Academia.eduAcademia.edu
THREADS OF ORNAMENT IN THE STYLE WORLD OF THE FIFTEENTH AND SIXTEENTH CENTURIES Anna Contadini riven by congruent needs and tastes that fos- within a quintessentially medieval, western European tered the production of goods for export, the ambon in trefoil shape against a background de<:oration relationship between Renaissance Italy and of verni gris. Here various interpretative problems arise. the Ottoman Empire was marked by a mutually beneficial adoption and adaptation of an array of including that of perception: were the Middle Eastern objects of particular symbolic significance, in the con- D designs and their constituent motifs. But only rarely, it text of trans/alio imperii. as representative of the cui· seems, did this process provoke reflection, so that tural glitter of the Islamic world. or were they thOUght although the Renaissance is better documented than earlier periods, we find that the ascription of meaning to be of Byzantine origin? Or. did they. rather, as I have argued elsewhere, primarily form part of an aesthetic remains elusive. Reception, beyond the evident valua- program determined by the concept of varietas?' tion of objects shown by the barometer of price, was During the Renaissance. new functions might still be found for exotic items (a perfume container might be certainly not verbalized in ways that might suggest recognition of an emerging cultural nexus with an articulated aesthetic in some degree connected to the reengagement with the world of Islam occurring in intel· lectual circles. As a result, the artifacts themselves pro· vide the primary and sometimes the only investigative resource. Yet however thorny the problems they may present, we can at least disentangle some of the complex strands of borrowing and mutation that mark the changes in Middle Eastern and Italian ornament during the Renaissance, tracking the ways in which the responses of each to the arts of the other would change. Previously, Middle Eastern artifacts acquired by the West did not serve as models to be imitated. Rather, they were assigned novel functions: rock crystal vessels, for example, might be used as reliquaries, often embellished with luxurious mounts, to "stage" them and acknowledge them as, usually, royal gifts. But if the process of adaptation in such cases is transparent, it is far less so with the ambon of Henry II in Aachen Cathedral, an early example of the integration of a variety of artifacts, including two Middle Eastern rock crystal vessels, 290 used as a hand·warmer). but thisaspect becomes less significant, and there is a major shift in emphasis toward what I have termed the Wfreeing of the motif. "' Italian textiles, for example. begin to incorporate Ottoman designs, and Ottoman production in turn adopts ltalianate elements, thereby presenting scholars. in addition to problems of provenance, with questions concerning the transmission of design as the industry evolved-and it also needs to be borne in mind that "Ottoman" design may be shorthand for a common vocabulary of ornament shared with the Persianate world. As with the rock crystals on the ambon, a motif may not always have a clear geographical provenance or "national" identity. We are, rather, confronted with the incorporation of imported features of ornament that are then creatively reinterpreted or reassembled to provide new variations to attract appreciative customers: Italian fabrics based on Ottoman models are thus not simple imitations either in terms of ornament or of technique, even if they might be aimed at the Ottoman market. Such fabrics illustrate well the seamless integration of motifs from various sources within a common design world, and if associated prob- unidirectional and insensitive to reciprocity. In tracing lems of attribution can now often be resolved, we are still this change we may point to trade itself as a vehicle of left with the more intriguing and important task of read- exchange and familiarization, but also to creativity in ing them as cultural texts, of following the local inflections of a common vocabulary, and, where possible, technology and design fo r purposes of emulation and teasing out their implications. competition. Transmateriality provides further evidence of adaptation. and the way in which it plays not just with There is, in addition, the phenomenon of transmateriality to consider. It is found both in the morphology of vessel shapes but also with decorative motifs serves as an index of reduced cultural localism and of an eclectic wid- objects (such as metal vessels in the shape of leather ones) and, in particular, in the vocabulary ofomament. Within ening of aesthetic horizons. the Islamic world, for example, thirteenth·century Abbasid manuscript illustration inspires Mosuli metalwork; Antecedents decorative motifs in fourteenth ·century Mamluk Our 'anic The European acquisition of Middle Eastern artifacts, illumination recur on the relief design of the domes of Mamluk mosques; and sixteenth-century Ottoman and whether by pillage, diplomatic gift. or trade. began long Safavid ornament is adapted to all media, from textiles to carpets to book illumination to ceramics.l In Europe, we find similar phenomena of both morphological adaptation and transmateriality (as between metal and glass, for example), and as far as perceptions are concerned, we may detect a parallel move toward nonspecificity in the trajectory of the Renaissance vocabulary of design. before the Renaissance. The rock crystal vessels con· verted into reliquaries and those on the ambon of Henry II, mentioned above, provide early examples. and there are others in different media, for instance, Middle Eastern textiles with ornamental bands «(iraz). sometimes decorative but usually consisting of text.· They provide evidence for the existence at this period of trade in lux- The material discussed below suggests, indeed, that ury goods, and that they were appreciated as precious objects is demonstrated by the fact that they might be by the sixteen th century, if not before. Middle Eastern ornament had become an integral part of an artistic used as wrapping or shrouds for Christian relics. An extraordinary example is the "tunic of Saint Ambrogio" vocabulary that was increasingly international. thereby (d. 397), used as a wrapping for the remains of the saint, calling into question, for this period, the validity of tradi· tional art-historical tropes such as "exoticism" and "imitation. セ@ The term "influence," too. needs questioning: while unavoidable, it must be understood here to operate in the context of a complex set of circulating elements, and not to denote a simple relationship between donor and recipient. that is, from a Eurocentric perspective, as made of indigo-dyed silk with an inscription in Arabic woven in yeUow silk. The blue silk has a lozenge pattern, and the inscription is in a double horizontal band, repeated in mirror image. Unfortunately, thanks to the Bock" for activities of Franz Bock, known as セs」ゥウッイ@ having systematically cut textiles to sell to museums and private collectors,Sit is now dispersed in different reposi- THREADS OF ORNAMENT .... 291 Reliquary of the Nails of Saint Claire, a rather beautiful tenth· century Fatimid vessel that was mounted in Italy, upside down, on a high, copper-gilt stem with a base embellished with semiprecious stones, probably in the fourteenth century.9 The carving in relief is sharp, and it exhibits mastery in the curved floral decoration, with one element seamlessly linked to the next, that is typical of the highest quality of rock crystal production from Fatimid Egypt. Drilled into the very clear crystal is a cylindrical hole, which suggests that the vessel originally must have served as a m:eptade for perfume or cosmetics, '0 but it now contains nail clippings of Saint Clare, the devoted disciple of Francis of Assisi. who died in 1253-a striking example of the radical transformations to which such early acquisitions were often subjected (fig. 23.1). Fig. 23.1. Reliquary of the Naill 01 Samt (Jore. Egypt. tenth century. fallmid rock crystal; copper golt chalice WIth precious stones, Italy. lovrteer1th century (?), PrOl0monastero di Santa Chiara. Assisi.ltaly, tories, with the Victoria and Albert Museum (V&A) in London having the most important fragment-the one containing the part of the inscription that tells us it was done for the noble prince Nasr al-Dawla Abu Nasr, the Kurdish ruler of Diyarbaklr, in southeastern Anatolia, between 1010 and 1061.0 The tunic was probably woven in Abbasid Baghdad, a major center of textile production at the time. Another early acquisition, now in BOOmin, Cornwall, is an ivory casket with a painted decoration, mainly in gold, now largely lost. Used to house relics of Saint Petroc, it shows connections with various Middle Eastern styles and is the work of Muslim craftsmen? It makes the point that the "Middle Eastern" geographical boundaries may at that time have been rather different from the current ones, fo r it was produced in Sicily or southern Italy under Norman rule and, as one of the so-called Siculo-Arabic caskets, bears witness to the extraordinary syncretic culture of Sicily in the twelfth and thirteenth centuries.' Particularly prominent among these early Euro· pean acquisitions are rock crystal vessels, ranging from the spectacular and finely carved ewers now displayed in, for example, the treasury of San Marco in Venice and the V&A, to a variety of smaller pieces. One such is the 292 CIRCULATIONS ANOTRANStATIONS Renaissance Acquisitions The above are just three examples from a wide range of artifacts that survive in European collections and church treasuries and demonstrate that Europeans started acquiring artifacts from the Islamic Middle East already during the Middle Ages." During the Renaissance, such acquisitions multiplied and became more varied, as trade assumed greater importance, facilitated by the growth of extensive and increasingly dependable mercantile networks. Artifacts were imported from various parts of the Middle East: from Fatimid (909-1171) and, later, Mamluk (1250-1517) territories, that is, prinCipally, from Egypt and Syria; from the Ilkhanid Empire (1256-1353), which controlled Iraq and Iran and also gave access to Central Asia (Turkestan) and China, especially with regard to silk; and, with the rise of the Ott omans as a new major power in the fifteenth century, increasingly from Turkish centers of production. Indeed, Ottoman rugs and textiles were to become a Significant import. In Italy, Islamic artifacts were transmitted not only through Sicily and southern Italy, as before, but also primarily through the commercial activities of the maritime republics and other pivotal mercantile