle itself and finally to its aftermath.25 There are indications that the animation of a scene was felt to contribute to its vividness.26 Ekphraseis of works of art show a tendency to ignore the static, spatial nature of the painting or mosaic, and to recount the events depicted as if they were unfolding in time, while imbuing the scene with human interest by expressing the describer’s emotional involvement.27 But while a painting can easily be turned into narrative—a verbal equivalent that tells the same story in a different medium—a building or a place posed different problems. There are no explicit comments in ancient treatises or Byzantine commentaries on how to deal with such subject matter. All the authors of progymnasmata mention “places” as a category of subject matter for ekphrasis, but as Ruth Macrides and Paul Magdalino See R. Webb, “Me´moire et imagination: Les limites de l’enargeia dans la the´orie rhe´torique grecque,” in Dire l’e´vidence, ed. C. Le´vy and L. Pernot (Paris, 1997), 229–48. 25 Hermogenes, Progymnasmata, in Hermogenis opera, ed. H. Rabe (Leipzig, 1913), 22–23; Aphthonios, Progymnasmata, ed. H. Rabe (Leipzig, 1926), 37. 26 Aristotle supposes a connection between animation of the subject matter and vividness of the language in his enigmatic remarks on metaphor, Rhetoric, 1411B.24–25; see G. Morpurgo-Tagliabue, Linguistica e stilistica de Aristotele (Rome, 1967), 256–66. Ioannes Sardianos conversely describes enargeia as the quality that lends animation to speech, making it as if alive (e“mpnou"), in his Commentarium in Aphthonii Progymnasmata, ed. H. Rabe (Leipzig, 1928), 217, 1.3. 27 L. James and R. Webb, “‘To Understand Ultimate Things and Enter Secret Places’: Ekphrasis and Art in Byzantium,” Art History 14 (1991): 1–17. 24
RUTH WEBB
65
note in their study of Paul the Silentiary, none gives any specific instructions.28 Despite Photios’s reference to the “canons” of ekphrasis, there are no rules comparable in detail to those for the full-scale epideictic speeches. However, the very definition of ekphrasis contains a valuable indication: most of the ancient rhetoricians agree in calling it a lo´ go" perihghmatiko´ ", literally a speech that “leads one around.” 29 The glosses on this term in Ioannes Sardianos’s commentary to Aphthonios show that the literal meaning was still active: it is as if, he explains, one were to show a new arrival around the city of Athens, adding that the metaphorical meaning is “to recount in order” (eJxh'").30 By this means the orator could turn an account of a static object into an account of a journey, representing space through the passage of time.31 The periegesis format is thus a convenient way of ordering details, and one that adapts the representation of space to the demands of the temporal flow of language. Indeed, the metaphor of language as journey is well established in the Greek language in terms such as dihge´ omai or peri´odo". A further advantage is the introduction of movement into the presentation of an otherwise static subject, which dramatizes and personalizes the description by suggesting a viewer whose responses may be noted and commented upon. The role of metaphorical guide adopted by speakers also contributes to the construction of a relationship between the speaker and the audience. This is most notable in Chorikios, who changes abruptly from the second person plural of his introduction to the more intimate singular form as he begins his periegesis of the church of St. Sergios, and in Mesarites, who frequently exhorts his audience to follow him or to direct their gaze toward certain elements of the decoration during his imaginary tour of the Holy Apostles. In adopting the periegesis as an organizing principle for their descriptions, authors such as Chorikios and Mesarites were following the standard procedure for ekphraseis of places. Such ekphraseis were a regular part of city encomia in the Roman Empire, the most detailed being Libanios’s Antiochikos.32 Aristeides similarly takes his audience on an imaginary tour of Smyrna, traversing the city from west to east, then moving up onto the acropolis and down again.33 But when he wishes to create an impression of the whole rather than the parts, Aristeides uses a different method. In order to convey the beauty of Smyrna he compares the city to shining pearls and a splendid many-colored cloak.34 Similarly, in his oration on Kyzikos, the size and splendor of the temple were evoked for the citizens through hyperbolic comparisons: the blocks of stone are said to be as large as temples, the temple as large as a sanctuary, the sanctuary as large as a city.35 The principle underlying such comparisons is very different from that of the periegesis, though Macrides and Magdalino, “Architecture of Ekphrasis,” 48–49. Hermogenes, Progymnasmata, 22.7, and Aphthonios, Progymnasmata, 36.21. 30 Sardianos, Commentarium, 216. 31 See James and Webb, “To Understand Ultimate Things,” 8; for a more detailed study of the periegesis in late classical literature and its connection with enargeia, see S. Dubel, “Ekphrasis et enargeia: La description antique comme parcours,” in Dire l’evidence, 249–64. Michael the Deacon uses the verb perihge´ omai to refer to his description in the passage quoted above, note 1. 32 On the development of ekphrasis within city encomia, see H. Saradi, “The Kallos of the Byzantine City: The Development of a Rhetorical Topos and Historical Reality,” Gesta 34 (1995): 37–56. 33 Aelius Aristeides, Oration 17, in Opera, ed. Keil (as above, note 3), 2:3–4. 34 Ibid., 2:4.2, 4.8–9. 35 Aelius Aristeides, Oration 27, in Opera, ed. Keil, 2:129–30. 28 29
66
THE AESTHETICS OF SACRED SPACE
the final aim is the same—vividness, or ejna´ rgeia. While the periegesis is a way of organizing references to the visible features of the subject, the recourse to simile introduces a new dimension by comparing the present sight (often visible to the original audience) to some other, absent entity. The dual function suggested by Mesarites in his speech was already implicit in the classical theory and practice of ekphrasis. Along with the periegesis, the constant reference to the unseen is adopted by Byzantine authors, but invested with a new significance. Chorikios’s speeches were still very much part of this late classical tradition, but later authors too were clearly aware of these earlier models, both Christian and pre-Christian. Mesarites, for example, signals his consciousness of the tradition of ekphrasis, and of his own place within it, through his use of Libanios’s Antiochikos in the early sections of his Ekphrasis.36 The generalized terms in which Libanios couched his description made it easy to adapt his phrases to the suburban setting of the Constantinopolitan church. But such models were of only limited use to authors of church ekphraseis, and it is notable that Mesarites confines his use of the Antiochikos to the introductory sections of his work in which the church is placed in its urban context. A building like the Holy Apostles involved an entirely new challenge connected with its function and architecture, requiring an extended rhetorical account of a sacred interior. For this there were no direct classical precedents: descriptions of temples like that in Aristeides’ oration on Kyzikos focused on the exterior of the building presented within its civic or ritual context.37 Moreover, the churches were visually rich in their architectural form and surface decoration as well as complex in terms of their symbolic function.38 As the Greek hymn on the church of Hagia Sophia in Edessa proclaims, the very idea of the church as a house of the uncircumscribable divinity is equal in its paradox to the incarnation itself.39 The earliest extant church ekphrasis, Eusebios’s Panegyric on the Church at Tyre, insists on this symbolic function, treating the church first as a building but then as a symbol of the living temple.40 For Eusebios, the very existence of Paulinos’s building at Tyre was tangible proof of the triumph of Christianity. Similarly, Hagia Sophia and the Holy Apostles also stood as signs of imperial power and patronage, themes that are particularly evident in Paul the Silentiary,41 Prokopios, and Constantine Rhodios. Thus, in addition to the problems of describing, the authors of church ekphraseis encountered the challenge of evoking sites replete with meaning. To do so they adopted and adapted traditional strategies like the periegesis. PERIEGESIS The periegesis format is the most frequently used method of organizing an account of the building. It may begin with some mention of the building’s location within the city, as in Mesarites and Chorikios; but, in contrast to the classical precedents, the exterior is For a full list of references, see Downey’s commentary to Mesarites, Ekphrasis, 862. On the evidence for speakers performing in front of temples, see L. Pernot, La rhe´torique de l’e´loge dans le monde gre´co-romain (Paris, 1994), 441. 38 Pseudo-Germanos, Historia ecclesiastica, PG 98:384A–453B, enumerates the symbolism of the church building, furniture, and vestments. 39 A. Palmer, “The Inauguration Anthem of Hagia Sophia in Edessa,” BMGS 12 (1988): 117–67. 40 See Macrides and Magdalino, “Architecture of Ekphrasis,” 52. 41 On Paul, see Macrides and Magdalino, “Architecture of Ekphrasis.” 36 37
RUTH WEBB
67
by no means the main focus of attention. Some authors, moreover, go far beyond the simple use of the periegesis as an organizing principle for their descriptions and develop the metaphor of the text itself, or the process of composition, as a journey. This tendency is evident, for example, in Photios’s reference to the “progress of my speech” (oJ tou' lo´ gou dro´ mo").42 The most elaborate conceit is found in Mesarites’ use of the term logos. Here the speech is not just a journey, but is personified and described as if it were itself the imaginary visitor to the church: the personified logos takes on the emotions appropriate to the various sights described, it runs away from the author, addresses the women at the tomb of its own accord, looks about curiously, and notices details by itself.43 In these passages Mesarites fully exploits the polyvalence of the term: word, speech, discourse, thought, and, of course, the divine Logos.44 The old metaphor is thus endowed with a new significance by Mesarites. Other authors manipulate the notion of the periegesis in different ways in order to express the nature of their subjects. Rather than following an unproblematic linear progression (as in Aristeides, for example), they frequently express difficulty, the threat of interruption, or the need to retrace their figurative steps. This may be used as a way of expressing the magnificence of the spectacle, as when Chorikios claims that the variegated splendor of the exterior of the church of St. Stephen may prevent the visitor from entering it.45 The topos is developed by Photios, who says that the beauty of the atrium of the church of the Virgin of the Pharos will cause the visitor to be rooted to the spot, even to be transformed into a tree in his amazement (qau'ma).46 The primary function of these hyperbolic statements is evidently to express the beauty of the building and its effect upon the listener; as such they are a variation on the standard expression of aporia by the speaker who claims that “words cannot describe the sight.” A similar effect is to be found in Constantine Rhodios’s ekphrasis. He begins to mention the figural decoration of the Holy Apostles only to remind himself that this is not the right moment (kairo´ "); he then breaks off to describe the construction of the building first.47 Here the passage echoes a long-established literary topos, creating the impression that the poem is an improvised performance that might develop in any of several different ways, rather than a carefully planned composition. At the same time one could argue that these artful impressions of disarray are also wholly appropriate to the subject matter of the ekphraseis. A centrally planned church, like the Holy Apostles, in reality imposes no simple itinerary upon the visitor. The way in which Constantine evokes the decoration, only to return to the architecture, is an acknowledgment both of the variety of spectacle offered by the church and of the fact that the author must impose his own order upon this material. Such hesitation is perhaps the closest an author can get to rendering the simultaneity of visual experience, which would otherwise exhaust and surpass the possibilities of speech. Photios, Homiliai 10.7. For this conceit in the earlier literature, see Dubel, “Ekphrasis et enargeia.” Mesarites, Ekphrasis 28. 44 On logos in Mesarites, see T. Baseu-Barabas, Zwischen Wort und Bild: Nikolaos Mesarites und seine Beschreibung des Mosaikschmucks der Apostelkirche in Konstantinopel (Vienna, 1992), 126–42. 45 Chorikios, Laudatio Marciani 2.21. 46 Photios, Homiliai 10.4. 47 Constantine Rhodios, Ekphrasis, ll. 534–40: all∆ ajmfi` tw'nde qauma´ twn kai` pragma´ twn/ta` nu'n siga´ sqw ta´ xew" ca´ rin lo´ go" / kairou' d∆ejpotru´ nonto" e“mpalin fra´ sw, / Qeou' qe´ lonto" kai` lo´ gon dwroume´ nou. 42 43
68
THE AESTHETICS OF SACRED SPACE
The imagined obstacles to entering the buildings evoked by Chorikios and Photios also serve to focus attention on the moment of transition between the exterior and the interior. The crossing of the threshold into the church interior is a dramatic moment in the journey. Paul the Silentiary, for example, begins his account of Hagia Sophia by imagining that he and his audience (who seem to have been in the Patriarchal Palace) are knocking at the doors of the church and requesting entrance.48 Mesarites uses a change in the register of his language to underline the point, ceasing to use classical quotations and citing the Gospels instead. Chorikios dramatizes the moment for his audience, claiming that the variety of things to see within the church will make them dizzy as they look around them.49 Once again, Chorikios’s idea is developed by Photios in his tenth homily: But when, having torn oneself away from the atrium, one looks into the church itself, with how much joy and trembling and astonishment is one filled! It is as if one were stepping into heaven itself with no one standing in the way at any point; one is illuminated and struck by the various beauties that shine forth like stars all around. Then everything else seems to be in ecstasy and the church itself seems to whirl around; for the viewer, with his twistings and turnings in every direction and his constant movements that the variety of the spectacle imposes on him, imagines that his personal experience is transferred to the church.50
In contrast to the exterior, the interior poses no obstacle to the viewer, offering instead almost (wJ") a vision of heaven itself. What is more, the intensity of aesthetic experience inside the church is such that normal modes of perception and normal distinctions between subject and object are said to be disrupted. This is suggested by Photios’s explicit statements about the effect of the church interior on the viewer’s senses, but it may also be implicit in the type of language used to evoke the appearance of the buildings. METAPHOR AND TRANSCENDENCE A monument like a church might seem to be the very model of the static and the solid; yet architectural features, particularly the curves of arches and vaults, are frequently described in ekphraseis as if their forms were traced not in static stone, but by a point in motion. In Paul the Silentiary’s account of Hagia Sophia the four arches between the four central piers are each said to “rise up little by little on well-curved airy paths” and thus to “separate from their former partner.” 51 Elsewhere, columns are said to dance, arches to grow up, and conches to spring forth.52 Some of these terms may well be “dead metaphors,” but the frequency and variety of their use invite us to look further. 48 Paul the Silentiary, Ekphrasis, ll. 350–52: oi“xate´ moi klhi'da qeoude´ e", oi“xate mu´ stai, / oi“xate d∆ hJmete´ roisin ajna´ ktora qe´ skela mu´ qoi". 49 Chorikios, Laudatio Marciani 1.23. 50 Photios, Homiliai 10.5: ejpeida` n de´ ti" ejkei'qen mo´ li" ajpospasqei`" eij" aujto` paraku´ yh to` te´ meno", hJli´kh" kai` o”sh" ou»to" cara'" te a”ma kai` tarach'" kai` qa´ mbou" ejmpi´platai. wJ" eij" aujto` n ga` r to` n oujrano` n mhdeno` " ejpiprosqou'nto" mhdamo´ qen ejmbebhkw` " kai` toi'" polumo´ rfoi" kai` pantaco´ qen uJpofainome´ noi" ka´ llesin wJ" a“stroi" perilampo´ meno" o”lo" ejkpeplhgme´ no" gi´netai. dokei' de` loipo` n ejnteu'qen ta´ te a“lla ejn ejksta´ sei ei«nai kai` ajuto` peridinei'sqai to` te´ meno"⭈ tai'" ga` r oijkei´ai" kai` pantodapai'" peristrofai'" kai` sunece´ si kinh´ sesin, a’ pa´ ntw" paqei'n to` n qeath` n hJ pantaco´ qen poikili´a bia´ zetai tou' qea´ mato", eij" aujto` to` oJrw´ menon to` oijkei'on fanta´ zetai pa´ qo". 51 Paul the Silentiary, Ekphrasis, ll. 463–65: ojrnume´ nh de` /hjeri´ai" kata` baio` n ejugna´ mptoisi keleu´ qoi" /th'" pri`n oJmognh´ toio dii´statai. 52 See, for example, Paul the Silentiary, Ekphrasis, ll. 400, 405; Michael the Deacon, Ekphrasis 4.105.
RUTH WEBB
69
The rhetorical motivation is clear: the attribution of movement and animation to a static entity was one means of making the subject vivid for the listener. This mode of describing also reflects the experience of the building. The movement attributed to the architectural features may also be a means of suggesting the actual experience of the visitor who is moving around the church and for whom the architecture is an intricate and ever-changing pattern of forms.53 At the same time this rhetorical technique was also an effective way of expressing the magnificence and otherness of the church. Michael the Deacon argues that the perception of static elements as being in motion is part of the viewer’s experience of Hagia Sophia and that this is the impression created by the physical nature of the building itself. The brightness of the gold in the narthex, he says, makes it almost appear to drip down (katasta´ zein).54 Like Photios, Michael offers a physiological explanation: the light reflected back by the gold affects the moisture of the eyes and creates this effect. The building is such that it transcends its material nature and cannot be perceived as an object. These examples show how rhetorical techniques may in fact contribute to the depiction of the church interior as a sight that transcends normal human experience, and may serve to reveal rather than conceal. Mesarites is the most explicit of the authors on this point. At the threshold of the Holy Apostles, the point of transition, he explains that the eyes of the senses (oiJ aijsqhtoi` ojfqalmoi´) must be supplemented by the eyes of the mind (oiJ noeroi` ojfqalmoi´) in a progression, not just from the exterior to the interior of the church, but from the evidence of the senses to the final mysteries and secret places.55 There is thus a further sense in which the text represents a journey. The listeners are not only taken on a periegesis around the physical building, but also receive the spiritual understanding necessary to lift their perception from the material to the spiritual. Seen in this light, the remarks of Photios and Chorikios on the danger of being distracted by the appearance of the exterior and thus failing to reach the glories of the interior take on a new significance. In Eusebios, the mere appearance of the church at Tyre, in contrast to its symbolic significance, is termed “the external impression” (hJ e“xwqen fantasi´a).56 The external appearance inspires wonder (qau'ma) in the viewer, but the archetypes and their prototypes to which it points—the spiritual renewal of which the building is just a sign—are more wondrous still. For these authors, then, to restrict one’s attention to the objective, material, aspects of the building is to fail to enter it. One way of leading the soul upward, of revealing the mysteries of the building, seems to have been by making explicit the significance of the architectural form. None of the ekphraseis provides anything like the systematic discussion of the elements of church architecture and of their significance that is to be found in the Historia ecclesiastica attributed 53 I am grateful to Robert Ousterhout for this suggestion. See also the remarks of O. Wulff, “Das Raumerlebnis des Naos im Spiegel der Ekphrasis,” BZ 30 (1929–30): 531. 54 Michael the Deacon, Ekphrasis 3. 55 Mesarites, Ekphrasis 12.1: h“dh de` kairo` " hJma'" procwrh'sai tv' lo´ gv kajpi` ta` e“ndoqen tou' naou' kai` katopteu'sai me` n tau'ta toi'" aijsqhtoi'" ojfqalmoi'", katanoh'sai de` kai` toi'" noeroi'" . oi«de ga` r kai` nou'" proko´ ptein ejk tw'n kat∆ ai“sqhsin kajk tou' ejla´ ttono" podhgou´ meno" katalamba´ nein ta` telew´ tera kai` pro` " ta` a“duta pareisdu´ nein. . . . 56 Eusebios, Historia ecclesiastica 10.4.55: qau'ma me` n ou«n me´ giston tou'to kai` pe´ ra pa´ sh" ejkplh´ xew", ma´ lista toi'" ejpi` mo´ nh th' tw'n e“xwqen fantasi´a to` n nou'n prosane´ cousin⭈ qauma´ twn de` qaumasiw´ tera ta´ te ajrce´ tupa kai` tou´ twn ta` prwto´ tupa nohta` kai` qeopreph' paradei´gmata, ta` th'" ejnqe´ ou fhmi` kai` logikh'" ejn yucai'" oijkodomh'" ajnanew´ mata.
70
THE AESTHETICS OF SACRED SPACE
to Germanos. Rather, the statements about the symbolism of the building tend to be general in nature, like Photios’s comparison of the interior of the church of the Virgin of the Pharos to heaven itself. Constantine Rhodios notes the significance of the crossshaped plan of the Holy Apostles, points out the numerical significance of the number of columns, and, like Mesarites, compares the dome, with its depiction of the Pantokrator, to heaven in a common simile.57 The most striking presentation of the architectural form is to be found in the text of Michael the Deacon. He offers the most unusual reading of the architecture of Hagia Sophia, using the image of pregnancy to express the form, size, and significance of the building as a whole: “The building opens up to immensity (wJ" eij" to` ajcane´ "); the breadth of its hollows is such that it could be pregnant with many thousands of bodies.” 58 The image creates a powerful visual impression of the volume and the rounded forms of the building, at the same time pointing to its theological significance. The idea of pregnancy recalls the comparison of the Virgin to the living church (nao` " e“myuco") as well as the parallel between the church and Theotokos, both receptacles for the boundless divinity.59 Such is the size, in fact, that the very notion of the interior is now surpassed and abolished in Michael’s ekphrasis; the very forms that close around the viewer are said to open up to immensity. The authors thus use the medium of language to refer to those aspects of the buildings that are not immediately visible but are implicit in the structures. Another type of meaning brought out by these texts can be seen in the narratives of past events, which in various ways seem “built into” the church buildings and which the authors of ekphraseis retell in words. NARRATIVES One type of past event made present through the description is the building’s construction, with the narrative telling how each element of the church was put in place.60 Naturally, the greatest emphasis on construction is to be found in the texts celebrating Justinian’s rebuilding of Hagia Sophia. The second half of Prokopios’s account of the church describes its architecture in terms of the rebuilding and Justinian’s divinely inspired role in this.61 Paul the Silentiary’s ekphrasis of Hagia Sophia has the most elaborate narrative framework of any of the texts in the iambic side panels that tell in epic fashion how Roma spurred Justinian to action.62 Paul’s description of the church is mainly in the periegesis form, but he is also careful to describe its precious marbles in terms of their origin, reminding his audience of the laborious process by which they were mined and Constantine Rhodios, Ekphrasis, ll. 462–71, 718–22, 501, respectively; Mesarites, Ekphrasis 13.5. On num¨ B 37 (1987): 56–57. ber symbolism, see P. Magdalino, “Observations on the Nea Ekklesia of Basil I,” JO 58 Michael the Deacon, Ekphrasis 4.88–90: wJ" eij" to` ajcane` " oJ oi«ko" ajne´ vge, ku´ tou" me` n eujru´ thta e“cwn wJ" polla` " a‘n swma´ twn muria´ da" ejgkumonei'n. 59 J. Meyendorff, “Wisdom-Sophia: Contrasting Approaches to a Complex Theme,” DOP 41 (1987): 391– 401, and Palmer, “Inauguration Anthem,” 148. 60 The technique goes back to Homer’s account of Hephaistos making the Shield of Achilles in Iliad 18, a description praised by G. E. Lessing in his Laokoon (1766) precisely because it respected the nature of poetry by presenting the material in narrative form. 61 Prokopios, Buildings 1.1.70–71. 62 M. Whitby, “Paul the Silentiary and Claudian,” CQ 35 (1985): 507–16. 57
RUTH WEBB
71
brought to Constantinople.63 Implicit in the columns is a narrative of their origin and, above all, of the action of Justinian who is credited with gathering their splendor together; the very presence of the columns in the church is treated as proof of the extent of his domain. Constantine Rhodios, whose work is dedicated to an emperor, similarly insists on the various origins of the stones in the Holy Apostles, treating their presence as proof of their past journey to Constantinople.64 In these brief accounts of stones, the authors evoke in words the narratives implicit in the objects themselves. In the case of the ekphraseis of the Holy Apostles, the description of figural decoration introduces a further layer of narrative complexity. When Constantine Rhodios finally allows himself to treat the figural decoration, he does so in a plain list form. On the other hand, Mesarites’ presentation of the images in the Holy Apostles is far more sophisticated in its interweaving of time and space, the events depicted, and their architectural framework. Mesarites speaks as if he were moving around the church from scene to scene, but the order he follows is dictated by the relationship between the scenes (not necessarily a chronological one) and not, as far as it is possible to tell from the lacunary text, by their architectural setting.65 The image of the Transfiguration, for example, is described just before that of the Crucifixion. After dwelling upon the significance of the events on Mount Tabor, Mesarites asks his audience to follow him from one scene to the next with these words: “But Him, whose glory the disciples saw just now as He was transfigured on Tabor and whose end, which He was going to fulfill in Jerusalem, the chiefs of the prophets spoke of, going on a little further in our speech let us see Him hanging on the cross in the eastern hall, fulfilling willingly in Golgotha the end, which was shortly before spoken of on Tabor by the prophets . . .”66 The relative pronouns make awkward reading, but they serve to illustrate the intimate connections between the two events and their depictions. As Mesarites moves between the two images in his periegesis he also moves in time from one event to the next; as he does so, the narrative of the life of Christ is replayed within the framework of a fictional movement through space. The images bring the events of the past dramatically into the present for both the speaker and the audience. But Mesarites’ verbal presentation also underlines the relationship between the two scenes, the relationship that transcends any normal temporal order. The Crucifixion is both cause and effect of the events on Mount Tabor. Using his material—the word—Mesarites is able not only to evoke the scenes depicted in the church but also to make explicit the theological connections between them. The images therefore introduce a further narrative complexity into these texts. On a purely literary level, one might suggest that the multiple layers of time evoked by these texts—the fictional time of the periegesis and the past times of the events evoked—could 63 Paul the Silentiary, Ekphrasis, ll. 376–80, 617–46. Paul naturally omits to mention that some of the columns were taken from earlier buildings. 64 L. James, Light and Colour in Byzantine Art (Oxford, 1996), 115. 65 See Epstein, “Rebuilding and Redecoration,” 84–85. 66 Mesarites, Ekphrasis 17.1: ajlla` ga` r ou» th` n do´ xan ei«don oiJ maqhtai` ejn tv' paro´ nti metamorfoume´ nou Qabw` r kai` ou» th` n e“xodon sunela´ loun, h’n e“melle plhrou'n ejn JIerousalh´ m, oiJ tw'n profhtw'n prou“conte", tou'ton tv' lo´ gv metaba´ nte" mikro` n kati´dwmen krema´ menon ejn staurv' ejn th' peri` th` n e”w stoa', plhrou'nta me` n eJkousi´w" ejn Golgoqa' th` n ejn Qabw` r pro` mikrou' para` tw'n profhtw'n sullaloume´ nhn e“xodon. . . . Translation adapted from Downey’s edition (p. 873).
72
THE AESTHETICS OF SACRED SPACE
be seen as a verbal, temporal equivalent of the spatial three-dimensionality of the building itself. In this way, literary structure echoes, on its own terms, the architectural structure.67 But the presentation of images in the ekphrasis also serves to bring out the latent significance of the building, illustrating the role of images in making the past present as well as the narrative resonance between the images themselves. A narrative framework in the form of the periegesis clearly provided a valuable organizing principle for describing the interior of a building, allowing space to be represented in terms of an ordered progression unfolding in time. The subsidiary narratives, whether of the events depicted within the church or of the process by which the building came to be constructed in the past, performed a somewhat different function in relation to the building, unlocking layers of significance implicit in the structures and illustrating how these buildings might conquer time itself.68 The interiors of the buildings are treated as microcosms in which ordinary time, as well as ordinary modes of perception, is surpassed and the past is made eternally present through images, through the structure itself, or through tangible signs of empire. That this was indeed how a church interior might have been conceived has been argued by Robert Ousterhout, on the basis of material remains. He concludes that the spatial arrangement of images in a church interior “serves to create a ritual space in which past, present, and future converge.” 69 The causal and theological resonances that Mesarites brings out in his account of the Transfiguration and Crucifixion scenes in the Holy Apostles must surely reflect the ways in which those images were experienced. CONCLUSIONS The interwoven narratives in this final example illustrate the extent to which the characteristics of the word, notably its temporal progression and its ability to express the implicit, are exploited by the authors in describing their subjects. In this article I have brought out some of the literary aspects of these texts—their structure, their manipulation of the traditional periegesis, or the introduction of different levels of narrative into the framing narrative of the tour around the church, as in Mesarites’ particularly complex presentation of the Holy Apostles. I hope to have suggested that these ekphraseis are worthy of analysis as texts in their own right. However, in the past it has always been the relation of the ekphraseis to their referents that has provoked the most interest, and this is a question that cannot be ignored. In the introduction I discussed some of the problems involved in the enterprise of description. A different type of relationship between text and referent, suggested by the authors themselves, is that of equivalence: an artfully constructed piece of poetry or prose constitutes a fitting response to the artistry of the building. This function of equivalence is in itself an encouragement to focus on the intrinsic qualities of the texts. One aspect of the texts to which the authors draw attention, and which must be important for our assessment of their relation to the monuments they describe, is the process of composition. A further function of the frequent expressions of aporia is to call Macrides and Magdalino, “Architecture of Ekphrasis,” 58–59, suggest that the structure of Paul’s Ekphrasis might reflect the process of construction. 68 Michael the Deacon, Ekphrasis 1. 69 R. Ousterhout, “Temporal Structuring in the Chora Parekklesion,” Gesta 34 (1995): 76. 67
RUTH WEBB
73
attention to the author’s art, his role in creating the order of his composition, as well as to the divine aid that makes this possible. Mesarites’ ekphrasis contains the clearest and most developed references to the role of the author. Breaking off after his account of the depiction of the Baptism in the Holy Apostles, he goes on to develop an extended metaphor that creates an analogy between the subject of the image and the act of composition: his pen is an oar, and the ink is the river Jordan into which the author has fallen.70 This striking metaphor effectively expresses the speaker’s involvement in the scene that he is placing before the eyes of his audience.71 At the same time, paradoxically, this unexpected reference to the tools of the author’s trade serves to disrupt the illusion—carefully built up by Mesarites’ words—that the speaker is composing his speech within the church as his audience listens in. Here he reveals that at the moment of composition the author is faced only with his pen and ink and is presumably evoking the scenes in memory. Although Mesarites is the only author making explicit reference to the material conditions of his task, this is an aspect that needs to be taken into consideration for all the texts when assessing their relation to their referents. These are not transcriptions of tours, or, as far as we can tell, spontaneous utterances. The authors clearly wrote from autopsy and deep familiarity with the monuments; but at the moment of composition they were, presumably, describing an intangible, mental image impressed in their memories. Mesarites, again, makes this most clear in his invocation to St. Bartholomew, in which he prays for his mind (nou'") to remain attached, as if crucified, to the church. The image of the church that he seeks to convey through his written discourse derives therefore primarily from the intellect and only secondarily from the material fabric of the church. The hints that Mesarites gives us about the composition process of his ekphrasis suggest both the limitations of these texts and what they can reveal about their subjects. The ekphraseis can tell us about the observable aesthetic qualities that were appreciated by the Byzantines. The attention paid to the effects of light and variegated color, for example, has been shown to reflect wider aesthetic theories.72 But the ekphraseis have more to tell; if language in the end falls short of a sight as marvelous as a church, it can still be used to state and comment on the qualities of the buildings that are not visible to the eye, as Mesarites again points out.73 The most striking feature of these ekphraseis is the exploitation of the capacity of the word to evoke the absent, to express the intelligible meanings implicit in the material sights. If we treat the ekphraseis as texts, the distinction between the passages that refer to the tangible aspects of the building and those that refer to the intangible disappears. The fact that this lack of distinction was an important feature of the Byzantine conception of the “aesthetic” is suggested by the extension of the term aijsqhtiko´ " in later Greek to include the intelligible in addition to the sensible.74 By constantly juxtaposing the intelligible and the sensible these ekphraseis unlock the inherent significance of the buildings they describe. 70 Mesarites, Ekphrasis 25.1. Compare with the invocation to St. Bartholomew (ibid., 12.18) with its reference to “pen and ink.” 71 Epstein, “Rebuilding and Redecoration,” 84, notes Mesarites’ characteristic involvement in his subject. 72 James, Light and Colour. 73 Mesarites, Ekphrasis 12.1, 12.18. 74 G. Lampe, Patristic Greek Lexicon (Oxford, 1961), s.v. aijsqhtiko´ " D.
74
THE AESTHETICS OF SACRED SPACE
The rendering of space through time and the appeal to the invisible in order to express the visible are two techniques that Byzantine authors inherited from antiquity and developed far beyond their antique models. They are techniques that exploit the characteristics of language and are well suited to their subjects, allowing the authors to unlock the significance that was hidden, more or less deeply, within the buildings. In such an enterprise the distinction between the perceptible and the imperceptible qualities of the buildings becomes unimportant. By juxtaposing the two aspects, the authors of the ekphraseis seek to convey the experience of sacred space in which the seen and the unseen, the tangible and the intelligible are equally real. Princeton University
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Epigrams in Context: Metrical Inscriptions on Art and Architecture of the Palaiologan Era ALICE-MARY TALBOT
INTRODUCTION he numerous epigrams on Byzantine art and architecture preserved only in anthologies offer tantalizing but hazy and elusive glimpses of buildings and works of art that have long since disappeared. Since the poems have become separated from their artistic context and the poet did not so much intend to describe as to evoke his subject,1 it can be difficult to visualize the artifact to which he alludes. Any investigation of such poetry of the Palaiologan period must necessarily focus on the most prolific poet of the era, Manuel Philes, who flourished in the first half of the fourteenth century; approximately thirty thousand of his verses are preserved. A rapid skimming of the 1,078 poems of Philes published by Emmanuel Miller and Emidio Martini2 indicates that almost half of them are devoted to works of art and architecture. This staggering figure of over five hundred dedicatory and ecphrastic epigrams testifies to the profusion of artistic activity in Constantinople during the early decades of the fourteenth century, while the verses themselves suggest a close collaboration among patron, artist, and poet.3 Careful analysis of verses that have lost their artistic context, combined with their comparison to pertinent works of art (a hypothetical exercise in which I engage briefly at the conclusion of the article), sometimes enables the reader to conjure up a visual image of the lost work of art alluded to in the poem, but never with any degree of preci-
T
I should like to thank my art history colleagues who assisted me in locating illustrative materials for this article, especially Martin Dennert, Sharon Gerstel, and Natalia Teteriatnikov. Thanks are also due to Henry Maguire and the two anonymous reviewers who read and commented on an earlier version of the article. Unless otherwise specified, the translations of the epigrams are my own. 1 On a related topic, see the article of L. James and R. Webb, “‘To Understand Ultimate Things and Enter Secret Places’: Ekphrasis and Art in Byzantium,” Art History 14 (1991): 1–17, which argues that ekphraseis were “attempts to convey the spiritual truth residing in art” (p. 14). For a nuanced response to this article, focusing on epigrams, see H. Maguire, Image and Imagination: The Byzantine Epigram as Evidence for Viewer Response (Toronto, 1996). 2 E. Miller, Manuelis Philae Carmina, 2 vols. (Paris, 1855–57), and E. Martini, Manuelis Philae Carmina Inedita (Naples, 1900); in the footnotes below each epigram of Philes is identified by its numbering in the manuscript in which it is found (e.g., Escurial, no. 281), as well as by the page number of the edited version. A critical edition of the entire corpus is currently in preparation by Gunther Stickler and Hans Veit Beyer. 3 For analysis of a group of Philes’s poems on ex-voto offerings to the Pege shrine in Constantinople, see A.-M. Talbot, “Epigrams of Manuel Philes on the Theotokos tes Peges and Its Art,” DOP 48 (1994): 135–65.
76
EPIGRAMS IN CONTEXT
sion. On occasion the title or lemma, which generally precedes Philes’s poems, may provide the reader with additional information. Here are a few examples: “On the great Demetrios represented on a stone with red veins,” “On the St. George carved from white stone at the Mangana monastery,” “On an icon of the Lord with a frame of pearls on a black background.” 4 These titles can be helpful as aids to comprehension of the verses, especially when they specify the identity of the sacred figure represented, the medium of the work of art, its location, or its patron. Readers of those poems of Philes that are no longer attached to works of art are faced with a number of puzzling questions, such as the problem of whether the epigrams found in the anthologies were “inscriptional” epigrams, once painted, carved, or incised on buildings and objects—or whether they were “literary” epigrams, intended as separate poems to accompany, rather than adorn, a work of art.5 The titles of the epigrams are ambiguous on this point: they generally begin with the preposition eij", as in the phrase eij" th` n eijko´ na tou' aJgi´ou Dhmhtri´ou, which can be variously interpreted as meaning “addressed to the icon of St. Demetrios,” or “on [in the sense of ‘about’] the icon of St. Demetrios,” or “located on the icon of St. Demetrios.” Apparently the preposition could have all three meanings, depending on the context; but in at least a few cases we can be certain that the title is describing the location of the epigram. Thus, Philes’s verses on the parekklesion of the Constantinopolitan church of the Virgin Pammakaristos are clearly labeled in the manuscript: “[Verses written] on behalf of the protostratorissa, on the cornice (eij" to` n kosmh´ thn) of the church which she built upon the death of her husband.” 6 The internal evidence of the epigrams strongly supports the supposition that many of them were intended for placement on the actual object described or on a surrounding frame. Some epigrams address the viewer directly as if the object is speaking; in others the poet uses such telltale words or phrases as “here,” “this man,” or “these verses,” implying that the epigram was inscribed on the object. Another puzzle is the varying length of the poems in Philes’s anthology. It is not difficult to imagine a couplet or a quatrain adorning a work of art even of modest size, but what are we to make of longer poems of twelve or even twenty-four lines? Could they be accommodated by an icon frame or a marble tombstone? Yet another problem, arising when one reads verses divorced from the work of art on which they were once inscribed, is that it is difficult to appreciate fully the relationship between words and image and to understand how the poem may have enhanced or reflected the viewer’s aesthetic or spiritual response. I. THE EVIDENCE OF SURVIVING INSCRIPTIONAL EPIGRAMS To address some of these issues, let us turn to a select group of surviving inscriptional epigrams by Philes and other, anonymous poets, whose verses are still preserved in situ, on either objects or buildings. It is not my intention to present a comprehensive survey 4 Escurial, no. 281, ed. Miller, Carmina, 1:136–37; and Florence, nos. 34, 230, ed. Miller, Carmina, 1:210, 433. 5 The terminology of inscriptional and literary epigrams is that of Marc Lauxtermann (“The Byzantine Epigram in the Ninth and Tenth Centuries” [Ph.D. diss., Universiteit van Amsterdam, 1994], esp. 21, 26, ¨randner’s thesis that many (or most) ecphrastic and 30–32, 55–70), whose research supports Wolfram Ho dedicatory epigrams were actually inscribed on works of art. 6 Escurial, no. 223, ed. Miller, Carmina, 1:117–18.
ALICE-MARY TALBOT
77
of such epigrams,7 but rather to examine a few poems that adorn churches and works of art of various functions and in different media, in an attempt to characterize the relationship among object, artist, poet, patron, and viewer. One of the aims of my inquiry is to demonstrate that poems of substantial length could be inscribed on structures and objects of all sizes and types, from funerary monuments to tiny reliquaries. A. Epigrams on Funerary Monuments 1. Parekklesion of the Church of Pammakaristos. The parekklesion of the church of the Virgin Pammakaristos in Constantinople (Fig. 1) vividly illustrates the incorporation of lengthy epigrams into an architectural context. The monastery church, originally built in the twelfth century, was restored in the thirteenth century by the general and military governor Michael Tarchaneiotes Glabas and his wife, Maria. Upon Glabas’s death, ca. 1305, his widow, who took the veil, added a parekklesion or funerary chapel to the south side of the church as a mausoleum for her husband. She evidently commissioned Manuel Philes to produce for the chapel a series of epigrams, some of which are still preserved in situ.8 Most impressive is the twenty-three–line epitaph carved on the exterior string course of the parekklesion. The first eight verses, on the west facade where the main door was situated, have now been obscured by the later construction of an ambulatory, but thirteen of the remaining fifteen lines on the south facade can still be read (Fig. 2).9 This epigram can safely be attributed to Philes since it is also preserved in the Escurial anthology of his poems.10 It is one of the very few epigrams by Philes that can still be seen on the object (in this case a building) for which it was destined, and it illustrates how an actual carved inscription, 21 m in length, makes a much greater aesthetic impact on the viewer than a text in a manuscript or a printed page of a modern anthology. The carefully carved letters, prominent punctuation marks, and decorative ligatures are as much an ornamental feature of the church facade (Fig. 3) as the brick patterns and blind arcades. Curiously, the epitaph, addressed to Michael by his widow, never mentions his name but alludes to him only by his title of protostrator. The poem says virtually nothing about the chapel itself except that it was built of stone; rather, it eulogizes the brave general who took the monastic habit before his death, and informs visitors that the parekklesion was built by Glabas’s widow, now the nun Martha, to house his remains. The visitor who wished to read the entire epigram would have to walk along two sides of the chapel, no doubt craning his neck to make out the letters. As suggested by Amy Cassens Papalexandrou in another context,11 the viewer would probably have spoken the verses out loud as ¨randner. Such a corpus is currently in preparation by Wolfram Ho H. Belting, C. Mango, and D. Mouriki, The Mosaics and Frescoes of St. Mary Pammakaristos (Fethiye Camii) at Istanbul (Washington, D.C., 1978), 3–22. 9 Ibid., 16, 20, 33; C. Mango and E. J. W. Hawkins, “Report on Field Work in Istanbul and Cyprus, 1962– 1963,” DOP 18 (1964): 327, 330–31. 10 Escurial, no. 223, ed. Miller, Carmina, 1:117–18. The inscription has never been properly published. For a line drawing of the surviving verses of the inscription and an English translation of the entire poem, see A. Van Millingen, Byzantine Churches in Constantinople (London, 1912), 157–60. 11 In a paper delivered at Ohio State University in October 1994, entitled “Medieval Wordpower and the Inscriptions of Skripou.” I am grateful to her for letting me read a typescript of this paper. She also pursues this line of argument in a recently completed Ph.D. dissertation at Princeton University, “The Church of the Virgin of Skripou: Architecture, Sculpture and Inscriptions in Ninth-Century Byzantium.” 7 8
78
EPIGRAMS IN CONTEXT
he slowly deciphered the ligatures and abbreviations; thus he would have performed a kind of ritual recitation in memory of the deceased protostrator. As the visitor proceeded inside the parekklesion, he would pass through the narthex into the chapel proper. Here he would no doubt first focus on the apse mosaic of Christ Hyperagathos, the Supremely Good, which lay straight ahead (Fig. 4). In the conch a three-line epigram frames the image of the seated Christ (Fig. 5); it reads: On behalf of her husband Michael Glabas, Who was a champion and worthy protostrator, Martha the nun [has offered] this pledge of salvation to God.12
With these verses the widow again reminds the visitor of her patronage of the chapel, names her late husband, and states that she constructed the parekklesion to help ensure his salvation. Since the image of Christ forms part of a Deesis scene, one can interpret the iconography of the apse as depicting the Virgin and John the Baptist interceding with Christ for the salvation of Glabas in response to Maria-Martha’s generous dedication to Christ of a splendidly decorated chapel. A third metrical inscription, painted in gold letters on a blue background, ornaments the lower and upper cornices of the church interior (Figs. 6, 7). It is so badly damaged that photos cannot do it justice, but it must have been a colorful decorative element when the parekklesion was new. According to the reconstruction by Arthur H. S. Megaw, the poem was twenty-seven lines long.13 Beginning on the south side of the lower cornice, it continued along the west and north walls of the parekklesion, with five verses in each section. One should note that the painted lower cornice on the south and west sides of the parekklesion, separating two zones of the wall decoration, corresponds exactly to the location of the carved inscription on the exterior cornice. The poem then moved to the cruciform upper cornice, with one verse on each of the twelve cornice sections. If my calculations are correct, the total running length of the inscription would have been 42 m! The fragmentary remains indicate that it was yet another epitaph on the death of Glabas, praising the departed general and praying for Christ’s blessing upon him. As on the exterior, the spectator who wished to read the epitaph would have had to exert some physical effort, either walking twice around the chapel, gazing upward at the two cornices, or perhaps, since the chapel is so small, merely turning around slowly in place. Sounding out the verses, the viewer would have thus uttered praises of Glabas, whose tomb was probably located in an arcosolium in the north wall,14 and would have addressed prayers for the protostrator to Christ whose mosaic image dominated the apse. Although the two inscriptions from the interior are not preserved in any anthologies of Philes’s poems, another poem, originally intended for the Pammakaristos parekklesion
12 On this mosaic and the inscription, see Belting, Mango, and Mouriki, Pammakaristos, 21, 45, 54–58. The translation is by C. Mango. 13 A. H. S. Megaw, “Notes on Recent Work of the Byzantine Institute in Istanbul,” DOP 17 (1963): 367–71. He includes a drawing of all surviving parts of the inscription (figs. M, N). See also Belting, Mango, and Mouriki, Pammakaristos, 16. 14 Ibid., 45.
ALICE-MARY TALBOT
79
but never used, does appear in the Escurial collection of Philes’s epigrams;15 it is therefore highly probable that the surviving mosaic and painted epigrams were by his hand. Also to be found in his anthologies are other metrical inscriptions commissioned by the Glabas family for the Pammakaristos complex, for example, its hospital and entrance gate, as well as a series of narrative paintings presenting the highlights of Glabas’s military career.16 To sum up, the three surviving inscriptions at Pammakaristos, executed in three media—carved stone, mosaic tesserae, and blue and gold paint—performed both a decorative and an organizational function. They added sculptural detail to the exterior cornice, and colorful ornament to the interior ones. Moreover, these inscriptions might frame a mosaic or accentuate the division into zones of exterior and interior wall surfaces. At the same time they served a commemorative and intercessory purpose: not only did they remind the viewer of the glorious achievements of Michael Glabas and the devotion of his widow who had dedicated his mausoleum to Christ in hope of her husband’s future salvation, but they also encouraged the visitor’s prayers for the soul of the deceased. 2. Tomb of Michael Tornikes at the Chora Monastery (Kariye Cami). At Pammakaristos the entire parekklesion served as a funerary monument for Glabas (and probably eventually for his widow). Other funerary chapels housed the tombs of several individuals, for whom separate arcosolia might be reserved. I limit myself to two examples from Constantinople, examining first the tomb of Michael Tornikes at the Chora Monastery (Kariye Cami).17 The Grand Constable Tornikes was a counselor of Andronikos II and served as ambassador to the Serbian court in 1327.18 His tomb, located in the south wall of the parekklesion, has lost a substantial part of its decoration. The main lunette panel, originally in mosaic, depicted Tornikes and his wife, in secular garb, flanking the Virgin and child (Fig. 8). The mosaic portraits of the couple were replaced with fresco images at some later date in the Palaiologan era. The soffit arch preserves, at least in part, the original mosaic portraits of the couple after they took the monastic habit; they are identified as the monk Makarios and the nun Eugenia. The elaborate sculpted decoration of the upper portion of the tomb is in much better shape; only the faces of Christ and the two angels have suffered damage. The accompanying epigram, incised in stone, is in virtually pristine condition. The poem is of twentyfour verses, arranged two to a line, so that verses 1–12 are on the left side of Christ, and verses 13–24 on the right. The end of each verse is neatly indicated by three dots, the letters are carefully carved, and great attention has been given to the spacing of the letters, especially on the left where the last four lines conform to the curve of Christ’s
15 Escurial, no. 219, ed. Miller, Carmina, 1:115–16. This is an example of “trial pieces,” multiple versions of epigrams composed by the poet from which the patron would select one; cf. Maguire, Byzantine Epigram, 8. 16 Florence, no. 98, ed. Miller, Carmina, 1:280–82; and Florence, no. 119, ed. Miller, Carmina, 1:310 (for the lemma, see Martini, Carmina, 47). For more on the Glabas family’s patronage of Philes, see Belting, Mango, and Mouriki, Pammakaristos, 12–13, 16. 17 P. Underwood, The Kariye Djami, 4 vols. (New York, 1966; Princeton, N.J., 1975), 1:276–80; see also plates, 3:537–39; the translation of the poem is by A. Van Millingen. See also Ø. Hjort, “The Sculpture of Kariye Camii,” DOP 33 (1979): 250–55. 18 PLP 12, no. 29132.
80
EPIGRAMS IN CONTEXT
halo. The poet, probably to be identified with Manuel Philes,19 eulogizes Tornikes for his military prowess, his service to the state, and his marriage to a highborn woman; then, at lines 17–18, he abruptly shifts the tone to comment: And leaving his life as a splendid example, He lies a poor monk among bones,
thus drawing the viewer’s attention to his monastic portrait and to his sarcophagus (now missing). Much of the poem is addressed to the visitor who seeks to learn the history of the man buried in the niche, but the final three lines shift course to invoke the name of Christ (whose bust is directly to the left), begging Him to be merciful to the deceased and to grant him admission to Paradise. There is one curious disjunction between the poem and the images: Tornikes’ wife (whose secular name we do not know) is given equal billing with her husband in the double portraits, but is scarcely mentioned in the poem, except for the two verses (15–16) that comment on Tornikes’ “highly born and seemly marriage connection.” 3. Funerary Stele of the Nun Maria. Another Constantinopolitan funerary epigram inscribed in stone is preserved on a fragmentary marble plaque presently located in the Archaeological Museum of Istanbul (Fig. 9).20 Its original provenance is unknown, but it may have come from the south church or mausoleum of the Lips monastery (Fenari Isa Cami).21 The first editor of the stele, William Buckler, argued that its unworn surface suggested that it had initially been built into a wall. A standing female figure is accompanied by an inscription, which originally was at least fourteen lines long, and may have been twice that length, if Buckler is correct in hypothesizing that the figure was once flanked by an inscription on both sides. On the basis of letter forms, he dated the stele to 1275–1325. The verses, in the first person, purportedly written by the nun Maria, boast of her Palaiologan lineage, lament obscure past sorrows, and supplicate Christ to receive her into His heavenly bridal chamber. Buckler assumed that the sculpted figure was Maria, while the authors of the recent sculpture catalogue of the Archaeological Museum of Istanbul are more cautious, suggesting that the figure might be the Virgin.22 Since Maria’s poem addresses Christ rather than the Virgin, I would argue that the image is more likely to be that of the deceased nun. It is noteworthy that this relatively small stele, originally perhaps 1 m in height, was able to accommodate a poem that may have contained twenty-eight to thirty verses. One should also remark the careful carving of the letters, decorative ligatures, and the exuberant letter x. Judging from the amount of space occupied by the epigram and the care of execution, one feels that the metrical inscription was deemed as important as the ˇevcˇenko, “Theodore Metochites, the Chora, and the Intellectual Trends of His Time,” in UnCf. I. S derwood, Kariye Djami (as above, note 17), 4:21 n. 14. 20 W. H. Buckler, “The Monument of a Palaiologina,” in Me´langes offerts `a m. Gustave Schlumberger, 2 vols. (Paris, 1924), 2:521–26; N. Firatlı, La sculpture byzantine figure´e au Muse´e arche´ologique d’Istanbul (Paris, 1990), 67, no. 115. 21 T. Macridy, “The Monastery of Lips (Fenari Isa Camii) at Istanbul,” DOP 18 (1964): 271 and n. 64. 22 A. Grabar (Sculptures byzantines de Constantinople, vol. 2, Sculptures byzantines du moyen ˆage, XIe–XIVe sie`cle [Paris, 1976], no. 128) also identifies the figure as the Virgin. 19
ALICE-MARY TALBOT
81
image by the patron who commissioned it. The first-person voice of the epigram also served to enhance the efficacy of the prayer that the nun addressed to Christ from her grave: Receive me, Christ, [my] handsome bridegroom; Heeding the intercession of Thy mother, Open for us the spiritual bridechamber. Clothe us in the garment of divine marriage, And place us in the ranks of your [fellow] banqueters. I, the nun Maria, faithful sebaste And daughter of a Palaiologos, write these words.
B. Epigrams on Icons and Icon Frames 1. Marble Bas-Relief of the Virgin Episkepsis from the Makrinitissa Monastery. Moving from funerary monuments to works of religious art, I next examine a selection of icons and icon frames that bear metrical donor inscriptions. One might begin with an extremely rare example of a marble icon, a bas-relief plaque of the Virgin as Oxeia Episkepsis, “swift visitation,” with a medallion of the Christ Child in front of her chest (Figs. 10, 11).23 It originally belonged to the Thessalian monastery of the Oxeia Episkepsis, founded by Constantine Maliasenos at Makrinitsa on Mt. Pelion in the beginning of the thirteenth century.24 The plaque is of special interest because it contains both the small figure of the monk donor praying at the Virgin’s feet and, on the marble frame, his metrical invocation of the Virgin. Unfortunately, the inscription was already damaged by 1924 when it was first published, so that it cannot be restored in its entirety. The poem was eight to nine lines in length,25 with the first verse, which invoked the Virgin, at the top of the marble slab. The poem continued down the left side of the frame, then went across the bottom, and concluded on the right side. The words on the sides of the frame are more difficult to decipher since each one is broken up into fragments of two to three letters—all that the narrow width of the frame would accommodate. A similar arrangement of letters on the side panels can be found on the painted thirteenth-century icon of the Virgin Dexiokratousa from Mt. Sinai.26 In the Makrinitsa epigram the monk donor prays to the Virgin for salvation from hellfire and for admission to the heavenly hosts in return for his commissioning of her image in marble. The monk’s name was either omitted from the poem or is illegible. Georgios Soteriou dates the plaque to the thirteenth century and suggests that the donor was a member of the Maliasenos family. 23 N. Giannopoulou, “AiJ para` th` n Dhmhtria´ da buzantinai` monai´,” jEp. JEt.Buz.Sp. 1 (1924): 237–40. The fullest publication, best photographic reproduction, and most accurate rendering of the inscription are to be found in G. Soteriou, “Buzantinai` ajna´ glufoi eijko´ ne",” in Recueil d’e´tudes de´die´es `a la me´moire de N. P. Kondakov (Prague, 1926), 133–36. I am indebted to Sharon Gerstel, who first drew my attention to this plaque. Unfortunately, no decent photograph exists because the relief, first published in 1924, has since been badly damaged in an earthquake. 24 See A.-M. Talbot, “Makrinitissa Monastery,” ODB 2:1273–74. 25 Most of the lines are dodecasyllables, but two can perhaps be restored as decasyllables. I thank Lee Sherry for his advice on the meter. 26 K. A. Manafis, ed., Sinai: Treasures of the Monastery of Saint Catherine (Athens, 1990), fig. 62.
82
EPIGRAMS IN CONTEXT
2. The Freising Icon Frame. Metal icon frames were frequently added to icons as adornment and protection. Usually of silver or silver gilt, they represented a substantial and long-lasting pious donation. Epigrams inscribed on these frames were often better preserved than those painted on the original icons, and thus ensured the perpetuation of the donor’s name and the sentiments of the verses. Such frames also protected the vulnerable edges of icons in wood and especially those in steatite, which was a soft and inherently fragile stone.27 Icon frames lent themselves particularly to the inclusion of epigrams. Not only could they accommodate quite lengthy poems, as we shall see, but, being an adornment added to an icon after its creation, they offered more time for the patron and the poet to devise a suitable theme related to its iconography and materials. I begin with the well-known silver-gilt example preserved at the cathedral in Freising (Fig. 12). Attributed to the thirteenth century, the frame surrounds an eleventh-century icon of the Virgin.28 The fourteen dodecasyllable verses may be translated thus: The yearning of my soul, and silver, and thirdly gold Are [here] offered to you, the pure Virgin. However, silver and gold by nature Could be stained since they are of perishable material, Whereas the yearning of an immortal soul Could not be stained nor come to an end. For even if this body should dissolve in Hades, It continues to entreat you for the mercy of its soul. These words are addressed to you By Manuel Dishypatos, kanstrisios29 and deacon. Receive them compassionately, O Virgin, And grant in return that through your entreaties I may traverse this ephemeral life without sorrow, Until you show the end of the day and light.
To summarize, Manuel Dishypatos,30 the donor of the silver-gilt icon frame, addresses the Virgin depicted in the icon, hoping that in exchange for his gift she will grant him a peaceful life on earth and eternal salvation. At the same time he notes that the third element of his gift, his spiritual love for the Virgin, is eternal, while the gold and silver are perishable. The fourteen lines of the poem are divided among ten panels, notable for their proportions; they are larger than the enamel medallions and together take up more than 27 Cf. I. Kalavrezou-Maxeiner, Byzantine Icons in Steatite (Vienna, 1985), 1:28–31; she also argues that the frames “made these icons of relatively small size more impressive and emphasized the preciousness of the objects within” (p. 30). Such frames can still be seen on steatite icons of St. Demetrios preserved in Moscow and Paris; ibid., 198–200 and 201–2, nos. 124 and 127, pls. 59 and 62. The Kremlin frame is original; the Louvre frame was made later, after the icon had been broken. Few such frames survive, but the reverse of some steatite plaques preserves indications that they were once framed; cf. ibid., 28–29. 28 ¨nstler der For the dates of the icon and the frame, see A. Legner, ed., Ornamenta Ecclesiae: Kunst und Ku Romanik (Cologne, 1985), 3:171–72. See also A. Grabar, Les reveˆtements en or et en argent des icoˆnes byzantines du moyen ˆage (Venice, 1975), 41–43, no. 16; C. Wolters, “Beobachtungen am Freisinger Lukasbild,” Kunstchronik 17 (1964): 85–91; M. Kalligas, “Buzantinh` forhth` eijkw` n ejn Freising,” Arc. j jEf. (1937): 501–6. 29 A patriarchal official; cf. A. Kazhdan, “Kastresios,” ODB 2:1111–12. 30 Perhaps to be identified with Manuel Opsaras Dishypatos, the 13th-century metropolitan of Thessalonike; PLP 3, no. 5543.
ALICE-MARY TALBOT
83
half the frame. The letters are nielloed, and often include breathings and accents; ligatures are used as necessary to compress a verse in order for it to fit into a panel. The top and bottom panels of the frame hold two lines apiece, while each panel on the sides accommodates one verse. The arrangement of verses differs from that on the Makrinitsa marble icon. The first four verses are at the top; the poem then continues with verses 5–7 in the three panels on the right side, while verses 8–10 are on the left side. The final four lines are at the bottom. On the Makrinitsa panel, in contrast, the poem moved from top to left to bottom, ending on the right side. As a result of the arrangement of the panels on the Freising frame, the verses in which Manuel states his name and titles are in the two lower left panels, a location that corresponds more or less to the placement of the figural donor panel on the icon frame in the Tretiakov Gallery, which was presented by Constantine Akropolites and his wife (Fig. 13).31 Akropolites, identified by a prose inscription, extends his hands in prayer and offering to the Virgin, whereas Manuel makes similar gestures of prayer and offering through the medium of verse. 3. The Vatopedi Icon Frame. Yet another metrical dedicatory inscription is preserved on a fourteenth-century silver-gilt revetment at Vatopedi which adorns an icon of the Virgin Hodegetria. The epigram commemorates the gift of the frame by a certain Papadopoulina in honor of her sister, and contains a prayer for the grace of the Lord to descend upon the two siblings. One should note the different spatial arrangement of the epigram: the twelve verses are divided between two panels at the bottom of the revetment (Fig. 14).32 If the patron had so wished, the artisan could have easily fitted another twelve lines at the top of the frame. I would conclude that there would have been no problem in accommodating poems of up to twenty-four verses on icon frames or revetments. C. Epigrams on Reliquaries 1. The Reliquary Casket from Trebizond. Reliquaries, like icons, were often enhanced by donors with inscriptional epigrams; a fine example is the silver-gilt reliquary casket of the late fourteenth or the fifteenth century from Trebizond, now in the Treasury of San Marco in Venice (Figs. 15, 16).33 The casket, 28 cm in length, has figurative imagery only on the lid, where an enthroned Christ is depicted, flanked by four standing saints, Eugenios, Valerianos, Kanidios, and Aquila, martyred at Trebizond under Diocletian. One can assume that the casket once held relics of these martyrs, and was originally housed in one of the principal churches of Trebizond, perhaps at the katholikon of the monastery of St. Eugenios.34 The four sides of the casket are ornamented with six bands of decoration, two of which contain an engraved and nielloed metrical inscription. Its verses may be translated as follows: Grabar, Reveˆtements, 45–46, no. 18; A. Bank, Byzantine Art in the Collections of Soviet Museums (Leningrad, 1985), 317 and figs. 252–54. 32 Grabar, Reveˆtements, 49–52, no. 21. 33 D. Buckton, The Treasury of San Marco, Venice (Milan, 1984), 201–3, no. 28. 34 On the monastery of St. Eugenios, see A. Bryer and D. Winfield, The Byzantine Monuments and Topography of the Pontos (Washington, D.C., 1985), 222–24, and J. O. Rosenqvist, The Hagiographic Dossier of St. Eugenios of Trebizond in Codex Athous Dionysiou 154 (Uppsala, 1996), 81–85. 31
84
EPIGRAMS IN CONTEXT You martyrs did not fear to shed your blood But contended with all your might; I am speaking of the unbending pillars of the East, The gleaming good fortune of the Trapezuntines, The prizewinner Eugenios together with Aquila, Valerianos, and Kanidios. And Christ Himself is providing you With the reward of your immeasurable labors; For He is giving out the crowns that you deserve. And I, wretched that I am, filled with sin, Make you intercessors for my salvation In my desire to escape condemnation.
The poem fulfils several functions—praise of the martyrs, evocation of the pride of Trebizond in possessing their relics, description of the iconography in which Christ offers crowns to the martyrs, and the donor’s prayer for intercession. The poet’s comparison of the martyrs with “unbending pillars” reflects their upright stance and perhaps the colonnettes that separate their images, while his use of the adjective lampro´ " to characterize the “gleaming good fortune of the Trapezuntines” evokes not only the martyrs’ fame but also the brilliant silver gilt of the casket. In the final three verses the donor asks the martyrs to intercede with Christ in order to obtain remission for his sins; the saints’ hands are thus stretched out not only to receive their martyrs’ crowns but also in prayers of intercession for the donor and for mankind. The letters of the inscription are handsomely incised and easily legible, in part due to the inclusion of breathings and accents. The use of ligatures gave the engraver flexibility, so he could accommodate verses of variant length in the same space. The poem, of twelve dodecasyllable verses, begins with the sign of the cross in the middle of the front side of the casket, just to the right of the security clasp, and continues in the same band around the four sides of the casket; then the viewer needs to move his eye to the middle of the lower inscribed band where verses 7–12 continue. Each inscribed band on the long sides of the casket accommodates two verses, while the short sides have one verse each. The viewer would have to turn the reliquary around two times in his hands in order to read the inscription or, if it were placed on a table, might perhaps walk around it twice. In so doing he might utter aloud praise of the four martyrs and make a prayer to them to intercede for his salvation. The prayer was of course originally intended to be the supplication of the donor but, since he (or she) remains anonymous, it has universal application and can be appropriately repeated by any wretched sinner. 2. The St. Demetrios Reliquary at Dumbarton Oaks. My second example, a gold and enamel pendant reliquary at Dumbarton Oaks (Fig. 17), is also the smallest object examined in this study. It is attributed to thirteenth-century Thessalonike.35 The obverse of the reliquary bears a bust of St. Demetrios, while the standing figures of Sts. Sergios and Bacchos are to be found on the reverse. Amazingly, this tiny object, less than 3 cm in M. C. Ross, Catalogue of the Byzantine and Early Mediaeval Antiquities in the Dumbarton Oaks Collection (Washington, D.C., 1965), 2:111–12, no. 160. J. Durand (ed., Byzance: L’art byzantin dans les collections publiques fran¸caises [Paris, 1993], 445) concurs in assigning the reliquary to the 13th century. For a similar reliquary, see D. Buckton, Byzantium: Treasures of Byzantine Art and Culture from British Collections (London, 1994), no. 200. 35
ALICE-MARY TALBOT
85
diameter, or the size of a pillbox, accommodates a four-verse epigram: two lines on the obverse and two running around the side. The quatrain may be translated as follows: The faith of Sergios carries the venerable receptacle Of Demetrios’s blood together with the balm. He asks to have you as protector, while he is living and when he is dead, Along with the two martyrs who have won the prize [of glory].
These verses inform us that the reliquary belonged to a certain Sergios, who included the effigy of his patron saint, together with his companion Bacchos, on the reverse, even though the reliquary was primarily dedicated to St. Demetrios and contained myron or perfumed oil from his tomb in Thessalonike. The poem is essentially Sergios’s prayer to all three saints to protect him in life and to help him attain salvation after death. D. Liturgical Vessels The Steatite Panagiarion from the Panteleimon Monastery on Mt. Athos. I conclude this survey of epigrams appearing in context with yet another category of objects—liturgical vessels. The fourteenth-century steatite panagiarion from the Panteleimon monastery on Mt. Athos (Fig. 18),36 unfortunately now lost, exemplifies an ideal collaboration between poet and artisan, and perhaps patron as well. As Ioli Kalavrezou has remarked, both in iconography and in form the green twelve-lobed bowl-shaped paten closely resembles the ribbed melon domes to be found in Palaiologan churches such as the Kariye Cami. As is appropriate for a panagiarion (a type of paten used for the bread that monks offered to the Virgin Panagia at mealtime or during the orthros service), the central medallion contains an image of the Virgin holding the Christ child; each lobe holds the bust of a prophet. Although only 9 cm in diameter, the paten teems with inscriptions, including eight dodecasyllable verses. The first two verses frame the central medallion and read as follows: O Mother without a husband, Virgin who nourishes an infant, May you protect Alexios Komnenos Angelos.
This couplet refers to the iconography of the central roundel, evoking a sense of wonderment at the mystery of the virgin birth; at the same time it is a prayer on behalf of the donor of the panagiarion, the emperor Alexios III Komnenos of Trebizond (1349–90). The second metrical inscription runs around the border of the panagiarion; the lobed design greatly increases the amount of space available for the carving of letters. Each of the six verses is divided between two lobes. The poem reads as follows: The meadow and the plants and the light with three rays. The stone is a meadow and the row of prophets are the plants. The three beams are Christ, the bread and the Virgin. The maiden lends flesh to the Word of God, And Christ by means of bread distributes salvation And strength to Alexios Komnenos Angelos.37 Kalavrezou-Maxeiner, Steatite, 83–85, 206–8; Talbot, “Epigrams,” 145–46 and n. 54, where I mistakenly assigned the panagiarion to Chilandar. 37 The translation is that of Kalavrezou-Maxeiner, Steatite, 206. 36
86
EPIGRAMS IN CONTEXT
As Kalavrezou has pointed out, the verses evoke the green color of the steatite in their allusion to the stone as meadow and the prophets as plants. The poem also explains the iconography and its suitability for a paten on which the bread of salvation was offered. The final couplet assures the imperial patron of his future salvation. These metrical inscriptions, combined with the identifying labels of each prophet and the text on each scroll, take up fully half of the paten and perform an important decorative and aesthetic function, accentuating the central roundel and the lobed border and defining the space for each image. At the same time they enhance the viewer’s appreciation of the steatite material and the understanding of the iconography. The circular and lobed design of the inscriptions adds yet another dimension: as with the Trebizond and St. Demetrios reliquaries, in order to read the lines of poetry the viewer would have to turn the paten around in his hands twice, no doubt sounding the verses out loud; in so doing he would utter a prayer to the Virgin for protection of the emperor Alexios and would reiterate Christ’s assurance of salvation. This process closely resembles the action of a visitor to the Pammakaristos parekklesion, who might sound out the lengthy inscriptions on exterior and interior cornices. This sample of surviving metrical inscriptions from the thirteenth and fourteenth centuries has demonstrated that epigrams were painted, carved, or engraved on objects of various types and in different media; in addition to the cited examples in or on stone, mosaic, painted icons, fresco, steatite, metal, and enamel, numerous epigrams are preserved in illuminated manuscripts.38 As we have seen, even relatively small objects could accommodate surprisingly long poems, and there seems to be no limit to the length of an epigram destined for a building: the epigram on the cornice of the sixth-century church of St. Polyeuktos was seventy-six lines,39 and the north facade of a Palaiologan church in Mistra once held eighty-seven dodecasyllable verses.40 I hope to have shown that metrical inscriptions had an aesthetic function in the overall decorative scheme of a building or object, being allotted a significant amount of space and a strategic location and being carefully and artfully executed. They might add color or ornament, and often accented spatial divisions. At the same time, with some exceptions the content of the epigrams that survive in situ does little to express aesthetic appreciation of the structure or artifact. The poems considered so far, especially those on the funerary monuments and the icons, are primarily dedicatory and commemorative, or prayers for intercession, intended to stimulate the viewer’s spiritual response. II. EPIGRAMS KNOWN ONLY FROM ANTHOLOGIES In only a few cases, such as the Panteleimon panagiarion and the Trebizond reliquary, does one find verses that are more exegetical and ecphrastic in nature. Thus, in order to understand better how inscribed epigrams might have reflected or stimulated a more 38 E.g., Patmos gr. 81, fols. 16v, 98v, 238v; cf. A. D. Kominis, Patmos: Treasures of the Monastery (Athens, 1988), 319–20, 322, pls. 41, 42, 44. 39 R. M. Harrison, Excavations at Sarac¸hane in Istanbul (Princeton, N.J., 1986), 1:5–8, 407–11; idem, A Temple for Byzantium (Austin, Tex., 1989), 33–34. 40 The church of the Panagia of Panori, now destroyed; cf. G. Millet, “Inscriptions byzantines de Mistra,” BCH 23 (1899): 150–54.
ALICE-MARY TALBOT
87
aesthetic form of viewer response to an object, one needs to supplement the foregoing sample of surviving epigrams by returning to Manuel Philes and looking at a few of his poems that have become separated from their context and are preserved only in anthologies. Our appreciation of these verses will be increased if we try to place them in context by pairing them with existing works of art similar to those evoked by the epigrams. Space permits examination of only two types of such epigrams—those that enhance the viewer’s response to the icons of certain saints and those that allude to the materials of carved precious stones. A. Epigrams on Icons Turning first to iconography, we might compare a quatrain entitled “On an icon of the Forty Holy Martyrs, adorned by Athanasios the Monk” 41 with the fourteenth-century miniature mosaic icon of the Forty Martyrs of Sebaste exhibited at Dumbarton Oaks (Fig. 19). These Early Christian martyrs were condemned to stand in an icy lake until they froze to death; the icon shows them half-naked, shivering and clutching themselves, as their martyrs’ crowns descend from heaven. Philes’s poem, probably engraved on a silvergilt frame, reads as follows: The spiritual ardor of Athanasios warms Your hands paralyzed by cold, So that in recompense to him you may Worthily extend [them in supplication] to the Lord.
The monk hopes that his fervent prayer may warm up the freezing martyrs enough that they will unclench their arms and extend them in prayer to the Lord above. This sentiment draws the viewer’s attention to the martyrs’ tense bodies and increases his empathy with their suffering in the frigid waters. The epigram might also encourage the viewer’s anticipation that at any moment the martyrs will miraculously relax their arms and raise them in prayer to heaven.42 Five of Philes’s poems43 relate to an icon of St. Onouphrios, an Egyptian hermit who is generally represented as totally naked, his only protection from the elements being his long white beard and his body hair,44 as can be seen, for example, in a fresco from the church of Calendzˇicha (Fig. 20). At least three of the poems were designed as alternative choices for a silver-gilt icon frame commissioned by a certain Basil.45 The epigrams focus on Onouphrios’s nudity; one quatrain includes the word gumno´ " or its cognate in each of its four lines: O naked one, thrice-blessed and well adorned, may you adorn me Who am garbed in nakedness of good [deeds]; Even better, O lover of nudity, may you show me Your suppliant stripped naked of shameful passions.46 Escurial, no. 79, ed. Miller, Carmina, 1:35. For an earlier epigram on the Forty Martyrs that encourages viewer response, see Maguire (Byzantine Epigram, 12–13), who discusses the verses accompanying a wall painting at Asinou. 43 Paris, nos. 52–54, ed. Miller, Carmina, 2:93–94; Florence, no. 38, ed. Miller, Carmina, 1:214; Florence, no. 129, ed. Miller, Carmina, 1:319 (on John Chrysostom and Onouphrios). 44 ODB 3:1527. 45 Paris, nos. 52–54, ed. Miller, Carmina, 2:93–94. 46 Florence, no. 38, ed. Miller, Carmina, 1:214. 41 42
88
EPIGRAMS IN CONTEXT
Philes delights in wordplay on the theme of nudity, commenting in another of the poems that the golden icon revetment will cover the saint’s nakedness.47 The poet also contrasts Onouphrios’s nakedness with that of Adam, and expresses the hope that the donor Basil, who has been “naked to shame” 48 and is “garbed in nakedness of good [deeds],” will be “stripped naked of shameful passions.” 49 Basil was evidently depicted on the frame as a standing donor figure;50 one can assume that the viewer’s response to the wordplay on “clothed” and “naked” would have been enhanced by the contrast between the fully garbed figure of Basil on the frame, decrying his spiritual “nakedness of good deeds,” and the figure of Onouphrios, who is totally nude yet garbed in silver-gilt revetment. B. Epigrams on Glyptics My final examples are drawn from a series of epigrams on glyptics, the carvings in semi-precious stone, which delighted Philes and the patrons who commissioned his poems. I assume that most of these epigrams were designed for metal frames surrounding the carvings and were not cut on the stones themselves. Rock crystal (called in Greek li´qo" kru´ o", “cold rock,” or kru´ stallo", “ice”) particularly intrigued Philes, as can be seen from a poem on an image of Christ, which may have been similar to the carving now housed in the Benaki Museum (Fig. 21):51 This stone is water, not really stone; He Who freezes flowing water into ice Also freezes this into the nature of stone Lest the rock melt and flow away.52
Thus it is Christ’s image on the crystal that miraculously keeps it from melting. Elsewhere Philes alludes to the sparkle of crystal, which he terms a “fiery coal,” 53 well illustrated in the late antique piece of crystal preserved in the curious assemblage in the Treasury of San Marco, called the Grotto of the Virgin.54 Another type of stone favored by carvers was jasper or heliotrope, with colored veins running through it; a particularly fine example is the heliotrope cameo of Christ now housed in the Kremlin in Moscow (Fig. 22).55 In a poem on a jasper enkolpion of Daniel, with green and red veins, Philes exclaims: The stone is wet, but I see fire within. The stone contains fire, the flame contains dew.56 Paris, nos. 52.6–7, 54.6, ed. Miller, Carmina, 2:93–94. Paris, no. 52.4. 49 Florence, no. 38.1 and 4. 50 Paris, no. 52.8: JO Basi´leio" ou»to" eJstw` " ejnqa´ de; Paris, no. 53.9: JO Basi´leio" ou»to" eJstw` " ejggu´ qen. 51 Benaki Museum, no. 2113; cf. A. R. Bromberg, Gold of Greece: Jewelry and Ornaments from the Benaki Museum (Dallas, Tex., 1990), 84, pl. 65. 52 Escurial, no. 86, ed. Miller, Carmina, 1:38. 53 Appendix, no. 59.3, ed. Miller, Carmina, 2:420. For other poems on rock crystal, see Escurial, no. 87, ed. Miller, Carmina, 1:38; and Paris, nos. 19–21, ed. Miller, Carmina, 2:65–66. 54 Buckton, Treasury of San Marco, 117–20, no. 8. 55 Cf. Bank, Byzantine Art, 298 and fig. 152. 56 Escurial, no. 107.1–2, ed. Miller, Carmina, 1:50. 47 48
ALICE-MARY TALBOT
89
Another epigram, entitled “On the great Demetrios represented on a stone with red veins,” reads as follows: Your throat is dyed red from your slaughter, The tip of the sword is also dyed red; For you in being slain yourself slew the error, O new kind of victim, O slayer who lives in stone!57
These verses suggest that the artisan may have masterfully designed his carving so that one red vein of the stone coincided with Demetrios’s slit throat, while another indicated the bloody tip of his slayer’s sword—or perhaps this effect was achieved by carving away the top layer of a stone like sardonyx, as was done in a cameo of St. George and St. Demetrios now in Paris (Fig. 23).58 CONCLUSION I would argue that most of the epigrams discussed here were created de novo, expressly to accompany the buildings or objects evoked. This is an obvious conclusion in the case of the dedicatory and funerary epigrams; it can also be inferred in such instances as the poem on St. Demetrios, cited immediately above, in which a carved stone with calculated placement of red veins is described, or the Panteleimon panagiarion in which the verses explain the iconography of the object in such detail. In cases where alternative versions or “trial pieces” of an epigram survive, as in Philes’s poem on an icon of St. Onouphrios, one can also assume that the verses were commissioned to order by a donor for a specific object. A possible exception to this pattern is Philes’s epigram on the cameo of Christ in rock crystal, which could in theory be reused for the frame of any similar carving.59 The majority of the epigrams examined here seem then to have been specifically commissioned, and thus involved an interaction among patron, artisan/architect, and poet.60 Recent scholarship has emphasized the dominant role of the patron in the creation of medieval works of art and literature; and, indeed, epigrams (and prose inscriptions as well) credit the patron rather than the artist with the creative act. Many patrons, speaking in the first person through the ghostwriter poet, take credit not only for writing the verses, but for producing the work of art as well.61 One must assume that the patron decided on the type of object or building he or she wished to commission, such as a commemorative monument, or an ex-voto offering in thanksgiving for healing or in hope of future salvation. The patron then selected the Escurial, no. 281, ed. Miller, Carmina, 1:136–37. Bibliothe`que Nationale, Cabinet des Me´dailles; see Durand, Byzance, no. 193. I am grateful to Martin Dennert for suggesting this comparison. 59 For further discussion of this phenomenon of reused epigrams with an “autonomous existence,” see Maguire, Byzantine Epigram, 6–14. 60 This is the second of the two scenarios postulated by Maguire in his Byzantine Epigram, according to which “the poet . . . compose[d] the epigram on the basis of close knowledge of the work of art on which it is inscribed” (p. 6). 61 See, for example, Escurial, no. 156, ed. Miller, Carmina, 1:66–67, for a poem on an icon of the Virgin written by Philes on behalf of Manuel Atzymes, in which Atzymes is made to say, “I paint thee with the hand of an icon-painter” (v. 16) and “Thy <servant> Manuel Atzymes says these words” (v. 23). 57 58
90
EPIGRAMS IN CONTEXT
appropriate artisan/architect and no doubt had some say in the materials, iconography, and design of an object or structure. The patron might have his role commemorated through a portrait as did Metochites at Chora and Akropolites on the Tretiakov icon frame, or through epigrams as did Maria-Martha at Pammakaristos and Manuel Dishypatos on the Freising icon frame, or perhaps through both portraiture and verse or prose inscription as on the Makrinitsa bas-relief. If the patron wished to have an epigram included in or on the object, he might hire a poet to produce verses with the desired sentiments, expressing family pride, thanksgiving for healing, prayer for intercession, exegesis of the iconography, or whatever the occasion demanded; it is also conceivable that the patron might have written the epigram himself. The writer would have to coordinate his efforts with the plans of the architect or artisan, to get some idea of how long an epigram might fit in the design. Then the skill of the engraver or stone carver would be called into play, to plan the cutting of the letters so that they would fit into the prescribed space. His task was eased by the flexibility provided by the use of ligatures and abbreviations, but he would still have to plan his work meticulously. The composition of epigrams for frames may have followed a somewhat different procedure. I would argue that frames were often commissioned for already existing works of art, such as icons or carved stones; as in the case of the Freising icon, some of these objects may in fact have been centuries old.62 Under these circumstances both patron and poet would be responding to a work of art in front of them, rather than one still in conceptual form; it is therefore not surprising that this would lead to a more aesthetic evocation of the iconography and materials. Thus I would argue that many of the Palaiologan ecphrastic epigrams known only from anthologies were originally intended for frames. Since these frames or revetments were primarily fashioned of silver gilt, they were particularly vulnerable to being melted down for the content of their precious metals.63 I would therefore suggest that it is not mere coincidence that few inscriptional ecphrastic epigrams are preserved in situ. Because of the relative paucity of ecphrastic epigrams surviving on the objects for which they were intended, one must resort to the pages of anthologies for fuller understanding of this genre of poetry. At the same time, examination of the preserved inscriptional epigrams where one can actually see the interplay between poetry and art can provide a better imaginative context for reading the epigrams of Philes and other poets. Dumbarton Oaks In addition to the Freising icon, one could cite the 11th-century steatite icon of St. Nicholas from Mt. Sinai, with a painted frame of the Palaiologan period; cf. Kalavrezou-Maxeiner, Steatite, 30, 106–7, and pl. 10. But the reverse could also occur, as in the example of the 15th-century icon of the Crucifixion in Moscow (cited by Maguire, Byzantine Epigram, 23), in which a 12th-century poem was reused, in revised form, on the frame. 63 Moreover, if a frame was made of wood revetted with silver gilt, it was susceptible to dry rot or being eaten by worms; cf. Kalavrezou-Maxeiner, Steatite, 29. 62
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Parekklesion of the church of Pammakaristos, Istanbul, south facade (photo: T. Mathews)
2 Parekklesion of the church of Pammakaristos, Istanbul, inscribed epigram on the south facade cornice (photo: T. Mathews)
3
Parekklesion of the church of Pammakaristos, Istanbul, detail of the inscribed epigram on the cornice (photo: C. Mango)
4 Parekklesion of the church of Pammakaristos, Istanbul, interior, looking toward the apse
5 Parekklesion of the church of Pammakaristos, Istanbul, apse mosaic of Christ Hyperagathos
6
Parekklesion of the church of Pammakaristos, Istanbul, interior view with cornices and the arcosolium for the tomb of Michael Tarchaneiotes Glabas
7 (a, b) Parekklesion of the church of Pammakaristos, Istanbul, details of the interior cornice with the painted epigram
8 Parekklesion of the church of the Chora monastery, Istanbul, tomb of Michael Tornikes
9
Funeral stele of Maria Palaiologina, Archaeological Museum, Istanbul (photo: I. Sevcenko)
10
Marble icon of the Virgin of Oxeia Episkepsis, formerly in Makrinitissa church, Thessaly (after G. Soteriou, “Buzantina‹ énãglufoi efikÒnew,” in Recueil d’études dédiées à la mémoire de N. P. Kondakov [Prague, 1926], 133, fig. 6)
11
Icon of the Virgin of Oxeia Episkepsis, line drawing of the inscription on the frame (after Soteriou, “Buzantina‹ énãglufoi efikÒnew,” 134, fig. 8)
12
Icon of the Virgin with a silver-gilt frame, Freising cathedral, Treasury (after A. Grabar, Les revêtements en or et en argent des icônes byzantines du moyen âge [ Venice, 1975], fig. 39)
13
Icon of the Virgin with a silver-gilt frame, Tretiakov Gallery, Moscow (after A. Bank, Byzantine Art in the Collections of Soviet Museums [ Leningrad, 1985], fig. 252)
14
Icon of the Virgin Hodegetria with a silver-gilt frame, Vatopedi monastery, Mt. Athos (after Grabar, Revêtements, fig. 47)
15
Reliquary casket from Trebizond, Treasury of San Marco, Venice (after D. Buckton, The Treasury of San Marco, Venice [ Milan, 1984], 201)
16
Reliquary casket from Trebizond, Treasury of San Marco, Venice, side view (after Buckton, Treasury of San Marco, 202)
17
18
Steatite panagiarion, formerly at Panteleimon monastery, Mt. Athos (after I. Kalavrezou-Maxeiner, Byzantine Icons in Steatite [ Vienna, 1985], pl. 65)
Enamel reliquary of St. Demetrios, Dumbarton Oaks, Washington, D.C.
19
Miniature mosaic icon of the Forty Martyrs of Sebaste, Dumbarton Oaks, Washington, D.C.
20
Fresco of St. Onouphrios by Manuel Eugenikos, church of Calendzˇicha (photo: N. Teteriatnikov)
21
22
Framed heliotrope cameo of Christ, Kremlin, Moscow (after Bank, Byzantine Art, fig. 152)
Rock-crystal cameo of Christ, Benaki Museum, Athens (after A. R. Bromberg, Gold of Greece: Jewelry and Ornaments from the Benaki Museum [ Dallas, Tex., 1990], pl. 65)
23
Sardonyx cameo of St. George and St. Demetrios blessed by Christ, Bibliothèque Nationale, Cabinet des Médailles, Paris (after J. Durand et al., Byzance: L’art byzantin dans les collections publiques françaises [ Paris, 1993], fig. 193)
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Ploys of Performance: Games and Play in the Ptochoprodromic Poems MARGARET ALEXIOU
THE PTOCHOPRODROMIC POEMS: LANGUAGE, HUMOR, AND EPISODIC STRUCTURE he value of the four so-called Ptochoprodromic Poems has long been recognized by historians for their unique vignettes of everyday life and their inventories of foodstuffs, items of clothing, and other realia.1 Philologists and linguists have mined the text for rare lexical and morphological forms.2 However, there has been far less unanimity as to their stylistic and literary merits. Three major stumbling blocks seem to have prevented them from being taken seriously as “literature”: the mixed levels of language, ranging from learned and semi-learned Greek to vulgar street slang; the coarseness of their humor, especially in matters relating to what goes into and out of the gut and belly; and their discursive, episodic structure. This article addresses each of these three issues, along with questions of authorship and dating, before moving on to a new approach to the poems—the poet’s use of games and play. As for language, it is now generally recognized that vernacular forms emerged from within court circles in the course of the twelfth century, as part of a new confidence in language as a means of exploring multiple levels of expression and signification, rather than as an end in itself. Particularly relevant is the following extract from a letter addressed by Theodore Prodromos to the nomophylax Alexios Aristenos. Language, style, and communication figure throughout the letter as important matters of debate, but it
T
This paper is based on research carried out for an edition of the Ptochoprodromic Poems, with Greek text and facing English translation, introduction, commentary, and glossary, to be completed in collaboration with Michael Hendy. I should like to acknowledge my gratitude to Dumbarton Oaks, where most of the research was carried out during my term there as a summer fellow (1994), and to thank the following people for their generous advice: John Duffy, Angeliki E. Laiou, Alexander P. Kazhdan,† Henry Maguire, Eric McGeer, Ian Rutherford, Lee Sherry, and Sarolta Taka´cs. I would also like to acknowledge special debts to Panagiotis Agapitos, Panagiotis Roilos, and Dimitris Yatromanolakis, who made invaluable suggestions. 1 See, especially, P. Koukoules, Buzantinw'n Bi´o" kai` Politismo´ ", 6 vols. (Athens, 1947–55); M. F. Hendy, Studies in the Byzantine Monetary Economy, c. 300–1450 (Cambridge, 1985), 514, 588; A. E. Harvey, Economic Expansion in the Byzantine Empire (Cambridge, 1989), 205. 2 D. C. Hesseling and H. Pernot, eds., Poe`mes Prodromiques en grec vulgaire (Amsterdam, 1910), passim; N. P. Andriotis, Lexikon der Archaismen in neugriechischen Dialekten (Vienna, 1974); H. Eideneier, “Zu den Ptochopro¨hrung, kritische Ausgabe, deutsche Uebersetzung, Glosdromika,” BZ 57 (1964): [300]; idem, Ptochoprodromos: Einfu sar, Neograeca Medii Aevi (Cologne, 1991).
92
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS
is clear from this passage that Prodromos was interested in the difference between the language of the “cloth seller” and the “wise man”: Oujk ejpainw' ga` r ejgw` tou'to to` me´ ro" to` n Strwmate´ a, eujglwtti´an mh´ pote zhlou'n le´ gonta mhde` rJhma´ twn eujge´ neian, ajrkei'sqai de` mo´ nv tv' aijni´xasqai to` noou´ menon. jAdia´ foro" ga` r a‘n ou”tw kai` oJ blatopw´ lh" ei“h kai` oJ sofo´ ". Ej gw` de` kai` aujto` " mikrou' a‘n h‘ oujde` tou' tuco´ nto" lo´ gou th` n glw'ttan hjxi´wsa, eij gumnai'" tai'" yucai'" diezw'men, to` n oJmo´ zugon tou'ton uJperanaba´ nte" phlo´ n. Ej pei` kai` oJ tou' sw´ mato" ou»to" oJlko´ ", oJ ojrganiko´ ", fhmi´, ajndria´ ", ta` " hJmete´ ra" yuca` " peripe´ plastai, kai` oujk ejxo` n ajme´ sw" ta` tou' noo` " hJmi'n ejmfanisqh'nai kinh´ mata, ouj deute´ ra" oi«mai dei'n ajxiou'n th` n glw'ssan timh'". I do not praise Clement where he claims in the Stromata never to strive for eloquence and nobility of diction but to be satisfied with merely touching upon the sense. For thus there would be no difference between the cloth seller and the wise man. I too would have counted language as of meager or no import if we transcended this clay that is yoked to us and could pass through life with naked souls. But since this burden of the body—I refer to this living statue—has been formed around our souls, and since it is not possible for mental processes to be intimated to us directly, I consider it imperative to rate language as of no secondary importance.3
Alexander Kazhdan and Ann Epstein suggest that within imperial circles, both vernacular and learned forms of language were employed, and not infrequently by the same author according to occasion and context.4 This raises the question: What degree of variation was permitted between learned and vernacular forms within the same text, as is indubitably the case with each of the four Ptochoprodromic Poems? Michael Jeffreys finds it “difficult to accept the picture of generations of Byzantine intellectuals who experiment in popular language yet fail to carry through their experiments to the logical conclusion of a completely vernacular poem. The education of such men was directed entirely to the elimination of mistakes from their writing, towards the preservation of a uniform linguistic level. If they decided to experiment with the vernacular, surely at least one man could have been found in several centuries to impose a similar uniformity on his ˇevcˇenko and Robert Browning have maintained, it is demotic writing?” 5 Yet if, as Ihor S more appropriate to think of the varieties of style in Byzantine literature in terms not of “language” but of “register,” then the issue of “linguistic inconsistencies” is infinitely more complex, especially if we are dealing with “genre poems,” as Kazhdan and Simon Franklin suggest.6 Moreover, with regard to the proliferation of forms, Browning notes that it is probable that there existed “a common tongue in which a great many alternative forms, belonging historically to different dialects, were acceptable. Men from all over the Greek world mingled in Constantinople as they did nowhere else.” 7 Was not Jeffreys parachronistically projecting onto twelfth-century Constantinople our own, modern, and Western, 3 PG 133:1265A–B. The passage is cited in A. P. Kazhdan and S. Franklin, Studies on Byzantine Literature of the Eleventh and Twelfth Centuries (Cambridge, 1983), 111. I have preferred to cite the full text in Greek, with a more literal translation. 4 A. P. Kazhdan and A. Epstein, Change in Byzantine Culture in the Eleventh and Twelfth Centuries (Berkeley–Los Angeles, 1985), 84. 5 M. J. Jeffreys, “The Literary Emergence of Vernacular Greek,” Mosaic 8.4 (1975): 176. 6 ˇevcˇenko, “Levels of Style in Byzantine Prose,” JO ¨ B 31 (1981): 289–312; R. A. Browning, “The LanI. S guage of Byzantine Literature,” Buzantina` kai` Metabuzantina´ , ed. S. Vryonis, vol. 1 (Malibu, 1978), 103; Kazhdan and Franklin, Studies, 91. 7 R. A. Browning, Medieval and Modern Greek, 2d ed. (Cambridge, 1983), 82.
MARGARET ALEXIOU
93
notions of linguistic homogeneity and purity and, in particular, notions of the Greek “demotic,” which go back only to the last decades of the nineteenth century?8 The following passage from poem I of the Ptochoprodromic Poems has been taken by some scholars as proof that such a learned writer as Theodore Prodromos could not have been the author of these poems: the genitive gunaiko´ " does not agree with the qualifying participle proba´ llousa, subject of the main verb profe´ retai.9 Yet the sudden switch of levels is both vivid and effective: having listed the diseases he does not suffer from, the poet now names his actual affliction (an awful wife), then “performs” her complaints in an alliterative nominatival clause: 15
20
25
Ka‘n fai´nwmai ga´ r, de´ spota, gelw'n oJmou' kai` pai´zwn, ajll∆ e“cw po´ non a“peiron kai` qli´yin baruta´ thn, kai` calepo` n ajrrw´ sthma, kai` pa´ qo", ajlla` pa´ qo". Pa´ qo" ajkou´ sa" toigarou'n mh` kh´ lh uJpola´ bh", mhd∆ a“llo ti ceiro´ teron ejk tw'n mustikwte´ rwn, mh` kerata'n to` fanero´ n, mh` tantantotraga´ thn, mh` no´ shma kardiako´ n, mh` periflegmoni´an, mh` skordayo´ n, mhd∆ u”deron, mh` parapneumoni´an, ajlla` maci´mou gunaiko` " pollh` n eujtrapeli´an, problh´ mata proba´ llousa kai` piqanologi´a" kai` to` dokei'n eujlo´ gw" moi profe´ retai plouta´ rcw". 20 tartanotarta´ nhn⭈ Eideneier | 21 kardiako´ nÚ heartburn, not heart disease; see Alexiou and Hendy, forthcoming.
15
20
25
Although I seem, lord, to laugh and play at once, I am oppressed by endless grief and burdensome affliction, by grave indisposition, and suffering—what suffering! Hearing of suffering, do not suppose I have a rupture, or any other of the graver, inward ills, it’s no eyesores, plain to see, nor shivering fever either, no heartburn, nor inflammation of the lung, no gut-rotting shit-face, no dropsy, nor bronchial ills. No, I have a warring wife, whose tongue wags on and on, pugnaciously parading parapets and predictions, redundantly recounting me the rightness of her cause.
“Mistakes” of agreement, particularly between subject and participle, had been in common usage since the second century A.D. (papyrus letters), and are consistent with the gradual disappearance of the declined active participle attested in Romanos the Melodist.10 The linguistic grounds for rejecting Theodore Prodromos as possible author of our four poems are based on false assumptions. We move now to our second area of doubt: Is our poet’s sense of humor too base to be attributed to Theodore Prodromos? Cyril Mango complains that he “tries to be clever See D. Tziovas, The Nationalism of the Demoticists and Its Impact on Their Literary Theory (Amsterdam, 1986). Hesseling and Pernot, Poe`mes Prodromiques; Eideneier, Ptochoprodromos. 10 Papyrus letters: mou' kindineu´ santo" eij" qa´ lassan e“swse eujqe´ w", cited in G. Thomson, The Greek Language (Cambridge, 1966), 46, no. 19.5. Romanos: see K. Mitsakis, The Language of Romanos the Melodist (Munich, 1967), 158–59, no. 306. A comparable lapse of syntax can be found in Prodromos’ Rhodanthe and Dosikles, IV.65 (Bryaxis’ letter to Mystilos): h‘ gou'n Brua´ xhn kata` sou' kinw'n ma´ qe. I owe the last example to P. Roilos. 8 9
94
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS
without being funny”; that the poems contain “too much slapstick,” or are “frequently obsessed with what everyone has to eat,” echoing the earlier strictures of Dirk Christian Hesseling and Hubert Pernot, and Henry Tozer. It is perhaps easier to appreciate another culture’s sense of history, or tragedy, than it is their sense of humor, simply because we take it for granted that what fails to amuse us cannot, by its very nature, be “funny.” 11 Yet, as Mikhail Bakhtin has indicated in his essays on Rabelais, premodern parody differs from its modern literary counterparts precisely in the regenerative nature of imagery drawn from “the human body with its food, drink, defecation, and sexual life.” Two types of imagery converge, Bakhtin argues, especially at the crossroads of medieval parody and early Renaissance realism, the one drawn from the culture of folk humor, the other from the bourgeois concept of the individual.12 Rabelais and his world may seem a far cry indeed from twelfth-century Constantinople, but I hope to show that the Ptochoprodromic Poems share the same features of dialogic interaction and regenerative humor as those discussed by Bakhtin, and in ways that were in important respects independent of, and prior to, the medieval West. As an example of comic humor, or games and play but with a serious purpose, let us examine what I regard as two paired “keystone themes” developed throughout the four poems: pain and disease, death and resurrection, which are counterposed to games and play. The theme of pain and disease is introduced at the beginning of poem I, with reference to po´nos and pa´thos, as we have just seen. Prodromos’ itemization of the afflictions, both external and internal, that do not beset him affords him his first chance for a scatological joke: one thing I do not have, he reassures his addressee and patron, John II, is skordapso´s. This is not “eye disease” (Augenleiden), as rendered by Hans Eideneier, but “gut-rot,” a vernacular form of cordayo´ " (cordh´ ⫹ a”ptw), pa´ qo" tw'n ejnte´ rwn (Hesychios, s.v.). Theophanes Continuator (cf. III.25j) adds the particularly nasty detail that the ailment involved obstruction of the intestines, causing the patient to vomit feces from the mouth.13 No, Prodromos continues, my pa´thos is not skordapso´s, but something much worse—a nagging wife! It is thanks to her, he implies, that he is obliged to vomit so much “shit language” from his mouth; all these ills could be cured if only he were given more money! By the end of poem I, the links between domestic discord and imperial malaise have been established by the implied contrast between Prodromos’ own dire straits and imminent death (unless rescued by his patron’s munificence), and by the comic but curious interlude of his baby son’s fall from “on high,” whence portents descend, with instant “resurrection” thereafter (I.219: tou' pa´ qou" katapau´ santo", tou' bre´ fou" d∆ ajnasta´ nto"). Baby’s resurrection is effected by his wife’s female neighbors, the illicit mandragourai, “sorceresses” (I.211).14 In poems II, III, and IV, Prodromos elaborates, especially in the 11 C. Mango, Byzantium: The Empire of New Rome (New York, 1981), 251; cf. Hesseling and Pernot, Poe`mes Prodromiques, passim; H. J. Tozer, “Byzantine Satire,” JHS 2 (1892): 233–70. 12 M. M. Bakhtin, Tvorchestvo Fransua Rable (1965), trans. Rabelais and His World (Bloomington, Ind., 1984), 18–24. 13 Eideneier, Ptochoprodromos, 122. For analogies to sound shifts from initial k- to sk-, cf. AG ko´ ni", MG sko´ ni… for sch- to sk-, cf. AG scolei'on, MG skolio´ . John Duffy has pointed out to me that both shifts may be assisted by a pun on skw'r, gen. skato´ ". 14 On the significance of portents from “on high,” see A. P. Kazhdan, “Byzantine Hagiography and Sex in the Fifth to Twelfth Centuries,” DOP 44 (1990): 140, who cites the 10th-century vita of St. Irene of Chrysobalanton, BHG 952 (Life of St. Irene, Abbess of Chrysobalanton, ed. J. O. Rosenqvist [Uppsala, 1986], 52–64): Irene, a girl from a Cappadocian family, settled in a convent, but her former suitor found a magician who bewitched
MARGARET ALEXIOU
95
proem, final episode, and epilogue, the theme of pa´thos in relation to his children (II), monastic life (III), and Constantinople’s artisanate (IV), but always returning to his key political complaint, or warning, that his imperial patron will suffer from his divine one unless he gives succor to subordinates. In return, Prodromos offers tina` " politikou` " ajme´ trou" pa´ lin sti´cou", É sunestalme´ nou", pai´zonta", ajll∆ oujk ajnascaitw'nta", É pai´zousi ga` r ge´ ronte", ajlla` swfroneste´ rw" (I.9–11). “Politikos” here carries more than one meaning. Beneath Prodromos’ humor lies a serious purpose; so much for the second in the litany of complaints against his literary style. As to the third issue, the episodic and discursive structure of the four poems, if we lay aside our preference for a linear, ordered narrative and analyze the episodes in relation to their historical and performative contexts, we shall see how much insight they afford across a wide and crucial range of twelfth-century issues in relation to the imperial court: gender and marital relations; household economy and authority; conjugal rights and role reversals; family life and cost of living; monastic inequities and nefarious practices; the low status of the scholar in comparison with the butcher, the baker, and the candlestick maker. But their literary qualities? I have chosen to summarize the episodic structure of poem I as a whole, and to illustrate the last two episodes in detail, because, although reviled by scholars as containing “non-sens indiscutables,” 15 they can be shown to make perfect sense—and afford us with some performative fun—if we are prepared to shed our preconceptions and do a little homework. It is precisely the episodic structure of the poems that permits Prodromos to develop to the full his use of games and play. Poem I is framed by a proem and closing address, as are the other poems, with frequent interventions to the emperor, John II, in the course of the narrated dialogues and events. Prodromos sets the scene cunningly. The tone sounds cringing, but nothing is quite what it seems. He wants to give “equal recompense” (ajntamoibh` n ejxiswme´ nhn) to the “bright benefactions” (lampra` " eujergesi´a", perhaps also meaning “glittering new coins”) he has received from his patron in the past (I.2–3), but he undercuts the implied compliment by offering in return his own ambivalent verses (politikou` " ajme´ trou" sti´cou" . . . sunestalme´ nou", pai´zonta", I.9–10). Politikos here, as before, refers simultaneously to the popular fifteen-syllable verse form (politikos stichos), to matters of state, to the city, and to currency.16 He begs his patron to listen to what he writes (a“kouson a”per gra´ fw), because, although his po´nos is real enough, he can be playful at the same time (gelw'n oJmw'" kai` pai´zwn, I.15), “clowning wisely” with verses that may lack perfume but not spice, thereby drawing attention from the outset to the four poems’ enduring theme of food, but with a subversive edge. What is more, he is no novice, but one who has enjoyed imperial favor her with an attack of lust. The Mother of God appeared to her in a vision, then sent the martyr Anastasia and Basil the Great, who came flying down with a package weighing three pounds and containing magic ´mata were condevices (including two leaden puppets resembling Irene and her suitor). Once these goeˆteu signed to the flames, Irene was liberated from her sexuality. The parallels with the Prodromic text include sexuality denied on earth and magic recuperation sent from the heavens. On the multiple and ambivalent meanings of mandragourai, see P. Koukoules, “ jEtumologika´ ,” Athena 57 (1953): 213; A. Papamichael, Birth and Plant Symbolism: Symbolic and Magical Uses of Plants in Connection with Birth in Modern Greece (Athens, 1975), 33–56. 15 Hesseling and Pernot, Poe`mes Prodromiques, passim; P. Speck, “‘Interpolations et non-sens indiscutables’: Das erste Gedicht der Ptochoprodromos,” Poikila Byzantina 4 (1984): 274–309. 16 See Hendy, Studies, 137, 138, 159, 573, on politikos as opposed to strategikos to imply favors and strife in public and private matters.
96
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS
in the past—a probable reference to an earlier poem addressed to John II by Theodore Prodromos—and now hopes to win it back.17 The core of poem I (lines 26–197) is built around four episodes, each linked by a transitional intervening address to John II, which serves to remind us of the narrating instant, the imperial court. The first two episodes are composed largely of dialogue, giving us not only a taste of Mrs. Prodromos’ lively tongue, but also some rare insights into household economy and authority, conjugal rights, and role reversal. In episode one (I.42–112), Mrs. Prodromos has four complaints. First (I.46–62), he never buys her clothes or jewelry for cash, but brings back instead useless old things from the palace. The items she mentions (dibi´kin, qa´ lassan) are hard to identify precisely, but probably infer precious cloths and textiles given out to favored clients by imperial patrons, of priceless value to any household but not necessarily exchangeable for cash.18 Beneath the humorous surface, Prodromos is telling the emperor how little respect his wife (and others of her upstart ilk) has for the court. He also decries her mercenary demands, while at the same time asking his patron for more money. Second, she utilizes a classic eristic comic topos, contrasting her “high” status with his “base” one—“you are x, I am y,” etc. She suggests that the inequity of their respective social and financial status invalidates their marriage:
Why gaze so into space, can’t you look me in the eye? I was of good household, you wore a soldier’s club; I was of noble stock, you were just a pauper. You are poor old Prodromos, I am from Matzouka. You used to sleep on mats of straw, I slept on a bed; I brought dowry gifts a-plenty, you just errand [-tips]; I brought gold and silverware; you your scrubbing boards, one board for kneading dough, and a great big cooking pot.
In fact, according to Angeliki Laiou, it was perfectly normal for a wife’s dowry to include all such items as are here named, while the general trend from the eleventh to the fourteenth century was for the value of the dowry, in relation to the groom’s portion, to increase from one-half to two-thirds.19 Once again, Prodromos is using his wife’s unreaSee A. Maiuri, “Una nuova poesia di Teodoro Prodromo in greco volgare,” BZ 23 (1914–18): 397–407. ¨randner, Theodoros Prodromos: Historische Gedichte, For parallels in Prodromos’ historical poems, see W. Ho WByzSt 11 (Vienna, 1974), 481–82. 18 On costume, see DOC 4.1: 159–61 (qa´ lassa), 157 (dibiti´sion), and footnotes; on the value of textiles to households, see N. Oikonomides, “The Contents of the Byzantine House from the Eleventh to the Fifteenth Centuries,” DOP 44 (1990): 204–15. 19 A. E. Laiou, Gender, Society and Economic Life in Byzantium (Nottingham, 1992), 203, 237. 17
MARGARET ALEXIOU
97
sonable complaint to upgrade his own remunerations. Her third complaint concerns their house (by implication, her gift on marriage). Once grand, it is now in a state of rack and ruin. A dilapidated house was a portent of ill boding not just for its inmates, but for army, emperor, and state alike.20 This is one of the very few detailed descriptions we have in Byzantine literature of a well-to-do house. Yet the medical metaphors (ajnerra´ yw, perirra´ yh) take us back to the key theme of the poem—disease and death. Fourth, there are the household chores. Here, Mrs. Prodromos’s complaints have a curiously modern ring, rather like “I run three companies, bring up the kids, and manage the household. I also drive the car!” Again, according to Laiou’s evidence, her position of having basement workshops and an estate within the household would have been enviable indeed in the twelfth century, and, still, all she does is complain.21 The transitional section to episode two (I.113–22) suggests the parodic treatment to follow of the twin Byzantine concepts of heroism and warfare, and the ambivalent nature of language and truth, that are central to the Prodromic oeuvre. Critics have ridiculed the apparent “non-sens” of lines 133 through 148, where Prodromos discloses to the emperor an awful incident from the distant past. What is meant by his insistence that he went home from the palace alone, without an armed retinue? And what do we make of his wife’s accusations that he beat her when, according to the text, he was just in a foul— and tipsy—temper at finding nothing to eat:
140
145
150
Ej gw` d∆ wJ" h“mhn nhstiko` " ajpo` to` filopo´ tin, mh` kru´ yw th` n aijti´an mou kai` e“cw polla´ ki" kri´ma, wJsa` n ejmelagco´ lhsa kai` hjgriola´ lhsa´ thn, kai` pa´ lin ta` sunh´ qh moi sumfw´ nw" ejpefw´ nei⭈ –to` ti´ qarrei'", to` ti´" ei«sai, to` ble´ pe ti´na de´ rei"⭈ poi´an uJbri´zei" pro´ sece kai` poi´an ajtima´ zei"⭈ oujk ei«mai sqlabopou'la sou, oujde` misqa´ rnissa´ sou. Pw'" h”plwsa" ejpa´ nw mou… to` pw'" oujk ejnetra´ ph"… ta` brw´ sima ejpeku´ rwsa" kai` ta` pota` wJsau´ tw", ta` pa´ nta ejxestra´ ggisa" kai` ejpoi'ke" me ejrhmh´ trian. “An i“dwsi ta` ojmma´ tia mou pote` tou` " ajderfou´ " mou, kai` ouj pia´ soun kai` ajyidw´ soun se kai` dei´xoun kai` tele´ soun, kai` dh´ sw sou eij" to` n tra´ chlon ta` te´ ssara paidi´a, kai` ba´ lw eij" th` n kardi´an mou ta` go´ nima´ mou ke´ rdh, kai` ejkba´ lw se ejk to` ojspi´tin mou meta` pomph'" mega´ lh", na` poi´sw kai` to` pro´ swpon kai` th` n uJpo´ lhyi´n sou, na` poi´sw th` n koudou'pan sou aujth` n th` n madisme´ nhn.
See Koukoules, Bi´o", 4:294–67; C. Bouras, “Houses in Byzantium,” Delt. JEt. JEll. 11 (1982–83): 1–16. Laiou cites evidence from the 11th and 12th centuries to show, first, that it was normal for aristocratic women to provide full household management (Gender, 186–93) and second, that “within the household, the model occupation for a woman was spinning, weaving and making cloth” (ibid., 242–43). Moreover, with the assistance of her servants (yuca´ ria, line 88), Mrs. Prodromos would have been in the enviable position of producing cloth not merely for household consumption, but for sale in the market to her own advantage, as Laiou notes of the Vlach woman in poem IV, line 249 (ibid., 190–91). Harvey (Economic Expansion, 190–91) testifies to the increase in urban building during the later 12th century, as also to the reconstruction of older mansions to accommodate workshops within domestic residences. Such workshops, situated on the ground floor, were lucrative sources for developed domestic economic production in Constantinople, especially among the families of senior officials; see Bouras, “Houses,” 22. 20 21
98
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS As for me, starving as I was after too much drink, I won’t conceal the cause lest I commit a sin, 140 I flew into a rage and spoke angrily to her. At that she started yelling in her usual fashion: –How dare you? Who are you? Watch out whom you thrash; and –Mind your tongue, take care whom you dishonor; I’m not your Slavic serf girl, not even your paid servant. 145 How dare you lay your hands on me! Are you not ashamed? You’ve polished off the food, finished all the drink, you’ve creamed off everything, I’m broke as some old hermit. If I set eyes one day upon my brothers, just see if they don’t catch you, break your back and kill you off, 150 or else I’ll fasten our four children right around your neck. I’ll keep the profits of my labors entirely to myself, I’ll turn you out of house and home, parade you in disgrace, I’ll ruin that fine face of yours, and your reputation, and fix your ugly, plucked-out mug for good and all!
Once more, Laiou’s legal evidence on coercion and consent in sexual relations helps us to make sense of this episode. First, if the man’s use of armed force can be proved on the occasion of marriage, even where the woman’s prior consent is shown, he will be deemed guilty of abduction. Where other evidence is absent, she can “cry out in the wilderness,” always summoning other (usually female) witnesses to her aid.22 Second, wife beating was by no means uncommon; but a wife was entitled to scream for help in case of severe abuse in order to secure the punishment of the offender. Prodromos is telling the emperor that his wife is determined to prove him guilty, although he didn’t lift a finger against her! What is more, the imperial court is to blame: too much drinking on an empty stomach (and pocket). Viewed in this way, the episode includes splendid racy dialogue, and also implies that the source of domestic troubles lies in the imperial court itself. If episodes one and two have attracted adverse critical comment, episodes three and four have been reviled as unworthy of any decent writer, let alone Theodore Prodromos. Hesseling and Pernot “have never come across such nonsensical buffooneries in appalling comic taste.” 23 But Prodromos, like other comic writers, was capable of more than one type of humor, and merely shifts comic mode from the predominantly verbal humor of the first two episodes to humor of situation and action, commonly known as slapstick, with incidents ranging from horseplay to coarse lavatory humor with strong sexual innuendoes. The predominantly dialogic mode of the first two episodes now turns to narrative, interspersed with interior monologue and dialogue, as Prodromos craftily directs his audience’s attention to his own pitiful state. There is one thread that connects all the seemingly disjointed incidents in the last two episodes with each other, and with the sense and structure of the poem as a whole: the frolics of carnival and entertainment, as performed at the imperial court and among the urban population, where roles are reversed, and where the sacra of imperial ceremony and liturgy might be mocked with impunity. The salient features of Byzantine A. E. Laiou, “Sex, Consent, and Coercion,” in Consent and Coercion to Sex in Ancient and Medieval Societies, ed. A. E. Laiou (Washington, D.C., 1993), 140–42, 158, 181–84 (use of force); 163–67 (woman’s cry of protest regarded as evidence of man’s guilt). 23 Hesseling and Pernot, Poe`mes Prodromiques, 14, 87–89. 22
MARGARET ALEXIOU
99
carnival may be summarized as follows. On certain occasions, including public holidays and saints’ days, it was customary for mock processions to take place, during the course of which participants would dress up (men in women’s costumes, women in men’s, people in animal masks and skins), hurl things at one another, engage in mock battle and races with staffs, spears, and other weapons.24 Such antics, already hinted at in lines 36 through 39, are given full expression in episode three. When Prodromos’ wife threatens to have him beaten up and cast out as a beggar, his first thought is to seize a staff, cast his hat askew (ri´yon to` kamelau´ cin . . . to` kamelau´ kin stra´ bwson, I.165–71), then roll a stone in her direction and run as if to catch her, roaring like a lion with a wild look in his eyes (I.161–71). There follows the hilarious incident in which Prodromos, failing to find a staff, grabs a broomstick from the privy. He and his wife have a fight on either side of the broomstick, which he pokes through a hole in her bedroom door; she loosens her hold just as he is “coming strong.” He ends up on his back:
175
180
185
190
195
W J " d∆ oujde` ra´ bdon ejfeurei'n oJ ta´ la" hjdunh´ qhn, ajpai´rw to` skoupo´ rrabdon gorgo` n ajpo` th` n crei´an, parakalw'n, eujco´ meno", kai` duswpw'n kai` le´ gwn⭈ –Pana´ crante´ mou, kra´ tei thn, ejmpo´ dize, Criste´ mou, mh` pai´xh kontogu´ risma kai` ejpa´ rh to` rabdi´n mou, kai` dw´ sh kai` poih´ sh me` strabo` n para` diabo´ lou. W J " dh` aujth´ , qeo´ stepte, pro` tw'n loipw'n aJpa´ ntwn, kai` to` ywmi`n ejklei´dwse kai` to` krasi`n ejnta´ ma, feu´ gei, lanqa´ nei, kru´ ptetai, kai` klei´sasa th` n qu´ ran, ejka´ qisen ajme´ rimno" kai` ejme` ajfh'ken e“xw. Kratw'n de` to` skoupo´ rrabdon, th` n qu´ ran ajphrxa´ mhn⭈ wJ" d∆ hjgana´ kthsa loipo` n krou´ wn sfodrw'" th` n qu´ ran, euJrw` n ojph` n ejse´ basa t∆ a“kron tou' skouporra´ bdou⭈ ejkei´nh de` phdh´ sasa kai` tou´ tou draxame´ nh ejtau´ rizen ajpe´ swqen, ejgw` de` pa´ lin e“xw⭈ wJ" d∆ e“gnw o”ti du´ namai kai` sterea` th` n su´ rw, cauni´zei to` skoupo´ rrabdon, th` n qu´ ran paranoi´gei, kai` par∆ ejlpi´da kata` gh'" katapesw` n hJplw´ qhn. W J " d∆ ei«den o”ti e“peson, h“rxato tou' gela'n me, ejkbai´nei kai` shkw´ nei me gorgo` n ajpo` tou' pa´ tou, kai` ta´ ca kolakeu´ ousa toiau'ta prosefw´ nei⭈ – Ej ntre´ pou, ku´ ri, na` swqh'"⭈ ejntre´ pou ka‘n ojli´gon, oujk ei«sai cwrikotou´ sikon, oujde` mikro` n nini´tsin⭈ kata´ leiyon th` n du´ namin, th` n perissh` n ajndrei´an, kai` fro´ nei–kalokai´rin e“n – ti´ma tou` " krei´ttona´ " sou, kai` mh` pallikarreu´ esai, mhde` laxofardeu´ h".25 182 hjraxa´ mhn propos. Kapsomenos |196 kalliote´ rin corr. Andriotis; e“n∆ Hesseling et Pernot; e”n∆ propos. Agapitos |197 lazofardeu´ h" corr. Legrand
See ODB, s.v. “carnival”; Koukoules, Bi´o", 3:263–68. On the vexing question of conjugal rights to sexual intercourse, much debated in the later 12th century, see Laiou, “Sex,” 109, 181–90. The sexual innuendoes in this passage have close verbal precedents in ancient comedy and literature. Examples include: ´ bdon (176, 182, 184, 188), “penis”; cf. Archil. 31.1: JUakinqi´nhi me rJa´ bdwi É calepw'" “Erw" rJapi´zwn É 172. rJa ejke´ leue suntroca´ zein; Diog. Laert. 25, 28. ` n qu´ ran (182, 183), “vagina”; cf. Ar. Lys. 310–11: a”yante" ei«t∆ ej" qu´ ran krihdo` n ejmpe´ soimen… É ka‘n 180. th mh` kalou´ ntwn tou` " moclou` " calw'sin aiJ gunai'ke". 24 25
100
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS
Well, as I couldn’t even find a staff, poor wretch, I quickly grabbed the broomstick from the privy, pleading, beseeching, importuning, crying out in misery: 175 –All-immaculate one, control her, stop the bitch, my Christ, or else she’ll turn around and take my stick away, and then use it to beat me more crooked than the Devil!– As for herself, God-crowned one, before what next transpired, she had placed the bread and wine under lock and key. 180 Sneaking furtively away, she closed the door behind her, and, leaving me outside, she just sat there regardless. With the broomstick in my hands, I began to ply her door; and, aroused with fury as I was, I battered hard upon it. Just then I found a little hole, and poked inside my stick end. 185 In quick response she got up and held fast onto my stick. We kept shafting, she from the inside, me from outside just as she knew my potency and I had her firmly grasped, she came, loosens on the broomstick, sets the door ajar, and there I was, quite suddenly, stretched out upon the ground. 190 When she saw I’d fallen down, she started mocking me, as she came to help me from where I was laid down. Feigning love and tender care, here is what she said: –Shame on you, lord, on your life! have you no pride? you are no village tomboy, nor no whining baby! 195 Leave off your potency, your excessive manliness, and have the sense this summer time, give honor to your betters, stop trying to play hero, cease your bellowing howls!
The curious incident of Prodromos’ infant son, who falls from “on high” to be miraculously restored by women mandragou´ rai (I.206–20), can likewise be explained in terms of games and play. “Wonderworkers” (qaumatourgoi´) were frequently called upon to perform at weddings, festivities, and imperial banquets, slitting throats and severing heads and then bringing their victims back to life. In Prodromos’ Rhodanthe and Dosikles, a magician named Satyrion entertains two satraps, Gobryas and Artaxanes, at a banquet. Satyrion slits his own throat with a sword, draws blood, and falls dead to the ground (kai` ne´ ku" eij" gh'n a“qlio" Saturi´wn É kei'tai pro` pa´ ntwn ejklelu´ meno" fre´ na"), until lamented by Artaxanes and touched by Gobryas’ wand at the same time as the words a“nqrwpe . . . ejxana´ sta kai` bi´ou are uttered, at the bidding of the mighty lord Mystilos, who is thereafter praised as the Sun king (IV.214–42). Recalling the dead to life, chief among the miracle acts, is the key theme in this episode, which not only afforded the then-starving Prodromos with the chance to sneak to the cupboard and steal a quick snack, but allows him now, in the telling of the tale, to hint that if street women can save his child, so too can the emperor save his Prodromos. ´ mhn, “get started; practice” (as on a musical instrument), LSJ. 182. ajphrxa ´ wn th` n qu´ ran, “bang at the door”; cf. Ar. Eccl. 316, 989: o”tan ge krou´ sh" th` n ejmh` n prw'ton qu´ ran. 183. krou ` n, “vagina”; cf. Ar. Lys. 720: th` n me´ n ge prw´ thn diale´ gousan th` n ojph´ n. 184. ojph ´ rizen (MG trabw'), “shaft”; tau'ro", “penis”: Ar. Lys. 81. 186. ejtau 188. cauni´zei, “open wide,” “gape.” Used of sexual release: Eust. Makr. Hysm. 3.22–35; of wanton hetaira’s kisses: Ephippos in Kock CAF 2.254.6, paraphrased as ti" kwmiko` " eijpw` n wJ" ejkola´ keusen ouj sumpe´ sasa to` sto´ ma w”sper pole´ mio", ajlla` toi'si strouqi´oi" caunou'sa oJmoi´w". Eust. Thess. 1411.8.
MARGARET ALEXIOU
101
Beneath the humor lurks the hint of the subversive parallel with the Resurrection of Christ.
210
215
220
210
215
220
Tou' gou'n hJli´ou pro` " dusma` " me´ llonto" h“dh kli´nai26 boh´ ti" a“fnw ªgi´netaiº kai` tarach` mega´ lh, e’n kai` ga` r tw'n pai´dwn mou e“pesen ejk tou' u”you", kai` krou'san ka´ tw e“keito w”sper nekro` n aujti´ka⭈ sunh´ cqhsan aiJ gei´tone" wJ" pro` " parhgori´an, aiJ mandragou'rai ma´ lista kai` prwtokourkousou'rai. kai` to´ te a‘" ei«de" qo´ rubon kai` tarach` n mega´ lhn. jAscoloume´ nwn toigarou'n tw'n gunaikw'n kai` pa´ ntwn tw'n sunelqo´ ntwn ejp∆ aujtv', wJ" fqa´ sa" ⫹ei«pon⫹ a“nw, tou' bre´ fou" tv' sumptw´ mati kai` tou' paido` " tv' pa´ qei, kruptw'" ajph'ra to` kleidi´n, kai` h“noixa to` ajrma´ rin⭈ fagw` n eujqu´ " te kai` piw` n kai` koresqei`" ejxai´fnh", ejxh'lqon e“xwqen kajgw` qrhnw'n su` n toi'" eJte´ roi". Tou' pa´ qou" katapau´ santo", tou' bre´ fou" d∆ ajnasta´ nto", ajpecaire´ thsan eujqu` " oiJ sundedramhko´ te" . . . The sun, forsooth, was on the point of setting to the west, when there arises a loud cry, and a great commotion: one of my children, indeed, had fallen from on high! Dashed to the ground, he lay there, good as dead, when up ran neighbor women with their condolences, all those prize peddlars in magic gossip came, then what hubbub and commotion were there for you to see! With womenfolk thus preoccupied, as were all who gathered round, having reached, as I said before, the scene of baby’s accident, the child’s great suffering, I furtively removed the key, and opened up the cupboard. After sudden assuagement of drink and hunger pangs, I went straight out again, lamenting with the others. When his suffering ceased, and baby resurrected, bystanders and onlookers dispersed and took their leave.
In the final episode, Prodromos is constrained by starvation and by exclusion from the family table to dress up as a Slavic mendicant pilgrim, only to be beaten off by his own children, and readmitted at the end by his wife, who thereby realizes her earlier threat to reduce him to a state of destitution. But, he concludes, if even she took pity on me, your best congratulator, so must you, crowned emperor, show me compassion now by giving me more money, just as you wish for succor from Christ:
270
Toiau'ta pe´ ponqa deina´ , krata´ rca stefhfo´ re, para` maci´mou gunaiko` " kai` trisalithri´a", wJ" ei«de me kenw´ tata ejlqo´ nta pro` " to` n oi«kon. ‘An ou«n mh` fqa´ sh me to` so` n fileu´ splagcnon, aujta´ nax, kai` dw´ roi" kai` cari´smasi th` n a“plhston ejmplh´ sh",
26 The phrase is deliberately portentous. The verb kli´nai humorously marks the transition from redundant emphasis on the noun kli´nh in the previous twenty-five lines (four times), as we turn from empty, hungry bed/bench (see Oikonomides, “Byzantine House,” 208–13) to the setting of the sun (and of day, life, empire?). For associations of sunset with old age, see Theodore Prodromos, RD I, line 1; Manass. Chron. 6335 (fqa´ sante" ejpi` dusma` " bi´ou); Arist. Po. 1.1457B (to` ge´ ra" dusmai` bi´ou).
102
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS tre´ mw, ptoou'mai, de´ doika mh` foneuqw' pro` w”ra", kai` ca´ sh" sou to` n Pro´ dromon, to` n ka´ lliston eujce´ thn.
270
Such were my dread sufferings, almighty crowned lord, my sore tribulations caused by a warring wife, when she saw me coming home with empty hands. Unless your loving mercy reaches unto me, sole ruler, unless you satisfy her lust with gifts and favors, I tremble, terror-stricken, lest untimely murder deprive you of your Prodromos, your best congratulator.
Interpreted in the context of games and play, the episodic structure of poem I fulfils Prodromos’ initial claim to “laugh and play” in earnest. By casting himself at the end in the victim’s role as a mendicant outcast, Prodromos assumes an aura of sanctity. POEM III: A DARKER HUMOR? We pass over the shortest of the four poems (II), in which the impecunious members of the householder’s family are reduced to eating their own clothes/estate (with strange and dangerous results, implying heresy and alchemy), and turn now to poem III, addressed to the emperor Manuel I (1143–80), by the novice monk, or “rag-wearer.” He is verbally and otherwise abused, while the hierarchs indulge in all conceivable (and inconceivable) delectations of bathtub, food, and wine. They get every fish and the very best of seafood “laid before them” (kei´mena), and drink from decorated cups of the best Samian and Chiot wines, while we are fed on what they baptize as “holy broth” in dirty claypots. They get a “true baptismal font” of finest fish stew, we just get twenty onionrings, with a few stale crusts and three drops of holy oil thrown into boiling water in a massive old cauldron, green with verdigris. The contents of the hierarchs’ repast, served in what is playfully but subversively termed a “baptismal font,” are described in detail tantamount to a recipe (III.174–94). But the “fish stew” is not as innocent as it may sound, coming as it does right after a full array of four rich courses—broiled, sauced, sweet-and-sour, grilled with spices—of every kind of fish the Black Sea could boast of (including turbot cooked with labrus, tender bream, gray mullet three palms long, and flatfish, or “citharus linguatula”). Prodromos has exploited Constantinople’s rich fish supplies to add to the range of sexual reference to “tasty bits of roasted meat” familiar from Attic comedy. What is more, he wishes a second “Akrites” (like Herakles) could enter their refectory fray and smash their heinous “dishes” into smithereens. Humor in this episode depends on wordplay and bilingual puns. What is aJgiozou'miî It is frequently mentioned in the monastic typika as a fasting “holy broth,” meant to be simple but nourishing; but Prodromos puns on ijozou'mi (“viral swill”). He also puns on the twenty “rings of onions” thrown into boiling water, calling them kalole´ onta" (“good lions,” throughout), which unimaginative editors have emended into kole´ nta" (“rings,” from the Venetian coleta, “ring,” “chain”). The suffix lio´ nta" suggests a champion wrestler or jouster. A few lines later he has “Good Lion” engaging with “Fat Thug” (Contro´ ") in a wrestling match, with clear sexual connotations, as in the carnival games and jousting with his wife of poem I, with the difference that in poem III
MARGARET ALEXIOU
103
the connotations are homosexual; indeed, the filthy cauldron of “viral swill” reminds us of the bathtub in which, as a novice, he was forced to “rub up” the two fat hierarchs (III.107–16). We are back with the theme of games and play with a funny but sinister and obscene twist. Now to our passage, transmitted in differing forms in most manuscripts, but relegated by Hesseling and Pernot to the critical apparatus (325a-u). Here is the version transmitted in manuscript H, probably the oldest witness of poem III: [311 325a b c d e f h i j k l m n o p q r s u
As in I.172–97, the sexual innuendoes have close ancient precedents. Examples include the following: ” ska", f.: a rare word, possibly to be associated with i“skai, “fungus growing on oaks and walnut 325c. u trees” (Latin esca “bait”), used for cautery (LSJ); seal, R. Volk, Byzantina Vindobonensia (Vienna, 1991), 299–311; sturgeon (A. Alexakis, Dumbarton Oaks, hagiography database project). It also suggests ijsca´ ", “dried fig,” and u”ssaka", “little piggies,” both attested for “cunt”: Hippon. 124; Ar. Ach. 802; Lys. 1001. On u”kh and u»" as “unidentified fish,” see D. W. Thompson, A Glossary of Greek Fishes (London, 1947), 272–73; more generally, see also C. F. Tinnefeld, “Zur kulinarischen Qualita¨t byzantinischer Speisefische,” in Collected Papers Dedicated to Kin-ichi Watanabe, Studies in the Mediterranean World, Past and Present 11 (Tokyo, 1988), 155–76. I thank my learned colleagues for their suggestions, but I suspect that Prodromos knew these different meanings too, and was playing upon them all. ´ kia su` n filomh´ lai" i“skai", the word 325k. filomh'le", f.: a kind of fish; cf. Meliteniotes 1425: xifi´ai kai` sua also suggests the mythical Philomela, and the epithet filo´ mhlo", “fond of apples” [?tits]; Doroth. ap. Athen. 7.276. ´ n, m.; cf. poem I.22: see above, p. 94. 325j. skordayo ´ lhkon, n.: this hapax legomenon, attested in a majority of manuscripts, is coined on the analogy 325o. uJdro of the prefix uJdro- ⫹ substantival suffix; cf. uJdro´ garon, “fish sauce cooked in water”; uJdroke´ falon, “water on the head; hydrocephalus”; uJdrokirsokh´ lh, “aneurysm of the vessels of the testicles”; uJdroko´ mion, “gum water”; uJdro´ koilo" (or uJgro´ koilo"), “having moist faeces.” As for the suffix -lhkon, the fem. lhkw´ indicates “penis”; cf. lhka´ w, laika´ zw (lh´ khma), for “sexual intercourse”; Pherekr. (LSJ). By consonantal metathesis, the suffix -ke´ lh" would indicate “courser, cock riding-horse; yacht” ⫽ “vagina”; Ar. Pax. 894. The suffix -lhko" phonetically invokes -luko"; cf. proverb ejk lu´ kou sto´ mato" of “getting something in
104 311 325a b c d e f h i j k l m n o p q r s u
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS [They are eating frogs, we get the “holy broth”] . . . They feed on the best sea bass, and huge [red mullet] too, we get that stinking smokey “holy poison broth”; they get fat fish and seal steaks, truffles, dabs, while we were eating that old what’s-its-name: it does have a crazy name, it’s just a trifle strange and a man gets sizzled up before he hits on it . . . I’m drowned, I’m dead, let the worm cut it out, look, they say “Slow down!” just as you want to come. Come now, shit head, I think you are bewitched! If only mullet, dabs, and tit-bits had found their way down here, then I might hit upon the word, –Behold, a mystery ensues!– Lord, with my blessing, you know its name so well, it’s here upon my tongue tips, –Devil, curses on you!– Enjoy the leaky bag, I just got it in the mouth, so here’s its food and mess, while I was being sizzled. They were quite determined to shove it up me, but I was no village boy, not to puke it out with fire. Eh, how much sore chafing, lord, have I upon my soul, yet do we speak out all things as have been uttered in.
