Studies in the Islamic Decorative Arts 1904597505, 9781904597506

Citation preview

Studies in the Islamic Decorative Arts

Studies in the Islamic Decorative Arts

Robert Hillenbrand

Pindar Press London 2019

Published by Pindar Press 30 Wentworth Drive Middlesex HA5 2PU · UK

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library

ISBN 978 1 904597 50 6

Printed by Short Run Press Ltd Exeter, Devon EX2 7LW This book is printed on acid-free paper

Contents Prefacei I

Images of Authority on Kashan Lustreware 

1

II

The Sarcophagus of Shah Ismail at Ardabil

38

III

Problems in Islamic Pottery

63

IV

Persian Lustre Ware 

90

V

Richard Ettinghausen and the Iconography of Islamic Art

97

VI

Figural Calligraphy in the Muslim World 

111

VII

Introduction: The Orient of the Imagination

VIII

For God, Empire and Mammon: Some Art-Historical Aspects of the Reformed Dinars of ‘Abd al-Malik

132

IX

The Soul of Islamic Art 

171

X

The Syrian Connection: Archaic Elements in Spanish Umayyad Ivories

174

XI

Islamic Figural Sculpture

264

XII

Islamic Art: ‘Infinite Riches in a Little Room’? 

271

XIII

What Happened to the Sasanian Hunt in Islamic Art? 

281





122

XIV

The Major Minor Arts of Islam 

302

XV

Recent Work on Islamic Iconography 

314

XVI

The Islamic Reworking of the Sasanian Heritage: Two Case Studies

349

XVII

A Leaf from my Rose Garden: Themes in Persian Art

378

XVIII Oleg Grabar, Distinguished Historian of Islamic Art 

420

XIX

Oleg Grabar: the Scholarly Legacy

428

XX

Wine in Islamic Art and Society 

469

Index495 Acknowledgements519

Preface

O

ne of the consistent pleasures of working in the field of Islamic art is the never-failing opportunity to escape from one’s comfort zone and explore the comfort zones of others. This is something that is not on offer for the much more heavily populated — and policed — fields of, for example, most of Western art. It is a direct consequence of the fact that the study of Islamic art has been late in coming of age. So it has been slow to acquire the dauntingly long bibliographies that are the norm for so much of art-historical scholarship. Of course, that is not an invitation to indulge in superficial op-ed pieces or sloppy research. But it does leave the door wide open to explore all sorts of material that has remained virtually untouched in detail even if it has been known about for decades. Much earlier scholarship contented itself with staking out a claim as distinct from providing in-depth analysis of some new discovery. There are, of course, numerous possible approaches to analysing works of art — historical, technical, philosophical, political, legal, religious, economic, sociological, theoretical and aesthetic, to mention only some. Many a work of Islamic art has not received detailed attention on some of these counts, or indeed on most of them. The mere fact that an item has been published is often only the beginning of a long story. The papers reprinted here, then, are the result of occasional forays into areas somewhat removed from my principal research interests, which focus on Umayyad art, the architecture and book painting of the medieval Iranian world, and Islamic iconography. The inspiration for them was very varied — a book review that gradually grew into a review article, a new undergraduate course that propelled me in unexpected directions, the need to make a contribution to team-taught courses, an invitation to speak at a conference in a cognate field or write a preface to a book, or contribute to a Festschrift or a collection of studies round a central theme. And there were still other reasons to embark on an unexpected topic, such as the sudden death of an esteemed colleague, or the desire to define how earlier scholars had impacted on the entire field of Islamic art. The

ii

historiography of Islamic art is still in its infancy and indeed is not much more than a century old. It is all the more relevant as a topic of study in view of the exponential growth of the field at large over the past couple of generations. Several papers explore the eternal conundrum of what makes Islamic art Islamic, a question that provokes animated discussion among every new generation of scholars of Islamic art but nevertheless obstinately eludes definition. All these occasions have one element in common: they represented an opportunity to do something different, to explore something new, or to make connections between media, themes, periods or regions that had hitherto escaped notice. It is surely appropriate to strike a personal note here. Many of these papers, like so much else of my work, have profited from the back and forth of seminar teaching in small groups, in which the prolonged close focus on a single work of art ensured that most if not all of those present contributed information, ideas, questions and comments, so that the discussion could take off in unpredictable directions. That unpredictability has made teaching a huge pleasure for me throughout my career. And as the articles collected here show, it has led to some strange places whose very obscurity is part of their attraction. Moreover, the sheer geographical and chronological extent of the pre-modern Muslim world ensured that the manifestations of the visual arts in that world, however united by a common faith, were so varied that they defy generalisations. These papers, then, explore various aspects of a huge and somewhat disordered field: writing, iconography, the nature of Islamic art, historiography and the minor arts (itself a term that, like “the decorative arts”, has justifiably become very contested, though no alternative to these terms has won general acceptance). The category of writing embraces topics as varied as figural calligraphy, Umayyad dinars and the epigraphy of Spanish ivories. That of iconography has yielded papers that survey the development of this field as well as considering the controversial topic of wine and pondering the Islamic reworking of the Sasanian heritage, including the theme of the royal hunt. I have returned repeatedly to the vexed question of the very nature of Islamic art, and half a dozen papers in this collection investigate aspects of this topic, including the interaction between the Islamic world and the worlds beyond it. Historiography is served by analyses of the two unchallenged giants in the field of Islamic art in the later 20th century, namely Richard Ettinghausen and Oleg Grabar. Finally, the category of “the minor arts” is represented by three papers on ceramics, including two on Persian lustreware, and one

preface

iii

on the wooden sarcophagus of the first Safavid ruler, Shah Isma‘il. All in all, then, these papers testify to some exciting voyages of discovery on the insufficiently charted seas of Islamic art. Robert Hillenbrand March 2019

I Images of Authority on Kashan Lustreware

T

he large lustre dish in the Ashmolean Museum depicting, it seems, a seated prince with attendants (fig.1) belongs to a well-known group of lustrewares with generalized courtly subject matter. Within that group it is of special importance by virtue of its remarkable size, the technical finish of its painting and its integrated, well-developed royal iconography. Only a very few pieces have survived which can rival it on all three 1 counts. Many smaller dishes with related subject matter could be cited, as could dishes of comparable size, but with the surface so broken up by individual scenes that the visual authority which can be exerted by a 2 seated monarch is frittered away. Given that hundreds of lustre pieces are known from the Saljuq period, and given too that these luxury wares tended to depict themes of courtly life – lovers, warriors, polo players, musicians, processions, cavaliers, hunting scenes, revellers – the dearth of pieces closely related to the Oxford dish brings one up short. Clearly the theme is not quite the cliché for which it has been taken. The most obvious candidates for inclusion in this select group are the ex-Havemeyer bowl in the Metropolitan Museum of Art, New York (fig.2), the dish in the Iran Bastan Museum in Tehran (fig.3), and pieces formerly in the 3 Rabenou (fig.4) and Debenham (fig.5) collections. Yet for all their intrinsic similarity, at least three sub-categories can be identified among them: the ruler seated with his consort, who offers him a drink (fig.5); the ruler flanked by an attendant at each side (figs 1, 3); and the ruler surrounded Pope (1939), pls 635D, 707A, 720A, 773B. Pope (1939), pls 710–11, 719B, 720B. 3   I have been unable to obtain any information as to the present whereabouts of these pieces. I would welcome any information on this matter. 1 

2 

2

by a double phalanx of courtiers (figs 2, 4). As with the frontispieces of contemporary thirteenth-century manuscripts, it is plain that there was scope for development of the core theme along a number of different lines. It must be admitted at the outset that to single out these five pieces from the whole body of Kashan lustreware has something arbitrary about it. It assumes that the image is what counts. But one might equally well argue that different affiliations – based on size, shape or style – can or should be made. The only answer to such remonstrations is that a body of high-quality material can usefully be approached from several quite distinctive points of view. These are not mutually exclusive, nor is any one of them the only right one. The same piece can therefore yield valuable information as to provenance, style and shape, for example, as well as iconography. It seems appropriate to consider these five pieces together (although, given the context of this volume, the Oxford dish will take pride of place), because they represent, so far as surviving pieces are concerned, the very best that potters could do with perhaps the major royal theme of the Saljuq period: the ruler enthroned amidst attendants or with his spouse. The inherently official and propagandist nature of the theme ensured 4 5 its frequent appearance in such media as book painting and metalwork, 6 though curiously not in textiles. Compared with these, pottery was a humble vehicle, even if lustre pottery was a luxury craft, and the cheap nature of the material itself brought it within the means of a much larger 7 body of consumers. The popularity of the enthronement theme has a further consequence, in that it enables one to plot how royal images might filter down to a wider clientele than the court itself, and how such ‘high’ subject matter fared at this level of patronage. The norm, it must 8 be confessed, is to trivialize the borrowed imagery, but the best of these images, such as the five discussed in this paper, lost nothing of their power or authority in the process of transference. They are not popularized or vulgarized in any way. Clearly the artists, and thus presumably their patrons,  Melikian-Chirvani (1970), pl.4 (see also pl.64); Ettinghausen (1962), 65, 91; Hayes (1976), 51. 5  Ettinghausen & Grabar (1987), pl.249; for coins, see pl.253. 6  The nearest approach is the royal hunter: Ettinghausen & Grabar (1987), pl.261; Blair et al. (1992), fig.5. 7  See Watson (1986), 205–206, 208; and Caiger-Smith (1986), 215. 8  See Bahrami (1949), pl.LXXIV for a typical large-scale example of this process; a further couple of dozen could be cited from this book alone. 4

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

3

took them seriously. Indeed, it is possible that the best lustre pieces were produced at the behest of the court – an issue that will be discussed in 9 10 detail later. The tiles found at Takht-i Sulaiman and Qubadabad show that royal patronage was indeed extended to lustre ceramics, and the key 11 role that they played in such major religious shrines as Mashhad and 12 Qumm scotches the theory that they were seen as an art form of the second rank. If there is little that is absolutely new in these images, where does their importance lie? The answer is two-fold. Within the context of contemporary ceramics, they represent, as noted above, the top of the range. And within the wider context of twelfth to fourteenth-century art in the Iranian world, each contains distinctive details which set it apart from related images in other media. Thus it seems profitable to examine these five pieces together, in the hope that their combined evidence will allow the parameters of such iconography on ceramics to be defined with greater precision, and enrich current knowledge of how this same theme is treated in other media. It is only by such close comparisons that the diagnostic features of this iconography can be highlighted effectively. Finally, why should there be so few large-scale versions of this theme, given that large-scale lustre ceramics are by no means rare and presumably cost much the same to produce? When one looks at the comparative material in other media more closely, one realizes that the standard procedure in lustreware was to simplify, and to simplify drastically. Hence the popularity of designs involving only one or two figures. Hence, too, the tendency to reduce the size of the main subject by enclosing it within 13 a series of concentric bands of ornament or inscription. This device also means that the main subject will for the most part be confined to the flattest area of the bowl or dish. Accordingly it will present no special difficulties of execution or of perspectival distortion. By the same token, the ambitious enthronement scenes stand out precisely because they confront and overcome daunting technical difficulties. A major caveat should be made at this point. It is not always practicable to examine pieces at first hand, yet even a good photograph of the whole  Naumann (1976), pls 5–8.  Colour plates in Öney (1976), 36, 38. 11  Watson (1985), 124, 126, pls 104a–105. 12  Watson (1985), 124, 154–55, pl.103. 13  Notable examples are the plate in Chicago (Pope [1939], pl.638) and several of the so-called ‘Gurgan’ wares (Bahrami [1949], pls XLIX-LIII, LV, LXVIII). 9

10

4

of one of these wares does not show up more than a small fraction of its constituent fragments. But all have been assembled from such fragments, whether large or small, and therefore published photographs presenting them as intact pieces give the wrong impression. This is, of course, well known to specialists, but it is not easy to bear such repairs in mind consistently when interpreting a scene, or to remember that restoring and repainting are also interpretation. In the case of the Oxford dish, for example, the empty right hand of the prince is most unlikely to represent the original design, and with the loss of that one detail the interpretation of the whole piece is rendered significantly more difficult. A comparison of how the New York bowl looked before (figs 6, 7) and after its recent reassembly (fig.2) administers the necessary shock to the system. It would be desirable, especially in the case of key pieces, to publish analytical drawings summarizing the findings of technical and scientific examination. The most that can be hoped for in the case of the wares under discussion is that the design as it now appears is not too far away from its original form. 14 This seems to be so for the Oxford dish and the New York bowl; for the others, it remains an open question. Formal aspects All five pieces obey a basic schema, though the departures from it are revealing. The central figures are flanked by attendants, and the upper and lower exergue is filled in each case by an element decisively differentiated from the rest of the background. It is an ancient formula whose basic geometrical division was never so clearly proclaimed as in Early Christian 15 and Byzantine ivory diptychs. That formula was devised for a flat surface, and a certain loosening of the boundaries between its component parts 16 can already be seen in late antique and Byzantine silverware. The lustre plates under discussion may indeed be distant descendants of these Byzantine works, especially if – as seems likely – they are closely linked to 17 the frontispieces of contemporary Islamic book painting, which in their

14  Close examination of the Oxford dish suggests that there is much reconstruction in its present state, some of it visible under ultraviolet light. I am most grateful to Dr James Allan for his helpfulnesss on the three occasions when I examined this piece. 15  Natanson (1953), pls 43, 45. 16  Such as some of the David plates; see Weitzmann (1979), nos 425, 427, 430–32. 17  Perhaps the closest such parallel is one of the Kitab al-Aghani frontispieces (Ettinghausen [1962], 65).

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

5

turn clearly derive from Byzantine prototypes in physical layout as well as 18 in iconography. Thus fig. 8 becomes fig. 9 or fig. 10. The departures from the basic framework found in the ivory diptychs are dictated principally by the physical form of these wares. Their circular shape encouraged the artists to devise the entire composition in curvilinear terms, so that its constituent elements, instead of being abruptly juxtaposed or confronted with each other, are integrated, blending together in continuous fashion. Such a relaxed re-interpretation of the five-part diptych formula did away with the traditional dividing lines 19 (except the upper and lower exergue), and the resultant visual informality removed some of the portentousness which is practically built into that formula. Thus the Kashan artists were clearly not oppressed by the dead hand of tradition. This recomposition can be seen to best advantage in the way that the outermost seated figures accommodate themselves to the curve of the dish or bowl. Indeed, the figures on these pieces seem to be made of rubber. It was simply not possible to transfer the originally twodimensional image as such on to the three-dimensional ceramic surface. The adjustments called for careful calculation. For all the apparent ease with which the figures interact, however, the artist sticks closely to the skeleton framework which underlies them. The circularity of the composition bears closer investigation. So marked is this feature that the prince is presented as the still centre of the turning world, with his attendants orbiting slowly around him. The impression is especially marked in the case of the two smaller attendants below. Given the strong three-dimensionality of the dish, too, the fact that he occupies almost all the flat surface at the centre, while his attendants are relegated to the curved sides of the dish, emphasizes alike their mobility and his stability. In the celebrated ‘Cup of Solomon’ the Sasanian shah occupies exactly the same flat central position, which makes him the obvious focus 20 of attention. The physical form of these wares had distinct advantages over their two-dimensional counterparts, notably book painting, in that they made it possible to devise illusionistic compositions. The monarch is physically furthest away from the viewer, and the effect is subtle or marked depending on the depth of the piece. Similarly, his flanking attendants, Compare the layout of the Vienna Galen frontispiece, c.1250 (Ettinghausen [1962]), 91), with such Byzantine ivories as the St Lupicinus diptych (Natanson [1953], pl.50). 19   The Oxford dish does away even with this feature. 20   Pope (1939), pl.203. 18 

6

thanks to the third dimension, can be understood not as participants in a two-dimensional group photograph but as being drawn up in ranks perpendicular to the monarch, like a guard of honour, creating a corridor at the far end of which he sits enthroned. Medieval literary sources attest 21 such practices, for example at the court of the Ghaznavids. This effect is not only lost but actually contradicted in a published reproduction, which tends to propel the ruler forwards, and to suggest that his attendants are behind him, whereas, when one looks at the object itself, the reverse is the case. In short, to make full sense of the image one has to hold, or at any rate confront the three-dimensional object and to change one’s angle of vision continually. The eyes of the prince’s attendants are turned towards him, thereby automatically deflecting the gaze of the viewer back towards him too. All these details show the artist turning to advantage the difficulties and distortions inherent in the shape of a bowl or dish with deeply plunging sides. It should also be remembered that the visual relationship between the figures varies as the viewer moves. It is neither as static nor as distorted as a two-dimensional photograph of it so easily suggests. The patterned ground, for which parallels in manuscript painting can 22 be cited, comprises severely abstracted foliate scrolls whose constant 23 sinuous movement provides an apt foil for the immobility of the figures. The other pieces under discussion use a similar background, and this type 24 of patterning is standard in Kashan lustreware. It is not likely to have any specific significance in the context of enthronement images. It even invades the rim of the Oxford dish (fig.11). Several functions could be suggested for it. In purely compositional terms, it fills the leftover spaces 25 in the design, and thus ministers to the celebrated Islamic horror vacui. It also removes the scene from the real world, lending it an exemplary and timeless quality. For good measure, it provides a subdued visual interest to 21   Al-Juzjani (1977), Vol.I, 83 mentions that the court of Sultan Mahmud ‘was guarded by four thousand Turkish slave-youths, who, on days of public audience, were stationed on the right and left of the throne’. For further references, under the Samanids and Tughluqids, see Schlumberger & Sourdel-Thomine (1978), Vol.I, 62; also the description of the 10th-century palace of the Armenian King Gagik of Vaspurakan at Aght’amar, in which were paintings of ‘gilt thrones, seated upon which appears the king in splendid majesty surrounded by shining young men …. ’(Artsruni [1985], 357).. 22  As in the Varqa va Gulshah manuscript: Melikian-Chirvani (1970), pls 21, 36, 467, 59. 23  Pope (1935), figs 3–5. 24  Pope (1939), 1590–96. 25  Ettinghausen (1979).

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

7

complement the main image and yet not compete with it. In that sense, it 26 27 may be likened to the gold or red background commonly used to set off images of authority in thirteenth-century Islamic book painting. But in fact the foliate background of these lustrewares is more closely integrated with the main figures than this description suggests. In the Oxford dish, in particular, the curvilinear patterning of the background is echoed by the similar but denser ornament which overruns the prince’s clothing; and, more generally, by the generous curves of the attendant figures, especially their haloes, faces and thighs. The same deliberate ambiguity between the background and the patterns on clothing can be noted on the other four pieces (e.g. fig.12); indeed, it is a leitmotif of Kashan lustreware. It could be interpreted as a means of integrating foreground and background in 28 such a way as to highlight the faces of the figures. Given the close kinship between the background ornament and that depicted on the clothing of the figures, it would be unwise to assume that 29 these patterns reflect the actual designs found on contemporary costume. Moreover, they change indiscriminately from one area to another, which still further reduces their evidential value. Spiral and scrolling patterns are interchangeable, and further variations on the theme are achieved by switching colours or tones. The decoration of several of these wares finds very close equivalents in 30 one medium, and one medium only: manuscript painting. The New York bowl has an obvious pendant in the Istanbul Kitab al-Aghani frontispiece, 31 produced in the very same decade. The Rabenou bowl obviously derives from the same kind of source, and even its most distinctive iconographic feature – the suspended crown placed redundantly above the alreadycrowned head of the prince, clearly a misunderstanding of the Sasanian hanging crown – has its close equivalent in manuscript painting, namely in the depiction of Sultan Berkyaruq in the World History of Rashid al32 Din. Admittedly this is of early fourteenth-century date. So too is the close manuscript parallel for the Debenham plate – one of several images   Such as the images of the Kitab al-Aghani frontispieces (Talbot-Rice [1976], col. pl.I).   E.g. Grube (1959), fig.9 (the Bodleian Dioscorides author portrait, Cod. Or. d. 138, f.2b, dated 637/1239). 28  Pope makes an illuminating comparison with the role of ornament in Gothic tapestry (Pope [1939], 1590). 29   Pace Pope (1939), 1587, 1595, 1601. 30  R. Hillenbrand (1994). 31   Ettinghausen (1962), 65. 32   Talbot Rice (1976), pl.66; cf. H. Erdmann (1983). 26 27

8

of Mongol rulers enthroned with their wives and talking to them, so that a sense of human contact alleviates the austerity of the enthronement 33 image. As for the prince flanked by attendants, this was the most popular 34 version of enthronement images not only in manuscript painting, but 35 36 37 also in wall painting, sculpture and metalwork. The curious placing and posture of the two attendants below the prince on the Oxford dish invite the supposition that they do not naturally belong there. Earlier related images, such as the post-Sasanian silver plates, perhaps of Buyid date, in the Hermitage, suggest that a pair of lions was originally 38 present – and that they had the function of guarding the throne, as on 39 the Rabenou bowl. As so often occurs, the replacement of a traditional element by a new one weakens the entire image, for it surrenders meaning in return for decoration, a poor exchange. A more popular method of treating this area was to seal it off by a horizontal line, thus creating a formal exergue which could be filled by some theme physically separate from, but related to, the principal subject matter. In common with many of the more ambitious Kashan lustrewares, attention is focused on the inner surface of all these pieces; the exterior is kept much simpler. The Oxford dish, for example, originally had at least part of the outer surface covered with a plain blue glaze, latterly eked out with a darker blue sprayed on. This suggests that the primary purpose of the dish was display, for if it were to be used for food the design would be obscured; and the inherent fragility of such wares would make them more 40 suitable for display than for use. Key components of the image In the Oxford dish the royal figure has an immediacy, a personal quality, totally lacking in the four other faces on the dish (fig.13). His pensive, slightly sad expression, subtly suggested by the careful detailing of the  Rogers et al. (1986), 38, 43–44; Ipşiroğlu (1964), Taf.VII.   Ipşiroğlu (1964), Taf.IV and Taf.XIII; Ipşiroğlu (1971), pl.16 (where this formula has been adapted for narrative purposes to the tale of Varqa va Gulshah) and 23. 35  E.g. Pope (1939), pl.554B. Compare also the Ghaznavid palace at Lashkar-i Bazar, though the depiction of the monarch has not survived: Schlumberger & Sourdel-Thomine (1978), Vol.II, pl.122. 36  Pope (1939), pl.517. 37  Hagedorn (1989), 103–108, pls 21, 24, 29, 62, 76, 79, 83, 85, 102, 107, 201. 38  E.g. Marshak (1986), 30, 33, 135; for the Sasanian source, cf. 194. Compare also the laqabi dish illustrated in Pope (1939), pl.603, and Otto-Dorn (1982), 159–60. 39   Though their maladroit placing weakens the idea. 40   Caiger-Smith (1985), 65–66. 33 34

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

9

mouth and by the figure’s own left eye, has nothing in common with the blank mask worn by the other four figures. It brings a religious image to mind. In a different context it could easily be taken for a depiction of Jesus, or as a forerunner of the standard devotional twentieth-century image of 41 ‘Ali, shown with his face half-turned and his eyes serenely focused on the 42 far distance. And Kashan, where this piece was in all probability made, was at this period a thoroughly Shi‘ite town, for all that it was marooned 43 in a Sunni countryside. Two other dishes in this group also treat the face of the central figure differently from those of his entourage. In both the Tehran and New York 44 pieces (fig.14) he is the only one to wear a beard, a detail which underlines his mature age and his masculinity, as against the youth and indeterminate sex of those flanking him. Even the prince on the Rabenou bowl, whose face has the conventional bland plumpness and pronounced jowls so dear to contemporary taste, is nevertheless singled out from his courtiers by his tiny goatee. In all three cases, too, he alone is depicted frontally, which adds an extra charge of authority to his pose. Finally, in the New York bowl the prince’s eyebrows take the form of two short thick crescents, in contrast to the extravagantly arched double bow used for his retinue. A similar distinction is made in the Rabenou bowl, except that here the 45 prince has his eyebrows rendered as separate thin bars. Both the principal attendants on the Oxford dish have a spot low down 46 on each cheek, and a third spot below the centre of the mouth. It is not clear what the purpose of these spots is -whether to suggest dimples or puffy features (and the attendants on the Tehran dish sport majestic 47 double chins), or to serve as beauty spots. Similar spots occur on the hands of the attendants in the New York, Rabenou and Tehran pieces; in 48 these cases they point unambiguously to beauty spots. Women had their I am grateful to Dr Ulrike al-Khamis for this suggestion. Compare the depictions of Sufis or saints in later Persian painting (Lentz & Lowry [1989], pls on 177, 263, 281), or indeed the Prophet himself (Lentz & Lowry [1989], pl. on 309). 43   Lapidus (1969), 57, n.11. 44   As in the frontispieces to the Kitab al-Aghani: e.g. Ettinghausen (1962), 65. 45  For the Sufi connotations of the numerous types of eyebrow, see the quotations from medieval poets assembled by Nurbakhsh {1980), 49–59. 46   This may be the mole, a multivalent Sufi symbol; cf. Nurbakhsh (1980), 84–89. 47  The double chin (ghabghab) was connected by Sufi poets such as ‘Iraqi and Tabasi to esoteric knowledge (Nurbakhsh [1980], 129). 48   Cf. also Ettinghausen & Grabar (1987), pl.373. 41

42 

10

hands elaborately tattooed in medieval Iran, and the more modest form of marking found on these pieces may have been deemed appropriate for young men. In the Rabenou bowl the ruler alone has a triple dot on each 50 hand. The double-tiered arrangement of the courtiers in the Rabenou and 51 New York bowls finds its closest parallel in contemporary book painting, for in no other art form of the Saljuq period did the technical capacity to execute such fine detail exist. The idea itself is not new; it can also be 52 53 found in Sasanian royal rock reliefs and in contemporary metalwork. But a distinctive difference may now be noted: the courtiers themselves express an ideal of beauty, so that they are not present merely as adjuncts to the monarch’s power. This emphasis on facial beauty is underlined by the densely ornamented clothing that they wear and to which the blankness of the face provides such a strong contrast. Yet despite this emphasis there is never a suggestion of portraiture. These are stereotyped images of beauty, the moon face (mahruy) celebrated in medieval Persian 54 poetry. The ruler image at this period symbolized, as it had for centuries, a concept; it did not aim to represent a specific person. For all this detail, it must be conceded that it is not always an easy matter to distinguish male from female in these images, whether by face or attire. A curious aspect of the design of all the pieces in this group is that 55 the figure of authority holds no attribute of power. Instead, one hand – whether his left or his right – is raised and placed across his chest. Sometimes, as in the Rabenou bowl, this gesture recalls a Christian 56 benediction, though it could also denote speech; it is strangely tentative. The open hand of the Debenham plate makes best sense as a gesture of 49

49   Talbot Rice (1976), pl.68; Lowry & Nemazee (1988), col. pl.9; cf. also Önder (1978), 655–56, fig.4 and, for an earlier period, Milstein & Brosh (1984), 20, with a colour plate of the celebrated Fatimid drawing of a tattooed courtesan (see principally Rice [1958], 33-37, figs 1-2, pls I–IV). 50   In his raised hand it is smudged. Compare the same motif on a servant’s cheek in a painting from the Cappella Palatina (Ettinghausen [1962], 45) and the brief discussion of it in Jones & Michell (1976), 82. 51  Once again the obvious parallel is the Kitab al-Aghani frontispiece in Istanbul (Ettinghausen [1962], 65). 52   As at Bishapur (for a colour plate, see Herrmann [1977], 91-92). 53  Whelan (1988), figs 13–16. 54  Melikian-Chirvani (1974), 51–55; Nurbakhsh (1980), 110; Esin (1979). 55 That is, in so far as his hands have been preserved. 56  In early photographs, the New York bowl has a similar theme; but this was added in modern times.

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

11

speech, because the ruler is leaning over to his consort taking her hand, but the same gesture on the Tehran dish looks aimless and somehow incomplete. Some echo of Buddhist mudras is also possible. Whatever the explanation of these empty hands, the key point is that these figures of authority are not holding the standard attributes of such images, namely 57 58 59 a wine-cup, a bow and arrow or a mandil. And with the exception of the Debenham plate, already noted, the other hand is also not holding anything. It is as if the image were being deliberately stripped of obvious royal connotations. The absence of a crown on four of the five pieces is a pointer in the same direction. If these various hand gestures are indeed 60 to be interpreted as denoting speech, the entire context of such scenes would require drastic revision, since the conventional interpretation sees them as royal enthronement images. Perhaps, then, they refer rather to meetings between some teacher, shaikh or Sufi master and his disciples, 61 pupils or associates. Or – and this is perhaps still more likely – the toning down or removal of specifically royal attributes was appropriate for a subroyal level of patronage, in which an element of ambiguity was actually desirable. Only the central figure on the Rabenou bowl is indisputably royal. For the other pieces, a suggestion of kingliness suffices. It is perhaps worth noting that the most elaborate of all surviving Kashan lustre dishes was made for no more exalted personage than an anonymous amir. Some of these issues will be dealt with in more detail later. All of the princes rest their backs against a decorative support. This is clearly not a full-scale throne of the kind that became popular in Iran in the wake of the Mongol invasions, and for which Chinese sources have 62 convincingly been proposed. Once again, parallels are to be found readily enough in thirteenth-century book painting, for example the frontispiece of the Schefer Hariri, in which two different objects of this kind are  As in the tabouret at Philadelphia (Watson [1985], pl.86b); cf. Roux (1982), 83–108.  As in one of the frontispieces to the Kitab al-Aghani (Ettinghausen [962], 65); cf. Roux (1982), 59–82. 59  As in the Vienna Maqamat frontispiece (Ettinghausen [1962], 148); cf. Rosenthal (1971), 63–99. 60   As in the figure on the left-hand page of the double frontispiece in the Schefer Hariri (Paris, Bibliothèque Nationale, ms. arabe 5847): Fares (1953), fig.4 and Ettinghausen (1962), 114. 61  The Schefer Hariri frontispiece shows how closely the two iconographies – one royal, the other intellectual – were linked, since the differences between the images facing each other on ff.1b and 2a are matters of nuance only. 62  Donovan (1988-89), 18–22, 27–31. 57 58

12

depicted. It is not easy to work out how this back-rest functioned. Did it, for example, ensure stability by being braced against the floor in some way, or was it attached to the bolsters which are always shown in association with it? Triangular projections at top right and left are a consistent feature; these seem to be purely decorative. The ornament applied to the back-rest is varied – braided designs of possibly apotropaic intent are common – but suggests that, whatever the material of construction, it was covered 65 with some kind of textile or patterned upholstery. The most elaborate of the five examples is that on the Debenham plate, whose differently patterned corner-posts topped by pointed finials suggest a more solid construction than is depicted on the other pieces. This may be due to the need to accommodate two figures – the prince and his consort – instead 66 of just one. While only the Debenham plate illustrates a throne-like object, in all four of the other pieces the ruler is singled out by the fact that he is sitting on something, and not just on the floor. Perhaps the bolster and the back-rest were the informal version of the throne. But the essential formality remains. Some kind of seat, at all events, is a royal prerogative not extended to the attendant courtiers. That is the key point; the actual form of the seat – cushions, screen, litter, couch, throne – is secondary. It elevates him, even if only slightly, and thereby singles him out. And it injects a sense of formality into an otherwise relaxed scene. Nevertheless, 67 68 the earlier role of the throne itself – as in Sasanian and Byzantine art – as an abstract symbol of royalty or authority has definitively gone. The evidence provided by the headgear of the principal figures in these five pieces is curiously contradictory. Of the five men concerned, no two wear the same kind of headgear, and in each case they are the only ones 69 on the piece to wear it. Thus the principal figure is singled out, even if the absence of a traditional crown is noticeable. And such crowns are found frequently enough in the art of the eastern Islamic lands between 1050 and 1250 to remove any doubt that suitable models were available 63

64

  Fares (1953), figs 3–4.   For a brief discussion of Saljuq thrones, see Otto-Dorn (1982), 158–59 and figs 14a–c. 65   Sadan (1976), 99–120, especially 115–19. 66   Sadan (1976), 36–41. 67  Von Gall (1971). 68  A. Grabar (1968); cf. also the associated studies (341–50, 351–64, 393–401). 69 In the Tehran dish the ruler has a larger version of the hat worn by his attendants. Is this how the piece looked originally? 63 64

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

13

to the painters of these lustrewares. What does this absence imply? Two options present themselves: either the figures of authority on these plates are not princely at all, or they are princes depicted in a setting which falls short of the extreme formality of a diwan scene. The concomitant absence of formal thrones and retainers with identifiable offices would fit either option. Nevertheless, it seems more likely on balance that at least three of these scenes are royal images, since if they were not royal, the presence of the entourage and bodyguards on the New York and Rabenou bowls and on the Debenham plate would be hard to explain. If these arguments are acceptable, the headgear depicted on these five lustre ceramics gives some indication of how wide the range of choice 71 was at this time, and how the crown had lost its earlier predominance as 72 a sign of royalty, let alone that capacity to indicate a specific ruler which 73 it had possessed in Sasanian times. Indeed, most monarchs depicted in Saljuq art wear something other than a crown on their heads. In the postSaljuq centuries this marked reduction in formality and in specific detail, amounting almost to a democratization of headgear, became the norm 74 in the royal iconography of the Iranian world. In the New York bowl, for example, the princely figure wears an elaborate turban, more stately and imposing than those of his entourage, but nevertheless plainly of the same general type. This same type recurs in the Istanbul Dioscorides, 75 where it is worn by master and student alike. None of the five pieces shows the central figure wearing a sharbush, the fur hat which the ruler 76 frequently wears in Mesopotamian painting, though it is not his exclusive 77 prerogative. But book painting shows that elaborate turbans were worn by medieval rulers, not only by prophets or by members of the learned 70

E.g. Melikian-Chirvani (1970), pl.42; Fares (1953), pls III–IV and colour frontispiece; and Atil (1973), nos 52–53. 71  Esin (1970a); cf. also Sims (1973), 296–97. For all his painstaking research into the types of headgear found in Persian painting of the 14th century, Schroeder was unable to come to any definitive conclusion about the meanings attached to them (Schroeder [1939], 120–23). 72  Shahbazi (1993). 73   K. Erdmann (1951), 87–90, Abb.1–9, 18. 74  Tiraz bands underwent the same process; in the Schefer Hariri of 1237, for example, most of the villagers in the rural scene on f.138r wear them (Ettinghausen [1962], 116). 75  Ettinghausen (1962), 68–69, 71. 76  For example, in the Kitab al-Aghani frontispieces: Fares [1961], pls I, VIII, X–XII. 77  In the Vienna Galen frontispiece, not only the ruler but also his cook is shown wearing it (Ettinghausen [1962], 91). 70 

14

class, and that a hierarchy of turban types operated. Perhaps the wearing of a turban by a ruler implied a degree of informality or signalled his participation in some intellectual or religious activity. This may be the intention of the New York bowl. One curious detail of the turban depicted is the finial or ‘alam attached to the top of it. At first glance this suggests a chain suspended from the canopy, so that the turban itself seems to hang above the head of the prince; but in fact it does not quite connect with the canopy and must therefore be interpreted, however unsatisfactorily, 80 as a turban ornament. It makes no sense as part of the throne. As with the throne or royal costume at this time, there was no specific type of crown which was passed on from one generation to the next and therefore acquired, like the later Peacock Throne, a symbolic authority. An unusual feature of the Oxford dish is the segment of a sunface, complete with rays, in the upper exergue. Complete sun-faces are encountered in Kashan lustreware, but in a context where their purpose is self-evident, namely at the centre of pieces decorated with planets and 81 zodiacal signs orbiting around them. In such cases the pervasive astral imagery requires some reference to the sun, and one must also reckon with a punning reference to the golden light generated by the lustre technique. More complex is the presence of a sun-face a century later in a narrative scene from the Great Mongol Shahnama, depicting Faridun’s grief as he 82 realizes that his son is dead. Firdausi comments sombrely ‘Such is the usage of the ancient sky’, and indeed the sun-face, placed centrally at the top of the picture, looks down impassively like an umpire upon this tragic scene. In the Oxford dish yet another strand of significance manifests itself. The presence of a sun-face – not just a sun – directly above the ruler’s 83 head seems to express some special relationship between the two. It may point to his divine right to rule. The presence of a half-face irresistibly brings to mind the combination of the Lion and the Sun (in which the sun is again a half-face) known in this period in the coinage of the Saljuqs 78

79

 For example, the scholars in the double frontispiece to The Epistles of the Sincere Brethren, dated 1287 (Ettinghausen [1962], 98–99). 79   For example, in the Maqamat manuscript in Vienna (Nationalbibliothek, A.F.9), dated 1334, figures of authority such as the ruler and governors wear turbans with two horns (Ettinghausen [1962], 148). 80 Björkman (1934), cols 887a, 888a; cf. Zygulski (1992), 70–99, 104–17. 81  Pope (1939), pls 712–13; R. Hillenbrand (1988), 32-37. Cf. also the use of the theme in mina’i ware: Pope (1939), pl.656 and Bahrami (1937), fig.3. 82  Lowry & Nemazee (1988), 77. 83   This connection is explored in Baer (1981) and R. Hillenbrand (1988). 78

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

15

of Rum, an image which was eventually destined to become the national 85 emblem of Iran, and was already popular there. It was particularly in the Saljuq and Mongol periods, and in contemporary Mamluk art, that 86 the ruler was invested with solar qualities, whether in his titulature, in 87 88 the panegyrics of poets or in the visual arts. The parallel between the sun shining on the ruler and the ruler in turn directing the light of his 89 benevolence and generosity on to his subjects is an obvious one. In the case of the Oxford dish, as noted earlier, the imagery is intensified by the use of lustre, whose reflected golden light mimics that of the sun. It is 90 perfectly possible that the spiritual associations of light, as exploited for 91 example in the Illuminationist school founded by al-Suhrawardi (d.1234), added an extra strand of significance to such an image. But that depends on the person for whom it was made. In some types of Islamic solar 92 symbolism the implication is that the ruler himself is the source of light, but here the nuance seems to be that light is conferred upon him from 93 above. Finally, the curved shape of the sun-face with inner and outer rim evokes a dome, and thereby in the medieval Persian mind the sky itself. In three of the five pieces in the group the importance of the central figure or figures is underlined by the sheer quantity of attendants – twelve or sixteen in number – waiting on them. It is part and parcel of the intended visual effect that these attendants should illusionistically seem more numerous than they really are. Hence the emphasis on their sameness. In the Debenham plate, a minor detail of hierarchy is noticeable, in that the 84

 Spuler & Sourdel-Thomine (1973), pl.267e.  Bahrami (1949), pl.LXVIII. 86 Especially in titles using the words siraj, shams, badr, najm, falak, nur, baha’ and diya. 87  Melikian-Chirvani (1994). 88   Pope (1939), pls 1314A, 1328; Allan (1982), 15–16, 23–29, 62–68, 88–89; Allan (forthcoming). [See bibliography]. 89   Caiger-Smith (1985), 191–92. His wide-ranging analysis of how symbols of light are used in various cultures in quite everyday contexts contains much of relevance to the present discussion. 90  See Schnyder (1994). 91  For a brief summary, with basic bibliography, see Schimmel (1975), 259–63. That same school developed associations between Sufism and the Shahnama (Melikian-Chirvani [1988], 46–49; Melikian-Chirvani [1991], 99–109); the relevance for the royal scenes executed in the resplendent medium of lustre is obvious. 92   For example, the famous incense-burner of the Mamluk Sultan Muhammad b. Qala’un (Allan [1982], 86–89); see, more generally, R. Hillenbrand (1988). 93  Compare the presence of angels in so many frontispieces of the 13th and 14th centuries: Ettinghausen (1962), 65, 148; Fares (1953), pls III–IV; Ipşiroğlu (1964), Taf. IV; Ipşiroğlu (1971), pl.23. This is discussed briefly in Otto-Dorn (1982), 161, 185, n.67. 84 85

16

courtiers in the rear rank wear simple striped robes instead of the robes richly patterned with arabesques worn by those in front. A comparably clear hierarchy of footwear may be observed in the Oxford dish. The ruler alone wears dark boots, while those of his principal attendants are light-coloured, with a little curvilinear embroidery just below the ankle. As for the two smaller attendants lower down, they have no boots at all. Similarly, on the Rabenou and New York bowls, the attendants in the rear rank are depicted with faces markedly smaller than those of their colleagues in front. But in all three pieces attention is concentrated upon the large, undeniably vacant, moon-like faces clustered so close together that they form two solid white bands across the surface of the dish or bowl; and it is the prince and his headgear alone that interrupt the even tenor of these bands. Thus even in the most basic design terms of solids and voids, the prince is set apart. It is standard practice for the courtiers to be of significantly wider girth than their master – indeed, the discrepancy is almost caricatured in the Tehran dish – and to show only one hand in 94 contrast to his two. All incline submissively towards him, and direct their gaze at him alone, unless bodyguards are depicted (figs 2, 4), in which case 95 they look outwards as their office demands. The Debenham plate shows the ruler with his consort, a common 96 theme in Saljuq luxury ceramics. This is a significant departure from earlier royal iconography, whether Islamic or Sasanian, which relegated 97 women to a secondary position. The reason for the change presumably lies in the very different role of women in Turkish society, where they had the right to control their own money, to invest in large-scale architectural 98 and charitable benefactions and to have a voice in political affairs. Saljuq luxury ware confirms this rise in status, for it frequently shows women 99 seated beside men on terms of equality, and the scenes themselves, or the poetry which encircles them, indicate that the women are there to be  Compare the Sasanian practice whereby courtiers folded their arms in the royal presence in such a way as to keep their hands out of sight (Herrmann [1977], 119). For further information on the etiquette of the attendants, see Esin (1970b). 95  On the New York bowl they alone wear a special kind of one-horned hat; on the Debenham plate only one bodyguard, at the consort’s side of the throne, is to be seen (the present layout may be a maladroit modern restoration in the case of this detail). Similar bodyguards occur in manuscript frontispieces, as shown for example in Ipşiroğlu (1971), 23. 96   Atil (1973), no.51. 97   Ghirshman (1962), pl.259, is a rare exception. 98   C. Hillenbrand (forthcoming). [See bibliography]. 99   Or, most notably, on a matching pair of plates: Atil (1973), nos 52–53. 94

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

17

pined for or to be won over. Echoes of the courtly love tradition may be detected here. Patronage In sum, then, the Oxford dish is a succinct statement of authority. The prince is central, set apart from his attendants or companions, selfcontained and remote as a Buddha. His eyes make no contact with his attendants or with the viewer; his gaze is turned away and fixed blankly, as if in meditation, on an indeterminate spot outside the world depicted on the plate. Yet for all this sense of otherworldliness, his rank in this world is assured, and is ostentatiously emphasized by his elaborately patterned clothing, from his headgear to his boots, his central position, his beard and moustache, the throne back, the bolsters against which he rests, and those who defer to his pleasure. Similar devices single out Badr al-Din Lu’lu’ in 100 the Kitab al-Aghani frontispieces. For all that the obvious analogies of the piece, at least at first glance, are with formal enthronement scenes, the cumulative impact of numerous details, small in themselves, makes it plain that an essentially private atmosphere is being created. To begin with, there are only five people present, which rules out a truly official atmosphere. Next, the glance of the prince is not frontal, but turned to one side. Introspection, not dominance, is the keynote. His principal attendants look not at him but somewhat vacantly beyond; they both make quite different gestures, nor is their headgear or the patterning on their clothes the same. So clearly, for all their mask-like faces, they are intended to represent individuals. This, 101 then, is a scene of the prince (or other generalized figure of authority) at ease with his companions. It is unclear whether some more specific subject is intended, though the fixed gaze and the lack of contact between the various figures suggests that it is not a conversation piece but rather, perhaps, a meditation or trance. A full explanation may never be possible, for key parts of the dish have been replaced by modern repairs. It would be blinkered to propose royal patronage for these pieces on the basis of their iconography, while denying a similar patronage to other wares of the same scale (and therefore expense), shape and style but with different themes. The problem of the patronage of the more 100  Fares (1961), pls I, VIII, X–XII. For a full discussion of this and related types of images of authority, see Esin (1968) and Esin (1971). 101  Caiger-Smith (1985), 73.

18

1. Dish, lustreware. Diameter 47.3 cm. Iran, Kashan style, 12th–13th century. Ashmolean Museum, acc. no. EA 1956.183, Barlow Gift.

2. Bowl, lustreware. Diameter 30 cm. Iran, Kashan style, dated Jumada II 607/ November 1210. New York, Metropolitan Museum of Art, acc. no.41.119.1, Gift of H. Havemeyer (photograph courtesy of the Metropolitan Museum of Art). Current state.

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

19

3. Dish, lustreware. Diameter 37.2 cm, height 4.2 cm. Iran, Kashan style, 13th century. Tehran, Iran Bastan Museum, museum no.3181 (after Bagherzadeh [1981]).

4. Bowl, lustreware. Diameter 38 cm. Iran, Kashan style, 13th century. Formerly in the Rabenou Collection (after Pope and Ackerman [1939]).

20

5. Bowl, lustreware. Diameter xx cm. Iran, Kashan style, 13th century. Formerly in the Debenham collection (after Pope and Ackerman [1939]).

6. Bowl, lustreware. Diameter 30 cm. Iran. Kashan style, dated Jumada II 607/ November 1210. New York, Metropolitan Museum of Art, acc. No. 41.119.1, Gift of H. Havemeyer (photograph courtesy of the Metropolitan Museum of Art). Restored state.