centers, with Venice particularly active in importing carpets and textiles. After the Ottoman conquest of Constantinople in 1453, the Genoese trade gradually became less significant; Venetian commercial activity increased, regardless of the political tensions-sometimes escalating into actual military conflict-between the Sublime Porte and the Serenissima; and Florence, in turn, was granted trading Fig. 23.2. (a) Velvet cope, Turkey. ca. 1500. Chiesa di Santa Maria Gloriosa dei Frari, Venice. (b) Velvet, Italy (pfOoobl y VeniCe), sixteenth century. Museo Nazionale del Bargello (inv. Franchetti 639), Florence. (el Velvet, Italy. second half of fifteenth century. VJ(toria and Alben Museum (inv. 1\0, ORC.346·1911), London, capitulations in 1460: a ready supply of imported material of various prices and qualities was assured alongside an equivalent range of locaUy produced material. likewise, Ottoman customers had access to European goods, and became increasingly keen to acquire them, fabrics espeCially, for the quality of Italian production made them particularly attractive. There are two resulting trends: a variety of economic factors encouraged the manufacture of similar materials in several locations, while on the other hand, homogeneity was countered by local specialization. Trade The dissemination of ornament through trade may be illustrated first by textiles. Ottoman exports were principally in the form of fatma (voided and brocaded) velvets, with Bursa as the main production center from the later fourteenth century onward. But Bursa also became an international center for trade in raw silk, and it was this that increasingly attracted Italian merchants supplying Italian centers of production. The consequent growth in the output of the Italian weaving industry resulted in a reduction in local demand for Ottoman worked silk, especially as Italian weavers had begun to explore Ottoman patterns. Indeed, Italian fabrics with design features of Ottoman or other Middle Eastern derivation would be imported in increasing volume by the Ottomans, as demonstrated by Ottoman court documents: of the velvet caftans in the Topkapl Palace, only a few are of local production." Not surprisingly, the ornamental repertoire of these fabrics shows a degree of interchange that can create problems of identification. For example, on grounds of design, the Santa Maria dei Frari cope of ca. 1500 (fig. 23·2a) was long thOUght to be Venetian, but technical analyses confirm that it is in fact Ottoman, testimony to the adaptability of Ottoman weavers in responding to imported fabrics. 'l On the other hand, a sixteenthcentury velvet in the Bargello Museum was once thought to be Ottoman, but it is now accepted that it is of Italian (and probably Venetian) manufacture (fig. 23.2b)." Although it incorporates well· known Ottoman motifs, their overall organization is rather atypical, as is the combination of colors, which can, however, be matched in textiles known to be Italian: the light red/pinkish color of the spreading tendrils, for instance, is found in Italian textiles from the end of the fifteenth century onward, as in an example in the V&A (fig. 23.2C), lS and further evidence for an Italian origin is provided by the fact that the fabric is pure silk (not normal for Ottoman velvets) and by dJfferences in the way the pile is treated. ,6 Italian fabrics based on Ottoman models would have been mainly aimed at the Ottoman market, but that THRE .... DS Of ORN .... MENT 293 Fig. 23.3. (a) (arrna, Bursa, Turkey, ウゥセエ・ュ ィ@ century, brocaded velvet. VICtoria and Albert M useum (mil. no. 100-1878) , Lo rxto n. (b) Velvet, Bursa. Turkey. la te sixteenth - seventeenth century. Muse<> NaziOOaledei Bargello (inv Franchet!l99l. Rorence. (Centaurea moschata), a floral Ottoman-derived motifs were also included in designs or possibly sweet sultan for the Italian home market is suggested by the presence of clothes with Ottoman patterns in paintings, such as the Portrait ofa Lildy by Parrasio Micheli (ca. 156S)" and Titian's The Burial of Christ (ca. 1572) with the 」ゥョエ。ュセ@ motif." It is instructive, however, to note that fabrics with Ottoman motifs in paintings cannot readily be identified as Ottoman, in contrast to the frequent presence in element that may have been derived from European paintings of Ottoman rugs, thus reflecting the disparity between the high level of demand for Ottoman rugs as against the low level of demand for Ottoman fabrics , given the abundance aflocal manufactories.'9 herbals and books of floriculture, but took on a rather abstract and instantly identifiable fanlike shape in its Ottoman manifestation.l1 And then we have capers. A document dated June 14.1555, in the National Archive in Florence, sent from Frankfurt by the merchant Francesco Carletti to the SaHti Company in Florence, contains a drawing for a textile design with an ogivallattice through which are threaded branches with capers, accompanied by a request for pieces like it to be manufactured for a The trajectory of a particular design motif is often "Frankfurt fair. ''OJ Although the organization of the ogi- complex, as shown, for example. by the diffusion of the ogivallattice, the origins of which are ultimately to be val lattice in this drawing is typically Italian, and the representation of the capers likewise, as may be seen in found in eastern Asia. It traveled westward with the textiles such as a Mongol Ilkhanids (1256-1353), reaching Mamluk Syria and Egypt and thence Re naissance Italy, and it is likely stole) in Palazzo Mocenigo in Venice,1.O both were also to that the Ottomans' adaptation of it was indebted to stoia da procuratore (a procurator's be found on Ottoman velvets, such as one in the Bargello セ@ and Carletti was presumably familiar with (fig. RS N S「IL Italian rather than Eastern models.' ° In a sixteenthcentury farma velvet in the V&N' (fig. 23.3a), the ogival such fabrics. But no Middle Eastern source is implied by lattice encloses yet another motif with a complex history, for it serves as a framework for rows of carnations, cho tane, for by the sixteenth centu ry, domascho had 294 CIRCUlATIONS AND TRANSlATIONS his specifying that the order should be made of domaslong lost any connection with Damascus : it refers to a locally produced fine, thin silk (while lane specifies a maroon/orange color). This document thus illustrates basis of the more clearly compartmentalized organization of the decoration. a conclusion confirmed by the well a design world marked by the seamless integration of elements from various sources, and quite possibly absence of the black organic compound used on Middle Eastern pieces to provide the background for the silver ignorance of. their ultimate origins. inlay." Another candlestick in the V&A demonstrates the reciprocal nature of such transfers. although in this case with regard to morphology rather than decoration (fig. 23.4b)."' Of its two component pieces (the third is At the same time. his letter provides an interesting insight into entrepreneurial activity and commissioning well beyond Italy. This international market also included the Middle East. for commissions involving the dispatch of drawings with textile designs were not just internal European affairs: documents recently published by Giilru Necipoglu contain orders of this kind from Ottoman pashas. one for Venetian fabrics to be sent to Cairo, the other. to go to Istanbul. involving various cushion designs.'· Similar complexities arise with metalwork that can be identified as European imitations of Middle Eastern models, mainly Italian objects demonstrating the desirability of such designs in Renaissance Italy. Examples are the two candlesticks in the V&A with the Foscarini coat of arms that strive toward a Middle Eastern typol· ogy in their decoration (fig. 23.4a)."' lndeed. the stylistic similarity of such pieces with ones of Middle Eastern origin is suffiCiently close for this group to have been identified as European only relatively recently, on the missing), the upper part is a later replacement and does not concern us.)O The morphology of the lower part. made in western Iran in the late sixteenth or seventeenth century. is derived from an Italian and most probably Venetian prototype. one demonstrated by the Foscarini candlesticks. The incised design, however. most of the inlay of which is now unfortunately lost, conforms faithfully to Safavid ornament of the period of Shah 'Abbas I (r. 1587-1629) as demonstrated. for example, by a flask of ca. 998 (1590) in the British Museum (fig. 23.4c).l' In both the V&A and BM pieces. we find an almost identical treatment of the cusped arches and split palmettes. For the interpretation of these phenomena, however, especially in order to make sound deductions about style preferences. much still remains to be done, in particular by taking into consideration a much larger corpus of artifacts than has hitherto been Fig. 23.4. (a) One of a pair of can dlesticks, Italy {prob<lblyVenicel, mid-Sixteenth century, brass engraved and inlaid wit h silver. VICtoria and Albert Museum (inv no. 553-1865), London. (b) Candlestkk (lower part), western Iran, latesャセエ・ョィ@ or seventeenth century, engraved bronze. Victoria and Albert Museum (inv. no. 4)01-1857), Lon don. (c) Flask. han, ca . 998 (1590). brass. British Museum. Henderson Bequest (inv. no. 78.12-)0.735). London. THREADS OF ORNAMENT 29S attempted and by identifying textual references more fully. In the interim, it may be suggested as plausible that, beyond curiosity, an aesthetic openness allowed a conceptual Bョ。エオイャゥコッセ@ of Middle Eastern orna- dome of his mausoleum and on a brass bowl inlaid with gold and silver in the V&A." The process by which a decorative motif migrates across different media may readily be iUustrated by the ment that allowed for the frictionless integration of certain novel elements. grotesque, which consists of fantastical human and animal forms interwoven with foliage designs. It derives from ancient Roman wall paintings that were discov- Transmateriality A further feature of the circulation of ornament is transmateriality, as a common pool of design elements appears in different media. 