The Elephant and the Ark: Cultural and Material Interchange across the Mediterranean in the Fighth and Ninth Centuries LESLIE BRUBAKER Pie the medieval Mediterranean, luxury goods were exchanged as objects of trade, as spoils of war, and as gifts.' For a cultural historian, the interest of exchange or ex- port of goods lies less in the fact of the exchange itself than in why a particular artifact (or type of artifact) was selected for export or import, pillage, or gift exchange and how that object was redefined once it was in a new context. The complexities of trade, war, and diplomacy give way to the ambiguities of socially constructed meaning, which is itself not static: moving an object changes its meaning. Sometimes the new meaning was calculated by the exporter. In 506 Cassiodorus or- dered Boethius to take a water clock to the ruler of the Burgundians, and to show the Bur- gundians how it worked, so that, in Cassiodorus’s words, “when they have turned from their amazement, they will not dare to think themselves the equals of us, among whom, as they know, sages have thought up such devices.”? In 757 the East Roman emperor Con- stantine V may have had similar hopes when he sent the Frankish king Pepin an organ, along with Byzantines to show the Franks how to use it.* Later described by Notker as the “most remarkable of organs ever possessed by musicians,”* the instrument—like the clock sent by Cassiodorus to Burgundy—represented technology not available to its recipients, and thus had the potential to demonstrate the superiority of the sender. It is a pleasure to thank participants in the Symposium on Realities in the Medieval Mediterranean, and, later, the Byzantine Seminar at Oxford, for their comments after this paper was delivered. I am grateful to the anonymous readers for helpful comments; to Mayke de Jong, David Ganz, Rosamond McKitterick, and Jinty Nelson for advice on Carolingian bibliography; and, as always, to Chris Wickham. ' For overviews, see A. Cutler, “Gifts and Gift Exchange as Aspects of the Byzantine, Arab, and Related Economies,” DOP 55 (2001): 247-78; and M. McCormick, Origins of the European Economy, Communications and Commerce AD 300-900 (Cambridge, 2001). 2 Variae 1.45: MGH, AA 12 (Berlin, 1894); trans. S. Barnish, Cassiodorus, Variae, Translated Texts for His- torians 12 (Liverpool, 1992), 20-23; I thank Chris Wickham for the reference. Cassiodorus would have ap- preciated the famous phrase of Marcel Mauss that gifts “are still followed around by their former owner”: M. Mauss, The Gift. Form and Reason for Exchange in Archaic Societies, trans. W. D. Halls (London, 1990), 66. ° See J. Herrin, “Constantinople, Rome and the Franks in the Seventh and Eighth Centuries,” in J. Shep- ard and S. Franklin, eds., Byzantine Diplomacy (Aldershot, 1992), 91-107. * Notker, Gesta Caroli Magni Imperatoris 2.7: MGH, ScriptRerGerm (Berlin, 1959); trans. L. Thorpe, Einhard and Notker the Stammerer, Two Lives of Charlemagne (Harmondsworth, 1969), 143. 176 CULTURAL AND MATERIAL INTERCHANGE Whatever the intention of the dispatcher, however, a new meaning is inevitably at- tached to the object by its receiver. In 802 the Abbasid caliph Harun al-Rashid sent Char- lemagne (among other things) an elephant named Abt ‘l-Abbas, which remained an ex- otic, and obviously foreign, marvel until he died in 810.° If Hartin reasoned as Cassiodorus had, however, he was doomed to disappointment, for while the Franks were indeed de- lighted with the elephant, they underscored its importance with claims that Abu ‘l-Abbas had been the caliph’s only elephant, and interpreted its arrival as an expression of Abbasid recognition of Carolingian status.° The elephant’s value lay in its uniqueness, which the court historian Einhard emphasized by the patently false assertion that Abt ‘1-Abbas had been as singular in Baghdad as he was in Aachen. Reinterpretation was not confined to recontextualized gifts. In 1049 the Byzantine emperor Constantine IX Monomachos sent a chrysobull—in scroll form, and validated by the imperial gold seal—to the German emperor Henry III.’ Constantine’s chrysobull was but one of many texts that we can document crossing the Mediterranean in the central Middle Ages. A century earlier, for example, Constantine VII sent a letter written in gold on blue-dyed parchment, valorized by his gold seal, to Abd al-Rahman III, Umayyad ca- liph of Spain;* and twenty years before that, Romanos I Lekapenos sent a bilingual pair of letters to the Abbasid caliph al-Radi, with the Greek text written in gold, the Arabic in sil- ver.’ Constantine IX’s letter to Henry was not in itself unusual; but it is one of the very rare examples of which we are told the subsequent history, and so can see precisely how the move across the Mediterranean changed its meaning. The chrysobull arrived in Germany in 1049. In 1050 Henry ITI sent it to the church of Sts. Simon and Jude in Goslar, along with relics, liturgical containers, and a letter of his own. Henry’s letter (a garbled version of which is preserved) explained that the church administrators were to melt down the golden seal and fashion it into a chalice, but to keep the parchment scroll whole, and use it as an altar cloth.'® This is cultural appropriation with a vengeance. If Henry’s instructions were carried out, objects that were intended as emblems of Byzantine authority were converted into something else entirely: an imperial letter became a tablecloth, an emperor's seal was transformed into a vessel for liquids. Of course, the relatively small amount of gold in a seal is not enough to make a chalice,"’ but > Annales Regni Francorum s.a. 802 and s.a. 810: MGH, ScripitRerGerm (Hannover, 1895); trans. B. W. Scholz, Carolingian Chronicles (Ann Arbor, 1972), 82, 92. ° Einhard, Vita Caroli Magni 16: MGH, ScriptRerGerm 25 (Hannover, 1911); PE. Dutton, Charlemagne’ Courtier: The Complete Einhard (Peterborough, Ont., 1998), 26. 7 Chronicon sanctorum Simonis et Iudae Goslariense: MGH, DtChron 2 (Hannover, 1877), 605. For discussion, see P. Buc, “Conversion of Objects,” Viator 28 (1997): 99-143, at 100, with earlier bibliography to which should be added O. Kresten, “Correctiunculae zu Auslandsschreiben byzantinischen Kaiser des 11. Jahrhunderts,” in M. Pippal, R. Preimesberger, F. Miitherich, and A. Rosenauer, eds., Festchrift fiir Hermann Fillitz zum 70. Geburtstag (= Aachener Kunstblatter 60) (Cologne, 1994), 143-58. I thank Peter Schreiner for the latter reference. 5 As recorded by Ibn ‘Idhari and al-Magqgari: discussion, trans., and bibliography in A. Vasiliev, Byzance et les arabes 2.2: Extraits des sources arabes, trans. M. Canard (Brussels, 1950), 218-19, 276-81; P. Soucek, “Byzantium and the Islamic East,” in H. Evans and W. Wixom, The Glory of Byzantium. Art and Culture of the Middle Byzantine Era, AD 843-1261 (New York, 1997), 408-9. ° As recorded by Ibn al-Zubayr: discussion, trans., and bibliography in O. Grabar, “The Shared Culture of Objects,” in H. Maguire, ed., Byzantine Court Culture from 829 to 1204 (Washington, D.C., 1997), 115-30, at 117- ol '0 See note 7 above. | This has suggested to at least one scholar that the letter has been corrupted beyond recovery: Kresten, “Correctiunculae zu Auslandsschreiben byzantinischen Kaiser.’ LESLIE BRUBAKER 177 Henry’s point was political: it expressed the Salian appropriation of Byzantine imperial symbols. The meaning of the chrysobull changed. As sent, it promoted the imperial authority of the Byzantines; as manipulated, it visualized the greater power of the Salian emperor.” Something produced as a unique, high-status object, beyond price, was treated as con- vertible.!? The conversion demonstrated—or was meant to demonstrate, if only to him- self—Henry’s control over Byzantine products, a metaphor for his superiority. In both forms, the chrysobull was a commanding imperial statement, and it remained part of the parlance of court culture. But the messages conveyed by the chrysobull as sent and the chrysobull as forwarded were diametrically opposed: its form, function, and meaning started by serving the Byzantines and ended by promoting the Salians. Henry treated Constantine’s chrysobull as an alien object and culturally redefined it.'* Few such blatant examples of cultural redefinition remain visible to us. Recent scholarship has promoted the idea of an international court culture, with shared values, that appreciated the portability of élite objects and the variable and fluid interpretations that such portable objects could be expected to carry.!° Luxury silks and expensive metalwork—the media usually cited in this context—fit the model neatly, but the fate of Constantine’s chrysobull (like the reinterpretation of Abt ‘l-Abbas the elephant) suggests the exercise of caution. The need for circumspection is underscored when we turn to the materialization of cultural exchange, for culturally conditioned reactions to the same phenomenon could result in diametrically opposed material responses. This is par- ticularly clear in the late eighth and ninth centuries. The years between ca. 750 and ca. 850 saw considerable mobility and cultural interac- tion across the Mediterranean.'® Theodulf, from 797/8 bishop of Orléans and abbot of Fleury (St. Benoit-sur-Loire), moved from Visigothic Spain to Francia in the 780s.'’ The future patriarch Methodios (843-847) moved from Sicily to Constantinople to Rome and then finally back to Constantinople.'* Embassies traveled back and forth with great regu- larity between Aachen, Constantinople, Jerusalem, and Damascus or Baghdad;’° pil- erimage continued to the Holy Land, North Africa, and into the Sinai peninsula.*° Cross- 12 So Buc, “Conversion of Objects,” 100. '8 The chrysobull’s new liturgical associations may have implicitly honored it (as suggested to me by Cather- ine Holmes, whom I thank for discussion of this point), but Constantine did not intend this use: the function was determined by Henry and required the destruction of the seal and the silencing of the letter. '4 See, e.g., I. Kopytoff, “The Cultural Biography of Things: Commoditization as Process,’ in A. Appadu- rai, ed., The Social Life of Things: Commodities in Cultural Perspective (Cambridge, 1986), 64-91, esp. 67. 'S See, e.g., R. Cormack, “But Is It Art?” in J. Shepard and S. Franklin, eds., Byzantine Diplomacy (London, 1992), 218-36; E. R. Hoffman, “Pathways and Portability: Islamic and Christian Interchange from the Tenth to the Twelfth Century,’ Art History 24.1 (2001): 17-50. 