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

21

7. Bowl in the Metropolitan Museum of Art (Fig.2): line drawing (V. Bernie).

8. Compositional schema: rectilinear layout.

9. Compositional schema: curvilinear layout (a).

10. Compositional schema: curvilinear layout (b).).

11. Rim of the Ashmolean dish (Fig.1), after Pope and Ackerman (1939).

22

12. Bowl in the Metropolitan Museum of Art (Fig.2): fabric patterns (V. Bernie).

13. Ashmolean dish (Fig.1): face, after Pope and Ackerman (1939).

14. Bowl in the Metropolitan Museum of Art (fig.2): face, after Pope and Ackerman (1939).

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

23

ambitious lustreware is a perennially contested one, and in default of 103 literary or epigraphic evidence a solution still eludes modern scholarship. But the statistical evidence overwhelmingly points to massive high-quality production aimed at an undifferentiated clientele. One may suggest that while the pervasive presence of courtly themes on costly lustrewares can be explained as an attempt to evoke a generalized ‘good life’ of leisure and privilege, enthronement scenes are at once too formal and too specific to fit comfortably into the same iconography of pleasure and gracious living. They are too clearly about rulership. Could they, then, be a formal present from a ruler, like a medal or a robe of honour, to some deserving subject? It is certainly feasible that the larger, more ambitious and more 104 carefully executed pieces were produced under court patronage; after all, this was the finest contemporary type of pottery. Decisive corroborative evidence for this is provided, as noted above, by the ceramic finds at Takht-i Sulaiman, the only major medieval Iranian palace of the period 1100–1300 of which substantial vestiges survive above ground. Here numerous lustre tiles and lustre fragments were found, and the same is 105 true of Qubadabad in western Anatolia. Given the very formulaic rendering of facial features in Kashan lustrewares, it is perhaps perilous to regard quirks of expression as clues to the meaning of these pieces. Attributes provide much safer evidence. Nevertheless, the difference in spirit between these ruler images and those of the Sasanian and post-Sasanian periods is very marked. In place of the earlier emphasis on aggression, dominance and the direct assertion of power, these Saljuq ruler images on pottery often depict the monarch as 106 gentle, pensive, even contemplative. His attendants are not miniaturized 102

Pope argues on grounds of size, gilding, execution of design and quality that these are actually pieces made for royal patrons (Pope and Ackerman [1939], 1599). See note 103.. See note 103. 103   The nearest – and very rare – pointer that epigraphy provides to exalted patronage for the top-quality large lustrewares is the mention of an anonymous amir on the Freer plate (Atil [1973], 68–69). Named sultans, amirs or high officials are also rare in metalwork. Ettinghausen has argued that ‘the Seljuq production in Iran was, on the whole, of a unified, anonymous character and was made to appeal to the large middle-class clientele of the bazaars, but was also acceptable to the aristocracy and even the courts and hardly ever created a resale problem’ (Ettinghausen [1970], 127). 104   Caiger-Smith (1985), 66, 73. 105   See notes 9 and 10, and Melikian-Chirvani (1984). 106   This is even the case with traditional equestrian images; thus the natural authority of the cavalier depicted on the Brangwyn plate (Pope and Ackerman [1939], pl.632) is weakened by the floral ornament that proliferates all over his clothing and by the fact that his horse is covered with clover-shaped spots. There is no sign of any game for him to hunt. 102 

24

into insignificance. One may note too the implied tendency to greater equality between ruler and ruled, in that the prince is depicted as a human being among many other human beings, and not markedly larger than them. The godlike remoteness of the ruler so carefully fostered in Sasanian art, often expressed by images showing him on his own, has disappeared. One may note an intimate, almost domestic quality in these scenes; the absence of chamberlains, cupbearers and other office-holders, formal thrones and crowns – indeed most of the official paraphernalia of kingship – all indicate that most of these pieces are not to be interpreted as formal diwan images. Altogether, a new informality reigns even in ostensibly royal images – indeed, in the Oxford dish, the ruler is virtually in mufti. The prince may be seated at his ease, flanked by attendants who 107 are of the same size and presumably his intimate companions. It could be argued that some levelling, some democratization, has occurred simply by virtue of transferring a kingly image from a royal or courtly medium to a popular one. The process can be documented most readily in the Nishapur buff wares, which often employ recycled Sasanian 108 royal images. Not only does this devalue the currency by making it too widely available; it also distorts, simplifies and drains the meaning from the original. Saljuq lustre ceramics are a different matter altogether. Since they were luxury wares, they were automatically directed towards the more 109 well-to-do members of the community. Their technical quality, especially in sheer draughtsmanship, is immeasurably superior. And they stick far more closely than do the Nishapur wares to their putative models – in this 110 case, book painting and, especially in their shapes, metalwork. Religious elements The cumulative effect of the details of face, pose, headgear, throne and hand gestures analysed above is to invest the central figure with a degree of authority which would be just as appropriate for someone wielding spiritual power as for a ruler of this world. But it is principally the sideways glance of the prince on the Oxford dish, his faraway look, and the trance-like stillness of his attendants, which suggest a mental and spiritual absorption at the opposite extreme to the standard icons of power which this tableau   Both the Oxford dish (Watson [1985], pl.B) and the Tehran dish are of this kind.   See Fitzherbert (1983). 109   Caiger-Smith (1985), 59. 110   Caiger-Smith (1985), 59–60, 71. 107 108

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

25

superficially resembles. The Tehran dish has the self-same rapt quality, 112 and traces of it can be sensed also in the New York and Rabenou bowls. In the Oxford dish, moreover, the figures on either side of the prince have not been demoted into faceless nonentities, as is so often the fate of the ruler’s attendants in formal enthronement scenes. They are differentiated from each other, they have a distinct presence and they yield little to him in size. All this makes him the first among equals, an impression more suited to a scene of mystics meeting for a common purpose than to a royal audience. The meditative, spiritual quality of most of the dishes under discussion here is in striking contrast with the otherwise popular 113 image of the figure of authority as a man of action, whether in combat 114 or some form of recreation. Images of this more contemplative kind are an appropriate visual reminder that it was part of the ruler’s duty to foster intellectual and literary pursuits, not just to engage in hunting, war, revelling and the transaction of state affairs. It is as if a new visual image of rulership were evolving in which the spiritual insight and piety of the 115 monarch comes to the fore. Contemporary caliphs showed the way here 116 – al-Nasir with his encouragement of the futuwwa organization and al117 Mustansir with his grandiose project of a universal madrasa. The strong Sufi elements in contemporary metalwork – which mainly take the form 118 of inscriptions – offer relevant corroborative evidence. Conversely, some Sufi masters were held in very high regard in the eastern Islamic world at this period, and kept well-nigh royal state. Contemporary authors note this phenomenon in Syria, for example, where, around 1180, according to Ibn 119 Jubair, Sufis were ‘really the kings in these parts’. Sufi themes have been 120 recognized in the inscriptions on Kashan lustre tiles and thus it would not be strange if similar ideas had infiltrated the pictorial repertoire of 121 other Kashan lustreware. 111

  Caiger-Smith (1985), 66, 68–69.   Caiger-Smith (1973), caption to pl.24. 113   Grube (1976), col. pl. opposite 216. 114   Watson (1985), col. pl.E. 115   This is illustrated by the so-called Muharram bowls (Bagherzadeh [1989]); for colour plates, see Pope and Ackerman (1939), pls 686–87, 689; cf. also pl.688). 116   Hartmann (1975), 92–108. 117  C. Hillenbrand (1992). 118   Melikian-Chirvani (1976), 287–91. 119  Trimingham (1971), 10. 120   Melikian (1966), 253–55. 121   Ettinghausen & Guest (1961), 60–63. 111 112

26

As noted earlier, the basic compositional scheme used on these wares is found long before the Islamic period. It was flexible enough to accommodate itself to drastically different formats, contexts and meanings, and often has religious connotations. It is found in various guises in late antique and Byzantine metalware, such as some of the David 122 123 plates or the missorium of Theodosius. The latter example represents perhaps its most formal and official rendering. A tripartite palatial building serves both as the backdrop to the scene and as a device for separating the three imperial protagonists – Theodosius himself, flanked by a coemperor on each side. In the pediment of the building there fly two putti, each bearing a wreath – clear references to heaven and to its approbation of the main scene, in which the emperor delivers a codicil of commission 124 to an official. Below, in a semi-circular exergue divided by the plinth of the building itself from the action above, reclines the goddess Terra, carrying the fruits of the earth and ready to present them to the emperor, to whom her gaze is directed. Effectively, then, the composition is divided into five segments, all of them defined not by arbitrary divisions but by the actual architecture of the building. In imperial and consular diptychs this elegant solution to the problem of compartmentalizing a complex image was rejected in favour of a simple set of linear divisions. And it is important to remember that the same physical layout was used both for 125 secular and for religious images. Moreover, this same basically five-part arrangement, sometimes even with the self-same association of the top part with heaven and the lowest with earth, occurs in medieval Buddhist art, for example in Tibet. It is a 126 127 128 standard feature of tangkas, embroideries or stelai in which the figure of power is flanked by a smaller figure on each side, with yet smaller figures in a predella below, and matched by similar heavenly figures above. Even more striking is the resemblance to figures of arhats – Buddhist holy men, often hermits or mystics – who are, as it happens, often depicted in a seated 129 position looking to the left with a faraway expression in their eyes. The inwardness of the princely figure on the Oxford dish comes surprisingly   See note 16.   Weitzmann (1979), 75. 124   King (1960). 125  Natanson (1953), pls 43, 45, 50. 126  Rhie & Thurman (1992), figs 11–12, cat.3. 127   Zwalf (1985), cat.311. 128   Rhie & Thurman (1992), cat.22. Cf. also Zwalf (1985), cats. 125, 132, 190, 353. 129   Rhie & Thurman (1992), cat.13. 122 123

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

27

close to such devotional images. An integral part of this connection is the marked downgrading of the flanking figures into mere acolytes. The princely figure, in other words, takes on authority in proportion to their surrender of it. In much the same way, the attendant figures in Buddhist art are set further back from the main figure, which is usually much larger than they are, and occupy a separate space. In the Oxford dish, it is noticeable that the ruler’s face bears the imprint of a personality. It would be going too far to suggest that the princely figure is intended to represent ‘Ali. After all, he does not bear the most distinctive attribute of ‘Ali, namely his two-pronged sword, nor does he 131 wear a turban drawn up under the chin in the Arab fashion. But the atmosphere created by this image would, as suggested earlier, be consistent 132 with, say, a scene of contemplation or of some Sufi gathering. Mystical connotations have indeed been noted in the Tehran dish, where only three figures are shown. It may perhaps be argued that such discussions of putative religious significance in these pieces are irrelevant, on the grounds that lustreware of this period probably had no devotional function. Yet the evidence of the Muharram minai plates signed by Abu Zaid should give one pause. Dr Firuz Bagherzadeh has demonstrated that scenes of Muharram ceremonies are repeatedly depicted on wares signed by this artist and dated, one after 133 134 another, to the month of Muharram – in 583, 586 and 587 for example – and these also show a figure of authority accompanied by attendants. Thus subject matter of religious significance was employed on other types of luxury ware at exactly this period. Certain elements in these compositions may carry more than one meaning. In particular, a whole series of motifs lend themselves to cosmological interpretations, especially when they are used in concert and in their appropriate locations. These include fish swimming in the exergue, sun-faces, canopies, birds and flying angel figures. In earlier occurrences of several such themes in Islamic art their symbolic significance is plain. Thus, below the enthroned prince at Qusayr ‘Amra, the early eighthcentury Umayyad hunting lodge in Jordan, there develops a Nilotic scene 130

  For the Buddhist element in pre-Mongol Iran, see Melikian-Chirvani (1974).   For examples from the copy of al-Biruni’s Al-athar al-baqiya of 1306, see Soucek (1975), fig.7. 132   Brend (1991), colour frontispiece. 133   Bagherzadeh (1989). 134   Watson (1994), 172–73. 130 131

28

with plenty of fish swimming about, while the extrados of the arch which 135 frames him is filled with birds. This same combination, now reduced to one bird above and one fish below a cross-legged figure holding a cup, 136 can be seen on a piece of Samarran lustreware. The connection of these creatures with the realms of the air and the sea is an ancient one, no doubt because it is so obvious, and the Samarran lustre bowl furnishes eloquent testimony in its very abstraction of how such symbols were understood. In later periods a pool of fish is often shown at the base of an object, either in the exergue of a dish (as in three of the pieces under discussion 137 here) or in the well of a bowl or other vessel. In the Debenham plate, these associations are made especially plain because the significant motifs have equally significant locations. At the very bottom of the plate is a pool full of fish. Immediately above is a file of animals and birds. The two main figures and their retinue are distributed above these bands, and at the very top of the plate can be seen two flying angels facing each other. 138 Thus heavenly beings watch over the royal pair, who clearly dominate the creatures of land, sea and air. Beneath the somewhat fussy and redundant detail may be discerned the idea of the monarch as cosmocrator. In much the same way the prince in the Tehran dish is depicted with fish below him, and birds completely encircle him. This is a very unusual feature; it seems to suggest some honorific intent, and may also imply that he has something celestial, and thus holy, about him. The ancient concept of apotheosis, for which birds were a popular symbol, may underlie 139 such images, especially when birds are depicted on the canopy above 140 the ruler. Certain Qur’anic inscriptions on Kashan tiles have the sacred 141 text surrounded by birds in much the same way. The combination of

 Spuler & Sourdel-Thomine (1973), pl.33.  Otto-Dorn (1967), 83, fig.28 (drawing) and Otto-Dorn (1982), Abb.4 (plate). 137  Baer (1968) and Ettinghausen & Guest (1961), 61; cf. Caiger-Smith (1973), pl.27. 138  See note 93. 139  L’Orange (1953), 69–72, 118-19; Ettinghausen (1972), 22, fig.D – a reconstruction of the diwan at Khirbat al-Mafjar – where a file of birds, apparently partridges, occupies the collar of the dome above roundels depicting winged horses, themselves symbols of apotheosis (Ettinghausen [1972], 14). 140  As in a star tile in the Metropolitan Museum of Art (Ettinghausen & Grabar [1987], pl.373). 141  Thus on a 13th-century tile in the Victoria and Albert Museum (acc. no. 729.1888: Lane [1960], pl.1b) can be read part of Sura 6: 165. For other examples, see Ettinghausen (1936), 44-45, fig.5. 135 136

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

29

birds and fish can be explained in a mystical sense too, though of course it could be read simultaneously at face value as a reference to a garden setting with a pool. As for the Rabenou and New York bowls, fish once again occupy the exergue, while the topmost reaches of the composition are filled with a canopy bearing rich scrolls. It has to be conceded that these same formulae of fish and canopy occur in other lustrewares where the central image 143 consists of an animal or a pair of animals, so it seems doubtful whether they always embody the same kind of solar and lunar connotations that seem appropriate when they are used in association with enthroned 144 monarchs. This kind of canopy seems to function as visual shorthand for the sky. It takes the same curved quadrant shape as the sun-face on the Oxford dish, and is elaborately decorated with vegetal scrolls. Such canopies are frequently depicted in Timurid painting in scenes of a ruler 145 relaxing in an open-air setting. The form of the canopy in the Rabenou bowl suggests a dome, not very accurately, but nevertheless with sloping sides and a segment scalloped out of the top in the manner of an oculus. Iconographically, therefore, if not physically, the canopy evokes a dome. In that sense a dome of cloth is an entirely adequate substitute for the real thing. The bowl form itself was, as contemporary Persian poetry shows, a favoured metaphor for the heavens, which were often referred to as an 146 upturned bowl. The sun-face on the Oxford dish could perhaps be seen as a version of that same theme. 142

Conclusion It will be evident that the issues raised by this small group of pieces are wide-ranging, and some of them are very complex. This paper is too short to enable definite conclusions to be reached on many of the topics it raises; it should rather be seen as a report on work in progress, and the tentative nature of its findings must be emphasized. But ambiguity lies at the heart of these luxury ceramics and it is this factor above all which   Schimmel (1976), 306-308 and Ettinghausen & Guest (1961), 61; cf. also Bagherzadeh (1981), 21, 154 (also col. pl.35). Pope suggests, citing the Bundahisn, an echo of ancient fertility cults (Pope and Ackerman [1939], 834, 1591). If so, it is a very distant echo indeed. 143  Bahrami (1949), pls XLIX–L. 144  Caiger-Smith (1985), 191. 145  Lentz & Lowry (1989), pls on 258, 260, 265. Cf. Sims (1973), 298–301, for a study of the related royal feature of the parasol. 146  Melikian-Chirvani (1994), 147. 142

30

invests them with that sense of mystery and power to which they owe their peculiar resonance. Bibliography Allan (1982) J.W. Allan, Islamic Metalwork. The Nuhad Es-Said Collection, London, 1982. Allan (forthcoming) = J.W.Allan, ‘Solar and celestial symbolism in medieval Islamic art’, in Image and Meaning in of Islamic Art, ed. R. Hillenbrand, London, 2005, 34–41. Artsruni (1985) T. Artsruni, History of the House of the Artsrunik, trans. R.W. Thomson, Detroit, 1985. Atil (1973) E. Atil, Ceramics from the World of Islam, exhibition catalogue, Freer Gallery of Art, Washington, DC, 1973. Baer (1968) E. Baer, ‘Fish-pond Ornaments on Persian and Mamluk Metal Vessels’, Bulletin of the School of Oriental and African Studies. XXXI/1, 1968, 14–27. Baer (1981) E. Baer, ‘The Ruler in Cosmic Setting: A Note on Medieval Islamic Iconography’, in Essays in Islamic Art and Architecture In Honor of Katharina Otto-Dorn, ed. A. Daneshvari, Malibu, 1981, 13–19. Bagherzadeh (1981) F. Bagherzadeh, The World’s Great Ceramics Collections. Vol. 4. Iran Bastan Museum Teheran, Tokyo, New York and San Francisco, 1981. Bagherzadeh (1989) F. Bagherzadeh, ‘Iconographie iranienne. Deux illustrations de Xel’at de l’Année 583H./1187 après JC’, in Archaeologia Iranica et Orientalis, Miscellanea in Honorem Louis vanden Berghe, ed. L. de Meyer and E. Haerinck, Vol.II, Ghent, 1989. Bahrami (1937) M. Bahrami, Recherches sur les carreaux de revêtement lustré dans la céramique persane du XIIIe au XVe siècle (étoiles et croix), Paris, 1937. Bahrami (1949) M. Bahrami, Gurgan Faiences, Cairo, 1949. Björkman (1934) W. Björkman, ‘Turban’, in The Encyclopaedia of Islam, 1st edn, 1934, cols 885a–92a.

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

31

Blair et al. (1992) S. Blair, J.M. Bloom and A.E. Wardwell, ‘Reevaluating the Date of the “Buyid” Silks by Epigraphic and Radiocarbon Analysis’, Ars Orientalis 22, 1992, 1–41. Brend (1991) B. Brend, Islamic Art, London, 1991. Caiger-Smith (1973) A. Caiger-Smith, ‘Tin-Glaze Pottery in Europe and the Islamic World. The Tradition of 1000 Years’, in A.CaigerSmith, Maiolica, Faience & Delftware, London, 1973. Caiger-Smith (1985) A. Caiger-Smith, Lustre Pottery. Technique, Tradition and Innovation in Islam and the Western World, London, 1985. Caiger-Smith (1986) A. Caiger-Smith, ‘Notes on the Colloquium on Ceramics and Metalware, Oxford, March 1985’, in Pots and Pans. A Colloquium on Precious Metals and Ceramics in the Muslim, Chinese and Graeco–Roman Worlds, Oxford, 1985, ed. M. Vickers, Oxford, 1986, 213–15. Donovan (1988–89) D. Donovan, ‘The Evolution of the Throne in Early Persian Painting: The Evidence of the Edinburgh Jami‘ alTawarikh’, Persica XIII, 1988–89, 1–76. K. Erdmann (1951) K. Erdmann, ‘Die Entwicklung der sasanidischen Krone’, Ars Islamica XV–XVI, 1951, 87–123. H. Erdmann (1983) H. Erdmann, ‘Die hängende Krone des Berk-Yaruk Ibn Malik Schahs’, Beiträge zur Altertumskunde Kleinasiens. Festschrift für Kurt Bittel, eds R.M. Boehmer and H. Hauptmann, I, Mainz, 1983, 171–74. Esin (1968) E. Esin, ‘Turkic and Ilkhanid Universal Monarch Representations and the Cakravartin’, Proceedings of the XXVIth Congress of Orientalists, II, Delhi, 1968, 86–132. Esin (1970a) E. Esin, ‘Bedük Börk: the iconography of Turkish honorific headgears’, Proceedings of the IXth Meeting of the Permanent International Altaistic Conference, Naples, 1970, 72–138. Esin (1970b) E. Esin, ‘Ay-Bitgi, The court attendants in Turkish iconography’, Central Asiatic Journal XIV, 1970, 78–117. Esin (1971) E. Esin, ‘Oldrug-Turug. The hierarchy of sedent postures in Turkish iconography’, Kunst des Orients VII, 1970–71, 1–29.

32

Esin (1979) E. Esin, ‘Turk-i Mah Chihrah (The Turkish Norm of Beauty in Iran)’, Akten des VII. Internationalen Kongresses für Iranische Kunst und Archäologie. München 7.–10. September 1976, ed. W. Kleiss, Berlin, 1979, 449–60. Ettinghausen (1936) R. Ettinghausen, ‘Evidence for the Identification of Kashan Pottery’, Ars Islamica, III/1, 1936, 44–70. Ettinghausen (1962) R. Ettinghausen, Arab Painting, Geneva, 1962. Ettinghausen (1970) R. Ettinghausen, ‘The Flowering of Seljuq Art’, Metropolitan Museum Journal, III, 1970, 113–31. Ettinghausen (1972) R. Ettinghausen, From Byzantium to Sasanian Iran and the Islamic World, Leiden, 1972. Ettinghausen (1979) R. Ettinghausen, ‘The Taming of the Horror Vacui in Islamic Art’, Proceedings of the American Philosophical Society 123/1, 1979, 15–28. Ettinghausen & Grabar (1987) R. Ettinghausen and O. Grabar, The Art and Architecture of Islam 650–1250, Harmondsworth, 1987. Ettinghausen & Guest (1961) R. Ettinghausen and G.D. Guest, ‘The Iconography of a Kashan Lustre Plate’, Ars Orientalis IV, 1961, 25–64. Fères (1953) B. Farès, Le Livre de la Thériaque. Manuscrit arabe à peintures de la fin du XIIe siècle conservé à la Bibliothèque Nationale de Paris, Cairo, 1953. Fares (1961) B. Fares, ‘Vision chrétienne et signes musulmans autour d’un manuscrit arabe illustré au XIIIe siècle’, Mémoires de l’Institut d’Egypte LVI, Cairo, 1961. Fitzherbert (1983) T. Fitzherbert, Themes and Images on the Animate Buff Ware of Medieval Nishapur, unpublished M.Phil. thesis, University of Oxford, 1983. Ghirshman (1962) R. Ghirshman, Iran. Parthians and Sassanians, trans. S. Gilbert and J. Emmons, London, 1962. Grabar (1968) A. Grabar, ‘Trônes épiscopaux’, in L’art de la fin de l’antiquité et du Moyen Age I, 365–92 and II, pls 90–102. Grube (1959) E.J. Grube, ‘Materialien zum Dioskurides Arabicus’, in

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

33

Aus der Welt der islamischen Kunst. Festschrift für Ernst Kühnel zum 75. Geburtstag am 26.10.1957, ed. R. Ettinghausen, Berlin, 1959, 163–94. Grube (1976) E.J. Grube, Islamic Pottery of the Eighth to the Fifteenth Century in the Keir Collection, London, 1976. Hagedorn (1989) A. Hagedorn, Ikonographie und Bedeutung islamischer Metallarbeiten des 13. und 14. Jahrhunderts aus Syrien, Mesopotamien und Ägypten, dargestellt am Beispiel der sog. Blacas-Kanne (London, British Museum, Inv. Nr. 66.12–69.61), unpublished D.Phil. thesis, Rheinische Friedrich-Wilhelms Universität zu Bonn, 1989. Hartmann (1975) A. Hartmann, an-Nasir li-Din Allah (1180–1225). Politik, Religion, Kultur in der späten ‘Abbasidenzeit, Berlin, 1975. Hayes (1976) J.R. Hayes, ed., The Genius of Arab Civilisation. Source of Renaissance, Oxford, 1976. Herrmann (1977) G. Herrmann, The Iranian Revival, Oxford, 1977. C. Hillenbrand (1992) C. Hillenbrand, ‘Al-Mustansir’, in The Encyclopaedia of Islam, 2nd edition, VII, 1992, columns 727a–729a. C. Hillenbrand (forthcoming) = C.Hillenbrand, ‘Women of the Seljung Period’, in Women in Iran from the Rise of Islam to 1800, eds G. Nashat and L. Beck, Urbana and Chicago, 2003, 103–20. R. Hillenbrand (1988) R. Hillenbrand, ‘The Symbolism of the Rayed Nimbus in Early Islamic Art’, Cosmos II, 1988, 1–52. R. Hillenbrand (1994) R. Hillenbrand, ‘The relationship between book painting and luxury ceramics in 13th-century Iran’, in The Art of the Saljuqs in Iran and Anatolia, ed. R. Hillenbrand, Costa Mesa, 1994, 134–45. Ipşiroğlu (1964) M.S. Ipşiroğlu, Saray-Alben. Diez’sche Klebebände aus den Berliner Sammlungen, Verzeichnis der Orientalischen Handschriften in Deutschland, VIII, Wiesbaden, 1964. Ipşiroğlu (1971) M.S. Ipşiroğlu, Das Bild im Islam. Ein Verbot und seine Folgen, Vienna and Munich, 1971. Jones & Michell (1976) D. Jones and G. Michell, eds., The Arts of Islam, exhibition catalogue, London, 1976.

34

al-Juzjani (1977), Tabakat-i-Nasiri: A General History of the Muhammadan Dynasties of Asia, including Hindustan; from A.H. 194 (810 A.D.) to A.H. 658 (1260 A.D.) and the Irruption of the Infidel Mughals into Islam, trans. H.G. Raverty, 2 vols, London, 1881–99, repr. Lahore, 1977. King (1960) N.Q. King, There’s such Divinity doth hedge a King, Edinburgh, 1960. Lane (1960) A. Lane, Victoria and Albert Museum. A Guide to the Collection of Tiles, London, 1960. Lapidus (1969) I.M. Lapidus, ‘Muslim Cities and Islamic Societies’, in Middle Eastern Cities, ed. I.M. Lapidus, Berkeley and Los Angeles, 1969, 47–74. Lentz & Lowry (1989) T.W. Lentz and G.D. Lowry, Timur and the Princely Vision. Persian Art and Culture in the Fifteenth Century, Los Angeles and Washington, 1989. L’Orange (1953) H.P. L’Orange, Studies in the iconography of cosmic kingship in the ancient world, Instituttet for Sammenlignende Kulturforskning, XXIII, Oslo, 1953. Lowry & Nemazee (1988) G.D. Lowry and S. Nemazee, A Jeweler’s Eye. Islamic Arts of the Book from the Vever Collection, Seattle and London, 1988. Marshak (1986) B. Marshak, Silberschätze des Orients. Metallkunst des 3.–13. Jahrhunderts und ihre Kontinuität, trans. L. Schirmer, Leipzig, 1986. Melikian (1966) S. Melikian [Melikian-Chirvani, A.S.], ‘The Sufi strain in the art of Kashan. Bab al-Ma‘rifat (The Gate of Knowledge)’, Oriental Art XII, 1966, 251–58. Melikian-Chirvani (1970) A.S. Melikian-Chirvani, Le roman de Varqe et Golšah. Essai sur les rapports de l’ésthetique littéraire et de l’ésthetique plastique dans l’Iran pré-mongol, suivi de la traduction du poème, Arts Asiatiques XXII, numéro special, 1970. Melikian-Chirvani (1974) A.S. Melikian-Chirvani, ‘L’évocation littéraire du bouddhisme dans l’Iran médiéval’, Le Monde Iranien et l’Islam II, 1974, 1–72. Melikian-Chirvani (1976) A.S. Melikian-Chirvani, ‘Iranian metal work and the written word’, Apollo CIII, 170, 1976, 286–91.

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

35

Melikian-Chirvani (1984) A.S. Melikian-Chirvani, ‘Le Shah-name, la gnose soufie et le pouvoir mongol’, Journal Asiatique CCLXXII/3–4, 1984, 249–337. Melikian-Chirvani (1988) A.S. Melikian-Chirvani, ‘Le livre des Rois, miroir du destin’, Studia Iranica 17/1, 1988, 7–46. Melikian-Chirvani (1991) A.S. Melikian-Chirvani, ‘Le livre des rois, miroir du destin. II – Takht-e Soleyman et la symbolique du Shah-Name’, Studia Iranica 20/1, 1991, 33–148. Melikian-Chirvani (1994) A.S. Melikian-Chirvani, ‘The Light of the World’, in The Art of the Saljuqs in Iran and Anatolia, ed. R. Hillenbrand, Costa Mesa, 1994, 146–55. Milstein & Brosh (1984) R. Milstein and N. Brosh, Islamic Painting in the Israel Museum, Jerusalem, 1984. Natanson (1953) J. Natanson, Early Christian Ivories, London, 1953. Naumann (1976) R. Naumann and E. Naumann, Takht-i Suleiman. Ausgrabung des Deutschen Archäologischen Instituts in Iran, Prähistorische Staatssamlung München, Munich, 1976. Nurbakhsh (1980) J. Nurbakhsh, What the Sufis say, New York, 1980. Önder (1978) M. Önder, ‘Le mode de se vêtir et de se parer chez les dames seldjoukides’, in Fifth International Congress of Turkish Art, ed. G. Fehér, Budapest, 1978, 655–63. Öney (1976) G. Öney, Türk Çini Sanati. Turkish Tile Art, Istanbul, 1976. Otto-Dorn (1967) K. Otto-Dorn, L’art de l’Islam, trans. J.-P. Simon, Paris, 1967. Otto-Dorn (1982) K. Otto-Dorn, ‘Das seldschukische Thronbild’, Persica X, 1982, 149–94. Pope (1935) A.U. Pope, ‘Notes on the Stucco Ornament in the Sanctuary of the Masjid-i-Jami‘, Qazvin’, Bulletin of the American Institute of Persian Art and Archaeology, IV/1 1935, 209–16. Pope and Ackerman (1939) A.U. Pope and Ackerman eds., A Survey of Persian Art from Prehistoric Times to the Present, Oxford, 1938–39. Rhie & Thurman (1992) M.M. Rhie and R.A.F. Thurman, Wisdom and

36

Compassion. The Sacred Art of Tibet, London, 1992. Rogers et al. (1986) J.M. Rogers, F. Çağman and Z. Tanindi, Topkapi. The Albums and Illustrated Manuscripts, London, 1986. Rosenthal (1971) F. Rosenthal, ‘A Note on the Mandil’, in F. Rosenthal, Four Essays on Art and Literature in Islam, Leiden, 1971, 63–99. Roux (1982) J.-P. Roux, Études d’iconographie islamique. Quelques objets numineux des Turcs et des Mongols, Paris and Leuven, 1982. Sadan (1976) J. Sadan, Le Mobilier au Proche Orient Médiéval, Leiden, 1976. Schlumberger & Sourdel-Thomine (1978) D. Schlumberger and J. Sourdel-Thomine, Lashkari Bazar. Une résidence royale ghaznévide et ghoride, 2 vols, Paris, 1978. Schimmel (1975) A. Schimmel, Mystical Dimensions of Islam, Chapel Hill, 1975. Schnyder (1994) R. Schnyder, ‘In search of the substance of light’, in The Art of the Saljuqs in Iran and Anatolia, ed. R. Hillenbrand, Costa Mesa, 1994, 165–69. Schroeder (1939) E. Schroeder, ‘Ahmed Musa and Shams al-Din: a review of fourteenth-century painting’, Ars Islamica VI/2, 1939, 113–42. Shahbazi (1993) S. Shahbazi, ‘Crown. iv. ‘Of Persian Rulers from the Islamic Conquest to the Qajar Period’, Encyclopaedia Iranica, VI, 1993, columns 421G–425a. Sims (1973) E.G. Sims, The Garrett Manuscript of the Zafar-Name: A Study in Fifteenth-Century Timurid Patronage, unpublished Ph.D. thesis, New York University, 1973. Soucek (1975) P.P. Soucek, ‘An Illustrated Manuscript of al-Biruni’s Chronology of Ancient Nations’, in The Scholar and the Saint, ed. P. Chelkowski, New York, 103–68. Spuler & Sourdel-Thomine (1973) B. Spuler and J. Sourdel-Thomine, Die Kunst des Islam (Propyläen Kunstgeschichte. Band 4), Berlin, 1973. Storm Rice (1958) D.S. Rice, ‘A drawing of the Fatimid period’, Bulletin of the School of Oriental and African Studies XXI, 1958, 31–39.

IMAGES OF AUTHORITY ON KASHAN LUSTREWARE

37

Talbot Rice (1971) D.T. Rice, Islamic Painting. A Survey, Edinburgh, 1971. Talbot Rice (1976) D.T. Rice, The Illustrations to the ‘World History’ of Rashid al-Din, ed. B. Gray, Edinburgh, 1976. Trimingham (1971) J.S. Trimingham, The Sufi Orders in Islam, Oxford, 1971. von Gall (1971) H. von Gall, ‘Entwicklung und Gestalt des Thrones im vorislamischen Iran’, Archaeologische Mitteilungen aus Iran N.F., 4, 1971, 207–35. Watson (1985) O. Watson, Persian Lustre Ware, London, 1985. Watson (1986) O. Watson, ‘Pottery and Metal Shapes in Persia in the 12th and 13th Centuries’, in Pots and Pans. A Colloquium on Precious Metals and Ceramics in the Muslim, Chinese and Graeco-Roman Worlds, Oxford 1985, ed. M. Vickers, Oxford Studies in Islamic Art IV, 1986, 205–12. Watson (1994) O. Watson, ‘Documentary mina’i and Abu Zaid’s bowls’, in The Art of the Saljuqs in Iran and Anatolia, ed. R. Hillenbrand, Costa Mesa, 1994, 170–80. Weitzmann (1979) K. Weitzmann, Age of Spirituality. Late Antique and Early Christian Art, Third to Seventh Century, New York, 1979. Whelan (1988) E. Whelan, ‘Representations of the Khassikiyah and the Origins of Mamluk Emblems’, in Content and Context of Visual Arts in the Islamic World. Papers from a Colloquium in Memory of Richard Ettinghausen. Institute of Fine Arts, New York University, 2–4 April 1980. Planned and Organised by Carol Manson Bier, ed. P.P. Soucek, University Park and London, 1988, 219–53. Zygulski (1992) Z. Zygulski Jr., Ottoman Art in the Service of the Empire, New York, 1992. Zwalf (1985) Buddhism. Art and Faith, ed.,W. Zwalf, London, 1985.

38

II The Sarcophagus of Shah Ismā‘īl at Ardabīl

T

he focus of this paper is the sarcophagus of Shah Ismā‘īl I, which is lodged in his extremely bijou mausoleum at the Ardabīl shrine in 1 north-west Iran, and can be dated soon after 930/1524. This splendid and unaccountably neglected work of art (pl. 1) is of special interest on several counts. First, it reflects, to a degree unusual in woodwork, the dominant aesthetic of its time, an aesthetic which pervaded the visual backdrop of the early Safavid court and can be traced in several media, such as 2 architecture, painting and perhaps especially metalwork. Second, so far as the surviving evidence tells us, it incorporates major innovations for this kind of object, notably in its technicolour palette. Third, it has its own distinctive religious message proclaimed by its inscriptions and referring to Shah Ismā‘īl himself. That last point is admittedly no more than an educated guess, for no historical inscription, apart from a craftsman’s signature, has been found on the sarcophagus. But the building in which it is placed is indeed the mausoleum of Ismā‘īl, so it seems reasonable to assume that the sarcophagus should also be his. I hope finally to suggest why its appearance is peculiarly appropriate to its time.   On this mausoleum, see my article entitled ‘The tomb of Shah Ismā‘il I, Ardabīl’ in S.R. Canby (ed.), Safavid Art & Architecture (London: The British Museum Press, 2002), 3–8. 2  See for example J.W. Allan, ‘Silver-faced doors of Safavid Iran’, Iran 33 (1995), 123–37. 1

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

39

It is curious to note that early Safavid art in general, despite the high profile allotted to court painting, such as the royal Shāhnāmeh and Khamseh manuscripts, is not at all well known in modern scholarship,. This alone is enough to lend this particular sarcophagus considerable rarity value as a clue to the nature of the earliest Safavid art. But it is of greater interest still because of its surroundings. The fact that the Ardabīl shrine has preserved several silver-plated doors, carpets, felt floor coverings, woodwork, metal screens, steel standards and wall painting, all placed within something close to their original setting, makes it nothing short of a time capsule by means of which one can take the measure of Safavid art, especially early Safavid art, and the aesthetic which it fostered. Here, and only here in all of Iran, the arts of the early Safavid period, from architecture to tombstones, are displayed side by side and medium by medium. As a result, the decorative arts can be appreciated to the full, subtly enhancing each other and almost bandying themes across the space of the shrine. There could be no better reminder that these works of art were deliberately complementary and are diminished if studied intensively in isolation. There is a further danger here that the very uniqueness of this survival of early Safavid works of art at Ardabīl—and what remains there today is certainly much less than there was two centuries ago, before the Russians removed some of its choicest items to St Petersburg, ostensibly for safe keeping but in fact for good—leads to the danger of misinterpreting this material, and in particular concluding that it was unique to late medieval Iran. Any such finding would be badly mistaken. The evidence of Ilkhanid and even more of Timurid painting reveals that the ambience of courtly luxury which these images document so precisely depended on all sorts of luxury objects which have totally disappeared. So much for the general context of the sarcophagus. But a brief digression on size is called for here. The mausoleum itself, as noted above, is bijou. It is worth dwelling on this dimension of scale for a moment. The total area of the tomb chamber is 6.4 square metres, while that of the sarcophagus is 2.1 square metres, that is, about a third of the entire space. Since it is set in the very middle of the room, its dimensions make it impossible for a visitor to walk around it in comfort, and at times there 3 is barely enough room to squeeze through. This adds illusionistically to its size. Indeed, it seems to fill all the room available—an effect not confined 3   This excessively cramped space greatly inhibits a close study of the entire object, and is exacerbated by the poor light in some areas; these factors combine to make a complete photographic survey impracticable.

40

to the horizontal plane, for the very low vault ensures that the height of the sarcophagus (1.44 metres) is also the dominant vertical accent of the interior space. It is in two stages, raised plinth and superstructure, a design that has honorific associations: the shah, even in death, is above the level of the rest of mankind. By these various means, then, the disadvantages attendant on a cramped space are triumphantly overcome, and the shah receives a worthy memorial. His dead presence literally fills the room. These facts have further implications. They suggest, first, that the sarcophagus was assembled within this very cramped space rather than being transported as an entity into the mausoleum. That, incidentally, turned out to be a blessing in disguise, for it ensured that the Russians were not able to make off with it when they pillaged the shrine so thoroughly in 4 1827. Next, it is clear that the sarcophagus was not intended to serve any rite of circumambulation or ziyārat. Moreover, entrance to the chamber was barred by a gilded metal grille, so that for most visitors to the shrine in the Safavid period, as is the case today, the immediate impact of this sarcophagus within its technicolour blue and gold tiled setting would have been at a distance and as a whole. Today in fact the door is kept locked, so visitors to the shrine have to content themselves with the view they get through the grille. The resultant aura of inaccessibility, apartness and mystery which the sarcophagus projects to this day would have been all the stronger at a time when the memory of the shah’s charisma was still fresh. So much for the digression. It will be convenient to begin with a detailed description of the sarcophagus and of its decoration before moving on to its inscriptions, to the issue of its patronage, and to the aesthetic to which it belongs. This last subject will entail some discussion both of earlier woodwork and of related work in other media. It should be noted that vandals have been busy all over the sarcophagus, prising out ivory   C.E. Bosworth, ‘Ardabīl. I. History of Ardabīl’, Encyclopaedia Iranica II/4 (London and New York: Routledge and Kegan Paul Ltd., 1986), 360; he notes the loss of many manuscripts. See too R.N. Frye, ‘Ardabīl’, Encyclopaedia of Islam, 2nd ed. (EI2), 1: 626, with a mention of the removal of ‘art objects’. In the previous century, the city was captured by the Ottomans in the course of their campaign in Azerbayjān (1137/1725); see S. Shaw, ‘Iranian relations with the Ottoman Empire in the eighteenth and nineteenth centuries’, in P. Avery, G. Hambly and C. Melville (eds), The Cambridge History of Iran. 7. From Nadir Shah to the Islamic Republic (Cambridge: Cambridge University Press, 1991), 299. But it is not clear whether the shrine or its contents suffered as a result. Āqā Muḥammad Khān Qājār visited the shrine in 1206/1791. See G.R.G. Hambly, ‘Agha Muhammad Khan and the establishment of the Qajar dynasty’, ibid., 122; but again it is uncertain whether he changed its aspect in any way. 4

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

41

inscription panels—the most grievous loss of all—turquoise-coloured insets, border strips and interlace work. First, then, a description of the sarcophagus. It is in two parts: a plinth measuring 2.03 m by 1.03 m on which is set the sarcophagus proper, which measures 1.85 m by 0.90 m. The height of the plinth is 0.41 m while that of the sarcophagus proper is 1.03 m, including an overhang of 4 cm at the top. These measurements place it in between the largest sarcophagus of the shrine, that of Shaykh ṣafī, and another sarcophagus dated 788/1386–7. The former has a plinth measuring 3.16 m (including a cornice of 1.5 cm on each side) by 1.31 m, with a height of 0.51 m, while the sarcophagus proper measures 2.96 m by 1.15 m with a height 5 of 1.23 m. The sarcophagus of 788/1386–7, encased in glass, measures approximately 2.08 m by 0.70 m, with a height of 0.67 m and a cornice on all sides with an overhang of 2 cm. The design of the sarcophagus of Shah Ismā‘īl is consistent on all four sides. It has a tripartite layout, crowned on its vertical face by an upper 6 inscription band carved in wood and set in oblong hexagons formed by strips of ivory with regular turquoise-coloured studs. Between these long hexagons are equilateral hexagons subdivided into six triangles, each containing a sacred name. The very top of the sarcophagus, seen from above, reveals the same design as at the upper face of the sides, except that 7 its centre bears a panel of khatamkārī design measuring 1.47 m by 0.41 m. Here the ivory is not flat but rounded, which gives the whole an added three-dimensional effect. The next layer down comprises a narrower band filled with oblong panels containing inscriptions in ivory; these measure 20.3 by 4.4 cm and terminate at each end with a 5-pointed star. These panels alternate with 8-pointed stars containing interlace. The bottom layer or plinth takes up by far the most room and comprises a continuous network of stars, rhomboids and polygons (of the type long familiar in minbar design), all filled with interlace. Finally,   Most of the sarcophagi at Ardabīl keep to the formula of plinth and sarcophagus.   Said to be sandalwood. See M.E. Weaver, Iran: Preliminary Study on the Conservation Problems of Five Iranian Monuments (Paris: UNESCO, 1970) (= Weaver I), 52. 7   For a brief history of this technique, see L. Honarfar, ‘Khatam-Kar’, in J. Gluck and S.H. Gluck (eds) and C. Penton (associate editor), A Survey of Persian Handicraft (Tehran: Survey of Persian Art under the auspices of Bank Melli Iran, 1977), 362–8. I am grateful to Dr Kjeld von Folsach for drawing my attention to this article. Honarfar cites Dawlatshah to the effect that the technique is first recorded in the late 14th century (ibid., 362). For a detailed examination of the technique see H.E. Wulff, The Traditional Crafts of Persia (Cambridge, Mass., and London: The M.I.T. Press, 1966), 92–7. See also n. 59 below. 5 6

42

the plinth is treated more simply than the area above it. It has the same stellar polygonal network, but this is much shallower. The inner strips of the hexagons are varnished and are of light-coloured wood. As for the hexagons themselves, they are smaller (18, 10 and 29 cm per side) so that there is less ivory. The colour scheme is also more muted, for both green and crimson accents are absent, let alone gold or silver thread. The borders of all the elements are plainer than those in the upper parts of the sarcophagus; indeed, the upper and lower borders are both entirely plain, with a dark brown veneer which is severely damaged in places. Nor does the plinth have any inscriptions. Thus the overall design establishes a clear hierarchy of the component parts. This basic description cannot do justice to the range of colour effects and grace notes which all make their contribution to this spectacular ensemble. The treatment of borders on all three tiers of the sarcophagus is a case in point. Some are flat, some bevelled or curved; some plain, some intricate, and yet others given a staccato rhythm by insets; some monochrome, others with three colours. The colour turquoise—as distinct from the stone itself—is frequently used. The effect of these insets or studs (for some stand proud of the surrounding surface) is cumulative, and they testify to a firm grasp of the overall design as well as to a delight in exploiting the potential of very small-scale ornament. The borders are not the only area notable for the vivid interchange of colours, as is shown by the field between the two kinds of hexagons in the upper tier. This is taken up by two triangles whose upper and lower points respectively meet to create a spindle shape. The field of each of these triangles has three turquoise-coloured studs separated by six tiny ivory triangles, each comprising three lozenges, set against a black ground. This creates a light and colourful accent against the dominant sombre tonality of the wooden panels in this tier. The infill throughout the sarcophagus tends to be inconsequential and dense, featuring tightly-woven arabesques with thin, spindly tendrils and three-leaved buds or five-leaved rosettes. With the art that conceals art, it effortlessly accommodates itself to the required space. In the main field of the sarcophagus, a rhomboidal hexagon with sides of 40 cm or 26 cm is the commonest form, but this is made to appear more varied by being 8 placed at various angles. The wood is painted black and has a bronze 8   It is not ebony, though ebony was indeed used in Safavid woodwork. See K. von Folsach, Art from the World of Islam in the David Collection (Copenhagen: F. Hendricksens Eftf., 2001), pl. 449.

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

43

tone where the black has faded. As for the ivory plaques of the middle tier, these are sometimes of one colour, with ivory being employed for inscription and background alike and left in its natural colour, so that such panels have the effect of a broad bar of light, as against the standard effect of white script emerging from a dark ground. Yet in the latter case the ivory relief carving is given extra interest by being set against a background of crimson or green silk, to luscious chromatic effect. A third way of variegating these inscriptions has to do with the letters themselves, 9 which are enriched by a central groove. The script remains the same in each of these three styles of execution. But perhaps the most impressive work is, as one would expect in such a hierarchical design, reserved for the topmost panel. Its inlay of ivory alternates with flattened spindle shapes in black-painted wood inlaid with a central turquoise roundel whose circumference is studded with twelve brass triangles whose high polish gives them a rich golden colour. Yet so small is the scale of these triangles that from even a foot away this feature simply looks like a green dot against a black ground. It is a sobering thought that the detail on this sarcophagus can be so fine that it requires a magnifying glass to appreciate its subtleties. The kinship with contemporary book painting in this regard is unmistakable. Close examination of the lower and main tier reveals that slivers of black-painted wood flank strips of turquoise colour with splinters of ivory inlaid at a deeper level. Several levels are used, so that within a narrow compass there is a constant variation of depth. The smallest work is sewn together with black thread, originally covered with gold. The resultant contrast is not just one of colours but also of media and their associated 10 textures—ivory, gold and silver thread, brass and silk all enrich the basic material, namely wood. Those panels where the background is unstained ivory have a distinctively different visual impact from that of the openwork panels. The framing strips of these panels are green and are probably, in view of their length, made of wood rather than bone stained that colour. The inner borders of these ivory panels comprise thin strips in which 9   This creates a strong sculptural effect, as seen in the doors of the Gūr-i Amīr (see A.U. Pope and P. Ackerman, eds, A Survey of Persian Art from Prehistoric Times to the Present [London and New York, Oxford University Press, 1939], pl.1470), and in Qur’ans of the 8th/14th and 9th/15th centuries whose gold script is outlined in black or vice versa, sometimes on the same page; see D. James, Qur’ans of the Mamluks (London: Alexandria Press in association with Thames and Hudson, 1988), 86, fig. 53. 10   This is also the material of the corner fittings.