1his is a phenomenon that appears within both European and Middle Eastern pro· duction as well as between them. European metalwork, for example, may imitate ornament on glass, which in its turn derives from textiles, as in the case of a late· seventeenth-century silver gilt beaker in Hamburg P that displays the same peacock-feather pattern as an early·seventeenth·century glass beaker in ViennaJJ on which the red dots, rendered in the metal beaker by punches, seem to have their origin in a textile pattern." For Renaissance Italy, to take just one instance, we may cite the decorative designs on the foil disks of a group of medallion·shaped, silver·gilt and enamel costume ornaments that exhibit similarities with manuscript illumination from Milan around 1380 to 1400.l!I The incorporation of designs found in manuscript illumination also occurs in Mamluk metalwork, while similarities between the figural images on Mosul metal· work and Arab and Syriac manuscript illustrations have been noted on a number of occasions.3/> In terms of orna· ment, one may observe parallels with manuscript and luster tile painting in the background decorations of MosuJi metalwork, which range from plain backgrounds to thick winding scrolls, hatching, spirals, and indepen· dent ornamental scrolls.17 Likewise. it has been noted that the designs on metalwork produced by Mahmud aI·Kurdi (see below) have elements in common with those found in Mamluk and Iranian architecture and manuscript ilIumination,- while earlier metalwork may also exhibit the phenomenon of imitating the decora· tive effects used on a different material: the Courtauld metal bag (ca. 1300), for example, has an overall decora· tion that recalls Chinese·like textiles.'9 A particularly striking example of trans materiality is shown, during the reign of the Mamluk sultan Oaytbay (1468-96), by certain motifs such as the three-petaled leaf, which appear on artifacts in various media, ,0 including on the ered in Rome during the fifteenth century. and thereafter began to be popularly used in the decorative arts not only in Italy but also across Europe. From its beginnings as wall decoration, it thus spread to a variety of media such as engravings, woodcarving, textiles, ceramics, and metalwork. where it appears on objects as diverse as German silver tankards and Italian armor... Such transferability of motifs can be partly explained by the fact that artists both produced designs for, and worked on, a variety of luxury objects, including tapestries, frescoes, stucco, and metalwork and were often commissioned to decorate entire residences, as in the case of Giulio Romano (ca. 1499-1546) and Perino Del Vaga (l501-47), both of whom had trained with RaphaeL OJ The wide dissemination of artists' designs was a significant factor in the circulation of ornament in Renaissance Europe, for while these drawings were initially private affairs between artist and patron. they later became a collection of stock samples, and sketchbooks were lent to friends and colleagues. The development of printing further increased their availability, and ornamental prints and pattern books were published to cater to craftsmen in various fields who were trying to keep up with the demand for luxury goods from the emerging bourgeoisie but did not have the necessary expertise to create their own designs. In Germany, for example, pattern books by artists such as Hans Brosamer (1495-l554) provided goldsmiths with ideas." Evidence for the existence of such pattern books in the Islamic world is scanty,'! but there are certainly parallels between Europe and the Middle East with regard not just to one person working in more than one medium but also, and more significantly, to the ways in which transferability was encouraged by the close relationships that sometimes existed between craftsmen working in different media. How extensive the former practice was is still a matter of investigation, but it is very likely, for instance, that the building superintendent of the Sultan Hasan mosque-madrasa complex in Mamluk Cairo, Muhammad b. Biylik. was also the scribe of a Our 'an in the Keir Collection (and connec- tions have been made between the decoration of that building and manuscript illumination)." There is dearer Mamluk evidence for the latter process, as we find familial ties between goldsmiths and manuscript scribes and illuminators: the scribe of a Mamluk Qur 'an dated 801 (1397),<7 for instance, was a goldsmith's son, while an earlier Mamluk Qur 'an, dated 701 (1302), was produced in the mosque of the goldsmiths' market (Suq 。ャMsァィINセ ᄋ@ a European market while adhering to their own decorative idiom. or that they were commissioned to decorate pieces of European manufacture, which implies either a back-and-forth trading process or the presence of craftsmen from the Middle East in Venice. The latter possibility has generally been discounted. the assumption being that these metalwork pieces were probably produced in Egypt or Iran with European buyers in mind. However, recently discovered documents confirm In the Ottoman world, likewise, a direct connection between manuscripts and metalwork is provided by the binding of the Divan of Sultan Murad III in the Topkapl Sarayl, which was done by the court goldsmith Mehmed. oI'J For a European parallel, 1cite the particularly strong connection, reinforced on occasion by social and familial ties, that existed between German armorers and the engravers and etchers who ornamented their suits of armor.IO work of high quality, and the possibility cannot be excluded that he taught craft skills to Venetian assis- Given the resulting transferability of design elements, we find that, just as with fabrics, metalwork tants, as Marco Spallanzanj speculates.57 But whatever his role, recent documentary evidence confirms the sometimes presents us with seemingly intractable problems with regard to provenance. Those pieces for which existence of a back-and-forth trade in metalwork: a Middle Eastern origin can be identified include both objects made for a local market, some of which were acquired by Europeans, and objects made for a European market, sometimes in response to commissions. Dubbed セv・ョエッᄋs。イ」ゥLB p@ they are typically brass objects distinguished by the use of silver inlay and may be assigned broadly to two types: One consists of pieces in which the decorative design can be identified as late Mamluk, typical examples being globular perfume burners.l' The other type, associated with Mahmud alKurdi, is of uncertain provenance. It is stylistically akin to late fifteenth- and sixteenth-century Iranian metalwork. but yet distinct. l l while being differentiated technically from the first type by the finely engraved arabesques of the background.;.o In both types, the background is covered with a black organic compound, a feature that points to Middle Eastern origins, as do the metallurgical analyses showing that the Mahmud alKurdi pieces contain much lower levels of nickel than do European ones.}$ One example of this second. Mahmud al-Kurdi-type, actually includes on one rim an Arabic formula identifying the maker and on the opposite side a corresponding transliteration in Roman characters,S(! dearly indicating that it was intended for Europe. Among the late Mamluk pieces, some have a European morphology, which suggests either that Middle Eastern craftsmen were consciously creating shapes to appeal to the presence in Venice in 1563 of a certain Armenian named Antonio Surian, thirty-five years of age. from Damascus. employed to recover ordnance from sunken vessels (artiglierie dalle navi affondare) but who is also noted as doing inlay work (all'agemina) better than any italian, implying that he was producing inlaid metal- pieces produced in Venice or arriving from elsewhere in Europe were dispatched to Damascus to be decorated. even on occasion incorporating a specific design feature commissioned by an aristocratic European buyer. such as a family coat of arms, and then brought back to Venice.sl The well-known Molino ・キイNセ@ for example, has a European shape and coat of arms,60 but its decoration is characteristically Middle Eastern in style, and the presumption is of a vessel of European manufacture with the surface worked in the Middle East or by a Middle Eastern craftsman :61 the decoration is in fact very similar to that of a tray made in Cairo in the second half of the fourteenth century now in the V&A·' Further evidence for transmateriality is provided by leatherwork, as exhibited by bookbindings and shields. Venetian gilded leather shields,"J for example, which are primarily decorative symbols of power, paraded on special occasions, exhibit Ottoman design features found on bookbindings and other material such as textiles and metalwork. On one of the shields, we find a twelve-point medallion in the center with an interlace of flowers and half-palmettes (fig. 23.5a). while the field is decorated with the cloud-collar motif. reminiscent of the cloud-collar border of Ushak carpets. such as the one in the V&A, so that the whole is a quite typical assemblage ofOttoman motifs (fig. 23_5b). Indeed, without knowledge of the differences in shape and materials it would require detailed analysis in some cases to deter- Fig 23.5. (a) Gilded leather shield. Venice, 1SSO- 1600. Almeria del Palano Ducale (inv. J20). Venice. (b) Small Ushak double-niche medallion rug. Turkey. 1500-1S5O. Vioori<l and Albert Museum セョカN@ no. T.Sl·19:ZO). London. ,,,b mine their Venetian provenance. On a buckler from the In the only painted buckler (fig. 23.7a) that does same group, in contrast, the organization of the various elements, together with the coloristic effect. is a creative Venetian reinterpretation of an Ottoman design (see below), its transformational strategies reminiscent of what we have seen happening on the Venetian velvet in the Bargello and the Frari cope. Similar processes are apparent on another shield. where the medallions on the field . with their poly lobed contour and the quadrilobed split palmette with a central Hower (fig. 23.6a). are very similar in shape and ornament to those on sixteenth-century Ottoman silks, and also have a similar coloristic effect (figs. 23.6b. 23.6c). However. another motif. the cloud-band. is used in a セウエケャゥコ・、B@ form quite fore ign to its Ottoman realizations. with the curves squeezed tighter. The lack of any pretense at precisely reproducing an Ottoman object is confirmed by the insertion of the Lion of St. Mark in the central medallion and, below it, the initials "A C" (probably for a member of the Contarini family) . The shields thus exhibit a variety of responses, including the reassembly of selected motifs in novel combinations. not have a relief ornament, we find links with yet other media. In the interlace of half-palmette. including the coloristic effect of blue and red. the decoration is close to Ottoman Iznik ceramics , as illustrated by a tile datable to around 1578 (fig. 23.7C), while t he shape of the split-palmette medallions recalls elements found in metalwork, as seen in a late-fifteenth- to earlysixteenth-century perfume burner, in Bologna, proba· bly made in Egypt or Syria (fig. 23.7b). A glimpse of the importance attached to painted shields (and other arms such as lances and quivers) is given by documents that Luca Mola has recently