16 See McCormick, Origins, 123-277. 7 A. Freeman, “Carolingian Orthodoxy and the Fate of the Libri Carolini,” Viator 16 (1985): 65-108; A. Free- man and P. Meyvaert, “The Meaning of Theodulf’s Apse Mosaic at Germigny-des-Prés,” Gesta 40.2 (2001): 125-39. '§ ODB 2: 1355. '8 McCormick, Origins, 138-47, 175-81, 254-61; P. Magdalino, “The Road to Baghdad in the Thought- World of Ninth-Century Byzantium,” in L. Brubaker, ed., Byzantium in the Ninth Century: Dead or Alive? (Alder- shot, 1998), 195-213; T: Lounghis, Les ambassades byzantines en Occident depuis la fondation des états barbares jusqu aux croisades, 407-1096 (Athens, 1980). 20 McCormick, Origins, 129-38, 197-210; L. Brubaker and J. Haldon, Byzantium in the Iconoclast Eva (ca. 680- 850): The Sources, Birmingham Byzantine and Ottoman Monographs 7 (Aldershot, 2001), 57-60. 178 CULTURAL AND MATERIAL INTERCHANGE cultural interest was high, but misunderstandings were constant and sometimes devel- oped into hostility.2! Pope Leo III’s letter to Charlemagne reporting on rumors of an at- tempted usurpation of the Byzantine throne by the patrikios Constantine—rumors that the pope knew were untrue, as he told Charlemagne—demonstrates that even false reports about the Byzantines were interesting in the West, but also how easily misunderstandings could arise and be circulated.” One of the classic misinterpretations involved this same tri- angle of the pope, the Carolingians, and the Byzantines, and was focused on the Frankish understanding of the Second Council of Nicaea in 787, convened by Empress Eirene, her young son Constantine VI, and Patriarch Tarasios to reinstate the veneration of sacred portraits. As is well known, when the Greek version of the Acts of Nicaea II reached Rome, it was so poorly translated into Latin that it contradicted the actual Byzantine position, most blatantly by translating the Greek proskynesis (veneration) with the Latin adoratio (adoration) so that it appeared that the Orthodox church was espousing idolatry. This Latin version was sent to the Frankish court, and sufficiently perturbed Charlemagne that he commissioned the Visigothic émigré Theodulf to write a response demonstrating why the Byzantine position was incorrect. To this end, Theodulf wrote the Opus Caroli Regis con- tra Synodum (the Libri Carolini) between 791 and 793, after which the Carolingians appar- ently discovered that the translation they had used was faulty and, more important still, that Pope Hadrian I had endorsed Nicaea II. Charlemagne prudently decided to archive the Opus Carolt.”° Beyond its demonstration of the potential complexities of international dialogue, this particular episode had a material consequence that feeds directly into our understanding of the dynamics of cultural exchange. One of the passages of the Acts of the 787 council that struck a chord with Theodulf also resonated with the Byzantines, and their distinct responses exemplify how the same concept could be visualized to radically different effect in the Byzantine East and the Latin West. The passage reconciled the second command- ment, which forbade the creation of “graven images,” with God’s order to Moses to deco- rate the tabernacle, recorded in Exodus 25:18—20. It reads: Thou shalt not make for thyself an idol. . . . However, when his faithful servant Moses was making the tabernacle ... under the commandment of God, he, in order to show that everything is to the service of God, made perceptible cherubim in the form of men—an- titypes of the spiritual ones. These cherubim were to overshadow the seat of expiation, a seat which was an antecedent type of Christ; for, as the divine apostle [ John] says, He is the expiation for our sins. Therefore, he introduced them to the knowledge of God through two actions: by saying “Thou shalt bow down to God and Him only shalt thou worship,” and by having made cherubim of molten gold which were overshadowing the seat of ex- piation, that is, bowing to Him. He led them up “to bow down to God the Lord and Him only to worship” by both sight and hearing.** 21 See M. T. Fégen, “Reanimation of Roman Law in the Ninth Century: Remarks on Reasons and Results,” and C. J. Wickham, “Ninth-Century Byzantium through Western Eyes,” both in Brubaker, ed., Byzantium in the Ninth Century (as above, note 19), 11-22 and 245-56. 22 MGH, Ep 5 (Berlin, 1899), 99-100; discussion in Wickham, “Byzantium through Western Eyes,’ 245-46. 3 TD). Ganz, “Theology and the Organisation of Thought,” in R. McKitterick, ed., New Cambridge Medieval History (Cambridge, 1995), 2:773-75; Freeman, “Carolingian Orthodoxy”; Freeman and Meyvaert, “Theo- dulf’s Apse Mosaic.” 4 Mansi 13, 285 a-s; trans. from D. J. Sahas, Icon and Logos. Sources in Eighth-Century Iconoclasm (Toronto, 1986), 110. LESLIE BRUBAKER 179 Byzantine authors of the eighth and ninth centuries returned again and again to this same selection from Exodus because they saw it as proof that, while the second com- mandment of Moses—“thou shalt not create graven images”—forbade idols, God had ap- proved, and ordered the production of, religious images for worship. The argument had emerged as part of the Christian defense of images directed against the Jews,”° and dur- ing the iconoclast debates (ca. 730-843) became the central proof text used by the icono- philes to demonstrate that the iconoclasts were wrong. Virtually every iconophile defense of images includes this argument. John of Damascus, writing around 750, provides a col- orful example: “On the one hand you say, ‘You shall not make for yourself a graven image, or any likeness’ and yet you yourself have cherubim woven on the veil and two cherubim fashioned of pure gold. But listen to what the answer of God’s servant Moses might be: ‘O blind and stupid people, listen to the force of these words, and guard your souls carefully. Yes, I said that .. . you should not make for yourselves molten gods [idols]... . I did not say, You shall not make images of cherubim’. . . . See how the purpose of scripture is made clear to those who search for it intelligently.’*° To the Byzantines, the Old Testament nar- rative that described God’s command to produce and decorate the ark and the tabernacle was the ultimate defense of Christian imagery. In the Opus Caroli Regis, Theodulf argued against this interpretation. He reasoned that the ark could not be used to justify religious images because God commanded Moses to have it made, and it thus differed fundamentally from human commissions.”’ Against the Orthodox belief that the ark supplied a rationale for Christian representation, ‘Theodulf instead saw the ark as a pale Old Testament prefiguration, now surpassed by the realities of the New.*® A decade later, Theodulf expressed his views in visual form by placing an im- age of the ark of the covenant in the mosaic above the altar in his oratory at Germigny-des- Prés (Fig. 1).2° The cherubim gesture toward an empty ark and, below that, the altar itself; and the hand of God in the center of the composition carries the stigmata of the risen Christ. Here the reality of the New Testament has replaced the symbolism of the Old; the future promised in the Old Testament, once prefigured by the contents of the now-empty ark, has been fulfilled by Christ, present at the altar in the form of the eucharist.*° The Byzantine visual response was quite different, as demonstrated by a miniature in the mid-ninth-century marginal psalter on Mount Athos, Pantokrator 61 (Fig. 2). Here the 25 §. Dufrenne, “Une illustration ‘historique’ inconnue du psautier du Mont-Athos, Pantokrator 61,” CahArch 15 (1965): 89; and esp. K. Corrigan, Visual Polemics in the Ninth-Century Byzantine Psalters (Cambridge, 1992), 34-35. 6 Against Those Who Attack Holy Images 2.9 (= 3.9): ed. B. Kotter, Die Schriften des Johannes von Damaskos, vol. 3, Contra Imaginum Calumniatores Orationes tres (Berlin, 1975), 96-97; trans. from D. Anderson, St. John of Dam- ascus, On the Divine Images, Three Apologies against Those Who Attack the Divine Images (Crestwood, N.Y., 1980), 56-57. 27 MGH, Conc 2, supp. 1 (Hannover, 1998), 175; trans. and discussion in Freeman and Meyvaert, “Theo- dulf’s Apse Mosaic,” 127. 8 See esp. MGH, Conc 2, supp. 1, 193; trans. and discussion in Freeman and Meyvaert, “Theodulf’s Apse Mosaic,” 131. *° "The extensive bibliography on this image is summarized and cited in Freeman and Meyvaert, “Theo- dulf’s Apse Mosaic.” °° For development of this interpretation, see Freeman and Meyvaert, “Theodulf’s Apse Mosaic,” 125-39. The concept of the eucharist as an image of Christ was also familiar in Byzantium (see, e.g., S. Gero, “The Eu- charistic Doctrine of the Byzantine Iconoclasts and Its Sources,” BZ 68 [1975]: 4-22), though it was not visu- alized in this way. 1380 CULTURAL AND MATERIAL INTERCHANGE miniaturist painted a gold tabernacle, surmounted by two gold cherubim that cant inward toward each other, within a blue-curtained court.*! The tabernacle image accompanies Psalm 113:12-15, which reads: “The idols of the nations are silver and gold; the work- manship of men’s hands. They have a mouth but cannot speak; they have eyes but they cannot see; they have ears but they cannot hear; they have noses but they cannot smell; they have hands but they cannot handle; they have feet but they cannot walk; they cannot speak through their throat. Let those that make them become like them, and all who trust in them.” These verses were cited by the iconoclasts as a biblical witness against religious images, but, as Suzy Dufrenne has shown, the miniature attacks this interpretation.* Three figures stand below the tabernacle and text: the iconoclast patriarch John the Grammarian, the psalmist David, and Beseleel, the artisan commanded by God to con- struct the tabernacle (Exodus 31:1-11). John represents the iconoclast reading of the ad- Jacent psalm text as a rejection of religious images. The miniaturist, however, has shown David, the author and most authoritative interpreter of the psalms, turning away from John in refutation of such iconoclast thinking.*® David turns instead toward Beseleel: through visual antithesis, the miniaturist reminded the viewer that God commanded the veneration of certain material manifestations of sanctity, the products of human hands. We are led to understand Psalm 113 as a condemnation of idols, not of all religious images.