In this passage, Prodromos comes nearest to what Jeffrey Henderson has termed “primary obscenity,” in that it is hard to read “innocently.” 28 Yet even here, the humor depends on double entendres, achieved through wordplay, punning, and above all by the metaphorical associations of food, wine, and sex. And beneath the games and play there lies a deadly serious purpose—to expose the filth, corruption, and abuse experienced by the lowly monks in the Philotheou monastery.29 He also wants to get transferred to another monastery (the Mangana), where Theodore Prodromos actually served. Read in the context of monastic reforms debated during the later twelfth century, poem III touches on issues no less topical than does poem I.30 As to the possible sources of Prodromos’ comic humor, both light and dark, there is evidence to suggest a connection between Prodromos’ use of the vernacular and the twelfth century’s rediscovery of ancient comic tradition. Literary discovery of the vernacular can be linked with the twelfth century’s renewed interest in Aristophanes and the the mouth, praeter spem”; Zen. 3.48, and lascivious connotations of prefix lu´ ko- (as in the case of Lykainion, who initiated young Daphnis in sexual matters, Long. DC III.17). 325p. brw'mo", m.: “stink, noisome smell”; LXX Jb. 6.7, Gal. 7.214. “Viral swill” seems to be none other than shssss. . . . , associated with the mire (borboros) to which the evil shall be damned eternally. Here, Prodromos adapts Aristophanic and Lucianic images of Underworld torments of the damned to a 12th-century context, as is also the case in Timarion, ed. R. Romano (Naples, 1974), chap. 46, 1173–74. 28 J. Henderson, The Maculate Muse: Obscene Language in Attic Comedy, 2d ed. (New York, 1991), 35–41. It is surely relevant to Prodromos’ revival of obscene humor in the 12th century that, as Henderson (ibid., 13) points out, with the exception of the Ionic iambic poets and the cults of Dionysos and Demeter, obscenity is not found elsewhere in the surviving literature of the time. 29 ¨y), was founded ca. The monastery, situated five miles up the Bosporos at Anaplous (modern Arnavutko 1022–53; see R. Janin, La ge´ographie eccle´siastique de l’Empire byzantin, vol. 1, Le sie`ge de Constantinople et le patriarcat œcume´nique, pt. 3, Les eglises et les monaste`res, 2d ed. (Paris, 1969), 494, and idem, Constantinople byzantine: De´veloppement urbain et re´pertoire topographique, 2d ed. (Paris, 1964), 510. 30 See Kazhdan and Epstein, Change, 87–92; P. Magdalino, The Empire of Manuel I Komnenos, 1143–1180 (Cambridge, 1993), 298–99.
MARGARET ALEXIOU
105
comic tradition, as is evident from John Tzetzes, author of voluminous scholia on Aristophanes’ plays. He describes in a letter his fascination, albeit disapproving, with the language of fisherwomen on the Bosporos.31 DATE AND AUTHORSHIP OF THE PTOCHOPRODROMIC POEMS We come now to the final and most controversial question: What evidence is there, for or against, for the dating of the four Ptochoprodromic Poems to the twelfth century and for or against authorship by Theodore Prodromos? Let us consider first the remaining evidence against Prodromos, since we have already dispensed with arguments of language and literary style. According to S. D. Papadimitriou, Paul Magdalino, and others, “incontrovertible evidence” that Prodromos was dead by the 1150s may be found in Mangana Prodromos’s poem 37.32 Nai´, sto´ ma, sta´ xon ÔUmhtto` n ejk glukerw'n ceile´ wn, eijpe` kai` lo´ gon zwtiko` n kai` zh´ sw kai` skirth´ sw, eijpe` kai` zw'sa´ n sou fwnh` n zwh´ n moi corhgou'san. ijdou` tele´ w" h“rghsa, kai` ga` r ejggwnia´ zw 25 kai` klh'ron e“cw patriko` n tou'to to` no´ shma´ mou⭈ tre´ mw kai` th` n ejkme´ trhsin tou' klh´ rou th'" zwh'" mou⭈ ptoei' me ga` r oJ Pro´ dromo", oJ prodramw` n ejkei'no", oJ rJh´ twr oJ peri´fhmo", oJ proteqrulªlºhme´ no", hJ celidw` n hJ mousourgo´ ", hJ lalista´ th glw'tta, 30 mh` to´ pon eJtoima´ zh moi kai` li´qon kai` gwni´an. mononouci` ga` r profwnei' kai` promartu´ retai´ moi kai` rJhtoreu´ ei kai` qanw` n para` nekrw'n keuqmw'ni.33
25
30
Yes, mouth, drip Hymettos’ honey from sweet lips, speak the word of life and I will live and skip, speak in your living voice, bestowing life on me. See, I’ve grown quite idle, keeping to my corner, and my father’s portion is this my sore affliction, while I tremble at the countdown of my lot in life. That Prodromos, the one who ran before, he frightens me, the renowned rhetor, whose fame was bruited as of yore, the music-making swallow, the most loquacious tongue, lest he is keeping me a place, a stone and corner. It’s as if he’s calling me ahead, bearing prior witness, rehearsing speeches, albeit dead, in the nooks of the departed.
It is true that the poem refers to Prodromos’ death, but, rather than postulate that it refers to Prodromos’ father, as do Kazhdan and Franklin, I suggest that it is just another Prodromic game with death and rebirth, as so often played out in the course of our four Ptochoprodromic Poems and in Theodore Prodromos’ known works.34 We may compare John Tzetzes, Epistulae 57, ed. P. A. M. Leone (Leipzig, 1972), 79–84. See S. Papadimitriou, “ JO Pro´ dromo" tou' Markianou' kw´ diko" XI.22,” VizVrem 10 (1903): 102–63; Magdalino, Manuel I, 440. 33 Theodori Prodromi, De Manganis, poem X, lines 21–32, ed. S. Bernardinello, SBN 4 (1972): 71. 34 Kazhdan and Franklin, Studies, 88–89. For another poem by the Mangana Prodromos, who claims to be both “dead and alive,” see ed. Bernardinello, I, 45–57, 70–71; cf. I, 6–9; VI, 211–17; VII, 1–15; X, 21–54 for similar ambivalence on allegorical references to language and animals. 31 32
106
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS
the poem, composed by Theodore Prodromos in Homeric (and heroic) hexameters and in classical style, which resembles in detail our first Ptochoprodromic poem, on disease and sickness, except that it is composed in archaizing language and meter, appropriately modeled on Gregory of Nazianzos: Tou' aujtou' hJrwi¨ka´ Nou'se, ti´ daimoni´h me dama´ xao… ti´pte´ me, deilh´ , w”ste ra´ ko" ri´knwsa" ajp∆ ojsteo´ fin de´ te pa´ nta muelo` n ejkmu´ zhsa", ajta´ r ge cala´ sao neu'ra, sarki´a d∆ ejxeda´ i¨xa" ijd∆ e“gkata pa´ nta la´ fuxa" 5 cei´lesi peinale´ oisin oJmoi´ia fwla´ sin a“rktoi" hje` kakoi'" dakte´ toisin, ejcidna´ sin hje` le´ ousi… zwo` " ejgw` neku´ essin, ajta` r zwoi'" ne´ ku" eijmi´, ajmfi´bion de´ m∆ e“qhke brotoi'" mero´ pessin Ej rinnu´ "⭈ ou“te li´hn bio´ ousi mete´ ssomai ajmf∆ ajne´ ressin, 10 ou“te li´hn fqime´ noisi, me´ shn d∆ ejpite´ mnw ajtarpo´ n. hje´ ra me` n pnei´w kai` hjeli´ou ble´ pw aujgh` n kai` lale´ w paqe´ wn t∆ aijsqa´ nomai o”ssa me kentei'⭈ toi'sin ejgw` mou´ noisin ejni` zwoi'si metreu'mai, t∆ a“lla de` pa´ nta ne´ kussin eji´skomai.35 Of the same, heroic verses Blight, why hast thy noxious might so smitten me? Cad, why hast from my bones stripped all to tatters like a rag, sucked forth my brain sap, softened sinews too, feasted upon my flesh, gulped down all the guts 5 with famished lips like unto lurking bears and foulest fanged monsters, vipers and lions? Alive among the dead, yet am I dead among the living, for ’tis a Fury set me to dwell amphibian midst mortal wights, not long shall I consort among the living, nor long among the dead, I cut a middle path. 10 Yes, I breathe the air, see the rising of the sun, I can speak my woes, feel what pricks me. Thus, and only thus, am I counted midst the quick, for the rest, I am to be likened to the dead.
There is no good reason to doubt either the attribution in almost all the manuscripts of the four poems to Prodromos (or Ptochoprodromos) or their dedication to John II (poem I), to an unnamed sebastokrator (poem II), and to Manuel I (poems III and IV). That would date them to some time between the 1140s and the 1170s, when Theodore Prodromos was arguably still alive. Details in the proem to poem IV in some manuscripts suggest the 1180s.36 By the mid-1100s, Prodromos’ poems were well known, as is made clear from a letter addressed to Theodore Prodromos by Michael Italikos: O J gou'n parw` n ouJtosi` papa'" Micah` l ple´ on aje´ ra" ajnapnei' tou` " lo´ gou" tou` " sou´ ", pa´ nta pezo` n lo´ gon, pa'n 35 See “Theodoros Prodromos. Historisches Gedicht LXXXVIII,” ed. M. Tziatzi-Papagianni, BZ 86/87 (1993–94): 363–82, lines 1–15. My thanks to Agapitos and Roilos. 36 For a careful review of Prodromos’ possible dates, see Kazhdan and Franklin, Studies, 87–104. On datable elements in poem IV (proem in manuscripts CSA), see J. Grosdidier de Matons, “Courants archaisants et ´ tudes byzantines (Athens, 1976), populaires dans la langue et litte´rature,” in Actes du XVe Congre`s international d’E 3–10, esp. 7, on excellence of critical apparatus in Hesseling and Pernot.
MARGARET ALEXIOU
107
ijambei'on ejpi` sto´ mato" e“cwn.37 Passages from Theodore Prodromos’ “Cat and Mouse War” can also be used to explicate the problematic proem to poem III (lines 8–17), where the narrator likens himself to the proverbial Ant, the rhetors and philosophers to Lions: Kai` qau´ mason tou' mu´ rmhko" th` n thlikau´ thn to´ lman, pw'" o”lw" e“xw ge´ gone th'" tou´ tou muwxi´a" 10 kai` tre´ cein i“sw" w”rmhse toi'" ijscuroi'" qhri´oi", ajkolouqw'n toi'" i“cnesin ajfo´ bw" tw'n leo´ ntwn, th` n tw'n ojnu´ cwn du´ namin posw'" mh` kekthme´ no". Ej me` ga` r sko´ pei mu´ rmhka, de´ spota stefhfo´ re, kata` tw'n po´ nwn th` n ijscu` n kai` th` n ajkthmosu´ nhn, 15 le´ onta" de` tou` " rJh´ tora" meta` tw'n filoso´ fwn, oi”tine" eijsi do´ kimoi stici´zein te kai` gra´ fein kai` suggrafa` " basilika´ ", nikhtika` " ejkpla´ ttein. Just marvel then at such daring as the ant’s, how he managed to get outside his mouse hole 10 and darted at a running pace against the fierce wild beasts, following fiercely in the tracks of lions, although in no way furnished by the power of claws. Think, then, upon me as an ant, O crowned lord, as to the hard spareness and poverty of travails, 15 and upon the rhetors and philosophers as lions, for verily are they esteemed in versifying and writing, and in composing heroic victory pieces.
But why should an ant have a mouse hole? Perhaps because muwxi´a suggests any kind of underground nook or cranny in which refuge from the high and mighty can be sought. In “Cat and Mouse War,” it is Tyrokleptes (“Cheese-snatcher”) who urges escape from the mouse holes upon Kreillos (“Fleshy,” or chief mouse) as the only path to freedom, not just access to food. In all these poems, hunger, deprivation, and death are rhetorical tropes that explore new paths to power and freedom, not least by means of varied linguistic registers and daring enjambment: Krei?llo"
Tupokle´ pth"
ajll∆ oijktro´ tatoi kai` fo´ bou peplhsme´ noi bi´on skoteino` n ajqli´w" muwxi´a" zw'men, kaqw´ sper oiJ pefulakisme´ noi . . . ka‘n mh` qe´ lwmen, ejsme` n ejn muwxi´ai". Eij ga` r proelqei'n, wJ" le´ gei", qarrale´ w" tolmh´ somen bai´nonte" ajsce´ tv dro´ mv.38
37 Michael Italikos, letter to Prodromos, from Philippopolis, after 1143, ed. P. Gautier, Michael Italikos, Lettres et discours, AOC 14 (Paris, 1972), 64. The text was first published by R. A. Browning, “Unpublished Correspondence between Michael Italicus, Archbishop of Philippopolis, and Theodore Prodromos,” BBulg 1 (1962): 283–86. See further, Browning, Studies on Byzantine History, Literature, and Education, Collected Studies 59 (London, 1977), art. 6. 38 Theodore Prodromos, “Katomyomachia,” lines 6–8, 14–16, ed. H. Hunger, Der byzantinische Katz-Ma¨useKrieg (Graz-Vienna, 1968), 80–82. There are further parallels between Ptochoprodromos and Katomyomachia, but in reverse, which lend support to my suggestion of a connection between the two texts: at Kat. 304, cat rushed at (w”rmhse) Kreillos’ son and killed him; at Ptochoprodromos III.11–12, ant rushes against (same verb); at Kat. 30, cat seeks out mice just as dogs “track down” (ijcnhlatou'si) hares; at Ptochoprodromos III.11, ant will follow in tracks (i“cnesi) of lions. Lion imagery, especially in the context of chasing wild beasts, is commonly used by Mangana Prodromos for Manuel I’s forays against the Turks (XXV, ed. Bernardinello, lines 58–61: campaign of 1146). For relevant Aesopean fables relating to the ant and lion, ant and beetle,
108
GAMES AND PLAY IN THE PTOCHOPRODROMIC POEMS
Katomyomachia But most piteously and filled with fear do we in mouse holes wretchedly a dark life live, as though we were imprisoned . . . Cheese-snatcher: Yes, we are in mouse holes, willy-nilly. If, then, we are to come out, courageously let us dare, treading in unchecked track. Fleshy:
Most remarkably, and to the best of my knowledge not previously noted with reference to the date and authorship of our four poems, is the evidence of Gregory Pardos. Pardos, bishop of Corinth, includes in his list of recommended reading (compiled before 1156) “Ptochoprodromos” and Nikolaos Kallikles among his contemporaries: Ej pei` ou«n kai` ta` ijambei'a logografi´a ti´" ejstin eu“ruqmo", zhlou´ sqw soi kai` to` ejnqumhmatiko` n ejn aujtoi'"⭈ e“cei" ajrce´ tupon to` n Pisi´dhn, newte´ rou" to` n Kalliklh'n, to` n Ptwcopro´ dromon kai` ei“ ti" toiou'to"⭈ ejn toi'" palaioi'" to` n Qeolo´ gon, to` n Sofoklh'n, ejkto` " tw'n poihtikw'n ijdiwma´ twn aujtou', ta` eujfrade´ stata tou' Luko´ frono" kai` ei“ ti toiou'ton.39 Since, then, iambics are a rhythmic art of writing, let them be emulated by you along with the memorable quality in them. You have as model Pisides, more recently Kallikles, Ptochoprodromos, and so on. Among the older ones, the Theologian, Sophocles, apart from his poetic idioms, the most eloquent writings of Lykophron and so on.
Among older writers, Pardos recommends the letters of Alkiphron, genre exercises composed probably between the third and the fourth century A.D., in the form of fictitious letters exchanged among farmers, fisherfolk, parasites, and courtesans in Alkiphron’s own re-created Attic dialect of the fourth century B.C.40 Pardos was the author of one of the first Byzantine treatises to address, with reference to Alkiphron, the use of dialect in ancient literature, a question discussed elsewhere by Prodromos.41 There is therefore a textual thread that links the twelfth-century creative rehandling of ancient genres with the use of the vernacular, and with our poems, coinciding with the twelfth-century revival of ancient comic, dialogic, novelistic, and satirical genres (Plato, the novels, Lucian). The lexical parallels between “Ptochoprodromos,” Aristophanes, and Alkiphron should be apparent from the parallels cited above and are all the more convincing because so many words in the Ptochoprodromic vocabulary are either very rare or new coinages based on words in older medical treatises.42 ant and dove, lion and mouse, see B. E. Perry, Aesiopica, vol. 1 (Chicago, 1952), 365, 379, 385–86. A final parallel: at Kat. 366–78, cat is struck stone dead by a block of ancient wood, fallen from “on high” (xu´ lon katelqo` n th'" uJperta´ th" ste´ gh", Ptochoprodromos I.207). We may not conclude that the concurrence of passages infers a single author, but we may infer that the author of our four vernacular poems was not unacquainted with learned texts, both Byzantine and ancient. 39 Peri` sunta´ xew" tou' lo´ gou, Gregorio Pardos, ed. A. Kominis (Rome-Athens, 1960), 129, lines 100–105. 40 Ibid., 128. On slender evidence, V. Becares dates Gregory to the reign of Leo VI: “Ein unbekanntes Werk des Gregorios von Korinth und seine Lebenzeit,” BZ 81 (1988): 247–48. 41 “On dialects,” ed. G. H. Schaeffer (Leipzig, 1811). For disdainful reference to Alkiphron and Pardos, see N. G. Wilson, Scribes and Scholars (London, 1983), 184–90. 42 I am currently compiling a glossary for the bilingual edition under preparation by myself and Michael Hendy. Before checking each item in the TLG, I am hesitant to pronounce on the number of actual hapax legomena, but unusual words used with significant allusion to specific contexts include ajnamanqa´ nw, III.7 (verify an oracle or statement with reference to military victory, Hdt. 9.101); ajnarra´ ptw, perirra´ ptw, I.85–86 (stitch up, repair, stitch all around, as after a lifesaving operation, Hierakos, ed. Hercher 484.22, Poll. 7.84). For a fascinating and learned analysis of 12th-century schedographiai, some composed by Theodore Pro-
MARGARET ALEXIOU
109
CONCLUSIONS? What conclusions can be drawn? Byzantinists have been all too ready to treat our poems as “non-sens indiscutables,” simply because they are in “mixed-up Greek.” They have also been reluctant to accept that vernacular texts, from the twelfth century on, can and should be treated with the same degree of seriousness as texts in the high style. Textual emendations and conflation of different manuscripts have been arbitrary. Yet, if Theodore Prodromos was the author of our four poems, as a mounting body of evidence suggests, the twelfth century provides the literary, cultural, and linguistic starting point for “modern” Greek, at the same time as “ancient” texts were rediscovered, edited, and performed. The twelfth century, as Michael Hendy, Alan Harvey, and Magdalino have shown from socioeconomic and cultural perspectives, was not one of decline, rather one of bewildering yet productive social diversification. Prodromos in the four vernacular poems spells out a timely if complex message for imperial rulers: they must pay serious attention to games and play in low-style language or else they will fall, as did indeed Constantinople to the Latins in 1204. Such is the wealth and specificity of detail afforded by the four poems here that we may be certain that they were not composed after that date, although they may have been revised by later scribes. The poems’ literary and performative richness invites comparisons, directly or indirectly, with comic topoi of ancient literature and modern popular traditions, so long as we are careful to take account of historical and performative contexts. If modern Greek literature has its roots in the twelfth century, the twelfth century had at the same time revitalized Byzantine roots in the Greek, Roman, and other pasts, not least through the ploys of performance—games and play in the diverse spoken languages of the time. A few final, yet unanswerable, questions remain: How may the poems have been performed at the Comnene court? Were they written, recited, or both, as the proem to poem I suggests? Did their performance include dialogue, mime, and “horseplay”? If we may answer the last question with reference to Theodore Prodromos’ Rhodanthe and Dosikles (IV.214–42), we may infer that imperial rulers then, as in mythical times, enjoyed a little slapstick and coarse humor, even if they were sometimes themselves the indirect targets.43 Harvard University dromos with use of the vernacular and evidence of diglossic puns and word play, see I. D. Polemis, “Problh´ mata th'" buzantinh'" scedografi´a",” JEllhnika´ 45 (1995): 277–302. 43 Byzantinists assure us that “there is no evidence” about how such texts as Prodromos’ four delightful poems may have been performed, and that during the 12th century the rhetorical theater excelled in literary and verbal humor alone. However, a careful reading of all texts attributed to Theodore Prodromos, and to the four 12th-century novels, suggests that low-life humor and slapstick comedy played a role.
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Dreams and the Spatial Aesthetics of Narrative Presentation in Livistros and Rhodamne PANAGIOTIS A. AGAPITOS In memory of Ole L. Smith (1943–95) You only live twice, or so it seems: one life for yourself and one for your dreams. –Leslie Bricusse
ritical interpretation is to a substantial degree the definition of the internal operative principles governing the composition of a text or an image.1 In cultures where rhetoric and mimesis play an overwhelming role in the production and reception of texts and images, the critic must unavoidably concentrate on matters concerning imagery, style, language, structure, and so on.2 This applies especially to longer, literary or visual, narrative compositions that arguably operate within the expectations of the reader or the viewer.3 Sustaining, rejecting, inverting, subverting, or even canceling the schemata of conventional composition in narrative are essential parts of the artist’s game with his clientele, particularly in the opening sequences of a narrative, where the artist needs to impress and captivate his audience in order to keep them interested in his work.4 In this sense, the Komnenian novels and the Palaiologan romances5 are excellent cases for studying the narrative function of rhetorical patterns. More specifically, the analysis of space provides the opportunity to investigate how the narrative presents itself at its most visual moments and how it establishes its own aesthetic concepts.
C
R. Ingarden, Das literarische Kunstwerk, 4th ed. (Tu ¨ bingen, 1972), 25–196, and H.-G. Gadamer, Wahrheit ¨ge einer philosophischen Hermeneutik, 5th ed. (Tu und Methode: Grundzu ¨ bingen, 1985), 107–74. 2 See, for example, the fruitful comparative analysis of text and image in H. Maguire, Art and Eloquence in Byzantium (Princeton, N.J., 1981). 3 On the interaction between the author and the reader as a governing principle, see W. Iser, Der implizite Leser: Kommunikationsformen des Romans von Bunyan bis Beckett, 2d ed. (Munich, 1979), and idem, Der Akt des Lesens: Theorie ¨asthetischer Wirkung (Munich, 1976). For an application of Iser’s method to the ancient novels, see, indicatively, J. J. Winkler, Auctor and Actor: A Narratological Reading of Apuleius’ The Golden Ass (Berkeley, Calif., 1984), and S. Bartsch, Decoding the Ancient Novel: The Reader and the Role of Description in Heliodorus and Achilles Tatius (Princeton, N.J., 1989). 4 See, for example, the intelligent analysis of Heliodoros by J. J. Winkler, “The Mendacity of Kalasiris and the Narrative Strategy of Heliodoros’ Aithiopika,” YCS 27 (1982): 93–158. 5 The terms novel and romance are used here in order to distinguish the 12th-century works, which feign a “bourgeois antique” setting, from the later texts, which are placed in a “contemporary aristocratic” environment. 1
112
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
The present article examines, from a structural and a spatial point of view, the opening dream sequence in Livistros and Rhodamne, a vernacular romance in approximately four thousand five hundred fifteen-syllable verses.6 The poem was probably composed in the middle of the thirteenth century at the court of the Laskarid emperors in Nicaea.7 Should this date be correct, Livistros and Rhodamne would be the earliest of the seven surviving late Byzantine love romances, and thus constitutes a bridge between the learned novels of the twelfth century and the other vernacular poems of the fourteenth century.
I The beginning of Livistros and Rhodamne is governed by a peculiar structural innovation, which, to my knowledge, has so far remained unnoticed. The anonymous poet has combined two traditions of opening a novelistic narrative. It may be useful to start this analysis with a brief summary of the romance’s opening sequences (N 1–560).8 I (N 1–24). Prince Klitovon invites Queen Myrtane and the court of the Armenian kingdom of Litavia to listen to a wondrous story about a man who was inexperienced in matters of the world, fell in love, and suffered thereafter. The reader, alerted through
6 The romance (hereafter referred to as L&R) survives in three different versions (a, E, V), of which the oldest (a) is transmitted by three manuscripts (N, P, S). My analysis is based on version a, which must have been circulating by the late 14th century, and which is closest to the lost original (on the whole question, see briefly P. A. Agapitos, “Libistros und Rhodamne: Vorla¨ufiges zu einer kritischen Ausgabe der Version a,” ¨ B 42 [1992]: 191–208). For a critical text of the dream sequence, see the Appendix. An English translation JO of the romance has been recently published, in Three Medieval Greek Romances: Velthandros and Chrysantza, Kallimachos and Chrysorroi, Livistros and Rodamni, trans. G. Betts, The Garland Library of Medieval Literature, ser. B, 98 (New York–London, 1995). 7 The date and the place of composition of the romance have been a matter of great dispute. For an overview of the debate and the proposal to date the romance to the mid-13th century, see P. A. Agapitos, “ JH cronologikh` ajkolouqi´a tw'n muqistorhma´ twn Kalli´maco", Be´ lqandro" kai` Li´bistro",” in Origini della letteratura neogreca, ed. N. M. Panagiotakis, Biblioqh´ kh tou' JEllhnikou' jInstitou´ tou Buzantinw'n kai` Metabuzantinw'n Spoudw'n th'" Beneti´a" 15 (Venice, 1993), 2:197–234. See also the reluctant acceptance of this proposal by R. Beaton, The Medieval Greek Romance, 2d ed. (London–New York, 1996), 219–20. Be this as it may, at least, a definite terminus ante quem has now been furnished by M. Manoussacas, “Le terminus ante quem pour la ¨ B 44 (1994): 297–306, since Leonardos Dellaportas, who composition du roman Libistros et Rhodamne´,” JO composed his poems between 1403 and 1411, quotes extensively from the romance; see also M. Manoussakas, Leona´ rdou Ntellapo´ rta poih´ mata (1403/1411): “Ekdosh kritikh´ , eijsagwgh´ , sco´ lia kai` euJreth´ ria (Athens, 1995), 74–82. On the basis of this evidence, a dating in the early 15th century, as proposed by D. Michailidis, “Ne´ e" cronologh´ sei" mesaiwnikw'n dhmwdw'n keime´ nwn,” in Origini della letteratura neogreca, ed. Panagiotakis (same as above), 2:148–55 (and supported by G. Kechagioglou, review of Origini della letteratura neogreca, Hellenika 44 [1994]: 530), is ruled out; see also the cautionary remarks by N. M. Panagiotakes, “The Italian Background of Early Cretan Literature,” DOP 49 (1995): 282 n. 4. 8 Summaries of L&R are provided by H.-G. Beck, Geschichte der byzantinischen Volksliteratur, HAW 12.2.3 (Munich, 1971), 122–23, and Beaton, Medieval Greek Romance, 114–16. Unfortunately, both summaries are too brief and in some cases contain errors; see P. A. Agapitos and O. L. Smith, The Study of Medieval Greek Romance: A Reassessment of Recent Work, Opuscula graecolatina 33 (Copenhagen, 1992), 62. It should be pointed out here that, though Beaton has published a revised version of his study, only minor errors have been removed from the main text. Therefore, the methodological criticisms of Agapitos and Smith stand as before.
PANAGIOTIS A. AGAPITOS
113
this signal, which evokes the tradition of Achilleus Tatios and Eustathios Makrembolites,9 expects Klitovon to go on with his personal tale.10 II (N 25–100). The reader is immediately thwarted. Klitovon inserts an internal title to his story (N 25–26: Loipo` n kai` th` n ajfh´ ghsin a“rxomai th'" ajga´ ph" | Libi´strou tou' polupaqou'" kai` ko´ rh" th'" Roda´ mnh")11 and reveals himself as the romance’s secondary male character. He then embarks on an in medias res narration, now picking up the tradition of Heliodoros and Theodore Prodromos.12 Klitovon’s narration runs as follows. He sees a young man wandering along a narrow path that crosses a beautiful meadow; the man is a warrior, crying and sighing in despair (N 27–43). Klitovon remarks that he also was chased away from his country because of a love affair.13 He then tries to make friends with the warrior and prompts him to tell his story (N 44–67). After some hesitation, the stranger is convinced to tell Klitovon his sorrowful tale; they swear friendship to each other (N 68–91). The reader must accommodate himself to meet the new situation, expecting some introductory action that will now truly lead to the story of the secondary character, this being the conventional pattern in the older novels.14 However, the author surprises the reader again by omitting any such action. The narrative changes direction (N 77–100), since the unknown warrior gives a brief overview of his present desperate situation (N 93–99) and then in his turn inserts an internal title to his story (N 100), thus introducing his ab ovo narration (N 100a ff). All these narrative shifts occur within the first hundred verses of the romance.15 By the time the young warrior has embarked on his story as a first-person narrator, the On the narrative frame in Tatios (dialogic setting between a first and a second narrator) and Makrembolites’ subversive use of it (switch to an epistolary dialogue), see P. A. Agapitos, Narrative Structure in the Byzantine Vernacular Romances: A Textual and Literary Study of Kallimachos, Belthandros and Libistros, MiscByzMonac 34 (Munich, 1991), 131–32. 10 The fact that Klitovon is the main narrator is established by the two rubrics framing the romance’s prologue (N 1–2: Sti´coi polu` ejrwtikoi´, ajfh´ ghsi" Libi´strou , | pw'" oJ fi´lo" Klitobw` n dihgei'tai th'" Murta´ nh"; and N 26a: “Hrxato th` n ajfh´ ghsin oJ Klitobw` n oJ fi´lo"). This situation shows that the rubrics, contrary to recent opinion, were part of the original composition, whatever their present state of transmission; on this matter, see P. A. Agapitos and O. L. Smith, “Scribes and Manuscripts of Byzantine Vernacular Romances: Palaeographical Facts and Editorial Implications,” Hellenika 44 (1994): 66–71. Their proposal has been accepted by C. Cupane, Romanzi cavallerreschi bizantini (Turin, 1995), who reintroduced in her text of the Naples Achilleid the rubrics omitted by D. C. Hesseling in his edition, L’Achilleide byzantine (Amsterdam, 1919). 11 On this device, see Agapitos, Narrative Structure, 79. 12 On the openings of Heliodoros, Prodromos, and Eugeneianos, see P. A. Agapitos, “Narrative, Rhetoric and Drama Rediscovered: Scholars and Poets in Byzantium Interpret Heliodorus,” in Studies in Heliodorus, ed. R. Hunter, Cambridge Philological Society, suppl. 21 (Cambridge, 1998), 125–56. 13 At a much later point in the romance (S 1467–1520), Klitovon will tell his own story. The reader will then discover that the young prince was in fact the lover of Myrtane, who was, however, married to a Persian nobleman. It was because of this adulterous relation that Klitovon finally had to flee from Litavia. This more “down-to-earth” relation between Klitovon and Myrtane is typical of the behavior of the secondary couples in the romances. 14 In Heliodoros, Prodromos, and Eugeneianos, the opening sequences are structured through a schema of “introductory action—story of a secondary character—beginning of the main character’s story.” 15 This is a further subversion of the conventional pattern in the older novels. In the Aithiopika it takes the reader quite some time until he understands at what point in the plot he actually is (Kalasiris’s narration begins at 2.23, some 75 pages after the novel’s famous opening scene). In the two Byzantine novels, the waiting time has been considerably reduced (25 pages in Eugeneianos and 16 pages in Prodromos), but a sense of suspense is still present. 9
114
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
reader has lost certainty about how the narrative will progress, while the poet has successfully blended the two novelistic traditions in order to present his “strangely made tale of love” (N 17: xenoca´ ragon ajfh´ ghma ajga´ ph"). The possibility that the poet of Livistros and Rhodamne had a reading familiarity with, or at least knew of, these two traditions represented by Tatios and Heliodoros and developed by the twelfth-century authors is not as far-fetched as it seems. Besides the quite obvious literary relations between Livistros and Rhodamne and the twelfth-century works,16 it is worth noting that nine of the most important manuscripts preserving the ancient and the medieval novels were written in the thirteenth century,17 and in two cases an ancient and a Byzantine novel were transmitted side by side.18 Moreover, one of these manuscripts, the Laurentianus Conventi Soppressi 627, also preserves the surviving letters of the emperor Theodore II Laskaris (1254–58) and his teacher Nikephoros Blemmydes (1198–1269). The codex is thus related to the Nicaean court,19 exactly the milieu where Livistros and Rhodamne was in all probability produced.20 III (N 100a–118). The young warrior reveals that he is Livistros, ruler of the Latin kingdom of Livandros (N 100a–103).21 As a youth he did not know what love was (N 104–14) and, in fact, scorned anyone who admitted having fallen in love (N 115–18).22 IV (N 119–85). One day he goes with his companions on a hunting expedition (N 119–26); toward dusk he kills with his arrow a male turtledove, and its mate commits suicide by falling from the sky onto a rock (N 127–36). Livistros is profoundly shocked by this incident and asks a close relative of his for advice (N 137–39). In a complex speech, the Relative discloses to Livistros the power of Eros (N 140–77). Livistros returns to his abode with disturbed emotions (N 178–85). See Agapitos, “Cronologikh` ajkolouqi´a,” 101–17. Ven. Marc. gr. 409, early 13th century (Heliodoros, Tatios); Ven. Marc. gr. 410, first half of 13th century (Heliodoros); Ven. Marc. gr. 412, 13th century (Eugeneianos); Vat. gr. 114, 13th century (Makrembolites, Tatios); Vat. gr. 121, 13th century (Prodromos); Vat. Barb. gr. 29, 13th century (Makrembolites); Oxon. Baroc. 131, ca. 1250–80 (Makrembolites); Laur. conv. soppr. 627, ca. 1270–85 (Longos, Tatios, Chariton, Xenophon); Vat. gr. 1390, late 13th–early 14th century (Heliodoros, Makrembolites). 18 For the Heliodoran and Tatian manuscripts, see briefly the respective introductions in He´liodore: Les ´ thiopiques (The´age`ne et Charicle´e), ed. R. M. Rattenbury and T. W. Lumb (Paris, 1934–40), 1:xxiv–xlvii, and E Achilles Tatius: Leucippe and Clitophon, ed. E. Vilborg, Studia Graeca et Latina Gothoburgensia 1 (Stockholm, 1955), xvii–xxxii. On the textual history of the three Komnenian novels in question, see M. T. Cottone, “La tradizione manoscritta del romanzo di Teodoro Prodromo,” in Miscellanea (Padua, 1979), 2:9–34; A. C. Palau, “La tradition manuscrite d’Eustathe Makrembolite`s,” Revue d’histoire des textes 10 (1980): 75–112; F. Conca, “Per una nuova edizione critica del romanzo di Niceta Eugeniano: Collazione dei codici Vat. Urb. gr. 134 e Laur. Acquisti e Doni 341,” in Graeco-latina Mediolanensia, Quaderni di Acme 5 (Milan, 1985), 161–205. 19 On Laskarid manuscript production, see briefly N. G. Wilson, “Nicaean and Palaiologan Hands: Introduction to a Discussion,” in La pale´ographie grecque et byzantine, Colloques internationaux du Centre National de la Recherche Scientifique 559 (Paris, 1977), 263–67. 20 A similar connection seems to be reflected in the Oxon. Baroc. 131, on which see N. G. Wilson, “A ¨ B 27 (1978): 157–79. Byzantine Miscellany: MS. Barocci 131 Described,” JO 21 On the function of the romance’s Latin color in accentuating a “historical-exotic” setting, see Agapitos, “Cronologikh` ajkolouqi´a,” 109–11. 22 Livistros suffers from a syndrome typical of medieval romance characters. Though a highly sexual being, he consciously negates his own erotic nature; Rhodamne suffers from it as well. This attitude to love is described in the romance as “haughtiness” (N 242: to` ajge´ rwcon [Livistros]; S 208: to` hjphrme´ non [Rhodamne]). On this symptom as suffered by the protagonist couple in the Achilleid, see O. L. Smith, “Some Features of Structure and Narrative in the Byzantine Achilleid,” Hellenika 42 (1991–92): 75–94, in particular 85–87. 16 17
PANAGIOTIS A. AGAPITOS
115
V (N 186–270).23 Night falls, Livistros goes to sleep, and a dream comes to him (N 176–89a). While riding across a beautiful meadow he is arrested by a group of winged and armed beings (N 190–223). As he is led, bound by the neck, to the palace of Eros, one of these warrior cupids gives Livistros sound advice on how to entreat the angry ruler, pointing out that it is inappropriate for a beautiful person not to submit to the power of the Realm of Eros, the dreaded Erotokratia (N 224–70). VI (N 270a–P 2757). Outside the courtyard gates of the palace of Eros, Livistros is confronted by a terrifying guard holding an inscribed sheet of paper that urges men to become slaves to Eros (N 270a–80 ⫹ P 2702–11). They enter, and Livistros is led through a triumphal arch (tropikh´ ) whose interior is covered with mosaics and marble reliefs. These depict birds singing and flying amidst thick foliage, Aphrodite giving birth to Eros, and the Judgment of Paris-Alexandros (P 2712–35). Speaking statues adorn the four corners of the arch’s exterior cornice (P 2736–57). VII (P 2758–N 431). Exiting the arch with his escort cupid, Livistros meets a beautiful man—he is Desire (Po´ qo")—who sternly reproaches him for his rebellious conduct. Livistros admits that he will swear vassalage to Eros (P 2758–63 ⫹ N 433–53). At that moment, a beautiful tall woman appears. She is Love ( jAga´ ph), and Livistros’s escort advises him to fall at her feet. She wonders at the young man; Livistros then falls in proskynesis in front of her, and begs her and Desire to become his mediators with Eros (N 454–65 ⫹ P 2796–97 ⫹ N 398–417). The two of them magnanimously accept and leave for the Hall of Judgment, having assured Livistros that he shall even obtain the noble love of a chaste maiden (N 418–31). VIII (N 432–N 395). Livistros and his escort leave as well. They first pass by a low terrace decorated with statues of cupids (N 432 ⫹ N 359–66). Next to the terrace is a pool with a fountain basin, at the top of which stands the statue of a man tormented by snakes because he has rebelled against Eros. The statue warns Livistros about the fate of rebels (N 367–95). IX (N 396–P 256). Finally, Livistros and his escort enter the crowded Hall of Judgment; in its center Eros holds court on a throne (N 396–97 ⫹ N 292). Livistros describes in detail the three-faced ruler (N 293–309). When he is summoned before Eros, he falls in proskynesis, begs for mercy, and declares that he will swear vassalage (N 310–27). Eros replies that he will ignore Livistros’s previous behavior and will give him the hand of the princess Rhodamne, daughter of the emperor Chrysos (N 328–35 ⫹ P 217–23). Livistros then notices two women, one dressed in white and the other in red, flanking Eros; he is informed that they are Truth ( jAlh´ qeia) and Justice (Dikaiosu´ nh). Eros addresses Livistros again and tells him to go with Desire and Love and swear his oath (P 224–56). 23 It will become obvious to the reader who studies the text of the romance in the appendix that the textual situation of N 186–560 is in a sad, though not irreparable, state. The narrative sequence in manuscript N is totally disturbed due to (a) the exemplar from which the scribe was copying, and (b) two major lacunae. Moreover, the disorder was made worse by the first editor (Wilhelm Wagner) who rearranged a number of passages and then numerated his text. At the same time, manuscript P also presents a serious disorder, because two quaternions from the beginning of the manuscript were by mistake bound at the end, which its first editor (Demetrios Maurophrydes) did not notice. The correct order of the text can be reconstructed with the help of versions E and V. When one leaves smaller lacunae aside, the main sequence of the disordered passage is as follows: N 186–280, P 2702–63, N 433–66, N 398–432, N 359–97, N 281–316, N 466ab, N 317–35, P 217–302, N 336–58, N 467–560.
116
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
X (P 257–311). The three of them go to the Room of Oaths; Livistros describes a painting of Eros and various inscriptions that are mounted on the door (P 257–70). He enters the room, finds a paper attached to the bow of Eros, reads the text of the oath describing the inescapable power of the mighty ruler, and swears by it (P 271–96). Immediately a prophet appears and announces to Livistros his fate, which is a brief summary of the romance’s plot. As soon as the prophet finishes, Livistros wakes up (P 297–302 ⫹ N 336–47 ⫹ P 311). XI (N 347a–P 360). Livistros is in a state of shock, remembering the wondrous things that he saw in his dream and believing that they were real (N 347a–58 ⫹ P 322 ⫹ N 467–72). He calls for the Relative who prompts him to tell his dream (N 473–82). As he names Rhodamne, the Relative interrupts him and gives him specific information about the unknown princess, advising Livistros to go out in search of her; they pass the whole day in such conversation (N 483–98 ⫹ P 359–60). XII (N 499–560). Night arrives, and the Relative leaves; Livistros dines with his companions, then falls asleep, and sees a second dream (N 499–503). It appears to him that he is walking in a garden decorated by Eros, Desire, and Love. As he wanders in astonishment, he sees from a distance Eros, now in his form as Cupid, holding the wondrous Rhodamne by her hand (N 504–29 ⫹ P 390–93). Eros calls out to Livistros to join his company; the latter is amazed at the sight of the young woman, wondering if she is a female cupid or even Aphrodite herself (N 530–39 ⫹ P 404–12). Eros explains to him that this is Rhodamne and hands her over to him. Livistros reaches out his hand, rushes toward the maiden, and wakes up in sheer agony, desiring to see the lost dream, hating the morning light, and wishing for the return of night (P 413 ⫹ N 540–60). II Before I examine more closely Livistros’s two dreams and their spatial aesthetics, it is necessary to consider some theoretical points concerning the nature and use of space in narrative, and in the Byzantine romances in particular. In contrast to the “objective” notion of time, space in ancient and medieval literature is perceived as something “subjective,” for, while the cyclical arrival of day and night appears as a phenomenon external to the subject, space is anchored in the concrete world of the subject’s existence. This difference explains, at least in the case of ancient and medieval texts, the absence of a fully developed system of spatial formulas that would structure the narrative process, as opposed to the obvious presence of temporal formulas. (One needs only to remember the formulaic appearance of “rosy-fingered dawn” in the Homeric poems.)24 Space manifests itself in narrative primarily through oppositions—here and there, above and below, left and right, closed and open, inside and outside. These spatial oppositions are often connected to the notions of movement or immobility. At the same time, space can be used to create an “atmospheric” setting. In any case, in whatever form space might appear in a narrative, it is part of the narrative situation, namely, a defined unit within the narrative process, which consists of time, space, and action.25 For a more detailed exposition, see Agapitos, Narrative Structure, 272–75; on temporal formulas in the romances and their history in Greek literature, see ibid., 227–35. 25 On space in narrative, see G. Hoffmann, Raum, Situation, erza¨hlte Wirklichkeit: Poetologische und historische Studien zum englischen und amerikanischen Roman (Stuttgart, 1978), 1–53, with further bibliography. 24
PANAGIOTIS A. AGAPITOS
117
In the Byzantine vernacular romances, some of which are based to a large extent on the rhetorical patterns developed in the Komnenian novels,26 space is used in a number of ways. For example, spatial notions are conveyed through the various signals that introduce the descriptive mode.27 Moreover, in Livistros and Rhodamne space assumes a particular coloring, because the romance is written in the first-person perspective, with space thus acquiring maximum subjectivity. Verbs of seeing (ei«da, ejfa´ nh me) or being (h«ton), employed by the characters as narrators, signal for the reader a visual (qua spatial) component in the narrative situation,28 and similar phrases also indicate the conclusion of such a visual section.29 A more complex formula for the conveying of spatial notions is what can be termed “passage of space,” 30 which is closely related to a similar “passage-of-time” formula.31 The passage-of-space device is a phrase consisting of an adverb, a verb of motion, and a space signifier.32 In Livistros and Rhodamne this phrase is mostly combined with some notion of time: either the adverb is temporal, or another sentence that includes temporal signifiers is added to the phrase (N 268–70: Kai` me` ta` " to´ sa" ta` " polla` " ejrwtonouqesi´a" | oJka´ mpote eij" tou' “Erwto" h“lqame th` n katou´ na | kai` th` n aujlh` n ejse´ bhmen th'" E j rwtokrati´a").33 A prominent device of the author in indicating the passage of space in the romance is the “spatial bridge”: the traversal of space is conveyed by inserting into the narrative mode various types of the discoursive mode,34 such as monologues, dialogues, speech26 See Agapitos, Narrative Structure, 141–93 (the three modes of narration), and idem, “Cronologikh` ajkolouqi´a,” 101–28 (structural, typologic, and stylistic aspects of the vernacular romances). To view the Palaiologan romances as completely separate from the Komnenian novels (as does C. Cupane, review of Narrative ¨ B 43 [1993]: 455–62), or to count them as Structure in the Byzantine Vernacular Romances, by P. A. Agapitos, JO part of early Modern Greek literature in order to study them from the perspective of Apollonios and Erotokritos (as does G. Kechagioglou, review of Narrative Structure in the Byzantine Vernacular Romances, by P. A. Agapitos, Hellenika 44 [1994]: 200–213), shows, in my opinion, a lack of understanding of the socio-political and cultural context in which these works were produced and read. 27 The descriptive mode signifies the sections of text in which the whole range of rhetorical ekphrastic devices is employed in a clearly defined manner and embedded in the narrative process. Signals of the descriptive mode are verbs of seeing or being that introduce or conclude a descriptive passage; see, for example, Kallimachos and Chrysorroe (hereafter K&C) 808 («Hn ga` r hJ ko´ rh pa´ nterpno", ejrwtoforoume´ nh), or Velthandros and Chrysantza (hereafter V&C) 292 (Ei«de, parexenw´ qhn ta, uJpereqau´ mase´ n ta); see further Agapitos, Narrative Structure, 177–93. 28 See, for example, N 189–90 (the beginning of the first dream) or P 2722 (the triumphal arch). On the group ei«da/ejfa´ nh me, see also N 271, N 371 ⫹ 374, N 292, N 504, N 513; on the group h«ton, see also P 2729, P 2737, N 361, N 367, N 513. 29 See, for example, P 2746 (connecting to P 2722 and ending the description of the triumphal arch). See also N 195 (⬍ N 190), N 380 (⬍ N 374), N 383–84 (⬍ N 371), N 305 (⬍ N 292), N 558 (⬍ N 504); in the parentheses the introductory verse is given to which the concluding verb connects. 30 Agapitos, Narrative Structure, 275–76, and 279–80 on L&R in particular. 31 The “passage of time” refers to a set of phrases that function as clearly defined joints between various narrative situations with the purpose of propelling the action forward in time. See, for example, K&C 918–19 (“Efqasen, h«lqe to` loipo` n kai` tri´to" hJ hJme´ ra, | ajne´ teilen oJ h”lio", h«lqe pro` " meshmbri´an), V&C 851 (kai` die´ bh w”ra perissh` to` n nou'n twn na` sumfe´ roun), L&R S 1197 (kai` meta` wJri´tsan ojligh` n h«lqen oJ Berderi´co"). 32 See, for example, K&C 173–74 (Kai` meta` th` n paradromh` n kai` tou' tosou´ tou to´ pou | eij" ka´ stron kathnth´ samen me´ ga, frikto` n kai` xe´ non). The phrase kai` meta` th` n paradromh` n kai` tou' tosou´ tou to´ pou is modeled on the passage-of-time formulas, such as meta` kairou' paradromh´ n (K&C 147). 33 For further examples in the dream sequence, see P 2712, P 2722, P 2758, N 434, N 454–55, N 432, P 246, P 257, P 271 (without a temporal notion), P 297. 34 The discoursive mode encompasses the clearly delineated succession of constructed monologues and dialogues between the characters, integrated into the narrative process (see Agapitos, Narrative Structure,
118
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
frame formulas,35 and narratorial statements.36 Thus, while the romance’s characters talk to themselves or to someone else, they have moved from one place to another. This motion is framed by appropriate signals (N 195–202: “Eblepa to` ajnali´badon . . . mo´ no" kai` mo´ no" e“lega . . . Kai` ejno´ sv to` ejpara´ treca).37 However, the most obvious carrier of spatial notions in the descriptive mode is description itself, the time-honored ekphrasis.38 In the vernacular romances ekphrasis is employed chiefly in two ways, as a “compact” or as a “broken” description.39 The description is compact when it is framed by specific signals and usually concentrates on one person or object, giving to the described subject a sense of formality and importance.40 The description appears as broken when the ekphrasis is divided into smaller units and intertwined with the narrative and the discoursive modes. In this way, description, narration, and dialogue are inextricably combined, creating an amalgam that lends a sense of fluidity to the rhetorical structure of the text. Such an extended broken description is Livistros’s first dream, with a total of some 460 verses (N 190–347 ⫹ P 311), which is introduced (N 188–89a) and concluded (N 347a–58 ⫹ P 322 ⫹ N 467–72) with all the necessary signals of the descriptive mode. The dream itself includes various fully signaled compact ekphraseis41 and shorter descriptions,42 all of them embedded in a fluid sequence of action and dialogue that is held together by passage-of-space formulas and spatial bridges.
159–76). For a splendid example from the dream sequence, see N 434–65 ⫹ P 2796–97 ⫹ N 398–415 (the conversation among Livistros, his escort cupid, Desire, and Love). 35 Within the discoursive mode, the “speech frame” represents the signaling system, originally a device from the oral epic tradition, by means of which a speech is introduced and concluded (see Agapitos, Narrative Structure, 64–73). For two examples from the dream sequence, see N 229–30 ⫹ N 268–70 (the speech of the escort cupid), N 405 ⫹ N 416–18 (Livistros’s petition to Love). 36 On authorial-narratorial statements, see Agapitos, Narrative Structure, 74–90. Two excellent examples from the passage under consideration here are N 227–30 and P 2714–21 (both cases combine an interjection in the form of a question and an intervention in the form of an explicative statement). 37 See also N 224–70 (Livistros and the cupids cross the meadow and reach the camp of Eros), P 2758–63 ⫹ N 433–34 (the escort cupid leads Livistros away from the triumphal arch), N 454–65 ⫹ P 2796–97 ⫹ N 398–400 (Love arrives while Desire and Livistros talk to each other), N 432 ⫹ N 359–97 (Livistros passes by the terrace and the pool on his way to the Hall of Judgment), N 473–75 ⫹ P 331 (a case of abstract space traversal: Livistros calls for the Relative), N 513–37 (a complex case with a set of inner delays: Livistros meets Eros and Rhodamne in the garden; he twice breaks the narration and addresses Klitovon). 38 On ekphrasis in general, see E. Mitsi and P. A. Agapitos, “Eijkw` n kai` lo´ go": JH perigrafh` e“rgwn te´ cnh" sth` buzantinh` logotecni´a,” Cronika` Aijsqhtikh'" 29–30 (1990–91): 109–26, and L. James and R. Webb, “‘To Understand Ultimate Things and Enter Secret Places’: Ekphrasis and Art in Byzantium,” Art History 14 (1991): 1–17. 39 For a detailed analysis of the stylistic and formal aspects of descriptions in the Digenes Akrites, the Komnenian novels, and the Palaiologan romances, see C. Jouanno, “L’ekphrasis dans la litte´rature byzantine d’imagination” (diss., Sorbonne, 1987), 63–142. I take here the opportunity to thank Dr. Corinne Jouanno for making a copy of her unpublished dissertation available to me. 40 For example, the description of the pool (fiski´na) in Rhodamne’s garden (L&R S 1311–71), or the description of Chrysantza (V&C 677–99); on compact descriptions in the romances, see Agapitos, Narrative Structure, 179–80, 184, 186–88. 41 P 2720–52 (the triumphal arch), N 359–95 (the terrace, the pool, and the basin), N 291–311 (Eros). 42 N 190–97 (the meadow), N 218–20 (the escort cupid), N 271–73 (the guardian at the door), N 435–37 (Desire), N 455–64 (Love), P 224–33 (Truth and Justice), P 259a–67 (the door of the Room of Oaths).
PANAGIOTIS A. AGAPITOS
119
III Let us now return to the protagonist’s first two dreams. Their chief narrative function is to introduce Livistros to the Realm of Eros and make him fall in love. In other words, they depict a conversion from rebellion and unfaithfulness 43 to submission and faithfulness.44 This conversion is part of a larger initiation process into the “religion” of love, a process that starts with this sequence and ends much later, when Livistros finally marries Rhodamne and enters her father’s castle. Livistros’s conversion at the beginning of the romance is carried out in three stages. The first stage is the mental preparation of the initiate, necessary before he is confronted with the power of Eros. The preparation is achieved through the turtledove episode. The sad incident provokes in Livistros a sense of wonder, which he describes as a musth´ rion fobero´ n (N 132).45 In seeking out the Relative, Livistros voluntarily wishes to find out about this awe-inspiring mystery (N 137–39, N 148–50). This voluntary act is of extreme importance, since it reflects a desire for knowledge that the initiate must display in order to exit from his state of “senselessness” (ajnaisqhsi´a).46 The speech delivered by the Relative on the all-encompassing power of Eros, which freely uses material from the older novels,47 is a didaskalia into the mysteries of Eros, as Livistros candidly acknowledges (N 151–52: kai` ejkei'no" ejpecei´rhse tou' na` me` ajnadida´ xh | tou' e“rwto" ta` musth´ ria kai` ta` desma` tou' po´ qou) and as the Relative emphatically points out (N 173–74: katalepto` n a‘n a“rxwmai tou' na` se` ajnadida´ xw | tou' e“rwto" ta` musth´ ria ta` dei´cnei eij" th` n ajga´ phn). At the end of this lesson, Livistros finds himself contemplating Eros, deeply anxious that he might get entangled in the sorrows of love (N 178–85). With the initiate’s mind and heart prepared, the second stage begins when night falls and Livistros has his first dream. Night as the time when the mind liberates itself and 43 See the phrases used to describe Livistros as a “rebel against desire” (P 2742: ajntista´ th" tou' po´ qou) and a “slave to unfaithfulness” (P 2743: th'" ajpisti´a" oJ dou'lo"). 44 See the core of Livistros’s oath, where he offers servitude to Eros, vassal allegiance to Desire, and faithfulness to Love (P 295–96: doulw´ nomai eij" to` n “Erwta, liziw´ nomai eij" to` n Po´ qon, | pisto´ " th" na` ei«mai ajpo` tou' nu'n th'" jErwtikoaga´ ph"). 45 The phrase also appears twice within the dream sequence, at P 2739 (musth´ rion ei«da fobero´ n) and P 218 (ei«da frikto` n musth´ rion), where it characterizes first an architectural feature and then Eros’s tripartite voice. In this sense, the phrase expresses an aspect of the author’s poetics, since the semantic field of fobero´ ", frikto´ ", para´ xeno", and xenoca´ rago" is applied in the romance to works of art, the dream, and narration itself. It is unfortunate that studies of poetic concepts and terms in Byzantine literature have not been undertaken at all. 46 Throughout the first part of the romance Livistros is described as insensible (or insensitive) to love (N 175: ajnaisthto´ tero"; N 250: a“mnhsto"; N 321: ajpo` ajnaisqhsi´a" mou). This concept is connected to inexperience in erotic affairs, which is described by the notions of boorishness and peasantry (N 23: a“nqrwpon a“groikon eij" to` n ko´ smon; N 411: a“nqrwpo" h«ton cwriko´ "; N 324: h“moun cwriko´ ")—notions that clearly reflect the aristocratic social perspective of the author. 47 The speech consists of the list of four exempla—three taken from nature’s paradoxes and one from myth—that illustrate the power of Eros. Such lists appear in Tatios 1.17–18, Eugeneianos 4.135–49, and ´ berlieferung, Rekonstruktion, Textausgabe Manasses frag. 21–21a (O. Mazal, Der Roman des Konstantinos Manasses: U der Fragmente, Wiener Byzantinistische Studien 4 [Vienna, 1967]); the list in L&R is identical with the one in Manasses’ novel (see Agapitos, “Cronologikh` ajkolouqi´a,” 107, and Agapitos and Smith, Medieval Greek Romance, 77).
120
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
dreams visit the sleeping person is, of course, an old convention in Greek literature.48 What is interesting in the case of our romance is that the first dream of Livistros represents a complex narrative sequence, with a number of topoi incorporated but rearranged. In his dream Livistros, being a rebel against imperial authority, experiences the terrifying power of the Erotokratia. This sense of anxiety and fear, established already from the first stage, forces Livistros to accept a second didaskalia, now offered to him by his escort cupid (N 232–67). The monologue, which picks up some of the images in the Relative’s speech,49 concentrates on the offense of rebellion (N 240–48), on the need for Livistros, whose sexual beauty is extolled, to learn about the affairs of love (N 249–58), and, lastly, on the cardinal virtue of humility in erotic servitude (N 259–63). From this point on and for the rest of the dream, Livistros sees persons and objects, reads texts, and hears voices with growing astonishment.50 The various speaking statues, the numerous inscriptions, the triumphal arch with its mosaics and marble reliefs, the admonitions of Love and Desire, and all other elements add up to a didaskalia on the vassalage of love. A closer look at the text of the first dream reveals that the author has used a repetitive pattern51 to underline the didactic function of what Livistros sees, reads, and hears. Four times a tripartite schema of “figure holding text—inscription read—text explained” is repeated: the guardian at the gate of the courtyard of the Erotokratia and his inscription explained by the escort cupid (N 271–80 ⫹ P 2702–8); the tormented man in the pool basin and his inscription again explained by the escort cupid (N 373–82); the painting at the door of the Room of Oaths and its inscription explained by Desire (P 260–267); and the bow of Eros and the text of the oath of love explained by the prophet (P 273–91).52 Livistros himself accentuates this didactic aspect of the narrative when, confronted with the three-faced Eros, he exclaims: “Who is the creator and what is this strangely made being I see, what is it really? Who will tell me what I see, who will inter48 On dreams and dream theory in antiquity, see the collective volume, D. I. Kyrtatas, ed., “Oyi" ejnupni´ou: JH crh´ sh tw'n ojnei´rwn sth` n eJllhnikh` kai` rwmai¨kh` ajrcaio´ thta (Herakleion, 1993), with full bibliography; on dreams in the ancient novels, see Bartsch, Ancient Novel, 80–108; and on the learned novels, see S. MacAlister, “Aristotle on the Dream: A Twelfth-Century Romance Revival,” Byzantion 60 (1990): 195–212, and eadem, Dreams and Suicides: The Greek Novel from Antiquity to the Byzantine Empire (London–New York, 1996), 115–52. 49 These images are the four exempla on the power of Eros (N 233–39 艐 N 161–77). 50 The astonishment of a character confronted with unfamiliar situations, especially in the first part of the romance, is a major concept used to convey the “didactic” function of narrative (Agapitos, Narrative Structure, 48–54). In this way, a romance becomes an explanation of and a guide to the mysteries of the ejrwtikh` ajsco´ lhsi", the “worries of love.” In L&R, this sense of astonishment to be experienced by the characters and the readers (and expressed through the verbs qauma´ zw, ajporw', ejxaporw', and xeni´zomai) is extremely prominent; from the dream sequence alone, note N 196, P 2721, P 2746, P 2747, P 2756, N 291, N 292, P 245, N 467, N 537, N 541. The concept of aporia forms, together with the notion of musth´ rion fobero´ n (see above, note 45), one of the romance’s key poetic concepts. 51 On such patterns and their importance in L&R, see P. A. Agapitos, “ JH ajfhghmatikh` shmasi´a th'" ajntallagh'" ejpistolw'n kai` tragoudiw'n sto` muqisto´ rhma Li´bistro" kai` Roda´ mnh,” Thesaurismata 26 (1996): 25–42, in particular 38–40. 52 On the importance of the written word in the form of a document for Byzantine mentality, see H. Hunger, “Die Herrschaft des ‘Buchstabens’: Das Verha¨ltnis der Byzantiner zu Schrift- und Kanzleiwesen,” Delt.Crist. Arc. j JEt. 4.12 (1984): 17–38, esp. 30–33, with reference to the romances. Obviously, the written text held in the hand of a figure also reflects the iconography of prophets, saints, and bishops in Byzantine art, shown holding similar scrolls of wisdom (I owe the suggestion to the kindness of Ioli Kalavrezou).
PANAGIOTIS A. AGAPITOS
121
pret it for me, what friend of beauty will teach me about this being?” (N 306–9).53 When, finally, Livistros has been brought into the Room of Oaths to swear, he has voluntarily accepted the power of Eros. The prophet, in yet another didaskalia, instructs him about his future fate and the woman Eros promised to give him. And, just as at the end of the first stage, so too now, when Livistros wakes up from the dream, he is only concerned, in amazement and fear, with what he saw during the previous night. He is again in need of his erotodidaskalos, who, knowing very well who the princess Rhodamne is, begins to direct Livistros’s thoughts to love through words of comfort and the suggestion that he should go in search of her. The initiate has by now understood the power of Eros and has accepted it as a governing force of his life, but he has not as yet experienced love. At this point, we enter the third stage in the process of conversion; for, now that Livistros is a vassal of Eros, he can be converted body and soul to his new religion. This happens in the second dream, when Livistros actually meets Rhodamne in Eros’s magnificent garden—the garden being the female erotic setting par excellence.54 The passage wherein Livistros falls in love with Rhodamne—a process described by him in a complex series of exits from and entrances onto different narrative levels (N 515–29 ⫹ P 390–93 ⫹ N 530–32)—captures all of the mental uncertainties involved in the process of conversion, which, however, once concluded, is unalterable and complete. Thus, Livistros exits his dream in a passage from perfect delight (N 551: ajpo` th` n to´ shn hJdonh´ n) to total agony (N 553: ojdu´ na" ajmetrh´ tou"). The only things left to him are a sense of sorrow and desire (N 557–58), exactly the feelings that he feared and wanted to avoid at the beginning of his initiation (N 552–60 艐 N 184–85). The Relative appears again, giving his final advice to the young king on how to set out from Livandros in search of Argyrokastron, the seat of the emperor Chrysos (N 561–606).55 Here ends Livistros’s conversion. It has been structured by the author with admirable clarity through the threefold repetition of the tripartite sequence of “appearance of external causative agent—releasing of disturbed emotions—provision of instructive teaching.” The world of dreams, simultaneously external and internal to the protagonist,56 then forms the medium through which the chief concern of the romance—the act of searching for one’s lover—is established at the beginning of the narrative and subsequently sustained in its further development.57 From the moment when Livistros wakes up, converted to love but suffering because of it, he tries to superimpose his two dreams— On the complex figure of “Erw" trimorfopro´ swpo", see C. Cupane, ““Erw" basileu´ ": La figura di Eros nel romanzo bizantino d’amore,” Atti dell’ Academia di scienze, lettere e arti di Palermo, ser. 4, 33.2.2 (1973–74): 243–97, in particular 290–91, and Beaton, Medieval Greek Romance, 156–58, with the cautionary remarks by Agapitos and Smith, Medieval Greek Romance, 84–85. 54 ˆle of the Garden in the Byzantine Romance,” BMGS 5 A. R. Littlewood, “Romantic Paradises: The Ro (1979): 95–114, and Smith, “Some Features,” 88. 55 See Agapitos, “Cronologikh` ajkolouqi´a,” 105–6, on the Relative’s function as a “helper,” which is a term used in the analysis of wonder-tales (C. Bremond, Logique du re´cit [Paris, 1973], 282–94). 56 On this aspect of erotic dreams in the romances, see Smith, “Some Features,” 93–94. 57 There are three further dreams in the romance: Eros appears to Livistros assuring him that he will instill love in the heart of the young princess (N 690–706); Eros pierces Rhodamne’s heart with his arrow (S 197–219); and Klitovon sees a prophetic dream on the happy outcome of Livistros’s quest for Rhodamne in Egypt (S 1535–76). 53
122
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
in other words, the inner world of desire—on reality, that is, the outer world of obstacles. IV Absent or present, the most impressive figure in Livistros’s dreams is Eros, “lord emperor, master of all the earth, commander of the inanimate world, ruler of animate beings, examiner of every soul, judge of the law of desire, helper of love, friend of respect” (N 317–20). This is how Livistros, in proper Byzantine form, addresses the mighty ruler, having fallen on the ground in front of his throne and begging for mercy. By the time the reader has reached this point in the dream, the image of Eros as a Byzantine emperor has fully established itself. The realm of the Erotokratia with its warriors, the concept of rebellion, Eros’s palace, the triumphal arch, the ruler as judge, the mediating figures of Desire and Love, the strictly observed protocol, and the act of proskynesis—these are all elements derived from Byzantine imperial imagery.58 Their function is to present for a Byzantine readership the terrifying and “autocratic” power of Eros. But once Livistros has begged for forgiveness, Eros shows his magnanimity, which is a chief virtue of the imperial monarch (N 329–35).59 Only then does Livistros notice the figures of Justice and Truth flanking the ruler’s side.60 Carolina Cupane was the first to point out that the dream sequence in Livistros and Rhodamne stands in close relation to the opening sequences of Makrembolites’ Hysmine and Hysminias, although she perceived the relation as one of intelligent imitation.61 I would like to suggest that a more profound and creative dialogue is carried out between the two texts. The situation in the Komnenian novel is as follows.62 Hysminias meets Hysmine in her parents’ house; she falls in love with him, but he is insensitive to this feeling. The setting for this and subsequent flirtation is the house’s beautiful garden, described by Hysminias at the very beginning of the novel (1.4–6). The next day, he and his cousin Kratisthenes go to the garden where they find a magnificent pavilion filled with paintings (2.1–11). These depict the four cardinal virtues in the form of four women, as a verse On these elements, see Agapitos, Narrative Structure, 189–90, 192 n. 149, 325 n. 137; P. E. Pieler, “Recht, ¨ B 20 (1971): 189–221, esp. 205–11. Gesellschaft und Staat im byzantinischen Roman der Palaiologenzeit,” JO 59 H. Hunger, “Philanthropia: Eine griechische Wortpra¨gung auf ihrem Wege von Aischylos bis Theodoros Metochites,” AnzWien 100 (1963): 1–21. 60 On these figures, see Agapitos, Narrative Structure, 325 n. 138; Cupane, ““Erw" basileu´ ",” 246 n. 15; Jouanno, “L’ekphrasis,” 250–56. See also the depiction of John II Komnenos and his son Alexios crowned by the enthroned Christ who is flanked by Mercy and Justice in Vat. Urb. gr. 2, fol. 19v of A.D. 1122 (V. Lazarev, Storia della pittura bizantina: Edizione italiana rielaborata e ampliata dall’autore, Biblioteca di storia dell’arte 7 [Turin, 1967], pl. 251) and the representation of David between Wisdom and Prophecy in Vat. Pal. gr. 381, fol. 2r of the 13th century (D. Talbot Rice, The Art of Byzantium [New York, 1959], pl. 177). 61 Cupane, ““Erw" basileu´ ",” 286. In a similar vein are the remarks on the relation between the two texts by Jouanno, “L’ekphrasis,” 232–35. 62 For a more detailed summary of the novel’s plot, see H. Hunger, Die hochsprachliche profane Literatur der Byzantiner, HAW 12.5.2 (Munich, 1978), 2:137–41. The novel has received a most sympathetic interpretation by M. Alexiou, “A Critical Reappraisal of Eustathios Makrembolites’ Hysmine and Hysminias,” BMGS 3 (1977): 23–43. See also Beaton, Medieval Greek Romance, 79–87, and Agapitos and Smith, Medieval Greek Romance, 39–40, 42–44, 81–82. 58
PANAGIOTIS A. AGAPITOS
123
inscription discloses; Hysminias attempts a bold—and wrong—interpretation of the fresco. Next, they see a painting of a naked youth sitting on a throne, flanked by two women. Hysminias makes another—equally wrong—interpretive attempt. At the enthroned youth’s feet stand thousands of people, commoners, rulers, and soldiers, men and women. Hysminias is bewildered, until a second verse inscription discloses the identity of the youth: it is Eros. Kratisthenes instructs Hysminias that the fresco depicts the all-conquering power of Eros, while the two women represent Night and Day. Then, at night, Hysminias is visited by a terrifying dream: Eros, as he was depicted on the fresco, appears to Hysminias, accompanied by a crowd of people holding torches. A terrible voice summons Hysminias the rebel; Hysmine as a suppliant asks for the ruler’s mercy, and Eros crowns Hysminias with a garland of roses. As the god tells Hysmine that she now has her lover, the dream disappears, an oppressive weight falls on Hysminias’s heart, and he wakes up in fear (3.1–4). He goes to sleep again and sees another dream, this time of a purely erotic character (3.5–7). He and Hysmine engage in playful sexual combat, which is about to lead to its final climax. Hysminias senses a languishing and sweet pain, but at that moment Hysmine flies from his hands, and sleep from his eyes.63 It has been argued that, although one is confronted with various Byzantine elements, the chief source of inspiration for this sequence is to be found in Western medieval, French in particular, romantic poetry.64 Be this as it may, I would like to point out the following in connection with this sequence. In the first instance, its imagery is placed in a clearly ancient Greek context, while Achilleus Tatios’s erotic garden and erotic painting function as important structural guides of the narrative.65 But for a twelfth-century Byzantine reader this ancient setting is “exotic,” in the sense that it represents a literary construct unrelated to social reality.66 At the same time, this “antique” world is suffused with Byzantine cultural associations, some of which, in my opinion, derive from Christian 63 On the novel’s dreams, see Alexiou, “Critical Reappraisal,” 40–42, and MacAlister, Dreams, 135–40, 144–46. 64 Cupane, ““Erw" basileu´ ",” 261–81; eadem, “Il motivo del castello nella narrativa tardobizantina: Evo¨ B 27 (1978): 229–67; eadem, “Topica romanzesca in oriente e in occidente: luzione di un’allegoria,” JO ‘Avanture’ e ‘amour,’” in Il romanzo tra cultura latina e cultura bizantina: Testi della III settimana residenziale di studi medievali (Carini, Villa Belvedere, 17–21 Ottobre 1983), ed. C. Roccaro, Biblioteca dell’Enchiridion 5 (Palermo, 1986), 47–72; Beaton, Medieval Greek Romance, 154–63. Objections have been voiced against this approach and some of its results; see, in particular, E. Jeffreys, “The Comnenian Background to the romans d’antiquite´,” Byzantion 50 (1980): 455–86; P. A. Agapitos, “The Erotic Bath in the Byzantine Vernacular Romance Kallimachos and Chrysorrhoe,” ClMed 41 (1990): 259 n. 11, 268–69; idem, Narrative Structure, 189–90; Agapitos and Smith, Medieval Greek Romance, 81–90; Agapitos, “Cronologikh` ajkolouqi´a,” 97–99, 133. Thus, Roderick Beaton (Medieval Greek Romance, 211–12, 216–17, 219–20) has retracted or modified some of his views. Since the dates of composition for both the learned and the vernacular works are still open to debate (see above, note 7), exact lines of dependency between the Western and the Byzantine works are very difficult to demonstrate. On the Komnenian novels, see S. MacAlister, “Byzantine Twelfth-Century Romances: A Relative Chronology,” BMGS 15 (1991): 175–210, and the cautionary remarks by P. Magdalino, “Eros the King and the King of Amours: Some Observations on Hysmine and Hysminias,” DOP 46 (1992): 197–204. 65 On Tatios’s opening sequences, see Bartsch, Ancient Novel, 40–79. 66 Agapitos, “Cronologikh` ajkolouqi´a,” 110. Already in the ancient novels one is confronted with “historical” settings (Chariton, Heliodoros) that may evoke or even criticize the classical Greek world from a Hellenistic or late antique perspective. Indicatively, see T. Ha¨gg, “Callirhoe and Parthenope: The Beginnings of the Historical Novel,” Classical Antiquity 6 (1987): 184–204; J. R. Morgan, “History, Romance and Realism in Heliodoros,” Classical Antiquity 1 (1982): 221–65.
124
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
iconography rather than directly from imperial imagery.67 The image of Eros as a basileu´ " on the throne owes some structural and stylistic debts to Byzantine depictions of the Last Judgment and of Christ in his glory as the King of Kings.68 And most importantly, Hysminias’s first dream makes use of two famous Gospel scenes: the thunderous voice descending from heaven and stopping the rebel is inspired by the conversion of Paul on the road to Damascus (Acts 9:1–9); and the handing over of Hysminias to Hysmine by Eros reflects Christ handing over Mary to John at the cross (John 19:25). Moreover, even though Eros is characterized as basileu´ ", he is always referred to in the novel as the painted image of a god worshiped by his believers.69 In this sense, the divine character of Eros is brought out by Makrembolites through a suffusion of the two images of rulership that his culture revered, namely, Christ as the heavenly ruler and the emperor as the earthly one. Turning now to Livistros and Rhodamne, one can discern both the similarities and the differences between the two texts. The anonymous poet has introduced two major changes in Makrembolites’ schema—one structural and one iconographic. Iconographically, Eros is not a god any more, but is consistently described as a ruler of a state, with all the accompanying apparatus. In Livistros and Rhodamne, therefore, Eros is secularized. Moreover, he has been removed from the world of representative art and thus given animated substance, since he is not a painted image but a real person. Yet, at the same time, his power has been restricted, since he now appears only in the world of dreams, functioning as a causative agent and a projection of the protagonist couple’s emotions, but decidedly not as a deus ex machina.70 Furthermore, in Livistros and Rhodamne the overpowering presence of Eros stands in inverted relation to the depiction of sexuality, which is limited to the use of specific imagery rather than fully represented,71 as is the case in the later romances72 (though not without the imposition of certain rules of conduct, especially concerning premarital sex).73 67 When bringing Christian iconography into the discussion, I do not in the least mean to imply that the Byzantine novels should be read as Christian allegories, as has been suggested for Makrembolites by K. Plepelits, Eustathios Makrembolites: Hysmine und Hysminias, Bibliothek der Griechischen Literatur 29 (Stuttgart, 1989), 29–69. On the other hand, Byzantine readers are not precluded from supplying Christian allegorical interpretations of the ancient novels and the later romances, though the subject is far from seriously studied (Beaton, Medieval Greek Romance, 190–92; Agapitos and Smith, Medieval Greek Romance, 106–8). See the forth´ou Siga´ la: Pracoming study by P. Odorico, “”Ena" monaciko` " ajnagnw´ sth" tou' Kalli´macou ,” in Mnh´ mh Antwni j ´ou ktika` th'" 6h" ejpisthmonikh'" suna´ nthsh" tou' Tome´ a Mesaiwnikw'n kai` Neoellhnikw'n Spoudw'n tou' Aristotelei j Panepisthmi´ou Qessaloni´kh", ed. V. Katsaros (Thessalonike, forthcoming). 68 See, for example, the Last Judgment in the famous Par. gr. 74, fol. 51v, dating from the third quarter of the 11th century (Lazarev, Storia, pl. 194). 69 3.1.2: to` n gegramme´ non “Erwta; the phrase is repeated at 3.8.2 and at 7.18.2 (Eros’s final appearance in Hysminias’s last dream). 70 Contrast Hysminias’s dream at 7.18, where Eros actually saves Hysmine from drowning in the sea storm. 71 ` shmasi´a,” 40–42, and idem, “Cronologikh` ajkolouqi´a,” 133. Agapitos, “ Afhghmatikh j 72 The study of sexuality in the vernacular romances still has to be undertaken (see Agapitos, “Erotic Bath,” 268–70). The lengthy article by L. Garland, “‘Be amorous, but be chaste . . .’: Sexual Morality in Byzantine Learned and Vernacular Romance,” BMGS 14 (1990): 62–120, is in many points oversimplifying and misleading (Agapitos and Smith, Medieval Greek Romance, 61 n. 148; Smith, “Some Features,” 93 n. 71), pace Beaton, Medieval Greek Romance, 216–17. 73 The prevailing opinion on the matter, as summarized by Beaton, Medieval Greek Romance, 109, is that “the taboo on sex before marriage, rigorously maintained throughout all the earlier medieval romances, is
PANAGIOTIS A. AGAPITOS
125
Beyond that, however, the structural, and ultimately symbolic, change between the two texts is quite revealing. Everything that takes place in Makrembolites’ account is connected to the novel’s narrative reality. Hysminias and Hysmine are seeing each other, the garden and its artistry are present, the frescoes are painted inside a real pavilion, the dreams stand in immediate relation to experienced reality,74 and the various stages of Hysminias’s conversion fluctuate between reality and dream. This is not the case in Livistros and Rhodamne. First, the stages of conversion are much more clearly defined through the tripartite repetitive pattern I have described here. Second, while everything connected with the protagonist’s conversion is apparently external to him, at the same time the whole process takes place within his mind. He does not see anything in reality since everything appears to him in dreams. To give but one example, even though Rhodamne appears in Livistros’s second dream, it takes both the protagonist and the reader some eighteen hundred verses after the dream sequence to actually see Rhodamne, as Livistros enters her father’s castle and describes the beautiful princess in a formal ekphrasis (S 1271– 1300). Within the dreams themselves, the various images have shifted, and everything related to the allegory of love has become animate. Eros, Love, Desire, Justice, Truth, the guardians, the warrior cupids, and the prophet are living beings and not painted images of an exotic scenery. Art, however, is still very much present in Livistros’s dreams; but it has become the testimony of the musth´ rion fobero´ n (P 2739) that has to be explained step by step to the ignorant protagonist by his “teachers.” The architectural features, the decorative frescoes and mosaics, the mythological reliefs, the statues, and the gardens make the setting of a fantastic world that has been created by what to Livistros seems to be the wondrous art (P 2730: ejk te´ cnh" paraxe´ nou) of a painter (N 193: ce´ ria zwgra´ fou). The complex relation between te´ cnh and reality (or fu´ si" in Byzantine terms)75 is most vividly illustrated in the case of the three-faced Eros. Livistros initially perceives the enthroned ruler as yet another work of art (N 296–97: eja` n to` ei«de", na` ei«pe" ejk panto` " ce´ ria kalou' zwgra´ fou | tecni´tou to` eJsto´ rhsan, ye´ go" oujde` n basta´ zei) and, as we have seen, asks for an interpreter to explain to him this wondrous creation (N 306–9).76 However, no flouted, frequently with gusto” (so also H.-G. Beck, Byzantinisches Erotikon [Munich, 1986], 182–83, and Cupane, “Topica romanzesca,” 64–66). However, in those vernacular romances where premarital sex is to be found (K&C, V&C, Achilleid, Byzantine Iliad), it is carried out only by the protagonist couples, and in such a manner as to suggest that the lovers perceive their relation as a private marriage (see O. L. Smith, “Literary and Ideological Observations on the N Version of the Achilleid,” in Origini della letteratura neogreca, ed. Panagiotakis [as in note 7], 2:182–87; idem, “Some Features,” 87–94; Agapitos, “Cronologikh` ajkolouqi´a,” 112–13). 74 MacAlister, “Aristotle on the Dream,” 198–205, and eadem, Dreams, 158–64. 75 On the issue, see Beaton, Medieval Greek Romance, 28, 65–68, with the objections of Agapitos and Smith, Medieval Greek Romance, 40–44. 76 This, in miniature form, is the motif of the interpreter of a work of art, as it is used in the novels, for example, Longos explaining the painting at the Grotto of the Nymphs on Lesbos, Kleitophon appearing to the anonymous narrator in Tatios, or Kratisthenes interpreting the frescoes for Hysminias in Makrembolites. The motif has also been used by Manasses in an ekphrasis of a mosaic in the imperial palace depicting the ¨B figure of Earth (O. Lampsides, “Der vollsta¨ndige Text der “Ekfrasi" gh'" des Konstantinos Manasses,” JO 41 [1991]: 189–205), where suddenly a connoisseur of art appears and gives vital technical information to the astonished narrator: Tau'ta´ mou le´ gonto" e”tero" ejggu´ teron parestw` " (h«n de` deino` " polupragmonei'n ta` toiau'ta kai` ta` musthriwde´ stera katanoei'n tw'n tecnw'n) “e“ti ple´ on,” e“fh, “qauma´ sei" to` n tau'ta diamorfw´ santa, eij kai` th` n u”lhn tw'n tupwma´ twn ejxakribw´ sh⭈ ouj ga` r uJgro´ th" ejpitrimma´ twn oujde` bafw'n ajnakerasmoi` oujde` crwma´ twn
126
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
such person appears, since Livistros immediately discovers that Eros, unlike the previous speaking statues, is a real being (N 310–35 ⫹ P 217–23). In his second dream, therefore, Livistros interrupts his narrative (P 391–93) to refer back to his omission of the didaskalia expected after N 309, and makes a note to Klitovon that the latter should remind him to furnish for him—and for the reader—an explanation of the nature of Eros. This Klitovon duly does at N 711–15, and then Livistros promptly delivers his explanation (N 716–24 ⫹ P 577–84 ⫹ N 725–29). In sum, the poet of Livistros and Rhodamne has created out of Makrembolites’ antique reality a modern fantasy in which spaces are presented as “landscapes of the mind.” V The imaginary landscapes in Livistros’s two dreams consist of three larger spaces: a meadow and the palace of Eros in the first dream, and the garden of Eros in the second.77 To a certain extent, the meadow and the garden, which open the two dreams, are described in similar vocabulary (N 190–203 艐 N 504–14), though their narrative and symbolic functions are very different. The meadow, an open and undefined space, creates a setting where Livistros aimlessly rides alone. The brief description of the meadow gives the reader a feeling of the setting but does not allow him to form a clear picture of it, since it is seen through the protagonist’s distracted gaze.78 At the same time, the meadow is traversed from two opposite directions. On the one hand, Livistros, in a typical spatial bridge, talks to himself and moves through the meadow (N 195–202); on the other, the warrior cupids appear from the distance, flying across the meadow and attacking the narrator (N 202–11). This combination of space as an atmospheric setting—what I would term “space presented”—and space as a three-dimensional notion, in other words, “space activated,” is something new in Byzantine novelistic tradition.79 The garden in the second dream—this time a closed and defined space into which Livistros enters (N 504)—is again an atmospheric setting that is created by Eros as an artist (N 506). Here the protagonist himself activates spatial notions by consciously exploring the garden (N sumfu´ rasi", ajlla` leptw'n yhfi´dwn eujfuh` " aJrmogh` tw'n eijkonismw'n tou´ twn o”lwn ejzwogra´ fhse” (ibid., 196.45–50). One should note the similarity of vocabulary used to describe the didactic process (musthriwde´ stera, qauma´ sei") to that in L&R; see above, notes 45 and 50. 77 On the three spaces, see also Agapitos, Narrative Structure, 322–28. 78 A concomitant result of the author’s insistence on aporia as a key concept to the narrative’s progress in the romance (see above, note 50) is what can be termed the motif of the “distraction-in-amazement” gaze. A character walks through a space, and his gaze distractedly falls on various parts of the setting, thus evoking for the reader a puzzling picture that he has to piece together himself. See, for example, passages within the dream sequence: N 195–97, P 2751–52, N 310, N 469–72, N 509–11 ⫹ P 374, N 555–60. This motif does not seem to be of any importance in ancient Greek and Latin literature, where one only finds a viewer’s gaze oscillating between two persons or objects (e.g., Theokritos 1.36–38, Heliodoros 7.4.2, Ovid Metamorphoses 5.164–67, Seneca Thyestes 707–11). 79 Compare, for example, the way in which space is “presented,” rather than “activated,” in similar situations in the three Komnenian novels: Eugeneianos, Drosilla and Charikles, 4.325–5.171 (Charikles finds Drosilla sleeping in a garden; she wakes up, and a long conversation ensues); Prodromos, Rhodanthe and Dosikles, 3.43–78 (Dosikles and Rhodanthe enter a vineyard where they have an amorous conversation); Makrembolites, Hysmine and Hysminias, 2.1–11 (Kratisthenes and Hysminias enter Sosthenes’ garden, where they see and describe the frescoes painted on the pavilion’s walls). In all three cases, both movement in space and description of space are included, but they are clearly separated, so that the reader cannot perceive space as three-dimensional. Obviously, the subject needs further study.
PANAGIOTIS A. AGAPITOS
127
514), only to come upon Eros and Rhodamne.80 The difference between the meadow and the garden, a spatial opposition of open versus closed, is of primary importance for understanding Livistros’s state of mind in the two dreams. The central space in the romance’s dream sequence is, of course, the palace of Eros. The atmospheric aspect of this space is very prominent, since what Livistros sees causes him to voluntarily acknowledge the power of the Erotokratia.81 The walled palace has a gate with a guard (N 271–75). Once in the courtyard (aujlh´ ),82 Livistros offers the reader yet another brief description of the setting seen through his astonished gaze (P 2714–17). In a narratorial intervention of the “garrulity” type (P 2718–21),83 he decides to focus on one architectural feature of the courtyard, namely, the impressive triumphal arch (P 2722–52), which is a three-dimensional version of the decorative porticoes used for the Eusebian canon tables in Byzantine Gospel manuscripts.84 Having passed through the arch, Livistros first meets Desire and then Love. He is then led past a marble terrace and a pool, next enters the crowded Hall of Judgment, and faces Eros on his throne. After his judgment he is taken to the Room of Oaths, the innermost space in the palace of Eros. Thus in his first dream Livistros moves from an open space (the meadow) to an enclosed space (the courtyard), then to a closed but large space (the hall), and finally to a closed and small space (the room). Through this spatial motion from open to closed, the poet has depicted Livistros’s growing anxiety and receding power of resistance as he finds himself gradually submitting to Eros.85 What at first appeared as an atmospheric setting ultimately proves to be, in narrative terms, a highly active space. The spatial opposition “open versus closed” is correlated to an opposition “freedom versus subjugation”—a point that is underlined throughout the dream, for example, in the inscription held by the guardian at the palace’s gate (N 274–80 ⫹ P 2702–6).86 Moreover, this opposition is connected to the opposition “movement versus immobility,” since movement does acquire in certain instances the meaning of “freedom (qua rebellion) from love” (N 190, P 280–81), while immobility is similarly correlated to the idea of “subjugation (qua submission) to love” (N 214, P 286). By looking back at the opposition “open versus closed” as expressed by the meadow of the first dream and the garden of the second, we may now conclude that this reflects in spatial terms the initial free state of Livistros as opposed to his final subjugated state at the end of the sequence. Note, in particular, the splendid traversal of space at N 530–36, with its dense verbal indentation (sunapantw' . . . sunapanta', ble´ pei . . . ble´ pei . . . ble´ pw , sta´ . . . sth´ kw). 81 See above, notes 28–29, for the references to “seeing.” 82 N 270, N 271, N 278, P 2708, P 2712, P 2715. 83 On this device, see Agapitos, Narrative Structure, 78–81. 84 Indicatively, see Athen. Bibl. Nat. 2364 of the 11th century (A. Marava-Chatzinikolaou and C. ToufexiPaschou, Kata´ logo" mikrografiw'n buzantinw'n ceirogra´ fwn th'" jEqnikh'" Biblioqh´ kh" th'" JElla´ do" [Athens, 1978], 1: pls. 122–26); Athous Meg. Laur. A 42 of the 11th century (S. M. Pelekanidis et al., OiJ qhsauroi` tou' Agi J ´ou “Orou", ser. 1, Eijkonografhme´ na ceiro´ grafa [Athens, 1979], 3: pls. 29–30); and Vat. gr. 1158 of the 13th century (H. Buchthal and H. Belting, Patronage in Thirteenth-Century Constantinople: An Atelier of Late Byzantine Book Illumination and Calligraphy, DOS 16 [Washington, D.C., 1978], pls. 18–20). 85 This device of growing anxiety joined to the traversal of an unfamiliar space has also been used in V&C and K&C, in a similar position within the narrative (Agapitos, Narrative Structure, 326–27). 86 Being outside the courtyard of the Erotokratia suggests freedom from love (N 249–52), while being inside the courtyard implies subjugation to love (N 256–64). 80
128
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
In concluding, I would like to suggest that the landscapes of the mind in Livistros and Rhodamne, far from being mere conventional patterns, are meaningful spatial entities that are firmly placed within the narrative process. The poet has succeeded in establishing his own, innovative spatial aesthetics of simultaneous formality and fluidity, where space is both presented and activated, where dreams are both realistic and fantastic, and where love’s instruction is both explicitly and implicitly inculcated. The peculiar sensation and the narrative importance of these dreams can be fittingly summarized by a phrase of the Canadian novelist Robertson Davies: these are “dreams in which something significant is told, not in bold Civil Service narrative, but in a puzzle of ambiguity and omission.” University of Cyprus
Appendix A Preliminary Critical Edition of the Dream Sequence in Livistros and Rhodamne, Version a (⫽ N 186–560)
Since no critical edition of the romance exists and none of the available printed editions of the individual manuscripts (S, N, P) or versions (E)1 is satisfactory,2 the foregoing analysis has been based on a preliminary edition of version a, of which I am preparing a full critical edition. The editorial method employed has been presented in a series of studies, to which the reader is referred for further information.3 In general, for the first 1,300 verses of the romance the text is based on manuscript N, whose lacunae are supplemented from P whenever there is an agreement between P, E, and V. An attempt has been made to keep as closely as possible to the wording of the manuscripts, without “smoothing out” grammar, syntax, or metre. Editorial corrections are restricted to passages where none of the manuscripts offers any solution.
1 The printed editions are as follows: J. A. Lambert, Le roman de Libistros et Rhodamne´ publie´ d’apre`s les manuscrits de Leyde et de Madrid avec une introduction, des observations grammaticales et un glossaire, Verhandelingen der Koninklijke Akademie van Wetenschappen te Amsterdam, Afdeeling Letterkunde, n.r., 35 (Amsterdam, 1935), for manuscripts N, S, and for version E; W. Wagner, Trois poe`mes grecs du Moyen-Age (Berlin, 1881), 242–349, for manuscript N; D. I. Maurophrydis, jEklogh` mnhmei´wn th'" newte´ ra" eJllhnikh'" glw´ ssh" (Athens, 1866), 1:324–428, for manuscript P. Corrections to the text of manuscript N by D. C. Hesseling are recorded in the apparatus of Lambert’s edition. Further corrections have been suggested by E. Kriaras, Lexiko` th'" Mesaiwnikh'" JEllhnikh'" Dhmw´ dou" Grammatei´a" (Thessalonike, 1969– ). 2 See the detailed, but far from exhaustive, criticism by M. K. Chatzegiakoumis, Ta` mesaiwnika` dhmw´ dh kei´mena: Sumbolh` sth` mele´ th kai` th` n e“kdosh´ tou", vol. 1: Li´bistro", Kalli´maco", Be´ lqandro" (Athens, 1977), 35–79. 3 ¨ B 42 P. A. Agapitos, “Libistros und Rhodamne: Vorla¨ufiges zu einer kritischen Ausgabe der Version a,” JO (1992): 191–208; idem, “ JH e“mmesh para´ dosh tou' dhmw´ dou" muqistorh´ mato" Li´bistro" kai` Roda´ mnh,” Hellenika 42 (1991–92): 61–74; idem,“”Ena ajko´ mh spa´ ragma tou' muqistorh´ mato" Li´bistro" kai` Roda´ mnh: JO batikano` " kw´ dika" Barb. gr. 172,” Hellenika 43 (1993): 337–59; idem, “Pro` " mia` kritikh` e“kdosh tou' muqistorh´ mato" Li´bistro" kai` Roda´ mnh: Problh´ mata meqo´ dou,” in Prosa y verso en Griego medieval: Rapports of the International Congress “Neograeca Medii Aevi III,” ed. J. M. Egea and J. Alonso (Amsterdam, 1996), 1–16; P. A. Agapitos and O. L. Smith, “Scribes and Manuscripts of Byzantine Vernacular Romances: Palaeographical Facts and Editorial Implications,” Hellenika 44 (1994): 61–80.
130
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE
Sigla Atque Breviata E N P V Ch Hs Kr <> {} [] †† a.c. add. cf. codd. coll. coni. corr. del. dist. edd. i.e.
Escurialensis Y-IV-22, saec. XV exeuntis Neapolitanus III-Aa-9, saec. XVI ineuntis Parisinus graecus 2910, saec. XV medii Vaticanus graecus 2391, ca. a. 1500 Chatzegiakoumis Hesseling Kriaras
iteravit omittit, omittunt post correctionem praebet putat scripsit seclusit supplevit temptavit traiecit transposuit
addenda delenda supplenda in lacuna codicis locus corruptus ante correctionem addidit confer codices collato, collatis coniecit correxit delevit distinxit editores id est
PANAGIOTIS A. AGAPITOS
131
AFHGHSIS LIBISTROU KAI RODAMNHS O J ka´ pote ejka´ lesen hJ eJspe´ ra th` n hJme´ ran kai` e“klinen oJ h”lio" kai` ejse´ bhken hJ nu´ kta⭈ ejnu´ staxa ejk th` n me´ rimnan th` n ei«cen hJ yuch´ mou, e“pesa na` ajpokoimhqw', kai` a“kouse ti´ me` ejfa´ nh.
N 186
O j nei´rwto" ajfh´ ghsi" Libi´strou polupo´ nou.
N 189a
Ej fa´ nh me o”ti mo´ no" mou peri´treca liba´ din, liba´ din panexai´reton murioanqisme´ non kai` kru´ on nero` n gluko´ bruton, cilio´ dendra gema´ ton⭈ ce´ ria zwgra´ fou na´ lj ege", a‘n ei«de" to` liba´ din, to` ejpoi'kan cilioe´ mmorfon, muriocrwmatisme´ non. “Eblepa to` ajnali´badon, ejpro´ seca ta` de´ ndrh, ejpiterpo´ mhn ta` futa´ , ejqau´ mazon ta` " bru´ sa", eij" ta` a“nqh oJ nou'" mou ejkre´ meton ta` ejgra´ mmizan to` n to´ pon, mo´ no" kai` mo´ no" e“lega kai` ejno´ sv periepa´ tounÚ “”Opou eij" te´ toion a“nqrwpo" liba´ din katoune´ yei kai` zh´ sei eij" te´ toia" ca´ rita" zwh'" tou ta` " hJme´ ra" kai` crh´ zei to` n para´ deison, loipo` n oujk e“n∆ stratiw´ th".” Kai` ejno´ sv to` ejpara´ treca to` e“mnoston liba´ din kai` uJph´ gaina´ to ejnh´ dona kai` ejsko´ piza eij" ejkei'non, ajpo` makra` ajnetra´ nisa kai` ejble´ pw aJrmatwme´ nou", ajnqrw´ pou" o”lou" pterwtou` " kai` e“rcontai eij" ejme´ nan, meta` qumou' ejpe´ tonto, e“trecan to` liba´ din⭈ kai` wJ" tou` " ejnetra´ nisa polla` ejyucofobh´ qhn, ejk to` a“logo´ n mou ejpe´ zeusa kai` su´ rnw to` spaqi´n mou⭈ kai` w’" ou» na` su´ rw to` spaqi´n, ejkei'noi ejpe´ pesa´ n me kai` triguri´a mou ejsta´ qhsan kai` ajgrioglwssofwni´zoun kai` le´ gounÚ “Ri´ye ta` a”rmata, mh` tw´ ra oJka´ ti pa´ qh".” Kai` ejgw` wJ" tou` " ei«da perissou´ ", o”lou" aJrmatwme´ nou", tou` " o”lou" na` e“cousin ptera` kai` na` ajnasai´noun flo´ ga, periesta´ qhn a“peira, scedo` n o”ti ajpe` tw´ ra eij" a”dhn ejkath´ nthsa kai` th` n zwh` n ajfei´qhn⭈
N 190
N 195
N 200
N 205
N 210
N 215
N 186–92 lacuna in P N 188 ejnu´ staxa Hs (cf. ejnu´ staxa E V) : -xen N N 189a polupo´ nou Lt : po´ nou N N 192 cilio´ dendra scripsi (cf. muriw´ dendron E : mhriodendra` V) : cei´lh dendrw'n N : ci´lia dendra` Wa N 193 ga` r na` e“lege", iJsto´ risen to` luba´ din P N 196 ejpiterpo´ mhn] periepo´ qoun P bru´ sei" P N 197 oJ nou'" e“keiton P ta` 2 om. N ejgra´ mmizan coni. Ch coll. ejgra´ miza E : ejgra´ mmhsen P : ejza´ lhna N to` n to´ pon N : oJ po´ qo" P N 198 kai` mo´ no"] ejkei'no" P kai`2 om. P N 199 wJsa` n o”pou N P : wJsa` n del. Wa a“nqrwpo" om. P N 200 ta` " Wa : th` " N th'" zwh'" tou hJme´ ra" P N 201 om. P N 201 kai` Lt : cai` N N 202 to` 1 om. P e“mnoston] ajnqo´ mnoston P N 203 kai` ajph´ gena th'" hJdonh'" kai` ejski´rtoun eij" ejkei'no P ejsko´ piza Wa : ejsko´ pisa N N 204 ajne´ bleya kai` ble´ pw P N 205 h“rconto pro` " P N 206 ejpe´ tonto P : kai` potapou' N e“trecan Wa : e“treca N : kai` e“trecon P N 207 om. P N 207 polla` ejyucofobh´ qhn scripsi : pollou` " ejyucoefobh´ qhn N N 208–11 hos versus post N 215 transp. P N 209 w’" ou»] w”sper P su´ rw P : ri´sw N ejpe´ pesan Wa : ejpo´ pesan N : e“pesa´ n P N 210 trigu´ rou ga´ r me e“sthsan kai` me qumou' me le´ goun P N 211 kri´yai P oJka´ ti tw'ra paqein qe´ lei" P N 211 post hunc versum ejgw` wJ" se ei“pa su´ ntrofe kale` sunodoipo´ re, tou` " o”llou" ei«da sobarou` " o“llou" aJrmatome´ nou" praeb.P N 213 a“llon na` e“cousi P flo´ gan P N 213 post hunc versum to´ ra ga` r na` basta´ zousin, spaqi´a gegumnwme´ na praeb.P N 214–15 periesta´ qhn a“peira kai` eij" a”dhn ejkate´ bhn⭈ kai` ajph´ lpisa tou' zh'n me e“lega mo´ no" kai` katanou'n mou⭈ ti´ne" kai` po´ qen e“rcontai kai` ti´ to` sau´ thn th` n spoudh` n poiou'sin pro` " ejme´ nan⭈ kai` ejn o“sw tau'ta eij" me´ rimnan mo´ no" mou ejlogizo´ mhn P N 214 periantista´ qhn a“peira"
Li´bistron e“rw" nouqetei' kai` poqoparagge´ llei. ““Anqrwpe, qe´ lei" na` se` eijpw' kai` de´ xou to wJ" qe´ lei". “An oujk ejpla´ sth" ejk th` n gh'n kai` oujk h“soun ejk to` n ko´ smon, kai` h“soun spora` ejk to` si´deron kai` ajpo´ komma ejk th` n pe´ tran, ouj mh` to` ei«ca para´ xenon posw'" a‘n oujk aijsqa´ nou th` n du´ namin th` n a“peiron tw'n ejrwtokrato´ rwn⭈ dio´ ti kai` pe´ tra kai` dendro` n kai` si´dhron kai` li´qo" kai` pa'sa fu´ si" a“yuco" kai` ejmyucwme´ nh pa'sa ejkto` " ejrwtou¨ polh´ yew" oujk e“ni oJdo` " na` zh´ sh. Kai` su´ , oJ tosou'to" a“nqrwpo", oJ ejxai´reto" kai` ne´ o", ajnaiscuntei'" to` n “Erwtan kai` Po´ qon ouj yhfi´zei"… “Arti a‘n me` ajkou´ ei", sugkli´qhse, ri´xe to` ajge´ rwco´ n sou, tra´ chlon kli´ne eij" to` n zugo` n th'" ejrwtodoulei´a", e“mpa eij" tou' Po´ qou to` n desmo´ n, de´ qhse eij" th` n jAga´ phn, pro´ spese eij" th` n jAsco´ lhsin, to` n Kremasmo` n ijde´ ton, kai` aujtoi` ka‘n na` eijpou'n to` n “Erwtan, na` to` n parakale´ soun, kai` ajpe` to` to´ son maniko` n to` kat∆ ejsou' ejkakw´ qhn, na` metape´ sh, na` ajllagh' kai` na` se` sumpaqh´ sh.
N 235
N 240
N 245
N 216 to` konta´ rin P N 217 e“dusa kai` ta` " cei´ra" mou, kai` le´ gw mh` ajpoqa´ nw P ei«mai secl. Wa N 218 tw' ei“dh P N 219 eu“morfo" P kai` sch'ma] eij" pla´ sin P N 220 kai` ajra´ smata ejba´ stan P N 221 om. P N 222 du´ nei P N 223 polu´ P N 224 ajrxa´ meqa P N 225–26 ajpe´ dw mou kai` ajpe´ kei mou kai` tou` " poinhlatista´ " mou P N 226 ajdia´ kritou" Wa (cf. P 2705) : ajdiko´ kritou" put. Ch : ajdikri´tou" N N 227 pw'" na` ] ti` na` se P ei“pw P : eijpw' N N 229 nouqeti´smata P N 230 om. P N 230 k∆ e“surne´ scripsi : kai` surne´ N N 231 atramento in N, sed minio exaratus in P lu´ bistron P : lu´ bestro" N nouqetei' kai` poqoparagge´ llei N : parenei' e“rw" kai` paragge´ llei P N 232 qe´ lei"1 om. P na` se` eijpw' scripsi coll. V : a‘n se ei“pw ti´pote P : nasiwpw' N N 233 ejpla´ sth" Wa : ejpla´ sqh" P : ejpla´ sthn N tou' ko´ smou P N 234 h“soun ajpo` to` si´dhron P ajpo´ komma Wa : ajpo´ koman P : ajpi´koma N ejk pe´ tra" P N 235 ei«con . . . eja` n . . . ijsqa´ nou P N 237 dendro` n P : drendro` n N si´dhro" P N 238 a“yuco" Wa (cf. a“yuco" V et a“yucon E) : e“myuco" P : a”pasa N N 239 cwri`" e“rwto", uJpo´ lhyin oujk e“cei P ejrwtoejpoli´yew" N N 240 oJ ne´ o" P N 241–43 a‘n ei«ce" kai` su` to` n e“rwta aujqe´ nthn eij" to` n ko´ smon, tou' po´ qou a‘n h‘sai douleuth` " lu´ zio" th'" ajga´ ph"⭈ eja` n o”lw" eijsakou´ sh" mou dou´ lwsin uJpogra´ yh", oujk ajstoca` se to` ejpiqumei'", ma'llon me` n ou«n ouj ca´ nei" P N 242 sugkli´qhse scripsi (cf. sunklh´ qhse E et klh´ qise V) : sunqli´bese N N 244 e“mpa] ejlqe` P de´ qhse P : de´ qhte N N 245 pro´ speson P N 246 aujtoi` ka‘n na` Ch : aujtoi´kana N : aujtoi` na` se` P N 247 ajpo` P N 248 na` metaqe´ sh th` n ojrgh` n P ajllagh' Wa : ajlgh' N
PANAGIOTIS A. AGAPITOS Kai` a‘n se` eijpw', to` n a“nqrwpon, qe` " o”ti oujde` n na` pa´ qh", a“mnhsto" na` ei«sai ajpo` e“rwtan, oujde` n na` to` n yhfi´zh"… Ti´ le´ gei" na` ei«sai parekto` " ejrwtikh'" ajga´ ph"… Pre´ pei se, ijde´ to mo´ no" sou, sko´ pei to` se` aJrmo´ zei. Ma` to` spaqi`n to` di´stomon th'" Ej rwtokrati´a", ouj le´ gw se´ to ti´pote dia` tro´ pon kolakei´a"⭈ o”son kai` a‘n ei«sai ejxai´reto" eij" su´ nqesin kai` pla´ sin, a‘n oujk ejmph'" eij" to` n zugo` n tou' po´ qou na` pone´ sh", na` paideuqh'" ta` ejrwtika` kai` ma´ qh" ta wJ" aJrmo´ zei, ei«sai oujdeti´pote, ajpo` ejme` n plhrofore´ qhse´ to. Kai` paragge´ llw se ajpeda` kai` le´ gw se kai` a“kouse´ mou⭈ a“rti a‘n uJpa'" eij" “Erwtan kai` qe´ lh" proskunh'sai, e“mpa klito` " to` n tra´ chlon kai` camhlo` " to` sch'ma, poi'se deino` n to` ble´ faron wJsa` n foberisme´ no", de´ se ta` ce´ ria sou sfikta` kai` pe´ se eij" gh'n ojmpro´ " tou kai` ajpo` kardi´a" sou stri´ggise kai` pareka´ lese´ ton. Plh` n o”tan e“mph", pro´ sexe ajpa´ nw eij" to` ajnw'flin th'" po´ rta", ejlefa´ ntinon a‘n i“dh" pinaki´din, ajna´ gnwse ta` gra´ mmata ta` gra´ foun eij" ejkei'non.” Kai` me` ta` " to´ sa" ta` " polla` " ejrwtonouqesi´a" oJka´ mpote eij" tou' “Erwto" h“lqame th` n katou´ na kai` th` n aujlh` n ejse´ bhmen th'" Ej rwtokrati´a".
133
N 250
N 255
N 260
N 265
N 270
O J Li´bistro" eij" u”pnon tou ta` " ca´ rita" ta` " ei«den, katalepto` n ejkfra´ ssei ta", le´ gei ta" meta` po´ nou.
N 249 a‘n se eijpw' kai` ajnqrw´ pina, qe` " o”ti oujde` n ejph´ rqh" P oujde` n na` Wa : oujde´ na N N 250 ajmnh´ mwn na` ei“sai tou' e“rwto" oujde e’n na` se yhlafh´ sh P to` n yhfi´zh" scripsi (cf. ajneyifh´ si" E) : se yufh´ zei N N 251 ti´] su` P kai` post ei«sai add. P N 252 ske´ you to eja` n ajrmo´ zh P N 253 ejrwtokratei´a" P (cf. oJrwtokratori´a" E et ejrotokrato´ rwn V) : ejrwtomani´a" N N 254 ouj le´ gw soi tou'to P N 255 o”son P (cf. o”sa E V) : wJsa` n N kai`1 om. P kai`2] eij" P N 256 ejmph'"] ejmpe´ sh" P N 257 wJ" e“cei na` to` ma´ qh" P N 258 plhrwforh´ qhsai ajp∆ ejme` n o”ti ti´pote oujk h‘sai P N 259 se . . . se] soi . . . soi P kai`2 om. P N 260 o”tan ejmbh'" P uJpa'" Wa : upa´ gei" N e“rwta P proskunh´ sin P N 261 e“mpa] se´ ba P camhlo` "] tapeino` " P N 262 poi´hsai P ble´ faron wJsa` n] pro´ swpon kaqa` P N 263 dei´sai P pe´ se kai` eij" th` n gh` n ejmpro´ " tou P N 264 kardi´a" P : kardi´a N sou om. P parakale´ se P N 265 e“mph"] sebh'" P pro´ sexon ejpa´ nw eij" to` n kosmu´ thn P N 267 ajna´ gnwson P gra´ mmata P : gra´ mma N eij" ejkei'non] pro` " ejkei´nhn P N 268 polla` "] fora` " P ta` " ante ejrwtonouqesi´a" add. P N 269 oJka´ pote P h“lqomen th` n katou´ nan P N 270 kai` eij" th` n aujlh` n ejfqa´ samen P N 270b kai` ante le´ gei add. P po´ nou] po´ qou P N 271 me` n P : ejme` n N mou] me P o”ti oJka´ ti" om. P N 272 h«ton om. P ejxespaqome´ no" P : xespaqwme´ no" coni. Wa N 273 eij" post polla` add. P sch'ma] ble´ man P N 274 ei«cen eij" ce´ rin tou carti`n P N 275 e“gemen om. P a“kouson P : a“ko N : a“kw coni. Wa N 276 ejrwtokratei´a" P N 277–78 om. P N 279 a‘n de` qelh´ sh na` ejmbh' P kai`2 om. P
134
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE a‘" uJpogra´ yh dou'lo" tou kai` a‘" gi´netai ejdiko´ " tou, kai` to´ te na` ijdh' ca´ rita" a’" e“cei oJ poqokra´ twr⭈ a‘n de` mourteu´ sh na` ejmbh', mh` uJpogra´ yh dou'lo", a‘" ejgnwri´sh dh´ mio" tou gi´netai to` spaqi´n mou, kai` ejgw` pikro´ " tou tu´ ranno", meta` ajdiakrisi´a" na` ko´ yw to` kefa´ lin tou, na` lei´yh ajpo` to` n ko´ smon.” Kai` paraka´ tw e“grafen ejk th` n grafh` n ejkei´nhnÚ “Aujlh'" porta´ rh" eu“morfo" kai` po´ rta th'" jAga´ ph".” Kai` oJpou' ta` " po´ rta" e“blepe, le´ gei meÚ “ jAna´ gnwse´ ta.” jAne´ gnwsa ta` gra´ mmata kai` sfo´ dra ejqlibo´ mhn, le´ gwÚ “ jApeda` doulw´ nomai eij" tou' “Erwto" to` to´ xon.” Kai` to´ te ajpe´ sw eij" th` n aujlh` n ejse´ bhmen ajnta´ ma, ejgw` kai` oJ dh´ mio" e“rw" mou kai` oiJ poinhlatistai´ mou. Kai` ti´ na` se` ei“pw, fi´le mou, kai` ti´ na` se` ajfhgou'mai, th` n su´ nqesi´n te th'" aujlh'", ta` zw'a a”per ei«cen, ta` kru´ a nera´ , ta` de´ ndra
, th` n zwdiofiski´nan, th` n iJstori´an tou' toi´cou th", ta` zw'a ta` periepatou'san… Kai` te´ w" katalepto` n a‘n se` to` qe´ lw le´ gein, qe´ lw kairou' paradromh` n kai` pla´ to" o”lou cro´ nou⭈ kai` a“fe" tou'ta ta` polla` kai` e”nan na` se` ei“pw, to` ei«da eij" to` ojspi´tin tou' “Erwto", , kai` ejxeni´sqhn. Ta` prw'ta ejse´ bhn tropikh` n kai` h«ton oJ pa´ to" o”lo" ajpo` crusoliqw´ mato" ejrwtoi¨storisme´ no", na` e“ni dia´ fora dendra´ , eu“morfa kai` wJrai'a⭈ kai` eij" tou` " klw´ nou" tw'n dendrw'n tou' kaqeno` " pro` " e”nan, pouli´a ta` me` n na` ka´ qhntai, ta` de` oJrmou'n peta'sai, kai` a“lla na` camaipe´ tountai ajpo` kladi`n eij" a“llon. Kai` me´ sa eij" to` mousi´wma to` ejrwtoi¨storisme´ non ma´ rmaron h«ton pra´ sinon kai` h«n lelatomhme´ non tou' “Erwto" ta` gene´ qlia ejk te´ cnh" paraxe´ nou, pw'" hJ jAfrodi´th to` n genna' to` n “Erwta tw'n to´ xwn kai` pw'" aujto` " ejdo´ xeusen pa´ lin th` n jAfrodi´thn. Kai` ei«cen ajpa´ nw ejnu´ steron th'" tropikh'" hJ te´ cnh th` n kri´sin th` n ejdi´daxen jAle´ xandro" tou' mh´ lou, kai` poi´an ejkatedi´kazen kai` poi´an to` mh'lon di´dei. Kai` eij" ta` tetrakogcw´ mata th'" tropikh'" ejkei´nh" h«san ejk te´ cnh" e“rwte" ajpo` yilou' guyi´nou, na` ste´ koun kai` eij" ta` cei´lh tou" na` kei´tetai kala´ min⭈ <musth´ rion> ei«da fobero` n ka‘n ajpo` pnoh'" ajne´ mou,
N 280 P 2702
P 2705
P 2710
P 2715
P 2720
P 2725
P 2730
P 2735
N 280 ejdiko´ " tou om. P P 2702–63 om. N P 2704 dhmi´o" P, corr. Ma P 2707 ejk scripsi (cf. ejk E) : ajk P P 2708–9 minio exarati in P P 2711 ajpeda` scripsi (cf. ajpeda` E V) : ga` r ajpa´ rth P P 2713 dh´ mio" Ma (cf. dh´ mio" E) : de´ smio" P punalhstai´ P, corr. Ma P 2716 th" addidi th` n zwdiofiski´nan scripsi (cf. ta` zw´ dia th'" fhsku´ na" E) : th'" zwopiskidi´a" P P 2718 to` pa'n supplevi ex E V P 2721 fi´le addidi (cf. E V) P 2722 th` n post ejse´ bhn praeb. P, quod delevi propikh` n P, corr. Ma P 2724 e“ni scripsi (cf. e”nan E) : ble´ pei P P 2726 na` add. Ma P 2728 to` mousi´wma Ch coll. E V : th` n mousikh` n P P 2731 ajfrodi´th" P tw'n to´ xwn] an toxo´ thn? (cf. toxo´ thn V et ejrwtodoxh´ thn E) P 2733 ejnu´ steron scripsi (cf. ejnu´ stron E et h“steron V) : iJstorisqh` n P hJ te´ cnh scripsi (cf. th'" te´ cnh" E V) : to` ge´ no" P P 2738 pro` " ejse´ nan post kei´tetai praeb. P, quod delevi kala´ min e versu sequenti huc traieci P 2739 musth´ rion supplevi ex E V
Eij" u”pnon tou ejsunh´ nthsen oJ Li´bistro" to` n Po´ qon. ”Ama to` ejbgh'n th` n tropikh` n sunapanta' me oJ Po´ qo", a“ndra" polla` paneu´ morfo", a“spro", xanqo` " th` n tri´can, ei«cen eij" to` kefa´ lin tou stefa´ nin ajpo` da´ fnh", ei«cen kai` eij" ta` ce´ ria tou kladi`n oi»on to` stefa´ nin⭈ kai` eujqu` " wJ" me` uJph´ nthsen, h“rxato na` me` le´ gh lo´ gou" wJ" dikasi´mato" dh'qen kai` nouqesi´a"Ú “Qauma´ zw ajpa´ rti to` n qumo` n th'" Ej rwtokrati´a", to` pw'" eij" e”na a“nqrwpon makroqumei' tosou'ton, to` pw'" oujk h“soun dou'lo" tou, pw'" to` n ajnaiscunth´ qh", to` to´ xon tou h«rgen eij" ejse` n kai` oujk ejto´ xeuse´ n se⭈ fri´ttw to` pu'r to` kaustiko` n to` eij" cei'ra" tou basta´ zei, pw'" ajko´ mh oujk ejflo´ ghsen th` n o”lhn sou kardi´an.”
135 P 2740
P 2745
P 2750
P 2755
P 2760
P 2763 N 433
N 435
N 440
N 445
P 2740 ka‘n Ma : eja` n P ajna´ basi P, corr. Ma P 2742 to´ pon Ch coll. E V : po´ qon P P 2744 to` n Ma (cf. to` n V) : to` P P 2748 fobera` n coni. Ma P 2755 me´ nan post me` praeb. P, quod delevi : me` ajnemh´ nusen Ma P 2760 pai´rnei Ch coll. ejpe´ rnh E : fe´ rnei P mia'" coni. Ma : eJno` " P N 433 atramento exaratus in N tou P : to` n N ejsunh´ nthsen P : ejsune´ nthsen N lu´ bestro" N N 434 ante a”ma add. kai` P eujgh' N N 435 ajnh` r polla` eu“morfo" P xanqo` "] lampro` " P N 436 eij" th` n kefalh´ n tou e“keito ste´ fano" P N 437 ta` " cei´ra" P o”moion w”sper stefa´ nin P N 438 wJ"] kaqw´ " P N 439 dikasi´mata P N 440 ajpa´ rti Ch : ajko´ mi P : ma´ rmh N N 441 pw'" eij" ejse´ nan a“nqrwpe P tosou'ton] tosau'ta P N 442 to` pw'"] pw'" wJ" P ejnescunti´qh" P N 443 to` to´ xon tou h«rgen] oujk oi«ton to` to´ xon P h«rgen scripsi (cf. h“rcisen E) : ei“rgen N : oujk oi«ton P kai` ejkateto´ xeusai P N 444 tou basta´ zei scripsi coll. P E V : sou basta´ zei" N fri´ttw eij" to` pu'r to` basta´ zei eij" ta` ce´ ria tou P N 445 pw'" ouj flogi´zei ajko´ mi P
136
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE Te´ w" le´ gei mouÚ “ Ej gnwri´zei" me…” kai` ei«pa tonÚ ““Oci” tau'ta⭈ “ Ej gw´ mj ai oJ Po´ qo",” le´ gei me kai` ei«pa tonÚ “Proskunw' se, tre´ mw th` n ejxousi´an sou, fri´ttw th` n du´ nami´n sou, doulw´ nomai eij" to` n “Erwtan, li´zio" tou na` uJpogra´ yw.” “ Ej de` a‘n fronh'" kai` doulwqh'",” ejkei'no" me` ajpekri´qhn, “kai` pe´ sh" eij" to` e“ggrafon th'" Ej rwtokrati´a", cro´ non gluku` n pare´ trece", hJme´ ra" ajnwdu´ nou", zwh` n polla` para´ xenon e“zhse" ajpe` tw´ ra.” Ej no´ sv ejsuntu´ caina to` n Po´ qon kai` e“blepa´ ton, e“fqasen a“llh ejrwtikh` gunai'ka ejxhrhgme´ nh, eu“morfo", kaloca´ rago", ajske´ pasto", makre´ a⭈ to` crw'ma tw'n malli´wn th" h«ton xanqo` n ojli´gon, eij" du´ o plemme´ na camhla` kai` eij" to´ pou" na` sgouri´zoun, stefa´ nin eij" to` kefa´ lin th" ei«cen ajpo` mursi´nhn, mursi´nh h«ton su´ gkarpo" kai` eij" to´ pou" ei«cen a“nqh, rou'con paneumorfo´ taton ejfo´ rei ajpoka´ tw, cruso´ aspron di´ca zw´ smato", ejpa´ nw meta` gou´ na"⭈ kai` eij" to` e’n to` ce´ rin th" trianta´ fullon ejkra´ tei kai` eij" to` a“llon th" carto´ poulon kai` to` e“grafen oujk oi«da. Kai` ejkei'no" o”pou me` e“surnenÚ “ H J jAga´ ph ei«nai,” me` le´ gei, “kai` a“rti prosku´ nhson aujth´ n, doulw´ qhse eij" au“thn, kai` eijpe` aujth` n na` ge´ nhtai mesi´th" wJ" dia` ejse´ na, oJmo´ noian eij" to` n “Erwtan na` poi´sh me` to` n Po´ qon, na` eijpou'sin ti´pote" dia` ejse` n kai` na` se` sumpaqh´ sh.” «Hlqen hJ jAga´ ph, ejsi´mwsen kai` ejsta´ qh me` to` n Po´ qon kai` le´ gei wJ" pro` " to` n “Erwta ta´ ca kai` pro` " to` n Po´ qonÚ “Dia` tou'ton ejqumw´ neton procqe` " oJ ejrwtokra´ twr, dia` tou'ton h«ton oJ biasmo` " kai` hJ e“kstasi" hJ to´ sh…” Kai` ei«pa thnÚ “Nai´,” kai` to` eijpei'n tau'ta ejprosku´ nhsa´ thn, e“pesa eij" ta` poda´ ria th" kai` te´ toia th` n ejla´ lounÚ “ jAga´ ph, dou´ lh tou' “Erwto", gnh´ sia tou suggeni´da, kai` Po´ qe, dou'le tou' “Erwto" kai` gnh´ sie suggenh' tou, oujk h“xeura to` n “Erwtan na` doulwqw' eij" ejkei'non, to´ son mh` me` kaki´shte, mh` me` ejxorisqh'te,
N 450
N 455
N 460
N 465 P 2796 P 2797 N 398 N 400
N 405
N 408 P 2807
N 446 te´ w" eja` n se eijpw' gnwri´zei" me kai` o“ci tau'ta ei«pon P N 449 boulw´ nomai P li´zio" P : zhlio´ " N uJpogra´ yw P : ge´ nw N N 450 ejde` a‘n] eja` n P ejpekri´qhn P N 451 e“ggrafon Wa : e“grafon N : pro´ grafon P N 452 pare´ trece" scripsi : ejpe´ trece" N : ejpare´ tuce" P N 453 kai` ante zwh` n add. P polla` P : pollh` n N e“zhsa" ajpo` to´ te P N 454 kai` ante ejno´ sv add. P kai` post Po´ qon om. N N 455 ejxhrhme´ nh P N 456 makre´ a] wJrai'a P N 457 xanqo` n ojli´gon scripsi (cf. uJpo` xanqh´ zh oJli´gon E) : wJsa` n to` ga´ la N : wJ" kastanou'con P N 458 plemme´ na scripsi : pleme´ nh N : plegme´ nh P sgouri´zoun P : guri´zoun N N 459 eij" om. P ei«cen] h“ton P mursi´nh" P N 460 mursi´nh P : smursu´ nh N kai` e“ni hJ mursi´nh su´ gkarpo" na` e“ch kai` eij" to´ pou" a“nqh P N 461 paneumorfo´ taton] latinoko´ kopon P N 462 cruso´ aspron scripsi : cruso` n a“spron N : cru´ sapron P a“neu gou´ nh" P N 463 sic praeb. P : kai` eij" to` ce´ rhn th" ejba´ stazen trianta´ fila hJ ko´ rh N N 464 carti`n P to` 2] ta` P N 465 kai` om. P o”pou scripsi : oJpou N : oJpou' P ei“pe moi o”ti ajga´ ph e“ni P P 2796/7 kai` a“rti doulw´ qhse kai` aujth' mesi´th" wJ" dia ejse´ na N P 2796 aujth´ n Ma : aujto` n P P 2797 wJ" dia ejse´ na N : eij" ejse´ nan P N 398 poi´sh me` ] poih´ swmen P N 399 ti´pote eij" ejse´ nan P kai`] eij" to` P N 400 ejsta´ qhn eij" to` n to´ pon P N 401 om. P kai` scripsi (cf. kai` E) : wJ" N N 403 om. P dia` tou'ton Wa : dietou'ton N to´ sh Wa : to´ sh" N N 404 thn1 om. P kai` proskunw' thn P N 405 toiau'ta P N 406 gnhsi´a P N 407 a‘n h“soun oujk ejgnw´ riza kai` oJlocwrikeuo´ mhn P Po´ qe Wa : po´ qen N suggenh' Wa : suggeno´ N tou scripsi coll. E : th" N edd. N 408 kai` ante oujk add. P P 2807 praeb. P E V : om. N
PANAGIOTIS A. AGAPITOS poi´sete ti´pote eij" ejme´ n, jAga´ ph mou kai` Po´ qe, suntu´ cete to` n “Erwtan, parakale´ sete´ tonÚ ““Anqrwpo" h«ton cwriko´ ", ti´" ei«sai oujk h“xeure´ se, økai`Ø di∆ aujto` n oujk ejdoulw´ neto eij" to` ejxousiastiko´ n sou.” Di∆ ejme´ nan poi´sete ejgguhtai´, lo´ gou" kalou` " eijpe´ te, kai` ejgw` na` mj o´ sw eij" tou' “Erwto" to` to´ xon kai` th` n flo´ gan, na` ei«mai ajpe` tw´ ra dou'lo" tou kai` tou' qelh´ mato´ " tou.” Plh` n wJ" ejla´ loun, fi´le mou, kai` tou` " ejpareka´ loun, a‘n ei«con fu´ sei" tw'n lhstw'n na` me` ejyucoponou'san. Kai` ajfo´ tou ejpareka´ lesa meta` pollw'n dakru´ wn, sku´ ptei hJ jAga´ ph, ajpo` th` n gh'n shkw´ nei me kai` le´ geiÚ “Po´ qe mou, dia` to` n a“nqrwpon tou'ton na` to` n eijpou'me to` n poqoerwtokra´ tora mh` na` to` n sumpaqh´ sh.” Kai` oJ Po´ qo" me` ei«penÚ “Sw´ pase, ti´pote mh` lupa'sai⭈ ejpei` ajpeda` doulw´ nesai kai` ojmnu´ ei" na` mh` ajpisth´ sh", e“cei" kai` ko´ rh" hjqikh'" eujgenikh` n ajga´ phn kai` ajpo` “Erwtan sumpa´ qeian kai` th` n ejmh` n fili´an.” Kai` ei«pen oJpou' me` ejfu´ latten to` n e“rwtan ejkei'nonÚ ““A" e“lqh tw´ ra meta` ejse` n ejdw' oJpou' dika´ zei, na` ge´ nhtai sumpa´ qeio" tou, na` mj o´ sh eij" to` ptero´ n tou, kai` th'" ajga´ ph" to` n desmo` n na` ejpa´ rh dia` th` n ko´ rhn.” Ej metesta´ qhn ajp∆ ejme` n hJ jAga´ ph me` to` n Po´ qon, ajllh´ lw" na` ceirokratou'n kai` na` krufomilou'sin. Kai` ajfo´ tou ejkei' ajpesw´ qhmen eij" th` n ejrwtodi´khn,
137 N 409 N 410
N 415
N 420
N 425
N 430 N 432
Pa´ lin lalei' katalepto` n oJ Li´bistro" oJ xe´ no" ta` " th'" fiski´na" sumfora` " ta` " ei«den ejn ojnei´rv.