found relating to a Hungarian. Nicolo Ongaro, who was invited to work in the Venetian arsenal. as he had a reputation of being a good shield and lance maker." He eventually complained of being underpaid, and was granted a yearly stipend of Sixty ducats on condition that he would supply thirty shields and thirty lances annually. Although the shields mentioned in these documents may not be the same as the ones discussed so far, as they were not destined to the Venetian aristocracy, the documents clearly describe 29 8 CIRCULATIO NS ANOTRANSLATIONS Fig. 23.6. (a) Gilded leaTh('f shield, Venice. IS50-16OO. Armeria del Palazzo Docale (inv )14 (fOfmefly inv. In/Sala El). VeniCe. (b) Ottoman Silk (kemryd), Bur!ol Of Istanbul. second half of SixteenTh century. Museodel Tessuto (inv. no. 75-01316), Plato,llaly. (c) Qnoman silk (kemrydJ, Bur!ol Of IST(lnbut Turkey, ca. lS4o-5O Museo NaziOnale del Bargello (inv. (arrand 2$14), Florence. them as painted, and the money and time that Nicolo Ongaro was granted suggest that items like these were of importance nevertheless. Technique Analogous combinations of ornamental features found on a wide range of media occur in another leather product, bookbinding, the study of which highlights again the importance of investigating the techniques used in order to understand modalities of transfer. The splendid Venetian stamped, painted, and gilt binding in the Newberry Library containing the document of appointment, by Doge Alvise Mocenigo, of Girolamo Mula as procurator (procuratore) of St. Mark in 1572 is made up ofvarnished upper covers and doublures, and within the clearly Islamic-derived design format of a central lobed medallion, corner pieces, and arabesques we find not only the Lion of St. Mark and the coat of arms on the reverse but also elements of Renaissance ornament in the Bーッオャ。エ・、セ@ border that contains not just birds and insects but also grotesque figures."l Another instance of the incorporation of features characteristic of fifteenth-century Mamluk bindings is provided by a copy of Cicero's Epistolae ad {amiliares, printed on parchment in Venice in 1475, and bound for Peter Ugelheimer (d. 1489), the owner of the Deutsches Haus Inn in Venice.1>6 Edged with knotwork motifs, it has at the center a typically Middle Eastern almondshaped medallion. But this contains Ugelheimer's coat of arms surrounded by the Y-shaped stamps that are usually found on Islamic metalwork, not on bindings, thus indicating that transferability of ornament might also be mediated technically. Similarly, the tools used on the Italian binding of a manuscript from Padua, copied in 1400, now in the Pierpont Morgan Library in New York, may have been modeled on metalwork tools, "a natural borrowing since tool cutting was generally the work of goldsmiths. セ VW@ Venetian bindings in Mamluk or Ottoman styles are never precise imitations, contrary to common assumptions. The outer cover of Leonardo Bruni's Com- mentarius rerum in !talia suo tempore gestarum (146465) shows segmented borders with gilded tool work, THREADS OF ORNA M ENT 299 Fig. 23.7. (al Buckler, Venice,late ">ixteemh century. Armeria del Palazzo Ducale {lnv 66/Sala E}. Venice. (hl Perfume burnet', Egypl or Syria,late fifteentn-e,uly silcteenth century. Muse<> (Meo Medievale (iOY. no. 2110), BoIogf"la. Ic} Tile. Iznik. Turkey. ca. 1578. Victor;3 300 Alben Museum {inv. no. 164S>1892l,loodoo. 300 Fig. 23.8. (a) Doublure,l. Bruni. CommemarlUS rerum in Italia suo remporegesrarum (Bologna m 1464- 65). Biblioteo Marcianil (LaI.X. 117 [=3844 ]), Venice. (b) Upper cover, Petrarch, Canzoniereand TriOnfi (FICKence. 1460-wl. Bodleian libfary (Ms.CanonJt(lI.78). OxfCKd while the doublures have elaborate filigree (fig. z3.8a)." Although the overall organization is derived from Mamluk bindings, the leather cutout constituting the design of the filigree is covered with little pearls, once thought to be made of glass, but actually, as recent analysis shows, of resin-a form of ornamentation not used by Middle Eastern binders.69 The tooling inside the segment ed borders of the outer cover is rather messy by comparison with the binding in the Bodleian Library of 1460-70 (fig. Z3.8b), a type that could have provided a possible inspiration.7<I This binding, which covers Petrarch's Canzoniere, was for a long time thought to be Italian, but recent analyses of the sewing show that it m ust be Mamluk,71 a conclusion reinforced both by stylistic considerations, as the design is elegant and rigorous in its organization, and by technical features, for the tooling is identical to that on other bind· ings known to be Mamluk. n Similarly, a volume in the Biblioteca Marciana containing two manuscripts (one of which, De vita et moribus philosophorum, is dated 1453) that was later owned by the Venetian historian Marin Sanudo the Younger (1466-1536) has a Mamluk or North African-style binding (with a flap) that Anthony Hobson believes was bound in Egypt.7J In addition, Hobson has noted two other European books-a copy of the Aldine Press edition of Catullus, Tibullus, and Propertius (1502) and the Horae Beatae Mariae Virginis, printed in Paris in 1s05-that have Ottoman bindings and sewing, which suggests to him that they were sent to Istanbul to be bound." At the same time, the collections of kings and scholars such as King Rene of Anjou (1409-80), the Italian philosopher Giovanni Pico della Mirandola(1463-94), and the Spanish ambassador to Venice Diego Hurtado de Mendoza (ca. 1503-75) attest to the presence of Arabic and Turkish manuscripts from the Islamic world (and their bindings) in Europe, and these could have provided models for local bookbinders, thereby faci litating the transfer of Mamluk and Ottoman ornaments into the European repertoire.75 Vet if we compare the tooling of the borders in the Bruni binding, we can see how the design ofthe Mamluk model has, seemingly, been misunderstood. Or has it? Another possibility, in the absence of tools capable of such fine detail, would be THREADS OF OANAMENT t. 301 approximation born of necessity, which might also in the intertwined vegetal motifs that often contain a explain why the problem of steering the design around the Corners is solved or, rather, evaded by the substitu- reinterpretation of the lotus flower. palmettes. tulips. and carnations. The outlines have been pricked for tion of little squares. Differences in both equipment and technique, transfer of the design by pouncing. Like the two embroi- whether enabling or inhibiting, may well be a factor contributing to stylistic shifts as a design feature travels. For example, on the varnished binding in the Newberry Library that was mentioned earlier, the binder resorts to riearni, dated 1570. contains material similar to Pellegrino's aforementioned drawings, published in 1530, pointing to the longevity of such designs. But Pellegrino was already producing designs rather painting, thereby allowing certain designs to be copied more easily; and where painting was combined with closer to the abstract "arabesques" found on early- relief, this was made not with small metal tools, but by sixteenth -century Venetian bookbindings. metalwork, pressure molding. With advances in technology, by the end of the fifteenth century. large stamps could be pro· and textiles than to those on Middle Eastern objects fea turing seminaturalistic identifiable flowers. There is duced by European binders (fig. 23.9C), facilitating the no attempt to identify the origin of specific designs, and transfer of ornament by drawing an imitation design, whether inspired by an object or a pattern book, from as with other sixteenth-century European pattern which the stamp would be cut. The pattern of the central in character, including also Renaissance grotesque ornament. Frequently a single term-usually セ。イᆳ medallion of the Venetian binding in Chatsworth library, the 1520-30 Fra Giocondo, SyIlO!}e (fig. 23.9a), for example, is very similar to the field of an earlier Ottoman binding made by the influential Turkmen binder Ghiyath aJ-Din in 14n (fig. 23.9b),76 but even without access to an Ottoman model incorporating this design in a medallion, the Venetian binder could have drawn it out himself, based on an original or copied pattern, in order to create a stamp for pressure molding. The printed drawing of an ヲ。セッョ@ dery books in the V&A by the Venetians Lunardo Ferro and Amadio Novello, both dated 1559. the Libra dei books, Pellegrino's designs are in fact strongly eclect ic besque" or "moresque"-is applied generically to a variety of styles: for Pellegrino, his patterns are in the ヲ。セッョ@ arabicque et ytalique (ArabiC and Italianate manner). With regard to both metalwork and textiles. the early da or di Darnasco (from Damascus) is gradually replaced, in the sixteenth century, by alia darnaschina (in the Damascene manner). a term now also applicable to pieces made in Italy.'" Although, as seen above, it could be a technical almond medallion by Francesco di Pellegrino in his La /leur de la science de pourtraicture: Patrans de brolierie, designation of a particular quality of silk, in other can· arabicque et ytalique (1530) demonstrates well this possibility (fig. 23.9d).n Such pattern books, together while no longer necessarily Signaling geographical provenance, it could be argued that it demonstrated at the with single-page ornament prinrs. were particularly in vogue in the sixteenth century; intended forcraftsmen of level of style an enduring awareness of a Middle Eastern association allied to the prolongation of a taste fo r "orien- texts it referred to a spectrum of design features. so that various fields, including bookbinding," they would have talizing" motifs up to the second half of the sixteenth facilitated the transmission of ornament across media and encouraged eclecticism. century. It is true that the Carletti drawing mentioned Renai ssance Eclectic Taste above contains no such motifs. but the term was certainly applied to the Middle Eastern objects and designs and their Italian emulations for which there was a taste The extraordinary Libra dei riearni (Book of embroidery) and even a fashion in cities like Venice and Florence from the fourteenth to the sixteenth century. Such artifacts. (fig. 23.10) by the Venetian Gaspare Novello. dedicated to Loredana Mocenigo (wife of the Doge Alvise Mocenigo. carpets especially. spread to various levels of society, not just the aristocratic and upper classes. Indeed, as Marco whose varnished binding was discussed earlier), is a pre- Spallanzani has shown. the type of carpet that we now label as Holbein (a ruote) became so fashionable that one cious document testifying to the circulation of embroidery models intended for women, whether printed, as in the case of Pellegrino's work, or drawn by hand in ink customer, in 1472, wanting variety, had to insist on having something different.I' and various watercolors, as here." Some of the drawings By the second quarter of the sixteenth century, show clear affinities with Ottoman ornament, especially Middle Eastern-derived ornament had become an 302 CIRCU LATIONS AND TRAN SL ATIONS • Fig. 23.9. (a) Upper cOYer. Flit Giocondo. 