*4 A less sophisticated, but in some ways more interesting, Byzantine reaction appears in the ninth-century Christian Topography now in the Biblioteca Apostolica Vaticana (gr. 699).*° This is a large (332 x 337/342 mm), deluxe book that was clearly expensive to produce and was almost certainly made in Constantinople.*® Yet, in his Bibliotheke, written in the capital at about the same time as the manuscript was made, Photios condemned the text as “vul- gar, “implausible,” and “absurd.”*’ A sequence of five miniatures of the tabernacle and its accoutrements (Fig. 3) illustrated the central chapter of the volume and offer one justifi- cation for its sumptuous production.*® The Topography is the only nonbiblical Byzantine text to concentrate so extensively on the tabernacle; it is the only Byzantine text with miniatures devoted to single-minded representations of the tabernacle and its related paraphernalia.*? The significance of the ark and of the tabernacle in iconophile rhetoric suggests that the opportunity to depict them, in all of their manifestations, overrode the basic incompatibility of the Topography text itself with ninth-century thought. The virtually square format of the manuscript buttresses this suggestion. Julien Leroy calculated that >! For more on the relationship between the psalter and Topography images, see Dufrenne, “Une illustration inconnue,” and Corrigan, Viswal Polemics, 25. °? Dufrenne, “Une illustration inconnue,” esp. 86; see too Corrigan, Visual Polemics, 33-35. °3 Miniatures that visualize dispute in this way, and thereby sanction a particular interpretation of a psalm passage, are common in the marginal psalters: see Corrigan, Visual Polemics, esp. 111-13. 4 Dufrenne, “Une illustration inconnue”; followed by Corrigan, Visual Polemics, 33-35. °° C. Stornajolo, Le miniature della Topografia Cristiana di Cosma Indicopleuste. Codice vaticano greco 699, Codices e Vaticanis selecti 10 (Milan, 1908). °6 See L. Brubaker, Vision and Meaning in Ninth-Century Byzantium. Image as Exegesis in the Homilies of Gregory of Nazianzus, Cambridge Studies in Palaeography and Codicology 6 (Cambridge, 1999), 26. 3” Codex 36: ed. R. Henry, Photius, Bibliothéque (Paris, 1959), 1:21-22. °8 Four survive: Vat. gr. 699, fols. 46v (Fig. 3), 48r (tabernacle with cherubim), 49r (curtains), 108v (taber- nacle precinct). °° The Octateuchs contain a sequence of virtually identical images, but these were copied from the Topogra- phy: L. Brubaker, “The Tabernacle Miniatures of the Byzantine Octateuchs,” in Actes du XVe Congres Interna- tional d’Etudes Byzantines, vol. 2, Art et archéologie (Athens, 1981), 73-92. 4. <a Tee x Ee WS ; ue et a ce a i . : (P= wy Ay ¥. “ : J o; . 2 . a fd : = ‘ e : ae A : . i € Bien am - - y ; ny) Fy . 4 a. . et gS ; 4 > ; Oe é 4: . FP . ' Hye ' 1 Germigny-des-Prés, oratory of Theodulf, apse mosaic (photo: author) ei re ‘dphe “arout qlee : ¥ 0G 5 GE Tod SusTes ra Tye,” “wa pTay ot F wet OL BY Bn aren ee : ae aly aa . ; ¢ a. a ps > gh vty ee whos bac 2 Mount Athos, Pantokrator 61 (psalter), fol. 165r (photo: after S. Pelekanides et al., The Treasures of Mount Athos. Illuminated Manuscripts, vol. 3 [Athens, 1979]) % pero . é NBA Br Nate sad dw dde 3 Vatican City, Biblioteca Apostolica, gr. 699 (Christian Topography), fol. 46v (photo: Biblioteca Apostolica Vaticana) | | | meen o> =. 4 Paris, Bibliotheque Nationale, gr. 437 (pseudo-Dionysios the Areopagite), fol Paris, Bibliothéque Nationale de France) . 98r (photo: ‘ake: ptt “id aviereieni ied sie AOL TUK ieAyE arageneeet ri MAAITIAPEY TOMA SPP Tey Ter Tigehans fie oo, ihe 16at TETipecAy Ver nie (eT Karey ve [Rin Nareeearnte ANAITE REA +. PERDUE RAE TION, FIRE “ae ITEC yrrente MACACA CIE" EO TORRY EAP RET if iyi zeren ert TAIMICATALED ANAL W ANE Peiivapnearsios a WVudimervitintaucwn grr eH Hie ISALECWIN AW Pret as “an 5 Paris, Bibliothéque Nationale, gr. 510 (Gregory of Nazianzos), fol. 319v (photo: Paris, Bibliothéque Nationale de France) , Nationalbibliothek) ienna Vv ond Bs . ’ Tis eal fe me? } = = > S ee & cn vo = bay = o) Z 2) om <=) — = ep go) v . av, v as —_ i ib] Nationalb 6 Vienna, bet ce OS as ae 4 oth : 7 % 3%: ea > i wt * eet Fw Cae per. x > fed A Anime ite SP ee * 5 RG SPRL: ch oo) a 6] >| way BL. 7 Paris, Bibliothéque Nationale, gr. 510 (Gregory of Nazianzos), fol. 250v (photo: Paris, Bibliothéque Nationale de France) ips mw HISAEEWALOYVTIOL-FICA . oe UV AIAN 2 ey cyl 3 pore Ae pind virven sexs lcacarrrafiarraan. re TOMA AAIIOW T'S Wear” tars ae a MALCOM LION Werenries riraensdriren fuer. TAAEIS MOAI pI TG. WANTOTE RAEANG Vericap mirdricea ORITECTIRGEL Exar Arne writen MAINT PCA Aprise rune neg Be THEY C TA MIM MAICOCOLE. wu | Siar Ménere Bins erway 8 Paris, Bibliothéque Nationale, gr. 510 (Gregory of Nazianzos), fol. 86r (photo: Paris, Bibliotheque Nationale de France) ‘o | wei mon oy pannen Sat a SS poy . | Awoy OMCT: AALpoye ry »* ae “wrargaevres THC wists Ma ar WoliG HOMME AN recor uc HOA GpACAPP OAONG * | a | aie TSVANOP.w Linley AKSY ORtenwoureo Oye oe _" es | renoye. discon MONG ~ as Pe: Sy no a . se BAKO WG rigs wept \ ASR TIA S ticweors,: “eG, hed AC Ae’s ¥PASCGwc ‘TAY TAG | "AG Ener CApt Nd ou as ae “oSMIRNGp Haein Vonee Le TIMEG PANO G PERS gi eaginnrensitvagy”. CINCLU AEM CIN Gort Unedy nr Pay’ Waly ARS en Ob: ith MPHCs A | “pavrsyaganguante, > HGIpEtc GLEWorr an INA $e PN TACHI OGACAC spAnig Pen TES Eun | pane PAL ONY HA OPAMP MIC Parr Ercan Ue NOK 3 9 Vatican City, Biblioteca Apostolica, gr. 1666 (Gregory the Great), fol. 136v (photo: Biblioteca Apostolica Vaticana) Museo Sacra) 10 Silk panel, Nativity (Volbach T1104) (photo: Vatican City, 11 Silk panel, “Samson” (DO 34.1) (photo: Washington, D.C., Dumbarton Oaks) SF oe PE ~ we : a SAA TET oF) 12 Silk panel, “Sasanian” hunters (photo: Cologne, Didzesanmuseum) Aachen, Munster Treasury) 13 Silk panel, Charioteer (photo 14 Silk panel, Dioskouroi (drawing) (photo: after D. Buckton, ed., Byzantium. Treasures of Byzantine Art and Culture from British Collections [London, 1994}) LESLIE BRUBAKER 181 the proportion of height to width of a representative sampling of eighth- and ninth- century Greek books ranged from 1.33 to 1.55, as opposed to 1.03 for the Vatican Topog- raphy, a ratio more characteristic of late antiquity than of the ninth century.*® The Vatican manuscript presents an expanded version of a sixth-century text, and it is possible that its creators followed the format of the book they were adapting.*! The difference between old and new texts was, however, an important concern in the eighth and ninth centuries, a pe- riod that also produced manuscripts designed to look ancient.*? In such a milieu, the square layout of the Vatican Topography is best understood as a conscious retention of an archaic format. It indicated that the miniatures of the Christian Topography recorded the past: some of the ideas contained in the book might no longer be accepted, but the ark and the tabernacle, with their cherubim commissioned by God, were sanctioned by tradition. The Topography miniatures were meant to be seen as a record of the past, and that aspect of the past represented by the ark and tabernacle validated the present—anzd its use of sa- cred images—in ninth-century Byzantium. The understanding and visualization of the ark of the covenant developed by Theo- dulf and the Byzantines in response to exactly the same passage— Exodus interpreted by the Acts of the 787 Council of Nicaea—could hardly be more divergent. For both, the ark represented the past, but for Theodulfit was an outmoded relic that had been superseded by the advent of Christ, whereas for the Byzantines it was a representative of venerable tra- dition that authorized the present. To Theodulf, as indeed most western authors, it was necessary to “disregard what is visible so that we may contemplate what is invisible”; he be- lieved that “we do not seek truth through images and paintings; we who attain to that truth which is Christ do so through faith, hope and charity.’** This sentiment would have sounded familiar to Byzantine iconoclasts,** but the Orthodox belief, expressed by Theo- dulf’s contemporaries John of Damascus, Patriarch Nikephoros, and Theodore of Stou- dios, among others, was different.*° “We are led to the perception of God and his majesty by visible images,” John wrote, for “it is impossible for us to think immaterial things unless we can envision analogous shapes.’*° The distinct attitudes toward material reality re- vealed in the words written by both sides are mirrored in their practical applications. Theodulf’s apse mosaic is about material absence: the ark of the covenant is empty, and Christ is unseen except for his wounded hand. The Pantokrator psalter and the Christian 40 J. Leroy, “Note codicologique sur le Vat. gr. 699,” CahArch 23 (1974): 73-74. 41 See W. Wolska, La Topographie chrétienne de Cosmas Indicopleustés. Théologie et science au VIe siécle, Biblio- théque byzantine, Etudes 3 (Paris, 1962); and W. Wolska-Conus, ed., Cosmas Indicopleustés, Topographie chréti- enne, 3 vols., SC 141 (Paris, 1968), 159 (Paris, 1970), 197 (Paris, 1973). #2 See, e.g., PR van den Ven, “La patristique et ?hagiographie au concile de Nicée en 787,” Byzantion 25-27 (1955-57): 325-62; G. Bardy, “Faux et fraudes littéraires dans l’antiquité chrétienne,” RHE 32 (1936): 290- 91; L. Brubaker, “Byzantine Art in the Ninth Century: Theory, Practice, and Culture,” BMGS 13 (1989): 52- 53. *8 MGH, Conc 2, supp. 1, 175 and 193; trans. and discussion in Freeman and Meyvaert, “Theodulf’s Apse Mosaic,” 127 and 131. For Theodulf, see further A. Freeman, “Scripture and Images in the Libri Carolini,” in Testo e immagine nell'alto medioevo, Settimane 41 (Spoleto, 1994): 163-88. “4 See esp. M. V. Anastos, “The Ethical Theory of Images Formulated by the Iconoclasts in 754 and 815,” DOP 8 (1954): 151-60. 45 K. Parry, “Theodore Studites and the Patriarch Nicephoros on Image-Making as a Christian Imperative,’ Byzantion 59 (1989): 164-83. 46 Against Those Who Attack Holy Images 1.30-31 (= 2.26-27) and 3.21: ed. Kotter, Schriften, 3:144—45, 128; trans. with modifications from Anderson, John of Damascus, 34-35, 76. 