44

turquoise-coloured studs alternate with six ivory dots. These form spindle shapes comprising two triangles, one pointing up, the other down, with their points touching. Outside these thin strip borders are thicker borders whose multiple elements—uneven hexagons and pentagons, spindle shapes and even hexagons containing a six-pointed star on a white ground with a central turquoise-coloured stud—are all carefully dovetailed. These shapes employ both openwork and marquetry techniques. The central tier of the sarcophagus maintains a constant visual interest because the shapes of the polygonal stellar network which form its principal accent are themselves continually changing. Their infill is also varied. Sometimes it comprises arabesques, but sometimes it creates a thicket of leaves, tendrils and buds in high-relief ivory. Still further contrast is assured by the background, which is sometimes the plain white of ivory and sometimes crimson or occasionally green silk. There is a tenacious oral tradition to the effect that the sarcophagus was a gift to Shah Ţahmāsp from the Mughal emperor Humāyūn at a time (c. 1545) when he was a temporary refugee from his own country and 11 living at the Safavid court by the grace and favour of the shah. Several factors conspire to cast some doubts on this tale. First, the issue of pietas. It is implausible that the grave of Shah Ismā‘īl would have lacked some suitable sarcophagus, especially in view of the veneration in which he was held. Although he died at a spot variously recorded by the chroniclers 12 13 as Mangutay and Sāyīn (Gadīkī) near Sarāb, he was not buried there. Instead, his corpse was transported to Ardabīl so that it could be laid to 14 rest in the shrine of Shaykh Ṣafī. This would not have been done without appropriate ceremony. Even though space was at a premium, room was found to squeeze his mausoleum into the most favoured location of all—right next to that of the shaykh himself. Given the tiny size of the shah’s mausoleum, however—it has the smallest surface area of any royal Islamic mausoleum in Iran—it was all the more important that its   F. Sarre, Denkmäler Persischer Baukunst (Berlin: Ernst Wasmuth, 1910), 41; Weaver I, 52; see n. 17 below. A.H. Morton, ‘The Ardabīl Shrine in the Reign of Shah Tahmāsp I’, Iran XII (1974) (= Morton I), 48, n. 67, says ‘the Indian story is to be regarded with scepticism’. 12   For the reports of the sources on the place of Shah Ismā‘īl’s death, see S. Abrahams, A Historiographical Study and Annotated Translation of Volume 2 of the Afzal al-Tavarikh by Fazli Khuzani al-Isfahani (unpublished Ph.D. thesis, University of Edinburgh, 1999), 222. 13   M.K. Yusuf-Jamali, The life and personality of Shah Ismail I (907–930/ 1489 (sic)–1524) (Esfahan: Amir Kabir Publication, 1998), 352. 14   Ibid., 352–3. 11

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

45

interior, cramped as it was, should nevertheless be as splendid as possible. Otherwise the gesture of giving it such a favoured location would have fallen flat. A magnificent sarcophagus was the obvious solution, and one can hardly doubt that it was constructed as soon as possible after his death. The idea that the mausoleum should have lacked a sarcophagus for two decades after the shah’s death—an implicit insult by Ţahmāsp to the memory of his father—does not recommend itself at all. If, then, there was already a sarcophagus in place in 952/1545 it would have been a supremely tactless move on the part of the homeless emperor to make an offer to his royal host to replace the sarcophagus ordered by Ţahmāsp with something better. Secondly—and this is a related point—the presence of so mary other medieval sarcophagi at the shrine for the shaykh and his descendants (at least seven sarcophagi of major art-historical interest), makes it all the more likely that the same method of honouring the illustrious dead was followed in the case of Shah Ismā‘īl. Indeed, there are significant points of contact between the sarcophagus of Shah Ismā‘īl and some of these other sarcophagi. They include the star-and-polygon network employed 15 for all four sides, the preference for inlay, the repeated use of a black and white colour scheme, and the deliberate contrast between various colours of wood. All this makes it still more probable that the sarcophagus of 16 Shah Ismā‘īl was constructed shortly after his death, if not before, in accordance with the well-established tradition of this shrine. Finally, Humāyūn was an exile, far from home, and in straitened political circumstances, with the accompanying threat of future financial hardship as an inescapable corollary. While it is known that he brought some of his treasures with him, notably jewels and manuscripts, a sarcophagus would 17 scarcely have formed part of his portable treasury. Nor is it likely that Humāyūn brought Indian woodworkers in his train. To have ordered the

  Weaver I, 136 and 138.   Though the unexpectedly early death of Ismā‘īl, when he was only 36 (Yusuf-Jamali, Life and personality, 1 and 352), argues against this. 17  Yet this is exactly what is suggested by J. Morier, Second journey through Persia Armenia and Asia Minor to Constantinople (London: Longman, 1818), 253 (which Morton [‘Ardabīl Shrine: I’, 48, n. 67, quoting S. Ray, Humayun in Persia (Calcutta: Royal Asiatic Society, 1948), 42] says is the earliest known reference to this tale) and by J.B. Fraser, Travels and adventures in the Persian provinces on the southern banks of the Caspian sea (London: Longman, Rees, Orme, Brown and Green, 1826), 297; the latter says of the sarcophagus that it is ‘said to have been brought from India’. 15 16

46

manufacture of the sarcophagus in Iran by Iranian artisans would of course have been a possibility, but it would remove the Indian element from the traditional story and reduce Humāyūn to the person who paid for the work rather than the monarch who brought some of India’s fabled splendour to Ardabīl. It would also have cost him dearly in ready cash—a commodity in short supply with him. This would be even more true if— 19 as some scholars note —Humāyūn also presented the sarcophagus of Shaykh Ṣafī. The fact that the cult of the shaykh had already flourished for so long in Ardabīl makes that latter story even more implausible. The most likely source of these legends is that, in the course of his attempts to 20 persuade Humāyūn to adopt Shi‘ite beliefs, Tahmāsp took him to Ardabīl and that this double royal visit entered the folklore of the shrine, to be embroidered in later years. It may be that Humāyūn’s gift of a Qur’an dated by colophon to 27 Jumada II 951/15 September 1544 marks this 21 occasion. That said, it is worth reflecting that the convergence of styles in the art of the three Muslim superpowers in the early 10th/16th century makes itself felt in the similarities between Iranian art, whether in luxury 22 Qur’ans or metalwork, and the production of Lahore and the Deccan. Thus there is a nugget of historical truth embedded in these legends. I turn now to the religious inscriptions. These are of various kinds; there is room in this paper only to indicate the broad parameters of the detailed study that they deserve. Moreover, so many of the inscriptions have gone— the piecemeal assembly of the sarcophagus would have made it easier to prise out individual elements of the design—that it is an impossible task to reconstitute the original epigraphic programme. Nevertheless, enough survives to give a fair idea of the flavour of these inscriptions. The tiny cartouche—located at the short end of the sarcophagus which is furthest from the entrance to the tomb chamber—with a signature identifying the artist as Maqṣūd ‘Alī is the only historical material in these inscriptions. Physically, the largest (and the most traditional) inscription is the riq’a text in high-relief wood carving close to the top of the sarcophagus (pl. 18

  Where the artists came from was perhaps not an issue.   Sarre, 83; Weaver I, 50. 20   Dr Sheila Canby, in a paper delivered at ‘Iran and the World in the Safavid Age’, University of London, September 2002, used the more appropriate word ‘extort’ in this connection. 21   Canby, ibid. This manuscript, which is in the British Library, is the earliest dated Mughal Qur’an. 22  These issues were discussed by Professor James Allan and Dr Sheila Canby in papers delivered at the conference mentioned in n. 20 above. 18 19

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

47

2). The Qur’anic passages are from at least two suras—50 and 69—and 23 are separated by hexagons that repeat holy names. Much of the long inscription is continuous; but it is a task for the future to record its entire text and comment in detail on its calligraphy. The band includes Sura 50: 1–2 and Sura 69: 70, but the longest consecutive text is Sura 50: 21– 25—’Now we have removed from thee thy covering, and piercing is thy 24 sight this day. And (unto the evil-doer) his comrade saith: This is that which I have ready (as testimony). (And it is said:) Do ye twain hurl to hell each rebel ingrate, Hinderer of good, transgressor, doubter, Who setteth up another god with Allāh.’ The terrifying references here to the Last Judgment and to eternal punishment are unmistakable. Pride of place, however, goes to the inscriptions executed in ivory (pl. 3). A chahar ‘Alī is found in one of the ivory eight-pointed stars which 25 frame the large ivory inscribed panels. Other ivory inscriptions consist of holy names repeated six times clockwise forming an interlace design in square Kufic: Allah, Muḥammad, ‘Alī. They are set in octagons of ivory against black wood and act as interstitial accents for the main inscriptions in ivory, which occupy elongated rectangular cartouches terminating at each end in the five points of an incomplete star. These are without doubt intended to be the main focus of interest in the sarcophagus; hence their prime setting, their high relief, and their bright white colour set off by the dark ground from which they emerge. These ivory openwork panels also bear Qur’anic inscriptions, and here too there is a measure of continuity. Nevertheless, even in multiple quotations from the same suras (namely 73 and 76) there are several gaps. 26 The texts include Suras 56: 17, 73: 1–4, 6–8 and 20 and 76:1, 3, 12–13, 27 15–17, 22, 25–6 and 28–9. One group refers to all-night prayer vigils: ‘O thou wrapped up in thy raiment! Keep vigil the night long, save a little—A half thereof, or abate a little thereof. Or add (a little) thereto—and chant 28 the Qur’an in measure’. Others focus on God’s creative power and on His showing man the way. But the bulk of them deal with the afterlife,   These inscriptions are positively curvaceous, full of dynamism and energy   This is a verse beloved of Sufis and therefore appropriate for use in a Sufi shrine. 25   In other cases these stars bear a lozenge with a superimposed saltire cross. 26  This plaque is in the Walters Art Gallery, Baltimore; I am grateful to Dr Barry Wood for this information and for a photograph of it. The last word of verse 12, hārirān, is here mis-spelled hārinān. 27   Qur’an 73: 1–4, 6–8 and 20. 28   Qur’an 73: 1–4. See M. Pickthall, The Meaning of The Glorious Koran. An Explanatory Translation (London: George Allen and Unwin, repr. 1957), 616. 23 24

1. Ardabil, mausoleum of Shah Ismā‘il I, sarcophagus: general view of two-tier layout.

48

2. Ardabil, mausoleum of Shah Ismā‘il I, sarcophagus: upper part of top tier with riq‘a inscription in wood on a ground of the same material.

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

49

3. Ardabil, mausoleum of Shah Ismā‘il I, sarcophagus: top tier showing part of an inscribed panel in ivory on a ground of the same material.

50

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

4. Ardabil, mausoleum of Shah Ismā‘il I, sarcophagus: top tier with part of an inscription in ivory against a dark ground.

51

5. Ardabil, sarcophagus dated 753/1352: detail of plinth.

52

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

53

promising ‘a Garden and silk attire; Reclining therein upon couches, they 29 will find there neither (heat of) a sun’, and elsewhere: ‘Lo! This is a reward 30 for you. Your endeavour [upon earth] hath found acceptance’; ‘There 31 wait on them immortal youths’; and the promise of ‘[Goblets] crystal clear, made of silver. They will determine the measure thereof. And they 32 will be given to drink there [of a cup]’. And later, from the same sura (v. 30), in the context of the statement that whoever wishes may choose a way to Allāh: ‘Yet ye will not, unless Allāh willeth. Lo! Allah is Knower, 33 Wise’. The shadow of the next verse lies over these words; verse 31, of which the opening words are quoted, speaks of the painful doom awaiting evildoers. Most of these quotations, then, allude to the blessed life after death which awaits Shah Ismā‘īl, but, as in the main inscription above, the threat of eternal punishment is implied. Thus the Qur’anic inscriptions in ivory broadcast the double message of joy for believers and misery for unbelievers in the hereafter. The reason for this fine balance remains an open question: were these inscriptions aimed at the living viewer by way of warning and encouragement, or did they refer to Shah Ismā‘īl himself, commending him to the mercy of Allah? Perhaps the ambiguity was intentional. At all events, the lengthy concentration on the joys and terrors of the afterlife go far beyond 34 the standard ‘Every soul shall taste of death’ so common in funerary 35 contexts. It is a great pity that the disappearance of so many of the cartouches means that it is no longer possible to reconstitute the entire epigraphic programme, though some of the missing panels can be traced 36 to the collections of the Iran Bastan Museum, Tehran and the Walters 37 Art Gallery, Baltimore. In the odd case, only a few words remain.

  Qur’an 76: 12–13; Pickthall, 624.   Qur’an 76: 22; Pickthall, 624. 31   Qur’an 56: 17; Pickthall, 561. 32   Qur’an 76: 16–17. See A. Yusuf Ali, translation and commentary, The meaning of the Glorious Qur’an II (Cairo: Dar al-Kitab al-Masri, n.d.), 1658. 33   Pickthall, 625. 34   Qur’an 21:35. 35   Among Iranian examples one may cite tomb towers at Rasgat and Maragha. See E.C. Dodd and S. Khairallah, The Image of the Word. A Study of Qur’anic Verses in Islamic Architecture (Beirut: The American University of Beirut, 1981), II, 77. 36  Accession numbers 20508–20512. For illustrations, see D. Jones and G. Michell, eds., The Arts of Islam (London: The Arts Council of Great Britain, 1976), 156, fig. 156. 37   Accession number 71.580; erroneously attributed to 9th/15th-century Egypt. It measures 20.1 by 4.1 cm. 29 30

54

It is now time to discuss the contemporary aesthetic. It is plain that in the early 10th/16th century there flourished an international Islamic princely style whose outstanding feature was a love of strong colour contrasts involving the use of different media and textures as well as extremely dense and small-scale vegetal ornament. It can be seen everywhere— from clothing to carpets, from fashion accessories to footwear. Naturally enough, there are differences between Ottoman, Safavid and Mughal 38 work in this vein, just as there are regional subsets in the style dubbed International Gothic; but the points of resemblance are more telling than the differences. Equally significant is the fact that some of these objects originated in one empire but were reworked in another, which is a further 39 indication of how widespread the taste for such work was. Unfortunately, the royal arts of the Safavid and Mughal empires have been scattered, and much has been lost. 40 Moreover, craftsmen migrated, voluntarily or otherwise, from one 41 empire to the other. Recent work by James Allan has revealed a sea-change in the field of metalwork which is of direct relevance to the study of the sarcophagus of Shah Ismā‘īl. The rich collections of the Topkapi Saray in Istanbul house an unrivalled collection of the finest Iranian metalwork dating from the opening years of the 16th century, and Allan argues convincingly that these objects—belts, armbands, weapons, shields, flasks, jugs—and many of the artists who produced them, were taken to Istanbul by Sultan Selīm as booty after he had defeated Shah Ismā‘īl at Chaldirān in 920/1514. Once installed in the Ottoman capital, these Persian artists of course influenced the course of royal Ottoman metalwork. Other examples of their work have found their way into European collections, 42 such as those in Stockholm and Vienna. The catastrophe of Chaldiran,   See C. Köseoğlu, The Topkapi Saray Museum. The Treasury, tr., expanded and ed. J.M. Rogers (London: Thames and Hudson, 1987), pl. 65 (covered box). 39   Ibid., pls. 65 and 77. 40   Persian painters in Hindustan are only the best-known example of this process at the time. 41   Presented in a paper delivered at the Safavid conference held in London in September 2002. 42   For Stockholm, see the wickerwork shield painted with such carpet motifs as a seated fox and a lion bringing down a bull, all against arabesque scrolls, and terminating in a steel boss inlaid with gold. The colour scheme of the wickerwork is indeed rich—white, yellow, grey and black against a red ground—and the contrast in texture between it and the steel boss is very effective. A dagger in the Museum für angewandte Kunst in Vienna (no. E 1 822) has a steel scabbard inlaid with gold, with a multi-foil cartouche termination at the hilt. I am grateful to Professor James Allan for sharing this material with me. 38

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

55

Allan suggests, destroyed Tabrīz as a centre of metalworking in general— armourers, specialists in zinc and in brass inlay, and artists in cognate fields. A particular characteristic of these metal objects, many of them made of 43 zinc, is the delight in using colourful studs or insets in precious stones. And it is this metalwork which, of all early Safavid art, is—despite the difference in medium—closest in style and aesthetic to the sarcophagus of Shah Ismā‘īl. That sarcophagus therefore has special value as one of the few survivals in Iran itself of an early Safavid aesthetic whose great masterpieces have in so many cases left the country. Thus the most representative survivals of this taste are Ottoman, since Istanbul was never sacked and successive generations of sultans preserved luxury objects of all kinds in the Topkapi Saray. Daggers and tankards, belts and book-covers, put over an unmistakable message of extravagant 44 wealth. The promiscuous mingling of textures, media, colours and ornament (especially vegetal) in such objects creates quite distinct visual effects and is central to this aesthetic. Familiar motifs were transformed by variegating the colours, textures and materials in which they were executed, and by the fashion in early Safavid times for extremely dense and detailed ornament. The aim was to create a busy surface. In media ranging from glazed architectural tilework to carpet design and Qur’anic 45 illumination, the repetition of identical motifs in different colours makes 46 an essentially simple design look far more complex than it actually is. The fact that so many specialists in all sorts of media worked together in the royal kitābkhāna—for which ‘design studio’ might be a free but appropriate translation—fostered the spread of the more popular designs across media, on various scales and in unexpected colour combinations. All this led quite naturally to an increased interdependence of the visual   See Köseoğlu and Rogers, 200–1 and pls. 74 (dish), 75 (jug), 76 (flask) and 77 (covered

43

jug).

  Ibid., pls. 32–3 (yataghan, scimitars and swords), 37 (parade helmet), 50 (hard-stone pot), 59 (penbox), 62 (footed cup), 78 and 81 (book-binding), 79 (reliquary-chest), 111 (box), 115 (belt) and 116 (armband). 45  There is no space in this paper to develop the connections between the sarcophagus of Shah Ismā‘īl and early Safavid Qur’ans; for the latter, see Dzul Haimi b. Muhammad Zayn, Qur’ans of the Safavid Period (unpublished Ph.D. thesis, University of Edinburgh, 1998). 46   C.G. Ellis, ‘The system of multiple levels’, in A. U. Pope and P. Ackerman, eds., A Survey of Persian Art from Prehistoric Times to the Present. XIV. New Studies 1938–1960. Proceedings, the IVth International Congress of Iranian Art and Archaeology, Part A. April 24– May 3, 1960 (Tehran, London, New York and Tokyo: Asia Institute of Pahlavi University, 1967), 3172–83. 44

56

arts. The Topkapi Saray Museum holds many examples of early Safavid luxury art which permit an assessment of how this fashion for mixing 47 textures and colours was interpreted in Iran. This delight in mixed textures, and above all in colour, is equally characteristic of the sarcophagus of Shah Ismā‘īl (pl. 4). Although made of wood, it contrives to hide that simple fact at almost every turn. Why? Because the craftsmen wanted to make it glisten with many colours—and wood is not a natural medium for lasting colour effects. The sarcophagus 48 uses various tones of red and green silk as a backing for openwork white ivory interlace. The effect is very close to that of Safavid steel plaques, where the bold, fat riq’a inscriptions stand out against spacious but thin 49 scrolling arabesques. In such work the background is sometimes gilded, which creates a luscious contrast with the silver colour of the inscription, and even the outline of the plaque itself can often be very close to that of 50 the ivory plaques on the sarcophagus. The colour contrast of these ivory plaques is constantly variegated by more ivory or bone stained black and green, often applied in tiny triangles in the marquetry technique. Longer strips are not glued onto the backing but sewn onto the underlying textile with silvered metal thread. Some of the inscriptions are of wood in high relief, but even here the letters themselves are stained black. In terms of colour alone, then, the craftsman disposes of white, black, yellow-brown, 51 dark brown, silver, green, crimson and russet. The striking use of white against a dark ground is a basic principle of the entire design. In particular, the interstices of the geometric strapwork bring the whole composition to life, for they are ornamented with grace notes applied almost parsimoniously at key points in black and white Köseoglu and Rogers, pls. 48 (gold-inlaid nephrite pot), 74 (dish), 75 (jug), 76 (flask), 77 (covered jug). 48   This use of textiles as a foil for other materials occurs frequently in early Safavid metalwork. See ibid., 206 and pls. 115 [belt] and 116 [armband], though it is also known in 10th/16th-century Ottoman art, for example bookbinding. See ibid., pl. 78. 49   Folsach, 328, pl. 526. 50   Sotheby’s. Islamic Works of Art, Carpets and Textiles, 15-16 October 1985, lot 218; Sotheby’s. Islamic Works of Art, Carpets and Textiles, 16 April 1986, lots 181-3; Sotheby’s. Arts of the Islamic World, 16 October 2002, 68–9, lot 62, with ample references to comparative material (the discussion mentions the shrine of Shah Ṭahmāsp, which probably refers to the Ardabīl shrine, in which Shah Ṭahmāsp took such an interest). See, in general, J.W. Allan and B. Gilmour, Persian Steel. The Tanavoli Collection. Oxford Studies in Islamic Art XV (Oxford: Oxford University Press, 2000), 264–81, 294–302. 51   The mother-of-pearl used in some earlier woodwork at Ardabīl has, however, dropped out of fashion. It became characteristic of Ottoman woodwork. See, for example, Sotheby’s. Islamic Works of Art, Carpets and Textiles. 16–17 April 1985, lot 137. 47 

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

57

marquetry work. Similar colour effects can be found in contemporary 52 53 Anatolian carpets. Tiny spots of green, presumably stained bone rather 54 than the turquoise mentioned in several sources, add a further touch of luxury. The borders are marked by further clusters of marquetry in a sprightly dancing rhythm. All this means that the immediate visual impact of the sarcophagus is not that of a large and weighty block of carved dark brown wood, which after all is the basic impression given by most earlier sarcophagi, but of a light and fragile object, with a surface of shimmering white on which there floats a green mist. Its basic shape, then, like its skeletal design framework, are both thoroughly traditional; but these elements are transformed by an infusion of striking, indeed ethereal, colour harmonies of a kind not hitherto encountered in surviving Iranian woodwork. Next, the question of innovation. This is bound up with the issue of colour. It is too early to define the role which polychromy played in medieval Iranian woodwork. But it is likely that in Iran, as elsewhere, wood was painted; portions of a probably Saljuq table recently acquired by the David Collection display several bright colours, among them an intense 55 cinnabar red. The sarcophagus of Mīr Qavvām al-Dīn al-Mar’ashī near Amul, dated 781/1379, has the letters of its inscription gilded and set 56 against a dark blue ground. A 9th/15th-century door from Kokand in the Metropolitan Museum of Art had a blue background while the main pattern was prinked out in red, green, brown and gold. But only traces 57 of these colours remain. Denike’s work on the late 8th/14th-century door to the mosque of the Shāh-i Zinda indicates that its polychromy was 58 constantly renewed. So it would have been obvious enough that paint did 52   Certain types of late 9th/15th-century Ushak carpets survive in which lobed lozenges of medallions in colours strikingly different from their background figure largely. See R. Pinner, ‘Oriental Rug Design. Multiple and substrate designs in early Anatolian and cast Mediterranean carpets’, Hali 42 (1988), pls. 26–7. 53   I am indebted to Dr Kjeld von Folsach for the suggestion that the material might well be camel-bone. 54  For example, Morier, 253 (he also mentions tortoise-shell) and Dibaj, tr. Emamy, 113. I am told by Robert Skelton that turquoise, as a hardstone, would not lend itself to work on this tiny scale; and it is not used by workers in khatamkari, which is the technique used here. 55   Folsach, pl. 427. 56   L. Bronstein, ‘Decorative woodwork of the Islamic period’, Survey, 2622. 57   Idem, 2623; M.S. Dimand, A Handbook of Muhammadan Art (New York: Hartsdale House, 1947), 121. 58  B. Denike, ‘Quelques monuments de bois sculpté au Turkestan occidental’, Ars Islamica II (1935), 83 and fig. 12.

58

not provide lasting colour, and thus medieval Islamic craftsmen developed other ways to create an enduring colour contrast. One was inlay; another was marquetry. In Islamic times this technique, already long familiar in the Mediterranean world and the ancient Near East, is recorded from the 59 8th century in Islamic Egypt. It involves thin strips of material—such as wood, ivory, or bone—being glued together in prefabricated patterns, and sliced horizontally. This is clearly related to millefiori technique. The resulting tesserae are then glued onto a wooden backing. Desks and doors from Andalusia show that the Western Islamic emphasis on inlay had shifted by the 15th century to the use of long thin strips of white and black to create the main design and a much greater reliance on marquetry 60 work for the infill. So much for the pan-Islamic picture. It is now time to look at earlier Iranian woodwork, first at Ardabīl itself—since this would have been the obvious source of inspiration—and then in a more general sense. The earlier funerary woodwork at Ardabīl proves conclusively that the sarcophagus of Shah Ismā‘īl belongs to a well-established tradition that flourished at Ardabīl itself, even though the name of at least one master shows that woodworkers from other areas were active at the shrine (e.g. a wooden headboard signed ‘amal Ustād ‘Uthmān ibn Aḥmad al-Maraghī). Another specimen, in which the plinth dwarfs the sarcophagus proper, bears the remains of an inscription. …Muḥammad, nawwara Allāh qabrhuma. Fī shahr Muḥarram san’at thalatha wa khamsīn wa sab’a mi’at (‘… Muḥammad, may God illumine the grave of the two of them. In the month of Muḥarram of the year 753’ [18 February–18 March, 1352] (see pl. 5). This has marquetry work in gold and black for the centrepieces and stars of the polygonal stellar networks which fill the panels of the plinth. Similar marquetry forms the decoration of the small sarcophagus itself. A third sarcophagus dated 788/1386–7, already mentioned briefly above, is somewhat more limited in its repertoire, confining its chromatic scale to light and dark brown and its carved ornament to dense interlace within horizontal, stellar or polygonal panels. Since the sarcophagi at the shrine include those of Shaykh Ṣadr al-Dīn Mūsā—the son of Shaykh Ṣafī himself—and of Mūsā’s grandson Shaykh Ibrahīm (d. 851/1447) and   For a colour illustration, see R. Hillenbrand, Islamic Art and Architecture (London: Thames and Hudson, 1999), 57. See also n. 7 above. 60   J.D. Dodds, ed., Al-Andalus. The Art of Islamic Spain (New York: The Metropolitan Museum of Art, 1992), 268–9, cat. 53 and 372–3, cat. 118. 59

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

59

great-grandson Shaykh (or Sultan) Ḥaydar (d. 893/1488), as well as those of Shaykh Ṣafī and Shah Ismā‘īl, the time-span of this woodwork covers the full two centuries of the order’s rise to power. It represents easily the greatest challenge for future research by specialists in this field of Iranian art. What of the woodwork of Iran in general in the immediately pre-Safavid period? The top-quality 9th/15th-century woodwork which survives can be divided into three major categories. The first is conservative to a fault. It favours long continuous inscription bands, and a framework of bold polygons and stellar motifs. The emphasis lies squarely on the principal 62 design, and infill is demoted to a minor role. The second category could scarcely be more different. It is known in both doors and boxes. Here the emphasis is on extremely small-scale, almost filigree work. The panels, which tend to be vertically oriented, are long and narrow. They feature a single continuous design rather than the repetition of a single motif. The artists delight in the interplay of different levels of carving, and parallels 63 with contemporary tilework readily suggest themselves. The third type is the one most closely related to the sarcophagus of Shah Ismā‘īl I. Its distinguishing characteristics are its strong colour sense and its use of marquetry. Very little of it survives. The prime example is furnished by 64 a pair of leaves for a door at the Gūr-i Amīr. Another instance can be 65 found on a sandalwood box made for Ulūgh Beg. A foretaste of this technique in Iran can be recognised in the marquetry using several kinds of wood, all of different tones of brown, on a Qur’an stand of 761/1360, though in that object the colour potential of the marquetry technique is 66 exploited only timidly. In the later 9th/15th century Josafa Barbaro saw 61

61   M.E. Weaver, ‘Ardabīl iii. Monuments of Ardabīl’, Encyclopaedia Iranica II.4 (London: Routledge and Kegan Paul, Ltd., 1986), 363. 62  Survey, pls. 1465C and 1472; for a colour plate of the latter sarcophagus, see T. Lentz and G. Lowry, Timur and the Princely Vision. Timurid Art in the Fifteenth Century (Washington, D.C.: Smithsonian Institution, 1989), 207. For colour illustrations of such woodwork see also Sotheby’s, Islamic Works of Art, Carpets and Textiles. 16 April 1986, lot 110; Sotheby’s. Islamic and Indian Art. Oriental Manuscripts and Miniatures. 29–30 April 1992, lot 89; Sotheby’s. Islamic and Indian Art, 19 October 1995, lot 75; Sotheby’s. Islamic and Indian Art. 25 April 1996, lot 67. 63   Survey, pls. 1467–9. Note that the door depicted on pl. 1468 has a small amount of ivory inlay. See Lentz and Lowry, figs. 68–9. An identical style was favoured for late Timurid tombstones. See ibid., 209, fig. 70. 64   Survey, pl. 1470. 65   Lentz and Lowry, 339 and cat. 49; for colour plates, see 142 and 207. 66   Ibid., 330 and cat. 9; for a colour plate, see 47.

60

in Tabrīz a sandalwood door with intarsia of gold wire and pearls, which clearly had a richer palette than the examples cited so far. And this is where the innovative quality of our sarcophagus may be recognised. For it is assuredly the outstanding, and also the earliest surviving, masterpiece of marquetry in Iranian woodwork. Admittedly, the evidence is badly skewed because almost all surviving work was made for a religious context—mosque or shrine doors or sarcophagi—whereas the many different kinds of woodwork depicted in pre-Safavid Iranian book painting tell quite another story. Unfortunately, all this secular woodwork made for the court has vanished. But the painted evidence suggests that the idiom employed for certain types of wood fittings in a courtly context—screens, doors, window-grilles, footstools, thrones—was much more daring in its use of colour than what survives in the religious sphere; that it made lavish use of marquetry; and that its overall designs favoured 68 small-scale, rather finicky patterns. The key question here is simple—can these painted images be taken as a reliable index of what contemporary woodwork of luxury type looked like? This is best answered by attempting to define their aims. The artists painted this woodwork as part of the environment for the figural subjects which were, for them, the main focus of pictorial interest. For them, therefore, it was an incidental element of the composition, though the obsession with the accurate depiction of detail which so characterised late Timurid painting ensured that these painters would pay close attention to the rendering of even minor elements of the composition. That said, a note of caution should be sounded here. The artists who painted this woodwork were of course not striving after naturalism in any Western sense, and their handling of architecture shows well enough that they sometimes chose colours which are not to be found in any surviving monuments and probably had no basis in reality. Equally, then, the woodwork which they depict may contain some element of fantasy. Nevertheless, the miniature painters consistently make a distinction between the kind of woodwork used for screens and for 67

  Bronstein, Survey, 2620, n. 2.   F. Suleimanova, Miniatures Illuminations of Nisami’s ‘Hamsah’ (sic) (Tashkent: Fan Publishers, 1985), pls. 8, 11 and 82; Lentz and Lowry, 220, cat. 147; 260, 286 and 294, cat. 146; 263, fig. 93; 267, cat. 147; 279, cat. 153; 281, cat. 155; A. Soudavar, Art of the Persian Courts. Selections from the Art and History Trust Collection (New York: Rizzoli, 1992), 99, 104–5 and 176. These examples could easily be multiplied, and their cumulative testimony is overwhelming: high-quality woodworking was indeed transformed in the late Timurid period. 67 68

THE SARCOPHAGUS OF SHAH ISMĀ‘ĪL I AT ARDABĪL

61

window grilles—which is of openwork type using fairly simple polygonal geometric designs, exactly like the surviving woodwork which served these same purposes—and the much more technically complex kind used for doors, thrones, footstools and panels below balconies. This suggests that the painters were recording a distinction which applied to objects in actual use at the time, even if the individual patterns may not have been based on fact. Moreover, not only are the patterns shown precisely those best known in marquetry work, but their colours—white, black, green—are again the favoured ones in that technique. All this tends to suggest that, in the case of luxury woodwork at all events, the painters were reproducing objects before them with a high degree of accuracy. The decision to execute a substantial amount of the sarcophagus in the marquetry technique made little difference to the overall design, since this followed the type of angular polygonal interlace which had for centuries been well-nigh canonical for sarcophagi. Doors, on the other hand, featured a much wider range of design types, notably those of curvilinear type. None of the marquetry doors in Timurid painting employ curvilinear motifs, which suggests that marquetry and curved designs were mutually exclusive. Thus the sarcophagus of Ismā‘īl combines conservative design with innovative technique. This is in line with the general contemporary preference for experiment in colour rather than in design. Time does not permit more than a brief allusion to the sarcophagus of Shaykh Ṣafī, which almost certainly dates from soon after his death on 69 12 Muḥarram 735/12 September 1334. Here too the design is strongly conservative in its emphasis on polygonal networks; here too there is innovation, as in its numerous and varied metalwork fittings and the silver panel with cloisonné enamel decoration which records that Shaykh Ṣadr 70 al-Dīn, Ṣafī’s son, built the tomb; and here too marquetry plays a part, though the tonality is reduced to silver and black (there is no green, though there are mother-of-pearl effects), and the marquetry work has to compete with densely drilled foliate carving as infill ornament. Lastly, let me return to the use of marquetry, for it raises a wider question, namely the source of the idea of using it, and this has a bearing on my final point. To my knowledge, the earliest known surviving specimen from 69   For this date, see Morton, ‘Ardabīl Shrine’, 47. Ṣadr al-Din himself died, aged nearly ninety, in 794/1391–2. See R.M. Savory, ‘Ṣadr al-Dīn Ardabīlī’, EI2, VIII: 753. So, in theory the sarcophagus could date from any time between 1334 and 1392; but a date not long after Ṣafī’s death is the most likely one. 70   Morton, ‘Ardabīl Shrine’, 47–8 and pl. IIa.

62

the Iranian world is the box of Ulūgh Beg, and this eastern connection is corroborated by a whole host of doors and thrones depicted in late Herati painting. Earlier Herati painting, incidentally, suggests that the fashion for 71 doors was plainer earlier in the century. If, then, this was an eastern Iranian fashion, and an up-to-date one at that, what was it doing at Ardabīl? This contrast is made still more marked by the preference in the eastern Iranian 72 world for small-scale curvilinear techniques in woodcarving, whereas the sarcophagus of Shah Ismā‘īl opts for the polygonal networks long standard in western Iran and Anatolia. Moreover, the comparative material in book painting suggests a Khurasani preference for all-over marquetry; and our sarcophagus falls far short of that. Where do these remarks leave us? The obvious parallel for this abrupt juxtaposition of east and west is the mixture of north-western and northeastern strands, of Turcoman and Timurid, in early Safavid painting. Recent research has shown that this commingling is not quite as easily explained as Cary Welch has suggested, but the existence of those two strands in 73 early Safavid painting, and what may be termed their shotgun marriage, 74 is indisputable. Whether the Maqṣūd ‘Alī who signed our sarcophagus was a Herati is not known; but what matters is that the style of choice for a royal sarcophagus at this time was, it seems, a combination of eastern and western Iranian modes. And that has unmistakable political resonances. For it evokes, with extraordinary aptness—just as both style and content of Ṭahmāsp’s Shāhnāmeh do—that decade and a half from 931/1524 when the young ShahṬahmāsp, himself raised in the east, was valiantly trying to maintain the territorial integrity of his eastern and western lands.

71 A.T. Adamova, ‘The Hermitage Museum Manuscript of Nizami’s Khamseh, dated 835/1431’, Islamic Art V (2001), figs. 17, 25 and 37 (though one should note fig. 39, which is much more elaborate, perhaps because it represents a foreign environment); B. Gray, Persian Painting (Geneva: Skira, 1961), 86–7. 72  This trend was already marked in the 14th century, as shown by the sarcophagus of Sayf al-Dīn Bakharzī; see Denike, figs 9–11. 73  A theory aired at length in S.C. Welch and M.B. Dickson, The Houghton Shjahnameh (Harvard University Press, Cambridge, Mass., and London, 1981). 74  On the eastern face: ‘amal-i Ustād Maqṣūd ‘Alī (Weaver I, 113, quoting a translation made by M. Emamy of the discussion of Ardabīl in I. Dibāj, Rahnāma-yi Āthār-i Tārīkhi-yi Āzarbāijān-i Sharqi (Tabrīz, 1334/1995).

III Problems in Islamic Pottery

T

he Keir collection, which has been published by Faber and Faber in a series of sumptuous volumes, contains probably the most representative body of medieval Islamic objects still in private hands. In certain areas, such as the painting and lustre pottery of the Fatimid period, it has only very few rivals, even among the great museums of the world. Its detailed publication, in fully illustrated catalogues, is therefore an event of major importance. These volumes will ensure that the collection becomes better known than the holdings of many a museum. Students of Islamic art will thus be compensated to some extent for the fact that the collection is not on public display. The published catalogues of the collection cover painting and the arts of the book; metalwork; carpets and textiles; and ceramics. It is this last volume, written by Professor Ernst J. Grube, which will be discussed in the present article. Its title – “Islamic Pottery of the Eighth to the Fifteenth Century in the Keir Collection” – is significant. In the pottery, as in the metalwork, of this collection, work from the Ottoman, Mughal, Safavid and later periods is unrepresented – a gap which is more curious in the case of pottery than of metalwork, for it means that Iznik ware, which for some tastes is the apogee of Islamic pottery, is excluded. Of the paintings in the Keir collection, on the other hand, over half belong to these later periods. These weightings of the material in the Keir collection have to be borne in mind in assessing its utility as a comprehensive guide to Islamic art in general. Magnificent as it is, the collection is uneven, apparently by design rather than accident; but as its owner Mr de Unger states, a collector – unlike a museum curator – need consult only his own

64

preferences in his purchases. He himself feels that the private collector belongs to a dying breed, but he may at least console himself with the reflection that his stature as a collector should be so fittingly expressed in these magnificent volumes. Only in this way can the essentially ephemeral nature of a private collection receive a lasting memorial. Mr de Unger ranks with such collectors as Martin, Goloubew, Eumorphopoulos and Chester Beatty and, as in their case, the catalogues of his collection will be of permanent value to scholarship. But this book is very much more than a catalogue. It is the most valuable contribution to studies of Islamic pottery which has appeared in the last twenty years. Thanks to the range of the Keir collection, the book has a well-nigh encyclopaedic scope, and its documentation is prodigal. The 264 pieces in the collection are all illustrated, 31 of them in fine colour plates, and basic catalogue information is given, including measurements, date, provenance, repairs and bibliographical references where necessary. But it is above all the way that the material has been presented which will stimulate further discussion and will make this such a useful handbook for students of Islamic pottery. Time and again, the author has lucidly outlined the current state of scholarship on a given group of wares and, by indicating new approaches, has provided the basis for a deeper understanding of them. This is especially true of the later period. New insights, and where relevant, new dating suggestions, are offered for Egyptian pottery and tilework in the Mamluk period, later medieval celadon, Tīmūrid pottery and blue and white ware. In the earlier period Professor Grube has carefully distinguished the significant from the trivial or conventional in the themes of ‘Abbāsid, Fātimid and Saljūq lustreware. Certain widely accepted datings, notably the attribution of Fāṭimid figural lustre to the 12th century, are challenged with cogent argument. Perhaps it is even more important that he has managed to break out of that straitjacket of provenance which has for so long paralysed discussions of Saljūq pottery. Drawing freely, as he candidly admits, on the conclusions of fieldwork carried out by the late Andrew Williamson, he has shown that the old concepts which assigned “Rayy lustre” or “Kāshān lustre” to those particular cities must now be discarded. The discovery that most types of Saljūq ware were made all over Iran opens up all sorts of new perspectives, and Professor Grube’s book is the first large-scale work to explore the implications of that finding. Finally, this discussion is underpinned by a triple bibliographical apparatus – general and classified bibliography plus footnotes – that is

PROBLEMS IN ISLAMIC POTTERY

65

simply beyond praise. The classified bibliography comprises twenty headings in which publications on Islamic pottery are organised according to area (e.g. “Seljuq pottery in Anatolia”), type of pottery (e.g. “ceramic figurines”), technique (e.g. “early unglazed pottery”) or dynasty (e.g. “’Abbasid pottery”). Each item in these detailed bibliographies is crossreferenced to the concise general bibliography. Thus for the first time it will be possible for anyone interested in a given branch of Islamic pottery to consult an up-to-date, exhaustive bibliography for that particular group. For this immensely time-saving aid alone Professor Grube deserves our gratitude. The very range of the material presented forbids a detailed review of the many issues tackled in the book. It seems a more sensible use of space to deal in some depth with only a few categories of this material, but at the same time to give the discussion wider relevance by extending it to related work in other periods. This article will therefore deal with some of the problems raised by: the earliest Islamic pottery; epigraphic decoration; the subject matter of certain lustre wares; architectural tilework; and dated ceramics. In every case the discussion is based on the material in the Keir collection; the remarks are intended to supplement Professor Grube’s findings. I The comprehensive range of the pottery in the Keir collection is illustrated at the very beginning of the catalogue by the first five pieces of the collection, for which an 8th-century date is suggested. Islamic pottery of this date is rare, so much so that no criteria for distinguishing Umayyad from early ‘Abbāsid wares have yet won general acceptance. The provenance of such pottery is apt to be equally nebulous. The excavations of Umayyad sites in Syria have produced notably little in the way of fine pottery, even though the buildings uncovered were of princely character; and in this respect the excavations in ‘Irāq, at Kūfa, Wāsiṭ and Ukhaiḍir, have been no more informative. At Sīrāf steatite bowls have been found which provide useful parallels for the decoration of no. 3 in this collection, but Dr Whitehouse 1 assigns them to the ninth century. It seems right, therefore, to date such pieces only approximately. A far more teasing problem is to discover why the Arab rulers of the eighth century, whose residences exuded wealth and ostentation, did not cultivate a taste for fine pottery. Much the same 1 

“Excavations at Sīrāf; First Interim Report”, Iran VI (1968), 20.

66

is true of metalwork, except that there the problem is even more puzzling since both Byzantium and Sasanian Iran, the two great external sources of 2 Umayyad art, produced quantities of fine silverware. As in fine pottery, there is little agreement about the date and provenance of the group of ewers which are apparently the earliest pieces of Islamic metalwork. Their material is nothing more precious than bronze. The dearth of fine glazed pottery in this early period could be explained by the fact that there was no flourishing tradition in this medium for the Umayyads to inherit; but in metalwork the reverse applied. The surviving material imposes the rather unsatisfactory conclusion that the Umayyad princes never developed a taste for luxurious tableware or display pieces in either ceramic or metal. The decoration of these early pieces permits more positive conclusions. Their glazes are all monochrome and in that respect alone offer a strong contrast to virtually all the other pottery in the collection. Where the basic element of colour is so timidly employed, the potter is clearly not yet alive to the potential of glaze. Purely on this ground the 8th-century date suggested for these pieces seems reasonable. Their decorative motifs invite a more precise dating because in several cases they have surprisingly close analogies to the carved ornament in later Umayyad architecture. The medallions on the hemispherical bowl (no. 2) recall details from the 3 4 Mshattā façade and from the lintel at Qaṣr al-Ṭūba. Similarly, the swaglike floral band on the pitcher (no. 5) echoes border designs at the base of the 5 Mshattā façade. A date of c. 750 for such pieces seems quite justified. More generally, the rosette is as much a leitmotif on these pieces as it is at Khirbat al-Mafjar and Mshattā. Its very popularity indicates that, in this context at any rate, it is a neutral motif. It is possible to go even further and to suggest that since it figures so prominently on rather modest pottery, the rosette would not have been particularly suitable as a symbol of majesty. The propaganda element in royal iconography demands that symbolic references to the ruler should be as clear and precise as possible. Ambiguity is counter-productive. By this reckoning, the rosette would be too equivocal an image to be useful, for however complex the associations that might be read into it, its past history as a bland decorative cliché 2  It might be argued that quantities of early Islamic objects in precious metals were made, but that one and all were subsequently melted down. But such a theory would have to account for the numerous surviving pieces of Byzantine and Sasanian silver. 3  K.A.C. Creswell, Early Muslim Architecture. Umayyads. A.D. 622-750 I/2 (Oxford, 1969), pl. 122. 4  Ibid., pl. 138a. 5   Ibid., figs 655a, d, f and 656 g, h, i, k.

PROBLEMS IN ISLAMIC POTTERY

67

would militate against its use in a precise symbolic sense. At most it might operate on a secondary level behind the major iconographic motifs – a 6 role which it might perhaps have served at Khirbat al Mafjar. No. 1 in the collection might be an illustration of this same process. It is a pot depicting a lion round whose neck floats the patev, the royal ribbon, and whose body is sprinkled with small rosettes. The very fact that these rosettes also 7 variegate the background indicates how unspecific the motif could be. The decoration on no. 4, a bowl, invites quite different speculation. The 8 main theme is purely Chinese – the ju-i head which connotes prosperity. No. 2 exhibits the pearl roundel, another Far Eastern theme. Such motifs show that Chinese elements had infiltrated Islamic pottery even before the 9th century, a period which witnessed the first major impact of Chinese modes on Islamic pottery. It is quite likely that the motifs on these early ceramics were borrowed from textiles, where they had become naturalised in the Iranian repertory under the Sasanians, but a direct source in imported Chinese pottery need not be excluded. II The Keir collection is rich in virtually all types of pottery where epigraphic decoration dominates. This type of ceramic ornament seems to have been an independent creation of Islamic potters; significantly enough, its genesis coincides with the burgeoning interest in decorative varieties of so-called “Kufic” in the early ‘Abbāsid period. The series begins with the ninth-century blue and white wares, in which the inscriptions spread like an ink-stain in blotting paper. The qualities of this epigraphic ware have always tended to be somewhat neglected in favour of the more spectacular epigraphic wares of Nīshāpūr and Samarqand. But they are fascinating in their own right, for they owe much to a source that later Islamic wares did not exploit – coins. No. 7 (fig. 1) preserves something of the classic 9 simplicity of later Umayyad and early ‘Abbāsid coinage. The parallel is not limited to the type of script, but extends to the spacing, the use of 6  The motif is given rather more emphasis in the interpretation proposed by R. Ettinghausen, From Byzantium to Sasanian Iran and the Islamic World (Leiden, 1972), 36–9. 7  Cf. no.18 in the collection. 8   For an analysis of this theme in Islamic art, see A.R. al-Gailānī, Islamic Art and the Role of China (unpublished Ph.D. thesis, Edinburgh, 1973). 9   One may perhaps speculate that the reason why the artist has written li ṣāḥibihā, which is a mistake, instead of li ṣāḥibihi is that the otiose alif plays a vital role in ensuring a harmonious visual rhythm for the inscription. Cf. The Arts of Islam, eds D. Jones and G. Michell (London, 1976), no. 255, for the same phenomenon.