5y1/ogue (Venice. ca. rS»-30). Chatsworth Library, Chatsworth House, Defbysh,re, England. (b) Binding. Fakhr ai-Din al- 'Irilqi,AHama 'or (Istanbul, 88, (1477), Museum ofTurkish ilnd Islamic Arts (MS. 20)1, forrnef/y MS. lSOlJ,ISlimbul. (e) Stamp fOf pressure molding from a modern 「ッォセイ、・N@ Private colle<:liOfl, Istanbul. (d) Pimern in Islamic style. from Francesco Pellegrino, La fltIJrde 10 ウ」Zセョ・、@ Patrons de broderie, fO{oo arobKque el yralique (Pilris, 1530) セョ|ャ・ィ@ century_ poortraicture; "" h .. • Fig. 23-10. Gaspare Novello, Librodet"cami (Book 01 embrOIdery) (Venice, 1570). Museo delTesSU!O (inv. no. 97.01.M. fol. 14f "rod detilil), Prato, Italy integral part of the ltalianate stylistic repertoire, a productively hybrid domain within the larger European and Mediterranean style world where concepts such as "influence no longer have traction. This is n underlined by the ernie perceptions that we can detect, however faintly: the generalizing vocabulary of Renais- sance ornament seems to indicate a gradual diminution in the signaling, not of Middle Eastern connections, but of non-European otherness. There is, indeed, a SUI- prising lack of commentary on the "foreign" nature of both Middle Eastern objects and the so-called arabesque. For example, Sabba da Castiglione, in his Ricardi (written in 1549), simply lists a wide range of objects to adorn the home that includes tapestries from Flanders, Turkish and Syrian carpets, leathers from Spain, and new and wonderful things from the Levant and Germany." Such eclectic acceptance and integration seems to be characteristic of the primarily nonrepresentational arts. Although there could, by definition, be no compa· rable dilution of otherness in figural painting, parallels might be anticipated in the acceptance and circulation of novel styles and techniques, yet these can be detected only sporadically. The early paintings in the Cappella Palatina, Sicily (1l43), demonstrate that Islamic-style figural representations might be integrated within a 304 CIRCULATIONS ANO TR .... NSL .... TlONS Christian setting, and a later self· conscious adaptation of techniques typical of painting in an Islamic tradition can be seen in the Seated Scribe (1479-81), attributed to either Gentile Bellini or Costanzo di Moysis (or da Ferrara). This in turn was to be copied by Persian artists, 'I and a Persian painting of The Virgin and Child, datable to the late fifteenth century, was also based on an Italian model. closely resembling one of Bellini's works." Yet such examples are rare, and later European depic· tions of people from the Islamic world remain firmly within Western artistic traditions of representation. 11 Having seen Western paintings, Mughal artists were prepared to copy aspects of the techniques that they employed." But apart from the painter of the Seated Scribe, it may be assumed that Western artists did not generally have access to representative examples of Islamic painting, and even ifthis had been the case, one can only speculate as to what their reactions might have been. Accordingly, comparison between figural representation and the circulation of ornament can only be taken so far: the former gives the occasional glimpse of a potential cultural openness and reciprocity with implications for an awareness of novel aesthetic norms, while the latter demonstrates an achieved inte· gration. The apparent ease with which this came about may be partially explained by a significant cultural shift during the Renaissance, the "rediscovery" of antiquity. Allied to the growing humanist concern with the languages, literatures, histories, and sciences of the terranean, while the Kufic inscriptions that merge beautifully with the rest of the decoration were often not understood to be Arabic at alL past, this also embraced an enhanced visual awareness Conceptually naturalized, Middle Eastern ornament was thus fused within an increasingly undifferentiated Renaissance design compendium, a unified world that allowed Sabba da Castiglione to arrive at a cultural vision with an ethical dimension, for he concludes that all these ornaments (and he actually uses the word ornamentj) are to be commended and praised because they sharpen the of Greco-Roman art, and with it of the elements of arabesque and their organizational possibilities that both classical and Byzantine ornament contained. Once familiar with such forms, the Western eye would hardly find their Middle Eastern manifestations unusual and would, indeed, be predisposed to react positively toward them. They could thus be both readily incorporated as design elements and naturalized to the extent that awareness of their origin might be erased. Even as late as the nineteenth century Middle Eastern objects such as the famous Fatimid rock crystal ewers were considered Byzantine: the vegetal interlace surrounding animals is a form that had long existed around the MOOi· intellect and induce politeness, Civility, and courtliness (e tutti questi ornamenti aneora commendo e laudo. perche arguiscono ingegno, poIitezza, civiltd e cortegiania).l? It would be nice to think that our enhanced awareness of the international movement of ornament and the creative local energies it helped to inspire might, in turn, itself foster such qualities. THREAOS OF ORNAME NT lOS Perosa (Lon cion: Warburg Institute, 1960-81), 1.5.5-22.5. For a different attribution of the archite<ture away from Alberti, セ@ Charles Mack, PiellZO: Tht! CI'eotion ofa Renai$sal\Ce Cicy (Ithaca, NY; Cornell University Press, 1987). 44· "Hanna I rittori un'altra sorte di pittura, 」ィ・セ@ Disegno & pittura insieme, & questo si domanda Sgraffito セ エ@ non serve ad aluo, che per ornamenti di facciate di case & palaui .... - Vasari, LA vilt, 1:142. "Sgraffio, 0 Sgraffito m. Una sorta di pittura (he l disegno. e pittura insieme; serve per 10 piil per ornamenti di fa(ciate di case, palaui, e tortili; ed セ@ sicurissimo all'acque, percM tutti i dintorni son tratteggiati can un ferro incavando 10 'ntonaco prima tinto di color nero, e poi coperto di bianco fatto di calcina di travertino; e (os\ can que' tratteggini, levato il bianco, e scoperto il nero rimane una pittura, 0 disegno, che vogliamo dire, co' suoi chiari e scuri, che avitata can alcuni 。」アオセイ・ャゥ@ $Curetti 1 un bel rilievo, e fa bellissima vista." Filippo Baldinucci. Vocobolario toscano !WU' am del drugno (norenct: Santi Franchi, 1681), 151 (my emphasis). 4.5. H. Sumner. ·OfSgraffito Work: in Arts and Crofts Essa!l5. by Membm of the Arts and Crafts £rhibition Societ!l: With a Preface by William Morris, Artsand Crafts Exhibition Society (London: Longmans, Grll't'n and Co.. 19(3). 161-71. 46. Vasari. Le vite, 3:766. Marabattini disagrll't's with Vasari on the quality of Polidoro's paintings, though he agrees on his qualities asdisegnatoTl!. On Polidoro's distgrlO, see Marabattini. Polidoro do Caravaggio, 14. 47. For a transcription of the de<ree. set' G. M. Urbani de Gheltof. Degli araui in Venezia con note sui teuuti atrtistici veneziani (Venice: Ferdinando Ongania, 1878),104-.5. 48. Erwin Panofsky, "Excursus: Two f。セ、・@ Designs by Domenico Be<cafumi and the Problem of Mannerism in aイ」ィゥエ・\ュセ@ (1930), in Meaning in VislUll Aru(Garden City, NY: Doubleday Anchor Books, 1955), 2.26-35. 49. Jacob Burckhardt, Gtschichteder Renoil:sance in Italien (Stuttgart: Ebner and Seubert, 1878; rev. ed., Munich: Be<k, 2000): "1m XV Jahrhundert war sowohl der edlere Prachtsinn au die Lust am hOchsten Putlund Prunk gewaltig gestiegen ... und eine fluchtige Uebersicht def wichtigeren Nathrichten ... wird zeigen welch ein Feld dieser Kunst offen キ。イセ@ (287). For a development of this topic in recent scholarship see llichard Goldthwaite, Wealth and the Dtmandfor Art in Italy MO: Johns Hopkins University Press. 1993). 13-40. An indication of the mrn toward an apprecia· tion of surfaces and artisanship is evident in Alberti·s definition of the origins of pleasure. which arises not only from inte!lected fonn but also from -the work of the hand- and treatment of material qualities (VI, 4). Alberti, On Building, 159. .50. Yuriko Saito. Ewryday Aesthetics (Oxford: Oxford University Press, l007). .51. Fernand Braude\, La Mediterraneeet It mondt mUitemmeen d npoque de Philippe If (Paris: Armand Colin, 1(49). 1300-1600 (Baltimore. Chapter 23 1. Anna Contadini, 'Sharinga Taste? Material Culture and Intelle<tual Curiosity around the Mediterranean. from the Eleventh to the Sixteenth Century," in The Renaissance and the Ottoman World. ed. Anna Contadini and Claire Norton (Farnham: Ashgate, 2013), 30. 2. Anna Contadini. "Artistic Contacts: in Current Scholarship and Future t。ウォULセ@ Islam and the Italian Renail:sance, ed. Charles Burnett and Anna Contadini (London: Warburg Institute, 1999), 9-lI. 3. Whether this is to be セョ@ as a centralization of the vocabulary of ornament during the Safavid period that would retle<t a political agenda is a matter of debate, and it is beyond the remit of this chapter. 4. For tirdz. Sll't' Anna Contadinl, Fatimid Art at the Victoria and Albert Museum (London: V&A Publications, 1(98), chap. 2. with relevant bibliography. Also rochen Sokoly, "Towards a Model CJf Early Islamic Textile Inst itutions in eァケーエNセ@ in Islamische Textilkunst des Mittelalters: Aktuelle Froblerne, Riggisberger Berichte. nCJ. 5 (Riggisberg: Abegg-Stiftung, 1(97). 5. For Franz Bock. セ@ Bil1litt Sorkopp. Resile, Der Aachtrltr KanDnikus From 8acIc und seine Tmilsllmmlunge:n: Eln Beitrag lUf Geschichte der lumstgewerbe im 19- /ahmun· tier! (Riggisberg: Abegg·Stiftung. :zOOS). 6. London, Victoria and Alben Museum (V&A). inv. no. 8560'1863. See Contadini, FotimidAr!, 6:z, pI. 16. 7. R. H. Pinder-Wilson and C.N.L Brooke, Jhe Reliquary of St. Petroc and セ@ in Pinderthe Ivories of Norman sゥ」ャケL Wilson, Studies in Islamic Art (London: Pindar Press, 1985; first published in Archaeolagio 104 [1973[: 261-305); Antony Eastmond, 'ihe SI. Petroc Casket, a Certain Mutilated Man, and the Trade in Ivories," in Siculo-Arabic Ivories and Islamic PointingllOO-IJOO. ed. David Knipp (Munich: Hirmer Verlag. 2011). S. Knipp, Siculo·Arobic ll/Ories. 9· Emma Z\Xca, Clita/ogo deUe co,w dane e di antichitd di Amsi (Rome: Libreria dello Stato, 1936), 203. fig. at 20.5; Kurt Erdmann, "Islamische Bergkristallameiten: lahrbuch der Preussischen Kunstsommlungen 61 (1940): 128-30 and fig. 3; Francesco Gabrieli and Umberto Scerrato. Gli Arabi in Italia (Milan: Garzanti·Scheiwiller, 1979), no . 