182 CULTURAL AND MATERIAL INTERCHANGE Topography are about material presence: the visual narrative encompasses the ark, its ac- coutrements, and the people whose thoughts and beliefs are invested in it. These intrinsically different attitudes toward imagery form a package with diamet- rically opposed written and visual responses to the same passage from Exodus, a funda- mental component of the Old Testament cultural framework of Christianity East and West. Whatever shared court values floated on the surface of élite cultural interaction with the visual, they did not run deep. The Acts of the Council of 787 that inspired such different reactions to Exodus in the Latin West and the Greek East exemplify the ambivalent impact of texts traveling across the medieval Mediterranean. Like Constantine IX’s eleventh-century chrysobull, the vary- ing responses to the eighth-century Acts suggest that the transmission of texts raises par- ticular issues of cultural exchange: texts are often more fraught than most other élite products of exchange because language barriers made them incomprehensible outside of a restricted environment. Hence books were normally involved in relatively local exchange networks for the obvious reasons that language and (for service books) liturgical practice were regional rather than international. There was nonetheless still some traffic in books across the Mediterranean in the eighth and ninth centuries. Some traveled as the spoils of war: most notably, the Abbasid caliph Hartn al-Rashid took Greek manuscripts as booty when he captured Amorium and Ankara in the 790s.*’ Others were purchased, or went as gifts. Books came into Constantinople, as we shall see, but written documentation for text transfer favors other export channels. To give three examples: books, icons, and silks were carried from Rome to England by Benedict Biscop; gospel books went to Rome from Con- stantinople in 824, 855, and 857/8 as gifts from the Byzantine emperors Michael II and III to Popes Paschal I, Leo IV, and Benedict III; and, in the 830s, books went from Constan- tinople to Baghdad, gifts from Emperor Theophilos to the caliph al-Ma mtn.* The cultural impact of books given at this level is not always easy to determine, and, when we can trace it at all, the results are often unexpected. They inevitably reveal the complexities and ambiguities of cross-cultural interchange; and, rather than highlighting shared values of élite court interaction, the exchange of books elicited inconsistent and random responses. The copy of pseudo-Dionysios the Areopagite (Paris, Bibl. Nationale, gr. 437) sent in 827 (along with ten pieces of silk) from Constantinople by the Byzantine emperor Michael IT as a gift to Louis the Pious provides a well-known example.* The vol- ume can only have been valued as an erudite text, for there is little decoration (Fig. 4). It was nonetheless a carefully chosen gift, for Louis and his advisor Hilduin believed that *” See L. Goodman, “The Translation of Greek Materials into Arabic,” in M. J. L. Young, J. D. Latham, and R. B. Serjeant, eds., Religion, Learning and Science in the Abbasid Period (Cambridge, 1990), 477-97, at 482. *® For Biscop, see Bede, Historia abbatum 9.1.373: ed. C. Plummer, Venerabilis Baedae Historia ecclestastica gentis Anglorum, vol. 1 (Oxford, 1896), 364-87. For books sent to the popes, see G. Haendler, “Der byzantinische Bilder- streit und das Abendland 815-828,” in H. Képstein and F Winckelmann, eds., Studien zum 8. und 9. Jahrhundert in Byzanz (Berlin, 1983), 159-62; and the Liber Pontificalis: ed. L. Duchesne, Le Liber Pontificalis: Texte, introduction et commentaire (Paris, 1955), 2:147; trans. R. Davis, The Lives of the Ninth-Century Popes (Liber Pontificalis), Translated Texts for Historians 20 (Liverpool, 1995), 186 (to Benedict III). For books sent to al-Ma’'min, see Goodman, “The ‘Translation of Greek Materials into Arabic,” 482-85; D. Gutas, Greek Thought, Arabic Culture (London, 1998), 77-103. * H. Omont, “Manuscrit des oeuvres de S. Denys l’Aréopagite envoyé de Constantinople 4 Louis le Débon- naire en 827,” REG 17 (1904): 230-36; L. Perria, “Palaeographica, I: In margine alla tradizione manoscritta dello Pseudo-Dionigi lAreopagita; II: Minuscole librarie fra IX e X secolo,’ RSBN n.s. 37 (2000): 43-72. LESLIE BRUBAKER 183 Dionysios was the patron saint of the imperial monastery of St. Denis, where Hilduin was abbot. Its arrival set off a chain of reactions. Three generations of Carolingian church- men—Hilduin, John Scottus thirty years later, and Anastasius Bibliothecarius in 875— translated the work into Latin: each belittled previous scholarship and used the transla- tion as a springboard to promote his own intellectual superiority.°° Understanding Greek was, in this instance, an important signal of status, and the imported works of pseudo- Dionysios allowed that status to be displayed. The manuscript itself, however, had no dis- cernible impact on Carolingian book production. In that arena, the chain of influence in fact runs the opposite direction, from West to East. Charlemagne’s elephant, Constantine’s chrysobull, Theodulf’s ark, and Louis’ pseudo- Dionysios expose the ambiguities and reinterpretations inherent in cultural exchange. When we can trace the direct impact of an import on local material production, these ten- sions do not disappear: the adaptations continue to envelop the import within the preoc- cupations of the receivers. The integration of imported techniques or formulae into local work nonetheless shows that cultural exchange had an impact not only on ideology but also on practice. One example is book ornament. The pseudo-Dionysios manuscript may be used to exemplify the characteristic early ninth-century Byzantine manuscript. As already noted, its ornament is restricted. Crosses that accompany chapter headings are drawn in red ink; there are some rudimentary ink division bars, a scattering of enlarged initials, in red or brown ink, and a handful of let- ters—always in the bottom line of text—with elongated base serifs (Fig. 4). The great ma- jority of ninth-century Byzantine books follow this minimalist model, both those written in majuscule and those using the new, faster, and cheaper minuscule.°' Three manuscripts from the Stoudite monasteries dated by colophon across the ninth century—the Uspen- skij Gospel of 835, the Moscow Basil of 880, and the Glasgow Basil of 899—exemplify this tendency: all follow the pattern established by the Paris pseudo-Dionysios.”* When ornament appears in ninth-century Greek books, it is a significant indicator of the impact of trans-Mediterranean exchange. Despite the lack of elaborate decoration in most ninth-century Byzantine manuscripts, we can trace the effects of this exchange in a handful of notable exceptions, two of them particularly relevant to this discussion. One is the famous manuscript of the Homilies of Gregory of Nazianzos in Paris (Bibl. Nationale, gr. 510), which was written and decorated in Constantinople between 879 and 882, and which has more than sixteen hundred painted or gilded initials (an average of two per side), along with headpieces and full-page miniatures (Fig. 5).°° The decorated initials °° For details, see Wickham, “Byzantium through Western Eyes,” 248-49. 5! See L. Brubaker, “Greek Manuscript Decoration in the Ninth and Tenth Centuries: Rethinking Centre and Periphery,’ in G. Prato, ed., I manoscritts greci tra riflessione e dibattito, Atti del V Colloquio Internazionale di Paleografia Greca (Florence, 2000), 2:513-33. 2 Petropol. gr. 219 (apparently the oldest dated minuscule manuscript); Moscow, GIM 117; Glasgow, University Library, Hunter.V.3.5: see B. Fonki¢é, “Scriptoria bizantini. Risultati e prospettive della ricerca,” RSBN n.s. 17-19 (1980-82): 73-116, at 84-85; N. Kavrus, “Studysky skriptory v IX v. (po materialam rukopise}j Moskvy i Leningrada,” VizVrem 44 (1983): 98-111, at 99-104; L. Perria, “Scrittura e ornamentazione nei mano- scritti di originie studita,” BollGrott n.s. 47 (1993): 245-60, at 248-53. 3 L. Brubaker, “The Introduction of Painted Initials in Byzantium,” Scriptortum 45 (1991): 22-46. Since that article was written, I have had the opportunity to examine the Paris Gregory in detail, and would like to thank Dr. Christian Forstel, curator of Greek manuscripts at the Bibliothéque Nationale de France, for his generos- ity and many kindnesses. 184 CULTURAL AND MATERIAL INTERCHANGE were a direct response to western imports, and provide real visual evidence of the impact of the exchange of élite products, in this case illuminated books, across the Mediter- ranean.”* Earlier Byzantine manuscripts had, as seen in the pseudo-Dionysios, enlarged initials and, occasionally, very limited decoration, always in the ink of the text. The upsilon in the lower line of text on fol. 10v of the Dioskourides in Vienna (Nationalbibliothek, med. gr. 1) of ca. 512 (Fig. 6) is the most elaborately decorated letter preserved in a Greek manu- script before the ninth century.°° The radical departure represented by the Paris Gregory requires explanation, and three factors argue that western imports are responsible for the change. First, decorated and painted initial letters appear in Latin manuscripts virtually without a break from late antiquity onward. When the “official” language of the East Ro- man Empire switched from Latin to Greek in the sixth century, initials continued in Latin books produced in the western half of the empire, and their use increased over time; but they failed to appear in Greek books until the ninth century. Only then, apparently, did western illuminators spur eastern imitators. Second, the location of the painted initials in the Paris Gregory follows Latin practice. When enlarged letters appeared in Greek texts, they were virtually always placed at the beginning of a naturally occurring line of text. Hence, to signal the start of a significant passage the scribe enlarged the first letter of the next line of text, which could fall any- where in the sentence. This is not how it works in the Paris Gregory where, with only rare exceptions, the first letter of the passage to be signaled is enlarged and painted, and the passage itself begins a new line of text so that the enlarged initial can expand into the mar- gin of the page. The significant passage begins flush with the left margin, leaving empty space at the end of the previous line (Fig. 7).