68

three lines and even to the carefully placed ornaments. The shape of the dish might itself have encouraged the artist to exploit this parallel. The inscriptions here, like those on coins, were meant to be read. Moreover, in many cases the inscriptions were the only decoration on these pieces, which ipso facto seems to indicate that epigraphy was regarded as significant, not casual, decoration. Many of them are signed, which may mean that they were more highly prized than most other contemporary wares. When the artist’s signature is the only decoration of a piece, one may deduce that the plain white glaze itself was an achievement on which the potter prided himself. Professor Grube notes that the inscribed pieces are all small and of one type, which could also imply that they had a limited market and function. But these signatures normally give only the ism of the artist, and sometimes his kunya too, but no nisba. The implication is that the ware was for local consumption and that those who bought it would at least know of the potter. In such a context the addition of a nisba would have been unnecessary and pretentious. In the more public medium of architecture it was of course common for the artist to give his nisba. Whatever the implications of the signatures, these pieces are of prime importance on several counts. They inaugurate a long and proud tradition of epigraphic ornament on Islamic pottery, a tradition for which there was no useful precedent in Byzantine, Sasanian or even Chinese ceramics. Yet it is difficult not to see a connection with Chinese calligraphy in the 10 freedom and speed of the writing, its deliberately blurred quality and its frankly decorative bias. Such an instinctive affinity matches the deliberate imitation of Chinese porcelain in the glaze of this ware. This is indeed a remarkably mature style given its revolutionary qualities. In the chaotic horror vacui of early ‘Abbāsid non-figural lustre ware, it is symptomatic that even writing is reduced to an indecipherable scrawl. Yet such pieces are remarkably close in time to the aristocrat of all epigraphic ware in Islam: the Sāmānid pottery of Khurāsān and Transoxiana. These Sāmānid dishes, paradoxically enough, are fascinating precisely because 11 of their appreciation of void space as a positive factor of the design. In them the oppressive surfeit of ornament just mentioned, which is such a recurrent feature of Islamic art in general, is set aside. As with the blue 10   Perhaps the artist had to write fast in order to prevent the writing from spreading too much on the wet surface and thus becoming totally illegible. 11   This extends also to such non-epigraphic pieces as. no. 59, where two medallions crammed with ornament are placed at opposite ends of a plain white dish but are linked across its centre by a thin dotted “life-line”; it is a piece to puzzle over.

PROBLEMS IN ISLAMIC POTTERY

69

and white style just discussed, the fashion for this type was short-lived. But this is in no sense strange. One has only to leaf through this catalogue to be struck by the speed with which decorative schemes changed in Islamic pottery. By contrast, techniques changed much more slowly. Even so, the conservatism of these pieces is surprising. Very few shapes are employed; the potters apparently had an instinctive understanding of the shapes which best suited this decoration. There seem, for instance, to be few ewers decorated in this style. The colours are reduced to the minimum – usually just dark brown on white. Clearly, in such cases the artist voluntarily imposed maximum restrictions on himself. There is no denying that the style derives much of its peculiar éclat from this self-discipline; it is this very restraint which lends such intensity to the occasional non-epigraphic 12 motif, like the beauty spot at the centre of no. 57 (fig. 2). Indeed, the abstraction of the script frequently extends to the ornament, which exploits the same dynamic, capricious asymmetry. No satisfactory source for the calligraphy on these pieces has yet been suggested. The nearest parallel is perhaps provided by Qur’āns executed in the so-called “Carmathian” Kufic, but these Qur’āns seem on the whole to post-date the ceramics under discussion. Nevertheless, in the mannered elongation of the uprights and the correspondingly crabbed style of letters with a horizontal bias, the two styles do seem akin. Certainly the inspiration of this Sāmānid calligraphy is in pen-made writing. Another source which might be canvassed is the Uighur alphabet. Geographically, of course, the Uighur empire bordered on that of the Sāmānids, and the silk route from China to the west, along which travelled porcelain and other luxuries for the Caliphal court, lay through Uighur territory. If the inscriptions on Sāmānid ware consciously evoked this alien script, it would not have been the last time that the Arabic alphabet was distorted almost beyond recognition to express an alien style of writing. The type of square Kufic developed from the early 12th century in the Iranian world owed much to China and culminated in the deliberately exotic seal script of the 13 Īlkhānid period. Whatever the source of this script, its sophistication is undeniable. It mirrors the taste of a highly literate, even precious, society. No more complex ductus was ever used on Islamic pottery. In the stark severity of   The same sensitivity to colour in otherwise monochrome ware manifests itself, again in Iran, in certain wares of the 12th century (nos. 107-9). 13   Cf. E. Herzfeld, Matériaux pour un corpus inscriptionum arabicarum. Deuxième partie: Syrie du Nord. Inscriptions et monuments d’Alep II (Cairo, 1954), pl. CLV a-c. 12

70

these inscriptions one can recognise (as did Arthur Lane) beauty of a high intellectual order. The question arises whether the potters designed these inscriptions themselves or whether they were the work of professional scribes or calligraphers. Such men certainly designed architectural inscriptions in the medieval period. To have accommodated a lengthy inscription in this style within the concentric circular bands demanded by the shape of the dish (e.g. no. 56) would certainly have taxed the resources of any professional calligrapher. The exigencies imposed by the shape of the dishes help to explain why no adequate parallels can be found in other media for this script; for no other medium would call for such unpredictable variations in the length of the ligatures – bold swoops alternating with huddled groups of letters. This was a style of writing 14 developed for pottery alone. The social context and function of this pottery has yet to be established precisely. The prominence given to epigraphy proves of course that it was made for the educated. But a particular connection with the Sāmānid court is far from certain. As Professor Grube perceptively remarks, very little Islamic pottery bears the name of a ruler, a situation largely reversed in Islamic metalwork. Thus very little can be made of the fact that none of these pieces bears the name of a patron or ruler. But the inscriptions themselves do not contain protocols or pompous benedictions: their subject matter comprises proverbs, maxims and quotations from the Qur’ān. The commonest single inscription is in baraka li-ṣāḥibihi (“blessing to its owner”); it would be hard to imagine a more neutral formula. Its frequency strongly suggests that this ware was made in quantity for a given class of patron. Sometimes the inscriptions break off short; this has been interpreted as a deliberately humorous device, but it could also be intended as a test of one’s religious and literary culture – almost in the spirit of a game. Thus the Qur’ānic inscription on no. 57 reads “When a creature is overcome by misfortune”, which means nothing unless the quotation 15 is continued: “... he calls upon his Lord ..” (Sūra 39,8). According to this interpretation, a single letter or a fragmentary word would then be a

Much work remains to be done in analysing the various sub-divisions of the most popular script used on these wares; and there are several other quite distinct types of script which appear more rarely and have been treated in even less detail (see e.g. A. Lane, Early Islamic Pottery [London, 1947], pls. 15A and 18A). 15  Cf. the same Sūra, vs. 49, for a slightly different version, which could also be the one intended here. 14

71

PROBLEMS IN ISLAMIC POTTERY

clue for the reader rather than a space-filler or a humorous device. The unexceptionable morality expressed in the aphorisms is of a piece with the sententious commonplaces which, embroidered on a sampler or inscribed on a plaque, have so frequently been a staple of interior decoration in the western world. The basically non-religious, even banal, content of the inscriptions, together with their stylish calligraphy, may perhaps be taken as adequate evidence of a purely aesthetic preference for this kind of decoration. Perhaps, then, one might suggest that the function of many inscribed Sāmānid pieces was display. Their difficult script would be no serious deterrent to this theory, for it would be intended as a challenge for the viewer. But it also establishes that this pottery was never geared to a truly popular taste. Never again were epigraphic themes to dominate an entire class of Islamic pottery. In Egypt, the roughly contemporary Ṭūlūnid epigraphic ware (nos. 74-5) betrays by its very inadequacy how difficult it was to combine bravura calligraphy with an innate harmony and balance. In Fāṭimid ceramics, inscriptions play a strangely minor role, and tend to be short. Indeed, it is not until the great efflorescence of ceramic types and techniques in the later Saljūq period that epigraphy comes into its own again. Yet even here strange anomalies exist. Why, for example, should mina’i pottery lack the lengthy inscriptions which occur so often on lustre, the other major type of luxury ceramic? The sheer expense of both types implies that both were made for the same class of patron. Only a rigid conservatism could keep the themes of mina’i and lustre so artificially apart. Such conservatism runs quite counter to the carefree innovation of ceramic techniques and designs in this period. Equally puzzling is the relative rarity of ambitious Kufic inscriptions as the major theme of decoration. Yet such inscriptions maintained their hold on contemporary metalwork and even more in the admittedly very conservative medium of architecture. In general, the Kufic script used on late Saljūq pottery is uninspired, full of mechanical repetitions and predictable rhythms. It has outlived its usefulness. The artist, bored with it, blunts and coarsens its forms. Significantly, it is used for empty benedictory phrases rather than 16

17

Cf. The Arts of Islam, 233. See L. Volov, “Plaited Kufic on Sananid Epigraphic Pottery”, Ars Orientalis VI (1966), 133. For a much more boisterous sense of humour expressed on medieval Iranian pottery, see no. 154 in the Keir collection, where the funnel-shaped spout of the bottle is decorated with a series of grinning faces, so drawn that they seem to be peering over the rim of the bottle. A pun on the genie of the lamp may even be intended. 16  17 

72

for the poetry which is the real subject matter of these pieces (fig. 3). For this poetry the medium of communication is a hurried, sketchy, cursive hand, strongly individual rather than formal – the “objective correlative” for the personal (and transient) emotions evoked. In analysing the subjects of these poems – which include the complaints of the lover to his beloved and reflections on the nature of grief – Professor Grube strikes an eloquent elegiac note. In these pieces, if anywhere in Islamic pottery, private feeling replaces the impersonal formality of most ceramic inscriptions. Yet the use of these particular verses betrays an unexpected element of convention, even of rote. For, as Professor Grube notes, the inscriptions often recur on different pieces. 18 19 The catalogue identifies 17 inscriptions on 7 pieces, but of these only a single text occurs just once, while one inscription is found no less than 20 four times in this small group of pieces alone. Clearly the artists dipped into a surprisingly small pool of verses when they came to select material for their inscriptions. The immense wealth and range of Persian medieval poetry remained unplumbed and we are left with these few verses which constant repetition would quickly have rendered hackneyed. Perhaps it would be more charitable to assume that in certain circles of society these represented the most quoted poems of the day rather than to infer that the potters and their patrons were badly read. The suggestion that these pieces were used as gifts from the lover to his beloved is admirably in tune with the romantic tenor of the inscriptions but does not recommend itself on other grounds; no record of such a custom appears to have been discovered as yet. But this repetition of a few verses does suggest that the pottery which they decorate was made for one class of society and probably for a limited period only. The quantity of surviving objects of this type need not in itself imply that they were intended for a middle-class milieu; minor courts abounded in Iran between c. 1150-c. 1250. The series of epigraphic wares in the Keir collection ends with the comparatively little known school of Mamlūk pottery. Its insistence, at this rather late date (14th century), on epigraphy as the major element of the design can be explained by the Mamlūk obsession with titulature, which finds ample expression in the architecture and metalwork of that dynasty. Poetry is conspicuous by its absence. The relationship of this pottery to   There are in fact a good many more.   This is the work of Oliver Watson. 20   The footnotes make it clear that the same inscriptions are found on many other pieces of contemporary pottery. 18 19

PROBLEMS IN ISLAMIC POTTERY

73

Mamlūk metalwork extends beyond shapes, armorial bearings, types of script and disposition of epigraphic bands to the unusual ochre and lemon glazes employed. This ware typifies the self-contained, inward-looking nature of Mamlūk art; it seems to owe very little to foreign influences. III The Keir collection provides an ample foundation for a study of Islamic architectural tilework. It is true that material from the Maghrib is unrepresented, but in the much more important Iranian tradition examples of virtually every technique are included. Many of these can no longer be found on any standing buildings and are thus liable to be ignored in the standard accounts of Persian medieval architecture. Many enjoyed no more than a passing popularity. Typical of these are the tiles from Ghazni which date from the Ghūrid period. They represent an experiment which deserved to fail (fig. 4). Coarsely designed and covered with glazes in muddy shades of russet and olive among other colours, they are among the ugliest medieval tilework which has yet come to light. It is hard to endorse the author’s speculation that such tiles could have been imported into Egypt all the way from Afghanistan. But Professor Grube’s suggestion that they formed the high spots in a predominantly plain tiled floor surface is very convincing; such tiles could not have competed with any of the established techniques of wall decoration. Moreover, their use as floor decoration would have exploited their strongest attraction – that of colour. Other early experiments in glazed tilework in the Iranian world during the pre-Mongol period also made the most of the contrast between glazed and unglazed elements by distributing the glazed elements parsimoniously against a monochrome surface. Īlkhānid figural lustre tiles were later to stand out in exactly the same way against a background of light or dark blue tiles. The tiles from Ghazni are precisely the kind of experiment which would be a by-product of the highly specialised terra cotta decoration of the Ghūrids and the Khwarazmshāhs. Monuments such as the minaret of Jām, the pishṭāq at Mashhad-i Miṣriyān and the southern tomb at Ūzgand all testify to the phenomenal versatility of this technique between c. 1180 and c. 1220. A moulded brick with an ambitious two-headed eagle design, very delicately and precisely delineated, was found in the Sāmānid levels at Nīshāpūr, while Saljūq moulded bricks with floral decoration and with inscriptions were discovered at Sīrjān and Ghubayrā respectively. The Ghazni tiles are lumpish by comparison but they do embody the technical advance of glaze. Their designs of single

74

1. Bowl with blue glaze decoration on white ground. Iraq, 9th century.

2. Bowl with slip-painted epigraphic decoration. Nīshāpūr, 10th-11th century.

PROBLEMS IN ISLAMIC POTTERY

3. Bowl with lustre-painted decoration. Iran. 12th-13th century.

4. Tile with moulded relief decoration. Ghazni, c. 1200.

75

76

5. Tile with lustre-painted decoration. Iran, 12th-13th century.

PROBLEMS IN ISLAMIC POTTERY

77

6. Tile with moulded relief and lustre-painted decoration. Iran, dated 860/1455.

78

7. Bowl with lustre-painted decoration. Iraq, 9th century.

8. Large bowl with lustre-painted decoration. Iraq, 10th century.

PROBLEMS IN ISLAMIC POTTERY

79

animals semi-rampant or passant, which also recall textiles, link them even more closely to the mainstream of domestic pottery rather than to its tributary of architectural tilework. The next two tiles in the collection, chronologically speaking, also did not serve as wall decoration. Both are Iranian and date from the 12th – 13th century. One, a hexagonal tile with gaping lion-head masks in high relief near each angle, is plausibly identified as a fountain cover and thus perpetuates that association between the lion and water which Islam took over from the classical world and which is found in places as far apart as Granada, the Kayseri area and Iṣfahān. The other tile depicts a miḥrāb enclosing a qandīl with the profession of faith above and a framing Qur’ānic inscription. If, as Professor Grube suggest, it was indeed used as 21 a tombstone – and numerous stone slabs of similar form, produced in Iran during the 12th century and bearing funerary inscriptions, still survive – it is not surprising that such objects were moulded. They would have commanded a captive market. But as with the Ghazni tiles, the rather poor quality of the moulded design (in comparison with what had been produced in 9th-century Nīshāpūr) is hard to explain though not to recognise. A comparison with typical stone prototypes of this tile reveals how virtually every element of the design has been ruthlessly streamlined and has thereby lost it original refinement – e.g. the decorative paired shafts at the upper corners, the segmental and double reversed curves of the inner arch framing the qandīl, or the decoration of the tympanum and extrados of the main arch. Even the apex of the arch is bungled; the arch profile does not come to a point. But such is the essential strength of the original layout that the tile still contrives to make a pleasing impression. The author’s promised study of a group of these monochrome-glazed tiles will fill a major lacuna in the field of Saljūq and Mongol pottery: for in these cheap, at least partly moulded and therefore mass-produced objects – if anywhere in Muslim ceramics – one could legitimately expect to meet a popular iconography with direct relevance to the perennial concerns of birth, marriage and death – to name only the most obvious subjects. With the mina’i and above all the lustre tiles in this collection we enter an entirely different imaginative world. Admittedly the mina’i tile (no. 150) exhibits only a neutral floral pattern, but the abundant Anatolian parallels, plus the few Iranian mina’i tiles carefully listed by the author, indicate that 21   The most thorough discussion of this question is by G. Fehérvári: “Tombstone or Mihrāb? A Speculation”, in Islamic Art in the Metropolitan Museum of Art, ed. R. Ettinghausen (New York, 1972), 241-54.

80

this tile probably formed part of a larger scheme of figural decoration. In this particular case that scheme must have been of unusual splendour, for even this subsidiary tile, whose function was to serve as a foil for the putative figural tiles, has an elaborate pattern and is not restricted to the monochrome undecorated glaze which characterised the cross tiles in the many contemporary lustre compositions of similar design. The lustre tiles published here are nothing short of spectacular. Two of them, nos. 182-3, are of the utmost rarity and their existence could scarcely have been predicted on the basis of surviving architectural tilework. Recent excavations of 13th-century palaces – e.g. at Kubadābād, Diyārbakr and Takht-i Sulaimān – have shown that the tiled dadoes of this period were even more scintillating and ambitious in palaces than they were in mosques, if the considerable surviving areas at Quhrūd and Mashhad are any guide. The tile with a scene from the Shāhnāma is welcome proof that the national epic did provide the subject-matter for large-scale mural 22 decoration. The ostensible subject here, Rustam about to rescue Bizhan, is set against a background of sprawling obtrusive foliage which threatens to engulf the whole scene. This feature is often encountered in 13th-and14th century book painting, but rarely in so extreme a form; Rustam’s body has to squeeze itself into the inadequate space left by these billowing foliate scrolls. The effect is absurd. Tile no. 183, which shows a ruler and his consort with fourteen attendants, all seated alfresco on a carpet by a stream, can lay claim to being the most astonishing piece in the whole collection (fig. 5). Its size alone (43 × 34 cm) is huge for this medium. Even the Ṣafavid tiles executed in the incomparably easier technique of haft rang do not approach this scale. Only in the Qājār period were tiles comparable in scale manufactured, but they could scarcely have less in common with this piece. Professor Grube is right to voice doubts about whether it is still in its original form. His surmise that it might consist of several pieces from separate tiles seems to be well founded. It would explain the otherwise arbitrary placing of the two courtiers in the frontal plane on the right, as well as minor inconsistencies in the border of the carpet and adjoining motifs,   The hero wears a curious striped turban. If this is intended to represent his traditional leopard-skin headgear, one might suggest that the leopard-stabbing figure on a 12thcentury lustre plate in the collection (no. 151) is also Rustam, for his turban is spotted in the same way as the leopard. The difference between Rustam’s headgear in ceramics and in miniature painting is – if these identifications are correct – quite marked, for the turbans have no suggestion of the leopard’s head. 22

PROBLEMS IN ISLAMIC POTTERY

81

and finally the variations in colour, which appear extreme even for lustre ware. These considerations naturally make any interpretation of the scene rather hazardous. But it is probably still justified to note that for all the sumptuous colour and refined detail of the tile, its organisation is chaotic. A remorseless will to pattern, which links it in spirit to the earliest lustre tiles from Sāmarrā’, removes any marked distinction between the bodies of the figures and their background. The edges of the tile promise no respite from this overcrowding and it seems likely that the neighbouring tiles would have made confusion worse confounded. The tile is an unusual phenomenon in Persian ceramics, in that design and form seem to be completely at cross purposes. Perhaps the difficulty results from the artist transferring to tilework the principles of design which were customary in curved lustre dishes, and then compounded his error by expanding the entire scene. Another source could also be canvassed. In all probability Saljūq wall painting included among its themes court scenes involving many figures. Indeed, the excavations of the Ghaznavid palace of Lashkar-i Bāzār have revealed paintings with serried rows of bodyguards. But there is no precedent for the mindless agglomeration of figures on this lustre tile. Such a crowd could easily be absorbed in a wall painting because the larger scale of that medium would automatically entail a more generous spacing. The mistake lies in transferring such compositions to the much smaller and more limited medium of ceramics, which cannot comfortably accommodate them. The celebrated mina’i bowl in the Freer Gallery is over-ambitious in exactly the same way as the Keir tile. The battle scene which it depicts is too lilliputian to be effective. No doubt the current convention of beauty, which dictated the rather swollen epicene faces and the tiny hands seen on the Keir tile, was not easily set aside, but it does seem strange that the artist should not have attempted to differentiate this mass of bodies by the type and patterning of their costumes. The same scrolls spiral over background and bodies alike. Professor Grube’s hesitation in identifying the bearded figure as the ruler, because he and the 23 figure beside him are not centrally placed, seems over-scrupulous. The composition is so crowded that a conventionally central position for this couple might well have looked rather artificial. They are central enough to attract the eye; they are, as the author notes, appreciably larger than the   It is perhaps strange, though, that the headgear worn by this couple is also found on several other figures in the scene. 23

82

other figures; the male is the only bearded figure on the whole tile, and his costume alone bears figural decoration – a flying bird of Chinese type. It is he and his consort that serve, if anything does, to draw the composition together. The Īlkhānid tiles, too, are almost all unparalleled on standing buildings. The exception is the lajvardlna type, which is attested at Takht-i Sulaimān 24 but which apparently enjoyed only a very brief popularity. Perhaps this was because its favoured themes were essentially alien to the Persian ceramic repertoire, and especially to architectural tilework. But in the Īlkhānid and early Tīmūrid periods a far wider variety of architectural tilework was deployed than even this collection indicates. Turbat-i Shaikh Jām and especially Pīr-i Bakrān both illustrate this, as do recent discoveries at Sulṭāniya. The group of blue and white hexagonal tiles from Egypt represents a curious by-way in the history of Islamic architectural decoration. It is notable, for example, that Iran, which boasts by far the most distinguished tradition of Islamic tilework, has virtually no tiles of this shape, decoration or colour. Even stranger is their Egyptian provenance, for no tradition of tiled decoration had developed there during the medieval period. When, in the previous century, glazed inscriptions of Iranian type had been introduced in Cairo, the fashion very quickly petered out. Although the case for Chinese influence has been made very convincingly, the immediate impetus for the motifs of these tiles must have come from much nearer home – perhaps Ottoman Turkey. The motifs are not only almost exclusively floral but also overbearingly so, as in much Iznik ware. The relationship of these tiles (or similar ones) to each other within a larger tiled panel is, however, rather problematic, for each is an entity in itself and thus differs fundamentally from Iznik tiles. An exception may be no. 252, where the design implies the existence of another tile to complete it, and where three adjoining sides have an extra border in light blue. Another curiosity is a moulded relief tile with lustre decoration which bears a building inscription and the year 1455 (fig. 6). This method of dating a building is highly unusual in Iran, though comparable tiles at 24  Almost equally short-lived was the fashion for the carved and glazed terracotta used in Samarqand in the later 14th century. Just as lajvardīna seems to have been the most spectacular form of tilework used for interiors, so was this technique paramount in external decoration. It is therefore doubly regrettable that these styles fell from favour so quickly. To the collections of Samarqand tilework in the U.S.S.R. and in London mentioned by Professor Grube may be added the substantial collection in the Royal Scottish Museum, Edinburgh.

PROBLEMS IN ISLAMIC POTTERY

83

Kūhpāya record the completion of a miḥrāb and a minbar. Small panels – not necessarily in the form of tiles – were a common way of dating newly completed or repaired sections of a building. But this tile seems a very modest memorial for an entire building, even if it was supplemented by the similar tile with the same inscription now in the Metropolitan Museum of Art. Other problems concern the poor quality of its script, the fact that the main subject matter of the tile is the vase and flowers rather than the historical inscription, and finally the unexpected and impractical device of beginning the inscription at the base of the tile and therefore upside down. The effect of this latter treatment is to draw attention to the name of the scribe and to the date of the building at the expense of words which are crucial to the sense of the text as a whole: “There ordered the construction of this building ...” Perhaps, though, this anomaly can be explained: if the tile had figured prominently in the building its inscription would probably in any case have been too distant from the viewer to be legible, whereas if it had been more modestly placed on a lower part of the building it would have been accessible enough for decipherment to present no problems. IV In his preface Mr de Unger confesses “Above all, I admire the lustre ware”, and in fact 86 of the 264 items in the catalogue are of this type. It is instructive to mark the gradually increasing confidence of the potter in manipulating this exacting technique. The early ‘Abbāsid pieces with highly stylised floral designs betray his uncertainty (fig. 7). This decoration is absurdly weak in comparison with what was being achieved at roughly the same time in architecture – the only relevant contemporary art form 25 of which sufficient examples have survived to make comparison possible. The designs perhaps follow a popular convention in late antique art in that no clear distinction is drawn between major and minor elements of the pattern. But very often the pattern seems a mere doodle, its density the result rather of the artist’s bankrupt imagination than of his desire to contrast colours. It is quite simply a waste of the medium (e.g. nos. 14, 16 and 17). Compared to such vacuous work the only slightly later ‘Abbāsid lustre ware with figural decoration represents a startling change of direction. Here the time-honoured themes of Coptic textiles and Sasanian  One has only to compare the infill of typical Sāmarrā’ stuccowork with that of early lustre pottery (e.g. no. 14) to be aware of the essential difference. 25

84

stucco plaques reappear in the distorting mirror of an alien tradition. The guiding principle behind these images is of course abstraction, a deliberate contrivance since the medium naturally allows ample detail. This abstraction announces itself in the reckless but supremely confident disregard of natural proportions, in the lack of internal modelling, and in the exaggeration of the features which identify a given animal. Thus the ears of a hare or the horns of a goat become as long as, or even much longer than, the animal itself. In the latter case (e.g. no. 24, fig. 8) the artist plays with exactly the same ideas as preoccupied the Iranian potters of the 4th millennium B.C. who produced the painted zoomorphic ware of Susa. In the ‘Abbāsid dishes a narrow plain white band loosely follows the contour of the images and sets it apart from a vigorously dotted background. This background is occasionally broken by enigmatic scrawls, blobs or other devices, also set against an enclosed white ground; given this prominence, they could almost be cabbalistic signs. ‘Abbāsid figural lustre ware illustrates a recurrent phenomenon of Islamic art: the sudden, unheralded appearance of an apparently novel and yet fully26 fledged style. In this particular case certain details can be recognised as borrowings from other media – for example, the white contour line of the form is a device familiar in Byzantine and Umayyad figurai mosaics, as is the animated background. But the important innovation here is that single, striking images of animals, at times distorted to the very limits of representational art, were chosen to serve as the principal decoration 27 28 of luxury ceramics. This theme became a standby of Islamic pottery , for reasons that are still largely unexplained. It is as yet impossible to say whether such animals were in any sense symbolic. Recent studies of the role of animals in Islamic iconography do suggest, however, that when a certain animal was intended to serve as a symbol it was placed in a context 29 which made this clear. Thus the stately tread and heraldic bearing of the royal animals on Islamic silks, displayed as they are in formal, richly   Compare the Dome of the Rock, the stucco of Sāmarrā’ or the Gunbad-i Qābūs, to choose only a few random examples. 27    No. 1 in this collection may be a minor exception. 28   They occur with equal frequency in unglazed pottery, such as Fāṭimid ceramic filters (nos. 78-81) or the moulded relief ware from the eastern Islamic world (nos. 65-67; cf. T. Allen, Reliefware from Rayy [unpublished M.A. thesis, University of Pennsylvania, 1974]). In the latter case a close relationship with roughly contemporary coins of this area is evident; compare, for example, no. 67 with a dirham of Kaikhusrau II (B. Spuler and J. SourdelThomine, Die Kunst des Islam (Berlin, 1973), pl. 267e). 29  E.g. Ettinghausen, From Byzantium, passim. 26

PROBLEMS IN ISLAMIC POTTERY

85

patterned roundels, convey the authentic flavour of princely life. These lustre pieces seem to lack any such context, though it is conceivable that in some cases the secondary features of the design may eventually prove to be identifying attributes. The variety of animals depicted on this ware is a further deterrent to assigning a symbolic value to them. In a style as homogeneous as this it would be bad method to single out a given animal as “symbolic” unless it differed from others of the group in its treatment or attributes. Similar problems attach to interpreting these subjects as part of a cycle. No special detail seems to demand this interpretation. At most, perhaps, some of the images could be understood as half-remembered echoes of an earlier, more directed iconography; the Sasanians in particular invested certain animals with royal attributes. There is of course no lack of debased Sasanian images used unthinkingly by early Islamic artists. But the concept of a cycle, when applied to ceramics, implies a group of objects made together and intended to be used or displayed together; in such a case even images that were rank clichés would acquire a meaning from their association with other images. Common sense suggests that such table settings must have existed. The difficulty in the way of such an interpretation for these animal pieces is twofold: first, no cycle of animal themes has yet been demonstrated in Islamic art; and second, the bowls which these animals decorate display marked discrepancies of diameter and height. It follows that each was created independently of other pieces in this style. The human figures depicted on this ware (nos. 35-38) are in an entirely different category, since many of them echo the themes of the cycle of courtly life, to be discussed shortly. For many tastes the nineteen pieces of Fāṭimid lustre ware will constitute the cream of the Keir collection. This material poses problems of technique, date and iconography. It is strange, for example, that despite the huge quantity of lustre pottery produced in Egypt during Fāṭimid times, the ware should be of such low quality technically. One cannot easily equate centuries of practice in this proverbially difficult technique, and a standard of pattern-making and figure drawing which far exceeds earlier Iraqi work, with inadequate expertise in the application of the pigment, in the composition of the glaze or in the firing process. Professor Grube is surely right to question the popular dating of so much of this ware to the 12th century. Such a date can be excluded for the mass of this material on grounds of sheer common sense; a dynasty that could afford to build as lavishly as did the Fāṭimids from 970 – c. 1100 could certainly have provided the patronage for vast quantities of luxury ceramics. The

86

subject matter of this ware does, however, entail major difficulties of interpretation. Themes that dominate in other styles of Islamic pottery are relegated to a secondary role. Abstract decoration is very rare as the 30 principal theme in Fāṭimid lustre, though not in other Fāṭimid pottery – which may well indicate that the figural content of the costlier ware reflects the concerns of the court. But it seems a little bold of Professor Grube to interpret most of such themes as part of the formal cycle of princely life; for in that case the cycle would lose its specific quality and instead simply express everyday life. A useful yardstick for the content of the princely cycle is provided by the subject matter of the sculptures on the Aght’amar church and of the ceiling paintings of the Cappella Palatina. In both cases the buildings were outside Islamic territory, but served a society with a powerful Islamic presence. In neither case is the Islamic iconography given pride of place – that would have been too much to expect in places of Christian worship. But despite its lofty setting in both instances, this decoration does seem to have symbolised political allegiance. A propaganda element may thus have been intended in the choice of decoration; this was art for export. Such a motivation may explain why in both cases the cycles are so detailed. They deliberately advertise the splendour of Islamic court life. Curiously enough, no equally full cycle survives on any Islamic building. While certain elements in the cycle – such as the seated monarch – quickly achieved an independent status which accounts for their frequent occurrence in other media, the existence of such clichés should not be taken to imply that in any given case they were intended as part of the wider cycle. In the case of pottery, the use of this courtly iconography could be demonstrated if it could be proved that a group of pieces formed part of a table setting; but this has apparently not been possible so far. Moreover, the much-vaunted realism of the figural scenes on Fāṭimid lustre ware might be considered an inappropriate manner for representing an essentially formal, official 31 theme. Many of the pieces may simply have been made for display. Such an interpretation has the merit of providing an umbrella for subjects which are clearly outside the princely cycle. Examples of such subjects are Christian themes – presumably explained by the presence of a strong

30   Nevertheless, intrusive foliate ornament marks many of these pieces (cf. Grube, Islamic Pottery, 140, for an analysis of this phenomenon). 31   Cf. Grube, op. cit., 130.

PROBLEMS IN ISLAMIC POTTERY

87

Coptic community, though one may also bear in mind the brief Crusader presence in Egypt. Christian subjects are quite frequently found in 13-th century Near Eastern metalwork. The zodiacal subjects which are so popular in that metalwork strangely find no echo in this Fāṭimid pottery, though no. 82, a bowl with a face identified as the sun itself, might be an exception. If this latter interpretation is correct, the context of the 33 bowl remains a puzzle. Some of the animals depicted on Fāṭimid lustre could refer to the zodiac, but the theory proposing astrological symbolism could only be clinched by inscriptions or by the discovery of a piece with a motif which could only be interpreted in astrological terms. It will be seen that much remains to be done in elucidating the meaning of Fāṭimid lustreware. 32

V Professor Grube devotes a fascinating and all-too-brief section to the subject of dated Islamic pottery in general. The fact that so little Islamic pottery is dated – some 62 pieces are mentioned in all – is of capital importance. Far fewer pieces still mention the name of the patron. In both these respects, particularly the latter, the contrast with metalwork is striking. As the author points out, earlier dated objects of metalwork and woodwork are known, while dated textiles of the 9th and 10th centuries abound. Coins and architecture were dated from the very first century of Islam. It is hard to suggest why ceramics should be the last of the major art forms in Islam to bear dated inscriptions, and why these inscriptions should always be so rare in that medium. In other ceramic traditions, however, the same phenomenon recurs. Dating implies not only a certain self-consciousness, but also an appeal to posterity, and this perhaps accords ill with so fragile a medium as pottery. The luxury wares of the first three centuries of Islam afford ample proof of the major status of this “minor” art at that time. But the ephemeral nature of much   As Professor Grube says, “there is ... no indication that Coptic workshops ever produced lustre-painted ware”, but it would be quite conceivable for richer members of the Coptic community to commission lustreware with Christian subjects. Christians were widely employed in the Fāṭimid administration and several on occasion even became viziers (M. Canard, “Fāṭimids”, Encyclopaedia of Islam, 2nd ed., II 858). It would therefore be perfectly natural for Christian subjects to appear intermittently in Fāṭimid art. 33   Solar symbolism is also proposed for two Iranian bowls of 12th-13th century date which exhibit rays or lines spreading from a central circle, like the spokes of a wheel. Here again the shape of the bowl is central to the interpretation, but the argument is untenable because the same design occurs on a contemporary pitcher (no. 127). 32

88

of Islamic art can scarcely have escaped the notice of medieval craftsmen. Splendid palaces would be erected only to collapse or to be pulled down within a generation or two. Lavish court festivities demanded an even more transitory magnificence to which the accounts of Fāṭimid court life testify. In this context it might well have seemed eccentric or pretentious to treat pottery on the same level as more permanent artifacts. Tiles clearly came into a different category, presumably because they were used in the context of architecture, which implied a degree of permanence. This was especially true of religious architecture, for which most of these tiles were made. Thus dated tiles far outnumber dated pottery. An awareness that pottery, as distinct from tilework, was by its very medium an art of the moment need not have inhibited the artist from giving of his best. If the meagre quantity of dated pieces can indeed be interpreted to indicate that the period c. 1150 – c. 1250 saw a change in this attitude and a growing readiness on the part of the artist to see luxury pottery as a permanent art form, then the advent of dating inscriptions can be associated with a deepening complexity of subject matter, often with symbolic content. The key surviving document in this change is perhaps the Freer lustre plate dated 1210. While Professor Grube is right to doubt whether a mere half-dozen dated inscriptions on pottery from the later 12th century are sufficient evidence to chart a marked change in iconography in that period, a further twenty are available before 1242. With such quantities it is possible to construct hypotheses. A change in the attitude of the artist – or the patron – to luxury pottery had certainly taken place by the midthirteenth century, even if we do not know when or why it began. The signposts of that change are two-fold: the occurrence of a date and the depiction of complex figural subjects. VI The foregoing remarks may hint at the range of topics discussed in Professor Grube’s book; that range is further anatomised in a useful index. As he himself repeatedly states, the format of a catalogue does not permit the detailed exploration of byways. This is especially and regrettably true of the discussions of iconography and symbolism. But pointers for further research abound; a random list might include the influence of the Maghrib on Fāṭimid pottery, the function of a small glazed head (no. 101), the rarity of Tīmūrid pottery or of ceramic beakers vis-à-vis glass ones, the use of decoration to differentiate the products of the various Saljūq ateliers, and the reason why so many conventions for facial depiction co-existed

PROBLEMS IN ISLAMIC POTTERY

89

in Fāṭimid pottery. Parallels for the objects described are provided in the footnotes. This discussion often entails a rich bibliography covering other Islamic arts or non-Islamic pottery traditions; such titles do not appear in the main bibliography. In many cases the author lists all the examples he knows of, including unpublished pieces in museums or private collections. These lists are themselves a remarkable achievement. They show – as do, for example, Professor Grube’s comments on the differences between superficially similar Iranian and Egyptian 15th-century ware – that there is no substitute for decades of experience in handling pottery. Combining as it does a sharply focussed, meticulous technical examination of the ceramics in the Keir collection with a constant awareness of the wider implications of their style and iconography, this volume has broken through the constraints of the catalogue form to become an indispensable handbook of Islamic pottery.

90

IV Persian Lustre Ware

I

this book Dr Watson has met a need that has existed for the best part of a century. In all that time there has been no lack of general discussion of Persian lustre ware in the context of handbooks and exhibition catalogues, and a few more ambitious studies of individual pieces or specific lustreware problems have also appeared, including several by the author himself. Yet there had been no serious attempt to see the evolution of Persian lustre steadily and see it whole—no monograph, in short. Here it is. That is reason enough to give this book a warm welcome—and would-be criticasters would do well to begin by reviewing the formidable difficulties of undertaking such a monograph, difficulties that were enough to put off even those of Dr Watson’s predecessors who wrote extensively on lustre. Not the least of these difficulties is the sheer quantity of material—many hundreds of pieces in public and private collections scattered throughout the world. The problem of mass is compounded by the still unresolved dispute over the authenticity of key pieces. The author does not tackle this issue head-on, though one may assume that he accepts the genuineness of the pieces which he illustrates. By the same token, the absence from the discussion of such pieces as the Brangwyn plate may mutely express his doubts about them. In the event, this is much more than a bald history of Persian lustre ware. Dr Watson has conscientiously attempted to set these wares within their wider ceramic context in Iran itself, and also within the more specific non-Iranian context of Islamic lustre generally. To this reviewer, it seems that he has achieved the task very well, maintaining a sharp focus on what is relevant and telling a complex story clearly. The obligatory chapter on n

1

 Oliver Watson, Persian Lustre Ware, London, Faber and Faber 1985.

1

PERSIAN LUSTRE WARE

91

technique is also briskly and competently handled. In the chapter which deals with actual and phantom production sites Dr Watson clears away the accumulated detritus of a century’s speculation and wishful thinking, and by the judicious application of Occam’s razor reduces the numerous production sites previously suggested to a single one—Kashan. This finding, first announced by the author in general terms more than ten years ago in a seminal article in the Transactions of the Oriental Ceramic Society, and now formulated more precisely and definitively, bids fair to represent his major contribution so far to ceramic history. It is big with implications: for example, it establishes that lustre was a monopoly, that the ware was widely exported, and that it was more of a luxury ware than has often been supposed. It suggests too that the subject-matter of lustre-ware decoration will bear closer analysis than it has received to date. All this entails, for good measure, a comparable re-assessment of mīnā’ī ware. It will of course take some time before these implications are digested and before the ensuing revaluation of the role of lustre finds its way into the general handbooks; but it is not going too far to say that a major problem of lustre ware has now been solved once and for all. This, then, is one of the principal achievements of the book, and it is good to have it so clearly spelled out. There follows a trio of short chapters in which an attempt is made to categorise Kashan lustre of the 12th and 13th centuries into three major groupings: the Monumental style, the Miniature style and the Kashan style. In Dr Watson’s view this is roughly their chronological order, and the Kashan style is an amalgam of its two predecessors. Since this period is an obvious high-water mark of Persian lustre it is a pity that the discussion of it should be so brief—for these chapters occupy a mere 4, 5 and 6 full pages of text respectively. It seems likely that this brevity was imposed on the author by the traditional constraints of the series in which the book appears—the prestigious Faber Monographs on Pottery and Porcelain. One must hope that Dr Watson will return to this subject before long and develop his views at much greater length than he does here. Meanwhile, among the numerous details which invite further research, one might single out the propensity of pieces in the Miniature style to have their surfaces divided into friezes and panels, a feature which further strengthens the analogy with book painting already suggested by their figural style. Similarly, in the Shāhnāma tiles from Takht-i Sulaimān the grotesques in the form of tendrils sprouting animal heads, though initially similar to the animated inscriptions of 12th- to 13th-century metalwork, also evoke

92

the “earth spirits” which were to become such a pervasive feature of later Persian painting. Such connections between lustre ware and other media could be multiplied, and indeed it is within this wider context of analogy and meaning that future research into lustre ware might most profitably be pursued. The chapter on Īl-Khānid wares is almost inevitably an anticlimax after the splendours of the “Saljūq” material, though the historical evidence it adduces as to the survival of lustre production right through the turmoil of the Mongol invasions and their grim aftermath opens perspectives much wider than those of ceramic history alone. The chapter on free-standing sculptural figures has a text too thin—a bare one and a half pages—to do more than tilt at the significant issues of form and meaning which it raises. This chapter, indeed, dips somewhat below the generally high level of the book and is conspicuously devoid of context—for example the stucco sculpture and the metalwork of the period, and the uses which these pieces served. The connection between lustre and metalwork, a topic mentioned glancingly here and there throughout the book, in particular deserves a much more detailed analysis than it receives here. With the tiles covered in impressive depth in Chapter 10, Dr Watson is clearly on familiar ground, and this is the most ambitious and meaty chapter in the whole book. He deals in turn with the great miḥrābs with the multiple implications of the many signed and dated pieces, and with various oddities like the Paris disc containing the foundation inscription of a vanished shrine. This is a chapter worth re-reading several times, for a multitude of pertinent and challenging observations are embedded in it. Chapter 12, on the lustre ware produced between the 15th and the 20th century, neatly wraps up the later stages of the story. It summarises and revises a much longer article on this topic published by Dr Watson a dozen years ago, adding some important new ideas and material. Of particular interest is a Ṣafavid tile fragment persuasively attributed to a fountain or watercourse. The wavy fronds of its decoration would have suited it admirably for this role, as the rippling water would have produced the illusion that the leaves were swaying in the current, while the sunlight on the water would have reinforced the inbuilt reflection of the lustre itself. This kind of elaborate pun seems to be an integral element of lustre ware generally. Chapter 11, entitled “Images, Inscriptions and the Use of Tiles”, is an attempt to grasp the nettle. What do these lustre wares mean? For part of this chapter—the section on inscriptions—Dr Watson is on relatively

PERSIAN LUSTRE WARE

93

solid ground. His comments on the popularity of quotations from the later sections of the Qur’ān (passages dealing with judgment and hell) as tomb decoration are thought-provoking, and he gives several examples of the type of poetic text found on lustre vessels and tiles. Yet the material he presents poses insistent questions, as he himself is aware, and it is disappointing that they are not answered. Why do Qur’ānic quotations occur only on tiles? Why is the Verse of Light, the one Qur’ānic verse whose appropriateness for miḥrāb decoration was accepted throughout the Islamic world, so sedulously omitted from lustre miḥrābs in which reflected light was an essential attraction? How is the use of living creatures in the background of lustre Qur’ānic inscriptions to be explained? Why does a somewhat colourless reference to Rustam preparing for the hunt recur so frequently in quotations from the Shāhnama when the image itself has nothing to do with him? The use to which Ṣūfīs put the Shāhnāma, a matter recently explored by Dr Melikian-Chirvani, may offer some clues here. Why is Anatolia, with its remarkable wealth of 13th- to 14th-century architectural tilework, entirely devoid of lustre tiles in religious buildings, even though abundant examples are known from secular ones? Of course it is much easier to ask such questions than to answer them. Nevertheless, one may hope that Dr Watson will before long provide at least the basis for further investigation from some of the raw material he must have accumulated. Thus a check-list of the Qur’ānic inscriptions on lustre tiles, and of the verses which have been read on lustre ware, would be very useful indeed. It is here that the extremely fruitful results of Dr Kiani’s excavations at Gurgān would help significantly to broaden the scope of current knowledge. One finding of Chapter 11 which deserves special discussion is the association of lustre tiles with funerary monuments, and specifically with those which have Shi‘ite associations. The connection is perhaps not as exclusive as Dr Watson suggests, in that the Ṣūfī element, as distinct from the narrowly Shi‘ite one, is somewhat more pronounced than he asserts— there are lustre triple pentagons in situ at Pīr-i Bakrān, while both Naṭanz and Bākū (pace his remarks) are Ṣūfī, not Shi‘ite, shrines. Nor is the case for Shi‘ite funerary associations strengthened by the fragments of a lustre dado in the Madrasa of Shihāb al-Dīn Qāsim Ṭarāz in Yazd, a building dated 737/1336, or the débris of a long thulth inscription in lustre tiles, forming a frieze 45 cm wide, found in the ruins of Hülegü’s observatory at Marāgha. Above all, and in addition to these examples, the material assembled by Dr Watson himself on p. 188 of his book mentions lustre

94

tiles, or fragments of them, found at a further 15 sites of which only two have a clear Shi‘ite connection. All this points to a rather less exclusive use of lustre tiles than Dr Watson’s interpretation would suggest. Perhaps the theory should be modified to indicate that in the popular mind the connection between lustre tilework and funerary buildings was perhaps stronger than with any other type of monument, and one should not exclude the possibility that they were used in important secular buildings for their general connotations of wealth and luxury. The lack of specifically Shi‘ite themes in lustre ware generally should in fact enjoin caution in formulating a connection between this material and Shi‘ite beliefs. If the same type of lustre tile occurs indiscriminately in palace, mosque, observatory, madrasa or tomb, the chances are small that it bears a specifically Shi‘ite charge. Rather does it suggest that the tiles were made with a wide spectrum of uses in mind. Apart from the Takht-i Sulaimān tiles, it would be hard to point to any evidence suggesting that a custom-made iconography was developed for lustre tiles. Perhaps, then, it was the lustre itself, replete with mystical and alchemical associations, which really counted; the images themselves may have been of relatively secondary importance in comparison. By that reckoning, lustre would have been chosen for architectural decoration when splendour of a more than usual kind was required. What better way to honour a saint than to clothe his tomb in that most costly material? And in the fervid religious atmosphere of the times, what more likely type of building for such an honour than the tomb of a saint, with all the local devotion which his memory inspired? Such considerations would go far to explain the frequent use of lustre tiles in Shi’ite shrines, and would obviate the need for a theory linking the two. This in turn raises interesting questions about workshop practice. Dr Watson notes the presence of 250 tiles datable to 700/1300 at Quhrūd, and another 60 which are datable to 707/1307; elsewhere he mentions 150 tiles produced in about three months by a single workshop, another 30 made in a month, and so on. These are substantial figures, and it is a pity that no serious attempt has so far been made to assess the implications they have for workshop practice and output. Dr Watson suggests that the presence of several styles on pieces made successively from a single mould implies that several painters were active in a single workshop. One may perhaps go further and suggest, with all due reserve, that tiles were prepared as part of the general stock and that the way they were used depended on the purchaser. When a specific order was received the

PERSIAN LUSTRE WARE

95

procedure would of course be different, and this may well explain the unimpeachable orthodoxy of the miḥrābs at Qumm and Mashhad, which lack the figures of birds and animals found in other lustre miḥrābs. It would also explain the peculiarly complex iconography of the Freer dish dated 607/1210, one of only three surviving pieces dedicated to named individuals. All this suggests that a hierarchy of sorts was in operation, as indeed common sense would suggest. It might even be proposed that most miḥrābs, being relatively small, were also not made to order, and that most patrons were indifferent to the presence of discreetly camouflaged living creatures on them. The notion that only a few lustre tiles were made for a specific building explains the generally bland tone of their iconography and should make one view with a certain reserve attempts to interpret that iconography as a whole. The best prospect of understanding what these images mean is a study of those that have remained in situ or whose provenance is assured. These remarks in turn raise perhaps the major question left unanswered by this book: the meaning of the images on lustre ware. Dr Watson states at the outset that his book is “solely a study of ceramic history”, and on pp. 20, 121 and 156 he expresses his reluctance to embark on speculation about meaning and iconography. Yet this is to let the baby out with the bathwater. Lustre ware was the most expensive luxury ceramic type of medieval Iran and here, if anywhere in the field of pottery, one might expect meaningful images to be developed. Abu’l-Qāsim, after all, who came from a family of lustre potters and wrote a famous treatise on pottery which in passing touches on the technique of lustre, went on to become a court historian to the Mongols. Thus even if lustre ware enjoyed less kudos than fine metalwork or glass, its ambience was emphatically not that of a mere bazar craft. It had a quality of mystery—indeed, Abu’l-Qāsim described it as “a kind of alchemy”—and its connections with gold (the noblest and most high-ranking of metals in the medieval mind, as well as the most precious), riches, light, and the sun would not have escaped the artists who so painstakingly manufactured it. Rather would one have expected them to capitalise on such associations. Readers interested in pursuing these ideas would do well to consult Chapter 11 of Alan CaigerSmith’s Lustre Pottery, published almost simultaneously with Dr Watson’s book. Signed lustre ware has long been a particular interest of Dr Watson, and the book spotlights the dominating role of a single figure: Abū Zaid, whose career spans the golden age of “Saljūq” lustre. Even though

96

nothing is known about his personal history, his many surviving signed pieces—lustre and mīnā’ī, miḥrāb, bowl and tile alike—his use of his own poetry on some of them, and above all the fact that his work is found in the two principal shrines of Iran, namely Qumm and Mashhad, suggest a man at the very pinnacle of his profession, with sovereign confidence in his own ability. Nor was his an isolated case; a little over a generation later, the work of ‘Alī b. Muḥammad b. abī Ṭāhir was to be found all the way from Najaf in Iraq to Varamīn in Central Iran and Mashhad in Khurāsān and perhaps as many as three sites in between (Qumm, Dāmghān and Gurgān). Given that the surviving evidence is so patchy, the likelihood is that such men produced very much more and were correspondingly still more eminent in their profession. It is perhaps not surprising that a monopolistic craft localised in a single town should from time to time produce a dominating personality of this kind. Much of the work reads with deceptive ease as if it were a handbook summarising familiar information. In actual fact, of course, the information is anything but familiar; yet it is a significant achievement to make it seem so, and this is due to the lucid organisation of the material and to the author’s practised expository style. It is in the three impressive appendices that the genesis of this book in its author’s doctoral thesis can be sensed. They are the iceberg of which the text is the tip. A vast amount of information has been condensed into these 20-odd pages, which discuss lustre potters and their works, buildings decorated with lustre tiles, and finally include incomparably the most detailed lists yet published of dated tiles (143 in all) and dated lustre vessels (77), with notes on the collections where they are to be found and earlier publications of them. A significant number of these pieces are unpublished. Whatever modifications subsequent scholarship brings to the main text, these appendices give the subject a far broader and more secure base than it had previously possessed; and for that, Islamic art historians will long remain profoundly grateful.