520; Anna Contadini, iranslocation and Transformation: Some Middle Eastern Objects in Europe," in The Power of Things and the F10w of Cultural Transformations, ed. Lieselotte E. Saurma-[eltsch and Anja Eisenbeiss (Munich: Deutscher Kunstver· lag, 2010). 43-46, pI. 1.1 and fig. 1.1. 10. As the Geniza documents testify; see S. O. Goi tein, A MediterraneanSociety: Tht! fewish CommunitiL'$ofthe Arab World as Portrayed in the Documents of the Cairo Gtni:ll, 6 vols.: vol. I, Economic Foundations (1967); vol. 2, Tht! Community (1971); vol. 3, 1ht Family (1978); vol. 4. Doily Ufe (1983); voL.5, Tht! Individualh98S); vol. 6, Cumulative Indicts (1993) (Berkeley: University of California Press, 1967-93; reprint (paperback), Berkeley: University of California Press, 1999). II. Deborah Howard. Veniceond tM East: The Impact of the II/omic World on Venetian Arr:hitedul'\!. 1100-)500 (New Haven. C'r: Yale University Press, 20(0), 59-62; Catarina Schmidt Arcangeli and Gerhard Wolf, Islamic Artefacts in the Mediterranean World: Trade. Gift Exchange and Artistic Transfer (Venice: Marsilio, 2010); see also Julian Raby. "ExCJtita from Islam." in The Origins of Museums: The Cabinet of Curiosities inSirteenth· and セ|ャエョ・ィMcオイケ@ Europe, ed. Oliver Impey and Arthur Macgregor(Oxford: Clarendon Press, 1985). 12. For a discussion on the importation ofItalian textiles in the Ottoman courts, see Nurhan Atasoy et aI., lpek: Tht! Crescent and the Rose; Ottomlln Imperial Silks and VelwlS (London: Azimuth Editions, 200l), 182-90, where some CJf these documents are discussed on 185-86; セ@ also Nevber Gursu, The ArlofTurkish Weaving: Designs throl'9h the Ages, ed. William A. Edmonds (Istanbul: Redhouse Press, 1988). 28; Carlo Maria Suriano and Stefano Carboni, La seta islamica/lslamic Silk (Florence: Museo del Bargello/9th International Conference on Carpets, 1999), no. 2.5. Examples of Ottoman-made caftans in the Topkapl Palace indude one that dates to the late NOTES TO PAGfS 281- 293 399 fifteenth century (inv. no. 13/0) and another from the first half of the seventeenth century (inv. no. 1)/1909). 13. Venice, Chiesa di Santa Maria Gloriosadei Frali. Set Stefano Carboni. ed., Venice and the Islamic World. 828-1797 (New York: Metropolitan Museum of Art; New Haven. CT: Ya le University Press, 2007; French 1St ed., 2006), cat. 70. iセN@ Museo Nuionale del Bargello. Florence, inv. Franchetti 639. See Suriano and Carboni. La seta is/arnica, no. 25; Anna Contadini, "Le- stoffe islamiche nel Rinascimento Italiano tfa it XV e i! XVI secolo." in /ntrecci Mediterranei: nte5ruta rome dizianario di rapporti economid. culrurali e sociali, ed. Daniela Degl"lnnocenti (Prato: Museo del Tessuto. 2006), fig. 4: and Contadini, ··Sharing a Taste?," pI. 17. 15· V&A. iov. no. CIRe. 346·1911. 16. As Suriano and Carboni. La seta is/arnica, 85, note. "... by the end of the 15th century Ottoman velvets were already being made using silk for warp and (often) pile and cotton or linen for the weft.." 17. Genoa, Palazzo Rosso; see Contadini. 45-46, fig. 2.10. "Sharing a t。ウエ・セLB@ 18. Madrid, Museo!U1 Prado, inv. no. P00441. Fordiscussions of the 、ョエセゥ@ motifin the context oflslamic and Western art, see Priscilla Soucek. Bcゥョセュ。NイエM in eョ」ケOッセ、ゥ。@ /ronica, ed. Ehsan Yarshater (London: Routledge and Kegan Paul. 1991). vol. 5. fase. 6; Jaroslav Folda, "An Icon ofthe Crucifixion and the Nativity at Sinai: Investigating the Pictorial Language of Its Ornamental Vocabulary; Chrysography, Pearl-Dot Haloes and <;intemani," in In Laudem Hierasolymitani: Srudies in Crusades and Medieval Culture in HonourofBenjamin Z. Kedar, ed.lris Shagrir, Ronnie Ellenblum, and Jonathan Riley-Smith (A1dershot: Ashgate, 2007). 170-79. 19. Contadini, "Sharing a Taste?," 45-46. 20. Louise Mackie. セ@ Sp/endlJrof Turi:ish Wt(lving: An Exhibition of Silks and Carpets of the IJth-18th Cenruriu, Nowmllfr 9, 197J through March 24. 1974 (Washington. DC: Textile Museum, 197J-74), 14; Walter セョケL@ B・クエゥャウNセ@ in Tu/ips. Arobesque.sll Thi"Vans; Dtccroliw Arts from the Ottoman Empire, ed. Yanni Petsopoulos (New York.: Abbeville Press, 1982), 128; Gursu, The Art ofTurldsh w・。セゥョァL@ 43, 67-68; Atasoy et aI., Iptk, 208, 227. On the transmission of Chinese motifs into Iranian and Turkish art, see Jessica Rawson, Chinese Or1lllment: The Latus and the DrogOll (London: British Museum, 1984), 145-98. 21. V&A, inv. no. 100·1878. Although it .00 NOTES TO PAGES 29) -296 is currently not possible to distinguish ber,o.·een the products of different centers (I. M. Rogers, ed. and tran$., 1ht Topkapi Saroy Museum.: Costumes. Embroideries. and Other Textiles, from the original Turkish by HUlya Tezcan and Selma セャゥ@ [London: Thamesand Hudson, 19861. IS), fCluno weaving was particularly associated with Bursa. whereas the Istanbul ateliers appear to have specialized more in brocaded silks and cloths of gold and silver. セョケL@ Bt・クエゥャウLセ@ 12.4; Gursu, セ@ Art a/Turkish Weaving, 19; Atasoy et aI., Iptk, 156. 22. For a discussion of the floral motifs in Ottoman art, see J. M. Rogers and Rachel Ward, SUleyman the Magnificent (London: British Museum Publications, 1988), 60; and Nurhan Atasoy and Julian Raby, tznik: The Pottery o/Ottoman Turkey, ed. Yanni Petsopoulos {London: Alexandria Press in association with Thames and Hudson, 1989),222-23· 2J. Marco SpaHanzani, "Le compagnie Salili a Norimberga nella prima metA del Cinquecento (un primocontributo dagli archivi fiorentini).8 in Wirtschaftskrafte und Wirtschaftswtgt; Festschrift for Hermann KtUenrenz, VOLI. Mirte/metrund /(ontinent, ed.lurgen Schneider (Stuttgart; KlettCotta, 1978), 609, 610, fig. I; Atasoy et aI., tpek, QXセN@ 24. Venice, Museodi Palauo Mocenigo. inv. no. 491/191. See Degi'lnnocenti, fntrecci Mediterranei, 80-81, cat. 14; Contadini, セsィ。イゥョァ@ a Taste?: pI. 15. 25. Museo Nazionale del Bargello, florence, inv. Franchetti 99. See Suriano and Carboni, La seta Isla mica, no. 35. Contadini, "Le stoffe,"' fig. 2. 26. See Gulru n・」ゥーッセオL@ "From International Timurid to Ottoman: A Change of Taste in Sixteenth ·Century Ceramic TIles,8 Muqamas 7 (1990): 155. 169n49; n・」ゥセャオL@ "Connectivity, Mobility and 'Portable Archaeology': Pashas from the Dalmatian Hinterland as Cultural Mediators," in Dalmatia and the Medittrra· nean: Portoble Archae%gy and the Poetfa of /nflwmce, ed. Alina Payne (Leiden: Brill. 201 4),3531164. 27. V&A. inv. no. 553-1865 and UセᄋQXVN@ See Anna Contadini, "Middle-Eastern o「ェ・」エウLセ@ inAt Home in RenawanCl Italy, ed. Marta Ajrnar-Wollheim and Flora Dennis (London: V&A Publications, 2006). 313-14. 360, cat. 135; Contadini, "Sharing a Taste?," fig. 2.14. 28. See note 55, below, for references. 29· V&A, inv. no. 4301·1857. See A. S. MeHkian·Chirvani, Islamic MetallllOrkfrom the Iranian World, 8-18th Centuries (London: Her Majesty·s Stationery Office, 1982), 321-32, no. QセV@ and the entry by A.R.£. North in Europa und der Orrent 800-1900, ed. Gereon Sievernich and Hendrik Budde (Berlin: Berliner Festspiele, Bertelsmann Lexikon Verlag, 1989), 606, no. 4/104. 30. The bronze (gun metal) candlestick was called by Melikian ·Chirvani a セ」ッューウェエ・NBャョ、L@ it is not only made of two pie<:es, but the upper piece, in the form of a glass. is different in style of de<:oration, and al$O, it seems, in metal composition, probably brass. 31. British Museum, OA 78.12.30.735, Henderson Bequest, H., 37.8 cm. Sheila Canby. The Goldlm Age of Pe rsian Art QUPMWセR@ (London: British Museum Press, 1999), 110, col. pI. 98. )2. Two silver beakers of this type are known, both made by the metalworker Johann Adolf Lambrecht ca. 1675, one in the Kremlin in Moscow, the other in a private collection in Hamburg: see Bernhard Heitmann, "Migration and Metamorphosis: The Transform.ation of Shapes, Ornaments, 8 and Materials. Metropolitan Mustum Journal 37 (200;r:): 112 and fig. 9. 3J. The beaker is from northern Bohemia (or northern Dechoslovakia). an area with a long tradition of glassmaking. Strasser Collection. Vienna: see Heitmann, "Migration," 112 and fig. 10. 34. Heitmann, "Migration," ll2, who suggests the connection (although does not give a comparative example). 35. These sets are in Washington DC, National Gallery of Art, 1961.9.186·194. New York, Metropolitan Museum of Art, 17.190.926·961; and the Schrode r Collection. See Alison Luchs, "Costume Ornaments with Profile Portraits," in Western DecoratiIltArf$, Part 1: Medieval, Renai.lsance. and Historicizing Style5 Including Metalwork, Enamels. and Ceramics, ed. RudolfDistelberger et al. (Washington. DC: National Gallery of Art, 1993); Timothy Schroder, Renaissance SilM'r from the Schroder Collection (London; Wallace Collection. 2007). cat. S. 36. For example, see Arma Ballian. "Three Medieval Islamic Brasses and the Mosul Tradition of Inlaid Metalwork," Movm:lo MIrCY<l:/C1f9 (;r:009): 121; Julian Raby, "The Principle of Parsimony and the Problem of the 'Mosul School of Metalwork:" in MetallllOri: and Material Cu/rure in the Islamic World: Art. Craft and Tt.rt; Essays Presented to lame5 W. Allan, ed. Venetia Porter and Mariam Rosser·Owen (London: J. B. Tauris, 2012), 44-52; Anna Contadini, A World of Beasts: A Thirteenth ·Century lIlustrated Arabic Book tm Animals(the KitAb Na' t al-ijayawAn} in the fbn lklkhtis/ul' Tradition (uiden: Brill, 2(12), 149-51, and also chap. 8. esp. 161-62.. 37, D. S. Rice, "Inlaid Brasses from lhe Workshop of AI:Imad al- Dhakl aI· m。セゥャLB@ ArsOritntalis 2. (1957): 32.3, no. 6; alsosee Raby, "The Principle of Parsimony," 45. Transmiuion from books to metalwork does not only involve decoration but also illustrations, as seen on the Masuli candlestick in ャィセ@ Khalili Collection that represent, among other things, a scene of a teacher with pupils writing on tablets, a tableau that can be identified only through knowledge of the illustrations of the manuscripts of the early and mid· such as in one thirteenth century m。アセュエN@ probably copied in Syria, dated 619 (lll l), now in Paris, bゥ「ャッエィセアオ・@ Nationale, Ms. Arabe 6094, fol. 167r; see Anna Contadini, "Ayyubid Illustrated Manuscripts and Their North Jazlran and 'Abbasid Neighbours," in AlfYubid /erusukm: The Hf1ly City in Context 1187-12.50, ed. Robert Hillenbrand and Sylvia Auld (London: AltajirTrust, 20(9), pI. 9.4. 38./ames W. Alian, "Venetian·Saracenic Metalwork: The Problems of Provenance," in Arte vtnu;arw e ortt iswmica: Art! del primo simposio internozionale 1U//ilrte wnuiano e /'o11e is/amico, ed. Ernst J. Grube, Stefano Carboni, and Giovanni Curatola (Venice: L'Altra Riva, 1989): and Sylvia Auld, "Master Mahmud and Inlaid Metalwork in the 15th Century," in Venice and the Islamic World. ed. Carboni. 2.18-19. 