°° This is the way we write, but it is not how previous or later Byzantine scribes organized their texts: the Paris Gregory is almost unique among Greek books in using this system. The Gregory illuminators did not, how- ever, invent the formula; they borrowed it from Latin manuscripts, where it had a long history. The third factor that argues for western impact is the selection of motifs attached to the painted letters. Some of these were straightforward, ninth-century Constantinopoli- tan favorites; others were long-lived motifs that can be traced from late antiquity, or ear- lier, straight through to the ninth century (though not in the medium of manuscript illu- mination).°’ Others, however, are new to Byzantium, but appeared earlier in the West. There are, for example, twenty-one fish-shaped omicrons in the Paris Gregory (Fig. 8), and they have many precedents in sixth- and seventh-century manuscripts produced in the West, especially in Italy.°* Similarly, the epsilon with its crossbar formed of a blessing hand 4 For similar arguments about cloisonné enamel, see D. Buckton, “Byzantine Enamel and the West,” in J. Howard-Johnston, ed., Byzantium and the West ca. 850-ca. 1200, Byzantinische Forschungen 13 (1988): 235-44; idem, “Enamels” [in Early Christian & Byzantine Art], Dictionary of Art (London, 1996), 9:659-63. °° C. Nordenfalk, Die spdtantike Zierbuchstaben (Stockholm, 1970), pls. 29-31; Brubaker, “Introduction of Painted Initials,“ 22-23. °® Occasionally, and only toward the end of the manuscript, small painted initials actually appear within a line of text. This exceptional practice underscores the idiosyncrasy of the Paris initials. 57 Brubaker, “Introduction of Painted Initials,” 28-36. °8 Ibid., 38-39. LESLIE BRUBAKER 185 first appears in Paris. gr. 510, but E’s with hand crossbars appeared in Latin manuscripts from the seventh century onward.*° Practice, usage, and form argue that the initials in the Paris Gregory follow western precedents. In this context, it is significant that the earliest book written in Greek with painted initials is not “Byzantine” at all, but was produced in Rome in the year 800. This manuscript, a Greek translation of the Dialogues of Pope Gregory the Great (Biblioteca Apostolica Vaticana, gr. 1666), was clearly directly inspired by western manuscript illumi- nation—the interlace forms and animal-head terminals (Fig. 9) find particularly close par- allels in Lombard and Merovingian books—and shows how Latin decorated initials could be translated into Greek letters.® Italy, with its strong Greek and Latin cultural compo- nents, was the obvious place for this to happen, and indeed ninth-century Rome, where the Dialogues was made, was the crucible.®! There were, of course, Greek monasteries in Rome in the eighth and ninth centuries; and, as Guglielmo Cavallo has shown, it is in late eighth- and ninth-century manuscripts written in these monasteries that we first find scribes writing in Greek but grafting in Latin letters.® The intrusion of Latin decorative forms seen in the Vatican Dialogues also began a trend, charted by John Osborne, of Greek manuscripts written in Rome with ornamental initials.’ The combined evidence of the Paris Gregory and the Vatican Dialogues leads to the almost inescapable conclusion that Latin initial forms reached the Byzantine East through the intermediary of Italo-Greek manuscripts. When this happened is relatively easy to chart. Table 1 shows all dated Greek books produced in the ninth century that contain decoration beyond the most rudimentary scribal enlarged letters or division bars. After the Dialogues (Vat. gr. 1666), the next manu- script is dated 861/2: all of the Greek books with decoration produced in the eastern Mediterranean postdate 850. We may reasonably conclude that books with decorated ini- tial letters traveled east across the Mediterranean, from Rome to Constantinople, to- ward the middle of the ninth century. The Sinai manuscript (row 2), a lectionary perhaps produced in Palestine,** and Meteora 591 (row 3), a manuscript of the sermons of John °° Thid., 36-38. The position of the hand, palm facing outward, differs from the position of the Hand of God (palm facing inward) in the miniatures, and is one indication that different teams were responsible for each, with different sources of inspiration. °° For a good recent discussion, with full bibliography, see J. Osborne, “The Use of Painted Initials by Greek and Latin Scriptoria in Carolingian Rome,” Gesta 29.1 (1990): 76-85, at 77-80. °! Ibid.; G. Cavallo, “Interazione tra scrittura greca e scrittura latina a Roma tra VIII e IX secolo,” in P Cock- shaw, M.-C. Garand, and P. Jodogne, eds., Miscellanea Codicologica F Masai Dicata (Ghent, 1979), 1:23—29; and in general A. Guillou, “Rome, centre transit des produits de luxe d’orient au haut Moyen Age,” Zograf 10 (1979): 17-21. °? Guillou, “Rome,” and more generally M. Gigante, “La civilta letteraria,” in G. Cavallo et al., eds., J Bizan- tint in Italia (Milan, 1982), 616-18. On Greek monasteries in Rome, see J.-M. Sansterre, Les moines grecs et ori- entaux a Rome aux époques byzantine et carolingienne (milieu du VIe stécle-fin du [Xe siécle) (Brussels, 1983). °° Osborne, “The Use of Painted Initials in Carolingian Rome.” °* Sinai. gr. 210 and NE Meg. Perg. 12: L. Politis, “Nouveaux manuscrits grecs découverts au Mont Sinai, rapport préliminaire,” Scriptorium 34 (1980): 5-17, at 10-11; D. Harlfinger, D. R. Reinsch, and J. A. M. Sonder- kamp, Specimina Sinaitica. Die datierten griechischen Handschriften des Katharinen-Klosters auf dem Berge Sinai, 9. bis 12. Jahrhundert (Berlin, 1983), 13-14; P. Nicolopoulos, ed., Holy Monastery and Archdiocese of Sinai. The New Finds (Athens, 1999), 144. sjieoy X xX sodins xX xX xX [Ig 2eWwI09s ~— sade pure soqojiy x syeumrue SI[-9p-Inopy xX Jd] 19} UI xX xX SIIPUZISOP Jes] gynosnurw = wi ajnosnfew = pw 668 L68 068 68/6L8 $/698 6/198 O08 NOILVYOOAC HLIM AYNLNAT) HLNIN] THE fO SLdTHOSONVIL Wada) GALVG *[ AIAV EL Ww vsi WI) ‘Moosopy pp is ‘jed 1e, QLPT pure OLFT 18 ‘sted OTG 18 “st1eg 16G P1I09I9/ 01g 18 pue GI ‘Bldg “SIN AN 18 “TeUuls gQQ| IS ‘1eA LESLIE BRUBAKER 187 Chrysostom produced in Bithynia,® are both decorated only with ink rather than paint, and neither includes ornament beyond headpieces and introductory initials. The earliest securely dated East Roman manuscript to follow the lead of the Vatican Dialogues in its use of painted letters is the Paris Gregory (row 4) from Constantinople. Two Constantinopoli- tan manuscripts that probably slightly predate Paris. gr. 510, the Khludov Psalter (Moscow, Historical Museum, gr. 129) and the Sacra Parallela (Paris, Bibl. Nationale, gr. 923), include a scattering of simple hollow-bar initials painted in single colors, and some in the latter book have two-strand interlace fill:®° these may be the first surviving responses to cross- Mediterranean manuscript exchange, but neither is securely dated. The Paris Gregory, distinguished both by the number of painted letters and by its range of imported motifs, remains the most compelling witness to the phenomenon. It did not, however, spawn many imitators: only the Leo Bible (Vat. Reg. Gr. 1) of ca. 920 and the Paris Psalter (Paris. gr. 139) of ca. 960 contain initials that significantly re- semble those in the Paris Gregory, and the heyday of Byzantine painted initials awaited the eleventh and twelfth centuries.*’ After ca. 880, many Byzantine manuscripts incorporated painted initial letters, but without the precise Latin analogues found in Paris. gr. 510. While the indirect effect of Latin books suggested the inclusion of painted letters and, in the end, changed the face of Byzantine manuscripts, direct western impact was a short- lived phenomenon; it seems to have been confined to court circles and was restricted toa trio of manuscripts that are closely interconnected in other ways as well.®* While this might seem, for once, to reveal international shared court values, the source of the Latin impact was neither the Carolingian nor the papal court, but rather, as we have seen, Greek monas- tic scriptoria in Rome. It was only in the tenth century that motifs associated with earlier Carolingian court manuscripts began to influence Byzantine illuminators, and even then it was a brief phenomenon, limited to a handful of manuscripts.®° The Byzantines were also selective about which motifs they borrowed, and how these were displayed. Comparison of the Roman Dialogues manuscript of ca. 800 (Fig. 9) with the far more restrained use of painted letters in the Constantinopolitan Paris Gregory of ca. 880 (Figs. 5, 7, 8) reveals essential differences. In the former, the initial displaces text and sprawls across the page; in the latter, the initial forms part of a hierarchy of ornament that is carefully used to clarify the structure of the text. This hierarchy was developed in °° I. Hutter, “Scriptoria in Bithynia,” in C. Mango and G. Dagron, eds., Constantinople and Its Hinterland (Aldershot, 1995), 379-96, at 381-82; Brubaker, “Greek Manuscript Decoration,” 516-23. °° For the date of the Khludov Psalter, see Corrigan, Visual Polemics, 124-34; I am indebted for information about its painted initials to Maria Evangelatou. For the date and decoration of the Sacra Parallela, see Brubaker, “Greek Manuscript Decoration,” 524 (with earlier bibliography). °7 See P. Canart and S. Dufrenne, “Le Vaticanus Reginensis graecus | ou la province 4 Constantinople,” in G. Cavallo, G. de Gregorio, and M. Maniaci, eds., Scritture, libri e testi nelle aree provincialt di Bisanzto, (Spoleto, 1991), 2:631-36; K. Weitzmann, Die byzantinische Buchmalerei des 9. und 10. Jahrhunderts (Berlin, 1935), figs. 4’7— 48, 278-82. I thank Monsignor Canart at the Vatican and, again, Dr. Forstel in Paris for allowing me to study the initials in both manuscripts, most of which remain unpublished. °° "The Paris Gregory was made for Basil I, on which see Brubaker, Vision and Meaning, 158-62, with com- ments on the relationship among the three books at 397. On Leo sakellarios for whom Vat. Reg. gr. 1 was made, see C. Mango, “The Date of cod. Vat. Regin. Gr. 1 and the ‘Macedonian Renaissance’,” Acta [IRNorv 4 (1969): 121-26; for the Paris Psalter and Constantine VII Porphyrogennetos, see H. Buchthal, “The Exaltation of David,” ]Warb 37 (1974): 330-33. °° The so-called blue-gold group: Weitzmann, Byzantinische Buchmalerei, 7-8. 