N 359 N 360
h«ton camo´ geio" hJliako` " ktisto` " ajpo` marma´ rou, gu´ rou tou leftoka´ lama latomhme´ na sth´ koun, kai` to` kaqe´ nan e“mprosqen ei«cen iJstorisme´ na zwdi´a ejrwtido´ poula e“mmorfa †ejstanwme´ na†, kai` zw'a mikra` ejkaqi´zontan eij" to` sthqai'on trigu´ rou,
N 365
N 409 poih'setai P ti´pote P : ti´pote" N Po´ qe] po´ te P N 411 h‘ton P : h“moun N kai` ante ti´" add. N N 412 kai` seclusi (cf. E 346) aujto` P ejdoulw´ nwmou N N 413 kai` ante di∆ add. N poi´sete] pe´ setai P N 414 na` oJmo´ sw eij" to` n e“rwta P N 415 ajpo` P dou'lo"] fi´lo" P N 416 kai` tou` "] kai` wJ" to` n P N 417 ei«con scripsi (cf. h“ca E) : h«ton N P fu´ sei" scripsi (cf. fu´ sin E) : fu´ si" N P edd. tou' lhstou' N ejyucoponou'san scripsi (cf. yucopone´ sh me h«can E) : ejyucopone´ qhn N P N 418 ajfo´ tou] ajf∆ ou” to` n (tou' p.c.) P N 419 pi´ptei ejpi` th` n gh'n hJ ajga´ ph P N 420 na` ] a‘" P N 421 to` n poqoerwtokra´ tora scripsi coll. V : to` n ejrwtopoqokra´ tora P : tou' po´ qou ejrwtokra´ toro" N N 422 kai` ei«pe moi oJ po´ qo" P luph'sai P N 423 kai` ojmnu´ ei"] w“mosa" P N 424 ko´ rhn hjqikh` n P eujgenikh'" N N 425 kai`1 om. P ajpoerwtosumpa´ qeian P ejmh` n fili´an Ch coll. E V : oJmofili´an N P N 426 kai` om. P oJpou N ejfu´ lasse P N 427 me post tw´ ra add. N metase` n P se post oJpou' add. P dika´ zw N N 428 na` ge´ nhtai scripsi (cf. nagu´ nete E) : na` ge´ nh eij" P : kai` ge´ nei hJ N sumpa´ qeio" tou] sumpa´ qeia P mw´ sh" P N 429 ejpa´ rh" P N 430 ajp∆ P : dia` N N 431 ajllh´ lwn P krufomilou'sin Wa (cf. krifwmulou'sin E et krifa` na` shnthce´ noun V) : krufwfula´ ssoun N : krifofilou'ntai P N 432 ajpesw´ qhmen P (cf. ejkei' ejpe´ samen E) : ajpesw´ sasin N th` n om. N N 359 lu´ bestro" N N 360 om. P N 361 hJliako` " camw´ geio" N : camai´gio" hjliako` " P ajpo` marma´ rou scripsi coll. ajpo` marma´ ron V : ajpo` trigu´ rou P : uJpe` r mersi´nh N N 362 gu´ rou tou N : trigu´ rou P leptoka´ lama lelatomhme´ na iJsth´ koun P N 364–65 om. P N 364 zwdi´a N ejrwtido´ poula Wa : ejrwteido´ pola N ejstanwme´ na N (cf. para´ xena th' qe´ sei E) : eJstanwme´ na Wa : aijstanome´ na Hs N 365 sthqai'on scripsi : sth'qo" et supra lineam t add. N : sth'qo" twn coni. Wa
138
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE o”la na` ptu´ ousin to` nero` n ajpo` ta` sto´ mata´ tou". Kai` parekto` " tou' hJliakou' fiski´na h«ton ktisme´ nh, nero` n na` ge´ mh oJlo´ glukon kai` kru´ on wJ" oJ pa´ go", kai` eij" th` n fiski´nan e“swqen iJsth´ keton ajna´ gwn kai` ajpa´ nw eij" to` n ajna´ gonta ma´ rmaron wJ" leka´ nh, kai` eij" th` n leka´ nhn e“swqen a“nqrwpo" me` ejfa´ nh⭈ e“myuco" h«ton e“lege", na` zh' kai` na` kinh'tai, ta` du´ o tou ce´ ria na` bastou'n ojmpro´ " tou eij" to` sth'qo" ta´ cate ejfa´ nh me carti`n kai` ei«ce gramme´ na tau'taÚ ““A" me` ponh' o”pou me` qewrei' kai` oJpou' me` ble´ pei a‘" pa´ sch, a‘" qli´betai oJpou' ta` ojmma´ tia tou guri´zoun pro` " ejme´ nan, tou´ thn th` n katadi´khn mou th` n e“cw kai` th` n pa´ scw, th` n uJpome´ nw ajpo` “Erwtan, th` n ejkatedika´ sthn, diati` to` n oujk ejgnw´ riza kai` oujde` n to` n ejfobou´ mhn.” Kai` ta` me` n gra´ mmata e“blepa kai` ta´ ca ajna´ gnwsa´ ta, kai` ejsei´oun to` kefa´ lin mou na` mh` kai` ejme´ nan ou”tw" katadika´ sh oJ fobero` " oJ poqoerwtokra´ twr. Ej kei'non o”pou ejba´ stazen ta` gra´ mmata ejpro´ seca´ ton, kai` e“blepa ajpo` ta` ojmma´ tia tou to` da´ kruon na` stala´ ssh kai` na` cocla´ zh wJ" to` qermo´ n, na` kai´h wJ" to` kami´nin⭈ ajpa´ nou eij" to` kefa´ lin tou na` kei´tetai wJ" ojfi´din, o”lhn na` periple´ ketai ta´ ca th` n korufh´ n tou, na` e“nai to` sto´ ma tou pukno` n eij" to` me´ twpon tou' ajnqrw´ pou. Kai` ejno´ sv to` n ejpro´ seca, na` ei«pa ejfw´ naxe´ meÚ “Fobou' mh` pa´ qh" ta` e“paqa kai` murioturannh'sai.” Kai` th` n fwnh` n wJ" h“kousa, na` ei«pe" uJpepa´ ghn, na` sunqrhnw' ejpecei´rhsa to` n a“nqrwpon ejkei'non, na` klai´w th` n katadi´khn tou kai` na` to` n ajntipa´ scw⭈ kai` le´ gei moi oJ e“rw" mou to` n ei«ca meta` me´ nanÚ “W J " dia` to` ptai'sman to` e“poiken dia` tou'to turannei'tai.” Kai` meta` wJri´tsan ojligh` n ei«da ajnoikta` " ta` " po´ rta" oJpou' hJ jAga´ ph ejse´ bhken ajpe´ sw me` to` n Po´ qon⭈
N 370
N 375
N 380
N 385
N 390
N 395 N 397
N 366 o”la] ajlla` P ptu´ ousin P : pi´nousin N tou' sto´ mato´ " tou P N 367–68 kai` parekto` " tou' hJliakou' kaqa´ rion w”sper pa´ go" P N 368 pa´ go" P : pa´ gwn N edd. N 369 ejsth´ keton P ajna´ gwn P (cf. E) : ajnw´ ghn N N 370 ejpa´ nw P to` n ajna´ gonta scripsi coll. E : to` n ajna´ gona P : to` ajnw´ geon N to` post ma´ rmaron add. P leka´ nhn N N 371 me` ] wJ" ejme´ nan P N 372 na` zh' kai` na` kinh'tai] zh' kai` ajnakinei'tai N N 373 ta´ ca ante ta` du´ o add. N basta' e“mprosqen pro` " P N 374 carth` n kai` e“grafen ou»toi e“grafon sti´coi N gramme´ na scripsi coll. E V : gra´ mmata P N 375 a‘"] wJ" P o”pou1 scripsi : oJpou N : oJpou' P a”" pa´ scei oJpou' me qe´ lei P me` 2 om. N pa´ cei N, corr. Wa N 376 na` qlu´ betai oJpou' oJmmatia (oJmmata a.c.) guri´zei eij" P N 377 mou om. N N 378 th` n2] to` n N ejkatedika´ sqhn p.c. P N 380 gra´ mmata e“blepa P : gra´ mma e“grafa N kai`2 om. N ajne´ gnwqe´ P N 381 e“sion P mh` kai` ejme´ nan poih´ sei P N 382 me post katadika´ sei add. P oJ poqoerwtokra´ twr P : ejrwtokra´ th" e“rw" N N 383 ejkei'non scripsi (cf. ejkei'no" E) : ejkei'na P : eij" de` to` n N o”pou scripsi : oJpou N : oJpou' P ejpro´ seca´ ton] proei'con P N 384 kai` om. P to` da´ kruon P : ta da´ krua N stala´ ssh scripsi : sta´ zei P : stala´ soun N N 385 kai` P (cf. E) : tou' N cocla´ zh scripsi (cf. coclw´ zh E) : kocla´ zh P : cocla´ zoun N na` kai´h wJ" to` kami´nin P (cf. E) : na bra´ zoun wJ" to` kaka´ bhn N N 386 pa´ nw P kh'te P ojfi´din P : to` fi´din N N 387 o”lhn] a“llon P N 388 eij" to` me´ topo´ n tou pukno` n to` sto´ man tou' ajnqrw´ pou P pukno` n P : pikto` n N N 389 kai` om. P na` ei«pa] i“ppo" (i.e. ei«p∆ wJ") P ejfo´ naze´ P N 390 fobou'mai P ta` ] to` P kai` pikra` timwrh'sai P N 391 na` P : wJ" N le´ gh" P uJpepa´ ghn P : uJpeta´ ghn N : ejpepa´ ghn tempt. Ch N 392 kai` sunqrhnh` n N a“nqrwpon P (cf. E V) : a“gwron N N 393 tou] mou P to` n P : th` n N N 394 e“rw" me le´ gei oJ fu´ lax mou P meteme´ nan P N 395 wJ" dia` ] o”lon P ejpoi´hse" P dia` tou'to P : di∆ au”ton N timwrh'sai P N 396 w”ran P ojligh` n P (cf. E) : ojli´ghn N ei«da] ble´ pw P
PANAGIOTIS A. AGAPITOS e“rw" ejxe´ bhn ajpekei' kai` le´ gei ton ejkei'non oJpou' me` ejfu´ lassen ejme´ nÚ ““Epar∆ ton, e“la ajpe´ sw.” Kai` ejkei'no" to` n ejrw´ thsenÚ “Ti´ le´ goun wJ" dia` tou'ton…” Ej kei'no" ajpekri´qhken krufw'" na` mh` to` ma´ qwÚ “H J jAga´ ph mesiteu´ ei ton, ejgguei'tai ton oJ Po´ qo", kai` sumpaqei' ton oJ “Erwta", ti´pote mh` luph'tai.” Kai` ei«pen moiÚ “Peripa´ thse, mh` qli´besai ajpe` tw´ ra.” Ej mpai´nw ajpe´ sw met∆ aujtou´ ", plh'qo" polu` n euJri´skw, «hsan muria´ de" a“nqrwpoi kai` hJ di´kh tou" toiau´ th di∆ ajga´ phn, e“rwto" storgh´ n, dia` po´ qou diakrisi´an⭈ kai` me´ sa eij" tou´ tou", fi´le mou, ma´ qe to` ti´ ejxeni´sthn, to` ei“dasin ta` ojmma´ tia mou ejxaporei' to oJ nou'" mou. “Erw" trimorfopro´ swpo" ka´ qhtai eij" to` n qro´ non, to` prw'ton tou to` pro´ swpon bre´ fo" mikrou' paidi´ou, aJpalosa´ rkou, truferou', kai` ei«cen xanqh` n th` n pla´ sin, eja` n to` ei«de", na` ei«pe" ejk panto` " ce´ ria kalou' zwgra´ fou tecni´tou to` ejsto´ rhsan, ye´ go" oujde` n basta´ zei⭈ to` deu´ teron ejfai´neton wJ" me´ sh" hJliki´a", na` e“ch to` ge´ nin stroggulo´ n, th` n o“yin wJ" to` cio´ ni⭈ kai` to` ajp∆ ejkei´nou pro´ swpon ge´ ronto" na` ei«de" o“yin, su´ nqesin, sch'ma kai` koph` n kai` pla´ sin ajnalo´ gw"⭈ kai` to` me` n prw'ton pro´ swpon ei«cen ejx oJloklh´ rou ta` ce´ ria, ta` poda´ ria kai` to` a“llon tou to` sw'ma, ta` de` ajp∆ ejkei´nou pro´ swpa mo´ non ajpo` tou` " w“mou". Ej qw´ roun ta o”ti ejkei´tontan wJ" h«san kat∆ ajxi´an, e“blepa th` n trimo´ rfwsin, e“legaÚ “Ti´" oJ pla´ sth" ⬍kai`⬎ ti´ to` xenoca´ ragon to` ble´ pw, ti´ e“nai ejtou'to… Ti´" na` me` ei“ph to` qewrw', ti´" na` me` to` eJrmhneu´ sh, ti´" a“nqrwpo" filo´ kalo" na` me` to` ajnadida´ xh…” Kai` ejno´ sv eij" te´ toian me´ rimnan oJ nou'" mou ejtriokopa'ton, oJka´ pote kai` hJ zh´ thsi" gi´netai hJ ejdikh´ mouÚ “⬍ O J ⬎ ajntista´ th" a‘" ejlqh',” fwna´ zei oJka´ ti" e“rw". Gi´netai ta´ xh" merismo` " ajpe´ dw mou kai` ajpe´ kei, le´ gei me oJpou' me ejfu´ lassenÚ ““Ela na` proskunh´ sh".”
139 N 281
N 285
N 290
N 295
N 300
N 305
N 310
N 282 ejme` n] ejme` P kai` ante e“la praeb. P N 283 kai` om. P hjrw´ thsen P ti´ le´ goun wJ"] ti´ le´ gousi iter. P N 284 ajpekri´qhken] uJpekri´qhn ton P krufa` P ma´ qw Wa (cf. E) : ma´ qh N P N 285 hJ om. P oJ po´ qo" to` n ejgkuh'tai P N 286 oJ “Erwta"] ajpo` e“rwta P N 287 ei“pe me peripa´ thsen P N 288 sebai´nw P aujtou` " P : aujth'" N N 289 h«san] wJsa` n P oiJ dh´ kh na` e”n tw'n o”lwn P N 290 storgh` n om. P diakrisi´an P : di∆ ajkrasi´an N N 291 me´ sa] ejme´ nan P tou´ tou"] aujtou` " P ejxeni´sqhn P N 292 kai` ti´ ei“dasi ta` jma´ tia´ mou P N 293 strwtrimorfwpro´ swpo" ejka´ qhto eij" qro´ non P N 294 wJ" bre´ fo" N : w”sper coni. Wa N 295 aJpalosa´ rkin trufero` n na` ejxanqh' eij" th` n pla'sin P truferou' Lt coll. E : trufero` n P : foberou' N N 296 eja` n to` ei«de" na` ei«pe" P : tou' nato` n i“de" N ce´ ria kalou' zwgra´ fou] ka´ llou" kalou' tecni´tou P N 297 tecni´tou Wa : tecni´te" N tecni´tou to` ejsto´ rhsen] zwgra´ fou cai´ria iJstorikou' P ye´ go" Wa : yi´go" N kai` yo´ gon ouj basta´ zei P N 298 ejfai´neto ejk P ge´ neion P to` cio´ ni] sito´ crou" P N 300 ajp∆ ejkei'qen P ei«den P N 301 koph` n . . . pla´ sin] skopo` n . . . glw'ssan P N 303 tou kai` post ce´ ria add. P to` a“llon] o”lon P N 304 ta` . . . pro´ swpa scripsi : to` . . . pro´ swpon N P ajp∆ ejkei'qen P ejqew´ roun P ta om. P ejkei´tenton P N 306 h“lega ti´" hJ pla´ si" P N 307 kai` suppl. Ch coll. E ejxenoca´ ragon P e“nai ejtou'to] e“ni tou'to P N 308 me1 om. P eijph' P N 310 kai`—nou'" mou] oJ nou'" mou eij" toiau´ thn me´ rimnan P ejtriokopa´ ton N P (cf. ejtriekopa´ ton E) N 311 ijdikh´ P N 312 oJ suppl. Wa to` n ajntista´ thn P ejlqh' P : e“lqh N N 313 ta´ xh" scripsi : ta´ xi" N P edd. merismw'" P : me´ rou" mou N mou om. P ajp∆ ejkei'qen P N 314 le´ gei me] kai` lo´ gou" P ejfu´ laxen P
140
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE Ej se´ bhn eij" to` n “Erwtan, pi´ptw eij" th` n gh'n ojmpro´ " tou, kai` hjrxa´ mhn meta` da´ krua mou na` klai´w kai` na` to` n le´ gwÚ
Eij" u”pnon tou to` n “Erwta oJ Li´bistro" prospi´ptei, prospi´ptei kai` parakalei' meta` pollw'n dakru´ wn. ““Erw", aujqe´ nta basileu', de´ spota gh'" aJpa´ sh", tw'n ajnaisqh´ twn ajrchge´ , tw'n aijsqhtw'n kata´ rca, pa´ sh" yuch'" ejreunhta´ , tou' po´ qou dikaiokri´ta, kai` th'" ajga´ ph" sunerge´ , th'" uJpolh´ yew" fi´le⭈ a‘n ajpo` ajnaisqhsi´a" mou th` n ei«ca pro` " ejse´ nan katefroni´sqh" ajpo` ejme´ n, de´ spota poqokra´ twr, mh` ejxeristh'" to` ptai'sma mou, to´ son mh` to` kakw´ sh", gnw´ rise, h“moun cwriko` " kai` suggnwmo´ nhse´ to⭈ ajrkei' to` me` ejfobe´ rise", ejle´ hse me ajpe` tw´ ra, ªnºa` oJmo´ sw na` ei«mai dou'lo" sou o”lo" tou' oJrismou' sou, li´zio" tou' qelh´ mato" kai` tou' prosta´ gmato´ " sou.” Kai` to´ te ajfou' to` ejplh´ rwsa to` to` n ejpareka´ loun, le´ gei meÚ “ Ej gei´rou ajpo` tou' nu'n dia` mesitei´a" tou' Po´ qou⭈ dia` th` n ejggu´ hsin, gnw´ rize, th` n ei«ca ejk th` n jAga´ phn, splagcni´zomai´ se ajpo` tou' nu'n, ejlew' kai` sumpaqw' se⭈ to` e“ptaise" ouj yhfi´zw to, ajmnhmonw' eij" ejkei'non, kai` ajpo` tou' nu'n para´ labe ajga´ phn eij" to` n nou'n sou kai` po´ qon ko´ rh" hjqikh'", ejrwtoexhrhgme´ nh", po´ qon Roda´ mnh" qugatro` " Crusou' tou' basile´ w".” Ej proshkw´ qhn ejk th` n gh'n, ejproseku´ nhsa´ ton, ei«da frikto` n musth´ rion, fi´le mou, eij" ejkei'non⭈ th` n mi´an fwnh` n ejme´ rizan ta` sto´ mata ta` tri´a, ejla´ lei ou»to" kai` na` le` " ejfw´ nazen ejkei'no", økai`Ø h“koue" øto` Ø te´ lo" th'" fwnh'" ejk tw'n triw'n to` sto´ ma, kai` aJplw'" oujk ei«ce" th` n ajrch´ n, oujde` to` te´ lo" pa´ lin, to` n lo´ gon to` n ejfw´ naxen po´ qen na` to` n eijka´ zh". jApe´ dw tou kai` ajpe´ kei tou tou' ejrwtotriprosw´ pou du´ o gunai'ke" i”stanto kale` " eij" carakth'ran, ⬍kale` " eij" ei«do" kai` eij" koph´ n, para´ xene" th` n o“yin⭈⬎ hJ mi´a stefa´ nin na` forh' margarita´ rin o”lon, a“spron wJ" to` cio´ nin to` kalo` n to` ajko´ mh oujk ejpath´ qh⭈
N 315 N 316 N 466a N 466b N 317
N 320
N 325
N 330
N 335 P 217 P 218 P 220 P 219 P 221
P 225
N 315 ejmpro´ " P N 316 kai`1 om. P meta` N P : me` ta` dist. Hs dakru´ wn P N 466a-b hanc rubricam atramento exaratam hic praeb. N P, sed post N 466 (艐 P 2797) perperam transp. Hs N 466b prospi´ptei kai` om. P N 317 aujqe´ nth P tw'n aJpa´ ntwn P N 318 ajnaisqh´ twn P : ajnesth´ mwn N aijsqhtw'n P : ajrchgw'n N N 320 th'" ajga´ ph"] gh'" aJpa´ sh" P sunerge` P : su´ nerge N N 321 a‘n om. P th'" post ajpo` add. P ajnaisqhsi´a" P : nhstei´a" N th` n ei«ca pro` " ejse´ nan] kai` ajpo` cwrismou' mou P N 322 katefroni´sqh" Wa : kafroni´sqh" N : katefronh´ qh" P ajp∆ ejmou' P aujtokra´ twr P N 323 ejxorisqh'" P kaki´sh" P N 324 om. P N 325 ejle´ hso´ n P ajpo` P N 326 na` suppl. Wa : om. P w“mosa P o”lo" N : dou'lo" P oJrismou' P : oJrismo´ " N N 328 ajfou' to` ] ajfo´ tou P ejparaka´ loun P N 330 th` n e“ggusin ejgnw´ rize P ejggu´ hsin corr. Kr : ejggu´ wsin N edd. ei«ca] oi«da P N 331 kai`] se P N 332 e“ptaise" P : e“pese" N kai` ante ajmnhmonw' add. P ejkei'no P N 333 kai` om. P N 334 hjqikh'" P : sunqhkh'" N ejrwtoejxhrhme´ nh" P P 217–302 om. N P 220 huc traieci coll. E V P 221 kai` . . . to` seclusi coll. V P 223 to` n2 scripsi (cf. tou E et oJpou V) : oujk P : o”n coni. Ma P 224 tou1 corr. Ma (cf. E) : to P P 225 unum excidisse versum videtur in P, quem e.g. supplevi ex E (etiam cf. kalai'" eij" h”do" eij" morfh` n eij" su´ nqeshn kai` eij" ka´ lo" V) P 226 o”lon scripsi coll. E V : mo´ non P P 228 aujth` scripsi coll. E V : a“llh P : a“llo coni. Ma
PANAGIOTIS A. AGAPITOS hJ a“llh eij" to` kefa´ lin th" kai` aujth` stefa´ nin ei«cen, ko´ kkina lucnita´ ria na` tsou´ zoun wJ" hJ flo´ ga⭈ oJlo´ aspra h«n ta` rou'ca th" th'" mia'" wJ" to` bamba´ kin, ⬍th'" a“llh" oJloko´ kkina na` la´ mpoun wJ" oJ h”lio"⭈⬎ kai` eij" to` pro` " e”nan go´ naton tou' ejrwtokratou'nto" tw'n du´ o ta` ce´ ria kei´tountai ejpa´ nw eij" ta` dexia´ twn, o”rkou shmei'on ejrwtiko` n eij" euju¨ polhyi´an. “Akouson ti´ me` eJrmh´ neusan dia` ta` " du´ o gunai'ka"Ú “Ble´ pei" th` n oJloko´ kkinon ejkei´nhn th` n wJrai´an oJpou' eij" to` me´ ro" i”statai †tou' “Erwto" to` dexio´ n†; Le´ goun thn hJ jAlh´ qeia, ejkei´nh ojmnei' eij" ejtou'ton na` mh` yeusqh' ⬍ta` ⬎ uJpo´ scetai dia` po´ qon eij" to` n ko´ smon⭈ oJpou' i”statai eij" to` n ajristero` n le´ goun thn Dikaiosu´ nhn, ojmnu´ ei kai` aujth` to` n “Erwta to` di´kaion na` fula´ ssh, na` mh` dia` proswpo´ lhyin polla´ ki" parakri´nh⭈ e“ni aujth` oJloko´ kkino" kai` oJlo´ leuko" ejkei´nh, kai` wJ" kru´ stallon parei´kase ejkei´nhn ejk ta` ne´ fh, wJ" flo´ ga tau´ thn ejk th` n gh'n dia` to` ajyeudh´ goro´ n tou⭈ i“de, a“nqrwpe, qau´ maze kai` ta` " morfa` " tw'n tri´wn. jAfou' ajpedw` metastaqh'" kai` uJpa´ gh" kai` pare´ kei, qe´ lei" sebei'n eij" tou' “Erwto" ta` be´ lh na` ojmo´ sh" kai` ajpe´ kei na` ta` didacqh'", na` i“dh" kai` to` n ma´ ntin, oJpou' se` qe´ lei ajfhghqei'n øto` Ø ti´ e“nai to` qe´ lei" pa´ qein kai` po´ te th` n ajga´ phn sou qe´ lei" ajpokerdi´sein.” O J pou' me` sunetu´ cainen ajfh'ke, metesta´ qh, pa´ lin qewrw' to` n “Erwta, pa´ lin aujto` " me` le´ geiÚ “Li´bistre, ti´ sth´ kei" kai` qewrei'"… “Agwme na` ojmo´ sh". jAga´ ph, su´ ton e“pare, Po´ qe, para´ labe´ ton, to` ejggutiko´ n tou poih´ sete, to` be´ baion a‘" ojmo´ sh, a‘" toxeuqh' øejkØ to` n e“rwta tou' po´ qou th'" Roda´ mnh".” Eujqu` " to` n proseku´ nhsa kai` ajpe´ kei metesta´ qhn⭈ oJ Po´ qo" h«ton ajpedw` kai` ajpe´ kei mou hJ jAga´ ph, eij" to` kelli`n ajph´ gamen o”pou h«ton to` oJrkwmo´ sin⭈
P 230 oJlo´ aspra scripsi (cf. oJla´ spra E V) : oJlo´ lampra P bamba´ kin scripsi (cf. bampa´ gkhn V) : stefa´ nin P (et E) P 230 post hunc versum lacunam unius versus suspicor, quam e.g. supplevi (insuper cf. P 235/242 et V ad v. P 229) P 231 fort. post hunc versum statuenda est lacuna duorum fere versuum P 234 a“kouson scripsi (cf. a“kouse E et a“kouson V) : h“kouse" P P 235 ble´ pei" scripsi coll. E V : ble´ pw P P 236 cf. oJpou' eij" to` me´ ro" to` dexiw'n h“steke tou' kratou'nto" E P 237 ejtou'ton scripsi (cf. ejkei'non E et au“thn V) : to` n e“rwta´ mou P P 238 na` Ch coll. E V : i”na P ta` supplevi e V (cf. to` E) P 249 to` 1 seclusi P 251 an ajfh'ke me kai` ejsta´ qh? (cf. ajfoi´ke me kai` ejdie´ bh E) P 253 sth´ kesai P, corr. Ma P 255 poih´ sete scripsi (cf. pi´setai E) : poi´hsai P a‘" ojmo´ sh scripsi (cf. na` wJmw´ sei E) : hJ oJmwsi´a aj" ge´ nh P P 256 a‘" suppl. Ma ejk seclusi coll. E tou' po´ qou] an th'" ko´ rh"? (cf. E) P 260–62 minio exarati in P P 261 post hunc versum lacunam aliquorum versuum suspicor, in quibus Libistrus verba inscriptionis refert; e.g., cf. oJpo´ ci po´ qon kai` filh´ an aj" tre´ ch eij" to` n ma´ nthn | sto` n e´ rotan to` n qaumasto` n to` n po´ qon th'" ajga´ ph" | hJ de ejkfeu´ gi ajpautou'" ajli`" ejkh´ non e“ci V
142
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE ⬍ ⬎ Pa´ li a“kouson to` di´quron ti´ ei«cen iJstori´an⭈ ei«ce to` n “Erwta gumno` n ejpa´ nw iJstorisme´ non, to` e”nan tou ce´ rin na` krath' spaqi`n hjkonhme´ non, to` a“llon oJloko´ kkinon aJptome´ nhn lampa´ da. jApeda` ma´ qe ejk th'" grafh'" to` h«ton paraka´ twÚ ““Erw" ajkatapo´ neto" oujranobuqofqa´ no".” “Loipo` n oujk e“ni oJpou' e“laqen to` n “Erwta eij" to` n ko´ smon, e“cei †plhroforh´ qhsai†,” pa´ lin le´ gei me oJ Po´ qo", “oujk e“ni oJpou' ejxeglu´ twsen th` n ejrwtotaxi´an.” jApe´ kei ejmetesta´ qhmen kai` ajph´ gamen ajpe´ sw eij" to` kelli`n to` ejrwtiko` n th'" poqoorkwmosi´a"⭈ euJri´skw eij" crusoko´ kkinon ejpa´ nw ajnalo´ gin ptero` n na` kei'tai tou' “Erwto" kai` to´ xon gemisme´ non, kai` me´ sa eij" aujto` carti´n, ei«ce tou` " lo´ gou" tou´ tou"Ú
“ Ej gw` ei«mai ⬍oJ⬎ no´ mo" tou' “Erwto", økai`Ø tou'to e“ni to` ptero´ n mou kai` tou'to e“nai to` doxa´ rin mou, kai` ojmnu´ ete oiJ pa´ nte" li´zioi na` ei«ste dou'loi tou, na` mh` to` n ajqeth'te. Ti´ plana'sqe ejxaporw' kai` ti´ øe“naiØ to` n ajqetei'te⭈ pou' na` to` n ejglutw´ sete… Fri´ttw o”ti feu´ gete´ ton. ‘An petasqh'te eij" øto` nØ oujrano´ n, ptero` n e“cei kai` fqa´ nei⭈ a‘n katabh'te eij" ⬍qa´ lassan, gumno` " wJ" to` n qewrei'te, kai` katafqa´ nei eij"⬎ a“busson kai` oujk ejglutw´ nete´ ton⭈ a‘n de` i“sw" pa´ lin eij" th` n gh'n kosmoperipatei'te, qewrei'te ⬍kai`⬎ to` to´ xon tou, polla` stoca` doxeu´ ei, kai` oujk e“ni oJdo` " na` fu´ gete th` n ejrwtotaxi´an. Loipo` n ejpifwnou'mai sa" oJpou' ei«stai ajpo` to` n ko´ smon, doulw´ nesqai⬍eij"⬎ to` n “Erwta, kai` oJpou' to` n qe´ lei ojmo´ sein, a‘" e“ni be´ baion to` lalei', mh` oJrkoparabath´ sh.” Kai` paraka´ tw e“grafen ta` ejk th` n grafh` n ejkei´nhnÚ “Be´ lh kai` to´ po" kai` monh` th'" poqoorkwmosi´a".” jAne´ gnwsa´ thn th` n grafh´ n, ei«da th` n iJstori´an, aJplw´ nw tau'ta eij" to` ptero` n kai` ejpa´ nw eij" to` doxa´ rin, kai` ei«paÚ “Ma` tou'to to` ptero´ n, ma` to` eu“stoco´ n sou to´ xon, doulw´ nomai eij" to` n “Erwta, liziw´ nomai eij" to` n Po´ qon, pisto´ " th" na` ei«mai ajpo` tou' nu'n th'" Ej rwtikoaga´ ph".”
P 265
P 270
P 280
P 283a P 283b P 285
P 290
P 295
P 267 oujranobuqofqa´ nou" P, corr. Ma P 269 plhroforh'sai se coni. Ma: an plhroforh´ qhse? P 271 ejmetesta´ qhsan P, corr. Ma P 275 aujto` Ma (cf. aujto` n E) : aujta` P P 276a-b rubrica lacunosa atramento exarata in P, quam e.g. supplevi ex E P 277 oJ supplevi (cf. oJno´ mosttou' E) kai` seclusi P 279 li´zioi Ch : su´ zioi P ei«ste scripsi coll. E : h«te P P 280 e“nai seclusi to` n scripsi : to` P P 281 ejglutw´ sete scripsi coll. E : ejglu´ sete P P 282 to` n seclusi P 283a-b lacunam e.g. supplevi ex E P 284 a‘n Ma : eja` n P kosmoperipatei'te scripsi : ko´ smon peripatei'te P P 285 kai` supplevi coll. E stoca` P : eu“stoca coni. Ch P 287 sa" . . . ei«stai scripsi coll. E V : se . . . h«sai P P 288 eij" suppl. Ma (cf. E V) P 289 a‘" put. Ma (cf. E) P 291 be´ lh kai` to´ po" scripsi coll. E : kelli`n po´ qo" P : Kelli`n tou' Po´ qou coni. Ma monh` scripsi coll. E : monh` n P Ma P 295 boulo´ nomai P, corr. Ma ajluzio´ nomai P, corr. Ma
PANAGIOTIS A. AGAPITOS
143
“Wmosa kai´, ajfo´ tou w“mosa, tau'ta kai` oJ ma´ nti" h«lqen, oJ ma´ nti" to` n ⬍me` ⬎ eJrmh´ neusan na` me` i“dh eij" to` n o”rkon, kai` tau'ta ejn prw´ toi" le´ gei me lo´ gou" kai` suntuci´a"Ú
P 298a
Ta` qe´ lei pa´ qei ⬍oJ⬎ Li´bistro" oJ ma´ nti" tou' ta` le´ gei. “Li´bistro", gh'" latinikh'" a“rcwn, topa´ rch" me´ ga", pollw'n pragma´ twn a“nqrwpo", rh´ ga" pollw'n ajnqrw´ pwn, me´ llei tou' to´ pou na` genh' tou' gonikou' tou xe´ no" dia` Roda´ mnhn th` n kalh´ n, th` n ejrwtikopo´ qon⭈ ⬍despo´ th" me´ llei na` genh' tou' jArgurou' tou' Ka´ strou,⬎ Cruso´ n, pate´ ra to` n lampro` n th'" ko´ rh" th'" Roda´ mnh", na` diadecqh' ejk th` n cw´ ran tou kai` na` kosmokrath´ sh⭈ meta` de` cro´ nou dia´ sthma na` ca´ sh th` n Roda´ mnhn uJpo` gunai´ka" ponhra'", th'" kakoma´ gou grai´a", kai` na` e“bgh eij" ajnazh´ thsin th'" ko´ rh" th'" Roda´ mnh"⭈ di´cronon qe´ lei perpatei' na` uJpa´ gh na` th` n eu”rh⭈ kai` ajfou' th` n eu”rei ajpo` kalou' fi´lou tou sunergi´an, eij" a“llon e”nan dia´ sthman na` paradra´ mh cro´ no" kai` pa´ lin Ka´ strou tou' jArgurou' na` gi´netai despo´ th" kai` ko´ rhn th` n para´ xenon na` th` n sunapoqa´ nh.” Ei«pen ejkei'no" oJ kalo` " oJ ma´ nti" oJ prognw´ sth", ejla´ lhse ta` me´ llonta kai` oJka´ pote ajnespa´ sqhn ajpo` tou' ojnei´rou th` n plokh´ n, fi´le mou, tou' tosou´ tou.
O J nou'" mou ei«cen tarach` n kai` fo´ bon hJ yuch´ mou, me´ rimnan hJ kardi´a mou kai` klo´ non ajll∆ oJpo´ son⭈ w”ran casme´ no" perissh` n na` kei´twmai na` ble´ pw to` po´ te na` e“lqh oJ logismo` " to` n ei«ca a‘n oujk ejca´ qh. O J ka´ pote ejpanh´ fera kai` h«lqan ta` logika´ mou, ajne´ bhn ajpo` to` n buqo` n tou' ojnei´rato" oJ nou'" mou, e“rriya ajpo` ta` ojmma´ tia mou to` n u”pnon parauti´ka⭈
N 348 N 350
P 297 ma´ nti" scripsi coll. E : protoma´ nti" P h«lqen e versu sequenti huc traieci coll. E P 298 me` suppl. Ma eJrmh´ neusan scripsi coll. P 234 : eJrmh´ neuse P P 298a le´ gei me scripsi coll. E V : le´ gw ton P P 298b atramento exaratus in P oJ supplevi coll. E P 299–302 minio exarati in P P 300 rh´ ga" Ch (cf. ku´ ri" E) : me´ ga" P P 301 versum correxi coll. E : me´ llei tou' to´ pou tou tou' gonikou' na ge´ nhtai kai` xe´ no" P P 302 post hunc versum lacunam in N P suspicor, quam e.g. supplevi ex E (etiam cf. despo´ th" a“lo" na genh` " gh` n ajrgirou' tou' ka´ strou V) N 336–39 om. P N 337 diadecqh' scripsi (cf. eu“gei E) : diadecqei`" N kosmokrath´ sh scripsi (cf. kosmwgureu´ sh E) : kosmwkrath´ " N N 338 de` cro´ nou] dicro´ nou E, an recte? (cf. S 1375–80 et S 1713–15) N 340 eu“gh P : eu“gh" N N 341 di´cronon] du´ o cro´ nou" P qe´ lei P : qe´ lei" N peripatei'n P uJpa´ gh . . . eu”rh P : uJpa´ gh" . . . eu”rh" N N 342 eu”rei P : eu”rei" N tou P : sou N N 343 e”na P cro´ non P N 344 gi´netai P : gu´ nese N N 345 sunapoqa´ nh P : sunapoqa´ nh" N N 346 ei«pen—kalo` "] tau'ta ei«pen ejkei'no" P N 347 kai` oJka´ pote ajnespa´ sqh (corr. Ma) P (cf. kai` oJka´ pote ajntista´ qh E) : kai` ejxu´ pnhsa eujqe´ w" N P 311 praeb. P E : om. N N 347a-b atramento exarati in N: om. P N 347a ojneiroplhsi´a" N, corr. Lt N 348 ei«cen] na` e“ch P N 349 klo´ non] fo´ bon P ajll∆ oJpo´ son scripsi coll. a“llw po´ son E : ajlla` po´ son P : ajllo` po´ qon N N 350 casme´ no" P : cai¨me´ no" N N 351 to` ante po´ te om. P ejca´ qhn P N 352 post N 353 praeb. P N 352 kai` h«lqan] h«lqon P N 354 parauti´ka] peri` bi´an P
144
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE ejka´ tsa kai` ejkloni´zeton oJ nou'" mou ajpe` to` n fo´ bon, ajneyhla´ foun dia` desmo` n mh` e“ch oJ sfo´ ndulo´ " mou, e“blepa a‘n ei«mai wJ" me` e“surnan ejkei'noi desmwme´ no". «Hton oJ nou'" mou eij" “Erwtan to` n tri´morfon ejkei'non, ejsko´ poun ta` " para´ xena" ta` " du´ o plagina´ " tou, to` n hJliako` n ejqau´ maza kai` th` n kalh` n fiski´nan, e“blepa na` ei«pe" aijsqhta` th` n poqoorkomwsi´an⭈ h“moun eij" tou'to kai` eij" aujto` kai` eij" to` a“llon to` pare´ kei, h“qela ejkei'na na` skopw' kai` ejtou'ta na` fronti´zw, to` a“llo na` e“cw me´ rimnan, na` mh` me` la´ qh ejkei'non, kai´, pi´steuse´ me, eij" eJkato` n ejko´ pteton oJ nou'" mou. Kai` te´ w" meta` w”ran perissh` n mhnw' to` n suggenh' mou, ejkei'non o”pou mou' e“legen ta` du´ natai hJ ajga´ ph⭈
N 355
N 358 P 322 N 467
N 470
O J Li´bistro" to` o“neiron le´ gei to` n suggenh' tou.
N 475
kai` a”ma to` n ei«da, ei«pa ton, oujde` ejcaire´ thsa´ tonÚ “ Ij de` " to` ti´ me` ejgi´neton th` n cqesinh` n eJspe´ ran. “Eplasen oJ “Erw" o“neiron to` na` ei«pe" o”ti ejble´ pw, kai` e“cei oJ nou'" mou me´ rimnan kai` ojdu´ nhn hJ yuch´ mou⭈ e“plasen oJ “Erw", suggenh', pla´ sma frikto` n ojnei´rou, to` ajko´ mh ble´ pw aijsthtw'" kai` ejntre´ cei eij" ojfqalmou´ " mou.” Kai` ejkei'no" me` le´ geiÚ “Li´bistre, de´ spota, gh'" topa´ rca, gh'" kosmokra´ twr th'" ejmh'", eijpe´ , ajfhgh´ sou me´ to.” Kai` hjrxa´ mhn th` n ajfh´ ghsin, fi´le mou, na` th` n le´ gw tou' ojnei´rato" tou' ejrwtikou' to` n suggenh'n mou ejkei'non⭈ kai` oJpo´ tan h«lqa eij" to` o“noma th'" ko´ rh" th'" Roda´ mnh", ei«pen moiÚ “Kra´ tei, Li´bistre, di∆ aujth` n th` n suntucai´nei" ei“dhsin e“cw kai` a“kouse, tw´ ra kairo` " parh'lqen. Nomi´zw e“ni di´cronon kai` eij" cw´ ran ajpedh´ moun dia` to` n rhgodespo´ thn th" kai` kosmoarchgo´ n th", o”ti ajpedh´ mhn eij" aujth` n th` n ko´ rhn h’n me` le´ gei"⭈ kai` sw´ pase, mh` qli´besai, ti´pote mh` luph'sai, ejpei` kai` ma´ nti" a“nqrwpo" proei'pen soi to` me´ llon. jAlh´ qeia th` n patri´da sou kai` o”lou" tou` " ejdikou´ " sou
P 331 N 476
N 480
N 485 N 487 P 345 P 347 N 488 N 490
N 355 praeb. N E V : om. P N 356 ejyhla´ foun P dia` desmo` n P : diadesmo` n N sfo´ ndulo" N : tra´ chlo" P, an recte? N 357 e“blepon ejkei' wJ" me e“surnan a‘n h«me ejmplegme´ no" P desmwme´ no" scripsi (cf. ejmplegme´ no" P) : desmwme´ non N edd. N 358 e“rwta P tri´morfon P : trimo´ rfwn N P 322 praeb. P E V : om. N N 467 ejqau´ mazon P fiski´nan P : fiski´an N N 468 e“blepon P aijsqhtw'" P N 469 h“mhn P pare´ kei P : pare´ xw N N 470 ejkei'na . . . ejtou'ta] ejtou'to . . . ejkei'no P kai` om. P N 471 ejkei'no P N 472 pi´steuso´ n moi P o”ti ante eij" add. P N 473 te´ w" om. N w”ra P mhnu´ w P suggenh´ n P N 474 o”pou scripsi : oJpou N : oJpou' P mou'] me P h“lege P du´ natai P (cf. du´ natai V et du´ netai E) : di´eta N N 475 om. P P 331 praeb. P E V : om. N N 476 ei«de" P to` ante ti´ om. P ejge´ neto P eJspe´ ran P : hJme´ ran N N 477 om. P ei«pe" Lt coll. E : i“de" N N 478 post N 479 praeb. P kai` e“cei scripsi coll. E : to` e“cei P : na` e“cei N me´ rimnan Lt : me´ rimna P (cf. me´ rimnan E) : o“neiron N N 479 suggenh' P : suggenh´ n (-gk- a.c.) N pla´ sma P (cf. E) : pra´ gma N N 480 aijsthtw'" Lt coll. E (etiam cf. ejstita` V) : aijsqhtw'" P : eJsthkw` " N N 481 le´ gei] ei«pe P li´bestre N de´ spota om. P N 482 gh'"] th'" P ajfhgh´ son me P N 484 ojnei´rou P suggenh' P N 485 h«lqa Lt coll. E (etiam cf. h“lqa V) : h«lqon P : h«lqen N to` om. P N 486 ei«pe P th` n suntucai´nei" P : th'" suntece´ nh" N N 487 a“kouson P oJ ante kairo` " add. P P 345–47 praeb. P (etiam cf. E V, qui vel similia vel diversa praeb.) : om. N P 346 kosmoarchgo´ n scripsi : ko´ smou ajrchgo´ n P N 488 siw´ phsai P ti´pote] lu´ bistre P N 489 proei'pe P N 490 ijdikou´ " P
⬍“Oneiron pa´ lin deu´ teron Libi´strou polupo´ nou.⬎ Ej fa´ nh me eij" para´ xenon ejse´ bhn peribo´ lin, eij" muriodendrofu´ teuton, øeij"Ø o”lon ejxhnqisme´ non, to` ejko´ smhsen “Erw" basileu` " kai` ejla´ mprunen oJ Po´ qo" kai` hJ jAga´ ph ejkateka´ llunen me` ta` para´ xena´ th"⭈ ejperiepa´ toun, e“blepa to` te´ toion peribo´ lin, e“treca eij" tou'to to` dendro´ n, hjkou´ mpiza eij" ejkei'non, e“pepta eij" tou'to kai` eij" aujto´ , mete´ treca eij" ejkei'non, kai` e“kopta ejk tou'to to` n karpo´ n, a“nqo" ajpe´ kei ejtru´ goun, kai` aJplw'" to` poi'on oujde` n ei«con to` prw'ton na` cwri´sw⭈ ejdw' oJpou' bru´ sei" kai` nera´ , nerw'n ejkei' fiski'nai, ejdw' Cari´twn su´ naxi", ejkei' coro` " Ej rw´ twn⭈ me´ tron oujk ei«can ta` kala` ta` ei«cen to` peribo´ lin. Ej fa´ nh me h«ton kai` polu` n to` ejrwtoperibo´ lin, kai` ejdw' ajnatre´ cw na` to` ijdw' kai` ejkei' na` to` gureu´ sw,
N 505
N 510 N 511 P 374 P 376 N 512
N 492 kalla` kai` eij" ajnazh´ thsin kinh´ sei" th'" wJrai´a" P N 493 ke´ rdaise th` n] ke´ rdhson h’n P uJpesce´ qhn P N 494 ou’" ejlu´ phsa" P N 495 h“rxato P : h“rxa N kai` ante h“rxato add. P fi´le—diko´ " mou] oJ suggenh` " fi´le mou oJ ijdiko´ " mou P N 496 post N 498 praeb. P dia` . . . di∆ om. P N 497 ejmpa´ zh] ba´ nh P kai` om. P eij" P (cf. E): om. N N 498 kai` lo´ gou" meta` stenagmw'n h“rxato na` me le´ gei P kai` seclusi coll. P P 359 praeb. P E V : om. N ejkto´ te scripsi coll. V : to´ te P ejnoia´ zeton scripsi coll. E V : ejbia´ zeton P P 360 praeb. P E : om. N P 360 post hunc versum lacunam unius versus suspicor in N P, quam e.g. in E 610 nisus supplevi N 499 om. P N 500 nu´ kta] nu` x P ejsko´ thsen P N 501 o’n h«con P N 502 sune´ tuca P P 364 praeb. P E V : om. N kai` seclusi coll. E V N 503 oJ e“rw" di∆ ejme` P kai` om. P nu´ kta] nu` x P N 503 post hunc versum rubricam unius excidisse versus videtur in N (et E), quam e.g. in N 692 nisus supplevi : oJnei´rwto" ajfh´ ghsi" lubi´strou polupo´ nou P (etiam cf. kai` pa´ lin a“llon o“niron to` n li´bistron ejfa´ nhn V) N 504-P 374 minio exarati in P N 504 me] mou P N 505 eij" secl. ejxanqhsme´ non P N 506 ejko´ smei P basileu` " P : basileu' N oJ po´ qo" P (cf. E) : ejde´ tzo N N 507 kai` om. P kateka´ llunen P me` ta` dist. Ma : meta` N P para´ xeno´ " N N 508 e“blepon P N 509 hjkou´ mpiza] kai` ejgkou´ piza P N 510 e“pipta ejk tou´ tou P kai` ejmete´ trecon P N 511 e“kopta om. N tou´ tou P a“nqh ejkei´nou P P 374–75 praeb. P E : om. N P 375 oJpou'] h«san coni. Ma nerw'n scripsi coll. nerw'n E : futw'n P fukei'nai P, corr. Ma P 376 praeb. P E V : om. N N 512 me´ tron Wa : me´ tran N : me´ tro" P ei«can] ei«cen P N 513 om. P N 514 kai` . . . kai` om. P ajnagureu´ w P
146
NARRATIVE PRESENTATION IN LIVISTROS AND RHODAMNE sunapantw' to` n “Erwtan, plh` n to` mikro` n to` bre´ fo", ejkei'non oJpou' ejkaqe´ zeton meta` prosw´ pwn du´ o⭈ ei«cen eij" w“mou" tou ptera´ , kai` eij" to` e”nan tou to` ce´ rin ei«cen doxa´ rin ajrguro` n kai` eij" to` a“llon tou th` n ko´ rhn, th` n ko´ rhn th` n para´ xenhn ejkei´nhn th` n Roda´ mnhn. Ei«cen th` n ko´ rhn—ba´ staze, polu´ pone kardi´a, mh` tw´ ra pa´ qh" kai` ragh'" kai` nekrwqh'" ejk po´ nou⭈ ei«cen th` n ko´ rhn—ba´ staze, yuch´ , mh` raqumh´ sh"⭈ ei«cen th` n ko´ rhn—pro´ sece, yuch´ , mh` ajnaisthth´ sh"⭈ ei«cen th` n ko´ rhn—logisme´ , mh` fu´ gh" ajpo` ejme´ nan⭈ ei«cen th` n ko´ rhn—th'" ejmh'" ajna´ pausin kardi´a"⭈ ei«cen th` n ko´ rhn—fi´le mou, to` fw'" tw'n ojfqalmw'n mou⭈ ei«cen th` n ko´ rhn—pw'" na` eijpw', ba´ sta, yuch´ , mh` ejxe´ bh". Sunapantw' to` n “Erwtan, to` n ge´ ronta, to` bre´ fo", to` bre´ fo" to` para´ doxon th'" me´ sh" hJliki´a", ejkei'non oJpou' ejkaqe´ zeton meta` prosw´ pwn du´ o. ”Enan pra'gma me` e“laqen, fi´le mou, paropi´sw, to` na` se` ajfhgh´ swmai kai` pa´ lin ejnqu´ mise´ to, kai` ajfou' plhrw´ sw to` o“neiron, na` se` ei“pw to` ti´ ei«don. Sunapantw' to` n “Erwta, to` n “Erwn kai` th` n ko´ rhn, th` n ko´ rhn kai` to` n “Erwtan tou' na` ceirokratou'ntai⭈ sunapanta' me, ble´ pei me prw'ton aujto` " ejme´ nan, ble´ pei me prw'ton, kra´ zei meÚ “Li´bistre, si´mwse´ me.” Ble´ pw, gnwri´zw ti´" e“ni, simw´ nw, proskunw' ton⭈ “Sta´ , mh` fobh'sai,” le´ gei me, “fe´ re ta` logika´ sou.” Sth´ kw, qwrw' to` n “Erwtan, th` n ko´ rhn ejntrani´zw, ejpro´ seca to` n “Erwtan, ejqau´ maza th` n ko´ rhn, e“lega mo´ no" kata` nou'nÚ ““Erw" kai` ejtou'to" e“ni, kai` mo´ no" metapla´ ttetai kai` gi´netai gunai'ka. Kai` pa´ lin di´ca to` ptero´ n⭈ pou' kru´ bei to` ptero´ n tou… ‘An de` gunai'ka, pi´steuson, to` pro´ swpon th'" ko´ rh" nika' th` n pla´ sin “Erwto" eij" th` n eujarmosti´an⭈ gunai'ka økai`Ø tou'to kai` loipo` n “Erwto" e“ni mh´ thr, pla´ si" aujth'" pro` " “Erwtan ajna´ logo" uJpa´ rcei⭈ na` zh' kai`⬍hJ⬎ mh´ thr “Erwto", th` n le´ goun jAfrodi´thn, kai` na` suntre´ ch met∆ aujto´ n, fai´netai ne´ on kai` pa´ lin⭈ tou'to to` pra'gma to` qewrw' kai` kora´ sin ⬍to` ⬎ ble´ pw,
N 515
N 520
N 525
N 529 P 390
P 393 N 530
N 535
N 539 P 404 P 405
P 410
N 515 e“rwta P N 516 ejkei'no" P N 517 tou` " w“mou" P tou P (cf. E) : ta` N to` ce´ rin] th` n cei'ran P N 518 doxa´ rion ajrgurou'n P ko´ rhn] ceio´ rhn a.c. P N 519 para´ xenon P N 520 ei«cen] ei«pe P N 521 po´ nwn P N 522–23 ei«pen th` n ko´ rhn ba´ staze yuch' mh` ajnaisqh´ tei P N 524 ajp∆ ejme´ nan P N 525 th` n ko´ rhn om. P N 526–27 om. P N 528 sunapantw' to` n ge´ ronta to` n e“rwta tou' bre´ fou" P N 529 om. P P 390–93 praeb. P E : om. N P 393 to` 1 scripsi : to` n P N 530 sunapantw' P (cf. E) : sunapanta` N to` n “Erwn kai` th` n ko´ rhn] to` n e“rwta th'" ko´ rh" P N 531 kai`] me` P tou' na` ceirokratou'ntai] kai` ejceirokratou´ san P N 532 ble´ pei] le´ gei P prw'ton N (cf. E) : prw'to" P N 533 prw'ton P (cf. E) : to´ son N lu´ bestre N me2] mou P N 534 om. P N 535 sta´ Ch coll. E : ajsi´sta N : ta` P N 536 sth´ kw Ma (cf. ste´ kw E) : sti´cw N : sh´ kw P qewrw' . . . e“rwta . . . ajnatrani´zw P N 536 post hunc versum ble´ pw to` ti´ ajna´ plasen hJ fu´ si" wJ" kali´an add. P N 537 ejpro´ seca P : pro´ seca N e“rwta P kai` ante th` n add. P N 538 ejtou'to"] plou'to" P N 539 metapla´ netai P P 404–13 praeb. P E (et partim V): om. N P 407 kai` seclusi mh´ thr scripsi (cf. mh'r E) : mu'ra P P 408 aujth'" scripsi : aujth' P P 409 hJ supplevi coll. E mh´ thr scripsi (cf. mh'r E) : mu'ra P P 411 to` addidi (cf. th` n ko´ rhn E)
PANAGIOTIS A. AGAPITOS wJ" fai´netai ejk tou' pra´ gmato", a“gamon di´ca lo´ gou.” Kai` ⬍me` ⬎ tou` " to´ sou" logismou` " tou` " ejlogomaco´ mhn, oJ “Erw" me` le´ geiÚ “Li´bistre, ble´ pei" th` n ko´ rhn tou´ thn… Th` n ble´ pei" kai` xeni´zesai kai` th` n polla` qauma´ zei", e“ni hJ Roda´ mnh tou' Crusou' quga´ thr basile´ w", th` n se` uJposce´ qhn, Li´bistre, th` n se` ei«pa na` sou' dw´ sw⭈ kai` na´ thn, e“par∆ thn ⬍loipo´ n⬎, na´ thn, cari´zw sou' thn, a”plwse ce´ rin, kra´ thse th` n ko´ rhn ajpe` tw´ ra, kai` zh'se cro´ nou" met∆ aujth` n kai` sunapo´ qane´ thn kai` so` n tra´ chlon a“kliton kli´ne pro` " e“rwta´ n th".” “Hkousa lo´ gou" “Erwto", aJplw´ nw mou to` ce´ rin, “Erw" th'" ko´ rh" di´dei me to` ce´ rin meta` qa´ rrou"⭈ katafilw' to` n “Erwtan, oJrmw' kai` pro` " th` n ko´ rhn, kai` ajpo` th` n to´ shn hJdonh` n e“xupno" ejgeno´ mhn, e“xupno" meta` qa´ naton, meta` pollh'" pikri´a", e“xupno" na` e“cw stenagmou` " kai` ojdu´ na" ajmetrh´ tou", e“xupno" na` flogi´zwmai, na` ko´ ptwmai ejk tou` " po´ nou". To` n kh'pon ajnegu´ reua, to` n “Erwtan, th` n ko´ rhn, ta` de´ ndrh kai` ta` " ca´ rita" ta` " e“blepa eij" to` n kh'pon⭈ katekopto´ mhn, e“pasca, th` n nu´ ktan ejpequ´ moun, to` o“neiron wjre´ geton oJ nou'" mou na` to` ble´ ph, to` fw'" ejparatou´ mhn to, ejmi´soun th` n hJme´ ran, th` n nu´ ktan ei«ca de´ spoinan, aujgh` n na` mh` ejntrani´zw.
147
P 413 N 540
N 545
N 550
N 555
N 560
P 412 post hunc versum aliqua excidisse videtur, de quibus cf. E 659–61 P 413 me` supplevi ex E ejlogomaco´ mhn scripsi coll. E : ejlogoma´ coun P N 540 oJ om. P lu´ bestre N tau´ thn P N 541 th` n1] a‘n P N 542 au”th ante e“ni add. P e“ni P : anei N tou' . . . quga´ thr om. P tou' ante basile´ w" add. P N 543 ejpesce´ qhn P ejgw` ante Li´bistre add. P th` n—dw´ sw] i”na kerdh´ sh" P N 544 lacunam in N e.g. supplevi : ajpo` tou' nu'n P na´ thn cari´zw sou' thn] cari´zw se´ thn na´ thn P N 545 kra´ thson P N 547 so` n tra´ chlon P : sw` n tracei´lwn N : an sou' trach´ lou to` a“kliton? (cf. sou' tra´ cilon a“khlte" E) N 549 th'" ko´ rh" . . . to` ce´ rin] to` cai´rein . . . th'" ko´ rh" P N 550 e“rwta P N 551 to´ shn] tosau´ thn P N 552 hunc versum hic praeb. N E V : post N 554 transp. P meta` 1] me to` n P meta` pollh'" pikri´a"] me th` n pollh` n pikri´an P N 554 ko´ ptomai P : ko´ ptw N ejk] eij" P N 555 ajnegu´ reusa P e“rwta P N 556 de´ ndra P ta` " e“blepa] a’" e“blepon P N 557 kai` ejkatekopto´ mhn kai` e“pascon met∆ ejkei´nwn P inter nu´ ktan et ejpequ´ moun lacunam statuit Ch coll. E 681–82, sed cf. kataikoptw´ mhn e“pasca th` ni´ktan ejpeqi´moun V N 558 om. P ble´ ph scripsi coll. E V : ble´ pw N N 559 ejparatou´ mhn scripsi : ejparate´ wmou N : ejpareti´qeto P : ejparatou´ mou coni. Wa N 560 aujgh` n P (cf. E V) : hJ aujgh` N mh` P (cf. E V) : me N ejntrani´zw scripsi coll. E V : ejntrani´zei P : trani´zei N
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
The Vita Icon and the Painter as Hagiographer NANCY PATTERSON Sˇ EVCˇENKO For Cyril Mango
he author of a vita of St. Nicholas, composed sometime in the thirteenth or fourteenth century, tells us how he views the various ways of honoring a saint. If someone celebrates the memory of the saint with all his heart and soul, says this anonymous author, he will not go away disappointed. If someone builds a chapel in the saint’s name, he will confound the devil as well as all his enemies, and God will increase his possessions, as He did for Job. If someone writes down the life and miracles of the saint, he will be granted release from sins on the Day of Judgment. And if someone expounds the saint’s life and miracles before other men, he will earn his reward in heaven and eternal life.1 In short, to honor the saint on his feast day is fine; to build something in his name is better; to write down his life is better still—but to declaim it before others is the best of all.
T
Versions of this article were presented at the University of Maryland, at the meeting of the Australian Association of Byzantine Studies in Canberra, before the Delaware Valley Medieval Association, and at the Dutch Institute in Florence; it has benefitted from discussions at each of these places. I wish especially to thank Mary Aspra and Joanna Cannon for their careful reading of a draft of this article and for sharing unpublished material with me, and Engelina Smirnova for introducing me to early Russian icons and for many helpful discussions on icons in general. I alone should be held responsible for the errors that remain. 1 The text is a version of the Bi´o" ejn sunto´ mv, ed. G. Anrich, Hagios Nikolaos: Der heilige Nikolaos in der griechischen Kirche. Texte und Untersuchungen, 2 vols. (Leipzig, 1913, 1917), 1:298.5–13 (cf. 277.5–12); 2:388– 89. Some Byzantine literary figures did in fact achieve sainthood through their hagiographical writings. At the death of Joseph the Hymnographer, for example, all the saints to whom he had composed hymns are said to have spilled out of heaven to greet him and escort him inside; see the vita of Joseph by John the Deacon, PG 105:973A–76A, and my article “Canon and Calendar: The Role of a Ninth-Century Hymnographer in Shaping the Celebration of the Saints,” in The Byzantine World in the Ninth Century: Dead or Alive? ed. L. Brubaker (Aldershot, 1998), 101–14, esp. 103. Symeon Metaphrastes, the 10th-century author and revisor of dozens of saints’ vitae, was celebrated as a saint within a century of his death, and Michael Psellos wrote an akolouthia for him; see E. Kurtz and F. Drexl, eds., Michaelis Pselli scripta minora (Milan, 1936), 1:108–19. Neophytos of Cyprus, also an author of saints’ vitae in the late 12th century, apparently believed he too would be well rewarded, to judge by the fresco he used to adorn the bema next to his rock-cut cell; see C. Mango and E. Hawkins, “The Hermitage of St. Neophytos and Its Wall-Paintings,” DOP 20 (1966): 121– 206, esp. 128–29, 165–66, and fig. 66. See also A. W. Epstein, “Formulas for Salvation: A Comparison of Two Byzantine Monasteries and Their Founders,” ChHist 50 (1981): 385–400; R. Cormack, Writing in Gold: Byzantine Society and Its Icons (London, 1985), 215–51, esp. 239–42; and, above all, C. Galatariotou, The Making of a Saint: The Life, Times and Sanctification of Neophytos the Recluse (Cambridge, 1991), who describes “how he [Neophytos] proceeded to imbue his own life’s story with the same structural characteristics of sanctity which he himself had used to give flesh and blood to his own saints’ stories” (p. 73).
150
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
The subject of this article is the Byzantine vita, or biographical, icon, a genre of hagiography that, in this author’s terms at least, would rank highly: not only does it honor the saint, and in a work of art at that, but it also recounts his life—and does so publicly. There are less than two dozen surviving Byzantine examples of vita icons, though the influence exerted by the genre was considerable. Byzantine icons of this type first appear in the early thirteenth century; they fade away somewhat in the fourteenth century, reappear in the fifteenth, and are popular in the post-Byzantine period.2 The most obvious characteristics of this genre of icon are the following: they are generally rather large panels, ranging from ca. 70 cm to 2 m in height; they are devoted to the vitae of quite well established saints; the scenes—generally between twelve and twenty—usually run around all four sides of the image of the saint; and the cycles of any one saint vary remarkably little from icon to icon.3 These scenes may echo, but do not consistently follow, 2 The earliest extant icon with a hagiographical cycle is probably an 11th-century icon of St. Nicholas on Mt. Sinai (now in two parts); it was originally a triptych, a form for which there are no later parallels among the surviving narrative icons. Also rare in its form is a very small late-14th-century icon in the Chilandari monastery on Mt. Athos (25 ⫻ 29.5 cm), which has scenes from the life of St. Mary of Egypt arranged in four horizontal rows; an unpublished icon on Mt. Sinai (64.4 ⫻ 49.8 cm) has comparable rows of scenes from the life of St. George. The latter two icons, which are the only two I know of this type, echo cycles in manuscripts such as Paris gr. 510 of the 9th century or the 11th-century Menologion on Mt. Athos, Esphigmenou 14. For the Sinai triptych, see K. Weitzmann, “Fragments of an Early St. Nicholas Triptych on Mount ˇevcˇenko, The Life of Saint Nicholas in Byzantine Art Sinai,” Delt.Crist. Arc. j JEt., ser. 4, 4 (1964): 1–23, and N. P. S (Turin, 1983), 162–63 and no. 1; for the Chilandari icon, see G. Babic´, Icons (New York, 1988), pls. 34, 35; for the icon of St. George on Mt. Sinai, see T. Mark-Weiner, “Narrative Cycles of the Life of St. George in Byzantine Art” (Ph.D. diss., New York University, 1977), no. 80. Other icons with hagiographical narratives in rare forms comprise an epistyle on Mt. Sinai containing some scenes from the life of St. Nicholas, and one ˇevcˇenko, Life in the same monastery illustrating miracles of St. Eustratios and others of the Holy Five; see S of Saint Nicholas, no. 5, and D. Mouriki, “Icons from the 12th to the 15th Century,” in Sinai: Treasures of the Monastery of Saint Catherine, ed. K. A. Manafis (Athens, 1990), 106, figs. 20–22 (hereafter Sinai). The oldest vita icon of the familiar sort, that is, one having scenes framing a central portrait, may well be the fragmentary St. Marina icon from Philoussa on Cyprus (see note 3 below). Athanasios Papageorghiou dates the icon to the 8th to 9th century (Icons of Cyprus [Nicosia, 1992], 8–9, 55), which is surely too early, but a late 12th-century date, which I would prefer, as this would place the icon neatly in the context of the other surviving vita icons of its type, may be just too late. The anonymous reader of this article believes that the icon can be no later than the early 11th century. I myself have not seen it, and must leave the question of its date temporarily unresolved. 3 ˇevcˇenko, Life of Saint Nicholas, 162–65, and eadem, “Vita Icons and On the subject of vita icons, see S ‘Decorated’ Icons of the Komnenian Period,” in Four Icons in the Menil Collection, ed. B. Davezac (Houston, Tex., 1992), 56–69; H. Belting, Likeness and Presence (Chicago, 1994), 249–60; and the invaluable study by H. Hager, Die Anfa¨nge des italienischen Altarbildes: Untersuchungen zur Entstehungsgeschichte des toskanischen Hochaltarretabels (Munich, 1962), esp. 91–100. On mechanisms for creating a vita icon cycle, see the interesting article by M. Vassilakis, “Eijko´ na tou' aJgi´ou Carala´ mpou",” Delt.Crist. Arc. j JEt., ser. 4, 13 (1985–86): 247–60. See also H. Maguire, The Icons of Their Bodies: Saints and Their Images in Byzantium (Princeton, N.J., 1996), esp. 169–94 (on St. Nicholas and St. George cycles). A comprehensive list of surviving examples has never been drawn up. For the purposes of this paper, I will cite as Byzantine the following icons: vita icons on Mt. Sinai of Sts. Nicholas, George, Catherine, John ˇevcˇenko, Life of Saint Nicholas, nos. 3 and 42; G. Soteriou and M. the Baptist, Moses, and Panteleimon (S Soteriou, Eijko´ ne" th'" monh'" Sina', vols. 1–2 [Athens, 1956, 1958], 2: pls. 165–70; K. Weitzmann, “Icon Programs of the 12th and 13th Centuries at Sinai,” Delt.Crist. Arc. j JEt., ser. 4, 12 [1986]: 63–116, esp. 94–107; Mouriki, in Sinai, 114–16; eadem, “A Moses Cycle on a Sinai Icon of the Early Thirteenth Century,” in Byzantine East, Latin West: Art-Historical Studies in Honor of Kurt Weitzmann, ed. D. Mouriki et al. [Princeton, N.J., 1995], 531–46; H. C. Evans and W. D. Wixom, eds., The Glory of Byzantium: Art and Culture of the Middle Byzantine Era, A.D. 843–1261 [New York, 1997], no. 249 [St. Panteleimon icon]); vita icons of St. Nicholas on
ˇEVC ˇ ENKO NANCY PATTERSON S
151
known written texts, and the order of episodes is erratic and jumpy—they start with a birth scene and end with a death scene, but otherwise there is no attempt to follow a story line through adjoining sequences of panels. Finally, there is an apparent absence of program in the choice of scenes, and no obvious connection to any particular sanctuary or to the saint’s relics.4 In content, these icon cycles emphasize—as had Passio cycles from the beginning of Christian time—what the saint did with his life, rather than the wonders he performed after his death. For these illustrated vitae are not miracle collections, but models for earthly behavior that will lead to a place in the heavenly kingdom. On St. Nicholas icons, for example (Fig. 1), scenes of the saint’s birth, education, and consecration join what are mainly stories of his good deeds and prompt response to need. His tomb at Myra is not depicted, and miraculous healings there are never illustrated, despite the fact that the tomb was known to exude a healing oil. Though collections of Nicholas’s miracles existed, the only posthumous one commonly represented on these icons is that of the boy Basil rescued by Nicholas from his service as cupbearer to his Arab abductors and returned to his family.5 Cycles devoted to martyrs such as St. George (Fig. 2) lay stress on the tortures overcome by the saint one by one, and, once more, award little space to the recounting of his miracles and none at all to the promotion of a particular sanctuary.6 The apparently unprogrammatic tone of these panels has meant that it has been hard to understand just what brought this genre of icon into existence and what it is all about.7 ˇevcˇenko, Life of Saint Nicholas, nos. 14, 37, 41, Patmos, in Kastoria, Skopje, and from Kakopetria, Cyprus (S 43, 44; N. Zias, “Eijko´ ne" tou' bi´ou kai` th'" koimh´ sew" tou' aJgi´ou Nikola´ ou,” Delt.Crist. Arc. j JEt., ser. 4, 5 [1969]: 275–98 [Kastoria icons]; Evans and Wixom, Glory of Byzantium, no. 263 [Kakopetria icon; see pp. 158, 159– 60, and note 48 below]); an icon of the Anargyroi in Kastoria (Byzantine Murals and Icons, National Gallery, September–December, 1976, exhibition catologue [Athens, 1976], no. 107; Affreschi e icone della Grecia X–XVII secolo [Athens, 1986], no. 28 [unavailable to me]); the relief icon of St. George in the Byzantine Museum in ˇevcˇenko, “The Representation of Donors Athens (K. Weitzmann, The Icon [New York, 1978], pl. 35; N. P. S and Holy Figures on Four Byzantine Icons,” Delt.Crist. Arc. j JEt., ser. 4, 17 [1993–94]: 157–64, esp. 158–60; Conversation with God: Icons from the Byzantine Museum of Athens (9th–15th centuries), exhibition catalogue, The Hellenic Centre, London [Athens, 1998], no. 4); icons of St. Marina on Cyprus (one from Pedhoulas now in Nicosia, one from Philoussa now in Paphos, a third in the catholikon of the monastery of St. John Lampadistes at Kalopanagiotes; Papageorghiou, Icons of Cyprus, 8–9, 55, figs. 2, 34; idem, Byzantine Icons from Cyprus [Athens, 1976], nos. 11, 35; S. Sophocleous, Icons of Cyprus, 7th-20th Century [Nicosia, 1995], no. 1 [for the Philoussa icon, see note 2 above]); an icon of St. Basil in Houston (L. Brubaker, “The Vita Icon of Saint Basil: Iconography”; A. W. Carr, “The Vita Icon of Saint Basil: Notes on a Byzantine Object,” both in Four Icons, ed. Davezac, 70–93, 94–105); and an icon of St. John Lampadistes in the Byzantine Museum in Nicosia (D. Mouriki, “The Cult of Cypriot Saints in Medieval Cyprus as Attested by Church Decorations and Icon Painting,” in “The Sweet Land of Cyprus”: Papers Given at the Twenty-fifth Jubilee Spring Symposium of Byzantine Studies, Birmingham, March 1991, ed. A. A. M. Bryer and G. S. Georghallides [Nicosia, 1993], 237–77, esp. 249, fig. 17). Icons with framing scenes from the life of Christ or the Virgin have not been included. They first appear somewhat later than the hagiographical ones, and most likely derive their form from them. 4ˇ Sevcˇenko, Life of Saint Nicholas, 162–65. An exception to the final statement is the iconostasis beam on Mt. Sinai (see above, note 2), which illustrates various miracles performed by the relics of the Holy Five. No written miracle collection describing the events depicted on this beam has survived. 5 ˇevcˇenko, Life of Saint Nicholas, 143–48. On the iconography of this episode, see S 6 Mark-Weiner, “Life of St. George,” esp. 302–9. On this Sinai icon of St. George, see Soteriou and Soteriou, Eijko´ ne", pl. 167; Weitzmann, “Icon Programs,” 98–99; and recently E. C. Constantinides, “Une icone historie´e de saint Georges du XIIIe sie`cle au monaste`re de Sainte-Catherine du Mont Sinai,” in Drevnerusskoe iskusstvo: Rus’, Vizantija, Balkany (St. Petersburg, 1997), 77–104. 7 ˇevcˇenko, “Vita Icons.” On possible origins of the form of the vita icon, see S
152
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
Almost none of the icons is accompanied by a revealing dedicatory inscription, and none has been found in what we would consider its original position, so our knowledge of their liturgical function, if any, is vague at best. The independence of these icon cycles from known literary texts has been frustrating to iconographers, and has led Kurt Weitzmann to assume the existence of manuscript originals with lengthy cycles of illustrations that served as the basis for the painted cycles.8 Unfortunately, no example of such a manuscript survives, and a close look at the manuscript traditions of the various vitae reveals that, in fact, they were never illustrated with cycles of any length and a saint’s vita in a Byzantine manuscript was rarely adorned with anything more than a standing portrait or a single scene of martyrdom.9 Fresco cycles are no more likely to have been an immediate source of the icon cycles than were manuscripts, for few ever include as many scenes as we find on the icons. Whatever its origin, the form spread quickly: already in the thirteenth century one finds variants in South and North Italy, Cyprus, Sinai, and Russia. If one compares an icon of St. Nicholas in Bari (Fig. 3) with the earlier thirteenth-century icon of St. Nicholas from Mt. Sinai (Fig. 1), one sees the obvious similarities.10 (The fact that Nicholas is a standing figure, not a bust, on the Italian panel is not relevant, since there are plenty of Eastern examples of standing saints.) The Western panel is quite a bit larger, but the scenes, sixteen in each case, run around all four sides of the central portrait on both. On the Apulian work, Nicholas is still clad as an Eastern bishop, and not, as one might have expected, as a Western one with crozier and mitre—this despite the fact that the captions to the scenes are in Latin and the panel was destined for a Latin church. Other than the fact that the icon is painted in a local style, that the captions are in Latin and that there are some oddities in the iconography, the only obvious Western feature here is the addition of a couple of scenes that, though known from the written legend in the East as well as in the West, are not illustrated in Byzantium: the scene of the welcoming of Nicholas as bishop (the top scene in the left row) and two episodes of the story of the exploding vial designed to blow up the saint’s church in Myra but defused by Nicholas on the open sea (left and right rows, further down).11 There are some slightly more subtle programWeitzmann, “The Selection of Texts for Cyclic Illustration in Byzantine Manuscripts,” in Byzantine Books and Bookmen, Dumbarton Oaks Colloquium (Washington, D.C., 1975), 69–109, esp. 84–86; idem, “Icon Programs,” 98. 9 ˇevcˇenko, Illustrated Manuscripts of the Metaphrastian Menologion (Chicago, 1990), esp. 181–96; eadem, N. P. S Life of Saint Nicholas, 165–71. 10 ˇevcˇenko, Life of Saint Nicholas, no. 3. On On the Sinai icon, see Soteriou and Soteriou, Eijko´ ne", pl. 165; S the Bari icon, see Evans and Wixom, Glory of Byzantium, no. 320, with bibliography. See also G. Kaftal, Iconography of the Saints in Central and South Italian Schools of Painting (Florence, 1965), figs. 935 ff; P. Belli D’Elia, Icone di Puglia e Basilicata dal Medioevo al Settecento (Bari, 1988), no. 25; V. Pace, “Presenze e influenza cipriote nella pitture duecentesca italiana,” CorsiRav 32 (1985): 259–98, esp. 271 n. 17; idem, “Icone di Puglia, della Terra Santa e di Cipro: Appunti preliminari per un’indagine sulla ricenzione bizantina nell’Italia meridionale duecentesca,” in Il medio oriente e l’occidente nell’arte del XIII secolo, ed. H. Belting (Bologna, 1979), pt. 2, 181–91, esp. 182, fig. 163. The Bari panel, which was apparently painted for the church of Santa Margherita at Bisceglie, measures 128 ⫻ 83.5 cm; the Sinai icon measures 82 ⫻ 57 cm. On the question of whether the panel was displayed on the altar of the Bisceglie church, see Hager, Anfa¨nge, 98. 11 Anrich, Hagios Nikolaos, 79, 83, 110. On the Western St. Nicholas tradition, see K. Meisen, Nikolauskult und Nikolausbrauch im Abendlande (Du ¨ sseldorf, 1931; repr. 1981), with introduction and expanded bibliography, esp. 269–75. Oddities in the iconography of the Bari panel include the three maidens whom Nicholas rescued from prostitution standing behind a high wall, separated from their sleeping father, and the barred 8
ˇEVC ˇ ENKO NANCY PATTERSON S
153
matic elements here: the painter has emphasized the episcopal content by transforming Nicholas’s tutor, usually a monk, into a bishop, and by placing scenes relating to his ordination high on the sides, across from each other; he has also delighted in painting the billowing sails in the various sea scenes, so that the marine aspects of Nicholas’s career certainly gain greater prominence. But otherwise this artist, whatever his background, has retold the traditional story of Nicholas from birth to death, seamlessly introducing an event rarely told in the East but popular in the West, and in so doing has woven the Eastern and Western aspects of the saint into a single figure acceptable to the mixed world for which this panel was created. A Pisan panel of the third quarter of the thirteenth century devoted to St. Nicholas, the one in Peccioli (Fig. 4), presents a different treatment of the basic cycle.12 Here the number of scenes has been drastically reduced, and these are limited to a position on the sides of the saint; the saint elbows his way into their midst in a thoroughly unByzantine fashion. The choice of scenes at first glance looks ordinary enough: there are the birth of the saint, two episodes of the Basil story, and the story of the three maidens saved by Nicholas from prostitution by his nighttime gift to their father, here told in barest outline. The selection does not appear so random, however, if read in connection with the portrait of the donor, a tiny figure, ostensibly female, who kneels near the saint. For, indeed, what has been chosen, to the exclusion of the usual consecration themes and marine or rescue scenes, is a sequence of family scenes, each involving in some way the birth and protection of children: the birth and miraculous first hours of the saint, his rescue of the child Basil, or his charity toward the three destitute young maidens. The donor has thus wrenched a personal sequence out of the general narrative that characterizes a proper Byzantine vita cycle; we find ourselves reading St. Nicholas’s stories as though they were the story of this donor herself, and we learn more about her, in fact, than we do about St. Nicholas. Each scene on this panel is a sort of personal thanksgiving for a miracle experienced or awaited by the kneeling figure of the donor. Such a concern for refashioning a vita cycle of the Byzantine type for a specific purpose can be seen if one compares, as has often been done, a mid-thirteenth-century Pisan panel of St. Catherine of Alexandria (Fig. 5) with a somewhat earlier Sinai panel of the same saint (Fig. 6).13 As Klaus Kru ¨ ger has noted, the Pisan panel, destined for display in a Dominican institution, stresses through its choice of scenes the learning and rhetorical prison housing the three generals. The vial episode, known as the Thauma de Artemide, is rarely repreˇevcˇenko, Life of Saint Nicholas, 96, 102–3. sented in Byzantine art; see S 12 G. Kaftal, Iconography of the Saints in Tuscan Painting (Florence, 1952), figs. 847, 848, 850, 863, 864; Hager, Anfa¨nge, 95, pl. 135; K. Weitzmann, “Crusader Icon and Maniera Greca,” in Byzanz und der Westen: Studien zur Kunst des europa¨ischen Mittelalters, ed. I. Hutter (Vienna, 1984), 156, fig. 17. The panel is thought to have been made for the church of San Verano in Peccioli, where it is currently housed. In the 19th century it belonged to a private collection. It measures 123 ⫻ 101 cm. 13 J. H. Stubblebine, “Byzantine Influence in Thirteenth-Century Italian Panel Painting,” DOP 20 (1966): ¨he Bildkult des Franziskus in Italien: Gestalt- und Funktionswandel 87–101, esp. 91–92, figs. 9, 10; K. Kru ¨ ger, Die fru des Tafelbildes im 13. und 14. Jahrhundert (Berlin, 1992), esp. 65–67, figs. 117, 222; Belting, Likeness, 377, 380, figs. 226, 227; Weitzmann, “Maniera Greca,” 154–55. See also E. Carli, Pittura medievale pisana (Milan, 1958), 52, figs. 77–81, and color pl. V; Kaftal, Tuscan Painting, fig. 249; Hager, Anfa¨nge, 95, pl. 134. The Pisan panel, which was probably made for the Dominican church of Santa Caterina in Pisa, measures 107 ⫻ 113 cm; the Sinai icon measures 75.3 ⫻ 51.4 cm. For the Sinai panel, see Mouriki, in Sinai, 173, fig. 46, and 386 nn. 70, 71; Soteriou and Soteriou, Eijko´ ne", pl. 166; Weitzmann, “Icon Programs,” 95, fig. 25.
154
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
skills of the princess, which are values consonant with the aims of the Dominican order.14 The Byzantine icon, by contrast, emphasizes her inexorable path toward martyrdom: her refusal to sacrifice, her tortures, and execution. The Pisan panel ends with the angels carrying the remains of the saint to Mt. Sinai; the Byzantine panel, though very likely painted on Sinai itself, does not include this particular reference at all and concludes instead with the scene of her death. The Tuscan panels devoted to the life of St. Francis show another variant on the traditional Byzantine vita icon. The earliest surviving dated panel is the one in Pescia, painted by Bonaventura Berlinghieri in 1235 (Fig. 7);15 the date means not only that these St. Francis panels are older than any of the Italian panels mentioned so far, but also that they were being produced very soon indeed after the saint’s death in 1226. The early St. Francis panels concentrate on miracles, especially miracles at the tomb of the saint, a theme foreign to the Byzantine vita icon tradition. However, around the year 1250 the content of St. Francis cycles changes and a new genre of cycle is introduced, one that stresses the biography of Francis, his ethical behavior, his assimilation to Christ, and the triumph of his now solidly institutionalized order (Fig. 8).16 What marks these St. Francis panels, and distinguishes them from Byzantine ones, is that the form here is being used to celebrate and develop the cult of very new saints, such as Francis, his disciple Clare, or Margherita of Cortona—saints who breathed their last only yesterday and whose vitae were at that very time being hurriedly composed and as hurriedly rewritten.17 The panels served as a means of making the new saint known to the public, revealing first the miraculous powers of the saint and eventually the virtues 14 Kru ¨ ger, Bildkult, 66–67. It should be noted, however, that the imbalance may be due to the number of scenes included in each cycle. The Sinai panel does include scenes of Catherine’s declaration of faith and her dispute with the philosophers, which are on the Pisan panel, but, since it makes room for a larger number of episodes, it appears to dwell less on these than on her tortures and martyrdom. 15 Kru ¨ ger, Bildkult, no. 2, figs. 1–7, 9; Hager, Anfa¨nge, 94, pl. 130. The Pescia panel measures 160 ⫻ 123 cm. The oldest known St. Francis panel, once in San Miniato al Tedesco, is known only through a 17thcentury engraving. The panel was reportedly dated 1228, the year in which the saint was canonized; see Kru ¨ ger, Bildkult, no. 1, fig. 8; Hager, Anfa¨nge, 94, fig. 128. 16 Kru ¨ ger, Bildkult, esp. 101–41. See also D. Blume, Wandmalerei als Ordenspropaganda: Bildprogramme im Chorbereich franziskaner Konvente Italiens bis zur Mitte des 14. Jahrhunderts (Worms, 1983), esp. 13–17; H. Van Os, “St. Francis as a Second Christ,” and idem, “The Earliest Altarpieces of St. Francis,” both in his Studies in Early Tuscan Painting (London, 1992), 203–32, 267–76; B. Bughetti, “Vita e miracoli di S. Francesco nelle tavole istoriate dei secoli XIII e XIV,” AFrH 19 (1926): 636–732; P. Scarpellini, “Iconografia francescana nei secoli XIII e XIV,” in Francesco d’Assisi: Storia e arte (Milan, 1982), esp. 91–99; C. Frugoni, Francesco e l’invenzione delle stimmate (Turin, 1993), esp. 321–98. Another book by C. Frugoni, Francesco: Un’ altra storia (Genoa, 1988), was unavailable to me. 17 On St. Francis legends, see S. Da Campagnola, Francesco d’Assisi nei suoi scritti e nelle sue biografie dei secoli XIII–XIV (Assisi, 1981); J. Moorman, The Sources of the Life of St. Francis of Assisi (London, 1966). On St. Clare, see J. Wood, “Thirteenth-Century Italian Painting: Clare of Assisi,” Art History 14 (1991): 301–28. On Margherita of Cortona, see J. Cannon, “Marguerite et les Cortonais: Iconographie d’un ‘culte civique’ au XIVe sie`cle,” in La religion civique `a l’e´poque me´die´vale et moderne (Rome, 1995), 403–13; J. Cannon and A. Vauchez, Margherita of Cortona and the Lorenzetti: Sienese Art and the Cult of a Holy Woman in Medieval Tuscany (University Park, Pa., 1999). As Joanna Cannon has pointed out, the Cortona panel with scenes devoted to the life of Margherita of Cortona actually preceded the completion of her written vita. Illustrated vitae of St. Dominic are, interestingly enough, all but non-existent in this period; see Kru ¨ ger, Bildkult, 141–47, and J. Cannon, “Dominic alter Christus? Representations of the Founder in and after the Arca di San Domenico,” in Christ among the Medieval Dominicans, ed. K. Emery Jr. and J. Wawrykow (Notre Dame, Ind., 1998), 26–48.
ˇEVC ˇ ENKO NANCY PATTERSON S
155
of the order to which he or she belonged, setting out for all to see the proper reading and interpretation of the events depicted. In a period in which differing written versions of these events began to proliferate, the work of art apparently offered the “truth,” showing what the saint actually looked like and establishing what had really happened. The scenes thus became the authoritative version of the life, rivaling the written vita as an authenticating document.18 The Western variants on the vita icon form, then, show the vita narrative being used in a variety of ways, namely, to integrate the veneration of a saint common to two cultures, to promulgate an institution’s values or voice a particular personal appeal, and to establish the virtues of an entirely new holy figure. A similar variety of uses of the vita icon form can be found in the Slavic world. A magnificent thirteenth-century icon of Elijah done in Pskov, west of Novgorod (now in the Tretiakov Gallery in Moscow), is the earliest of the Slavic vita icons (Fig. 9).19 By the fourteenth century there are numerous Slavic vita icons of St. Nicholas (Fig. 10), as well as of a few other familiar saints such as the archangel Michael and St. George.20 As in Italy, but in contrast to Byzantium, relatively new heroes were also honored with the production of vita icons. As early as the fourteenth century, icons were being produced that outlined the story of the princes Boris and Gleb, sons of the tsar Vladimir the Great, who had been reportedly murdered by order of their brother Svjatopolk in 1015.21 On an icon from the church of Sts. Boris and Gleb in Kolomna, now in the Tretiakov Gallery (Fig. 11), one finds few conventional hagiographic topoi other than death and burial, and many of the scenes one has to consider essentially “historical,” representing events in the history of a people, such as Boris returning from the war against the Petchenegs or the combat of Jaroslav the Wise with the evil Svjatopolk.22 Such an attempt to present the viewer with a common, nationalist, heritage is not something one finds in the Byzantine icons. The vita icon as historical chronicle reached its apogee in the early sixteenth century, when splendid icons of this type were created to celebrate the great founders of Russian monasticism who had lived two centuries earlier—figures such as Sergej of Radonezˇ, Kiril Belozerskij, and others (Figs. 12, 13). The long icon cycles devoted to their lives became visual chronicles of the founding, the difficulties surmounted, the miracles attached, and the ultimate triumph of a single monastic institution through the efforts On the painted vita as an authenticating document, see Blume, Wandmalerei, 17, 20–21. M. Alpatov and I. Rodnikova, Pskovskaja ikona XIII–XVI vekov (Leningrad, 1990), no. 1, pl. 1. The icon measures 141 ⫻ 111 cm. 20 For St. Nicholas icons, see A. Boguslawski, “The Vitae of St. Nicholas and His Hagiographical Icons in Russia” (Ph.D. diss., University of Kansas, 1982); E. Smirnova, Zhivopis’ Velikogo Novgoroda (Moscow, 1976), nos. 9, 13, 35; Alpatov and Rodnikova, Pskovskaja ikona, no. 7; V. Lazarev, Moscow School of Icon Painting (Moscow, 1971), pls. 9–11; V. Antonova and N. Mneva, Katalog drevnerusskoj zhivopisi (Moscow, 1963), 1: nos. 13, 164, 175, 302, etc.; Vizantija, Balkany, Rus’, exhibition catalogue (Moscow, 1991), no. 105. For the archangel Michael icons, see E. Smirnova, Icoˆnes de l’e´cole de Moscou (Moscow, 1989), pls. 42–45, 68–76. For the St. George icon, see V. Lazarev, Novgorodian Icon Painting (Moscow, 1969), 15, pls. 17, 18. 21 ¨ber die heiligen Boris und L. Mu ¨ ller, Die altrussischen hagiographischen Erza¨hlungen und liturgische Dichtungen u Gleb (Munich, 1967); G. Lenhoff, The Martyred Princes Boris and Gleb: A Socio-Cultural Study of the Cult and the Texts (Columbus, Ohio, 1989). 22 Smirnova, L’e´cole de Moscou, pls. 46–52; Lazarev, Moscow School, pls. 3–8. 18 19
156
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
of its holy founder.23 And it was not just monastic institutions that were celebrated: the genre included ecclesiastical cycles as well, prime examples being the vitae of the metropolitans of Moscow Peter and Alexis, founders of the Muscovite church, which Peter had moved from Kiev to Moscow in 1326 (Figs. 14, 15).24 Like the monastic icons for their institutions, the large vitae panels devoted to famous churchmen acted as chronicles of the creation and later history of the church of Moscow. They included scenes of such historically certifiable events as the building of the Dormition Cathedral in Moscow and the visit of Peter to Patriarch Athanasios I in Constantinople. An extreme step in this direction can be seen in the production in 1577 of a huge icon for the church of Decˇani in Serbia (Fig. 16).25 The icon is again devoted to the life of a founder of a monastery—a founder who in this case happened to be the king of Serbia, Stefan Urosˇ III. Stefan Decˇanski, as he is known, had died over two hundred years earlier, in 1331; he was buried in his foundation at Decˇani, and his relics had become the object of a cult. In this work themes related to the development of the saint’s cult at his tomb and to the chronicle of a royal ruler are inextricably woven together. Although the form used is Byzantine, the message it serves to convey is unparalleled in Byzantium, where Constantine the Great had been the only ruler to become an important saint,26 and where the scenes on vita icons, as we shall see, never celebrated a specific place of cult. By contrast with the panels from other cultures, which apparently initially borrowed the vita icon form from Byzantium, the Byzantine panels come across as almost bland, being biographical and ethical in content, comprehensive in their choice of scenes, and utterly unspecific with regard to topographical references. The use of the form for personal, historical, or political purposes was a potential apparently never exploited in Byzantium. What, then, was their purpose? At this point I turn to the earliest Byzantine panels we have, in order to look more closely at their date and provenance, their size, their choice of saints, and the character of their donors, as well as the particular form chosen for the recounting of the vitae. There is little internal evidence to be gleaned, but what there is should be investigated carefully for any insights into the origin and purpose of this most influential genre. Date and provenance: Most of the surviving Byzantine vita icons date from the thirteenth century.27 Of these thirteenth-century examples, six come from Mt. Sinai, one from Cyprus, and two from Kastoria. Similarities in the reverses of the Sinai panels sugFor St. Sergej, see Lazarev, Moscow School, pls. 86, 87. For St. Kiril, see Smirnova, L’e´cole de Moscou, pls. 155–57; Pu´ le" tou' Musthri´ou, exhibition catalogue (Athens, 1994), no. 20. See also the icon of St. Dimitri Priloutski, ibid., pl. 148. 24 For Peter, see Smirnova, L’e´cole de Moscou, pl. 151; for Alexis, see ibid., pl. 152. 25 V. Djuric´, Ikona svetog kralja Stefana Decˇanskog (Belgrade, 1985); French trans., BalkSt 24 (1983): 377–401. 26 We should not, however, ignore the empresses Irene and Theodora, even though their vitae were never illustrated. 27 Icons on Mt. Sinai of Sts. Catherine, George, Nicholas, John the Baptist, Panteleimon, and Moses; icon in Nicosia of St. Nicholas (originally from Kakopetria); icons in Kastoria of St. Nicholas and of the Anargyroi. On these icons, see note 3 above. 23
ˇEVC ˇ ENKO NANCY PATTERSON S
157
gest that all of the early vita icons on Sinai known today may in fact have been made at the monastery itself.28 Size: Most of the panels are quite large, that is, a meter or so in height.29 The large size of the early panels leads one to conclude that these icons had some sort of public position and perhaps function: despite the presence of donor portraits on some of the icons, they were not intended for private contemplation. Although in post-Byzantine times they are sometimes put in the iconostasis, this does not seem to have been the case in earlier times; Weitzmann in his study of the early Sinai vita icons proposes that they were simply hung on the wall of an appropriate chapel.30 Choice of saints: If one compares the choice of saints on the early vita icons with saints chosen for contemporary narrative fresco cycles, one sees that the choice of the vita icons is not particularly representative of hagiographic illustration as a whole, but seems to reflect a connection with Mt. Sinai itself.31 Weitzmann pointed out that the extant vita icons on Sinai could all be connected with chapels located either in the basilica and within the monastery walls, or near the monastery and under its supervision.32 Those icons found outside the monastery also primarily celebrate saints who were highly venerated at Sinai; one thinks especially of Marina (Fig. 17) and Elijah. Although there is no vita icon of Marina on Sinai, there are dozens of plain icons of Marina there, and one of the three chapels attached to the north aisle of the basilica at Sinai was dedicated to her.33 The panel of Elijah in Moscow (Fig. 9) has no exact Sinai counterpart, but the site of Elijah’s succor by the raven was, of course, one of the three holiest spots around the monastery.34 Furthermore, across the top of the Pskov icon there is a seven-figure 28 The design on the reverse of the icons of Sts. Catherine, Nicholas, and John the Baptist is described by Doula Mouriki as “superimposed bands of alternating red and blue-black wavy brushstrokes” (in Sinai, 386). The icon of St. Panteleimon has nested lozenge-shaped designs in red and blue-black (see Evans and Wixom, Glory of Byzantium, 379). The designs on the reverse of the icons of George and Moses are unknown to me. 29 Of the 13th-century panels, the St. Basil panel, housed in Houston, is the smallest, at 66 cm; the Kakopetria icon, by far the largest, at 203 ⫻ 158 cm. Mouriki in her catalogue of donors portrayed on Sinai icons has included, and dated to the second half of the 13th century, a pair of vita icons of Sts. George and Nicholas, usually considered to be 15th-century; D. Mouriki, “Portraits de donateurs et invocations sur les ˆnes du XIIIe sie`cle au Sinaı¨,” in Modes de vie et modes de pense´e `a Byzance, Etudes balkaniques, Cahiers Pierre ico Belon 2 (Paris, 1995), nos. 10, 11 (see also Soteriou and Soteriou, Eijko´ ne", nos. 169, 170). These icons are smaller still: each measures 44 ⫻ 34 cm. 30 Weitzmann, “Icon Programs,” 94–107, 113. 31 T. Gouma-Peterson, “Narrative Cycles of Saints’ Lives in Byzantine Churches from the Tenth to the Mid-Fourteenth Century,” GOTR 30 (1985): 31–44. She has counted 36 cycles of St. George, 31 of Nicholas, 6 of Forty Martyrs, 4 of Demetrios (and 2 for Peter and Paul, and 2 for Stephen), 2 of Basil, 2 of Kosmas and Damianos, 2 of Sabbas, and 1 of Panteleimon (and a single cycle for several other saints). See also the ˇevcˇenko, “Cycles chapter “A Survey of Hagiographical Cycles through the End of the 12th Century,” in N. P. S of the Life of St. Nicholas in Byzantine Art” (Ph.D. diss., Columbia University, 1973), 13–72. 32 Weitzmann, “Icon Programs,” 94–107, 113. 33 K. Weitzmann, The Icon (New York, 1982), 206; J. Folda, “The Saint Marina Icon,” in Four Icons, ed. Davezac (as in note 3 above), 107–33, esp. 117. 34 There was a chapel of Elijah on Mt. Horeb, much visited by pilgrims. The pilgrim Antonius of Cremona, who visited it in 1331, reported that one hundred ravens were fed every day at the monastery in memory of ¨hricht, ZDPV 13 (1890): Elijah; Antonius de Cremona, Itinerarium ad Sepulcrum Domini (1327, 1330), ed. R. Ro 167; cf. L. Eckenstein, A History of Sinai (London–New York, 1921), 158. The typikon composed by the abbot Symeon of Sinai in 1214 (see below, p. 163) celebrates above all Moses, Aaron, Elijah, Elisha, and Catherine; A. Dmitrievskij, Opisanie liturgicheskikh rukopisej (Kiev, 1917), 3:394–419.
158
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
Deesis, apparently one of the earliest anywhere in Russia, which bears a remarkable resemblance to the ones reconstructed for Sinai by Weitzmann and most recently by Mary Aspra.35 Although the connections of vita icons found on Cyprus and elsewhere with the Sinai saints may be coincidence, the coincidence is worth watching. Character of the donors: Not very many portraits of donors are to be found among these thirteenth-century vita icons, but all the Sinai figures, at least, seem to be foreigners.36 The St. George icon on Sinai was given by John the Iberian, that is, John the Georgian, who was both monk and priest (Fig. 2);37 the donor of the St. John the Baptist icon wears the same white robes and has the tonsure of a Georgian cleric (Fig. 18).38 Even the abbot of Sinai who donated the huge Moses icon has the name of Neilos ho Kooueri, which is probably Quirini, a Venetian name known on Crete from shortly after the Venetian takeover (Fig. 19).39 And the Kakopetria icon, an example from the later thirteenth century, shows a Latin knight and his lady kneeling by the feet of St. Nicholas (Fig. 20).40 Finally, the form chosen to display the vitae: The form—vita scenes as a framing element around a central portrait—associates them with the decorated periphereia, the precious frames of gold and silver that were being added to icons during the Komnenian period.41 In works such as these, the central panel is envisaged as a separate unit from Weitzmann, “Icon Programs,” 86–94; M. Aspra-Vardavakis, “Three Thirteenth-Century Sinai Icons of John the Baptist Derived from a Cypriot Model,” in Medieval Cyprus: Studies in Art, Architecture, and History in ˇevcˇenko and C. Moss (Princeton, N.J., 1999), 179–93. Memory of Doula Mouriki, ed. N. P. S 36 For a listing of the donors, see Mouriki, “Portraits de donateurs,” 103–35, nos. 3, 4, 7, 10, 11. 37 For the inscription, see Soteriou and Soteriou, Eijko´ ne", 2:150; Mouriki, “Portraits de donateurs,” no. 7. 38 Mouriki, in Sinai, 115–16. See R. Janin, “Les Ge´orgiens `a Je´rusalem,” EO 16 (1913): 211, quoting Jacques de Vitry; see also G. Peradze, “An Account of the Georgian Monks and Monasteries in Palestine as Revealed in the Writings of Non-Georgian Pilgrims,” Georgica 1.4–5 (1937): 181–246. The pilgrim Magister Thietmar dressed as a Georgian monk and grew a long beard to be able to travel more safely through the Holy Land to Sinai; Thietmar, Peregrinatio, VIII, ed. J. C. M. Laurent, Peregrinatores Medii Aevi Quatuor, 2d ed. (Leipzig, 1873), 20: “Accon igitur iter arripiens, habitu tamquam Georgianus monachus et longa barba simulavi quod non eram.” See Eckenstein, Sinai, 150–51. Later Magister Thietmar describes Georgians in Jerusalem as having “coronas,” laymen and clergy alike—the laymen’s being square, the clergy’s round—and tall hats (“ferentes in capitibus pillea unius ulne longa”) (ibid., XXVIII, ed. Laurent, Peregrinatores, 51). 39 Mouriki (“Moses Cycle,” 531; “Portraits de donateurs,” no. 3) reconstructed the name as Booue´ ri, while older Sinai publications, probably rightly, transcribed the name as Koueri'no"… see K. Amantos, Sinai¨tika` mnhmei'a ajne´ kdota, suppl. 1 to JEllhnika´ (Athens, 1928), 53; H. L. Rabino, Le monaste`re de Sainte-Catherine du Mont Sinai (Cairo, 1938), 62, 86 no. 49, 112 no. 165. The full inscription as transcribed by Amantos reads: oJ eujtel(h` ") (mon)ac(o` ") Ne(i'l)o(") Kouer(i'no"…) ajrciepi´skopo" kai` kaqh(gou´ )meno" tou' aJg(i´ou) o”rou" Sina' (p. 53). Amantos also cites an inscription on a 17th-century icon on Sinai, bearing the name of Nicholas Kouri´no" (p. 52; see also Rabino, Monaste`re, 61). On Venetian gentlemen on Crete by the name of Querini, see G. Fedalto, La chiesa latina in oriente, 2d ed. (Verona, 1981), 1:382, 386 (Paolo Querini, duke of Crete in 1217), 391 (Giovanni Querini, Latin archbishop of Candia in 1248). See also S. Xanthoudidou, “Krhtika` sumbo´ laia ejk th'" JEnetokrati´a",” Cristianikh` Krh´ thÚ Creta christiana 1 (1912): 35. I owe this reference to the kindness of Mary Aspra, and I take this opportunity to thank her for her generous advice on the reading of the inscription. The Latin name is indeed puzzling for an icon done apparently early in the Latin domination, and one wonders whether it could conceivably be connected with the struggles over jurisdiction, which the Greek archbishop of Sinai, Symeon, managed to resolve by 1217 with the assistance of the pope and authorities in Venice (see below, p. 163). But see, most recently, S. McKee, “Women under Venetian Colonial Rule in the Early Renaissance: Observations on Their Economic Activities,” Renaissance Quarterly 51 (1998): 34–67, esp. 39–40, on the difficulties of ascertaining ethnic origin from proper names. I wish to thank David Jacoby for alerting me to this article, and for other helpful and relevant information. 40 On this icon, see note 48 below. 41 ˇ Sevcˇenko, “Vita Icons.” 35
ˇEVC ˇ ENKO NANCY PATTERSON S
159
the frame; in many cases, the frame is in fact a later gift to an older panel.42 The frame both adorns and comments upon the portrait; the possibilities inherent in such a dialogue become highly developed in the Komnenian period.43 The study of the vita icons suggests the same distinction, with the vita being something added to the central image. It is possible to read the vita icon of St. George in Athens, which comes originally from Kastoria (Fig. 21), in this way. The central panel, no ordinary image of George but a wooden relief sculpture that contrasts strongly with the rest of the icon, which is painted, may well refer to a specific wooden image of St. George, such as the extraordinary xoanon, nearly 3 m tall, in the Omorphecclesia church in the hills above Kastoria (Fig. 22).44 I have suggested elsewhere that the central panel was meant to evoke such a wooden figure, and that the donor, the lady from Kastoria in proskynesis at the lower left, is therefore appealing to the saint through a very specific, identifiable image, which she has had copied and then adorned by surrounding it with a vita of the saint.45 Even when the central panel seems a straightforward portrait of a saint, one should consider the possibility that it is meant to be reproducing a known image. The central figure of Moses on the Sinai vita icon (Fig. 19) quite closely resembles another icon on Sinai—that of Moses receiving the Law, painted by the artist Stephen on one of his two large panels, which were designed as a pair and are now hanging on the north and south walls of the basilica (Fig. 23).46 Though the two Moses icons differ in style, the rare composition, which combines the reception of the Law with the image of the Burning Bush and the prominent sandals, is shared by both; but on one a vita cycle has been added. However, once the vita border is subtracted, the panels are nearly identical in size.47 Again, one has the impression that the artist of the vita panel is deliberately reproducing a known original, to which he has added a framing vita. An icon from Cyprus, done later in the thirteenth century, makes this process even clearer. The famous icon of St. Nicholas from Kakopetria has intrigued scholars because of its large size—at over 2 m tall, it is the largest Byzantine vita icon in existence, and Examples can be found in A. Grabar, Les reveˆtements en or et en argent des icoˆnes byzantines du moyen ˆage (Venice, 1975). 43 See A. W. Carr, “The Presentation of an Icon on Sinai,” Delt.Crist. Arc. j JEt., ser. 4, 17 (1993–94): 239–48. 44 On the Athens icon, see note 3 above; on the xoanon, see G. Soteriou, “La sculpture sur bois dans l’art byzantin,” in Me´langes Charles Diehl (Paris, 1930), 2:178–80; on the Omorphecclesia church, see S. KalopissiVerti, Dedicatory Inscriptions and Donor Portraits in Thirteenth-Century Churches of Greece (Vienna, 1992), 48–49, with earlier bibliography. The church was probably built in the 11th century; it also bears an inscription on the west wall of the exonarthex that gives a date between 1296 and 1317. 45 ˇ Sevcˇenko, “Donors and Holy Figures,” 158–60. 46 On the icons of Moses and Elijah by Stephen, see Mouriki, in Sinai, 109–10, 164, fig. 34 (Elijah); Soteriou and Soteriou, Eijko´ ne", pl. 75; Weitzmann, “Icon Programs,” fig. 31. The icon of Elijah (and perhaps the Moses one as well?) exhibits the same pattern of wavy red and blue-black lines on its reverse as the vita icons mentioned earlier (see note 28 above). 47 The dimensions of the Moses vita icon are 142 ⫻ 90 cm; without the vita (11 cm on each side), the height of the icon would be 120, and its width 68 cm. The icons by Stephen measure 129 ⫻ 69 cm. One wonders what were the images of Moses described as located on either side of the golden Burning Bush by Magister Thietmar, Peregrinatio, XVIII, ed. Laurent, Peregrinatores, 42: “Rubus quidam sublatus est et inter Christianos pro reliquiis distractus, ad instar autem illius rubi factus est aureus rubus ex laminis aureis, et ymago Domini aurea super rubum, et ymago Moysi aurea stans ad dexteram rubi, discalcians se. Stat et alia ymago Moysi aurea in sinistra parte rubi tamquam discalciata et nudis pedibus. Ubi Dominus dedit ei legationem ad Pharaonem, regem Egypti, de educendo populo suo.” Does he mean the Justinianic mosaics or more recent panels? If the latter, the pair of icons must consciously echo the arch mosaics. 42
160
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
one of the largest icons anywhere—and because of its combination of a Byzantine vita cycle and Latin donors (Fig. 20).48 The central figure is inscribed as St. Nicholas oJ th'" Ste´ gi", that is, Nicholas “of the roof.” The icon, in fact, used to be in the church of St. Nicholas tes Steges, an eleventh-century foundation known by this name because of its particular double roof.49 The icon, now transferred to Nicosia, was formerly affixed to a wall of the narthex of this church. Of relevance here is a fresco uncovered only quite recently, an image of St. Nicholas painted on the walled-up entrance to the diakonikon, and serving in this position as the end panel of the iconostasis (Fig. 24).50 The twelfth-century fresco is also very large, over life-size, in fact. It must certainly have been the model for the icon, and the unusually large size of the icon may have been prompted by a desire to reproduce the fresco exactly, to scale. Here, then, is a sure case of a vita icon reproducing as its central panel what is essentially the title-saint icon of a particular church. As the inscription makes clear, the central panel of this icon is not a general image of Nicholas, but a very particular one, namely, the fresco image in the church of Nicholas tes Steges. It is through this local image of the saint, adorned by the donors with a vita, that the kneeling figures hope to reach St. Nicholas and achieve salvation. This brief survey of the earliest vita icons has shown that the form appears rather suddenly in the early thirteenth century, and is used at first for a rather narrow selection of saints, many of them having connections with Sinai, where, in fact, the majority of surviving examples are to be found. The icons are large-scale and surely intended for public display, but they do not apparently belong in the iconostasis or have a precise liturgical function. The donors are often foreigners. The form chosen fits nicely with the taste of the late twelfth century and indicates that the vita cycles were being viewed as a form of donation to a particular image of the saint.51 Keeping in mind the admittedly rather scanty concrete evidence, I shall now focus on the possible origin and function of these icons, and suggest that the main impetus for 48 ˇevcˇenko, Life of Saint Nicholas, no. 14. See also PapageorEvans and Wixom, Glory of Byzantium, no. 263; S ghiou, Icons of Cyprus, 46–49, pl. 32a, b; Weitzmann, “Maniera Greca,” 155, fig. 16; D. Mouriki, “The Wall Paintings of the Church of the Panagia at Moutoullas, Cyprus,” in Byzanz und der Westen, ed. Hutter (as in note 12 above), 171–213, esp. 210 and fig. 32; and, especially, J. Folda, “Crusader Art in the Kingdom of Cyprus, 1275–1291: Reflections on the State of the Question,” in Cyprus and the Crusades, ed. N. Coureas and J. Riley-Smith (Nicosia, 1995), 209–37, esp. 216–21 and fig. 6. 49 On the church, see A. Stylianou and J. Stylianou, The Painted Churches of Cyprus (London, 1985), 53–75, figs. 18–30. 50 Ibid., 62, 65, fig. 25. 51 To this category should be assigned the companion icon to that of St. Nicholas of Steges, namely, the panel by the same artist showing the Virgin surrounded by depictions of miracles performed not so much by the Virgin Mary herself as by the particular image reproduced in the central panel; Papageorghiou, Icons of Cyprus, 46–49, fig. 31; Sophocleous, Icons of Cyprus, no. 25; Folda, “Reflections,” 216–21; D. Mouriki, “Thirteenth-Century Icon Painting in Cyprus,” The Griffon, n.s., 1–2 (1985–86): 38–47; A. W. Carr and L. Morrocco, A Byzantine Masterpiece Recovered: The Thirteenth-Century Murals of Lysi, Cyprus (Austin, Tex., 1991), 106. Compare also Figure 2 above: in the icon of St. George on Sinai, the central figure is clad in military costume, while in the scenes around the edge he wears court dress (except when riding against the dragon); Soteriou and Soteriou, EiJko´ ne", pls. 167, 169. One has the impression that the central image could be a copy of a known icon of St. George, which has been then surrounded by scenes of his life. And the icon of the Anargyroi in the Kastoria museum should be compared with the fresco icon on a pier in the church of the
ˇEVC ˇ ENKO NANCY PATTERSON S
161
the development of the vita icon should be sought in the Eastern Mediterranean, perhaps at Sinai itself, and be seen as a response to the particular multilingual, multiethnic environment of the region, as well as to the perceived role of Sinai in the fragmented political world of the Eastern Mediterranean in the first half of the thirteenth century. I would like to characterize these icons as a new form of vita expressly designed to be understood by the diverse groups that constituted this society. First of all, it is worth remembering that the vita icon first appeared at the end of the twelfth century, which was a period of decline in both the composition and the widespread circulation of written saints’ vitae.52 There was no liturgical need for more vitae, since few slots for readings were left vacant: Metaphrastes had seen to that, by providing a vita to be read at orthros on many days of the year.53 The writing of saints’ vitae was becoming a more personal affair, motivated by the desire to honor a saint through one’s own writings, as well as the wish to provide a favored sanctuary with a special occasional piece for its own use. By contrast to this shrinking of expectation and audience of the literary production, the vita icon offered a fresh and versatile medium for hagiography, appropriate to the mixed society of the thirteenth-century Levant. The secret of the success of the vita icon genre becomes apparent when one looks at two cycles of St. Nicholas, the earlythirteenth-century Sinai one and the late-thirteenth-century Kakopetria icon (Figs. 1, 20). The Sinai icon devotes many scenes to the cycle, beginning with the birth and ending with the death of the saint; in between, it weaves stories from several different sources, maintaining a careful balance between the biographical topoi and the ethical and the miraculous themes, and exhibiting a signal reluctance, noted earlier, to attach the veneration of the saint to relics in any specific sanctuary.54 The result is a vita that is balanced, broad, and informative, as well as independent of local concerns. The Kakopetria cycle takes things a bit further. The vita here presents all the tradiAnargyroi at Kastoria; S. Pelekanidis and M. Chatzidakis, Kastoria (Athens, 1985), 32, fig. 10. There are many later examples of icons with scenes that surround a specific miraculous image of the Virgin relating the story of that very image: see, for example, an icon of the Virgin of Vladimir from the Stroganoff school, or icons of the Virgin Portaitissa. See E. Bakalova, “Zwei Ikonen der Muttergottes Portaitissa (von Iviron) in Bulgarien,” Delt.Crist. Arc. j JEt., ser. 4, 17 (1993–94): 347–58. 52 See P. Magdalino, “The Byzantine Holy Man in the Twelfth Century,” in The Byzantine Saint, ed. S. Hackel (London, 1981), 51–66. On the revival of hagiography in the Palaiologan period, see A.-M. Talbot, “Old Wine in New Bottles: The Rewriting of Saints’ Lives in the Palaeologan Period,” in The Twilight of ´ urcˇic´ and D. Mouriki (Princeton, N.J., 1991), 15–26. Byzantium, ed. S. C 53 ¨ berlieferung und Bestand On the readings from Symeon Metaphrastes’ vitae of the saints, see A. Ehrhard, U ˇevder hagiographischen und homiletischen Literatur der griechischen Kirche, 3 vols. (Leipzig, 1937–40), 2:314–17; S ˇcenko, Metaphrastian Menologion, 3–6. The readings originally took place at monastic orthros, but were eventually transferred to the refectory, which meant that their audience would have been even more restricted. Just when this change took place is unclear, but refectory readings of supplementary, non-Metaphrastian texts are documented in 12th-century Sicily; see M. Arranz, Le typicon du monaste`re du Saint-Sauveur `a Messine (Rome, 1969), 67 (a text of the miracles of St. Nicholas was read aloud in the refectory on December 6). 54 ˇ Sevcˇenko, Life of Saint Nicholas, no. 3. The scenes are the following: the birth of Nicholas; the schooling of Nicholas; Nicholas is consecrated priest, then bishop; Nicholas celebrates mass; Nicholas fells the cypress of Plakoma; Nicholas saves mariners; Nicholas saves three men from execution; the three generals in prison; Nicholas appears to Constantine, then to Ablabius; the three generals come before Constantine the Great; the three generals thank Nicholas; the death of Nicholas; Nicholas rescues Basil from the Saracens.
162
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
tional Byzantine episodes, slightly altered, to be sure, in the retelling. But it also introduces a posthumous miracle known only from Latin sources, that of the three salted boys (Fig. 25).55 It is capable of combining elements, then, not only from different written sources but also from two independent literary traditions, the Greek and the Latin. The result is a hybrid vita that would probably never have come into existence in written form, and would have had no place in the liturgy of either culture. Here it adorns and celebrates a specific Nicholas image located in a specific place, but its eclectic language serves to glorify a saint who can be venerated by all regardless of language, race, or creed. The final two scenes on the Kakopetria icon show the death of the saint and then his ´ ]FOS TOU AGIOU NIKOLAOU BRHON TO MIRON (sic), that is, “pouring forth tomb as TA its oil,” with pilgrims shown marveling at the tomb (Fig. 26). The need to identify a final resting place, or source of healing, is arguably more Western in spirit: we saw the same thing in the comparison of the Byzantine and the Western St. Catherine panels. What is interesting here, though, is that the caption to the first scene identifies it as K]HMHSIS TOU AGIOU NIKOLAOU EN THS MIRHS or “the death of the saint at Myra” (Fig. 27; italics mine). Which sanctuary, then, are the pilgrims visiting—Bari, where the relics had been taken in 1087, or Myra, which apparently never admitted the loss? The donors are Latin, which would suggest a preference for the Bari sanctuary; yet, despite its generic character, the architecture of the two sites does seem to be deliberately presented as identical and one is clearly identified as Myra. This ambiguity, intended or not, is in perfect accord with the mixed character of this work; the vita cycle here interweaves the two literary traditions, and perhaps even the two grave sites, thereby giving Nicholas a universal validity. Where could this genre of icon have originated? It does not seem to have been Constantinopolitan in origin, although we lack any icons from the capital from just this period, the first half of the thirteenth century.56 But the evidence we do have suggests that the form first became popular in the Eastern Mediterranean, and was possibly promoted by Sinai itself. Sinai in the early thirteenth century was under the leadership of a very active abbot, the archbishop Symeon.57 The monastery had been given substantial holdings on Crete ˇevcˇenko, Life of Saint Nicholas, no. 14. The scenes represented are: the birth of Nicholas; the schooling S of Nicholas; Nicholas is consecrated priest, then bishop; the story of the three maidens; Nicholas fells the cypress of Plakoma; Nicholas saves mariners; Nicholas saves three men from execution; the three generals in prison; Nicholas appears to Constantine, then to Ablabius; the three generals come before Constantine the Great; the three generals thank Nicholas; the death of Nicholas; the tomb of Nicholas; Nicholas rescues Basil from the Saracens; Nicholas resurrects three murdered schoolboys. On the latter episode, which first appears in Latin texts of the 12th century, see Meisen, Nikolauskult, 289–306; C. W. Jones, Saint Nicholas of ˇevcˇenko, Life of Saint Nicholas, 153. Myra, Bari and Manhattan (Chicago, 1978), 128–40; S 56 The fresco cycle of St. Francis from the Kalenderhane church in Constantinople (ca. 1250, now in the Istanbul Archaeological Museum) uses the vita panel format of a large central figure flanked by ten smallscale scenes. I wish to thank Cecil Lee Striker warmly for showing me, in advance of publication, the chapter on these frescoes; now see C. L. Striker and Y. D. Kuban, Kalenderhane in Istanbul: The Buildings, Their History, Architecture, and Decoration. Final Reports on the Archaeological Exploration and Restoration at Kalenderhane (Mainz, 1997), 128–42. The choice of format should be studied within the context of the function of the vita icon itself. 57 On the history of Sinai, see J. Kamil, The Monastery of Saint Catherine in Sinai: History and Guide (Cairo, 1991); A. Paliouras, The Monastery of St. Catherine on Mount Sinai (Athens-Cairo, 1985); Eckenstein, Sinai; 55
ˇEVC ˇ ENKO NANCY PATTERSON S
163
by the archbishop of Crete in 1203, which the new Latin archbishop was threatening to expropriate after the Venetian acquisition of the island shortly thereafter in 1204. Symeon traveled to Venice and Rome in 1211, where he not only won reparations from the Doge Pietro Ziani for the harm done to the Sinai monks on Crete, but also succeeded in persuading Pope Honorius III to provide papal guarantees of protection for all Sinai properties throughout the Mediterranean.58 Symeon also undertook to rewrite a typikon for the monastery in 1214, and from this document one learns that an earthquake had struck the monastery on 1 May 1201, collapsing the fortress walls and the cells of the monks (a calamity which, incidentally, does not seem to have been noted in any architectural history of the monastery).59 Symeon was most likely the abbot responsible for the reconstruction. Additionally, the monastery was emerging in this very period as an important center of icon production: its extraordinary activity there in the early thirteenth century has been chronicled in the researches of Weitzmann and Doula Mouriki.60 Sinai’s properties lay in Jerusalem, Syria, Crete, Cyprus, and Constantinople,61 and Sinai was housing an ever more diverse body of monks and visitors within its walls. Leaving aside the evidence provided by the numerous Georgian, Syriac, and Arabic manuscripts held in the library,62 the inscriptions found scattered throughout the monastery,63 Rabino, Monaste`re; H. Skrobucha, Sinai (Olten-Lausanne, 1959); Amantos, Sinai¨tika` mnhmei'a, 1–118; idem, Su´ ntomo" iJstori´a th'" iJera'" monh'" tou' Sina', suppl. 3 to JEllhnika´ (Thessalonike, 1953), 1–114. A list of the archbishops of Sinai in this period can be found in Rabino, Monaste`re, 83. There is some doubt whether the Symeon known to have been archbishop in 1258 is the same individual as the archbishop Symeon of 1203. At any rate, there is a puzzlingly rapid succession of abbots: there were six in the second quarter of the century, with three between 1223 and 1228 alone. 58 On Symeon’s Cretan difficulties, see Fedalto, Chiesa latina, 1:383–89; G. Hofmann, “Sinai und Rom,” OrChr 9.3 (1924): 226, 243; idem, “Lettere pontificie edite et inedite intorno ai monasteri del Monte Sinai,” OCP 17 (1951): 283–303. See also Mouriki, in Sinai, 384 nn. 6–9. Honorius’s bull is dated 1217; he sent confirming letters to the bishop of Sinai in 1223, 1225, and 1226, as well as letters to various civil and Latin ecclesiastical authorities in Crete on behalf of the Sinai monks there, first in 1217, and again in 1224, 1225, and 1226. Several of Honorius’s successors as pope in the course of the 13th and 14th centuries also signed documents confirming Sinai’s rights to its properties on the island and elsewhere. After 1217, the documents are addressed simply to the bishop of Sinai, without giving a name; the early ones, however, are addressed to Symeon, whom Pietro Ziani, the Doge of Venice, addresses as archbishop in 1212, which apparently is the first use of the title. The codex Sinai gr. 2246 contains copies, in both Greek and Latin, of Cretan Venetian documents relevant to Sinai from 1212 to 1669. One Sinai manuscript contains a sermon in Arabic attributed to a Symeon, bishop of Sinai (Eckenstein, Sinai, 149). 59 Dmitrievskij, Opisanie, 3:394–419, esp. 402–3, 415–16. The liturgy for the anniversary of the earthquake included prayer services in the desert outside the walls of the monastery. The manuscript containing the typikon, a redaction of the Typikon of St. Sabas, is Sinai gr. 1097; in this text the monastery is still dedicated to the Virgin, not yet to St. Catherine, but the larnax of Catherine has already been brought down from the mountain into the monastery (cf. Dmitrievskij, Opisanie, 409, 411). 60 Weitzmann, “Icon Programs,” esp. 112–16; idem, The Icon (1982), 201–7; Mouriki, “Thirteenth-Century Icon Painting,” esp. 61–62. 61 The properties are listed in the papal bull of 1217; see Hofmann, “Lettere,” 284. 62 According to Hofmann, there are on Sinai 2,289 Greek manuscripts, 580 Arabic ones, 276 Syriac, 98 Georgian, 40 Slavic, 6 Ethiopian, and 5 “Palestinian” ones; “Lettere,” 296. The numbers will probably need to be revised upwards. See M. Kamil, Catalogue of All Manuscripts in the Monastery of St. Catherine on Mount Sinai (Wiesbaden, 1970); G. Garitte, Catalogue des manuscrits ge´orgiens litte´raires du Mont Sinai (Louvain, 1956). 63 Rabino, Monaste`re, 101–13.
164
THE VITA ICON AND THE PAINTER AS HAGIOGRAPHER
and the reports of pilgrims,64 and looking only at the linguistic evidence provided by the icons, one sees inscriptions on the Sinai icons being written in two languages—Georgian and Greek already in the eleventh century (Menologion icons), and Arabic and Greek in the thirteenth (e.g., Fig. 23).65 On one icon of Moses before the Burning Bush, the donor wears an Arab turban.66 Weitzmann has posited the existence of a Latin Crusader icon workshop actually living and working in the monastery in the later thirteenth century.67 Buildings were erected within the monastery walls to accommodate these diverse groups: an older structure was turned into a mosque sometime in the eleventh or twelfth century, and in the same century a Gothic hall with pointed arches, of unclear function, was constructed.68 A chapel dedicated to St. Catherine provided Latin services, according to Felix Fabri, a late-fifteenth-century pilgrim.69 At any rate, these various groups had been encouraged to conduct services in their own languages for some time. The typikon of the monastery of St. Sabas outside Jerusalem—this being the rule followed at Sinai as well—states that the Georgians, Syrians, and Franks in that monastery are allowed to recite the hours and read the Gospels and Apostles in their own “dialect” and in their own chapels, although they are not to celebrate the liturgy on their own: for that they are all to congregate in the main church and celebrate with the brotherhood the holy mysteries.70 And the abbot Symeon, in his typikon for Sinai, says that the Apocalypse of St. John may be read on Sundays, if the abbot so desires—possibly a concession to Latin interests.71 The saints whose vita icons I have discussed—John the Baptist, George, Marina (alias Margaret), Moses, Catherine, and Nicholas—were, of course, being venerated by all these various groups, with Moses, Catherine, and John the Baptist reportedly venerated even by the Muslims.72 East and West divided or disputed the possession of the relics of some Magister Thietmar reported in 1217 that the monks consisted of Greeks and Syrians (Peregrinatio, XVIII, ed. Laurent, Peregrinatores, 41). Ludolph of Suchem wrote in 1350 that there were more than four hundred Greeks, Georgians, and Arabs associated with the monastery; Ludolph von Suchem’s Description of the Holy Land (Written in the Year A.D. 1350), trans. A. Stewart, Palestine Pilgrims’ Text Society 12 (London, 1895), 85. 65 Mouriki, in Sinai, 102–3, 384 nn. 10–12, where she lists icons on Sinai with non-Greek inscriptions. 66 Evans and Wixom, Glory of Byzantium, no. 250; Mouriki, in Sinai, 110, pl. 36; and eadem, “Portraits de donateurs,” no. 6. 67 K. Weitzmann, “Thirteenth-Century Crusader Icons on Mount Sinai,” ArtB 45 (1963): 179–203; idem, Icon, 202–7. 68 For the mosque, see Fratris Felicis Fabri, Evagatorium in Terrae Sanctae: Arabiae et Egypti Peregrinationem, ed. C. D. Hassler (Stuttgart, 1843–49), trans. A. Stewart, The Wanderings of Felix Fabri, Palestine Pilgrims’ Text Society 10 (London, 1893), 4:614; Rabino, Monaste`re, 39–43. According to George Forsyth, the structure belongs to the 6th century and was the original monastery guest house; G. Forsyth and K. Weitzmann, The Monastery of Saint Catherine at Mount Sinai: The Church and Fortress of Justinian (Ann Arbor, Mich., 1968), 7, 8, pls. XVI, XVII. For the Gothic hall, see Rabino, Monaste`re, 54–57 (this structure eventually became the refectory of the monastery). 69 Felix Fabri, who visited Sinai in 1484, describes the Latin chapel near the pilgrims’ guest house in the monastery; Fabri, Wanderings, 4:547, 549, 599, 612, 624. A Franciscan Antonius de Fano could celebrate Mass in the basilica in 1425, a privilege no longer granted Latins in Fabri’s time, which was sorely resented by him. But Fabri and his fellow pilgrims were allowed to recite their own prayers at all the holy spots; see, e.g., Wanderings, 607. 70 Dmitrievskij, Opisanie, 1:222–23. 71 Ibid., 3:401–2 (except in the Easter season). 72 Ludolph of Suchem, who visited Sinai in 1341, describes how the Saracens joined the Latins in kneeling before the relics of St. Catherine; Description, 86. Magister Thietmar in 1217 found the Burning Bush vener64
ˇEVC ˇ ENKO NANCY PATTERSON S
165
of these saints, and new legends sprang up right and left. The relics of St. Catherine, all but a skull and a hand, were taken from Sinai itself to France in the eleventh century; the relics of St. Nicholas were claimed by Bari after the expedition of 1087, though Myra never did concede the loss.73 A diverse community, sharing saints but having divergent literary traditions for their stories, would have been a fertile ground for the growth of a new form of hagiography.74 The immense icon production activity on Sinai in the early thirteenth century, including the almost serial-like production of the vita icons we have been studying, its diversity of population, and its fame as a pilgrimage site, as well as its wide connections through its properties all over the Mediterranean, do make Sinai one possible center for the promotion and dissemination of this new form of icon.75 Perhaps the vita icon was not, after all, as had been thought, a form invented in one milieu, namely, the East, and borrowed by the other, that is, the West, but instead one designed from its inception to be a mode of transmission between cultures sharing the same saints but venerating them in different languages and liturgies. In the monastery of Sinai, saints’ vitae continued to be translated from Greek into Arabic and into Georgian for the use of these various groups76 —but it was the vita icon that could provide the one truly international language in which to tell the tale, namely, the language of art. Philadelphia, Pa. ated by Saracens and Christians alike, all of whom, regardless of rank, had to remove their shoes before entering the chapel; Peregrinatio, XVIII, ed. Laurent, Peregrinatores, 42. Another pilgrim, Frescobaldo (1384), claimed that the Saracens venerated the Virgin, John the Baptist, St. Catherine, and all the patriarchs of the Old Testament; cf. Eckenstein, Sinai, 163. 73 Relics of St. Catherine were peddled at Rouen by a Sinai monk Symeon “the Five-Tongued,” who died in Tre`ves (Trier) in 1035; see Amantos, Su´ ntomo" iJstori´a, 29–31, 34. According to Eckenstein, Sinai, 139, there once existed an order of knights devoted to the protection of pilgrims traveling to visit St. Catherine’s monastery at Sinai, although I am unable to verify this assertion. On St. Nicholas, see Anrich, Hagios Nikolaos, 1:435–49; 2:514–26 (Translatio Barim Graece). 74 See G. T. Dennis, “Schism, Union, and the Crusades”; D. M. Nicol, “The Crusades and the Unity of Christendom”; and A. W. Carr, “East, West, and Icons in Twelfth-Century Outremer,” all in The Meeting of Two Worlds: Cultural Exchange between East and West during the Period of the Crusades, ed. V. P. Goss (Kalamazoo, 1986), 181–87, 169–80, 347–59. See also the perceptive articles of A. W. Carr, “Byzantines and Italians on Cyprus: Images from Art,” DOP 49 (1995): 339–57, and J. Folda, “Crusader Art in the Twelfth Century: Reflections on Christian Multiculturalism in the Levant,” Mediterranean Historical Review 10 (1995): 80–91, along with the works cited in his n. 9. S. Griffith, “From Aramaic to Arabic: The Languages of the Monasteries of Palestine in the Byzantine and Early Islamic Periods,” DOP 51 (1997): 11–31, deals with mixed communities in Palestinian monasteries, albeit in a much earlier period. 75 Cyprus is an alternative, especially if the choice of saints on the vita icons will turn out not to be quite as tied to Sinai as I have claimed, but simply to reflect figures popular throughout the various cultures of the Eastern Mediterranean of this time. 76 For Georgian translations of Greek saints’ vitae, see Garitte, Catalogue, passim.
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
1 St. Nicholas vita icon, monastery of St. Catherine, Mt. Sinai (photo: courtesy of the MichiganPrinceton-Alexandria Expedition to Mount Sinai)
2 St. George vita icon, monastery of St. Catherine, Mt. Sinai (photo: courtesy of the Michigan-PrincetonAlexandria Expedition to Mount Sinai)
3
St. Nicholas panel, Pinacoteca Provinciale, Bari (photo: Pinacoteca Provinciale, Bari)
4
St. Nicholas panel, church of San Verano, Peccioli (photo: Alinari/Art Resource, New York)
5 St. Catherine panel, Museo Nazionale di San Matteo, Pisa (photo: Alinari/Art Resource, New York)
6 St. Catherine vita icon, monastery of St. Catherine, Mt. Sinai (photo: courtesy of the Michigan-PrincetonAlexandria Expedition to Mount Sinai)
7 St. Francis panel by Bonaventura Berlinghieri, church of San Francesco, Pescia (photo: Alinari/Art Resource, New York)
8
St. Francis panel, Cappella Bardi, church of Santa Croce, Florence (photo: Alinari/Art Resource, New York)
9
10
Elijah vita icon, Tretiakov Gallery, Moscow (photo: courtesy of Lev Lifshits)
St. Nicholas vita icon, Russian Museum, St. Petersburg (photo: courtesy of Engelina Smirnova)
11
Sts. Boris and Gleb vita icon, Tretiakov Gallery, Moscow (photo: courtesy of Engelina Smirnova)
12
St. Sergej of Radone∑ vita icon, Rublev Museum, Moscow (after V. Lazarev, Moscow School of Icon Painting [ Moscow, 1971], pl. 86)
13
St. Kiril Belozerskij vita icon, Russian Museum, St. Petersburg (photo: courtesy of Engelina Smirnova)
14
Metropolitan Alexis vita icon, Tretiakov Gallery, Moscow (after E. Smirnova, Moscow Icons, 14th–17th Centuries [ Leningrad, 1989], fig. 152)
15
Metropolitan Peter vita icon, Kremlin Museums, Moscow (after Smirnova, Moscow Icons, fig. 151)
16
King Stefan Decanski vita icon, monastery of Decani (after V. Djuri{, Ikona svetog kralja Stefana Decanskog [ Belgrade, 1985], fig. 1)
17
St. Marina vita icon from Pedoulas, Museum of the Archbishop Makarios III Cultural Foundation, Nicosia (photo: D. Mouriki)
18
St. John the Baptist vita icon, monastery of St. Catherine, Mt. Sinai (photo: courtesy of the Michigan-PrincetonAlexandria Expedition to Mount Sinai)
19
Moses vita icon, monastery of St. Catherine, Mt. Sinai (photo: courtesy of the MichiganPrinceton-Alexandria Expedition to Mount Sinai)
20
St. Nicholas vita icon from the church of St. Nicholas tes Steges, Kakopetria, Museum of the Archbishop Makarios III Cultural Foundation, Nicosia (after A. Papageorghiou, Icons of Cyprus [ Nicosia, 1992], fig. 32a)
21
St. George vita icon from Kastoria, Byzantine Museum, Athens (photo: courtesy of the museum)
22
Wooden icon or xoanon of St. George, Omorphecclesia church, near Kastoria (photo: S. Gerstel)
23
Icon of Moses receiving the Law, monastery of St. Catherine, Mt. Sinai (photo: courtesy of the Michigan-Princeton-Alexandria Expedition to Mount Sinai)
24
St. Nicholas fresco icon, church of St. Nicholas tes Steges, Kakopetria (after A. Stylianou and J. Stylianou, The Painted Churches of Cyprus [ London, 1985], fig. 25)
25
Detail, St. Nicholas revives the three salted boys, St. Nicholas vita icon from Kakopetria, Museum of the Archbishop Makarios III Cultural Foundation, Nicosia (photo: I. Sevcenko)
26
Detail, Tomb of St. Nicholas, St. Nicholas vita icon from Kakopetria, Museum of the Archbishop Makarios III Cultural Foundation, Nicosia (photo: I. Sevcenko)
27
Detail, Death of St. Nicholas at Myra, St. Nicholas vita icon from Kakopetria, Museum of the Archbishop Makarios III Cultural Foundation, Nicosia (photo: I. Sevcenko)
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Narrative on the Golden Altar of Sant’Ambrogio in Milan: Presentation and Reception CYNTHIA HAHN
he golden altar of Sant’Ambrogio in Milan is the liturgical focus and the spiritual center of the archiepiscopal city. It is also an extraordinary monument of early medieval art (Fig. 1). The narratives that adorn its front and back—stories of the lives of Christ and of Ambrose—are just two of the many features that secure the altar’s power and prestige. Other features include the altar’s precious materials, its consummate artistry, its status as the tomb of both the bishop Ambrose and martyrs, and its theatrical setting in the church, framed as it is by a ciborium, an apse mosaic, and an elevated floor with a crypt below (Figs. 2, 3).1 Among these many potent elements, however, the story of the sainted bishop on his altar/tomb assumes a particularly important and active role in asserting the altar’s prestige. It purposefully reaches out to the viewer, claiming both that the saint still “lives” and that sacred power is continued in the person of the archbishop of Milan, “quasi sanctus Ambrosius.”2 Indeed, by means of its hagiographic narrative, the golden altar effectively supports Milanese episcopal claims to primacy. Because pictorial hagiography can be such a powerful tool in the service of political interests,3 this essay seeks to analyze precisely how Ambrose’s story constructs its persuasive sacred and political rhetoric. Although the use of narrative on the altar is highly significant, it is not the most striking visual element of the altar, nor is it unprecedented. There is evidence for early use of the saint’s vitae on frontals at prominent “apostolic” sites, and some altars employed
T
This article was originally delivered at the Dumbarton Oaks Symposium as a paper titled “Saints, Stories, and the Sacred: The Discourse of Reliquaries.” I would like to thank the two readers, Larry Nees (who identified himself to me) and the other who remains anonymous, for their many valuable suggestions for revising this paper. I would also like to thank Areli Marina and Louis Demos for their timely assistance. 1 Although the present finished ciborium is of a later date, the altar always had a ciborium; C. Bertelli, “Mosaici a Milano,” in Atti del 10. Congresso internazionale di Studi sull’alto Medioevo, Milano, 26–30 settembre 1983 (Spoleto, 1986), 345. 2 Lothar said as much of Angilbertus II; A. Ambrosioni, “Gli Arcivescovi nella vita di Milano,” in Atti del 10. Congresso internazionale (as in note 1 above), 110. Bibliography on the effects and effectiveness of narratives, hagiographic and otherwise, is ample. See C. Hahn, “Picturing the Text: Narrative in the Life of the Saints,” Art History 13 (1990): 1–32; and eadem, Portrayed on the Heart (forthcoming). 3 B. Abou-el-Haj, The Medieval Cult of Saints: Formations and Transformations (Cambridge, 1994), 11–12. She implicates the altar in similar political maneuverings among Pavia, Venice, and Ravenna.