39. James W. Allan, "Chinese Silks and Mosul Metalwork," in Court and Craft: A Masterpiece from Northern Iraq, ed. Rachel Ward (London: Caurtauld Galle ry in association with Paul Holberton Publish· in&, RNPQセI@ 40. Oleg Grabar, "Reflections on Mamluk Art," in "The Art oflhe Mamluks," speclall$Sue, Muqarnos 2. (1984)' 7, Grabar, howevtr, cautions against attributing ornamental features that were common dUrin& Qaytbay's rtign as a style, as lhey wert not tKClusive to this period. 41. London. V&A, inv. no. 132.5-1856. See A. S. Melikian-Chirvani, "CuivIeS ゥョセエウ、・@ iGセーッアオ・@ de Qil 'itWy," Kunstdes Orients 6, no. 2 (1969): fig. 2.8; also Tim Stanley et al .. Polacr and Mosque: Iswmie Art from the Middle East (London: V&A Publications, 2004), fig. 1Il. 42.. M, B, PiotrovskiI, and T. N. Kosourovil, The Magic World of the GrotiSqut: 16th- and 17th·Century Grotesques in the Applied Art ofWesrern Europt! from the Hermitage Cf1lleclwn; Catalogue (Saint Petersburg: Siaviia, 2.000). 43. For brief discussions on thtir ornamental drawin&s, set Stuart W. Pyhrr and Jose·A. Godoy, Heroic Arrrwro{IM Italian Rtrwil:roncr: Filippo Negroli ami His Contemporaries (New York: Metropolitan Museum of Art, 1<)98), 109"10: and Marjorie Connell, "Pietro del Vaga," in Desig/lS of Desirt: Architectural and Ornamental Prints and Drawings 1500-1850. exh. cat., ed. Timolhy Clifford (Edinburgh: National Galleries of Scotland, 1999). 44. For an overview of ornament draw ings and prints during the Renaissance, see Janet S. Byrne, Renaissance Ornament Prints and Drawings (New York: Metropolitan Museum of An, 1981), 11-2.1. For example, a print by Brosamer, first now in the V&A, E.2.3S·1914 published Q UセッL@ (this copy printed ca. 1570). shows designs for two cups with the ornamental motif of lhe acanthus leaf and molded decoration. 45./. M. Rogers, "Ornament Prints, Patterns and Designs, East and West," in Islam and the Italwl! Renaissanct, ed. Burnett and Contadini. 46. David James, "More Qur 'Ansofthe Mamluks," Manuscripta Oritntalia 13, no. 2 (2007): 7-8. セWN@ Cairo, Dilr al·Kutub, no. 11. Martin Lings and Yasin Hamid Safadi, The Qur 'lIn (London: World of Islam Publishing for the British Library, 1976), no. 88, 48. Sofia, SS Cyril and Methodius National Library, OP 2707. Z. [vanova and A. Stoilova, The Holy Qur 'dn through the Cen turies, A Catalogue of the Exhibition of Manuscripts and Printed Editions Preserved in the 55 Cyril and Methodius National Library, Sofia, February 1995. Sofia: SS Cyril and Methodius National Library, Centre for Manuscripts and Documentation, Oriental Department. 1995. 49-50, cat. 2.; James, "More Qur 'ilns," . "5. 49. Thpkapl SaraYI Museum, 2.12107; Zeren Tanmdl, "Bibliophile Aghas (Eunuchs) at Topkapl Saray," in "Essays in Honordl. M. Rogers," special issue, Muqarnos 2.1 (2.004): 338. 50. Marina Btlouf$kaya, Lu.r1lry Arts of the RellOIsmnCf (London: Thames and Hudson, 2.005), 180. 51. However, now that the scholarly COI15!'nsus is that lhey were not made in Venice, this tenn Is best avoided. For a review of the scholarship on them, set Sylvia Auld, Renaissance Venia, Islam and Mahmud the Kurd: A Metal1llOrk;ng Enigma (London: Altajir World of Islam Trust, 20(4),7-8.36-43; Doris Behrens·Aboustif, -Veneto·Saracenic Metalware. a Mamluk Art," Mamluk Studies Review 9, nO.:I (2.005) (who has argued that all the pieces come from Mamluk Egypt); and Contadini, "Middle-Eastern Objects," 309-15. where lhe term -Veneto·Saracenic" is avoided. 52.· For examples of these, see Auld, RenaissallU Venict, 108-40. 53. The Iranian provenance was already suggested by Rachel Ward, Islamic Mttal1llOrk (London: British Museum, 1993), IOl-3. Auld puts forward the hypothesis that these masters might have been itinerant Aqqoyunlu Turkmen working in and around northwest Iran or Anatolia, on the grounds of stylistic comparison with early Ottoman andAqqoyuniu material: Auld, Renaissance Venice , 8-9, and chap. 7, see also Auld, "Master Mahmud," 218-19. 54. Auld, Renaissance Venice, 60. 55. Although other Islamic metal ..... ork of lhis type contains high levels of nickel. For the scientific analysis of these objects, set Rachel Ward et aI., WVeneto-Saracenic Metalwork: An Analysis of the Bowls and Incense Burners in lhe British Museum," in Truth and Discollfry, The nntific Study of Artefacts from Post·Mefiiellal Europeand Beyond, BM Occasional Paper 109, ed. O. R. Hook and D.R.M. Gaill1$ter (London: British Museum Pre$S, 1995): and Susan La Niece, "Master Mahmud and Inlaid Metalwork: A Scientific Perspective," in Venia and the Islamic World, ed. Carboni; see also Auld, Renaissance Venice, 60-61. 56. The Roman transliteration is not, as often reported in the literature, a Persian version, "AMALEI MALEM MAM UO" (for example, B. W. Robinson, "Oriental Metalwork in the Gambier-Parry Collection," Burlington MfUJozine 109, no. 768 [March 19671: 170-73: Auld, Renaissam'e Venice, and Auld, セm。ウエ・イ@ Mahmud," cat. no. 103; and Rosamond E. Mack, Bazaar to Piazza: Islamic Trade and /talian Art. 1300-1600 [Berkeley: University of California Press, 2002.], 214017), but rather "AMAL ELMALEM MAMUD.· from lhe Arabic inscription on the olher side of lhe rim: "lhe work of the master Mahmud who hopes for forgiveness from his lord" ramal aI·mu 'allim ma.1tmUd yarju al-ml19hfira min mauldhi). 57. Marco Spallanzani, Metalli iswmid a Firtnze nel Rinascimento(Florence: Studio per Edizioni 5<:elte, 2(10), 11-12, and n. RNセ@ 58. Marco Spallanzani points out (ibid., 7-10) that lhe Florentine documents of the fourteenth century that refer to metal cargoes from a port of the Near East fail to give any further specification. They cast NOTES TO PAGES 296 -291 401 light on other aspects such as prices and usage, and even sometimes refer to ornamental motifs, but not to places of origin. However, thert is mention of a back·and-forth mOVf.'mf.'nt of objf.'cts to bf.' decorattd in the Middle East. 59. V&A. inv. no. mNSBャMQYセV⦅@ Contadini, MMiddle·Eastern o「ェ・」エウNセ@ )10-11, 356, cat. 62. fig. 21.2. 60. See Contad ini, MSharing a Taste?" fig. 2.13 for the detail. 61. In the entry by A.R.E North in Sievernich and Budde, eds., Europa und der Orient, 601, no. セOYWL@ it is stated that the ewer was crafted in Europe (Netherlands?) and decorated in an Islamic workshop. either by a Middle Eastern craftsman in Venice or, more probably. in Cairo. In the V&A catalogue of 1951 (Fifty Mo.sterpiece.<; of Meta/work), it is stated that the ewer would have reachtd Venice from the -Low Countries or Germany" in the fift\'"!'nth century and been decorated there by a group of Venetian craftsmen, while its RPセ@ publication (Sta nley et aI., PaltIct and Mosque, QRWMセL@ fig. 152), $IIYS that it was a Late Gothic ewer from the Netherlands or Germany, which was sent to the Middle Easl, probably by a member of the Molino family, for the inlaid ornament to beadded before it was re-exporttd back 10 Italy. 62. V&A, inv. no. セRPMQXUN@ Slanley et aI., l'uillCl and Mosque, fig. 105. 6). For these shields, see Anna Contadini, "'Cuoridoro': Tecnica I.' decorazione di cuoi dorali veneziani e italiani can influssi Isla mid," inArte veneziano, ed. Grube, Carbon, and Curatola, 231-51. In Contadini, "Middle· Eastern Objects," 3"lO-21. some of these shields are published in color. 64. He was invited by a certain Nicol/) Drasdovich of the Signoria. For "targhe aU'usanUi di Ccovatia, percM quelle chI.' si facevano in questa dttA, 1.'1 a Modena, non $Olamenle non aggiongevanodi gran 10nga alia perfettionedi queste. rna buona parte di quelle sonostate conosciule inutele da faltione. Vtduto poi con I'occhio proprio Ie targhe, chI.' da sopradetlo sono state fatte in questa casa per mandar in Cipro.le qual, oltra chI.' sono laudate da periti, lIOn costano pili delle modenese, 10 riputiamo perfetto et perito maestro non sola mente di far et depinger targhe, ma etiam di far una bella sorte di lancie da cavallo buse innervate pili longhI.', pili legieri, et pili forte delle altre, che sono massieI', Ie quali reputiamo habbino ad esser molto a proposito nelle fattioni per l'avantaggio della longhezza. Per/) essendo V. S.tA di parerI' di far una 402 NOTES TO PAGES 297-305 .. quantila di questa sorte di targheet lancie dadispensardove farA bisogno. massime alle cavalieriI.' de stradiotti chI' si trovano sopra Ie sue isole et fortezze da mare, 51 come sopra !'isola di Cipro セ@ Sialo novamente introelotto." Ar<:hivio di Stato di Venezia, Senato Mar, fiIza iセZ@ incartamenlo November 29, 1560. and May RセL@ 1561.1 thank Luca Mol.1i who hasgiven me the opportunity to mention this document here. 65. Chicago, Newberry Ubrary, Wing MS ZW 1.575. See Grube, Carboni. and Curatola, Arte wneziana, cover; Mack, Bazaar, fig. IH and Ernst J. Grube. "Venetian Lacquer and Bookbindings of the Sixt\'"!'nth Century," in Venice and the [slamic World, ed. Carboni, fig. I. 66. By Nicolas Jenson HjセiッMXIN@ Paris, Bibliotheque National, vセャゥョウ@ QiセYZ@ published in Anthony Hobson, Humanists and Bookbinders: The Originsand Dijfu.swn of Humanistic Bookbinding. iセ sYMi@ (Cambridge: Cambridge University Pre$S, 1989),40-41, fig. H. pI. I; see also discussion in Alison Ohta. "Binding Relationships: MamJuk. Ottoman and Renaissance Book-Bindings," in 1hf Renaissonceond the Ot/omon World. ed. Contadini and Norton, 1I6. 67. MS M. 859. Hobson, Humanists and Bookbinders, 16. and fig. 9. 68. Venice, Bibliote<:a Marciana, Lat.X, ll7 (·3844). Hobson, Humanists and Bookbinders, 43, figs. 37. 38, pI. I; Mack, Bazaar, 129, fig. 1)5; Ohta, -Binding Relationships," pI. セRN@ 69. O. Granzotto, "A1cune note su Felice Feliciano Legatore," in L·"Antiquario· Felice Feliciano wronese. tra epigra/ia antica, letteroturo e arti dellibro: Alti del Convegno di Studi, Verona, 3. giugnOl993. I'd. A. Cont/) and L. Quaquarelli (Padova: Antenore, 1995); see also Rogers, MOmament Prints," 139. 70. Oxford, Bodleian, Ms.Canon.lta1.78. Hobson, Humanists and Bookbinden, 13, figs. 15, 16; Ohta, "Binding Relationships," fig. II_I. 71. Ohta. "Binding Relationships." II3. 7I. lhis indicates toAiison Ohta that the manuscript was bound in the Mamluk region (rather than Istanbul as suggesttd by Hobson). Ohla, "Binding Relationships," RSMセ[@ Hobson, HUmllnists and Bookbinders, 13-24. 73- Venice, Bibtioteca Marciana, Lat. VI 270 ("3671). Hobson, Humanists and Bookbinders, 22-13; also Ohta, "Binding Relationships," 223-Z4. 74. Hobson, Humanists and Bookbinders, QセXN@ The AldinI.' Catullus is in the Vatican Ubrary, AldinI.' II1.19, and the Horae is in a private collection; both are published in Hobson. Humonists and Bookbinders, figs. 111 and 118, rt'5pectively. 75. Hobson. Humanists and Bookbinders. II, QセYMU[@ alsoOhta, MBinding Relation· ships," 22). 76. Giocondo: Chatsworth Library, Chatsworth House, Derbyshire. Hobson. Humanistsllnd Bookbinders. 151, fig. ll9, pI. 3: Ohta, "Binding r・ャ。