188 CULTURAL AND MATERIAL INTERCHANGE the ninth century,’”° and the enlarged and decorated initial played into the new system: in eastern manuscripts, unlike western ones, the initial is always secondary to the script, and is of less ornamental importance than the headpieces that are used to separate discrete sec- tions of the text. However patchy the influence, the introduction of painted initials to Byzantium is not our only evidence for shared scribal or textual interests across the Mediterranean. The Vatican Dialogues was, as noted above, a Greek version of a Latin text, and returns to trans- lation, a level of cultural exchange already noted in connection with the pseudo-Dionysios sent to Louis the Pious. Another well-known Greek example survives in two volumes now in Paris (Bibl. Nationale, gr. 1470 and 1476), dated by colophon to 890. These contain a series of martyr stories, some of which are copies of texts written by the future patriarch Methodios when he was staying in a Greek monastery in Rome between 815 and 821, and some of which are translations of Latin martyr texts.”! The urge to translate was not confined to Christian centers. At more or less the same time as Gregory the Great’s Dialogues were translated from Latin to Greek, Harun al- Rashid began the great Abbasid project of translating Greek manuscripts into Arabic.” This endeavor escalated under Hartn’s son, al-Ma min, who opened an official translating institution in Baghdad called the House of Wisdom (Bayt al-Hikmah) in 830.” In the years on either side of 800, in other words, there was a pan-Mediterranean awareness of the written artifacts of other cultures. This awareness extended to certain tools of the trade, for, like manuscripts, papyrus traveled in the ninth century, though it moved along a north-south axis rather than an east-west one. Papyrus was used for certain legal texts in Byzantium, and continued to be shipped from Egypt at least intermittently after the Arab conquest. Justinianic law re- quired that it be stamped with a protocol, and while Egypt was a Byzantine province this had named the emperor and invoked Christ. This changed after the conquest, and by the second half of the ninth century, papyrus made for Abbasid use named the caliph and in- voked the prophet; papyrus made specifically for export, however, instead invoked the ‘Trinity.”* The transmission of initials from west to east, the importance of translations across the Mediterranean, and the willingness to modify tools of the scribe to suit the export trade show the strength of cultural interchange in word production in the ninth century. Though texts are not obviously the most explicit form of cultural exchange—and in other periods did not, in fact, participate very much at all in the system—their use in the ninth century is significant. This use extended to the courts, as witnessed by Michael II’s gift of the works of pseudo-Dionysios to Louis the Pious, the translation project instituted by Harun al-Rashid, and on a more pragmatic level the stamps required by imperial law on 7° See further Hutter, “Scriptoria in Bithynia.” 71 P Canart, “Le patriarche Méthode de Constantinople copiste a Rome,” Palaeographica Diplomatica et Archi- vistica. Studi in onore di Giulio Battellt (Rome, 1979), 1:343—53; the extensive bibliography on the manuscript is cited in Brubaker, “Greek Manuscript Decoration,” 518 n. 22; see also Table 1. 72 Goodman, “The Translation of Greek Materials into Arabic,” 478-79; Gutas, Greek Thought, 55-56. 73 Goodman, “The Translation of Greek Materials into Arabic,” 484; Gutas, Greek Thought, 75-83. 74 R. Lopez, “Mohammed and Charlemagne: A Revision,” Speculum 18 (1943): 14-38, at 23. On the distri- bution of papyrus during the first millennium, see further N. Lewis, Papyrus in Classical Antiquity (Oxford, 1974), 84-94; and McCormick, Origins of the European Economy, 704-8. LESLIE BRUBAKER 189 papyrus imported into Byzantium. But it was not restricted to the artisans and scribes working under the kings, popes, emperors, and caliphs: parallel and interlocking inter- change came through monasteries, especially those of Rome. Trade in words and their or- nament was not dependent on a shared aesthetic of élite culture. As always, whether an im- port had an impact on local production or whether it did not was entirely dependent on the needs of the receivers, not on the wishes of the exporter; and, as with the materializa- tion of cultural exchange, exports were adapted to suit the needs of their importers. ‘Turning from texts to textiles, we find many of the same patterns appearing. New uses for imported silks, and unintended and ambivalent reactions to them, remain constant re- sponses. Because fabric was a more visible part of daily cultural life, however, imported materials could also play a role in the articulation of political claims and social tension that texts rarely manage. This is nicely demonstrated by reports of the dress of the West Frank- ish king Charles the Bald (840-877) in contemporary sources. The Annals of St.-Bertin, which detail aspects of the history of West Francia between 830 and 882, had a series of compilers some of whom were sufficiently interested in tex- tiles to note their use. The compiler for 875, Hincmar of Reims, tells us that Charles the Bald went to Rome in 875, was crowned emperor by the pope on Christmas Day, and was back in Francia by Easter of 876; then, on 20 June 876, Charles “in a gilded robe and clad in Frankish costume,” opened the Synod of Ponthion, with “the whole interior of the build- ing and the seats . . . covered in fine cloths.”” On 16 July, the final morning of the synod, “the emperor [Charles] entered, clad in Greek fashion and wearing his crown, led by the papal legates clad in the Roman fashion.”’® This fairly neutral account establishes three modes of costume: Frankish, Greek, and Roman. The first of these recalls descriptions of Charlemagne, “who rejected foreign clothes” and “normally wore the customary attire of the Franks”: for daily wear, a linen shirt under a silk-fringed tunic with an otter or ermine vest in the winter; for high feasts, “clothes weaved with gold, bejewelled shoes... [and] a golden, gem-encrusted crown.’”’ The second refers to Byzantine imperial dress, appar- ently assumed by Charles the Bald for special occasions after he had been crowned as em- peror by Pope John VIII;” and the third—Roman costume—normally indicates Byzan- tine clothing as well but may in this case signal some form of papal insignia. The Annals of Fulda cover much the same period (838-901), but from the perspective of East Francia, where feelings ran against Charles the Bald. Its account of his clothing in 876 is somewhat different from that in the St.-Bertin annals: “King Charles returned from Italy to Gaul and is said to have adopted new and unaccustomed modes of dress: for he used to go to church on Sundays and feast days dressed in a dalmatic down to his ankles and with a sword-belt girdled over it, his head wrapped in a silk veil with a diadem on top. For, despising all the customs of the Frankish kings, he held the glories of the Greeks to be the best, and so that he might show his overweeningness more fully he put aside the ’”° MGH, ScripiRerGerm, 5 (Hannover, 1883), s.a. 876; J. L. Nelson, The Annals of St-Bertin, Ninth-Century His- tories, vol. 1 (Manchester, 1991), 189-90. 7° MGH, ScriptRerGerm 5, s.a. 876; Nelson, Annals of St-Bertin, 194. ” Einhard, Vita Caroli 23: MGH, ScriptRerGerm 25; trans. Dutton, Charlemagne’s Courtier, 31. ’8 His spectacles too were staged “in the style of Roman [Byzantine] emperors”: see J. L. Nelson, Charles the Bald (London, 1992), 17; for further examples, ibid., 242-43. 190 CULTURAL AND MATERIAL INTERCHANGE name of king and ordered that he should be called emperor and augustus of all kings on this side of the sea.”’? As in the St.-Bertin account, the Fulda annals describe Charles’s Greek regalia, but, unlike the former, here Greek attire counts against the emperor. In- deed, throughout the Fulda annals, Charles is described in pejorative terms normally re- served by the Franks for the Byzantines themselves. In the accounts of 875 and 876 alone, he is described as a tyrant, crooked, fearful, greedy, goaded by avarice, treacherous, cow- ardly, and as proceeding with his customary trickery—all terms also applied by Carolin- gian authors to the Byzantines.*° Just as ninth-century Byzantine historians tarred their antagonists by comparing them with the Saracens,®' so the East Frankish annalist smeared Charles by comparing him with the Greeks. What is interesting, in this context, is that re- porting an imported clothing style was considered an appropriate vehicle for conveying this disapproval.*° Imported Byzantine textiles, normally silk, were prized in western Europe. If we as- sume that Charles’s garments were made from imported fabric—and the veil at least must have been, for silk was not produced in Francia—the purpose to which these had been put overrode the value of the material in the mind of the author of the Annals of Fulda. The foreignness of the import could rebound two ways: it could convey prestige and exclusiv- ity, but it could also, as here, be used to condemn by analogy. Both require familiarity, and for silk—as for texts—that was no problem: though sericulture was not practiced in the Christian West before the tenth century, silk was imported from Byzantium, the Islamic East, or, from the eighth century onward, Islamic Spain. And, as with Egyptian papyrus, silks produced for exchange and export were not always identical to those kept at home. Great quantities of eighth- and ninth-century silks remain in the West, and many pre- served patterns find documentary parallels in the Roman Book of the Popes, the Liber Pon- tificalis. The great majority of these silks are eastern—only one group is possibly Span- ish*°—and many are specified as Byzantine. Before 772, silk is mentioned in the Liber Pontificalis only three times. Then, under Hadrian I (772-795), the number of silks re- ported picks up dramatically: more than one thousand are recorded as gifts to various churches in and around Rome.* Of these, however, only two are representational.*®° One 2 MGH, ScriptRerGerm 7 (Hannover, 1891), s.a. 876; T: Reuter, The Annals of Fulda, Ninth-Century Histories (Manchester, 1992), 2:79. 