168
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
narratives of the life of Christ.4 Nevertheless, the majority of altars were decorated with symbolic ornament alone. Such ornamentation served not only to beautify the altars to the “greater glory of God,” but also to locate them in a context of reception that amplified their sanctity and significance.5 The Milan altar is not lacking in conventional ornament; on the contrary, enameled bands, applied gems, and other beautiful and precious materials are present in abundance. Moreover, on the front of the altar, all narrative and figural material is clearly subordinate to the great jeweled, equal-armed cross at the center (Fig. 1). Notably, when the artist represented the altar on the altar, he found it sufficient to depict a box adorned with prominent bands in a generalized cross shape (Fig. 13). The primary message of the altar is thus inarguably dogmatic and hierarchical.6 It centers on the meaning and truth of the Passion of Christ as expressed in the Eucharist, on its implications for the church, and on its continuing significance as conveyed through the saints to whom most altars are dedicated.7 Ambrose himself explicated the meaning of this altar/tomb when he wrote about the fourth-century original: 4 Joseph Braun is the best source for a survey of altars and their ornamentation; see his Der christliche Altar in seiner geschichtlichen Entwicklung, 2 vols. (Munich, 1924). Altars of precious metal were particularly characteristic of the Constantinian donations; Liber Pontificalis 1.34, trans. R. Davis, The Book of Pontiffs (Liber Pontificalis): The Ancient Biographies of the First Ninety Roman Bishops to A.D. 715, Translated Texts for Historians, Latin ser., 5 (Liverpool, 1989). Constantine gave seven silver altars to the Lateran (ibid., p. 16), and one to the basilica of Sts. Peter and Marcellinus (p. 23). Martyria did not generally have fixed altars, but over the graves of Peter and Paul Constantine placed precious metal crosses (pp. 18, 20), and in the crypt of Lawrence he placed the saint’s “passio in medallions chased with silver” (p. 22). By the time of the Ambrose altar, however, there were, of course, altars in St. Peter’s basilica. Beginning with Hadrian I, the 8th-century popes were generous with gold and silver at the altar of Peter’s confessio; The Lives of the Eighth-Century Popes (Liber Pontificalis): The Ancient Biographies of Nine Popes from A.D. 715 to A.D. 817, trans. R. Davis, Translated Texts for Historians 13 (Liverpool, 1992), 97.83, 93: pp. 166, 170. Leo III even added a gold pavement (ibid., 98.51: p. 203). Following the sack of 846, Leo IV (847–855) found it appropriate to replace stolen liturgical treasures, including “imagines argenteas totasque exauratas” (Liber Pontificalis) at Peter’s grave and, at the main altar, an altar frontal with gold and silver tablets, with subjects from the Old and New Testaments. See J. Croquison, “L’iconographie chre´tienne `a Rome d’apre`s le Liber Pontificalis,” Byzantion 34 (1964): 35–37, who notes a similarity to the Ambrose altar. Perhaps there might be said to have been a wave of precious altars or frontals produced in the Carolingian era. A lost example from Reims celebrated Nicaise, Remi, and the Virgin; it was decorated with precious metals and gems, and similarly celebrated “this seat,” ca. 845(?)–862; G. Kornbluth, Engraved Gems of the Carolingian Empire (University Park, Pa., 1995), 121–22. Braun mentions other examples in Italy and France, including that, of course, of St. Denis given by Charles the Bald (Altar, 2:87–90). S. Bandera, “L’altare di Sant’Ambrogio: Indagine storico-artistica,” in L’altare d’oro di Sant’Ambrogio, ed. C. Capponi (Milan, 1996), 79, cites a number of Carolingian altars and ciboria. (S. Bandera also wrote L’altare d’oro di Sant’Ambrogio [Milan, 1995], which I have not been able to consult.) Other than those at saints’ shrines in Rome, the precious altars tend to be episcopal or imperial gifts and usually celebrate episcopal seats. (Although St. Denis is not the cathedral of Paris, Denis himself was the bishop.) Carolingians could have been following the lead either of altars in Rome or of lavish imperial donations made in the East, such as altars in Hagia Sophia and more than one at the Holy Sepulchre; Braun, Altar, 1:110, 116. It is interesting to note that by the 11th century the cathedral of Milan also had a golden altar; A. Ambrosioni, “L’altare e le due comunita` santambrosiane,” in L’altare d’oro, ed. Capponi (as above), 70 n. 13. 5 E. Dahl, “Dilexi decorem domus Dei: Building to the Glory of God in the Middle Ages,” Acta ad archaeologiam et artium historiam pertinentia 1 (1981): 157–90; and O. Grabar, The Mediation of Ornament (Princeton, N.J., 1989). 6 Some interesting recent works on medieval narrative in constructed and meaningful settings include R. Brilliant, “The Bayeux Tapestry: A Stripped Narrative for Their Eyes and Ears,” Word and Image 7 (1991): 98–126; and M. Kupfer, Romanesque Wall Painting in Central France: The Politics of Narrative (New Haven, Conn., 1993). 7 An interesting document to consider in a discussion of the typical ornamentation of altars is the Codex Calixtinus. Although it dates well after the Ambrose altar, it may be representative. It lists a number of altars,
CYNTHIA HAHN
169
Let the triumphant victims [martyrs] take their place where Christ is the victim. Let Him be above the altar who suffered for all; let them be beneath the altar who were redeemed by His suffering. This is the spot that I had destined for myself because it is fitting that a bishop rest where he was wont to offer the Holy Sacrifice. But I yield the right-hand portion to the sacred victims, that place is owed the martyrs [Gervasius and Protasius].8
The same multilayered equivalence of Christ’s sacrifice, martyr’s sacrifice, and bishop’s sacrifice expressed through hierarchical spatial organization clearly orders the ornamentation of the Milan altar.9 In contrast to the powerful but conventional significance of the altar, Ambrose’s story offers the appeal of specific, complex, and discursive meanings. Amidst the dominant chords of symbol and ornament, pictorial narrative, as it were, plays a quiet but distinctive melody. Yet, in order to be heard, that melody must be carefully attended. The altar’s reception depends on the abilities of its audience to discern the significance of the saint’s story in relationship to its complex environment.10 The altar makes a presentation, but any particular reading is only one of many that are possible—a momentary intersection among an infinite number of stories, histories, and symbols. CIRCUMSTANCES OF THE ALTAR The ninth-century reliquary-altar-tomb of St. Ambrose in Sant’Ambrogio is rare among early medieval altars in both its state of preservation and its relatively full documentation. We know something about the circumstances of its production: the altar was made by the artist Wolvinus at the behest of Bishop Angilbertus II (824–859). Furthermore, it has been investigated archaeologically and recently restored;11 and its civic setting, Milan, has been the subject of much recent research.12 The altar is often called a paliotto, but, rather than being an altar frontal, it is a complete altar enclosing the tomb of Ambrose, fabricated entirely of gold and silver, with added gems and enamels. most of which represent Christ, sometimes with the college of apostles; see J. Vielliard, Le Guide du Pe`lerin de Saint-Jacques de Compostelle (Paris, 1984). See also the critical edition and translation, P. Gerson et al., eds., A Pilgrim’s Guide to Santiago de Compostela: A Critical Edition (London, 1998). 8 Ambrose, Epistola 22.13: PL 16:1066; Saint Ambrose: Letters, trans. M. M. Beyenka (New York, 1954), 380, 381. 9 Although at the time of the manufacture of the altar, and perhaps even earlier, Ambrose was tellingly shifted to the center, between and above the two martyrs; J.-C. Picard, Le souvenir des ´eveˆques: Se´pultures, listes ´episcopales et culte des ´eveˆques en Italie du nord des origines au Xe sie`cle, Bibliothe`que des ´ecoles franc¸aises d’Athe`nes et de Rome 268 (Rome, 1988), fig. 59. 10 See C. Hahn, “Seeing and Believing: The Construction of Sanctity in Early Medieval Saints’ Shrines,” Speculum 72 (1997): 1079–106. 11 Picard (Souvenir, 625 n. 158) cites F. M. Rossi, Cronaca dei ristauri e delle scoperte fatte nell’insigne basilica di Sant’Ambrogio dall’anno 1857 al 1876 (Milan, 1884); L. Biraghi, I tre sepolchri santambrosiani scoperti nel gennaio 1864 (Milan, 1864); and F. Reggiori, La basilica di Sant’Ambrogio (Milan, 1966), 101. The restoration of the altar in 1996 under the direction of Carlo Capponi is documented and celebrated with a full set of color photographs and a series of articles in L’altare d’oro, ed. Capponi. 12 As well as research on the Christian tradition and geography of Milan. See Picard, Souvenir; F. Monfrin, ` propos de Milan chre´tien: Sie`ge ´episcopal et topographie chre´tienne, IVe–VIe sie`cle,” CahArch 9 (1991): “A 7–46; C. Bertelli, ed., Il Millennio Ambrosiano, 3 vols. (Milan, 1987–88), esp. C. Bertelli, “Sant’Ambrogio da Angilberto II a Gotofredo,” in ibid., 1:19 ff; V. H. Elbern, “Mailand–Spa¨tantike Kaiserstadt,” Aachener Kunstbla¨tter 58 (1989–90): 11–30; and R. Krautheimer, Three Christian Capitals: Topography and Politics (Berkeley, Calif., 1982). Recently see also M. L. Gatti Perer, ed., La basilica di S. Ambrogio: Il tempio ininterrotto (Milan, 1995), with numerous articles summing up the state of research on the church.
170
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
The public side of the altar, facing the nave, proclaims the life of Christ, with particular focus on the Passion, in twelve framed narrative scenes in gold repousse´ (Fig. 1).13 On the other side of the altar, facing the liturgical celebrant (according to the Ambrosian rite),14 the life of St. Ambrose is presented, in twelve framed scenes in partially gilded silver repousse´ (Fig. 4). In his published dissertation on the altar, Viktor Elbern proposed that the two narratives functioned in typological opposition.15 Although there is some general relationship of this kind between the front and the back of the altar, their scenes lack a one-to-one correspondence that is essential to typology. Instead, there are complex relationships among the scenes on the back, which collectively, and occasionally specifically, refer to the life of Christ on the front of the altar. Rather than establishing a set of typological links, the relationship works to locate the saint as part of the church, the institution founded on the Passion of Christ and realized in the celebration of the Mass. The life of Christ therefore provides a rhetorical “frame” within which the hagiographic narrative on the altar promises a continuing history of salvation, culminating in the guarantee of Ambrose’s powerful intercession for his supplicants. The specifics of the altar’s layout are crucial to reading its narrative. In contrast to the scenes from the life of Christ, which read left to right in rising registers on the left side and begin again from the bottom on the right (Fig. 1), the scenes of Ambrose’s life begin at the lower left and progress from left to right in each of the three horizontal registers—each register “jumping” across the space of a large central, non-narrative area—leading to a de´nouement in the upper right (Fig. 4). In a second difference between the back and the front, the less familiar narrative from the life of Ambrose includes an additional discursive level of labels, or tituli, which are not included in the scenes from the life of Christ. These labels both secure and direct the meaning of the story of Ambrose. Ten of the episodes are drawn from the fifthcentury vita of Ambrose by Paulinus, as the wording of the tituli indicates;16 the remaining two derive from a story in the Libri de virtutibus sancti Martini episcopi by Gregory of Tours.17 Notably, Paulinus’s vita and the episode from Gregory are also combined in a Carolingian vita of Ambrose, written shortly after the altar was made.18 In general, this latter text matches the narrative on the altar less well than the earlier vita, but it can serve as a useful reflection of contemporary reception of the altar’s story. As the discussion below shows, although each scene on the altar can be located in a textual source, the motives behind the selection of the specific scenes from Ambrose’s vitae are to be found in con13 The scenes from the life of Christ include, on the left, the Annunciation, the Nativity, the presentation in the Temple, the miracle at Cana, Christ called by Jairus, and the Transfiguration; and on the right, the cleansing of the Temple, the healing of the blind man, the Crucifixion, the Pentecost, the Resurrection, and the Ascension. The last three are the Renaissance replacements of the Carolingian originals. 14 V. Elbern, “Der Karolingische Goldschmiedekunst,” in Atti del 10. Congresso internazionale (as in note 1 above), 302, cites E. Tea, Arti minori nelle chiese di Milano (Milan, 1950), 25, noting that the celebration of the mass “al ambrosiana” was “versus populum.” 15 V. Elbern, Der Karolingische Goldaltar von Mailand (Bonn, 1952), 23 ff. 16 PL 14:65–114, trans. J. A. Lacy, “Life of St. Ambrose by Paulinus,” in Early Christian Biographies, ed. R. J. Deferrari (New York, 1952), 27–66. 17 Libri de virtutibus sancti Martini episcopi, ed. B. Krusch, MGH ScriptRerMerov (1885), 1:562–84; trans. R. Van Dam, Saints and Their Miracles in Late Antique Gaul (Princeton, N.J., 1993), 207. 18 A. Paredi, ed., Vita e meriti di S. Ambrogio: Testo inedito del secolo nono illustrato con le miniature del Salterio di Arnolfo (Milan, 1964). Ambrosioni dates the text to ca. 880 (“Arcivescovi,” 112 n. 101).
CYNTHIA HAHN
171
temporary events occurring in the archbishopric of Milan and efforts on the part of that institution to bring glory to the newly acknowledged “Ecclesia ambrosiana.” 19 The unusual choice of the scenes selected can be highlighted by contrasting them with the more typical emphases in other texts concerning Ambrose’s life. A hymn to Ambrose by Ennodius from the early sixth century focuses on three groups of events from the saint’s life: his Orthodox faith and his development of the cult of martyrs; his conflict with the empress Justina; and his role as a “shepherd” and theologian.20 The later ninth-century Carolingian vita, mentioned above, similarly focuses on the cult of martyrs, politics, and theology, as does the Paulinus vita, if to a lesser degree. As parallels to these textual examples, any number of images of St. Ambrose depicted as a theologian might be cited, such as, for example, the author portrait from the eighth-century Egino Codex.21 Despite this clear pattern of hagiographic assessment of Ambrose as sainted theologian and able politician, almost no reference to these themes can be found on the golden altar, which clearly reconfigures the life of Ambrose in a new reading. There is no indication of conflict with heretics or secular authority, nor is there direct depiction of Ambrose as a theologian and writer. But even more surprising than these lacunae is the absence of any reference to Ambrose’s involvement in the cult of martyrs, which, in Peter Brown’s estimation, was an essential element of the saint’s importance.22 One should recall that, in addition to the body of Ambrose, this reliquary-altar-tomb also contains the relics of Gervasius and Protasius, the martyrs whom Ambrose discovered and translated to his basilica—yet they are barely acknowledged on the altar, appearing only on one end, in simple, tiny portrait roundels (Fig. 17). THE NARRATIVE SCENES OF THE LIFE OF SAINT AMBROSE What scenes from the saint’s life are chosen for representation on the altar (Figs. 5, 6)? The first one seems to be a logical starting place. Ambrose is born; he is shown in his cradle (Fig. 5, lower left). Despite the apparent simplicity of this first scene, a number of important themes are introduced here. The label across the bottom of the panel reads, “In which a swarm of bees filled the mouth of the boy Ambrose.” 23 The titulus is taken from Paulinus and identifies the first miracle of Ambrose’s life: not only did the bees not harm the infant, they filled the baby’s mouth with honey. This topos is generally used to foretell the honeyed words of the gifted orator—the same astounding story was told of the infants Pindar, Hesiod, and Plato24 —but there is more here. The bees, as Paulinus tells us, afterwards ascended to heaven, above the clouds. The clouds in the altar scene are thus included in order to remind the viewer of the bees’ celestial ascent. Picard, Souvenir, 628. Ibid., 614. 21 P. Courcelle, Recherches sur saint Ambroise: “Vies” anciennes, culture, iconographie (Paris, 1973), 156, pl. II. This miniature is part of a book that was written for Egino of Verona, but decorated in a northern style related to the Ada school, providing an interesting comparison for the Milan altar with its northern ties. 22 P. Brown, The Cult of the Saints (Chicago, 1981). 23 UBI EXAM(EN) APU(M) PUERI OS CO(M)PLEVIT A(M)BROSI(I). See G. B. Tatum, “The Paliotto of Sant’ Ambrogio at Milan,” ArtB 26 (1944): 29; Lacy, “Life of St. Ambrose,” 3, p. 35: “well-ordered words are as a honey-comb” (Prov. 16:24.) 24 Elbern, Goldaltar, 43. 19 20
172
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
The Carolingian vita further specifies that the bees came from God.25 The insects are not a test of the infant saint, but his attribute: their honey is his by birthright, denoted by his very name, Ambrosius/ambrosia.26 In congruence with his name, Ambrose’s writings are repeatedly referred to in the Middle Ages as “nectar.” 27 Instead of harming him, the insects of celestial origin mark a foreordained source of doctrinal sweetness, depositing the miraculous nectar as the infant’s first taste of the divine and returning to heaven from whence they came. The attentive viewer might also note the elaboration of the legs of the cradle. The same beading is repeated as a decorative motif throughout the reliefs, especially on holy objects such as church furniture.28 Elbern compared the birth image to a scene of the Nativity of Christ, such as, for example, the one on the front of the altar. Even more tellingly, he also compared the Ambrose birth scene and the Nativity from the front of the altar to Ottonian manuscript images, such as that in the Lectionary of Henry II, in which Joseph and Mary present the Christ child while standing to either side of a crib that in the manuscript is clearly meant to double as an altar (Fig. 7).29 If one understands Ambrose’s parents as making the same gesture of presentation, the child Ambrose similarly becomes the sacrifice on the altar. Such a metaphorical consonance—Christ as sacrifice, and the saint in his stead—is typical in hagiography and essential to the meaning of the altar.30 Indeed, the sacramental tone of the image explains the absence of the nurse of the Paulinus vita, who was said to have discovered the bees and wished to chase them away: she would become an unwelcome anecdotal distraction in the altar version.31 Thus, already in this first scene, the themes of the altar have been introduced: it will present a discourse about the bishop’s place in the church, especially his role as font of nourishing (and sweet) doctrine and his eventual sacrifice through his service. The next two scenes, a pendant pair on either side of the central area, depict Ambrose on horseback, seemingly as two parts of the same journey (Figs. 8, 9). In fact, two different journeys are represented here. The first is Ambrose’s departure from Rome as a young lawyer, as described by Paulinus (although, interestingly, the titulus specifies that he is called, “petit,” to his new vocation).32 The second is a much later episode concerning Ambrose’s attempted escape from his episcopal duty in Milan. By this time the people of Milan had already acclaimed him bishop—an incident that was considered miraculous Paredi, Vita, 5–6, p. 26. Elbern, Goldaltar, 43. 27 Courcelle, Recherches, 132, esp. n. 2. 28 Other examples of beading include the font in the fourth scene (baptism); the altar in the sixth scene (sleeping at the altar); the sarcophagus in the seventh scene (funeral of St. Martin); the altar in the ninth scene (foot healing); the beds in the tenth and eleventh scenes (visions of Ambrose and Honoratus); and the bed/altar in the twelfth scene (death of Ambrose), as well as objects and architecture on the opposite side of the altar. 29 Elbern, Goldaltar, 43. 30 See above, pp. 168–69, and also C. Hahn, Passio Kiliani; Ps. Theotimus, Passio Margaretae; Orationes: Vollsta¨ndige Faksimile-Ausgabe im Originalformat des Codex Ms. I 189 aus dem Besitz der Niedersa¨chsischen Landesbibliothek Hannover (Graz, 1988), 116. 31 Although Pierre Courcelle argues that the artist left her out because she did not fit in the small space of the panel (Recherches, 172). 32 UBI A(M)BROSI(U)S EMILIA(M) PETIT AC LIGURIA(M). See Tatum, “Paliotto,” 29; Lacy, “Life of St. Ambrose,” 5, pp. 35–36. 25 26
CYNTHIA HAHN
173
at the time of the creation of the altar,33 since it meant that the people had recognized Ambrose’s abilities and duty even before he was baptized. However, the miraculous recognition by the people is specifically not shown on the altar; in its stead, the saint, on horseback, is turned back by God’s call. The titulus below reads, “In which, while fleeing, he is turned back by the breath of the Holy Spirit,” a detail missing from Paulinus.34 The episode is reminiscent of the calling of Saul/Paul on the road to Damascus, perhaps thus implicating an apostolic reference.35 Indeed, Paulinus notes that at this point in the life of Ambrose God who “was preparing . . . a tower of David against the face of Damascus . . . checked his flight.” 36 In typical hagiographic fashion, Ambrose’s many sins (even feigned lechery), which are detailed in the fifth-century narrative of Paulinus (but only vaguely alluded to in the Carolingian vita), are represented here symbolically.37 Thus, Ambrose’s willful character is conveyed solely through a depiction of his proud stallion with its enormous genitalia. Furthermore, by means of this horse Ambrose is characterized as a noble, a status that implicated pride and unchecked power.38 In turning to God’s call in the scene parallel to but reversed in direction from the previous, more civic scene, Ambrose is specifically shown to give up his elevated secular status: he uses the reins and the prominently pictured stirrups to turn the horse forcibly and accept God’s will. In the next scene, Ambrose is baptized (Fig. 10).39 Following Carlo Romussi, Elbern argued that the image conformed to the early Milanese practice of combining immersion with infusion.40 With the baptism as the final scene on the lower register, the depiction of Ambrose’s secular life is concluded. In this last moment, by virtue of the ceremony of baptism, not only is Ambrose reborn to a new life as a Christian, but, humble and nude, he also visibly renounces the accoutrements of status and power of his past. Both Paulinus and the Carolingian hagiographer are careful to note that the bishop who performed the baptism was Orthodox.41 On the next register, Ambrose assumes a new garb, that of liturgical vestments. The titulus of the first scene specifies that “on the eighth day he is ordained bishop.” 42 In the relief, the saint is acclaimed and ordained by his fellow bishops (Fig. 5, middle left). For the first time a halo marks his sanctity. Notably, Paulinus’s vita reads very differently than Courcelle, Recherches, 126. UBI FUGIENS SP(IRIT)U S(AN)C(T)O FLANTE REVERTITUR. See Tatum, “Paliotto,” 29. One possible reason for the pairing of the two scenes is that the Carolingian vita notes that Ambrose is also “called” to his secular life; see Paredi, Vita, 5–6, p. 26. 35 L. Eleen, “Acts Illustration in Italy and Byzantium,” DOP 31 (1977): 255–80. 36 Lacy, “Life of St. Ambrose,” 7–8, p. 37. 37 Ibid.; Paredi, Vita, 10–14, pp. 31–36. 38 A horse often signifies the vice superbia: J. Traeger, “Pferd,” in Lexikon der Christlichen Ikonographie (Rome, 1971), 3:413; although most examples date later than the altar. 39 The titulus reads, UBI CATHOLICO BAPTIZATUR EP(ISCOP)O. See Tatum, “Paliotto,” 30; Lacy, “Life of St. Ambrose,” 9, p. 38. 40 Elbern suggests that the image combines two formulae for the baptism of Christ; Goldaltar, 45–46. He cites C. Romussi, Milano nei suoi monumenti (Milan, 1912), 1:14. For the baptismal rite in Milan, see E. Cattaneo, “Storia e particolarita` del rito ambrosiano,” in Storia di Milano (Milan, 1954), 3:831. 41 Lacy, “Life of St. Ambrose,” 9, p. 38; Paredi, Vita, 11, p. 34. 42 UBI OCTAVO DIE ORDINATUR EP(ISCOPU)S. See Tatum, “Paliotto,” 30; Lacy, “Life of St. Ambrose,” 9, p. 38. 33 34
174
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
does the sequence of events implied by these last three reliefs. In the fifth-century vita, the people spontaneously acclaimed Ambrose, recognizing him as the politic solution to the strife in the city caused by dissension between the Arians and the Orthodox. Liturgical ordination is not mentioned at all. In contrast, in the Carolingian vita, the process of the bishop’s election is initiated by the emperor but left to the bishops, who, the emperor insists, will make a better choice. In the process, a public meeting is held, but in place of the spontaneous acclamation by the people, a single “celestial” voice is heard to nominate Ambrose to office.43 Not surprisingly, the relief is more liturgically and institutionally oriented than the fifth-century vita. As does the ninth-century account, the image emphasizes Ambrose’s reception by and place among other clergy, especially other bishops, initiating the liturgical theme that dominates the second register. It is significant that the Carolingian vita emphasizes the emperor’s recognition of the dignity and independence of episcopal election, which is essentially what is depicted here, especially in the absence of the emperor.44 The celestial recognition of Ambrose’s fitness for the episcopal state, marked by miraculous events in each version of the vita, is represented on the altar first by the hand of God that points Ambrose to his calling and then by the halo that shows the fulfillment of the divine plan.45 The next two scenes (Fig. 5, middle right; Fig. 11) are again linked across the central panel, as were the equestrian scenes directly below. In this episode taken from the Libri de virtutibus sancti Martini episcopi, Ambrose’s brotherhood with St. Martin is demonstrated as well as his importance in the community of bishops. While celebrating a mass, just before the reading of the Epistle (the lector on the left is poised, ready to read), Ambrose fell asleep. The people and the clergy waited patiently for two or three hours, but finally moved to wake the bishop. Upon awakening, Ambrose explained that the great Bishop Martin had died and that he (Ambrose) had gone to Tours and helped to celebrate the funeral.46 The titulus of the first scene reads, “In which as he sleeps at the altar he visits Tours” 47; and that of the second relief continues, “In which he buried the body of the blessed Martin.” 48 In the scenes on the altar, Ambrose is much more active than simply chanting the Psalms, as he does in Gregory’s text: here he actually handles the shroud and the holy haloed head, as the titulus implies. As noted above, the Carolingian vita includes this scene, although the author seems to be uncertain about its chronology and squeezes it in at the end.49 On the altar, the depiction of this extraordinary episode as two scenes that are placed in the very center of a narrative limited to twelve scenes signals its importance. See note 45. This is reminiscent of an incident that occurred between Bishop Angilbertus II and Lothar. In refusing to bow to the emperor, Angilbertus claimed a special status based on the honor of his church. Lothar admitted that he was “quasi sanctus Ambrosius.” See Ambrosioni, “Arcivescovi,” 110. 45 Similarly, in the Fulda manuscript of his vita, Kilian of Wu ¨ rzburg is first shown with a halo only when he assumes a liturgical role; Hahn, Passio Kiliani, 50. Barbara Abou-el-Haj discusses episcopal ordination in somewhat later cases, citing the Ambrose image as an example, in “Consecration and Investiture in the Life of Saint Amand, Valenciennes Bibl. Mun. Ms. 502,” ArtB 61 (1979): 342–58. 46 Van Dam, Saints, 207. 47 UBI SU(PE)R ALTARE DORMIENS TURONIAM (P)ETIT. See Tatum, “Paliotto,” 30. 48 UBI SEPELVIT CORPUS BEATI MARTINI. See ibid. 49 As well he should, since Ambrose died before Martin and could not possibly have attended his funeral. 43 44
CYNTHIA HAHN
175
Remarkably, the subject was also chosen for the ninth- or tenth-century mosaics that are included in the apse of the church (Fig. 2, just above the windows, left and right).50 In those mosaics Ambrose assumes an even more prominent role in the funeral ceremonies, seemingly leading the liturgical celebration of the funeral.51 The appearance of the person of Bishop Martin in these narratives seems puzzling at first, but a number of explanations come to the fore. In the ninth (and also the tenth) century, Martin was the premier saint of Carolingian France, a status that would have been recognized by the Frankish patron of the altar, Angilbertus.52 Martin was known in Italy as the “hammer of heretics” and was thus a spiritual brother to Ambrose, who was known as a scourge of the Arians.53 The close accord of the Frankish Martin and the Milanese patron may in some sense also reflect a desire of the two populations of Milan—Frankish overlords and Milanese citizens—to reach a similar accord.54 Ultimately, however, this scene may have been included in order to call attention to Ambrose’s status as equal to the Frankish bishop and therefore as part of a sort of spiritual brotherhood of sainted bishops.55 Contemporary to the altar, numerous other bishops’ shrines were constructed throughout the empire at monasteries practicing laus perennis, creating a series of foci of intercessory prayer for the empire and its citizens, and especially for its clergy. Exchanges among these monasteries, formalized as confraternities of prayer, dealt in the valuable commodity of saintly patronage through the medium of liturgical commemoration.56 Sant’Ambrogio was an important part of this network of prayer. In the eighth century, Frankish monks had been brought to Milan to found a monastery at Sant’Ambrogio that specialized in the laus perennis and the accompanying prayer for the dead.57 The monks 50 Carlo Bertelli dates these mosaics to the 9th century, although many scholars would have them date to the 10th. He connects the pseudo-Greek of the inscriptions with similar 9th-century Graecisms, and also notes a Byzantine use of color. I would think that the mosaics represent the completion of the program of the apse and the choir, and postdate the altar. Cf. Bertelli, “Mosaici a Milano,” 333–51; also see bibliography ` in Courcelle, Recherches, 180–82; and summary in C. Bertelli, “Percorso tra le testimonianze figurative piu antiche: Dai mosaici di S. Vittore in Ciel d’oro al pulpito della basilica,” in S. Ambrogio, ed. Gatti Perer, 364. 51 Surprisingly, the mosaic episodes are in reverse order in comparison to the story as it appears on the altar. This cannot be the result of a mistake, as has been suggested, but must be purposeful. In fact, the episode of Ambrose at the altar does not take place solely before the Touronian episode, but acts as a frame— Ambrose is awakened and explains where he has been after the event. The present arrangement in the mosaics does result in the final episode occurring in Milan, with the effect that, perhaps, Milan is seen as heir to the saintly spirit of the great Bishop Martin. 52 Angilbertus II was Frankish and the patron of other Frankish ecclesiastics; Picard, Souvenir, 242, 625. Sandrina Bandera (“L’altare,” 84–91) even links the altar to the Carolingian image controversy and Touronian manuscripts. 53 O. von Simson, Sacred Fortress: Byzantine Art and Statecraft in Ravenna (Chicago, 1948), 83, 138. Martin is even cited in the litanies as an enemy of heresy; Bandera (“L’altare,” 74) prefers this explanation. 54 Annamaria Ambrosioni (“Arcivescovi,” 106) speaks of the inclusion of the Martin episode in the Carolingian vita of Ambrose. 55 Imagery in the vita recalls the saint’s status among the “stars”; Paredi, Vita, 4, p. 24. This is a topos in saints’ vitae. It occurs in the epigrams at St. Martin at Tours; see Van Dam, Saints, 312. 56 These were shrines such as those of Germanus or St. Denis. See M. McLaughlin, Consorting with Saints: Prayer for the Dead in Early Medieval France (Ithaca, N.Y., 1994), 88–89; and the essays in K. Schmid and J. Wollasch, eds., Memoria: Der geschichtliche Zeugniswert des liturgischen Gedenkens im Mittelalter (Munich, 1984). 57 McLaughlin, Consorting with Saints, 60–67 and passim.
176
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
were specifically to pray for the king, his people, the church of Milan, and the founder of the monastery (the archbishop Peter).58 Specific evidence of a confraternity survives from the ninth century. Angilbertus himself, for example, wrote to Fulda, asking to be included in the monk’s prayers at the shrine of Bonifatius and promising prayers in return.59 Perhaps, in the scenes on the altar St. Ambrose is pictured as initiating such a commemoration and spiritual confraternity with the monastery of St. Martin. This narrative of brotherhood thus serves to tie Milan to the empire while at the same time emphasizing a certain independence and spiritual prestige. The second register of the altar is completed with a deceptively simple preaching scene (Fig. 12): Ambrose, on the left, speaks to a group of three standing listeners. The image surprises the viewer by its lack of an ecclesiastical context. No architecture or church furniture is present, as they are in so many of the other reliefs; and, although the Carolingian vita did emphasize Ambrose’s role as a teacher, especially of the clergy,60 this audience is lay. Elsewhere I have argued that this scene and others like it in the vitae of missionaries are based on the images from apostolic Lives originating in Early Christian Rome, specifically known to us from the walls of St. Peter’s and St. Paul’s basilicas.61 These images were still extant in the time of the ninth-century Milanese viewers and may therefore have been familiar to some of them. If so, Ambrose’s preaching is not merely simple but takes on apostolic simplicity. One other remarkable aspect of the image, of course, is that a seemingly fleshly angel stands behind the saint, leaning heavily upon the saint’s back and whispering intimately into his ear. Ambrose seems almost not to react, but, as the titulus specifies, he serenely “preaches with the words of an angel,” 62 a point on which both Paulinus and the Carolingian vita are in agreement.63 This angelic inspiration of the saint’s speech and the miracle of the visible angel are to be differentiated from Ambrose’s native eloquence as demonstrated in the initial scene of the bees, in that the angelic words indicate divine inspiration. In keeping with the liturgical emphasis of the whole second register, Ambrose is liturgically garbed, despite the lack of a formal setting. The final and topmost register begins with what must be considered a eucharistic miracle (Fig. 13). Nicetius, a man of the tribune class, approached the altar to receive the sacraments and was trodden upon by the celebrant Ambrose.64 The inadvertent trampling cured the man’s crippled feet. This episode is not an important feature in the text of Paulinus, and does not occur at all in the Carolingian vita. Wolvinus’s depiction of the scene “in which Ambrose treads on the foot of a sick man,” as it is labeled by the titulus,65 perhaps explains the saint’s clumsiness as a result of his focused attention on the Eucharist. The sacred elements are prominent upon the altar; Elbern notes that the carefully Ambrosioni, “Arcivescovi,” 108. E. Cattaneo, “La tradizione e il rito ambrosiani nell’ambiente lombardo-medioevale,” in La Chiesa di Ambrogio: Studi di storia e di liturgia (Milan, 1984), 13 n. 25, citing MGH Ep, 5:532. Bandera cites other, more general connections with Fulda (“L’altare,” 94). 60 Paredi, Vita, 16, p. 38. 61 Hahn, Passio Kiliani, 75. 62 UBI PREDICAT A(N)G(E)LO LOQ(UE)NTE A(M)BROSIU(S). See Tatum, “Paliotto,” 30. 63 Paulinus specifies that the sight of the angel converted an Arian; Lacy, “Life of St. Ambrose,” 17, p. 43. In the Carolingian vita, the angel is also visible and whispers “celestial doctrine”; Paredi, Vita, 90, p. 144. 64 Lacy, “Life of St. Ambrose,” 44, p. 60. 65 UBI PEDE(M) A(M)BROSIUS CALCAT DOLENTI. See Tatum, “Paliotto,” 30. 58 59
CYNTHIA HAHN
177
depicted four-part fractio panis follows the Ambrosian rite.66 Furthermore, the altar in the relief conspicuously matches the altar represented in Angilbertus’s dedication image (see Fig. 16). It appears that the glorious nature of the Eucharist, celebrated at the famous altar according to the revered Ambrosian rite, is the true subject of the image. If so, in this miracle, which is the only one in the narrative focusing on the miraculous cause and effect, the altar itself may be part of the cause.67 The Carolingian vita dwells at length on another episode concerning a layman’s approach to the altar. In that story, the emperor Theodosius approached the Milan altar, as was his custom at Hagia Sophia. Ambrose reprimanded the emperor, saying that his place was outside the cancelli; in Milan only priests were allowed to approach the altar.68 Is it a coincidence that in the relief a priest approaches the altar from the right, while Nicetius seems almost to get a gentle reprimand from Ambrose along with the cure of his foot?69 The remainder of the topmost register concerns Ambrose’s death and celestial ascent. The fact that it begins with the celebration of the Eucharist, which is presented as the most sacred of the events, seems fitting. It ends also with an altar and the sacrifice of the Mass. The two scenes in between are, once more, constructed as paired pendants framing the central opening (Figs. 14, 15).70 They represent Ambrose’s vision of Christ at his deathbed, and Bishop Honoratus of Vercelli’s vision of an angel telling him of Ambrose’s impending death. The scenes are reversed but alike, down to the representation of the shoes left at the side of the bed. Indeed, these shoes are often noted as a charming detail. They may, however, have greater significance. Given that the details of saints’ vitae were invariably charged with meaning rather than whimsy, that the pontifical sandalia were a mark of sanctity of the celebrant and often emphasized in images,71 that Ambrose had just performed a miracle with his foot, and that in both cases the footwear was “displayed” on a jeweled footstool like that in the fourth scene of the baptism or that behind the altar in the sixth scene, one must conclude that these shoes are purposefully represented. I would suggest that they stand for the active and liturgically embedded life of the bishop, that is, a life that is quintessentially repetitive and shared. They are a mark of the earthly episcopal office, put aside while each bishop as an individual receives a heavenly vision.72 Elbern, Goldaltar, 50. For the Ambrosian rite, see Cattaneo, “Storia e particolarita`,” 811. This miracle, with the saint’s inattention to the cure, might be compared to a miracle from the life of Christ, namely, the healing of the woman with the issue of blood. In that miracle, Christ felt the touch that effected the cure, but did not consciously will the miracle. 68 Paredi, Vita, 58, p. 98. 69 Paulinus writes that the man testified to the secretive cure after Ambrose’s death. He heard the saint say, “Go, and be well”; Lacy, “Life of St. Ambrose,” 44, p. 60. 70 The tituli read, UBI IE(S)UM AD SE VIDET VENIENTE(M); and UBI AM(M)ONIT(US) HONORAT(US) EP(ISCOPU)S D(OMI)NI OFFER(T) COR(PUS). See Tatum, “Paliotto,” 30. 71 Elsewhere the altar simply depicts generalized footwear, but some manuscripts show slipperlike shoes similar to those on the footstools. See, for example, the portrait of Ambrose in the Egino Codex cited above, note 21. See also J. Braun, Die liturgische Gewandung im Occident und Orient (Darmstadt, 1964), 393–98. Although Braun discusses shoes that show elaborate straps, more like sandals than the slippers of the Ambrose altar, his evidence shows that the episcopal shoe type was considered a privilege and was at times granted to abbots (p. 398). 72 A similar sort of theology of office that separates the person from the office is explicated by Ernst Kantorowicz. In particular, the feet of the king are said to be on earth while the head is in heaven; see idem, The King’s Two Bodies: A Study in Medieval Political Theology (Princeton, N.J., 1957), 71. 66 67
178
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
In Ambrose’s case, the saint receives the vision as a sort of dialogue with Christ: his eyes are open, and he gestures with his hands. This is emphatically unlike the typical saint’s vision. For example, in the vita of St. Adalbert, represented on the doors at Gniezno in Poland, the saint dreams; his eyes are closed, and he is passive.73 In another variant, in Honoratus’s case on the altar, the sleeping bishop looks stunned, almost flattened by an angelic revelation. Furthermore, in other instances on the altar, the artist represents the conveyance of divine messages yet in two more ways—once picturing an angel whispering in the saint’s ear, and once diverging from textual sources to use the symbolic image of the hand of God. All of these various possibilities for the representation of the delivery and reception of divine messages are familiar but nonetheless imbued with meaning. An angelic messenger and a dream vision are quite comparable, although dreams may give more leeway for the interpretation of the message.74 Alternately, the hand of God represents something more immediately divine than do either of the first two possibilities, at the same time being quite abstract; it may more often represent hearing God rather than seeing him. In contrast to such symbolic visions, dreams, and disembodied or angelic voices, Ambrose’s vision of Christ represents direct contact with the divine. Christ is depicted as physically present, actually standing at the side of the bed. Paulinus even notes that he was “smiling.” 75 Venetian hagiography concerning Mark’s comparable vision of Christ pointedly presents such a bodily vision and dialogue as equivalent to the apostolic experience of Christ in the flesh.76 But, in the end, Ambrose’s open-eyed revelation is best understood as a visio Dei, usually reserved until after death except in the case of the holiest of saints.77 Finally, in the last scene (Fig. 6, upper right), Ambrose’s soul is carried to heaven by an angel, “anima in celum ducitur,” a detail which Elbern compares to the Ambrosian liturgy for the dead.78 The scene leaves no doubt that the bishop now resides in heaven— he is welcomed by the hand of God. The titulus concludes, in the last words of the altar narrative, that the body remained behind, “corpore in lecto posito.” 79 Once again, as in the scene of his infancy, Ambrose’s bed is, in effect, designated as an altarlike site of offering, this time draped with lavish fabrics, and the celebrating bishop Honoratus is dressed in the pallium. Abou-el-Haj, Medieval Cult, 30, 43, 46–48, fig. 52. S. F. Kruger, Dreaming in the Middle Ages (Cambridge, 1992). 75 Lacy, “Life of St. Ambrose,” 47, p. 61. For the Honoratus episode, see ibid., 47, p. 62. 76 T. A. Dale, “Inventing a Sacred Past: Pictorial Narratives of St. Mark the Evangelist in Aquileia and Venice, ca. 1000–1300,” DOP 48 (1994): 65–66. 77 Paschasius Radbertus wrote, “If souls still clinging to bodily chains, as if bound tight in a prison of limbs, can in the still of the night perceive higher things and things separated from their appearances, how much more do they see who are free from all earthly corruptions, for as one of the saints said, they see with a sense both pure and ethereal.” Epitaphium Arsenii 1.5, ed. E. Du ¨ mmler, Philologische und historische Abhandlungen der ko¨niglichen Akademie der Wissenschaften zu Berlin (1900), 2:27; cited in P. E. Dutton, The Politics of Dreaming in the Carolingian Empire (Lincoln, Nebr., 1994), 43. Also see C. Hahn, “Visio Dei: Changes in Medieval Visuality,” to appear in Visuality before and beyond the Renaissance, ed. R. Nelson (Cambridge, in press). 78 Elbern, Goldaltar, 53. 79 UBI ANIMA IN CELUM DUCITUR CORPORE IN LECTO POSITO, translated as “In which his soul is borne up to heaven, his body remaining on the bier”; Tatum, “Paliotto,” 30. See also Lacy, “Life of St. Ambrose,” 47, p. 62. 73 74
CYNTHIA HAHN
179
Remarkably, a relic bed of St. Ambrose of similar shape still exists in Milan, although it was not deposited at Sant’Ambrogio. Although it was not documented before 1598, its construction and design have been suggested to be consistent with fourth- or perhaps ninth-century manufacture.80 Such a relic bed would have gained prestige through its use as a means of display for the body, and it may be specifically referred to in the titulus. Although Bishop Honoratus seems to be witnessing the miracle of the ascension of Ambrose’s soul, in the previous scene he was commanded by the angel to offer the sacrament, the corpus Domini, as viaticum. The recurrence of the word corpus in the two tituli emphasizes the parallel between the saint’s body and that of Christ. It also implicates the sacrifice of Ambrose’s body and the bishop’s continuing earthly service as the very foundation of the altar. To summarize, a number of striking aspects of the construction and presentation of the Ambrose narrative on the Milan altar have been noted. First, the hagiographic narrative faces the celebrant and the choir, while the more public side is occupied with scenes from the life of Christ. Secondly, an altar, or altarlike element, is included in at least four of the scenes, as well as in the dedication scene in the center (Angilbertus II with the altar, Fig. 16), and four other scenes include additional liturgical elements or actions. Only the two linked horseback scenes of Ambrose’s calling and the two linked vision scenes (to which I return below) can be excluded from the specific realm of liturgical celebrations, actual or implied. Clearly, the bishop’s life is depicted on the altar as fully and intrinsically fashioned by the requirements of the liturgy, in this case specifically the Ambrosian liturgy. Finally, and perhaps most intriguing, a series of three pairs of scenes have been identified, each of which connects to its pendant across the central area of the altar. The three pairs are the only such linked scenes in the narrative (excepting perhaps the link of the angelic summons of Honoratus with the funeral scene), and yet their connection is left to be established by the viewer across a chasm that is equal to a third of the altar’s width. The viewer nevertheless makes the association, because in each case the connection is not a simple link, but is doubly forged. The link is firmly established both by means of the artistic devices of repetition, reversal, and composition, and by the narrative device of significant causal connection. Furthermore, rather than logical or historical cause and effect, it is divine causality—either a miracle or divine summons—that makes the link between each pair. Thus, the disruption of the narrative continuity that might seem intrusive or even aggressive performs a positive function instead, decisively locating divine power and establishing a sacred center for the narrative and the altar. THE DOORS AS THE “GATES OF JUSTICE” Remarkably, the central third of the altar’s back has no literal center (Fig. 16). In contrast to the front of the altar, which finds its focus in Christ and the cross (Fig. 1), the middle of the back is marked by a void, or, rather, the thin line created by the meeting The relic was located in the Capella di San Vittore in Ciel d’Oro; F. Reggiori, La “lettiera di Sant’Ambrogio,” in A. de Capitani d’Arzago, Antichi tessuti della Basilica Ambrosiana, Biblioteca de l’arte 1 (Milan, 1941), 107–16. Also see E. Cattaneo, “La devozione a Sant’Ambrogio,” Archivio ambrosiano 27 (1974): 85–110. Bandera (“L’altare,” 78) also notes the existence of the bed. 80
180
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
of the two doors of a very large and elaborate fenestella. Rather than being an image or a place, the sacred center thus consists of space and movement—space for the devotion to the saint, and movement of the soul toward God through the intervention of the saint. Both of these are clearly articulated in the visual construction of the doors. The space beckons the pious body. The functioning doors of the fenestella open to an area large enough for the insertion of a good part of a supplicant’s body into the saint’s confessio, as was the case at famous tombs, notably that of St. Peter.81 Within the altar the devotee would have encountered the saint’s porphyry sarcophagus, a reused imperial object positioned over the graves of the martyrs.82 The movement that represents the spiritual salvation of the soul through Ambrose’s patronage is the culmination of the narrative and the subject of the imagery on the doors. In the past, attention has been focused on the relatively static images of the portrait of the patron Angilbertus II (824–859), the bishop of Milan, and the self-portrait of Wolvinus, magist(er) phaber, the artist. Angilbertus wears a square halo, indicating a living subject of the portrait.83 Each figure is also crowned by Ambrose and, along with the saint, is encircled by a jeweled wreath. Although these two tondi are the major foci for the imagery (Fig. 16), circling the artist and the bishop are a “host” of angels. Eight are represented in enamel button-like clipea; and in their own tondi above, two large figures of the archangels Michael and Gabriel point with staves to the line created by the juncture of the doors and seemingly gesture with welcome, one to the left and the other to the right.84 Twelve gems are inset into the enamel band that frames the doors, and a large emerald is placed just to the right of the center. Surely, these elaborate doors are meant to represent the “Gates of Justice” (Ps. 118:19) and the entry to Paradise, opened to Christians because of the sacrifice of the Mass and accessible especially through the intercession of the saints. At Santa Prassede in Rome, on the triumphal arch of the 820s, a similar iconography of angels who lead souls to welcoming saints in the Heavenly Jerusalem is based on the funerary antiphon In paradisum.85 Elbern argued that Angilbertus would have seen Santa Prassede when he had attended the coronation of Louis II in Rome in 844 and that the apse mosaics of Elbern (Goldaltar, 20–21) draws a parallel with Gregory of Tours’ description of Peter’s tomb with its fenestella, through which Gregory inserted his head. The 8th-century version of Peter’s tomb also had doors; ¨seneder, “Fenestella,” Reallexikon zur deutschen Kunstgeschichte 7 (1981): 1227–53, see Y. Christes and K. Mo esp. 1228; also see references to the Liber Pontificalis in note 4. 82 I describe the 9th-century configuration. In the present configuration of the altar, one would encounter a space and a floor slab with holes through which one has limited access to the sarcophagus; see G. Righetto, “Scavi ottocenteschi in S. Ambrogio: La basilica ambrosiana in eta` paleocristiana e alto medioevale nella ‘Cronaca dei restauri’ di Mons. Rossi,” in S. Ambrogio, ed. Gatti Perer, 145 and fig. 18; Picard, Souvenir, 626. 83 Tatum (“Paliotto,” 27–28) argues that the recipient of such a halo could be dead, but he is countered by Elbern, Goldaltar, 57. For a more recent discussion of such haloes, see H. Belting, “Eine Privatkapelle im fru ¨ hmittelalterlichen Rom,” DOP 41 (1987): 55–69. 84 Elbern (Goldaltar, 56) calls them “welcoming.” I would say they are also indicating the holy nature of the interior of the confessio. 85 The antiphon is a response to Psalm 118; see M. Mauck, “The Mosaic of the Triumphal Arch of S. Prassede: A Liturgical Interpretation,” Speculum 62 (1987): 814, 824. For the Ambrosian funeral rite, see Cattaneo, “Storia e particolarita`,” 837. The altar as a whole has been referred to as the Celestial Jerusalem by G. Ravasi, “Iconografia Biblica dell’altare d’oro,” in L’altare d’oro, ed. Capponi (as in note 4 above), 43. The inscription on the altar has been compared to the apse inscription at Santa Prassede and that at S. Cecilia; see M. Ferrari, “Le iscrizioni,” in ibid., 150. 81
CYNTHIA HAHN
181
Sant’Ambrogio had been inspired by the apse of that church.86 This particular argument is problematic considering the disagreements over the dating of the mosaics; yet the possible relationship between the iconography of the altar and the triumphal arch of Santa Prassede might be much more specific. It is not, however, a relationship of copying: the saints that welcome souls at Santa Prassede are martyrs, not a bishop as on the doors of the altar. In fact, rather than offering a close parallel to the altar’s doors, the funerary and intercessory iconography of Santa Prassede corresponds more precisely to the language of the funeral antiphons as well as to the imagery of the two, nearly identical, ends of the Milanese altar. It appears that intercessory iconography on the ends of the altar is integrated with an abbreviated imagery of the Milanese episcopal list.87 On each end, two bishops and two martyrs are represented in repousse´ roundels. Ambrose is paired with Simplicianus, his sainted successor, and with Gervasius and Protasius, the two martyrs whom he buried under the altar (Fig. 17). On the opposite end are Bishop Martin, once more (above), and Maternus, who, according to ninth-century belief, was the first bishop of Milan.88 The two martyrs represented here, Nabor and Nazarius, were also “invented” by Ambrose. In fact, the three Milanese bishops who are singled out for representation on the altar ends appear in Carolingian Milanese missals among the Communicantes, the saintly community that is remembered during the intercessory prayers at the outset of the Mass,89 while Martin, as above, may represent confraternal intercessory prayer. Each end of the altar also includes eight angels and four unexplained vested or togate figures praying toward central images of a jeweled cross. Elbern notes that the Ambrosian liturgical preface specifies eight angel choirs, as did certain Carolingian litanies.90 It has also been suggested that the rotulae that the majority of the angels carry represent the “book of life” of Revelation 20:12.91 I would suggest that the standing figures are meant to be saints as citizens of heaven praying in intercession as part of the funeral liturgy, as at Santa Prassede. Thus, the ends of the altar correspond more precisely to the liturgy of intercession than do the doors. The imagery on the doors must then be seen as a variation on the established intercessory iconography and a refocusing on the possibilities of episcopal intercession. Martyrs are replaced by the bishop Ambrose, and, in a further transgression of norms, Ambrose himself crowns Angilbertus and Wolvinus as he receives their efforts on his Elbern, “Mailand-Spa¨tantike,” 19. Episcopal lists often represent a first step in claiming privileges such as patriarchal or metropolitan status for an archiepiscopacy; see Picard, Souvenir, 689–700. Some of these lists are visual (505–520), and a number of the visual lists take the form of altar cloths (515–518). 88 At Tours, many of the bishops were buried in St. Martin’s, which was not the cathedral, just as the bishops’ burial place in Milan; see S. Farmer, Communities of Saint Martin (Ithaca, N.Y., 1991), 22. One other similarity has been noted between Martin and Ambrose, namely, the recurrence of the topos of the bishop sent to the sick bed of another bishop. In Martin’s vita, Martin was sent to the bed of Bishop Liborius; Elbern, Goldaltar, 52. 89 Picard, Souvenir, 624; J. A. Jungmann, Mass of the Roman Rite: Its Origins and Development, trans. F. A. Brunner (1949; repr. Westminster, Md., 1986), 2:170–79. 90 Elbern, Goldaltar, 61. 91 Gianfranco Ravasi (“Iconografia,” 54) identifies the vested figures as deacons, and possibly the four deacons of Ambrose. 86 87
182
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
behalf. Byzantinists are familiar with such crownings occurring in imperial images of a later date,92 and saints are often depicted as similarly crowned by Christ when they enter heaven;93 but this crowning of a living archbishop and his artisan by a saint instead of Christ is unprecedented and quite startling. It can only be understood in terms of the expanded use to which the altar was put and the tremendous prestige of Ambrose’s cult. The cult seems to center on these doors. One must ask, however, how often the doors were actually opened and for whom. The fact that their opening was considered “a great event in the life of the basilica” 94 indicates that they were opened only rarely. Appropriate occasions seem to have included important feast days and the first mass of a new sacerdos, or guardian, of the church. As a result of his special privilege, the new guardian must have been able to experience the mass he celebrated with special poignancy, “sacrificing himself on the altar” in memory of Ambrose and the martyrs, just as Ambrose had done himself, according to the Carolingian vita.95 In turn, he became the intimate recipient of the grace of salvation that the altar promised. In contrast, other aspects of the doors’ use seem more ceremonial, and the ceremonies would have had one particular audience—the clergy. The very act of the opening of the doors seems to have been celebrated as solemn liturgy by the cimiliarchia, the guardian who had been entrusted with the keys in a special ceremony called tollite claves.96 Once the doors were open, because of their position and the arrangement of the altar in the choir space, the best visual access to the interior of the confessio would have been enjoyed by the archbishop. His ninth-century stone sedia is positioned directly behind the altar, in line with the doors.97 Whether or not the altar’s doors were actually opened to supplicants, they were represented as the very nexus of a system of prayer through and with the saints, which was sponsored by the clergy. The inscription on the back face of the shrine begins, “The beneficent ark [reliquary] shines forth, lovely with its glittering panoply of metal and dressed with gems. But more potent than all its gold is the treasure with which it is endowed by virtue of the holy bones within it . . .” The inscription then names the patron Angilbertus and asks for the divine blessing upon the bishop who gave it while serving “this brilliant see.” 98 Ambrose is invoked as the treasure and center of this holy archbish92 I. Kalavrezou-Maxeiner, “Eudokia Makrembolitissa and the Romanus Ivory,” DOP 31 (1977): 305–28; there the emperor and the empress are standing on footstools in the same sort of undefined space. On the Ambrose altar, only Ambrose (and the angels above him) stands on a footstool. Elsewhere the footstool may be liturgical (in scenes 4, 7, and possibly 10 and 11), but in the center it seems quite specifically to indicate a heavenly location. 93 Hahn, Passio Kiliani, 100–103. Elbern (Goldaltar, 57) also notes the unusual nature of crowning. 94 E. Mazza, “La liturgia nella basilica di S. Ambrogio in epoca medioevale,” in S. Ambrogio, 304. 95 Paredi, Vita, 86, p. 140. This topos also occurs in Bede’s Life of Cuthbert, in Two Lives of Saint Cuthbert: A Life by an Anonymous Monk of Lindisfarne and Bede’s Prose Life, ed. B. Colgrave (Cambridge, 1940), xvi, 213, and is similar to the phrasing in Augustine, Civitas Dei 20.19. For the sacerdotes, see Mazza, “Liturgia,” 304. 96 On the other hand, the abbots of the monastery had to fight for their right to have the doors opened; see Mazza, “Liturgia,” 304, 307. Surely the “key liturgy” was reminiscent of the handing of the keys to Peter— represented on the 10th-century ciborium above the altar—and also argued for the likeness of the doors to the Gates of Justice. 97 S. Ambrogio, ed. Gatti Perer, 67, fig. 76. The inscription reads, PRESUL MAGNIFICUS RESIDENS IN SEDE DECORUS/⫹SITU ROMANA VERO QUAE SEDE SECUNDA. 98 Æmicat alma foris rutiloque decore venust(a) Arca metallorum gemmis quae compta corusca(t) Thesauro tamen haec cuncto potiore metall(o)
CYNTHIA HAHN
183
opric; and the imagery on the doors makes clear that he is able to guarantee his intercessory power. A document purported to have been written by Angilbertus gave the monks control of the altar and later became the center of a controversy between the monks and the guardians/canons of the cathedral.99 Although the controversy extended from the eleventh to the eighteenth century, its roots were founded in jealousies between the monks and the guardians of the basilica that had existed from the time of the altar’s creation.100 The altar had become the focus of a sacred necropolis: the majority of the bishops of Milan were buried in close proximity to it, rather than in the cathedral.101 The controversy is testimony to the power of the altar and its function as the very center of Milanese intercessory and commemorative prayer. As images of the liturgical functioning of saintly intercession, the back and the ends of the altar hold out a promise to any Christian that prayer to the saints, and specifically prayer to Ambrose, will advance them toward salvation. Nevertheless, as noted above, one portion of the audience is especially privileged and would presumably move most swiftly along the path toward salvation (or work as mediators for others). The imagery confirms this privileged position of the clergy and especially the bishop. Not only is Ambrose defined as a bishop by his participation in the liturgy, but he is also carefully placed within a succession and a community of bishops—being baptized by an Orthodox bishop and ordained by two bishops, serving to bury Bishop Martin, naming his successor on his deathbed (according to one vita),102 and being buried in turn by Honoratus of Vercelli, one of Milan’s suffragan bishops.103 Finally, the saintly bishop’s promotion to the place of the welcoming martyrs in the structure of intercessory iconography holds promise of remarkable powers residing in his hands. THE ALTAR’S POLITICAL AND SOCIAL SETTING It is not surprising that such an altar should have been commissioned by Archbishop Angilbertus II. The episcopacy of Milan enjoyed unprecedented ecclesiastical and secular Ossibus interius pollet donata sacrati(s). (Æ)gregius quod praesul opus sub honore beat(i) Inclitus Ambrosii templo recubantis in isto Optulit Angilbertus ovans, Dominoque dicavi(t) Tempore quo nitidae servabat culmina sedis. (A)spice, summe pater, famulo miserere benign(o), (T)e miserante Deus donum sublime reporte(t). Letters in parentheses occur at corners and are used in two words, one written horizontally and one vertically. See Tatum, “Paliotto,” 26. Tatum translates the rest as follows: “This work the noble bishop, famed Angilbert, offered with joy in honor of the blessed Ambrose who lies in this temple and dedicated to the Lord in the time when he held the chief place of this brilliant see. Father on high [i.e., Ambrose], look upon and pity thy loving servant, by thy intercession may God bestow his divine blessing.” 99 The document’s authenticity has been challenged. See Tatum, “Paliotto,” 26–27; Abou-el-Haj, Medieval Cult, 11–12. 100 A. Ambrosioni, “Monaci e canonici all’ombra delle due torri,” in S. Ambrogio, ed. Gatti Perer, 244–46. The guardians were reformed as canons in the 11th century. 101 Picard, Souvenir, 358, 624. See also A. Rovetta, “Memorie e monumenti funerari in S. Ambrogio tra Medioevo e Rinascimento,” in S. Ambrogio, ed. Gatti Perer, 269–93. 102 Paulinus does not mention this episode, but the Carolingian vita does; see Paredi, Vita, 82, p. 132. 103 The bishop of Vercelli was a suffragan of Milan at least by the time of the altar’s manufacture; see Picard, Souvenir, 627 n. 165.
184
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
power under the Carolingians. Angilbertus II is known for his close “adhesion to the program of Carolingian reforms” and his acceptance of the “universalist ideal of empire.” 104 Nevertheless, he was a strong leader for Milan as an important city in its own right. He reformed the Ambrosian liturgy of the cathedral,105 was intimately involved in judicial matters, and apparently used his moral authority to eradicate superstition and “evil.” He founded hospitals and monasteries, brought in trans-Alpine monks, and supported the poor.106 He also seems to have initiated a rebuilding of the apse of Sant’Ambrogio, constructing a typically Carolingian crypt.107 I have already noted that the altar is literally surrounded by the bodies of the former bishops of Milan (this pattern of episcopal burial near a saintly bishop, albeit not in the local cathedral, was increasingly popular at this time; cf. Tours, Auxerre, Ravenna/Classe, and, of course, Rome).108 However, it is also significant that, in the apse above the altar, the suffragan bishops of Milan were painted above their episcopal seats (Fig. 2).109 The seated bishops visually commemorated Milan’s episcopal prestige.110 During the episcopacy of Angilbertus II, the number of Milan’s suffragans had risen from four to seven, and then to an all-time high of eleven, later possibly reduced to nine. Moreover, the depiction of this chorus of suffragans in Sant’Ambrogio rather than in the cathedral once more confirms Ambrose’s premier importance for Milan. He is the center, perhaps even the source, of an episcopal network that extends in both space and time, including his suffragans as well as his successors and his brothers throughout the empire. As his ninthcentury hagiographer wrote, “All that this Milanese church owns in merit and grace, it owes to the magisterium of Ambrose.” 111 At that time he was even said to be the author of the Ambrosian liturgy, an assertion contrary to fact, but which thereby credited him with one of the great glories of Milan.112 Indeed, the altar must have even been intended in part as confirmation and visualization of the prestige of the Milanese liturgy. Although hagiographic texts and the imagery of the altar allow no doubt about Ambrose’s status—and through him the status and Ambrosioni, “Arcivescovi,” 99. Ambrosioni, “L’altare,” 58; also Bandera, “L’altare,” 78. 106 Ambrosioni, “Arcivescovi,” 102–6; Bandera, “L’altare,” 106. 107 Although the body remained directly under the altar rather than being buried in the crypt below; see Ambrosioni, “L’altare,” 58. Also see Hahn, “Seeing and Believing,” 1101–6. 108 See notes 88 and 107. 109 There were eighteen bishops with inscriptions, which were in poor shape already in the 19th century. ´ tude sur l’architecture lombarde et sur F. de Dartein cannot ascertain a date but gives inscriptions and labels, in E les origines de l’architecture romano-byzantine (Paris, 1865–82), 86–88. Bandera (“L’altare,” 75–78) believes that the presence of the bishop of Coira among the suffragans helps to date the rebuilding of the apse and thus the construction of the altar, since the bishop of Coira ceased to be a suffragan of Milan between 842 and 847. She also compares this iconography to that of Saint-Germain at Auxerre, which is comparable in other ways as well (cf. above, p. 175 and note 56). 110 These represent the bishops who accompanied the archbishop (or the metropolitan in 850) to various councils; see Ambrosioni, “Arcivescovi,” 104. There seem to have been eighteen suffragans represented in the apse, each with a synodal decree beneath; see Bandera, “L’altare,” 103, citing C. Ferrarrio, Monumenti sacri e profani della basilica di Sant’Ambrogio (Milan, 1824), 159. 111 Courcelle, Recherches, 121; Picard (Souvenir, 630) notes that this quote derives from a sermon on Eusebius of Vercelli and marks a beginning of Ambrose as city patron. 112 The 8th-century Versum de Mediolano civitate (a text important for its celebration of Milan’s episcopacy) cites the liturgy as one of the glories of Milan; see Picard, Souvenir, 628 n. 167. Picard (ibid., 697) shows that other cities developed similar cults around their founding bishops and martyrs; nonetheless, Milan is confident in its prestigious saint and bishop. 104 105
CYNTHIA HAHN
185
position of the Milanese episcopacy—apparently the Milanese liturgy was briefly challenged earlier under the Carolingians. A ninth-century miracle story recounted by Paul the Deacon and an eleventh-century one reported by Landulfus Senior both express a need to defend the validity of the Ambrosian rite; the latter even explicitly defends it against the liturgical standardizations of Charlemagne.113 The matter seems, however, to have been resolved by the time Walafrid Strabo wrote ca. 840 (that is, approximately contemporary with the altar), “Ambrose, bishop of Milan, determined the organization of the mass as well as the other offices for his church and the other churches of Liguria, an organization that is still respected today in the Church of Milan.” 114 Moreover, the altar does not only claim equality with other altars and their liturgies— it claims superiority. It does so, as noted in its inscription, through its very materials and form; the altar is made entirely of precious metals, enamels, and gems. Although many tombs and altars were ornamented with such riches, few were entirely gold and/or silver. Those that were singled out in medieval sources as being so constructed stood among the most famous shrines and altars of Christendom. Furthermore, for the most part they were located in three major centers renowned for liturgy, namely, Rome, Constantinople, and Jerusalem: in Rome, examples are recorded by the Liber Pontificalis, including particularly the altar in St. Peter’s; in Constantinople, both the Constantinian and Justinianic Hagia Sophia had lavish golden altars; and in Jerusalem, more than one golden altar was recorded in the Holy Sepulchre.115 Indeed, gold even has celestial resonance, for in Revelation 8:3 an angel offers “the prayers of all saints upon the golden altar.” Just as the precious materials resonated with apostolic, liturgical, and heavenly significance, so large fenestellae were characteristic of apostolic altars, particularly those in Rome.116 Other references to Rome might also be found in the burial patterns of the bishops, in the inscription on the contemporary episcopal throne in Sant’Ambrogio, which claims a comparison to Rome, and, as argued by Carlo Bertelli, in the later stucco ornament of the ciborium, which depicts Ambrose and his clergy as comparable to Peter and Paul.117 Finally, as we have seen, there are apostolic parallels between the imagery of the altar and the texts of Ambrose’s vitae. On the altar, Ambrose is likened to Paul in the scene of his calling, preaches with apostolic simplicity, and witnesses Christ in apostolic fashion, “in his body.” The Carolingian vita in particular is filled with comparisons of Ambrose to the apostles, especially Peter.118 Elsewhere in Italy, for example, in Venice, Aquileia, and Picard, Souvenir, 628–29. Ibid., 629: Libellus de exordiis et incrementis quarumdam observationibus ecclesiasticis rerum, 23; MGH Capit, 2:497. 115 See note 4. Braun also discusses the confessio ornamentation; see Altar, 1:563 ff. 116 See note 81. Throughout the Liber Pontificalis, it is noted that popes renewed the gold or silver on confessio doors of various saints. Also see Elbern, Goldaltar, 20; and idem, “Karolingische Goldschmiedekunst,” 302. 117 See note 97 for the inscription on the episcopal throne. Bertelli argues that the stuccoes make the claim that Milan is equal to Rome and that Ambrose was chosen like Peter and Paul; see his “Il ciborio restaurato,” in Il ciborio della basilica di Sant’Ambrogio in Milano (Milan, 1981), 37, 59. Ambrose and his clergy face the clerical side or choir, while Peter and Paul face the nave. I have not been able to consult A. Columbo, “Milano ‘secunda Roma’ e la lapide encomiastica dell’antica Porta Romana,” AStLomb 83 (1956): 149–52. 118 Paredi, Vita, 16, p. 38: Ambrose built the church like the apostle on the rock of Christ; ibid., 43, p. 76: Ambrose wrote in his own hand at night like Paul; ibid., 76, p. 124: like the apostle, Ambrose wanted to be liberated from his body; ibid., 86, p. 138: Ambrose had apostolic virtues. 113 114
186
NARRATIVE ON THE GOLDEN ALTAR OF SANT’AMBROGIO
Ravenna, apostolic prestige was sought for archiepiscopal sees, and even claims to the dignity of the patriarchate were made through the use of texts and images.119 The altar in Milan does not so much claim apostolicity for the saint as it asserts Ambrose’s status as fully equal to the greatest of saints. Thus, in sum, the ornament and the setting of the altar in Sant’Ambrogio construct a sanctity of enormous power and prestige: Ambrose is depicted as a leader of bishops among bishops, the equal to any saint, and the very font and origin of sanctity in Milan. It is therefore not surprising that the altar narrative foregoes a historical and anecdotal Ambrose in favor of the depiction of a patron fully able to guarantee access to salvation— to crown his supplicants from the city of Milan as citizens of heaven. OTHER STORIES At the outset of this essay, I argued that a narrative could tell many stories. The historical context and the audience of Ambrose’s life seem to determine the altar’s story as a single story written for a specific clerical audience. Clerics certainly were the primary patrons and audience of the altar; however, narratives always allow other stories, and hagiographies, being the stories of individuals, always allow for a more private and intimate perception. In my opinion, at least one pathway of the narrative, unrelated to the highroad posted with the signs of episcopal and ecclesiastical power, is accessible and should be explored here. It leads more directly to concerns about the reliquary status of the altar as a tomb, which is a question of great interest to me. Perhaps it begins to trace the way in which reliquaries may operate in providing access to the divine and succoring the individual. This pathway is a path of the senses. In the lowest register, the first miracle emphasized the mouth of the baby Ambrose. Although I concentrated earlier on the meaning of the miracle for the church, one could also read it as a miracle of taste—the baby Ambrose was given his first intimation of the Lord through a sweetness in his mouth. Likewise, in the scenes of baptism and ordination, Ambrose experienced contact with the divine through liturgical touch. In the last scene of the second register, Ambrose heard the words of the Lord as an angel whispered in his ear. Finally, in the center of the uppermost register, Ambrose is granted a vision of God before death. The selection of miracles and their depiction describe an unmistakable hierarchy of the senses in approaching the knowledge of God.120 The focus on the body of the saint is central to hagiography and to the meaning of tombs. As Peter the Venerable taught in a sermon concerning relics, “And he teaches us to know from his body, and he shows by his own body what you ought to hope for yours.” 121 The narrative recommends an ascent toward Dale, “Inventing a Sacred Past,” esp. 103. Sight is always at the apex of a system of the senses. Augustine used the image of the Five Wise Virgins as a model for the proper use and refinement of the senses. See C. Hahn, “Icon and Narrative in the Berlin Life of St. Lucy (Kupferstichkabinett MS78A4),” in The Sacred Image East and West, ed. R. Ousterhout and L. Brubaker (Urbana, Ill., 1995), 72–90. The five senses were explicitly represented in art as early as the 9th century; see C. Nordenfalk, “Les cinq sens dans l’art du moyen ˆage,” Revue de l’art 34 (1976): 17–28. 121 “Sermo cuius supra in honore sancti illius cuius reliquiae sunt in presenti,” in “Petri Venerabilis sermones tres,” ed. G. Constable, Revue be´ne´dictine 64 (1954): 271; cited and trans. in C. Solt, “Romanesque French Reliquaries,” Studies in Medieval and Renaissance History 9 (1987): 181. 119 120
CYNTHIA HAHN
187
God through the senses, and it feeds directly into the ascent promised through the saint’s advocacy as discussed above. Perhaps the viewer can hope, like Ambrose, to know God. It may be that one is invited by the fenestella to approach and come into contact with the holy body through touch, but above I expressed scepticism about how often and for whom the doors would have been opened. Rather than touch, taste, or even hear the sacred, the supplicant is invited by the imagery of the altar to seek the sight of the holy as the highest experience of the divine. The altar does so through its narrative, and also, of course, through its very existence, “lovely with its glittering panoply of metal and . . . gems.” 122 However, even were this message legible, it must have been rather faint for the laity. Just as today, so also in the medieval period the altar was difficult to approach—witness the miracle of the foot-trampling in the upper left corner. Whose altar is this? Maybe it is that of the monks, maybe that of the guardians, but it is certainly not a place for the average Christian. Even the nobility were to keep their distance from this holy altar; perhaps, as in the Carolingian vita, the emperor himself was enjoined not to approach. Ultimately, the altar of Ambrose wields powerful weapons of holy rhetoric—it calls upon the life of Christ, the power of the liturgy, the significance of its materials, the persuasions of space, the representation of prayer, the power of words, and even the witness of the senses. Its discourse is almost overwhelming in its complexity and power, a battering ram of holiness, if you will. Clearly, the original intention was to reserve the intercessory power of the saint’s body and of the narrative for the service of the church, specifically the church as an institution located in Milan. Nevertheless, narratives resist control, and some of the power of Ambrose’s story spills out and is able even to touch the viewer of today. Florida State University 122
The quote comes from the altar’s inscription (see note 98).
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
1 Christological narrative, view of the altar from the nave (photo: courtesy of the Conway Library, Courtauld Institute of Art, University of London)
2 View of the choir (after F. de Dartein, Étude sur l’architecture lombarde et sur les origines de l’architecture romano-byzantine [Paris, 1865–82], pl. 28)
3 Floor plan (based on de Dartein, Étude)
4
Ambrose narrative, view of the altar from the choir (photo: courtesy of the Conway Library, Courtauld Institute of Art, University of London)
5
Back of the altar, left portion (photo: Bildarchiv Foto Marburg)
6
Back of the altar, right portion (photo: Bildarchiv Foto Marburg)
7
8
Ambrose arriving in Milan (photo: Bildarchiv Foto Marburg)
Nativity, Lectionary of Henry II, Staatsbibl. Clm. 4452, fol. 9r, Munich (photo: Bayerische Staatsbibliothek)
9 Ambrose leaving Milan (photo: Bildarchiv Foto Marburg)
10
Ambrose baptized into the Orthodox faith (photo: Bildarchiv Foto Marburg)
11
Ambrose celebrates the funeral mass of St. Martin (photo: Bildarchiv Foto Marburg)
12
Ambrose preaches and converts an Arian (photo: Bildarchiv Foto Marburg)
13
Ambrose steps on Nicetius’s foot and heals it miraculously (photo: Bildarchiv Foto Marburg)
14
Ambrose’s vision of Christ (photo: Bildarchiv Foto Marburg)
15
An angel speaks to Bishop Honoratus (photo: Bildarchiv Foto Marburg)
16
Back of the altar, central portion, fenestella (photo: Bildarchiv Foto Marburg)
17
Altar end (photo: courtesy of the Conway Library, Courtauld Institute of Art, University of London)
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
The Profane Aesthetic in Byzantine Art and Literature HENRY MAGUIRE
n the cloisters, under the eyes of the brethren engaged in reading, what business has there that ridiculous monstrosity, that amazing mis-shapen shapeliness and shapely mis-shapenness? . . . Those fierce lions? Those monstrous centaurs? Those semihuman beings? . . . Here you behold several bodies beneath one head; there again several heads upon one body. Here you see a quadruped with the tail of a serpent, there a fish with the head of a quadruped. . . . In fine, on all sides there appears so rich and so amazing a variety of forms that it is more delightful to read the marbles than the manuscripts.” With these famous words the Cistercian leader, St. Bernard of Clairvaux, condemned the imaginative variety of profane art that was to be found in the twelfth-century Cluniac monasteries (Figs. 1, 2).1 St. Bernard clearly considered this art to be worthless, and yet, equally clearly, he was attracted by its allure. As Erwin Panofsky wrote, “A modern art historian would thank god on his knees for the ability to write so minute, so graphic, and so truly evocatory a description of a decorative ensemble in the ‘Cluniac manner.’” 2 Partly as a result of St. Bernard’s diatribe, Western medievalists have paid considerable attention to the type of art he describes; indeed, the monstrous forms of the capitals and the initials in the manuscripts have been, for many modern scholars, the defining features of the Romanesque style.3 But in Byzantium the case is different. There was no St. Bernard to focus our attention on the profane elements in Byzantine art, and so these features have stayed in the margins of our vision. The present-day view of Byzantine art is that it is either religious or political, or both. Only a few scholars have drawn attention to its more subversive elements.4
“I
I thank Eunice Dauterman Maguire for her many contributions to this article, and Alice-Mary Talbot for reading the manuscript in draft. I am also grateful to members of the Dumbarton Oaks Symposium, at which the paper was originally delivered, for their helpful observations, especially Panagiotis Agapitos, Marˇevcˇenko, Ruth garet Alexiou, Jeffrey Anderson, Kathleen Corrigan, Anne Derbes, Cynthia Hahn, Nancy S Webb, and the late Alexander Kazhdan. 1 Trans. E. Panofsky, Abbot Suger on the Abbey Church of St.-Denis and Its Art Treasures, 2d. ed. (Princeton, N.J., 1979), 25. 2 Ibid. 3 See, for example, M. Schapiro, Romanesque Art (New York, 1977), esp. 4–10. 4 Most notably, A. Cutler, “On Byzantine Boxes,” JWalt 42–43 (1984–85): 32–47. See also the new publication A. Iacobini and E. Zanini, eds., Arte profana e arte sacra a Bisanzio (Rome, 1995).
190
THE PROFANE AESTHETIC
The following article attempts to define the profane aesthetic in Byzantine art. Since the topic is a large one, there is space to consider only three of its most important aspects: first, innovation in art; second, unbridled movement; and, finally, nudity. Another restriction that I place upon myself is the concentration, by and large, on Byzantine art in the middle period, that is, from the ninth to the twelfth century (although it is necessary to range rather wider in the texts). I. INNOVATION The discussion thus begins with innovation. It was, of course, axiomatic in Byzantium that orthodox Christian artists did not invent. The very legitimacy of the holy image depended upon its adherence to tradition and its supposed accuracy in reproducing the prototype.5 It was their lack of invention that distinguished Christian images from the fanciful and arbitrary creations of the pagans. In 787, the council of Nicaea quoted from a seventh-century dialogue on this subject, written by John of Thessalonike. In this exchange, which takes place between a Christian and a pagan, the Christian says, “We . . . make images of men who have existed and have had bodies—the holy servants of God— so that we may remember them and reverence them, and we do nothing incongruous in depicting them such as they have been. We do not invent anything as you [pagans] do.” 6 A later passage, from the first Antirrhetic by the early-ninth-century iconodule Patriarch Nikephoros, specifies what these “incongruous” inventions of the pagans were: “The idol is a fiction of those things that do not exist and have no being in themselves. Of such a kind are the shapes that the pagans fatuously and irreligiously invent, such as of tritons, centaurs, and other phantoms that do not exist.” 7 Byzantine secular writings also characterized such composite creatures derived from pagan mythology as implausible or absurd. In the twelfth-century Byzantine satire, the Timarion, the protagonist says that the unlikely event of his release from hell is “as unrealistic as the things sculptors and painters create . . . , hippocentaurs, sphinxes, and all the other mythological fabrications of the ancients.” 8 Likewise, the tenth-century Life of Basil the First speaks scornfully of the centaur Chiron, saying that the future emperor was educated by his father for, unlike Achilles, he had no need of a semihuman tutor.9 In the eleventh century, Psellos, in an allegorical treatise on the sphinx, treats the creature as pure fantasy, although without condemning it. He says that the sphinx has the form of a 5 L. Brubaker, “Byzantine Art in the Ninth Century: Theory, Practice and Culture,” BMGS 13 (1989): 23–93, esp. 48–49. On the general problem of innovation in Byzantium, see A. R. Littlewood, ed., Originality in Byzantine Literature, Art and Music (Oxford, 1995). 6 Kai` oujde` n ajpeiko` " ejrgazo´ meqa gra´ fonte" toutou` ", oi»oi kai` gego´ nasin. Oujde` ga` r platto´ meqa kaq∆ uJma'" . . . Mansi 13:165; trans. C. Mango, The Art of the Byzantine Empire, 312–1453 (Toronto, 1986), 140. 7 To` de` ei“dwlon ajnupa´ rktwn tinw'n kai` ajnuposta´ twn ajna´ plasma, oJpoi´a" dh´ tina" ”Ellhne" uJp∆ ajsunesi´a" kai` ajqei?a", Tritw´ nwn tinw'n kai` Kentau´ rwn kai` a“llwn fasma´ twn oujc uJfestw´ twn, morfa` " ajnapla´ ttousi. Antirrheticus 1.29, PG 100:277B. See K. Parry, “Theodore Studites and the Patriarch Nicephoros on Image-Making as a Christian Imperative,” Byzantion 59 (1989): 164–83, esp. 180. 8 “Ta` me` n ou«n o”sa nu'n fh'",” h«n d∆ ejgw´ , “komyo´ tate, a“pista pro` tou' gene´ sqai o“nta, tera´ stia´ moi dokou'si kai` ajlhqw'" aijni´gmata, oJpoi'a liqoxo´ oi kai` zwgra´ foi ejn oijki´ai" pla´ ttousin, iJppokentau´ rou" dhladh` kai` sfi´gga" kai` ei“ ti a“llo muqw'de" toi'" palaioi'" ajnesth´ lwto.” Timarion 27, ed. R. Romano (Naples, 1974), 74.674–78; trans. B. Baldwin, Timarion (Detroit, Mich., 1984), 60. 9 Vita Basilii 6, PG 109:236A.
HENRY MAGUIRE
191
beautiful maiden as far as the navel, but its lower parts are covered with thick hair and have the feet of a wild beast and a long tail; nevertheless, its voice speaks atticizing Greek. Such is the monster of the myth, declares Psellos, and let license be given to the poets to create whatever they like; but as for him, he is concerned not with the appearance of the monster but with its symbolism. He goes on to explain that the sphinx represents man, who is composed of both rational and irrational natures.10 The invention of composite monsters by pagan artists was seen in more negative terms by the Byzantine saints’ Lives. The Life of St. Andrew the Fool has a story about the saint standing in front of the great bronze doors of the Senate House and looking at its reliefs, which portrayed the battle of the giants against the gods. These reliefs may be reflected in a miniature from a tenth-century copy of the Theriaka of Nikander, where one sees the legs of the giants represented in the usual classical manner, as writhing snakes (Fig. 3).11 As St. Andrew was gazing at the giants on the Senate House doors—the text calls them “thong-legs”—a passerby saw him and gave him a slap on the neck, saying, “You idiot, what are you staring at?” The saint answered back, “You fool in your spirit! I am looking at the visible idols, but you are a spiritual ‘thong-leg,’ and a serpent, and of the viper’s brood, for your soul’s axles and your heart’s spiritual legs are crooked and going to Hades.” 12 For St. Andrew, then, the snake-legged giants were not only idols but also symbols of evil. Another type of composite creature is recorded in the Lives and encomia of the stylite St. Alypios. Here we read of a monstrous stone statue of a tauroleon, a combination of a lion and a bull, which the saint found sitting on top of a column in a pagan cemetery that was deserted by all except demons. Showing considerable acrobatic prowess, St. Alypios scaled the column and prized this reasonless creature off its pedestal with a crowbar; he then replaced it with an image that was true, namely, an icon of Christ. He did this, we are told, so that the enemy army of demons might be laughed at and made fun of.13 The god Pan, half human and half goat, was singled out for special censure by Byzantine writers. Another encomium of St. Alypios, written by a certain Antony of St. Sophia, perhaps in the tenth century, describes how the noble man, acquiring the feet of a deer, mounted the column by leaps and bounds until finally he reached its top and threw down the complex and beautifully worked sculpture of a lion and an ox that had stood there. The saint, says the encomium, made himself into a living statue on top of the column. Now the column was no longer serving as a pedestal for a mute sculpture of a 10 J. M. Duffy and D. J. O’Meara, eds., Michaelis Pselli Philosophica minora (Stuttgart-Leipzig, 1992), 1:158.15–26. 11 Bibliothe`que Nationale, Paris suppl. gr. 247, fol. 47. See H. Omont, Miniatures des plus anciens manuscrits grecs de la Bibliothe`que Nationale du VIe au XIVe sie`cle (Paris, 1929), 40, pl. 68; Byzance: L’art byzantin dans les collections publiques franc¸aises, exhibition catalogue, Muse´e du Louvre (Paris, 1992), 349, no. 259. 12 “Sale´ , ti´ i”stasai ble´ pwn…” JO de` maka´ rio" e“fh pro` " aujto´ n⭈ ““Exhce tv' noi?, tw'n eijdw´ lwn toi'" aijsqhtoi'" ejntrani´zwn e”sthka, kai` ga` r kai` aujto` " nohto` " lwro´ pou" kaqe´ sthka" kai` o“fi" kai` ge´ nnhma ejcidnw'n⭈ oiJ ga` r a“xone" th'" yuch'" sou kai` ta` nohta` diabh´ mata th'" kardi´a" sou diestramme´ na eijsi` kai` ejpi` to` n a” dhn badi´zonta.” Ed. and trans. L. Ryde´n, The Life of St. Andrew the Fool (Uppsala, 1995), 2:140.1924–28. 13 H. Delehaye, Les saints stylites, SubsHag 14 (Brussels, 1923), 154.1–26 (vita prior); cf. ibid., 190.10–17 (encomium by St. Neophytos). On this episode, see H. Saradi-Mendelovici, “Christian Attitudes toward Pagan Monuments in Late Antiquity and Their Legacy in Later Byzantine Centuries,” DOP 44 (1990): 47–61, esp. 55.
192
THE PROFANE AESTHETIC
pagan god, such as Sarapis, Dionysos, or Apollo. Nor—far from it—did the column bear Pan, who, says the writer, was “the most ludicrously laughable of the lot—a mixture of different natures and faculties.” Rather, the column now carried the saint himself, a divinely shaped image, an icon of piety, and a statue with reason.14 Tritons, centaurs, hippocentaurs, sphinxes, snake-legged giants, bull-lions, and the goat-shanked Pan—all of these pagan inventions were given negative connotations by Byzantine writers.15 Were such monsters avoided, then, in Byzantine art? Not one bit of it; they were portrayed frequently and with relish by Byzantine artists—and not only in secular contexts, but on churches and in religious manuscripts as well. Even the goatfooted Pan, or his relatives, found a place in Byzantine art, as did other composites drawn from mythology, such as sirens. A small selection of such images is considered presently, drawn first from secular contexts and then from church art. One can start with a tenth- or eleventh-century casket covered with plaques of bone and ivory, housed in the collection at Dumbarton Oaks. On the lid one finds several interesting inventions, including a satyr provided with a long tail and goat’s hooves and legs (Fig. 4, center).16 There is also a bird with a human head, which one would call a siren, were it not for the dog’s head that grows out of the back of the human head (Fig. 5).17 In the fourteenth-century illuminated manuscript of the Alexander Romance, now in Venice, the kynokephaloi are shown in this way, with their dogs’ heads growing from the backs of their human heads (Fig. 6, upper right).18 The strange creature on the casket is thus really a mixture of a siren and a kynokephalos, a composite of two composites. Finally, one may note the appearance of a composite creature derived from Persia, with the foreparts of a winged lion and the tail of a peacock (Fig. 7, second from the right).19 This animal had entered the repertoire of Byzantine silk design from Iran; it can be seen ´es d’Art et d’Histoire in Brussels (Fig. 8).20 woven into a Byzantine silk from the Muse The next example is a rosette casket decorated with plaques depicting a marine thiasos, housed in the Walters Art Gallery in Baltimore. Here one finds boys riding hippocamps, animals that have a good pedigree in antique art. But there are also some less authentically classical creatures, such as dolphins whose twisted tails turn into the heads of beasts, which are inventions of both the land and the sea (Fig. 9, lower left).21 The final example of the rosette casket is provided by the Muse´e de Cluny in Paris (here I am giving only a small selection from the rich variety of images provided by these Oujde` pollou' ge kai` dei' Pa'na to` n tw'n proeirhme´ nwn eij" ge´ lwto" ajformh` n geloio´ taton, mi´gma diafo´ rwn fu´ sewn kai` duna´ mewn . . . Ed. F. Halkin, Ine´dits byzantins d’Ochrida, Candie et Moscou, SubsHag 38 (Brussels, 1963), 189.7–191.60. 15 The only composite creature from ancient art that escaped censure was the griffin, which was accepted because many Byzantines considered it to be not an invention of human artists, but an authentic work of the creator. It was praised, for example, by George of Pisidia in his Hexaemeron, PG 92:1505A. 16 K. Weitzmann, Catalogue of the Byzantine and Early Mediaeval Antiquities in the Dumbarton Oaks Collection, vol. 3, Ivories and Steatites (Washington, D.C., 1972), 54, pl. 30C. 17 Ibid., 53, pl. 31E. 18 Fol. 107; A. Xyngopoulos, Les miniatures du Roman d’Alexandre le Grand dans le codex de l’Institut Helle´nique de Venise (Athens, 1965), 122, fig. 129. 19 Weitzmann, Ivories and Steatites, 54, pl. 31C. 20 O. von Falke, Kunstgeschichte der Seidenweberei (Berlin, 1913), 2:11, fig. 236. 21 A. Goldschmidt and K. Weitzmann, Die byzantinischen Elfenbeinskulpturen des X.–XIII, Jahrhunderts (Berlin, 1930), 1:38–39, no. 40, pl. 22c. 14
HENRY MAGUIRE
193
boxes). On the back of this casket one finds a centaur with his arms around a boy who is riding on his back—these are probably Chiron and Achilles. We also see another hippocamp, this time being ridden by a winged Eros (Fig. 10, second and fourth panels from the left).22 A somewhat similar repertoire of composite creatures may be found on Byzantine tableware, both metal bowls and their down-market imitations in pottery. A silver bowl from the Hermitage Museum in St. Petersburg, which may date to the eleventh century, has a single Christian image on its inside—a relief icon of St. Theodore Tiron (Fig. 11). On the outside of the bowl, an inscription is engraved upon the rim: “Lord help your servant Theodore Tourkeles.” Beneath this invocation there is a zoo of fearsome beasts, including several man-eaters, as well as sphinxes with lions’ bodies and human heads (Fig. 12).23 From the many examples in pottery, I illustrate a twelfth-century engraved slipware bowl found at Corinth, which is decorated with a siren seizing a large water bird, presumably in order to carry it off as prey (Fig. 13).24 Another bowl from Corinth depicts a centaur in the classical manner, bearded and with horse’s hooves (Fig. 14).25 In ecclesiastical art, sculptures of mixed creatures appeared on the outside of churches. The late-twelfth-century church of the Little Metropolis in Athens presents the best-known gallery of inventions, such as the relief on the facade showing four sphinxes, two with wings and two without (Fig. 15).26 Ecclesiastical manuscripts, such as the canon tables from a Gospel Book in Parma, Biblioteca Palatina 5, contain some of the most engaging inventions in Byzantine art. Here, at the top, a siren and a centaurlike creature engage in a duet beside a fountain set in a garden, with the siren playing a harp, and the centaur—actually half man and half leopard—clashing a pair of cymbals (Fig. 16).27 In another manuscript, a winged centaur may be seen playing a lute; this initial decorates a twelfth-century copy of the Homilies of Gregory of Nazianzos, commissioned by the abbot of the Pantokrator monastery in Constantinople, Joseph Hagioglykerites (Fig. 17).28 Manuscripts containing such motifs were both commissioned and painted by monks. This is demonstrated by an illuminated Gospel Book of the twelfth century, now in the National Gallery of Victoria, Melbourne. The frontispiece miniature of this book shows a structure with a double arcade framing both the donor, a monk standing on the left, and the recipient, the Virgin holding the blessing Christ child on her left arm (Fig. 18). 22 Ibid., 39–40, no. 41, pl. 23c; K. Weitzmann, Greek Mythology in Byzantine Art (Princeton, N.J., 1951), 165–66, fig. 205. 23 A. Bank, Byzantine Art in the Collections of the USSR (Leningrad-Moscow, 1966), 368, figs. 205–7. 24 C. H. Morgan, The Byzantine Pottery, vol. 11 of Corinth: Results of Excavations Conducted by the American School of Classical Studies at Athens (Cambridge, Mass., 1942), 94, 239, no. 668, fig. 70; H. C. Evans and W. D. Wixom, eds., The Glory of Byzantium: Art and Culture of the Middle Byzantine Era, A.D. 843–1261, exhibition catalogue, The Metropolitan Museum of Art (New York, 1997), 267–68, no. 189. 25 Morgan, Byzantine Pottery, 237, no. 651, pl. 27a. 26 A. Grabar, Sculptures byzantines du moyen ˆage, vol. 2, XIe–XIVe sie`cle (Paris, 1976), 96–99, pl. 69b. 27 Gospel book, Biblioteca Palatina 5, fol. 5; P. Eleuteri, I manoscritti greci della Biblioteca Palatina di Parma (Milan, 1993), 3–13, pl. 1. 28 Homilies of Gregory of Nazianzos, Sinai 339, fol. 344v; K. Weitzmann and G. Galavaris, The Monastery of Saint Catherine at Mount Sinai: The Illuminated Greek Manuscripts, vol. 1, From the Ninth to the Twelfth Century (Princeton, N.J., 1990), 140–53, pl. 161, fig. 569.
194
THE PROFANE AESTHETIC
Fitted neatly on either side of the triangular roof are four dodecasyllable verses, which read as follows: O queen of all, as Mother of God the Logos, Theophanes is the donor and the scribe of this book, as well as the executor of the ornaments it contains, Theophanes your Nazarite servant.
Thus, the poem specifies that the monk Theophanes was not only the donor of the book, but also the writer and the artist who executed the painted ornaments.29 When one looks at these ornaments, one finds that they include some inventions, such as the epsilon that serves as an initial to St. Luke’s Gospel: it is composed of an unfortunate hare being devoured by two falcons and also, on the left, by the disembodied head of a fox that appears to grow out of the birds’ tails (Fig. 19).30 Certainly, there is a striking contrast between the austere self-image of the monk, as he presents himself in the company of Christ and the Virgin on the dedication page, and the relative frivolity of his ornaments inside. Obviously, then, the Byzantines liked looking at these composite creatures, whatever the official view of their church was on such creations. Do we, then, find any reflections of this appreciation in their literature, that is, a positive as opposed to a negative evaluation of innovative and unnatural forms in art? The Byzantines did indeed express such an appreciation, and it is where one might expect to find it—not in church writings, but in learned descriptions of classical monuments and in the secular romances. I cite three examples. The first example comes from the long poem written by Constantine the Rhodian in the tenth century and describing the church of the Holy Apostles and its mosaics. Constantine prefaces his ekphrasis of the church with an account of the Seven Wonders of Constantinople, among them the marvels of the Senate House at the Forum of Constantine (the third wonder). Here he describes at some length the bronze doors with their reliefs of the battle between the gods and the giants—the same reliefs that were criticized by Andrew the Fool. Constantine’s description of the snake-legged giants is vivid: “The giants [are shown] with their feet turned inwards and coiled underneath them like serpents . . . and the snakes, as if with flickering tongues, bellow terribly. They are grim to look at, and their eyes flash fire, so that those who gaze at them are in fright and trembling, and their hearts are filled with horror and fear.” Immediately after this dramatic description, Constantine makes a disclaimer: “With such errors was the stupid race “Anassa pa´ ntwn wJ" qeou' mh´ thr lo´ gou doth` r kat∆ aujto` kai` grafeu` " th'" puxi´do" kai` tw'n kat∆ aujth` n ejrga´ th" poikilma´ twn so` " nazirai'o" oijke´ th" Qeofa´ nh". H. Buchthal, An Illuminated Byzantine Gospel Book, Special Bulletin of the National Gallery of Victoria (Melbourne, 1961), 1. The identification of Theophanes as scribe and painter was questioned by R. S. Nelson, “Theoktistos and Associates in Twelfth-Century Constantinople: An Illustrated New Testament of A.D. 1133,” The J. Paul Getty Museum Journal 15 (1987): 53–78, esp. 63–64, but reaffirmed by I. Hutter, “Decorative Systems in Byzantine Manuscripts and the Scribe as Artist: Evidence from Manuscripts in Oxford,” Word and Image 12.1 (1996): 4–22, esp. 5–6 and nn. 24, 32. The careful distinction of Theophanes’ three roles suggests that we should take his statement at face value. 30 Buchthal, Illuminated Byzantine Gospel Book, 10, fig. 7. 29
HENRY MAGUIRE
195
of [pagan] Greece deceived, and gave an evil veneration to the indecency of vain impieties. But the great and wise [emperor] Constantine brought [the sculptures] here to be a sport for the city, to be a plaything for children, and a source of laughter for men.” 31 This statement is a paraphrase of Eusebius’s attempt to explain why Constantine the Great had decorated his new foundation with pagan statues.32 But, in the case of Constantine the Rhodian, one wonders if the topos in his ekphrasis is not inserted pro forma; certainly, his characterization of the pagan idols at the Senate House is in some respects more vivid than his following accounts of the Christian mosaics in the church of the Holy Apostles, many of which show less interest in colorful physical description than in theological commentary. Here, for example, is how Constantine begins his passage on the mosaic of the Annunciation: “The first miracle is that of Gabriel bringing to a virgin maiden [news of ] the incarnation of the Logos and filling her with divine joy.” 33 Other feast scenes from the life of Christ are given an equally terse treatment: “The fifth is the Baptism received from the hands of John by the stream of the Jordan; the Father testifying to the Logos from above, and the spirit coming down in the guise of a bird, in the resplendent form of a dove.” 34 The juxtaposition of profane and sacred in the ekphrasis of Constantine the Rhodian certainly served to enhance the solemnity of the latter, but the strategy carried with it the danger of making the devil’s works more fascinating than the stereotyped sanctity of the familiar Christian images. Later Byzantine authors penned equally vivid descriptions of the monstrous creations of pagan sculptors; now, however, the disclaimers are distinctly fainter. By way of Kai` tou` " Gi´ganta" wJ" dra´ konta" tou` " po´ da" ka´ twqen ejnstre´ fonta" ejspeirhme´ nou", rJiptou'nta" u”yei tw'n petrw'n ajpospa´ da", kai` tou` " dra´ konta" w”sper ejllicmwme´ nou", deino` n bru´ conta", blosuro` n dedorko´ ta" kai` pu'r ajposti´lbonta" ejk tw'n ojmma´ twn, wJ" tou` " oJrw'nta" deimatou'sqai kai` tre´ mein, fo´ bon te` frikto` n ejmbalei'n th' kardi´a. Toi´ai" pla´ naisin JElla´ do" mwro` n ge´ no" ejxhpatei'to kai` se´ ba" kako` n ne´ men th' tw'n matai´wn dussebw'n bdeluri´a, ajll∆ oJ kra´ tisto" kai` sofo` " Kwnstanti'no" h“negken w»de pai´gnion pe´ lein po´ lei, paisi´ t∆ a“qurma kai` ge´ lwn toi'" ajndra´ sin. ˆtres de Constantinople,” REG 9 Ed. E. Legrand, “Description des œuvres d’art et de l’e´glise des Saints Apo (1896): 40.139–52. 32 Vita Constantini 3.54. See C. Mango, “Antique Statuary and the Byzantine Beholder,” DOP 17 (1963): 56–57, 67. 33 Prw'ton me` n ou«n ge qau'ma, parqe´ nv ko´ rh to` n Gabrih` l fe´ ronta sa´ rkwsin Lo´ gou kai` carmonh'" plhrou'nta tau´ thn ejnqe´ ou . . . Ed. Legrand, “Description des œuvres,” 58.751–53. 34 Ba´ ptisma pe´ mpton cersi` prosdedegme´ non tai'" jIwa´ nnou pro` " rJoai'" jIorda´ nou, patro´ " t∆ a“nwqen marturou'nto" tv' lo´ gv kai` Pneu´ mato" foitw'nto" ojrne´ ou di´khn peristera'" eij" ei«do" hjglai¨sme´ non . . . Ibid., 59.792–60.796. The poet does, however, treat the Passion at more length, giving to the Virgin of the Crucifixion scene a moving lament; ibid., 63.916–65.981. 31
196
THE PROFANE AESTHETIC
example, one may cite Niketas Choniates praising the ancient statues of Constantinople that were destroyed by the crusaders in 1204. He calls the Latins barbarians and “haters of the beautiful,” for they did not allow “marvelous works of art to escape destruction.” 35 Among these marvelous works he describes a Nile hippocamp, with the front body of a horse joined to a scaly, spiny, tapering tail. He also describes sculptures of sphinxes, which he praises for their novelty—in other words, for the very qualities of innovation that church writers had condemned. He says that the sphinxes are “like comely women in the front, and like horrible beasts in their hind parts, moving on foot in a newly-invented manner, and nimbly borne aloft on their wings, rivalling the great winged birds.” 36 The most explicit appreciation of novelty in the visual arts is to be found in the twelfth-century novel Rodanthi and Dosiklis by Theodore Prodromos. In a remarkable passage toward the end of the work, the author compares a fond embrace uniting the two lovers and their two fathers to certain textiles that he has seen: “I have often seen in many weavings . . . such a depiction by an innovative artist, the invention, that is to say, of the weaver’s art, one head dividing itself into a quartet of bodies, or a quartet of bodies as though joined together in a single head—a four-bodied animal, or conversely, a onefaced creation [made up of ] four animals, both lion and lions. For the bodies of the beasts were displayed separately from the necks to the tails, but they all came together into the face of one lion.” 37 What kind of weaving did Prodromos have in mind when he wrote this description? In my opinion, it was probably a western rather than a Byzantine textile, and specifically one from Venice. Although in Byzantine art of this period one can find examples of oneheaded monsters with two bodies, there are very few with four.38 Such a composition is, however, frequently found on sculptures in, or coming from, Venice; one example is the All∆ j oujde` tw'n ejn tv' iJppikv' iJstame´ nwn ajgalma´ twn kai` ajlloi´wn qaumastw'n e“rgwn th` n katastrofh` n parh'kan oiJ tou' kalou' ajne´ rastoi ou»toi ba´ rbaroi. Ed. J. A. van Dieten, Nicetae Choniatae Historia (Berlin, 1975), 649.79–81. 36 Ta` " Sfi´gga" ejpi` tou´ toi", ta` " eujeidei'" wJ" gunai'ka" ta` e“mprosqen kai` frikta` " wJ" qhri´a ta` o“pisqen, kainote´ ra" de` wJ" kai` pezh' bainou´ sa" kai` kou´ fw" tv' pterv' ferome´ na" kai` diamillwme´ na" toi'" tw'n ojrni´qwn megalopte´ ruxi. Ibid., 650.22–26. 37 Ei«don kajgw` polla´ ki" ejn polloi'" pe´ ploi" . . . toiou'ton eijko´ nisma kainou' zwgra´ fou, uJfantikh'" eu”rhma dhladh` te´ cnh"⭈ mi´an kefalh` n eij" tetraktu` n swma´ twn diaireqei'san, h‘ tetraktu` n swma´ twn oi»on sunizhkui'an eij" ka´ ran mi´an⭈ zv'o´ n ti tetra´ swmon, h‘ toujnanti´on monopro´ swpon tetta´ rwn zv´ wn pla´ sin, le´ onta kai` le´ onta"⭈ oiJ ga` r aujce´ ne" a”pan to` loipo` n sw'ma th'" oujra'" me´ cri tou` " qh'ra" ejplh´ qunon th' diasta´ sei⭈ tv' de` prosw´ pv pa´ nte" h«san ei»" le´ wn. Rodanthi and Dosiklis, 9.320–34; ed. F. Conca, Il romanzo bizantino del XII secolo (Turin, 1994), 294. See the discussion of this passage in R. Beaton, The Medieval Greek Romance (Cambridge, 1989), 65. 38 A rare example of a one-headed, four-bodied lion is found in University Library, Turin, MS. C.I.6, fol. 77, where the animal forms the letter chi; a one-headed lion with two bodies forming the base of the letter tau is found on fol. 148v of the same manuscript. See J. C. Anderson, “The Illustration of Cod. Sinai. Gr. 339,” ArtB 61 (1979): fig. 19c, d. 35
HENRY MAGUIRE
197
marble roundel that is set into the north face of San Marco (Fig. 20).39 There are also several similar paterae, all from the Veneto.40 The ultimate source of the four-bodied, one-headed lion seems to have been northern Romanesque art of the type criticized by St. Bernard, such as an initial Q in a manuscript produced in the first half of the twelfth century in northern France, at St. Omer (Fig. 21).41 Whatever the source of the textile seen by Prodromos, however, one can note that the criteria by which he praised this composition, that is, innovation and invention, are precisely those that the official church doctrine condemned. II. MOVEMENT My second, and shorter, section is devoted to movement.42 By movement, I mean not the sacred movement—not, that is, the solemn processions of the liturgies, nor the decorous dances of the Hebrew women of the Old Testament, nor even the animation of holy witnesses, such as the apostles at the Ascension of Christ—but rather the unbridled gyrations of maenads and satyrs, the posturings of idols, and the ridiculous antics of actors. Here again, the profane aesthetic can be defined in negative terms, through its antithesis, namely, the ordered art and ceremony of church and palace. For example, in both ecclesiastical and political writing, an effective way of blackening an enemy was to make a synkrisis, or contrast, between the solemn movements of the Christian liturgy and the wild antics of one’s opponent. Thus, in the Life of the emperor Basil, much of which was composed as an invective against his predecessor Michael III, Michael is presented as a drunkard and compared to Dionysos in his pursuit of “what was soft, loose, voluptuous, and without rigor or moral fiber.”43 Meanwhile, his followers have been characterized as “unbridled satyrs, ready for all shameless conduct.” 44 By way of example, we are told how Michael’s protege´ Gryllos staged a bogus procession in mockery of a liturgy conducted by the patriarch Ignatios. According to the text, the most holy Patriarch Ignatios, together with his whole ecclesiastical retinue, was processing to some church with the customary liturgy and chanting. Then they met with the emperor’s impious pseudopatriarch Gryllos, who had wrapped himself in priestly attire and was riding upon an ass, together with his own retinue of fake metropolitans. In scandalized tones, the text describes how this chorus of satyrs mocked the true patriarch’s procession with an exaggerated theatrical parody, lifting up their church vestments and strumming on their stringed instruments, while singing obscene songs in competition with the chanting of 39 ´wiechowski and A. Rizzi, Romanische Reliefs von venezianischen Fassaden (Wiesbaden, 1982), 36, no. 6, Z. S pl. 3. 40 Ibid., 51, no. 104, pl. 10; 161–62, no. 743, pl. 56; 188, no. 922, pl. 69. The same composition was also woven into Italian textiles; see A. Muthesius, Studies in Byzantine and Islamic Silk Weaving (London, 1995), 219, pl. 71 (Aachen Munster Treasury, silk twill of the 13th or 14th century). 41 ´wiechowksi and Rizzi, Romanische Reliefs, 231, Boulogne-sur-Mer, Bibliothe`que Municipale 36, fol. 124; S fig. 13. 42 On this topic, see A. P. Kazhdan, “The Concept of Motion in the Vocabulary of the Byzantine Historian Nicetas Choniates” (in Russian), in Odysseus, Man in History (Moscow, 1994), 95–116. 43 Plh` n ouj to` meili´cion mo´ non kai` luai'on kai` trufhlo´ n te kai` ajneime´ non kai` aJpalo` n kai` parakekinhko` " ejk th'" me´ qh" ejke´ kthto tou' carido´ tou Dionu´ sou, o’n mimei'sqai v“ eto kai` ejspou´ dazen. Vita Basilii 26, PG 109: 268A. 44 Sa´ turoi´ tine" ou»toi kai` pro` " pa'san aijscrourgi´an ajko´ lastoi. Theophanes Continuatus 4.38, PG 109:216A.
198
THE PROFANE AESTHETIC
the genuine clergy, so that “leaping about like Pan and the satyrs, and beating their cymbals, and mocking the [true] priests and the archpriest as if they were their rivals in art, they accomplished their diabolic procession and dance.” 45 Another story that involves mockery of the liturgy is told in the Life of Leo, Bishop of Catania, which was in circulation by the tenth century. Here the opponent of the saint, the pagan magician Heliodoros, is trying to subvert the liturgy performed by the bishop on a feast day. We are told that Heliodoros came into the church together with a crowd of believers as if he were a pious Christian, but, once inside, he began to leap about in a disorderly manner, playfully jumping upon people and imitating the kicking of mules. The congregation, says the text, were by turns moved to laughter and vexed at his disgraceful nonsense and blasphemy.46 The same antithesis—of rivals in art, of order and unrestrained movement—can be observed in the carvings on the tenth- and eleventh-century Byzantine ivory and bone boxes. One can contrast the severe posing of the saints on a tenth-century casket at Dumbarton Oaks (Fig. 22)47 with another box in St. Petersburg (Fig. 23).48 On the second casket, the main point of the imagery seems to be motion rather than the reproduction of particular pagan myths. While it is certainly true that Byzantine religious art is not entirely devoid of motion, there is no doubt that the pursuit of movement and of contortion for its own sake is taken much further in these secular carvings than in any examples of church art. Some examples of carved plaques from other secular caskets, such as a box in the Museo Nazionale of Florence, show how the artists went out of their way to portray figures in motion, whether seen from the front, from the back, or upside down (Fig. 24, at the top).49 Much of this repertoire of moving figures was drawn from Dionysiac imagery; the god himself can be seen on the Veroli casket in the Victoria and Albert Museum in London (Fig. 25),50 while his cavorting maenads appear, for example, in a panel from a box in the Archaeological Museum at Pula (Fig. 26).51 On a plaque in the Victoria and Albert Museum, one even finds a drunken Herakles who has to be supported, which is a reference to the drinking contest between Herakles and Dionysos (Fig. 27).52 Centaurs Kai` Panikw'" te kai` Satirikw'" skirtw'nte", kai` kumbali´zonte", kai` wJ" ajntite´ cnou" tou` " iJerei'" kai` to` n ajrciere´ a mukthri´sante", thn` diabolikh` n corei´an kai` porei´an dih´ nuon. Vita Basilii 22, PG 109:260D–61A. On this passage, see F. Tinnefeld, “Zum profanen Mimos in Byzanz nach dem Verdikt des Trullanums (691),” Byzantina 6 (1974): 323–43, esp. 330–33; Cutler, “On Byzantine Boxes,” esp. 45; Ja. N. Ljubarskij, “Der Kaiser als ¨ B 37 (1987): 39–50; and K. Corrigan, “The ‘Jewish Satyr’ in the 9th-Century Byzantine Psalters,” Mime,” JO in Hellenic and Jewish Arts: Interaction, Tradition and Renewal, ed. A. Ovadiah (Tel Aviv, 1998), 351–68, esp. 357–58. On the characterizations of Basil and Michael in the Vita Basilii, see also P. Agapitos, “ JH eijko´ na tou' aujtokra´ tora Basilei´ou A⬘ sth` filomakedonikh` grammatei´a, 867–959,” Hellenika 40 (1989): 285–322. 46 Ed. V. V. Latyshev, Neizdannye grecheskie agiograficheskie teksty (St. Petersburg, 1914), 25.1–7. 47 Weitzmann, Ivories and Steatites, 73–77, no. 30, pls. 44–49. 48 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1:42–43, no. 51, pl. 32. 49 Ibid., 1:37–38, no. 33, pl. 20b. 50 Ibid., 1:30–32, no. 21, pl. 10e; Weitzmann, Greek Mythology, 180, fig. 229. 51 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1:34–35, no. 28, pl. 15b; Weitzmann, Greek Mythology, 180, fig. 231. 52 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1:29, no. 16, pl. 7; Weitzmann, Greek Mythology, 163, fig. 199. 45
HENRY MAGUIRE
199
also appear in a variety of strange and undignified positions, as illustrated by another plaque from the box in Florence (Fig. 28).53 A similar kind of aesthetic, celebrating movement and joie de vivre, was found in other media of Byzantine art. Recently Jeffrey Anderson and Michael Jeffreys have published a mid-twelfth-century description of a tent that was evidently embroidered with subjects very similar to those appearing on the caskets (compare with the lid of a casket in the Louvre, Fig. 29).54 The ekphrasis appears in a poem by Manganeios Prodromos; the tent was owned by his patron, the sebastokratorissa Irene. The verses read, in part: My lady, muse of muses, akropolis of beauty, the porch of your tent is filled with delights. Cupids are plucking strings and quietly strumming the cithara, satyrs seem to play, the centaurs gambol, the muses join in the dance, the nereids are leaping.55
As in the case of the invented creatures, these images celebrating frivolity and freedom of movement, with their strong Dionysiac and erotic overtones, were subversive in character. They opposed the ideals of decorum upheld by both ecclesiastical and imperial protocol. Ecclesiastical opposition to this kind of subject appears most clearly in the proscriptions of the canonists, who forbade not only the depiction of profane imagery in art, but also its performance in the theater, where the antics of mimes and buffoons were singled out for special criticism. I will cite two passages from the twelfth-century commentaries as examples. Balsamon wrote as follows on the hundredth canon of the Council in Trullo: “Since certain erotomaniacs . . . depicted in paintings, on walls, or elsewhere, erotes or other kinds of abominations, in order to sate their carnal desires by looking at them, the holy fathers . . . decreed that such things should cease completely, because they bewitch and deceive the sight, and . . . corrupt what is in the image of God.” 56 As for theatrical performances, the official view is exemplified by the following commentary of Zonaras on the fifty-first canon of the Council in Trullo: Correct Christian discipline requires the faithful not to indulge in loose and dissolute living, but to live in a manner that befits the saints. Therefore, this canon forbade whatever gives unnecessary merriment to the soul or weakens and enfeebles its [moral] fibers, and whatever causes shaking with laughter and loud guffaws. Of this kind are the actions 53 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1:37–38, no. 33, pl. 20c; Weitzmann, Greek Mythology, 167, fig. 210. 54 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1:33–34, no. 26, pl. 11a, e. 55 De´ spoina, mou'sa tw'n mousw'n, ajkro´ poli" tou' ka´ llou", ta` pro´ qura´ sou th'" skhnh'" peplh´ rwntai cari´twn. “Erwte" plh´ ttousin corda´ ", sigh' kiqarwdou'sin, dokou'si pai´zein sa´ turoi, skirtw'sin iJppokra´ tai, aiJ mou'sai sugcoreu´ ousi, phdw'si nhrh´ ide". Ed. and trans. J. C. Anderson and M. J. Jeffreys, “The Decoration of the Sevastokratorissa’s Tent,” Byzantion 64 (1994): 8–18, esp. 11.1–5. 56 jEpei` toi´nun tine` " ejrwtomanou'nte" . . . ejn pi´naxin, h‘ ejn tei´coi", h‘ ejn a“lloi" tisi`n ei“desin eijko´ nizon ejrwti´dia, h‘ kai´ tina musara´ , o”pw" ta` " sarkika` " aujtw'n ejpiqumi´a" dia` th'" pro` " tau'ta oJra´ sew" ejkperai´nwsin, w”risan oiJ a”gioi Pate´ re", crhsa´ menoi kai` marturi´ai" grafikai'", scola´ sai tau'ta panta´ pasin, wJ" katagohteu´ onta h“toi ajpatw'nta th` n o”rasin, . . . kai` parai´tia o“nta aijscrw'n kai` ajse´ mnwn tolmhma´ twn, kai` diafqeiro´ ntwn to` kat∆ eijko´ na qeou'. PG 137:861C. See H. Saradi, Aspects of the Classical Tradition in Byzantium (Toronto, 1995), 31.
200
THE PROFANE AESTHETIC of the mimes . . . who incite unseemly laughter with slaps to the temples and loud noises, and who, as it were, incite their more simple-minded and heedless [spectators] to Bacchic frenzy.57
Cupids, slapstick, immoderate merriment, Bacchic frenzy 58 —all this was to be avoided by Christians. And yet, obviously, they liked it. Again, it is in the secular romances that we find a frank appreciation of unrestrained dancing and drunken merrymaking, as is demonstrated, once more, by Prodromos’s Rodanthi and Dosiklis. In book two, there is a description of a party given in the garden of a house in the town of Rhodes, at which one of the diners, Nausikrates, “rose from his cups, and began to dance a sailor’s dance.” The narrator, Dosiklis, describes how the other guests admired the movements of the dance, whereas he had eyes only for Rodanthi: “For the others, the contortions and twistings of Nausikrates were even more delightful than the melody; they were rustic, . . . but not without grace and comic effect—but for me, [only] Rodanthi’s complexion was beautiful.” 59 More remarkable still is a farcical passage in the Drosilla and Chariklis by Niketas Eugeneianos, which describes how the drunken old woman Baryllis, picking up a hand towel, attempted to execute a Bacchic dance, but in her twistings and turnings tripped herself up on her own legs, fell, and landed upside down on her head, to the accompaniment of loud farts—all of which caused much laughter among her spectators (compare Fig. 24, top right).60 III. NUDITY The third and last aspect of the profane aesthetic to be considered here is nudity. Today the nude is more often associated with Renaissance and post-Renaissance art rather than with Byzantine art. Nevertheless, the Byzantines appreciated nudity in art more than might be supposed, even though the connotations that they gave to it were quite different from those applied in the Renaissance. In the official Byzantine view, nudity in art was condemned unless it served a clearly Christian purpose, such as portraying the nakedness of Adam and Eve before the Fall, Tou` " pistou` " hJ ajkri´beia th'" eujaggelikh'" politei´a" mh` ajneime´ nw" kai` diakecume´ nw" biou'n bou´ letai, ajll∆ wJ" pre´ pei aJgi´oi". Dia` tou'to ou«n oJ kanw` n ou»to" o”sa dia´ cusin th' yuch' ejmpoiou'si para` to` ajnagkai'on, ejklu´ ousi´ te kai` caunou'si to` n to´ non aujth'", kai` pro` " ge´ lwta" brasmatw´ dei" parakinou'si kai` kagcasmou´ ", ajphgo´ reusen, oi»a´ eijsi ta` tw'n mi´mwn . . . ejni´ote d∆ e”ter∆ a“tta toi'" ejpi` ko´ rjrJh" rJapi´smasi kai` yofh´ masi, ge´ lwta" ajprepei'" kinou'si, kai` oi»on ejkbakceu´ ousi tou` " ajfeleste´ rou" h‘ ajprosektote´ rou". PG 137:693B–C. See Tinnefeld, “Zum profanen Mimos in Byzanz,” 337–38. 58 Bacchic frenzy, as an image of disorder, was also applied to heretics. See, for example, Photius’s description of the iconoclasts as “bacchantes and harpies of the heresies and schisms,” who filled their followers with “corybantic frenzy”: B. Laourdas, Photiou homiliai (Thessalonike, 1959), 176.10–14; C. Mango, The Homilies of Photius, Patriarch of Constantinople (Cambridge, Mass., 1958), 310. 59 “Alloi" ejrasto` n th'" melwdi´a" ple´ on to` stre´ mma kai` lu´ gisma tou' Nausikra´ tou", ajgroikiko` n me` n (ti´ ga` r h‘ Nausikra´ tou"…), ouj mh` n gelw´ twn ejndee` " kai` cari´twn⭈ ejmoi` de` kalh` th'" JRoda´ nqh" hJ cro´ a. Rodanthi and Dosiklis, 2:115–19; ed. Conca, Il romanzo bizantino, 96. 60 Drosilla and Chariklis, 7:272–89; ed. Conca, Il romanzo bizantino, 460. I owe the reference to the kindness of Panagiotis Agapitos. On this passage, see Beaton, Medieval Greek Romance, 75; A. Giusti, “Nota a Niceta Eugeniano (Dros. et Char. 7:247–332),” Studi italiani di filologia classica, 3d. ser. 11 (1993): 216–23. 57
HENRY MAGUIRE
201
the human nature of Christ at his Baptism and Crucifixion,61 or the self-deprivation of certain ascetic saints.62 Otherwise, nude images were associated with paganism. Older accounts of Christian victories over idols often stressed their shameful nakedness. For example, the Life of Bishop Porphyrios of Gaza by Mark the Deacon tells us of a statue of Aphrodite that stood upon an altar in the city of Gaza, and which represented a naked woman, who, in the author’s words, “allowed all of her shameful parts to be seen.” The statue was particularly pernicious, because she was venerated by the women of the city, who hoped to receive from Aphrodite dreams giving advice concerning their matrimonial prospects; but the demon’s advice was bad, and the resulting marriages frequently ended in divorce.63 Even Niketas Choniates, who wrote the lament on the statues of Constantinople destroyed by the crusaders, elsewhere praised the bronze of Athena that stood in the Forum of Constantine because the goddess was fully dressed. Her bronze garment covered her completely, “so that no part of the body which Nature has ordained to be clothed should be exposed.” 64 However, there was another viewpoint. One can see this in a curious piece, an ethopoiia, or character study, written around 1178 by Eustathios, shortly to become Archbishop of Thessalonike. The ethopoiia is entitled “What the monk Neophytos of Mokissos might have said when, on the day after the death of his patron, the patriarch Michael III, who had done many good things for him, he was robbed of his outer garment and the rest of his clothes at the baths at the instigation of the Megas Oikonomos Pantechnes, the stolen garments then being given as a joke to the poor.” 65 The text is basically a long lament of the now naked monk, beginning with an inventory of what he had lost and continuing with a graphic description of his present predicament. At one point during his speech, the robbed man refers to art. “Painters,” he says, “portray the [three] graces naked; and they do this for the sake of symbolism. But nudity does not seem graceful to us, only disgraceful.” 66 But if nakedness was disgraceful, it was also funny, as the poor man’s complaints make clear. He describes the crowd that gathers to mock him: 61 On Christ shown naked in Crucifixion scenes, see A. P. Kazhdan and H. Maguire, “Byzantine Hagiographical Texts as Sources on Art,” DOP 45 (1991): 1–22, esp. 10–11. See also A. Cutler and A. P. Kazhdan, “The Nude,” ODB 3:1500–1501. 62 See the poems of Manuel Philes devoted to the nakedness of St. Onouphrios, cited by Alice-Mary Talbot in this volume. 63 ´ veˆque de Gaza (Paris, 1930), 47–48, Ed. H. Gre´goire and M.-A. Kugener, Marc le Diacre, Vie de Porphyre, E par. 59. 64 JW" mh´ ti tou' sw´ mato" parafai´noito, o”per hJ fu´ si" periste´ llein ejpe´ taxe. Ed. van Dieten, Nicetae Choniatae Historia, 558.53–55. 65 Poi´ou" a‘n ei«pe lo´ gou" oJ monaco` " Neo´ futo" oJ Mwkhsou', o”te th' ejpau´ rion meta` qa´ naton tou' polla` eujergeth´ ´ lou, louo´ meno" ajfhre´ qh ejx ajpostolh'" tou' santo" aujto` n aJgiwta´ tou patria´ rcou kuri´ou Micah´ l, tou' tou' Agcia j mega´ lou oijkono´ mou tou' pantecnh' to` e“xw strw'ma kai` ta` iJma´ tia kai` loipa´ , ta` ajstei´w" aujti´ka doqe´ nta ptwcoi'". Ed. T. L. F. Tafel, Eustathii Metropolitae Thessalonicensis Opuscula (Frankfurt, 1832; repr. Amsterdam, 1964), ¨rt die Ethopoiie Poi´ou" a‘n ei«pe lo´ gou" ktl. zum Briefcorpus 328.58–66, with emendations by P. Wirth, “Geho des Erzbischofs Eustathios von Thessalonike?” Classica et Mediaevalia 21 (1960): 215–17, esp. 216. The piece is discussed by A. P. Kazhdan, Studies on Byzantine Literature of the Eleventh and Twelfth Centuries (Cambridge, 1984), 127–28, 187. 66 Gumna` " oiJ zwgrafou'nte" ta` " Ca´ rita" gra´ fousi⭈ kai` tou'to aujtai'" sumboliko` n para´ shmon. All∆ j hJmi'n oujk ejpi´cari" hJ gumno´ th", ajll∆ e“xw ca´ rito". Ed. Tafel, Eustathii Opuscula, 330.46–48, with emendation from Wirth, ¨rt die Ethopoiie,” 217. “Geho
202
THE PROFANE AESTHETIC Surely, I have become a laughing stock, and have become a performer in a trial of unseemly nakedness. . . . Now the great crowd of bathers is leaving the baths, having heaped up their laughter. And so the evil that has befallen me will be heralded and spread everywhere by rumor. And these thousands of people at the baths talk about me and look me up and down, on this side and on that side . . . their faces appear amazed, their eyes stare down, and they look as if they are seeing some monster. The bath keeper’s pay has increased alright, but as for me, I am paraded about with a full escort, mocked for my unseemly condition, and made a show of.67
Thus the naked man compares himself to a monster and an object of ridicule. He has become a sideshow in the theater, like the bear led on a leash on an eleventh- or twelfth-century relief discovered at Constantinople (Fig. 30), or the plainly nude kynokephalos depicted on the relief ’s other face (Fig. 31).68 Since this kynokephalos is armed with a shield, he is of the aggressive variety that, according to the Alexander Romance, attacked Alexander the Great when he reached their territory.69 Apart from the fact that they had dogs’ heads, another strange feature of the kynokephaloi, recorded in some Byzantine sources, was that they wore no clothes.70 Thus, we have here a multiple freak— dog-headed, naked, and armed. It must have been in such sense, as freaks, that the Byzantines understood the nude fighters who disported themselves on the rosette caskets, beside the composite beasts and other monsters. As Anthony Cutler has observed,71 the pot-bellied warriors striding with their weapons into battle were surely intended to be ridiculous. One such fighter is found on a box in the collection at Dumbarton Oaks (Fig. 32);72 another one, who sticks out his bottom, is from a casket in the Metropolitan Museum of Art (Fig. 33).73 Similar naked warriors appear on Byzantine pottery, where they must also have been intended to be humorous. A twelfth-century bowl, excavated at Thebes (Fig. 34), is a parody of the common image of the heroic dragon slayer, such as appeared, for example, in a fragmentary bowl found at the Athenian agora (Fig. 35). The latter sherd also dates to the twelfth, or possibly the early thirteenth, century; it is well known because it probably represents Digenes Akritas, who shot his dragon through the neck with five darts or arrows, a feat that is referred to in some of the Akritic Songs.74 Digenes is appro-
67 Oujkou'n qe´ atron gino´ meqa ejpi` ge´ lwti, kai` gumniko` n oi»on a«qlon tou'ton ajschmosu´ nh" ajnu´ omen . . . “Oclo" oiJ louo´ menoi kai` oiJ plei´ou" ajpe´ rcontai, ge´ lwta sumforhsa´ menoi⭈ kai` to` eij" ejme` kako` n ou”tw khru´ ssetai, kai` th' fh´ mh pantacou' ski´dnatai⭈ kai` loutrou' pro´ sfasin dh'mo" ou»to" muri´o" o”so" peri` ejme´ , kai` perierga´ zontai ta` a“nw, ta` ka´ tw, ta` ejk dexiw'n, ta` eujw´ numa. . . . to` pro´ swpon eij" e“kplhxin ejschma´ tistai⭈ uJpoba´ llontai oiJ ojfqalmoi´, kai` dia´ keintai wJsei` kai´ ti te´ ra" ble´ pousi. Kai` tv' me` n balanei' plhqu´ netai oJ misqo´ ", ejmou' de` plh´ qwn kata´ getai qri´ambo", kai` eij" ajcaristi´an skwpto´ meno" qeatri´zomai. Ed. Tafel, Eustathii Opuscula, 329.87–88, 331.30–42. 68 Grabar, Sculptures byzantines, 2:39, no. 7, pl. 3; N. Firatlı, La sculpture byzantine figure´e au Muse´e Arche´ologique d’Istanbul (Paris, 1990), 161–62, no. 320, pl. 98. 69 Alexander Romance, 2:34–35 (text C); G. Bounoure and B. Serret, Pseudo-Callisthe`ne, Le Roman d’Alexandre (Paris, 1992), 191. 70 See the Vita of Makarios of Rome, ed. A. Vassiliev, Anecdota graeco-byzantina (Moscow, 1893), 139. 71 Cutler, “On Byzantine Boxes,” esp. 45. 72 Weitzmann, Ivories and Steatites, 51–52, no. 23, pl. 29C. 73 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1:28, no. 12, pl. 6b; Weitzmann, Greek Mythology, 182, fig. 243. 74 M. A. Frantz, “Digenis Akritas: A Byzantine Epic and Its Illustrations,” Byzantion 15 (1940–41): 87–91, fig. 1; eadem, “Akritas and the Dragons,” Hesperia 10.1 (1941): 9–13, fig. 1.
HENRY MAGUIRE
203
priately dressed for the task, in armor. But the hero on the bowl from Thebes accomplishes his feat in the nude, and this surely must be a Byzantine joke.75 As in the case of the invented beasts, nudity made its way into church art as well as into the art of the home. I do not speak here of the Christian subjects; rather, I am referring to the decoration of churches with frankly pagan reliefs depicting naked figures, such as those found on the north wall of the Little Metropolis church in Athens (Fig. 36).76 One also discovers initials in manuscripts portraying naked acrobats, as can be seen in an eleventh-century copy of the sermons of St. Gregory of Nazianzos, now in Turin. Here two nude men make up the letter tau (Fig. 37),77 while in another vignette, somewhat reminiscent of the carvings on the secular boxes, a running naked boy clashes a pair of cymbals at the same time as he tramples a snake (Fig. 38).78 In such images one sees how the Byzantine taste for the profane literally rubs against the prescriptions of the church fathers. IV. CONCLUSIONS In the foregoing pages some of the principal characteristics of profane art in Byzantium have been discussed. This art depicted creatures that were monstrous and unnatural in form, contorted in pose, and indecent in attire. Its attributes of recombination, movement, and nudity were not in themselves good or bad, since they appeared in some Christian contexts as well; but they acquired negative connotations when applied to forbidden subject matter. Moreover, these features were certainly accentuated and exaggerated in the profane context: composite creatures were more frequent, movement more contorted and extreme, and nakedness more explicit (see, especially, Figs. 5, 7, 10, 23, 24, and 31).79 Profane imagery was a source of both fascination and amusement. It could be found not only in ancient pagan art, as one might expect, but in medieval productions also. Whether it was the snake-legged giants on the doors of the Senate House, the statue of the tauroleon thrown off the column by St. Alypios, the drunken dance described by Prodromos, or the naked man who lost his clothes—all these forms were presented as both monstrous and ludicrous. It is true that it was a topos to say that pagan idols were the objects of ridicule; but a repeated topos is not a lie. In the Middle Ages, the Byzantines really did find pagan art to be funny, as well as sinister. It was outlandish, in the same manner as were the inventions produced by medieval artists. It is for this reason that one finds, for example, the seemingly strange juxtaposition of forms on the church of the Little Metropolis in Athens, where there are both ancient reliefs portraying naked figures See E. Dauterman Maguire, “Ceramic Arts of Everyday Life,” in Evans and Wixom, Glory of Byzantium (as in note 24), 256–57. For other examples of ceramic bowls with clothed dragon slayers, see Frantz, “Digenis Akritas,” figs. 2 (Corinth), 3 (Athenian agora), and 5 (Corinth); Morgan, Byzantine Pottery, 153, 315, no. 1502, fig. 131; 319, no. 1532, pl. 49a; 163, 333, no. 1681, fig. 141. 76 K. Michel and A. Struck, “Die mittelbyzantinischen Kirchen Athens,” AM 31 (1906): 279–324, esp. 308, 334–35, no. 87, fig. 22. 77 Homilies of Gregory of Nazianzos, University Library C.I.6, Turin. See A. Grabar, “Une pyxide en ivoire `a Dumbarton Oaks,” DOP 14 (1960): 123–46, esp. 143–44, pl. 35c; G. Galavaris, The Illustrations of the Liturgical Homilies of Gregory Nazianzenus (Princeton, N.J., 1969), 259–60. 78 Galavaris, Liturgical Homilies, pl. 9, fig. 50. 79 Cutler, “On Byzantine Boxes,” esp. 46. 75
204
THE PROFANE AESTHETIC
from pagan mythology and medieval reliefs showing composite creatures. To us these works come from different worlds, the classical and the medieval, but to the Byzantines they were from the same world of profane art. Nor did the Byzantines make a rigid division between secular art and the art of churches; the profane imagery appears in both places. This fact was noted by Niketas Stethatos, in his eleventh-century Life of Symeon the Theologian: “One can see scenes of the theater depicted in some of the churches and on the holy veils for the sake of adornment, I mean images of wrestling, of hunting, and of dancing, and different varieties of dogs and monkeys and creatures that walk and fly . . . these things are considered to be a delight and an adornment.” 80 Even though he was quoting from an earlier iconodule polemic, we have seen that he spoke the truth. Although Byzantine attitudes toward profane art remained much the same from the tenth to the twelfth century, the Byzantines tended to express their appreciation of it more frankly toward the end of the period. In the tenth century the appreciation, though certainly there, was more veiled, with stronger disclaimers. And here it is important to make a distinction between appreciation and admiration. The Byzantines of our period liked this art, because it was funny, freakish, and titillating—and they needed to laugh, because their saints could never laugh but only smiled.81 But the Byzantines did not yet admire profane art. I know of nothing written between the tenth and the twelfth century that matches the attitude of Manuel Chrysoloras, a protohumanist of the early fifteenth century who revered ancient sculpture for its ennobling qualities, saying that the beauties of the bodies depicted there reflected the mind of the supreme Creator himself.82 We have seen that the repertoire of invented creatures in Byzantine art was relatively restricted; it was certainly much smaller than in Western medieval art.83 To a large extent, Byzantine artists confined themselves to portraying the very composites inherited from pagan antiquity that their church authorities had condemned, such as centaurs, sphinxes, sirens, and satyrs.84 Only occasionally does one find an import, such as the winged lion with a peacock’s tail (derived from Persia) or the four-bodied lion described by Prodromos (which was probably Western). Thus, in Byzantium even the opposition to authority had prescribed forms. Profane art was regimented in an inverse way; even when the Byzantines were disobeying, their imaginations could not escape the discipline of their church. Finally, the profane art of Byzantium illustrates the value of synkrisis, or comparison. 80 Kai` ta` me` n th'" skhnh'" wJ" e“n tisi tw'n ejkklhsiw'n kai` iJerw'n katapetasma´ twn kallwpismou' e”neka gra´ fetai, palai'strai´ fhmi kai` kunhge´ sia kai` ojrch'strai, kunw'n de` ge´ nh kai` piqh´ kwn kai` eJrpetw'n qhri´wn te kai` peteinw'n . . . , te´ ryi" ejlogi´sqh kai` ko´ smo" aujtoi'". Ed. I. Hausherr and G. Horn, “Vie de Syme´on le Nouveau The´ologien (949–1022) par Nice´tas Ste´thatos,” OCA 12 (Rome, 1928): 128.12–16. 81 On the Byzantine sense of humor, see L. Garland, “‘And His Bald Head Shone Like a Full Moon . . .’: An Appreciation of the Byzantine Sense of Humour as Recorded in Historical Sources of the Eleventh and Twelfth Centuries,” Parergon 8 (1990): 1–31. 82 Letter addressed to Demetrius Chrysoloras, PG 156: 57C–60B. See Saradi-Mendelovici, “Christian Attitudes toward Pagan Monuments,” 59. 83 On the Western repertoire, see the classic study by J. Baltrusˇaitis, Le moyen ˆage fantastique (Paris, 1955). For a more recent treatment, see M. Camille, Image on the Edge: The Margins of Medieval Art (London, 1992). 84 Compare the observations by Jacques Le Goff on the medieval West: “It is not farfetched to say that the marvelous was one form of resistance to the official ideology of Christianity”; The Medieval Imagination (Chicago, 1985), 32. I thank Helen Saradi for this reference.