エゥッョウィーNセ@ fig. n.). Ghiyath ai-Din: Istanbul, Museum of Turkish and Islamic Arts, MS. 2031, formerly MS. ISOI. Julian Raby and Z. TaOlndl, Turkish Bookbinding in the 15th Century: The Foundation of an Ottoman Court Style (London: Azimuth Editions, 1993), cat. 33. n. Contadini, "Sharing a Taste?." fig. I.15. 78. Sue Budden, trans., Arabesques: Decoro/illt p。セOウ@ oftM Renaissance (Paris: Bookking International, 1995), 12. 79. Prato, Museo del Tessuto, inv. no. 97.01.M. Set' Degl·Innocenti.lntre«i m・、ゥエュjセL@ 76-n, cat. II_ 80. For a discussion of the lenn Ollll domoschina, set' Valentina Catalucci, -Gli oggetti 'islamic!' a Firenze nell'eu della 」ッョエイ イゥヲッョ。 Lセ@ in cッョエイカ・セZ@ Dispute letterarie, storiche, religiose dal/(mtichitd 01 Rino.scimento, ed. Gloria Larini (Padova: libereriauniversitaria.it edizioni, 2013). Note that as Marco SpaJlanzani observes, the term alia domaschina applies not only to objects in the lslamk style being made in Europe but also to those imported from the Middle East; see Spallanzani, Oriental Rugs in Renaissance Rorence, Bruschettini Foundation for Islamic and Asian Art, Textile Studies, no. I (Florence: Studio per Edizioni Scelte, 2(07), 60, 67- 68. 81. "Non gli vogJio a ruote": SpallalUani, Oriental Rugs, 63. 8z. Sabba da Castiglione, Ricordi (Venice: Paolo Gherardo, isUセIN@ 8). Boston. Isabella Stewart Gardner Museum, P1seB. Alan Chong, -Seated Scribe, QセWYMXLB@ in Bellini and the Ellst, ed. Caroline Campbell and Alan Chong (London: National Gallery Co.; Boston: Isabella Stewart Gardner Museum, I005). XセN@ Istanbul University Library, fiセRL@ fol. l"ri. Alan Chong, MGentile Bellini in Istanbul: Myths and Misunderstandings," in Btllini and the EIlSt, ed. Campbell and Chong, ll2-13, fig. 40. 85. Contadini, "Artistic Contacts," 7-8. 86. Ibid., 10-ll. 87. Castiglione. Ricord!, chap. 109, on "Cerca gli Ornamenti della Casa," 53: "Alcunl altri apparano セ@ adornano iセ@ lora ウエ。イョZセ@ di panna di razza I' di celani venuti di Fiandra, fatti Aヲゥァオイセ@ セ@ Afogliami, echi a セ@ カ・イ、オセL@ chi can tepeti I' mO$Chetti エオイ」ィセウゥ@ I' soriani, セ」ィゥ@ con earpette e ウー。ャゥセイ@ barbaresche, chi di エ・ャセ@ di mana di booni maestri, chi con corami ingegnosa· mente lavorati カセョオエゥ@ di Spagna. セ@ aleuni allr! con COSt' ョオッカセ@ fantastiche e bizarre, rna ingegniOSt' venutI' di levantI' 0 d'Alemagna, sottile inventrice di moltecose belle e artificiose I' tutti questi ornamenti ancora commendo I' laudo, perthI' argulscono ingegno, politezza, civiltA I' cortegiania." Chapter 24 I. Edmund L Sterling, HistufY uf Hendersun Cuunty. Kentucky (Henderson, KY,IB87),15o · 2. By the later nineteenth century, most national governments had intervened to unify and nationalize their paper currency. See Eric Helleiner, The Making of National Money: Te"irorial Cummcies in Historical Perspectiw (lthaca, NY: Cornell University Press, :1(03), Q YMセQN@ 3- Stephen Mihm,A Natjqn afCaunterftiters: Capitalists, Con Mtn. and the Making afthe United States(Cambridge, MA: Harvard University Press, 2007),3. 4· Ibid.,). S. On these serials, sef! William H. Dillistin, Bank Nott Reporters and Counterfeit Detectars, 1826-1866. with a Discourse on Wildcat Banks and Wildcat Bank Notes (New York: American Numismatk Society, 1949). 6. Jane Kamensky has written eloquently of the "reformulation of distance" that emerged from the geography of paper mOnty. See Kamensky, The &changt Artist: A Tale of High-Flying Speculation and America's First Banking Callapst (New York: Viking. 2,ooS), 52- 72,. 7. Helleiner, Making of National MontY,3 1. 8. On Asa Spencer, see Greville Bathe and Dorothy Bathe, Jacob Perkins: His Inlltntions. His Times, and His Contemporuries (Philadelphia: Historical Society of Pennsylvania. 19-43), 72-7), 108; and セb。ョォ@ Note EngraVing," Franklin Journal and Amtrican Mechanic's Magazine 2, no. 2 (August IS26): 106-S. On Cyrus Durand (brother of the important American artist Asher B. Durand), see Alan A. Siegel, Outof Our Past: A History ofIrving/ull. New Jmey (Irvington, NJ: Irvington Centennial Committee, 1974); "Bank· Note Engraving in aュ・イゥ」。Lセ@ llIustroted Magazineof Art 3 (IS54): 30S-12; セcケイオウ@ Durand,the Machinist and Bank· Note Engraver." illustrated Magazint of Art 3 (ISH): 267-70: and セhゥウエッイケ@ and Progress of Bank Note Engraving," Croyon I, no. 8 (ISS5): 116-17. For a discussion of the relationship between Cyrus's banknote engravings and his brother Asher's paintings, see Jennifer L Roberts, Tronsporting Visions: The Mowmtnt uflmagts in Early America (Berkeley: University of California Press, RPQセIL@ 119-37· 9. See A. D. MackenzIe, The Bank of England Note: A History of Its Printing (Cambridge: Cambridge University Press, 1953),47- 107. 10. Laban Heath, Heath'S Infalliblt Counterfeit Deteetllfat Sight (Boston: L Heath,IS64),9-1I. II. For details on this lathe, see the object description on the New-York Historical Society's website at www .nyhistory.orglnode/15026, accessed August 31, 2013. 12. ForgeneraJ discussions ofanticoun· terfeiting patterns in banknote engraving, see Frances Robenson, -rite Atsthetics of Authenticity: Printed Banknotts as Industrial Currency," Technology and Culture 46 (January 2,OOS): 31-50; Stephan for Security," Wilkinson, セo・ウゥァョ、@ Connllweur 210 (April 1982): 24-26; Basil Hunnisett, "lhe Quest for the Unforgeable Document," in Engrawd on Steel: The History Ilf Picture Production Using Steel Plates (A1dershot: Ashgate. 1998), 30-62: and Granvil!e Sharp, The Gilbart Prize Enay on the Adaptation of Recent Discoll4lries and Inventions in Science and Art to the Purposes ofProctical Banking, 3rd English ed. (London, 1854). 13· "Bank Note Engraving," 107. 14. Jacob Perkins, Gideon Fairman, and Charles Heath, セpイ・カョエゥッ@ of Forgery," Tronsactionsofthe sッ」ゥセエケNャョウオ・、@ at London. for 1M Encuu1'Ogemtnt of Arts. Manufactures, and CommelU 3S h82,1): <47-56. 15. Period discussions dwelled frequently on the engraver's impotence tn the face of the superhuman perfection of the geometrical lathe: "lhe engraver cannot imitate the labour of the geometrical lathe" ("Bank-Note Engraving in America." 310). One might argue that rather than attempt to make manual copies of bank note ornament, counterfeiters would need only to get a hold of a lathe. But the lathes and their associated presses and transferpresses were extremely expensive, bulky, and noisy, making them difficult to acquire and nearly impossible to conceal from authorities. 16. John Holt Ibbetson.A Ptactico/ View afon lnl'l!ntion for the Better セエ・」ゥョァ@ of Bank Notes against Forgery, 2nd ed. (London. 182.1), 1. 17. Ibid., 1-2. IS. Ibid., 15-16. 19. Joe Conway, セm。ォゥョァ@ Beautiful Money: Currency Connoisseurship in the Nineteenth·Century United sエ。・ウLセ@ Nineteenth·Century Contexts 34. no. 5 (2012): 427-43. 10. Heath's Infallible Counterfeit Detectur.7. 11. Ibid., 8. 22. Edgar Allan Poe, !he Daguerreo. type." Alexanders Weekly Messenger (Philadelphia), January IS, 1840: "For, in truth. the Daguerreotype<! plate is infinitely (we use the term advisedly) is infinitely more accurate in its representation than any painting by human hands.lfwe examine a work of ordinary art, by means of a powerful microscope, all traces of resemblance to nature will disappear-but the closest scrotiny of the photogenic drawing discloses only a more absolute truth, a more perfect identity of aspect with the thing represented." 'Z3. On the process of steel-engraving and transfer (siderography) see Perkins, Fairman, and Heath, セpイ・カョエゥッ@ of Forgery." 41-56; "Bank Note Engraving." 107; and Mark D. Tomasko, The Feel ofSteel: The Art and History of Bank note Engraving in the United States (Newtown, PA: Bird and Bull Press. 1009), IS-20, 75-76. 2,4· Tomasko, Feel ofSteel. 75. 25. See Walter Benjamin. !he Work of Art in the Age of Its Technological Reproduc ibility," in The Work IlfArt in the Age offts Technological Reproducibility, and Other Writings on Media, ed. Michael William Jennings et al. (Cambridge. MA: Harvard University Press, 'ZooS), 19- 56; Gilles Deleuze, Difference and Rtpetitilln, trans. Paul Patton (New York: Columbia University Press, 1994); and Stephen Bann, Paralltl Lines: Prinlmaktrs, Painttrsond Plwtogrophers in Nineteenth·Century Franct (New Haven, CT: Yale University Press, 1001). 16. William Ivins, Prints and Visuol Communication (Cambridge, MA, MIT Press. 1953), 3. 27. Banknote engraving, which stood at the center of developments in reproduction generally in the nineteenth century, complicates the Benjamintan dictum about NOTES TO PAGES 108-11 5 403 FROM GLOBAL TO LOCAL Copyright CI 2016 by Prince ton University Press Published by Princeton University Press. 41 William Street. Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press, 6 Oxford Street. Woodstock, Oxfordshire aX20 ITW press.princeton.edu Jacket arl: (front) Top left: Cosimo Fanzago, decoration of door frame in the cloister of the Certosa di San Martino. Napl('s. Top right: Binding, Fakhr aI-Din al- Iraqi. AI-Lama -jjt (Istanbul, 881/!4n). Museum of Turkish and Islamic Arts (MS. 2031. formerly MS. 1501), Istanbul. Bottom: Herzog & de Meuron. Eberswalde Technical School Library. 1994-99. Eberswalde. Gwnany. (back) Top left: Cristofano Gherardi. Palazzo Vinelli alia Cannoniera. 1534. Castello. Italy. Top right: Herzog & de MeuTon. de Young Museum. 2005. San Francisco. Middle: Herzog & de Meuron. Pfaffenholz Sports Centre. 1989-93. Saint- Louis, France. Bottom left: IJ€'taii from an illuminated page with margins decorated by a split palmette scroll. Signed by Hasan. Amir Ghayb Beg Album. Safavid Iran. bt>forr 1566. Topkapl Palace Museum Library (H. 2161, fiJI. 93u).lstanbul. Bottom and hOI/om right: IJ€'tail of a ninth-century stucco dado from Samarra showing Style C-typ" ornament. All Rights Reserved. LIBRARY Of CONGRESS CATALOGING-IN·PUBLICATION DATA Histories ofOrnamrnt: From Globa! to Local / e、ゥエセ@ by Gulru Necipoglu and Alina Payne: With contributions by Michele Bacci. Anna Contadini. Thomas B.F. Cummins, Chanchal Dadlani, Daniela delPesco. Vittoria Di Palma, Annr Dunlop. Marzia Fairtti. Maria judith Feliciano, Finbarr Barry Flood. jonathan Hay. Christopher 1', Heurr. Remi Labrusse, Gulru Necipoglu, Marco Rosario Nobile, Spyros Pap;Ipetros, Oya Panearoglu, Alina Payne, Antoine Picon. David Pullins, Jennifer L. Roberts. David J. Roxburgh, Avinoam Shalem, Hashim Sarkis, Robin Schuldenfrei, and Gerhard Wolf. pages em Includes bibliographical references and index. ISBN 978-0-691-16728-2 (hardcover: alk. papt'r) l. Decoration and omaml'nt. Architectural. J. Necipoglu. Gulru, editor. JI. iG。ケョセN@ Alina Alexandra. editor. NA3310.HS72016 WRYᄋPセ、cS@ 2015022263 British Library Cataloging-in-Publication Data is available This book has been composed in Vesper Pro Light and Myriad Pro Printed on acid-free paper. "" Printed in China 10987654321