8° MGH, ScripRerGerm 7, s.a. 875-876; Reuter, Annals of Fulda, 76-82. On Carolingian terms for the Byzan- tines, see further Wickham, “Ninth-Century Byzantium through Western Eyes.” *! E.g., Theophanes on “the Saracen-minded Leo [III]”: Theophanis Chronographia, ed. C. de Boor (Leipzig, 1883-85), 405; C. Mango and R. Scott, eds., The Chronicle of Theophanes Confessor, Byzantine and Near Eastern His- tory AD 284-813 (Oxford, 1997), 560. *? This pattern is also found in late antique sources: see M. Harlow, “Clothes Maketh the Man,” in L. Brubaker and J. Smith, eds., Gender in Early Medieval Societies 300-900 (Cambridge, 2004), 44-69. *5 “Fourteen Spanish veils with silver” are recorded in the Liber Pontificalis as gifts from Pope Gregory IV to St. Mark’s in 829/31 but the fabric type is not specified: ed. Duchesne 2: 75; trans. Davis, Lives of the Ninth- Century Popes, 54. On the early silk industry in Islamic Spain, see O. R. Constable, Trade and Traders in Muslim Spain: The Commercial Realignment of the Iberian Peninsula, 900-1500 (Cambridge, 1994), 173-81, esp. 177-78; and in general F. May, Silk Textiles of Spain, Eighth to Fifteenth Century (New York, 1957). 84 See McCormick, Origins, table 24.1. 85 One showed Christ’s Passion and Resurrection; the other the Annunciation (identified as cheretismon, a Latin transliteration of its Greek name), the Nativity, and the Presentation. Though the latter is not identified as silk, the use of the Greek title suggests that it was a Byzantine work. Both were donated in 793/4. Duchesne 2:2; Davis, Lives of the Ninth-Century Popes, 180-81. LESLIE BRUBAKER 19] of these showed the Nativity, and the well-known Byzantine silk still preserved in the Vat- ican, dated to around the year 800, suggests how this may have looked (Fig. 10).°° Under the next pope, Leo III (795-816), there are more than seven hundred donations of silks recorded; and nearly three dozen of them are described as portraying animals or scenes from the life of Christ.8” One, for example, showed “wheels” of silk (presumably medal- lions, as seen on the Vatican Nativity silk) depicting the Annunciation, Nativity, Passion, and Resurrection. The Nativity fragment is still matched by another that shows the An- nunciation, and the pair presumably resembles the multiscene silks described in the Liber Pontificalis. These particular silks are of extremely high quality, and, because they are among the few silks to have been chemically analyzed, we can identify the dyes used, which include the very expensive kermes (cochineal red) and murex-purple, the use of which was restricted to the imperial workshops in Constantinople.* After Leo’s death in 816, and on up through the middle of the ninth century, the num- ber of representational silks drops off again markedly. The great majority of figured silks recorded in the Liber Pontificalis before ca. 850 thus date to the papacy of Leo III; nearly three-quarters of them appear in the donation lists of 798-800 and 812-814.°° While the compilers of the Liber Pontificalis had distinct reportorial styles, and some may simply have ignored papal gifts of silk, the pattern is nonetheless striking and appears to reflect large acquisitions of eastern silks in the years immediately preceding 798 and 812. The figured silks, with Christian subject matter, almost certainly came from Byzantium, and the two dates in question fall into the period when Iconoclasm (ca. 730-787, 815-843) was tem- porarily suspended. We know from other sources that the silk industry in Constantinople continued during Iconoclasm,” and the evidence of the Liber Pontificalis suggests that soon after the temporary suspension of the image ban in 787 old patterns were revived (or new ones were developed) to satisfy a demand for silks like the Vatican Nativity, with religious subject matter. Whatever the case, the silks still preserved in the Vatican collections, and others spread across Europe, conclusively demonstrate that Byzantine textiles were ex- ported to the West throughout the eighth and ninth centuries.”! Outside Rome, too, there was a great demand for silk textiles—or, at least, consider- able care was taken to record transactions in prestige silk. In 1935 Etienne Sabbe cata- logued hundreds of references to eastern and Spanish silk in ninth- and tenth-century documents from western Europe. Byzantine silk, for example, was sent by Emperor Con- stantine V to the Frankish king Pepin in 757 (Pepin in turn donated it to the church at 86 M. Martiniani-Reber, “Nouveau regard sur les soieries de l’annonciation et de la nativité du Sancta Sanc- torum,” CIETA 63-64 (1986): 12-21; A. Muthesius, Byzantine Silk Weaving AD 400 to AD 1200 (Vienna, 1997), 67, 175; A. Starensier, “An Art Historical Study of the Byzantine Silk Industry” (Ph.D. diss., Columbia Uni- versity, 1982), 571-76. 87 Listed in tabular form in L. Brubaker and J. Haldon, Byzantium in the Iconoclast Eva (ca. 680-850): The Sources, Birmingham Byzantine and Ottoman Monographs 7 (Aldershot, 2001), 104-7; see also McCormick, Origins, 719-28, who independently made some of the same points. 88 References in note 86 above. 89 See Brubaker and Haldon, Byzantium in the Iconoclast Era, 108. °° Tbid., 81-82. °1 FE. Sabbe, “Limportation des tissus orientaux en Europe occidentale au haut moyen age (IXe et Xe siécles),” RBPH 14 (1935): 811-48, 1261-88; D. Jacoby, “Silk Crosses the Mediterranean,” in G. Airaldi, ed., Le vie del Mediterraneo. Idee, uomini, oggetti (secoli XI-XVI) (Genoa, 1997), 55-79; repr. in D. Jacoby, Byzantium, Latin Romania and the Mediterranean (Aldershot, 2001), no. 10. 192 CULTURAL AND MATERIAL INTERCHANGE Mozac in 764 for use as a relic shroud); Theodulf of Orléans acquired Arab silk at Arles and used it as interleaf fill in his precious Bible, written on murex-purple stained parch- ment, where it is still preserved;*? Louis the Pious gave silk from Spain (stragulum hispan- icum) to the abbey of St.-Wandrille in Fontanelle sometime between 823 and 833." Pieces of the so-called Samson silk (Fig. 11) were found at seven sites in Europe, four of them so closely associated with ninth-century manuscripts or relic translations that the fabric can be dated to the years around 800 with some assurance.” Silk also traveled between Spain and the East, between the eastern courts of Constan- tinople and Baghdad, and from both of those courts outward in diplomatic circles. It is of- ten, in fact, difficult to tell where a piece of silk came from: some examples, such as the so- called Sasanian hunters of ca. 800 that was found in the reliquary of St.-Calais when it was opened in 1943 (Fig. 12), “look” ‘Abbasid, but technical analysis assures an origin in Byzan- tium.°*° Other silks, seemingly identical, are distinguished only by a variation in their bor- der ornament: crosses (evidently for Christian use) or invocations to Allah (evidently for Islamic ownership).°’ Whether these were made in one place but for different audiences, or whether cartoons for the patterns were carried from one shop to another, is unclear. Both practices find parallels: cartoons for weaving patterns still survive from Egypt,” and, like the papyri discussed earlier, ceremonial cloths manufactured in, especially, the fabric factories of Marw, in northern Persia, produced fabrics (usually a combination of cotton and silk) embroidered with an Islamic inscription and caliphal name if they were destined for local use, and with an invocation to the Trinity if they were to be exported to the Chris- tian world.® In any event, contrary to what Byzantinists sometimes assume, silk was sold and traded widely outside imperial circles.'°° Exchange and interchange of prestige silk were widespread. ® R. de Micheaux, “Le tissu dit de Mozac, fragment du suaire de Saint-Austremoine (8éme siécle),” CIETA 17 (1963): 14-20; Starensier, “Byzantine Silk Industry,” 146-47; Muthesius, Byzantine Silk Weaving, 68-69. For the date, see L. Duchesne, “Sur la translation de S. Austremoine,’ AB 24 (1905): 106-14. 3 M. Viseux, “Les tissus de la Bible de Theodulf,’ CIETA 71 (1993): 19-25; R. Pfister, “Les tissus orientaux de la Bible de Théodulf,’ in Coptic Studies in Honour of W. E. Crum (Boston, 1950), 501-29; Sabbe, “Limporta- tion des tissus orientaux,’ 816. °* Sabbe, “Limportation des tissus orientaux,” 821. °° Starensier, “Byzantine Silk Industry,’ 561-64; Muthesius, Byzantine Silk Weaving, 67-68. % G. Vial, “Le suaire de Saint Calais,” CIETA 20 (1964): 27-39; Starensier, “Byzantine Silk Industry,” 554—- 56; Muthesius, Byzantine Silk Weaving, 69-71. For further discussion of the relationship between Byzantine and Islamic technique, with additional bibliography, see J. Shepard, “Silks, Skills and Opportunities in Byzantium: Some Reflexions,” BMGS 21 (1997): 246-57, esp. 253-54. °7 Starensier, “Byzantine Silk Industry,” 545. °° A. Stauffer, “Cartoons for Weavers from Graeco-Roman Egypt,” in D. M. Bailey, ed., Archaeological Re- search in Roman Egypt, Journal of Roman Archaeology suppl. ser. 19 (Ann Arbor, 1996), 223-30; I thank Mary Har- low for this reference. See further A. Stauffer, “Une soierie ‘aux Amazones’ au Musée Gustav Liibcke 4a Hamm: A propos de la diffusion des cartons pour la production des soies figurées aux VIIe/Xe siécles,” CIETA 70 (1992): 45-52. °° Lopez, “Mohammed and Charlemagne,’ 23. 100 See Jacoby, “Silk Crosses the Mediterranean”; idem, “Italian Privileges and Trade in Byzantium before the Fourth Crusade: A Reconsideration,” Anuario de estudios medievales 24 (1994): 349-69; repr. in idem, Trade, Commodities and Shipping in the Medieval Mediterranean (Aldershot, 1997), no. 2; idem, “Silk in Western Byzan- tium before the Fourth Crusade,” BZ 84-85 (1991-92): 452-500; repr. in idem, Trade, Commodities and Shipping in the Medieval Mediterranean, no. 7. Luxury exchange, especially of silk, provided the rare exception to the breakdown of large-scale trans-Mediterranean exchange in the 8th century: see C. J. Wickham, “Overview: Production, Distribution and Demand II,” in I. Hansen and C. Wickham, eds., The Long Eighth Century (Lei- den, 2000), 345-77, esp. 369-71. LESLIE BRUBAKER 193 Silk textiles woven with patterns (as opposed to raw or unembellished silk) traveled so widely and, often, with such attention to their movements, because of the high material value of the worked fabric. This was based, in part, on actual costings: the level of techni- cal skill, the specialized machinery necessary to weave complex patterns, the time it took to do so, and the rare dyes often used were all extremely expensive.'' The material value of silk was further enhanced by manipulation of the cultural value of prestige silks, which was elevated by formal restrictions on its exchange. Silks passed between the eastern courts at Constantinople and Baghdad engaged in— or were described as engaging in—complicated games of one-upmanship.'