HENRY MAGUIRE
205
The profane motifs on the outside of the churches, or in the margins of the sacred texts, made what was inside more holy by contrast. Just as the stereotyped piety of Leo of Catania was enhanced by the demonic antics of Heliodoros, so too the disorder of the reliefs on the outside of a church enhanced the good order of the images inside, which adhered strictly to the rules.85 In Byzantium official art and profane art were mutually reinforcing; it is impossible to cut the one from the other. University of Illinois, Urbana-Champaign 85 In later Byzantine art, especially during the Palaiologan period, profane motifs began to invade the margins of the christological scenes themselves, such as the Baptism (nude bathers, sea monsters), but the aesthetic effect remained the same—namely, the solemnity of the sacred figures at the center of the composition was enhanced. See G. Millet, Recherches sur l’iconographie de l’e´vangile (Paris, 1960), 170–215; D. Mouriki, “Revival Themes with Elements of Daily Life in Two Palaeologan Frescoes Depicting the Baptism,” in ˇevcˇenko, eds. C. Mango and O. Pritsak (Cambridge, Mass., 1983), 458–88. So, Okeanos: Essays Presented to Ihor S also, the serenity of Christ was contrasted with those who mocked him; see, for example, A. Derbes, Picturing the Passion in Late Medieval Italy (Cambridge, 1996), esp. 99–107, fig. 62 (Staro Nagoricˇino).
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Lions with two bodies, capital, cloister of St.-Pierre, Moissac
2 “Quadruped with the tail of a serpent,” capital, cloister of St.-Pierre, Moissac
3
Giants, Theriaka of Nikander, Bibliothèque Nationale suppl. gr. 247, fol. 47, Paris (photo: courtesy of the library)
4 Casket, Dumbarton Oaks, Washington, D.C.
5 Bird with human and canine heads, casket, detail of the lid, Dumbarton Oaks, Washington, D.C.
6
Alexander and the kynokephaloi, Alexander Romance, fol. 107, Istituto Ellenico di Studi Bizantini e Postbizantini di Venezia, Venice (after A. Xyngopoulos, Les miniatures du Roman d’Alexandre le Grand dans le codex de l’Institut Hellénique de Venise [Athens, 1965], fig. 129)
7
Casket, back of the lid, Dumbarton Oaks, Washington, D.C.
8
Silk, Musées d’Art et d’Histoire, Brussels (after O. von Falke, Kunstgeschichte der Seidenweberei [ Berlin, 1913], 2: fig. 236)
9
10
Casket, Walters Art Gallery, Baltimore (photo: courtesy of the gallery)
Casket, Musée de Cluny, Paris (photo: courtesy of Réunion des Musées Nationaux)
11
12
St. Theodore Tiron, silver bowl, detail of interior, Hermitage Museum, St. Petersburg (after A. Bank, Byzantine Art in the Collections of the USSR [ Leningrad-Moscow, 1966], fig. 206)
Silver bowl, Hermitage Museum, St. Petersburg (after Bank, Byzantine Art in Collections, fig. 205)
13
14
Centaur, ceramic bowl, Archaeological Museum, Corinth (after C. H. Morgan, The Byzantine Pottery, vol. 11 of Corinth: Results of Excavations Conducted by the American School of Classical Studies at Athens [Cambridge, Mass., 1942], pl. 27a)
Siren and water bird, ceramic bowl, Archaeological Museum, Corinth
15
Sphinxes, Panaghia Gorgoepikoos (Little Metropolis), detail of the west facade, Athens
16
Canon tables with musical inventions, Gospel Book, Biblioteca Palatina 5, fol. 5, Parma (after P. Eleuteri, I manoscritti greci della Biblioteca Palatina di Parma [ Milan, 1993], pl. 1)
17
“Centaur” with a lute, Homilies of Gregory of Nazianzos, Sinai 339, fol. 344v, St. Catherine’s monastery, Mt. Sinai (photo: courtesy of the Michigan-Princeton-Alexandria Expedition to Mount Sinai)
18
Monk Theophanes with the Virgin, Gospel Book, Felton 710/5, fol. 1v , National Gallery of Victoria, Melbourne (after H. Buchthal, An Illuminated Byzantine Gospel Book, Special Bulletin of the National Gallery of Victoria [ Melbourne, 1961], cover)
19
Opening of St. Luke’s Gospel, Gospel Book, Felton 710/5, National Gallery of Victoria, Melbourne (after Buchthal, Illuminated Byzantine Gospel Book, fig. 7)
20
Lion with one head and four bodies, marble roundel on the north facade, San Marco, Venice (photo: Osvaldo Böhm)
21
Initial Q with four-bodied lion, Bibliothèque Municipale 36, fol. 124, Boulogne´ sur-Mer (after Z. Swiechowski and A. Rizzi, Romanische Reliefs von venezianischen Fassaden [ Wiesbaden, 1982], fig. 13)
22
23
Deesis with apostles and saints, casket, Dumbarton Oaks, Washington, D.C.
Figures in motion, casket, Hermitage Museum, St. Petersburg (after A. Goldschmidt and K. Weitzmann, Die byzantinischen Elfenbeinskulpturen des X.–XIII. Jahrhunderts [ Berlin, 1930], 1: pl. 32)
24
Figures in motion, casket, Museo Nazionale del Bargello, Florence (photo: courtesy of Giraudon/Art Resource, New York)
25
Dionysos, Veroli casket, detail, Victoria and Albert Museum, London (photo: courtesy of the museum)
26
Maenads, plaque from a casket, Archaeological Museum, Pula (after Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1: pl. 15b)
27
Drunken Herakles, plaque from a casket, Victoria and Albert Museum, London (photo: courtesy of the museum)
28
Centaur, casket, detail, Museo Nazionale del Bargello, Florence (after Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, 1: pl. 20c)
29
Playing cupids and centaurs, casket, lid, Musée du Louvre, Paris (photo: courtesy of Réunion des Musées Nationaux)
30
32
31
Bear on a leash, bilateral stone relief, Archaeological Museum, Istanbul (after A. Grabar, Sculptures byzantines du moyen âge, vol. 2, XIe–XIVe siècle [ Paris, 1976], pl. 3a)
Naked warrior, casket, detail of the back, Dumbarton Oaks, Washington, D.C.
33
Armed kynokephalos, bilateral stone relief, Archaeological Museum, Istanbul (after Grabar, Sculptures byzantines, pl. 3b)
Naked warrior, casket, detail, Metropolitan Museum of Art, New York (photo: courtesy of the museum)
34
35
Naked dragon slayer, ceramic bowl, excavated at Thebes
Digenes Akritas slaying the dragon, ceramic bowl, excavated in the Athenian agora (after M. A. Frantz, “Akritas and the Dragons,” Hesperia 10.1 [1941]: fig. 1)
36
Pagan relief with added crosses, Panaghia Gorgoepikoos (Little Metropolis), detail of the north wall, Athens
37
Naked acrobats, Homilies of Gregory of Nazianzos, University Library, C.I.6, Turin (after A. Grabar, “Une pyxide en ivoire à Dumbarton Oaks,” DOP 14 [1960]: pl. 35c)
38
Naked performer, Homilies of Gregory of Nazianzos, University Library, C.I.6, fol. 67, Turin (after G. Galavaris, The Illustrations of the Liturgical Homilies of Gregory Nazianzenus [ Princeton, N.J., 1969], fig. 50)
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Exe´ge`se, typologie et rhe´torique dans l’hymnographie byzantine CHRISTIAN HANNICK
n comparaison avec l’ampleur des textes qui nous sont parvenus et dont beaucoup sont encore en usage dans les ´eglises orthodoxes, l’hymnographie byzantine est mal ´etudie´e et repre´sente peut-eˆtre dans le domaine the´ologique grec le genre litte´raire le moins connu.1 Ce qui ne signifie pas que son niveau soit la raison de ce manque d’inte´reˆt. Autant la forme que le contenu des milliers de stichera, troparia, kathismata, kontakia, heirmoi et autres types d’hymnes confe`rent `a ces pie`ces, joyaux de la poe´sie et de la musique, une valeur artistique dont il nous manque peut-eˆtre les crite`res pour en appre´cier la juste valeur. Si l’on tient compte du fait que poe´sie et musique ne font qu’un dans l’hymnographie, on comprendra ainsi la porte´e de titres de publications comme Tre´sor de musique byzantine de Egon Wellesz (Paris, 1934) ou P. Lorenzo Tardo e la musica bizantina: Tesori della melurgia bizantina (Grottaferrata, 1985), ouvrage posthume du re´novateur du chant byzantin `a l’abbaye grecque de Grottaferrata (1883–1967).2 En tout cas, les lettre´s contemporains ne me´prisaient pas les compositions hymnographiques; The´odore Prodromos, Jean Zonaras, pour ne citer que deux personnalite´s bien connues du XIIe sie`cle, n’ont pas cru superflu de re´diger d’abondants commentaires, l’un sur les canons des grandes feˆtes, l’autre sur les huit canons de la re´surrection, attribue´s `a Jean Damasce`ne.3 Ces commentaires de toute grande importance pour un historien de l’hymnographie byzantine, sont loin d’avoir ´ete´ ´etudie´s et mis en valeur jusqu’a` pre´sent. Les deux ´editions signale´es sont partielles et de´pourvues de toute annotation. Un des meilleurs connais¨randner, constatait seurs de l’he´ritage litte´raire de The´odore Prodromos, Wolfram Ho justement il y a une vingtaine d’anne´es: “Wir haben es hier mit einem Werk zu tun, das sofort nach seinem Erscheinen allgemeine und zweifellos auch offizielle Anerkennung
E
1 ¨ve´rffy, A Guide to Byzantine Hymnography: A Classified Bibliography of Texts and Studies, Voir, en ge´ne´ral, J. Szo 2 vols. (Brookline, Mass.-Leyde, 1978–79). 2 Cf. O. Strunk, “Padre Lorenzo Tardo e il suo Ottoeco nei mss. melurgici,” BollGrott 21 (1967): 21–33; version anglaise dans Essays on Music in the Byzantine World (New York, 1977), 255–67. 3 Voir l’e´dition partielle de H. M. Stevenson, Theodori Prodromi Commentarios in carmina sacra melodorum Cosmae Hierosolymitani et Ioannis Damasceni (Rome, 1888); S. Lauriotes, “ jIwa´ nnh" Zwnara'", JErmhnei´a tw'n ajnastasi´mwn kano´ nwn tou' jIwa´ nnou Damaskhnou',” JO “Aqw" 1.3–4 (Athe`nes, 1920).
208
L’HYMNOGRAPHIE BYZANTINE
fand und durch die Jahrhunderte hindurch ein Handbuch ersten Ranges fu ¨ r den theologischen Unterricht blieb.” 4 Tout paradoxe que cela puisse paraıˆtre, l’hymnographie repre´sente une veine litte´raire encore vivante de nos jours. Il y a quelques anne´es (7 de´cembre 1991) le ce´le`bre ´ glise, le moine Gerasimos de la skite de ste. Anne, mourait hymnographe de la Grande E 5 au Mont Athos, apre`s avoir re´dige´ nombre d’acolouthies pour de nouveaux saints, et prenait ainsi place dans une ligne´e de poe`tes religieux illustre´e par des noms comme Joseph Sikelos ou Theophanes, qui, tout comme Gerasimos, ont compose´ des hymnes sur des me´lodies de´ja` connues, sans donc ˆetre `a la fois me´lographe, `a l’oppose´ de poe`tesmusiciens comme Jean Damasce`ne ou du patriarche Sophrone de Je´rusalem. Il n’est pas de notre propos ici de juger de la qualite´ litte´raire et artistique des compositions du moine Gerasimos. Les de´buts de la poe´sie religieuse `a Byzance se laissent mal saisir. Karyophiles Metsakes et Michael Lattke y ont consacre´ chacun un gros volume riche de mate´riaux,6 dont il apparaıˆt que les parties hymniques du Nouveau Testament, les psaumes, les odes bibliques, certains passages des prophe`tes, puis la prose rythme´e des pe`res de l’e´glise, `a commencer par Me´liton de Sardes jusqu’a` Gre´goire de Nazianze, sont `a l’ore´e de cette litte´rature immense qui se de´veloppera en rapport avec l’e´volution des formes liturgiques. Si les de´buts de l’hymnographie byzantine restent en beaucoup de points dans l’ombre—je ne ferai qu’e´voquer ici les proble`mes relatifs `a la poe´sie religieuse relie´e `a la liturgie de Je´rusalem et dont le tropologion ge´orgien est l’une des sources les plus anciennes7 —il en est de meˆme pour l’histoire de plusieurs des diffe´rents genres hymnographiques ainsi que pour la composition et le de´veloppement des collections d’hymnes, c.-a`-d. des recueils hymnographiques destine´s `a l’usage liturgique. En effet, que sait-on de la formation et de l’exe´cution musicale du kathisma, atteste´ de`s le Xe sie`cle dans un recueil appele´ sticherokathismatarion (par exemple, Sinai gr. 779, fols. 11r–43v, Xe sie`cle; ou Sinai gr. 778, fols. 1r–54v, XIe sie`cle) dont il n’existe aucun exemplaire neume´ et dont il est donc impossible d’appre´cier le genre musical et les rapports avec d’autres types d’hymnes. Connait-on mieux l’histoire du tropologion, une appellation de manuscrits liturgiques non neume´s, du reste ambivalente, mais dont on a maintenant des te´moins irre´futables graˆce aux de´couvertes de 1975 au monaste`re du Mont Sinai,8 un type de ¨randner, Theodoros Prodromos: Historische Gedichte, WByzSt 11 (Vienne, 1974), 44. W. Ho Ire´nikon 65 (1992): 105. 6 K. Metsakes, Buzantinh` uJmnografi´a A´. Apo j ` th` n Kainh` Diaqh´ kh e”w" th` n eijkonomaci´a (Thessalonike, 1971; ¨ B 39 (1989): 320–21; M. Lattke, Hymnus: Materialien zu einer Geschichte 2e ´ed., Athe`nes, 1986); cf. J. Koder, JO der antiken Hymnologie, Novum Testamentum et orbis antiquus 19 (Fribourg, 1991). 7 ˇaraknoc arme´nien,” REArm 24 (1993): 89–112; P. Jeffery, Cf. C. Renoux, “Le Iadgari ge´orgien et le S “Jerusalem and Rome (and Constantinople): The Musical Heritage of Two Great Cities in the Formation of the Medieval Chant Traditions,” dans Cantus Planus: Papers Read at the Fourth Meeting, Pe´cs, Hungary, 3–8 September 1990 (Budapest, 1992), 163–74. 8 P. G. Nikolopoulos, JIera` monh` kai` ajrciepiskoph` Sina´ Ú Ta` ne´ a euJru´ mata tou' Sina' (Athe`nes, 1998), 146 (cod. MG 28); cf. H. Husmann, “Hymnus und Troparion: Studium zur Geschichte der musikalischen Gat¨r Musikforschung Preußischer Kultungen von Horologion und Tropologion,” Jahrbuch des staatlichen Instituts fu turbesitz (1971): 7–86; voir aussi, en ge´ne´ral, C. Hannick, “Hymnen. II: Orthodoxe Kirche,” dans Theologische Realenzyklopa¨die (Berlin–New York, 1986), 15:762–70; idem, “Byzantinische Musik,” dans Die Musik in Geschichte und Gegenwart (Kassel, 1995), 2:288–310. 4 5
CHRISTIAN HANNICK
209
ˆ repre´senter un large recueil d’hymnes avant la re´forme liturgique manuscrit qui a du ` l’on retravailla ope´re´e au milieu du IXe sie`cle apre`s les luttes iconoclastes, `a l’e´poque ou 9 ´egalement dans une large mesure le syste`me des lectures bibliques. Dans d’autres domaines on se trouve confronte´ `a des proble`mes insurmontables, vu la tradition lacunaire comme dans le cas de l’hypakoe,10 atteste´e avec me´lodie dans les psaltika provenant de l’Italie me´ridionale au XIIIe–XIVe sie`cle, et de´ja` mentionne´e dans le kanonarion ge´orgien.11 Comme l’hymnographie est une poe´sie chante´e (sangbare Dichtung comme le disait Jacques Handschin il y a environ cinquante ans)12 et exe´cute´e dans le cadre liturgique, une appre´ciation de ces textes doit tenir compte ´egalement des aspects musicaux et de l’histoire de la liturgie. Ce serait toutefois de´border le cadre de cette pre´sentation si l’on cherchait `a prendre en conside´ration, dans la mesure requise, des ´ele´ments musicaux et liturgiques. Je ne ferai que les effleurer, le cas ´eche´ant, en soulignant d’emble´e que pre´cise´ment la` re´side une grande lacune dans l’e´tude de l’hymnographie byzantine. Il serait ne´cessaire en effet de reprendre et d’approfondir l’approche qu’en donnait le cardinal Jean Baptiste Pitra il y a 130 ans,13 en tenant compte des re´sultats acquis dans le domaine de la musicologie et qui ont fait plusieurs fois l’objet d’un aperc¸u critique pre´sente´ aux congre`s internationaux de byzantinologie. Un des savants pre´sents `a cette confe´rence ne me contredira pas quand j’exprime une fois de plus mon ´etonnement sur le fait que les recherches musicologiques sur le kontakion, ayant produit des re´sultats significatifs voici trente ans de´ja`, restent pratiquement ignore´es dans les ´etudes philologiques sur Romanos et dans les deux grandes ´editions des dernie`res de´cennies.14 Il ne faut pas oublier toutefois que les travaux sur le genre musical kondakarien, et, de la sorte, tout essai d’interpre´tation musicologique des kontakia se fondent sur des manuscrits neume´s, grecs et slaves, que plus d’un demi-mille´naire se´parent de Romanos. Les cinq te´moins vieux-russes conserve´s datent du XIe–XIIe sie`cle, les psaltika grecs provenant de l’Italie du Sud et qui seuls conservent une version pleinement lisible, parce que diaste´matique, de la me´lodie et de la tonalite´ des kontakia datent du XIIIe–XIVe sie`cle. Ils repre´sentent certes une tradition plus ancienne, mais qu’il serait impossible de faire remonter `a Romanos, au VIe sie`cle. Ce proble`me chronologique doit ˆetre pris en conside´ration dans tout essai d’une approche musicologique du kontakion. Cela ne justifie toutefois aucunement l’ignorance—ou meˆme le de´dain—des aspects musicaux de la
9 ¨ berlieferung der katholischen Briefe,” Cf. K. Junack, “Zu den griechischen Lektionaren und ihrer U ¨ bersetzungen des Neuen Testaments, die Kirchenva¨terzitate und Lektionare, ´ed. K. Aland, Arbeiten dans Die alten U zur neutestamentlichen Textforschung 5 (Berlin–New York, 1972), 498–575, surtout “Versuche zur Altersbestimmung des byzantinischen Lektionssystems,” 532 sq. Voir aussi, en ge´ne´ral, Ch.-B. Amphoux et J.-P. Bou´ glise ancienne (Lausanne, 1996). hot, ´eds., La lecture liturgique des Epitres catholiques dans l’E 10 Cf. L. Calı`, “Le ipacoe` dell’octoichos bizantino,” BollGrott 19 (1963): 161–74. 11 Cf. H. Leeb, Die Gesa¨nge im Gemeindegottesdienst von Jerusalem (vom 5. bis 8. Jahrhundert) (Vienne, 1970), 221 sq. 12 J. Handschin, Das Zeremonienwerk Kaiser Konstantins und die sangbare Dichtung (Basel, 1942). 13 J. B. Pitra, Hymnographie de l’e´glise grecque (Rome, 1867). 14 Cf. K. Levy, “Die slavische Kondakarien—Notation,” dans Anfa¨nge der slavischen Musik (Bratislava, 1966), ¨r Herbert Hunger zum 70. 77–92; C. Hannick, “Zur Metrik des Kontakion,” dans BUZANTIOSÚ Festschrift fu Geburtstag (Vienne, 1984), 107–19. Les exemples musicaux sur lesquels mon ´etude est fonde´e sont tire´s du cod. Ashburnh. 64, copie´ en 1289.
210
L’HYMNOGRAPHIE BYZANTINE
poe´sie byzantine qui affecte moins les autres genres hymnographiques comme, par exemple, le sticheron auquel par ailleurs aucune ´etude philologique n’a ´ete´ consacre´e.15 Il reste encore un point `a signaler en ce qui concerne l’inte´reˆt porte´ `a l’hymnographie byzantine par les contemporains de l’effloraison de cette litte´rature, c’est sa diffusion dans la tradition slave `a partir du Xe sie`cle et plus encore en Rus’ kie´vienne de`s la fin du XIe sie`cle. On souligne volontiers l’importance des rapports byzantinoslaves dans le domaine de l’hagiographie,16 ce qui explique aussi qu’une section du Congre`s byzantinologique ˆ t 1996 ait ´ete´ consacre´e `a ce sujet, bien qu’il faille constater que de Copenhague d’aou beaucoup de pie`ces de l’hagiographie byzantine, surtout les enkomia de l’e´poque patristique, ne sont pas passe´es en slave avant le XIVe sie`cle, et ceci graˆce aux traducteurs de ˘ rnovo.17 Quant `a l’hymnographie, qui re´pondait aux meˆmes bel’e´cole d’Euthyme de Tu soins liturgiques que l’hagiographie, les manuscrits provenant des terres russes de`s le XIIe sie`cle te´moignent de l’ampleur du travail de traduction et de re´ception. A l’e´poque de la “seconde influence slave me´ridionale en Russie,” `a la charnie`re du XIVe au XVe sie`cle, tout le re´pertoire hymnographique ´etait pratiquement traduit. Ici aussi beaucoup reste `a ´etudier; un livre central de la liturgie comme le sticherarion attend encore d’eˆtre ˆt18 publie´ en vieux-russe. Une ´etude de Nicolas Schidlovsky annonce´e pour tre`s biento devrait apporter de nouvelles lumie`res sur ces proble`mes et servir ainsi de pendant au beau travail de Milosˇ Velimirovic´ sur la tradition slave de l’Hirmologion.19 On pourra s’e´tonner aussi que compte tenu des traditions textuelles diversifie´es, nombre d’hymnes d’origine byzantine soient conserve´es en vieux-slave sans que l’on puisse en retrouver l’original grec. Un projet d’inventaire de ce mate´riel important est actuellement en cours `a Wu ¨ rzburg. Mais quittons maintenant ces conside´rations sur l’e´tat des recherches pour nous pencher sur les aspects principaux de cette pre´sentation, l’exe´ge`se, la typologie et la rhe´torique dans les hymnes byzantines. Le lieu de la naissance de Je´sus d’apre`s Luc 2.7 preˆte sujet `a deux interpre´tations diffe´rentes releve´es de´ja` dans l’exe´ge`se patristique a` partir d’Orige`ne.20 Il est question ou bien d’une creˆche (fa´ tnh) ou d’une grotte (sph´ laion). Les textes de l’e´vangile (Luc 2.7) ne mentionnent que la creˆche alors qu’une partie de la tradition patristique, suivie en cela par beaucoup d’hymnographes, situe le lieu de la naissance dans la grotte. Les 15 L’e´tude de J. Raasted, “Some Observations on the Structure of the Stichera in the Byzantine Rite,” ¨ve´rffy, Guide, 2:231 sq. Byzantion 28 (1958): 529–41, reste capitale; cf. la bibliographie pre´sente´e par Szo 16 Cf., par exemple, I. Dujcˇev, “Les rapports hagiographiques entre Byzance et les Slaves,” dans Medioevo bizantino-slavo (Rome, 1971), 3:267–79 (rapport pre´sente´ au Congre`s byzantinologique d’Oxford en 1966). 17 Cf. C. Hannick, Maximos Holobolos in der kirchenslavischen homiletischen Literatur, WByzSt 14 (Vienne, 1981), ˇ ertorickaja, Vorla¨ufiger Katalog kirchenslavischer Homilien des beweglichen 57 sq. L’important catalogue de T. C Jahreszyklus: Aus Handschriften des 11.–16. Jahrhunderts vorwiegend ostslavischer Provenienz, Abhandlungen der Nordrhein-Westfa¨lischen Akademie der Wissenschaften 91 (Opladen, 1994), ne donne aucune indication sur les sources grecques. 18 L’e´dition du Sticherarion St. Petersbourg, Bibliothe`que de l’Acade´mie des sciences (BAN) 34.7.6 (Svodnyj ˇ ukovskaja et al. [Moscou, katalog slavjano-russkich rukopisnych knig, khranjasˇˇcichsja v SSSR, XI–XIII vv., ´ed. L. P. Z 1984], no. 98) dans les Monumenta Musicae Byzantinae. Voir aussi N. Schidlovsky, “The Notated Lenten Prosomoia in the Byzantine and Slavic Traditions” (The`se de doctorat, Princeton Universite´, N.J., 1983). 19 M. Velimirovic´, Byzantine Elements in Early Slavic Chant: The Hirmologion, Monumenta Musicae Byzantinae, Subsidia 4 (Copenhague, 1960). 20 R. Stichel, Die Geburt Christi in der russischen Ikonenmalerei (Stuttgart, 1990), 52 sq.
CHRISTIAN HANNICK
211
plus anciens te´moignages qui se rapportent `a une grotte se trouvent chez Justin et dans le Prote´vangile de Jacques.21 Parmi les nombreux exemples retenons un troparion du deuxie`me mode plagal pour les veˆpres du 25 de´cembre: jAne´ teila" Criste` ejk parqe´ nou * nohte` h”lie th'" dikaiosu´ nh" * kai` ajsth´ r se uJpe´ deixen * ejn sphlai´v cwrou´ menon to` n ajcw´ rhton * ma´ gou" oJdhgh´ sa" eij" prosku´ nhsi´n sou * meq∆ w»n se megalu´ nomen * zwodo´ ta do´ xa soi (MR II 655).22 La grotte (sph´ laion) est de´ja` reprise chez Romanos dans le ce´le`bre kontakion de la Nativite´: H J parqe´ no" sh´ meron to` n uJperou´ sion ti´ktei * kai` hJ gh' to` sph´ laion tv' ajprosi´tv 23 prosa´ gei. Le meˆme Romanos, dans le deuxie`me oikos de ce kontakion, mentionne ´egalement la creˆche: oJ swth` r tw'n brefw'n bre´ fo" ejn fa´ tnh e“keito… ou mieux encore, dans le troisie`me oikos, il combine les deux concepts comme c’est souvent le cas dans l’iconographie: sphlai´ou hjra´ sqh" h‘ fa´ tnh ejte´ rfqh" (Es-tu ´epris d’une grotte, amoureux d’une creˆche?).24 La meˆme combinaison des deux termes caracte´rise ´egalement un troparion d’un canon anonyme pour l’avant-feˆte de Noe¨l: O J lo´ gv tei´na" oujrano` n * uJpeise´ rch sphlai´v * kai` ajlo´ gwn ejn fa´ tnh ajnake´ klisai (MR II 600). Ce the`me de la grotte apparaıˆt de´ja` dans le Prote´vangile de Jacques: hJ Mari´a . . . e“laben to` n pai'da kai` ejsparga´ nwsen aujto` n kai` e“balen ejn pa´ qnh bow'n…25 ainsi que dans un autre canon anonyme de l’avantfeˆte: No´ mwn se hJ parqe´ no" * tw'n th'" sarko` " di´ca ku´ rie * paragi´netai ajpokuh'sai * ejn tv' sphlai´v kai` fa´ tnh * sarki` ajnaklinei' se wJ" nh´ pion (MR II 615). Compte tenu des spe´cificite´s du genre litte´raire de l’hymnographie, on cherchera souvent en vain dans ces textes une attestation incontestable pour une variante significaˆ r en de nombreux passages tive du texte biblique. Les hymnographes reprennent bien su des citations directes et explicites a` l’Ecriture sainte, rehausse´e, comme je l’ai de´ja` signale´ ailleurs, par des proce´de´s musicaux qui rappellent le style ekphone´tique.26 C’est surtout dans les stichera idiomela que l’on rencontre des citations e´tendues et ceci conforme´ment aux re`gles de composition de ce genre d’hymnes. En effet, le caracte`re intrinse`que d’un idiome`le re´side dans la me´lodie propre, unique, de ce texte, ce qui veut dire que dans ce cas l’hymnographe, qui est tout `a la fois le me´lographe, n’est pas astreint `a reprendre un sche´ma me´trique pre´existant. Aussi est-ce difficile de de´finir les caracte´ristiques poe´tiques d’un stiche`re idiome`le ou de pre´ciser ce qui distingue un tel texte de la prose ´eleve´e, par exemple, d’une home´lie. Les stichera du temps pascal abondent en citations bibliques, qui, conforme´ment au cycle de lecture, sont reprises a` l’e´vangile de s. Jean. L’exemple le plus significatif, peutˆetre, pour notre propos est fourni par un doxastikon du premier mode plagal pour les ` l’on lit la veˆpres du dimanche du paralytique, le troisie`me dimanche apre`s Paˆques, ou pe´ricope de la gue´rison du paralytique `a la piscine de Bethsaida (Jean 5.1–15). Le doxas21 Cf. E. De Strycker, La forme la plus ancienne du Prote´vangile de Jacques, SubsHag 33 (Bruxelles, 1961), 146, par. 43.1; commentaires, 416–17. Dans le Prote´vangile, la creˆche apparaıˆt dans un autre contexte (ibid., 174, par. 22.2; commentaire, 175, no. 2). 22 Abre´viations des livres liturgiques d’apre`s E. Follieri, Initia hymnorum ecclesiae graecae, 5 vols., ST 211– 15bis (Rome, 1960–66). 23 J. Grosdidier de Matons, Romanos le Me´lode, Hymnes, SC 110 (Paris, 1965), 2:50; MR II 666. 24 Grosdidier de Matons, Hymnes, 2:52–53. 25 De Strycker, Prote´vangile de Jacques, 174, par. 43 (avec variante fa´ tnh - pa´ qnh). 26 C. Hannick, “Ele´ments ekphone´tiques dans le Sticherarion,” dans Actes du XIVe Congre`s international des ´ tudes byzantines (Bucarest, 1976), 3:537–41. E
212
L’HYMNOGRAPHIE BYZANTINE
tikon en question est attribue´ `a un hymnographe Kumulas dont on ne sait rien27 et qui est conside´re´ comme l’auteur de quatre ou cinq idiomela pour le dimanche des myrophores et le dimanche du paralytique. Ce nom est atteste´ de`s les plus anciens sticheraria conserve´s, comme, par exemple, le cod. Vind. theol. gr. 136 de la premie`re moitie´ du XIIe sie`cle, et le cod. Athous Vatop. 1488 de la premie`re moitie´ du XIe sie`cle.28 On a ainsi un terminus ante quem, le XIe sie`cle, mais il est ´evident que ces compositions et leur auteur sont bien plus anciens. Avant de passer aux textes attribue´s `a Kumulas il est indispensable de prendre en conside´ration le verset biblique central dont il sera question ensuite (Jean 5.7): a“nqrwpon oujk e“cw i”na, o”tan taracqh' to` u”dwr, ba´ lh me eij" th` n kolumbh´ qran ⭈ ejn v» de` e“rcomai ejgw´ , a“llo" pro` ejmou' katabai´nei { ⫹ kai` lamba´ nei th` n i“asin } { ⫹ ejgw` de` ajsqenw'n poreu´ omai }. Donnons d’abord le texte complet de ce sticheron E j pi` th' Probatikh' kolumbh´ qra d’apre`s l’e´dition romaine (PeR 140) de 1883 et soulignant en italique les ´ele´ments repris au texte ´evange´lique (Jean 5.2–8): Ej pi` th' Probatikh' kolumbh´ qra * a“nqrwpo" kate´ keito ejn ajsqenei´a * kai` ijdw´ n se, * Ku´ rie, ejbo´ a⭈ * a“nqrwpon oujk e“cw, * i”na, o”tan taracqh' to` u”dwr * ba´ lh me ejn aujtv'⭈ * ejn v» de` poreu´ omai, * a“llo" prolamba´ nei me, * kai` lamba´ nei th` n i“asin, * ejgw` de` ajsqenw'n kata´ keimai. * Kai` eujqu` " splagcnisqei`" oJ Swth` r * le´ gei pro` " aujto´ n⭈ * Dia` se` * a“nqrwpo" ge´ gona, * dia` se` * sa´ rka peribe´ blhmai, * kai` le´ gei"⭈ * a“nqrwpon oujk e“cw… * a«ro´ n sou to` n kra´ bbaton kai` peripa´ tei. * Pa´ nta soi dunata´ , * pa´ nta uJpakou´ ei, * pa´ nta uJpote´ taktai⭈ * pa´ ntwn hJmw'n mnh´ sqhti, * kai` ejle´ hson, ”Agie, * wJ" fila´ nqrwpo".
Notons, d’abord, que les sticheraria neume´s de Vienne et de Vatopedi donnent la lec¸on poreu´ omai au lieu de kata´ keimai et que l’e´dition de Vitali de 1738 (Pent 98) retient kata´ keimai mais pre´sente eujsplagcnisqei`" au lieu de splagcnisqei´". Tillyard 29 propose une transcription de cette pie`ce d’apre`s trois manuscrits en notation me´diobyzantine (Cardiff, University College; Cantabr. Trinitatis 256 et Cod. Vind. theol. gr. 181) du XIIIe–XIVe sie`cle.30 La partie centrale du sticheron, kai` lamba´ nei th` n i“asin et ejgw` de` ajsqenw'n poreu´ omai, n’est atteste´e que respectivement dans un manuscrit du Nouveau Testament, l’ajout kai` lamba´ nei th` n i“asin dans le cod. Oslo/London, the Schøyen Collection 230 (e 64) du XIIe sie`cle et ejgw` de` ajsqenw'n poreu´ omai dans le cod. Leicester, Leicestershire Record Office 6 D 32/1 du XVe sie`cle, l’un des repre´sentants de la famille 13.31 Compte tenu de l’aˆge des manuscrits cite´s et de la disparite´ de leurs attestations, il est vraisemblable que l’hymne de Kumulas soit la source des ajouts dans le texte biblique et non l’inverse. Dans un autre sticheron du premier mode plagal pour le meˆme dimanche du paralytique, jAne´ bh oJ Ij hsou'", Kumulas cite de nouveau abondamment Jean 5.1–7, relie´ en fin de texte `a Luc 8.43: 27 H. J. W. Tillyard, The Hymns of the Pentecostarium, Monumenta Musicae Byzantinae, Transcripta 7 (Copenhague, 1960), 25. 28 G. Wolfram, ´ed., Sticherarium antiquum vindobonense, Monumenta Musicae Byzantinae 10 (Vienne, 1987), fol. 238v (anonyme); E. Follieri et O. Strunk, ´eds., Triodium athoum: Codex Monasterii Vatopedi 1488 phototypice depictus, Monumenta Musicae Byzantinae 9 (Copenhague, 1975), fol. 140r (anonyme). 29 Tillyard, Hymns of the Pentecostarium, 36. 30 ¨nfzigta¨giTraduction allemande de ce sticheron par K. Kirchhoff, Osterjubel der Ostkirche: Hymnen aus der fu gen Osterfeier der byzantinischen Kirche (Mu ¨ nster, 1940), 1:160. 31 Cf. K. Aland, Kurzgefaßte Liste der griechischen Handschriften des Neuen Testaments, Arbeiten zur neutestamentlichen Textforschung 1, 2e ´ed. (Berlin–New York, 1994), 50.
CHRISTIAN HANNICK
213
A j ne´ bh oJ Ij hsou'" eij" IJ eroso´ luma * ejpi` th` n Probatikh` n kolumbh´ qran, * th` n legome´ nhn kata` Ij oudai´ou" Bhqesda´ n, * pe´ nte stoa` " e“cousan⭈ * ejn tau´ tai" ga` r kate´ keito * plh'qo" tw'n ajsqenou´ ntwn⭈ * a“ggelo" ga` r tou' Qeou' * kata` kairo` n ejpifoitw'n, * dieta´ ratten aujth` n, * kai` rJw'sin ejcari´zeto * toi'" prosiou'sin ejn pi´stei. * Kai` ijdw` n oJ Ku´ rio" * croniou'nta a“nqrwpon, * le´ gei pro` " aujto´ n⭈ * Qe´ lei" uJgih` " gene´ sqai… * oJ ajsqenw'n ajpekri´nato⭈ * Ku´ rie, * a“nqrwpon oujk e“cw, * i”na, o”tan taracqh' to` u”dwr, * ba´ lh me eij" th` n kolumbh´ qran⭈ * Ij atroi'" kathna´ lwsa * to` n a”panta´ mou bi´on * kai` ejle´ ou" tucei'n oujk hjxiw´ qhn. * jAll∆ oJ ijatro` " tw'n yucw'n kai` tw'n swma´ twn * le´ gei pro` " aujto` n⭈ * «Aro´ n sou to` n kra´ bbaton kai` peripa´ tei, * khru´ ttwn mou th` n du´ namin, * kai` to` me´ ga e“leo" ejn toi'" pe´ rasin (PeR 139).
La typologie ou pre´figuration du Nouveau dans l’Ancien Testament est une me´thode the´ologique tre`s courante `a l’e´poque patristique, de`s les pe`res apostoliques comme Justin et, par la`, dans l’hymnographie byzantine. Un proce´de´ litte´raire tre`s fre´quent pour introduire des ´ele´ments typologiques consiste dans la question rhe´torique, Ti´ se kale´ swmen… Le plus fameux et le plus connu de ces textes est un theotokion re´cite´ chaque jour `a l’heure de Prime ainsi qu’aux Grandes Heures du Vendredi Saint, de la Nativite´ et des The´ophanies. Nulle part le mode musical n’est indique´, ce qui souligne l’antiquite´ de ce texte, `a ce qu’il semble, non soumis au syste`me de l’Oktoechos: Ti´ se kale´ swmen, w« kecaritwme´ nh… * oujrano´ n * ajne´ teila" to` n h”lion th'" dikaiosu´ nh" * para´ deison * o”ti ejbla´ sthsa" to` a“nqo" th'" ajfqarsi´a" * parqe´ non * o”ti e“meina" a“fqoro" * aJgnh` n mhte´ ra * o”ti e“sce" sai'" aJgi´ai" ajgka´ lai" uiJo´ n, * to` n pa´ ntwn qeo´ n * aujto` n iJke´ teue swqh'nai ta` " yuca` " hJmw'n (MR II 632). Les rapports typologiques mentionne´s dans ce theotokion sont bien connus et ne soule`vent aucun proble`me d’interpre´tation, car ils ressortissent au re´pertoire habituel des ´epithe`tes mariaux, dont le me´tropolite Sophronios Eustratiades a pre´sente´ un re´pertoire en 1930.32 Ce qui ne signifie aucunement que ce riche patrimoine hymnologique soit en tout point compre´hensible d’emble´e. Tout qui a la pratique des home´lies et des hymnes mariales sait combien il est difficile parfois de de´tecter le lieu biblique approprie´ et plus encore de reconstruire pas `a pas le raisonnement logique sur lequel toute typologie repose. Dans le cas du theotokion cite´, les rapports `a Malachias 3.20 (h”lio" th'" dikaiosu´ nh") et Isaı¨e 11.1 sont ´evidents. La typologie oujrano´ " - h”lio" est reprise de fac¸on presque identique dans un canon du troisie`me ton `a la Me`re de Dieu, anonyme, quatrie`me ode: “Allo" oujrano` " * qeoto´ ke w“fqh" to` n h”lion * th'" dikaiosu´ nh" ajfra´ stw" * ejpi` th'" gh'", aJgnh´ , ajnatei´lasa (ETh 281). Prenons un autre exemple du type Ti´ se kale´ swmen, une se´rie de stichera prosomoia d’apre`s la strophe mode`le Ti´ uJma'" kale´ swmen du quatrie`me mode plagal, applique´ aux anges (PaR 644). Ces stichera sont chante´s le 12 novembre (MR II 122) `a la feˆte de s. Nil d’Ancyre, auteur asce´tique ayant ve´cu vers l’an 400, connu pour ses lettres spirituelles (CPG 6043). L’auteur anonyme des stichera en faisant allusion `a l’homonymie entre le nom de personne et le nom du fleuve d’Egypte compare Nil aux fleuves du paradis (Gen 2.10): Ti´ se nu'n kale´ swmen, a”gie… potamo` n th'" nohth'" ejkporeuo´ menon E j de´ m. Cette typologie est compre´hensible si l’on se souvient que les quatre fleuves de l’Eden—Pisˇon, Gihon, 32 S. Eustratiades, JH Qeoto´ ko" ejn th' uJmnografi´a (Paris, 1930). Ce re´pertoire n’a pas rendu inutile les commentaires de T. Toscani et I. Cozza-Luzi, De Immaculata Deiparae conceptione hymnologia graecorum ex editis et ` kai` JUmnolomanuscriptis codicibus cryptoferratensibus (Rome, 1862). Voir aussi N. B. Tomadakes, “ Agiologika J gika´ BÚ Qew´ numa kai` iJerw´ numa, h“toi hJ JEllhnikh` oJrologi´a eij" th` n JEllhnikh` n cristianikh` n glw'ssan,” jEp. JEt.Buz.Sp. 46 (1983–86): 119–56.
214
L’HYMNOGRAPHIE BYZANTINE
Tigre et Euphrate—symbolisent les quatre ´evange´listes; ce rapport, connu de`s l’e´poque patristique, est expose´, par exemple, dans un colophon d’e´vange´liaire arme´nien du XIIe sie`cle dont j’ai traite´ nague`re.33 La diffe´rence entre typologie et me´taphore est sensible: dans le meˆme groupe de stichera prosomoia, Nil d’Ancyre est compare´ `a un crate`re (krath´ r) de la sagesse et de la connaissance ou `a un canal (ojceto´ ") des dons prodigue´s par Dieu, deux concepts qui, ou bien n’appartiennent pas au langage biblique ou bien y sont repris dans une autre acception. ` il est fait mention explicitement des concepts tu´ po", Nombreuses sont les hymnes ou tupo´ w et qui permettent ainsi de comprendre mieux le syste`me typologique applique´ `a l’hymnographie. Quelques exemples suffiront, sans vouloir exposer cette question de fac¸on exhaustive. Bien plus explicite que le theotokion Ti´ se kale´ swmen, dont il vient d’eˆtre question, s’ave`re un staurotheotokion du deuxie`me mode plagal pour le mercredi de la quatrie`me semaine du careˆme (TR 390), une composition de l’hymnographe Jo` le poe`te explique la typologie du buisson ardent (ba´ to"Ú Ex 3.2) applique´e `a la seph, ou Me`re de Dieu. Le rapport typologique consiste dans le fait que Marie en concevant le Christ de´signe´ par feu (pu'r) est demeure´e inconsume´e (ajfle´ ktw"), tout comme le buisson ardent dans lequel Dieu se manifesta `a Moı¨se dans Exode 3.2. Le syllogisme devient complet si l’on se re´fe`re `a He´breux 12.29 kai` ga` r oJ qeo` " hJmw'n pu'r katanali´skon repris `a Deute´ronome 4.24: E j tu´ pou ba´ to" * to` musth´ rion * tou' uJpe` r nou'n sou to´ kou * ko´ rh pana´ mwme * wJ" ejkei´nh ga` r die´ meina" * ajfle´ ktw" pu'r tekou'sa * Cristo` n swth'ra * to` n ajnuywqe´ nta ejn staurv' * o’n dusw´ pei rJusqh'nai me * puro` " tou' aijwni´ou * . . . (TR 390). Cette typologie de Marie et du buisson ardent en rapport avec la conception du Christ est de´ja` de´veloppe´e chez Gre´goire de Nysse, Jean Chrysostome et The´odoret de Cyrrhe.34 Une autre image typologique de la Me`re de Dieu nous est fournie par un sticheron prosomoion du deuxie`me mode plagal pour les laudes, qui fait suite `a un tetraodion de l’avant-feˆte de Noe¨l, une composition de Bartholomaios de Grottaferrata.35 Ce sticheron fait partie d’une se´rie alphabe´tique de 24 stichera pour l’avant-feˆte de Noe¨l, attribue´e `a Romanos le Me´lode (MR II 564), mais les textes pre´sente´s par Vitali dans son Anthologion de 1738 (Anth I 386 sq.) diffe`rent souvent du texte de MR. On a donc `a faire `a une ` plusieurs stichera ont ´ete´ remplace´s et interpole´s, comme c’est composition mixte ou plusieurs fois le cas dans les se´ries d’hymnes des avant- et apre`s-feˆtes, comme, par exemple, dans les textes construits sur l’automelon Oi«ko" tou' E j fraqa'.36 Marie est compare´e ici `a l’urne (sta´ mno") qu’Aaron remplit de manne et de´pose dans l’arche d’alliance (Ex 16.33): Tu´ po" kibwtou' * proetu´ pou th` n parqe´ non * te´ xasan qeo´ n * iJlasth´ rion tou' ko´ smou * ejn h» ga` r h«n hJ sta´ mno" * hJ to` ma´ nna kate´ cousa * di∆ h»" Ij srah` l ejpodhgei'to * dio` kai` hJmei'" th' qeoto´ kv * u”mnon ei“pwmen * eujloghme´ nh pe´ lei" su´ * parqe´ ne panu´ mnhte (Anth I 405). Cette description de l’urne contenant la manne est reprise dans He´breux 9.4 dans un C. Hannick, “Bibelexegese in armenischen Handschriften-Kolophonen,” dans Armenia and the Bible: Papers Presented to the International Symposium Held at Heidelberg, July 16–19, 1990, ´ed. C. Burchard (Atlanta, Ga., 1993), 79–86. 34 Quelques textes dans J. C. Suicerus, Thesaurus ecclesiasticus e patribus graecis (Amsterdam, 1682), 1:672 (s.v. “ba´ to"”). 35 Publie´ avec traduction italienne par G. Giovanelli, Gli inni sacri di s. Bartolomeo juniore confondatore e IV egumeno di Grottaferrata, Innografi italo-greci 3 (Grottaferrata, 1955), 115–17. 36 Voir `a ce sujet mon ´etude `a paraıˆtre dans le second volume du Me´morial Ivan Dujcˇev (Sofia). 33
CHRISTIAN HANNICK
215
contexte christologique, ce qui conclut le syllogisme, base indispensable du raisonnement typologique. Dans sa concision insurpassable, un sticheron prosomoion baˆti sur la strophe mode`le Oi«ko" tou' E j fraqa' du second mode, sticheron du genre syntomon attribue´ `a Kyprianos, dont les dates restent incertaines,37 pre´sente dans l’octave de la feˆte de la Nativite´ de la Me`re de Dieu le 11 septembre un autre aspect de la typologie mariale, la verge d’Aaron qui bourgeonne sans ˆetre arrose´e (Num 17.23): Tu´ po" qeopreph´ " * tou' qei´ou toketou' sou * qeo´ fron “Anna w“fqh * tou' jAarw` n hJ rJa´ bdo" * ajni´kmw" ejkblasth´ sasa (MR I 133). Cf. la me´lodie de la strophe mode`le d’apre`s le cod. Ambros. A139 sup. fol. 73r, datant de 1341 (Fig. 1). Le meˆme passage de He´breux 9.4 (hJ rJa´ bdo" jAarw` n hJ blasth´ sasa) situe cet ´episode dans la dimension christologique. La porte´e messianique de l’oracle de Balaam, fils de Be´or, dans Nombres 24.17, “Un ` dans Matthieu astre se le`vera de Jacob,” a ´ete´ comprise de`s le Nouveau Testament, ou 2.2 les termes principaux a“stron et le verbe ajnate´ llw sont change´s en ajsth´ r et ajnatolh´ .38 Les pe`res de l’e´glise `a partir de Justin ont fait large usage de ce verset de sorte qu’il n’est pas ´etonnant que dans les textes hymnographiques se rapportant `a la Nativite´ du Christ l’astre de Jacob soit souvent mentionne´. Le rapport typologique est ´enonce´ de fac¸on explicite dans un canon de l’avant-feˆte: Ma´ ntew" crhsmolo´ gou * problh´ mata * Balaa` m nu'n plhrou'ntai * ajne´ teile ga` r a“stron * ejx Ij akw` b kai` wJdh´ ghse * pro` " to` n th'" do´ xh" h”lion * dw'ra komi´zonta" * ma´ gou" ejk Persi´do" a“nakta" (MR II 619). Le terme technique pro´ blhma au sens de “e´nigme” rappelle Orige`ne dans une gradation comprenant ai“nigma, skoteinoi` lo´ goi, parabolh´ , pro´ blhma et se re´fe´rant `a Proverbes 1.6 et Psaumes 77.2.39 Depuis les ´etudes, vieillies, de Wilhelm Christ et de Hubert Grimme sur la composition des hymnes byzantines, mises en rapport avec la poe´sie religieuse syriaque,40 on n’a pratiquement pas ´etudie´ les ´ele´ments rhe´toriques de l’hymnographie byzantine, `a part dans les textes attribue´s au prince des me´lodes, Romanos.41 Aussi serait-il pre´mature´ de vouloir pre´senter ici plus que des remarques, somme toute des notes de lectures. Parmi les textes consacre´s `a la re´surrection, les stichera anatolika (dont l’appellation, comme le proposait de´ja` Christ,42 se rattache plus `a Je´rusalem qu’a` un ´eventuel hymno37 Cf. M. Cappelli Arata, “Some Notes in Cyprian the Hymnographer,” Studies in Eastern Chant 5 (1990): 123–36. 38 Cf. La Bible d’Alexandrie, vol. 4, Les Nombres, trad. et ´ed. G. Dorival (Paris, 1994), 453. 39 Contra Celsum III 45, PG 11:980 A; Origenes Werke, vol. 1, ´ed. P. Koetschau, Die griechischen christlichen Schriftsteller 2 (Leipzig, 1899), 242.2; Orige`ne. Contre Celse, vol. 2, ´ed. M. Borret, SC 136 (Paris, 1968), 108. 40 W. Christ et M. Paranikas, Anthologia graeca carminum christianorum (Leipzig, 1871), surtout “De stropharum compositione,” 101 sq.; H. Grimme, Der Strophenbau in den Gedichten Ephraems des Syrers (Fribourg, 1893), surtout “Anhang u ¨ ber den Zusammenhang zwischen syrischer und byzantinischer Hymnenform,” 77 sq. Cette question a ´ete´ reprise re´cemment, et traite´e de fac¸on convaincante, par W. L. Petersen, “The Dependence of Romanos the Melodist upon the Syriac Ephrem,” Vigiliae Christianae 39 (1985): 171–87. 41 Cf., par exemple, H. Hunger, “Romano il Melode—poeta, predicatore, retore—ed il suo pubblico,” Ro¨mische Historische Mitteilungen 25 (1983): 305–32 (repr. dans H. Hunger, Epidosis: Gesammelte Schriften zur byzantinischen Geistes- und Kulturgeschichte [Mu ¨ nchen, 1989]); voir aussi idem, “Rhetorik als politischer und gesellschaftlicher Faktor in Byzanz,” dans Rhetorik zwischen den Wissenschaften, ´ed. G. Ueding (Tu ¨ bingen, 1991), 103–7. 42 Christ et Paranikas, Anthologia graeca, 41; cf. ´egalement C. Hannick, “Le texte de l’Oktoechos,” dans Dimanche: Office selon les huit tons, La prie`re des ´eglises de rite byzantin 3 (Chevetogne, 1972), 44 sq.
216
L’HYMNOGRAPHIE BYZANTINE
graphe Anatolios), te´moignent d’une diction audacieuse et pleine de force, avec phrases courtes, asynde´tiques, avec re´pe´tition des meˆmes constructions, comme, par exemple, dans un sticheron du second mode dans lequel la gradation ternaire, ejn tv' staurv' sou ejn th' tafh' sou - ejn th' ejge´ rsei sou, est renforce´e par les formes verbales semblables, kath´ rghsa" - ejne´ krwsa" - ejfw´ tisa"Ú E j n tv' staurv' sou kath´ rghsa" * th` n tou' xu´ lou kata´ ran * ejn th' tafh' sou ejne´ krwsa" * tou' qana´ tou to` kra´ to" * ejn de` th' ejge´ rsei sou * ejfw´ tisa" to` ge´ no" tw'n ajnqrw´ pwn * dia` tou'to soi bow'men * eujerge´ ta Criste´ * oJ qeo` " hJmw'n do´ xa soi (PaR 101). Quand il est question de rhe´torique en rapport avec la poe´sie religieuse, il faut au moins s’attarder quelques instants sur la forme de cette poe´sie religieuse, meˆme si c’est un sujet bien connu. Dans la plus grande majorite´ des textes, l’hymnographie byzantine n’a pas de forme me´trique qui soit se´parable de la musique. Les quelques cas de canons 43 festifs de Jean Damasce`ne en trime`tres ¨ambiques ı sont connus et ce sont pre´cise´ment ˆt artificiel, qui ont ´ete´ commente´s par The´odore Prodromos.44 ces textes, d’aspect pluto Les onze exaposteilaria attribue´s `a Constantin Porphyroge´nne`te, en vers politiques de quinze syllabes, ne nous retiendront gue`re, sauf pour souligner que leur traitement musiˆ ge si connu qu’il n’y a pratiquement aucune source manuscrite qui cal ´etait au Moyen-A les transmette, `a une ou deux exceptions pre`s.45 Moins connu s’ave`re par contre un genre poe´tique cultive´ dans les derniers sie`cles byzantins, les megalynaria alphabe´tiques consistant en quatre ou cinq vers heptasyllabiques. Eustratiades avait e´dite´ l’un de ces textes voici 60 ans, jAkata´ flekte ba´ te, qui porte le nom du patriarche Germanos de Constantinople et qu’il avait trouve´ atteste´ dans cinq manuscrits athonites (Lavra I 190, I 79, I 185, L 165, et Pantokrator 214).46 En collaborant avec Herbert Hunger au catalogue des manuscrits grecs de Vienne, j’ai retrouve´ ce meˆme texte, cette fois neume´, dans un codex musical du type des jAkolouqi´ai, le cod. Vind. theol. gr. 185, fol. 230v, et ai propose´,47 `a l’encontre du me´tropolite Sophronios, de l’attribuer au patriarche Germanos II de Constantinople (1223–40), connu comme hymnographe, mais dont l’he´ritage litte´raire n’est pas encore assure´. La structure musicale confirme la division presque constante en vers heptasyllabiques48 et requiert une analyse qui de´passe les limites de cette pre´sentation (Fig. 2). Aussi ne donnerai-je ici que la premie`re strophe de ce long poe`me d’apre`s le manuscrit de Vienne copie´ vers l’anne´e 1400: jAkata´ flekte ba´ te * ajlato´ mhton o“ro" po´ li" tou' mega´ lou * basile´ w" cai're ko´ rh 43 N. Somma, “Considerazioni sui canoni giambici bizantini delle festivita` del Natale, Teofania e Pentecoste,” BollGrott 21 (1967): 35–40; T. Xydes, Buzantinh` uJmnografi´a (Athe`nes, 1978), 68–86 (ijambikoi` kano´ ne" tou' Damaskhnou'). 44 Cf. note 3. 45 G. Wolfram, “Ein neumiertes Exaposteilarion anastasimon Konstantins VII.,” dans BUZANTIOS (voir note 14), 333–38. 46 S. Eustratiades, “ JH ajkolouqi´a tou' mega´ lou sabba´ tou kai` ta` megaluna´ ria tou' ejpitafi´ou,” Ne´ a Siw´ n 32 (1937): 273–75. 47 ¨ sterreichischen NationalH. Hunger, O. Kresten, et C. Hannick, Katalog der griechischen Handschriften der O bibliothek, vol. 3.2, Codices theologici 101–200 (Vienne, 1984), 372. 48 Comme l’a souligne´ J. B. Pitra, Hymnographie de l’e´glise grecque (Rome, 1867), 15, le canon du me´lode Gabriel, ´egalement auteur de kontakia, pour la feˆte de l’icone de la Me`re de Dieu Portaı¨tissa, dans le quatrie`me mode plagal, est baˆti dans la sixie`me ode sur un sche´ma heptasyllabique. Pitra y reconnaıˆt “le vers familier des Syriens” (p. 17).
Oi
-
kow
h
po -
tvn
pro - fh
eu -
tre
en
v
tou
liw
-
Eu -
h
a -
- tvn
h
pi
to
-
gi
do
son
yei
fra - ya
ton
-
on
-
a
-
ja
oi
-
kon
tik - te
-
1 Ambros. A 139 sup., fol. 73r. Sticheron automelon du 2e mode pour l’avant-fête de la Nativité du Christ.
tai
A - ka - ta
a - la - to
po
ba 2
-
liw
-
si
-
-
tou
-
flek
-
te
ba
mh - ton
me
le
o
-
-
-
ga
vw
Alagma du 3e mode du patriarche Germanos de Constantinople.
te
-
-
xai - re
row
lou
ko
-
rh
Crypt. E. g. II, fol. 309v
O - ti
eiw
ba
ptais
kai
e
tou
-
-
ne
bu
-
no
-
thw e
-
je
vw
-
-
ya
-
ma
-
tvn
a
pa
-
ghn
eiw
a
-
fou
h ra
yh
-
-
lou
I
-
tou fu
-
las
-
-
per
i
-
gnv
-
xei
-
mv
-
-
aw
ri
-
se
- vw
now xris
me
v
-
nan
tou
fh
Hirmos de la 6e ode, 1er mode, de l’acolouthie pour saint Markianos.
mai
lun
mou
Sab. 83, fol. 12r 3
-
po
mi
shw
-
row
-
te
CHRISTIAN HANNICK
217
Si nous nous tournons maintenant vers la masse des hymnes byzantines, le kontakion ˆt que les crite`res sur lesquels reexclus, comme je l’ai de´ja` signale´, on s’aperc¸oit aussito pose la distinction entre prose et poe´sie font ici de´faut. Comment caracte´riser, par exemple, la tessiture poe´tique des onze hymnes de la re´surrection attribue´es `a l’empereur Le´on VI?49 Solomon Hadjisolomos y a consacre´ nague`re une ´etude musicologique soulignant le contexte modal de ces hymnes.50 La division en kola par lui propose´e, par exemple, pour la premie`re hymne dans le premier ton, Eij" to` o“ro" toi'" maqhtai'" (PaR 706) avec 14 kola,51 qui d’ailleurs n’a rien de re´gulier, est parfois contredite par une partie de la tradition manuscrite,52 de sorte qu’ici aussi, seule la tonh´ , ou harmonie intrinse`que entre texte et musique au-dela` des caracte´ristiques formelles, semble confe´rer `a ces hymnes leur aspect hymnique. Il ne faut par ailleurs pas oublier que quelques hymnes, et parmi les plus anciennes, sont reprises inte´gralement `a un texte homile´tique, comme, par exemple, le kathisma-troparion du quatrie`me ton OiJ ma´ rture´ " sou ku´ rie (PaR 348; MR I 212), ou bien l’hirmos du premier ton pour la Nativite´ du Christ, Cristo` " genna'tai (MR II 662).53 La structure de la prose rythme´e de l’homile´tique, `a la meilleure ´epoque, re´git les grandes lignes de l’hymnographie et, rehausse´e par la tessiture musicale, fait de ces textes, d’abord au plan atomise´, puis au plan de l’ensemble correspondant au genre hymnographique donne´, parfois des petits bijoux qui aligne´s l’un apre`s l’autre sont comparables aux meilleures re´ussites de l’orfe`vrerie. L’image est ose´e, elle n’est pas neuve, le terme a”lusi" crush', “chaıˆne d’or,” est connu des auteurs byzantins eux-meˆmes. Pour ce qui regarde la structure, un exemple montrera d’abord la pre´cision de la construction d’un tel texte, par ailleurs peu connu aujourd’hui, un hirmos du premier ´ glise, compose´ pour un canon, qui n’est ton, attribue´ `a Nikephoros, diacre de la Grande E ´ glise et feˆte´ le pas conserve´, pour saint Markianos, preˆtre et oikonomos de la Grande E 54 10 janvier (Fig. 3). Un tel paralle´lisme des membres, souligne´ tant par la me´lodie que par la rime, se retrouve dans beaucoup d’hymnes, kontakia inclus. J’ai analyse´ re´cemment un tel exemple, un sticheron pour la feˆte des The´ophanies, attribue´ `a Ioannes Monachos, To` n fwtismo` n hJmw'n.55 Pre´senter des conclusions de ce rapport serait artificiel. Il n’y avait rien `a de´couvrir, 49 Cf. H. J. W. Tillyard, “ JEwqina` ajnasta´ simaÚ The Morning Hymns of the Emperor Leo,” BSA 30 (1928– 29): 86–108; ibid., 31 (1929–30): 115–47. Transcription par H. J. W. Tillyard, The Hymns of the Octoechus, 2e pt., Monumenta Musicae Byzantinae, Transcripta 5 (Copenhague, 1949), 59–84. 50 S. J. Hadjisolomos, The Modal Structure of the 11 Eothina Anastassima Ascribed to the Emperor Leo († 912): A Musicological Study (Nicosia, 1986). 51 Hadjisolomos, Modal Structure, 80. 52 Cf. Tillyard, Hymns of the Octoechus, 61. 53 Sur ce domaine, encore mal ´etudie´, cf., entre autres, B. Pseutonkas, “AiJ oJmili´ai tou' mega´ lou Basilei´ou wJ" phgh` eij" th` n uJmnografi´an,” Klhronomi´a 6 (1974): 261–76; et idem, AiJ peri` staurou' kai` pa´ qou" tou' kuri´ou oJmili´ai ajnatolikw'n pate´ rwn kai` suggrafe´ wn ajpo` tou' 2ou me´ cri kai` tou' 4ou aijw'no", Ana j ´ lekta Blata´ dwn 53 (Thessalonike, 1991), 68–71. 54 Structure musicale d’apre`s Crypt. E. g. II fol. 309v, a. D. 1281 et Hieros. Sab. 83, fol. 12r, du de´but du XIIe sie`cle: L. Tardo, ´ed., Hirmologium cryptense, Monumenta Musicae Byzantinae 3 (Rome, 1951); et J. Raasted, ´ed., Hirmologium sabbaiticum, Monumenta Musicae Byzantinae 8 (Copenhague, 1968). 55 C. Hannick, “Early Slavic Liturgical Hymns in Musicological Context,” Ricerche slavistiche 41 (1994): 19–22.
218
L’HYMNOGRAPHIE BYZANTINE
sinon `a rede´couvrir. Les auteurs byzantins cite´s, tout inconnus qu’ils restent trop souvent au plan chronologique, ont ´ete´ resitue´s dans le contexte esthe´tique et artistique qui e´tait le leur, pour autant que l’on puisse le reconstruire. Nouvelle appre´ciation des hymnographes byzantins? Ce serait te´me´raire et fanfaron, car beaucoup n’a ´ete´ qu’effleure´. J’aurais atteint mon but si j’avais pu montrer qu’une compre´hension du riche he´ritage hymnographique de l’e´poque byzantine passe par la voie du ve´hicule musical de cette poe´sie religieuse, pour autant que l’on puisse en saisir les fondements, parfois trop peu clairs encore, avant les re´formes des temps modernes, et en tenant compte du fait qu’a` la pe´riode byzantine, les the´ories musicales elles-meˆmes n’ont pas ´ete´ ´epargne´es par le ˆ t des transformations, comme le de´montrent trop bien les traite´s de´ja` publie´s dans le gou Corpus scriptorum de re musica. Universita¨t Wu ¨ rzburg
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Narratives of the Fall: Structure and Meaning in the Genesis Frieze at Hagia Sophia, Trebizond ANTONY EASTMOND
he church of Hagia Sophia stands 2 km to the west of the walled citadel of Trebizond on an outcrop overlooking the Black Sea (Fig. 1).1 It is the most impressive surviving monument of the Grand Komnenoi emperors of Trebizond, and is dated by a donor portrait, now lost, to the reign of Manuel I Grand Komnenos (1238–63).2 The church contains the finest Byzantine wall paintings of the thirteenth century, but it also has many peculiarities that have never been fully explained, such as the unusual plinth on which the church stands, the three porches that precede all the entrances to the church, its architectural layout, its sculptural decoration, and many aspects of the wall-painting program. This article examines one of the peculiarities—the sculptural frieze of scenes from Genesis that runs across the south porch of the church.
T
The Genesis frieze, which appears in a series of carved blocks on the south porch of the church, is one of the more unusual artistic products of the Byzantine world (Fig. 2). Despite this, it has received relatively little attention from scholars. It was first published in an engraving by Charles Texier in 1864 as part of his general description of the church. He was relying on his memories of visiting Trebizond some thirty years earlier, and the image is correspondingly vague. His representation includes many errors, and he entirely misreads the inscriptions.3 When George Finlay, the traveler and early historian of the empire, visited the church in 1850, he noted in his journal only that “there is a very strange frieze with figures in high relief on the southern wall and an inscription.” 4
1 The principal publication of the church is that produced after its restoration by the Russell Trust in 1957–64: D. Talbot Rice, ed., The Church of Haghia Sophia at Trebizond (Edinburgh, 1968). 2 The portrait was described by G. Finlay (ed. J. M. Hussey, The Journals and Letters of George Finlay, vol. 1, The Journals [Camberley, Surrey, 1995], 301); and a copy of it was made by G. Gagarine (see Sobranie vizantijskikh, gruzinskikh i drevnerusskikh ornamentov i pamjatnikov arkhitektury [Recueil d’ornements et d’architecture byzantines, ge´orgiens et russes] [St. Petersburg, 1897], pl. 25). The image was missing by 1866. 3 C. Texier and R. Pullan, Byzantine Architecture Illustrated by a Series of the Earliest Christian Edifices in the East (London, 1864), 199–201 and pl. LXa–LXV. G. Millet, “Les monaste`res et les ´eglises de Tre´bizonde,” BCH 19 (1895): 419, notes the inadequacies and inaccuracies of Texier’s drawings. 4 Ed. Hussey, Journals of George Finlay, 300.
220
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
The first (and only) detailed account of the frieze was produced by Mikhail Alpatov in 1927, and this was very heavily relied on by David Talbot Rice in his 1968 description.5 This article reexamines every aspect of the frieze, including its composition, liturgical importance, and function within the church, and makes use of the accompanying inscriptions, which were generally ignored by previous writers. No complete photographic record of the frieze has ever been published, and I am very grateful to David Winfield for permission to publish the photographs taken by him from scaffolding during the Russell Trust restoration of the church. I. DESCRIPTION The Genesis frieze is approximately 8 m long and 75 cm high (Figs. 3–6). The scenes are carved on twenty blocks of differing sizes, and above them run two inscriptions incised on thirteen blocks of stone that interlink with those below. The inscriptions and the frieze are divided in two by the architecture of the porch—the pointed center arch, which rises above the bottom line of the frieze, and a small quatrefoil window, which cuts into it from above. Between these there is a small carved shell conch, displayed pointing upward. Around the frieze are a series of other carved figures and inlaid designs that are not linked directly to the frieze and are not discussed here. The frieze runs from right to left across the porch and contains seven scenes. Many of the details of all the figures are now badly weathered, but enough remains to reconstruct the sequence with some certainty: 1. The first scene (Figs. 4, 6), at the right end, shows the creation of Eve. The hand of God appears from an aureole in the top right corner of the main block and conjures Eve with a gesture of blessing from the body of Adam, who is shown reclining on his left elbow amongst the trees and plants of Paradise. The figure of Eve has now almost completely disappeared, but her approximate outline can be made out rising from the ribs of Adam. This scene is the only one to have been carved on a horizontal block; it has been enlarged by the addition of four small blocks that extend the foliage background to fill gaps above and to the right of the main stone. 2. The next scene shows the temptation of Eve. Eve, facing to the right, reaches up with her left hand to pick the forbidden fruit from the tree of knowledge. The serpent has entwined itself around the tree, and looks closely into the eyes of Eve. Eve has long hair and wears a robe that descends to her feet in a series of sketchily delineated folds. 3. The third scene, now in very bad condition, is carved over two blocks and shows Adam and Eve facing each other as Eve hands over the fruit for Adam to eat. Again the two figures are shown in long robes, but all the details of Eve are now lost. In the background of all the scenes on the right half of the porch, the trees of Paradise burst forth with luxuriant foliage; these trees continue alone across the center of the porch’s arch. The base line of the trees on these blocks rises successively toward the center of the arch, showing that there was not enough room for full figures in this section. 5 M. Alpatov, “Les reliefs de la Sainte-Sophie de Tre´bizonde,” Byzantion 4 (1927–28): 407–18; Talbot Rice, Haghia Sophia, 47–49. More recently, L. Safran, “The Genesis Frieze at Hagia Sophia at Trebizond,” BSCAbstr 20 (1994): 29, has examined the liturgical nature of the frieze, noting many of the same points that I do here. Unfortunately, Safran has only published her work as a brief abstract; I have not had access to her whole paper.
ANTONY EASTMOND
221
4. The frieze starts again on the left side of the arch with the closed gate of Eden (Figs. 3, 5). The door is depicted as a splendid architectural device with an elaborately carved frame surmounted by a magnificent shell-headed niche. Within it, an angel with extended wings and a spear in each outstretched hand blocks and guards the entrance. The size of the door decoration is such that the angel is half the size of all the other figures in the frieze. 5. Following this, and carved on three blocks, is the expulsion from Eden. An angel is shown driving the errant humans from Paradise. The angel leans forward on his right foot, with his left heel raised and his right hand out toward the humans. He is the most dynamic figure in the frieze, but this sense of movement, designed to emphasize the act of expulsion, is undermined by the very static poses of the two humans. Adam and Eve face each other and thus negate further any consistent sense of narrative direction. They are now shown naked; they stand opposite each other with their heads bowed and their hands raised to their faces in gestures of horror and despair at what they have done. Eve’s body is curiously twisted, with her right leg in front of her left (perhaps in a display of modesty or shame) whereas Adam stands in a more straightforward pose. The angel’s left wing is cut off by the right edge of the block, producing an awkward link with the scene to the right, which again suggests that the overall visual coherence of the frieze was not carefully established in advance. 6. The sixth scene shows Adam and Eve lamenting again. Now the pair sit facing each other on two low mounds, but otherwise their poses are much the same as in the previous scene, with their hands raised to their faces and their heads bowed lower. 7. The final scene, which has been constructed much more carefully in order to fit into the side of the porch’s arch, is again badly eroded, and is the hardest to identify. In 1895, Gabriel Millet proposed that it showed the birth of Cain,6 but in 1927 Alpatov suggested that it showed Cain’s murder of Abel,7 and the latter identification has now been generally accepted. One figure can be seen lying on the ground facing right, cut off from the knees down by the edge of the arch. His right hand is raised to his face, and his left is draped across his chest. He wears a long robe with a hem running horizontally along his body. Looming over him is a second figure in whose outstretched hands there appears to be a shadow of a rock or boulder. This scene, then, shows the moment of the killing. The instrument of the murder is not mentioned in Genesis, but most of the early commentators concluded that Cain had used either a rock or a staff.8 The frieze, when examined as a single composition, is uneven and inconsistent, even messy. It appears to be divided into a series of paired tableaux; and it seems that each block has been carved separately, with little consideration of the appearance of those around it. The figures vary in size considerably both between scenes and within them, especially in the scene of the expulsion from Eden. The depth of carving appears to vary considerably as well, although this is harder to determine now, given the erosion that the frieze has suffered: the nude figures on the left-hand side look to be in higher relief and Millet, “Les monaste`res et les ´eglises de Tre´bizonde,” 457. Alpatov, “Les reliefs de la Sainte-Sophie,” 412. 8 V. Aptowitzer, Kain und Abel in der Agada, den Apokryphen, der hellenistischen, christlichen und muhammedanischen Literatur (Vienna-Leipzig, 1922), 44. 6 7
222
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
more rounded than the clothed figures on the right. There is no consistent sense of movement across the frieze. Only one figure, the angel of the expulsion scene, is shown in a dynamic pose; the rest are in static postures. However, the depiction of dress is fairly consistent across the frieze, with all robes carved in a single plane, with many shallow, incised lines to indicate folds. Gestures are also repeated with little variation. But, other than these relatively minor points, there seems to have been little attempt at creating a unified whole. Each half of the frieze, nonetheless, is clearly themed and has some internal cohesion. On the right-hand side, the glory and richness of life before the Fall are indicated by the clothed figures of Adam and Eve and by the luxuriant foliage that fills every inch of the background to the center of the arch and beyond to the gate of Paradise. On the lefthand side, the aridity and pain of life outside Eden manifest themselves in the nudity of the figures, the barrenness of the setting—there is no foliage here—and the oft-repeated gesture of lamentation, that of the hand raised to the face.9 A few surviving fragments of paint were noticed on the frieze by its modern restorers, although none is visible from the ground.10 This suggests that the frieze was originally painted, which would have made the visual opposition between Eden and the exile much starker. The themes of each half of the frieze are emphasized by the two inscriptions, each incised above the carving in a clear line of capitals (majuscule) with few abbreviations or ligatures. The inscription on the left, which would be read first, comes from the Lenten Triodion, specifically the service of vespers on the eve of the Sunday before Lent (the Sunday of Forgiveness): † E j k[a´ qis]e[n] jAd[a` ]m ajpe´ [nanti tou' Par]a[dei´]sou kai` th` n ijdi´an ´ ´ gumnwsin qrhn[w']n wjdureto (“Adam sat before Paradise and, lamenting his nakedness, he wept”).11 That on the right comes from Genesis 2.8: † E j fu´ teusen oJ q[eo` ]" Para´ deison ejn ` ` ` ` ` E j dem k[a]t[a] ajnatola" kai e“qet[o] ejkei' ton a“nq[rwpo]n o’n e“plase (“And the Lord God planted a garden eastward in Eden; and there he put the man whom he had formed”). Each can be associated with the section of the frieze below it. The order of construction of the frieze is difficult to establish. The irregular size of the blocks and the fragmentary nature of the narrative suggest that they were carved before being installed on the south porch.12 However, some scenes were clearly carefully designed for their future positions: the composition of the murder of Abel, for instance, was certainly designed to match its location against the frame of the arch; and the bases of the trees of Paradise in the central portion were also angled to accommodate the rising 9 On this gesture, see H. Maguire, “The Depiction of Sorrow in Middle Byzantine Art,” DOP 31 (1977): 123–74, esp. 132–51. 10 I am very grateful to David Winfield for bringing this to my attention. Supporting evidence can be found in the polychrome appearance of other elements of the south porch, such as the inlaid panel with a star and crescent motif to the right of the central quatrefoil. Polychrome sculpture is known elsewhere in the Byzantine and eastern Christian world; see, for example, A. Grabar, Sculptures byzantines de Constantinople, vol. 1, IVe-Xe sie`cle (Paris, 1963), esp. 100–122. For a recent but controversial view on Byzantine polychromy, see C. Connor, The Color of Ivory: Polychromy on Byzantine Ivories (Princeton, N.J., 1998). 11 Triodion (Rome, 1879), 100; trans. Mother Mary and K. Ware, The Lenten Triodion (London, 1978), 169. My reading of the inscriptions is taken from their recording by Millet, “Les monaste`res et les ´eglises de Tre´bizonde,” 457 n. 2. Millet gives his reference only as Triodion, 58a, without ever quoting an edition; and this has been repeated by all subsequent writers. 12 This order of construction was proposed by Talbot Rice, Haghia Sophia, 47.
ANTONY EASTMOND
223
base line of the frieze at this point. On the other hand, four blocks were needed to enlarge and change the shape of the creation of Eve scene in order to make it fit its irregular location by adding more background foliage, which indicates that this end of the frieze had not been predetermined. It suggests that the frieze was set in place from left to right, with the additional blocks added to the creation scene to fill up the few final gaps. If this is correct and the order of installation was thus opposite to that of the narrative, it shows that the content and limits of the narrative had been determined before the frieze was carved. The inscriptions were certainly carved after the frieze was installed. The carver took into account the narrowing of the inscription blocks above the angel of the expulsion scene in the left half, and reduced the height of the attendant inscription as a result. The right-hand inscription is carved on blocks of consistent width, and all the letters are consequently taller. II. ANALYSIS AND INTERPRETATION As all commentators have noted, there are a number of oddities about the frieze, such as its right-to-left narrative movement; the decision to show the protoplasts clothed before the Fall and naked afterward; the unusually prominent use of sculpture for a Byzantine narrative frieze; and the appearance of a liturgical inscription alongside the Genesis quotation on the facade of the church. Scholars have sought explanations for all these elements, and have generally looked to “Oriental,” mostly Syriac, sources for possible models.13 This article aims to reinvestigate all the elements of the frieze, reevaluating these peculiarities. In my opinion, the principal theological meaning of the frieze and the interpretations of many of its elements can be found within the mainstream of Byzantine culture, in particular by reference to the Lenten liturgy. The frieze is not as alien as some commentators have assumed.14 Nonetheless, I also reexamine the possible eastern models that may have had some influence on the frieze, in order to show that it is possible to find convincing parallels to the frieze in sources closer to the empire of Trebizond, and that these can be tied in with the political position and cultural orientation of the empire in the mid-thirteenth century. In particular, I investigate here Georgian, Armenian, and Islamic sources, which can be shown to have had definite links to the empire, rather than vaguer Syriac models. Iconography The most noticeable aspect of the Genesis frieze at Hagia Sophia is the range of the narrative, particularly its unusual choice of scenes at the beginning and the end. The frieze starts with the creation of Eve and ends with the murder of Abel. We are not shown the creation of Adam—which would be the normal start for any account of the Fall—nor do we see the events that led up to Cain’s fratricide, namely, the brothers’ sacrifices to God. The result is a very abbreviated cycle that makes a specific point. If one looks at other Genesis cycles, it is possible to see how carefully selected the Trebizond scenes are. It should be noted that Alpatov, “Les reliefs de la Sainte-Sophie,” 414, raises Syriac models only as a possible hypothesis, whereas Talbot Rice, Haghia Sophia, 49, rephrases this to make the link more concrete. 14 C. Mango, Byzantine Architecture (Milan, 1976), 166. 13
224
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
In general, these cycles begin with the creation of Adam, followed by that of Eve, and include the sacrifice of Cain and Abel before the first murder (Figs. 7, 8). Many, of course, start even earlier, with the creation of the cosmos, and include the Fall simply as part of their long narrative cycles that cover large portions (if not all) of the text of Genesis, as in the case of the nave mosaics of the Cappella Palatina in Palermo,15 the thirteenthcentury atrium mosaics in San Marco, Venice,16 or the comprehensive cycles in Octateuch manuscripts.17 These cycles clearly have different functions from the more selective Trebizond frieze. Moreover, the narrative of Trebizond appears unusual even among shorter depictions of the Fall. Those that begin with the creation include that of Adam, and those that show the murder of Abel also show the preceding sacrifice, which explains its cause and ¨ lrich of the appearance of Cain and significance. The catalogue and analysis by Anna U Abel in art show that the murder is almost invariably portrayed with this accompaniment.18 In these symbolic depictions, it becomes a eucharistic parallel in which the nature of the sacrifice is emphasized by the scene of the brothers’ offerings to God: the shepherd Abel with his sacrificial sheep is the prototype for Christ.19 It can be seen that the Cain and Abel image at Trebizond cannot be fitted into either the narrative or the eucharistic scheme, and, combined with the first scene in the frieze, it suggests that one must look to other sources for the frieze’s meaning. The only objects on which a remotely comparable selection of Genesis scenes is shown are a number of Byzantine ivory and bone caskets, variously dated between the tenth and twelfth century.20 The iconography of these varies, but generally includes the creation of Adam and Eve, the temptation, the expulsion, Adam and Eve lamenting outside of Eden, and the murder of Abel. The caskets do not show the image of the brothers’ sacrifices. The scenes of the creation and murder generally appear together on their lids, as on the Cleveland casket (Fig. 9), with the other scenes around the sides (Figs. 10, 11). One plaque in the Muse´e des Beaux-Arts de Lyon (D 312) explicitly links the creation and 15 E. Kitzinger, I mosaici del periodo normanno in Sicilia, vol. 2, La Cappella Palatina di Palermo: I mosaici delle Navate (Palermo, 1993), 26, figs. 39, 44–46, 49, 50. 16 O. Demus, The Mosaics of San Marco in Venice (Chicago-London, 1984), 2:117–18, figs. 134–43. This cycle is, of course, based on the extensive illustrations of the Cotton Genesis (London, British Library, Otho B.VI). The work by P. H. Jolly, Made in God’s Image? Eve and Adam in the Genesis Mosaics at San Marco, Venice (Berkeley, 1997), is an unsuccessful attempt to interpret the Creation dome at San Marco in terms of 13th-century Venetian ideas. 17 See, for example, F. Ouspensky, L’Octateuque de la Bibliothe`que du Se´rail `a Constantinople (Sofia, 1907), pl. ´cole ´evangelique de Smyrne IX, figs. 23–27; D. C. Hesseling, Miniatures de l’Octateuque grec de Smyrne: Manuscrit de l’e (Leyden, 1909), figs. 1–24; C. Hahn, “The Creation of the Cosmos: Genesis Illustration in the Octateuchs,” CahArch 28 (1979): 29–40. The non-Genesis manuscripts with images of the Fall are investigated below. 18 ¨ lrich, Kain und Abel in der Kunst (Bamberg, 1981). Although this account does include the Salerno A. U ivory plaques, it excludes Byzantine ivory caskets showing Cain and Abel, which are considered below. 19 See J. B. Glenthøj, “Cain and Abel in Syriac and Greek Writers (4th-6th Centuries)” (Ph.D. diss., Oxford University, 1991), 41–47, who quotes examples such as Clement of Alexandria and Irenaeus, as well as other Early Christian and Gnostic writers. In Cosmas Indicopleustes, Topographie chre´tienne, SC 159, ed. and trans. W. Wolska-Conus (Paris, 1970), 5:75, Abel is described as a type of Christ, and in the Vatican copy (Vat. gr. 699, fol. 55r) he is depicted as a shepherd: C. Stornajolo, Le miniature della Topografia christiana di Cosma Indicopleuste: Codice Vaticano Greco 699 (Milan, 1908), 33, pl. 18. 20 A. Goldschmidt and K. Weitzmann, Die byzantinischen Elfenbeinskulpturen des X.–XIII. Jahrhunderts, vol. 1, Ka¨sten (Berlin, 1930), nos. 67–81, 86, 87; A. Cutler, “On Byzantine Boxes,” JWalt 42–43 (1984–85): 32–43.
ANTONY EASTMOND
225
the murder even more closely by depicting them on the same plaque (Fig. 12).21 These plaques show that the theological (or narrative) message that was being conveyed at Trebizond by the choice of these end points belonged to a wider tradition in Byzantium. However, the ivories do not help to establish what the message was. None of the ivory examples exactly matches the scenes at Trebizond (they all include the creation of Adam at the start), and the meaning and function of all the caskets are very uncertain.22 It is not even known whether the choice of the Genesis scenes is linked to a religious or a secular function: were the depictions of the Fall designed to be understood in association with some liturgical setting, or were they to be read simply as a narrative in the same way as were the mythological stories on other caskets of similar date, size, and design? It is possible that the frieze at Trebizond may help to elucidate these caskets, rather than the other way round. In any case, what is significant is that the narrative range used at Trebizond was already well established in the Byzantine world. It is obvious that the frieze is concerned with the Fall of Man, but the choice of scenes gives it an unusual, even unpleasant, emphasis. The frieze presents an especially misogynistic view of the Fall. The opening scene shows the creation of woman as the start of all man’s problems, with its inexorable result—sin and murder—depicted at the end. The link is reinforced by the poses of Adam and Eve being repeated in those of Cain and Abel at each end of the frieze. The association of Eve with death was a recognized theme in Christian theology from Tertullian onward.23 Ephrem the Syrian, for example, wrote that Eve “was the virgin, whose wall the Evil One pierced with her own hands and by whose fruit he made men taste death. And Eve, the mother of all the living, became the source of death for all the living”;24 and Gregory Nazianzos noted in Homily 13, “On the Canticle,” that Eve “introduced death through her sin.” 25 A similar interpretation can be found in many other writers, such as Ephrem the Greek, who also wrote of Eve as the cause of death.26 However, it is extremely rare to find this link depicted so clearly in art. The only comparable example that I have been able to find is the bronze doors of Hildesheim cathedral, commissioned by Bishop Bernward in 1015.27 On the left door, the story 21 Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, no. 70; J. Durand, ed., Byzance: L’art byzantin dans les collections publiques franc¸aises (Paris, 1992), 261–62, no. 170. 22 H. C. Evans and W. D. Wixom, eds., The Glory of Byzantium: Art and Culture of the Middle Byzantine Era, A.D. 843–1261 (New York, 1997), 222, 234, classify them as luxury, secular objects, probably designed as wedding gifts. This is based on the decoration of the Darmstadt casket, which includes a further figure with a moneybag, identified as O PLOUTOS (Wealth), who acts as a reminder of the transience of good fortune. Henry Maguire has recently expanded this idea in “Magic and Money in the Early Middle Ages,” Speculum 72 (1997): 1037–54, esp. 1047–50. He suggests that the scenes are antithetical, and that the protection of wealth is assured through penance, as modeled by Adam and Eve. While this argument provides a convincing function for the caskets, it ignores their precise narrative, which seems as much concerned with death as with penance. 23 Many of the texts are collected together in H. Graef, Mary: A History of Doctrine and Devotion, vol. 1 (London, 1963); also J. A. Phillips, Eve: The History of an Idea (New York, 1984). 24 Quoted in Graef, Mary, 61. 25 Gregory Nazianzos, PG 44:1053B. 26 Quoted in Glenthøj, “Cain and Abel,” 372. 27 R. Wesenberg, Bernwardische Plastik: Zur ottonischen Kunst unter Bischof Bernward von Hildesheim (Berlin, 1955), 65–80; W. Tronzo, “The Hildesheim Doors: An Iconographic Source and Its Implications,” ZKunstg 46 (1983): 357–66. One further example may exist in the Creation/Fall page of the 9th-century Carolingian Moutier-Grandval Bible (London, British Library, Add. 10546, fol. 5v). In the lowest register, hanging from
226
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
of the Fall is depicted in seven scenes that descend from the creation of Eve to the murder of Abel (although an image of the sacrifices of Cain and Abel is also included).28 I return to this emphasis on Eve later, but for now it is clear that, in general, the frieze on the south porch was to prepare those entering the church: it was a reminder of man’s flawed nature and his need for redemption. The south porch is a call to penitence.29 It would have been the first element of the church that worshipers saw as they entered the monastic complex through its original entrance to the southwest of the enclosure (where the modern entrance is today). The richness of the porch decoration, and especially the imperial eagle that crowns the keystone of the arch, indicates that this (rather than the west entrance) was the principal door into the church. This would bring Trapezuntine practice into line with that of Georgia, where the south entrance generally acted as the principal door into the church.30 After being reminded of their weakness, worshipers entered the naos of the church where the frescoes all around would have described the life of Christ and his sacrifice to redeem all mankind. Indeed, the first view of the north wall of the church explicitly showed this, as it was there that the Passion was played out (these frescoes are in very bad condition now, but fragments of the Crucifixion and Anastasis survive as well as figures from either the Entombment or the Lamentation).31 And as viewers then looked at the central features of the church—the dome and the apse—they would see images of the Ascension and Christ Pantokrator. The frieze therefore forms part of a progression from exterior to interior, from sin to salvation. A compositional structure similar to this narrative can be found in other images of the Fall. In an eleventh-century copy of John Klimakos’s Heavenly Ladder (Vat. gr. 394, fol. 77r), an image of the temptation and expulsion accompanies Homily 14, “On Gluttony” (Fig. 13).32 This rung of the ladder reminds the reading monk that Adam and Eve would not have sinned if they had succeeded in resisting their appetite for the forbidden fruit.33 The narrative moves from left to right, but makes a parallel division between interior and exterior: life in Paradise is depicted within the border of the image and the text, whereas the expulsion pushes Adam and Eve literally beyond the frame and into the empty margin. The margins of the page are used in exactly the same way in a the garland over the seat on which Eve sits breastfeeding Cain, is a minute picture within the miniature (ca. 8 mm2), which, Herbert Kessler has argued, depicts Cain’s murder of Abel: H. L. Kessler, “An Unnoticed Scene in the Grandval Bible,” CahArch 17 (1967): 113–20. 28 The interpretation of this door has always been regarded as inseparable from the narrative of Christ’s incarnation and life that ascends on the right door, from the Annunciation to the Noli me tangere. The location of the Hildesheim Fall cycle at the entrance to the cathedral, its use of narrative direction, and the direct parallels drawn between the Fall and the incarnation of Christ through Mary are all echoed at Trebizond. Hildesheim provides an excellent, if indirect, model of ways in which to read and interpret the narrative in the frieze at Trebizond. 29 This is the meaning of the frieze stressed by Safran, “The Genesis Frieze,” 29. 30 The majority of medieval churches in Georgia have their south entrance accentuated by means of a porch or additional decoration, and this can be traced back to 6th-century churches, such as Jvari. For examples, see A. Alpago-Novello, V. Beridze, and J. Lafontaine-Dosogne, Art and Architecture in Medieval Georgia (Louvain-la-Neuve, 1980): Jvari (p. 389), Betania (p. 296), Cinarexi (p. 311), Kvataxevi (p. 370), Manglisi (p. 376), Samtavro in Mcxeta (p. 392), and Nikorcminda (p. 411). 31 Talbot Rice, Haghia Sophia, 117–20, figs. 80, 81. 32 J. R. Martin, The Illustration of the Heavenly Ladder of John Climacus, Studies in Manuscript Illumination 5 (Princeton, N.J., 1954), 68–69, fig. 106. 33 John Klimakos, PG 88:864–72.
ANTONY EASTMOND
227
fourteenth-century copy of the Homilies of Gregory Nazianzos (Paris, Bibliothe`que Nationale, Cod. gr. 543, fol. 116v), in order to divide Eden from the world and innocence from sin (Fig. 24).34 In the luxuriously illuminated twelfth-century copy of the Homilies of James of Kokkinobaphos in the Vatican (Vat. gr. 1162, fols. 35v, 37r), the same sense of interior and exterior, abundance and aridity, is spread over a series of images (Figs. 14, 15).35 On folio 33r, in the bottom right corner of the page, the expulsion is shown, followed on folio 35v by Adam and Eve lamenting their Fall on a brown, bare hill, while Cain murders Abel below them. This is immediately followed by an image of the Garden of Eden, now empty, its door barred by three angels. In contrast to the image of life outside Eden, this page is a riot of plenty, placed on a rich, golden ground. At Trebizond, it is only by entering the church that the viewers were able to reenter Paradise. The shell design over the main arch of the south porch, which echoes the design of the gate of Paradise in the image on the frieze (albeit upside down), reinforces this idea of the porch being the entrance to Paradise on earth. The location of the image at the threshold of the church thus creates part of its meaning and forces its viewers to take part in the narrative as they prepare to enter the church. The two inscriptions allow the circumstances of the frieze’s progression to be examined with greater precision. As noted earlier, the first inscription forms part of the Lenten Triodion. Monumental liturgical inscriptions are very rare on the facades of Byzantine churches, which gives this inscription definite emphasis. It was intoned as part of the vesper service on the eve of the Sunday of Forgiveness.36 The right-hand half of the inscription, the verse from Genesis 2.8, was read out at vespers on the first Thursday of Lent.37 We also know that Adam, Eve, Cain, and Abel were all featured prominently during Lenten homilies: John Chrysostom’s Homilies on Genesis, which examine the meaning of all these events, were read out during Lent.38 These considerations indicate that the church of Hagia Sophia must have played an important role in the liturgical calendar of the empire of Trebizond at the beginning of Lent. Anthony Bryer has shown how, in the fourteenth century, the Feast of the Transfiguration was the patronal day of the church,39 and I propose that the start of Lent was also specially celebrated at Hagia Sophia. Only during this one week of the year would the full significance of the frieze and its inscriptions become apparent as the meanings of the images and texts were brought to life by the service. The emphasis demonstrated by the frieze—it is the only element of the church’s narrative decoration to be built into the architecture—shows that it must have been regarded as a central feature of the church’s (or at least the porch’s) function at the time when it was built. The uniqueness 34 G. Galavaris, The Illustrations of the Liturgical Homilies of Gregory Nazianzenus, Studies in Manuscript Illumination 6 (Princeton, N.J., 1969), fig. 462. 35 C. Stornajolo, Miniature delle Omilie di Giacomo Monaco (Cod. Vatic. gr. 1162) e dell’ evangeliario greco urbinate (Cod. Vatic. Urbin. gr. 2) (Rome, 1910), pls. 11–13. 36 In other words, this phrase was recited on the Saturday evening, the liturgical beginning of the Sunday of Forgiveness: Triodion, 100; trans. Mother Mary and Ware, The Lenten Triodion, 168. 37 Triodion, 155; trans. Mother Mary and Ware, The Lenten Triodion, 254. 38 John Chrysostom, PG 53:164. 39 A. A. M. Bryer and D. Winfield, The Byzantine Monuments and Topography of the Pontos, DOS 20 (Washington, D.C., 1985), 232–33.
228
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
of the Trebizond liturgical inscription, and of the porch itself, means that it is not possible to examine this relationship in greater detail; however, this is the only logical explanation of the presence of these various elements. Clothed/Naked The reversal of the clothed and naked states of Adam and Eve is the problem that has exercised writers most on the Genesis frieze. Alpatov proposed that the reversal could be best explained by “Oriental” parallels,40 yet it is not necessary to look to such remote sources for models: the phenomenon is not as unusual as has been presumed. For example, in two Octateuch manuscripts (Vat. gr. 746, fol. 37r; and Istanbul, Topkapı gr. 8, fol. 42v), the protoplasts are shown clothed before the Fall (Fig. 16).41 Herbert Broderick has noted the reasons for this: Genesis 3.7 states that after Adam and Eve had eaten the forbidden fruit they “perceived themselves to be naked,” which raises the question of how it had been hidden from them until that moment.42 There evolved a theology about prelapsarian garments, robes of grace or light, discussed by John Chrysostom and others.43 The apocryphal Books of Adam also mention this story very prominently: “At that hour I learned with my own eyes that I was naked of the glory with which I had been clothed.” 44 It seems, then, that the artist was following an existing, if uncommon, trend in order to emphasize the nature of the Fall and the loss of the divine glory that had surrounded Adam and Eve. A more immediate source for this reversal can also be found. This is the text of the Lenten Triodion itself. In the course of the vesper service on the eve of the Sunday of Forgiveness, Adam is no less than four times described as being naked outside Paradise: “In my wretchedness I have cast off the robe woven by God . . .”; “Naked he sat outside the garden, lamenting . . .”; “Adam sat before Paradise and, lamenting his nakedness, he wept”; “Woe is me! In my simplicity I was stripped naked, and now I am in want.” 45 The third quotation was, of course, used as the inscription to accompany the frieze. It is clear from this that the artists and craftsmen had been instructed to illustrate the liturgical, not the biblical, account of the Fall, and so needed to show the move from clothed to naked, rather than the other way around. The frieze does not represent a literal rendition of Genesis 2, as Alpatov suggested in 1927;46 rather, it is dependent on the liturgical text. If part of the service took place before this frieze, it simply would not have made sense for the biblical version of the Genesis narrative to have been followed. Alpatov, “Les reliefs de la Sainte-Sophie,” 415–18. H. Schade, “Adam und Eve,” in Lexikon der christlichen Ikonographie, ed. E. Kirschbaum (Rome-Freiburg, 1972), 1: cols. 44–51. The reversal also occurs in the frescoes of the Otranto cathedral: L. Safran, San Pietro at Otranto: Byzantine Art in South Italy (Rome, 1992), 110. Alpatov notes a further example of this phenomenon in the Novgorod Bible (a reference copied by Talbot Rice, Haghia Sophia, 48); he cites this manuscript as Moscow, Synodal Library 1147, but I have been unable to find any other reference to it, and Alpatov gives no photograph. 42 H. R. Broderick, “A Note on the Garments of Paradise,” Byzantion 55 (1985): 250–54. 43 John Chrysostom, PG 53:125. 44 G. A. Anderson and M. E. Stone, A Synopsis of the Books of Adam and Eve (Atlanta, Ga., 1994), 46. See also J. Issavardens, trans., The Uncanonical Writings of the Old Testament Found in the Armenian Manuscripts of the Library of St. Lazarus, Venice (Venice, 1901), 21, 22, 41, 44. 45 Triodion, 134; trans. Mother Mary and Ware, The Lenten Triodion, 168–69. 46 Alpatov, “Les reliefs de la Sainte-Sophie,” 412. 40 41
ANTONY EASTMOND
229
The decision to emphasize this aspect of the Fall fits in with the overall interpretation of the frieze, but it may also have been influenced by the particular local circumstances of the empire of Trebizond, in particular Armenian and Islamic ideas. The Books of Adam, for example, existed in Greek but were more common in Armenian, and it is known that many Armenians fled to Trebizond in 1239, after the sack of Ani by the Mongols, and that a number of Armenian monasteries were founded in and around Trebizond.47 Prelapsarian garments are mentioned in the Qur’an and hadith as well,48 and Islamic ideas associating nudity with shame may also have influenced the iconography. An Ilkhanid translation into Persian of Ibn Bakhtishu’s Manafi-i hayawan [Benefits of animals] (New York, Pierpont Morgan Library, M500, fol. 4v), which was copied at Maragha in northwestern Iran in 1297–98 or 1299–1300, shortly after the construction of Hagia Sophia, shows a human couple covering their nudity with (admittedly rather revealing) cloaks (Fig. 17). The manuscript provides scientific descriptions of animals, but it is clear that its illuminations were unable to be as frank about human biology as about any other species.49 Barbara Schmitz has recently shown that these two figures are not, in fact, Adam and Eve, but rather just two human animals, Mard va Zan (“man and woman”).50 However, certain details of the iconography, such as the haloes, suggest that the image is based on depictions of the first humans. It is impossible to detect any consistency in ˆ nıˆ’s al-Athar alIranian views on nudity and shame, and in the Edinburgh copy of al-Bıˆru baqiya [Chronology of ancient nations] (Edinburgh, University Library, Or. Ms. 161, fol. 48v), which was copied a decade later, in 1307–8, the Zoroastrian equivalent of Adam and Eve, Misha and Mishyana, are shown naked when tempted by Ahriman (Fig. 18).51 Although both of these examples date later than Hagia Sophia, it is possible that Iranian ideas could have moved to Trebizond from the east, along the trade routes that were strengthened under the Mongols in the second half of the thirteenth century.52 Reverse Narrative The other peculiarity of the Genesis frieze, the right-to-left direction of the narrative, has also traditionally been explained by reference to “Oriental” art.53 Syriac art, in partic47 Bryer and Winfield, The Byzantine Monuments, 210, also note the problems with dating the monasteries’ foundations. 48 Qur’an, trans. M. Pickthall (London, 1992), 7.22, 27; al-Tabari, Ta’arikh al-rusul wa’l-muluk [The history of al-Tabari], vol. 1, General Introduction and from the Creation to the Flood, tr. F. Rosenthal (New York, 1989), 276: “But by tearing their clothes, Iblis wanted to show them their secret parts, which had been concealed from them.” 49 For other illustrations, see S. S. Blair and J. M. Bloom, The Art and Architecture of Islam, 1250–1800 (New Haven, Conn.–London, 1994), 25–26. 50 B. Schmitz, Islamic and Indian Manuscripts and Paintings in the Pierpoint Morgan Library (New York, 1997), 17; I am grateful to Roger S. Wieck for providing me with this information in advance of publication. 51 ˆ nıˆ’s Chronology of Ancient Nations,” in The Scholar and the P. Soucek, “An Illustrated Manuscript of al-Bıˆru ˆnıˆand Jalal ad-Din al-Ru ´mi, ed. P. Chelkowski (New York, Saint: Studies in Commemoration of Abu’l-Rayhan al-Bıˆru 1975), 103–65, esp. 111. This image was also influenced by Christian iconography of the Fall. 52 Under the Mongols, the western end of these trade routes moved north to the ports of the Black Sea, and principally to Trebizond, away from the Mediterranean coast of Syria, which resisted the Mongol rule more successfully. On trade routes, see Bryer and Winfield, The Byzantine Monuments, 56, 92; H. Manandian, The Trade and Cities of Armenia in Relation to Ancient World Trade (Lisbon, 1965), 189–97. 53 Talbot Rice, Haghia Sophia, 48–49. As before, Talbot Rice has converted a hypothesis of Alpatov, “Les reliefs de la Sainte-Sophie,” 414, into a bold assertion.
230
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
ular, has been proposed as a source for this feature, as, for example, the early-thirteenthcentury Syriac Gospel Lectionary (London, British Library, Add. 7170, fol. 115r), in which the entry to Jerusalem has a right-to-left narrative (Fig. 19).54 In other words, the choice of the narrative direction has generally been seen as a formal characteristic of the frieze, and so has been explained as an artistic decision, rather than one with any theological or other meaning, or any interpretative rationale. This approach has seriously limited the analysis of the Trebizond frieze. Moreover, if we accept the Syriac artist explanation, then we are led to some bizarre, simplistic conclusions about medieval artistic practice in general. Such an interpretation would assume that artists automatically and invariably used their “normal” compositional devices brought by them from their training at home (in this case, the right-to-left narrative), and that they were unable (or unwilling) to adjust this idiom to suit the predominant left-to-right reading pattern of their new location. (This issue is raised again by the style of all the geometric plaques and stalactite impost blocks on the facade of the church, and is addressed at length elsewhere.) It also implies that the designer of the church (let us say, for the sake of argument, its patron Manuel I) either especially wanted reverse narrative, and so had to invite foreign artists (which implies that local artists were somehow incapable of carrying it out), or took no interest in the appearance of the frieze and allowed the artists to carry out the commission in their native tradition and in ignorance of normal Greek practice. One is left with either very inflexible artists, or an uninterested, laissez-faire patron. However, the contrast between the direction of the frieze and that of the inscriptions, which must have been very apparent to the artists and planners of the frieze, suggests that this difference was a much more deliberate decision. It is, of course, possible to find examples of reverse narrative throughout the Byzantine and eastern Christian worlds. In the Old Church at Tokalı Kilise, each half of the christological cycle on the vault of the cave begins with the initial scene reversed (Fig. 20);55 and the Hałbat Gospels (Erevan, Matenadaran, 6288, fol. 16v), copied and illuminated in Mqargrdzeli/Zakharid-controlled northern Armenia in 1211, contain an image of the entry to Jerusalem in which the narrative runs from right to left (Fig. 21).56 However, in none of the examples is there any consistency: in Tokalı Kilise the majority of other scenes move from left to right, and the Hałbat Gospels image is exceptional. Even in the Syriac Gospel Lectionary, narrative direction is variable, and the Annunciation on folio 15r moves from left to right (Fig. 22).57 None of these cases is really comparable to the extended nature of the narrative frieze at Trebizond. The reversed scenes are either isolated examples in longer cycles or separate iconic images to be viewed and considered individually. These examples show that right-to-left narrative movement should not be simply linked to a cultural or artistic origin, with the scheme determined by the preconceptions J. Leroy, Les manuscrits syriaques `a peintures conserve´s dans les bibliothe`ques d’Europe et d’Orient (Paris, 1964), pl. 86, fig. 1; Evans and Wixom, The Glory of Byzantium, 385. This explanation also requires us to find concrete links between Trebizond and Syriac communities that might explain the transmission of ideas. 55 A. W. Epstein, Tokalı Kilise, DOS 22 (Washington, D.C., 1986), 14–15, fig. 16, argues that the reason for this is to mark out the beginning of each section of narrative. 56 T. F. Mathews and A. K. Sanjian, Armenian Gospel Iconography: The Tradition of the Glajor Gospel, DOS 29 (Washington, D.C., 1991), 60–61, in which it is argued that this image is itself influenced by Syriac sources. 57 Leroy, Les manuscrits syriaques, pl. 73, fig. 3. 54
ANTONY EASTMOND
231
or limitations of the artist. It has been used many times in Byzantine, Armenian, and Syriac art, often without any apparent meaning. If one is to find a meaning in this feature, one must look elsewhere for a system whereby a choice such as this can have a distinct explanation. I wish to suggest here that the narrative of the frieze is, in fact, more carefully organized than previous analysis has supposed, and that it is not dependent on some posited but imprecise connection between Trebizond and Syriac communities in Mesopotamia or Anatolia. Instead, it can be read coherently within mainstream Byzantine art. This does not mean that the visual parallels listed above may not have been influential, especially given the mixed ethnic and cultural makeup of the empire of Trebizond—only that the previous approach limits explanation to a purely artistic decision and thus cannot provide a complete interpretation of the frieze. The interpretation I propose is based on creating a parallel model for the appearance of the narrative of the Fall in Byzantine art. There is no direct link between the images discussed below and those at Trebizond, and therefore no absolute system can be established. However, the parallels of iconography and context suggest that the use of narrative direction could be an important element in the construction of meaning. They also allow us to establish models for the interpretation of the Fall and its relation to the liturgy. The most important comparison is with the fourteenth-century manuscript of the Homilies of Gregory Nazianzos (Paris, Bibliothe`que Nationale, Cod. gr. 543).58 This is one of the later copies of the so-called liturgical edition, in which sixteen of Gregory’s homilies were presented in calendar order of the feasts and events discussed in the texts. The copy was made sometime after the building of Hagia Sophia, and I do not wish to suggest any direct connection between the two, only to present them as parallel phenomena. In this copy, each of the sixteen homilies is preceded by a full-page introductory miniature with two registers, related to the event discussed in the sermon.59 A typical example is folio 74v, which precedes Homily 15, “On the Martyrdom of the Maccabees” (Fig. 23). In the image the various acts of martyrdom are shown in both registers. A strong symmetry to the scenes is dictated by the landscape (with two mountains on either side and a third one in the center), but the images read from left to right (as is shown by the movement of the figures and the fact that the mountains recede from left to right). Both registers present conventional images with a straightforward narrative structure, overriding symmetry, and known iconographic origins and parallels.60 The system is common to all the other surviving miniatures in the manuscript—with one exception. This is the miniature that precedes Homily 38, “On the Nativity/Theophany” (Fig. 24). The upper register shows the Nativity, conflating the birth of Christ, the adoration of the Magi and shepherds, and the first bath of the Christ child, all of which are common enough. The lower scene introduces the Fall of Man in three scenes: the quickening of Adam/creation of Eve, Adam and Eve in Eden, and Adam and Eve lamenting after the Galavaris, Liturgical Homilies, 14–15, 140–42. Two of the miniatures are missing, having been removed. 60 In the case of this scene, the images are more dependent on the details from the fourth book of Maccabees than on the details from Gregory’s homilies, which give only generalized accounts of the martyrdoms; but the relation between the image and the homily is very straightforward. See Galavaris, Liturgical Homilies, 109–13. 58 59
232
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
expulsion.61 What is interesting about this image is that it reads from right to left, unlike any other in the manuscript. The artist has also used the psychological device of placing the scene of Adam and Eve after the expulsion beyond the frame of the miniature, in the margin, to emphasize man’s distance from Paradise. This provides a very interesting visual interpretation of the homily. In Homily 38, Gregory proclaims the Nativity as man’s chance to reenter into obedience and harmony with God, the contract dissolved by the Fall.62 Christ is the new Adam. The image demonstrates this by placing the Nativity and the scenes in Eden within the rich, broad decorative frame of the miniature, reflecting the parallel between the two chances that God has offered to man, as well as the parallel between the creations of Christ and Adam. The placing of the scene of life outside Eden outside the frame reinforces the message by demonstrating the barrenness and pain of life after disobedience.63 The choice of right-to-left narrative in the image acts to highlight this message. Since it is at odds with every other image in the manuscript, it forces the reader to reevaluate the scenes and their meaning. Instead of a normal left-to-right narrative progression, one sees here a narrative regression in which harm, rather than good, is the outcome of the events. Equally, to read from left to right is to see a return from the state of sin to purity—a return to Adam and the state of innocence. Gregory explicitly states this to be one of the purposes of the Incarnation: “But [all the mysteries of Christ] have a sole principle: to lead me to perfection, to remodel me, to bring me back to the First Adam.” 64 It would seem that the frieze shares some of the qualities of a palindrome in that it can be read in either direction (although, in this case, a change in the direction of reading produces a different reading).65 The second comparison is with another piece of sculpture in the empire of Trebizond. It is located high on the east wall of the citadel, across the ravine from the church of St.
Ibid., 118–20. This is the only copy of the liturgical homilies in which the Nativity and the Fall are combined. For a concordance of all the images accompanying the liturgical homilies, see ibid., 14–17. The interpretation of the first of the Genesis scenes in this miniature is disputed. H. Omont, Miniatures des plus anciens manuscrits grecs de la Bibliothe`que Nationale du VIe au XIVe sie`cle (Paris, 1929), 57, thought that it represented the creation of Eve; but Galavaris, Liturgical Homilies, 118–19, argued against it, although conceding that the scene may be a conflation of the quickening of Adam and the creation of Eve. The image appears to show Adam lying rigid on the ground and Christ standing over him, bending to pull Adam up by his left hand and making a gesture of blessing with his right hand. It does not seem, therefore, directly to show the creation of Eve, although the iconographic pose is very similar to images of this scene. 62 Gregory Nazianzos, Homily 38, in Gre´goire de Nazianze: Discours 38–41, ed. and trans. C. Moreschini and P. Gallay, SC 358 (Paris, 1990), 104–48. 63 The 9th-century complete edition of Gregory Nazianzos’s homilies (Paris, Bibliothe`que Nationale, Cod. gr. 510, fol. 52v) provides yet another way of interpreting the Fall in terms of homiletic literature. In this case, however, it has been shown to have specific references to 9th-century thought. See S. Der Nersessian, “The Illustrations of the Homilies of Gregory of Nazianzus Paris Gr. 510: A Study of the Connections between Text and Images,” DOP 16 (1962): 197–228; L. Brubaker, “Politics, Patronage, and Art in NinthCentury Byzantium: The Homilies of Gregory of Nazianzus in Paris (B.N. Gr. 510),” DOP 39 (1985): 1–13. 64 Gregory Nazianzos, Homily 38.16. 65 S. Pe´tride`s, “Les ‘karkinoi’ dans la litte´rature grecque,” EO 12 (1909): 86–94. R. Aubreton and F. Buffie`re, eds., Anthologie grecque, pt. 2, Anthologie de Planude (Paris, 1980), 13: bk. 16, 387b-c, record nine palindromes; and H. Hunger, Die hochsprachliche profane Literatur der Byzantiner, Handbuch der Altertumswissenschaft 12.5.2 (Munich, 1978), 2:105–6, notes the “suitability” of palindromes for a monastic setting. I am grateful to Sarah Ekdawi and Maria Vassilaki for help with this point. 61
ANTONY EASTMOND
233
Eugenios, where it is now barely visible (Fig. 25).66 It is another image of the expulsion from Eden. The plaque has clearly been moved there from an unknown location (along with a second plaque showing Elijah and the raven).67 Stylistically, this image is different from the one at Hagia Sophia: it is carved in much lower relief on a single stone within a carved frame, and the figures of Adam and Eve are shown clothed; it is also more dynamic, and the figure sizes are all compatible. But, like the frieze, it portrays a pronounced foliage background, and the narrative direction is from right to left. Divorced from any context and displayed so awkwardly, this plaque is now impossible to study or date, and only the Hagia Sophia frieze survives to provide any form of comparison. The modeling of the angel and the general consistency of the figures suggest that the image is of higher quality than the one at Hagia Sophia, but I would propose that the latter provided the model for this plaque. Although there are several iconographic differences, it is interesting to note that reverse narrative has been retained in a second image at Trebizond. Second Eve/Typology It has been noted above that the frieze appears to concentrate on the figure of Eve. This idea can now be investigated more thoroughly in the context of the church. Typological comparisons between Christ and Adam (as well as Abel) have already been noted, but it is possible that similar comparisons were also being made between Eve and Mary, the Mother of God. The idea of Mary as the second Eve is well known, and allusions to it are often made in art.68 In the miniatures that accompany the twelfth-century copy of the Homilies of James of Kokkinobaphos (Vat. gr. 1162, fol. 35v), it appears well illustrated.69 Homily 2, “On the Nativity of the Most Holy Mother of Our Lord God,” is accompanied by images of the Fall and the first murder, the results of Eve’s temptation (Fig. 14). What Eve gave birth to can only be redeemed by Christ, the son of Mary.70 In the Nazianzos Nativity/Genesis miniature, a similar link is made between the birth of Christ and the creation of Eve: in the same way that Christ redeems Adam, Mary redeems Eve. And on the twelfth-century icon of the Mother of God with the Christ child surrounded by Old Testament prophets and saints, in the collection of St. Catherine’s monastery on Mt. Sinai (Fig. 26), an inscription beneath the Virgin quotes from Romanos the Melode’s hymn on the Feast of Mary’s Nativity, proclaiming that “Joachim and Anna conceived and Adam and Eve were liberated.” 71 This system of typology has a direct parallel in Hagia Sophia. The frieze of the Fall with its emphasis on Eve is counterbalanced elsewhere in the church, namely, in the frescoes that survive in the north porch.72 These show an interesting and unusual choice Bryer and Winfield, The Byzantine Monuments, 193, pl. 130a. Bryer and Winfield, The Byzantine Monuments, 193, pl. 130b, suggest that the two reliefs may be associated, but this connection cannot be demonstrated. 68 On this subject, see Graef, Mary; and S. Benko, The Virgin Goddess: Studies in the Pagan and Christian Roots of Mariology (Leiden, 1993), 234–45. 69 Liturgie und Andacht in Mittelalter, Bibliotheca Apostolica Vaticana (Stuttgart, 1992), 132–37, esp. 134. 70 James of Kokkinobaphos, Homily 2, PG 127:580–84. 71 Evans and Wixom, The Glory of Byzantium, 372: IWAKEIM K[AI] ANNA ETEKNOGONHSAN K[AI] ADAM K[AI] EUA HLEUQERWSAN. 72 Talbot Rice, Haghia Sophia, 149–55. No identifiable frescoes survive in the south porch. 66 67
234
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
of subjects, including Jacob’s ladder, Jacob’s struggle with the angel, Moses and the burning bush (Fig. 27), the prophet Gideon, and the Tree of Jesse.73 All these images were recognized as prototypes of the Virgin, and are reflected on the Sinai icon.74 It is also noteworthy that the Tree of Jesse culminates in the figure of Mary, not in that of Christ.75 The introduction of monumental typological images at Trebizond is unusual, and only one earlier cycle of such frescoes is known, from the Georgian monastery of Betania, southwest of Tbilisi.76 The typological frescoes seem to establish another line of progression through the church—from Eve as the bringer of sin in the south porch to Mary as the bringer of salvation in the north porch. This arrangement raises two points. One is that the church and its porches may have had their own processional structure; given the scale of the porches, they must have been designed for large gatherings of people, possibly moving about as part of the service. The second point is that the links between the porches argue that the program of the church was designed in advance and included both frescoes and sculpture. Medium and Regional Context A final, but important, issue that must be addressed is the question of why sculpture in particular should have been chosen as the medium in which to portray this image. After all, sculpture is unusual in Byzantine art, and narrative reliefs are extremely rare. Here we find ourselves on firmer ground by looking at the regional context of the frieze and its possible models. However, first it is necessary to discount one famous and much quoted possible source, the Armenian church of the Holy Cross at Ałt’amar on Lake Van, erected by King Gagik of Vaspurakan between 915 and 921. It has often been cited in conjunction with Trebizond since it too has Genesis scenes carved on the exterior of the church.77 On the north facade two scenes from Genesis 3 appear: the temptation of Eve Images of Job and the hospitality of Abraham also survive in the north porch. These have typological meanings as well, but are generally associated with Christ rather than Mary. 74 On Marian typology, see S. Der Nersessian, “The Program and Iconography of the Frescoes of the Parecclesion,” in The Kariye Djami, vol. 4, Studies in the Art of the Kariye Djami and Its Intellectual Background, ed. P. A. Underwood, Bollingen Series 70 (Princeton, N.J., 1975), 303–49, esp. 310–13. 75 Talbot Rice, Haghia Sophia, fig. 115. 76 Earlier examples of individual images are, of course, known, such as the 12th-century Tree of Jesse in the church of the Nativity at Bethlehem; but Betania contains the earliest typological cycle. On this church, see A. Eastmond, Royal Imagery in Medieval Georgia (University Park, Pa., 1998), 154–69; E. Privalova, “Betaniis moxat’uloba” [The paintings of Betania], in Sabcˇ’ota xelovneba [Soviet art], pt. 8 (1980): 55–62. This may point to links between Trebizond and Georgia. Betania was taken from its owners, the Orbeli family, and partially decorated by Queen Tamar (1184–1213), who was involved in the foundation of the empire of Trebizond. The nature and extent of the links between Trebizond and Georgia after 1204 have long been disputed, but this may tentatively point to the survival of some intellectual ties between the royal families of the two countries. Manuel I was married to the princess Rusudan of Georgia; see M. Kursˇanskis, “Relations matrimoniales entre Grands Comne`nes de Tre´bizonde et princes ge´orgiens,” BK 34 (1976): 112–27. 77 For example, Talbot Rice, Haghia Sophia, 49; Alpatov, “Les reliefs de la Sainte-Sophie,” 417. Another possible source to be discounted is the sculpture of Rus’. Vladimir-Suzdal’ had a tradition of external sculptural decoration in the 12th and 13th centuries, in the form of a mass of individual images that encrusted the fac¸ades of the churches. Again, there are few formal similarities, and little evidence of any artistic transmission between the two states. On the sculpture, see G. K. Vagner, Skul’ptura Drevnej Rusi, XII vek: Vladimir, Bogoliubovo (Moscow, 1969). 73
ANTONY EASTMOND
235
(Fig. 28) and Eve offering the forbidden fruit to Adam (Fig. 29).78 No other scenes from the Fall are shown on the exterior, and in both reliefs the protoplasts are depicted naked in Paradise. The style and the scale of the reliefs are completely different from those of the Trebizond frieze, and the discrepancy in date is large and obvious. Additionally, the contemporaneous fresco cycle of the Fall painted between the windows of the drum of the dome inside the church runs only from the creation of Adam to the expulsion from Eden.79 Thus, Ałt’amar cannot be seriously considered in conjunction with Trebizond. A chronologically more reasonable comparison can be found in another Armenian carving of Adam and Eve, this time on the west face of the dome of the nearly contemporary church of Ganjasar, dated to ca. 1230 (Fig. 30).80 However, once again, iconographic and stylistic comparisons with Trebizond are few: the couple are shown naked before the Fall, and the figures are schematic and flat. Still, Ałt’amar and Ganjasar, like the Books of Adam, provide more evidence of Armenian interest in the story of the Fall. These examples also demonstrate the tradition of sculptural fac¸ade decoration in Armenia continuing into the thirteenth century. Sculptural decorations appear on church tympana in both Georgia and Armenia in this period. This shows that carving was still part of the regular artistic vocabulary of the eastern churches. But the carvings tend to be iconic or symbolic representations rather than narrative accounts, and do not present a realistic model on which to base Trebizond.81 However, there is one little-known example that can be more closely tied to Hagia Sophia in Trebizond. This is a pair of reliefs on the so-called Georgian church in Ani, which have never been properly published (Fig. 31).82 On the interior north wall (which is all that survives of the church) are two reliefs, each located under a blind arcade. The scenes show the Annunciation and the Visitation (Figs. 32, 33). It seems likely that they once formed part of a longer cycle of scenes from the early life of Christ that ran around the walls of the church. The images are carved in the tufa used throughout Ani, which results in a slightly cruder style than that of the reliefs at Trebizond, but they have roughly the same scale and depth of relief. We know that they were made by or for the Georgian community in the city by 1218, at a time when Ani was the capital of a semiautonomous fiefdom held by the Mqargrdzeli/Zakharid family from the Georgian crown. In 1239, Ani was sacked by the invading Mongols under Chamarghan, which resulted in a large exodus of its population to Trebizond. I do not intend to propose a direct transmission from Ani to Trebizond of an artist who would then be responsible for both churches, but I do think that this context provides the best parallels for the use of sculpS. Der Nersessian, Church of the Holy Cross, Aght’amar (Cambridge, Mass., 1965), figs. 41, 47, 48. T. F. Mathews, “The Genesis Frescoes of Ałt’amar,” REArm 16 (1982): 245–57; N. Thierry, “Le cycle de la cre´ation et de la faute d’Adam `a Ałt’amar,” REArm 17 (1983): 289–329; B. Outtier, “Le cycle d’Adam `a Ałt’amar et la version arme´nienne du commentaire de S. Ephrem sur la Gene`se,” REArm 18 (1984): 589–92. 80 P. Donabedian and J.-M. Thierry, Les arts arme´niens (Paris, 1987), pl. 86, fig. 715. 81 For 13th-century Armenian tympana sculptures, see Donabedian and Thierry, Les arts arme´niens, figs. 335–46. Figural sculpture in Georgia is less common after the 11th century; see N. A. Aladashvili, Monumental’naja skul’ptura Gruzii (Moscow, 1977). 82 The church inscription was published by N. I. Marr, “Nadpis’ Epifanija, katalikosa Gruzii (iz raskopok v Ani, 1910 g.),” Izvestija Imperatorskoj Akademii Nauk 4 (1910): 1433–42. The reliefs are mentioned in P. Cuneo et al., Ani, Documenti di architettura armena 12 (Milan, 1984), 91, but are described as deriving from 12thcentury Byzantine art. 78 79
236
THE GENESIS FRIEZE AT HAGIA SOPHIA, TREBIZOND
ture at Trebizond. More specifically, it further emphasizes the regional character of the art in Trebizond. The principal features of the frieze—its iconography and the choice of medium—have had strong precedents in Georgia and Armenia. It is to the Caucasus that we must turn in order to understand properly the context of the decoration of the church. III. CONCLUSION I have attempted to show in this article that, in order to understand and interpret the frieze most fully, it is necessary to find its place in the mainstream of Byzantine artistic practice. Its relation to the liturgy that it must have accompanied provides the fullest explanation of the frieze’s function, and other images of the Fall in different contexts also help to explain the possible ways in which it may have been understood. Nevertheless, the frieze remains highly unusual when seen in a purely Byzantine context. Here, the many parallels drawn from Georgian, Armenian, and Islamic sources provide many possible points of comparison. I have shown that none of these can be considered a direct source of inspiration for Hagia Sophia at Trebizond, yet taken as a group, they show how deeply the individual elements of the frieze were influenced by local regional sources. The frieze provides a concrete expression of the links between the Grand Komnenoi, their neighbors to the east—Georgia and Armenia—and their trading partners of the second half of the thirteenth century. This opens up interesting possibilities for using the church of Hagia Sophia and its decoration to investigate more deeply the questions of the cultural orientation of the empire of Trebizond in this period. The precarious position of the empire on the fringes of the Greek world, surrounded as it was by Armenian, Georgian, and Seljuq states, and its political turmoil in the wake of the Mongol invasion mean that the self-identity constructed by the Grand Komnenoi has never been fully explored. To what extent did they create a pocket Greek empire, imitating Constantinople; and to what extent did they absorb local institutions and ideas? Hagia Sophia, the only surviving piece of contemporary evidence, can help to fill in many of the gaps in the history of an empire driven, like Adam and Eve, into exile on the margins of the Byzantine world. University of Warwick
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Ivory plaque with Adam and Eve, scene with Cain and Abel. Byzantium or Italy, 11th century. Musée des Beaux-Arts de Lyon, D 312 (photo: Studio Basset, 69300 Caluire)
13
John Klimakos, Heavenly Ladder, temptation and expulsion (Vat. gr. 394, fol. 77r). Byzantium, 11th century (photo: Biblioteca Apostolica Vaticana)
14
Homilies of James of Kokkinobaphos, Adam and Eve outside Paradise, Cain and Abel (Vat. gr. 1162, fol. 35v). Byzantium, 12th century (photo: Biblioteca Apostolica Vaticana)
15
16
Homilies of James of Kokkinobaphos, Paradise (Vat. gr. 1162, fol. 37r). Byzantium, 12th century (photo: Biblioteca Apostolica Vaticana)
Octateuch, naming of the animals, creation of Eve (Istanbul, Topkapı gr. 8, fol. 42v). Byzantium, 11th century (photo: courtesy of John Lowden)
17
18
Ibn Bakhtishu, Manafi-i hayavan, man and woman (New York, Pierpont Morgan Library, M500, fol. 4v). Maragha, Iran, 1297 or 1299 (photo: David A. Loggie)
Al-Bîrûnî, al-Athar al-baqiya, Misha and Mishyana tempted by Ahriman (Edinburgh, University Library, Or. Ms. 161, fol. 48v). Iran, 1307–8 (photo: courtesy of the library)
19
Gospel Lectionary, entry to Jerusalem (London, British Library, Add. 7170, fol. 115r). Northern Mesopotamia, 1216–20 (photo: courtesy of the library)
20
21
Tokalı Kilise, Old Church, Cappadocia, scene of the Annunciation, 10th century (photo: A. Wharton)
Halbat Gospels, entry to Jerusalem (Erevan, Matenadaran, 6288, fol. 16v). Northern Armenia, 1211 (after E. Korkhmazian, I. Drampian, and G. Hakopian, Armenian Miniatures of the 13th and 14th Centuries from the Matenadaran Collection, Yerevan [ Leningrad, 1984], pl. 13)
22
Gospel Lectionary, Annunciation (London, British Library, Add. 7170, fol. 15r). Northern Mesopotamia, 1216– 20 (photo: courtesy of the library)
23
Homilies of Gregory Nazianzos, martyrdom of the Maccabees (Paris, Bibliothèque Nationale, Cod. gr. 543, fol. 74v). Byzantium, 14th century (photo: courtesy of the library)
24
Homilies of Gregory Nazianzos, nativity and expulsion (Paris, Bibliothèque Nationale, Cod. gr. 543, fol. 116v). Byzantium, 14th century (photo: courtesy of the library)
25
26
Citadel, Trebizond, east wall showing the expulsion, 13th century (after A. A. M. Bryer and D. Winfield, The Byzantine Monuments and Topography of the Pontos, DOS 20 [ Washington, D.C., 1985], pl. 130a)
Icon of the Mother of God of Kykkos with prophets, monastery of St. Catherine, Mt. Sinai. Byzantium, 11th century (photo: courtesy of the Michigan-PrincetonAlexandria Expedition to Mount Sinai)
27
Hagia Sophia, Trebizond, north wall of the north porch showing Jacob’s ladder, Jacob’s struggle with an angel, and Moses and the burning bush (photo: courtesy of David Winfield and the Russell Trust)
“Georgian church,’’ Ani, general view of the interior of the north wall, before 1218 (photo: courtesy of the Conway Library, Courtauld Institute of Art, University of London)
32
“Georgian church,” Ani, scene of the Annunciation (photo: courtesy of the Conway Library, Courtauld Institute of Art, University of London)
33
“Georgian church,” Ani, scene of the Visitation (photo: courtesy of the Conway Library, Courtauld Institute of Art, University of London)
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
The Lost Royal Portraits of Gerace ` Cathedrals and Cefalu MARK J. JOHNSON
he city of Gerace is a quiet, unassuming collection of buildings atop a table mountain in southern Calabria (Fig. 1). It is dominated by its Norman cathedral, which is the largest church in the region.1 Constructed with spolia from nearby Locri, it has lost the medieval decoration that once graced its walls. Two obscure texts, however, give evidence that this artistic program contained a very significant work of art, a mosaic panel depicting King Roger II and Leontius II, bishop of Gerace. Across the Straits of Messina on the island of Sicily is another Norman jewel, better known among those who study ` and the Byzantine influence in this part of the Mediterranean. The cathedral of Cefalu its mosaic decoration are celebrated masterpieces of the twelfth-century flowering of art in the Norman kingdom of Sicily and southern Italy. The church was a royal foundation, erected by order of King Roger II and intended eventually to have served as his final resting place.2 A little-known part of the decoration of the cathedral was a series of now destroyed panels depicting Roger and his successors, located on the west facade of the building and described in a fourteenth-century text. These descriptions permit a glimpse at what must have been remarkable expressions of royal and episcopal patronage, recognizable at once as part and parcel of the corpus of royal images in the decoration of
T
For their helpful comments on earlier drafts of this study, I wish to thank Ernst Kitzinger, Dale Kinney, ´ urcˇic´, who brought several Serbian examples and references to my attention. I am also and Slobodan C grateful to the Department of Visual Arts and the College of Fine Arts and Communications at Brigham Young University for financial support in paying for photographs and allowing me to conduct research in Sicily and Calabria during the summer of 1996. Some of the ideas found here were presented in “The Lost ` Cathedral,” BSCAbstr 18 (1992): 40–41, and “A Lost Mosaic Portrait of Roger II Porch Decoration of Cefalu and Bishop Leontius II of Gerace,” BSCAbstr 23 (1997): 61. 1 In general, see S. Gemelli, ed., La Cattedrale di Gerace: Il monumento, le funzioni, i corredi (Cosenza, 1986), with bibliography. 2 In general, see O. Demus, The Mosaics of Norman Sicily (London, 1949), 3–24 and passim; T. Thieme and `, Analecta Romana Instituti Danici, Suppl. 8 (Odense, 1977); R. I. Beck, La cattedrale normanna di Cefalu Calandra et al., Materiali per la conoscenza storica e il restauro di una cattedrale: Mostra di documenti e testimonianze ` (Palermo, 1982); G. Aurigemma, ed., La Basilica Cattedrale di Cefalu `: figurative della basilica ruggeriana di Cefalu Materiali per la conoscenza storica e il restauro, 8 vols. (Palermo, 1989); E. Borsook, Messages in Mosaic: The Royal Programmes of Norman Sicily, 1130–1187 (Oxford, 1990), 6–16; M. J. Johnson, “The Episcopal and Royal Views ` ,” Gesta 33 (1994): 118–31. at Cefalu
238
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
Norman churches in Sicily, yet unique in their own right.3 Though not unknown, the texts have never been completely analyzed before now, and the full significance of these lost pictures and of their relationship to Byzantine and Western prototypes has gone unrecognized. I. THE MOSAIC OF ROGER II AND LEONTIUS II IN GERACE The descriptions of the lost mosaic panel of Roger II and Leontius II, bishop of Gerace, have, to my knowledge, gone entirely unnoticed in discussions of the surviving images of Norman rulers in Sicily.4 Neither text is particularly detailed, but enough information is given to offer a tantalizing glimpse of the appearance of this mosaic. The first description is an eyewitness account written in the sixteenth century. After years of service in the Roman curia, Ottaviano Pasqua, a humanist scholar, was appointed by Pope Gregory VII as bishop of Gerace, where he served for a number of years (1574–91).5 His historical interests led him to compile a work on his predecessors, the Vitae episcoporum ecclesiae Hieracensis, which was published in an obscure book only in 1755.6 In discussing the life of Leontius, Pasqua noted that “in the cathedral, near the altar of the Santissimo Salvatore one may still admire a mosaic in which are depicted, in poses of devotion, on the right Leontius and on the left Count Roger. The former [is shown] with a gold miter and cope, the other with a gold crown on his head, the royal scepter in his hand and clothes covered with gold lilies.” 7 A second, even briefer, description is found in a book written by Giovanni Fiore in the middle of the seventeenth century and published a century later. Confirming the report of Pasqua, Fiore mentions “the chapel of the Savior with its ancient mosaic image, For the royal images of Norman Sicily, see S. Steinberg, “I ritratti dei re normanni di Sicilia,” La Bibliofilia 39 (1937): 29–57, who does not mention the images under discussion here. On the larger issue of medieval royal and imperial images, no single study has yet been attempted. For Byzantine imperial images, see A. Grabar, L’empereur dans l’art byzantin (Paris, 1936); K. Wessel, “Kaiserbild,” RBK 3:722–853; T. Velmans, “Le portrait dans l’art des Pale´ologues,” in Art et socie´te´ `a Byzance sous les Pale´ologues: Actes du Colloque organise´ par l’Association internationale des ´etudes byzantines `a Venise en septembre 1968 (Venice, 1971), 91–148; eadem, La peinture murale byzantine `a la fin du Moyen Age (Paris, 1977), 62–74; P. Magdalino and R. Nelson, “The Emperor in Byzantine Art of the Twelfth Century,” ByzF 8 (1982): 123–83, repr. in P. Magdalino, Tradition and Transformation in Medieval Byzantium (Brookfield, Vt., 1991), pt. 6; C. Jolivet-Le´vy, “L’image du pouvoir dans l’art byzantin `a l’e´poque de la dynastie mace´donienne,” Byzantion 57 (1987): 441–70; J. Lafontaine-Dosogne, “Les the`mes iconographiques profanes dans la peinture monumentale byzantine du Ve au XVe sie`cle,” in Arte profana e arte sacra a Bisanzio, ed. A. Iacobini and E. Zanini, Milion 3 (Rome, 1995), 189–219. For the West, see P. Schramm and F. Mu ¨ therich, Die deutschen Kaiser und Ko¨nige in Bildern ihrer Zeit, 751–1190, 2d ed. (Munich, 1983); W. J. Diebold, “The Ruler Portrait of Charles the Bald in the S. Paolo Bible,” ArtB 76 (1994): ¨ppingen, 1977). 6–18; C. A. Willemsen, Die Bildnisse der Staufer (Go 4 They have been briefly mentioned in literature concerning the church at Gerace. See, for example, C. Garzya Romano, La Basilicata: La Calabria, Italia romanica 9 (Milan, 1988), 213. 5 E. Cochrane, Historians and Historiography in the Italian Renaissance (Chicago, 1981), 448. 6 O. Pasqua, Vitae episcoporum ecclesiae Hieracensis, ed. G. A. Parla`, in Constitutiones et acta Synody Hieracensis ab illustrissimo et reverendissimo domino Caesare Rossi Episcopo celebratae diebus 10, 11 et 12 Novembris 1754 (Naples, 1755). 7 “Eiusdem Leontii imago opere vermiculato in basilica cathedrali ad altare SS. Salvatori dicato ad dexteram, Rogerii autem Comitis ad laevam pie expresse spectatur adhuc, quorum ille mitra auro intexta ac pluviali indutus visitur, hic auream coronam capiti impositam, manu regale sceptrum gestans, vestitu aureis liliis circumfuso.” Constitutiones, 248–49; quoted in E. D’Agostino, “I vescovi,” in Cattedrale di Gerace, ed. Gemelli (as in note 1), 209. 3
MARK J. JOHNSON
239
with, on the left, King Roger, dressed in royal clothes, and on the right, Bishop Leontius, his relative.” 8 Pasqua indicates that the mosaic was located near the altar of the Santissimo Salvatore. This altar was, and still is today, located in the north apse of the church, but it is unclear where exactly the panel would have been (Fig. 2). It may have been on the apse wall itself, below or to one or the other side of the window. Perhaps it was actually near the apse, for example, on the pier separating the main and the north apses. As the walls are now covered with plaster, no physical evidence for the placement of the panel is visible. The date of the mosaic can be roughly determined from the description, which makes it clear that Roger wore a crown. Although he ruled Sicily and southern Italy from 1101, his coronation as king occurred in 1130, while Leontius was bishop from about 1124 until his death in 1144(?). Therefore, the mosaic would have been completed between 1130 and ca. 1144. Its destruction apparently happened during the bishopric of Domenico Diez (1689–1729), in order for a crucifix to be hung in its place.9 The editor of Pasqua’s Vitae, Giuseppe Parla`, bemoaned its loss and noted that there were still people alive in 1755 who remembered seeing the mosaic. The descriptions clarify that the mosaic depicted King Roger on the left and Bishop Leontius on the right, presumably standing and identified by inscriptions. Pasqua noted that both were in “poses of devotion.” What was meant by this phrase is open to interpretation—perhaps, their heads were slightly bowed, or maybe their hands were raised, or their arms were outstretched in some manner. Whatever the exact pose was, somehow the two were made to look pious. Leontius wore a gold miter and cope; Roger was shown with his royal regalia of the gold crown, the scepter, and clothes strewn with gold lilies. The appearance of Roger is, of course, well known from his depiction in the famous investiture mosaic in Santa Maria dell’Ammiraglio, also known as the Martorana, in Palermo, done during the 1140s and therefore a close contemporary to the mosaic at Gerace (Fig. 3).10 There Roger is shown in the dress of a Byzantine emperor, with his head bowed and arms raised toward Christ in a gesture of prayer. It may be assumed that his facial features would have been similar in the Gerace mosaic, as well as the crown, though his dress was somewhat different. In Santa Maria, Roger wears a loros, a long under-tunic, and a dark blue outer tunic decorated with fleur-de-lis. As for Bishop Leontius, no other image exists of him—or, for that matter, of any other bishop in Norman southern Italy. His dress, however, was likely to be similar to that worn by the bishops depicted in twelfth-century illuminated manuscripts produced in southern Italy.11 The overall composition of the mosaic at Gerace, as described by Pasqua and Fiore, calls to mind that of the enamel plaque from the church of San Nicola in Bari, showing G. Fiore, Della Calabria illustrata (Naples, 1748; repr. Sala Bolognese, 1974), 305. So says Giuseppe Parla`, the editor of Pasqua’s text: “Has Leontii et Rogerii imagines vivunt adhuc qui spectavere: sed pessum iverunt ipsae, quod est dolendum, Diezio Episcopo, qui ibi magnum crucifixi statuam voluit apponere, ubi illae erant, etsi temporum injuria aliqua parte attritae, at quae tamen possent, et dignum era, restaurari.” See Constitutiones, 249 n. 1; quoted in D’Agostino, “Vescovi,” 217. 10 For the Martorana mosaic, see E. Kitzinger, The Mosaics of St. Mary’s of the Admiral in Palermo, DOS 27 (Washington, D.C., 1990), 189–97. 11 For example, the Chronicon Vulturnense and the Liber in honorem Augusti of Petrus de Ebuli, for which see below. 8 9
240
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
Roger being crowned by St. Nicholas (Fig. 4).12 Though its date is uncertain, the period soon after 1139—the year when Roger suppressed a rebellion and asserted control in Bari—seems likely for its manufacture. Though somewhat schematic in its details, it depicts Roger standing on the left with a crown, labarum, and orb, wearing a tunic, mantle, and loros decorated with fleur-de-lis. To the right stands the bishop saint in ecclesiastical garments. Notwithstanding the differing details, the small enamel image is close to what must have been visible in the mosaic at Gerace. The origins of the cathedral of Gerace are very uncertain, as no documentation for its foundation and construction survives. It seems to have been built either in the late eleventh century under the reign of the Great Count Roger I, who died in 1101, or in the early part of the twelfth, during the reign of his son, Roger II. Though no role in the building of the church is documented for either ruler, the younger Roger is known to have collaborated with Leontius I in the foundation and construction of a Basilian monastery near Gerace.13 A mosaic depicting King Roger and Bishop Leontius suggests that the bishop may have been responsible for building the church or decorating its interior, presumably with assistance from the king; if Roger had been the sole patron of the church, he would have likely been shown alone. As the likely founder or donor, Leontius was honored in a monumental mosaic. No evidence exists to support Fiore’s claim that Leontius and Roger were relatives. This may or may not be true; if true, it may explain the king’s interest in this project. Nevertheless, related or not, Roger’s presence in the mosaic certainly conveyed some meaning of royal support for the bishop and his church. The king possessed the right to appoint bishops in his kingdom.14 Leontius, therefore, owed his position and any temporal powers as vassal to the king; Roger was his benefactor, ecclesiastically and temporally for sure, and possibly financially as well, especially when one considers the great size of this church. One is again reminded of the mosaic at Santa Maria dell’Ammiraglio, the presence of which in that church is understood when it is considered through its pendant, the dedicatory panel of George of Antioch, Roger’s minister and founder of the church (Fig. 5). Roger is depicted here in order to show the source of George’s authority. Perhaps Roger was depicted at Gerace for a similar reason—to show from whom the bishop, Leontius, derived his position and authority. If this was indeed the case, then the likely patron for the mosaic would be Leontius. There may have been one other mosaic depicting a bishop in the Norman kingdom of Sicily. The cathedral of Palermo, rebuilt by Bishop Walter at the same time that the cathedral of Monreale was under construction, and dedicated in 1185, apparently had a mosaic in its apse. It was destroyed, with no surviving depictions or descriptions, but its inscription was recorded and seems to indicate that its decoration consisted of a large 12 E. Bertaux, “L’e´mail de St. Nicholas de Bari,” Acade´mie des inscriptions et belles-lettres, Commission de la Fondation Piot: Monuments et me´moires 6 (1899): 61–90 and pl. VI; P. Belli D’Elia, La Puglia, Italia romanica 8 (Milan, 1986), 172–73; M. Andaloro, ed., Splendori di Bisanzio (Milan, 1990), 190–91. 13 A diploma in the Syllabus of Trinchera, written in Greek and dating from 1101, documents this action, for which see E. D’Agostino, “Osservazioni e note su un documento geracese del XII secolo,” BollGrott, n.s., 36 (1982): 43–59. 14 I. S. Robinson, The Papacy, 1073–1198: Continuity and Innovation (Cambridge, 1990), 388.