V Richard Ettinghausen and the Iconography of Islamic Art

T

his chapter will begin by trying to define where Richard Ettinghausen belonged in the context of the German museum tradition; it will then summarize briefly where his strengths lay; next, it will attempt to situate his 2 development within the elusive matrix of German refugee scholarship; and finally, it will seek to pin down his approach in the study of Islamic iconography (see fig. 51). In many of the obituaries and appreciations that appeared after 3 Ettinghausen’s death in 1979, the formative influences on him as a scholar are held to be such men as Wilhelm von Bode, Friedrich Sarre and Ernst Kühnel, all of whom had distinguished careers as museum men in Berlin. 1

1  I should like to record my warm gratitude to Oleg Grabar for his generously full comments on an early draft of this paper. An earlier version of part of this paper was published as a review of Miriam Rosen-Ayalon (ed.), Richard Ettinghausen: Islamic Art and Archaeology: Collected Papers (Berlin, 1984), in Bibliotheca Orientalis XLVI/1–2 (1989), cols. 207–10. 2  This is the most speculative part of the paper, for it touches on what happened to Richard Ettinghausen in the turbulent times of the 1930s and 1940s, and tackles some of the wider implications of being a refugee. I write here from personal experience. 3  See, in particular, the booklet Richard Ettinghausen: February 5, 1906 – April 2, 1979. In Memoriam (n.p., 1979), with appreciations by Stephen E. Ettinghausen, John H. Marks, Kurt Weitzmann, Adelyn D. Breeskin, Oleg Grabar, Priscilla P. Soucek, R. Bayly Winder, Angelica Zander Rudenstine, S.D. Goitein, Philippe de Montebello, Craig Hugh Smyth, Philip D. Minasian, Norbert Schimmel, John C. Sawhill and Marilyn Jenkins.

98

And it is true enough that Ettinghausen knew these men and their work well. He certainly understood their ideas. But it is a serious mistake to hold that he fitted snugly within this tradition, a tradition to which his almost exact contemporary Kurt Erdmann belonged. Ettinghausen did not have the lengthy museum apprenticeship that would have been required for membership of that club, for while he did indeed begin as an assistant to Kühnel in the Islamic department at the Kaiser-Friedrich Museum in Berlin after completing his doctorate in 1931, he left as early as 1933 to spend a year in England, at Cambridge, and by 1934, at the age of 28, he had already emigrated to the United States. For the next ten years he was employed in and around American universities and research institutes. It was only in 1944 that he was appointed to a museum post, at the Freer Gallery in Washington, and by that time his scholarly formation was complete, as the major articles on Fatimid painting and the Bobrinsky 4 ‘bucket’ demonstrate. The German museum tradition in the field of Islamic art, and the scholarship that it had generated, was altogether too pedestrian to inspire him. As one can see in the work of Sarre and Kühnel (and later Erdmann), it dealt too much with trivia – the brief publication of this or that ‘new’ object. In such articles, the basic documentation is unexceptional, and one cannot deny that a stone has been added to the cairn of knowledge. But the wider and subtler faculties of the mind are not seriously engaged. Ettinghausen, by contrast, had the instincts of an explorer rather than those of a recorder of information. Beside him, these others, frankly, 5 look dull. He had far more in common intellectually with Ernst Herzfeld and, more generally, with an adventurous, speculative, even diffusionist tradition of intellectual enquiry in art history – in other words, the tradition represented by so many other émigré German art historians of the 1930s. Yet his instincts as a museum man and his usually unerring feel for objects preserved him from the worst excesses of speculative scholarship. He was never in danger of turning into some Josef Strzygowski, foaming at the

Richard Ettinghausen, ‘Painting in the Fatimid Period: A Reconstruction’, Ars Islamica, IX (1942), pp. 112-24; idem, ‘The Bobrinski “Kettle”: Patron and Style of an Islamic Bronze’, Gazette des Beaux-Arts XXIV (1943), pp. 193–208. 5  Though he did not have Herzfeld’s Classical background, his philological expertise, his archaeological and architectural bent and his fluent draughtsmanship – all of them qualities that constantly fed into Herzfeld’s scholarship. 4 

ETTINGHAUSEN AND ICONOGRAPHY

99

mouth as he propounded some cherished theory in defiance of common 6 sense. A clue to Ettinghausen’s own preferences as a scholar may be found in the extremely generous tribute that he paid to Kühnel in the preface he wrote to the Festschrift (1959) for that scholar, a work which he organized and edited. He singles out for special praise the ‘wide range of subjects covering every branch of Islamic art from Moorish Spain to Muslim India and every field with which Islamic art had any relations’, the ‘originality of his research… his pioneering spirit… his rare ability to penetrate to the essentials of a broad subject… [and] to write general works in a lucid and 7 stimulating manner’. All this could, with much more justice, be an epitaph for Ettinghausen’s own scholarly achievement. It may also indirectly shed light on what he was trying to achieve in his own published work. Unlike von Bode and his colleagues, Ettinghausen was by temperament, and also to a significant degree by training, an orientalist in the broad sense rather than a museum curator alone. The word ‘alone’ is necessary because, while the museum side was indeed an important element of his overall make-up as a scholar, it was not chronologically the first nor perhaps even the crucial one. And the phrase ‘orientalist in the broad sense’ – the term is not intended to evoke the negative connotations that it has acquired in 8 the post-Said era – can be justified by the fact that he was not at base a philologist. His work is not founded on close and extensive reading of original sources in Arabic, Persian or Turkish, and it is perhaps significant that the topic of his German Ph.D., anti-heathen polemic in the Qur’an, did not act as a base for his future work in any obvious sense. That said, he did have a particular fondness for the Persian language, at which he 9 worked hard in the early 1940s, and which he loved to speak. It is worth noting, however, that he did not have many of the experiences which nowadays are taken for granted as part of the intellectual formation of Islamic art historians – for example, archaeological fieldwork, architectural surveys, wide travelling in the Muslim world and living for a substantial period in a Muslim country. Thus his very strengths as a museum man were offset by certain lacunae. 6   See Robert Hillenbrand, ‘Creswell and Contemporary Central European Scholarship’, Muqarnas VIII (1991), pp. 27–8, 30. 7  Richard Ettinghausen (ed.), Aus der Welt der islamischen Kunst: Festschrift für Ernst Kühnel zum 75. Geburtstag am 26.10.1957 (Berlin, 1959), p. 9. 8   See Edward Said, Orientalism (New York, 1978). 9   I am grateful to Oleg Grabar and to Jerry Clinton for information on this topic.

100

The sheer range of Ettinghausen’s expertise leaves one with an abiding sense of awe. The days are now definitively gone when a scholar could not merely aim to cover the entire chronological and geographical span of Islamic art, from the seventh to the twentieth century and from Spain to India, but also succeed in the attempt at the level of the scholarship of the time. The system has now geared itself not just to prefer the specialist but also to reject the would-be ‘Renaissance man’. While much has been gained in the process – the data base for Islamic art historical scholarship is now incomparably bigger than it was, say, 40 years ago, and this has made it possible to create a much more nuanced picture of the past – there have been corresponding losses. These make themselves felt at both the general level represented by handbooks of Islamic art and by general accounts of specific media, dynasties or periods, and at the level of specialist pioneering research in which new information is unearthed and interpreted. Specialization has clogged the sustained flow of information, association, intuition and speculation which, when wielded by a master, can so enrich both the overall perspective of a field and the close-focus study. The breadth of sympathy that made virtually nothing to do with Islamic civilization alien to Ettinghausen is fostered no more, indeed, is positively restricted in today’s scholars by the need to cultivate intensively a limited field of enquiry. That represents a quantum leap backward as well as forward in Islamic studies. After all, such a commitment to a limited field usually engenders a corresponding and mounting unease the further away one ventures from it. The result is all too often intellectual impoverishment. Such a figure as the Muslim geographer Ibn Battuta, who in the fourteenth century could travel for most of his adult life from his native Morocco to India and beyond, and yet remain the whole time within a single civilization united by similar institutions and forms of worship, is a reminder that the approach adopted by Ettinghausen has the sanction of medieval example behind it. Yet there was emphatically nothing superficial about Ettinghausen’s work. He was, if not a ‘Renaissance man’, one of the few scholars of Islamic art who was a generalist and specialist alike, and it is doing him no more than justice to salute him as a genius in the Oxford English Dictionary’s definition of that word: ‘Native intellectual power of an exalted type, such as is attributed to those who are esteemed greatest in any department of art, speculation or practice; instinctive and extraordinary capacity for imaginative creation, original thought, invention or discovery. Often contrasted with talent.’

ETTINGHAUSEN AND ICONOGRAPHY

101

But of course there is much, much more to the making of a great scholar than a first-class intellect. Such a person needs to possess an everflowing fountain of inspiration, and an even more copious fountain of perspiration, and the wisdom to know the difference. Ettinghausen had that vital gift. It kept his feet on the ground, and it kept his head the proper size. Above all, it seems that he had some elusive extra quality that set him apart from the crowd. One may recognize well enough, for example, that the artist Jan Vermeer achieves something with the play of light in his paintings, or in the balance between lemon yellow and powder blue, that magically transforms everyday reality; but the secret of that mysterious alchemy simply will not be pinned down. We can all think of musical examples of the same phenomenon, when a passage somehow slips under our guard and transports us to another place, a place of revelation of whose existence we otherwise have only rare intimations. And the same is true of literature – poetry, as they say, is what gets lost in translation. Insight of this superior kind is also possible in creative scholarship, even if it is encountered only rarely, and it is worth considering how Ettinghausen achieved it – though the road to it perhaps leads through dark places. There are, no doubt, those who may find the paragraphs that follow inappropriate in a scholarly appraisal of Ettinghausen. It might be argued that essentially the man is the work that he has left behind. But this would be to deny a more holistic approach that is peculiarly apt in the context of the turmoil of the 1930s and 1940s and the massive disruptions and suffering that it inflicted on so many people. Some people weathered these storms; others were scarred for life, although they did not speak of their experiences, preferring perhaps at some subconscious level to bury the pain. What then, of Ettinghausen’s place in the context of German refugee scholarship? It is common knowledge that German, mostly Jewish, scholars poured out of the Third Reich in the 1930s. While some of them, such as Hugo Buchthal, Ernst Gombrich, Otto Pächt, Nikolaus Pevsner, Fritz Saxl and Rudolf Wittkower, settled in Britain, the majority went to the United States. ‘Settled in Britain’; ‘went to the United States’. Consider for a moment what lies behind those innocuous words. Imagine what it might be like to leave your loved ones, your native country, your house, your job and your books, in other words literally everything you have, and to have to do all this perhaps at fearfully short notice, in secret and in dread. For a scholar after all is first a human being and does not, cannot, live entirely

102

1. Richard Ettinghausen

ETTINGHAUSEN AND ICONOGRAPHY

103

in the study. Think of the countless blows to your pride, self-esteem, confidence that such a harrowing experience brings in its train, and the gnawing uncertainty of what tomorrow will bring. For the resettlement process was not straightforward. In the United States, someone, some fairy godmother, had to persuade the authorities to allow these refugees to enter the country, had to find a place for them to live, money, a job that Americans in the Depression could not do. The nightmare did not end when they left Germany. Bitter the bread that exiles eat. To become a refugee in adult life challenges the whole of what you are. In the field of art history, the end result is familiar: Ernst Herzfeld, Ernst Kitzinger, Richard Krautheimer, Erwin Panofsky, Kurt Weitzmann. As Walter Cook in Princeton said, ‘Hitler is my best friend; he shakes the 10 tree and I collect the apples’. But it really was not as simple as that. It is true that for the most part they were still youngish men, in their thirties. Ettinghausen, as it happens, was younger still. Even so, some never managed to recover their mental, emotional and spiritual equilibrium. They could not come to terms with the changes that had been forced 11 on them. Bernard Malamud has written movingly about these casualties. Those who made it were changed, changed utterly. And this is where it is necessary to use one’s imagination. Was it that they had been able not only to digest the atrocious upheavals in their lives but also in some way to transmute them, to turn them to good? That means more than simply starting again in a new environment, where they could breathe the air of freedom. It is hard not to sound pompous or banal saying this, but the events of the 1930s and what followed forced these scholars to confront, in their own personal lives, questions of human destiny. Those are not questions that come naturally to mind in the study of art history. But perhaps they should, and we in turn should be the richer for it. The picture that I have drawn here may seem too highly coloured, too emotional or even romanticized. Assuredly it is not the whole truth. Nor, of course, does it apply to all the German art historians who found asylum in Britain or the United States. But for those refugee academics in the field of art history who do in general fit the category loosely defined in   See Erwin Panofsky, ‘Three Decades of Art History in the United States: Impressions of a Transplanted European’, in Meaning in the Visual Arts (New York, 1955), p. 332. I am grateful to Katya Weil-Garris Brandt for an illuminating conversation on this whole issue. 11   Notably ‘The German Refugee’, in The Stories of Bernard Malamud (London, 1984), pp. 93–108. 10

104

the preceding paragraphs – and some of the scholars already mentioned, irrespective of race, do not fit it and had a distinctively different personal history – the effect was that now they thought not just intelligently but profoundly, as it were from within. This was a victory for the human spirit. It gave them insights into human nature denied to those who had lived safer, softer lives, and had merely their intelligence and their learning to offer. Personal tragedy, in other words, transformed their scholarship and gave it a heightened awareness. There was another element at work too: the switch from German to English. Those non-Germans who have to read academic German will know that its difficulties are not those of vocabulary but of syntax, and that they often obscure rather than reveal the writer’s meaning. The issue is much more than the change of gear involved in English when moving from conversation to lecture, or even from the spoken to the written word. Some heightening of tone is inevitable in this process. In German, the change is quite radical. In the hands of too many German academics, especially in the early twentieth century, sentences lengthen to almost comic proportions. They trail subordinate and adjectival clauses like so many railway carriages attached to the parent locomotive. The prose is clotted and opaque. First it slows you down. Then it swallows you alive. Mark Twain has extracted the full humorous potential of that 12 process. Yet this kind of language was de vigueur in German scholarship in the 1930s, and even now it is not entirely dead. To write in a much lighter and transparent way was not a serious option. It was not respectable. Of course one can be pretentious and obscure in any language; but it certainly does not help if the language itself – or rather one of its sub-sets – seems to encourage this. Personally, I doubt whether some of the great German refugee scholars would have become the towering international figures that they did if they had continued to write in German. As the sociologist Marshall McLuhan has taught us, the medium is the message. Those of us who know what it is like to think and write in two languages will be aware that a language imposes certain patterns of thought and expression. In another language the patterns are different. In the case under discussion it is striking to 12  See ‘The Awful German Language’, in A Tramp Abroad (The Writings of Mark Twain. Definitive Edition. Volume X [New York, 1923]), Appendix D, pp. 267–84; and, for a selection of his views on this issue (which he could not leave alone), see Everyone’s Mark Twain, compiled by Caroline T. Harnsberger (Cranbury, NJ, and London, 1972), pp. 202–11.

ETTINGHAUSEN AND ICONOGRAPHY

105

note that once these men had discarded the baroque superstructure of academic German they developed (with help) a good plain English style. In the process, of course, they acquired an audience. In Ettinghausen’s 13 case one has only to compare his German article of 1934 on the Ka‘ba 14 with his English article of 1935 on Seljuq Qur’ans. These remarks should help to explain that for such refugee scholars the act of writing in English, however difficult and frustrating it was for them, 15 was also a liberating experience. And it had wider ramifications than this, for it was the high road to assimilation. They were, after all, outsiders – foreigners whose tongues gave them away with every word that they uttered. So they were driven by quite disparate motives: the urge to belong, the desire to forget the past in work, the need to make up for lost time, the compulsion to escape from the threatening world outside, the sense of intellectual freedom which they experienced in America. Whatever the motives, their productivity was astonishing, and Ettinghausen was no exception. Between 1934 and 1939, he published 30 titles on topics ranging from Qur’ans to the iconography of shadow figures, from dated 16 tiles to glass weights, from Egyptian woodwork to Persian painting. His contributions to Arthur Upham Pope and Phyllis Ackerman’s A Survey of 17 Persian Art (1938-9), taken together, are of book length. Almost all this work is in English – no mean achievement for a man uprooted from his personal and academic environment and forced to fend for himself in a foreign land and in a foreign tongue, even if, in his particular case, he already had some facility with English. And then it stops. With the outbreak of the Second World War there began for him not just the biblical seven lean years but a lean decade. He spent its early years in Michigan, but it is likely that his mind and heart were elsewhere. He published only two articles of real consequence, one 13   Richard Ettinghausen, ‘Die bildliche Darstellung der Ka‘ba im islamischen Kulturkreis’, Zeitschrift der Deutschen Morgenländischen Gesellschaft LXXXVII (1934), pp. 111–37. 14  Richard Ettinghausen, ‘A Signed and Dated Seljuq Qur’an’, Bulletin of the American Institute for Persian Art and Archaeology IV/2 (1935), pp. 92–102. 15   C.f. Panofsky, op. cit., pp. 321–46, especially pp. 329–30. 16   For a bibliography of Ettinghausen’s writings, see June H. Taboroff, ‘Bibliography of the Writings of Richard Ettinghausen’, in Peter J. Chelkowski (ed.), Studies in Art and Literature of the Near East in Honor of Richard Ettinghausen (New York, 1974), pp. 5–25. 17   Ettinghausen first began to work for the Survey of Persian Art in London in 1934, and married Arthur Upham Pope’s secretary, Basilia Gruliow. I am grateful to Stephen Vernoit for this information.

106

on Fatimid painting, ‘Painting in the Fatimid Period: A Reconstruction’ (1942), the other a classic exploration of the Bobrinski ‘bucket’, ‘The 18 Bobrinski “Kettle”: Patron and Style of an Islamic Bronze’ (1943). It is grimly appropriate that this latter article, which launched him in the new direction of iconography, failed to have the impact that it deserved in America because the ship which was transporting that issue of the Gazette des Beaux-Arts across the Atlantic was sunk by a submarine. It is hard to explain this ten-year lag in productivity, which – it is important to remember – was both preceded and succeeded by intense activity in published research, in professional terms alone. This was, after all, a scholar who was sold out to research and publication, who was in his prime and who had a safe, relatively untaxing job as a university teacher followed by a curatorial post in a museum. These should have been his most productive years. 19 Instead, they were the opposite. The lengthy gap in his output therefore seems much more likely to reflect his personal situation. Since this period coincides exactly with the war years and their immediate aftermath, that is the obvious place to look. One thinks immediately of the loss of his first wife in the late 1930s. I remember in 1979 talking to Ettinghausen’s fellow Islamicist and friend Shlomo Goitein, then nearly 80, about these years, and he recalled that Ettinghausen himself did not talk about them, though Goitein well knew that his friend had lost loved ones. He had a vivid memory of meeting Ettinghausen in 1946 or so, and being deeply struck by how ‘his lovely face’, as Goitein termed it, was transfigured by grief. I simply have no information about how Ettinghausen coped with his bereavements. But what is plain from the published record is that he only very gradually regained the impressive productivity of his early thirties, though now at a markedly higher intellectual level. As noted earlier, between 1933 and 1939 his publications – reviews and obituaries excluded – totalled 30; between 1940 and 1949, the corresponding number is ten. In other words, he published roughly one-fifth as much per year in the 1940s as he had in the middle and later 1930s. The turn-around was signalled 20 by the publication of his monograph The Unicorn (1950). His reputation depends principally on what he published thereafter.   See note 4 above.   Though this was also the time that he made a serious effort to learn Persian,

18 19

much as Jean Sauvaget at more or less the same period turned his attention to Turkish. I am grateful to Oleg Grabar for this parallel. 20   Richard Ettinghausen, The Unicorn (‘Studies in Muslim Iconography, 1’, Freer Gallery of Art Occasional Papers, 1, 3) (Washington, DC, 1950).

ETTINGHAUSEN AND ICONOGRAPHY

107

And so to the last topic for discussion. It will be clear by now that there is no intention here to present Ettinghausen as merely one member of an undifferentiated mass of refugee German scholars; but through that prism at least some aspects of his achievements may in part be understood. Moreover, there operated between these scholars an informal freemasonry far stronger than the normal bonds of collegiality. This is not to suggest that Ettinghausen was a team player, but merely to highlight that the shared educational, cultural and linguistic background of these refugee scholars gave them a ready source of intellectual companionship and professional expertise. And if there was one thing that bound these émigré German art historians together, it was an interest not in connoisseurship, not in the recording of objects, not in problems of dating or provenance, but in the meaning of works of art, that is in iconography. Here, Western art history had a flying start, and it was left to Ettinghausen virtually to create this subject in the study of Islamic art. He certainly had precious little foundation to build on in the United States; the many articles on Islamic art in the bulletins of the various American museums before 1940 are remarkable for the arid and meagre quality of their discourse. How did Ettinghausen operate? Although a great many students passed through his hands over the years, they do not seem to have been a vital element in Ettinghausen’s life as a scholar. Few completed a doctorate under his care. One has the impression that he taught students rather than that students taught him. They were not an integral part of his modus operandi as a scholar. It seems instead that in the first instance he drew inspiration not from freewheeling discussions with his students but from the objects with which he communed so closely. The close, rigorous analysis to which he subjected these objects allowed him to distinguish the elements that mattered from those that did not, and thus to seize the essence of the work of art that he was studying. That may sound rather obvious, but, in fact, very few people, art historians included, have both the X-ray eyes and the intellectual discrimination to use them creatively. He then contextualized these objects with all manner of supplementary information drawn from Islamic religion, history, literature and culture. Here was the orientalist at work. But all this was merely the raw material for the article itself. Ettinghausen’s particular skill lay in the arrangement of the raw material. It was not his way to bludgeon the reader by sheer weight of data. He preferred to create an argument. Typically, he would begin by identifying a puzzle. Often this puzzle related to material that was, at one level, very well known:

108

illustrations and descriptions had been published, the approximate date and likely provenance had been established. And there earlier scholarship had left it. No further questions had been asked. And then Ettinghausen would draw attention to some hitherto overlooked aspect of the piece and invite you to study it, trailing before you a series of interconnected questions or comments. You feel as if you have been taken into his confidence. At this stage there begins a treasure hunt, with clues artfully scattered along the route. And this is exciting. For in the manner of his presentation, Ettinghausen leads you to expect that he has the answer up his sleeve. A voyage of discovery is under way, but the pilot knows the destination. The method is to cast light on the problem by viewing it from several different angles, and its effect is to reveal the solution very gradually. The evidence is presented slowly, step by step. It is essentially a version of the Socratic dialogue; the reader is fully involved in the quest and has to work actively, not merely absorb passively and wait for the answer to be presented on a plate. As with a Socratic dialogue, a rigorous logic controls but does not dampen the speculative activity of the author’s intellect. Part of the method is to render unbreakable each separate link in the chain of reasoning. Understandably, then, it calls for a powerful effort of the will to challenge the cumulative weight of evidence so persuasively and rationally presented. Once that effort has been made, it is equally difficult to find any chink in the reasoning – witness his account 21 of the Freer lustre plate, which seems continually to teeter on the brink 22 of over-interpretation but never quite topple over. Not for Ettinghausen the wild leap into the unknown. Not for him either, as noted earlier, the half-baked speculation. His speculations were always fully baked; one feels 21   Richard Ettinghausen, ‘The Iconography of a Kāshān Luster Plate’ (with Grace D. Guest), Ars Orientalis IV (1961), pp. 25–64. 22   Ettinghausen’s study of Khirbat al-Mafjar is perhaps the boldest of all his ventures into iconography, and while most of the discussion is at once pioneering and plausible, his handling of the architectural material seems somehow to miscarry. Parallels that look convincing enough when only ground plans are studied will not hold once absolute size and the third dimension are brought into play (‘The Throne and Banquet Hall of Khirbat al-Mafjar’, in From Byzantium to Sasanian Iran and the Islamic World: Three Modes of Artistic Influence [Leiden, 1972], pp. 17–65). This is perhaps also the place to note that Ettinghausen consistently avoided writing about architecture; apart from the piece on Khirbat al-Mafjar, his best attempt in this field was a review article of Oleg Grabar’s chapter ‘The Visual Arts, 1050–1350’, in J.A. Boyle (ed.), The Cambridge History of Iran. V, The Saljuq and Mongol Periods (London, 1968), pp. 626–58 (‘Some Comments on Medieval Iranian Art’, Artibus Asiae XXXI [1969], pp. 276–300).

ETTINGHAUSEN AND ICONOGRAPHY

109

he would not have wished to present material for which he did not have an explanation that satisfied him. Wrong turns are meticulously documented; less likely solutions are presented in detail, only to be discarded. You do not feel, therefore, that inconvenient evidence has been suppressed: it has been confronted and dealt with honestly. So you trust the result. The method also makes allowances for not arriving at a single clear-cut answer but for presenting a range of possibilities, or – more accurately – a field of circumstances within which the answer is likely to lie. His articles on the 23 24 ‘Wade Cup’ or the lustre ware of Spain are examples of this technique. And the range of supplementary documentation, both literary and visual, that he brings to bear on such key pieces as the Boston ‘hunting carpet’ 25 transforms the way we look at them. He even makes us believe that we half-knew all this already. When one surveys Ettinghausen’s published output, the most striking aspect of it is the absence of books. He wrote only two full-sized monographs, and they could scarcely be more different from each other. One was The Unicorn, an extremely detailed iconographic study, still the only one to have been published on the significance of a single animal in Islamic art, though there are, of course, shorter studies galore. The other 26 was Arab Painting (1962), surely his masterpiece, a book that is still, almost 40 years after its publication, miraculously up-to-date. At one level, it is a mature summation of decades of research and thinking on the subject. At another, it charts, with sovereign assurance, a huge field of Islamic art in its entirety, and does it in such a way that the beginner as well as the specialist can read it with profit. It is a pity that this gift of synthesis, of writing a survey in a challenging manner, was not exercised more in his career. His contribution to The Art and Architecture of Islam, 650–1250 (with Oleg Grabar, 1987) in the Pelican History of Art series, was in the works for 20 years or more, and death prevented him from putting the finishing 23   Richard Ettinghausen, ‘The “Wade Cup” in The Cleveland Museum of Art, its Origin and Decoration’, Ars Orientalis, II (1957), pp. 327–66; idem, ‘Further Comments on the Wade Cup’, Ars Orientalis, III (1959), pp. 197–200. Ettinghausen discussed his ‘discovery’ of this piece in a film on the collecting of Iranian art made by Oleg Grabar. I am grateful to the latter for this information. 24   Richard Ettinghausen, ‘Notes on the Lusterware of Spain’, Ars Orientalis I (1954), pp. 133–56. 25  Richard Ettinghausen, ‘The Boston Hunting Carpet in Historical Perspective’, Boston Museum Bulletin, LXIX (1971), pp. 70–81. 26  See, however, the review by Ernst J. Grube in Art Bulletin XLVII (1965), pp. 375–6. I am grateful to Stephen Vernoit for this reference.

110

touches to it. The other books he wrote were either brief introductions or, in the case of his last book – From Byzantium to Sasanian Iran and the Islamic World (1972) – essentially discrete articles published in book form. It is no accident that he avoided the book as a means of expression. His instinct was rather to engage with specific problems in the context of specific objects; and that pointed to the article rather than the book. It is in the article form that he could best investigate in appropriate detail the meaning of a given object. Thus his articles are his principal legacy to the scholarship of Islamic art. They document his precocious recognition that there was such a thing as Islamic iconography. Previously, this was the subject that dared not tell its name. His exploration of this field exposed the threadbare quality of the traditional taxonomic and aesthetic approaches. It is a tribute to his prescience that no other field of Islamic art has grown so rapidly in the last 50 years as that of iconography. And so it is that the key to his scholarly achievement, and to its cumulative power, 27 is the mammoth 1300-page volume of his Collected Papers (1984). It is a fitting memorial to a man whose like we shall not see again.

  Cited in full in note 1 above. All of Ettinghausen’s articles mentioned in the notes to the present paper are reprinted in that volume. 27

VI Figural Calligraphy in the Muslim World fashioned by Jila Peacock from Persian

tpoetry fall naturally into a long and distinguished tradition within he images of living creatures

Islamic art. Unlike many such traditions of that art, this one is still full of vitality in our own day, and has shown an impressive capacity to regenerate itself. Indeed, it could well be argued that the art of beautiful writing — for that is what ‘calligraphy’ means — has, alone of the major Islamic visual arts, continued its creative evolution without a break from the first Islamic century until the present day. In other words, it has been less subject to failing inspiration or to the dominance of ideas from outside the Islamic world than have all its sister arts, from architecture to painting, from pottery to carpets. If any one art can claim to evoke the essential character of the Islamic world, in medieval as in modern times, this is it. It is worth pondering why this should be so. Part of the answer must lie in the continuously high regard in which calligraphy has always been held throughout the Islamic world. Its traditional association with the Qur’an has conferred on it a particular reverence. God, ‘the Supreme Pen’, first taught man to write. And Islam has always seen itself as a religion of the Word; indeed, the Word is its essential icon. This nexus of ideas helps to explain why calligraphers to this day ensure that when they copy the Qur’an they are in a state of ritual purity. As the saying goes, ‘Purity of writing is purity of soul’. No wonder that the copying of the sacred text has always been regarded as meritorious. Moreover, the need to ensure absolute accuracy in the transmission of that sacred text — and here the copying of the Torah by Jewish scribes offers an illuminating parallel — puts a premium on clarity and thus on the precise execution of the letters.

112

Once that degree of care has been learned, its effects naturally make themselves felt in the writing of other, non-sacred texts as well. But there are practical reasons as well as religious ones which help to explain why calligraphy should hold such a special place in the hearts of Muslims. It is an art open to all. Its presence is pervasive in the Islamic world to this day; the quality of signage in an ordinary street in a commercial quarter, whether in Morocco or Pakistan, is much higher than in a comparable street in a Western city. In the visual arts, inscriptions can be found everywhere, and they illustrate an endless range of variations. People who can write well are widely admired, and the basis of expression for much modern art in practically all Muslim societies is writing. Comparisons are odious, and there is no need to attempt to establish a pecking order among the world’s scripts. But even to an untutored eye, the extreme flexibility of the Arabic script — which with only very minor changes was adopted for writing such other ‘Islamic’ languages as Persian, Turkish and Urdu — is obvious at first glance. On closer inspection, numerous other intrinsic features of this alphabet make themselves felt. They include its constantly shifting interplay between angular and curved letters, its capacity to handle with equal facility compression and prolongation alike, the visual unity that is a by-product of the absence of a majuscule (capital letter), and the way it seems to lend itself with equal ease to a very large and a very small scale, or can accommodate a sudden shift from the baseline to the upper register. Nor should one forget such apparently innate characteristics as dynamism, energy and rhythm. All this makes such calligraphy a wonderfully responsive instrument for aesthetic expression. It has a chameleon-like quality of adapting itself to all manner of purposes and moods. What follows from that flexibility? The malleability of the constituent letters of a word can allow the calligrapher to employ such devices as symmetry, echo, antithesis and the like, and even to make words look like each other although in any modern typeface they would look distinctively different. It is here that the distinctive curvilinearity of the Arabic script — so different from the angularity of Roman or the squareness of Hebrew lettering, or for that matter the pictographic density of Chinese characters — comes into its own. It somehow allows the calligrapher access to another dimension; it is liberating. The letters can develop in virtually any direction. In many scripts they acquire fanciful terminations, whether geometric, floral or zoomorphic. Or the centre of the letter will develop an ornamental life of its own. No manual attempted to

FIGURAL CALLIGRAPHY IN THE MUSLIM WORLD

113

control or systematise these idiosyncrasies. Even Victorian copperplate — the closest that Western handwriting in standard use (as distinct from professional calligraphy employing a variety of italic hands) comes to that curvilinearity — is far more constrained by the rule-book than the Arabic script. Moreover, formal inscriptions in the West are almost uniformly executed in Roman (or Greek) capitals, and the conventions that apply in that context are extremely restrictive in comparison with those that operate in formal contexts in Islamic inscriptions. Of course this does not mean that there are no rules which underpin Islamic calligraphy. But there is no need to engage in deep study of this art to realise that, while there are indeed rules to follow, these rules sit very lightly on the masters of the craft. We are told, for example, that the celebrated calligrapher Ibn al-Bawwab followed the standardisation of scripts introduced by his predecessor Ibn Muqlah a generation earlier in the tenth century. A key feature of that attempt at uniformity was the system whereby each letter was composed of a certain number of dots. Thus there was in theory an ‘ideal’ way to form each letter, almost as if the calligrapher were turning himself into a typewriter. But one has only to study closely a page of the Qur’an in the Chester Beatty Library, Dublin, dated 397/1000, which is signed by Ibn al-Bawwab, to note that he can write the same letter (e.g. kaf) in as many as six different ways in a short passage. That variety is not only what distinguishes a page of his work from that of a typewriter, but also what gives it life. This small example shows how profoundly unreliable the manuals on calligraphy can be. They can plainly be contradicted by the evidence of the work itself. This is also the clearest proof that mastery can as well express itself in creative disobedience to the rules as in following them to the letter. As it happens, the calligraphic tradition honours both kinds of masters, though to judge by the histories of calligraphy it is those who can best copy their predecessors who form the great majority. As in classic Persian painting, one mark of having reached the status of master was the ability to copy closely the great work of the past. Hence the genealogical nature of so many of the accounts of the great calligraphers. The citation of the past masters in the same style is an integral part of describing the work of a given great calligrapher. What we are never told in such frankly formulaic and adulatory accounts is precisely how the work of one master differed from that, say, of his immediate predecessors in that particular style. At any rate, the relationship to earlier masters is the principal element of the calligraphy that is recorded for posterity — not whether that person

114

devised a new way of constructing a given letter or ligature, or introduced a new layout for the page. It is almost as if the reader is expected to know that kind of thing without any help. This situation highlights a perennial difficulty in how calligraphy is assessed within the Islamic tradition: the lack of a useful and generally understandable vocabulary of criticism. What exactly is it that constitutes a weak, average or superlative hand? The great calligraphers themselves have left very little in the way of clues. This is not surprising, for one very rarely finds in the same person the capacities of the creative artist and the discriminating critic. Those who produce masterpieces are seldom able to explain how they did it. But there is also a strong tradition of masters deliberately concealing the secrets of their craft, for example hiding from sight the way that they cut their reed pens. In part this conforms to the custom of passing on craft techniques from master to pupil rather than putting them into the public domain. Whatever the reason, the result is plain: there is no easy guide to what makes great calligraphy. Inductive reasoning is required. One innate characteristic of the flexibility noted above is that the letters can suggest something beyond their verbal meaning. This further significance may be purely abstract, for example if it creates some kind of pattern, whether through the design and placing of individual letters or of blocks of text, which themselves might be so arranged as to create further texts; but it can also enrich the meaning of the words themselves, or set up quite other meanings. An example from the Ibn al-Bawwab manuscript just mentioned is the preternaturally elongated ba in the bi’sm of bi’smillah (‘in the name of God’) placed at the opening of each sura (chapter). This serves as a visual marker, like punctuation or underlining. It indicates the beginning of a new part of the text. But it has an aural quality too, as if it denoted a long-drawn-out and possibly louder sound. This is the closest that Islamic calligraphy gets to the historiated initial letter so beloved of the tradition of medieval Western illumination. The delight in constantly expanding the boundaries of expression explains why so many contrived scripts were invented in the course of the centuries — scripts named after the full moon or the crescent moon, after dust or the locks of the bride, after trembling or after flowers or peacocks. But perhaps the most radical experiments with calligraphy were those which involved figural designs. The earliest attempts to do this can be seen in the inlaid metalwork made in the eastern Iranian world in the twelfth century. Perhaps the first major example of this trend is to be

FIGURAL CALLIGRAPHY IN THE MUSLIM WORLD

115

found on the so-called Bobrinski bucket made in 1163 in Herat (now in the Hermitage, St Petersburg). This provenance is significant. Herat is sufficiently distant from Armenia, the easternmost part of the Christian world to experiment (as did other Christian traditions) with the notion of making letters of the alphabet come to life with figural or zoomorphic elements, to make it likely that this was an independent local invention. Indeed, it was the Iranian world that was destined to produce the richest variety of work in this sophisticated subset of calligraphic invention. The beginnings of this mode trend in Herat are in Herat are modest, but they already display two distinct modes. The simpler of the two (significantly placed at the base of the bucket’s curvature, not a prime location, and interwoven with a frieze of running animals) involves incising faces onto the thickened upper shafts of the tallest letters. These faces could hardly be simpler: an eyebrow, an eye and a mouth, with a neck below it. But in a larger band, advantageously placed at the top of the bucket’s curvature and thus the natural first focus of attention, the artist takes advantage of the extra space available to animate the letters still further. Now, the shafts of the taller letters have morphed into legless bodies, some upright, some leaning forwards or backwards, some bearded, some clean-shaven, some with long kiss-curls or pointed hats. They interact with unmistakable humour, chucking each other under the chin, patting each other on the shoulder or tweaking each other’s ears, shaking hands, holding beakers or jugs, and stirring bowls. Thus the upper storey of the inscription is shot through with narrative while its ground floor spells out a message of good wishes and happy life. In this way the figural elements tacked on to the letters themselves not only enrich but also reinforce the content of the inscription. This fashion of animated inscriptions lasted for the best part of a century and a half. It ranged from shafts ending in a head, giving the visual effect of a long row of circles at the very top of the band, to tightly packed seated figures gesticulating vigorously, and finally the Arabic letters themselves either give birth to, or are integrated almost seamlessly with archers, spearmen and swordsmen in combat, revellers, and dancers performing to the music of tambourine, harp and wind instruments. All this occurs in the context of letters ending in the heads of dogs, hares, bears, harpies and birds, as in the famous Wade Cup (in The Cleveland Museum of Art). The next experiment in bringing living creatures into the ambit of calligraphy was even more radical, and it seems also to have originated in the Iranian world. Here we meet the ancestors of modern masters like

116

Ahmad Mustafa and Jila Peacock. This new mode was not a matter of script metamorphosing into living forms which are also readable letters, but of using script to delineate such forms. Seldom had the flexibility of the Arabic alphabet been so tested. This practice established itself only relatively late in Islamic art, when the taboos outlawing religious iconography had lost some of their power. Indeed, it developed in the very same areas — Ottoman Turkey, India and Qajar Iran — where iconographic cycles of complex narrative scope involving the lives of Muhammad, ‘Ali and certain Islamic saints gradually won a degree of popularity. Clearly, then, the practice is related to a loosening of earlier restraints. When did this practice begin? According to a treatise on calligraphy written by Qadi Ahmad around 1606, it was once again in the city of Herat that this invention was made. The alleged inventor, Maulana Mahmud Chapnivis, ‘wrote the hemistich: “The price of sugar and candy has come down because of the lips of the beloved,” on two sides [in mirror writing?], in the shape of three or four men standing one under the other, and both the figures and the writing were executed with perfect skill and charm’. No date is given, but the context suggests the reign of Shah Tahmasp (1524–76). This all sounds convincing enough, but once again it is contradicted by the facts, which show that zoomorphic calligraphy was known at least as early as 1458, when a certain ‘Ata Allah b. Muhammad al-Tabrizi produced a scroll (now in the Topkapi Saray in Istanbul) which included this motif. Perhaps the most celebrated of the early attempts in this mode is a probably sixteenth-century pacing open-mouthed lion, largely created in gold both within the body and in outline by a prayer to ‘Ali, the Prophet’s cousin and son-in-law, known as Nad ‘Ali. Since ‘Ali was known as Haidar, ‘lion’, the design is also a visual pun. The piece is signed by the scribe ‘Ali who ‘cut it out’. The calligraphy fills the body only loosely, and scattered white and red blossoms, sprays and tendrils, as well as a continuous lapisblue ground, serve to fill the many gaps. Thus script and decoration counterbalance each other, and to some extent are even at cross purposes; this is by no means a purely calligraphic exercise. This lavish use of infill, and the sudden thinning of the outline at breaks of sense, suggests that the conventions of this kind of calligraphic tour de force were not yet fully established. In particular, the emphasis on outline, which involved many awkward curves, points — despite the calligrapher’s obvious skill — to a certain unease with this form of expression. Perhaps the clearest sign of

FIGURAL CALLIGRAPHY IN THE MUSLIM WORLD

117

this is the fact that the tail is only feebly incorporated into the text of the prayer; indeed, the tail remained a problem in many subsequent renderings of calligraphic lions. Later essays in this manner, whether of tigers, parrots, ostriches, storks or cockerels, though often constrained to add such necessary details as feet, tail or crest without recourse to writing, tended to minimise if not exclude added ornament, and indeed polychromy, so that the calligraphy itself reigned supreme. Gradually calligraphers learned to use outline better, for example in the depiction of fruit or the conical hat of the poet Maulana Jalal al-Din Rumi. The calligraphy of such designs, whether of living beings or other subjects, was usually of pious content, even if (as in the case of a human face created entirely by writing) this consisted entirely of venerated names such as Allah, Muhammad, ‘Ali and his two sons Hasan and Husain. This goes also for the many essays involving cursive hands, and also rectilinear Kufic, to create images of mosques, boats (the ship of salvation and Noah’s Ark), helmets, lamps and vases. Very often the image thus created has its own symbolic associations which are independent of the writing of which it is formed, though that writing may further deepen the meaning of the image. Thus a mihrab may be formed of a quotation from the Qur’an attacking unbelievers, or a multi-domed mosque complete with minarets may appropriately be constructed out of the words of the shahada, the Muslim profession of faith. In a still more ambitious design, a huge horse composed of the Throne Verse (Qur’an 2:255) carries a dignitary whose tiny scale underlines both the pride of man — perhaps also symbolised, in yet another pun, by his throne-like saddle — and his insignificance before God, as represented here by His Word. Given the complexity of these ‘word-pictures’ and the clearly symbolic use to which they put the Arabic script, it is not surprising that magic powers were often ascribed to them and that they served as amulets. In such images one encounters an out-and-out violation of Islamic norms, in which blasphemy is added to injury because it is so often the very words of the Qur’an which are tortured into forming the previously hated images. In his defence against such an accusation the scribe would no doubt argue that to create forms by means of actual letters would so distance those forms from reality that no-one could reasonably believe that he was trying to breathe life into them. Even when the texts are not Qur’anic, their normal content is prayers or invocations and benedictions, rather than secular matter. It is this detail which clinches the religious intent of such

118

images, for at the time that they began to be made figural art had already been established in the Islamic world for a millennium. There would have been no difficulty in forming images from secular as distinct from religious texts, and the contemporary fashion for creating images out of smaller figures or objects in the style of Arcimboldo (the Renaissance painter famed for his fantastical portraits made up of vegetables and fruit) shows that the idea had already long been in the air at the time. The choice of religious texts for these figural images, then, betrays the desire both to flout and to circumvent the orthodox disapproval of religious images. In many of these calligraphic images, one senses that writing is being pushed to its furthest limits so as to make it express unnaturally what it cannot do naturally. With their elaborate mirror-writing and the flavour of secrecy and cipher which pervades them, these arcane images offer striking evidence as to the slightly perverse outlets into which Islamic artists channelled their frustrated desires to create religious pictures. It was a current which proved too strong for Islamic orthodoxy to dam, but that same orthodoxy compelled it to take a remarkably roundabout route before it found the expression it craved. This, then, is the wider context of the images that Jila Peacock has created out of the poems of Hafez. Where, one might ask, does she belong within this tradition? She differs from most of her predecessors on many points: her choice of text; her use of nasta‘liq rather than some other cursive script; the density of calligraphic mass which makes up her images; the sheer range of creatures she depicts; and the way she glories in that dimension of colour which most practitioners in this specialised field discarded. These features bear further discussion. The texts are taken from the Divan of Hafez (d. 1390), Iran’s premier and most quoted lyric poet, whose status in his own country can be compared with that of Shakespeare in the English-speaking world. As the saying goes, every Persian home has a Qur’an and a Hafez. His magical use of language, his towering spirituality, the passion with which he explores the seemingly inexhaustible themes of love and wine, his vehement satire and invective, the mysticism which suffuses so many of his lyrics, and the subtlety and allusiveness which permit so many contradictory interpretations depending on the reader’s mood, have endeared him to generations of Persian speakers from Turkey and Iraq to India and Central Asia. Not for nothing is his tomb in Shiraz to this day a magnet for lovers and for connoisseurs of poetry. He is, moreover, popularly regarded as a seer and soothsayer, and his lyrics are often used even now for purposes

FIGURAL CALLIGRAPHY IN THE MUSLIM WORLD

119

of divination, much in the manner of an astrological column in a modern newspaper. Thus the text of Hafez, while not holy writ like the Qur’an, is — like that of no other writer of poetry or prose — part of the warp and weft of Persian life. What follows from the use of nasta‘liq for these images? To begin with, this is a ductus not often used for calligraphic pictures. This type of script, above all other Islamic hands, encourages the most complex rhythms and asymmetries, with its natural tendency towards the layering or tiering of words and phrases, its swooping lines, its constant switching from razor-thin to fat, thick strokes, its full-bottomed curves and its powerful momentum — the very image of rapid thought. These characteristics allow the artist to vary the nature of the outline not only from one creature to the next but also within a given image. Thus in the case of the fish (whose outline itself recalls a flourish of nasta‘liq, for example a terminal letter nun) only a very few letter forms suffice to suggest the upward curve of its back, the flick of its tail and the jut of a fin. This function of nasta‘liq suggests an affiliation to the changing thickness of line in superlative draughtsmanship. Traditionally, most calligraphic pictures rely heavily on outline, often of a very contrived kind, in order to establish the shape of the object which is being portrayed. That in turn demotes the writing inside the outline to infill. Jila Peacock’s technique does not privilege any part of the text in that way. It builds up the form of the creature as if techniques of modelling were being employed. Thus sometimes the letters and words are densely packed, while at other times the arrangement is looser, though never so loose as to break down into incoherence or to leave awkward gaps. It is a delight to admire the high-stepping canter of the horse, the proud sweep of the peacock’s tail or the branching antlers of the stag, details which are perfectly captured in this technique. Some scholars argue that there is a subtle but almost invisible unity that links each hemistich to the next in the ghazals of Hafez; others prefer to see them as ‘Orient pearls at random strung’. But in either case, the choice of creature for illustration picks out one element — often the key image of the poem — for special attention. When the image and the poem from which it is taken are considered together, and the reader meditates on the connection, the words of Hafez come alive, often in an unexpected and gripping way. ‘Neither noose nor net shall capture the bird of wisdom’, for instance, seems a peculiarly apposite caption to the image of a bird whose very body is made up of wise words. Since Jila Peacock has herself

120

translated the text which she has fashioned into these shapes, she is uniquely placed to plumb the subtleties of the poet’s diction, and to bring word and image together. The poem which begins ‘O Heaven show me your redemptive fire’ contains the stanza: O fellowship of that blissful flame, Pray call on God to tell me Whose is the burning butterfly, whose? The insistent questioning of this poem, its anxious uncertainties, its intensity, its repeated images of fire, find apt expression in the tremulous fluttering of the outspread wings of the butterfly. It is the very image of mutability. It is here that the very lack of firm outline in these pictures comes into its own, for the continuously broken silhouette conveys the pulse of life itself. One last example must suffice. The couplet The falcons of the path repose content as flies, Such is the sweetness of this world. speaks of repose, not action; but a few lines later the mood changes abruptly: How many bells must ring To rouse you from your daze? Pity a bird like you Imprisoned in a cage. Spread your wings and sing From the Tree of Paradise. and it is this moment of dynamic energy that the artist has captured in her image of the hovering falcon with its wings beating powerfully and its talons outstretched. The rendering of each of the last three outer primaries by an outsize letter ya epitomises the inherent flexibility and expressive power of this nasta‘liq script. Finally, what of the dimension of colour? As suggested above, artists in earlier centuries who experimented with figural calligraphy tended to work in monochrome. But recent practice in this genre has eagerly embraced the potential of colour. In colour, as in other respects, the images in the original handprint book operate on more than one level. The use of different colours for successive images gives each of them its own distinctive character. But the sequence of strong colours also has a cumulative enriching effect on the whole book. And when an image is

FIGURAL CALLIGRAPHY IN THE MUSLIM WORLD

121

viewed at an oblique angle, it comes to life in a dramatically different way, seeming to leap off the page. The shimmering, reflective quality of the writing as it catches the light evokes the quintessentially Islamic medium of lustre pottery or glass, whose fitful iridescence suggests the glint of such precious materials as silver and gold. Thus the dimension of colour sets up a certain ambiguity whose character changes from one image to the next. This is because not only the background colour changes, but also the colour used for the script itself, which is by turns emerald, magenta, silver, violet, turquoise, vermilion and so on. One might catch an echo of the famed allusiveness and ambiguity of Hafez himself in this constantly shifting panorama. Jila Peacock, then, like a host of Islamic scribes before her, illustrates in our own day the untiring virtuosity of those who use the Arabic and Persian script as their preferred form of visual expression. Thanks to her choice of Hafez as a text, her innovative use of nasta‘liq, her sensitive handling of script as mass, her imaginative bestiary and her radical approach to colour, she has revealed new riches in this traditional genre.