*? How silk was redefined when it arrived in the West is more difficult to say. Silk was used for specialized clothing—from imperial ceremonial attire to bridal kerchiefs—and for curtains and wall coverings in the West and in the Byzantine and Islamic East.'°? Kings, emperors, and caliphs alike used silk in various diplomatic maneuvers.’ Two features do, however, seem to distinguish western from eastern use: the association of silk with the dead of western Europe, as a shroud for important people or for relics; and the use of silk to line or pro- tect the pages of Latin religious books. The so-called Aachen charioteer silk of ca. 800 (Fig. 13), for example, was recovered from the tomb of Charlemagne, though it is not certain when it was placed there,’ and the famous Dioskouroi silk, also of ca. 800 (Fig. 14), was found wrapped around the remains of St. Servatius of Maastricht.'° Pepin, as we have seen, sent silk to Mozac for relic enhancement, and the 836 Council at Aachen requested faithful Christians to donate silk to churches for, apparently, this same purpose.'®’ The silk used by Theodulf to protect his Bible, donated by him to Notre Dame du Puy-en-Velay, has already been mentioned, as has the association of pieces of the Samson silk with ninth- century (Latin) books.'® '©! Good discussion in Jacoby, “Silk in Western Byzantium”; for looms, see J. Flanagan, “The Origin of the Drawloom Used in the Making of Early Byzantine Silks,” Burlington Magazine 35 (1919): 167-72; and in gen- eral, Muthesius, Byzantine Silk Weaving, esp. 5-33. '02 Grabar, “The Shared Culture of Objects”; Soucek, “Byzantium and the Islamic East”; Hoffman, “Path- ways and Portability,’ esp. 25-38; and in general R. Serjeant, Islamic Textiles. Material for a History up to the Mon- gol Conquest (Beirut, 1972). '03 For discussion of the use of silk see, e.g., S. Goitein, A Mediterranean Society: The Jewish Communities of the Arab World as Portrayed in the Documents of the Cairo Geniza, vol. 1, Economic Foundations (Berkeley, 1967), 46, 84- 90, 101-3; A. Muthesius, “The Impact of the Mediterranean Silk Trade on the West before 1200 A.D.,” in Tex- tiles in Trade, Textile Society of America, Biennial Symposium (Washington, D.C., 1990), 126-35; repr. in ea- dem, Studies in Byzantine and Islamic Silk Weaving (London, 1995), 135-45; Starensier, “Byzantine Silk Indus- try,” 183-213; Sabbe, “Limportation des tissus orientaux.” 104 See, e.g., M. Hendy, Studies in the Byzantine Monetary Economy c. 300-1450 (Cambridge, 1985), 268-69, 307-8; A. Muthesius, “Silken Diplomacy,’ in J. Shepard and S. Franklin, eds., Byzantine Diplomacy (Aldershot, 1992), 235-48; repr. in eadem, Studies in Byzantine and Islamic Silk Weaving (London, 1995), 165-72. '05 Description, with bibliography, in Muthesius, Byzantine Silk Weaving, 173; discussion in Brubaker and Haldon, Byzantium in the Iconoclast Era, 93-94. For other examples, see Sabbe, “Limportation des tissus orien- taux,” 825. '06 Muthesius, Byzantine Silk Weaving, 175-76; H. Granger-Taylor, in D. Buckton, ed., Byzantium: Treasures of Byzantine Art and Culture from British Collections (London, 1994), 123-24, both with extensive bibliography. 107 Discussion in Sabbe, “Limportation des tissus orientaux,” 817; other examples at, e.g., 823, 830, 838. See further M. Martiniani-Reber, “La réle des étoffes dans le culte des reliques au moyen age,” CIETA 70 (1992): 53-58. '08 For another example, see Sabbe, “Limportation des tissus orientaux,” 817; for a partial catalogue of silks associated with books, with discussion, see Muthesius, Byzantine Silk Weaving, 127-32. 194 CULTURAL AND MATERIAL INTERCHANGE It is because imported silks were used as relic shrouds that so many of them have been preserved in the church treasures of Europe.'°° The problem is, of course, that the church treasures of Byzantium no longer survive, so it is very hard to tell whether they served as repositories for silk in the same way, or to the same extent, as did western church treas- ures. The Byzantine cemetery excavations that have been published are mostly from sites dated before the seventh century; those from later sites only occasionally list silk among the grave-goods found,'’° and no high-status Byzantine sarcophagi have been discovered with surviving remains. The important relic of the Virgin’s robe was apparently wrapped in silk,'!' which suggests that the practice may have been followed in Byzantium. All the same, there is little other documentary evidence for Byzantine use of silk to encase relics or important corpses. Nor does silk seem to have been associated with sacred books in the Christian East. And, whatever the case in Byzantium, what we see in the West is the delib- erate use of imported silk to enhance the status of the dead and of holy words: the exotic import was used to convey sacral character. The results of cultural exchange that have dominated this article are twofold: response to an import itself, and the impact of an import on local production. The import acquires new meaning within a specific cultural network, and its impact on local production works within that same frame. Moving an object into a different cultural grid changes its mean- ing. The new context imposes its own set of rules; and, beyond this, the imported object can have significance and value—either positive or negative, as we have seen—as an im- port, precisely because it is rare, exotic, and foreign. The eighth-century entries in the Liber Pontificalis distinguish between the (relatively common) “purple cloth” and the (rela- tively rare) “gold-studded cloth of Byzantine purple” in their listings of the prized gifts given to the churches of Rome by the popes, to demonstrate the cultural value of the lat- ter. At the same time, the Carolingian chroniclers distinguish between Frankish and for- eign dress, but now sometimes seek to display the decadence of the import. The silks en- casing relics responded to and reinforced their sanctity; the rarity-value of exotics like Abt ‘l-Abbas the elephant or Constantine’s chrysobull were manipulated to express the status of the recipient. The impact of an import on local production is equally multivalent and ambiguous. In terms of the materialization of cultural exchange, comparison of Theodulf’s mosaic at Germigny-des-Prés and the miniature in the Pantokrator Psalter demonstrates how dif- ferent attitudes toward the visual can result in radically different responses to the same stimulus. The Byzantine response to painted initial letters was perhaps less extreme, but '09 Sabbe, “Limportation des tissus orientaux”; Martiniani-Reber, “La role des étoffes dans le culte des re- liques”; Muthesius, Byzantine Silk Weaving, 119-39. '10 "The largest collections have come, not surprisingly, from Egypt—e.g., the so-called Antinoe and Achmim silks—but others have been found in Syria (e.g., 3d-century sites at Dura Europos and Palmyra, a probably 7th-century site at Halabiyeh) and in the Negev: see Muthesius, Byzantine Silk Weaving, 66, 81-82 for discus- sion and bibliography. For an early10th-c. example see L.-A. Hunt, “For the Salvation of a Woman’s Soul: An Icon of St. Michael Described within a Medieval Coptic Context,” in Icon and Word: The Power of Images in Byzan- tum, ed. A. Eastmond and L. James (Aldershot, 2003), 205-32. '1) A text recounting events of 619/20, usually attributed to Theodore Synkellos, describes the Virgin’s robe as wrapped in imperial purple suffused with myrrh: see A. Cameron, “The Virgin’s Robe: An Episode in the History of Early Seventh-Century Constantinople,’ Byzantion 49 (1979): 42-56, at 52-53; repr. in eadem, Con- tinuity and Change in Sixth-Century Byzantium (London, 1981), no. 17. LESLIE BRUBAKER 195 equally distinct. The decorated initial moved east in response to, or as part of, a rise in book culture that is also evident in the introduction of new minuscule scripts in both the Carolingian West and the Byzantine East,!'? and in the translation movements men- tioned earlier that occurred across the Mediterranean world in the late eighth and ninth centuries. But in Byzantium, the rather chaotic, and sometimes monumental, western painted initial was tamed, reduced in size, and put to use organizing documents, presag- ing what Paul Lemerle called the encyclopedist movement—the urge to classify and orga- nize texts—of the tenth century by several decades.!'° We do not have many examples like Henry III’s cultural reframing of Constantine IX’s chrysobull, but all of the artifacts I have considered show the way that features of élite ma- terial culture were understood differently at their point of arrival from how they were at their point of departure. The particular cultural interpretations of élite goods were de- pendent on fairly small and conscripted groups of people, and although these groups may sometimes have shared beliefs, they did not always. Rulers wished to send out the prestige objects of their own culture in order to show off their wealth and political status, but they could not ever control how these objects were understood and (ab)used by their recipients. However global the élite culture across the Mediterranean pretended to be—and often was—prestige objects that move end up by telling us about two sets of cultural represen- tations, not just one. University of Birmingham, Edgbaston '® On the possible relationship between these parallel developments, see C. Mango, “Lorigine de la mi- nuscule,” in La paléographie grecque et byzantine, Colloques internationaux du Centre National de la Recherche Scientifique 559 (Paris, 1977), 175-80. "> P Lemerle, Byzantine Humanism, the First Phase. Notes and Remarks on Education and Culture in Byzantium from Its Origins to the 10th Century, ByzAus 3, trans. H. Lindsay and A. Moffatt (Canberra, 1986), 309-46; rev. ed. of idem, Le premier humanisme byzantin. Notes et remarques sur enseignement et culture d Byzance des origines au Xe siécle, Bibliotheque byzantine, Etudes 6 (Paris, 1971).