MARK J. JOHNSON
241
figure of Christ with a small prostrate Walter at one of his feet.15 Though a very different work of art from the mosaic at Gerace, the apse decoration of the Palermo cathedral demonstrates that episcopal patronage of mosaic art did indeed exist in the Norman kingdom. ` II. THE ROYAL PORTRAITS AT CEFALU
` panels is preserved in a compilation of documents The description of the Cefalu made in 1329 and known as the Rollus rubeus.16 Thomas de Butera, the bishop of the time, had ordered the collection of evidence concerning the privileges of his see, which for the most part meant the copying of various instruments of donations and privileges granted to the church and the bishopric from their inception. The job was given to Rogerius de Mistretta, a notary, who completed his commission in 1330. Rogerius began his compilation with a recounting of the legendary foundation of the church as a fulfillment of a vow following the deliverance of King Roger from a storm at sea.17 This was followed by a series of five documents, each describing a single panel of the facade decoration and containing a preface and signatures of several witnesses (see Appendix).18 According to the source, this decoration consisted of at least six figural panels, five of which depicted rulers who held scrolls containing some of the privileges and donations that they had granted to the church. Concerned about the possible effects of age and water on the panels and wishing to preserve knowledge of them for posterity, Bishop Thomas ordered Rogerius and Primus de Primo, a city judge, to document their appearance and, especially, to record the privileges granted to the cathedral that were depicted in the panels. The document gives the wall of the Porta Regia—or west facade of the church (“in pariete porte regum”)—as the position of the panels, and adds the phrase “in the bell tower of the same church” (“in ipsius ecclesie campanario”), which sounds confused since the door is not in a tower, but placed in the center of the wall between two towers (Fig. 6).19 Presumably, the author was using the phrase “in campanario” to refer to the part of the church in which the bell towers are located (i.e., the west facade), rather than to a 15 Demus, Mosaics, 188, 190 n. 13. The early-13th-century apse mosaic at San Paolo Fuori le Mura in Rome, showing Pope Honorius III (1216–27) at the foot of Christ and set by mosaicists from Venice, provides a parallel. See W. Oakeshott, The Mosaics of Rome from the Third to the Fourteenth Centuries (Greenwich, 1967), 295–96 and fig. 186. 16 Rollus rubeus, privilegia ecclesie Cephaleditane, a diversis regibus et imperatoribus concessa, recollecta et in hoc volumine scripta, ed. C. Mirto, Documenti per servire alla storia di Sicilia, ser. 1, 29 (Palermo, 1972). 17 Ibid., 24–26. See Johnson, “Views,” 118, 128–29. 18 Rollus rubeus, ed. Mirto, 26–32. Brief comments on the description may be found in Demus, Mosaics, 10; `: La sua origine e i suoi monumenti (Palermo, 1888), 130–36; H. Schwarz, “Die R. Salvo di Pietraganzili, Cefalu ¨r Kunstgeschichte 6 Baukunst Kalabriens und Siziliens im Zeitalter der Normannen,” Ro¨misches Jahrbuch fu ` e trasportati (1942–44): 99–102; G. Agnello di Ramata, “I sarcofagi donati da Ruggero II alla chiesa di Cefalu a Palermo per ordine di Federico II,” AStSic, ser. 3, 7 (1955): 260–65; C. Valenziano and M. Valenziano, “La ` pour la se´pulture du roi Roger,” CahCM 21 (1978): 3–30, supplique des chanoines de la cathe´drale de Cefalu ` nel periodo normanno (Palermo, 137–50, esp. 140–41, 146; trans. and repr. as La basilica cattedrale di Cefalu `: Memorie storiche (Palermo, 1988), 43. 1979), 46–47, 54; D. Potera, Cefalu 19 Rollus rubeus, ed. Mirto, 26.
242
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
single tower. As at the time there was no portico, this being added in 1471/2, the pictures would have indeed been exposed to the elements.20 The panels are described in chronological order, beginning with a dedicatory image of Roger II (1130–54), who founded the church in 1131, and continuing with panels representing his successors: William I (1154–66); William II (1166–89); Constance (1194–98), the daughter of Roger and wife of Henry VI; and her son, Frederick II (1198– 1250). There is nothing in the document that indicates their specific arrangement on the wall, though it may be assumed that the earliest panels would have been placed near the portal, and the others added to either side. It is not clear from the document whether these were fresco paintings or mosaics, as they are referred to simply as “pictures” or “images.” Given the known parallels from the period, however, mosaic seems likely at least for the earliest panels. The precise dating of the individual panels is largely a matter of conjecture, though it seems likely that the first two panels would have been placed there during the reign of Roger, with the others being added later. The date of their destruction is unknown, but it may have occurred during the construction of the portico and its roof, and certainly by the end of the sixteenth century.21 Before I turn to an examination of the panels, a few other observations are pertinent. It will be seen that the descriptions of the figures are not particularly detailed. However, the text of the document assures the reader that the inscriptions have been faithfully copied, with nothing added to or subtracted from what was depicted in the pictures. This is a strong indication that the primary reason for creating the descriptions was not to document the works of art but to record the inscriptions. No mention is even made in the Rollus rubeus of the other decoration of the church, and if these panels had no texts with privileges, the bishop and Rogerius would have ignored them as well. The images are described in a minimal way—just enough to lend authority to the texts. Each description is followed by the signatures of Primus and Rogerius, along with a variety of other witnesses. The first panel is described by the notary Rogerius and the judge Primus in this manner: Our Savior sitting in his majesty, with his left hand receiving the depicted church [i.e., model], and with his other, right, hand he makes the sign of the cross; the church is held by Roger, the builder of this church, with his right hand. He wears royal vestments and is crowned. And above the head of said king is written, “King Roger.” And with his left hand he holds a written scroll. The writing on this is, “Accept, Savior, the church and the ` with all of its rights and liberties; to ourselves and our successors we reserve city of Cefalu [the right of ] punishing felonies, seditions, and homicides.”
This description brings to mind the three famous surviving royal mosaics of Norman Sicily, namely, the panel of Roger crowned by Christ in the church of Santa Maria dell’Ammiraglio in Palermo (Fig. 3) and the two dedicatory panels of William II in the cathedral of Monreale, one showing him crowned by an enthroned Christ (Fig. 7) and R. Calandra, Aggiunte, modifiche e restauri degli ultimi sette secoli, vol. 2 of Basilica, ed. Aurigemma (as in note 2), 29–40. 21 B. Carandino, Descriptio totius Ecclesiae Cephaleditanae (Mantua, 1592), 23, states that there was nothing left to see of the decoration in his time. 20
MARK J. JOHNSON
243
` the other with him offering a model of the church to a seated Mary (Fig. 8).22 The Cefalu panel would seem to have included elements found in these mosaics—those of a standing figure of the king dressed in his fine silk robes, with one hand extended to offer a model of the church to an enthroned Christ. The basic composition must have been similar to that of the Monreale panels, with the major difference being the pose of the king, since ` he also held an open scroll in his left hand. in Cefalu As its founder, Roger had particularly strong ties to the cathedral. Having begun construction of the church in 1131, he urged to elevate its status to that of a cathedral and, in 1145, donated two porphyry sarcophagi to the church with the intention that he would eventually be entombed in one, with the other acting somewhat mysteriously as a monument to his memory.23 The church was equipped with a royal throne in the sanctuary, and its mosaic decoration included elements designed to stress its links with the king.24 Following his death, however, Roger was buried in Palermo, and, although a few years after the death of William I in 1166 Bishop Boso and the canons of the church petitioned William II and his regent mother for the permission to transfer the bodies of ` , they were unsuccessful. The last hope that the church would the two kings to Cefalu serve its intended function of the royal mausoleum ended when the sarcophagi were removed to Palermo by order of Frederick II, probably in 1215.25 The privileges accorded to the church that were recorded on the scroll held by Roger are documented elsewhere in identical terminology. They are named in a document of Roger of 1132 and reiterated in the donation instrument of 1145.26 This indicates, of course, that the source of the text found on Roger’s scroll was the actual diploma issued by the king to the church. The same is most likely true for the texts in the other panels. The document then describes what it identifies as the third panel, “cuius tertie figure imago talis est.” 27 The second panel is not described, presumably for its lack of a depicted legal document. Since the third one showed Roger II’s successor and son, William I, it is probable that Roger was pictured once more in the second panel, this time likely in an investiture scene, depicted receiving his crown from Christ in the manner of the Martorana panel or that of William II in Monreale.28 Thus, it appears that the portal was flanked For the Monreale panels, see Demus, Mosaics, 123–26; Borsook, Messages, 67–68; E. Kitzinger, The Mosaics of Monreale (Palermo, 1960), 18. 23 ` , perg. no. 44, in Rollus rubeus, ed. Palermo, Archivio di Stato, Tabulario della chiesa vescovile di Cefalu Mirto, 41–45. See also C. Bru ¨ hl, ed., Rogerii II. Regis diplomata Latina, Codex diplomaticus regni Siciliae, ser. 1, Diplomata regum et principum e gente Normannorum 2.1 (Cologne, 1987), 197–200, no. 68. See J. Dee´r, The Dynastic Porphyry Tombs of the Norman Period in Sicily, DOS 5 (Cambridge, Mass., 1959), 1–3. 24 Johnson, “Views,” 125–29. 25 ` made in 1170 by the canons of the On the petition for the burials of Roger and William I in Cefalu church, see Valenziano and Valenziano, “Supplique,” 3–7, 143–50; eidem, Basilica, 50–60, 69–76; Dee´r, Tombs, 5–14. For the removal of the sarcophagi, see J. L. A. Huillard-Bre´holles, Historia diplomatica Friderici Secundi (Paris, 1852), 1:426–27; Dee´r, Tombs, 18–19; Agnello di Ramata, “Sarcofagi.” 26 For the document, issued in Palermo in March 1132, see Bru ¨ hl, Rogerii, 53, no. 19; for the donation instrument, see Rollus rubeus, ed. Mirto, 43, and Bru ¨ hl, Rogerii, 200. 27 Rollus rubeus, ed. Mirto, 28. 28 As suggested by Valenziano and Valenziano, Basilica, 47 n. 56. For this type of image, see Grabar, Emper´ urcˇic´ has suggested to me that perhaps it eur, 112–22; Wessel, “Kaiserbild,” 745–51. Professor Slobodan C was Roger’s father shown in the panel, noting that in a later period King Milutin and his deceased father were depicted on either side of the entrance of the early-14th-century church of the Virgin Ljevisˇka at Prizren, for which see below. For father-son pairings in Serbian art, see S. Mandic´, “Dvojno ktitorstvo,” in 22
244
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
by two representations of Roger, one as donor and the other in an image of royal authority, as was later done at the cathedral of Monreale with the images of William II. The next panel to be described in the document, as has been noted, is the third. Concerning this picture, the notary Rogerius wrote: Pictured is the king, his face turned wholly to the right toward the people looking on, or to the opposite wall. He wears royal vestments, is crowned, and with his right hand holds the royal scepter, and with his left, a painted and written scroll. The writing on the scroll ` we conis, “That which Our Father of divine memory conceded to the church of Cefalu firm, ratify, and approve. And by our pious clemency we add to these gifts and we donate the church of Saint Lucia of Syracuse with its villages and all that belongs to it.” Above his head is written, “William I King of Sicily.”
This panel represented the only known monumental depiction of William I in Norman art. It must have been similar to those of his father and his son, namely, an image of a standing crowned figure in court dress, holding a symbol of his power in his right ` in the form of a document hand and the proof of his concern for the cathedral of Cefalu in his left. The passage is somewhat perplexing with its references to the “people looking on, or . . . the opposite wall.” The people would be those who viewed the image, but, if the image were on the facade, then there would be no “opposite wall”; though, perhaps, the phrase can be taken to mean that the king’s head was turned to the right toward the wall of the bell tower. ` . His interest in the church William I also had personal ties to the cathedral of Cefalu is cited in the canons’ petition of 1170, which states that upon the death of his father he confirmed that Roger should be buried there once the church was consecrated. During ` , he stopped in front of the sarcophagus intended for his father and stated a visit to Cefalu his desire that both he and his father should be buried in the church and that, as the people of the city worshiped in the church, they should pass by the two sarcophagi and pray for his soul and that of his father.29 In the end, he too was buried in Palermo, though his remains were later transferred to the new dynastic church, the cathedral of Monreale, built by his son. Although William’s confirmation of the church’s privileges is alluded to in the canons’ petition and in a privilege issued by Constance, no separate document issued by William himself and stating this survives.30 Apparently, no such document was available to Rogerius to be copied into the Rollus rubeus, so the pictorial representation must have taken on added significance for him and Bishop Thomas. There is some question about the affirmation of the gift of the church of St. Lucia in Syracuse to the cathedral. A later scribe crossed out “Lucia” and replaced it with “Mary,” but, once again, there is no independent confirmation of this gift either. A papal bull of 24 November 1169, issued by ` and to extend recogniAlexander III, was the first to recognize the bishopric of Cefalu Drevnik: Zapisi konzervatora (Belgrade, 1975), 146–54. In either scenario, the message would have been one of royal authority. 29 Valenziano and Valenziano, “Supplique,” 144–45; eidem, Basilica, 57–58; Dee´r, Tombs, 7. 30 `, For the document of Constance (Palermo, Archivio di Stato, Tabulario della chiesa vescovile di Cefalu ¨lzer, ed., Constantiae perg. no. 34), issued at Palermo in May 1198, see Rollus rubeus, ed. Mirto, 56–58; T. Ko imperatricis et reginae Siciliae diplomata (1195–1198), Codex diplomaticus regni Siciliae, ser. 2, Diplomata regum e gente Suevorum 1.2 (Cologne, 1983), 204.
MARK J. JOHNSON
245
tion to the cathedral’s rights and possessions. Among the possessions are listed the churches of St. Mary at Camarata and St. Lucia in Syracuse, yet no mention is made of any role played by William I.31 To further complicate matters, a charter given by Countess Adelicia, the niece of King Roger, in June 1140 presented the monastery of St. Lucia in Syracuse to the cathedral.32 Perhaps, William simply reconfirmed its dependence on Cef` , and this is what was commemorated in the picture. alu The image of the fourth panel is described in this manner: Pictured here was a king wearing royal vestments and crowned; above his head is written, “William II King of Sicily.” He holds in his right hand the royal scepter, and with his left, a scroll depicting writings. The writing was, “By royal clemency we, heirs of our ` progenitors, concede what they conceded by their benevolence to the church of Cefalu and we confirm it with the present writing.”
This figure must have been a virtual mirror image of the third panel, if one assumes that both William I and William II were depicted alone. It is also easy to imagine the figure of William II as being very close to that depicted in his coronation panel at Monreale (Fig. 7), though with his head raised and arms extended to hold the scepter and the scroll. ` , and there is no outside William II had no strong affinity for the church at Cefalu evidence for any specific donations made or privileges granted to it on his part other than the brief mentions of his confirmation of the rights of the church in the canons’ ` and in the privilege of Conpetition for the burials of Roger and William I at Cefalu stance. As noted, the petition was unsuccessful, which is evidence of William II’s lack of interest in the church. He soon turned to his own church-mausoleum project, building the new cathedral church at Monreale and seeing to the burial of his father there. Given ` fulfill its intended function, his panel may well have his lack of interest in seeing Cefalu been placed there as part of the canons’ effort to curry favor with the king. Unable to secure the remains of Roger and William I, the canons were nevertheless successful in holding onto the two sarcophagi, at least for the time being. The fifth panel was unique in the series and in the art of Norman Sicily: There was depicted here the image of a woman, her face turned toward the observers. She is dressed in royal vestments and with her right hand holds the royal scepter, and with her left, a painted and written scroll. The writing on it was of this tenor: “By our innate benevolence, we, Constance, Empress of the Romans, confirm to the church of ` that which our father, Roger King of Sicily, had confirmed, and in addition we Cefalu donate in perpetuity to it the village of Odosuer with its men and belongings.” This image was crowned and above was written, “Constance, daughter of King Roger, Queen of Sicily and Empress of the Romans.”
Constance ruled Sicily with her husband from 1194 until her death in 1198, but the text may provide a means of dating the panel even more precisely. The act of donating 31 Valenziano and Valenziano, Basilica, 48–49, 54, claim that the text in the Rollus rubeus naming the church of St. Lucia was a scribal error. However, Rogerius was simply copying the text that he saw in the picture, so it is more likely that either the text of the painting was in error, or William did have some role in donating ` or perhaps in confirming the donation. the church to the cathedral of Cefalu 32 L. T. White, Jr., Latin Monasticism in Norman Sicily (Cambridge, Mass., 1938), 192, 202–4.
246
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
the village of Odosuer is also preserved in a document of May 1198 issued by the queen.33 The panel may have been added to the facade during the last few months of her life, though the possibility remains that it was added posthumously, during the reign of her son. This image is the only recorded monumental representation of the queen in Norman Italy. There are a limited number of images of Constance to provide a comparison, or better, an idea of her appearance in the mosaic. A small likeness is found on a wax seal in Palermo, thought to have come from one of the diplomas issued in her name to the ` cathedral.34 On the seal, she is shown seated, dressed in a richly decorated robe, Cefalu and wearing a crown with pendants, while in her left hand she holds a scepter topped by a lily. Constance is depicted eleven times in the illustrations of Peter of Eboli’s Liber ad honorem Augusti, written and decorated in 1195/6, which includes the scene of her departure from the palace at Salerno on her way to claim the throne from Tancred (Fig. 9).35 Although the quality of these illustrations leaves much to be desired, they do provide at least a general sense of her appearance and apparel. Here she wears a green tunic with a red trim and a red cloak and shoes. She is crowned and holds a scepter in her right hand; in her covered left hand is a gold “half-moon,” part of the imperial regalia.36 One can imagine a similar, if more carefully rendered, image of the queen flanking the portal `. at Cefalu An interesting point is raised by the fact that it was Constance alone who was depicted here, rather than her husband Henry VI or the two of them together. To show the queen only was extremely unusual in either Byzantine or Western art of this period. It may be explained in light of the close connections between the church and members of her family. Her document of donation recalls her father’s role as founder and builder of the church and mentions the confirmation of privileges by her brother and her nephew. It is this familial link that is stressed in the document and recorded in the decoration. It is apparent that in making the gift of Odosuer and confirming previous privileges, the queen was continuing to evidence the interest in the church expressed by her father and other family members. The sixth and final panel is described by the notary Rogerius in this way: Depicted is a king wearing regalia and crowned seated on a throne, holding a royal scepter in his right hand. He turns to face a bishop wearing pontifical robes and a miter, above whose head is written, “Iohannes Episcopus,” and who receives a scroll from the left hand of the king, in which is written, “Go to Babylon [Cairo] and Damascus and seek 33 ` , perg. no. 34: “. . . casale quod Palermo, Archivio di Stato, Tabulario della chiesa vescovile di Cefalu dicitur Oddesver cum tentimentis et pertinenciis suis libere et absque ullo servicio eidem prefate Cephaludensi ecclesie perpetuo duximus concedendum . . .” 34 C. Damiano Fonseca, ed., Federico II e l’Italia: Percorsi, luoghi, segni e strumenti, exhibition catalogue (Rome, 1995), 190, no. II.1, with color illustration. 35 Fol. 119r. For editions of this text, both with complete illustrations, see Pietro da Eboli, Liber ad honorem Augusti, secondo il codice 120 della Biblioteca civica di Berna, ed. G. B. Siragusa, FStI 39–40 (Rome, 1905–6); E. Rota`, ed., De rebus Siculis carmen, RIS, n.s., 31.1 (1904–10). For a new edition with German translation, commentary, and excellent color illustrations, see Petrus de Ebulo, Liber ad honorem Augusti sive de rebus Siculis: ¨lzer and M. Sta¨hli (Sigmaringen, 1994), with additional bibliogCodex 120 II der Burgerbibliothek Bern, ed. T. Ko raphy. A crude portrait on the fragment of a wooden ambo from Salerno has been identified as Constance. See S. Steinberg, “A Portrait of Constance of Sicily,” JWarb 1 (1937–38): 249–51, pl. 31d. 36 ¨lzer and Sta¨hli, 129–30. Petrus de Ebulo, Liber, ed. Ko
MARK J. JOHNSON
247
the sons of Saladin asking of them in my name if they would want to improve the condition of their lands.” Above the head of the king is written, “Frederick I, Emperor.”
Actually, as Holy Roman Emperor he was Frederick II, the designation we recognize, but to the local artist he was the first Frederick to have ruled in Sicily. Frederick was the king of Sicily from 1198 until his death in 1250, but the reference to the mission of Bishop Johannes de Cicala gives us a more specific time frame for the addition of this panel to the series.37 The Cicala family had provided several allies and counselors to the king, and it is known that the bishop was part of a group sent by Frederick on a mission of peace to Cairo and Damascus in 1213.38 The decoration must have been done between this time and the bishop’s death in 1216, perhaps in 1215, when Frederick donated an estate to the cathedral in recompense for having had the sarcophagi removed to Palermo.39 This panel represents the only known monumental depiction of Frederick II in Sicily, though his images in a variety of media, including sculpture, were relatively numerous in southern Italy.40 Several images, particularly those on seals, show him seated on a throne, as he was depicted in the panel.41 The most famous portrait of an enthroned Frederick appears on the frontispiece to the text that he authored in about 1220, De arte venandi cum avibus, which was copied and illustrated within a few years after his death (Fig. 10).42 In a depiction that is probably based on his images on seals, he is seen sitting in a frontal position, holding a scepter in his right hand, in a manner similar to the picture described by Rogerius. He is shown beardless as he usually is in his portraiture. In fact, having been born in 1194, he would have still been a young man at the time of `. the creation of the panel at Cefalu The scroll is the only one in the group that does not allude to some donation or privilege, though Frederick did confirm the privileges of the cathedral in 1201.43 The nature of this panel is therefore different, as it records a historical event that links the On Johannes de Cicala, see N. Kamp, Kirche und Monarchie im staufischen Ko¨nigreich Sizilien, vol. 1, Prosopo¨mer und Bischo¨fe des Ko¨nigsreichs, 1194–1266, pt. 3, Sizilien, Mu graphische Grundlegung: Bistu ¨ nsterische Mittelalter-Schriften 10 (Munich, 1975), 1049–55. 38 The date of this mission is disputed, though it could not have been as late as 1226, as argued by Demus, Mosaics, 10. Agnello di Ramata, “Sarcofagi,” 263–65, does not believe that such a mission could have occurred before 1217, but this would have been too late for Johannes. Perhaps, the mission took place as early as ` . See Huillard-Bre´holles, Historia, 1:156. 1209, when Frederick was in Cefalu 39 Huillard-Bre´holles, Historia, 1:426–27; Dee´r, Tombs, 18. 40 ` Mariani and R. Cassano, eds., Federico II, immagine e potere, See Willemsen, Bildnisse, 20–37; M. S. Calo exhibition catalogue (Venice, 1995); V. Pace, “Il ‘ritratto’ e i ‘ritratti’ di Federico II,” in Federico II e l’Italia, ed. Fonseca (as in note 34), 5–10. 41 Die Zeit der Staufer, exhibition catalogue (Stuttgart, 1977), 1:29–35, nos. 43–51; 3: figs. 14–20; Willemsen, Bildnisse, figs. 45–60. 42 C. A. Willemsen, ed., De arte venandi cum avibus: Ms. Pal. Lat. 1071, Biblioteca Apostolica Vaticana, vol. 1, Kommentar; vol. 2, Facsimile (Graz, 1969); Zeit der Staufer, 1:658–59, no. 824; F. Cardini, “Federico II e il De arte venandi cum avibus,” in Politica e cultura nell’Italia di Federico II, ed. S. Gensini (Pisa, 1986), 213–32; M. A. Coppola, “Il De arte venandi cum avibus dell’imperatore Federico II di Svevia,” Helikon 31–32 (1991–92 [1993]): 127–88. 43 June 1201, for which see the Rollus rubeus, ed. Mirto, 53–55; Huillard-Bre´holles, Historia, 1:77; C. Valen` nei documenti d’archivio e nella epigrafi, vol. 4 of Basilica, ed. Aurigemma ziano, La Basilica ruggeriana di Cefalu (as in note 2), 11. Frederick would have been, of course, a very young boy at this time, but the privilege was issued in his name. 37
248
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
ruler and the bishop, rather than repeating the usual theme of privileges found in the previous panels. Based on the inclusion of Johannes in the panel, as well as the fact that it is his mission that is commemorated, one may conclude that the bishop rather than the emperor was in all likelihood the patron of this panel. Notwithstanding the divergent nature of this scene, ultimately it fit within the overall decoration of the facade, for, once again, the underlying message was that of the links of the church to royalty. It has been suggested that all of the decoration actually belonged to the period of Frederick. The arguments include the fact that there are no known Norman parallels of a dynastic assemblage, whereas others are found in the patronage of Frederick, and the fact that it is Constance, not Henry VI, who is depicted here in order to provide the genealogical link between Frederick and his grandfather, Roger II.44 Although such a dating may seem attractive to some, there are valid reasons for believing otherwise. Constance was depicted rather than Henry for the simple reason that she, not her husband, ` . She stands in her own right as an honored donor to the issued the privilege to Cefalu church, irrespective of what her son may have later done. Given the prominence of Bishop Johannes in the last panel, it is unlikely that its patron was Frederick, and, therefore, it is unlikely that the emperor would have been patron of the other representations as well. Returning to the description given in the Rollus rubeus, one will recall that the panels were listed in chronological order and that the second panel, placed between the images of Roger II and his successor, William I, was not described. Dating these panels in the Frederican period would not allow for any reasonable interpretation of who would have been represented in that panel. If, however, the first and the second panels are dated to the period of Roger II, then, based on Norman parallels, the logical assumption is that the second panel also depicted Roger II, presumably in an investiture scene. The two dedicatory panels at Santa Maria dell’Ammiraglio provide the clearest idea of the imagery that would have been expected for this period.45 Given the emphasis on donations seen in the other panels, and the fact that all other rulers were depicted just once, it is doubtful whether any image of Roger—other than that showing him as donor— would have been added more than a half century later. Furthermore, it is also possible that the images of the two Williams could have been contemporary to their actions in regard to the cathedral, or, perhaps, added in 1170 when Boso and canons of the church were stressing its ties to the ruling family and petitioning to have it fulfill its designated function as royal mausoleum. The image of Constance may have been added late in her life following her own donation to the church, though it could have also been executed as a pendant to the final panel, that of Frederick. In summary, the panels depicting Roger had likely been done during his reign, while the other four could have been added individually or as pairs—or even all at the same time under Bishop Johannes. It is impossible to ascertain the exact date of each part of the decoration.
44 Demus, Mosaics, 10 and 22 n. 36, suggests this as a possibility, excepting the dedicatory panel of Roger. See also H. Schaller, “Il rilievo dell’ambone della cattedrale di Bitonto,” AStPugl 13 (1960): 55; for German ¨r Kulturgeschichte 45 (1963): 295–312; repr. in Stupor Mundi: Zur Geschichte Friedrichs II. von trans. see Archiv fu Hohenstaufen, ed. G. Wolf (Darmstadt, 1966), 591–616. 45 Kitzinger, Mosaics of St. Mary’s, 197–206, for the dedicatory panel of George; 206–21, for the panels as a pair.
MARK J. JOHNSON
249
III. ANALYSIS The image of Roger and Leontius at Gerace and the collection of royal and imperial ` cathedral were rich and meaningful displays for portraits near the entrance of Cefalu the medieval visitor to these churches. Moreover, these pictures held a unique place in the art of this period, for there was nothing exactly like these decorations—an image of the ruler and the bishop, and a series of individual royal portraits holding scrolls with texts of privileges, which was located on the west facade of a church—to be found in the other churches of Norman Sicily or, apparently, anywhere else in twelfth-century Europe. Although this decoration in its totality appears to have been unique, in its particular details precise parallels may be found. In order to gain a clearer picture of the appear` and their place in the art of the period, ance of the decorations at Gerace and Cefalu several aspects must be examined, such as the tradition and use of royal donor portraits; the royal family or group portrait type; the inclusion of scrolls with privileges; and the location of such decoration in the church. As is well known, the art and culture of Norman Sicily resulted from the mixture of various cultures. The same is true for the panels ` , which included elements from both the Latin and the Byzantine of Gerace and Cefalu traditions. Nevertheless, the combination of all these elements in a single program was `. unique to Cefalu ` began with the portrait(s?) of Roger, with the first one As noted, the series at Cefalu described by Rogerius as showing the king holding a church model that he presents to Christ. This type of image had already had a long tradition in art before Roger’s time.46 From at least the sixth century on, the donor holding a model of his foundation is a common theme in the church art of both the East and the West. In Norman Sicily the only extant examples of this type are found in Monreale, where there are two: the mosaic panel of William II, discussed previously, and a sculpted capital in the cloister that shows the king presenting a model to the enthroned Virgin and the child.47 ` also The placement of the dedicatory panels on the facade of the church at Cefalu has Byzantine parallels from this period. The earliest dedicatory images often were to be found in the sanctuary of the church, which was a practice continued in later papal foundations in Rome.48 In Byzantine art from the tenth century on, however, it is common to find donor portraits in the narthex in the west part of the church. The donor images in the narthex of Hagios Nikolaos tou Kasnitzi in Kastoria may be cited as examples from the twelfth century.49 In Sicily, the dedicatory panels of Roger and George of Antioch in Santa Maria dell’Ammiraglio in Palermo, which are contemporary with ` , were originally located in the now destroyed narthex of the first decoration at Cefalu 50 the church. 46 See E. Lipsmeyer, “The Donor and His Church Model in Medieval Art from Early Christian Times to the Late Romanesque Period” (Ph.D. diss., Rutgers University, 1981). For imperial donors in Byzantine art, see Grabar, Empereur, 106–11; Velmans, “Portrait,” 121–23; Wessel, “Kaiserbild,” 818–21. 47 R. Salvini, The Cloister of Monreale and Romanesque Sculpture in Sicily (Palermo, 1962), 135–39. 48 For the papal donor images, see G. B. Ladner, Die Papstbildnisse des Altertums und des Mittelalters, 3 vols. (Vatican City, 1941–70); H. Belting, “Papal Artistic Commissions as Definitions of the Medieval Church in Rome,” in Light on the Eternal City, ed. H. Hager and S. Munshower (University Park, Pa., 1987), 13–30. 49 S. Pelekanidis and M. Chatzidakis, Kastoria (Athens, 1985), 50–65. 50 ´ urcˇic´, “The Architecture,” in ibid., 43. Kitzinger, Mosaics of St. Mary’s, 189, 208–11; S. C
250
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
` is a basilica without a narthex or an atrium, so its panels The cathedral of Cefalu were placed on the west facade, near the entrance, as an alternative. Similar solutions were used in other Byzantine churches that had no narthex.51 A contemporary example may be seen in the small basilica church of St. George at Kurbinovo, where dedicatory imperial images were placed high on the west facade.52 It is also true that royal imperial images during the twelfth century were not limited to facades and narthexes. The Gerace mosaic was in the north apse of its church, while the two royal mosaic panels at Monreale were located at the entrance to the sanctuary. Imperial portraits in Byzantine churches appear on facades and in narthexes but also elsewhere in churches. The fresco decoration of the rock-cut church known as the Great Pigeon House at C ¸ avus¸in in Cappadocia includes standing figures of the emperor Nikephoros Phokas (963–969), his wife Theophano, and other members of his family— who were not donors to this church—placed on the wall of the north apse.53 The imperial images at Hagia Sophia are scattered about the galleries and in the narthex, while Justinian’s image at San Vitale in Ravenna appears in the sanctuary. In short, there seems to have been no set policy regarding the placement of such images in the Byzantine world, and none in the Norman kingdom either. The mosaic at Gerace included two figures, a ruler and a bishop, which was an uncommon combination in the Byzantine or Western traditions, though not entirely unknown. The earliest surviving example of such a work of art is the Justinian panel in the mosaic decoration of San Vitale in Ravenna, datable to the period of 540–548.54 The emperor is shown together with a bishop (whose head was changed into the portrait of Maximianus after 546) as participants in a procession with several other men. Here, too, the messages of imperial support and authority are linked with, most likely, episcopal patronage. In nearby Classe a similar mosaic was installed in the apse of Sant’Apollinare sometime after 666. Shown are Constantine IV Pogonatus (668–685), his brothers Heraclius and Tiberius, and his son Justinian II, together with Archbishop Maurus of Ravenna, who has a nimbus, with his successor, Reparatus, and three deacons. All stand formally as Constantine hands a closed scroll, inscribed “privilegia,” to Reparatus.55 An example of a picture showing a ruler and an ecclesiastical leader together in Constantinople is 51 See L. Hadermann-Misguich, “Une longue tradition byzantine: La de´coration exte´rieure des ´eglises,” Zograf 7 (1976): 5–10. 52 L. Hadermann-Misguich, Kurbinovo: Les fresques de Saint-Georges et la peinture byzantine du XIIe sie`cle (Brussels, 1975), 267–75, figs. 2, 138–41, 146; C. Grozdanov and D. Bardzieva, “Sur les portraits des personnages historiques `a Kurbinovo,” ZRVI 33 (1994): 61–84. 53 ¨ B 33 (1983): 301–39; C. Jolivet-Le´vy, Les ´eglises L. Rodley, “The Pigeon House Church, C ¸ avus¸in,” JO byzantines de Cappadoce (Paris, 1991), 20–21, pl. 23, who also notes imperial images in sanctuaries in the chapel of St. Erasmus at Ohrid and in the cathedral at Faras. I wish to thank Sarah Brooks for calling the Pigeon House church to my attention. 54 The literature on the mosaics of San Vitale is vast. An important recent contribution is made by I. Andreescu-Treadgold and W. Treadgold, “Procopius and the Imperial Panels of San Vitale,” ArtB 79 (1997): 708–23. 55 O. von Simson, Sacred Fortress: Byzantine Art and Statecraft in Ravenna (Chicago, 1948; repr. Princeton, N.J., 1987), 59–60; M. C. Pela`, La decorazione musiva di S. Apollinare in Classe (Bologna, 1970), 160–68; F. Deichmann, Ravenna: Hauptstadt des spa¨tantiken Abendlandes. Kommentar (Wiesbaden, 1976), 2:273–79; J. Beckwith, Byzantine Art, 2d ed. (Harmondsworth, 1979), 119–20.
MARK J. JOHNSON
251
provided by a description of the decoration of the destroyed Chrysotriklinos in the Great Palace. Near its door was an image of the emperor Michael III (842–867) and Patriarch Photios.56 Two examples that are closer in date to the mosaic at Gerace may be cited. Isaac II Angelos (1185–95) is the emperor depicted in both. The first is a contemporary epigram of Theodore Balsamon that describes an icon in possession of the bishop of Sidon having portraits of the emperor, the patriarch (probably Dositheos), and the bishop himself.57 The second example is found on the west facade of the church of St. George in Kurbinovo. The upper part of the facade is decorated with damaged frescoes of four figures (Fig. 11). The two on the left are probably Isaac II Angelos and his wife, Empress Margherita, while those on the right are a cleric, most likely the archbishop of Ohrid in whose jurisdiction Kurbinovo was, and a nobleman, presumably the donor.58 Although secular and ecclesiastical leaders appearing together in Western art can be seen in such early medieval examples as the Vivian Bible of Charles the Bald and the Gospel Book of Otto III, in neither case can it be said that the ruler and the ecclesiastic are equals.59 In both, the ruler is depicted on a much larger scale, with the bishop shown smaller and off to one side. The Byzantine examples, on the other hand, stress the unity of church and state, as manifested in the persons of the emperor and the bishop depicted on the same scale. Clearly, the mosaic at Gerace is to be seen as part of the Byzantine tradition—limited as this tradition may have been—in which the secular and the religious powers are linked in an image of united authority and shared prestige. ` cathedral consisted of Although the earliest decoration of the facade of the Cefalu the image(s) of the founder, the subsequent addition of the other images, whether by 1170 or 1215, resulted in a group of individual figures that collectively formed a dynastic group portrait. For this aspect of the decoration, too, parallels may be found in the Byzantine and the Western artistic traditions.60 Imperial Byzantine group portraiture can be divided into two types: images of indi` , and images of couples or families, somevidual rulers linked by proximity, as at Cefalu times in the presence of Christ, Mary, or some other saint. Two examples of the first type may be noted in Constantinople. Anthony of Novgorod visited the capital in 1200 and stated that in the gallery at Hagia Sophia “are painted all the patriarchs and emperors, as many of them as there have been at Constantinople. . . .”61 It is not clear, however, if he had truly seen portraits of “all” the emperors, which is a series now lost, or if he was simply referring to the series of mosaic panels that decorated the gallery and were exeAnthologia graeca 1.106, trans. C. Mango, The Art of the Byzantine Empire, 312–1453 (Englewood Cliffs, N.J., 1972), 184. 57 Epigram 17, in K. Horna, ed., “Die Epigramme des Theodoros Balsamon,” WSt 25 (1903): 184–85; trans. Magdalino and Nelson, “Emperor,” 152–53. 58 See note 52. 59 See Diebold, “Ruler Portrait”; Schramm and Mu ¨ therich, Kaiser. None of the known ruler images in the West is monumental; instead, most are found in manuscripts. 60 Grabar, Empereur, 26–30; idem, “Une pyxide en ivoire `a Dumbarton Oaks: Quelques notes sur l’art profane pendant les derniers sie`cles de l’Empire byzantin,” DOP 14 (1960): 121–46, esp. 127–34, repr. in idem, L’art de la fin de l’antiquite´ et du Moyen Age (Paris, 1968), 229–49; I. Spatharakis, The Portrait in Byzantine Illuminated Manuscripts (Leiden, 1976), 251–53. 61 Trans. Mango, Art of the Byzantine Empire, 237. 56
252
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
cuted in different periods, of which a few remain extant. These include the portrait of a standing, frontal emperor Alexander in the north gallery, done in 912–913; the panel of Constantine IX Monomachos (1042–55) and Zoe flanking an enthroned Christ (Fig. 12); and a panel depicting the Virgin and the child flanked by John II Komnenos (1118–43) and Irene, with their son Alexios, named co-emperor in 1122, shown on the return of a buttress at a right angle to the rest of the panel (Fig. 13).62 The other series was found in the church of St. George Mangana founded by Constantine IX Monomachos. According to the Russian Anonymous, who visited the church in about 1390, its decoration included an icon painted with portraits of eighty emperors and placed in the northern part of the narthex.63 There were few imperial group portraits in churches during the early centuries of Christianity, but during the eleventh and twelfth centuries imperial family portraits became an important part of church decoration in the Byzantine world. The emperors and members of their families are normally shown in the role of donors, either as founders or to commemorate gifts that they made to churches. This is the function of the family panels in Hagia Sophia, in which the emperors hold bags containing gifts of gold, while the empresses hold closed scrolls of donation. The emperor Manuel I Komnenos (1143–80), King Roger’s contemporary and rival, seems to have especially favored the practice of having himself depicted in his foundations, and embellished the idea by adding portraits of his ancestors. Such was the case in the monastery of St. Mokios in Constantinople, to which he added a hall to serve as a refectory. In it were representations of himself, his grandfather Alexios I, his father John II, and Basil II (976–1025), all of whom had been patrons of the monastery. The decoration no longer survives but is described in a thirteenth-century text.64 The same text contains a description of a similar series of imperial representations that were once found in the monastery church of the Virgin Hodegetria built by Manuel in the capital. In the narthex were depicted Manuel’s deeds and a group of seven emperors and ancestors: Manuel, Constantine X Doukas (1059–67), Romanos IV Diogenes (1068–71), Michael 62 C. Mango, Materials for the Study of the Mosaics of St. Sophia at Istanbul, DOS 8 (Washington, D.C., 1962), 27–29, 46–47; idem, “The Mosaics of Hagia Sophia,” in H. Ka¨hler, Hagia Sophia, trans. E. Childs (New York, 1967), 55–58; P. A. Underwood and E. Hawkins, “The Mosaics of Haghia Sophia at Istanbul, 1959–60: The Portrait of the Emperor Alexander,” DOP 15 (1961): 187–217; T. Whittemore, The Mosaics of Haghia Sophia at Istanbul: Third Preliminary Report. Work Done in 1935–1938: The Imperial Portraits of the South Gallery (Oxford, 1942); R. Cormack, “Interpreting the Mosaics of S. Sophia at Istanbul,” Art History 4 (1981): 131–49, repr. in idem, The Byzantine Eye: Studies in Art and Patronage (London, 1989), pt. 8, “Additional Notes and Comments,” 13–14; idem, “The Emperor at St. Sophia: Viewer and Viewed,” in Byzance et les images, ed. A. Guillou and J. Durand (Paris, 1994), 223–53. The panel of Constantine and Zoe has elicited the most scholarly interest; see, most recently, I. Kalavrezou, “Irregular Marriages in the Eleventh Century and the Zoe and Constantine Mosaic in Hagia Sophia,” in Law and Society in Byzantium, Ninth–Twelfth Centuries, ed. A. Laiou and D. Simon (Washington, D.C., 1994), 241–57. 63 “The ‘Anonymous Description of Constantinople,’” ed. and trans. G. Majeska, Russian Travelers to Constantinople in the Fourteenth and Fifteenth Centuries, DOS 19 (Washington, D.C., 1984), 140–41, 366–71 (commentary). See also C. Mango, “The Legend of Leo the Wise,” ZRVI 6 (1960): 59–93, esp. 76–78, repr. in idem, Byzantium and Its Image: History and Culture of the Byzantine Empire and Its Heritage (London, 1984), pt. 16, with addendum, 3–4, who suggests that the icon should be dated “by the middle of the twelfth century” (p. 76). 64 S. Lampros, ed., “ JO Markiano` " Kw'dix 524,” Ne´ o" JEll. 8 (1911): 127–28; trans. Mango, Art of the Byzantine Empire, 226–27. Additional examples are cited in Magdalino and Nelson, “Emperor,” 135–47.
MARK J. JOHNSON
253
VII Doukas (1071–78), Nikephoros III Botaneiates (1078–81), Alexios I Komnenos (1081–1118), and John II.65 This type of decoration was also found outside of the capital and became increasingly common in later centuries. The cathedral of St. Sophia in Kiev, which is so dependent upon Byzantine models for its design, construction, and decoration, has as part of its original program of ca. 1043 a series of fresco portraits of Prince Jaroslav the Wise and his family.66 The south facade of the Panagia Mavriotissa in Kastoria is covered with frescoes in which the contemporary portrait of Michael VIII Palaiologos (1259–82) is combined with a posthumous image of Alexios I Komnenos, the original founder of the church.67 The richest iconographic usage of this tradition is found in the painted decoration, beginning in the thirteenth century, of royal Serbian foundations where donor images are combined with displays of dynastic lineages.68 The monastery at Studenica has several such images. Remains of a royal family portrait were recently found in the passageway of the gate tower and dated to 1208–9. In the main church of the Virgin there is a dedicatory fresco of about 1235 that shows King Radoslav holding a model, with two of his predecessors. The small church of SS. Joachim and Anna, also known as the King’s Church, was added to the monastery by King Milutin and decorated in 1313/4. On the south wall the king and his wife, Simonis, are led to Christ by the saints; on the opposite wall are two Serbian royal saints, Sava and Simeon, the king’s ancestors.69 Such pictorial displays were especially favored by King Milutin as part of a conscious attempt to legitimize his rule. Among his other works one may cite the cathedral church of the Virgin Ljevisˇka at Prizren, which was rebuilt and redecorated between 1307 and 1313. Its narthex contains a collection of images of the king and his royal ancestors.70 In Italy, such family images were also found during the thirteenth century in the patronage of Frederick II. A group of figures painted in fresco in the grotto church of Santa Margherita outside of Melfi have recently been dated to about 1237 and identified as portraits of Frederick II; his wife, Isabella of England; and Conrad, the son of his 65 Cod. Marcianus gr. 524, fol. 108r; ed. Lampros, “Markianos,” 148 ff; trans. Mango, Art of the Byzantine Empire, 227–28. See also R. Janin, La ge´ographie eccle´siastique de l’Empire byzantin, vol. 1, Le sie`ge de Constantinople et le patriarcat œcume´nique, pt. 3, Les ´eglises et les monaste`res, 2d ed. (Paris, 1969), 199–207, esp. 200. 66 V. Lazarev, Old Russian Murals and Mosaics from the XI to the XVI Century (London, 1966), 47–48, 272; idem, La pittura bizantina (Turin, 1967), 154–55. 67 Pelekanidis and Chatzidakis, Kastoria, 50–65. For the identification of the emperors, see S. KalopissiVerti, Dedicatory Inscriptions and Donor Portraits in Thirteenth-Century Churches of Greece, DenkWien 226 (Vienna, 1992), 28–29, 96–97; T. Papamastorakis, “”Ena eijkastiko´ jEgkw´ mio tou' Micah´ l H⬘ Palaiolo´ gou,” Delt.Crist. Arc. j JEt., ser. 4, 15 (1989–90 [1991]): 221–40. 68 Wessel, “Kaiserbild,” 844–48; Velmans, “Portrait,” 106–14; S. Radojcˇic´, Portreti srpskih vladara u srednjem veku (Skoplje, 1934). These paintings are also discussed and illustrated in Velmans, Peinture; V. Djuric´, Byzantinischen Fresken in Jugoslawien (Munich, 1976); R. Hamann-MacLean and H. Hallensleben, Die Monumen¨hen 14. Jahrhundert (Giessen, 1963). talmalerei in Serbien und Makedonien vom 11. bis zum fru 69 ´ urcˇic´, “The Nemanjic´ Family S. Cirkovic´ et al., Studenica Monastery (Belgrade, 1986), 82–85, 113. Also S. C Tree in the Light of the Ancestral Cult in the Church of Joachim and Anna at Studenica,” ZRVI 14–15 (1973): 191–95. 70 ˇ ivkovic´, Bogorodica D. Panic´ and G. Babic´, Bogorodica Ljevisˇka, 2d ed. (Belgrade, 1988), 108–9, 128; B. Z Ljevisˇka: Crtezˇi freska (Belgrade, 1991), pls. 51, 53. Another significant example of a royal “portrait gallery” is found in the inner narthex of the katholikon of Hilandar monastery on Mt. Athos where figures of Serbian rulers are juxtaposed with Byzantine emperors. See V. Djuric´, “Les portraits de souverains dans le narthex de Chilandar: L’histoire et la signification,” Hilandarski zbornik 7 (1989): 105–32.
254
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
second wife, Yolanda of Brienne (d. 1228).71 Another dynastic Hohenstaufen portrait is found on a side panel of the ambo of Bitonto cathedral, which is dated to 1229. Four figures in imperial regalia are shown there, but the quality of the representation is so crude that the figures’ identification is both difficult and disputed, though most interpretations are that Frederick II is depicted here, with at least one ancestor.72 This interest in family group portraiture that was exhibited during the reign of Frederick may be evi` were done during his reign. Conversely, it could be ardence that the panels at Cefalu gued that the presence of such portraits there may have influenced these later works of art. ` that merits consideration is the fact that The final aspect of the decoration at Cefalu the royal figures all hold scrolls with various privileges and gifts recorded on them. Figures with scrolls are not uncommon in Byzantine and Norman art, with the founder’s panel of George the Admiral and Mary at Santa Maria dell’Ammiraglio providing a notable example. There is, however, a major distinction to be made here. The text on the scroll that is held by the Virgin is a prayer of intercession on George’s behalf, while all other examples of figures with scrolls in the art of Norman Sicily are saints or prophets ` , with the exception of Fredwho bear scriptural texts. As discussed, the figures at Cefalu erick, hold scrolls that are charters of donations or privileges, thus taking on the character of “pictorial legal documents,” which is a distinction shared by none of the other royal images of Norman Sicily. While not completely unknown, the image of a royal or imperial donor holding a scroll was not frequent in Byzantine and Western medieval art, though it became more common in the twelfth century.73 Perhaps the most significant example was found in the papal audience hall that was built and decorated by Calixtus II (1119–24) as part of the Lateran palace complex. In a display of secular powers unequaled in papal art, the hall was decorated with a series of frescoes that showed the deeds of Calixtus. Among these was a painting, known from a sixteenth-century drawing executed for the antiquarian Onofrio Panvinio, that depicted an enthroned Calixtus resting his feet on the back of one of his enemies, Burdinus, and flanked by several clerics who held raised crosses and staffs like so many ancient Roman soldiers with standards. The pope reached his left hand out to receive an open scroll from the standing emperor Henry V (Fig. 14). The text of the scroll bore the opening words of the Concordat of Worms, which had been A. Ciarallo and L. Capaldo, Federico II a Melfi: Ritrovato il vero volto dell’Imperatore (Naples, 1994). Schaller, “Rilievo,” identifies the figures as Frederick I, Henry VI, Frederick II, and Conrad IV. H. Thelen, “Ancora una volta per il rilievo del pulpito di Bitonto,” in Federico II e l’arte del Duecento italiano, ed. A. M. Romanini, Atti della III settimana di studi di storia dell’arte medievale dell’Universita` di Roma (Galatina, 1980), 217–25, argues that the first, seated, figure is female and may be a personification of the city of Bitonto. See also E. Paratore, “L’ambone di Bitonto e la predica dell’Abate Nicola di Bari,” in ibid., 227–35; Belli D’Elia, Puglia, 260–63; P. C. Claussen, “Bitonto und Capua: Unterschiedliche Paradigmen in der Dar¨ppingen, stellung Friedrichs II,” in Staufisches Apulien, Schriften zur staufischen Geschichte und Kunst 13 (Go 1993), 77–124. 73 In general, see V. Djuric´, “Portraits of Byzantine and Serbian Rulers Granting Charters,” in Esfigmenska povelija despota Djurdja: The Esphigmenou Charter of Despot Djuradj (Belgrade, 1989), 80–105, which is a revision of his earlier study, “Portreti na poveljama vizantijskih i srpskih vladara,” ZbFilozFak 7.1 (1963): 251–72, with French summary. 71 72
MARK J. JOHNSON
255
the product of a compromise between the pope and the emperor but was depicted here as a triumph for Calixtus.74 This fresco may have influenced other art in the twelfth century, particularly in the realm of manuscript illumination, such as in the Chronicon Vulturnense, which was decorated in 1124–30 and produced in a monastery that was then in Norman territory in southern Italy. It contains more than twenty donation images, which typically show the grantor of a privilege offering an open scroll to either the abbot or St. Vincent, the patron saint of the monastery (Fig. 15).75 Other contemporary Western examples of the same period may be cited. The Codex Traditionis of the monastery at Formbach has two illustrations of donors and privileges. In one a seated emperor, Lothar III (1133–37), is shown handing a scroll of privileges to a monk, while on the next page Pope Innocent II is depicted with the scroll.76 A Norman example is found in the Chartulary of Mont-Saint-Michel of 1154–58, in which the donation of Duke Richard II of Normandy is depicted in an image with the seated ruler, an open scroll held by a scribe seated to his left, and, below, a group of monks who receive the gift. In the same book is a similar representation of Duchess Gonnor, the second wife of Richard I, who hands her open scroll of donation to an abbot. Both are posthumous portraits, depicting donations made in the 1020s.77 Perhaps the most striking surviving example of a pictorial legal document is to be found in the fresco decoration of the lower church of the Sacro Spello monastery at Subiaco. In 1202 Pope Innocent III spent August and September at the monastery and reformed it. It was perhaps due to his stay that later, on 24 February 1203, he issued a bulla to the monastery, granting it additional income.78 To commemorate this honor, an Biblioteca Apostolica Vaticana, Barb. lat. 2738, fol. 104r. See O. Panvinio, De praecipuis urbis Romae sanctioribusque basilicas quas septem ecclesias vulgo vocant (Rome, 1570), 175–76; G. B. Ladner, “I mosaici e affreschi ecclesiastico-politico nell’antico palazzo lateranense,” RACr 12 (1935): 265–92, repr. in idem, Images and Ideas in the Middle Ages: Selected Studies in History and Art, 2 vols. (Rome, 1983), 1:347–66, 2:1026–27 (addendum); idem, Papstbildnisse, 1:195–201; C. Walter, “Papal Political Imagery in the Medieval Lateran Palace,” CahArch 20 (1970): 155–76, esp. 163–65; ibid., 21 (1971), 109–36, esp. 119–23; idem, “Political Imagery: Osmosis between East and West,” BSl 54 (1993): 211–17, where Ladner reiterates his earlier study and suggests a Byzantine origin for the iconography of this scene; I. Herklotz, “Die Beratungsra¨ume Calixtus’ II. im Lateranpalast und ihre Fresken: Kunst und Propaganda am Ende des Investiturstreits,” ZKunstg 52.2 (1989): 145–214; M. Stroll, Symbols as Power: The Papacy Following the Investiture Contest (Leiden, 1991), 16–35. 75 Biblioteca Apostolica Vaticana, Barb. lat. 2724; Chronicon Vulturnense, ed. V. Federici, FStI 58–60 (1925– ˜ oz, “Le miniature del ‘Chronicon Vulturnense’ (Cod. Barb. lat. 2724),” 40). For its illuminations, see A. Mun Bullettino dell’Istituto storico italiano 30 (1909): 75–90; M. Avery, The Exultet Rolls of South Italy (Princeton, N.J., 1936), 2:41–43, pl. CC; Ladner, Papstbildnisse, 1:232–40, pl. 23; F. Riccioni, “Un codice da rivalutare: Il Chronicon Vulturnense,” Miniatura: Arte dell’illustrazione e decorazione del libro 3–4 (1990–91): 33–50. 76 Munich, Bayerisches Hauptsarchiv, Abt. I, Formbach KL 1, 4–5. See Schramm and Mu ¨ therich, Kaiser, ¨nder im Bild 1100 bis 1350, Vero ¨ffentlichungen 257–58; C. Sauer, Fundatio und Memoria: Stifter und Klostergru ¨ttingen, 1993), 66–88. des Max-Planck-Instituts fu ¨ r Geschichte 109 (Go 77 Avranches, Bibliothe`que Municipale, Ms. 210, fols. 19 and 23, respectively. See A. Boinet, “L’illustration du Cartulaire du Mont-Saint-Michel,” BEC 70 (1909): 335–43; M. Dosdat, L’enluminure romane au Mont SaintMichel, Xe–XIIe sie`cles (Rennes, 1991), 71–75, figs. 3–4. 78 For the pope’s stay at the monastery, see Chronicon Sublacense (A.A. 593–1369), ed. R. Morghen, RIS, n.s., 24.6 (1927): 34. For the text of the privilege, see C. Margarini, Bullarium Casinense seu constitutiones summorum Pontificum, imperatorum, regum, principum . . . (Venice, 1650–70), 2:232–33; also V. Federici, “La biblioteca e l’archivio,” in I monasteri di Subiaco (Rome, 1904), 2:51, no. CCLII. 74
256
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
artist was commissioned to paint the text of the privilege on an open scroll. Holding the scroll are Innocent to the right and Romanos, the then abbot (1192–1216), to the left, next to a seated St. Benedict who helps to hold the scroll (Fig. 16). For some unknown reason, a bust painting of the pope was added above the scroll during the third quarter of the century.79 This later figure also appears to be holding the scroll, and is so large that it dominates the painting. In Byzantine art, several examples of combining portraits of donors with documents may be cited, including two of the imperial panels of Hagia Sophia. The panel of Constantine IX Monomachos and Zoe is offertory in nature, with the emperor holding a moneybag, and Zoe a scroll that seems to be a charter of donation, but is unopened.80 The inscription simply identifies the emperor but has the function of giving authority to the charter. In the panel of John II Komnenos and Irene, it is again the empress who holds a closed scroll of donation. ` is a fresco found in the exonarthex of the Very similar to the decoration at Cefalu monastery church of the Virgin in Apollonia, Albania, which contains a family portrait of Michael VIII Palaiologos with his wife, Theodora, their son, Andronikos II (1282– 1328), and his son, Michael IX (1294/5–1320) (Fig. 17).81 The emperor hands a model of the church to the seated Virgin and the child; above and to the sides of the figures is the text of the donation chrysobull, though it is not contained within a scroll. The date of the exonarthex is disputed, leaving the possibility that the dedicatory fresco was originally on ` are notable: in both cases one the west facade of the church. The parallels with Cefalu finds a dedicatory panel with several family members, the founder who presents a church model, and the pictorial legal document, all depicted near the entrance on the west wall of the church. A Serbian image, though later in date, also provides a close parallel to the facade ` . This is a fresco panel of the despot Stefan Lazarevic´ in his foundadecoration of Cefalu tion, the church of the Holy Trinity at Manasija (Resava) in Serbia, which was consecrated in 1418. The dedicatory picture combines several of the features found in the two panels ` , which date two and a half centuries earlier. It shows the despot standof Roger at Cefalu ing in a frontal position, being crowned by Christ, as angels hand him royal regalia in an investiture scene. In his left hand he holds both a model of the church, which is being presented to the Holy Trinity (depicted as seated angels), and an open scroll of donation with text written on it (Fig. 18).82 Although there are no additional panels that show 79 Ladner, Papstbildnisse, 2:68–72; G. Matthiae, Pittura romana del Medioevo, 2 vols. (1966; repr. Rome, 1988), 2:112–13, 284; M. T. Cristiani Testi, “Gli affreschi del Sacro Speco,” in I monasteri benedettini di Subiaco, ed. C. Giumelli (Milan, 1982), 95 ff, esp. 106, figs. 94–95; pp. 110–11, 130, 237 n. 71; idem, “Consolo: Il maestro del busto di Innocenzo III e i collaboratori negli affreschi del S. Speco di Subiaco,” in Roma anno 1300: Atti della IV settimana di studi di storia dell’arte medievale dell’Universita` di Roma “La Sapienza” (19–24 maggio 1980), ed. A. M. Romanini (Rome, 1983), 403–7. 80 Mango, “Mosaics,” in Ka¨hler, Hagia Sophia, 56; N. Oikonomides, “The Mosaic Panel of Constantine IX and Zoe in Saint Sophia,” REB 36 (1978): 219–32, repr. in idem, Byzantium from the Ninth Century to the Fourteenth Century (Brookfield, 1992), art. 15, 224. 81 For Apollonia, see H. and H. Buschhausen, Die Marienkirche von Apollonia in Albanien: Byzantiner, Normannen und Serben im Kampf um die Via Egnatia, ByzVindo 8 (Vienna, 1976), 143–82. 82 V. Lazarev, Storia della pittura bizantina (Turin, 1981), 392; Velmans, Peinture, 72–73, fig. 33; Djuric´, Fresken, 150–52; S. Tomic´ and R. Nikolic´, eds., Manasija: Istorija—zˇivopis, Saopsˇtenja 6 (Belgrade, 1964), 55–56. For additional Serbian examples of rulers with charters, see Djuric´, “Portraits of Rulers.”
MARK J. JOHNSON
257
descendants holding scrolls as well, the similarities between the images of Stefan and Roger are striking and point to the common ultimate origins of all such scenes. In discussing the similarity between the panel of Roger at Santa Maria dell’Ammiraglio and an ivory relief of a Byzantine emperor, probably Constantine VII, receiving his crown from Christ, Ernst Kitzinger noted that the close resemblance between the two would be inexplicable unless it were assumed that a Byzantine design served as model.83 The same argument may be made here. The panels that once decorated the facade of ` cathedral, the imperial painting in Apollonia, and the royal images of Serbia all Cefalu had a common ancestor. In all probability, this ancestor was in Constantinople, though no comparable image in the capital now survives. There was, however, a church that ` as any example. This was the depossessed a decoration as close to the one at Cefalu stroyed church of St. Mary Peribleptos, previously mentioned for its exterior decoration.84 The church was founded by Romanos III Argyros (1028–34), who was buried there. Nikephoros Botaneiates restored the building, and his tomb was also found in the church. A final restoration was undertaken by Michael VIII Palaiologos. Two descriptions of imperial images in the church exist, though differences in these accounts have caused some confusion regarding the identity of their figures. On his visit to the church in 1403, Clavijo saw and described in detail an imperial panel: As one enters the body of the church, on the left-hand side are represented many images, among them one of St. Mary, and next to it, on one side, is an image of an emperor, and on the other side, the image of an empress, and at the feet of the image of St. Mary are represented thirty castles and towns, the name of each one being written in Greek. And they said that these towns and castles belonged to the domain of this church and had been granted to it by an emperor called Romanus. . . . Here are placed certain privileges on leather [parchment?], sealed with wax and lead seals, which are said to be the privileges received by this church over the aforesaid towns and castles.85
In the sixteenth century, Johannes Leunclavius wrote that toward the western part of the church was a “picture with Michael Palaiologos, Theodora, and their son, Constantine, placed between them,” with an inscription identifying the members of the family.86 There is some confusion as to which imperial couple is represented, with some scholars conflating the two descriptions and suggesting that it was Michael and Theodora who were depicted here.87 However, it is clear that the descriptions refer to two different pictures. Leunclavius omits any mention of Mary, the castles and towns, and the privileges, whereas Clavijo says nothing about the son. In fact, the son and Mary impossibly occupy the same position in the two pictures. Therefore, the panel described by Clavijo was Kitzinger, Mosaics of St. Mary’s, 191. ´ glises, 218–22; S. Cirac Estopan ˜ an, “Tres On the church, see Majeska, Russian Travelers, 276–83; Janin, E ˜ oles en el an ˜ o 1403,” REB 19 (1961): 358–81, esp. monasterios de Constantinopla visitados por Espan 374–77. 85 “. . . Estaban figurados trenta castillos e ciudades . . . ; e que estaban colgados unos privilegios de cuero ˜ an, “Monasterios,” 374–75; R. Gonzalez de sellados con sellos de cera e de plomo . . .” See Cirac Estopan Clavijo, Embassy to Tamerlane, as translated in Mango, Art of the Byzantine Empire (as in note 56), 217. 86 J. Leunclavius, Pandectes historiae Turcicae, chap. 51, PG 159:773 ⫽ idem, Annales sultanorum Othmanidarum, 2d ed. (Frankfurt, 1596), 137. 87 ´ glises, 218. Mango, Art of the Byzantine Empire, 217 n. 164, says that “they may have been” Michael Janin, E ˜ an, “Monasterios,” 374, says that Romanos and Zoe were represented. and Theodora. Cirac Estopan 83 84
258
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
different, and, since it was a foundation panel, it seems more likely that Romanos and Zoe were depicted there, together with the evidence of their donations. It may be remembered that the panel of Zoe and Constantine IX in Hagia Sophia has been altered, and it is possible that it originally depicted Zoe with Romanos, thus making it a striking pendant to the panel at the Peribleptos monastery. ` , providing an interesting This panel would then have preceded the images at Cefalu and important antecedent. It contained a dedicatory image of Mary and the founder, who was accompanied by his wife in a family portrait. The representations of the castles and towns may be likened to the models held by many donors, though here there were too many to be held by one person. The fact that the panel contained a record of the donation of the castles and towns to the church is related to the presence of information about the donation of a town in the inscription in the panel of Constance. The inclusion of the actual documents of donations in the picture at the Peribleptos church is apparently unique, but it helps to explain why some later donor panels would include painted versions of such documents. The original documents could be better preserved and protected elsewhere, but the link between the donor and the privilege remained strong and tangible. The image of the donor in this context was certainly meant to stress the authenticity and authority of the privilege given in the written document. For this same reason, in the fourteenth century the practice evolved of including images of the emperor, sometimes with his family, in chrysobulls.88 It is not possible to state, and probably unreasonable to maintain, that the decoration of the Peribleptos church was the only source for the similar scenes at Apollonia, Man` . It is important, however, to recognize that this type of decoration did asija, and Cefalu exist in Constantinople in at least one church, and may well have once existed in other imperial foundations in the capital. It is this tradition that was ultimately followed in other Byzantine churches outside the capital, as well as in areas such as Serbia and Sicily where rulers modeled themselves and their concepts of royalty and its expression in art on Byzantine prototypes. An important distinction must be drawn with regard to the images in the Norman kingdom. While many of the ruler images discussed here were the result of direct imperial patronage, such involvement was not necessarily the case for the mosaic at Gerace or ` . Given the importance accorded to Leontius in the for the facade decoration at Cefalu mosaic, it was probably the product of episcopal patronage. While it is likely that the ` was (were) probably part of the original decoration that image(s?) of Roger at Cefalu resulted from patronage of that king, there is no proof that any of the later rulers were involved in the creation of their own images. Rather, it is possible that patrons of the last four images were the bishops of the cathedral. The images of the two Williams might be attributable to Bishop Boso and done at a time when the bishop and his canons were fighting to maintain the ties between the kings and their church and to fulfill the church’s function of royal mausoleum. The representation of Frederick and possibly that of Constance could be attributed to Bishop Johannes, being executed at a moment when, al` , the bishop though it had become clear that no ruler was going to be buried at Cefalu 88 For these images, see Velmans, “Portrait,” 104–6; Spatharakis, Portrait, 184–89; Djuric´, “Portraits,” passim.
MARK J. JOHNSON
259
and his clergy were still proud of the church’s royal foundation and ties and may have been attempting to exploit them. The emphasis given to the texts of donations in the panels is, most likely, evidence of their episcopal rather than royal patronage. Nevertheless, even if the bishops were indeed the patrons of the last four panels, the models for ` do these images were royal. Though certain individual aspects of the program at Cefalu share similarities with Western, and especially papal, examples, the closest parallels have been seen to be with the Byzantine world. ` The recognition and examination of the lost decoration at Gerace and Cefalu broaden our understanding both of Norman art in Sicily and of Byzantine art. While previous discussions about royal images in Norman art have focused exclusively on the three surviving mosaics and one carved capital, it is now clear that there were three, possibly four, monumental images of Roger II in his kingdom, as well as a total of ten monumental images of Norman and Hohenstaufen rulers of southern Italy. The lost panels demonstrate once again the importance that the royal image had in Sicily, and the dependence of such images upon Byzantine models as well as their unique combination of individual elements from that tradition. In this type of decoration, the series of royal images bore witness to the importance of the links between the cathedrals of Gerace ` and the ruling family, and to the significance attributed to those links by the and Cefalu royal donors and by the bishops and canons of these cities. It was for this reason that the decoration was created, and it was for the same reason that Bishop Thomas commis` panels and the transcription of sioned Rogerius to produce the description of the Cefalu their pictorial legal documents, thus preserving them in a fashion to the present day. Brigham Young University
Appendix Excerpts from the Rollus rubeus (ed. Mirto, 26–32)
Cuius picture tenor per me, qui supra, notarium puplicum adnotatus per omnia talis est: In nomine Domini, amen. Anno dominice incarnationis millesimo tricentesimo vicesimo nono, menssis septembris vicesimo sexto eiusdem tertiedecime indictionis. Nos Primus de Primo iudex civitatis Ceph(alud)i et Rogerius notarii Gulielmi de Mistretta puplicus eiusdem civitatis notarius in presentia infrascriptorum testium ad hoc vocatorum et rogatorum notum facimus et testamur quod rev(er)endus in Christo pater et dominus, dominus Thomas, Dei gratia cephaludensis episcopus electus et confirmatus, fecit nos ad sui presentiam evocari asserans quod timens ne scriptura infrascripta modo aliquo deleatur [aqua vel] antiquitate et memoria regalium donationum factarum [sancte] / cephaludenssi ecclesie depicto albo pariete valeat deperire, ad perpetuam rei memoriam reservanda nobis obnixe requisivit nostrum officium implorando ut talis scriptura in pariete porte regum in ipsius ecclesie campanario picta in publicam deberemus redigere notionem. Nos autem, attendentes iustam esse requisitionem ipsius dicti d(omi)ni episcopi, adimplere curavimus quod quesivit. Et quia ipsam scripturam vidimus et legimus non deletam nec etiam vitiatem, nichil adendo vel minuendo nec etiam inmutando, [set in forma propria seu figura existendo, transcripsimus et in forma puplica redegimus]. Cuius figura talis est: Salvator noster in sua maiestate sedens cum sinistra manu recepit ecclesiam pictam et cum alia destra signat cruce, quam ecclesiam rex Rogerius conditor ipsius ecclesie offert cum manu destra eandem ecclesiam, indutus regalibus vestimentis et coronatus. Et super capud dicti regis scriptum est: Rogerius rex. Et cum sinistra cartam tenet in manibus scriptam. Cuius scripture tenor per omnia talis est: Suscipe, Salvator, ecclesiam et civitatem Cephaludi cum omni iure et libertate sua. Nichil in civitate preter felloniam, proditionem, homicidium nobis et nostris successoribus reservamus. Unde ad certitudinem presentium et futurorum memoriam presens scriptum testium amminiculo roboramus. ⫹ Ego Primus de Primo, qui supra, iudex, predictam scripturam in pariete porte regum in ipsius ecclesie campanario depictam seu scriptam vidi, legi, interfui et testor. ⫹ Ego Nicolaus de iudice Iacobo predictam scripturam in pariete porte regum in ipsius ecclesie campanario depictam seu scriptam vidi, legi, interfui et testor. ⫹ Ego Franciscus de Bonaquisto predictam scripturam in pariete porte regum depictam vidi, legi, interfui et testor. ⫹ Ego frater Andreas de Sancto Mauro canonicus cephaludensis ecclesie vidi, legi et testor. ⫹ Ego Rogerius notarii Gullielmi de Mistretta puplicus civitatis Cephaludi notarius predictam picturam parietis porte regum vidi et legi et in formam puplicam redegi et meo signo signavi.
MARK J. JOHNSON
261
In nomine Domini, amen. Anno dominice incarnationis millesimo tricentesimo vicesimo nono, menssis septembris vicesimo eiusdem tertiedecime inditionis. Nos Primus de Primo iudex civitatis Cephaludi et Rogerius notarii Gullielmi de Mistretta puplicus eiusdem civitatis notarius in presentia infrascriptorum testium ad hoc vocatorum et rogatorum notum facimus et testamur quod reverendus in Christo pater et dominus, dominus Thomas Dei gratia cephaludenssis episcopus electus et confirmatus fecit nos ad sui presentiam evocari asserens quod timens ne scriptura infrascripta modo aliquo deleatur aqua vel antiquitate et memoria regalium donationum factarum sancte cephaludenssi ecclesie de picto albo pariete valeat deperire, ad perpetuam rei memoriam reservanda nobis obnixe requisivit nostrum officium inplorando ut talem scripturam in pariete porte regum in ipsius ecclesie campanario pictam in puplicam deberemus redigere notionem. Nos autem, attendentes iustam esse requisitionem ipsius dicti domini episcopi, adimplere curavimus quod quesivit et quia ipsam scripturam vidimus et legimus non delatam nec etiam vitiatum, nichil adendo vel minuendo nec etiam inmutando, set in forma propria seu figura existendo, transcripsimus et in forma puplica redigimus. Cuius tertie figure imago est: Pictus est ibi quidam rex recta facie et integra v(er)sa ad populum inspectorem seu ad oppositum parietem, indutus vestimentis regalibus, coronatus et cum manu dextra tenet regalem virgam et cum sinistra cartam pictam et scriptam. Cuius scripture tenor talis est: Quod dive memorie pater noster ecclesie cephaludenssi concessit confirmamus, ratificamus et approbamus et de pia clementia nostra addicimus predicto dono et donamus ecclesiam sancte Lucie de Seracusia cum casalibus et pertinentiis suis. Super cuius capud scriptum erat: Guillelmus primus Sicilie rex. Et versus scripti sic: Ut rata sit bona res qui sum successor et heres, que prebet genitor pariter dare cum patre nitor. Unde ad certitudinem presentium et futurorum memoriam presens scriptum testium amminiculo roboramus. ⫹ Ego Primus de Primo, qui supra, iudex, predictam scripturam in pariete porte regum in ipsius ecclesie campanario seu scriptam vidi, legis, interfui et testor, etc. (as above). In nomine Domini, etc. (as above). Cuius quarte figure imago talis est: Pictus erat ibi quidam rex indutus vestimentis regalibus et coronatus, in cuius capite scriptum est: Gulielmus secundus Sicilie rex. Hic tenet cum manu dextra virgam regalem et cum sinistra cartam depictam scriptam. Cuius scripture tenor talis erat: Regali clementia nos heres progenitorum nostrorum concedimus que concesserunt de solita benignitate cephaludenssi ecclesie et presenti scripti robore confirmamus. Versus vero super capud eius in spatio hii sunt: Ne successores rapiant que dant genitores firmo patrum mores, nostros superaddo favores. Unde ad certitudinem presentium et futurorum memoriam presens scriptum testium amminiculo roboramus. ⫹ Ego Primus de Primo, etc. (as above). In nomine Domini, etc. (as above). Cuius imago talis est: Depicta erat ibi quedam imago mulieris, versa et recta facie ad populum inspectorem, induta regalibus vestimentis, cuius manus dextra tenebat virgam regalem et cum sinistra cartam pictam et scriptam. Cuius scripture tenor talis erat: De innata benignitate nos Costantia Romanorum imperatrix, que concessit pater noster Rogerius Sicilie rex confirmamus ecclesie Cephaludi et addicientes donamus perpetuo eidem casale Odosuer cum viribus et pertinentiis suis. Ipsa autem imago coronata erat et epyttaphium desuper tale erat: Constantia, Rogerii regis filia, regina Sicilie et Romanorum imperatris.
262
` CATHEDRALS ROYAL PORTRAITS OF GERACE AND CEFALU
Unde ad certitudinem presentium et futurorum memoriam presens scriptum testium amminiculo roboramus. ⫹ Ego Primus, etc. (shorter version). In nomine Domini, etc. (as above). Cuius imago talis est: Depictus est ibi rex regalibus indutus et coronatus sedens in falidistorio, tenens virgam regiam cum dextra. Cuius facies versa est ad quemdam episcopum mitratum et pontificalibus indutum, in cuius capite scriptum est: “Iohannes episcopus,” et recipit cartam de manu sinistra regis in qua scriptum est: vade in Babiloniam et Damascum et filios Saladini quere et verba mea eis audacter loquere ut statum ipsius terre valeas in melius reformare. In capite cuius regis scriptum est: Fredericus primus imperator. Unde ad certitudinem, etc. (as in previous section).
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
2 Gerace cathedral, plan (after C. Bozzoni, Calabria normanna: Richerche sull’ architettura dei secoli undicesimo e dodicesimo [ Rome, 1974], fig. XVIII)
3 Roger II crowned by Christ, Santa Maria dell’Ammiraglio, Palermo (photo: Alinari/Art Resource, New York)
4 Enamel plaque, Roger II crowned by St. Nicholas, San Nicola, Bari, after 1139
5 George of Antioch, Santa Maria dell’Ammiraglio, Palermo (photo: Alinari/Art Resource, New York)
6 Cefalù cathedral, general view from the west
7
William II crowned by Christ, Monreale cathedral (photo: Alinari/Art Resource, New York)
8
William II presenting the church model to the Virgin, Monreale cathedral (photo: Alinari/Art Resource, New York)
9 Petrus de Ebulo, Liber in honorem Augusti, departure of Constance for Sicily, Bern, Burgerbibliothek, Cod. 120 II, fol. 119r (photo: Burgerbibliothek)
10
De arte venandi cum avibus, Frederick II enthroned, Biblioteca Apostolica Vaticana, Ms. Pal. Lat. 1071, fol. 1v (photo: Biblioteca Apostolica Vaticana)
11
Church of St. George, Kurbinovo, facade (after L. Hadermann-Misguich, Kurbinovo: Les fresques de Saint-Georges et la peinture byzantine du XIIe siècle [ Brussels, 1975], fig. 141)
12
Constantine IX Monomachos and Zoe with Christ, Hagia Sophia, Istanbul
13
John II Komnenos and Irene with the Virgin and the child, Hagia Sophia, Istanbul
14
Calixtus II and Henry V with the Concordat of Worms, Lateran Palace (destroyed), 16th-century drawing, Biblioteca Apostolica Vaticana, Barb. lat. 2738, fol. 104r (photo: Biblioteca Apostolica Vaticana)
15
Donation of emperors Romanos I Lekapenos and Konstantinos to St. Vincent, Chronicon Vulturnense, Biblioteca Apostolica Vaticana, Barb. lat. 2724, fol. 169v (photo: Biblioteca Apostolica Vaticana)
16
Privilege of Pope Innocent III, lower church, Sacro Spello monastery (photo: Scala/Art Resource, New York)
17
Michael VIII Palaiologos with family, presenting the church model to the Virgin, church of the Virgin, Apollonia (after H. and H. Buschhausen, Die Marienkirche von Apollonia in Albanien: Byzantiner, Normannen und Serben im Kampf um die Via Egnatia, ByzVind 8 [ Vienna, 1976], fig. 19)
18
Dedicatory image of Stefan Lazarevi{, church of the Holy Trinity, Manasija (drawing after S. Tomi´c and R. Nikoli{, eds., Manasija: Istorija—Zˇivopis, Saopstenja 6 [ Belgrade, 1964], pl. II)
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Organization, Market Structure, and Modus Operandi of the Private Silk Industry in Tenth-Century Byzantium GEORGE C. MANIATIS
INTRODUCTION he private silk industry played an important role in the Byzantine economy. Yet, our understanding of the way it actually worked within the prevailing economic and institutional milieu of the tenth century is deficient. The scope of guild activity in the industry remains controversial since the enforceability of a mandated division of labor has been questioned. Nor has the impact of key institutional and operational parameters on the conduct and performance of the individual firms operating under the guild umbrella been fully understood, as past analyses have viewed the guild either as an organization protective of the interests of a group of businessmen operating under a licensed monopoly and acting in a monolithic fashion, or as an aggregation of aggrieved, downtrodden craftsmen. More important, the structure of the segmented by fiat markets, the nature of market competition, and their effect on the decision-making processes of individual firms have hardly been explored. In addressing these issues, this study challenges long-standing theories, identifies misconstructions and unsupported assertions, fills in lacunae, and puts forward new hypotheses about the likely manner in which the silk industry operated within its economic, political, and cultural constraints. Emphasis on disaggregation and the operational dimensions of probative value offers new perspectives for deciphering the rationale for institution of the guild system, the behavior of the players involved—guild members and enforcement organs alike—and the economic activities of the industry. Further, this approach filters our current knowledge on the subjects under scrutiny, providing more cogent answers to an array of collateral questions that have puzzled historians over the years. Having put the organizational setup in proper perspective, we can then analyze the structure and modus operandi of the silk industry’s functionally segmented markets, which were conditioned by the dictate of the mandated guild organizational framework and institutional arrangements. By establishing the structural characteristics of each market in the productive circuit, by delineating the operational functions of the guilds concerned, by ascertaining the impact of the restrictions imposed by the regulatory re-
T
264
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
gime, and by tracing input-output flows and market exchanges, a deeper understanding can be gained about the nature of the competitive process, the extent of state intervention, the calculus underlying business conduct, and critical aspects of enterprise performance. The nature of seller-buyer interaction, including the dynamics of the inter-seller and inter-buyer relations, are focal points of this inquiry since they affect the functioning of markets along the spectrum of the industry’s activities, which encompasses the stages of sourcing, manufacturing, and distribution. Market structure, as shaped by the degree of seller and buyer concentration, the degree of product differentiation, and the conditions for the entry of new participants, determines the potency of operative market forces, defines the nature of competition, and affects the behavior of firms. Market conduct, referring to the set of principles, strategies, and methods that businesses employ in deciding production policies and pricing, influences in turn the firms’ market performance, as reflected, inter alia, in the mix (design, quality, or variety) of products marketed, enterprise and output growth, and profit margins. During the period under review, Constantinople was the predominant locus of silk manufacturing and marketing in the empire. Sericulture and the trade in cocoons, raw silk, yarn, and silk fabrics were in private hands. In the public domain, silks were produced in the imperial workshops (basilika` ejrgodo´ sia)1 but this output was strictly nontradable.2 The bulk of marketable silks was manufactured in private workshops located in the capital, which were organized into guilds (susth´ mata) under close state surveillance.3 Outside the guild system, non-commercial home production for personal use was confined to the nobility and the wealthy. The Book of the Eparch set forth the rules of conduct for the private silk industry in the capital, as well as for guild members, dignitaries, and the well-to-do.4 The notion of imperial silk guilds has been challenged: S. Vryonis, Jr., “Byzantine Dhmokrati´a and the Guilds in the Eleventh Century,” DOP 17 (1963): 300 n. 46. Vryonis has pointed out that R. S. Lopez mistakenly identified the dhmo´ sia sw´ mata (public guilds) as imperial guilds: “Silk Industry in the Byzantine Empire,” Speculum 20 (1945): 3–8. What are referred to in the Basilics (LIV) as public guilds are actually private guilds. Imperial workshops are referred to in the sources as basilika` ejrgodo´ sia. 2 Silks of exquisite quality produced in the imperial workshops satisfied primarily the needs of the court. They were also offered by the emperor as presents to high-ranking government officials, favorite friends at home and abroad, foreign dignitaries, churches and monasteries and, on occasion, as tribute to hostile foreign leaders. 3 In the capital, the private silk industry was organized under five trade and manufacturing guilds: metaxopratai (dealers in cocoons, raw silk, and yarn), katartarioi (yarn producers), serikarioi (manufacturers of silks), vestiopratai (merchants of domestically produced silks), and prandiopratai (merchants of imported silks). 4 The Book of the Eparch ( jEparciko` n Bibli´on), probably promulgated in 911 or 912, codified earlier decrees. On the various views regarding the compilation, interpolations, subsequent additions, and the date of issue of this compilation, see A. P. Christophilopoulos, To` jEparciko` n Bibli´on Le´ onto" tou' Sofou' kai` aiJ Suntecni´ai ejn Buzanti´v (Athens, 1935), 7, 10–27; G. Ostrogorsky, History of the Byzantine State (Oxford, 1968), 215–16. ´ dit de l’Empereur Le´on le The Greek text was published with emendations by J. Nicole, Le Livre du Pre´fet ou l’E Sage sur les corporations de Constantinople (Geneva, 1893), and was reprinted in J. and P. Zepos, Jus Graecoromanum (Athens, 1931), 2:371–92. J. Nicole’s 1894 French translation was reprinted in To` jEparciko` n Bibli´onÚ The Book of the Prefect. Le Livre du Pre´fet (London, 1970), 120–203. English translations are by A. E. R. Boak, “The Book of the Prefect,” Journal of Economic and Business History 1 (1929): 597–619; and E. H. Freshfield, Roman Law in the Later Roman Empire: Byzantine Guilds Professional and Commercial (Cambridge, 1938). Reference to the Greek text is absolutely necessary as neither translation has always captured nuances and subtleties. A recent critical edition of the Greek text and German translation is by J. Koder, Das Eparchenbuch Leons des Weisen (Vienna, 1991). 1
GEORGE C. MANIATIS
265
ORGANIZATIONAL SETUP: FACTS AND FALLACIES Sericulture and Local Cocoon/Raw Silk Trade Commercial sericulture, which entailed raising the silkworm and producing the cocoon, was primarily pursued within the frontiers of the empire as a household industry in villages and rural towns. The organization and location of production was determined by the availability of feed, space, and seasonal labor.5 The basic production unit was the family. For the most part, sericulture was a seasonal (April–July) sideline activity, usually employing family members, and provided supplemental income to the producer. Those in possession of mulberry groves6 and space but lacking labor hired seasonal workers or met their needs through sharecropping arrangements. Within the boundaries of the empire the trade in cocoons and raw silk had a distinct element of seasonality dictated by the production cycle.7 Since the capital was the manufacturing center par excellence, stifled cocoons8 or reeled silk9 had to be transported there by land or sea and marketed through the intermediation of provincial wholesale merchants. Producers sold their fresh cocoons to traders in nearby towns,10 many of whom maintained facilities for stifling and, possibly, unraveling and reeling. Most small producers probably did not stifle their own cocoons, as small volume and processing risks rendered this task uneconomical.11 Besides, the expected gain could be easily lost, if the The cultivation of the silkworm from the egg through the spun cocoon requires extreme care, involving scrupulous cleanliness, control of temperature, humidity, and noise, and constant watch for infectious diseases, rendering sericulture a highly labor-intensive activity. 6 Silkworms (Bombyx mori), fed leaves of the mulberry tree, yield high-quality commercial silk. Several species of wild silkworms, found in China, India, and elsewhere, produce marketable silk but not of such fine quality. Moriculture requires particular soil conditions and a moderate climate. This suggests that cultivation of mulberry trees was most likely confined to the littoral provinces of the empire and the immediate hinterland. On the controversy over the territorial location of sericulture in the empire, see N. Oikonomides, “Silk Trade and Production in Byzantium from the Sixth to the Ninth Century: The Seals of the Kommerkiarioi,” DOP 40 (1986): 42–45; D. Jacoby, “Silk in Western Byzantium before the Fourth Crusade,” BZ 84/ 85 (1991/92): 453–54 and references in nn. 6 and 7; A. Muthesius, “The Byzantine Silk Industry: Lopez and Beyond,” Journal of Medieval History 19 (1993): 26–28; eadem, “Constantinople and Its Hinterland: Issues of Raw Silk Supply,” in Studies in Byzantine and Islamic Silk Weaving (London, 1995), 315–35. 7 In Europe and Asia Minor, where seasons are defined, the silkworm produces one harvest a year. This implies that there could not have been local production of multiple crops of silkworms, as Muthesius has hypothesized: “Byzantine Silk Industry,” 33. In parts of India and China reproduction is continuous, but the quality of the silk produced is inferior and inversely related to the number of hatchings. 8 Within ten to twelve days after the cocoon is spun, the chrysalis inside the cocoon must be destroyed to prevent it from damaging the threads. The most practical and economical method to achieve this was by heating the cocoons in ovens—a process called stifling. However, stifling often failed to kill the chrysalis since temperature could not always be fully controlled. This damaged the filament, affected the elasticity and color of the silk, and reduced the marketability and price of the cocoons. 9 Reeling is the process whereby the filament is unraveled from the cocoon, the strands from several cocoons are brought together to form a continuous, uniform, and rounded thread, and the fiber thus obtained is wound on reels. Reeled silk is referred to as raw silk. Three-quarters of a cocoon can be reeled; the remainder is surface floss and husk. To facilitate the unraveling and reeling, cocoons were boiled in water to soften the cericin, the natural gum that surrounds the silk fiber. Cericin acts as an adhesive, aiding in holding the several filaments together while they are combined to form a single thread. At this stage, only 1% out of the 20% cericin content is removed. The gum is retained until the yarn or fabric stage, because it affords protection during further processing of the delicate filament. 10 Itinerant merchants may also have been involved in the collection of cocoons from small producers at harvest time, reselling in turn to wholesale town merchants. 11 Large producers probably stifled, reeled, and even marketed their own cocoons. 5
266
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
buying merchant was uncertain as to how carefully the stifling process had been performed. Stifling obviated the need for immediate reeling, allowing time for storage and transport. However, since only one-sixth of a cocoon’s weight is silk, shipping reeled silk over long distances was cheaper.12 Traders therefore had to compare the potential savings in transport, given the distance from the origin of their shipment to the capital and their access to the cheapest mode of transport available,13 with the costs and risks of reeling.14 To be sure, limited utilization of reeling facilities (sheds and equipment) due to the seasonality of cocoon production could reduce perceptibly the return on investment. On the other hand, reeling could enhance profitability by making it possible to obtain higher selling prices in the capital, since storability enabled traders to time shipments to avoid periods of import peaks and depressed prices. Thus, distance, mode of transport, capacity utilization, carrying charges, and price expectations significantly affected the economic calculus and ultimately dictated the form in which the silk would be shipped. And, since shipment distances varied and production, initial processing, and transport conditions differed considerably among provinces and shippers, it is reasonable to assume that silk was shipped in both forms, depending on particular circumstances. Mickwitz maintains that raw silk was brought into the capital in the form of cocoons.15 Goitein and Jacoby also present convincing evidence that shipments of stifled cocoons in the Mediterranean basin were common, at times reaching distant points (e.g., exports from Spain to Egypt).16 Simon, on the other hand, questions the existence of a cocoon trade in the empire on the grounds that, because unraveling has always been closely linked with sericulture, trade must have been conducted in the form of reeled silk.17 In the same vein, Muthesius asserts that the cocoon trade was confined to the vicinity of production centers since cocoons had to be reeled within a matter of days to prevent spoilage.18 The fact that stifling made possible the storage and safe transport of cocoons, Although bulky, as a high-value raw material and in demand, the stifled cocoons fetched a market price that could absorb relatively high transport costs. Even so, deadweight could become a factor with distance, tending to restrict shipments of stifled cocoons to rather short distances. Nevertheless, lower freights as a result of intense competition, e.g., along heavily trafficked sea routes, and shipment of the cocoons as return cargo could extend the traveling distance and mitigate considerably the adverse impact of the cocoons’ deadweight. 13 Sea transport had a clear cost advantage over land transport. See M. F. Hendy, Studies in the Byzantine Monetary Economy, c. 300–1450 (Cambridge, 1985), 554–61. 14 The reeling process is extremely delicate. It requires inordinate skill not only to prevent defects in the filament but also to blend the filaments into a strand of uniform thickness. For details on moriculture, silkworm rearing, cocoon production and initial processing, apart from the standard reference works, see A. Muthesius, “From Seed to Samite: Aspects of Byzantine Silk Production,” in Studies in Silk (as above, note 6), 119–26; eadem, “Byzantine Silk Industry,” 16–18, 32–33. 15 “Die Organisationsformen zweier byzantinischer Gewerbe im X. Jahrhundert,” BZ 36 (1936): 71 and n. 3. 16 S. D. Goitein, A Mediterranean Society: The Jewish Communities of the Arab World as Portrayed in the Documents of the Cairo Geniza (Berkeley, 1967), 1:102; Jacoby, “Silk in Western Byzantium,” 484 n. 183; idem, “Silk Production in the Frankish Peloponnese: The Evidence of Fourteenth-Century Surveys and Reports,” in Travellers and Officials in the Peloponnese: Descriptions-Reports-Statistics, ed. H. A. Kalligas (Monemvasia, 1994), 45–47, 53, 61. 17 “Die byzantinischen Seidenzu ¨ nfte,” BZ 68 (1975): 25–26, 39. 18 “Byzantine Silk Industry,” 34, 39; eadem, “Crossing Traditional Boundaries: Grub to Glamour in Byzantine Silk Weaving,” BMGS 15 (1991): 346–37; eadem, “Raw Silk Supply,” 325. 12
GEORGE C. MANIATIS
267
historical evidence that trade in cocoons did occur (sometimes over fairly long distances), opportunities for obtaining competitive freights on major sea lanes, and the realization that, in the end, various economic factors determined the form in which the silk would be traded and shipped all suggest an active trade in cocoons. It was probably greater over shorter distances, although apparently the cocoon trade was also drawn to more distant ports when conditions were conducive. Cocoon/Raw Silk Trade in the Capital Trading Regulations. Silk reached the capital from within and outside the empire in the form of stifled cocoons or reeled silk. The Book of the Eparch mandated that it be stored in the warehouses of the mitata (VI.5), which served both as guest houses for foreign merchants and as the marketplace for the imported goods. Foreign (ejqnikoi´) and provincial (ejxwtikoi´) merchants were treated in the same manner as far as the disposal of their wares was concerned. The eparch’s deputy (legatarios) apprised him of all the merchants who had entered the capital and of their origin, inspected the goods they brought, instructed them as to the manner in which sales were to be conducted (diorizo´ meno" o”pw" ojfei´lousi pipra´ skesqai), and set the time limit within which they had to dispose of their products. As a rule, no one was permitted to stay longer than three months. Non-compliants were subjected to severe corporal punishment and expelled from the city, and their goods were confiscated (XX.1, 2; V.5; X.2). Suppliers negotiated with representatives of the guild of metaxopratai (dealers in cocoons/raw silk), the only persons officially authorized to purchase imported silk, although katartarioi (yarn producers) could join them under certain conditions (VI.1; VII.4, 5; VIII.8). As market day approached (ejn kairv' ajgora'"), the members of the guild of metaxopratai contributed according to their means to a fund for the statutorily mandated collective purchase of the imported silk, which was then allocated among them in proportion to their contributions (VI.8). Ad hoc contributions were required because the guild had no standing trading fund. The metaxopratai were forbidden to pre-empt the purchase of silk by traveling outside the capital (VI.12). Had they been allowed to do so, individual guild members operating outside the mandatory collective purchase system would have undercut the effectiveness of the officially enjoined quasi-monopsony.19 Evidently, the collective purchase mandate reflected the authorities’ strong belief that the exercise of monopsony power would substantially lower the price of this crucial input.20 The fact that the large and competent external suppliers were at times few in number may have contributed to the decision to legislate concerted action that neutralized their If purchases in local markets by traveling metaxopratai were allowed to become normal practice, local merchants could band together and thereby increase their bargaining power and selling prices, particularly as they were few in each location. Hence, a purchasing scheme based on coercive funneling of all imported cocoons and raw silk in the capital and prevention of guild members from trading individually would allow the monopsony power of the metaxopratai to operate with full force. Cf. also G. Mickwitz, Die Kartellfunk¨nfte und ihre Bedeutung bei der Entstehung des Zunftwesens (Helsingfors, 1936), 211, 212 and n. 2. In tionen der Zu light of the above remarks, there is no foundation to S. Runciman’s assertion that the metaxopratai were allowed to buy raw silk privately from itinerant importers: “Byzantine Trade and Industry,” in Cambridge Economic History of Europe (Cambridge, 1987), 2:154–55. 20 For an array of factors that could potentially attenuate the actual monopsony power of the collectively acting metaxopratai, see below, pp. 305–10. 19
268
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
bargaining power. The measure was also prompted by a concern to uphold fundamental principles of the guild organization, namely to ensure equal access to commercial opportunities for all members and prevent the monopolization of imports by entrenched members who might otherwise buy up the silk before other guild members had a chance to make purchases (“forestalling”). Finally, the localization of transactions, development of specialized traders, and prohibition of business on non-market days increased market efficiency by reducing the effective costs for both buyers and sellers of searching for information.21 As the wholesale nature of the import trade required substantial amounts of ready cash, some metaxopratai may have borrowed to supplement their own resources or formed partnerships with wealthy individuals when their credit limits were exhausted. Metaxopratai unable to buy wholesale could get supplies in smaller lots from entrenched (ejn eujpori´a) importer colleagues whose profit could not exceed 8 1/3 percent of the import price (VI.9). Rather than being a reflection of poverty of the metaxopratai, as has been alleged,22 this provision actually points to variability in the size of firms within the guild, the absence of coordination among guild members in the resale market, and the state’s subtle effort to foster inter-seller competition by enhancing the staying power of financially weaker members. Katartarioi also could join the metaxopratai in these collective purchases, but only if they were invited by the latter and met certain preconditions, and if the quantity they intended to buy was commensurate with their processing capacity (VII.1, 4, 5).23 The price that the participating katartarioi paid was fixed in advance by mutual agreement and could not be changed subsequently (VII.4).24 Apparently, such invitations were extended only exceptionally, for example during import peaks or to meet shortfalls in contributions by metaxopratai, instances when additional capital was urgently needed to ensure the purchase of the entire quantity of silk that had entered the market and prevent diversion of cargoes to rival manufacturing centers.25 The occasional inclusion of katartarioi not only increased the available capital but also forestalled competitive bid21 G. J. Stigler, “The Economics of Information,” Journal of Political Economy 69 (1961): 213 and 216. The concentration of a large number of buyers and sellers in one location significantly reduces information costs since prices in an organized market embody the requisite information for decision making. See also F. A. Hayek, Individualism and Economic Order (Chicago, 1948), 33–56 and 77–91. 22 Metaxopratai in this position have been invariably considered as literally “poorer” (pene´ steroi): Lopez, “Silk Industry,” 16. However, the notion of poor (pe´ nh", ptwco´ ", a“poro") needs to be put in proper perspective. Byzantine law (Basilics, LX.34.10) considered as poor those whose assets were valued at less than fifty nomismata—no mean sum in the tenth century. Apparently, the metaxopratai referred to in provision VI.9 were smaller, less liquid traders unable to buy wholesale but certainly not impoverished. In the context of provision VI.9 (and VII.2 as well), one ounce is equivalent to one-twelfth of the nomisma, i.e., one miliaresion. The Byzantine currency and its equivalent higher denominated subdivisions were: 1 nomisma ⫽ 12 miliaresia ⫽ 24 keratia ⫽ 288 folleis. A charge of one miliaresion per nomisma amounted to 8 1/3%. 23 The katartarioi had to convince the eparch that they were freemen, in good standing, and had the requisite financial resources (mh` ei«nai pantelw'" a“poroi) for the upcoming purchase (VII.5). 24 This suggests that the participating katartarioi were charged more than the import price but less than that charged to non-participating katartarioi of servile status (see below, pp. 276–78), which included a profit of 8 1/3% (VII.2). Since the final import price was not known with certainty beforehand, the purchase price most likely was fixed by adding a certain percentage on the realized import price. 25 See note 167. Since the katartarioi were approached by the metaxopratai and were invited to join in the collective purchase, it is unlikely that they were pressured to pay a high price differential.
GEORGE C. MANIATIS
269
ding between the two groups, which could only have benefited the suppliers. The fact that the serikarioi (silk manufacturers) were also forbidden to make direct purchases of raw silk from external suppliers (VIII.8) was further meant to prevent competition with the metaxopratai and preserve the latter’s purchase monopsony. The metaxopratai were not allowed to act as agents for the purchase of raw silk on behalf of influential and wealthy persons (VI.10). Nor could they sell raw silk to Jews or traders for resale outside the capital (VI.16) or market unprocessed silk from their homes; they were forced to sell openly in the marketplace (VI.1, 13). These restrictions were reinforced by similar provisions that applied to katartarioi. The latter also were forbidden to buy raw silk as surrogates for wealthy persons (VII.1), purchase more than they could process (VII.1, 5), or sell it unprocessed (VII.1). The main purpose of these restraints was to forestall the growth of commercial silk manufacturing outside the guild system by channeling all activity through legitimate conduits and curbing unseemly practices. By funneling all commercial transactions through a controllable environment, the strict division of labor among guilds could be enforced, while the emergence of monopolistic market structures, concentration of economic power, and potential defiance of imperial authority could be thwarted.26 It would be simplistic to suggest that a mandatory guild system was instituted only to allow parallel development of commercial silk activities outside the purview and control of the authorities, whether by noblemen, enterprising wealthy individuals, or small-scale traders and craftsmen. It would serve no purpose to establish an organizational structure, designate operational functions at each stage of the production process, and enact elaborate regulations concerning admission, obligations, and conduct, only to let silk activities be conducted outside the guild system. Had this happened, the guild system would have effectively been scuttled. In addition, the imposed constraints aimed at deterring diversion of supplies to low-priority uses through sales to unauthorized persons or in minuscule quantities, thereby assuring high-priority (guild-organized) producers adequate inputs (VI.10, 13, 16; VII.1, 5). Finally, the stipulation that raw silk could not be sold to traders who intended to export it unprocessed (VI.16) reflects a primary concern to avoid unwarranted shortages of a vital raw material and consequent price increases that could disrupt production and undermine the critical statutory function of the metaxopratai as a reliable source of supply. It also points to a conscious policy of exporting from the capital only finished products, a policy that generated local employment and a much higher added value. The fact that katartarioi were forbidden to acquire marketable surpluses, act as proxies for wealthy individuals, or sell unprocessed silk undoubtedly safeguarded not only the mandated import raw silk monopsony but also the metaxopratai’s sales monopoly in the domestic market. However, the existence of a statutory guild sales monopoly does not lead ipso facto to monopolistic pricing behavior by individual guild members and The power of the guilds in the capital, always a challenge to the emperor’s authority, was suppressed by the Macedonian dynasty during the 10th century. Thus, keeping in check potentially antagonistic economic and political activities by the numerous prospering traders and manufacturers who were part of the middle class (me´ soi) was a major concern of the authorities. In fact, the influence of the business community and the public at large reemerged in the 11th century, and their involvement in political decisions increased thereafter. Cf. P. Charanis, “On the Social Structure and Economic Organization of the Byzantine Empire in the Thirteenth Century and Later,” BSl 12 (1951): 147–49; Vryonis, “Byzantine Dhmokrati´a,” 289–314; Hendy, Studies, 571–80, 590. 26
270
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
the elimination of intra-guild competition, as has been argued.27 A sharp distinction should be made between the exclusive right to sell raw silk of the guild of metaxopratai as a group and the ability of individual guild members to take advantage of this prerogative and actually wield monopoly pricing power in the marketplace. As long as the metaxopratai were free to set their own prices, intra-guild competition could not be thwarted. Furthermore, effective exercise of monopoly pricing requires the existence of highly concentrated market structures, collusive action by guild members with the ability to set prices and enforce price discipline on the membership, closed entry to membership, and a supportive or quiescent regulatory regime. These conditions did not exist. The market structure was characterized by a large number of firms, a wide range of enterprise sizes, and unimpeded entry of new firms—conditions that fostered a competitive attitude and potent inter-seller rivalry among guild members. Express or tacit agreements leading to monopolistic arrangements were illegal and largely unenforceable, while actions that threatened to drive the katartarioi out of the market would most certainly invite their forceful reaction and state intervention.28 In short, securing a purchase monopsony did not by itself guarantee monopoly pricing power in a market structure characterized by low or negligible concentration and without opportunity for concerted action.29 Current thinking suffers from a “fallacy of division,” since what is true collectively for the entire guild membership as a class is not necessarily true for each member of the class. The metaxopratai sold the imported silk to katartarioi, but some metaxopratai entered into `a fac¸on arrangements with katartarioi to produce yarn to order on their behalf.30 They also supplied at capped profit margins their less liquid fellow guild members (VI.9), as well as metaxarioi and katartarioi who, because of their servile status, were ineligible to participate in the collective purchase of the imported silk (VII.2).31 The yarn the metaxopratai procured from the katartarioi was sold to the serikarioi, who were required to buy yarn only from metaxopratai (VIII.8).32 The guild of metaxopratai therefore retained statutory monopoly in the yarn resale trade as well, although this prerogative did not assure them as individuals control over selling prices.33 Mickwitz, Kartellfunktionen, 217; idem, “Organisationsformen,” 73; Lopez, “Silk Industry,” 18; N. Oikonomide`s, “Entrepreneurs,” in The Byzantines, ed. G. Cavallo (Chicago, 1997), 156; M. Angold, The Byzantine Empire, 1025–1204 (London, 1997), 93; Simon, “Seidenzu ¨ nfte,” 24, 41–42, 44. Simon pointedly refers to the purchase and sales monopoly (“Einkaufs und Preismonopol”) of the metaxopratai, ibid., 42. 28 See note 190. 29 The nature of competition in the domestic cocoon/raw silk market and the derivative ability of the metaxopratai to exercise monopoly pricing power in the resale yarn market is discussed below, pp. 310–19. By the same token, the reason the metaxopratai were not allowed to process the imported silk was to maintain the statutory “monopoly” of the katartarioi in yarn production and avoid sales competition between the two groups in the yarn market. Mickwitz argues that a strict division of labor between guilds was mutually advantageous because it prevented the development of a hybrid craftsman-trader class: “Organisationsformen,” 73–74. Nevertheless, the fact that yarn production was the prerogative of the guild of katartarioi did not mean that individual katartarioi could exercise monopoly power in the sale of yarn, for the same reasons that individual metaxopratai had no such power. See below, pp. 316–17. 30 Yarn procurement by metaxopratai through `a fac¸on contracts is examined below, pp. 274–76. 31 The prevailing view that a segment of metaxarioi and katartarioi were assisted because they were poor and not members of their respective guilds is refuted below, pp. 276–81. 32 The mainstream belief that the metaxopratai were the exclusive suppliers of yarn to serikarioi has not remained unchallenged. The controversy is critically reviewed and the challenge rebutted below, pp. 284–89. For an explanatory hypothesis regarding the circuitous yarn distribution pattern see below, pp. 290–91. 33 See below, pp. 317–19. 27
GEORGE C. MANIATIS
271
Although they were allowed to manufacture silks for their personal use, it has been suggested that the nobility and wealthy in the capital were forbidden to import or buy raw silk in the market, even through the intermediation of guild members; instead, it is argued they had to rely on their own home-grown production.34 However, it would have been very difficult, if not impracticable, for most members of the upper class residing in the capital to produce the raw silk they needed.35 Moreover, the notion that every person of means was eager to embark on home production of silks is unrealistic. Organizing and running such an involved undertaking took significant effort. Besides, restrictions on the kind, variety, and quality of the silks that could be produced (VIII.2), and unlimited access to a great variety of high-quality domestically produced and imported goods detracted much from the fascination with homemade silks. The fact that noblemen, highranking officials, and wealthy persons were preferentially allowed to join the prandiopratai in the purchase of imported silks (V.4) suggests not only that their quality was on a par with those produced locally but that home production in the capital was nominal. Also, home production would not have enabled the upper class to manufacture otherwise forbidden silks, as has been asserted.36 Finally, if influential persons were allowed to manufacture silks for non-commercial purposes (VIII.2), it would have made no sense to forbid them to buy raw silk or yarn retail to satisfy household needs. In fact, the Book of the Eparch did not restrict such direct purchases by final consumers for their own needs.37 Forward Integration into Yarn Production—A Specious Theory. The metaxopratai were not allowed to be involved in the processing of the imported silk (VI.1, 14). However, according to Mickwitz, they were involved, hiring workers to perform the task. These Lopez, “Silk Industry,” 15–16; Mickwitz, “Un Proble`me d’influence: Byzance et l’e´conomie de l’Occident me´die´val,” Annales d’Histoire ´economique et sociale 7 (1936): 27. 35 A. Harvey maintains that it must have been difficult for the authorities to enforce regulations aiming to curtail private production of expensive silks by the powerful and the wealthy since they could produce raw silk on their estates: Economic Expansion in the Byzantine Empire, 900–1200 (Cambridge, 1989), 183–84. However, estate production of raw silk in the immediate vicinity of the capital could not have been extensive— hence the restrictions on their access to imports. Only in the provinces of the empire where the provisions of the Book of the Eparch were not applicable could the magnates produce silks based on home-grown inputs, and then only if they were located in the right climatic zone. 36 Lopez maintains that the emperor could not prevent noblemen from manufacturing prohibited silks (kekwlume´ na) in their own workshops any more than he could impede the growth of the large estates: “Silk Industry,” 16. The analogy is strained, and the implicit assumption that the authorities tolerated such infractions is unfounded. Flaunting forbidden silks in the capital would be an affront to the emperor and a challenge to his authority because of their symbolism, strictly hierarchical order of display, and exclusive use by the imperial court and those on whom the emperor bestowed the privilege. It is therefore very unlikely that irreverent and defiant transgressors would get away with impunity. Besides, Lopez belittles the deterrent effect of the severe penalties inflicted on lawbreakers; see Basilics, XIX.1.30; Book of the Eparch, VIII.2, 4. On the other hand, potential infractions by regional noblemen who opposed the emperor and wanted to challenge his authority should be viewed as isolated incidents, at best. These alleged violations contrast sharply with the widespread and subtler legal maneuverings and machinations by means of which the powerful in the provinces could appropriate the land of defenseless smallholders because their exercise of power remained unchecked and was reinforced by the state’s confiscatory tax policy. On these issues see the references in note 225. 37 Mickwitz, Kartellfunktionen, 200. Simon also thinks that there was no ban on the purchase of raw silk or yarn by influential or wealthy people, provided that such transactions were aboveboard, and, in fact, he extends this possibility to all private consumers: “Seidenzu ¨ nfte,” 40–46 and esp. 41–42. Muthesius believes that the metaxopratai were allowed to sell raw silk to the wealthy for their limited private use and—mistakenly—to traders and craftsmen who were not guild members: “Byzantine Silk Industry,” 34 and 39. 34
272
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
workers were poor katartarioi, drawn from an abundant reservoir in their guild, who produced yarn for metaxopratai in their sheds for piece-rate wages (Stu ¨ cklohn).38 Mickwitz’s hypothesis, derived basically from the experience of medieval Italian cities, is untenable on several grounds. In the first place, even if the work was performed in the hired workers’ sheds, as long as the katartarioi were employees of the metaxopratai, a controlling principal-agent (i.e., employer-laborer) relationship existed, constituting a direct and illegal involvement of the metaxopratai in silk processing. Second, in Byzantium wage-earning craftsmen were not admitted to the guild; membership was restricted to independent operators exclusively.39 Third, these hired workers (misqwtoi´) are nowhere identified as katartarioi in the Book of the Eparch, which suggests that they were not guild members.40 This inference is reinforced by the fact that the notion of misqwto´ " implies compensation of highly skilled operatives or employees occupying responsible positions in the form of fixed payments at regular intervals, whereas piece rates are linked to ordinary workers’ output. Finally, if these workers were in such an abundance, there was no point in dealing with them in the Book of the Eparch, let alone imposing restrictions on contract lengths and salary advances, or prohibiting “poaching” of other members’ workers (VI.2, 3).41 In the same vein, Lopez has maintained that, although the metaxopratai were theoretically forbidden to take over silk processing, in practice the prohibition was rendered moot by the fact that the metaxopratai could hire workers or use their slaves to engage in yarn production. To this end, they could draw on a pool of “ill-favored” katartarioi, who preferred to give up their independence and work for the metaxopratai as employees. In Lopez’s view, it is hardly credible that workers were employed to assist the metaxopratai in buying and selling. And although the law forbade the metaxopratai to enter into a contract with a hired worker for longer than one month, it did not prohibit them from renewing the contract or hiring more than one worker. Alternatively, the metaxopratai could set up their slaves in the business of yarn production as proxies.42 Lopez’s assertion, too, is conjectural, and the arguments advanced in refuting Mickwitz’s theory apply with equal force. In addition, by setting up his slaves in yarn production, the metaxoprates did not sever the master-slave bond; he was liable for the slaves’ commercial debts or wrong-doing (VI.7). More important, the conduct of silk processing activities by a slave on behalf of his master clearly implied an on-going employerworker relationship and the slave’s proxy status. On both counts the law was violated. It is inconceivable that the Book of the Eparch created, inadvertently or deliberately, such a huge loophole undermining the division of labor among guilds—the cornerstone of the “Organisationsformen,” 72–73; “Un Proble`me,” 25–26. Nicole committed the initial error by characterizing the katartarioi as poor (“pauperiores confectorum”): Livre du Pre´fet, 34, and “peu fortune´s,” jEparciko` n Bibli´on 160, misinterpreting the term eujtelei'", which refers to their servile social status and not to their financial situation. Translators (see note 4) and historians have adopted the incorrect rendition. On this important issue see below, pp. 276–78, 310–17. 39 See below, p. 280 and note 66. 40 B. Mendl, “Les Corporations byzantines,” BSl 22 (1961): 308–9 n. 17 (M. Loos); Simon, “Seidenzu ¨ nfte,” 29 and n. 30, with references to concurring views of M. Sjuzjumov and A. Kazhdan in their respective works cited therein. 41 See below, pp. 275–76, 293. 42 Lopez, “Silk Industry,” 18, 19 and n. 1. 38
GEORGE C. MANIATIS
273
guild’s organizational structure. Were this the case, the de jure prohibition of vertical and horizontal integration within the guild system would have been defeated. Finally, it is not credible that the metaxopratai would openly hire workers or employ their slaves in workshops owned and run by them or their surrogates in defiance of the law, which imposed stiff and humiliating penalties (flogging, tonsure of the hair, and exile: VI.1, 14). If they ventured into silk processing, the metaxopratai must have been in a position to circumvent the law unscathed, either because they could remain undetected or could rely on venal public officials to cover up infractions or quash indictments. Yet, there is no evidence to suggest that individual metaxopratai had the power or political clout necessary to influence authorities and remain above the law during the period under review. The provisions of the Book of the Eparch undoubtedly allude to the potential for firms to infringe upon other guilds’ activities, something that can be expected in a free-market economy, but the law’s precautionary provisions cannot be construed as proof of a prevalent pattern by aggressive firms. More importantly, the very purpose of the legislative reform introduced by the Book of the Eparch was precisely to frustrate such transgressions, ushering in a “new order” that reflected a selectively interventionist industrial policy aimed at advancing the interests of the state and curbing the market power of guild members. Doubters have yet to show that the state failed in this effort.43 Thus, Lopez supplies no evidence to substantiate his claim that the guilds of metaxopratai, serikarioi, and prandiopratai encroached upon the other two guilds after the promulgation of the Book of the Eparch. He simply conjectures that these pressures persisted, dismisses offhand the willingness and ability of the authorities to enforce the law, underestimates the deterrent effect of the severe punishment of transgressors, and slights the capacity of the offended guilds to fend off infringement of their prerogatives. Nor does Mickwitz, for that matter, go beyond avouching presumptively the expansionary tendencies of these groups and, implicitly, the administration’s lack of will to enforce the law. In particular, the doubters fail to appreciate that the source of power and the enforcement mechanism materially affect the spirit and the manner in which the law is administered. The locus of the coercive executive power in Byzantium was the emperor, who had absolute authority over his subjects. Rank, reputation, or wealth afforded no immunity against the imperial will. The emperor could imprison, banish, discharge from office, or demote any person who betrayed his trust, confiscate assets and personal property, and inflict corporal punishment. In short, the emperor embodied and personified the law. To implement the Book of the Eparch, the emperor employed professional civil servants—not officials who had purchased office expecting to extract a satisfactory return on their investment or self-serving guild members.44 The assignment of the supervisory 43 Cf. Mickwitz, “Organisationsformen,” 75; idem, “Un Proble`me,” 26–27; Lopez, “Silk Industry,” 18–20; Mendl, “Les Corporations byzantines,” 306; Runciman, “Byzantine Trade,” 155; Simon, “Seidenzu ¨ nfte,” 32–33. 44 ¨ckle, Spa¨tro¨mische und byzantinische Zu ¨nfte (Leipzig, 1911), 84; Christophilopoulos, jEparciko` n A. Sto Bibli´on, 47–48, 49 and n. 1; Boak, “Book of the Prefect,” 599; and Oikonomide`s, “Entrepreneurs,” 155, maintain that the chiefs of the guilds, though officials of the state, were chosen from among guild members. However, this is true only for the association (su´ llogo": I.2, 3, 4, 13) of the notaries who, because of their legal training (I.1, 2) and quasi-judicial functions, were considered an elite group and were subject to a different set of rules. The Book of the Eparch explicitly states how the selection of the chief is to be made: the
274
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
task to a disinterested, full-time, ad hoc body of officers loyal to the emperor and paid by the fiscus ensured their diligence and evenhandedness, all the more so since they carried out their duties in a highly localized market and amidst the ubiquitous presence of informers. The central authorities watched carefully the performance of the executive officers, and there was no doubt about the dire consequences of dereliction of duty. Citizens were encouraged to lodge complaints against those in authority before the imperial tribunal, and officials were obliged to remain in their posts for fifty days after the termination of their office to respond to citizens who might want to bring suits against them. In sum, lacking convincing evidence about on-going transgressions, a corrupt administrative apparatus, and suborned administrative justice during the era in which the Book of the Eparch was operative, it cannot be cogently inferred that the division of labor among guilds was not upheld or that the rules prohibiting vertical and horizontal expansion were not effectively enforced.45 A Fac¸on Arrangements: A Sideline Activity. There was no compelling reason for the metaxopratai to resort to questionable practices in order to expand their commercial activities. They could lawfully supply cocoons or raw silk to katartarioi to be processed into yarn in their workshops under a charge for service arrangement—a` fac¸on. Absent any legal stipulation to the contrary, `a fac¸on transactions must have been legitimate commercial activities. What the Book of the Eparch explicitly forbade was the parallel conduct of silk processing activities by metaxopratai, through the hiring of laborers to perform the work in the metaxopratai’s own or the workers’ workshops. Under a verbal or written `a fac¸on contract, the metaxopratai provided raw silk to interested katartarioi, specified the tensile strength of the yarn to be produced, paid an agreed-upon price for the services rendered, and collected the final product on a set delivery date. Under this contractual arrangement, the participating katartarioi operated in their own workshops, employed their own equipment, tools, and auxiliary materials, and hired outside labor as needed. The compensation the katartarioi received was not a wage, but comprised the costs of processing the raw silk into yarn (wages, operating costs, overhead) plus profit. In other ranking notary is promoted to the vacant position by the eparch, provided all members vouch for him, a maximum of twenty-four due to the numerus clausus (I.23). If he is found unworthy of the position, the next in rank is advanced to the post, and so on (I.22). It is evident that this ad hoc procedure and internal organization was tailored to the distinct quasi-judicial function discharged by the notaries and certainly cannot be applied by analogy to the unrelated business activities of the other guilds. In dealing with the guilds proper (susth´ mata), the Book of the Eparch nowhere indicates that the chiefs were guild members. This is understandable, as the authorities could not trust active guild members to perform their duties impartially because of their vested interest. If the assessor (su´ mpono": XIV.2; XVIII.1, 4; XIX.1), senior inspectors (ejpiskepth'tai and ejpo´ ptai), sealers (boullwtai´: IV.4, VI.4, VIII.3, XI.9, XII.9, XIII.2, XVI.6, XIX.4), and examiners of silk fabrics (mitwtai´: VIII.3) were career public functionaries, apparently to ensure effective implementation of the rules and regulations, a fortiori the chiefs of the guilds had also to be outsiders (or at least guild members who had permanently given up their craft and were considered trustworthy), given the nature of their tasks, and their salaries must have been paid by the fiscus. Runciman, however, believes that the chiefs’ salaries were charged to the guilds: “Byzantine Trade,” 159. That such an arrangement would apply only to the chiefs seems very unlikely, as administrators in Byzantium were paid directly from the treasury. Cf. A. Kazhdan and G. Constable, People and Power in Byzantium: An Introduction to Modern Byzantine Studies (Washington, D.C., 1982), 151; A. Guillou, “Functionaries,” in The Byzantines, ed. Cavallo (as in note 27), 217. 45 See also the discussion below, pp. 284–89.