VII Introduction: The Orient of the Imagination

O

rientalism has many faces. That is inevitable, for the reality which those faces reflect is complex and many-layered. The papers in this volume explore some facets of this cultural phenomenon in appropriate detail; the purpose of the present introduction is to outline a general context for them. Yet that in itself is a tall order. The material is so rich and varied, and infiltrates so many disparate disciplines, that it is beyond the competence of any one scholar to master it. By the same token, one has an embarras de choix so far as relevant examples are concerned – clear evidence of the enduring fascination which ‘the Orient’ (that romantic and conveniently vague term) has held for the West over many centuries. Perhaps the only way to make sense of this data overload is to isolate certain key themes and to focus the discussion on them. To some extent the choice of themes is arbitrary, but in an introduction of this kind any attempt to cover the whole field would be counter-productive. In the same spirit, the argument here will be confined to the Islamic East rather than extending to South-East Asia, China and Japan, areas which constitute a distinct subset in the field of Orientalism and where perhaps somewhat different criteria apply. The terms ‘Orientalism’ and ‘the East’ are accordingly used here to connote only the Islamic world, including North Africa. The discussion that follows, then, is structured around three core themes: exoticism (the principal focus here), authenticity and politics. Some overlap between these three themes is inevitable. The association between the East and the exotic in European culture is many centuries old. The literature of marvels which found expression

THE ORIENT OF THE IMAGINATION

123

in the medieval bestiaries and in the 14th-century Travels of Sir John de Mandeville was a potent cocktail of fact, fiction and fantasy. Such literature invited the belief that just over the horizon there lay wonders beyond imagination. Even as the age of discovery pushed back the borders of the unknown, cartographers still decorated the huge blanks that remained with images of beasts and phrases like ‘here be dragons’, just as medieval mapmakers had placed the many varieties of Monstrous Races and other fabulous creatures – monopods, manticores, mermaids, ‘anthropophagi, and men Whose heads do grow beneath their shoulders’ in the outer limits of the known world. It is thought-provoking to reflect that Marco Polo was long regarded as a fraud, for the world he depicted was so different from that authoritatively but fallaciously presented by de Mandeville. The trade in luxuries such as silks, enamelled glass, rock crystal and exotic animals – the caliph Harun al-Rashid sent Charlemagne an elephant called Abu’l-‘Abbas – fleshed out these fabulous accounts. Rulers of the 16th and 17th centuries, fired with curiosity in the widest sense, and blessed with the wealth to indulge that curiosity, built up the type of collection known as a ‘cabinet of curiosities’ or Wunderkammer, the ancestor of the modern museum. Perhaps the most valued Islamic artefact that made its way to Europe in substantial quantities was the carpet. Then, as now, it functioned not just as an item of furnishing but as a status symbol. As such it features prominently in many late medieval and Renaissance paintings, so much so that the painstakingly detailed depictions in such paintings have made it possible for carpet historians to reconstruct the later medieval history of many rug types, even though no actual specimens survive. In Europe these carpets, which came principally from Spain, Egypt, Turkey and Iran, were regarded as far too valuable to be used as ordinary floor coverings, though they do function as such in western religious paintings and royal portraits. It is far commoner to encounter paintings in which they are shown draped over tables, artfully placed so that the full splendour of their colour, texture and design could be appreciated to the full. The power of the East to liberate the imagination survived intact in the teeth of all the accurate information about these countries that was steadily accumulating from the time of Marco Polo (c. 1254-c. 1324) onwards. The Portuguese voyages of discovery in the 15th and 16th centuries brought generations of Europeans into direct contact with Islamic societies far from the previously familiar Mediterranean world, namely those of India and South-East Asia. In Portugal, the wonder and excitement of those

124

voyages can be sensed in Manueline architecture, as at Batalha, Setubal, Belem and Tomar. Indeed, on the west front of Tomar abbey motifs of cordage, seaweed, coral and rank, fantastical foliage inspired by the distant Indies are employed as framing devices for the windows. In the 17th and 18th centuries, an increasing number of Europeans travelled for years at a time in the Middle East and India, some of them, like Jean-Baptiste Tavernier, returning repeatedly and others, including Sir John Chardin, Sir Robert Sherley and Lady Mary Wortley Montagu, living there for lengthy periods. That experience involved a surrender of glamour and myth in exchange for down-to-earth reality; as Lady Mary noted, ‘the manners of mankind do not differ so widely as our voyage writers would make us believe’. Portraits of Sherley and Lady Mary in Persian and Ottoman costume respectively (see fig. 2, p.23) offer a striking contrast to the many likenesses of Portuguese grandees in Goa and elsewhere in Portugal’s colonial domains, who are depicted wearing the stiff, heavy European fashions so thoroughly unsuited to the South Asian climate. Was the decision to wear oriental clothing a matter of courtesy to the host country? A desire to preserve, as far as that was possible, a degree of anonymity there? A ploy to ensure greater success in the worlds of business and diplomacy? Or should one ascribe it to a certain narcissism, a pleasure in dressing up, or even the ambition to start a new fashion back home? It was in this same period that the idea of the Orient began to invade European literature. A benchmark in this process was the free translation of The Thousand and One Nights into French by Antoine Galland (published in 12 volumes from 1703-13), which generated a flood of translations into other European languages in the following century or so. It is significant that Galland judged it necessary to tone down the flavour of many of the stories in the interests of decorum and of observing the narrow canons of good taste in contemporary French literature. The Arab world that emerges from his pages is, like the Iran of Montesquieu’s Lettres persanes, unseasonably and unreasonably French in its polite circumlocutions and stilted dialogue. Gotthold Ephraim Lessing’s Nathan der Weise (1779), itself also widely translated, takes up a parallel tone characterised by distance and good breeding; it too is immovably anchored in the language and mores of its time and betrays not the slightest effort to enter the 12th-century world evoked by the action of the play. In defence of Lessing, it has to be admitted that a consistent sensitivity to anachronism has developed in European culture only in the last couple of centuries. Nevertheless, the

THE ORIENT OF THE IMAGINATION

125

theme of inter-confessional dispute between Christian, Muslim and Jew and the message of tolerance broadcast by the play marks an attempt to handle an issue which was as important, then as now, in the East as in the West. It would therefore not be entirely accurate to stigmatise Lessing’s use of the Orient for the framework of his morality tale as a misuse of the oriental world for occidental purposes. But the aftertaste of the exotic is unmistakable. A similar, and perhaps slightly unthinking, adoption of the Orient to create an exotic effect, lightly assumed and as lightly shrugged off, can be seen in some 18th-century music. Gluck’s opera The Pilgrims to Mecca is an example and the trend reaches maturity in the work of Mozart, at a time when Turquerie in Europe was at its height and when a reciprocal process can be traced in the Ottoman lands, with performances of chamber music in the Sultan’s harem. Mozart uses Turkish themes derived from the military music of the janissaries to dashing effect in his 5th Violin Concerto (K.219) and his Piano Sonata in A (K.331) and he also experimented with Turkish themes – which were very popular in the contemporary Austrian theatre in the aftermath of the Turkish defeat south-west of Vienna in 1683 – in his Singspiel later known as Zaide and subtitled Das Serail (K.344). But it was in his opera, Die Entführung aus dem Serail (1782; K.384) that he deployed the full panoply of exotic Turkish colour – sultan, slaves, harem, janissaries and the bowstring. It was an ambience which Rossini in turn exploited in his L’italiana in Algieri (1813) and which continued to inspire later composers. In a related vein one may cite orchestral tone poems such as Borodin’s In the Steppes of Central Asia (1880) or Rimsky-Korsakov’s Scheherazade (1888). Later still, Puccini took a story from the medieval Persian lyric poet Nizami for the plot of his opera Turandot. Some of the literature of the 19th century followed similar paths, from Sir Walter Scott’s novel The Talisman (1825) to Pushkin’s languorous ode The Fountain of Bakhchiseray (1822). The difference between them, however, is significant – Scott dreamed up his oriental locations while safely ensconced in Scotland, whereas Pushkin sought his inspiration on the spot. One might note the similar contrast between Flaubert, whose Salammbô (1862) is a creation of the study, unencumbered by precise topographical knowledge of the historical Carthage, and Gerard de Nerval, who wrote up his actual experiences in his Voyage en Orient (1851). Yet Flaubert, as it happens, did make a Middle-Eastern journey and Scott followed the Knights of St John as far as Malta, the setting for his last novel, The Siege of Malta. In Germany the massive prestige of Goethe ensured that his

126

explorations of Islamic poetry, collected together in his West-östlicher Divan and based entirely on translations, brought medieval Islamic material to a wider public – though they had to experience it through the filter of a poet who had never travelled to the Orient and knew none of its languages. In the course of the 19th century the trickle of European visitors to the Orient became a flood. A series of highly individual, often eccentric and obsessive British travellers – Alexander Kinglake, William Palgrave, the Blunts, Richard Burton, Charles Doughty and others – were drawn to the Middle East, and frequently to Arabia in particular, and it is due to them that the idea of ‘heart-beguiling Araby’ – essentially a construct of the imagination and scarcely more real than Shangri-La – took root in the British consclousness. That almost numinous idea inspired later generations of British men and women too, irrespective of whether their interest was a professional or an amateur one, and in the person of the film star Rudolph Valentino, alias ‘The Sheikh’, it invaded popular culture too. More seriously, the case of T. E. Lawrence (himself later ‘immortalised’ on film) shows what portentous political implications this heightened image of Arabia could have. It is in such personalities that the theme of the exotic shades over into that of authenticity. The figure of the 19thcentury polymath and traveller Richard Burton is crucial here, for however exotic his travels were in themselves, his accounts of them show how scientific enquiry – religious, philological, anthropological, sociological – was enriching if not replacing travel for thrills. Perhaps the most obvious transitional area in this respect is architecture. As far back as the 15th century, Venice – dubbed ‘the Jerusalem of Christendom’ by Thomas Coryate in the 17th century – had enthusiastically plagiarised ideas from the Islamic architecture of the Mediterranean, especially Egypt (whence the bones of St Mark himself had come). From polychromy to mashrabiyya work and roof-top terraces, from markets and fondaci to such details of vernacular architecture as balconies, oriel windows and shutters, and from grand buildings such as the Ca’ d’Oro (begun 1421) to humble alleyways, the city steadily looked east. But Venice at this period was still ahead of its time and Islamic architecture first had to be misunderstood before it could be copied with confidence. Impudently free variations on Islamic themes were created in the later 18th century by such architects as William Chambers (with a delightfully bijou Alhambra in Kew, and a mosque of Turkish type there too), J. H. Müntz (another Alhambra, also at Kew) and Henry Keene, whose mosque and tent designs (V&A) draw impartially from several

THE ORIENT OF THE IMAGINATION

127

Islamic sources. A far more considered attempt at Islamic colour can be seen at Schwetzingen near Heidelberg, where in the 1770s the French architect Nicholas de Pigage laid out for the Elector Karl Theodor a bath pavilion and a mosque, the latter complete with Qur’anic quotations and saws (‘Speech is silver, silence is gold’). The numerous Islamic touches included two minarets and an ambulatory of ogee bays screened off with gilded latticework and crowned by gilded crescent moons. Mughal architecture was evoked (somewhat ineptly) at Brühl; and elsewhere in Europe, the notion of the garden as a museum of world architecture, replete with Tatar tents, Chinese pagodas, Moorish pavilions, Turkish mosques and Egyptian obelisks, captured the public imagination. As early as 1651-4 the diplomat Adam Olearius, fresh from a lengthy official mission to Iran, designed a ‘Persian House’ in the park of Schloss Gottorf near Schleswig. The range of buildings in some kind of Islamic style grew ever wider as time passed, from a caravansarai in the Moorish style by Grohmann in the 1770s (a charming confection with upper gallery and four corner minarets, published in the Ideenmagazin) to a Turkish room in the Palais Beauharnais in Paris (c. 1805) and Krafft’s ‘Oriental Pavilion’ (1809-10), an octagon festooned with crescents and lozenges and crowned with an ogee roof. Ballrooms, Turkish temples (!), mosques and minarets flourished throughout Germany, Poland and the former Czechoslovakia in the 19th century. Across the Channel in the same period the emphasis was naturally rather different, given Britain’s increasing political and economic interest in India. Gifted draughtsmen like William Hodges and the Daniell brothers produced splendidly evocative images of Indian monuments which, while broadly accurate in architectural detail, downplayed the cramped, jostling and often squalid context of these buildings. The press of people is, so to speak, airbrushed out. The resultant sense of ample vistas projected by the acquatint engravings of Oriental Scenery (published by the Daniells between 1795 and 1808) sufficiently explains the comparable amplitude of S. J. Cockerell’s Indian-style country house at Sezincote (1803-8) and the architecture and landscape designs of Humphry Repton, Henry Holland, William Porden and John Nash at Brighton, notably the Royal Pavilion. This idiom continued for much of the century in modest domestic architecture, including Amon Henry Wild’s own house in Brighton (before 1830), baths, mausolea, railway stations and of course, stately homes galore.

128

Examples from the field of architecture also serve to introduce the second theme of authenticity. Here one name stands above all others. The magisterial work of Owen Jones on the Alhambra, the fruit of years of obsessively careful recording, represents a rejection of the type of fantasy architecture noted above, even though it was in turn responsible for a further series of ‘Moorish’ extravaganzas, from Matthew Digby Wyatt’s Moorish billiard room in Kensington Palace Gardens (1864) to the grandiose Wilhelma, near Stuttgart, the Moorish villa built by Ludwig Zanth for King Wilhelm of Württemberg (1855 onwards), or Tafel’s Moorish chamber in Schloss Castell near Tägerwilen (1891-2). Jones meticulously drew what he saw in an effort to be as authentic as possible, and his work has remained fundamental to modern scholarship on the Alhambra. It is ironic to reflect that his Plans, Elevations, Sections and Details of the Alhambra was published, in two volumes, in 1836-45, in other words at the very point when photography was poised to take over precisely this kind of task, though its colour plates were a landmark in the new chromolithography process. Buildings prinked out in Islamic detail continued to be erected all over Europe, including would-be Turkish, Moorish, Arab or Algerian cafés, Turkish baths (Stuttgart, Paris, Manchester, Leeds, London), steam installations (Potsdam), factories (Dresden and Vienna), kiosks (Linderhof), synagogues (Vienna, Berlin, Cologne, Wiesbaden), town houses (Frankfurt and Barcelona), and Arab halls (London and Cardiff), French villas and Petersen’s Tivoli complex in Copenhagen. But as Islamic architecture became better known, especially through the medium of photography, the surface detail of these buildings became steadily more accurate, so that a spirit of antiquarianism sometimes edged out untrammelled fantasy. These multiple experiments in an Islamic idiom – for which ample parallels could be cited in the United States – culminated in the numerous oriental structures erected in the Exposition Universelle in Paris in 1867 and its successors in Vienna (1873) and again in Paris (1878 and 1889). In Paris in 1889 an entire Cairo street (Rue du Caire) was set up illusionistically, complete with balconies, awnings, shops and a distant minaret. It was enlivened by actual Arab tradesmen and artisans, the smell of authentic Arab cooking and Arab dances. This idea was gratefully plagiarised with state-of-the-art technology in the Festival of Islam in London in 1976, when the Museum of Mankind put on a very similar exhibition evoking the Yemeni city of San‘a’.

THE ORIENT OF THE IMAGINATION

129

The desire for an authentic evocation of the Orient itself has a long history. A Moorish architectural vocabulary – rib vaults, mouldings, intersecting arches and two-tone masonry – invaded the religious buildings of Romanesque France. Carefully copied and entirely legible Islamic inscriptions with such Arabic messages as ‘Glory to our Lord the Sultan’ gave authenticity to Renaissance religious paintings by Gentile da Fabriano and others. The desire to create a credible Islamic context runs through many paintings by Carpaccio and reflects the frequent diplomatic and commercial exchanges between the Italian city states and the Mamluk and Ottoman polities. The anonymous Venetian artist who painted The Reception of the Ambassadors in the Louvre encapsulated contemporary western knowledge of and interest in Mamluk society. And for all its authentic detail, the choice of that detail – from spoked turbans and heraldic blazons to camels, monkeys and gazelles – betrays the perennial allure of the exotic. Contemporary and later painters such as Bellini, Veronese and Rembrandt continued, insofar as they were able, to use authentic Islamic detail, such as turbans, to create a credible context for subjects set in the Levant. But for the most part they were unable to draw upon their own personal observation of life in these lands and the almost pervasive indifference to anachronism, already mentioned above, resulted in Old Testament personages being decked out in contemporary Muslim dress. Authenticity took on a new guise from the early 19th century onwards, when a new generation of painters made their way to North Africa and the Near East. Some, like Delacroix, were in search of local colour and what might be termed pictorial inspiration, though the emergent French empire in North Africa gave a political tinge to their work whether or not they intended it. Many of the painters who visited the Holy Land, however (the usage is significant), were in search rather of a theological authenticity, a desire to recapture the atmosphere of the biblical narratives. Naturally they were drawn also to the sights and personalities that they encountered en route. In the cases of Wilkie and Lewis, for example, this search for local colour was a paramount factor. Others, like Holman Hunt, focused more narrowly on religious subject matter, and thus the blasted landscape within which The Scapegoat is set, or the Palestinian clothing worn by biblical figures, lends the spiritual message an added immediacy. From 1839 onwards, the resources of photography were increasingly harnessed for similar ends; of the roughly one hundred photographers who worked in the Middle East in the 19th century, special mention

130

must be made of Francis Frith, John Cramb, Maxime du Camp, James Robertson, Francis Bedford, Felice Beato and Félix Bonfils. There is no space here to investigate in any detail the role of imitation in securing authenticity, but the copies made of Islamic inscriptions, ceramics (especially Iznik and lustreware), glass, carpets and gardens, especially in the 19th century, speak for themselves – for as the proverb says, imitation is the sincerest form of flattery. Finally, the intrusion of political factors in the European response to the East deserves some comment. Many of the specific examples cited above could be interpreted as expressions of a political agenda. Edward Said’s classic Orientalism, first published in 1978, and the huge quantity of further studies which it has generated, has made this the bestknown aspect of Orientalism, as well as the one which has attracted the most polemic. As early as 1529 Altdorfer painted The Battle of Issus as a confrontation between East and West. But it was the sustained irruption of the West into the Islamic world, symbolised most effectively by Clive’s victory at Plassy in 1757 and by Napoleon’s invasion of Egypt in 1798, which set the tone for future exchanges. The mighty Description de l’Égypte, published in 24 volumes between 1809 and 1822, did more than any other study to spark the Egyptian Revival fashion in Europe. But alongside its multifaceted scientific studies can be discerned a serene confidence in the natural superiority of the European. The same apparently unconscious assumption can be detected in the work of the military and civilian servants of the crown in India, men like Sir William Jones (1746-94), effectively the founder of Sanskrit studies in Europe, and Sir William Muir (18191905), sometime Principal of the University of Edinburgh and author of a classic Life of Mahomet, who in a professional way studied the languages, religion, history and society of the people whom they governed. By way of a coda, let it suffice to draw attention to a book which for some 50 years has served as the most popular and influential introductory text for art history, whether taught at schools and universities or enjoyed by the general public – Sir Ernst Gombrich’s The Story of Art, first published in 1950. This classic survey has been reprinted in almost 30 editions, yet in none of these reprints has the title been changed to reflect the essential subject of the book, namely the story of western art. That extra word says it all. The art of Islam (dismissed in less than 500 words) is bracketed with that of China in the same short chapter, and neither topic attracts as much coverage as 20 years of art in Renaissance Italy. As for India, the author candidly admits: ‘I may as well confess that I

THE ORIENT OF THE IMAGINATION

131

have found no room for Hindu or Etruscan art, or for masters of the rank of Quercia, Signorelli or Carpaccio …’ To speak in the same breath of a minor Venetian painter and the art which is the heritage of almost a billion people, and which stretches back for five millennia, reveals the very essence of that Eurocentricity and Orientalism which Edward Said exposed in his book. In art history, as in many other fields of enquiry, that is an inadequate way to prepare students to enter the multicultural world of the 21st century.

VIII 1

For God, Empire and Mammon : Some ArtHistorical Aspects of the Reformed Dīnārs of ‘Abd al-Malik2

T

Introduction

he general introduction of a purely Islamic epigraphic coinage by the Umayyad caliph ‘Abd al-Malik in the year 77 H./696–7 (for gold dīnārs) and 79 H./698–9 (for silver dirhams) has long been regarded, and rightly so, as a hinge of Islamic history. Its details have, however, largely remained the province of numismatists. This paper investigates its arthistorical aspects so far as dīnārs are concerned. But first, a brief disclaimer may be in order. Michael Bates, the doyen of Islamic numismatists, very properly notes the desirability of comparing like with like in analyzing Umayyad coins, and thus of studying the production 4 of a mint as a whole. The latter task can be approached from various 3

  I am grateful to my friend Dr. Shireen Mahdavi for giving me the inspiration for this title, which is purloined from that of her own work, at once scholarly and diverting: For God, Mammon and Country. A Nineteenth-Century Persian Merchant. Haj Muhammad Hassan Amin al-Zarb (1834–1898) (Boulder, Colorado, 1999), which I have read with pleasure and profit. Since her book is a biography of a mintmaster, the borrowing seemed all the more appropriate. 2  I acknowledge with gratitude the extensive practical and bibliographical help which I have received from Luke Treadwell in the preparation of this article. 3  In this paper the new type of coinage generally introduced from 77 A.H. is referred to either as “reformed” or as “purely epigraphic”. Both terms are intended to denote the coinage also known as “post-reform”. 4  M. Bates, “The Coinage of Syria Under the Umayyads, 692–750 A.D.”, in The Fourth International Conference On The History Of Bilad al-Sham During The Umayyad Period. Proceedings Of The Third Symposium 2–7 Rabī’ I 1408 A.H.124–29 October 1

FOR GOD, EMPIRE AND MAMMON

133

angles – metrology or fineness, for example, or individual details of the design such as pellets/dots or annulets and their positioning. He himself 5 has shown the way in his publications. But the issue of overall design and of epigraphic style – what one might call the aesthetics of coinage – has apparently not yet been treated in detail in this way. The present article is an attempt to suggest some tentative findings based on the unbroken sequence of surviving standard Umayyad reformed dīnārs between 77 H. 6 and 132 H. None of these bear a mint name and the scholarly consensus 7 is that all of them were struck in Damascus. Thus they belong together and can be studied as a group, the output of a single mint over a couple of generations. Special attention will be paid to the dīnārs of a few specific years – 77, 78, 79, 87, 97, 123, 127 and 132 H. – so that key stages of the evolution in terms of epigraphy and design alike can be isolated. My intention is to deal elsewhere with the larger and more complex issue of the design of Umayyad dirhams. The case of Umayyad dīnārs is simpler than that of Umayyad dirhams, not least because the epigraphy itself is less varied on the gold than on the silver coins. Another reason is, as just mentioned, the fact that the dīnārs generally lack a mint name. To treat each dated dīnār as representative of all the dīnārs minted in that year is of course only a rough and ready method, because it ignores the differences between the various dies for a given year. Thus the formal analyses in this paper are only preliminary. But the remarkable lack of orientating surveys by professional numismatists of the whole area of design and epigraphic change means that even a tentative survey of this field should bring to light material which, though 1987. English Section – Vol. II, eds. M.A. Bakhit and R. Schick (Amman, 1989), 210, n.26. 5   M. Bates, “History, geography and numismatics in the first century of Islamic coinage”, Schweizerische Numismatische Rundschau/Revue Suisse de Numismatique 65 (1986), 231–61; idem, “Coinage”, 195–228; and A.S. DeShazo and M. L. Bates, “The Umayyad Governors of al-‘Iraq and the Changing Annulet Patterns on their Dirhams”, Numismatic Chronicle (1974), 110–18. Cf. also P. Grierson, “The monetary reforms of ‘Abd al-Malik. Their metrological basis and their financial repercussions”, Journal of the Economic and Social History of the Orient 3 (1960), 241–64. 6  The minor exceptions will be dealt with shortly but do not affect the discussion of epigraphic style. 7   G.C. Miles, for example, stated that the epigraphic dīnārs were “undoubtedly” struck at Damascus (“The Earliest Arab Gold Coinage”, American Numismatic Society, Museum Notes 13 [1967], 205). This is also the opinion of Bates (“Coinage”, 212) – though earlier in the same article (210) he speculates that production of dīnārs was shifted to the mint of al-Jazīra after 127/744–5.

134

perhaps thoroughly familiar to numismatists, has still not been properly exploited to reveal its art-historical significance. Indeed, it seems scarcely 8 to have been published at all. Naturally, the next step is to subject some of the generalisations offered in this paper to intensive scrutiny based on 9 a study of the various dies recorded for a given year. That should clarify several remaining problems and fine-tune – or indeed revise – many of the theories proposed in this paper. The problems raised by the reformed gold coinage will be considered under five separate headings of very unequal length: first, the contemporary context; next, the sources of the design; third, the content; fourth, the design and epigraphic style, a topic to which most of this paper will be devoted; and finally, some implications of this reform. The contemporary context This is familiar territory. ‘Abd al-Malik reigned from 65/685 to 86/705. The decade of 65/685 to 76/695 witnessed his suppression of several anti-Umayyad uprisings in Iraq and northern Syria. It is therefore not surprising that in their wake he resolved to take measures to assert caliphal authority in both secular and religious terms. A few examples will make this clear. The 690s and the immediately following decade saw, as is well known, a general tightening of administrative procedures. The language of the 10 dīwāns was gradually changed from Greek and Pahlavi to Arabic; the 11 diverse tax-systems of the provinces were systematised; the barīd or 12 imperial postal service was instituted; a state archive was developed in 13 14 Damascus; weights and measures were standardised; and qāḍīs were A first (but largely dirham-based) attempt in this direction was made by S.E. Lane Poole, in a rather misleadingly entitled article published well over a century ago: “On mint-characteristics of Arabic coins”, Numismatic Chronicle, New Series, XI-II (1873), 54–9, and pl.1. 9  But not, in the case of dīnārs, a given mint, for the vast majority of the surviving coins bear no mint name; see note 75 below. 10   For brief surveys of this issue see J. Wellhausen, The Arab Kingdom and its Fall, tr. M.G. Weir (repr. of 1927 ed., Beirut, 1963), 219–20; G.R. Hawting, The First Dynasty of Islam. The Umayyad Caliphate AD 661–750 (London and Sydney, 1986), 63–4; P.K. Hitti, The Origins of the Islamic State (repr. of 1916 ed., Beirut, 1966), 465–6. 11  Idem, History of Syria including Lebanon and Palestine (London, 1951), 474–5. 12  Hawting, First Dynasty, 64. 13   Hitti, Syria, 478. 14   Ibid., 455. 8 

FOR GOD, EMPIRE AND MAMMON

135

appointed so that disputes between Muslims could be resolved on a 15 properly Islamic basis. But other actions were taken at this time which refer more particularly to the written word. Inscribed milestones in the name of ‘Abd alMalik were set up along major roads, for example those radiating from 16 Damascus. Much more importantly, the Dome of the Rock, completed 17 in 72 H./692, had its interior encircled by an inscription some 240 metres 18 long which largely comprised selections from the Qur’an affirming the key principles of Islam and attacking the Christian doctrine of the Trinity. Hitherto it has not been sufficiently noted that this inscription is only the sixth recorded in the Répertoire Chronologique d’ Épigraphie Arabe in the entire 19 seventy years following the Hijra. This provides the clearest evidence that there was effectively no established tradition at this time of using the Arabic language for public inscriptions in an Islamic context, let alone on such a grand scale. Moreover, the Dome of the Rock inscription is almost exclusively Qur’anic – the first use of the Qur’an in epigraphy, and a very public, indeed overwhelming, use at that. It is widely believed that it was under al-Hājjāj, governor of Iraq from 74/694 to 95/714 – and thus perhaps also in the 690s – that action was taken to ensure greater accuracy in Qur’anic recitation by eliminating ambiguities in the way the 20 text was written. The celebrated luxury Qur’an of which portions were  M.G.S. Hodgson, The Venture of Islam I (Chicago and London, 1974), 226.  E.g. M. Cohen et al., Répertoire Chronologique d’Épigraphie Arabe, I (Cairo, 1931),

15 16

no. 14. For a detailed account of these milestones as a group, see M. van Berchem, Matériaux pour un Corpus Inscriptionum Arabicarum. Deuxième Partie. – Syrie du Sud. Jérusalem “Ville”. Tome Premier, Premier Fascicule [Mémoires publiés par les membres de l’Institut Français d’Archéologie Orientale du Caire, 43] (Cairo, 1922), 17–29, especially 21, figs. 1–4. 17  For a discussion of the completion date of the Dome of the Rock, see S. Blair, “What is the Date of the Dome of the Rock?” in J. Raby and J. Johns (eds), Bayt al-Maqdis. ‘Abd al-Malik’s Jerusalem. Part One (Oxford Studies in Islamic Art, IX) (Oxford, 1992), 59–87. 18   RCEA I, 8; O. Grabar, “The Umayyad Dome of the Rock in Jerusalem”, Ars Orientalis III (1959), 38. 19  Subsequent discoveries have only marginally affected the accuracy of this statement. Note that the fifth (RCEA I, no.7), which purports to be a foundation inscription from the Dome of the Rock dated to 65/684–5, has been conclusively proved by Creswell to be spurious (K.A.C. Creswell, Early Muslim Architecture. Umayyads A.D. 622–750, 2nd ed. [Oxford, 1969], I, 72–3), so that the inscription of 72/691–2 is actually the fifth, and this number includes inscriptions no longer extant. 20   T. Nöldeke, Geschichte des Qorans, 2nd ed., ed. F. Schwally (Leipzig, 1909), III, 262. Cf. Hitti, Syria, 476.

136

found within the roof of the Great Mosque of Ṣan‘ā’ may well date from around this time, if the Carbon 14 tests that have been carried out on its 22 parchment are accurate. What of the situation in coinage at this time? A ten-year truce which ‘Abd 23 al-Malik had concluded with the Byzantine emperor in 69/689, and which had involved the payment of Muslim tribute, was abrogated in 72/692. Several theories have been put forward to explain this action. Gibb states that it was because the Byzantine Emperor refused to accept the new gold 24 currency struck by ‘Abd al-Malik, but this notion has been convincingly 25 rebutted by Breckenridge. Others suggest that the provocation was that the Emperor had threatened to overstrike the Byzantine gold coins 26 exported to Islamic territory with anti-Muslim formulae. Yet others argue that the changes introduced into Byzantine coin design by Justinian II, and especially the image of Christ with a cruciform nimbus, were seen as 27 too aggressively Christian for Muslim taste. At all events, it seems that, at least to some extent, matters connected with Byzantine currency provided 28 a trigger for the Muslims to design their own coinage. Moreover, there is a clear pattern of the increasing Arabisation and Islamicisation of both Byzantine and Sasanian coins by means of writing in the issues preceding the currency reform. The written content of these coins was varied. It could be purely neutral, as in the case of a date. Or it could have political ramifications, such as the name of a governor or a mint. Or, 21

 H.-C. Graf von Bothmer; ”Architekturbilder im Koran. Eine Prachthandschrift der Umayyadenzeit aus dem Yemen”, Pantheon XLV (1987), 4–20; O. Grabar, The Mediation of Ornament (Washington, D.C., 1992), 155–74, colour pls. 16–17. 22  Personal communication from Graf von Bothmer. 23   H.A.R. Gibb, “’Abd al-Malik”, Encyclopaedia of Islam (2nd ed.), I, 76. 24   Ibid., I, 77. 25   J.D. Breckenridge, The Numismatic Iconography of Justinian II (685–695, 705– 711 A.D.) (New York, 1959), 69–70. He argues that this report, which he believes was accepted by Gibb at face value on the basis of the account of Theophanes (Theophanis Chronographia, ed. C. de Boor [Leipzig, 1883], 365) was not the principal casus belli, but he does not deny the accuracy of the report itself. See also Bates, “History”, 247. 26   Hawting, First Dynasty, 65. 27   Breckenridge, Numismatic Iconography, 75 and pl. V/30. But see the doubts expressed by Bates (“History”, 252–3). 28   Miles succinctly summarises the sequence of events between 72 H./691–2 and 74 H./693: the issue of the modified Heraclian type with three standing figures; the new type of Justinian II; and ‘Abd al-Malik’s response to this, namely the “Standing Caliph” type (Miles, “Earliest”, 227). 21

FOR GOD, EMPIRE AND MAMMON

137

most important of all, it could be religious in tone. The earliest examples have only the bismillah. Later they also bear the shahāda with several minor 29 variations, and other messages also appear. It has to be emphasised that the dating of these intermediate issues – neither fully Islamic nor fully 30 Byzantine or Sasanian – is fraught with difficulties. Nevertheless, the overall message of these issues is clear: their restless experimentation, the constant tinkering with this or that element of the design, is a barometer of Muslim frustration, or at any rate unease, with the currency situation. Was the caliph master of his own house or not? If he was, why was he still using coins with Byzantine or Sasanian designs? Retaining the status quo in currency matters had earlier been acceptable as a stop-gap policy adopted in haste at the time of the conquests. In the 630s it had made excellent sense in the interests of continuity and economic stability. But it became progressively harder to justify as the Islamic empire became ever larger and more securely established. It was an extra source of humiliation that the Umayyads were dependent on Byzantium for bulk shipments of ready-made coinage. The time had come for the Muslims to separate themselves from those they had conquered, and to do so in a form that was both symbolic and final – a coming of age. The well-nigh frenetic innovation which characterized the Islamic coinage of the 690s prepared the ground well for the coup de théâtre of the reform coinage. Its date is conventionally given as 77 H. because this is the date of the earliest surviving specimens, all of them dīnārs. The survival of “Standing Caliph” issues in gold also dated 77 H. suggests that it was this year, and probably 31 the latter part of it, that witnessed the striking of the first epigraphic 32 33 coins. The medieval Arabic sources tend to give an earlier date, but are contradicted by the physical evidence just cited.  J.Walker A Catalogue of the Muhammadan Coins in the British Museum. A catalogue of the Arab-Byzantine and Post-Reform Umaiyad Coins, 2nd ed. (London, 1956), (hereinafter CMC) II, lxvii (copper) and ciii. 30  For a lucid summary of the position, see Bates, “History”, especially 231–55. 31  See Sotheby’s Coins and Paper Money 29 September 1988, 9 (T. Eden). 32  It has been suggested that the immediate cause of the reform was Muslim protests against the Byzantine nature of the “Standing Caliph” coins (ibid.). 33  Such as al-Tabari, who gives the year 76 H. and states that dīnārs and dirhams were issued together (The History of al-Tabarī (Ta’rīkh al-rusūl wa’l-mulūk). Volume XXII. The Marwanid Restoration, tr. and annotated by E.K. Rowson [Albany, N.Y., 1989], 90–1; for other dates in the Arabic sources see ibid., 91, n.360). Bates comments on the unreliability of the Arabic sources on the question of when the reform coinage was issued (“History”, 246, notes 22–3). 29

138

That the historians have simplified a complex process, which comprised not just two stages but several, is suggested by at least five dirhams, broadly of reform design, minted in various locations (Ādharbāijān, Arminiya, Jayy, Kūfa and Shaqq al-Taimara) in 78/697–8 – i.e. a year before the general reform of the dirham, as testified by surviving specimens from 35 numerous mints, can be shown to have occurred. 34

The sources of the design Was the reformed coinage an innovation? The answer, with some reserve, seems to be “yes”. But phrased in that way, the question is too crude. After all, the mind that could conceive of the Dome of the Rock as a building that was also a sacred book would have no difficulty in conceiving of a coin too, as a sacred book, though much shorter. In other words, the key process is likely to have happened in the mind first, on a theoretical level, and was only thereafter translated into visual terms. Perhaps a model was needed; perhaps not. As the case of the Dome of the Rock shows, that idea of using the text of the Qur’an in a public way was gathering momentum in the 690s and found various expressions – architecture, coins and, it seems, display Qur’ans, the latter perhaps a way of publicising the newly introduced but by no means universally employed scriptio plena of the sacred text. Now to the models that might have served the Umayyad designer. At first sight, apparently, there were none. It is important to remember that the coinage reform occurred in two stages: first gold, then silver. It is of course perfectly conceivable that the new design was intended from the first to be applied to both gold and silver issues. But still, the gold dīnār was the first denomination to be changed, and thus the obvious place to look for a model is in Byzantine gold coins, since there were practically no Sasanian issues in that metal. Now while some later Byzantine non-gold 34  As discussed in the previous note, they give various dates for the reform, but the fact that these are earlier than the dates on the coins themselves need not entirely invalidate their accounts, for presumably the reform took some time to plan and execute. 35  L. Ilisch and M. Rashad, Alte Münzen Dr Busso Peus Nachf. Münzhandler Auktion 31. Oktober 2001 Frankfurt am Main, Katalog 369, 80–1, lot 1467 (with an extended discussion of a further 78 H. dirham from Basra whose current whereabouts are not known). Cf. also the lengthy discussion of the dirhams of Ādharbāijān and Kūfa (texts attributed by Drs. Ilisch and Ramad to R. Darley-Doran: Sotheby’s Coins, Medals, Decorations and Banknotes 16 November 2000, 16–17, lot 7 and Sotheby’s Coins and Medals 27 May 1999, 14, lot 132 respectively).

139

FOR GOD, EMPIRE AND MAMMON

issues, notably the mid-eighth century silver milaresion, do have a reverse entirely occupied by an inscription, Byzantine gold coins predating ‘Abd alMalik’s reform do not have this feature. Nor was it the practice in Roman or Greek coins, though at times a reverse might contain a substantial inscription as well as an image. Thus neither the past nor the present, whether in the Mediterranean or the Middle Eastern world, offered a ready parallel for purely epigraphic coinage. China had been producing such coins for well over a millennium by this time, but no evidence has been produced to suggest that the Arabs knew about such coins. Perhaps, though, it is a mistake to be so very literal in the search for a model, for example by limiting that search to coins. It is a characteristic practice of Umayyad art to take a secondary theme and promote it to primary status, as the mosaics of Jerusalem and Damascus show. This capacity to change the visual impact of a work of art dramatically by exaggerating a feature which had hitherto been inconspicuous, latent or at any rate of minor visual importance recurs in many guises in Umayyad 37 art. The process seems simple, when baldly described like that, but its effect is dramatic. It transforms the model beyond recognition. By that reckoning, it would have taken only a modicum of imagination to remove the images from either Byzantine or Sasanian coin types, or indeed from the experimental Islamic issues, and simply increase the amount of writing to fill up the space thus created. And that may have been what actually happened. Another possibility may, however, be canvassed. It builds on another trademark of Umayyad art, namely the readiness to transpose ideas from one medium to another. It is worth recalling that the pre-reform experimental issues attest strong elements of Persian origin, such as the use of silver, the bust of the Sasanian monarch and the tripartite design of the reverse. These borrowings suggest that those responsible for the coinage reform were open to ideas from the east. And relevant ideas were not confined to coins. Indeed, several Sasanian seals are known whose 36

36   G.C. Miles, “Byzantine milaresion and Arab dirhem: some notes on their relationship”, ANS, MN IX (1960), 189–218. Most of this article deals with milaresia which are overstruck on Arab dirhams, but Miles notes that for almost three centuries this coin was “the only silver issue of the Constantinople mint” (ibid., 192). 37   The decision to encase the exterior of the Dome of the Rock in wall mosaic, a medium hitherto largely confined in the Byzantine context to the interiors of sacred buildings, is a case in point. So is the much more important role accorded to inscriptions in the Dome of the Rock vis-à-vis any of its Byzantine analogues.

140

design is entirely made up of writing. Their inscriptions identify the rank and name of the owner; sometimes a location is also given. Admittedly they are not coins, but the parallel in size and general appearance is very close. Moreover, the disorderly way in which these Pahlavi inscriptions are laid out would have sent a clear challenge to the Muslim designers to 39 devise a better layout: more disciplined, more symmetrical, more stately. Byzantine seals pre-dating the Umayyad coinage reform, and with one 40 side of purely epigraphic content in Greek, are also known. Could the transitional Muslim issues, then, have included a type of which either 41 obverse or reverse was purely epigraphic? 38

The content of the reformed coinage Content and design are intimately related, and it is worth lingering briefly over this. Since these two factors work rather differently in the dīnār from the way they do in the dirham, it is a misleading simplification to treat both these coin types together in this particular context. For reasons of space alone the discussion will be confined to Umayyad reformed dīnārs. Some general introductory comments may be in order. When one considers these dīnārs as a continuous sequence minted over a period of more than fifty years, certain features stand out. The deliberate originality of the layout invites a close focus on the text itself. The problems of design have been triumphantly overcome. The letters are eminently legible 38  See, for example, M.G. Morony, Iraq after the Muslim Conquest (Princeton, 1984), 49, fig. 2 d–g. 39  The unique coin with a Pahlavi version of the shahāda on the reverse, dated 72/692-3 (see M.I. Mochiri, “A Pahlavi Forerunner of the Umayyad Reformed Coinage”, Journal of the Royal Asiatic Society [1981, No.2], 168–72), might be regarded as an early experiment foreshadowing the purely epigraphic coinage; but it has subsequently been argued that it dates from some time between 79/698 and 81/700 (S.D. Sears, “A hybrid imitation of early Muslim coinage struck in Sijistan by Abu Bardha’a”, American Journal of Numismatics, Second Series, I [1989], 155–67), in which case its reverse would be a free copy in Pahlavi of the Arabic legend on the reverse of the standard reformed dirham. 40 Spink Auction 127: Byzantine Seals from the collection of George Zacos, Part I (7 October 1998), lot 12, with a six-line legend in Greek. It dates from 683–4 or 686–7. Here, too, the epigraphy is undistinguished. See also N. Oikonomides, “Silk Trade and Production in Byzantium from the Sixth to the Ninth Century: The Seals of the Kommerkiarioi”, Dumbarton Oaks Papers, 40 (1986), 33–53. 41  That such a coin – perhaps minted in small quantities like some other transitional issues – could have been struck around this time is proved by the coin discussed in n.39 above; whatever its date, a significant aspect of its importance is that 50 % of it is purely epigraphic. This underlines its tentative, experimental and transitory nature.

FOR GOD, EMPIRE AND MAMMON

141

and the inscription has sufficient “room to breathe”, thanks to the ample space between letters. Yet there is on the whole little attempt to lengthen or contract given letters for visual effect or emphasis. Like the layout and execution of so many Roman inscriptions, it is an exercise in good 42 manners. So carefully are the inscriptions designed that every word, even 43 every letter, counts. There is no code to be broken, no abbreviations; and it is tempting to speculate, therefore, that these inscriptions may also have a proclamatory or even a proselytising function. This is especially true of the shahāda, since it is written so much larger than the inscriptions of the margin; and it does, after all, encapsulate the message of Islam. Next, the religious content of these coins is, like their basic principles of design, consistent from the moment that they were first struck until the 44 fall of the Umayyad dynasty 54 years later. This is an impressive tribute to the steadfastness of purpose and policy in Umayyad coinage. Nor is this all. The decision, at once simple and profound, to devote the whole of the coin to writing removes at a stroke the canonical visual distinction between obverse and reverse. For centuries, in Byzantine, Sasanian, Roman and Greek coinage – and, incidentally, in the early Arab-Byzantine and ArabSasanian issues too – it had been the norm to place generically distinct designs on the obverse and reverse of coins. Thus an image, commonly that of the ruler in bust form, occupied one side of the coin, in contrast to some other kind of non-portrait representation, commonly of religious import, on the other side. In this way state and church both received due emphasis, but their symbols were, for the most part, kept physically distinct. Over the three centuries immediately preceding the Umayyad coinage reform this practice had hardened into a fixed convention in both Byzantine and Sasanian issues. Now that convention had been exploded. Obverse and reverse were no longer readily distinguishable. All the visual fixed points of the coinage had been removed. The speaking symbols that had served the Mediterranean and Middle Eastern world so well for a millennium and more on coins of all kinds had now been utterly 42   In this respect, the contrast with the reverse of the hybrid coin from Sīstān discussed earlier (n. 39 above) is striking. 43   The prevalence of such abbreviations in other numismatic traditions, even into modern times (e.g. on current British coins such legends as DG and FID DEF) makes this all the more remarkable. 44  This is presumably what Bates means when he states that “The first gold dīnārs of 77 are exactly like the last Umayyad ones of 132” (“History, geography and numismatics”, 256); he acknowledges considerable changes in their epigraphy (ibid., 257).

142

expunged, to be replaced by the written word alone. And that written word gave pride of place – in location, in size, in legibility – to the Word 46 of God. This was the case, moreover, on both obverse and reverse. Could one 47 indeed tell the difference? That is not a casual question; it is a signal that in design terms the purely epigraphic coinage represented a root-andbranch reform of previous practice, about as radical as it was possible to be. Any answer can only be speculative at this stage. Modern scholarly convention among numismatists identifies as the obverse the side whose field reads: “There is no god but/ God alone./He has no associate”, of which the first two lines contain the essence of Islamic monotheism. The margin is Qur’anic, namely part of Sura 9:33: “Muhammad is the Apostle of God whom He sent with guidance and the religion of truth that he may make it victorious over every other religion”. This side, then, is entirely religious in content. Not so the other side; and, since in pre-reform ArabSasanian coins the date was given on the reverse (Arab-Byzantine coins are undated), it seems reasonable to use the fact that this other side contains secular information in the margin: “In the name of God. This dīnār 48 was struck in the year…” as an indication (no more!) that it should be 49 identified as the reverse. Yet here too the principal emphasis is a religious one, for the all-important field is again Qur’anic; it reads “God is one./ God is the Eternal. He begets not/ and He is not begotten”. The content of the coins, then, is strongly Qur’anic and in a literal sense marginalises all other information. The latter is confined to the date 50 of minting, though a few coins give the mint as well. The absence of the ruler’s name is nothing short of revolutionary and ensures that there is, so to speak, no competition for the Word of God. Thus the absolutist 45

  That this change was well received in at least some Muslim circles is suggested by the report that the new dīnārs were suspect only because they weighed the same as the old worn dīnārs (Wellhausen, Arab Kingdom, 218). 46   Before the coinage reform, the pious reproached al-Hajjaj for putting his own name on the coins after the name of God (ibid.). 47   See M.L. Bates, “Islamic Numismatics”, Middle East Studies Association Bulletin XII/3 (1978), 6–7. 48  The word fī (“in”) disappears on dīnārs after 80 H. (Lane Poole, “Mintcharacteristics”, 58). It seems that the motive was to create extra space in the margin, though it was only after 110 H. that this became a significant issue. 49   Yet it could be argued that earlier convention made the obverse the side with secular content. 50  Walker, CMC, 99–103. 45

FOR GOD, EMPIRE AND MAMMON

143

flavour of the experimental coinage has been definitively discarded. The messages are crystal clear and, as already noted, overwhelmingly religious in tone. Reading continuously from obverse field to reverse field, the total concentration on a single message is unprecedented: “There is no god but God alone. He has no associate. God is one. God is the Eternal. He 51 begets not and He is not begotten”. This is a fivefold repetition of that same attack on the divinity of Jesus and on the Trinity that is encountered in the inscriptions of the Dome of the Rock. In the context of the aggressive Christology of exactly contemporary Byzantine coinage, the obvious conclusion is that the content of these inscriptions was beamed principally at the Christian population in the Near East, especially in Syria. There is, at any rate, nothing specifically anti-Sasanian in these words. It is also noticeable that, as befits a text that is either Qur’anic or has powerful Qur’anic resonances, the language should be Arabic, not Greek. The exact sequence of the text just cited is not found in the Qur’an but some parts 52 of it are, for example in Sura 112. There is also a direct quotation from Sura 9:33 in the obverse margin: “Muhammad is the Apostle of God whom He sent with guidance and the religion of truth that he may make it victorious over every other religion”. The unique and uncreated quality of God is at the forefront of the minor 53 variations of texts found on the reform coinage. The only human name 54 found on these gold reformed issues is that of the Prophet Muhammad. Moreover, the use of the phrase bismillah preceding the statement that this dīnār was struck in a given year sanctifies the act of minting coinage, removing it from the purely secular sphere and underlining that the   The last three of these sentences are taken from Sura 112.   O. Grabar, The Formation of Islamic Art, 2nd ed. (New Haven and London,

51 52

1987), 59–60. 53   See Walker, CMC, lvii-lviii for a convenient listing which distinguishes his socalled “Eastern legend” (used on mints of the Levant and further east) from the “Western legend” used in Ifrīqiya and al-Andalus. The latter formula omits the words lā shārik lahu (“He has no associate”) in the obverse field and the words liyuẓhirahu ‘alā al-dīn kullihi (“that He may cause it to prevail over all religion”) in the obverse margin. The reverse field in the Western Islamic dīnārs is totally different from that of the mintless dīnārs discussed so far; it reads bismillah al-raḥmān alraḥīm (“In the name of God, the Compassionate, the Merciful”), while its reverse margin omits the bismillah. For the flaws in Walker’s treatment of Western Islamic coins, see Bates, “History”, 259. 54  Though it is confined to the (obverse) margin, and is thus visually not centre stage. Nevertheless, his importance as the last of the prophets is sufficiently emphasised by this device (cf. Sotheby’s, 29 September 1988, 9).

144

right of sikka belongs to God alone. Finally, it is worth noting that the choice of the shahāda as the centrally placed message indicates that for the contemporary Muslim authorities this was the most important of the five pillars of the faith. It has been moved from the margin, the space which it occupied in various earlier experimental issues, to the field. It is thus the immediate centre of attention. The specific attack on key Christian doctrines contained in the texts of this purely epigraphic coinage has an unmistakably political dimension. Most of the subjects of the Umayyads were Christian, and these same subjects had no choice but to accept the new legal tender, however 56 they might detest its messages. The resultant humiliation, repeated 57 on a daily basis wherever cash transactions were made, made the new coins instruments – on a psychological level – of religious and political subjection. It is as well to remember what the purely epigraphic coinage in gold and silver omits. Perhaps most radically of all, it makes no reference to the ruler, whether caliph or governor, not even in the most abbreviated terms. No bust, no upright figure, no name, no title. In this respect it departs decisively from the precious metal coinage of the Byzantines and the Sasanians, and indeed from earlier Islamic coinage. This innovation was not to last; the ‘Abbasids within a generation re-instated the name of the ruler, and indeed eventually cited rulers (e.g. a governor and a caliph) 58 59 in both gold and silver. But in the all-important first half-century of purely epigraphic coinage, this remarkable modesty on the part of the Umayyad rulers concerned gave still greater glory to God. The novelty of content can be summarized in the decision to make of the coinage in precious metal a pugnacious rebuttal of Christian doctrine, a summary of Islamic doctrine, a reduced miniature Qur’an, a metaphor of state control and perhaps even a vehicle of conversion. 55

R. Darley-Doran, “Sikka”, EI, 2nd ed., 593, argues that this right is vested in the Prophet too. 56  Cf. the fuller text on Umayyad dirhams: “however much the idolaters may be averse”. It is not hard to read a contemporary significance into those words. 57  And not only in precious metals – for many Umayyad fulūs have similar messages. 58  E.g. G.C. Miles, Rare Islamic Coins (New York, 1950), 38, no. 148, pl. VII. 59  Ibid., 74, no. 262. 55 

FOR GOD, EMPIRE AND MAMMON

145

Design and epigraphic style These are the two features of crucial importance in the visual aspect of the reformed dīnār. The design formula looks deceptively simple – a central field comprising three lines of relatively large writing on both obverse and reverse, with a circular margin in relatively smaller writing enclosing that field, again on both obverse and reverse. The immediate impression it 60 gives is of a confident finished design. It is perhaps reasonable enough to expect, given that the median diameter of the dīnār (c. 20 mm) is about one-third less than that of the contemporary dirham (c. 29 mm), with a proportionate reduction in the space available for tinkering with the basic design, that such a formulaic layout should present fewer options to the designer of the dīnār and that this design should therefore undergo fewer 61 changes than did that of the dirham. Such is indeed the case. Nevertheless, in the dīnār too one may note many minor adjustments, and some of them have to do with the design formula. Quite apart from minor differences in letter forms from die to die and year to year, the exact placing of identical legends could differ quite markedly, resulting in coins which were visually unbalanced in various ways. These include uneven 62 horizontal spacing of the legends in either field or margin, the intrusion 63 64 of the legends of the field into those of the margin or vice versa – it is common to find less room between field and margin at the end of a line in the field than at its beginning – or an imbalance between the size of 65 the exergue and that of the corresponding “tympanum”. All of these variations are of course independent of the epigraphic style followed in Bates notes the near absence of fumbling or experimentation (“Coinage”, 208; “History”, 256). 61  The several dramatic changes that transformed the epigraphy and the layout of the dirham over a period of some fifty years are far more marked than any of the alterations in the dīnār. And they have scores of mints – even if in fact these coins were actually struck at relatively few locations (H. Kennedy, The Armies of the Caliphs. Military and Society in the Early Islamic State [London and New York, 2001], 70; M. Bates, “The Dirham Mint of the Northern Provinces of the Umayyad Caliphate”, American Numismatic Journal, Series I, 15 [1989], 89–111). 62  E.g. in some dīnārs of 90 H. For the purposes of this analysis of Umayyad dīnārs I have used the specimens from 78 – 132 A.H. illustrated (reverse only, and one from each year) in Spink Auction 13. Islamic Coins from the Turath collection Part I, 25 May 1999, lot 16, and in what follows these particular coins will be identified by the year in which they were struck. Dīnārs from other sources will be identified as such. 63  80, 81, 100 and 107 H. 64  94, 101, 108, 109, 114, 117, 119 and 129 H. 65  Or the upper curved segment at the top of the field. 60 

146

these coins. That said, the balance between obverse and reverse fields in the dīnār was well calculated, with three lines of text in each case and a similar number of letters in both fields (24 on the obverse and 28 on the 66 reverse). The coins minted from 77 H., then, opted for a new emphasis in design and content alike. The novelty of design, so far as coinage (as distinct from seals) is concerned, is plainly apparent, most notably in the decision to apply the same principle to the obverse as to the reverse and thus to do away with figural images altogether. Many of the minor epigraphic changes that come to light when this sequence of dīnārs is intensively scrutinized seem to be independent of the evolution of distinctive epigraphic styles. Some may have to do with legibility, for example the addition of dots to the ḍād or bā’ of ḍuriba, or of pairs of vertically arranged dots under the yā of yalid or yūlad – though the profusion and irregular placing of such dots argues against this, and indeed the dots on these coins have long been a controversial 67 topic. Other features clearly have an aesthetic dimension – for example 68 the tapering tip of the long bā’ of ḍuriba, or the treatment of that letter so that, in markedly exaggerated form, it follows the convex curve of the 69 coin profile itself. It is noticeable that these elements come and go, as do other minor differences in letter forms, spacing and the like between coins from the same year but different dies. Perhaps they reflected the personal 66  Incidentally, the uneven length of the messages which the reformed dirham carried made it impossible to replicate this harmony. 67   S. Lane Poole, “On mint-characteristics of Arabic coins”, 59, mentions Soret’s comments on points, but goes no further than this; cf. also idem, The Coins of the Eastern Khaleefehs in the British Museum, ed. R. Poole (London, 1875), x. For a fuller discussion, see M. Jungfleisch, “Les points secrets en numismatique: une innovation due aux Arabes (?)”, Bulletin de l’Institut d’Égypte XXVIII (1945–6), 10115. G.C. Miles, “Some Early Arab Dīnārs”, American Numismatic Society, Museum Notes 3 (1948), 97–114, records the points on some Umayyad dīnārs from 80 to 132 H. and finds that of these 120 coins 57 have points, in some cases multiple points; and that these points are clustered in the issues between 80 and 93 H. (they do not occur on the coins in his sample dated 77, 78 and 79 H.), and then again between 100 and 110 H. Thereafter they tail off sharply, though they are found spasmodically until 131 H. (pace Bates, “Coinage”, 212). With 7 exceptions, his sample shows that the points were on the reverse. Bates, “Coinage of Syria”, 211–12, speculates that points on dīnārs reflect changes in the administration of the minting establishment, and that the frequent congruence of points on dīnārs and Damascus dirhams suggests that the dīnārs were struck at a caliphal mint in or near Damascus. 68  E.g. 87, 91 (the most striking example) and 97 H. 69   E.g. 104 and 120 H.

FOR GOD, EMPIRE AND MAMMON

1a. Reverse of dinärs from 78H. to 104H.

147

148

1b. Reverse of dīnārs from 105H. to 132H.

FOR GOD, EMPIRE AND MAMMON

2. Obverse and reverse of dīnār of 77H.

3a. Obverse of dīnār of 77H. (different die)

3b. Reverse of dīnār of 77H. (different die)

149

150

4. Reverse of one third dīnār, al-Andalus, 102H.

5. Reverse of dīnār of 105H. inscribed ma’din Amīr al-Mu’minīn bi’l-Ḥijāz

6. Obverse of dīnār of 92H. inscribed ma’din Amīr al-Mu’minīn

FOR GOD, EMPIRE AND MAMMON

7. Reverse of half-dīnār of 96H.

8. Reverse of one-third dīnār of 91H.

9. Reverse of dīnār of 78H.

151

152

10. Reverse of dīnār of 79H.

11. Reverse of dīnār of 86H.

12. Reverse of dīnār of 87H.

FOR GOD, EMPIRE AND MAMMON

13. Reverse of dīnār of 97H.

14. Reverse of dīnār of 127H.

15. Reverse of dīnār of 123H.

153

154

16. Reverse of dīnār of 132H.

17. Obverse and reverse of dīnār of 153H.

FOR GOD, EMPIRE AND MAMMON

155

preferences of individual die-cutters rather than official policy, whether of the central mint or of the government itself. It is not easy to state with certainty when other, more significant changes – the turning points, small as they were, in the epigraphic style 70 of mintless Umayyad dīnārs – took place. A really thorough study would begin with separating out the various dies for each year and analyzing their differences. This must be a task for the future. Meanwhile, it must suffice here, as a first step in that direction, to identify some of the stages of the evolution and to assign them a rough date. Significant changes were indeed made over the years in the epigraphy of the Umayyad dīnār (pl. 1 a–b), and they did gradually change the appearance of the entire coin, so that the last of the series (132 H.) could not possibly be confused with the first (77 H.). The obvious place to begin the analysis is therefore the latter 71 coin, and indeed its characteristics are of absorbing interest (pl. 2). What are the orthographic peculiarities of this, the earliest purely 72 epigraphic coin in the Islamic world? Among the diagnostic details of importance are the rā’ or terminal nūn written as a slightly concave stroke; dāl/dhāl written in three distinct ways, one in which the letter encloses a “v”-shape, another in which it encloses an open rectangle, and a third, a compromise between these two, in which the base of the “v” is cut short in rectilinear fashion; and the fact that some letters – wāw, mīm, ṣād, terminal hā and tā’ marbūṭa – are written without the hole which facilitates their identification. One may note too the fudged execution of the terminal hā/tā’ marbūṭa, the tailless mīm, the uneven size of wāw (sometimes with a short tail, and the whole letter on the baseline; sometimes with a tail, long 70  Bates notes that the turning points are hard to identify within less than a decade (“Coinage of Syria”, 210). The investigation of the epigraphy of the dīnārs with named mints – i.e. al-Andalus, Ifrīqiya and Ma’din Amīr al-Mu’minīn [bi’lḤijāz] – is a separate task. 71  The specimen illustrated in Sotheby’s 29 September 1988, lot 22 will be the basis of this analysis. Cf. Sotheby’s Coins, catalogue of 18 April 1994, lot 290, for a different reverse die, and Walker, CMC, 84, no.186, pl.XII. For an easily accessible colour plate of the obverse and reverse of another dīnār of 77 A.H., see T. Falk (ed.), Treasures of Islam (London, 1985), 356, no.369. Yet another dīnār of 77 H. is distinguished by a pellet on the reverse at 3:00. This is presumably to mark the beginning of the marginal legend, like the similar pellet at 3:00 on the obverse of pl.2. The coin is illustrated in colour in Spink Auction 133. Islamic Coins, 8–9, lot 16 (part). 72  For perhaps the finest known specimen, see Coins of the Islamic World in gold, silver and copper. Spink &Son Numismatics Ltd. Zürich. Auction 31, 20 June 1989 (catalogue prepared under the direction of A. P. de Clermont), 38, lot 205; see pl. 3 a–b.

156

or short, which reaches below the baseline; and sometimes a misshapen “o”); the awkward execution of the lām-alif  ligature in three different ways; and a kāf which has the form of a dāl/dhāl but crucially without the final upward or diagonal stroke. Many of the letters are spatulate. A telling if isolated presage of future developments, however, is the exaggerated length of the bā’ of ḍuriba. And even as early as this, the custom of elongating the tooth of a letter preceding or following sīn/shīn (as in sanat, sab‘īa, saba‘īn and bi-ism), or in other cases of possible ambiguity (as in the nūn of dīnār) had entered the repertory of the die-engraver. Nevertheless, the overall impression is one of clumsy execution, if not clumsy design, and indeed the ensemble leaves an overall impression of crudity and roughness, as if the letters had been hacked or carelessly moulded into shape. The ambiguous forms of several letters, and their somewhat tentative construction – terminal rā’ and nūn are obvious examples, being not only interchangeable in shape but also as it were halfformed – points perhaps to inexperience in designing inscriptions of this length for coins. These features have suggested to some the influence 73 of Pahlavi epigraphy, which indeed has a distinctly invertebrate quality. The messages on this coin are not inscriptions which strive for elegance, let alone beauty. Nor, more to the point, are they as legible as they were 74 shortly to become on subsequent dīnārs. In these respects they differ distinctly from other contemporary expressions of Islamic epigraphy such 75 as the milestones of ‘Abd al-Malik and the inscriptions of the Dome of 76 77 the Rock, though there are links with some early but undated Qur’ans. Good calligraphy of course involves more than just the shape of the letters. It is also a matter of spacing. The length of the opening line of the reverse field – Allah aḥad Allah – betrays miscalculation, for the final Allah 78 is significantly shorter than the first, and thus is written in a distinctly cramped way, which also results in the word coming uncomfortably close  Sotheby’s 18 April 1994, 30; Bates, “History”, 249.

73

The difference in script between gold and copper is that between high and low art. 75   See n.16 above. 76  S. Nuseibeh and O. Grabar, The Dome of the Rock (London, 1996), 82–105. 77   See H.-C. Graf von Bothmer, “Frühislamische Koran-Illuminationen”, Kunst und Antiquitäten I (1986), 25 (fig.3), 26 (fig.5) and 27 (fig.6) – though it should be emphasised that in each of these cases the resemblances are confined to certain letters and do not extend to the ductus as a whole. It is certain that closer study of very early Qur’an leaves will produce further parallels. 78   By some 23%. 74 

FOR GOD, EMPIRE AND MAMMON

157

to the bā’ of ḍuriba in the margin. In much the same way, the first line of the inscription in the obverse field – lā ilah illā – is unevenly composed. This time the first sequence, namely lā ilah, is spaced quite tightly, while the next, comprising the one word illā, is laid out with inappropriate looseness, thereby destabilizing that line. In the next line, the designer allows a distinct caesura to develop between the two words allah waḥdahu. This creates a much more pleasing effect, especially as this line extends at both ends beyond those above and below it, even though it has one letter less than they do. Thus visually as well as in terms of content (for its message says “God alone”) it is the centrepiece of the obverse. The last line – lā shārīk lahu, “He has no associate”) – is once again more tightly written, but this time without any lack of internal balance. This same privileging of the centre line in terms of visual impact and content alike recurs in the reverse field, where this line again extends beyond those above and below it at both ends; it reads al-ṣamad lam yalid, “the Eternal; He begets not”. The survival of several obverse and reverse dies for the 77 H. dīnār, each with its own distinctive differences from its peers, makes it hazardous to attempt generalisations. Nowhere is this more true than in the matter 79 of design. The specimen exhibited in Geneva in 1985, for example, contrives, as specimens from other dies do not, to make the obverse field a fairly even square, admittedly at the cost of a spatial imbalance between 80 the exergue and the corresponding space (or “tympanum”) above. These comments on the spacing problems of the first purely epigraphic dīnār raise interesting general points, and it would be as well to consider them at this stage before tracing the subsequent evolution of Umayyad dīnārs. The different hijrī dates of Umayyad dīnārs of course take up a varying amount of space, for they range from 3 letters (mi’at, “hundred”, for 100 81 H.) to 13 letters (tis‘a wa ‘ashrīn wa mi’at, for 129 H.). This corresponds in the marginal design to the difference in space (reading the coin as a clockface) between 1:00–2:00 and that between 1:00 and 5:00, in other words from some 8% of the margin to some 32%. It is not hard to imagine that these swings would challenge the ingenuity of a designer charged with creating a consistently spaced marginal inscription. The figures just quoted  Falk, Treasures, no.369; see Sotheby’s 29 September 1988, p.9 for further references. 80  The same effect, incidentally, was occasionally achieved by the designers of Umayyad dirhams. 81  In the dates between 111 and 119 H. the wa before ‘ashara (“ten”) was dropped. 79

158

of course represent only the polarities; but even lesser discrepancies in the length of the date have disruptive repercussions on the rest of the marginal legend, notably in the way that ḍuriba is written, but also hādha’ldīnār. Clearly the axis, or point of departure, for the marginal legend – the bismillah – was not fixed. Its first letter is always to be found between 12:30 82 83 84 and 2:00, and it normally terminates between 10:30 and 11:50. There are exceptions, as is to be expected, such as 124 H., where the bismillah occupies 12:15–2:15. These irregularities indicate in particular that the designer did not take a hypothetical 12 noon, or for that matter any other spot, as his regular point of departure on the reverse marginal legend. This seems less true of the obverse marginal legend, whose opening word, 85 Muḥammad, conventionally begins as far down as 3:00, or slightly before. As is well known, most Umayyad dīnārs do not bear the name of a mint. The principal exceptions are the dīnārs of the Muslim West. Those which bear the names of mints mention either al-Andalus (presumably Cordoba) or Ifrīqiya (presumably Qairawān). These two mint names (both recorded 86 irregularly from 102 H./720–1 onwards) are accommodated in the reverse marginal legends. The displacement involved is significant even in the case of the standard dīnār: the eight letters of bi’l-Andalus occupy 5:00 87 to 7:00, but in the case of the much smaller one-third dīnār, where space is at a premium, it is really substantial: the same letters occupy roughly from 4:30 to 7:30 – a quarter of the available space (pl. 4). One further mint name is recorded, and this is by far the longest mint name in all the reformed Umayyad coinage: ma’dan Amīr al-Mu’minīn bi’lḤijāz (“the mine of the Commander of the Faithful in the Ḥijāz”). This legend, with its 23 letters, simply could not be accommodated in the standard reverse margin. Thus for the few issues of this design, and this 88 design alone, the arrangement of the reverse field was totally reconfigured 82  Usually 1:00; but 77 (see n.71) and 79 H. have 1:30, while 78 H. has 2:00. This suggests that from the very beginning of the series there was no concept of a standard die axis. For a general discussion of this subject, see Sears, “Pahlavi Imitation”, 151–2. 83 As in 100 H. 84  As in 122 H. 85  Sotheby’s 15 October 1998, lot 114, dīnār of 127 H.; Miles, “SEAD”, pl.XVIII, dīnār of 92 H. In the 77 H. dīnār sold at Sotheby’s on 18 April 1994, there is a small pellet above the mim of Muḥammad which has been interpreted as a device “to show the die sinker where to begin the legend” (Catalogue, 30). 86  Walker, CMC II, 99–102. 87  Walker, CMC II, 102, Ox. 2, fig.19. 88  Miles, RIC, 20–1, no.66 and pl.IV. See the following note.

FOR GOD, EMPIRE AND MAMMON

159

(pl. 5). Instead of 3 comfortably spaced lines of totally religious content – the normal message on the reverse of Umayyad dīnārs, namely Allah aḥad Allah/ al-ṣamad lam yalid/ wa lam yūlad (“God is One. God/ is the Eternal. He begets not/ and He is not begotten”) – the reverse field bears 5 cramped lines of which the first 3.7 have just that message. But thereafter, without even the visual break announced by a new line, the inscription continues directly after yūlad with ma’dan/ Amīr al-Mu’minīn/ bi’l-Ḥijāz (“the mine of the Commander of the Faithful in the Ḥijāz”). The seamless (and totally inappropriate) combination of religious and secular (in this case trivially secular) content in the field of the reverse is striking. This is in fact the only example among Umayyad dīnārs in which secular information is contained in the field, whether obverse or reverse. Clearly, then, it was regarded as a crucial element in the design of the reformed coinage that the obverse and reverse fields should proclaim key religious messages, the very essence of Islam, in legible script. It might be argued that room for the name of the mint – even a name as long as this – might have been made in the reverse margin, which was by convention less crowded than the obverse margin. For example, in the first reformed dīnār of all, that of 77 H., the obverse margin, with an abbreviated quotation from Sura 9:33 ending at kullihi, has 48 letters, while the reverse margin, reading Bismillah. Ḍuriba hādha’ l-dīnār fi sanat sab‘a wa saba‘īn (“In the name of God. This dīnār was struck in the year seventy-seven”) has 33 letters. But this long mint name has 23 letters, and to accommodate an extra legend of this length might well have stretched the reverse margin beyond the bounds of legibility. Some of these dīnārs are inscribed simply ma’dan Amīr al-Mu’minīn, leaving out the location (bi’l-Ḥijāz); they are known for the years 91/709 90 and 92/710 and are very rare. They solve the problem of cramped space in a different way, for they place the mint legend in the obverse rather than the reverse field, which is divided into five lines reading lā ilah illā 89

 Yet these coins were struck at Damascus, for Bates (“Coinage”, 209) has noted that they have the same obverse dies as regular dīnārs of the same year. 90  G.C. Miles, “A Unique Umayyad Dīnār of 91 H./A.D.709–10”, Revue Numismatique, sér. VI, 14 (1972), 264–8; for an illustration, see 265, fig.1. He suggests that “Perhaps the die-engraver in 91 H. did not realize that there was space enough on the reverse to consolidate the Qur’anic quotation and insert a fifth line, bi’l-Ḥijāz” (ibid., 266). Cf. Bates, “Coinage”, 209, n.24. For a colour plate of the issue dated 92 H./710–11, see Spink. Auction 133, 7, lot 8. 89

160

– Allah waḥdahu – lā shārīk lahu – ma’dan amīr /al-mu’minīn. This solution (pl. 6) obviates the uncomfortable mingling of religious and secular information on the same line that mars the coins of 105/723–4 which include the phrase bi’l-Ḥijāz, but the field itself does combine both types of information. With the exception of the coins that mention the caliph’s mine, which broke all rules, the amount of space taken up by the date could vary even more dramatically than that required for the name of the mint on those rare dīnārs that named one. In the context of the limited space available in the margin of a standard dīnār, even the difference between say 90 H. (tis‘īn) and 98 H. (thamān wa tis‘īn) was substantial. Small wonder that it would thereby in some cases – e.g. 129 H. – displace the opening bismillah, which thus begins just before 12:00 rather than the more usual 1:00. It may be that after 100 H., the extra letters needed to give the date imposed greater challenges on the die-cutters. This may itself have been enough to create the impetus for a more tightly ordered design for the whole coin. The layout of fractional and therefore smaller gold coins – half-, onethird- and quarter-dīnārs – sheds further light on the design principles which governed the purely epigraphic coinage. The obverse field of the half-dīnār reads lā ila i/ llā Allah/ waḥdahu (“There is no god but God alone”), while the reverse field reads Bismillah/ al-raḥmān/ al-raḥīm (“In the name of God, the Merciful, the Compassionate”) (pl. 7). The one-third dīnār confines its obverse field to the legend lā ilah i/ llā Allah, leaving out waḥdahu (pl. 8), though its reverse field is identical to that of the half-dīnār, while the quarter-dīnār (though this puzzlingly reads al-thulth 92 93 [“one-third”], like some half-dīnārs) is identical in its legends to the one94 third dīnār, despite its reduced size. The marginal legend on the obverse 91

  It would be interesting to compare the reverse dies of these coins with those of regular dīnārs of the years 91 and 92 H. to determine whether these exceptional coins, like those of 105/723–4, were also struck in Damascus (see n.89 above). 92  Miles, RIC, 19, no. 63, pl.IV. 93  Miles, RIC, 19, no. 4, pl.IV: “both size and weight conform to the standard of a half dīnār”. Other half-dīnārs are appropriately called al-nisf (“half ”): see Walker, CMC, 88–9, no.203, pl.XII. 94  The tables given in Miles, SEAD, 97–104, show that the 120 dīnārs listed varied in diameter between 18.5 mm (two examples) and 24 mm (one example – Miles, 103, suggests that this atypical coin “is possibly a contemporary forgery, but more probably the issue of some temporary mint”), with all the rest measuring between 19 and 21 mm. In the same list, half-dīnārs varied in diameter between 15 and 16.5 mm, with one-third dīnārs measuring between 14.5 and 12.5 mm, while the sole quarter-dīnār measured 12 mm. 91

FOR GOD, EMPIRE AND MAMMON

161

of the half-dīnār is, like that of the standard full Umayyad dīnār, from Sura 9:33, but it omits the last six words of the inscription on the dīnār, 95 ending with the words dīn al-ḥaqq; and, astonishingly enough, this same legend is accommodated not only on the smaller one-third dīnār but also – 96 somehow – even on the quarter-dīnār. A group of these fractional coins have been published by Miles; he records that the half-dīnār measures 16.5 mm in diameter, the one-third dīnār 14 mm and 15 mm, and the quarter97 dīnār 12 mm – as against the full dīnār of c.20 mm. It was therefore an increasingly tough challenge to fit essentially the same message onto ever 98 smaller surface areas. That effort, and the progressive omissions that were nevertheless necessary, show that the die-cutters (or mint-masters) were eager to maintain as much as possible of the sense of the inscriptions selected by the authorities for the standard dīnār. Now that some of the general factors which control the overall design of Umayyad dīnārs have been discussed, it is time to return to the chronological sequence of these coins in the years following the very first issue. The dīnār of the following year, 78 H. (pl. 9), has already corrected 99 some of the infelicities of this first (and hurried?) issue, which may reflect the creation of many more dies for this year instead of only a 100 few for the previous year. In place of four letters with their diagnostic pinholes wiped out, some coins of 78 H. have only one. Dāl/dhāl has been standardized. The uprights are thinner and slightly taller. Moreover, although this is a matter of layout rather than epigraphic form, the spacing  For a discussion of this term, see Hawting, First Dynasty, 61.  For the quarter-dīnār, see Miles, RIC, 19, no.63 and pl.IV – but the coin is

95 96

clipped, so between 1:00 and 5:00 the legend is only partially preserved. Since the coin in its present state is only marginally less wide than the standard one-third dīnār – the main difference is one of weight – there may have been little extra difficulty in fitting this legend on to the quarter-dīnār. 97  Miles, RIC, 19, nos. 62–5, pl. IV. 98  Cf. Walker, CMC, lix; for illustrations, see 102, fig.19 (a dīnār whose obverse has the so-called “Eastern legend” in the field and the margin, but the mint name al-Andalus in the reverse margin, thereby disproving Walker’s own contention – ibid., lvii – that the Western legend “is found in all cases on the gold of North Africa and Spain”), to be compared with pl. XII, nos. 203 and 208. The reasons why the Muslim West had different legends on its coinage would repay further investigation. 99  Sotheby’s 29 September 1988, 9. 100  Thus Walker, CMC, no.186 has different obverse and reverse dies from the 77 H. dīnār sold at Sotheby’s on 29 September 1988, while the 77 H dīnār sold at Sotheby’s on 18 April 1994, lot 290, had the same obverse die as the latter, but a different reverse die.

162

between the bā’ of ḍuriba and of the final Allah at the end of the first line in the field is distinctly more comfortable than in the coins of the previous year. That said, however, it recognisably belongs stylistically with the 77 H. issue. Not so the dīnārs of 79 H. (pl. 10). For by then, a still more confident, consistent and finished style has evolved. The forms of many of the 101 letters are more distinct. Wāw now has a well-developed tail. The upper horizontal stroke of dāl/dhāl, formed on a slightly rising diagonal, projects well beyond the lower stroke. Ṣād/ḍād has a pronounced oblong central 102 opening. Mīm tends to have a marked perpendicular tail pointing down. Terminal hā’/tā’ marbūṭa now has a distinct slender initial upward stroke; in the word Allah this mirrors and continues the massed uprights of all the other letters, and creates in the second Allah of the first line, where the marginal legend leaves less room, a handsome downward continuous curve for the whole word. Terminal rā’ and nūn now have a more pronounced return curve. Not one letter on the reverse blots out the pinhole that helps to identify it, and indeed the pinhole itself is found only once, in the hā’ of hādha. Elsewhere it has become larger, thereby making the whole inscription clearer. The cumulative effect of these changes is that in at least some of the dies engraved for the dīnārss of 79 H. a more assured and soigné epigraphy had developed. It is in the nature of things that the component elements in this change should surface irregularly from one die to the next and that a confluence of the most characteristic changes should itself occur only irregularly. The next change of gear in epigraphic style seems to have occurred around 86 – 87 H.(pls. 11 and 12). The most diagnostic feature of this style is its preference for blocklike shapes for certain letters, notably ṣād and dāl/dhāl. The latter form loses both the slight incline of the upper 103 horizontal stroke, and the upright tail at its upper end. It thus resembles a square or box with its left side open. Since this letter form occurs with more frequency (like alif, six times) than any other on the reverse except lām (which occurs twelve times), this dramatic change in its formation becomes the dominant visual accent of the design, especially since it occurs at least once in each line of the field. The form of the letter 101   For the evidence of other dies see Walker, CMC, no.189 and Falk, Treasures, 360, no.389. 102   The mīm in bismillah is the exception. 103  Nevertheless, these features turn up again in later issues; for example, some dīnārs of 131 H. have both forms (Spink, Auction 133, under that year).

FOR GOD, EMPIRE AND MAMMON

163

dāl/dhāl selected here imparts a strong forward horizontal drive to the 104 inscription, including the marginal legend. This change continued in some dies for several subsequent years – from 88 to 92 H. inclusive at least – and intermittently thereafter. Enough has been said to show that in the first twenty years in which epigraphic dīnārs were produced, numerous refinements were applied to individual letters. These had been added piecemeal to the epigraphic repertoire thus far, and were irregularly maintained. In that irregularity can be detected the personal preferences of individual die-cutters as distinct from a policy decision taken by the mint-master, or whoever 105 was responsible for fixing the general style of these inscriptions. These disparate features included a central downward perpendicular tail added 106 to mīm, a strong sublinear flourish for terminal nūn, creating a bowl107 like shape which embraces part of the next word, a distinct downward 108 109 taper in lām, and a dramatic lengthening of the ḍād in ḍuriba. Seen as a whole, these refinements show that epigraphic style was in a state of flux, but equally they indicate that overall control over the look of these inscriptions was still somewhat loose. A further significant change can be pinpointed to the year 97 H. or so (pl. 13). This marks a moment when the decision had been taken to add an extra element of ornamentation to the epigraphy at large. That measure can fairly be described as a quantum leap of the imagination. It was aimed at making the inscriptions not more legible, but more striking visually. And since it is a change that affected many letters, it transformed the whole look of these inscriptions. It therefore has to be distinguished from the spasmodic refinements discussed in the previous paragraph. So what is new in some dīnārs of 97 H.? 110 Up to this year, it seems, letters – especially uprights, which in Islamic epigraphy were destined to become the vehicle for the most elaborate  Especially between 6:30 and 8:00.  For a survey of how Muslim mints worked, see Bates, “Islamic numismatics”,

104 105

10–13. 106  At least from 86 H. 107  From at least 83 H. onwards. 108  From at least 89 H. onwards. 109  At least from 81 H. onwards. 110  Though of course this is pre-eminently a case where a detailed enquiry based on a close comparison of the different dies for the year 97 H. and those years close to it is needed to substantiate the observations made below. There is no intention here to suggest that the innovations attributed to 97 H. occur on

164

ornament that was applied to inscriptions – were devoid of any added ornament. Nor did this austerity entirely die out, as certain later issues 111 attest, though in some of these the uneven thickness of the shafts is noticeable and betrays a desire to introduce low-key variations of rhythm 112 into the layout of the inscription. This shows that Umayyad die-cutters initiated some quite radical experiments, some of them dead ends, in the search for visual variety. From at least 97 H. onwards, however, notably in the field (where it 113 would be more effective visually) but also in the margin of the reverse, 114 certain letters, especially alif, lam and dāl/dhāl, acquired a bulge at the top or at the end(s) of the letter. It could also be described as a kind of topknot or beading. Sometimes this little bulge is cut off from the rest of the letter by a thin line. Some issues use this device lavishly, others sparingly. Its effect can be compared to that of the serif in classical Roman epigraphy – a way of adding extra emphasis and dignity to a letter, and lending the inscription as a whole extra plasticity. This beading is often found, as in the dīnār of 97 H., in association with a distinct tapering in the lower part of the upright, and with a further bead at its base, for example in the opening “Allah” in the first line of the reverse field. As time passed, its form became more marked, projecting strongly from the main shaft on 115 both sides and sometimes resembling a tiny lozenge. The opening lam-alif of the obverse field was an obvious candidate 116 for this ornamentation, which served to emphasise the resounding proclamatory negative of the message itself – lā ilah illā – Allah, “there is no god but – God”. The intention to stress this key opening message may, incidentally, explain why on the obverse of the standard Umayyad dīnār the first line of the field typically takes up much more room than the two all dies of that year – see for example Walker, CMC, no.212, also of 97 H. but without the key characteristics outlined below. 111   E.g. 107, 122, and 127 H. 112   E.g. 107 H. 113  Further research is needed not only to establish whether this feature was equally characteristic of the obverse, but also to chart how widely and quickly it became an accepted feature of epigraphic style on Umayyad dīnārs. 114   This feature was perhaps first used, though only in the letter dāl, and then only for the termination of the upper horizontal stroke of that letter, in 96 H. (see Walker, 90, no. .210, pl.XII). 115   As in some dīnārs of 101 H. 116   E.g. Walker, CMC, 95, no.230 of 110 H., pl.XIII. Yet a slightly later dīnār of 114 H. (ibid., 96, no. 234, pl.XIII) has much less marked terminals. This illustrates the characteristic ebb and flow of new ideas in Umayyad dīnār design.

FOR GOD, EMPIRE AND MAMMON

165

lines which follow it, as can be seen by comparing the opening lam-alif in the first line with the same but much smaller form at the beginning of the third line. Much of the beauty of that opening line, of course, lies in its balance, for it opens and closes with a lam-alif, always potentially one of the most majestic ligatures in Arabic calligraphy. The greater height of the uprights in the first line vis-à-vis those in the second and third lines is also a potent ingredient in the success of this sequence, lending extra dignity 117 and grandeur to these words. With this attempt to dress up the appearance of most of the letters on the coin by modest ornamentation in a single recognisable mode, the 118 epigraphic evolution of the standard Umayyad dīnār stalled. This is not to deny that certain grace notes still lay in the future. Among them may be cited the increasingly dramatic and flamboyant terminations for the letters 119 120 yā and ‘ain in the dates, or a very marked wāw with an elongated straight 121 back and a horizontal extension along the baseline, or the haphazard adoption of the much-flattened and much-elongated type of dāl/dhāl with 122 an upward perpendicular extension on the upper stroke of the letter. This last form marks one of the few points of congruence between the epigraphy of very late Umayyad dīnārs and the dirhams of what might be called the Wāsiṭ style. Moreover, it is plain that within the (probably) Damascus mint, more than one style of epigraphy was adopted from one year to the next in the closing years of the Umayyad dynasty. The differences could be radical. One of the most handsome issues in the entire sequence was struck in 127 H. (pl. 14). It contrives, perhaps better than any other specimen, to use the surface area of the field to the fullest advantage. The result is that the uprights are as tall as they well can be within the parameters of a long three-line text in a confined space. This gives the epigraphy a stateliness rare in Umayyad dīnārs. Nor is this all. The rhythm of the main uprights –   Bates notes that the height of the inscription in the obverse field of Umayyad dīnārs increases markedly, but that it remains unclear whether this process is gradual or spasmodic (“Coinage”, 211). The relatively greater height of the uprights in the opening line is maintained even in the cramped space of the coins whose obverse field mentions the “mine of the Commander of the Faithful”; see n.94 above. 118  That is, the dīnārs with no mint name, and thus presumably struck in Damascus. 119   E.g. 112 H. 120   E.g. 109, 114, 117, 119, 124 and 129 H. 121   Circa 103 H. onwards. 122  127 H.; less marked in 123 and 126 H. 117

166

lam and alif – is taken up by such subsidiary accents as the terminations of the dāl, the tail of the mīm and the long straight back of the wāw, so that the overall impression of powerful verticality is strengthened. Thus the letter forms support each other and play their full part in the overall design. At the same time, the die-cutter managed to open up enough room on each line to create a small but crucial space between most, though not all, of the words in the inscription. He also contrived somehow to create extra space between the lines themselves. Thus this coin’s reverse field looks miraculously uncluttered, a pleasant change from many Umayyad dīnārs. In the third line there is even a marked caesura between the two syllables of yūlad, which recalls that characteristic feature of many early Qur’ans in which the calligrapher transcribed the words so as to leave spaces between the syllables rather than only between words. This in turn recalls the way that the words of the shahāda are cut up and radially distributed, with each successive sequence below the next, in certain Arab-Sasanian dirhams struck 123 under al-Hajjāj. Finally this coin, with its preternaturally exaggerated verticality, the thinness of its letters and its greatly drawn-out dāl in yūlad, and mīm in mi’at – letters which help to establish the continuous circle of the marginal legend – partakes of the mannered epigraphy of the Wāsiṭ school. And it has no trace of the beaded terminations so characteristic, say, of the dīnārs of the flanking years 126 and 128 H. An almost equally handsome coin of 123 H. (pl. 15) makes quite another kind of impact, even though the differences seem slight. Yet their cumulative impression is what counts. The spacing is slightly less generous, both within and between the lines, so that the letters seem to make not separate words but a single long one, especially in the first two lines; the verticality of the uprights is significantly less marked; the letters are chunky rather than attenuated; and many of them lean forward distinctly, which imparts a marked horizontal drive to the inscription of the field. Moreover, the ample space above and below the field, as well as around it (an unusual feature and one calling for very careful design), draws the text block together. Thus while its epigraphic style is less distinguished than that of the 127 H. dīnār, it displays exceptional clarity, simplicity 123  E.g. an example struck at Bīshāpūr in 77/696–7; see J. Sourdel-Thomine and B. Spuler, Die Kunst des Islam (Berlin, 1973), pl.72c. G.C. Miles, “Some ArabSasanian and Related Coins”, ANS, MN VII (1957), 192, publishes two coins of the “miḥrāb and ‘anāza” type whose obverses give two variants of how the shahāda is parcelled out among the four constituent quadrants of the margin.

FOR GOD, EMPIRE AND MAMMON

167

and harmony and may claim to attest that rare moment when design and content are interdependent and in perfect counterpoise. The last of the series, a dīnār of 132 H. (pl. 16), is as different from those of 123 and 127 H. as each is from the other. Here the fashion for terminal beading has been taken to excess, with the result that the letters of the entire reverse field are uncomfortably studded with bulging dots. Since this feature extends also to the bases of the letters, the baseline is corrugated continuously in the first two lines and towards the end of the third. The minimal spaces between words make the legend of the field resemble one continuous word. The beading and corrugation extend also 124 to the reverse marginal legend. Yet as an early ‘Abbasid dīnār of 153/770 indicates, this particular fashion reached its apogee a generation after the 125 fall of the Umayyads (pl. 17). Here, most upper and lower terminations of letters are marked with a stud, and in some cases there are further studs in the body of the letter itself. These brief analyses, preliminary and tentative as they are, and explicitly 126 not based – as ideally they should be – on extensive die analysis, should suffice to indicate that the style of writing on the reformed Umayyad dīnārs minted, presumably at Damascus, in the last fifty years or so of the dynasty’s rule did undergo some significant changes, though a tolerably strict series of conventions ensured that no dīnār strayed very far from the template laid down from the outset, in 77 H. For some reason the formula employed for dīnārs permitted a greater degree of experiment and variety than that employed for the dirhams struck at Damascus over much the same time-span. Equally, it is hard to miss the fact that the specific tricks of epigraphic style that developed in the mintless dīnārs are not reflected in the dirhams of Damascus, which by contrast are conservative to a fault. One further obvious conclusion to draw from these findings is that there was very little contact between the diecutters who specialised in dīnārs and those who specialised in dirhams, even though in all likelihood they worked in the same city. The reformed coins of the Muslim West, incidentally, tell a rather different story. The Umayyad dīnārs that bear the mint names of Ifriqiya and al-Andalus stick closely to the style of the Damascus dīnārs, which is not surprising. But the later Umayyad dirhams from these two mints adopt the peculiarities of the Wāsiṭ mint, and not  Cf. Walker, CMC, no.252 and pl.XIII, a very similar reverse die.  Sotheby’s, 30 March 1995, 14, lot Ex 56. 126  A task more practicable for numismatists than for art historians. For general 124 125

comments, see Bates, “Islamic numismatics”, 5–7.