GEORGE C. MANIATIS
275
words, the contracted katartarioi operated as principals, not agents, that is, as independent craftsmen within the guild structure and not as subordinate misqwtoi´ under a employer-worker framework. As this activity did not take place on the premises of the metaxopratai or in the workshops of the katartarioi under their or their proxies’ management and control, there was no direct involvement of the metaxopratai in any facet of the yarn production and, in consequence, no transgression of the law. In this fashion, the functional division of labor was preserved intact.46 ` fac¸on arrangements were incidental sideline activities probably prompted by meA taxopratai eager to ensure that commitments to their clients (serikarioi), particularly at times of peak yarn demand, would be fulfilled by securing the services of dependable katartarioi; or they were seen as an opportunity by more aggressive metaxopratai to procure yarn at a lower cost, enhance their competitiveness, and capture a larger share of ` fac¸on deals may have been initiated by katartarioi as well. Those the resale market. A lacking access to working capital could operate on interest-free advances, while others might secure work when their own sales were off. The arrangement enabled katartarioi otherwise unable to compete by purchasing raw silk at the going or even capped market prices to stay in business. Some katartarioi, lacking a competitive spirit, may have given up their entrepreneurial function and confined themselves to the strictly technical aspects of their craft. Finally, `a fac¸on agreements enabled larger katartarioi to economize on working capital and utilize more fully their productive capacity and workforce. To organize and supervise this sideline activity, metaxopratai with sizable ongoing operations needed the assistance of competent, experienced, and trustworthy operatives. Because of their qualifications, such operatives were in great demand and short supply.47 Apparently, these individuals formed a pool of fairly independent and mobile workmen, who wielded some bargaining power, as suggested by the fact that they could command advance payment. They were the very persons the metaxopratai were eager to entice into service, not to be employed as common laborers, as has been alleged, but to execute this largely managerial function. That is why the Book of the Eparch, for the sake of fairness and equal opportunity, stipulated that such persons could not be hired for more than a month, although contracts evidently could be renewed, probably after they had been 46 Although they have certain similarities, `a fac¸on business deals differ in important respects from the ` fac¸on agreements involved independent guild members (katartarioi) domestic or putting-out system. A working in their own sheds, payment of a price for their services, and concentration only on cocoon/raw silk processing. The merchants (metaxopratai) did not control the whole process from start (raw silk) to finish (silks). By contrast, the putting-out system in the West aimed at bypassing the guild system and entailed control of the entire process by the merchant-employer putter-out. Under this system, the work was done by hired labor receiving mostly piece-rate wages at the craftsmen’s homes or the putter-out’s workshop, and mainly with tools, equipment, and auxiliary materials supplied by the latter. To the extent that work was put out to guilds, the master craftsmen themselves were employed as wage earners (maıˆtres ouvriers), hiring additional workers if they had to meet contractual obligations exceeding their personal capacity: H. Pirenne, A History of Europe (New York, 1958), 2:102; E. F. Heckscher, Mercantilism (London, 1955), 1:149–50, 187, 189; M. Weber, General Economic History (New York, 1961), 122; S. B. Clough and C. W. Cole, Economic History of Europe (New York, 1952), 183. Thus, the putting-out system was based on a dependent employer-worker relationship—on labor contracts—whereas `a fac¸on agreements represented commercial contracts between independent businessmen. 47 Apparently, these operatives had hands-on practical experience in the various tasks of raw silk processing, including basic managerial skills. See also below, p. 293 and note 112. Conceivably, some were Jews or other foreigners; see note 66.
276
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
renegotiated. Nor could these workmen be hired away from another employer before they had rendered fully the services for which they had been paid (VI.2, 3).48 Clearly, the so-called misqwtoi´ were not ordinary workers involved in silk processing; rather, they possessed special skills and were retained by metaxopratai to implement demanding a` fac¸on contracts with independent katartarioi. The conjunction of legitimate `a fac¸on transactions with the existence of a body of operatives capable of managing this sui generis type of operation suggests that such activities did take place during the period under review, especially since they worked to the mutual advantage of metaxopratai and katartarioi, although their extent cannot be ascertained. The Enigma of the Metaxarioi. The function, social status, and guild affiliation of the metaxarioi remain controversial because of diverse interpretations of the relevant passage of the Book of the Eparch (VII.2). According to one hypothesis, the terms metaxarioi and metaxopratai cannot be synonymous since the text makes a clear distinction between them. The metaxarioi were a special class of raw silk traders dealing in silk of inferior quality that they procured from the metaxopratai and, like the katartarioi, only by special permission of the eparch directly from external suppliers. Most likely, they formed a section of the guild of metaxopratai. This view mistakenly identifies the metaxarioi with the melathrarioi.49 Mickwitz considered the metaxarioi to be small-scale entrepreneurs playing a modest role and eking out a living, a description that hardly helps in identifying the object of their activity.50 Mendl, on the other hand, maintains that the metaxarioi could have been either traders in raw silk or raw silk processors since the text permits both interpretations.51 But, as has been convincingly pointed out, if the metaxarioi were both traders and yarn producers, it would be difficult to explain the concurrent use of two terms (katartarioi and metaxarioi) in the same provision to convey the same notion.52 According to the prevailing view, metaxarioi and metaxopratai were interchangeable terms: the former being an earlier expression of the latter, as in the case of serikarioi (VIII.1–13) and serikopratai (IV.7). Metaxarioi and metaxopratai plied the same trade, being merchants in raw silk. The only difference between them, it is suggested, was that the metaxarioi were not enrolled in the guild of metaxopratai.53 However, the assertion that the metaxarioi were not guild members is based on two fundamental misconstructions: 1) the misperception of them as indigent and thus unable to join the guild, following the erroneous translation of the term eujtelh´ " in reference to katartarioi as “poor” in provision VII.2 but applicable by analogy to metaxarioi as well, since both faced the same 48 Certainly, the aim of these restrictions was not to preserve the small-scale character of the trade and ´ tat et les me´tiers `a Byzance,” BSl 23 (1962): 239–41. manufacturing activities, as E. Frances argues: “L’E 49 Christophilopoulos, jEparciko` n Bibli´on, 83–84, 85 and nn. 1 and 2. For a critique of this hypothesis see below, pp. 280–83. 50 “Organisationsformen,” 74. 51 “Les Corporations byzantines,” 312. 52 See Loos’s attribution to Sjuzjumov in Mendl, “Les Corporations byzantines,” 315 n. 28. 53 ¨ckle, Byzantinische Zu ¨nfte, 8 ff and n. 5; K. Gehring, “Das Zunftwesen KonsNicole, Livre du Pre´fet, 80; Sto ¨cher f. Nationaloeconomie u. Statistik 3.38 (1909): 580; E. Mayer, review tantinopels im 10. Jahrhunderte,” Jahrbu ¨ckle, Byzantinische Zu ¨nfte, in BZ 21 (1912): 532; Boak, “Book of the Prefect,” 608 n. 5; G. Zoras, Le of Sto Corporazioni bizantine, Studio sull’ jEparciko` n Bibli´on dell’ Imperatore Leone VI (Rome, 1931), 172; Lopez, “Silk Industry,” 16 and n. 3; Runciman, “Byzantine Trade,” 154 n. 12, 156; Simon, “Seidenzu ¨ nfte,” 36–39; Muthesius, “Byzantine Silk Industry,” 32.
GEORGE C. MANIATIS
277
predicament and were treated alike;54 and 2) reading of the passage oiJ mh` ejn th' ajpografh' o“nte" as indicating those metaxarioi who were not members of the guild of metaxopratai in provision VII.2. These misconstructions present a distorted picture of the social status of the metaxarioi (and of the katartarioi as well) and give rise to serious inconsistencies with other relevant provisions of the law. A faithful translation of the original text presented below offers a more cogent portrayal of their true social standing and preserves statutory consistency. Furthermore, although the mandatory enrollment only of workshop owners or operators in a guild for the practice of a trade or craft is not explicitly mentioned in the Book of the Eparch, the tenor of the statute leads to this conclusion. Provision VII.2 stipulates: OiJ eujtele´ steroi katarta´ rioi kai` ajpo` tw'n metaxari´wn oiJ mh` ejn th' ajpografh' o“nte", ei“te a“ndre" ei“te gunai'ke", mh` duna´ menoi ejk th'" e“xwqen meta´ xh" ejxwnei'sqai kai` ajpo` tw'n metaxopratw'n ejxwnou´ menoi, ojfei´lousin oujggi´an mo´ nhn ajpolei´pesqai ejn tv' ajpo` th'" ejxwnh´ sew" tw'n metaxopratw'n nomi´smati.
It should be rendered as follows: Katartarioi of a lower social status (eujtele´ steroi) and those of the metaxarioi [⫽metaxopratai] not included in the eparch’s list (oiJ mh` ejn th' ajpografh' o“nte") because of their lowly social standing, whether men or women, being ineligible to buy raw silk directly from external suppliers and having to make purchases from metaxopratai, cannot be charged more than 8 1/3 percent over the import price.
The long-standing rendition of eujtelei'" as “poor” is thus clearly incorrect, while the reading of the passage “oiJ mh` ejn th' ajpografh' o“nte"” as “those operating outside the guild of metaxopratai” is infelicitous at best. The characterization of katartarioi (and metaxarioi) as eujtelei'" in provision VII.2 pertains to their inferior social standing as slaves (oijke´ tai), not to their financial condition.55 To mitigate the disadvantages suffered by this segment of katartarioi and metaxopratai, because they were denied the right to partake in joint purchases from external suppliers, they were allowed to procure their requirements wholesale or retail at capped profit margins.56 The provision was not meant to allow the financially weak, the “poorer,” as has been argued, to take advantage of this concession if they were freemen. By the same token, rendition of the passage “kai` ajpo` tw'n metaxari´wn oiJ mh` ejn th' ajpografh' o“nte"” as “those metaxarioi not in the register of the guild” is arbitrary, lacks foundation, and leads to fundamental contradictions. The first inconsistency arises from Cf. Lopez, “Silk Industry,” 16 and n. 3. A statutory provision that metes out penalties to felons convicted of preempting and hoarding goods in times of scarcity is quite probatory: if the culprits belong to the merchant class (pragmateutai´), i.e., are of higher social standing (⫽freemen), their merchandise (pragmatei´a) is confiscated or they are exiled; if eij de´ eujtelei'", i.e., if they are of inferior social standing (⫽slaves), they are sentenced to hard labor in public works. See Basilics, LX.22.6; Hexabiblos, VI.15.7. Cf. also Basilics, XIX.1.82, LX.39.1.3, LX.51.34; Synopsis Basilicorum, F.II.2. It is telling that provision VI.9 refers to the poorer (⫽less liquid) metaxopratai as peneste´ roi" and not as eujteleste´ roi" . See also note 22. 56 Christophilopoulos maintains that the metaxopratai provided raw silk wholesale to those katartarioi and metaxarioi who were not allowed to procure it directly from external suppliers: jEparciko` n Bibli´on, 62. Lopez, on the other hand, believes that they met their requirements by buying retail: “Silk Industry,” 16. Probably, sales took place at both levels, depending on the scale of the buyers’ operations, as liquidity was not necessarily a constraint. 54 55
278
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
the different interpretations of ajpografh´ (list) in provisions VII.2 and VII.5. In provision VII.2, “oiJ mh` ejn th' ajpografh' o“nte"” has been applied to those metaxarioi who were not members of the guild of metaxopratai.57 This interpretation, however, is inconsistent with the generally accepted rendition of provision VII.5, which stipulates that katartarioi willing to purchase imported raw silk, inter alia, had first to register with the eparch (ajpografe´ sqwsan pro´ teron para` tv' ejpa´ rcv), that is, they had to be included in the list of applicants drawn up by him. Evidently, there cannot be two conflicting interpretations regarding the meaning of ajpografh´ in dealing with the same circumstance. A second inconsistency derives from the misconstrued social status of the katartarioi and metaxarioi referred to in provision VII.2. Concerning the katartarioi, provision VII.3 stipulates that katartarioi can enroll in the guild of metaxopratai if they are freemen, while provision VII.5 enables katartarioi to join the metaxopratai in collective purchases of imported raw silk provided they are not of servile status (oijke´ tai). However, provision VI.7 empowers metaxopratai to set up their slaves to carry on this trade in their place, whereas provision VI.8 stipulates that all members of the guild of metaxopratai (pa'sa hJ koino´ th" tou' susth´ mato") contribute to the fund for the common purchase of the raw silk. The conjunction of these provisions suggests that the slaves of the metaxopratai could participate in the collective purchase whereas katartarioi of servile status could not, either directly or by joining the guild of metaxopratai, even though they might meet other preconditions (e.g., integrity or financial means). Such a state of affairs, however, would have created two classes of slaves in the same industry subject to diametrically opposite treatment—an untenable situation. With respect to metaxarioi, a more sensible and consistent interpretation of provision VII.2 would be, as already alluded to, that those “not in the register” actually refers to those of the metaxarioi (⫽metaxopratai) who are not freemen and, because of this encumbrance, are not eligible to participate directly in the collective purchase.58 Put differently, slaves were not allowed to participate directly in the purchase of imported raw silk whether metaxarioi or katartarioi, even if they were persons of means. The fact that the metaxarioi are mentioned only once in the Book of the Eparch, and only in provision VII.2 in conjunction with the lowly katartarioi, cannot be mere chance. Rather, it reflects their inferior social standing among silk traders, a notion that goes back to earlier times (Codex Justinianus, VIII.13.27), and an attempt to draw a sharp distinction between freemen and slave metaxopratai in regard to this particular transaction. Although provision VI.8 states that all guild members must contribute to the fund, the fact of the matter is that some metaxopratai (⫽metaxarioi) could not participate in the purchase because of their inferior social standing.59 “All” therefore should be read as meaning all those eligible, that is, not of servile status. In short, the restriction affected both katartarioi and metaxarioi because they shared the same inferior social status (oijke´ See note 53. Simon points out that the precise character of the eparch’s list cannot be determined. Nevertheless, he disputes the notion that the metaxarioi were not included in the list because they lacked financial resources: “Seidenzu ¨ nfte,” 36–38. 59 It is very likely that even freemen metaxopratai did not always participate in the collective purchase, as frequency of arrivals, projected market demand, inventories on hand, liquidity position, and price expectations determined an individual buyer’s requirements and procurement timetable. Any shortfalls in contributions would be made up by inviting katartarioi to join in (VII.4). 57 58
GEORGE C. MANIATIS
279
tai) and, in consequence, were subject to identical treatment. As a concession, these lowly groups met their requirements at a fixed mark-up.60 The third inconsistency emerges from the characterization of the metaxarioi as poor. If indeed the metaxarioi were poorer metaxopratai not enrolled in the guild, as Lopez maintains, they would have been classed together with the poorer metaxopratai referred to in provision VI.9, where they properly belonged, since the same preferential treatment was extended to both. The reason for dealing with them together with the katartarioi in provision VII.2 was their common low social standing as slaves. The metaxarioi are dealt with in section VII, not because they were poor or operated outside the guild of metaxopratai, as has been alleged, but because they suffered from the same predicament: as slaves they were ineligible for this specific transaction, albeit for reasons that remain elusive. Furthermore, it is unlikely that the same privilege would have been extended to both guild and non-guild members in the face of a policy objective that aimed to discourage the development of a silk industry outside the bounds of the guild system. In point of fact, their conjoint treatment supports the notion that there was no trade in raw silk outside the guild structure and, in consequence, no unorganized traders. Based on his theory that the metaxarioi were metaxopratai who were not members ¨ckle surmises guild membership was not mandatory for silk merchants, of the guild, Sto a view that is shared by many others.61 Lopez, in particular, maintains that there were two groups who did not belong to private guilds. They stood above and below the middle class of craftsmen. The first group included the nobility and the wealthy, while the second comprised the lowly melathrarioi and the poorer metaxarioi (⫽metaxopratai) and katartarioi. Espousing the contested view that the metaxarioi were poor and not members of the guild, and bolstered by the fact that the wholesale trade in raw silk required considerable financial resources, Lopez hypothesizes that guild members were a rising oligarchy within the middle class. Around them was a fringe of poorer traders and craftsmen who could not join or had dropped out of their respective guilds because they could not afford the entrance fee or underwrite their share in the collective purchase of imported raw silk. Even though they were not guild members, these impoverished groups were still protected by the law, being entitled to purchases at capped profit margins.62 This thesis is untenable. First, the Book of the Eparch nowhere makes class distinctions or suggests that these groups operated outside the guild system. The very fact that they are dealt with within the legal framework of the guild attests to their being part of the system. Second, as noted earlier, the katartarioi and metaxarioi referred to in provision VII.2 have been mistakenly perceived as poor; they were of servile status and for that Even if the number of ineligible metaxarioi and katartarioi was small, a problem could arise as to how the financial burden would be shared equitably by the other metaxopratai when the market price exceeded the capped price. The problem would be exacerbated if the price charged was well below the going market price. For, in this circumstance, buyers would have an incentive to increase their purchases at the capped price, especially since there was no question of affordability, whereas the sellers would be disinclined to be accommodating and forego legitimate profits. The same would be true whenever a large number of claimants was involved. Probably, the chiefs of the guild intervened, assessing requirements and devising ad hoc allocative mechanisms to ensure “fair” treatment. While the solutions may not have always satisfied all parties to the exchange, the fact remains that the matter could not have been left to the parties’ discretion. 61 ¨nfte, 8. For concurring views see note 53. Byzantinische Zu 62 Lopez, “Silk Industry,” 15–16. 60
280
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
reason ineligible to take part in the collective purchase. Third, the existence of metaxopratai unable to participate in the wholesale common purchase (VI.9) or of katartarioi able to join (VII.5) confirms only the varying size of firms in the respective guilds and the concomitant differences in financial strength of the membership—not the existence of unorganized, needy tradesmen and craftsmen operating outside the guild framework. Fourth, silk was an expensive commodity requiring substantial capital; wholesale deals probably reached double- and triple-digit figures in nomismata. Even retail sales exceeded single digits, judging from the prices of silks (IV.2, VIII.1). If indeed these businessmen had been so poor, they would have been unable to set up shop in the first place; nor can it be presumed that they were trading from their homes, as this was not allowed (VI.13). Hence, there was no question of traders or craftsmen being unable to afford the one-time, nominal entrance fee of two nomismata. It is possible, indeed understandable, that some guild members could not muster the funds to buy wholesale, at least not at all times, but this does not mean that they were indigent. The definition of poor in Byzantine law should always be kept in mind.63 Inevitably, those who either failed, not an unusual circumstance in the business world, or lacked the financial resources to operate as independents had to seek employment as hired workers.64 Finally, if the contested provision VII.2 is interpreted, as it should be, as pertaining to those not registered with the eparch, then the conclusion to be drawn is that, in those economic activities like the silk industry that were organized into guilds, membership was a precondition.65 However, guild membership was mandatory only for the proprietor of a business establishment, not for the craftsman employed as a laborer, who was rather denied admission. This fine distinction has eluded historians.66 See note 22. Simon questions the possible involvement of women in the silk trade because careers of women as salespersons were viewed as subordinate, and believes that provision VII.2 refers only to activities involving raw silk processing: “Seidenzu ¨ nfte,” n. 96. On the other hand, P. Koukoules maintains that women were involved both in the raw silk and yarn trade: Buzantinw'n Bi´o" kai` Politismo´ " (Athens, 1948), 2:233. To be sure, given the status of women at the time, their active involvement in running businesses was probably limited. Nevertheless, their participation, as implied in the Book of the Eparch, was apparently prompted by their need to take over the businesses of their incapacitated or deceased spouses to support themselves and their families, although undertakings by women possessing skills and business acumen should not be ruled out altogether. Besides, as already discussed, the reference in provision VII.2 of the Book of the Eparch is not to “poor” women but to women of servile status, for whom societal perceptions were not a matter of great concern and who probably ran shops in the raw silk trade as widows of slave-proxies and possibly as independent operators in raw silk processing since the guild of katartarioi accepted slaves (VII.3). The assertion that ¨ckle, Byzantinische Zu ¨nfte, 58; Zoras, Corporazioni the katartarioi did not admit slaves as guild members (by Sto bizantine, 175; and A. Vogt, Basile Ier et la civilisation byzantine `a la fin du IXe sie`cle [Paris, 1908], 391) has been convincingly refuted by Mickwitz: “Organisationsformen,” 73 and n. 2, and Christophilopoulos, jEparciko` n Bibli´on, 52 n. 1. 65 This view is also held by Christophilopoulos, jEparciko` n Bibli´on, 4, 36, 50; Mickwitz, “Organisationsformen,” 72–74; A. D. Sideris, JIstori´a tou' oijkonomikou' bi´ou (Athens, 1950), 264; Mendl, “Les Corporations byzantines,” 302, 304, 312–18. 66 Put differently, the guild was an association of owners and operators of trade or manufacturing establishments, not a halfway labor union, as wage workers were not members. This, inter alia, explains the involvement of Jews in the silk industry despite the fact that they were not allowed to become guild members. They were hired as workmen, apparently from among those independent, mobile, and highly skilled operatives (misqwtoi´) whom the metaxopratai and serikarioi employed on a fixed-term basis, as discussed in pp. 275–76 and 293. Benjamin of Tudela found among the Jews in the capital “artificers in silk”: The Itinerary of Benjamin of Tudela, trans. M. A. Adler (London, 1907), 14. J. Starr maintains that the Jews of Constantinople did not 63 64
GEORGE C. MANIATIS
281
¨ckle and The interpretation just advanced accepts the basic notion propounded by Sto espoused by most historians, namely that metaxarioi and metaxopratai were synonymous terms, but it rejects the hypothesis that the metaxarioi were poor and not members of the guild of metaxopratai. Instead, this new interpretation views the metaxarioi as metaxopratai in servile status—a notion that renders accurately the animus of provision VII.2 and preserves the internal consistency of all provisions of the law concerning the trade in raw silk. Melathrarioi: An Unappreciated Lot. Views concerning the work performed by the melathrarioi differ considerably. As already mentioned,67 the melathrarioi have been mistakenly identified with the metaxarioi, on grounds that the former is an inaccurate transcription of the latter and that the term is incomprehensible, is not found in the sources, and is not connected etymologically with silk (me´ taxa).68 Yet, the term must have been familiar and was probably associated with a certain aspect of the product traded or its processing. Moreover, the melathrarioi cannot be regarded as metaxarioi because the formulation of provision VI.15, the “so-called” melathrarioi (oiJ lego´ menoi melaqra´ rioi), clearly refers to a distinct group of traders. Furthermore, identification of the melathrarioi with the metaxarioi would implicate the metaxopratai in the trade of inferior quality silk, which explicitly was the preserve of the melathrarioi (VI.15). By the same token, the ¨ckle69 that melathrarioi was a popular term for hypothesis advanced by Nicole and Sto those metaxopratai dealing in damaged or soiled silk is equally indefensible. On the other hand, the identification of melathrarioi as “lathrarioi” (laqra´ rioi), a derivation from (la´ qra ⫽ secretly), suggestive of illegal merchants trading on the sly,70 entirely misses the mark. A final hypothesis, that the melathrarioi were spinners who processed the waste from the unraveling of cocoons (flock silk) or even dyers,71 is also unlikely, as their inclusion in the section dealing with the metaxopratai suggests that they were regarded as traders and not as katartarioi or serikarioi. As traders, not unlike the metaxopratai, they would not have been permitted to process raw silk or spin waste (VI.14). Imports of substandard raw silk and, especially, domestically generated waste silk include silk manufacturers but had among them tailors making silk garments: The Jews in the Byzantine Empire, 641–1204 (Athens, 1939), 28–29. A. Andre´ade`s also is of the view that the majority of Jews in the capital and later in other provincial towns were engaged as workers or dyers in the silk industry: “The Jews in the Byzantine Empire,” Economic History 3.9 (1934): 9. On the other hand, S. W. Baron (The Jewish Community [Philadelphia, 1942], 1:365), Zvi Ankori (Karaites in Byzantium: The Formative Years, 970–1110 [New York, 1959], 142), D. Jacoby (“Les quartiers Juifs de Constantinople `a l’e´poque byzantine,” Byzantion 37 [1967]: 181), and Lopez (“Silk Industry,” 23–24), all maintain that the Jewish silk manufacturers in the capital were organized into a guild. However, Muthesius (“Byzantine Silk Industry,” 10) has convincingly contested the existence of the Jewish silk guild, arguing that the only way Jews could be involved in the silk industry was as employees of metaxopratai or serikarioi. Furthermore, given the enforced division of labor among guilds, this hypothesis would either have confined all Jews to one stage of the industry’s operations or have allowed their guild to integrate vertically—both untenable positions. 67 See above, p. 276. 68 Christophilopoulos, jEparciko` n Bibli´on, 83 and 85 n. 2. 69 ¨ckle, Byzantinische Zu ¨nfte, 25–26. Nicole, Livre du Pre´fet, 81; Sto 70 Attributed to Sjuzjumov by Loos in Mendl, “Les Corporations byzantines,” 310 n. 17. 71 Mickwitz, “Organisationsformen,” 74–75; Muthesius, “Byzantine Silk Industry,” 33, 39. Zoras, Corporazioni bizantine, 173, viewed the melathrarioi as spinners, while Freshfield thought they were either dyers or dressers of raw silk: Byzantine Guilds, 23 n. 1.
282
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
made the involvement of the melathrarioi in the silk trade worthwhile. The quality of the silk is affected by the color, configuration, and fineness of the cocoons, as well as by its source as cultivated or wild. Also, cocoons could be damaged and reeled silk soiled during transport.72 The melathrarioi bought such blemished silk directly from external suppliers, participating along with the metaxopratai in the collective purchase. It is improbable that the melathrarioi were compelled to buy the low-grade silk through the intermediation of the metaxopratai, as has been asserted,73 because such transactions would run counter to the statutorily mandated division of labor between the two groups. Besides, because of their special knowledge in handling waste silk, the involvement of the melathrarioi in the negotiations with the external suppliers, whether as members of the negotiating team or in an advisory capacity, was essential to detect and avoid underhanded deals. In turn, the melathrarioi sold the imported low-grade and damaged silk to, or entered into `a fac¸on contracts with, specialist katartarioi who spun it into yarn.74 A more important source of waste was the material discarded by the katartarioi in the course of processing raw silk into yarn.75 This waste material was a valuable byproduct and gave rise to profitable processing and trading activities. Probably, not all katartarioi processed the accrued waste material themselves, preferring to turn it over to katartarioi specializing in the spinning of waste. Either way, the coarser, inferior quality yarn thus produced was sold back to melathrarioi as the solely authorized dealers, who in turn sold it to serikarioi for the manufacture of cheaper and blended mix-fiber silks.76 Contrary to expressed views,77 there is no evidence that the melathrarioi operated outside the guild system. If the melathrarioi were picking up the crumbs of the silk trade and were not guild members, as Lopez maintains, there would be no particular reason for the authorities to demarcate their function and levy penalties for deviant conduct (VI.15), deal with them along with the metaxopratai in the Book of the Eparch, or try to curb the alleged ambitions of these upstarts. Also, no evidence is adduced to prove the assertion that the social standing of the melathrarioi was inferior to that of the metaxopratai or that they were not part of the middle class.78 Trading in inferior quality silk need not be demeaning; nor need it imply that the merchants involved were worse off or belonged to a lower class. Use of the expression “so-called” (lego´ menoi) in provision VI.15 should not be viewed as connoting a condescending or pejorative reference to melathrarioi but rather as an ascription to a commonly known specialized trade. The fact remains that the authorities were eager that as much as possible of this waste material For details, see note 162. Loos asserts that the melathrarioi purchased silk of inferior quality either directly from abroad or from metaxopratai: in Mendl, “Les Corporations byzantines,” 310 n. 17. 74 See above, pp. 274–76. 75 Waste silk comprised floss silk (i.e., rough loose filaments of short length obtained from damaged cocoons and parts of the cocoon that cannot be reeled), flock silk (i.e., short and fluffy fibers left over during the reeling process), and noils. These fibers were combed and spun to make a coarser yarn. See references at end of note 14. 76 Muthesius hypothesizes that the melathrarioi spun the waste silk and sold the yarn to metaxopratai or to serikarioi: “Byzantine Silk Industry,” 33. As already argued, the melathrarioi were not spinners. Also, there is no apparent reason to believe that the statutorily mandated division of labor among and within guilds was not adhered to in the case of melathrarioi. 77 Lopez, “Silk Industry,” 16; Muthesius, “Byzantine Silk Industry,” 39. 78 ¨ckle, Byzantinische Zu ¨nfte, 25–26; Lopez, “Silk Industry,” 15; Zoras, Corporazioni bizantine, 172–73. Sto 72 73
GEORGE C. MANIATIS
283
be recovered because of its high value, and, as this activity formed an integral part of the silk industry’s operations, it was only logical that it be incorporated in the guild organizational structure and be dealt with in the Book of the Eparch. Manufacturing Structure Yarn Production: Dimensions of the Activity.79 The katartarioi engaged in all facets of cocoon and raw-silk processing that led to the production of yarn.80 Depending on the condition of the silk when acquired,81 they unraveled and reeled stifled cocoons, degummed reeled silk,82 and twisted silk threads to produce yarn of the desired tensile strength.83 Some katartarioi may have specialized in the spinning of coarser yarn from waste or low-grade silk, while others very likely processed raw silk `a fac¸on on behalf of metaxopratai. These successive and distinct sub-processes resulted in an elemental division of labor at the shop-floor level, which required a skill-mix acquired through apprenticeship and long practice. The primary source of labor was the family and possibly its slaves. Depending on the size of its operations, the family hired qualified workers or trained unskilled laborers as necessary. The katartarioi operated as principals, that is, as independent craftsmen, always within the guild system, owning or leasing their workshops.84 In more integrated establishments, different space arrangements were necessary because of the particular working conditions and environmental demands of individual sub-processes (e.g., degumming as opposed to reeling).85 At times, established katartarioi had direct access to imported raw silk but only up to their processing capacity.86 For the 79 Yarn production involved the following main tasks: unraveling and reeling; degumming; doubling and twisting; spinning of waste and floss silks. See Muthesius, “Byzantine Silk Industry,” 31–32; eadem, “From Seed to Samite,” 124. 80 Christophilopoulos, jEparciko` n Bibli´on, 85; Mickwitz, “Organisationsformen,” 71 n. 3; Koukoules, Bu¨ckle believed that the katartarioi were involved only in twisting and not in unravzantinw'n Bi´o", 187, 233. Sto ¨nfte, 26. Muthesius asserts that the katartarioi eling because they acquired the raw silk reeled: Byzantinische Zu only degummed the filament silk, while the metaxopratai performed the reeling and initial twisting with hired labor when the raw silk was not imported reeled; the serikarioi added twist to meet their own requirements for tensile strength: “Byzantine Silk Industry,” 32–33, 35, 39. The notion that the metaxopratai were involved in any stage of the raw silk processing using hired labor has been shown to be untenable (see above, pp. 271–74). 81 As stifled cocoons or reeled silk. See above, pp. 265–67. 82 Degumming is a process whereby the skeins of raw silk are soaked in warm water with soap to remove the cericin (see note 9). This makes the thread lustrous, soft, pliable, and workable, and enhances its dyeing properties. Degumming reduces the silk’s weight by about 25%. 83 Twisting or throwing involves doubling and twisting several strands of silk together to increase the tensile strength of the yarn. Although the task can be performed before degumming, most yarn is twisted after degumming. See the references at end of note 14, for details regarding the degumming and twisting processes. 84 See above, pp. 279–80 and note 46. 85 See below, p. 289 and note 99. 86 See above, p. 268. The Book of the Eparch does not specify the timeframe of such purchases: a fortnight’s supply, a month’s, several months’? The quantities in each instance differ by multiples, and this has important implications for inventory management. Mickwitz asserts that the katartarioi joining the purchase cartel met yearly requirements: “Organisationsformen,” 73. However, such a pattern-setting conjecture seems impractical, if only because it would have been very costly for craftsmen to carry such a large inventory of a high-priced raw material over such a long period, particularly since katartarioi were not allowed to sell unprocessed silk (VII.1). Although irregular, shipments to the capital were not infrequent, as imports originated in regions outside the empire as well. As mentioned in note 59, an array of factors could also affect
284
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
most part, the katartarioi procured raw silk obtained from the metaxopratai for cash or on credit.87 The Book of the Eparch enjoined the serikarioi to buy the silk (me´ taxan) from the metaxopratai, but it did not specify the silk’s form (VIII.8). This ambiguity has given rise to ¨ckle, the prevailing view holds that, in this a long-standing controversy. Following Sto instance, me´ taxa means yarn, which the katartarioi could sell only to metaxopratai. The rationale is that, while the law nowhere indicates that the serikarioi were obliged to buy the yarn from the katartarioi, it explicitly requires them to purchase it from the metaxopratai. Hence, the katartarioi, having processed the raw silk, must have sold the yarn back to metaxopratai in the open market, who in turn supplied the serikarioi.88 Attempts have been made to debunk the predominant theory, but, although astute and intriguing, the hypotheses advanced cannot stand scrutiny. Backward Integration into Yarn Production: A Strained Thesis. Simon has challenged the prevailing thesis and asserts that the katartarioi were allowed to sell yarn directly to serikarioi who either did not have the requisite processing facilities or were not interested in processing the raw silk themselves. In his view, the hypothesis that the serikarioi processed raw silk as well does not conflict with provision VIII.8, because the provision simply states that the serikarioi must meet their demand for raw silk from metaxopratai and not from external suppliers. A stipulation enjoining the serikarioi to procure silk they did not prepare themselves from katartarioi would have been superfluous, as they were the only source of supply. Besides, the term me´ taxa has various shades of meaning and is used loosely in the Book of the Eparch. Thus, serikarioi unable or unwilling to produce yarn in-house made their purchases directly from katartarioi; those who had the procurement requirements and timing. Finally, the katartarioi had ready access to metaxopratai to cover shortfalls, which was likely to have been less expensive than carrying excess inventories that tied up capital and space. In this connection, it should be noted that raw silk requirements can be determined only grosso modo. Estimates based on the equipment in operation or number of workers employed are unreliable. Capacity utilization of existing facilities and workforce output can vary considerably depending on total factor productivity which, inter alia, is affected by organizational arrangements, space, working methods and work ethic, the tensile strength demanded of the yarn to be produced, and market demand. Apparently, enforcement rested on bona fides, spot checks, and the deterrent effect of the severe penalties for violators (VII.1). 87 According to Mickwitz, the katartarioi relied heavily on suppliers’ credits because most of them were poor: “Organisationsformen,” 73. The market position of the katartarioi and their likely economic situation is discussed below, pp. 310–19. For the potential savings in working capital to be reaped by entering into `a fac¸on agreements, see above, p. 275. 88 ¨ckle, Byzantinische Zu ¨nfte, 27–28; Christophilopoulos, jEparciko` n Bibli´on, 83, 86; Mickwitz, “OrganiSto sationsformen,” 71–72; idem, “Un Proble`me,” 26; Lopez, “Silk Industry,” 19; Muthesius, “Byzantine Silk Industry,” 35, 39. While Simon, “Seidenzu ¨ nfte,” 33 nn. 47 and 48, points out the concurring views of Sjuzju¨ckle’s mov and Kazhdan, he himself dissents (see below, pp. 284–89). Mayer (BZ 21 [1912]: 532) accepted Sto view that the katartarioi were twisters (see note 80), but maintained that they sold yarn directly to serikarioi. In his view, the term metaxopratai in provision VIII.8 was much broader and meant to differentiate between all indigenous silk merchants and processors, i.e., including the katartarioi and external suppliers. Hence, he suggested, the serikarioi could buy yarn directly from the katartarioi without the intermediation of the metaxopratai. This hypothesis runs contrary to the sharp distinction made in the Book of the Eparch between metaxopratai and katartarioi and their respective guilds, as Christophilopoulos has pointedly observed: jEparciko` n Bibli´on, 86 n. 1. Moreover, this conception blurs the clearly distinct functions of the two groups and, in so doing, runs counter to the all-embracing provision (XVIII.5), which prohibits the concurrent practice of two trades.
GEORGE C. MANIATIS
285
requisite productive capacity bought the raw silk from metaxopratai in the form of reeled silk; and those in need of gummed yarn procured it from metaxopratai who had imported it in this form. However, gummed yarn was used only for the manufacture of silks to be degummed and dyed after they had been woven and their production accounted for a small fraction of the total output. Therefore, the bulk of the yarn was produced in degummed form jointly by katartarioi and serikarioi.89 Simon goes on to question the notion that the silk industry operated on strict rules of division of labor, as the Book of the Eparch does not define in precise technical terms the respective functions of the metaxopratai, katartarioi, and serikarioi. The provisions about the serikarioi, in contrast to those for leather workers (XIV), do not differentiate between specific tasks performed during the manufacturing process. Hence, the serikarioi could also unravel, degum, and twist the silk filaments, performing in parallel the same tasks as the katartarioi. Also, no stipulation forbade the serikarioi from processing raw silk (katarti´zein), while the degumming and dyeing processes are closely connected and difficult to separate. Furthermore, if the metaxopratai had been allowed to be involved in yarn production, they would have deprived the katartarioi of their livelihood, appropriating work otherwise reserved for the katartarioi and, in addition, triggering competition between the two groups. The serikarioi, on the other hand, produced an intermediate product, which they did not sell but used themselves in the manufacture of finished goods. Compulsory guild membership (Zunftzwang) did not imply that the members of a guild were not allowed to perform tasks that other guilds made a business of, as long as this did not lead to market competition. Raw silk processing by serikarioi could not possibly lead to direct sales competition with the katartarioi. Nor could yarn production by serikarioi lead to competition for work, because they were not the only customers of the katartarioi; the latter also met the demand of private households. Besides, the katartarioi were small producers who hardly ever employed hired workers, and it is not certain whether their output even began to satisfy the requirements of the serikarioi. Finally, the involvement of the serikarioi in yarn production explains their need to hire workers, as well as the necessity for special provisions to regulate such hirings (VIII.10, 12). In contrast, no such provisions applied to the katartarioi.90 In the same vein, Mendl hypothesized that the katartarioi were both yarn producers and weavers, who manufactured simple undyed silk fabrics, as opposed to the high-value dyed silks produced by the serikarioi. Although distinct, the two processes were not performed by different guilds as was later the case in Venice; they were undertaken in parallel and not as successive phases of the productive process. Put differently, the serikarioi did not take over where the katartarioi left off. Mendl based his argument on the following logic: the Book of the Eparch does not stipulate that the serikarioi should buy yarn (soie tordue) from the katartarioi; on the other hand, me´ taxa refers to raw silk (soie gre`ge) in all other instances, and must do so in provision VIII.8; hence, there is no reason to resort to a strained interpretation of me´ taxa as yarn. Clearly, the serikarioi bought raw silk. 89 “Seidenzu ¨ nfte,” 26–32, 34, 44–46. Muthesius, too, does not exclude the possibility that the katartarioi sold yarn directly to serikarioi: “Byzantine Silk Industry,” 33. 90 Simon, “Seidenzu ¨ nfte,” 24, 26–28, 32–34, 44–46.
286
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
Furthermore, it has never been demonstrated that katarti´zein is not a technical term applicable to weaving as well.91 These are bold but unconvincing hypotheses. Mendl depicts the katartarioi as weavers as well, stretching the meaning of katarti´zein (to dress) to include weaving, a meaning without etymological foundation. Katarti´zein clearly denotes actions pertaining to a preparatory stage. Moreover, the law’s section on katartarioi contains no provisions similar to those in the section on serikarioi regarding the kind of silks to be manufactured, dyes to be used, inspection of facilities and output, disposal of silks, labor hiring procedures, or sale of skilled slaves to foreigners (VIII.1–5, 7, 9–12), as would be expected from a vigilant bureaucracy eager to enforce the rules strictly and uniformly. This very omission underscores the perception of katartarioi as the sole producers of an intermediate product, that is, yarn, making unnecessary the surveillance of such activities by the eparch to ensure compliance. Furthermore, the alleged division of labor between katartarioi and serikarioi, the former producing less valuable and the latter more expensive silks, is arbitrary and unenforceable, since, in the absence of an explicit prohibition, nothing could stop the katartarioi from manufacturing dyed silks or the serikarioi from producing undyed articles, if this was to their best interest. Finally, there are instances in the Book of the Eparch where it is quite evident that me´ taxa means yarn, as in provisions VII.6 and VIII.4, suggesting that the term does not always imply raw silk, as has been alleged. The serikarioi, according to Simon and Mendl, were also raw silk processors, performing the entire gamut of tasks otherwise assigned to katartarioi. Yet, the respective sections on katartarioi and serikarioi in the Book of the Eparch provide no indication that this was indeed the case. Section VII about katartarioi nowhere indicates that they manufactured silks (ejrga´ zontai iJma´ tia), whereas it clearly states that they processed raw silk ((kat)ejrga´ zesqai th` n me´ taxan). On the other hand, section VIII on serikarioi does not even hint at their involvement in raw silk processing (katarti´zein th` n me´ taxan). In both instances, critical terms are absent. Further, the alleged need of the serikarioi to hire workers to produce yarn in-house has been misconstrued. The employed workers were very skilled, highly paid, and mobile, were hired for the duration of a particular job, and performed tasks strictly related to weaving, dyeing, and tailoring; they were not hired to do routine jobs on a steady basis.92 Also, although consecutive, degumming and dyeing are altogether different processes requiring distinct skills and are technically conducive to a division of labor. There is therefore no compelling reason that they be performed under the same roof, especially since dyeing, unlike degumming, is an intermittent activity. Smaller serikarioi probably did not have in-house dyeing facilities.93 Finally, the fact that the katartarioi could participate in the collective purchase of raw silk from external suppliers (VII.4, 5), while the serikarioi, though presumed wealthier, were explicitly excluded (VIII.8), indicates the law’s intention that the serikarioi not expand their activity into yarn production, and not simply to protect the purchase monopoly of the metaxopratai. If the serikarioi had also been yarn producers, it would be only natural that they
Mendl, “Les Corporations byzantines,” 306 and n. 16; see also Loos’s comments in idem, 309–11 n. 17. Runciman also shares this view: “Byzantine Trade,” 155. 92 See below, p. 293. 93 Ibid. 91
GEORGE C. MANIATIS
287
too be invited to participate along with the katartarioi and under the same conditions. The fact that they were not suggests that they had no involvement in raw silk processing. In this context, Simon queries why the metaxopratai were explicitly denied the right to branch out into yarn production (VI.14) while the serikarioi were not? Or, why there were no explicit restrictions preventing katartarioi from becoming serikarioi, or serikarioi from becoming metaxopratai? His answer is that the rules governing the division of labor were lax and that one guild could take over another’s functions as long as this would not lead to market competition.94 However, the absence of explicit restrictions on every possible inter-guild move is not surprising. The law’s intention was to curb the parallel practice of trades or crafts in whatever form, not the movement from one occupation to another within the guild system. This explains the law’s prohibition of vertical integration (expansion into preceding or succeeding phases of production or distribution: IV.7, VI.14, VII.3, VIII.6) and horizontal integration (expansion into similar or diverse activities conducted by other guilds: IV.1, V.1), whether individually or through a partnership leading to a union of trades. Insistence on a strict division of labor prevented enterprise growth unrelated to market demand, dominance of the market by a few large firms, and weakening or elimination of competitors based on sheer market power—in short, monopolization or lessening of competition. The specific provisions enjoining members of the silk guilds, as well as members of other guilds (II.1, X.6, XI.2, XIII.1, XV.1), from practicing simultaneously another trade or craft were indicative and conditional: if the guild member relinquished his current occupation, he could take up the desired activity. Serikarioi could therefore legally engage in yarn production (or any other trade for that matter), but only if they were prepared to give up weaving. Guilds not explicitly enjoined were subject to the blanket provision XVIII.5.95 Indeed, the for“Seidenzu ¨ nfte,” 32–33, 45. The provision of the Book of the Eparch enjoining guild members from enrolling in more than one guild at a time is a repetition of the same stipulation in the Basilics, LX.32.1: eJno` " de` mo´ non ojfei´lei susth´ mato" ei«nai kai` ajnacwrei'n tw'n loipw'n. Several hypotheses have been advanced concerning the rationale for this prohibition. J. P. Waltzing believes the rule was enacted to keep guilds from banding together and becoming ´ tude historique sur les corporations professionnelles chez les Romains (Louvain, 1895), 1:147 a threat to the state: E ¨ckle argues that the rule aimed to foster labor specialization, facilitate state n. 1, 150, 354 and n. 3. Sto supervision of industrial, trade, and labor activities, make easier the protection of industrial secrets, and ¨nfte, 98–99. C. M. Macri maintains thwart theft and the purchase of prohibited merchandise: Byzantinische Zu that the restriction intended to ensure political stability by keeping the industrial and trade communities under constant submission to the imperial will; to simplify the detailed supervision of manufacturing processes so as to sustain the industry’s high reputation; to further labor specialization and productivity; to curb inter-craft rivalry as antithetical to Christian ethics; to facilitate the levy of taxes and prevent fiscal fraud; and to facilitate the exercise of the purchase monopsony in imports: L’Organisation de l’e´conomie urbaine dans Byzance sous la dynastie mace´doine (Paris, 1925), 35, 40–41, 43, 46, 57–58, 61. Zoras asserts that the provision simplified administrative control, promoted the division of labor and specialization, and impeded the coalescence of guild interests that could threaten public order and the state itself: Corporazioni byzantine, 76–78. In reality, these hypotheses seem peripheral at best. As far as silks were concerned, state control did not focus on maintaining quality standards but on preventing the production of prohibited articles. Quality was largely determined by market demand, innovative activity, and competition: see below, pp. 319, 323, 325, and note 196. In a few instances, a guild’s designated domain was broadened: subspecialties with a limited number of craftsmen (e.g., dyers, melathrarioi) that did not justify the creation of a separate guild were placed under the domain of a closely related guild (e.g., serikarioi, metaxopratai) in order to meet technico-economic imperatives; see also below, pp. 292–93. On the other hand, administrative convenience seems to have been more apparent than real. Finally, the potential threat to law and order has been based largely on social 94 95
288
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
mulation of this provision, designated as cardinal (kefalaiwdesta´ th dia´ taxi"), that no one could enter the craft of another and practice both concurrently but had to choose between them and inform the eparch of his choice, leaves no doubt that the simultaneous practice of weaving and raw silk processing, whether or not the output (yarn) was marketed, contravened the law because it violated a guild’s inalienable right to practice exclusively its designated craft.96 Simon’s argument that the output of the katartarioi was barely sufficient to satisfy the needs of the serikarioi and that the latter had therefore to take up in-house yarn production lacks foundation. First, no evidence has been adduced to prove persistent shortages that would justify the involvement of the serikarioi in yarn production. More important, continued shortages would have led to higher prices and increased yarn production through expansion of existing capacities and/or new enterprises. The guild of katartarioi included establishments of a wide range of sizes, as is evidenced by those katartarioi who were invited to partake in the wholesale collective purchases of imported raw silk. Entrenched katartarioi would have seized the opportunity to increase their productive capacity through self-financing, borrowing, and partnerships or a combination of these methods. Higher profit margins would have attracted new entrants. Metaxopratai very likely would also have stepped in, extending credit or entering into `a fac¸on arrangements with less liquid katartarioi to help them build up their capacities, as the gestation period for investments in raw silk processing was rather short. By being forthcoming, the metaxopratai would have kept the serikarioi at bay and secured the yarn market for themselves, and they would have done so without breaching the law. Simon’s dim view of the katartarioi emanates from the common misperception of them as an underclass of small craftsmen unresponsive to market signals and without financial means.97 Even if, as Simon suggests, shortfalls induced the serikarioi to meet their yarn requirements through in-house production, it is unlikely that all serikarioi could afford or would be willing to organize and operate raw silk processing facilities. Larger serikarioi would have seen this as an opportunity to exploit a niche in the yarn market and would considerations and not on the more critical implications of concentration of economic power in the hands of a few in the capital (see above, p. 269). Fundamentally, the principle of “one man, one trade” was intended to curb the cumulation of monopoly power by entrenched guild members. 96 The Book of the Eparch takes the same approach when it only sparingly prohibits in an explicit way any attempt by a guild member to raise the rental of another’s workshop deceptively or even openly (IV.9, IX.4, X.3, XI.7, XIII.6, XIX.2), relying in other instances on the blanket provision XVIII.5. 97 Historians have invariably depicted the katartarioi as lower-class craftsmen, largely servile, poor, and of ill repute: Mickwitz, Kartellfunktionen, 212–13; idem, “Un Proble`me,” 25–26; idem, “Organisationsformen,” 72–73; Lopez, “Silk Industry,” 16 and n. 3, 18–19; Simon, “Seidenzu ¨ nfte,” 24, 45–46; Muthesius, “Byzantine Silk Industry,” 35. This unrealistic perception derives from a misreading of provisions VII.2 and 5, as discussed above (pp. 276–78) and below (pp. 310–17), and a misunderstanding of provision VII.6 which aimed to set standards for the orderly conduct of commercial transactions, requiring the katartarioi to conduct their affairs in a businesslike manner. Specifically, the katartarioi were forbidden to adulterate or sell their wares dishonestly (mh` kaphleu´ ein th` n me´ taxan), e.g., by adding sizing to increase its weight. They were also forbidden to conduct business in a noisy, boorish, irascible, or importunate manner. Certainly, the provision did not aim to hinder haggling (Schachern⫽kaphleu´ ein), as Simon hypothesized: “Seidenzu ¨ nfte,” 44. Bargaining is an inseparable and essential part of trade negotiations, and it would be unthinkable to try to stop it by fiat. On the other hand, Simon’s dismissal of Sjuzjumov’s view that provision VII.6 intended to hinder katartarioi from drinking away their earnings is well-taken: ibid., n. 95. Generalizations based on such misconceptions are clearly wide of the mark, and have unjustifiably maligned the katartarioi as a professional group.
GEORGE C. MANIATIS
289
have been tempted to build excess capacity to satisfy the needs of fellow serikarioi, possibly at more competitive prices than offered by the katartarioi since the serikarioi were in a position to price at marginal cost. This would have created market competition between serikarioi and katartarioi that not only was against the law but that would have seriously threatened the latter’s very existence—an intolerable situation to be resisted by the guild of katartarioi and the authorities. To guard against such an eventuality, avoid de facto validation of such an encroachment, and prevent legitimization of such transactions, the raw silk processing activities of the serikarioi would have been policed to ensure that their yarn production did not exceed their own requirements. Yet, no such concern is voiced in provisions VIII.3 and 4, which refer only to inspections relating to weaving and dyeing, stamping of fabrics, and conformity of the cloth produced with statutory regulations. Legislative silence on such an important matter suggests that the issue never arose. In consequence, it is unlikely that the serikarioi ever engaged in yarn production. Nor does it follow that the involvement of the metaxopratai in yarn production would have threatened the livelihood of the katartarioi, in a way that the serikarioi would not, because the katartarioi could have fallen back on household demand. Despite Simon’s assertions, it is immaterial which, if any, group encroached on yarn production because the outcome would have been the same. In either case, the displaced katartarioi (whether owners of workshops or laborers) would have been absorbed by the intruding group, and only the employer and possibly the place of employment would have changed. Furthermore, the alleged substitute demand from private households, which was insignificant in volume, could hardly have made up for the loss of business incurred by allowing the serikarioi (or the metaxopratai, for that matter) to engage in yarn production. Although silks undoubtedly were coveted, most could not afford them, and few of those could resort to home production. Simon believes further that impoverished and wretched katartarioi, producing small quantities of yarn without outside help, supplied a contingent of equally small-scale serikarioi, also operating with modest equipment and without hired workers. In his view, both groups performed their work at home.98 However, institutional and operational imperatives militated against home-based activity. The tenor of the regulatory apparatus fostered the organization of silk manufacturing as a workshop industry, in particular, the formation of clusters of specially built sheds in designated areas. Such concentration, unlike a dispersed household industry, facilitated the exercise of state control over the firms’ activities (IV.1, 3, 4, 8; VI.13, 16; VIII.5, 7, 9). Space requirements and environmental regulations concerning fire hazards, smoke, and noxious fumes precluded the conduct of such activities in residential areas.99 Finally, the technological and production requirements of tasks that could not be performed efficiently, if at all, in cramped home settings and the need to ensure management control over production processes liable to costly human errors reinforced this tendency.100 “Seidenzu ¨ nfte,” 45–46. ´tou ,” jEp. JEt.Buz.Sp. See D. Ghinis, “To` jEparciko` n Bibli´on kai` oiJ No´ moi jIoulianou' tou' Askalwni j (1937): 187. 100 See below, pp. 319–20. Muthesius (“From Seed to Samite,” 127, 129) disputes that one serikarios alone could perform the range of tasks involved at this stage of production, as Simon implies. It is doubtful that he could even handle a single process, e.g., weaving or dyeing. 98 99
290
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
The Circuitous Yarn Distribution Pattern: An Explanatory Hypothesis. A nagging question still remains: what special function(s) were the metaxopratai expected to discharge that the authorities entrusted their guild with a monopoly in the sale of yarn, not allowing the katartarioi to sell yarn downstream directly to the serikarioi as would seem to be more rational? One could hypothesize that the metaxopratai as a group somehow asserted themselves when the ordinance was drawn up and managed to wrest this prerogative from the authorities. By steering the distribution of yarn their way, the metaxopratai would expand their activity and profits without encroaching on the exclusive right to process raw silk that was held by the katartarioi (and to whom it was immaterial whether yarn was sold to metaxopratai or serikarioi). However, viewing this seemingly unwarranted arrangement as simply a case of the authorities yielding to pressure is a facile explanation, as it avoids important underlying issues and ignores the possibility that a set of practical considerations may well have been the determining factor. The following three-pronged hypothesis attempts to provide a more plausible explanation. The circuitous distribution pattern that was adopted was probably prompted by the need to establish a workable information gathering network to conduct efficiently the raw silk import trade from which all parties stood to gain. To exercise more intelligently their purchase monopsony, the metaxopratai had to have current information about the proximate level of the prospective aggregate demand (⫽total demand less inventory) for raw silk before each market day. This information enabled them to estimate the gap between domestic demand and external supply,101 establish negotiable base prices and an initial bargaining position, calculate the contributions of the membership, and determine whether katartarioi should be invited to secure supplemental funds for the upcoming purchase. In a metaxopratai → katartarioi → serikarioi model, the information supplied to the metaxopratai would basically reflect the estimates of the katartarioi, which, apart from the inherent difficulties in assessing raw silk requirements,102 were subject to inadvertent or intended distortions. In addition, the interposition of the katartarioi might impede the timely collection and transmission of the information. By contrast, in a metaxopratai → katartarioi → metaxopratai → serikarioi model, where the metaxopratai are simultaneously in contact both with the suppliers of the intermediate input and its end-users, the requisite information regarding demand requirements, productive capacities, and approximate inventory levels could be obtained promptly and largely unfiltered. This would enable the metaxopratai to determine aggregate requirements with much greater accuracy and enhance their bargaining power vis-a`-vis suppliers. By the same token, a model whereby the metaxopratai sold raw silk to both katartarioi and serikarioi, while the katartarioi sold yarn to a segment of the serikarioi (Simon’s model), would diffuse further the information network and exacerbate the shortcomings encountered in the first model. Another useful service rendered by the metaxopratai pertains to inventory carrying, as inventory adjustments had significant balancing effects. In the face of irregular inflows The approximate level of the external supply of raw silk was known to the metaxopratai, as the eparch’s office apprised the guild of the quantities that had entered the mitata (see above, p. 267). 102 See note 86. 101
GEORGE C. MANIATIS
291
of raw silk and a fluctuating demand for silks, internal demand for and supply of input (raw silk and yarn) could be balanced in the short run only through inventory and/or price adjustments. But the katartarioi were not allowed to carry marketable surpluses, while the serikarioi could shift the burden of carrying inventories to the metaxopratai.103 Therefore the financial and other costs associated with maintaining an expensive inventory sufficient to act as a buffer against disruptions in production and wide swings in prices fell squarely on the metaxopratai. Hence the need to defray the high costs involved.104 Finally, the official thinking may have been that in a competitive market the combined profit of the metaxopratai from handling both raw silk and yarn trade activities would still be lower than the sum of the profits extracted by the metaxopratai and the katartarioi if the two activities remained separate. The compression of the metaxopratai’s mark-up, the tax they were obligated to pay to the state, which was based on the purchased quantities of raw silk (VI.4) and could not be readily passed on under competitive conditions, and the possibility that the serikarioi might be able to shift at least part of the increased cost to foreigners and the well-off (i.e., persons whose demand would not be influenced perceptibly and could afford higher prices) would make the palindromic trade pattern adopted by the authorities more palatable to all concerned. The authorities’ decision to interpose the metaxopratai between yarn and cloth production was not devoid of pragmatism and a certain rational basis considering that market intelligence was vital in negotiations with external suppliers; that the advantageous buffer function performed by the metaxopratai was not without cost, while the alternative would have been that it be discharged by the state via market interventions (e.g., price controls) and hence much less efficiently; that the metaxopratai could not shift the entire amount of the tax levied on them; that the serikarioi for the most part could offset the mark-up on the price of yarn by the reduced costs of inventory carrying and by passing it on to foreign traders and ultimately (through the retailers) to wealthy buyers; and that competitive forces were likely to keep the profit margins of the metaxopratai at the lowest possible levels.105 Weaving, Dyeing, Tailoring: Organization of the Production Process. The serikarioi manufactured silks with yarn produced by katartarioi but procured through the intermediation of the metaxopratai. The serikarioi were not integrated backward into yarn production or forward into retailing.106 Still, silk manufacturing required fixed investments in weaving and dyeing facilities,107 as well as working capital to meet operating Since the serikarioi could rely on the metaxopratai for immediate and uninterrupted access to yarn, they could reduce their inventories to a bare minimum and economize on working capital. 104 Aside from storage costs, interest on commercial loans ran at over 11% a year in the tenth century: Basilics, XXIII.3; J. B. Bury, Appendix 13, in E. Gibbon, The History of the Decline and Fall of the Roman Empire (London, 1898), 5:533–34; G. Cassimatis, Les Inte´reˆts dans la le´gislation de Justinien et dans le droit byzantin (Paris, 1931), 120–23. 105 See below, pp. 310–19. 106 See pp. 284–89, 294–95, 297. 107 There was a long-standing imperial monopoly in purple-dyed silks and the use of purple dyes of the finest quality (murex purple); see Codex Theodosianus, X.21.3; Codex Justinianus, IV.40.1; Basilics, ¨ckle to infer that the dyeing of silks both for imperial and XIX.1.80; Book of the Eparch, VIII.4. This led Sto public use was carried out in the imperial dyeworks as a state monopoly. Accordingly, the serikarioi were not 103
292
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
expenses, including costly inventories in inputs (e.g., silk yarn, dyestuffs), goods in process, and finished products. Also, it was a fairly risky business, subject to potentially costly human errors in design, dyeing, or weaving and to the fluctuations of market demand. Serikarioi had therefore to be persons of some substance although it is unlikely that all of them could afford to combine all of the various sub-processes, own their workshops, or employ large numbers of workers. The size of establishments in silk manufacturing must have varied widely. The nature of the tasks performed by the serikarioi required special space arrangements, for example, sheds in separate but adjacent locations, and a mix of worker skills. Those who could not afford to own their workshops leased space from wealthy property owners.108 Apparently, industrial space was in short supply at times, as is evidenced by the fact that those who surreptitiously or openly sought to raise rental rates for workshops were severely punished (XVIII.5). Skill specialization and the practical need for separate sheds to perform the different tasks and comply with safety and environmental requirements,109 would, however, certainly not require that each sub-process or specialty allowed to dye themselves but had to procure dyed yarn from imperial dyers through the intermediation of ¨nfte, 28–31. Boak holds the same view: “Book of the Prefect,” 609 n. 6. the metaxopratai: Byzantinische Zu ¨ckle was amended by Novel 80 of Leo VI. Thus, However, the law (Basilics, XIX.1.80) referred to by Sto although the provision still prohibited the private manufacture and sale of large pieces of silk fabrics dyed with murex purple (porfu´ ran), it abolished every restriction regarding the sale of small pieces of purpledyed silks. Besides, inferior quality purple dyes, which were freely marketed, could produce a range of red shades, e.g., purple, scarlet, or crimson, by applying different dyeing techniques. Cf. Jacoby, “Silk in Western Byzantium,” 482–84. The confluence of certain compelling factors, namely that the prohibition to use murex purple to dye yarn or cloth (except for trimming clothes with purple pieces or borders) is repeated in the Book of the Eparch only in connection with the serikarioi (VIII.4); that the serikarioi are explicitly allowed to manufacture certain types of dyed silks (VIII.1, 2), while they are not expressly forbidden to engage in dyeing; the nonexistence of a dyers’ guild; and the lack of any evidence that the katartarioi dyed silk yarn leaves no doubt that the serikarioi engaged in dyeing as well. Cf. Gehring, “Zunftwesen Constantinopels,” 584; Zoras, Corporazioni bizantine, 176; Christophilopoulos, jEparciko` n Bibli´on, 65, 86; Lopez, “Silk Industry,” 10–11 and 14–15; Simon, “Seidenzu ¨ nfte,” 34; L. Bre´hier, Le Monde byzantin: La civilisation byzantine (Paris, 1950), 188. The work was largely performed by hired expert dyers, possibly freelance (see below, p. 293). Dyers specializing in purple dyeing were called porfurobafei'" or ojxeoba´ foi: Koukoules, Buzantinw'n Bi´o", 186. The serikarioi procured dyestuffs of inferior quality from dealers in perfumes (X.1) and murex purple dyes from state-authorized vendors (Porfuropw'lai). See B. Koutava-Delivoria, “Les ojxe´ a et les fonctionnaires nomme´s tw'n ojxe´ wn: Les sceaux et les ´etoffes pourpres de soie apre`s le 9e sie`cle,” BZ 82 (1989): 181–82 and fig. 3, no. 39. 108 See N. Oikonomide`s, “Quelques Boutiques de Constantinople au Xe s: Prix, loyers, imposition (Cod. Patmiacus 171),” DOP 26 (1972): 345–56. In this context, Jacoby maintains that the sale by metaxopratai and prandiopratai of their shops (ejrgasth´ ria) to dignitaries and wealthy individuals was prompted by their illiquidity or their subjection to increased pressure from creditors who, taking advantage of their wealth and social status, brought them under growing financial dependence: “Silk in Western Byzantium,” 477 and 479. Jacoby’s assertion is gratuitous. In the first place, out of the five transactions mentioned by Oikonomides, only one involved sale of property by a guild member, that by a metaxoprates. The other four transactions involved wealthy individuals or members of the aristocracy. Secondly, it cannot be confidently concluded that financial straits were the real reason for the sale of the metaxoprates’ property and, even so, that he had been squeezed by his lender, assuming indeed that lender was the buyer, something for which there is no proof. The metaxoprates might well have been the victim of adverse economic circumstances, old or sick, or willing to retire; he might have needed the money to meet obligations unrelated to his trade; or he might have wanted to increase his liquidity in order to expand his business while retaining use of the premises as a tenant. In any event, one case out of five does not provide grounds to infer the exercise of undue pressure by unscrupulous upper-class creditors over members of the silk guilds. 109 See above, p. 289.
GEORGE C. MANIATIS
293
be organized as an independent guild with the guild of serikarioi forming the superstructure, as has been asserted.110 If indeed other guilds were involved in silk manufacturing, the Book of the Eparch would have not failed to mention them, as they were important constituent elements of the silk industry’s organizational makeup. Significantly, the processing of raw materials, which is an integral part of the silk manufacturing process, is dealt with as a separate guild (katartarioi) in the law, and this is inconsistent with the notion that the guild of serikarioi was an amalgamation of several guilds. Similar subsumption of closely related activities under one guild also occurred in tanning and leatherworking (XIV.2) and the trade in pure and waste silk (VI.15). The prevailing setup implies only that managing larger undertakings with fairly unified operations, involving procurement, dyeing, weaving, tailoring, and marketing activities, entailed a degree of sophistication in production and financial planning, a developed sense of coordination of the various subprocesses, and the existence of a rudimentary pyramidal management structure. Dyeing units were much smaller than weaving units and operated in shorter and intermittent intervals. This could cause process balancing problems and could have resulted in suboptimal utilization of capacity. Larger producers were able to achieve a proximate balance more easily because of high production levels, but smaller producers may have found it uneconomical to own and operate a dyeing facility, as equipment and highly skilled labor might have remained grossly underutilized. In such circumstances, it would have been beneficial for them to contract out this phase of their operations to larger serikarioi, an arrangement that would also have ensured fuller utilization of the large producers’ facilities and cut-rate charges to small producers. Medium-sized producers, on the other hand, may have owned dyeing sheds but hired expert dyers on a fixedterm basis (misqwtoi´) to operate them as required. Depending on the level of their activity, the serikarioi employed themselves, family members, their slaves, and hired labor. As in the case of metaxopratai, the law forbade them to enter into labor contracts exceeding one month or knowingly to take into their service the worker of another before the worker had completed the tasks for which he was paid by his current employer (VIII.10, 12). These provisions obviously refer to independent operatives commanding special, highly marketable skills in short supply: proficient carpenters, dyers, pattern and prototype designers, weavers, printers, embroiderers, or tailors, on whom the serikarioi had come to rely for the production of highvalue fabrics and whose services were available only on a temporary basis, although such contracts could be renewed or renegotiated. Certainly, the Book of the Eparch was not referring to low-skill steadily employed workers whose numbers could be easily increased through training on the job.111 The state was eager to maintain an adequate supply of trained workers, considered Muthesius, “Byzantine Silk Industry,” 14–15, 35, 39–40. See also above, pp. 275–76. The only way a person could acquire a skill was to practice the craft as an apprentice to an experienced craftsman. Though unstructured and probably unsupervised by the state, this was apparently a workable institutional arrangement for vocational training. The mandated and strictly hierarchical organizational structure of the industrial workforce of apprentices, journeymen, and masters found in the West, as well as the detailed regulations concerning apprenticeship, advancement, and conditions of work, did not exist in Byzantium. 110 111
294
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
to be a valuable community asset. To this end, the law inflicted severe penalties on those who sold outside the capital any slave employed as misqwto´ " or a foreman (VIII.7). At the same time, the provision aimed to curb the export of know-how to foreigners and provincials alike, thereby hampering efforts of existing or prospective competitors to gain a competitive edge.112 As a defensive and not uncommon tactic aimed at protecting industrial secrets, the measure made business sense, as thriving silk manufacturing centers already existed in the Islamic world. The prohibition of such sales to provincials as well was prompted by concern that these then-knowledgeable persons would eventually emigrate abroad, an action that was beyond the control of the authorities outside the capital. It is unlikely that the main thrust of the provision was to forestall external competition, as evidenced by the unrestricted importation of silks of comparable quality. It is also unlikely that the provision’s extension to provincials was intended to shelter local industry and thwart the growth of rival manufacturing centers within the empire, as Lopez maintains,113 as long as there were no restrictions on imports. Besides, the measure would have been ineffective in that regard because of the possibility of legal emigration of skillful free men enticed by sufficient incentives. Indeed, the measure could not, and did not, impede the development of silk manufacturing in Corinth, Thebes, Patras, Euripos, or Andros when the right economic conditions were met from the late tenth century onward.114 The ability of the authorities to thwart the tendency of more aggressive serikarioi to integrate vertically or horizontally has been dismissed uncritically. Lack of movement in this direction has been attributed to the thin market for silks rather than concern for the legal protection of guild equality.115 Nevertheless, there is convincing evidence that the demand for silks was strong and growing.116 Moreover, if the serikarioi dominated the other guilds, if there was a rising oligarchy of masters and the law could be flouted with impunity, as has been alleged,117 a small market would have provided a strong economic incentive to serikarioi to expand their activities into other stages of production The rendition of provision VIII.7: oJ oijke´ thn [,] h‘ misqwto` n h‘ ejkle´ kthn [,] ejxwtikoi'" h‘ ejqnikoi'" pipra´ skwn as “whoever sells a slave, a hired worker, or an overseer to persons outside the city” is misleading, because it conveys the fallacious notion that the sale of both slaves and freemen was prohibited and that the sale of a workman, regardless of the level of his skills, was banned. See Nicole, Le Livre du Pre´fet, 37, and his French translation of the law in jEparciko` n Bibli´on, 164; Boak, “Book of the Prefect,” 609; Freshfield, Byzantine Guilds, 26; Lopez, “Silk Industry,” 23. The misunderstanding arises because a comma is missing after the words oijke´ thn and ejkle´ kthn in the Greek text. After the insertion, the passage reads “Whoever sells a slave, whether a hired worker or a foreman,” as Koder has correctly emended: Eparchenbuch, 107. The provision could not possibly refer to freemen since only slaves could be sold. Also, the provision clearly implies that misqwtoi´ were salaried persons possessing valuable skills and technical knowledge, since it places them on an equal footing with those who had supervisory capabilities, and explains the eagerness of the authorities to prevent their loss. The provision did not prohibit the migration or emigration of freemen, an equally convenient vehicle for the transfer of skills and know-how, probably because such voluntary departures were not anticipated. People with skills and ambition gravitated toward the capital; they were not leaving the city. 113 “Silk Industry,” 23. 114 Harvey, Economic Expansion, 213–20; M. Angold, “The Shaping of the Medieval Byzantine ‘City,’” ByzF 10 (1985): 11–16 and 24–25; idem, The Byzantine Empire, 86–87 and 281–82; A. P. Kazhdan and A. Wharton Epstein, Change in Byzantine Culture in the Eleventh and Twelfth Centuries (Berkeley, 1985), 41; Jacoby, “Silk in Western Byzantium,” 461–71. 115 Lopez, “Silk Industry,” 19. 116 See below, pp. 326–28. 117 Lopez, “Silk Industry,” 16, 18, 19. 112
GEORGE C. MANIATIS
295
and distribution to capture a greater share of the market. That they did not should be ascribed to the effective enforcement of the law’s provisions.118 Weaving, Dyeing, Tailoring: Market Outlets. The market of the serikarioi comprised foreign and provincial wholesale traders, the local vestiopratai, and the Treasury (eijdiko´ n). The Treasury probably bought silks retail in the open market from vestiopratai and prandiopratai, but probably the bulk of its purchases was procured wholesale by placing orders with serikarioi.119 This would have enabled the Treasury not only to pay lower prices but also to stipulate its own specifications regarding design, color, and quality, and to fix convenient delivery dates. Such transactions are clearly alluded to in provision VIII.11, which penalizes the serikarioi if they delivered to the imperial warehouse garments that they had not manufactured themselves but had acquired from outside the capital (ejxwka´ mata). In this connection, Lopez asserts that the cloth the serikarioi supplied to the Treasury was a tax in kind (munus),120 a view that is indefensible. Provision VIII.11 does not even hint at this notion, as it talks about delivery (didou´ "), implying discharge of a commercial contractual obligation, not payment of a tax liability. If this had been such a levy, it would have been specified and quantified, as was the case for metaxopratai (VI.4). In any event, basic taxes had been monetized since the early ninth century, attenuating further the force of Lopez’s argument. Rather than imposing a tax obligation, the stipulation enjoins the serikarioi from a certain course of action, defining in that way a behavioral pattern the serikarioi ought to follow in their commercial dealings with the Treasury. The provision requires no financial sacrifice of the serikarioi, nor does it generate a revenue for the fiscus. There can be no doubt that these deliveries reflected commercial transactions only and were devoid of the character of a munus. The view has been expressed that the serikarioi sold only garments to vestiopratai, as the latter were forbidden to deal in unsewn silk fabrics (IV.1: ojfei´lousin ejxwnei'sqai shrika` " ejsqh'ta") to avoid competition with the serikarioi. By the same token, the serikarioi were not allowed to sell garments to consumers in direct competition with the vestiopratai, as the silk clothing trade was the latter’s exclusive domain. If they wished to enter the clothing trade, the serikarioi had to give up manufacturing because they were not permitted to practice both concurrently (VIII.6). On the other hand, it has been thought that the serikarioi could sell unsewn silk fabrics freely, not only to foreigners but also retail to natives who made their own clothes at home or had the clothes made for them. This conclusion is based on provision VIII.5, which, in stipulating that silks cannot be sold to foreigners (ejqnikoi'") without the eparch’s knowledge, uses the term “silks in general” (pragmatei´a) rather than clothes (iJma´ tia). The inference is that, since the requirement for the eparch’s permission is confined only to foreigners, the serikarioi were allowed to sell retail silk fabrics to natives. Since the sale of silk clothes to foreigners by
See above, pp. 273–74. See J. F. Haldon, Constantine Porphyrogenitus: Three Treatises on Imperial Military Expeditions (Vienna, 1990), C289–90 for purchases of silks in the open market, and C226 for silks produced to order by the private sector. 120 “Silk Industry,” 8. Muthesius also talks about an “obligation” of the serikarioi to deliver certain silk garments to the imperial storehouse: “Silken Diplomacy,” in Byzantine Diplomacy: Papers from the Twenty-fourth Spring Symposium of Byzantine Studies, ed. J. Shepard and S. Franklin (Aldershot, 1992), 246. 118 119
296
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
vestiopratai was forbidden (IV.1, 4), provision VIII.5 is seen as directed at dealings in silk fabrics.121 Closer examination of the relevant provisions, however, suggests that the serikarioi did not retail unsewn fabrics to either foreigners or natives but that they did wholesale unsewn fabrics not only to foreign and provincial silk traders but to vestiopratai as well.122 The latter in turn retailed them to sojourning merchants. Specifically, provision IV.4 enjoins the vestiopratai to have garments and unsewn fabrics that are to be sold to foreigners (th` n e“qnesin doqh'nai ojfei´lousan pragmatei´an) stamped by the eparch.123 Merchants sojourning in the mitata (mitateuo´ menoi), on the other hand, were not allowed to purchase certain articles (IV.1) or unsewn garments (a“rrafa iJma´ tia) except for their personal use, and these were required to have been cut and tailored in the capital (ejn th' basileuou´ sh sugkoptome´ nhn, IV.8). The conjunction of these provisions leaves no doubt that the vestiopratai also dealt in unsewn silk fabrics, which they had to procure from the serikarioi.124 Use of the term “silks in general” (pragmatei´a) concerning sales to foreigners (VIII.5) implies that the serikarioi were dealing in both fabrics and clothing. Although the provision does not indicate whether such transactions were wholesale or retail, it did not have to. If the vestiopratai retailed unsewn fabrics, as they evidently did, then the serikarioi would not have been allowed to since such direct competition ran counter to the tenet of functional division among the silk industry’s different activities. Therefore, the logical conclusion to be drawn from the passages referred to above is that the serikarioi sold unsewn fabrics and garments wholesale to vestiopratai and silk merchants, foreign and provincial.125 As wholesalers, however, the serikarioi were not allowed to retail unsewn fabrics to natives, as has been mistakenly inferred. Distribution of Silks Domestically Produced Silks. The vestiopratai (merchants of domestically produced silks) routinely obtained their supplies of clothing and unsewn fabrics wholesale from the Simon, “Seidenzu ¨ nfte,” 34–35. Provision VIII.3 only forbids the serikarioi to sell cloaks whose price exceeds ten nomismata. There is no mention of unsewn fabrics. The only requirement (VIII.5) was that the eparch be apprised of all sales to foreigners (ejqnikoi'"). 123 Apparently, the requirement did not apply to Byzantine citizens. 124 See also Muthesius, “Byzantine Silk Industry,” 37. Some tailoring may have been carried on in the shops of the vestiopratai (IV.6). 125 As a rule, there was no limit to the quantity of silk fabrics and clothes that foreign and provincial merchants could buy, as long as their sale was not explicitly forbidden (IV.1, VIII.3, IX.6); the sale was reported to the eparch (IV.2, 3; VIII.5); and the articles were stamped by his staff (IV.4, VIII.9). The purpose of having the silks stamped when produced and sold was not to guarantee product quality or trade mark; it served only to certify that the goods were tradable, i.e., not prohibited by the authorities for domestic consumption or export. Nevertheless, quantitative restrictions were imposed at times by commercial treaties. For instance, according to the 944 treaty, Russian merchants were not allowed to buy silks whose total value exceeded fifty nomismata: The Russian Primary Chronicle: Laurentian Text, trans. and ed. S. H. Cross and O. P. Sherbowitz-Wetzor (Cambridge, 1953), 75; I. Sorlin, “Les traite´s de Byzance avec la Russie au Xe s (II),” Cahiers du monde russe et sovie´tique 2 (1961): 449 and 457–58. Lopez’s view, following earlier authors, that the Russian traders were allowed to purchase large silk fabrics valued up to fifty nomismata (“Silk Industry,” 35 and references in n. 2), apparently derives from a misunderstanding of the relevant provision of the treaty. Muthesius also shares Lopez’s opinion: “Silk, Power and Diplomacy in Byzantium,” in eadem, Studies in Silk Weaving (as in note 6), 239. 121 122
GEORGE C. MANIATIS
297
serikarioi. Occasionally, noblemen and wealthy individuals used the vestiopratai as an outlet for small surpluses of their home-produced goods or for silks acquired as in-kind compensation for serving in the ranks of the imperial administration (IV.2). Purchases of purple and violet garments and garments whose price exceeded ten nomismata, from whatever source, had to be reported to the eparch so that he might know where these articles were to be sold (IV.2, 3). In competition with the prandiopratai, the vestiopratai retailed their wares to residents of the capital, as well as to provincials and foreigners, mostly sojourning merchants not involved in the silk trade. Non-residents were subject to restrictions in their purchases: they were not allowed to buy purple or red silks of large sizes, unsewn garments except for their personal use, or garments valued in excess of ten nomismata (IV.1, 8; VIII.3). On departure, the purchased articles had to be declared to the eparch (IV.8).126 These restrictions did not apply to the residents of the capital, who could buy in the open market silks of any size, quality, or price, and in unlimited quantities.127 Lopez maintains that the commercial activity of the vestiopratai was constricted, as both serikarioi and prandiopratai encroached upon their field despite the prohibition of the edict (V.1, VIII.6, XVIII.5). In his view, the very fact that the serikarioi were enjoined from acquiring raw silk and retailing silks proves that they tried to control all stages of production. Similarly, the prandiopratai infringed upon the domain of the vestiopratai, as the authorities were inclined to ignore their infractions because of the vestiopratai’s ability to strike favorable deals in their wholesale purchases from foreign suppliers.128 In short, Lopez seriously doubts that the eparch enforced the law rigorously and that, as a result, the tendency of dominant guilds to invade the province of weaker guilds remained unchecked. However, as has already been demonstrated, there is no indication that the principle of division of labor among guilds, as reflected in the prohibition of vertical and horizontal integration, was violated with impunity.129 Lopez argues further that the legal restrictions on trade opportunities prevented the vestiopratai from engaging in large-scale trade. As a result, they remained in an “inferior standing.” 130 However, these presumed legal constraints could hardly have thwarted the expansion of an individual firm’s activities within the guild’s domain. Firms at all stages of the production process could increase their market shares by exploiting the growth in demand and by chipping away at the shares of fellow guild members through intra-guild The inspection and confiscation by imperial customs officers of the silks acquired by Liutprand of Cremona, a bishop and diplomat, on his departure from the capital (The Embassy to Constantinople and Other Writings/Liutprand of Cremona, trans. F. A. Wright, ed. J. J. Norwich [London, 1993], 202–4), attest to the state’s readiness to enforce the regulations. But the incident alludes to occasional infractions of the legislation as well. Still, the possibility that the silks Liutprad acquired may have been smuggled out of the imperial silk factory cannot be dismissed: Jacoby, “Silk in Western Byzantium,” 490 and n. 219. If, indeed, this was the case, it would imply that the strict supervision by the eparch’s inspectors effectively precluded the stealthy production of prohibited silks. 127 Lopez, “Silk Industry,” 22. 128 Ibid., 19–20. 129 See above, pp. 273–74, 284–89, 294–95. 130 ´ tat,” 239–43; N. Oikonomide`s, “Le Marchand byzantin des provinces “Silk Industry,” 19. Frances, “L’E (IXe–XIe s.),” in Mercati e mercanti nell’ alto medioevo: L’area euroasiatica e l’area mediterranea (Spoleto, 1993), 658; idem, “Entrepreneurs,” 156–57; Mickwitz, “Organisationsformen,” 76; and Runciman, “Byzantine Trade,” 153, also entertain the view that guild regulations impeded the development of large undertakings. 126
298
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
competition, as the provisions of the law neither secured a member’s market position nor prevented the exit of uncompetitive enterprises. The law did not forbid price or nonprice competition, nor did it limit the number of apprentices and workers a guild member could employ, prohibit trading on borrowed funds, impose quantitative or qualitative restrictions on the equipment used, or regulate production techniques and work processes, all of which it would have to do if the intention was to curb growth of individual enterprises. The policy aim was that increased market demand be shared by as many firms as possible and that this demand be channeled through cascaded, close-knit markets at each stage of the production process. Increased demand would be met by existing firms and/or new entry, as the number of firms that might be established within a guild was not predetermined. What the restrictions did do, as already indicated, was to curb the growth in the size of a firm at the expense of other enterprises in guilds either upstream or downstream—a development that was not the outcome of a genuine market growth and could lead to restraint of trade and monopolization. Besides, while integration would likely yield pecuniary advantages to the consolidated firms in the absence of price controls, it is debatable whether it would also have measurably improved economic efficiency, or at what social costs, at this stage of development in the silk industry.131 It is therefore unrealistic to presume that the status quo impeded firms from exploiting their growth potential, so long as they were enterprising, competitive on price and quality, and forward-looking. From this perspective, the notion that the vestiopratai were in an inferior position seems untenable. Imported Silks. The prandiopratai specialized in the wholesale import and retail sale of an impressive array of silks from Syria, Seleucia, and other regions (V.1, 2). Imported silks had to be stored and traded in one of the mitata (V.2). As in the case of imported raw silk, at the time of the market, all guild members, as well as Syrian merchants who had domiciled in the capital for at least ten years,132 contributed according to their means to a fund for the collective purchase of the imported silks (V.2, 3). This mandated quasimonopsony aimed to secure the lowest possible prices, although it is unlikely that the monopsonistic buying power of the prandiopratai always remained unchallenged.133 Noblemen and other dignitaries were allowed to make purchases directly from the foreign merchants but only to satisfy their personal needs (V.4). The prandiopratai were obligated to buy the entire lot of silks of Syrian origin offered for sale during the market day, irrespective of makeup, quantity, or sizes (V.4). Foreign suppliers had up to three months to complete the sale of their merchandise.134 Syrian silks unsold by then had to be reported to the chief of the guild, who undertook his best effort to dispose of them (V.5). Ostensibly, the chief took it upon himself to find a market for the unsold silks, in the interest of maintaining a constant flow of luxury items for the upper class; but more See pp. 287–88, 320–22. Local Syrian merchants were allowed to trade but not to become members of the guild of prandiopratai: ¨ckle, Byzantinische Zu ¨nfte, 118; Christophilopoulos, jEparciko` n Bibli´on, 82 and n. 4; Zoras, Corporazioni Sto bizantine, 79–80. 133 See note 134 and pp. 305–10, below. 134 The implication is that external suppliers of silks were not obligated to sell their wares on arrival en bloc. This enabled them to get a feel for the market, space out the quantities offered for sale over several market days, and increase their chances of obtaining higher prices. 131 132
GEORGE C. MANIATIS
299
subtle policy aims were to sustain the raw silk supply from Syria135 and to keep the pressure on local producers to ensure high quality of output and competitive prices. The Book of the Eparch does not indicate who represented the guild in the negotiations, or the mode in which the transactions were conducted. Mickwitz believes that negotiation must have been the chief ’s responsibility,136 but this need not be so, especially if the chief lacked commercial experience or bargaining skills. Also, it is not known whether the representative(s) of the prandiopratai negotiated with the suppliers on a one-to-one basis, set the price for the entire consignment, or solicited quotations. The manner in which the transactions were carried out, as well as the extent of the suppliers’ bargaining power, could well influence the price-setting ability of the prandiopratai.137 The imported silks were not standard products but encompassed a range of highly differentiated luxury articles of varied type, value, and marketability. After the purchase, the silks were allocated by the chief in proportion to the respective contributions of the participants (V.3). While dignitaries could pick and choose pre-emptively, the prandiopratai had no such choices. However, it is unlikely that the use of allotment in lieu of the market mechanism satisfied individual buyers’ preferences and commercial interests at all times. Because the formula employed was crude, the system must have posed nagging problems of fairness for the participants.138 Lower cost due to the exercise of monopsonistic buying power for a collection of silks of unknown mix and quality did not necessarily amount to a real gain for the individual buyer, especially since suppliers may have wielded a measure of countervailing power. It is simplistic to infer, as Mickwitz does, that individual buyers were not pressured, and that low cost reflected the monopsony gain and the workability of the system.139 The rough-and-ready allocative formula violated the buyer merchant’s “sovereignty,” in the sense that he could not necessarily obtain the particular assortment of silks he thought would make the best profit on and he may also have been forced to pay prices well exceeding the going market value for silks he unwillingly acquired. Although the prandiopratai were enjoined from practicing the trade of vestiopratai (V.1, XVIII.5), as mentioned earlier, it has been asserted that they encroached on the prerogatives of the vestiopratai in defiance of the law.140 However, this action would have entailed collusion with the serikarioi as the sole suppliers of domestically produced silks, and they were not likely to have been forthcoming. Aside from the risk of detection, the serikarioi had a legitimate outlet for their production through the vestiopratai and silk merchants, foreign and provincial. Involvement of the prandiopratai was not likely to result in a measurable increase in sales, this assuming the serikarioi had difficulty in Muthesius, “Silk, Power and Diplomacy,” 240. Other reasons advanced for this concession are reciproc¨ckle, Byzantinische Zu ¨nfte, 117 ff) and sharing the 10% sales tax with the rulers of Aleppo (Lopez, “Silk ity (Sto Industry,” 31). On the other hand, there is no evidence that the importation of silks was ever regulated or subjected to strict controls in order to stem the outflow of gold from the Empire, as Muthesius speculates: “Byzantine Silk Industry,” 67. 136 Kartellfunktionen, 210. 137 See below, pp. 305–10. 138 A more equitable method of allocation consistent with the buyers’ preferences would have been to auction the silks after the collective purchase. 139 Kartellfunktionen, 210–11. 140 Lopez, “Silk Industry,” 19–20. See also above, p. 297. 135
300
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
selling accumulated stocks through lawful channels. Lopez is unwarrantedly dismissive of the importance of the Book of the Eparch as a policy instrument and of the eparch’s will to enforce the statute, while he overstates the ability of the prandiopratai to remain above the law. Moreover, the rightful reaction to such transgressions by the guild of the vestiopratai with vital interests at stake is conveniently overlooked.141 MARKET STRUCTURE AND MODUS OPERANDI The mandated functional division of the private silk industry’s operations led to market fragmentation, with different structures displaying varied enterprise behavioral patterns and modes of operation. The following commodity markets and their respective subdivisions can be distinguished: 1) the silk raw materials market, comprising the provincial cocoon, the import cocoon/raw silk, and the domestic cocoon/raw silk markets; 2) the two-tier yarn market, reflecting the intermediation of the metaxopratai in the yarn trade, which encompassed the yarn producer-trader and trader-cloth producer stages; and 3) the silk fabrics market, subdivided into wholesale and retail distribution segments. Based on this typology, an attempt is made in the next sections to identify the constituent elements of these diverse market structures, ascertain plausible patterns of market conduct by the different sets of sellers and buyers involved, and outline the most likely pricing policy decisions and performance outcomes. The Silk Raw Materials Market The Provincial Cocoon Markets. Silk was an expensive raw material that had to be shipped from production centers to the capital at a high transport cost and marketed there wholesale in unprocessed or semi-processed form. In addition, market uncertainties, spoilage, and unsafe transport,142 conditions inherent in distant trade, rendered the enterprise risky. Such venturesome commercial undertakings therefore made heavy demands on financial resources, organizational ability, and marketing skills, and the merchants involved in the purchase and shipping of the cocoons had to be enterprising, resourceful, and persons of means.143 Consequently, there were few in each locality, a situation that is suggestive of a relatively high buyer concentration. On the other hand, cocoon producers were, by and large, numerous and small, and at a distinct disadvantage because it was prohibitive, if feasible at all, to carry their produce to the capital and dispose of it themselves, and unprofitable to have it shipped and sold on consignment. Nor could they withhold their produce from the market for a long time, as it was liable to spoilage. Moreover, although the income from cocoons was supplementary, sericulture was an important cash crop for the families concerned and the need for this extra money put pressure on the producers to sell quickly, adding to their vulnerability. All in all, producers had no alternative but to sell hurriedly to the dominant local merchant(s) and on the latter’s terms.144 See above, pp. 273–74. Storms, piracy, and uncharted seas placed too many risks in the way of maritime trade. 143 This lot may have included enterprising landowners as well. 144 Conceivably, they could sell to a larger local producer, but it is unlikely that this would have changed materially their negotiating position. 141 142
GEORGE C. MANIATIS
301
In those rare instances where individual producers dealt with a relatively large number of merchants, for example, in a large town or at a trade fair, they could dispose of their cocoons under more competitive conditions and fetch higher prices.145 More often than not, however, producers had no option but to sell to the local merchant(s) and probably to a merchant to whom they were already in debt. Producers borrowed money or made purchases of consumer goods, agricultural implements, or building materials on credit, pledging their upcoming cocoon crop in repayment, at times at a pre-fixed price. Under these circumstances, and particularly if they counted on the same merchant to obtain new credit, producers had limited, if any, bargaining power and were inclined to accept a price below the going market level. In short, limited storability, limited market outlets, debt repayment obligations, the immediate need for cash, and insufficient knowledge of distant market conditions rendered small and unsophisticated producers vulnerable and enabled the buying merchants to dictate prices. On the other hand, cocoon merchants were in most instances able either to set the price at which they were prepared to buy or to negotiate from a position of strength. They could influence prices because they were few in number and bought in large quantities. By restricting their purchases or threatening to do so, they could force producers to accept a lower price or better bargain.146 Concentrated buying and the attendant market power gave rise to market structures leading to diverse price-setting behaviors.147 In the case of a simple monopsony, in which a single buyer of cocoons is competitively supplied by many small sellers, the monopsonist sets a single market price and buys all the produce offered at that price.148 By exercising his monopsony power, the monopsonist is able to pay the producers a price lower than the competitive price under the same conditions of demand,149 realizing excess profits. In the event that the monopsonist is threatened by potential new entrant(s), he may be led to temper his monopsony power to forestall the new entry. He may then set a price higher than he otherwise would, deter entry, and have the smaller excess profit all to himself. As sellers face buyers’ demand curves, so buyers face the sellers’ supply curves. In an atomistic market structure, in which there are many buyers of a homogeneous product (see note 186) and no one buys a significant proportion of the total output offered, buyers necessarily, because they are unable to exert perceptible influence on price, accept as given the going market price as determined by the aggregate demand and supply. In technical terms, to the small buyer the supply curve is perfectly elastic (horizontal) at the prevailing market price, regardless of the shape of the aggregate supply curve. Below this price he is unable to make any purchases, while it is not in his interest to pay a higher price. 146 In concentrated, i.e., monopolistic or oligopolistic, markets, the supply curve buyers face when they purchase from many small sellers will slope upward, being elastic and possibly inelastic, reflecting the fact that sellers are willing to take various prices the buyers may set, supplying more (less) of the good at increased (decreased) buying prices. If a seller’s supply curve is perfectly inelastic (vertical), as when he has no alternative but to dispose of his produce without delay, the intensity of the buyer’s demand alone will determine the prevailing price. 147 Large size per se, whether measured by the number of workers employed, assets in command, or turnover, does not necessarily imply the ability of a firm to exercise monopoly or monopsony power. For instance, a small firm by any standard that operates uncontested in an isolated area can effectively exercise monopoly or monopsony power. It is the unavailability of close substitutes, impeded new entry, legal concessions to a single firm, control over a strategic input, or effective exercise of concerted action by dominant firms that lead to market control—in short, the absence of competition. 148 By offering a lower price, the monopsonist will not lose his entire source of supply because the sellers’ supply curves slope upward (see note 146). 149 See also below, pp. 308–9 and note 172, and Figure 2 for a diagrammatic presentation of the argument. 145
302
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
A variant of the simple monopsony is perfect monopsonistic price discrimination, in which the monopsonist deals with the sellers one at a time, strikes a potentially different bargain with each one of them, and makes all-or-nothing offers.150 By exploiting each individual producer’s particular circumstances, which are usually common knowledge in small localities, the monopsonist is able to purchase the entire quantity of cocoons being offered at the minimum average supply price. This price would be lower than that established under a simple monopsony and would enable the monopsonist to enhance his excess profits.151 Nevertheless, for practical reasons, perfect discrimination is achievable only when the monopsonist deals with a limited number of sellers. Also, it may not be in the monopsonist’s long-term interest to follow an aggressive pricing policy that could alienate many producers and force them to discontinue production. In such instances, the force of monopsonistic buying power is attenuated, and higher purchasing prices are likely to prevail, albeit still below competitive levels. Another form of concentrated market structure—and one perhaps more likely to pertain to our current study—is oligopsony. It may encompass situations in which very few large buyers are supplied competitively by many small producers or in which a few large buyers coexist with a limited number of smaller buyers. In either case, price-output determination would tend to be dominated by the recognized interdependence and rivalry of the large buyers, whose pricing policies would likely set the policy parameters for the group as a whole. However, since oligopsonists must assume that any price change they make will elicit retaliatory or compensatory price changes by their rivals—with uncertain outcomes—independent pricing or active rivalry may not be enticing. Oligopsonists may then be amenable to settling on some determinate and mutually satisfactory price. It is, however, by no means certain that they will do so, because in an oligopsonistic (oligopolistic) setting two antithetic attitudes tend to vitiate such a decision. Common interest in maximizing the joint monopsony profit would induce larger buyers to act jointly or collusively, a motivation strongest when there are few dominant buyers. Such concurrent or collusive pricing action can be accomplished either by express agreement or by price leadership. In the latter case, through a tacit agreement or unspoken understanding, one of the larger buyers takes the lead, setting a low buying price that the others follow closely.152 Price leadership is more likely to occur in oligopsonies (oligopolies) dealing with a homogeneous product, such as cocoons and raw silk, because market shares shift drastically in response to price changes and affect all members of the group. Where collusive oligopsony is successful, as it could have been in the provinces,153 the impact on the price producers receive is similar to that of a simple monopsony.154 To the extent, however, that rivalry develops within the oligopsonistic camp with respect to the buying price, as when a large buyer bids a price above the monopsony level to secure added supplies at the expense of his competitors, the outcome would tend to shift away For instance, the monopsonist would be in a position to make an all-or-nothing offer when small producers were his debtors or had no other option than to dispose of their produce. 151 See also below, pp. 309–10 and note 173, and Figure 3 for a diagrammatic presentation of the two outcomes. 152 The idea is that the price leader will direct the price to the most advantageous level and that the market will be shared by all at a single price. 153 See note 190. 154 Figure 2 indicates that the price paid to the producers will be OPm. 150
GEORGE C. MANIATIS
303
from the monopsonistic and toward the competitive, and price would be indeterminate within a considerable range. Threat of new entry would have the same effect and reinforce the process. Therefore, inter-buyer rivalry and low or nonexistent entry barriers (i.e., insignificant advantages for established firms over potential entrants) would tend to push buying prices higher than under a simple monopsony, and to the producers’ benefit.155 In view of these conflicting tendencies, it is not possible in an oligopsony to discern a unique pattern of pricing policy. The uncertainty regarding the likely course of monopsonistic buying behavior makes it impossible to ascertain where the balance among these conflicting forces will be struck. There is a distinct possibility, in this instance, that the short trading period narrowed the window of opportunity and imposed a definite limit on the oligopsonists’ posturing, making them less intransigent and more disposed to reach a negotiated price level—more so at times of poor harvests—rather than to follow an uncharted course of action. Such a stance would result in a rigid but determinate price that can fall anywhere between monopsonistic and competitive limits. In the event that in a local market the number of both buyers and sellers was small, as when a few large producers supplying the bulk of the market were not dependent on merchants for the marketing of their produce, an oligopoly cum oligopsony situation would emerge, appropriately referred to as bilateral oligopoly. In this instance, impersonal market forces are totally absent, as both sides command and exercise substantial market power, resulting in an indeterminate market price. The final outcome depends on bargaining strength, maneuvering skills, or competitive price rivalry, and the price could vacillate within wide limits.156 The Import Cocoon/Raw Silk Market: Strengths and Weaknesses of the Competing Groups. Historians have maintained that the metaxopratai acted as a “cartel” in purchasing the silk brought into the capital by external suppliers, the implication being that in this way they were able to influence the terms of exchange to their own advantage and to strike deals at much lower prices than if each guild member negotiated individually.157 To be sure, the mandated collective purchase effectively created a quasi-monopsony, which was reinforced by restrictions limiting the maneuverability of the suppliers: by obligating them to sell in a designated and narrowly defined marketplace (the mitata) and on scheduled market days,158 forcing them to negotiate only with the guild’s representative(s), and limFigure 2 shows that the price will settle between OPm and OPc. The case of bilateral oligopoly can be conveniently subsumed under the case of bilateral monopoly because the results are similar. See the discussion below, pp. 307–8 and note 170, and Figure 1 for a diagrammatic presentation. 157 Mickwitz, Kartellfunktionen, 211; Christophilopoulos, jEparciko` n Bibli´on, 58–59; Lopez, “Silk Industry,” 14; R. S. Lopez and I. W. Raymond, Medieval Trade in the Mediterranean World (New York, 1955), 126; Muthesius, “Byzantine Silk Industry,” 34 and 36. The inference is based on provision VI.8 of the Book of the Eparch, which enjoins the guild members to contribute to a fund to effect collectively the purchase of the imported cocoons and raw silk. 158 It is not known whether cocoon/raw silk transactions in the mitata were taking place on pre-fixed days or if the frequency of market days was determined by the eparch at his discretion and in the light of current circumstances. For practical reasons, it is doubtful that deals were made even weekly. The nature of the trade, the institutional setup, and expediency must have had a bearing on the frequency of market days. Distant points of origin, mode of transport, weather conditions, and the seasonality of shipments were not conducive to fixed scheduling. Collection of advance information about quantities offered for sale and in155 156
304
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
iting their stay in the capital. The eparch’s prerogative to postpone opening the market when it was thin could also work to the advantage of the metaxopratai by augmenting the available supply through increased participation, while a prolonged stay would raise the carrying costs of the suppliers and cut into their profitability by decreasing stock turnover. Still, the bargaining power of the metaxopratai in these wholesale transactions was not unalloyed, as has been presumed. The market position of the external suppliers, as defined by their number, size, and attendant behavioral patterns, would also have to be taken into consideration in assessing the effective monopsony power of the metaxopratai. The high-price, high-volume character of the trade and the risks involved very likely limited the number of suppliers that entered the marketplace each time, suggesting that each cohort of suppliers comprised a core of fairly large merchants and possibly a fringe of smaller ones. The confluence of a small number of sizable suppliers, an array of potent factors affecting the demand/supply relation (e.g., excess demand, supply shortages, absence of substitutes, seasonality of shipments), and the sophistication and bargaining skills of seasoned suppliers could effectively mitigate, if not countervail, the market dominance of the metaxopratai. Furthermore, by timing their entry into the market, staggering the quantities offered for sale over several market days, or withdrawing part or all of their wares when prices were depressed, suppliers could perceptibly enhance their bargaining power. Opportunity to act in concert or re-export their wares could have a similar effect. Finally, import dependence and the legitimate concern that the exercise of inordinate monopsonistic leverage might result in unprofitable sales and disrupt the regular flow of silk imports in the longer run, for example, through a decline in domestic production or a shift of shipments to rival manufacturing centers, could further ease the pressure on suppliers.159 These remarks clearly suggest that the suppliers of raw silk were not inescapably trapped or constantly at the mercy of the metaxopratai and that the latter’s alleged overwhelming monopsonistic buying power could be readily reined in under a special set of circumstances. The effectiveness and assertive power of the contesting groups, as determined by their respective strengths and weaknesses and the manner in which the transactions were conducted in the mitata, defined different market structures, varying modes of operation, and varying levels of competitiveness that led, in turn, to diverse patterns tended to be purchased during the upcoming market day, needed by the metaxopratai so they could make deposits to the fund, invite katartarioi to join if necessary, and allow the latter time to prove their eligibility to the eparch, was time-consuming. That the metaxopratai allowed time for the arrival of additional traders and the accumulation of larger supplies in the hope of lowering the buying price cannot be ruled out. The confluence of these factors suggests that market days were probably fixed at regular intervals, say once a fortnight with a carry-over to the next day to complete unfinished business, but that the eparch, in consideration of particular circumstances, could postpone or interpolate market days. The fairly long time (three months) foreign merchants were allowed to stay in the capital reinforces this view. And so does provision X.2 of the Book of the Eparch, which stipulates that the dealers in perfumes shall buy the imported articles of their trade on the appointed days (kaqw` " a“n hJ hJme´ ra th` n wjnh` n e“ch tou' ei“dou"). The eparch’s staff would notify the guild members through their chiefs and the external suppliers who stayed in the mitata of the set market day. 159 To be sure, forward planning can hardly be expected, let alone be implemented, by buyers acting individually. However, when buyers act collectivelly, it is very likely that long-range issues affecting them as a group will come to the forefront and possibly be acted upon.
GEORGE C. MANIATIS
305
of pricing behavior. Furthermore, elusive and ever-changing market conditions placed a high premium on the contestants’ agility, circumspection, and positioning. It would seem therefore that the price imported silk ultimately fetched could not be fixed unilaterally by the metaxopratai, certainly not with the ubiquity that the notion of a cartel seems to imply, since price concessions extracted on each market day from external suppliers could vary appreciably and sometimes could even be negated. The Import Cocoon/Raw Silk Market: Organization and Management of Transactions. Intriguing questions of operational significance can be raised concerning the manner in which business transactions were organized and conducted. Regarding the metaxopratai, there is no information about who designated the person(s) responsible for making purchases on behalf of the guild: the chiefs160 or the membership? Were the chiefs ex officio the only authorized persons to be actively involved in the negotiations or were they titular bystanders? Did the chiefs share this responsibility with guild members, forming a committee for deliberation and decision-making?161 If so, by what criteria were these persons selected: buying experience,162 bargaining skills, contribution to the fund, rotation? Were all committee members actively involved in deal-making, or was one (or more) spokesman designated? Were the designated representative(s) fully authorized to consummate deals, or did they have to consult with and obtain the consent of the committee? How were the decisions made in case of disagreement among members? For instance, reserve prices (i.e., maximum buying prices above which demand becomes ineffective) may have differed substantially among committee members, and timidity or boldness could have affected appreciably the closing price and, by extension, the interests of the entire guild membership. To what extent was market intelligence regarding the volume of the collective demand for raw silk compiled, so as to enable the delegate(s) of the metaxopratai to exercise more effectively the mandated purchase monopsony?163 Were all suppliers treated alike, regardless of their ethnic origin during the negotiations, or were separate and more favorable deals negotiated with native merchants?164 Did the guild’s representa160 Apparently, because of its large membership, the guild of metaxopratai had more than one chief of high ranking in the Byzantine administrative hierarchy (e“xarco"), although the Book of the Eparch does not specify how many (VI.4). Since the law explicitly indicates that the guild of the prandiopratai was to be overseen by one chief and the guild of metaxopratai by more than one, it is not unlikely that the remaining guilds in the silk industry, on which the Book of the Prefect is silent, also had more than one chief, especially in view of their large membership since guilds in other industries with numerous members did (XVI.3, XVII.4, XIX.1). 161 ´ conomie urbaine, 64. For instance, Macri believes that a delegation was authorized to make the purchases: E 162 The quality of the silk is affected by the color, size, shape, and texture of the cocoons. There is a material difference in quality between cultivated silk, which is a narrow fiber with no markings, and wild silk, which is coarser and thicker with markings. The skill with which the reeling has been done also affects the silk’s quality. Further, silk is hygroscopic and can retain as much as one-third of its weight in moisture without any change in appearance or feel. In addition, cocoons and reeled silk could be soiled or damaged during transport. As a high-value raw material, stifled cocoons and reeled silk had to be sorted, graded, and all the elements affecting quality and weight had to be scrutinized at the time of the purchase by knowledgeable persons to detect blemishes and foil deceptive practices. Probably, the screening and grading were performed prior to negotiations by melathrarioi, who in turn bought the blemished raw silk (see above, pp. 281–82). 163 See above, p. 290 and note 101. 164 The authorities were anxious to ensure a constant flow of raw silk into the capital from all places and at the lowest possible price. The fact that imports of raw silk were tax exempt (VI.5) suggests a rather significant dependence on external supplies and, possibly, an indirect incentive to stimulate domestic pro-
306
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
tive(s) deal on an individual basis with each supplier, or did they negotiate with their representatives? If the latter, were the representatives designated at the request of the authorities, or on the suppliers’ initiative?165 Were quotations solicited from individual suppliers, or was a price set for the entire lot? Was the set price final (“take it or leave it”), or were the suppliers allowed to counter-offer, setting in motion a bargaining process? Evidently, the mindset of the metaxopratai as a group, the approach taken in dealing with these matters, and the suppliers’ reaction affected the decision-making process and the determination of the price paid. Similar questions can be raised with respect to the external suppliers. Could they somehow group together and act in concert as a defensive move against the collectively acting metaxopratai? Obviously, they had legitimate cause to try to neutralize the buyers’ advantage, adequate time to do so, and opportunity, as their stay in the mitata facilitated the exchange of information, collective assessment of market conditions, and development of a business strategy with fellow traders. Circumstances and personal acquaintances, as some traders came from the same region, if not the same town, or had met during previous journeys, could help forge mutual trust and alliances—all conducive to effective concerted action. The chance to develop a common position through cooperative schemes was further enhanced because the number of suppliers was probably not particularly large. They were, moreover, bound by common interest and had some room to maneuver. The fact that the suppliers were allowed to remain in the capital up to three months suggests that they were not obligated to dispose of their wares immediately on arrival and could instead enter the market at their own pace.166 Possibly an encumbrance, largely because within this time span merchants had also to complete deals for return cargo, the ruling nonetheless allowed a degree of flexibility in timing their sales and conduct of their affairs. Also, suppliers apparently had the option, as a last resort, to reexport unsold raw silk to other manufacturing centers—certainly not a preferred alternative to either party because of adverse economic repercussions.167 In sum, different duction, albeit with an uncertain impact on producer prices due to the considerable market power of the intermediating merchants, who were not compelled to pass on the savings from the waived tax to producers. Price discrimination against foreign suppliers based on ethnicity, although technically feasible, since sellers could be segregated in the mitata and dealt with separately, could have proven to be myopic and counterproductive in the long run, as shipments could be diverted to other manufacturing centers in the East, such as Seleucia, Antioch, or Alexandria, resulting in shortages and price increases. It would therefore appear that, as long as the silk industry relied on foreign supplies and the threat of diminished shipments to the capital was real, ethnic discrimination and differential pricing could not be practiced with impunity. See also notes 159 and 167. 165 This would expedite the negotiating process by reducing the number of one-to-one contacts. However, the hemming in of the suppliers could potentially enhance their bargaining power by facilitating concerted action, e.g., price leadership. 166 It is unlikely that provision XX.3 of the Book of the Eparch, which stipulates that merchants hoarding imported goods in times of scarcity shall be severely punished and have their goods confiscated, was applicable to external suppliers of raw silk. Rather, the provision was meant to apply to local merchants handling imported necessities and medicines that affected the public at large (XIII.4, X.2), not to dealers of a raw material intended for the production of luxuries whose scarcity had no real impact on the masses. Besides, hoarding implies that supplies are hidden, whereas the quantities of raw silk stored in the mitata were known to the authorities and their storage temporary. 167 In the case of the Syrian external suppliers, the Book of the Eparch makes it clear that, if part of the imported silks remained unsold at the end of the three months, the chief of the guild of prandiopratai was to make his best effort to see to their disposal (V.5). In the absence of an explicit provision, we cannot
GEORGE C. MANIATIS
307
factors could produce downward adjustments in the quantities offered on the market day and thereby enhance the suppliers’ bargaining power. Concerted action could take the form of an express or tacit collusive agreement and require the participation of a limited number of relatively large suppliers who controlled a substantial proportion of the raw silk offered for sale. In an overt or sub rosa agreement, a price floor would be set below which suppliers would refuse to sell. Under a tacit agreement, the unified group of suppliers would recognize and follow a price leader who would take the initiative to set the initial price and whose subsequent price changes would be matched by the other sellers. In theory, suppliers resorting to collusive behavior could have been indicted for violating the general provision of the Book of the Eparch (XX.3), whereby persons conducting business in a manner detrimental to the public interest were to be punished and have their wares confiscated, or the existing antimonopoly legislation.168 It is debatable, however, whether such overt or covert actions could be detected, proven, or effectively litigated, even if these legal strictures applied, which is doubtful. Ironically, prosecutorial action could raise the issue of fairness in commercial circles, with possible adverse implications for trade, since concerted action by the metaxopratai was officially mandated and sanctioned, while counteractive actions by aggrieved suppliers were impeachable. These observations clearly suggest that, as a result of changing market conditions, varying degrees of supplier concentration, and supplier bargaining power, the monopsonistic pricing power of the metaxopratai, though present, was not unbridled. Its final impact cannot be determined without an analysis of surrounding circumstances. As monopsonistic attitude entails different patterns of pricing behavior in response to market dynamics and seller reaction, likely price outcomes and their viability are explored below by considering three plausible market structures: bilateral monopoly, simple monopsony, and monopsonistic price discrimination. The Import Cocoon/Raw Silk Market: Market Structures and Behavioral Patterns. Effectively, whenever the external suppliers of raw silk succeeded in banding together and forming conclude that, by analogy, the same held for imports of cocoons and raw silk as well, because they do not lend themselves to the same handling as silk fabrics, and, in addition, such a concession would discriminate against domestic producers and other foreign suppliers. About the reasons for this concession see above, pp. 298–99 and note 135. Goitein reports that there is no indication from the Geniza records in Cairo that raw silk was traded outside the frontiers of the empire: A Mediterranean Society, 1:103. A. Guillou, on the other hand, maintains that part of the Calabrian raw silk production was exported to Cairo: “Production and Profits in the Byzantine Province of Italy (Tenth to Eleventh Centuries): An Expanding Society,” DOP 28 (1974): 94. In fact, there is no evidence of legal restrictions on raw silk being shipped outside the empire or of stipulations mandating that the only port of destination for domestic producers was the capital. For instance, Byzantine Calabria exported cocoons and raw silk to Arab Sicily: A. Guillou, “La Soie sicilienne au Xe–Xle s.,” in Bizantino-Sicula, 2: Miscellanea di Scritti in Memoria di Giuseppe Rossi Taibbi (Palermo, 1975), 287. Exportation of raw silk from the capital was explicitly prohibited only after it had been purchased by the metaxopratai for the reasons mentioned above, p. 269. Evidently, the authorities were convinced that policing would be ineffective, and only market forces could determine the direction of trade in raw silk. If shipments of raw silk converged on the capital, the primary reason was that the demand was strong and traders were confident that they would not be squeezed and that they would obtain a fair price. This suggests that the metaxopratai took pains to avoid aggressive monopsonistic pricing that would force suppliers to exit the market or divert shipments. 168 See note 190.
308
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
a quasi-monopoly to confront the quasi-monopsony of unified buyers, a market structure referred to as bilateral monopoly would emerge. In this situation, the interests of buyers and sellers are diametrically opposed and in sharp conflict. The monopsonistic group of metaxopratai would have wanted to pay the lowest possible average price, or just enough to cover the suppliers’ average cost per unit,169 keep them in business, and ensure their return to the market. The monopolistic group of suppliers, for their part, would have wanted to charge the highest possible average price for the units sold, or just enough to prevent the buyers from leaving the market. To achieve their respective objectives, both groups would have attempted to drive an all-or-nothing bargain for the marketed quantity of the raw silk at a specified average price. However, since both groups exercised control over price, the market price mechanism was not operative and the price at which this quantity changed hands was indeterminate within a wide range. Bargaining power and negotiating skills determined the outcome. Buyer dominance, seller dominance, or balanced power were all possible, and the price could therefore fall within the extreme limit of either group, whichever was dominant, or fall uncertainly between the two extremes. If neither group dominated, there was a good chance that the price gravitated toward competitive levels.170 A market structure referred to as a simple monopsony arose when the metaxopratai, acting as a single buyer, were supplied competitively by many sellers. Assuming that the Average cost includes the minimal profit necessary to induce the seller to continue in operation. In Figure 1, the unified monopsonistic buyers have a tentative demand curve (dd⬘b) in the import market, because they can set a buying price and not offer to buy various amounts at different prices as happens in conventional demand curves. Only if they had no control over price would this be a genuine demand curve. Each price shown on dd⬘b simply represents the net value to the monopsonistic group of an increment to their purchases of raw silk at the corresponding quantity. Since the monopsonistic group resells the purchased raw silk, dd⬘b also represents their marginal revenue curve (MRb) derived from the demand curve DD⬘ they face in the resale market. However, the unified monopsonists’ demand curve dd⬘b (MRb) is, in effect, the unified monopolistic sellers’ marginal revenue curve (MRs), reflecting the additions to their revenue from incremental sales. Similarly the monopsonistic buyers’ average outlay (AOb) and marginal outlay (MOb) curves are also the monopolistic sellers’ average cost (ACs) and marginal cost (MCs), respectively. It is evident that both groups will wish to drive a perfectly discriminatory, i.e., an all-or-nothing, bargain. To do this, the monopsonistic buyers will purchase that quantity OQ at which their marginal revenue curve intersects their marginal outlay curve (MRb ⫽ MOb). They will wish to pay OP for the last unit, but lower prices for all preceding units they buy, and an average price OPb for all units. By reselling this quantity at OPs (as found on the demand curve DD⬘ for their sales), they will reap a combined profit (selling price minus average cost), as shown by the rectangle abPbPs. The monopolistic sellers will sell that quantity at which their marginal cost equals their marginal revenue (MCs ⫽ MRs), which is the same quantity OQ desired by the monopsonistic buyers. They will wish to sell the last unit at OP but all preceding units at higher prices, receiving an average selling price OPs for the quantity OQ, appropriating the total available revenue of the buyers to themselves. So, when the raw silk is resold at OPs, the monopsonistic sellers will reap the total profit abPbPs. This sets the stage for bargaining. Both contesting groups have an interest in the same quantity of raw silk, which is such as to maximize their joint profits. But they disagree over the average price at which the raw silk will pass from sellers to buyers. This average price cannot go above OPs or below OPb but is indeterminate between these limits that define the bargaining range. Depending on their relative bargaining strength, the price may fall anywhere between OPb and OPs. For an in-depth analysis of the theoretical underpinning of the participants’ pricing behavior under bilateral monopoly, see J. S. Bain, Pricing, Distribution and Employment (New York, 1953), 394–96, and 432–36; E. Schneider, Pricing and Equilibrium (London, 1962), 299–313; W. J. L. Ryan, Price Theory (London, 1964), 353–62; C. E. Ferguson, Microeconomic Theory (Homewood, Ill., 1969), 281–82; E. Mansfield, Microeconomics: Theory and Applications (New York, 1970), 270–72. 169 170
1 Bilateral monopoly
2 Simple monopsony
3 Perfect monopsonistic price discrimination
4 Monopolistic competition, short run
5 Monopolistic competition, long run
GEORGE C. MANIATIS
309
monopsonistic buyers could not discriminate in buying price,171 they would have set a single market price depending on their room for maneuver and accepted all the raw silk offered for sale at that price. The effect of this action would have been that the monopsonistic buyers paid the sellers a lower price than competitive buyers under the same conditions of demand.172 In the absence of a prevailing countervening power or under conditions of excess supply, the chances that a monopsonistic pricing policy would be exercised effectively were enhanced. Yet, the force of the exercisable monopsonistic power could be tempered measurably if such action militated against the collective selfinterest of the metaxopratai, as it would have in the face of concern that it would lead to a disruption of the regular flow of imports. A paucity of suppliers or excess demand, factors that tend to bolster the bargaining power of sellers, also could have a mitigating effect on the monopsonists’ buying pricing power. Perfect monopsonistic price discrimination occurred when the metaxopratai, acting together as a single monopsonist, dealt with each seller (or a group of sellers) individually and made an all-or-nothing offer. Unlike the case of simple monopsony where a unit price was set at which each seller could supply as much as he wished, perfect price discrimination specified the quantity to be purchased from each seller (or set of sellers) and the total outlay. The resulting price tended to be the minimum each seller would be willing to accept rather than leave the market. The quasi-monopsonistic buyers were then able to buy raw silk at a supply price lower than that established under simple monopsony. Perfect price discrimination therefore entails maximum exploitation of each seller, as the monopsonistic buyers, by acting in unison, are in a position to dictate the price they want, capitalizing on individual supply conditions and arrogating a take-itor-leave-it posture.173 The extent to which perfect price discrimination actually occurs, That is, buying some units at a low price and further units at a higher price without raising the price of the previous units. 172 In Figure 2, dd⬘b is the unified monopsonistic buyers’ tentative demand curve as well as their marginal revenue curve (MRb) from the resale of the raw silk, as discussed in note 170. SS⬘ is the supply curve of raw silk, or the summed marginal cost curves of the sellers (MCs). In the absence of control over price, this is a conventional supply curve in atomistic competition. The marginal outlay curve (MOb), drawn to the supply curve, shows the rate of increase of the monopsonistic buyers’ total outlay with increases in purchases. Upward supply curves, as is reasonable to expect, denote that by increasing their purchases the monopsonistic buyers raise their total outlay steeply because they must pay a higher price for all intra-marginal units. Thus, the marginal outlay to them is greater than the supply price that represents their average cost or outlay (AOb). To maximize their profit on raw silk operations, the monopsonistic buyers will purchase such a quantity that their marginal outlay (MOb) just equals their marginal revenue (dd⬘b ⫽ MRb). This will be the quantity OQm, for which the monopsonistic buyers will pay the uniform price OPm per unit. If dd⬘b was a genuine demand curve, i.e., the aggregate demand curve of many small buyers or, alternatively, if the monopsonistic buyers did not exercise their monopsony power, the intersection of SS⬘ and dd⬘b would determine the competitive price (OPc) and quantity (OQc). As can be seen, the price the monopsonistic buyers pay is lower than the competitive price. For more detail, see Bain, Pricing, 382–88; J. Robinson, The Economics of Imperfect Competition (London, 1954), 218–23. 173 A comparison of the outcomes under simple monopsony and perfect monopsonistic price discrimination is presented in Figure 3. The curve dd⬘b is the unified monopsonistic buyers’ marginal revenue curve (MRb); SS⬘ is the supply curve the monopsonistic buyers face, reflecting their average outlay (AOb) and the sellers’ marginal cost (MCs); MOb is the monopsonistic buyers’ marginal outlay curve; and ACs is the sellers’ average cost, or the average price at which various quantities can be obtained under perfect discrimination. Under simple monopsony, with market pricing, i.e., setting a market price for all units and accepting what is offered at that price, the monopsonistic buyers would equate their marginal revenue (dd⬘b ⫽ MRb) to their 171
310
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
however, depends on the number of separate sources into which the total supply can be divided and on the supply conditions of each source. Thus, when monopsonistic buyers face too many sellers (whether individuals or groups), establishment of the degree of direct contact with each seller necessary to assess his individual circumstances may be quite impracticable. Further, it is questionable whether in the normal course of events market conditions would have always been favorable to the monopsonistic buyers, or, even if such conditions were propitious, whether it would have been in the long-term interest of the metaxopratai to take an extreme position and practice ruthless price discrimination. Finally, it is important that no transfers took place between sellers174 and that the monopsonistic buyers had to estimate the minimum prices sellers would be willing to accept, no easy task, particularly if the sellers had room to maneuver, for example, by withdrawing temporarily from the market. It would seem, then, that perfect monopsonistic price discrimination was achievable only under rather circumscribed conditions, namely, if the monopsonistic buyers dealt with a limited number of sellers who were unable to act in concert and if the buyers were able to ascertain and take advantage of the suppliers’ waning ability to wait for better prices, as, for example, when they approached the end of their sojourn in the capital. The Cocoon Import/Raw Silk Market: The Domestic Sector. The metaxopratai sold the imported cocoons and raw silk to katartarioi, who held the processing monopoly, just as the former had a monopoly on import and domestic sales. Nonetheless, the view is advanced that the metaxopratai overpowered the katartarioi, allegedly because the latter were an impoverished and helpless lot, and because the interposition of the metaxopratai between katartarioi and serikarioi in the sale of yarn considerably enhanced their market position. According to Lopez, the metaxopratai in effect “controlled” the entire guild of katartarioi, and raw silk processing was rendered an “ill-rewarded” craft. His conclusion is based on the fact that the katartarioi could join in the collective purchase of the imported raw silk only if invited to do so by the metaxopratai, because they could buy only as much as they could process, and because they could not sell yarn directly to serikarioi but only to metaxopratai. In fact, this fiat monopoly in yarn production was illusory, as it remained unenforceable.175 Earlier, Mickwitz stressed the close dependence of the katartarioi on the metaxopratai, on grounds that the latter provided the only outlet for the sale of silk yarn and, in consequence, the katartarioi would have had no alternative marginal outlay (MOb), purchase the quantity OQ1, and pay a uniform price OP1 per unit. Under perfect price discrimination, SS⬘ becomes the monopsonistic buyers’ marginal outlay (SS⬘ ⫽ MOb) and ACs their average outlay (ACs ⫽ AOb). This is so because the price of inframarginal units is not altered by purchasing added units and this makes the marginal outlay curve coincide with the supply curve, whereas the average outlay curve lies correspondingly below the supply curve. The monopsonistic buyers will equate their marginal revenue (dd⬘b ⫽ MRb) to their marginal outlay (MOb ⫽ SS⬘), purchase the quantity OQ2, and pay an average price OP2, which is lower than OP1. For more detail, see Bain, Pricing, 426–29; Robinson, Imperfect Competition, 225–27. 174 The Book of the Eparch does not explicitly prohibit inter-supplier trade before the market day. However, since such deals could lead to concentration of the imported raw silk in the hands of a few suppliers and thereby enhance their bargaining power, a concerned legatarios probably gave oral instructions that trading among external suppliers was prohibited. Covert transfers could be detected because of the narrow confines of the mitata and because merchants had their wares inspected on arrival. 175 Lopez, “Silk Industry,” 16 and 18–19. On the fallacy of forward and backward integration into spinning by metaxopratai and serikarioi, see above, pp. 271–74, 284–89.
GEORGE C. MANIATIS
311
but to accept the price offered by the metaxopratai. He further maintained that the katartarioi relied on the metaxopratai for suppliers’ credit and that the metaxopratai held back those poor katartarioi to whom they farmed out the processing of raw silk by paying them in advance and by squeezing their piece-rate wages.176 The notion that the katartarioi held a weak market position and were economically oppressed is persistent and generally accepted.177 However, analysis of the institutional setting, market fundamentals, business calculus, and the nature of competition that resulted, as presented in this and the next section, indicates that it is unlikely that the metaxopratai were in a position to exercise controlling power over the katartarioi and that the impression of the katartarioi as disempowered and prostrate at the feet of the metaxopratai is unwarranted. As already suggested, a guild’s sales monopoly established by fiat does not necessarily translate into the ability of individual guild members to exercise monopoly pricing power in the marketplace, as such faculty rested on an array of preconditions.178 This notion is predicated on the assumption that the metaxopratai formed a monolithic block of likeminded businessmen who acted in a coordinate fashion, since allegedly one of the purposes for instituting the guild system and the provisions of the Book of the Eparch was to limit price competition among members of the same guild.179 In contrast to practice in the West, the Byzantine guild system did not aim to raise or maintain prices through concerted action, since such conduct ran counter to the anti-monopoly tenor of the law. The law did not impose price discipline on guild members to thwart intra-guild competition and protect individual members’ shares in the total business. This is evidenced by the facts that neither prices nor profit margins were fixed at the manufacturing or distribution stages of the production process, while the guilds had no exercisable influence over the imperial authorities,180 that no external or internal regulations ensured equality “Organisationsformen,” 72–73; idem, “Un Proble`me,” 26. On the nature of the skills and the type of work performed by the workmen hired by the metaxopratai, see above, pp. 271–72, 275–76. 177 See note 97. 178 See above, pp. 269–70. 179 Lopez, “Silk Industry,” 18. 180 On the lack of political power by the silk guilds, see note 26 and pp. 273–74, 330. That price competition was fostered and price determination left up to the parties to the exchange is also evidenced from the following stipulations in the Basilics: ejpi` th'" pra´ sew" kai` th'" ajgorasi´a" e“xesti kata` fu´ sin ajllh´ lou" ejn tv' timh´ mati perigra´ fein tou` " sunalla´ ssonta" (X.4.16[4]); fusikw'" ejn th' pra´ sei kai` th' ajgorasi´a, kai` th' misqw´ sei kai` ejkmisqw´ sei, perigra´ fomen ajllh´ lou" ejn tv' posv' (XX.1.22[3]). The scholium reinforces the argument (ibid.): ejpi` th'" pra´ sew" kai` th'" ajgorasi´a" fusikw'" sugkecw´ rhtai to` plei´ono" a“xion h”ttono" ajgora´ zein, kai` to` h”ttono" plei´ono" pipra´ skein (goods of higher market value may be purchased at a lower price, while goods of lower market value may be sold at a higher price, if market conditions so dictate). Lopez’s own statement that “in the rising oligarchy of the masters there was no room for the unsuccessful craftsman” (“Silk Industry,” 16) clearly implies the existence of intra-guild competition. Even in the case of basic staples handled by guilds, such as bread (XVIII.1), fish (XVII.1), meat (XV.2), wine (XIX.1), or groceries (XIII.5), the authorities fixed maximum profit margins, not maximum retail prices. This is quite a progressive pricing policy, since profit capping, though intrusive, is much less disruptive of the functioning of the market mechanism and of the price formation process than outright price fixing. The reason is that the wholesale price structure of the consumer goods affected—the bedrock for retail price formation—is allowed to reflect the prevailing demand and supply conditions as well as prospective price movements. Besides, price fixing would have been counterproductive, as it would have led to shortages and black markets. Also, setting profit margins did not preclude competition, as some vendors might have accepted lower margins in order to increase their volume of sales, aiming to maximize their total, instead of unit, profits. At the same time, price competition was reinforced and consumers’ interests were safeguarded through anti-monopoly legislation (see note 190), while the Book of the Eparch addressed specific instances of unconscionable bargains resulting in excessive 176
312
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
in the scale of individual enterprise operations, that there were no restrictions on the quantity or quality of inputs that could be used in manufacturing, that new entry into the industry was not legally impeded,181 and, finally, that the exit of inefficient firms under normal competitive conditions was not hindered. On the other hand, incipient or potential competition provided a powerful deterrent to price-fixing agreements. Nor, as has been argued, were the regulations of the Book of the Eparch designed to protect guild members from the competition of unorganized craftsmen and noble owners profits and criminalized such business conduct. Thus, businessmen were forbidden to hoard imported commodities at times of scarcity in order to raise prices and exact a profit exceeding what is “fair” (mh` ajpoqhsauri´zein de` tau´ thn [pragmatei´an] eij" kairo` n ejndei´a" pro` " para´ logon ke´ rdo", mh´ te mh` n poluolkei'n ta` " tima` " pe´ ra tou' de´ onto"): X.2, XX.3. In certain transactions—e.g., contracts related to the building trade, and provided that the work involved was originally miscalculated by the contractor, the agreement was subsequently modified by the employer, the extent of the undertaking proved to be uncertain (Book of the Eparch, XXII.3; Hexabiblos, III.8.42), or the sale of real property prompted by financial need or ignorance (Basilics, XIX.10.66; Peira, XXXVIII.5 and 12; Hexabiblos, III.3.69, 71, 72)—if the price paid was less than half of the “just price” (dikai´a timh´ ), the contractor or seller (or their heirs) could request (within four years) either the difference between the amount paid and the fair price or the invalidation of the transaction on grounds of excessive injury (uJpe´ rogko" bla´ bh, laesio enormis). In sales of assets the seller had to reimburse the buyer, whereas in the building trades the value of the work already performed was reassessed. Nonetheless, in some cases sales could still be effected at prices less than the just price, as in, e.g., disposal of an unprofitable asset or opportune sale made to avoid an anticipated decline in the price of the asset; but in these instances the seller could not seek restitution (du´ nasqai ejpi` sumfw´ noi" tisi` pwlei'n e”kaston kai´ h”ttono" tou' dikai´ou timh´ mato"; Ecloga, XVI.30; Hexabiblos, III.3.73 and scholium). Provision XIX.10.66 of the Basilics was applicable only to landed property, and certainly not to ordinary commercial transactions, e.g., in dry or perishable goods, since such notions as ignorance of the much higher value of the commodity sold, negative (even if excessive) profit, or need are insubstantial in business deals and contrary to trade practices. Also, the four-year statute of limitations for recovery is hardly in the interest of commerce, which is concerned with the speedy resolution of disputes. The Byzantines did not express the concept of just price in precise economic terms. However, from the economy of the law it appears that in both commercial and non-commercial transactions they had identified just price with the going market price established under competitive conditions. On the origin, socioeconomic foundation, economic definition, scope, operationalization, and enforcement of the just price, see the author’s “Operationalization of the Concept of Just Price in the Byzantine Legal, Economic and Political System” (forthcoming). The notion of “just price” (or versions thereof) had permeated the agrarian and interest-rate legislation in efforts to curb the injustices inflicted upon weaker individuals. See ODB, s.v. “Just Price”; Kazhdan and Constable, People and Power in Byzantium, 44–45; A. A. Vasiliev, History of the Byzantine Empire (Madison, 1958), 1:346–49; Ostrogorsky, History of the Byzantine State, 272–76 and 280–82; P. Lemerle, “Esquisse pour une histoire agraire de Byzance,” Revue historique 219 (1958): 254–84; Cassimatis, Les Inte´reˆts, 49–66 and 112–27. 181 Although conditioned on certain qualifications (integrity) and requirements (payment of a small fee), legal entry into the silk manufacturing and trade guilds was unrestricted, and the decision on accepting new entrants was not made by guild members or even their chiefs but by the eparch (IV.5; VI.6, 7; VII.3; VIII.13; XII.2). Cf. Christophilopoulos, jEparciko` n Bibli´on, 37–38 and 51. At times, new entry was probably challenged by existing members, but it is unlikely that they could sway the eparch’s decision, as he was keenly interested in increasing productive capacity and in fostering intra-guild competition. Entry was probably denied only when it would result in chronic excess capacity and oversupply leading to cutthroat price cuts, subnormal earnings to enterprises and labor, and high rates of business mortality—i.e., to prevent excessive or destructive competition. Such a situation would arise in an industry of atomistic market structure where privately organized collusive price-output determination is effectively unavailable as a counter-measure against general industry distress. On the other hand, as argued above (p. 280), the entrance fee levied by the guilds (IV.5, VI.6, VIII.13) was nominal and could hardly have been intended to forestall new entry to deter competition between existing and new members, as Christophilopoulos has asserted: jEparciko` n Bibli´on, 54.
GEORGE C. MANIATIS
313
of workshops.182 As already discussed, the reasons the law discouraged the development of silk manufacturing activities outside the controllable organizational structure of the guild had nothing to do with fending off potential competition.183 The silk industry in the capital was expanding, and there was no need to protect member market shares from outside competition through restrictive measures. Moreover, a competitive mindset was fostered by relatively unconcentrated market structures, low barriers to new entry, unimpeded imports, and the institutional setup. Contrary to what has been asserted,184 guild members could ill afford to become lax or complacent; they were not immune from competition. The degree of market concentration was an important factor in determining the different players’ attitudes and pricing behavior. The evidence points to the existence of a large number of metaxopratai who were running enterprises of varying sizes and to unrestricted opportunities for augmenting their ranks.185 But even in such an unrealistic situation as a highly concentrated oligopoly, it is unlikely that the metaxopratai’s alleged dominance would obtain. In such a market structure, a few large metaxopratai would ostensibly control the entire market (or at least the bulk of the marketed raw silk, sharing the remainder with a limited number of smaller sellers, all supplying a homogeneous product to atomistically organized raw silk processing establishments).186 Since each large oligopolist controlled a significant fraction of the marketed output and was in a position to influence the market price by output adjustments, each seller would have determined his price and output in light of the actions of rival firms, which would have given rise to price uncertainty. The final outcome would have depended on whether inter-seller rivalry was active, which would have allowed prices to slide toward competitive levels, or if the rivalry ended in a compromise at some mutually satisfactory price level, permitting a measure of excess profits.187 Persistent excess profits would have invited new entry, since there were no legal barriers to new competition and economic barriers were low, confined essentially to capital availability.188 In the long run, therefore, prices and profits were unlikely to remain substantially above atomistically competitive levels. 182 ´ tat,” 239–41; Lopez, “Silk InKazhdan and Constable, People and Power in Byzantium, 31; Frances, “L’E dustry,” 15–16; Mickwitz, Kartellfunktionen, 228–31 and 234. 183 See above, pp. 269, 279–80, 297–98. 184 Oikonomide`s, “Le Marchand byzantin,” 658; idem, “Entrepreneurs,” 157 and 160; Kazhdan and Constable, People and Power in Byzantium, 48. 185 See notes 160 and 218. 186 Product homogeneity means that the product of any one seller is viewed by buyers as identical to the product of any other. This means that, for the buyer, any firm’s product is a perfect substitute for the product of any other firm in that market; in consequence, he is indifferent as to which firm he buys from. Raw silk and yarn are such commodities. In an atomistic market setting, product homogeneity implies the inability of sellers to raise prices without loss of market share. In an oligopolistic setting, none of the larger sellers can independently cut prices without triggering a chain of retaliation, nor can any such firm independently raise prices and hope to maintain its sales volume unless it can induce its rivals to increase their prices as well. 187 See the discussion on oligopsony above, pp. 302–3, which is the logical counterpart of oligopoly. 188 If they encountered a conflict of interest, wished to remain inconspicuous, or were unwilling to be involved in person in day-to-day operations, venturesome wealthy individuals and noblemen could enter the silk industry at any stage of the production cycle by using their slaves as surrogates. Such involvement has been criticized as an infiltration of lucrative businesses or as a stealthy, forcible, and exploitative action, deriving from the exercise of political or economic power: Jacoby, “Silk in Western Byzantium,” 477; Frances,
314
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
A more plausible market setting would be one of low oligopolistic concentration, involving quite a few large metaxopratai in competition with many small sellers who supplied a substantial proportion of the market. As each seller faced an elastic demand, raising the possibility that he could lose all his sales to others if he raised his price perceptibly (or, alternately, that he could appropriate his rivals’ customers if he lowered his price and his move met with no response), the price-output pattern likely to emerge was one in which the price was close to the minimal average cost with low or no excess profits. Low entry barriers would reinforce this outcome. In general, in oligopolies (oligopsonies) with homogeneous products facing unconcentrated buyers (sellers), the price is likely to be lower (higher) the larger (smaller) the number of firms in the group, until in the end there are enough firms for atomistic competition to emerge.189 A market structure equally, if not more, likely to prevail would be one defined by a fairly large number of metaxopratai (and katartarioi), encompassing establishments of quite different sizes, in which no one firm controlled a substantial proportion of the marketed output. Large membership, product homogeneity, and low entry barriers would have weakened monopolistic tendencies and, in consequence, the metaxopratai would have had little measurable influence over the price they charged, at least in the medium and long term, as such conditions were conducive to atomistic competition. Market dominance could also be achieved if the members of the guild of the metaxopratai could enter into formal or covert agreements that covered every transaction, setting uniform prices or sales quotas and driving all-or-nothing deals. To be effective, however, collusive agreements had to have enforceable schemes to restrict sales, maintain prices, and discipline fractious members, and have had the consent or acquiescence of the overseeing chiefs (and, by extension, of higher authorities). It would, however, have been extremely difficult to effect and sustain such schemes, even if initially successful, because they are not durable. Collusive agreements tend to be fragile and break down even in highly concentrated markets, because antagonistic tendencies among rival mem´ tat,” 239–40; Muthesius, “Byzantine Silk Industry,” 34–37. Yet, such entry, far from being surreptitious, “L’E suspect, and inimical, should rather be viewed as a conscious effort on the part of the authorities to provide a vent to latent entrepreneurship and tap a source of capital for the expansion of the silk industry. New entry would promote competition through the growth of guild membership, expand production for domestic consumption and exports, and increase the chances of employment and better pay for the capital’s workforce—policy objectives yielding significant economic and political dividends. At the same time, this open invitation encouraged potential entrants who did not share the prevailing elite bias against less socially respected economic endeavors to participate, albeit under the state’s terms and under the eparch’s watchful eye. It is noteworthy in this context that the privilege of the nobility to purchase imported silks directly (V.4) and the prohibition against using surrogates to acquire raw silk (VI.10, VII.1) did not intend to exclude them from participating in the silk trade or manufacturing, as Vryonis has hypothesized (“Byzantine Dhmokrati´a,” 298). The rationale for these stipulations is discussed in pp. 269–71. 189 Realistically, a certain level of intermittent excess (or subnormal) profits is unavoidable in a free enterprise system. Atomistic or pure competition is an ideal normative standard, and its conditions are unlikely to be fulfilled in a dynamic market environment. Indeed, it is highly improbable that at any moment all firms in an industry will be earning normal profits. A distinction therefore should be made between persistent excess profits arising from the exercise of sheer monopoly power and those arising periodically because of unsatisfied demand, since prices rise to restore the equilibrium between supply and demand. In the latter instance, price increases reflect shortages and signal the need to expand capacity. As supply increases, excess profits are eliminated. Persistent subnormal profits, on the other hand, signal the need to contract capacity. As supply decreases, profits are restored to normal levels. Hence, reference to atomistic competition literally implies the prevalence of conditions approximating atomistic market structures.
GEORGE C. MANIATIS
315
bers lead them to deviate from perfectly collusive behavior. Typically, some metaxopratai would have soon found that different selling prices were more profitable since cost structures and market shares varied among firms. Unequal financial strength among guild members would have tended to undermine the cohesiveness of the group: weaker members would have had a lower threshold of compliance and a stronger incentive to cheat, particularly during depressed business conditions. High prices encouraged veiled price cuts in the form of special concessions, discounts, or rebates, prompted defections, and invited new entry. For the same reasons, such informal devices as price leadership were unlikely to be sustained, especially in a multi-player setting. Finally, uncertainties surrounding explicit or tacit cooperative actions, in particular, the difficulties in keeping such agreements from collapsing, increased with the number of establishments involved, to the point that such arrangements became unmanageable. In general, internal and external pressures in contestable markets tend to undermine the effectiveness of collusive arrangements. Another potent deterrent of collusive business behavior has been conveniently overlooked. Express or tacit agreements to restrain competition and fix prices were forbidden and punishable by law. Conspiratory actions would have prompted authorities to crack down ipso jure or at the urging of the aggrieved katartarioi.190 It would have been difficult to put into practice, let alone enforce, unfair methods of competition under the watchful eye of state-appointed chiefs who served as overseers of guild members’ conduct. And, had official protestations or legal action from the katartarioi failed owing to venality, some retaliatory show of force would have to be expected, possibly leading to disruptions of silk production and trade. This would have agitated the serikarioi and vestiopratai as injured third parties, created social unrest, and triggered government action to protect the interests of the state and downstream guilds. At this point a question can be raised about the extent to which `a fac¸on arrangements were conducive to exploitation of the participating katartarioi. The answer is very little, if at all, because such commercial transactions were conducted between principals, that is, independent businessmen, and the agreements were voluntary and mutually beneficial. No party could be coerced into joining. Nevertheless, as commonly happens in the 190 Monopoly, i.e., dominant market control by one firm or by several acting in concert, and monopolization, i.e., market conduct involving actions undertaken to secure and maintain a monopolistic market position by effecting the weakening, elimination, or exclusion of competitors (e.g., through such predatory or exclusionary tactics as monopolization of inputs, predatory price cutting, entry-forestalling pricing, vertical or horizontal agreements, or conspiracies) were forbidden and punishable by law, as were express or tacit agreements to fix prices. The penalties inflicted ranged from stiff fines to confiscation of property and permanent exile. Tribunals were also liable to penalties if, because of venality, discrimination, or other failings, they did not enforce the law: CI, IV.59; Basilics, XIX.18.1. The preamble of the Book of the Eparch, in setting forth the animus of the statutory provisions (ratio legis), emphatically states that the parties to commercial transactions should not brazenly take advantage of one another, the financially stronger should not do injustice to the weaker, and no person oppress his fellow man, but that all actions should be governed by the rule of law: Nicole, Livre du Pre´fet, 13. In essence, the Book of the Eparch not only reflected the official industrial and commercial policy, but was also an extension of the pertinent legal provisions that sought to restrain monopoly and promote competition. To be sure, anti-monopoly legislation is difficult to implement because it is hard to detect and prove such actions, especially when illicit agreements are informal. Political influence could also attenuate the force of statutory provisions, particularly in the provinces. Nevertheless, the situation is quite different under a guild organizational structure, where patterns of illegal conduct and flagrant transgressions of the law could hardly elude the attention of the vigilant chiefs.
316
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
marketplace, strong competition among katartarioi to enter into such contracts could have decreased their profit margin and resulted in subnormal profits. The opposite could have occurred, however, when for example strong demand for such arrangements from the metaxopratai enabled the katartarioi to reap excess profits. To summarize, the nature of emerging market structures, the inauspicious prospects of inter-seller coordination, and the unobstructed possibilities for new entry strongly suggest that the metaxopratai wielded no appreciable monopoly power of a lasting nature. The prevailing market setting and institutional arrangements unleashed robust competitive forces and shaped individual behavior conducive to establishing conditions that approximated atomistic competition, although, in light of ever-changing market fundamentals, short-lived aberrations resulting in excess profits cannot be ruled out. The fact remains that the Byzantines relied on competitive forces to a much greater extent than we have been led to believe. Strained parallels regarding the purpose and tactics of the Byzantine guilds drawn from much later experience in the West are clearly unwarranted. The Two-Tier Yarn Market The Backdrop of the Katartarioi-Metaxopratai Rivalry. In the first stage of the two-tier yarn trade, the katartarioi supplied the metaxopratai with yarn under market conditions substantially approximating atomistic competition, as discussed above. It is not a foregone conclusion, however, that all metaxopratai were involved in the yarn trade. Possibly, some metaxopratai were unwilling to enter the market or were unable to compete in the second-tier resale market, since some firms could obtain discounts on cash or highvolume purchases or acquire yarn at a lower cost because of `a fac¸on arrangements. If this had occurred on a large scale, the yarn trade would have been concentrated in fewer hands. Even if the katartarioi had faced a somewhat concentrated group of metaxopratai, who provided the only outlet for their products, this would not necessarily have meant that the metaxopratai were in a position to dictate at will the buying price or subject the katartarioi to odious exploitation. Market fundamentals and legal restrictions exercised a powerful disarming effect on their behavior, as has been demonstrated in the preceding section. Even if the metaxopratai could somehow seize an opportunity to fix prices collusively and exercise monopsony power with impunity, concerted action is unlikely to have been enduring, as it would have been extremely difficult to coerce the compliance of all members of the group. Moreover, the fragility of collusive agreements and the unimpeded entry of new players would have heightened inter-seller rivalry, as competing metaxopratai tended to raise prices offered for yarn in order to secure needed supplies, thus driving prices up toward atomistically competitive levels. More importantly, it would have been contrary to the best interests of the metaxopratai to price aggressively and drive the weaker katartarioi out of the market. Business failures would have resulted in reduction of raw silk processing capacity, concentration of market power in the hands of fewer but more entrenched katartarioi, and their ability not only to resist downward pressures on prices but even to raise prices. The fact is that the metaxopratai relied on the katartarioi for uninterrupted delivery of high-quality yarn, and fulfillment of this function required not only that the katartarioi stay in busi-
GEORGE C. MANIATIS
317
ness but also that they be able to expand their productive capacity to meet the growing demand, something that was all the more critical since the metaxopratai were prohibited from involvement in yarn production. These economic imperatives underscore the advantages of a “live and let live” approach over an aggressive pricing policy, advantages the metaxopratai fully recognized. To be sure, such a business conduct may have entailed a significant attitudinal shift, but it was motivated by sheer economic calculus, not compassion. Failure rates were probably higher among smaller raw silk processing enterprises, although it was probably not very different from that experienced in other guilds, since small firms have greater vulnerability to economic vicissitudes, limited capacity to service debt, and vary widely in their ability to compete. The authorities were fully aware that some business failures must be expected in a competitive system and that the exit of anemic firms should not be prevented through ad hoc measures of dubious efficacy (e.g., price controls or restriction of new entry). Eventually, existing firms and new entrants would fill the gap left by the defunct enterprises, at a pace dictated by market conditions and entrepreneurial capacity. Operating on this premise, the guild system did not stifle competition among members or interfere with the exit of uncompetitive firms; it sought only to provide equality of opportunity for all members so they could compete fairly. It should be appreciated that yarn production was a vital part of the silk manufacturing process and that the katartarioi continued to discharge this function without interruption or depletion of their ranks as evidenced by the continued growth of silk manufacturing over the years and sustained levels of silk production and trade. Although ebbs and flows were inevitable, the secular trend of demand for silks remained strong and apparently rose,191 naturally increasing the demand for the services of the katartarioi and strengthening their market position. Operating within such a dynamic economic environment, as a group the katartarioi could hardly have been forced into abject poverty or become defenseless objects of exploitation in the hands of the metaxopratai. The Setting of Metaxopratai-Serikarioi Competition. The serikarioi were not permitted to be involved in yarn production and were obligated to buy yarn only from metaxopratai, as by law the latter retained the monopoly in the yarn resale market. This meant that the serikarioi had no access to raw silk and could not enter into `a fac¸on arrangements with katartarioi to process raw silk. As in the case of the metaxopratai, membership in the guild of serikarioi was quite extensive, the range of enterprise sizes broad, legal entry into the guild unobstructed, and the barriers to new entry apparently low or moderate.192 This market morphology formed the basis and shaped the conduct of the members in these two guilds as they traded in the second-tier yarn market. Emergence of an oligopoly with moderate seller concentration, involving quite a few large metaxopratai and a large number of small firms facing competitive buyers or, alternatively, an oligopsony with a moderate buyer concentration, involving quite a few large serikarioi and a substantial number of small enterprises supplied by competitive sellers, was not outside the realm of possibility. However, in either instance the ability of the See below, pp. 326–28. On the conditions of entry at the manufacturing stage of the silk industry, see below, pp. 322–23 and note 181. 191 192
318
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
established sellers or buyers to take advantage of their respective market position would have been substantially restrained. As has already been pointed out, the fact that the guilds of metaxopratai and serikarioi held fiat monopolies in their respective fields certainly did not enable individual guild members to exercise monopoly or monopsony pricing at their discretion. Recognized interdependence regarding price-output policies, threat of new entry, product homogeneity, independent pricing, and unworkable collusion schemes were likely to induce established firms to price at a point lower than the group’s profit maximizing level—indeed with a good chance that such a price might even come fairly close to competitive levels. A market structure more likely to emerge, though, was one in which a core of quite a few sellers and buyers had significant presence in the yarn market, while the remainder, representing a substantial proportion of the marketed output, was held by a large number of small firms—a bilateral oligopoly with a large competitive fringe. In such low oligopolistic cum oligopsonistic concentrations, price rivalry and imperfect collusion are more likely. In particular, the strong presence of a large competitive fringe of small firms, predisposed to act like competitors in an atomistic market and not easily policed or coerced into concurrence on a price set collusively by the large firms, would tend to drive the price toward the competitive level. For, while each small firm acts on the legitimate assumption that its own moderate price changes will induce no reaction from rivals since they will not perceptibly affect other sellers’ sales or buyers’ purchases, the combined effect of such tactics by many small firms tends to erode the market shares of the large firms and enhances the probability of independent pricing by them. This tendency will be stronger the greater the number of large firms and the smaller the proportion of the market they control, because the force of recognized interdependence tends to be attenuated, and as it weakens independent pricing is encouraged since the effects on rivals become less definite. An equally likely market structure would be one approximating atomistic competition. In such a setting, both sellers and buyers are so many in number and so small in size that no one of them has appreciable control over price, while concerted action by both groups would not be sustainable even if temporarily achievable. As buyers, the serikarioi would have been able to shop around and negotiate on an individual basis, exploiting comparative cost advantages among sellers, while as sellers the metaxopratai would have been able to select their clientele only as long as they took steps to remain competitive. Product homogeneity would have reinforced the ease with which the serikarioi could change suppliers, forcing potentially aggressive metaxopratai eager to retain the patronage of the serikarioi and maintain market share to price competitively. Moreover, the institutional setup, that is, anti-monopoly legislation and the guild system, afforded both groups a potent defense mechanism against monopolistic/monopsonistic tendencies, in that it enabled the aggrieved party to frustrate non-competitive behavior initiated by the other. The fact that the serikarioi were not allowed to produce in-house yarn to meet their requirements but depended exclusively on the metaxopratai for supplies may sustain lingering assumptions about the latter’s potential to sway prices. However, aside from the restraining effect of market fundamentals and institutional arrangements already mentioned, an important element the metaxopratai had to factor into their economic
GEORGE C. MANIATIS
319
calculus was that the demand for yarn derived from, and depended on, the demand for silks—the primary source of demand.193 Since yarn accounted for a significant proportion of the final product’s cost and no close substitute for silk yarn existed, a substantial price increase could measurably weaken the demand for yarn and force the serikarioi to curtail production, unless offset by an increase in productivity or a reduction in profit margins, to both of which actions there were limits.194 In the face of a fairly elastic demand for many final products, an attempt to pass cost increases on to the final consumers could result in a decline in the sales of silks,195 production cuts, and a loss of business to metaxopratai. The upshot is that the metaxopratai could not remain indifferent to the potential impact of their pricing policies on the salability of the end products. The Silk Fabrics Market Mode of Production, Plant Scale, and Output Features. Although production of the large highly prized purple silks associated with imperial authority was forbidden, the private sector was nevertheless allowed to produce silk fabrics of fine quality and in large variety.196 Apparently, cloaks and garments were the mainstay of silk production.197 These silks differed considerably in terms of weave, type, design, color, size, cut, weight, and fiber-mix (e.g., silk weft, cotton or linen warp), resulting in a high degree of product differentiation198 and considerable variation in price. Probably, there was a degree of product specialization among the serikarioi, in part dictated by capacity limitations. Focus on selected types of silks would also have simplified the organization of production and resulted in greater efficiency. In a handloom mode of production, the manufacture of low-volume, high-quality, imaginatively designed, and expensive articles ensured maximum profitability. They also had the additional advantage of being easily exported. Sale of low-quality, low-priced, low-markup silks could be profitable only if they could be standardized and produced in large volume. However, demand for mass-produced silks was virtually nonexistent because end users desired fine silks for personal display.199 Moreover, in an industry charac193 The elasticity of the derived demand for silk yarn depended on the elasticity of the primary demand for silks, the proportion of the unit cost of silks spent on yarn, the degree of substitutability of other raw materials for silk yarn, and the response of the price of silks to changes in the costs of production. 194 See below, pp. 319–22. 195 See below, pp. 327–28 and notes 223, 226. 196 Silks that the serikarioi were allowed to manufacture are listed in provisions IV.3 and VIII.1, 2, whereas those forbidden (kekwlume´ na) are found in provisions IV.1 and VIII.1, 2, 4. 197 Silk fabrics also included patterned brocades, scarves, tapestry and upholstery, quilts, shrouds, bed spreads, pillow cases, cushions, horse trappings, and the like: Jacoby, “Silk in Western Byzantium,” 473 and references in nn. 114 and 115. Also included were church furnishings, hangings, vestments, and icon, Bible and altar covers: A. Muthesius, “The Hidden Jewish Element in Byzantium’s Silk Industry: A Catalyst for the Impact of Byzantine Silks on the Latin Church before 1200 AD,” in Studies in Silk (as above, note 6), 247. Macri erroneously thought that the restrictions were so pervasive that commercially produced silks lacked ´ conomie urbaine, 57. variety and were dull and inelegant: E 198 Products are differentiated if buyers distinguish clearly the product of one seller from that of another. The distinction may be real or imagined, so long as it is of importance to buyers and leads to a preference of one variety of the product over another. Product differentiation, which can range from strong to weak, may also be based on such intangible factors as a seller’s way of doing business or his reputation for fair dealing. 199 See below, pp. 327–28 and note 226.
320
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
terized by wide product variation, skill intensiveness, painstaking labor, and emphasis on quality, the workshop form of industrial organization, as opposed to a putting out or a mass production factory system, fulfilled important desiderata: development of exceptional craftmanship, effective control of the work process (the entrepreneur usually being a hands-on worker as well), work discipline, and production of highly individualized items.200 Finally, small plant scale, short production runs, and product diversification afforded flexibility and enabled producers to make adjustments as needed in the face of changing market conditions. To be sure, the division of labor among guilds and attendant market fragmentation prevented the emergence of large and potentially more efficient economic units through vertical integration.201 In theory, units of larger size are more likely to reap the advantages of economies of scale arising from team production, better techniques, and higher specialization, and benefit from more effective coordination of sourcing, production, and marketing activities, better balancing of output rates in successive processes, fuller utilization of equipment and skills, and reduction of intermediate inventories. Integration would also eliminate transaction costs in moving goods from one stage to the next by avoiding the need to monitor inputs for quality, preventing post-contractual opportunism (e.g., sizing, fraud, delays in delivery),202 and eliminating the costs of enforcing unfulfilled or partially fulfilled contracts. In practice, however, it is debatable whether appreciable real economies could be attained in the face of the strong emphasis on product differentiation and given the stage of the industry’s technological development. Judging from the high quality of the silks produced, the margins for increasing specialization and improving production techniques, work methods, shop-floor organization, and overall efficiency do not seem to have been particularly significant. Apparently the degree of specialization in raw silk processing, dyeing, and weaving was already fairly advanced, and these advantages were likely reinforced by product specialization among firms. More importantly, there was no grass-roots demand, and, thus, no market pressure on the industry, to reorient production toward standardized, mass-produced silks. Product standardization was not a lucrative proposition because of strong cultural barriers, since silks were objects of distinction of a class-conscious clientele who expressed strong preference for variety and high quality. Hence, the industry had no economic incentive to seek novel production methods that would have promoted technological breakthroughs entailing large-scale production units and a fully integrated enterprise organization. Also, given the skill-intensive nature of production, capital deepening (i.e., increases in the quantity of capital per unit of labor) and plant scale were not essential to efficiency, as the level of technological development did not provide opportunities for further economies in production. The equipment was simple, the scale of the optimum technical unit in each production stage was small, and production runs were short. See above, p. 289. Lopez has been critical of the guild regulatory system as having an inimical effect on enterprise size, insinuating a missed opportunity for the silk industry to advance to a capitalistic stage: “Silk Industry,” 18–20. Others have expressed similar views as well; see note 130. 202 Opportunism has been aptly defined as “the ability of one party to an exchange to benefit at the expense of the other party by violating the agreement in his or her post-contractual behavior”: D. C. North, Structure and Change in Economic History (New York, 1981), 36. 200 201
GEORGE C. MANIATIS
321
Finally, a heavily labor-intensive mode of production, constant returns to scale, and the ebbs and flows of demand favored small and flexible units. Under these circumstances, dramatic gains from vertical integration were not very likely. Replacement of market transactions by process integration, that is, internalization within the firm, was not critical to ensure quality control either. The general pattern of exchange in the domestic trade was personalized,203 owing to individual contacts, repetitive dealings, and a common set of values. The notion of property rights on the objects of exchange was deeply ingrained, an impersonal judicial system existed to settle differences, and procedures were in place to enforce contracts. In addition, opportunism in the organized and localized silk market of the capital was further constrained by competition engendered by the large numbers of players, as sellers were forced to meet contract specifications and contractual obligations for fear of losing business to competitors.204 Further, although exchange in external trade was impersonal, it had a distinct local character, as transactions were always consummated in the capital. Again, competition constrained the behavior of the parties to the exchange, while the promulgation of a code of conduct by the state ensured the enforceability of property rights in a court of law, safe passage to foreign traders, and certainty in the handling of their affairs—a milieu that provided the requisite underpinnings for widening of the market and the overall growth of trade. In this scheme of things, transaction costs were reduced substantially, since the uncertainty and distrust inherent in market transactions were minimized and prices could be acceptably determined. Therefore any supposed advantages of setting up a long hierarchy within a firm, effectively substituting factor markets for output markets, were negated. In any case, process internalization is not necessarily less costly than input purchases on the market, as elaborate in-house monitoring devices must be set up in large firms to effect coordination of activities, oversee worker performance and prevent shirking, and secure quality control. Significantly, by the tenth century, the growing size of the market had brought about important changes in the economic organization of the private silk industry, involving a shift away from the undivided performance of consecutive tasks typical of home and handicraft production, toward an organizational form with central workshops of a respectable size and characterized by extensive separation of tasks and coordinated teamwork. This transition ensured not only that productivity gains from team production and specialization but also quality control could be achieved at a comparatively low cost, because firms involved in small batch production had a short hierarchy. As processes were not complex, the firms relied primarily upon worker self-discipline, and management was very close to production work. This simplified considerably the task of 203 On the distinction between personalized and impersonal exchanges, see North, Structure and Change, 40, 182, 204. 204 Personalized exchange cum third party (i.e., state) enforcement in the Byzantine domestic trade adds a fourth dimension to North’s triadic construct of historical patterns of economic exchange: D. C. North, Institutions, Institutional Change and Economic Performance (Cambridge, 1990), 34–35. North distinguishes the following evolutionary patterns in the West: personalized exchange, involving local trade and characterized by the absence of third party enforcement; impersonal exchange, involving long-distance and cross-cultural trade, in which the state’s role as protector and enforcer of property rights was ambiguous at best, since the state’s engagement led to greater insecurity and higher transaction costs; and impersonal exchange with effective third-party enforcement characteristic of successful modern economies.
322
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
supervision and assessment of the agents’ performance in principal-agent (i.e., employerworker) relations. In sum, two important advantages of vertically integrated firms, namely reduction in production costs due to scale economies and avoidance of transaction costs by retaining ownership of materials throughout the manufacturing process, eluded the small factory firm, which was the organizational form prevalent in Constantinople’s silk industry. Further, the industry’s capacity for incremental productivity improvements—achieved by solving practical production problems, improved techniques, taking advantage of widening markets through new investment, or adaptation of organization and work methods— was not impeded under the guild system, as enterprise growth, new entry, competition, and profit motive were not suppressed. Moreover, there is no certainty that a vertically integrated firm will operate at rated capacities in all stages of production in which they are engaged or at all times, as is necessary for the avowed benefits of integration to accrue. Unstable demand, balancing problems, irregular input inflows, the temptation of firms operating in concentrated markets to restrict output and raise prices to gain monopoly profits, and other imponderables are all factors unconducive to full capacity utilization, negating the benefits of integration, raising real production costs, and increasing social costs. Nature of the Barriers to New Competition. Established firms tend to have an advantage over potential new entrants, and this could measurably affect the degree of market competition. Economic barriers to new entry in silk fabric manufacturing were likely to include superior techniques possessed by existing firms (e.g., dyeing), possibly maintained by secrecy; lack of access by entrant firms to capital, skilled labor, or managerial capacity; scarcity of suitable workshops; and insufficient entrepreneurial talent. Prima facie, the required fixed investment in sheds, weaving equipment, and dyeing facilities affected prospects for the new entrant, raising barriers to new competition and enabling entrenched serikarioi to forestall new entry and reap excess profits. Nevertheless by opening a small plant, confining start-up production to one sub-process (e.g., only weaving), and renting industrial premises, initial capital requirements could be reduced substantially, while the contracting out of dyeing might not appreciably raise production costs. Shortage of industrial space would tend to raise rentals but, in the absence of rent controls, this would spur construction, increase supply, and alleviate the pressure on rents.205 Product variation could make it possible for potential entrants to enter the field at a minimum economical plant scale and secure a niche in the market. Expectations of high returns in the face of growing market demand could attract local and provincial entrepreneurs, including seasoned operatives striking out on their own with financial backing
The law (Basilics, XX.1.22[3] cited in note 180) imposed no ceilings on rentals. Leases were freely negotiable, and rentals could be renegotiated at agreed upon intervals or at the expiration of the lease. Nonetheless, the Book of the Eparch criminalized the acquisition of rental commercial properties (see note 96), mostly in prime locations and short supply, by deceitfully bidding up the rent. Apparently, the authorities were concerned that such a practice could create an unsettling business environment by unwarrantedly strengthening the bargaining power of landlords and forcing aggrieved tenants to accept excessive rent increases or face eviction. Clearly, the law did not prohibit rental adjustments that reflected changes in market conditions; rather, the aim was to frustrate opportunistic behavior. 205
GEORGE C. MANIATIS
323
from wealthy individuals.206 Finally, availability of a pool of qualified workmen and onthe-job training of local and immigrant workers could augment labor supply. All in all, the barriers to new entrants in silk manufacturing were probably low or moderate—at best, exerting a catalytic effect on potentially high market concentrations. The Wholesale Distribution Market. The serikarioi disposed of their output wholesale in a diffused and competitive market composed of numerous small- to medium-size vestiopratai and a flock of sojourning merchants in the capital, no one of whom could perceptibly influence his buying price.207 As each producer strived to differentiate his products, the silk goods that reached the marketplace ranged from slightly differentiated to very distinctive and, perhaps, at times even “unique.” This afforded individual producers a degree of “monopoly” power, the extent of which depended on the ability of rival producers to market substitutes,208 the degree and pattern of seller concentration (number and size distribution of firms), and the condition of entry. The relative importance of each of these constituent elements set the tone for the structure, conduct, and performance likely to emerge in the wholesale distribution market. A market structure in which a large fringe of small firms co-existed with a lowconcentration oligopolistic core facing competitive buyers was not impossible.209 In such a differentiated oligopoly with a large competitive fringe, the force of recognized interdependence of price policies among large serikarioi, which is characteristic of concentrated oligopolies, tended to be diluted, as moderate variations in price and sales did not perceptively affect the others. Active rivalry and independent pricing were encouraged, as cost structures and market shares differed among firms. Also, the possibility of imperfect collusion increased, as collusive arrangements became far more difficult to conclude and police, and defections and price cuts were encouraged since they would not have attracted retaliation. This pattern of market behavior becomes more pronounced as the For instance, a partnership could be formed in which the investing partner would supply the capital and, possibly, the place of work, while the managing partner would contribute his entrepreneurial and technical skills under an agreed-upon profit-sharing arrangement. For the legal foundation of partnerships (koinwni´a, eJtairi´a) see Basilics, XII.1.1, XVIII.1.13(2), XIX.8.76, XXIII.3.1, L.5.9; Hexabiblos, III.10.1–30, 11.1. Possession of technical skills was not a precondition for setting up shop in silk manufacturing, apparently because expertise could be obtained through hiring or partnership. The basic qualifying criteria of admission, which held for all silk guilds, were integrity (uJpo´ lhyi"), capability (a“xio"), and financial means (eu“poro"): VIII.3, IV.5, VI.6. 207 See below, pp. 323–26. 208 Products are close substitutes if a change in the price of any of them noticeably affects the quantity of the others purchased at given prices. This implies that the seller of a close-substitute product will be able to attract a significant portion of buyers of the other products by reducing his price because the products are competitive (and vice versa). The opposite holds when products are distant substitutes. See also note 213. 209 It is doubtful that oligopolies of even moderate concentration, let alone highly concentrated markets, could have emerged in the silk manufacturing industry in the absence of: economies of large-scale production, in which a relatively small number of firms may be able to supply the entire market at a lower cost than most firms can attain; dominant advantage of product differentiation, such that buyers prefer the products of a few firms to other varieties of the same good to the degree that these firms are able to secure a major share of the market against all competitors; merger moves among sellers to eliminate competition among themselves and bring about a sufficiently concentrated market structure that competition among the remaining sellers could be easily suppressed; and ability of a few firms to secure absolute cost advantages permitting them to operate profitably at a price at which the others could not survive (e.g., control of strategic raw material supplies, legal or institutional prerogatives). 206
324
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
number of firms in the competitive fringe and the market share they supply increase, and is reinforced by actual or threatened new competition due to low or moderate entry barriers.210 The confluence of these factors is likely to drive prices close to atomistically competitive levels and appreciably reduce excess profits. A more likely market structure, however, was one in which many sellers with varying productive capacities, producing close substitutes but not identical products, and no one of whom controlled a significant proportion of the total output supplied a motley of foreign and provincial wholesale traders and the multi-buyer guild of vestiopratai. Such a setting is suggestive of a silk manufacturing and distribution industry operating under conditions of monopolistic competition, in which competitive and monopolistic traits were combined.211 In this instance, small individual market shares and the virtual absence of conscious interdependence of the sellers’ price policies mean that each seller, facing a demand curve that tends to be quite elastic,212 will pursue an independent course, raising or lowering his price just enough to reduce or expand his sales without eliciting a rivalrous reaction from any other seller in the group. A monopolistic competition setting, however, could also include sellers with a substantial degree of discretion in setting prices. By making their product(s) distant substitutes, for example, through distinctive designs or color combinations, they could render the demand for their product(s) much less elastic,213 thereby ensuring a considerable degree of discretionary control over price, effective insulation from price competition, and an opportunity to realize and sustain excess profits. It is noteworthy in this context that, since the sellers’ demand curves could have slopes ranging from significantly negative (fairly inelastic) to approximately horizontal (very elastic) as a result of the varying degree of product differentiation, a range of prices and profit margins might have arisen, reflecting the buyers’ differing valuations of the marketed products. Strong product differentiation could ensure persistent excess profits even in the presOligopolistic serikarioi enjoying cost advantages vis-a`-vis potential entrants could either forego price increases and forestall entry or raise prices, accept new entrants, and hope that their profit margins would improve despite smaller market shares. 211 Monopolistic competition is distinguished from both monopoly and atomistic competition. In atomistic competition, the products of all sellers are identical and therefore perfect substitutes. The demand curve each seller faces is perfectly elastic, or nearly so, signifying that he cannot perceptibly influence the market price of the product. The monopolist’s product, on the other hand, has no close substitutes and his demand curve is inelastic, suggesting substantial control over price. In monopolistic competition, the individual seller has a product that no other seller duplicates, and in this sense he is a monopolist. But because there are close substitutes for his product, the demand for the monopolistic competitor’s product will be more sensitive to a relatively small range of price changes than that of the monopolist, i.e., his demand curve will be more elastic, reflecting less discretionary influence on price. 212 The more closely competitive substitutes there are, the more elastic the demand for the product of any one firm in the group will be. 213 The elasticity of the individual seller’s demand is a function of the degree of product differentiation and the number of sellers in the group. The strength and implications of the close or distant substitute relation can be captured by the notion of cross-elasticity of demand, which provides an index of the readiness with which buyers substitute one product for another. Thus, a distinction can be made between 1) a high cross-elasticity of demand, in which a small proportionate price reduction for any one product reduces the sales volume of every other product by a relatively large proportion, attracting buyers to the cheapest product, suggestive that the products are close substitutes; and 2) a low cross-elasticity of demand, as when a small proportionate price reduction by any one reduces the sales volume of every other product by a small proportional amount, suggestive that the products are distant substitutes. 210
GEORGE C. MANIATIS
325
ence of new entry and import competition. However, to the extent that production was centered on close substitutes, barriers to entry were relatively low,214 and import competition was unrestricted, monopolistic competition tended to yield results not substantially different from those prevailing under atomistic competition since, in the longer run, excess profits tend to remain small or disappear.215 Moreover, in this setup the serikarioi would have had great difficulty in making any concerted action given the large numbers, differentiated products, and unrecognized interdependence—a set of potent counteracting forces. Besides, any attempt at monopolization through collusion, even if practicable, would have been hard to enforce because of antagonistic tendencies among guild members, the attractiveness of defection from such agreements, likely retaliation from the guild of vestiopratai anxious to fend off aggressive pricing tactics of the serikarioi, and the almost certain intervention of the state to curb such conspiratory conduct. In contrast to later experience in the West, production and sales cartels were unknown in Byzantium. The Retail Distribution Market. The retail market in silks comprised the vestiopratai and the prandiopratai. The vestiopratai retailed unsewn fabrics and garments to residents of the capital and sojourning traders not involved in the silk trade. The vestiopratai faced intense competition from imported silks distributed by the prandiopratai, which were close substitutes in terms of quality, pattern, color, and fiber-mix. These silks very likely were competitively priced, as the prandiopratai acted collectively in the purchase of the imported silks.216 This practice, reinforced by the availability of locally produced silks, facilitated monopsonistic buying tactics and possibly resulted in lower import prices, thereby strengthening the competitive position of the prandiopratai vis-a`-vis the vestiopratai, since both catered to the same clientele.217 On the other hand, unrestricted imports of silks not only enhanced consumer choice but also forced the serikarioi to Relatively easy entry almost always accompanies markets populated by a large number of sellers. In Figure 4, the short-run demand curve SRDD⬘ of the monopolistic competitor lies at a level determined by the prices of all competing products, which are in turn determined by the total group demand and supply relation. The seller will independently try to maximize profits by choosing that price (OP) and output (OQ) at which marginal cost (SRMC) equals marginal revenue (SRMR), as drawn from the demand curve. If the demand curve slopes noticeably, in the short run, where the number of sellers in the group is fixed, the individual seller could reap an excess profit as measured by the rectangle Pabc. But in the long run, a confluence of factors, such as simultaneous group price-output adjustments to expand sales, new entry (as long as net revenue is positive), shifts in buying habits, and the appearance of more close substitutes would make the individual seller’s demand curve LRDD⬘ more elastic and shift it downward to the point that it becomes tangential to the average cost curve (LRAC). As a result, the seller realizes sales OQ at a price OP and earns only normal profits because the price is lowered, as shown in Figure 5. In reality, however, some sellers will ordinarily enjoy more advantageous demand to cost relations than others, and potential entrants may be at some disadvantage relative to established firms. At any moment, therefore, some sellers will probably be earning normal profits and others earning excess profits. On the theory of monopolistic competition see E. H. Chamberlin, Theory of Monopolistic Competition (Cambridge, 1958), 71–100; Bain, Pricing, 65–70 and 350–67; Ryan, Price Theory, 297–304; Ferguson, Microeconomic Theory, 285–301. 216 See above, p. 298. 217 The forcefulness of the monopsonistic power of the prandiopratai, however, may have been mitigated at times by the countervailing bargaining power external suppliers of silks may have been able to wield and concern about the repercussions of an aggressive pricing policy on the flow of imports. The dynamics that shaped the behavior of the metaxopratai and the external suppliers in the raw silk trade may apply equally in this case, mutatis mutandis. 214 215
326
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
produce high-quality silks at competitive prices, and the ripple effect prompted productivity improvements and cost cuts further upstream. This cascaded effect is reflective of the intensity of the competitive forces that permeated the silk industry. The organizational structure, institutional setting, and the attendant behavioral pattern suggest that the retail market in silks most likely operated under conditions of monopolistic competition, with results approximating those of atomistic competition. The large number of establishments of varying size,218 unimpeded entry, import competition, and inability to devise and maintain overt or covert collusive arrangements point to an absence of appreciable seller concentration and consequent monopoly power to influence prices significantly. Even so, where product differentiation was strong and buyers exhibited a marked status consciousness, sellers could influence prices and reap excess profits. In general, the retail market probably functioned much like the wholesale, and the discussion of it is applicable here as well, mutatis mutandis. THE DEMAND FOR SILKS: AN OVERVIEW In the absence of statistical evidence, any assessment of the demand for silks must be aggregative, impressionistic, and tentative. Basically, the private silk industry was outward-looking, albeit not aggressively so, and had a firm foothold in the domestic market. Strong external demand for these much coveted luxuries is attested by the active trade with the Italian cities, Islamic states, the Russians and the Bulgarians; export restrictions on particular types of silks; treaty restrictions on exportable quantities of silks; and strict measures to curb smuggling to foreign countries.219 In fact, claims have been made that production could not satisfy growing external demand.220 Similarly, the fact The Book of the Eparch does not mention the title of the chief of the guild of vestiopratai. In another source, however, he is referred to as an exarch (e“xarco"): Christophilopoulos, “Zhth´ mata tina` ejk tou' jEparcikou' Bibli´ou ,” Hellenika 11 (1939): 133–34. The appointment of high-ranking exarchs in the guilds of metaxopratai (see note 160), vestiopratai, and prandiopratai (V.1), and probably also in the guilds of katartarioi and serikarioi, where the law is silent, primarily signifies their large membership and the higher (middle class) social status of their members as economically better off compared to members of the other guilds, rather than the perceived importance or reputation of their profession, as Nicole, Livre du Pre´fet, 88, and Christophilopoulos, jEparciko` n bibli´on, 47–48, 49 n. 1, assert. Since there is no practical way to establish unequivocally each guild’s contribution to the public weal, there is no logical basis for asserting that because, say, the bakers’ chief was a lower-ranking official (prosta´ th"), bakers’ socioeconomic contribution was less important than that of the raw silk traders simply because their guild was headed by a higher-ranking official (e“xarco"). Reputation and esteem are not quantifiable, in any case, and hence are not useful in establishing discriminate professional traits or making meaningful distinctions among occupational guilds. 219 Lopez, “Silk Industry,” 3, 8, 22–23, 28–30, 32, 37–38; Goitein, A Mediterranean Society, 103; Harvey, Economic Expansion, 183. 220 Angold, “Medieval Byzantine ‘City,’” 29; Runciman, “Byzantine Trade,” 141; Bre´hier, Civilisation byzantine, 188. Angold (ibid., 29–30) maintains that the guilds were unable to meet the growing demand for silks because of regulatory constraints. As a result, “[i]t had in the end to be met by those working outside the gild framework, whether in the capital or in the provinces. This failure of the gilds boosted the ‘unofficial’ economy.” Angold’s assertion probably holds for developments during the eleventh century, when provincial towns entered the silk trade in earnest. As far as the capital is concerned, the guild system aimed precisely to prevent the parallel development of silk manufacturing activities outside the guild framework, and there is no evidence that it failed to do so during the tenth century (see pp. 269 and 273–74). Moreover, the Book of the Eparch did not in any way impose quantitative restrictions on the productive capacity of individual firms or interfere with their business decisions, and part of the domestic demand was met by unrestricted imports (see pp. 287–88, 294, 297–98, 311–12, 322). It follows that the guilds would have had no difficulty 218
GEORGE C. MANIATIS
327
that the local citizenry had access to a wide range of silks and that certain types were reserved exclusively for them (IV.4, 8; VIII.3), the constant need of the imperial court for silks to compensate government officials, the unrestricted import of silk fabrics, and their extensive use for ecclesiastic purposes and as a store of value,221 betoken an equally robust local demand. Finally, the emergence of important silk-manufacturing centers in the western provinces of the empire by the mid-eleventh century 222 is further evidence of an expanding demand for silks during the preceding century since their establishment and gradual development must be dated long before. To be sure, the low-income strata that comprised the bulk of the population did not enter the market, as the prices of silks were prohibitive for them. The primary consumers of silks remained the wealthy, state officials, the Church, the upper-middle class, and their counterparts abroad. Conspicuous display of status was a potent force behind the demand for high-priced silks. Moreover, the demand for luxuries, on which well-to-do and status-conscious consumers spent an increasing proportion of their income as they got richer, did not remain static over time.223 Although not immune from the vicissitudes and random shocks that afflicted the economy as a whole, the secular trend in the demand for silks was probably rising during the tenth century.224 It was fostered by a slowly expanding economy and the growing “feudalization” of the economic and social structure, by the increasing wealth and income of the nobility and the dominant classes (which was reinforced by skewed income distribution and incidence of taxation),225 the fact that in expanding their capacity and production as the market dictated and thereby meet any growth in demand (see below, p. 328). Therefore, the alleged failure of the guilds to meet this challenge and the consequent development of an unofficial silk manufacturing sector hardly seem plausible. 221 Harvey, Economic Expansion, 184. 222 K. M. Setton, “Athens in the Later Twelfth Century,” Speculum 19 (1944): 195–96; Jacoby, “Silk in Western Byzantium,” 460–62, 464–70, 490, 499. 223 Generally, the demand for luxuries has a relatively high income elasticity, i.e., given percentage increases (decreases) in income are accompanied by relatively larger percentage increases (decreases) in the amount spent on such articles. 224 In technical terms, over time the demand curve was shifting to the right, implying that at any given price a larger quantity was demanded or, for a given quantity, consumers were prepared to pay a higher price. The price level would also depend on the elasticity of supply: the less elastic the supply the higher the price, and vice versa. Apparently, the cultivation of mulberry trees remained profitable enough to encourage the expansion of sericulture in the empire, while shortfalls were met through imports. See Guillou, “Production and Profits,” 92–95. 225 On the territorial expansion of the empire during the Macedonian dynasty (867–1059), the concentration of land ownership in the hands of the powerful (dunatoi´) and the limited success of successive emperors to curb this trend, and the excessive tax burden of the small landholders, which led to their economic ruin and the patrocinium movement, see Ostrogorsky, History of the Byzantine State, 272–76, 280–82, 286–88, 306–7, 371–72; idem, “Agrarian Conditions in the Byzantine Empire in the Middle Ages,” in Cambridge Economic History of Europe (Cambridge, 1966), 1:215–22; S. Runciman, Byzantine Civilization (London, 1933), 194; P. Charanis, “On the Social Structure of the Later Roman Empire,” Byzantion 17 (1946): 51–55 and references cited therein. On the impact of population growth on the expansion of the area under cultivation and the increase in agricultural production; the opportunity afforded by the economic recovery for landowners to build up large estates, make yield-raising productive investments, and strengthen their economic position; and the effect of rising agricultural production on the growth of towns and markets, see Harvey, Economic Expansion, 47–48, 56–58, 78–79, 121, 159–62, 213–15, 218, 225–27; idem, “The Middle Byzantine Economy: Growth or Stagnation?” BMGS 19 (1995): 243–61, and references cited therein; Angold, The Byzantine Empire, 84–88 and 280. J. L. Teall also maintains that agricultural production intensified during the 9th and 10th centuries and the first half of the 11th century: “The Byzantine Agricultural Tradition,” DOP 25 (1971):
328
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
silks remained always in fashion, and the prestige and social distinction silks conferred upon status-conscious buyers.226 Given strong consumer preference, the quantity of silks bought over the relevant price ranges for local consumption and exports must have been substantial, and so must have been the silk industry’s corresponding share in total industrial output. And since productive effort among the various commodities in open economies is apportioned according to what buyers are prepared to spend on them, the silk industry was apparently able to muster the requisite financial (own capital, retained earnings, borrowings) and human resources (technical skills, organizational capacity) to expand as demand and profit opportunities unfolded. Correlatively, the market demand for good quality but moderately priced silks on the part of the upper classes tended to be price elastic, as the expenditure amounted to a considerable proportion of their income (although not their wealth)—at least for the majority of them.227 This meant that, for the bulk of the tradable silks produced by the private sector, referred to as “second-quality precious textiles” to differentiate them from the prohibited ones,228 a given decrease in price would result in a more than proportionate increase in output and sales.229 CONCLUDING REMARKS Rather than summarize the entire range of issues raised in this article, I conclude by highlighting its most important points. Where the findings of this study dissent from the literature, they do so because of differences in the framework in which the analysis is 53–59. Cf. also A. Andre´ade`s, “The Economic Life of the Byzantine Empire,” in Byzantium: An Introduction to East Roman Civilization, ed. N. H. Baynes and H. S. B. Moss (Oxford, 1961), 57–60; W. Treadgold, A History of the Byzantine State (Stanford, 1997), 569–79, 699–706; R. S. Lopez, “The Role of Trade in the Economic Readjustment of Byzantium in the Seventh Century,” DOP 13 (1959): 69. 226 Part of the demand for silks of exceptional workmanship and artistic distinction, perceived by buyers as somehow “unique,” is really for display, high-price distinction, or aesthetic value—for conspicuous consumption. The utility of possessing such an article derives not only from its inherent qualities, which reflect the actual price effect, but also from the price the consumer believes other people will think he paid for it, which reflects the conspicuous price effect. The demand curve for buyers whose conspicuous price effect is very strong and exceeds the actual price effect will be positively inclined (upward sloping), implying that consumers will buy more at higher prices because of the increased utility derived from the article and less at low prices, as the reduced utility due to the substantial reduction of the conspicuous effect at low prices induces a number of buyers to leave the market. See H. Leibenstein, Beyond Economic Man (Cambridge, 1980), 51–52, 62–65, 67. However, for the consumers of most of the high-quality silks with “non-unique” characteristics, the demand curve sloped downward in the conventional manner, implying that less was purchased at high prices and more at low prices. 227 Any judgment on the price elasticity of demand for an article must take into account three determinants: number and closeness of substitutes, importance in the buyer’s budget (i.e., the fraction of total expenditures it represents), and alternative uses. The more narrowly an article is defined, the more close substitutes it has and the more elastic the demand for it will tend to be. Also, the more important the good is in the buyer’s budget and the more uses an article has, the more elastic its demand. These factors can reinforce each other or work in opposite directions. For a complete discussion, see A. Marshall, Principles of Economics (London, 1920), 86–95. 228 Lopez, “Silk Industry,” 8–9. 229 Lopez asserts that the demand for precious silks was small and could not be increased beyond a certain point and that, if exports were not restricted, prices probably would have fallen without a proportionate increase in sales: “Silk Industry,” 19 and 42. Both assertions, implying a thin market and an inelastic demand for widely tradable silks, are at variance with the analysis presented here.
GEORGE C. MANIATIS
329
cast. Primary emphasis has been placed on the impact on the economic agents’ conduct and performance of basic economic and institutional parameters (e.g., import and domestic trade environment, industry organization, market structure, regulatory regime), operational dimensions (e.g., mode of production, output characteristics), and market dynamics. “Fallacy of division” arguments, strained hypotheses, conclusions based on insufficient evidence, unwarranted analogies rooted in the experience of the West, and abstraction from market dynamics has led to misunderstandings concerning the functioning of the guild superstructure and the behavior of guild members. These received notions of the guild system’s purpose and the effect of market fragmentation invited reconsideration, reevaluation, or confutation. Misreadings of, or unjustified readings into, certain provisions of the Book of the Eparch that led to erroneous conclusions have been elucidated. Allegations that the law could be scoffed at with impunity were challenged and found to lack foundation. Finally, following the development of a market typology, the likely nature of competition, its underlying forces, and their potential impact on the silk industry’s dynamics were delineated. The hypothesis that those metaxarioi who were denied direct access to imported raw silk because they were not included in the eparch’s list as being poor and hence not members of the guild of metaxopratai is specious. The term eujtelh´ " in provision VII.2 of the Book of the Eparch refers to inferior social standing—servile status—not to a lack of financial means. The ineligibility of the metaxarioi and katartarioi to participate in the collective purchase in this particular instance and their consequent entitlement to purchases from metaxopratai at capped profit margins was not motivated by their poverty, as has been alleged, but because both groups shared the same predicament: as slaves they occupied an inferior social position, although the reasons for their exclusion from this specific transaction remain elusive. The principle of the functional division of labor among guilds in the private silk industry, whereby no two guilds could compete in the purchase or sale of the same goods, was upheld—claims to the contrary notwithstanding. The view that the division of labor was unenforced and that more powerful guilds encroached upon weaker guilds’ activities in defiance of the law is unwarranted; this view slights the willingness and capability of the authorities to enforce the law and glosses over the ability of the offended guilds to ward off intrusions in the face of high stakes. Significantly, the persons hired by the metaxopratai in the course of their alleged invasive activities into yarn production were not poor and vulnerable workers as we have been led to believe. These workers possessed special skills and were employed to manage a` fac¸on contracts with katartarioi acting as principals, a sideline activity that enabled the metaxopratai to procure yarn beyond direct purchases lawfully without being involved in yarn production. In the same vein, the hypothesis that members of a guild were allowed to perform tasks that other guilds made a business of so long as this did not lead to market competition is indefensible. The regulatory apparatus aimed fundamentally to thwart parallel development of commercial silk manufacturing outside the guild system by channeling all related economic activities through a controllable setting, to enforce the mandated division of labor among guilds, and to forestall enterprise growth through obtrusive vertical or horizontal integration. The goal was to prevent the emergence of monopolistic market structures, concentration of economic power, and potential threats to the regime. It would make no
330
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
sense to set up a compulsory guild structure, designate operational functions, and enact elaborate rules, only to allow the same activities to be conducted outside the system. Permitting this would have defeated the purpose of the guild system and the state’s industrial policy. By the same token, implementation of policy objectives and enforcement of regulatory provisions dictated mandatory guild membership for proprietors of commercial and industrial establishments, although not for craftsmen employed as laborers, a notion that runs counter to widely held opinion. The growth of firms in response to expanding demand was not hampered by legal restrictions, as has been alleged. The objective of the guild structure and official policy was that growth in demand be shared by as many firms as possible, be met through expansion of existing firms and/or new entry, and be channeled through the cascaded markets. The restrictions imposed aimed solely to prevent the creation of large multistage firms through extension of their activities laterally or into preceding and/or succeeding phases of production, a move that was viewed as unrelated to market growth and that could have had inimical effects on competition. Such inroads would have negated a major objective of the regulatory regime, namely the preservation of competition and prevention of monopolies—surely a sound policy in light of the likely social costs and the dubious impact of integration on economic efficiency at this stage of the industry’s development. The guild system did not aim to fix prices through concerted action among guild members or thwart the free play of market forces to protect individual market shares. Contrary to expressed views, legally mandated group sales monopolies or buying monopsonies did not confer, ipso facto, exercisable monopolistic or monopsonistic pricing power on individual guild members, as members did not act on command, in unison, or in compliance with internal disciplinary regulations. More important, relatively unconcentrated market structures, unimpeded legal entry, moderate or low economic barriers to new competition, lack of control over the price of yarn and silks, opposition from wronged guilds reacting in self-preservation, and anti-monopoly legislation were potent countervailing forces, sustaining competitive rivalry among guild members and deterring collusive behavior. The prevailing market conditions, institutional arrangements (e.g., organized markets, body of laws, guild setup), and economic calculus all fostered antagonistic rather than cooperative attitudes in inter-seller and inter-buyer relations. Moreover, unlike the pattern typical in the medieval West, where the guilds often arrogated to themselves powers once wielded by the local government, the silk guilds in the capital had no control over the imperial authorities. In consequence, they could not with impunity exercise monopoly power in manufacturing and trade, counting on the acquiescence of authorities—even when market conditions were conducive. Nor did the guild system purport to shield the industry from external competition, as has been maintained. An expanding silk industry, well beyond the “infant” stage, was under no pressure to maintain its position, and unrestricted imports attest to the absence of protectionist tendencies. The silk industry was far more open to internal and external competition than is usually thought, as the authorities were aware that its survivability and growth depended on a policy that refrained from controlling prices, skill supply, conditions of employment, or wage rates. Instead, the government relied on the dynamic interaction of market forces and the market mechanism. Fundamentally, competition,
GEORGE C. MANIATIS
331
unimpeded entry and exit, free trade, and equality of opportunity rather than equality of economic results were the foundation of the state’s industrial policy. Market structure at each stage of the silk industry’s operations determined the nature of competition, influenced the behavior of participants, and shaped pricing decisions and performance outcomes. Yet, swayed by the vicissitudes of the economy and market dynamics, both structures and business conduct could vary over time, reflecting different productivity trends, growth rates in established firms, and the pace of new entry and exit. Such changes inevitably led to modifications of production and pricing patterns and, by extension, to changes in performance outcomes. Nevertheless, in the absence of economies of large-scale production, dominant advantage of product differentiation, and fortuitous absolute cost advantages, it is unlikely that market structures in silk manufacturing and distribution ever reached levels of high seller or buyer concentration in the capital. Among several logically possible outcomes in terms of market structure, conduct, and performance, certain ones appear more likely to have prevailed longer and with greater frequency. The provincial cocoon markets, characterized by a relatively high concentration of buyers supplied competitively by many small producers, very likely experienced price structures hovering below atomistically competitive levels and persistent excess monopsony profits. Contrary to long-standing belief, the monopsonistic pricing power of the collectively acting cocoon/raw silk and silks importers, although not absent, could be materially weakened since the price concessions extracted could vary significantly each market day, depending on the prevailing market conditions and ad hoc circumstances that could enhance the bargaining strength of the external suppliers. In the domestic cocoon/raw silk and two-tier yarn markets, relatively low market concentration, product homogeneity, low entry barriers, prevailing institutional parameters and economic calculus, and inauspicious prospects for collusive behavior all point to the prevalence of forces conducive to outcomes not substantially different from those produced by atomistically competitive conditions, largely resulting in normal or limited excess profits. Finally, the wholesale and retail markets for silks most likely operated under conditions of monopolistic competition. In this structure, prices of distant-substitute silks could remain above atomistically competitive levels and excess profits could persist, while prices of closesubstitute silks would tend in the longer run to gravitate toward competitive levels. In reality, however, as some sellers would have enjoyed a more advantageous price-cost relation and potential entrants might have been at some disadvantage relative to established firms, at any moment some sellers probably earned normal profits and others excess profits. The guild structure as applied to the silk industry provided a system of checks and balances within the parameters of the prevalent political, economic, and cultural milieu, and ensured a mutually accommodating modus vivendi among all parties concerned— the state, the business community, and the upper class. The state felt that it had the inalienable and uncontested right to intervene selectively and to control certain aspects of economic activity in the common interest. To be sure, the restrictions, inspections, and reporting requirements imposed must have been unwelcome and annoying to guild members—as they are still today. Yet, intrusive and cumbersome as these rules and regulations may have been, they did not result in micromanagement of the industry. Once
332
PRIVATE SILK INDUSTRY IN TENTH-CENTURY BYZANTIUM
the permissible economic activities and tradable silks were specified, the regulatory regime did not interfere with the strictly technico-economic dimensions of the firms’ operations, thus affording entrepreneurs ample room for independent action. As a result, state intervention did not perceptibly affect economic incentives, discourage private initiative, or impede the growth of firms. Once in place, the regulatory framework provided a set of non-shifting parameters that clearly defined the rules of the game and allowed guild members and sojourning merchants to conduct their affairs with certitude, as is attested by the continued expansion of the silk industry and trade.230 Bethesda, Maryland 230 Failure to put the functioning of the regulatory regime in proper perspective and to factor in the impact of market dynamics has led to sweeping pronouncements: “ce paradis du monopole et du privile`ge qui ´ dit de l’empereur Le´on VI Le Sage sur les corps de me´tiers de s’appelle Constantinople” (J. Nicole, “E Constantinople,” Revue ge´ne´rale du droit 17 [1893], repr. in idem, jEparciko` n Bibli´on, 293); “Free trade and free production were unknown in the Byzantine Empire” (A. A. Vasiliev, History of the Byzantine Empire [Madison, Wisc., 1958], 1:344); “There was no economic freedom in Byzantium, everything was regulated. Here the true socialist city was in being before Karl Marx and Lenin” (R. Guerdan, Byzantium, Its Triumphs and Tragedy [New York, 1957], 88).
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
The Amorium Project: The 1997 Study Season C. S. LIGHTFOOT ET AL. INTRODUCTION The year 1997 was the final one in the second stage of the program of research at Amorium. As a consequence, the season was largely devoted to the consolidation and assessment of work carried out since 1993 (when the present author succeeded the late Prof. R. Martin Harrison as field director of the project). This article, therefore, not only provides a preliminary report on the results obtained in 1997, but also contains an overview of the project’s achievements during the last five years, since recent excavations have enabled us to refine as well as substantially expand our understanding of the history and archaeology of the site (Figs. 1, A). The team spent most of the season studying and cataloguing material excavated in previous years, and little actual fieldwork was carried out.1 However, a major development of 1997 The work was generously sponsored by the British Institute of Archaeology at Ankara and in part also funded by Dumbarton Oaks, partly by donations from the Friends of Amorium, among whom Yu ¨ ksel Erimtan (of EMT, Ankara) deserves special mention. The project participants are grateful for the continued support of the Turkish auˇ, and for the help thorities both in Ankara and in Emirdag and advice provided by Prof. Cyril Mango and Dr. Marlia Mango. Special thanks must also go to Melih Arslan, Richard Ashton, Doc¸. Dr. Ebru Parman, and Dr. Hu ¨ seyin Tanrıkulu for their unstintingly generous support for the project. Visitors to the site during the excavations included Prof. Peter Kuniholm, Prof. Thomas Drew-Bear, Prof. Kenneth Harl, Dr. Stanley Ireland, and Prof. John Devreker (together with other members of the Pessinus team). We were also delighted to welcome Michael Mace (the Canadian ambassador), Thomas Wheeler (the South African ambassador), and Cafer Okray (of UDAS ¸, Ankara), all of whom showed great interest in the progress of our work. 1 The team comprised twenty-seven archaeologists and students, of whom nineteen were Turkish, six British, one German, and one Greek: Prof. Dr. Eric Ivison, Assistant
was the initiation of a program of geophysical surveying, conducted by separate teams led by two Turkish geophysicists. Other projects in and around the site included the continuation of the village survey, during which several new carved stones came to our attention (Fig. 2), and the start of a non-intrusive survey of the extensive necropolis that surrounds the ancient and medieval city (Fig. 3). In addition, an ethno-archaeological survey of the modern village was conducted by two anthropology students from Ankara University. This was intended not only to record the physical structure of the village, which is now in an advanced state of decay and depopulation, but also to preserve the villagers’ recollections and impressions of the archaeological ruins. As part of the program of site enhancement, work was carried out throughout the season in an area on the southern slope of the Upper City mound, which in recent years had been turned into an unsightly rubbish dump by some of the villagers. The accumulated trash was cleared away and removed entirely from the site, Director (College of Staten Island, City University of New York), Yalc¸ın Mergen (University of Anatolia, Eskis¸ehir), Dr. Margaret A. V. Gill, Dr. Mahmut G. Drahor (September 9th University, I˙ zmir), Dr. M. Ali Kaya (Su ¨ leyman Demirel University, I˙ sparta), Beate Bo ¨hlendorf (University of ¨ zkul (Gazi University, Ankara), Heidelberg), Nurs¸en O Olga Karagiorgou (University of Oxford); Mu ¨ cahide Ko¨ lker, I˙ rfan Yazıcı, ¸cak, Sultan S ¸en, Ays¸e Tas¸kin, Feruzat U and Hasan Yılmazyas¸ar (all from the University of Anatolia); S ¨ rel and Ebru S ¨ l (both from September ¸afak Gu ¸engu 9th University); Betu ¨l S ¸ahin, Is¸ıl Bayraktar, Aylan Erkal, and Aylin Zor (all from Ankara University); Fulya Adıyaman, Tolga Gu ¨ rbu ¨ z, Paola Pugsley, Julie Roberts, Yasemin Tok (Seljuk University, Konya), and Simon Young (University of Durham). The team was joined by the government ¨ yu representative, Mevlu ¨t U ¨ mez of the Afyon Archaeological Museum, who kindly provided much advice, help, and support throughout the season.
334
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
thereby enabling us to investigate the pit into which the material had been dumped. It was thus also possible to test the reliability of the information about the pit provided by men who admitted to taking part in the illicit excavation. In this way the work constituted a valuable corollary to the ethno-archaeological survey. The study of the human remains, largely from the rock-cut tomb excavated in 1995, was completed at the Dig House, while catalogues of the Roman and Byzantine terracotta lamps, the Ottoman tobacco pipes, the study collection of brick and tile, the loom weights, jar stoppers, and tokens were compiled. Many of the small finds were drawn by Paola Pugsley in preparation for inclusion in the final report on the excavations between 1987 and 1997, and at the same time a preliminary survey of the various groups of metal finds was completed.2 Work continued on the massive task of recording the carved stone fragments, with the total recorded since 1992 now reaching 1,256 items, while Thomas Drew-Bear gathered together the epigraphic material in preparation for a corpus of Amorium inscriptions. Further progress was also made in the study of the Byz¨hlendorf, assisted by antine pottery. Beate Bo Yasemin Tok, continued her work on a detailed study of the pottery from the Upper City (trenches TT and UU, Fig. A), while Nurs¸en ¨ zkul completed an inventory of the OttomanO period glazed pottery.3 One unexpected result of sorting through the pottery shards was the identification of a small number of handmade fragments from the Upper City. These may belong to the Early Bronze Age; if so, they would constitute the earliest finds from Amorium and provide a glimpse of the early history of the settlement on the mound. Twelve coins were recorded as surface finds during the course of the season, thus making a total of 243 coins re2
In conjunction with the cataloguing of the metalwork, Mu ¨ cahide Koc¸ak has undertaken a study of bronze belt buckles in the Afyon Archaeological Museum. In addition to the two examples found at Amorium in 1995 (cf. C. S. Lightfoot, E. A. Ivison, et al., “Amorium Excavations 1995: The Eighth Preliminary Report,” AnatSt 46 [1996]: 101–2, fig. 5), she has recorded a further forty-five examples, all recently acquired. A full report on this group of material is now being prepared for publication. 3 C. S. Lightfoot et al., “The Amorium Project: The 1996 Excavation Season,” DOP 52 (1998): 332–35 and figs. E–H.
covered from the site since 1987. In conjunction with the study of the numismatic material at Amorium itself, a study was carried out (together with Richard Ashton) of the coins in the Bolvadin Municipal Museum. The Byzantine material in particular provides a valuable record of low denomination issues that were circulating in the area, further reinforcing the view gained from the finds at Amorium, namely, that a monetary economy survived and indeed flourished in central Anatolia during the Byzantine period. It is to be hoped that such information will enable scholars to move toward a reassessment of the monetary economy in the Byzantine Empire and of the impact that the presence of a major thematic capital such as Amorium had on the local economy. GEOPHYSICAL SURVEY (by Mahmut G. Drahor and M. Ali Kaya)4 (a) History and Research Plan Since Amorium is a large site, geophysical investigation has long been recognized as the most practical and efficient way of obtaining information about the general layout of the settlement area.5 However, such non-intrusive survey work is only of value if one is able to recognize the significance (and probable date) of the buried features that, given the right conditions, it can reveal. While geophysical survey is often seen as a prelude to excavation (with its results only then tested by trial trenches), at Amorium it was possible to use the information obtained from the excavations that had been carried out in previous years to help with the interpretation and understanding of the geophysical results. The area selected for study in 1997 was the so-called enclosure in the Lower City, which lies between the church and the southeastern slope of the Upper City mound. 4 All of the equipment used in this work was supplied by the two teams, and it is fitting to record the gratitude of the Amorium Project to both Dr. Drahor and Dr. Kaya and their respective universities for their help and cooperation. Dr. Drahor was assisted by the two students from September 9th University, S ¨ rel and Ebru S ¨ l, ¸afak Gu ¸engu ¨ lker from the Amorium team volunteered while Feruzat U to act as Dr. Kaya’s assistant. 5 R. Martin Harrison attempted to exploit the use of magnetometry in the very first year of excavation, but, sadly, no results were obtained; cf. R. M. Harrison, “Amorium 1988: The First Preliminary Excavation,” AnatSt 39 (1989): 171 n. 11.
C. S. LIGHTFOOT ET AL. Part of the surrounding earth-covered bank had been excavated in 1996, revealing the remains of a massive defensive wall. The excavations had shown no sign of any Turkish-period activity in Trench XA/XB (Fig. A), and this encouraged us to believe that the enclosure contained only Byzantine (and earlier) features. Indeed, in the light of the preliminary findings a hypothesis was formulated that the enclosure may have been a middle Byzantine military compound.6 The aim of the work carried out during a period of intensive work on 1–16 August was to identify traces of the structures and layout within the enclosure.7 This would enable the archaeologists to formulate better a plan for the excavation of the enclosure in future seasons. Three principal methods of geophysical survey have been adapted for archaeological applications—resistivity, magnetometry, and the self-potential (SP) method—and all three were tested at Amorium during the season’s work. They detect either changes in ground resistivity (and, conversely, conductivity) or variations in magnetic intensity. Each method has its own advantages (and disadvantages) and gives different results, which, when all the results are correlated, provides for a more thorough picture of the area surveyed.8 Resistivity: Resistivity involves transmitting an electrical current through the ground in order to detect buried features by measuring the varying amounts of resistance that are encountered. It is suitable for most types of sites, but those that have very stony soils or extremely dry (or wet) conditions can give poor and/or misleading readings.9 Magnetometry: When clay and certain other materials are heated to a temperature of about 700° Celsius, they ac6
Lightfoot et al., “1996 Excavation Season,” 335. Initial attempts to conduct a resistivity survey of another, unexcavated section of the enclosure wall proved unsuccessful, since the density and spread of the stone rubble in the earth bank meant that signals were not transmitted between electrodes. 8 See M. G. Drahor, “Arkeolojik Alanlarda Jeofiziksel ¨ nemi,” in IX. Aras¸tırma Sonuc¸ları Prospeksiyonun O Toplantısı, C ¸ anakkale, 27–31 Mayıs 1991 (Ankara, 1992), 235–50. 9 See M. Joukowsky, A Complete Manual of Field Archaeology: Tools and Techniques of Field Work for Archaeologists (Englewood Cliffs, N.J., 1980), 48–49. Soil conditions at Amorium are usually very dry in midsummer, so the unseasonal rains facilitated greatly in obtaining good results from the resistivity work. 7
335
quire a magnetic force, or “signature,” of their own. A magnetometer measures the differences of magnetism in buried remains and can thus detect anomalies caused by the presence of material with an individual “signature.” The method is, therefore, particularly useful for detecting features such as kiln sites, rubbish pits containing quantities of ash, and burnt layers.10 Self-Potential Method: The self-potential effect is a natural voltage measured by two electrodes and a voltmeter. Variations in soil porosity caused by digging and backfilling should give rise to variations in the voltage measured by an SP survey. Likewise, using the SP method it should be possible to detect the presence of a relatively non-porous body such as a buried solid stone wall, while a loose stone or brick assemblage that allows water to pass through it should give quite a different reading. Thus, the SP method can be used for mapping disturbed features and buried structures.11 (b) Methodology The area investigated was divided into ten squares measuring 20 ⫻ 20 m (Fig. B). The resistivity survey was carried out using a twin probe array in squares A1, A2, B1, B2, B3, B4, C1, and C2. For this survey the grid intervals were set at 1 ⫻ 0.5 m, while the electrode spacing was at 1 m. Geomagnetic data were also collected for squares A1, A2, A3, B1, B2, B3, C1, C2, and C3, with grid intervals set at 1 ⫻ 1 m. In addition, self-potential measurements were taken for the area designated as “SP,” where very high resistivity and magnetic values had been recorded. The data for “SP” were collected using a gradient measurement technique, and for this the electrode spacing was set at 3 m, while the grid intervals were 1 ⫻ 0.5 m. For the geomagnetic survey, a separate grid of nine 20 ⫻ 20 m squares was used, and readings were taken at 1 m intervals. The sensor height of the magnetometer was initially set at a minimum depth of 0.6 m and a maximum of 1.8 m; further readings were then taken with 10
Joukowsky, A Complete Manual, 49–50. See J. C. Wynn and S. I. Sherwood, “The SelfPotential (SP) Method: An Inexpensive Reconnaissance and Archaeological Mapping Tool,” Journal of Field Archaeology 11.2 (1984): 195–204, esp. 195–200. 11
336
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
the sensors set at 0.6 m and 1.2 m, respectively. At daily intervals, readings were also taken at certain points to check if the selected area showed any signs of major fluctuation in its magnetic field. (c) The Resistivity Survey Once the readings had been taken, they were processed on site in order to create an apparent resistivity map from the raw data. This map revealed that high resistivity contours were concentrated in the southern and the southeastern parts of the area surveyed, whereas the northern and the northeastern sectors produced low resistivity contour levels (Fig. C). Very high resistivity values indicate that structures lie very close to the surface, and from this it would seem that there is a large building in the southern sector of the enclosure. In addition, a signal detection filter was used to locate features such as walls and to determine the effects of the resistivity signals with regard to the increase of the signal/noise ratio. As a result, a signal detection filter map was produced, which showed clearly the presence of the walls of this large structure (Fig. D). (d) The Self-Potential Survey The area in which very high magnetic and resistivity readings were detected was selected for further study using the self-potential method (Fig. E). The data were collected first by the gradient measurement technique with a pair of non-polarized copper-copper sulphate electrodes, then by contact resistance, and finally by the gradient self-potential method using metallic electrodes. The results obtained from the gradient self-potential survey correspond closely to those produced by the contact resistivity survey. The two maps are virtually identical, confirming the accuracy of the various readings with regard to both the polarization focal depth distribution and the horizontal projection of the polarization centers. Although the gradient self-potential map disclosed some deficiencies in the use of this technique for revealing archaeological features, the analysis of anomalies detected by the forward and inversion techniques clearly showed direction lines with some polarities.
(e) The Geomagnetic Survey Approximately 3,650 points were mapped within the 60-meter square grid, and the readings were processed on site, using different signal analysis methods (Fig. F). Several magnetic anomalies with high amplitudes were detected in the southern, southeastern, and northeastern parts of the enclosure; these indicated the presence of burnt material. ( f) Preliminary Results First and foremost, the season’s work has proved conclusively that the physical conditions at Amorium are conducive to geophysical survey, both resistivity and magnetometry methods. The success of the operation was in part due to the selection of an area that was both large enough to give a clear overall picture and known from surface observation and previous excavation to contain features that in all probability belonged to a specific period and formed part of the same complex; but in part it was also the lucky result of favorable weather conditions.12 Now that a start has been made, it should be possible to hone the use of equipment and techniques in the future in order to apply the results most effectively to the understanding of the site. Knowledge obtained from the various trenches opened in previous seasons suggests that the best results are likely to come from the Lower City where relatively undisturbed Byzantine structures, often containing burnt strata, have been found. The Upper City mound possesses a much more complex stratigraphy, stretching right up to the eighteenth century, and so is likely to provide a more confused picture. It is hoped, however, that over the coming years a program of survey in the Lower City will supply the archaeologists with a good idea of the layout both of individual structures and of the city as a whole. The specific results obtained in 1997 were also very satisfactory. Traces of several structures within the enclosure were revealed, although no regular plan was detected in them. A large building, measuring some 20 m square, 12 One may compare these results with the conclusions drawn from similar work at Sardis; cf. C. H. Greenewalt, Jr., and M. L. Rautman, “The Sardis Campaigns of 1994 and 1995,” AJA 102.3 (1998): 501.
C. S. LIGHTFOOT ET AL. was identified not far from the northwestern balk of Trench XB. It lies immediately below the modern ground surface; and expectations are that excavations in 1998 may provide evidence that the structure is contemporary with the enclosure wall and give some indication of the role that it served within the compound. CARBON-14 RESULTS Two ash samples were taken from Trench XA /XB in 1996 in order to obtain more information about the dating of the enclosure wall.13 Sample 1 was taken from Context XA25 in the trench outside the wall. The ash filled a small pit beside the end of the stone trough that protruded from under the enclosure wall (Context 4). The pit was located between the trough (Context 32) and the rubble stone wall foundations (Context 21) running parallel to the enclosure wall. Its fill comprised gray-black ash and a small quantity of bone. The pit was sealed before the robbing of the cross wall (Context 18), and belongs stratigraphically with the trough. The ash sample should, therefore, predate the construction of the enclosure wall. The radiocarbon “BP” age is given as 1480 ⫹/⫺100 BP (the Measured Radiocarbon Age with an assumed 13C/12C ratio of ⫺25 per mil relative to PDB-1). This may be converted into a calendar date of A.D. 470. The calendar calibrated results (2 sigma, 95% probability) are given as cal. A.D. 395 to 720 and cal. A.D. 735 to 760. Sample 2 (Context XB16) was taken from a thick ash layer in the south section of the trench within the enclosure. This burnt layer spread out in a decreasing thickness from immediately beside the enclosure wall at a depth of 1.54 m below the surviving top of the wall. It was associated with a horizontal row of blackened stones that was noted in the wall, above which there was a new row of larger blocks, taken to mark the foundation course of the Phase 2 wall. The ash deposit lay below Context 15, a layer of packed earth next to the 13 The samples were processed by Beta Analytic Inc. of Miami, Florida. The project participants are extremely grateful to John Giorgi and the Museum of London Archaeology Service for arranging the dispatch of the samples to the laboratory.
337
enclosure wall, which may be interpreted as a surface created during the construction of the Phase 2 wall. Below Context 16 was found another surface (Context 17), which contained many pottery and bone fragments. This lay above the rubble wall (Context 24), which appears to be stratigraphically contemporary with the two troughs found on this side of the enclosure wall. The ash layer is, therefore, placed between the construction of Phase 1 and Phase 2 of the enclosure wall. The radiocarbon “BP” age is given as 1160 ⫹/⫺40 BP (the Measured Radiocarbon Age with an assumed 13C/12C ratio of ⫺26.1 per mil relative to PDB-1). This converts into a calendar date of A.D. 790. The calendar calibrated results (2 sigma, 95% probability) are given as cal. A.D. 785 to 985. The intercept of the radiocarbon age with the calibration curve offers different dates for the two samples: Sample 1 ⫽ cal. A.D. 605, and Sample 2 ⫽ cal. A.D. 885. These dates should, however, be regarded as very speculative, and are not to be taken on their own as proof for the dating of the contexts from which the samples were taken. There is obviously a need to clarify these results by taking further samples and also comparing them with the information supplied by the other finds. Indeed, such scientific evidence is valuable principally as a means of providing independent confirmation of the dating sequence that has been arrived at during excavation by close attention to the details of stratigraphy, architectural features, and the various small finds. In this respect, the results of the C-14 tests would seem to support the view that the enclosure is a feature of the Byzantine city and that its defensive wall was probably not built before the eighth century.14 BONE REPORT (by Julie A. Roberts) The season’s work included the continuation and completion of the analysis of the human remains from the rock-cut tomb (MZ01, Fig. A), located in the west necropolis.15 Preserva14 For details of the excavation, see Lightfoot et al., “1996 Excavation Season,” 327–28, figs. A–D and 9–15. 15 Lightfoot, Ivison, et al., “Amorium Excavations 1995,” 97–102; Lightfoot et al., “1996 Excavation Season,” 328–29.
338
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
tion of the remains was generally poor, since all the larger fragments had been examined during the previous season. All skeletal elements were represented, indicating that these were probably primary burials of whole individuals; but the assemblage consisted mainly of vertebrae and bones from hands and feet. All data are presently being integrated with the information obtained in 1996 in preparation for a full report. A total number of 6,256 fragments was recorded, and from these a further twenty individuals were identified, bringing the total number of burials within the tomb to ninetythree. This is a conservative estimate based on the femora and humeri alone; other skeletal elements have yet to be considered, but, once all the data have been processed into an MNI (Minimum Number of Individuals) computer program, it should be possible to ascertain more precisely the number of individuals who found their last resting place in the tomb. The calculation of age at death was based on epiphyseal fusion, dental development and attrition, the appearance of the auricular surface and pubic symphysis of the pelvis, and the sternal ends of the ribs. Ectocranial suture closure was also taken into consideration but used only as a guide since it can be an unreliable indicator. Wherever specific age ranges could not be assigned, the general categories of Young Adult (18–30 yrs.), Middle Adult (31–45 yrs.), and Mature Adult (46⫹ yrs.) were applied. In some instances it was possible to state only that an element was Adult or Immature (⬍18 yrs.). At the present time, fifty-eight of the individuals are thought to be adult, and thirty-five to be sub-adult. The precise number of individuals within each age group has not yet been calculated, but it appears that the majority of the immature individuals were aged between 0 and 5 years at death. Sex was determined by the identification of morphological differences in the male and female pelvis and cranium, as well as by the consideration of postcranial metric data. A record of the numbers of males, females, and adults of unknown sex was compiled, and it is hoped that the analysis of these figures will reveal any sex-specific mortality patterns. Each individual fragment was examined for evidence of pathology, which was then classi-
fied according to its cause, as either infectious disease, degenerative disease, congenital abnormality, trauma, dental disease, neoplastic disease, or autoimmune disease. Relatively little pathology was identified, but among the conditions present were osteoarthritis (degenerative joint disease), dental disease, periostitis (superficial bone disease), minor traumatic injury, and cribra orbitalia (evidence for iron deficiency anaemia). Of these, the most frequently occurring conditions were degenerative joint disease and dental disease. COINS AND LEAD SEALS Since no proper excavations were conducted this summer, the number of coins recovered during the season was smaller than usual. Only twelve copper alloy coins were recorded, of which all but one were surface finds. The frequent heavy rains that bedeviled the season’s work actually aided the recovery of coins and other small objects from the surface of the site. ˘ another two coins In addition, in Emirdag were recorded as having been found in the nearby village of Demircili.16 These were subsequently acquired by the government repre¨ yu sentative, Mevlu ¨t U ¨ mez, and taken to the Afyon Archaeological Museum. Casts were also taken to be added to the project records. Since the project’s conservator, Karen Barker, did not attend the excavations this summer, the coins found at Amorium in 1997 still await cleaning and consolidation. Permission to take the finds directly to Ankara for this conservation work was not immediately forthcoming from the Directorate of Monuments and Museums, and, consequently, all of the coins were deposited in the Afyon Archaeological Museum at the end of the season. The present report thus represents only the results of an initial inspection of the coins, and further work will be required once they have been cleaned. However, it has been possible to identify eight of the coins with some confidence. Of particular note are two further issues of the Amorium mint: one (SF3703) is an Augustan coin bearing the name of the magistrate Kal16 ¨ mer U ¨ nlu The coins were in the possession of O ¨, a ˘, who stated that he had acquired shopkeeper in Emirdag them from a relative living in the village. For the position of Demircili in relation to Amorium, see Harrison, “Amorium 1988,” 168, fig. 1.
C. S. LIGHTFOOT ET AL. lippos, while the other, from Demircili, belongs to the reign of the emperor Nero.17 As in previous years, however, the majority of the finds belong to the Byzantine period; they include issues of Nikephoros I (SF3720), Michael II (SF3705), and Romanos I (SF3709), as well as an anonymous follis of class A2 (SF3704) and two signed folles of Constantine X (SF3700 and SF3707).18 The other coin from Demircili was also in good condition and was later identified from ¨ zme. It proved photographs and casts by Adil O to be of a very unusual type, depicting on the obverse a figure seated on a throne. No exact parallel has been found, but it has been attributed to the Seljuk sultan Kılıc¸ Arslan III, who occupied the throne as a six-year-old child for a mere eight months in 1204–5.19 In addition, three lead seals (SF3702, SF3711, and SF3719) were found during the season, whereas previously only two examples had come to light in ten years of work at the site. All three probably date to the middle Byzantine period (tenth–eleventh centuries).20 The last (SF3719, Fig. 4), a particularly impressive example, is the only seal recovered so far from the Upper City,21 while most of the others have been found in the vicinity of the Lower City church. The obverse depicts a bust of the Virgin Mary holding the infant Christ on her left arm; in the field to the left and right are the letters MHRÉQV; there is a border of dots. The reverse bears a six-line inscription, also within a border of dots: QKERQ É [N]IKHFORW / [M]AGICTRW / [R]ETARCSK É [TE]PANTW É MEL[C]N, that is, Q(eoto´ )ke, b(oh´ )17 RPC 3231/1 and RPC 3240/1, respectively (A. M. Burnett, M. Amandry, and P. Ripolle`s, Roman Provincial Coinage, vol. 1, 44 BC—AD 69 [London-Paris, 1992]). 18 After the close of the dig season, several more coins were picked up from the surface of the site by the excava˘it; they included an issue of the emtion guard, Bilaˆl Eryig peror Arkadios, a class 1 follis of Theophilos, and another anonymous follis. These will be properly recorded and studied in the next season. 19 A report on this coin has been prepared for publica¨ zme and C. S. Lightfoot, “Afyon tion separately; see A. O ¨ zerMu ¨ zesi’deki Figu ¨ rlu ¨ Bir Anadolu Selc¸uklu Sikkesi U inde Etu ¨rk Arkeoloji Dergisi (forthcoming). ¨ t,” in Tu 20 In the identification of these, help was received from Olga Karagiorgou, who plans to prepare the final report on all the seals found so far. 21 It was found on the surface in Trench TT on 12 August; it had clearly washed out of the balk during the heavy rain on the previous day. It measures 29 mm in diameter and weights 14.6 g.
339
q(ei) Nikhfo´ rv, magi´strv, be(s)ta´ rc(h), (kai`) k(a)tepa´ n(v), tv' Mel(is)s(h)nv'. The seal can thus be seen to have belonged to Nikephoros Melissenos, who was a vestarches, magistros, and katepano; it may be dated to 1067–78.22 This find is significant because it proves that powerful military men can be associated with Amorium even in the third quarter of the eleventh century; it indicates that the city was still an important enough place to find people there who were worthy of receiving letters from such as Nikephoros.23 Over thirty years ago Speros Vryonis stated in a lecture delivered in Ankara that “it would be of considerable interest to have a general tabulation of Byzantine coins in the various collections and museums of Anatolia at the present.” 24 Despite the importance of such work for the study of the survival of a monetary economy and the continued existence of urban sites in Anatolia during the Byzantine period, little progress was made in that direction until the last few years. Now, however, the Byzantine coin collections in Afyon, Amasya, Fethiye, and Sinop museums have been the subject of study by Turko-British teams.25 To add further to this body of evidence, a short season of work was carried out in Bolvadin at the kind invitation of Muharrem Bayar, the headmaster of the Bolvadin Anatolian High School. The Bolvadin Municipal Museum houses a small but significant collection of coins, most of which are believed to be local finds; it was thought worthwhile to study this material, both to compare finds from the area of ancient and Byzantine Polybotus, which lies 5 km west of Bolvadin, with those from Amorium and to provide a useful supplement to 22 G. Zacos and A. Veglery, Byzantine Lead Seals (Basel, 1972), 1.3:1480–81, no. 2697 and pl. 180. By a strange coincidence, another example was recently put up for auction in New York; see Classical Numismatic Group, sale catalogue (2–3 December 1997), lot 1815. 23 Another seal with an identical obverse but different reverse legend indicates that Nikephoros was also a dux of Triaditza (Serdica); cf. Zacos and Veglery, Byzantine Lead Seals, 1481, no. 2697bis. It is possible, therefore, that the letter sent to Amorium by Nikephoros had come from Bulgaria. 24 S. Vryonis, “Problems in the History of Byzantine An¨ niversitesi: Dil-Tarihi-Cog˘rafya Faku atolia,” Ankara U ¨ltesi Tarih Aras¸tırmaları Dergisi 1(1963): 113–32, esp. 126. 25 S. Ireland, “The Ancient Coins in Amasra Museum,” in Studies in Ancient Coinage from Turkey, ed. R. Ashton (London, 1996), 115–37.
340
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
the catalogue of the coin collection in the Afyon Archaeological Museum, being prepared by Ashton and John Casey. Indeed, the impression was gained from the material still in Bolvadin that the best coins from the area had probably been transferred to Afyon some years earlier. The Bolvadin collection thus forms part of a larger corpus of numismatic finds from this part of the Afyon province. In addition, the visit to Bolvadin provided the opportunity to view coins in two private collections, from which a number were selected for study, so that in total 285 coins were recorded (with dimensions, weight, and die axis) and casts were made for future use and photographic purposes.26 Of these, eighty-four were identified as Byzantine coins from the sixth to the eleventh century, and, as at Amorium, a large proportion are anonymous and signed folles of the late tenth and eleventh centuries. There are, however, also a fair number of issues from the reigns of Anastasios, Justin I, Justinian I, and Maurice Tiberios, whereas at Amorium coins of the sixth century are extremely rare. GLASS FINDS (by Margaret A. V. Gill)27 The site continues to produce large numbers of bangle fragments, both from the excavation trenches themselves and as surface finds, making this one of the most important collections for the study of medieval glass bracelets. With such a large sample, it has been possible to assemble a group of pieces illustrating a variety of different types, colors, shapes, and sizes, and by careful inspection of each fragment to identify features that help to explain how the bracelets were formed, which is a process that is still imperfectly understood. In all, fragments from some 1,284 bracelets 26 ¨ zme, “Ancient and R. Ashton, C. S. Lightfoot, and A. O Mediaeval Coins in Bolvadin (Turkey),” in Anatolia Antiqua (forthcoming). 27 For earlier reports, see R. M. Harrison et al., “Amorium Excavations 1991: The Fourth Preliminary Report,” AnatSt42 (1992): 219, 221, and fig. 6, nos. 18–30; R. M. Harrison, N. Christie, et al., “Excavations at Amorium: 1992 Interim Report,” AnatSt 43 (1993): 161 and fig. 4; C. S. Lightfoot et al., “Amorium Excavations 1993: The Sixth Preliminary Report,” AnatSt 44 (1994): 123–26 and fig. 3; Lightfoot, Ivison, et al., “Amorium Excavations 1995,” 107–9 and fig. 8; C. S. Lightfoot and E. A. Ivison, “The Amorium Project: The 1995 Excavation Season,” DOP 51 (1997): 296 and fig. C.
have come to light so far, including 527 from earlier seasons, which will be published in the final excavation report for the years 1987 to 1992, and 105 surface finds collected after the close of the main 1997 season and still waiting to be examined. Of the 652 pieces from the 1993–97 seasons, 103 showed traces of painted decoration; and among these, of particular interest are a small group of fragments that provide further evidence for the process of manufacture. These suggest that some (probably most) middle Byzantine painted bracelets were decorated at the straight-rod stage, before being reheated, curved into a circle, and having the overlapping ends fused. In seven cases, it is the section of the overlap that has been preserved. As soon as a softened rod was removed from the heat, the surface began to lose its viscosity more quickly than the interior so that, when the overlapping ends were pressed together, glass from the more fluid interior spread out beyond the edge of the surface. On one such fragment (Fig. G, no. 1) the tail of the tadpole-like blob of paint coincides exactly with the end of the rod; the overlap, however, extends further (with the weathering of the surface emphasizing the flow of the more viscous interior), indicating that the join was made after the application of the paint. Similarly, on a second fragment (Fig. G, no. 2) the lines on one side of the join terminate abruptly at the original end of the rod but at a distance from the edge of the overlap, while on the other side of the join, the tip of the cross-andquirk pattern is just visible as it disappears under the spread from the interior of the rod. A similar feature was noted on other fragments; the central zigzag line on one (Fig. G, no. 3) and the ends of two lines on another (Fig. G, no. 4) are clearly visible beneath the overlap, while the patterns on two further examples (Fig. G, nos. 5–6) disappear beneath the overlapping end of the bracelet. On the former fragment (Fig. G, no. 5), the final spiral has been distorted by tooling associated with work on the join. When a straight rod was held to the heat, it was attached at its midpoint to a pontil, the scar from which is visible on many of the Amorium fragments. On four examples (Fig. G, nos. 7 [not illustrated] and 8–10), part of the design has been damaged and obliterated by the application of the pontil and its
C. S. LIGHTFOOT ET AL. subsequent removal, thus confirming that the painted decoration was applied before the pontil was used to form the rod into a bracelet. More indirect evidence for the painting of bracelets at the straight-rod stage is provided by two further fragments (Fig. G, nos. 11–12), with elaborate decoration completely encircling the rods: such ornamentation would have been difficult to execute had the bracelets already been formed, yet there is no hint on either of the surviving pieces of any problem that the craftsmen might have experienced in dealing with what would have been the less accessible inner faces. Finally, one fragment (Fig. G, no. 13) demonstrates not only the craftsman’s skill but also his fallibility. It is of a more common type, of square cross section, decorated on three faces with a series of ovals between groups of endlines on either side of the overlap and with a reserve area on the opposite (pontil) side. This type was probably intended to be worn on its own so that all three painted faces would be visible; it was unnecessary for the fourth face to be decorated, as it would be unseen against the wrist. The actual painting of this example is characteristic, but the orientation of the sides is peculiar: while one of the three exposed faces is plain, one of the patterned faces is hidden on the inside. Clearly, the rod was painted as a normal bracelet, and the error took place after the rod had been decorated and while it was being bent into a circle, with a careless craftsman inadvertently giving it a 90° twist. This fragment, found in a secure middle Byzantine context in Trench XA just outside the Lower City enclosure wall, also indicates that such malformed bracelets were not always rejected by the manufacturer but were sent for sale, and indeed found a purchaser who was clearly not very fastidious. One can assume that it would have been easy for the craftsman to spot his error and discard the object, putting it in a pile for recycling. The fact that he did not do so may have economic implications: perhaps, the craftsman who formed the bracelet was different from the one(s) who created the glass rod and painted it. This may go some way toward explaining the general uniformity of Byzantine painted bracelets and their wide distribution, and it may not be too wild to hazard the suggestion that the rods
341
were created and decorated at a limited number of centers, from which local glassworkers purchased stock to turn into bracelets. This would certainly explain why the maker of the present example was reluctant to reject it despite his clumsy error: he would not have been able to reuse the piece of glass himself but would have had to write it off as waste. However, it is, of course, possible that some completely different reason explains the fortuitous survival of this malformed bracelet. A study of the glass vessel fragments from the early excavation seasons showed that the most common form used at Byzantine Amorium (second only to wine glasses) was the blue coil bowl. Blown in paler shades of bluish green and light green metal, the vessels are roughly hemispherical in shape, with a blue coil base ring, a fire-polished blue or light blue coil around the edge of the rim, and occasionally further trails lower down the side. However, more recent finds prove that Blue Coil Ware was produced in a wider range of body color, including colorless and various shades of yellow, that in some instances two techniques of decoration were combined, and that forms other than bowls were also manufactured in the ware.28 There are a few shards from Amorium with a band of relief decoration just below the rim; in one case the repeat impressed pattern was clearly made using patterned tongs, but in other cases the technique is less certain. Three of these shards are from blue coil bowls, with relief decoration below the blue coil rim. The outlined ellipse motif (Fig. H, no. 15) appears on both plain and blue coil rims. Despite the apparent difference in color (which is due to a difference in the thickness of the glass), it is possible that the base fragment (Fig. H, no. 16) may belong to the same vessel as one or another of the rim fragments (Fig. H, nos. 14–15). Another fragment (Fig. H, no. 17) represents a form new at Amorium. The lobed mouth was probably from a jug or, possibly, a small quatrefoil beaker; while one 28 Similar fragments of Blue Coil Ware have been found in the excavations at St. Nicholas church, in Demre, Lycia; ¨ tu see Y. Olcay, “Cam Buluntular,” in S. Yıldız O ¨ ken, “1995 Yılı Demre Aziz Nikolaos Kilisesi Kazısı,” in XVIII. Kazı Sonuc¸ları Toplantısı II, 27–31 Mayıs 1996 (Ankara, 1997), 477 ¨ tu and ill. 7; S. Yıldız O ¨ ken, “Demre Aziz Nikolaos Kilisesi ¨ ni˘ Aras¸tırmalarına Katkıları,” Ege U Kazısının Ortac¸ag versitesi Sanat Tarihi Dergisi 9 (1998): 95–96 and fig. 2.
342
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
lobe is certainly spouted and the neck appears narrow, the mouth is not a regular trefoil shape and the lobed sides are more characteristic of a beaker. There is another enigmatic shard with blue trail decoration: it is completely flat, which suggests that the fragment may have come from the wall of a square bottle. Another type of luxury glass from middle Byzantine contexts is represented at Amorium by a mere dozen shards that are fragments from at least two different shapes of goblet. This Red Streak Ware (Fig. I, nos. 18–23) is characterized by a bluish green translucent metal streaked with red. One fragment (Fig. I, no. 23) from the belly of the bowl in addition has remains of yellow enamel decoration in two wavy bands around the circumference; it may have come from the same vessel as the base fragments (Fig. I, nos. 20–21). Selected Glass Finds Glass bracelets: painted decoration (Fig. G) 1. (B30) Upper City, Trench L, Context 369. Fragment. Estimated diameter: 8 cm. Cross section flattish; with tooling at join. Light green with painted decoration in red: spiral linked to part of a second spiral; dab of paint in the tooled indentation above the overlapping join. 2. (B35) Upper City, Trench L, Context 370. Fragment. Diameter uncertain. Cross section flattish rectangular; with tool mark at join. Greenish blue with opaque red thread and painted decoration in white: St. Andrew’s cross with quirks between arms on one side of the join, and two curved lines on the other side. 3. (B274) Upper City, surface find. Fragment. Estimated diameter: 8 cm. Cross section rectangular; with join. Midgreen with painted decoration in red and white: white zigzag along the center, red zigzag on one side, and a line probably with beginning of a second zigzag on the other side. 4. (B193) Upper City, Trench TT, Context 59. Fragment. Estimated diameter: 5.5 cm. Cross section flattish oval; with tool mark at join. Surface of rod uneven. Pale green with painted decoration in red
and white: part of a white spiral with tendril, red angular spiral, and red hatched oval; crudely painted. 5. (B92) Upper City, Trench L, Context 409. Fragment. Estimated diameter: 4.5 cm. Cross section flattish triangular, with inlaid thread along carination; one end flattened with tool mark at join. Green with opaque red thread and painted decoration in white: row of running loops on either side of inlaid thread. 6. (B291) Dig House, surface find (from earth used in roof of dismantled outhouse). Fragment. Estimated diameter: 7.5 cm. Cross section flattish oval, with exterior groove; band shaped with shoulders and constriction toward join. Reddish brown with painted decoration in uncertain color, probably originally silver: row of running spirals, with groups of two or four end-lines on either side of the groove.29 7. (B87) Upper City, Trench L, Context 409 (not illustrated). Fragment. Estimated diameter: 7.5 cm. Cross section semicircular; with pontil scar. Light blue with painted decoration in uncertain color: parts of three oval medallions, with traces of tendril in one. 8. (B506) Upper City, Trench UU, Context 89. Fragment. Estimated diameter: 8.5 cm. Cross section round; with pontil scar. Greenish blue with painted decoration in creamy white: two (possibly three) oval medallions with line-and-dot borders, containing cross-pomme´es. 9. (B391) Lower City, Trench LC4, Context 6. Fragment. Estimated diameter: 8.5 cm. Cross section semicircular with flanges; with pontil scar. Light blue with painted decoration in creamy white: series of five medallions with knots and sprigs in spandrels; leafy scrolling motifs in three medallions, with intervening medallions probably left blank; row of dots along each flange. 10. (B111) Upper City, Trench L, Context 417. Fragment. Estimated diameter: 7.5 cm. Cross section round; with pontil 29 Lightfoot, Ivison, et al., “Amorium Excavations 1995,” 108, no. 5 and fig. 8, no. 5 (where a misprint occurs in the description).
C. S. LIGHTFOOT ET AL. scar. Mid-blue with painted decoration in white: parts of two elongated lozenges containing scrolls and tendrils, with leafy filling scrolls between. 11. (B670) Upper City, surface find. Fragment. Estimated diameter: 7.5 cm. Cross section round. Dark blue with painted decoration encircling rod, in gold, red, and yellow: two registers with similar designs, consisting of two medallions and part of a third containing crosses in quatrefoils; interstices between medallions filled with dotted lozenges; to side of medallions, two panels containing spiraling sprig motifs. Pattern in gold with red crosses and yellow dots. 12. (B38) Upper City, Trench L, Context 377. Fragment. Estimated diameter: 7 cm. Cross section round. Opaque appearing black with painted decoration encircling rod, probably in gold: panel delimited by end-line, containing spiral and curvilinear elements. 13. (B433) Lower City, Trench XA, Context 11. Fragment. Estimated diameter: 7.5 cm. Cross section square; with twist at either end of fragment. Light green with painted decoration in creamy yellow: similar design on three sides, consisting of a series of five or six ovals between pairs of end-lines; reserve section; hastily painted with some ovals linked and others open. Glass vessels: Blue Coil Ware (Fig. H) 14. (L380b) Upper City, Trench L, Context 380. Fragment from upper part of bowl. Estimated diameter of rim: 8 cm. Greenish yellow with blue coil. Rim rounded and thickened; colored coil at edge and relief decoration below: part of wavy line with superimposed ellipse. 15. (L/surface97a) Upper City, Trench L, unstratified (from surface in pithos pit). Fragment from upper part of bowl. Estimated diameter of rim: 8 cm. Greenish yellow with blue coil. Rim rounded and thickened; colored coil at edge and relief decoration below: parts of two outlined ellipses. Side convex, almost vertical. 16. (L419a) Upper City, Trench L, Context 419. Base of bowl. Diameter of base ring: 3.5 cm. Yellowish brown with blue coil.
343
Flat bottom with thickened center and pontil mark; colored complex coil base ring: inner spiral linked to surrounding circle. 17. (LC4/4g) Lower City, Trench LC4, Context 4. Fragment from mouth of jug. Estimated width of rim: 4 cm. Purple colorless with blue coil. Quatrefoil mouth; flaring wavy edge, and one lobe more pinched and pressed down to form spout. Rim folded in; colored coil at edge. Neck cylindrical. Glass vessels: Red Streak Ware (Fig. I) 18. (UU107a) Upper City, Trench UU, Context 107. Fragment from upper part of goblet. Estimated diameter of rim: 6 cm. Bluish green, streaked with red. Rim rounded and thickened, outsplayed. Concave neck; sloping shoulder. 19. (XB14a) Lower City, Trench XB, Context 14. Fragment from upper part of goblet. Estimated diameter: 6 cm. Bluish green, streaked with red. Tubular rim, folded inward; side flaring upward. 20. (TT56a⫹TT76a-b) Upper City, Trench TT, Contexts 56, 76. Three joining fragments from base of goblet. Estimated diameter of base: 6 cm. Colorless bluish green, streaked with red. Splayed foot with rounded edge. 21. (TT67a) Upper City, Trench TT, Context 67. Fragment from base of goblet. Colorless bluish green, streaked with red. Splayed foot with rounded edge. Probably from the same vessel as the preceding fragments. 22. (Surface96g) Lower City, surface find. Fragment from base of goblet. Estimated diameter of base: 4.5 cm. Bluish green, streaked with red. Tubular foot with pushed-in bottom and pontil mark. 23. (TT86d) Upper City, Trench TT, Context 86. Fragment from bowl of goblet. Bluish green with touch of red. Opaque yellow decoration: wavy festoon around circumference of bowl; traces of the second festoon below. MOSAIC TESSERAE, TERRACOTTA LOOM ¨ lker) WEIGHTS, AND TOKENS (by Feruzat U The study and conservation of the large quantity of glass mosaic tesserae recovered
344
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
from the site during the excavations were begun in earnest in 1994. In that year more than 17,800 individual tesserae were cleaned, sorted, and counted.30 Since then, however, further material has been found, again principally in the church during work undertaken in 1995 and 1996. From the recorded contexts it was observed that the majority of tesserae came from the area of the apse and the south aisle; noticeably fewer were found in the north aisle. This distribution suggests that certainly the eastern sanctuary and, perhaps, the eastern bays of the aisles had vaults decorated with glass mosaics. Colorless gold-glass and opaque red tesserae are prevalent. The proportion of tesserae found elsewhere on the site was relatively small, but the fact that the majority of these finds came from the Lower City suggests that other large, richly decorated Byzantine buildings were located in this part of the city.31 In addition to loose individual tesserae, the depots contain a number of small pieces of mosaic, which are fragments of ceiling mosaic that had presumably fallen as chunks into the accumulating debris on the floor of the church.32 Among these one example is noteworthy: the glass tesserae are arranged in tiered rows in the mortar bed, indicating that the piece came from a curved or arched section of the ceiling. Each tessera is thus set at an angle in order to display its face to the viewers standing below. Other fragments from the Lower City church appear as solidified blobs or drips of glass that are embedded in mortar. They would seem to have been formed when the ceiling mosaics suffered damage from a fire (or a series of fires) lit within the ruined church by its later Turkish occupants. A number of individual tesserae show the effects of a similar phenome30 C. S. Lightfoot, E. A. Ivison, et al., “Amorium Excavations 1994: The Seventh Preliminary Report,” AnatSt 45 (1995): 130–31 and fig. 7. 31 A total of 5,573 tesserae, from all contexts, were sorted and counted. 32 For further observations about the collapse of the church’s superstructure, see Lightfoot et al., “Amorium Excavations 1993,” 108, and Lightfoot, Ivison, et al., “Amorium Excavations 1995,” 97. In addition, during the excavations in 1993, traces of what appeared to be thin pools of glass mixed with gold were noticed adhering to the surface of the templon stylobate on the south side of the bema. These were taken to be the remnants of goldglass tesserae that had melted and dropped onto the stone slab.
non, having melted and then solidified into teardrop shapes or amorphous blobs (Fig. 5). Forty-six objects, provisionally recorded as loom weights, have been recovered from the site since 1993. In a preliminary study carried out in 1997, these finds were divided into four different groups according to their shape, fabric, and manufacturing technique. Most are circular, flattened disks with a central hole, but several have a rounded, globular body.33 Round and “doughnut-shaped” loom weights are said to have been used in Europe until the fifth century, after which the warp-weighted loom was replaced with the vertical twobeamed loom.34 It is likely, however, that the older type of loom, and therefore loom weights too, continued to be used in Anatolia until much later.35 Oddly, only one example of the other common form of loom weight, namely, the pyriform type with a horizontal hole for suspension in its upper part, has been found at Amorium.36 It may be that some of the “doughnut-shaped” objects found at Amorium, particularly those from secure Byzantine contexts (for example, SF3124 and SF3654 from Trench LC, and SF3651 and SF3658 from Trench XA, both in the Lower City), are in fact not loom weights at all. A recent study carried out in Israel concluded that such objects were probably stoppers for wine jars, with the central hole serving to allow gas produced during fermentation to escape.37 Further study of the different groups found at Amorium, their contexts, and the associated material is required before firm conclusions can be drawn about their date and purpose. Another enigmatic group of small finds consists of the forty-four items found since 1993 that belong to a class of objects described variously as disks, tokens, stoppers, or gaming 33 Similar objects have been recorded from the Phrygian highlands; see C. H. E. Haspels, La cite´ de Midas: Ce´ramique et trouvailles diverses, vol. 3 of Phrygie: Exploration arche´ologique (Paris, 1952), 91 and 108, pl. 38b. 34 E. Broudy, The Book of Looms (London, 1979), 23–27; R. J. Forbes, Studies in Ancient Technology (Leiden, 1956), 4:195–98. 35 Forbes, Ancient Technology, 199–202. Elsewhere it is claimed that this is not the case; see Z. Gal, “Loom Weights or Jar Stoppers?” IEJ 39.3–4 (1989): 281. 36 For this type, see Haspels, La cite´ de Midas, 92 and pl. 38c-d. The Amorium example, found before 1993, is now in the Afyon Archaeological Museum. 37 Gal, “Loom Weights,” 281–83, esp. nn. 11–12.
C. S. LIGHTFOOT ET AL. pieces. Such disks, of terracotta or stone, are plentiful on most ancient sites but are infrequently published.38 It is hoped that a detailed catalogue of the Amorium examples will be published in the final report for the years 1993 to 1997. All but one of the tokens were made of terracotta and produced by recycling broken pottery. Some shards were carefully selected to fit the purpose; thus, for example, SF2722 (a surface find) was chipped out of the flat base of a pottery vessel. The majority of the Amorium examples, however, were fashioned out of body fragments, which usually were heavily scratched and weathered pieces of red- and black-slip ware. These, although all roughly circular, are sometimes crudely made and of quite irregular shape. A number of other finds from Amorium (for example, SF2731 and SF2803) bear a close resemblance to these plain tokens, but they have two drilled holes near the center.39 Similar examples are known from Corinth.40 In addition to the uncertainty about the use of these small but ubiquitous objects, there is the question of whether suitable shards were collected specially by someone who made a living from creating these disks, or whether they were merely produced randomly by members of the household. TILE AND BRICK (by I˙ rfan Yazıcı) Another category of material is baked brick, which is relatively unknown and rarely studied, despite the fact that on most Roman and Byzantine sites tile and brick fragments constitute some of the most common finds.41 In 38 G. Davidson Weinberg, ed., Excavations at Jalame: Site of a Glass Factory in Late Roman Palestine (Columbia, Mo., 1988), 253, no. 181a-h, pls. 8–13, for eight terracotta examples that are very similar to some of those found at Amorium. 39 Another find (SF2729) has a single central hole. For parallels, see Haspels, La cite´ de Midas, 91 and pl. 38b. 40 G. R. Davidson, The Minor Objects, vol. 12 of Corinth: Results of Excavations Conducted by the American School of Classical Studies at Athens (Princeton, N.J., 1952), 297, 304, nos. 2631 and 2632. Gladys Davidson is probably correct in stating that the objects’ crude appearance precludes their use as dress buttons. 41 Published examples are scarce; see, for example, C. Foss, “Sites and Strongholds of Northern Lydia,” AnatSt 37 (1987): 84 and pl. XIXa (intact roof tile, measuring 1 m by 0.3 m, roughly curved, found at S ¸ahan Kaya, possibly of ¨ zcan, “Sulusuray—1990 kurtarma late antique date); B. O kazısı,” in II. Mu ¨ze Kurtarma Kazıları Semineri, 29–30 Nisan 1991, Ankara (Ankara, 1992), 172 and fig. 18a-b (square bricks found during the excavation of a church at Sebasto-
345
1994, a study collection was formed at Amorium, and during the 1997 season another sixty items were recorded and added to the catalogue.42 A number of intact or nearly complete examples have been found (Fig. 6), and fragments were likewise selected for addition to the collection on the basis of their unusual shape, decoration, or other distinguishing features.43 Several of these objects display marks that had been made accidentally before firing, such as foot marks, both human and animal; the paw marks of large dogs are particularly prevalent. Several more stamped and inscribed fragments have been recovered from the site since four examples were published in 1993.44 There is now a total of six fragments bearing the stamp ⫹ELIAN[O]U (Aijlianou'), all but one of which were found in the Lower City church (Fig. 7).45 A new example of another brick stamp, which reads [E]UGENI[OU] (Eujgeniou), has been recorded, while two roof tiles, also from the church, display a small circular stamp that contains a monogram composed of the letters HA.46 Since the latest brick stamps in Constantinople belong to the reign of Maurice Tiberios in the 580s–590s, the Amorium examples can in all likelihood be placed before this date; they presumably belong to the first phase of construction at the Lower City church.47 One final tile fragment has a fingerimpression inscription, with the letters EO polis). Discussion of the material is even more scarce; see H. Dodge, “Brick Construction in Roman Greece and Asia Minor,” in Roman Architecture in the Greek World, ed. S. Macready and F. H. Thompson (London, 1987), 106–16; for ¨. O ¨ zyig ˘it, “Alaturka Kiremidi Oluancient roof tiles, see O ¨ niversitesi Arkeoloji–Sanat Tarihi Dergisi 5 ¸sumu,” Ege U (1990): 149–79, esp. 159–67 (for Byzantine tiles). 42 K. Barker, “Brick and Tile,” AnatSt 45 (1995): 131–32. 43 B187: wall brick with protruding dogteeth and traces of mortar on all surfaces; length 0.269m, max. remaining width 0.257m, thickness 0.050–0.045m. 44 C. Mango, “Brickstamps,” AnatSt 43 (1993): 155 and fig. 1. 45 Cat. no. B093: ⫹E[LIANOU], found in the narthex in 1994; length (as extant) 0.141 m, width 0.102 m, thickness 0.035 m. 46 Cat. nos. B097 and B098; diameter of stamp 0.043 m. 47 From information kindly provided by Dr. Jonathan Bardill (personal communication). The (tentative) association of the ⫹ELIAN[O]U stamp with Aelianos who was a praefectus praetorio Orientis of the emperor Zeno (Mango, “Brickstamps,” 155) would push the dating of some of these bricks back to ca. 480; cf. J. R. Martindale, The Prosopography of the Later Roman Empire, vol. 2, A.D. 395–527 (Cambridge, 1980), 14, s.v. “Aelianus 4.”
346
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
drawn on its surface; similar objects have been recorded at the neighboring city of Pessinus.48 TOBACCO PIPES (by Simon F. Young) During the course of the season, the thirtyfive Ottoman pipe fragments that had been found at Amorium since 1987 were studied in detail. The main purpose was to create a comprehensive cataloguing system for the material, and subsequently to analyze the relationships of style within the assemblage and the implications of the location and context of the objects’ discovery. Furthermore, although Ottoman clay pipes are fairly ubiquitous, very little published material is available for sites in Anatolia.49 The work was thus intended to provide additional material for a proper study of the production, distribution, typology, and dating sequence of this interesting group of objects. Most of the fragments belong to clay pipe bowls, but two examples are made of meerschaum (Fig. 8) and so provide evidence for local use of the stone found exclusively near Eskis¸ehir, some 100 km from Amorium.50 The majority of the fragments (18 of the 35 pieces) originated as topsoil or surface finds, while eight came from mixed subsoil deposits 48 Cat. no. B217; length (as extant) 0.134 m, width 0.117m, thickness 0.034m. See J. Devreker, “Nouveaux inscriptions et monuments de Pessinonte (V),” Epigraphica Anatolica 28 (1997): 99, no. 4 and pl. 20. 49 M. T. Tarhan, “Van kalesi ve Eski Van ¸sehri kazıları, 1987,” in X. Kazı Sonuc¸ları Toplantısı I, Ankara, 23–27 Mayıs 1988 (Ankara, 1989), 378 and pl. 50; T. Mikami and S. Omura, “1986 yılı Kaman-Kaleho ¨yu ¨ k kazıları,” in IX. Kazı Sonuc¸ları Toplantısı II, Ankara, 6–10 Nisan 1987 (Ankara, 1988), 3 and pl. 8, 8–12 (dated to the 17th century); R. Meric¸, “1986 yılı Alas¸ehir kazı raporu,” in IX. Kazı Sonuc¸ları Toplantısı II, Ankara, 6–10 Nisan 1987 (Ankara, 1988), 243 and pl. 3 (a large number of examples, recovered from the Roman theater that had been used as a rubbish dump until the 20th century); H. Karpuz, “Konya Dokuzun Hanı Kazı ve Restorasyon Calıs¸maları, 1993,” in XVI. Kazı Sonuc¸ları Toplantısı II (Ankara, 1995), 382 and fig. 6; Y. Garlan and I˙ . Tatlıcan, “1994 ve 1995 Yılları Zeytinlik (Sinop) Amphora Ato ¨lyeleri Kazıları,” in XVIII. Kazı Sonuc¸ları Toplantısı II, Ankara, 27–31 Mayıs 1996 (Ankara, 1997), 340. 50 AM94/TT004/SF2852 and AM94/TT003/SF2881 (now in the Afyon Archaeological Museum). For the Eskis¸ehir mines and the use of meerschaum for pipes, see R. C. W. Robinson, “Tobacco Pipes of Corinth and of the Athenian Agora,” Hesperia 54.2 (1985): 167–68. The same article includes a catalogue containing eleven examples from Corinth (pp. 192–93, nos. C127-C137) and two from the Athenian agora (p. 210, nos. A43–A44). However, these Greek examples, most of which are dated to the 17th and the early 18th century, probably derive from mines at Thebes (p. 170).
and only nine so far have been recovered from firmly sealed contexts. In addition to indicating the necessity for further discoveries, this evidence also suggests that a number of the ¨y. finds relate to the modern village of Hisarko The refugees from the Balkans who founded the village in 1892 presumably left their share of material in the years until the introduction of cigarettes after the First World War. An iron tobacco tin, found in the subsoil in Trench ST on the northern slope of the Upper City mound, may be associated with these modern deposits.51 Despite this, the present assemblage has some important implications. First, the presence of tobacco pipes in post-Byzantine Amorium is a firm indication of the late nature of the Turkish occupation at the site. The introduction of tobacco to the region has been dated to the beginning of the seventeenth century.52 Secondly, the overwhelming concentration of pipe finds, from both the surface and deeper stratigraphy, on the Upper City mound (33 of the 35 pieces) would seem to suggest strongly that the Turkish settlement, which had preceded the modern village, had been located there and not in the Lower City. Two fragments found in Trench UU bear a close stratigraphic relationship to a silver coin of Mustafa III, dated to 1769, which was recovered from a room that also contained a large number of charred beams from the collapsed roof.53 Dendrochronology has also supplied a date of 1768 for the felling of the timber used in the construction of the roof.54 On present evidence, therefore, the final phase of habitation on the Upper City mound should be dated to the late eighteenth century. Given William Hamilton’s assertion that the site of Amorium was totally deserted in 1836, it would suggest that many of the other pipe fragments may belong to the eighteenth rather than the nineteenth century.55 In and around Trench L, three pipes of very similar construction have been found. They vary marginally in terms of shank decoration and dimension, as 51 The object is recorded as AM93/ST006/SF2524 (unpublished). 52 Robinson, “Tobacco Pipes,” 151–52. 53 AM96/UU072/SF3432 and AM96/UU087/SF3433. 54 From information kindly provided by Peter Kuniholm (personal communication). 55 W. J. Hamilton, Researches in Asia Minor, Pontus and Armenia (London, 1842), 1:449–51.
C. S. LIGHTFOOT ET AL. well as the presence or absence of a stamp, but are otherwise apparently indicative of a single style. The presence of a pipe fragment in the deeper contexts of the Lower City church probably indicates intrusive activity. THE UPPER CITY SONDAGE (UCS) There are a number of pits and piles of disturbed earth across the site that testify to a certain amount of illegal digging activity at Amorium prior to the start of the present excavation project in 1987. One such scar appears on the southern side of the Upper City mound, close to the principal track up onto the ¨k (Figs. 1, right, and A). According to some ho¨yu of the village elders, a deep pit was dug here in either 1947 or 1948, revealing a number of walls laid out in the form of a square. The bottom of this structure was apparently not reached, and the excavation was abandoned at a depth of some 6 m when water began to appear at the bottom of the pit. Apart from the discovery of some large blocks of masonry, claimed by some to bear inscriptions, no significant or valuable finds are reported to have been made. It is clear, however, from the surface material on the nearby spoil heaps that a considerable amount of Roman sigillata was thrown up in the course of the digging. The pit was then partially backfilled, and later came to be used as a refuse dump by the inhabitants of the nearest village houses; this practice continued right until the start of the present work. The initial purpose was to clear away the accumulated rubbish, which formed an unsightly blemish in what is otherwise a very clean and attractive village. In the process, it was hoped both to enhance the appearance of the site and to investigate the nature of the structures found fifty years ago. During the course of the season, the pit was cleared to a depth of approximately 4 m; at the bottom, modern rubbish (plastic, paper packaging, tin cans, etc.) was still being encountered. A roughly square area was exposed, flanked on all sides by walls of differing construction and date (Fig. 9). These can be defined as follows: Phase 1: The west wall, made of four courses of large, well-dressed blocks, continuing behind the north cross wall. Phase 2: The north cross wall, comprising two courses of finely dressed limestone blocks
347
with moldings at top and bottom, apparently hollow behind. Phase 3: A rubble-and-mortar wall built on top of Phase 2 and cutting Phase 1, perhaps the foundations of a defensive wall. Phase 4: A rough rubble-and-mortar wall above and behind Phase 1. Phase 5: Another rubble wall facing of small flattish blocks in front of Phase 4 and on top of Phase 1. Phase 6: The east wall, comprising large, roughly dressed blocks and mortar, built across in front of Phase 2, perhaps also foundations. Phase 7: A number of large limestone blocks on top of Phase 6, placed loosely and irregularly together. Similar blocks are located near the west side of the trench, while others are visible on the slope of the mound further to the east and west. They may all, perhaps, form part of the Upper City’s outer circuit wall. Phase 8: A stretch of rubble-and-mortar wall, not apparently connected with Phases 1–7, running from southeast to northwest and undercut, appearing as a vault or an arch. Phase 9: A number of large blocks lying either flat or in disturbed positions within the trench, evidently moved from their original location when the illegal digging was carried out. The dating and interpretation of these various elements remain problematic, not least because of the lack of any proper stratigraphy from premodern times. Several fragments of Eastern Sigillata Ware of early imperial date were recovered from the pit, but, since one of these was heavily encrusted with mortar, it would seem that this material had been residual and was reused during the construction of one or other of the structural phases. The fine masonry blocks, however, bear witness to the existence of a substantial (public?) building that cut deep into the side of the mound. Phases 1 and 2 may, therefore, date back to Roman or even Hellenistic times, but the later walls provide further evidence for the long and complicated multiphase building history of the site. CONCLUSIONS The first ten years of excavation have provided a wealth of new information about Amorium. The archaeological history of the site can now be seen to stretch from the Early Bronze
348
THE AMORIUM PROJECT: THE 1997 STUDY SEASON
Age to the late eighteenth century. Despite the fact that Amorium has been associated with the Hittite town of Aura, there is as yet no clear evidence from the Upper City mound that it was occupied in the second millennium B.C.56 However, it is likely that the site had existed in the Phrygian period and gradually developed into a thriving urban settlement in Hellenistic times. Thereafter it was continuously occupied throughout the first millennium, constituting one of the more important Roman cities in Phrygia and then becoming the capital of the Byzantine province of Anatolikon. The finds indicate that the city retained its strategic importance well into the eleventh century, but went into a rapid decline in the last quarter of the same century as a result of the unsettled conditions that affected central Anatolia in the wake of the Battle of Manzikert and the First Crusade. It is likely that a large proportion of the population fled from both the city and the surrounding countryside, taking the opportunity to find safer homes further west. So far, no pottery, coins, seals, or other objects have been recovered from the site that would provide evidence of occupation at the end of the eleventh and in the twelfth century. The total abandonment of the site during that period would help to explain why the long-standing name of Amorium was forgotten. After the site was reoccupied by Turkish settlers, probably in the early thirteenth century, it was given the new name of Hisarcık, which is later attested in the Ottoman archives. Apart from the use of prominent ruined buildings in the Lower City (as evidenced by the church), the Turkish settlement was confined to the Upper City. The newcomers took over and adapted for their own use the fortifications and dwellings of the middle Byzantine city on the mound. But it is also clear that during the Byzantine revival in the tenth and eleventh centuries occupation had not been restricted to the Upper City. The people of Amorium had spread out to occupy, perhaps in a rather scattered manner, the whole site (as encompassed by the ruined Lower City walls), and several major middle Byzantine public buildings were located in the Lower City area 56 See R. M. Harrison, “Amorium 1987: A Preliminary Survey,” AnatSt 38 (1988): 175 n. 1.
rather than in the refortified Upper City. Since there is evidence for pottery and tile production at Amorium in the tenth and eleventh centuries, it can be assumed that the city continued to act as a center for trade and commerce. It is likely that several other industries existed at Amorium—the most obvious candidate is a glass workshop where the extremely abundant bracelets would have been made. Amorium presumably also played a part in the marble- and stone-carving trades that continued to be a source of revenue for the Afyon region in Byzantine times. The large numbers of copper coins that have been found on the site confirm the fact that the local economy was thriving between 970 and 1080.57 The evidence from the necropolis may also be taken to show that parts of the site continued to function in Byzantine times in much the same way as they had in the Roman period. The rock-cut tomb had been originally built for only three occupants, so the discovery of skeletal remains belonging to no less than ninety-five individuals is remarkable and requires explanation. Detailed analysis of the bone material shows clearly that it cannot be taken as evidence of a mass burial, since there are no signs of traumatic injuries that could be associated with a violent end such as a massacre. Rather, the massive quantity of human bones indicates that the tomb had been used over a very long period, probably extending from the Roman to the middle Byzantine period (second-eleventh century).58 This accords well with the sparse amounts of shards found mixed in with the bone; some of these pottery fragments may well be of middle Byzantine date. Such late reuse of the tomb would be highly unusual, for at most sites extramural burials cease in the early seventh century; but the survival of Amorium as a major urban center in the Dark Ages may help to explain this anomaly. As yet, however, the excavations have failed to reveal a great deal of the Dark Age city. This is largely due to the unexpected extent and 57 Of the 243 coins found so far (i.e., between 1987 and 1997), 81—or precisely a third—have been identified as anonymous or signed folles. 58 Eric Ivison has expressed scepticism on this point, believing that the tomb did not continue in use much after ca. A.D. 600.
C. S. LIGHTFOOT ET AL. richness of the middle Byzantine remains. The devastating effect of the siege of 838, which has been attested in finds from the Lower City fortifications and church, may also partly explain the meagre nature of the archaeological remains. There are also some puzzling lacunae in the material found at Amorium. Perhaps the most surprising is the complete absence of Early Christian funerary stelae. Simple tombstones with a standard formulaic text are widespread in Anatolia; examples exist, for example, in the Bolvadin Municipal Museum (presumably from Polybotus) and are recorded in the Phrygian highlands, northwest of Amorium.59 By contrast, the only Christian funerary monuments known from Amorium are much more sophisticated. One is the inscribed sarcophagus, dated 591/2, which was found on the southern slope of the Upper City mound in 1933; the other, probably originally from Amorium, is the erudite verse epitaph of Pientios, now built into the minaret of the village ˘ılcık.60 Other groups of material mosque at Ag 59 C. H. E. Haspels, The Highlands of Phrygia: Sites and Monuments (Princeton, N.J., 1971), 342, no. 114 and fig. ¨ zu 629 (found at Sandık O ¨ , ca. 25 km west of Seyitgazi/Nacoleia). 60 The sarcophagus is now on display in the garden of the Afyon Archaeological Museum (Inv. no. E1440); see W. H. Buckler and W. M. Calder, Monuments and Documents from Phrygia and Caria, Monumenta Asiae Minoris antiqua
349
are also noticeable by their absence, such as, for example, Byzantine reliquary crosses. On the other hand, the site has already supplied some important examples of unique material, among them the first excavated middle Byzantine potter’s kiln in central Anatolia, the first published example of a new class of anonymous follis, and a fortified middle Byzantine compound, discovered in the Lower City, that was possibly used as the headquarters of the “Army of the Anatolics.” The lesson of the last ten years is that a large site such as Amorium, with its long and complex history, holds many surprises. Despite the best-laid plans, it is not always possible to achieve objectives that have been set, since discoveries in the field have conspired to deflect attention away to other goals. Nevertheless, the priorities for the coming seasons are clear. On-site, the main objective is to investigate further the central area of the Lower City in and around the enclosure, while off-site it is essential to push forward the program of publication as rapidly as possible so that the results of the past decade’s work can be made available to others. University of Durham, England 6 (Manchester, 1939), 134–35, no. 386, pl. 68. For the Pientios inscription, see Lightfoot, Ivison, et al., “Amorium Excavations 1994,” 135–36 and pl. XXb.
This is an extract from:
Dumbarton Oaks Papers, No. 53 Published by
Dumbarton Oaks Research Library and Collection Washington, D.C.
Sketch plan of Amorium, 1987–97 (adapted from H. G. Welfare, assisted by H. Dodge and A. Wilkins, in R. M. Harrison, “Amorium 1987: A Preliminary Survey,” AnatSt 38 [1988]: 178, fig. 2)
Fig. B
Geophysical research plan of the Lower City enclosure (prepared by M. G. Drahor)
Fig. C
Resistivity survey of the Lower City enclosure (prepared by M. G. Drahor)
Fig. D
Resistivity and self-potential maps of the Lower City enclosure (prepared by M. G. Drahor)
Fig. E
Self-potential maps of the Lower City enclosure (prepared by M. G. Drahor)
Fig. F Magnetic gradient map of the Lower City enclosure (prepared by M. G. Drahor and M. A. Kaya)
11
8
4
Fig. G
1
12
13
6
Glass bracelet fragments with painted decoration, scale 3:4 (drawing by M. A. V. Gill)
9
5
2
10
3
14
15
16
Fig. H
17
Glass vessel fragments, Blue Coil Ware, scale 3:4 (drawing by M. A. V. Gill)
18
19
22
23 20
Fig. I
21
Glass vessel fragments, Red Streak Ware, scale 3:4 (drawing by M. A. V. Gill)
1 Upper City mound, general view looking northeast (Neg. AM97/03/33)
2 Middle Byzantine colonnette capital built into a village house, T1072 (Neg. AM97/03/04)
3 Monumental chamber tomb, MZ02, incorporating two Phrygian doorstone stelae, T1073A shown to right (Neg. AM97/03/07)
4
Lead seal of Nikephoros Melissenos, SF3719 (Neg. ColPr AM97/07/16–17)
5
Glass tesserae from the south aisle of the Lower City church, showing the effects of a fire (Neg. AM97/08/33)
7 Stamped brick fragment, B093 (Neg. AM97/05/36) (photo: I. Yazıcı)
6
Wall brick with the dogtooth edge, B187 (Neg. AM97/05/20) (photo: I. Yazıcı)