168

the Damascus one, nor for that matter an individual style different from 127 both of these. Implications of the reform These are numerous, and it must suffice merely to indicate them here in general terms. First, the reform implies the existence of an effective, highly organised bureaucracy. Without such an instrument to hand, there would have been no point in launching the reform in the first place. And this was no tentative reform; once silver had been included, it was a knock-out blow, whose repercussions were very widespread indeed. And when the purely epigraphic design began to spread to the copper coinage as well, 128 it reached practically every citizen. The reform thus implies a tightening 129 grip of the central authority on the various aspects of government. Next, it implies an urgent need for ready cash. Building activities were an obvious case in point. These involved not merely high-profile structures like the Dome of the Rock and the Great Mosque of Damascus, the latter 130 of which gobbled up the land tax revenue of Syria for seven years, but also the very extensive and therefore costly exploitation of agricultural land in Syria. That included, for example, the building and maintenance of canals, cisterns, dams and aqueducts. All this was not just a matter of extravagant living but an integral part of an Umayyad policy to colonise the land itself. The standing army was a still more significant consumer of cash. Indeed, the coinage reform highlights the power of the military in Umayyad society, for – as Kennedy has demonstrated – the need to pay salaries or pensions to those soldiers whose names appeared on the military register, and to pay them in cash at an average minimum rate of 200 dirhams annually, created a pressing need for vast quantities of minted money. Obviously, much depends on exactly how many men appeared on that register – but a mere 5,000 men, say, would create a need for a million  See P. Balog’s review of Walker, CMC (RN Sér. VI/1 [1958], 230) for the discovery in excavations at Wāsiṭ of jars filled with uncirculated dirhams bearing the mint names al-Andalus and Ifriqiyya. 128  See H. Bone, The Administration of Umayyad Syria: the Evidence of Copper Coins (unpublished Ph.D. thesis, Princeton University, 2000). 129 In this context Wellhausen (Arab Kingdom, 218, n.2) quotes a ḥadīth ascribed to the Prophet and cited in Yaḥyā b. Ādam, Kitāb al-Kharāj, 52: “Iraq obstinately sticks to its dirham and qafîz, Syria to its dinar and modius, and Egypt to its dīnār and ardab; ye are returning to your old divisions and lack of unity, to the old particularism”. 130  For a detailed discussion of this matter, with detailed citation of the sources, see Creswell, EMA (2nd ed.), I, 151. 127

FOR GOD, EMPIRE AND MAMMON

169

dirhams annually, and on occasion the numbers on the register must have 131 been very much more than that. Those soldiers would be the largest single body of consumers in the economy, and their military authority en masse as well as their purchasing power would have made them the ideal instrument for putting this mountain of specie into rapid circulation. An anecdote about the capture of Paikand in Central Asia in the early 8th century, which resulted in the melting down of a golden idol there and the minting of 150,000 gold dīnārs from its bullion, clearly points in this 132 direction. A third implication of the reform is that the Muslims had reached a new plateau of confidence in matters economic, political and religious. The reform reveals a desire not only to control the economy but to dominate the population psychologically. On the political plane, it reveals the confidence to confront the subjects of the Umayyad state with a coinage that very few of them could have relished. It also reveals a hardening of attitudes to Byzantium. Clearly the government felt that it could handle trouble from either source. And in the matter of religion, this coinage marks the moment when Islam moves decisively onto the offensive. The coins no longer balance the secular and the religious; the stakes are now higher, with religious content dominating both sides of a coin. The coin, thanks to its “nation”-wide circulation, becomes an instrument for proclaiming the faith, even for conversion. In modern terms, it is like a commercial, or a party political broadcast, on TV – with the difference that you cannot turn it off. Islam was what bound this religion, society and polity together, and the coins reflect that reality. Finally, from the point of view of an art historian, it is fascinating to note that the entire period of initial copying, slow transformation and eventual creation of a distinctively Muslim style – a process that in the visual arts as a whole took the best part of two centuries – is telescoped in the medium of coinage into little over sixty years. The reformed coinage in gold (and silver) has one language, one era, one visual theme; and, unlike earlier Arab-Sasanian 131   Kennedy speculates that if the 80,000 men on the dīwān or military register of Baṣra at the end of the 7th century had been paid at this rate, 16 million dirhams would have been needed annually (Armies, 67). He calculates that the total number of names on the register throughout the Umayyad empire was of the order of 300,000 men, which would represent an annual salary or pension bill of some 60 million dirhams (ibid., 68). 132   Ibid., 70. The anecdote suggests that a temporary mint was set up (see n.99 above), which further implies that Damascus was not the only mint to issue gold coins.

170

and Arab-Byzantine coins, it contains no mention of the ruler, giving the Qur’an pride of place. It is, oddly enough, in coins that Islamic art first found its feet; and that is a striking reminder of the confident selfawareness of early Muslim society.*

*It is a pleasure to dedicate this article to Christian Ewert, whose life’s work has been dedicated to the minute exploration of western Islamic architecture. The remarkable results that he has achieved prove the accuracy of the old adage that “le dieu est dans les détails”. This article is an attempt to apply the same principle to coins.

IX The soul of Islamic art

H

ow should Westerners approach Islamic art? There is no single, simple answer, for such art is the product of a complex and sophisticated civilisation spanning a vast area from Spain to Indonesia, and set within a time frame of almost fourteen centuries. Within those spatial and temporal boundaries lie a bewildering variety of languages, societies and ethnicities. Why, then, should the art produced within the Islamic world be readily reducible to a few easy generalisations? After all, we do not apply such a sound-bite approach to the art of Europe from the end of antiquity, yet the time frame is comparable and the geographical area is much smaller. It is only when the shoe is on the other foot, and that foot is pinched, that one begins to detect some of the unacceptable assumptions underlying even the apparently innocent categories of art history. For art history is a discipline invented in the West, and to date practised almost exclusively by and for Westerners. This makes it ill-equipped to explain the art of nonWestern cultures. Even the term ‘non-Western’ is very loaded. Categories such as ‘Islamic art’, ‘Buddhist art’ or ‘Hindu art’ continue in regular use today in art-historical writing, but the equivalent umbrella term ‘Christian art’ has long been discarded in favour of its multiple sub-categories. These reflections add up to something more than a mere nod in the direction of political correctness. A serious effort of the imagination is needed if Westerners are to venture outside their comfort zones and attempt to take on the art of another culture on its own terms. In the case of the Muslim world, even the terminology is unhelpful. Previous generations played with, and eventually discarded, such definitions as ‘Moorish’, ‘Saracenic’ and

172

‘Muhammadan’ art. Yet the term in current use – ‘Islamic art’ – suffers from a serious ambiguity. Some interpret it as the art of the Islamic religion, and confine it to a sacred context; others (and they are in the majority) use it to describe the art of the entire culture of the Muslim world. As such, it covers art produced in that world by non-Muslims, notably Christians and Jews. But in default of a better term, it has achieved general acceptance. Perhaps the key point to note is that all the terms that have ever been used to describe this art have been general rather than particular. The time, it seems, is not yet fully ripe to deploy even the most basic subdivisions, except for a specialist audience. Thames and Hudson’s invaluable World of Art Library series includes any number of titles featuring specific Western artists, schools, periods or national traditions. But there are no titles on, say, Arab or Persian or Turkish art; nothing devoted even on a pan-Islamic scale to architecture, painting, ceramics or carpets; nothing on the art of a single major Islamic dynasty. And this list of gaps could easily be extended. All this neglect has resulted in a pretty meagre basis for an intelligent understanding of Islamic art, and an unthinking Eurocentricity – simply unacceptable in today’s globalised world – is to blame. So how should Westerners eager to get to grips with Islamic art prepare themselves? First, they need to jettison all kinds of preconceptions about what constitutes art, since those preconceptions have been fashioned to explain Western art. The named artist, with a well-documented personality and abundant surviving works, scarcely figures in the annals of Islamic art. Nor does easel painting (the staple of art galleries throughout the Western world) or sculpture, or the classical tradition (whether as form or myth) or naturalism as an ideal. Instead, the ‘minor’ or ‘decorative’ arts, such as ceramics, metalwork, glass, ivory, textiles and so on, are given major status. Some persistent misconceptions about Islamic art must also be demolished. They include the ideas that it forbids the depiction of living figures – a notion easily dispelled by leafing through any illustrated general book on Islamic art; that its content is entirely religious; and that its principal characteristic is the even filling of all the available space. In fact, Islamic art is very much more varied than these misconceptions might suggest. Not surprisingly, the religious element is strong and finds expression in the perennial emphasis placed on calligraphing and illuminating the Qur’an; on the primacy within Islamic architecture of the mosque, the madrasa (theological college) and the funerary shrine; on the pervasive role of paradisal symbolism, for example in carpets and gardens; and on Sufism (Islamic mysticism), which generated specific building

THE SOUL OF ISLAMIC ART

173

types and whose ideas permeate much of Persian painting. But the secular element is equally strong. Some of the most famous Islamic buildings, such as the Alhambra in Granada and the Topkapi Saray in Istanbul, are royal palaces. Princely titles decorate both buildings and objects of daily use, from textiles to incense burners, thereby stamping royal authority on objects large and small in the private and public domain alike. Citadels and city walls expressed that same authority, which was memorialised even in death by mausolea, from the tombs of nomadic chieftains to the Taj Mahal. By degrees there developed a set of images that expressed the good life of the elite, and this became a standard iconographic cycle in most varied contexts, from palace frescoes to luxury ceramics. These scenes included banqueting, drinking, dancing, music-making, chess- or backgammonplaying, wrestling, hunting and similar entertainments. They were echoed in epigraphic form by decorative inscriptions calling down blessings on the patron. Equally popular were images that evoked magic and the heavens, whether through the symbolic imagery of light, zodiacal and talismanic themes, or hybrid creatures like harpies and griffins. In manuscripts, illustrations could show genre scenes, or visual narratives (often heroic or romantic in nature) accompanying poetic or prose texts, or scientific subjects, dealing for example with botany, star lore or military techniques. So much for the subject matter of Islamic art. But the role of ornament is perhaps even more important. The world of Islamic art has been termed ‘a draped universe’, and it is ornament that transforms the surface of things. The modes of that ornament are vegetal, geometric and epigraphic. Plant décor, particularly the arabesque – whose name betrays its origins – is pervasive, for it can be used with equal facility on a huge or a tiny scale. Mathematical calculation of a sophisticated order underlies much Islamic architecture, while geometrical ornament, with its implications of infinity, its innate sense of order, its balance between solid and void, its smooth repetitions, graces many a surface. As for writing, it transforms the objects it covers into vehicles for religious or political statements, so that many of them – from coins to monuments – become a kind of book. Qur’anic quotations in particular sanctify the humblest of objects as well as the grandest mosques. And this ornament, whatever mode it uses, whatever its scale or context, serves to channel energy, to infuse a surface with colour and movement, to dissolve mass and to encourage contemplation. Here, if anywhere, is the soul of Islamic art.

X The Syrian Connection: Archaic Elements in Spanish Umayyad Ivories

T

his paper will seek to identify the sources of certain aspects of Spanish Muslim ivories in the art of Umayyad Syria. The notion of the links between the Muslim arts of Spain and Syria is a familiar one in general terms, but it has not been investigated in appropriate detail. Perhaps the best approach is to start by sketching a context for this artistic dependence of Muslim Spain on Umayyad Syria so as to explain why it occurs in the first place. Most of this paper will then be devoted to looking at the evidence for those links provided by two specific aspects of these ivories, namely epigraphy and vegetal ornament. It is worth pointing out, however, that the connections extend well beyond these two themes. To limit the inquiry to Spanish ivories, they include details as disparate as 1 2 3 hairstyles (including the cut of the beard); decorative borders; roundels 1  Schlumberger 1986, pl. 36. I should like to express my gratitute to Margaret Graves for executing the beautiful drawings that accompany this article. 2  Cf. cat. no. 20d with the detail of the face of the prince from Khirbat alMafjar (Hamilton 1969, pp. 62-65) or some of the sculptures from Qasr al-Hair West (Schlumberger 1986, pls. 67a and 70b and 70e). 3  For braided mouldings, compare cat. nos. 4a, 12a, 13a, 14a, 15a, 16a, 18a, 21a, 22a, 23a, 25a, 25b, and 28a with Creswell 1969, I/2, pls. 82d (al-Muwaqqar) and 108c (Khirbat al-Mafjar); for mouldings which change their design in the course of a single band, compare cat. nos. 2a and 13a with Creswell 1969, I/2, pls. 122 (Mshatta; three changes) and 138a (Qasr al-Tuba) – one might add here examples where the direction of the design is inverted even though the design itself is not changed (cat. no. 7a), while in other cases the design is constant within a single arch although it varies from one arch to the next; for mouldings of pearl roundels, compare cat. nos. 17a and 24a with Creswell 1969, I/2, pls. 89a (Qasr al-Hair West) and 102e (Khirbat al-Mafjar, among many other examples);

THE SYRIAN CONNECTION

175

that are lobed, braided, and both at once, and which may have taken their 7 inspiration from textiles; birds and animals trapped in a thicket of dense 8 9 leafy growth; affronted mythical creatures; the theme of the conquering 10 11 12 lion; the lion throne; inhabited scrolls; and, most pervasively, the cycle 4

5

6

for those which feature pearl roundels with curved tails resembling the repeated figure “9”, compare cat. nos. 10a and 11a-11e (on both ivories they are addorsed) with Creswell 1969, I/2, pl. 89b (Qasr al-Hair West; here they are also addorsed); for mouldings resembling the repeated letter “c” or a comma, compare cat. nos. 3a, 4a, and 7a-c with Creswell 1969, I/2, pl. 130 (Mshatta); and for dot mouldings, compare cat. nos. 2a and 16a, 19a, 25b, and 28a, with Creswell 1969, I/2, pl. 135b (Mshatta) and Schlumberger 1986, pl. 57b. 4  Compare cat. no. 5b with Creswell 1969, I/2, pls. 109b (Mshatta façade) and 138a (Qasr al-Tuba). 5 Compare cat. nos. 13a, 22a, and 23a with Creswell 1969, I/2, pls. 111c and 111f (Khirbat al-Mafjar). 6  Compare cat. nos. 11a-e, 12a-b, 20a-c, and 21a with Hamilton 1959, pls. XXIV (P6), LXXIX (no. 25), and XCI. 7  Baker 1995, pp. 10-11, 42-43. 8  Compare cat. nos. 4a, 10f, 15a, 16a, 24a, 26a, and 29a with Hamilton 1959, pl. XLIX/7, and Creswell 1969, I/2, pl. 129 (Mshatta). 9  Here again, textiles are a likely source of inspiration; see Weibel 1972, pls. 5960, 63-64, 67-68, and 73. For examples on Cordovan ivories, see cat. nos. 10c, 11c, 12a, 13a, 20c, 22b, and 25a; for Umayyad Syrian ones, see Creswell 1969, I/2, pls. 123 and 125 (Mshatta). In what follows, the term “Umayyad”, unless qualified by the prefix “Spanish”, will denote the Umayyads of Syria. 10  One should draw a distinction here between the use of this theme at centre stage, notably in the famous mosaic in the diwan of Khirbat al-Mafjar (Hamilton 1959, frontispiece), and its use as one figural theme among many, with no particular attempt to single it out, as at Qusair ‘Amra, Almagro et al. 1975, Pl. XXXIII, bottom right (indistinguishable); for a drawing of this image, see Musil 1907, II, pl. XXX, reproduced in Sourdel-Thomine and Spuler 1973, pl. 34. For discussions of the lion mosaic at Khirbat al-Mafjar, see Hamilton 1959, pp. 337-339; Ettinghausen 1972, pp. 44-46; and Behrens-Abouseif 1997. For the conquering lion theme in general, see Ettinghausen and Hartner 1964; for the same theme in the Cordovan ivories, see cat. nos. 10d, 11a, 11c, 19a, 20c, and 25b. 11   Here the connections are, it must be admitted, a little weaker, for while this theme turns up in straightforward guise in the Spanish ivories, as in cat. nos. 11b and 20d, at Khirbat al-Mafjar the prince stands on a lion-borne pedestal rather than on a proper throne, Hamilton 1959, pl. LV/1 and 5. 12   As befits a theme which dominated the decorative repertoire of the Levant from early Christian to early Islamic times (see Dauphin 1974), this motif pervades the iconography of Spanish Muslim ivories. Yet while it occasionally occurs in its authentic Near Eastern form, i.e. as a series of scrolls issuing from the stem of a plant, with each scroll containing some living creature (cat. no. 16a), for the most part the theme has metamorphosed into a sequence of inhabited roundels that have no connection whatever to a plant (cat. nos. 11e, 12a, 13a, 20a, and 22a). Most truly vegetal scrolls on these ivories contain leaves rather than creatures (cat. nos. 2a, 7a, 9b, and 14a). For representative Umayyad examples, see Creswell 1969, I/2, pls. 119b-130 (Mshatta) and, for the gradual disintegration of

176

of princely entertainments as developed at Qasr al-Hair West and Qusair 14 15 16 17 18 ‘Amra: wrestlers, musicians, flabellum-waving attendants, dancers, 19 20 21 peacocks, hunters, and figures of authority seated in majesty. That vast panorama of themes is emphatically not the subject of the present paper, but clearly it is more than rich enough to justify a study all to itself. Moreover, precisely because the borrowings from Umayyad Syria 13

this ancient theme under the pressure of the innovative tendencies of Umayyad sculptors, Hamilton 1959, pl. XLIX/7. 13  Schlumberger 1986, pp. 15, 21-23 and pls. 64, 65c, 67, 68a-b, 68d-e, 69c-d, 70, 81a-b, 81d-e, and 82. 14  Almagro et al. 1975. 15  See Almagro et al. 1975 (clearer in the drawing by Mielich: see Musil 1907, II, pl. XX, and Spuler and Sourdel-Thomine 1973, colour pl. IX) and cat. no. 11d. 16   The most famous examples of this theme in Syrian Umayyad art are the flautist and lutanist on one of the floor frescoes at Qasr al-Hair West; see Schlumberger 1946–1948, pp. 88-90 and 98 and color pl. B. See also the lute-playing “bear” at Qusair ‘Amra, Almagro et al. 1975, pl. XXXIX, as well as lutanists and flautists; Almagro et al. 1975, pls. IIIb and XXVIIb and VIb. For Cordovan examples, see cat. nos. 11b and 22a. 17  The best Syrian example is the attendant fanning the prince in the throne niche at Qusair ‘Amra; see Almagro et al. 1975, pl. XI, and Spuler and SourdelThomine 1973, pl. 33. For this theme in the ivories of Córdoba, see cat. no. 20d. Cat. no. 11b perhaps shows a flabellum, or some kind of fan, but it is doubtful that the figure holding it is a servant, since he is seated. 18   Once again, Qasr al-Hair West (Schlumberger 1986, pl. 82) and Qusair ‘Amra (Almagro et al. 1975, pls. VIa and XXVIIb) offer the key material, although this particular theme is found only once in the Cordovan ivories (cat. no. 24a). 19  Traditionally a symbol of luxury; compare cat. nos. 4a, 11d, 11e. 13a, 19b, 21a, 22a, 22b, and 24a with an example from Qasr al-Hair West (Schlumberger 1986, pl. 69c, whose diagnostic tail, though very fragmentary, suggests that this image represents a peacock). 20   It has to be admitted that, while both the Syrian and the Spanish Umayyad traditions use the theme of hunting in their royal iconography, they do so in rather different ways, and their closest link is in their joint use of the theme of the hunting dog (compare cat. nos. 10e and 12a with Almagro et al. 1975, pls. XXVIII-XXIX). Apart from Qusair ‘Amra, where the hunt is depicted repeatedly, the outstanding example of this theme in Umayyad Syrian art is to be found in a floor fresco from Qasr al-Hair West; Schlumberger 1946–1948, pp. 97-98 and pl. B. But this aspect of the princely cycle developed apace after 750, which is why the falconer, whether on foot (cat. no. 11a) or mounted (cat. nos. 11e, 12b, 13a, and 19b), is not to be found in Syrian Umayyad art. Nor is the wielder of the flywhisk (cat. no. 20d), the jouster (cat. no. 20b), or the wine-drinker (cat. nos. 11b, 12a, 20d, and 22a), despite the well-known partiality for wine shown by several Umayyad caliphs; Hillenbrand 1982, pp. 12-14 and pp. 28-29. 21  Compare cat. no. 20d with the examples from Qasr al-Hair West (Schlumberger 1986, pls. 64a, 67a, and 81b) and Qusair ‘Amra (Almagro et al. 1975, pls. IXc and X); cf. Musil 1907, II, XV, and XIX; and Spuler and SourdelThomine 1973, pl. 33.

THE SYRIAN CONNECTION

177

in these remarkably disparate areas are not slavish, but rather take each inherited theme in new directions, their close study would shed much light on the interaction between the Syrian and Spanish traditions and on the mechanics which governed the transfer of ideas. But all that is for a future study. Context Now for the more general context. Creswell, with typical thoroughness, long ago tabulated the startling dependence of the Great Mosque of Córdoba – the key work of art in 10th-century Andalusia – on Syrian models, which he sees in the horseshoe arch (which quickly became a logo for Spanish Islamic identity), the parallel gable roofs, the arcades in which two columns alternate with a pier, the stepped crenellations with 22 plant ornament, and the axial placement of the tower. To these factors should be added several others. Whether they include the lofty square tower itself is perhaps doubtful, though there is no clear evidence either 23 way, but other more likely borrowings can be cited: the dome in front of 24 25 the qibla, the geometrically patterned panels, claustra, or window grilles, 22 

Creswell 1979, p. 156.

  For perhaps the earliest example of this form in the area of Greater Syria,

23

i.e. the tower of Harran, see Bloom 1989b, pp. 57-58 (where he records the belief of Herzfeld and of Creswell that the adjoining mosque was Umayyad, and that Creswell also suggested that some of the work dates from the period of alMa’mun; Bloom himself dates it to c. 830) and figs. 25-26. But he argues, “By 170 / 786 multiple tall towers called manara surrounded both of the holiest shrines of Islam” (Bloom 1989b, p. 44). For the pre-850, probably square, northern tower at the Great Mosque of Damascus, see Bloom 1989b, pp. 60-61; cf. Bothmer 1987, p. 6, colour pl. II and pl. 1 for the depiction of a corner tower in the early (most probably Umayyad), but unfortunately undated architectural frontispiece to a Koranic manuscript found in the Great Mosque of Sanä. For a meticulous assessment of the introduction of the lofty square minaret into Muslim Spain, and of the role of the 47-metre tower attached to the Great Mosque of Córdoba in 346 / 958 by ‘Abd al-Rahman III, see Bloom 1989b, pp. 103-106 and figs. 4650. 24  As at the Aqsa mosque in Jerusalem; see Hamilton 1992, p. 143. The massive and lavish rebuilding involved in “Aqsa II” makes much better sense in the context of the patronage of al-Walid I than in that of the niggardly al-Mansur. See also the Great Mosque of Damascus: Creswell 1969 I/1, pp. 168-169, figs. 83-84, 86-87, and 90 and pls. 40a and 42a. Córdoba offers perhaps the earliest well-preserved example of how this feature developed to become a transept – the T-shape that was destined to characterize Western Islamic architecture. 25  Schlumberger 1986, pls. 72, 74, 75b, 76-79d, 80a-e, 81a, 81c, and 84b-e; Hamilton 1959, figs. 185-210 and 238-250 and pls. LIII, LVIII-LXIV/1, and LXIX; Creswell 1969, I/1, pp. 173-174 and pp. 202-204, figs. 92, 118, 127, and

178

the ajouré style of densely carved stone- and stucco-work, the two-tone 27 28 voussoirs, the two-tier elevation within the mosque, as well as the types of epigraphy and of vegetal ornament which are the subject of the present paper. Perhaps one should focus above all on the gross disparity of the qibla of the Córdoba mosque, which favours the south rather than the 29 west, a constant reminder, throughout successive transformations, of the Syrian heritage. Then there is the tenacious if dubious account of how in Córdoba ‘Abd al-Rahman I followed the example of his forebear al-Walid 30 I in Damascus, buying the local Christian church and then demolishing it 31 to build his Great Mosque on the same site. Even if this story is not true, it is still powerful evidence of how Syria was the role model for Umayyad Spain. The overtly Syrian character of the Cordovan mosque asserted to the city at large every Friday where the loyalties of the Umayyad rulers lay in their multi-tribal and multi-confessional state. It should be remembered, too, that the annual celebration of the Hajj, which often involved Muslims from distant lands visiting Jerusalem and even Damascus, would have kept memories of the great Syrian buildings alive. Creswell has also suggested a context for these borrowings in the flooding of Spain with Syrians even before the advent of the first of 32 33 the Umayyad rulers, ‘Abd al-Rahman I, who was himself born in Syria. He notes that the Berbers and urban Arabs cried to Abu ’l-Khattar, the 26

416 and pl. 59 (the marble and stucco window-grilles of the Great Mosque of Damascus); compare Brisch 1968. 26   As at Mshatta (Creswell 1969, I/2, pl. 120, is a representative example), Qasr al-Hair West (Schlumberger 1986, pls. 72-80) and Khirbat al-Mafjar (Hamilton 1959, pls. XLVII and LX/1 are typical). Compare the work at Madinat al-Zahra (Ewert 1996; Gómez-Moreno 1951, pls. on 78-79, 61, 83, and 85-87) and the panel beside the mihrab in the Great Mosque of Córdoba (best large-scale colour picture in Goitia 1965, pl. 24). 27   As at the Dome of the Rock; Nuseibeh and Grabar 1996, pls. 112-113. 28   Admittedly the version of this idea at Córdoba is distinctively different from the clerestory solution adopted at Damascus (Creswell 1969, I/1, pl. 62a), and one must also take into account the possibility that the builders at Córdoba were inspired by Roman aqueducts of the type found in Mérida and Segovia; Creswell 1979, II, p. 157 and fig. 150; New York 1992, pp. 13-14. 29   Creswell 1979, II, fig. 143 and Creswell’s own plan (fig. 146) suggests that the qibla faces due south. King 1999, p. 125, cites a number of possible orientations for mosques in Córdoba mentioned in a medieval treatise from Andalusia; these include the readings 180 (i.e. due south), 150, 135, 120 and 113 degrees. 30  Creswell 1969, I/1, pp. 152, 181, and 196. 31  Creswell 1979, II, p. 139; New York 1992, pp. 11-12. 32  Creswell 1979, II, pp. 156-157; Lévi-Provençal 1948a, pp. 55-59. 33  Lévi-Provençal 1950-1953, vol. 1, p. 95.

THE SYRIAN CONNECTION

179

governor of Andalusia between 743 and 747, “We swear obedience and fidelity to you, but we can no longer endure in our midst these hordes of 34 Syrians.” Groups of Syrian imigrants from Damascus, Hims, Qinnasrin, as well as those from Jordan and Palestine, were all settled in areas well removed from each other so as to lessen this nuisance and to minimize 35 36 their own inter-tribal strife. The Cordovans in particular hated them, though one would never guess it from the alacrity with which they adopted Syrian modes in the visual arts. That process began early on. It was ‘Abd al-Rahman I who dubbed his country villa Munyat al-Rusafa after his 37 grandfather’s favourite Syrian town. He stocked it with Syrian plants, including Spain’s first palm tree, and wrote a nostalgic ode celebrating 38 both its beauties and the pain of exile. He was, as this poem shows, in no 39 doubt as to where his real home was. The psychological importance of Syria to the Umayyads of Spain did not diminish in later centuries. It was especially marked during the golden 40 age of the Spanish caliphate, the 10th century. The astonishing wealth of ‘Abd al-Rahman III was legendary; his treasury of 20 million gold pieces made him by the mid-10th century the richest Muslim prince of his time, along with the perhaps significantly Syrian prince, the Hamdanid 41 ruler Ghadanfar Abu Taghlib ibn Hasan ibn ‘Abdallah. It was in that same 10th century that Córdoba became both the largest and the most cultured city in Europe, with a population probably rivalled only by 42 43 Constantinople – though its exact size has caused much controversy.  Creswell 1979, II, p. 156.  Creswell 1979, II, pp. 156-157, and Lévi-Provençal 1950-1953, I, pp. 34-38,

34 35

41-51, 82-85, and III, pp. 51 and 203. They were liable for military duties and thus formed a special corps; Lévi-Provençal 1950-1953, III, pp. 67-70, 76-77. 36  Creswell 1979, pp. 156-157. 37  Ruggles 2000, pp. 42-43. 38 Ruggles 2000, pp. 42 and 232. 39  For further examples of the Syrian connection in the period of ‘Abd alRahman I, see Hitti 1968, pp. 62, 68 and 72. 40  As shown, for example, by the Syrian lineage of the great families of alAndalus: Lévi-Provençal 1950-1953, III, pp. 192-193, and Lévi-Provençal 1932, pp. 99-111. 41 This is the report of Ibn Hauqal in his book Kitab Surat al-Ard, Ibn Hauqal 1964, I, p. 111. 42  Kennedy 1996, p. 107. He regards these two cities as the largest in Europe. 43  Lévi-Provençal 1950-1953, III, pp. 359-363, assembles the contradictory documentation of the Arabic sources as to the size of Córdoba in its prime and its probable population. He shows an impressive scepticism as to the higher range of estimates but still concludes that Córdoba was five cities rather than one (see his fig. 11) and that it was about eight times the scale of the city in

180

It was sustained by a hinterland extensively irrigated by water-mills and 44 norias (water-wheels), the na’uras that can still be seen at Hama – which 45 takes the story back to Syria once more. This prosperity provides an appropriate context for the luxury art of ivory carving, which reached its peak in this same century. Other arts which flourished there included caskets of repoussé silver (figs. 11 and 12). The Christian Rabi‘ b. Zaid, later to become bishop of Granada, was sent by ‘Abd al-Rahman III to Syria (where else?) to obtain objets d’art for the embellishment of Madinat 46 al-Zahra. And when in 961-965 al-Hakam II transformed the maqsura of the Great Mosque of Córdoba, the obvious material to choose was glass mosaic, as had been used in the Dome of the Rock and the Great Mosque of Damascus, and the obvious source for it was once again, as in the time 47 of the Umayyads of Syria, the Byzantine emperor and his craftsmen. In the hothouse atmosphere of the court, where the fashion for things Syrian was so pronounced, it should not cause surprise that the ivories (and for that matter the architectural detail of the Great Mosque and of Madinat al-Zahra) should echo an aesthetic developed by the Umayyads of Syria and now subtly transformed by their descendants. And that leads to the last introductory point, which must be strongly emphasised, for it colours all of the following discussion. It would be a travesty of the truth to regard the art of Umayyad Spain – in whatever medium – as in any sense a clone of Umayyad art in Syria. Rather does it suggest how that art might have developed had the ‘Abbasid takeover not occurred, had there been no Samarra. his own time (c. 1950), with a population far in excess of 100,000 (he quotes the estimate of Carande – 500,000 – without seeking to contradict it). L. Torres Balbás, in his posthumous work completed by H. Terrasse, Torres Balbás 1971, I, pp. 103-104, suggests how medieval populations might be calculated, but declines to suggest a figure for the surface area occupied by Córdoba, which renders his population estimate of 100,000 questionable because its basis is not revealed. P. Chalmeta, basing himself on urban perimeters and inhabited areas, plumps for the mean between the highest figures (based on perimeter walls 15.83 miles long and 273,377 households, and even so ignoring 21 extramural suburbs: 850,000) and the lowest figures (perimeter walls 10 miles long: 220,000) and thus arrives at 500,000; see Chalmeta 1992, p. 755. 44 Glick 1992, pp. 980-982, and García Sánchez 1992, p. 996. See Ettinghausen 1962, p. 127, for an illustration of a Spanish water-wheel from the romance known as Hadith Bayad wa Riyad. 45  Burns 1992, pl. 9. 46  Lévi-Provençal 1950-1953, III, p. 222. 47  Nor was this the last of such contacts; see Lévi-Provençal 1948b, pp. 81-108 (previously published in a less complete form in Byzantion XII, 1937, pp. 1-24).

THE SYRIAN CONNECTION

181

Epigraphy The rest of this paper will tackle the question of the archaizing tendency of Spanish Islamic art through the prism of two major themes: epigraphy 48 and vegetal ornament. First, epigraphy. Two quite distinct epigraphic fashions flourished in parallel in the ivories of the late Caliphal period. One may be termed the “severe” style, while for the other the epithet “foliated” may be proposed, since that somewhat specialised elaboration was its trademark. Yet this foliation, far from invading the main body of the inscriptions which display it, is confined to the top of the band, and most of the letters themselves, as distinct from their terminations, are executed in the “severe” style. 49 The very fact that a round score of inscriptions can be fitted into no more than two categories – and those two, as just noted, very closely related in their essentials – even though they represent very costly work of high quality produced for elite patrons, reveals the sluggish pace of change in contemporary Spanish epigraphy. It was conservative to the core. By the same token, this conservatism underlines the gulf between the styles of Andalusia and those of the rest of the Islamic world in the matter of inscriptions. In particular, given the rapid development of Islamic epigraphy and calligraphy in the 3rd and 4th centuries of 50 the Hijra, it is remarkable how tenaciously some 10th-century Spanish Umayyad artists – and in particular those who practised the “severe” style as just defined – conserved the shape and the spirit of Umayyad formal writing as it had developed in Syria during the period 690-750. The largest body of such writing is found in the inscriptions of the Dome of 51 the Rock. Despite the lapse of centuries, the different medium and the 48   This subject has been somewhat neglected, as is typified by the two lines devoted to the subject in Fernández-Puertas 1992; see p. 647. Meanwhile, see Ferrandis 1935-1940, I, pp. 81-89, though here the emphasis is on the content of inscriptions rather than on their form. Ocaña Jiménez 1970, p. 49, cites only the ivories of Zamora (briefly discussed on p. 36; pl. XX) and Pamplona (discussed on pp. 43-44; pl. XXIX) in his list of 17 major examples that document the evolution of Kufic epigraphy in Spain. 49  Sheila Blair, in her book Islamic Inscriptions, Blair 1998, p. 188, mentions a total of 33 objects, but the present study is based on the 19 most important examples of this epigraphy. 50  Well summarized by Grohmann 1957; since so much Kufic in its early centuries evolved precisely in the direction of foliation and floriation, this article has a wider significance than its seemingly very specialised title suggests. 51 Now best studied in the colour close-ups in Nuseibeh and Grabar 1996, pp. 82-105 and 161.

182

much smaller scale of the inscriptions on ivory, the similarities are striking. And the parallel is reinforced by the fact that in both cases elite patrons and expensive materials were involved. Indeed, these particular Syrian Umayyad inscriptions, rather than contemporary work from the Mashriq, are the closest parallels available for the epigraphy of these ivories – apart of course from other Western Islamic inscriptions, for example in stone, 52 which themselves bear the imprint of Umayyad Syria. Syrian Umayyad 53 54 inscriptions in other media, such as stone slabs and coins, have more 55 distant connections with Spain. The first priority here is to look in more detail at the defining characteristics of these two styles with reference to specific examples, and the discussion will then move to their links with earlier Umayyad epigraphy in Syria. The “severe” style Let us begin, then, with the so-called “severe” style. Judged by the most 56 forward-looking inscriptions of their time in the Islamic world at large, this type of epigraphy is simple to the point of austerity. It avoids both elongated ascenders and swooping curves for those letters that descend below the line, and thus these inscriptions operate within quite narrow boundaries. In much the same reductive spirit, this style eschews the horizontal extension of individual letters; their internal transformation, for example by braiding; wilful illegibility; and variations in sub-linear or supra-linear rhythm. Many inscriptions leave the background totally plain. Elsewhere, when the inscription continues for a substantial space without ascenders, bifurcated leaves (not related to any of the letters) float above the inscription and unobtrusively fill the space (figs. 1a and 1b). Such a  Lévi-Provençal 1931a, pls. Ia-b (Córdoba, 241 / 855-856 and 244 / 858, both stone) and IIa-b (Córdoba, 329 / 940 and 333 / 944-945, both stone). The Cordovan material is currently in the process of comprehensive and painstaking reworking by J. A. Souto; see, for example, Souto 1998, pp. 303-324; Rodríguez and Souto 2000, pp. 359-391; Souto 2001, pp. 283-307; Souto 2004. 53  Berchem 1922, pp. 17-29, especially p. 21, figs. 1-4. 54  Walker 1941, pp. cliv-clv and pls. XX and XL (representative examples); Walker 1956, p. cii and pls. XII-XXXI; Hillenbrand 2004. There are, however, close links between the epigraphic styles of Syria and al-Andalus – a matter for future study. 55 Given the uncertainties of dating, this account does not deal with Qur’anic manuscripts, which in any case fall into a markedly different epigraphic group. 56  E.g. on the tomb-tower of Radkan West in northern Iran, dated 407 / 10161017; Flury 1920, pp. 26-31, fig. 8 and pl. XIV. But note that the Gunbad-i Qabus has inscriptions that are remarkably similar to those of the “severe” style in ivories of the late caliphate in Spain; Diez 1918, p. 100, fig. 40 and pl. 5/1-2. 52

THE SYRIAN CONNECTION

183

practice suggests that the very narrow register within which the letters were confined was felt on occasion to be irksome. It may be that the material itself, so notoriously difficult to work and necessitating a much reduced compass which itself made these inscriptions more challenging to produce, was a powerful factor in keeping these inscriptions formally simple. But given the complexity of the figural and vegetal themes, this explanation has little to recommend it. 57 Other devices then fashionable elsewhere in the Islamic world, especially in the East, while not entirely absent, are used so sparingly in the “severe” style that they make relatively little visual impact. One such feature is the incorporation of decorative devices and motifs, such as 58 59 knots or rosettes, into the letter forms themselves, rather than their mere presence in the background as a counterweight to the epigraphy proper. Another is the massing together of certain letters, notably uprights, for 60 visual emphasis. A third is the development of a busy background as 61 a foil for the inscription itself. Yet another is the creation of subsidiary 62 rhythms by the careful placing of forward- or backward-leaning letters. In comparison with the innovative epigraphy of the contemporary eastern Islamic world, then, the “severe” style found in the inscriptions on some late 10th-century Spanish Islamic ivories is remarkably jejune. It is worth emphasizing these points, because these inscriptions, as already noted, are found on top-of-the-range artifacts made for members of the ruling elite, not – say – on a set of hurriedly executed provincial tombstones made for humble patrons. On the other hand, one should bear in mind that the original appearance of these bands might have been very different from what it is now, thanks to the use of polychromy. Moreover, the suggestion has been made that, for example, the tiny perforations on the edge of the leaves on the pyxis in New York, in the collection of the Hispanic To judge by the evidence of tombstones, as early as the 9th century A.D. Egyptian epigraphy was developing a range of devices that in certain respects (lam-alif ligature, terminal ya’, mashq and splayed crockets) is significantly bolder than what can be found in Spanish ivories a century later. See Miles 1957, figs. 7, 9, and 13. 58  Volov 1966, pp. 111, 129, 131, and figs. 3-4, 10-14; Flury 1925, p. 67, fig. 2; p. 73, fig. 5; p. 76, fig. 6; p. 77, fig. 7 and pl. IX. 59  Flury 1925, p. 111, figs. 4 and 7; Flury 1930, pl. X; Viollet and Flury 1921, pls. XXXIII-XXXIV. 60  Flury 1925, pls. VI-VII and XXI-XXIII/1-2. 61  Flury 1925, pl. XXI; Blair 1992b, pls. 48 and 147-148. 62  Blair 1992b, pls. 38, 73, 107-108, 146, and 150. 57 

184

Society of America, might have held jewels (cat. no. 14a). The use of a background whose colour was different from that of the inscription itself would have served to make the letters stand out more strongly. Would it be fair, then, to stigmatise this “severe” style as an inappropriately careful clone of a ductus that had emerged in the earliest years of Islamic art, that had somehow travelled to the Islamic West – perhaps through the 64 medium of much-venerated Syrian Koran manuscripts – and that by this time had become totally outmoded? Probably not. The key lies in some quite subtle changes of emphasis, and it is now time to explore these at some length, using the Louvre pyxis made in 968 for al-Mughira, the son of the Caliph ‘Abd al-Rahman III, as a test case (cat. no. 11 and figs. 1c 65 and 1d). The principal concession to decorative effect here is the sparing embellishment applied to the terminations of letters – the nearest that Islamic epigraphy gets to the use of the serifs that add such dignity and authority to Roman lettering. To widen the scope of the discussion for a moment, it is noticeable that not one of the inscriptions on the Spanish ivories confines itself to simple horizontal or vertical endings for its letters. That would make words and letters seem to collide quite abruptly. So at the very least they include bevelled endings, mostly of 45 degrees, but occasionally with steeper angles or featuring a slight concavity scooped out of the termination. This pronounced bevel has the effect of smoothing the transition from one word or letter to the next, and thus implies fluid movement. The lack of consistency in the very limited decorative devices applied to the letter forms is worth stressing, for it suggests that a distinctive style is in the making. On most of these 10thcentury ivories, bevelling is by no means the only decorative device in the epigraphy, for the designers were eager to highlight letter endings in more marked fashion. On the Louvre pyxis this discreet enrichment varies from one letter to the next, but in its standard form the letter tapers to a triangle set on its side; its hypotenuse is incorporated into the letter itself. One point of the 63

 Ettinghausen and Grabar 1987, p. 152.  A much-venerated Qur’an manuscript that included leaves from a copy

63 64

which had belonged to the Caliph ‘Uthman was preserved in the Great Mosque of Córdoba in medieval times; Grabar 1992, p. 585. An ‘Uthmanic Qur’an was a much-treasured relic of the Great Mosques of Ma’arrat al-Nu’man and Damascus in succession, and was also treated with extreme reverence; Hillenbrand 1999a, p. 305. 65  For an attempt to decipher its iconography, see Prado-Vilar 1997, pp. 21-30.

THE SYRIAN CONNECTION

185

triangle, sometimes needle-sharp, sometimes wedge-shaped, terminates the letter. In an ascender, the second point is turned either to the left or to the right, like a vertically bisected arrow or a short pennant flying from a staff. In a horizontal letter, such as the base of a lam, it is oriented upwards (cat. no. 11d and fig. 1d). This thickening and sharpening of the termination applies in one way or another to most letters; medial mim is one of the rare exceptions. The designer sometimes plays with the decorative effect of repeated spikes pointing left, often at different levels (three examples on cat. no. 11a; see also fig. 31d), thereby enhancing the forward momentum of the script. Or he exploits the impact of spikes pointing in opposed directions, whether placed close together or further apart (cat. no. 11a, 11d, and fig. 31c). Such juxtapositions, as in terminal ayn in the word sab‘a (cat. no. 11d and fig. 1d), where one ending nestles within the other, can be used to arrest the visual flow of the inscription. Other low-key modulations involve increasing the height of certain shafts so that their tips touch the top of the inscription band, as in the ta’ of ghibta (fig. 1c), and correspondingly allowing the tails of certain letters, such as the ra’ of li ‘l-Mughira, to graze its base (cat. no. 11b); or linking words by elongating the last letter of the previous word in sub-linear fashion so that it underlines the first letter of the next word, as in ‘amal and sanat (cat. no. 11d and fig. 1d). One may also note the uneven baseline, which results in a pleasing undulating effect as one follows the inscription around the pyxis, the varied forms sometimes used for the same letter, and the uneven thickness of the letters, features which are all illustrated in the 66 eccentric carving of the words wa surur (cat. no. 11b and fig. 1c). These comments suggest that even the stylistically “severe” inscriptions on certain 10th-century Cordovan ivories have their own distinctive felicities, and that the local artists were well capable of building in innovative fashion on the Syrian Umayyad heritage while maintaining intact the essential spirit of that heritage. But it cannot be denied that in general these “severe” inscriptions seem somewhat dry and precise, old-fashioned and unadventurous. They follow the spirit of the Dome of the Rock inscriptions – some even have the same total lack of added ornament (cat. no. 5a) – but modulate the almost primitive directness of the sans-serif simplicity favoured by the Umayyad designers in that building (fig. 1e). This side displays altogether four different forms of waw, plus two different forms of ra’ in the single word surur (fig. 31c). 66 

186

Enough has been said to show that the “severe” epigraphy of certain 10th-century ivories, such as the four pyxides now in Madrid (ex-Zamora) (cat. no. 4a), Paris (cat. no. 11a-11d), and London (cat. nos. 5a, 12a, and 12b), and all dated or datable to the reign of al-Hakam II, that is between 961 and 976, as well as the two caskets of 966 now in Paris (cat. no. 9a and 9b; see also fig. 1f) and Fitero (cat. no. 7a-7c; see also fig. 2a) may (with the most minor exceptions) fairly be termed archaic. Even when it employed extra ornament that was entirely or largely unrelated to the letters themselves, the type and expressive range of that ornament were notably limited. It included oval leaves in cruciform disposition (cat. no. 4a), pendant lozenges (cat. no. 12a), and fragmentary portions of leaves with multiple striations (cat. nos. 7a, 7c, 8a, 9a, and 9b). The fullest expression of this applied ornament is in the inscription band of the pyxis of al-Mughira (figs. 1c and 1d), and even here it is applied with relative parsimony. Moreover, its character differs markedly from that of 67 the vegetal designs on the bodies of these pieces. Splayed out fanwise, with a row of closely packed rounded parallel lines like the teeth of a comb giving a strongly sculptural gadrooned effect, the leaves in these compositions are further enlivened by an array of rounded perforations. Their outer edges have a tendency to curl inwards and crinkle. These leaves, for all that they have Syrian Umayyad antecedents, are effectively a Spanish Muslim innovation, and the numerous variants of this form 68 dominate the stone sculpture of the region. The “severe” style in epigraphy was not confined to ivories and stone carving. The version of it that is most frequently encountered in local metalwork favours thick spatulate letters, as in the Gerona silver casket of 69 976 (figs. 11 and 12), and the celebrated Griffin of Pisa – whose epigraphy 70 has been relatively neglected as a key to its dating and provenance. But 67  For typical examples on the bodies of these ivories, as distinct from their inscriptions, see cat. nos. 2a, 3a, and 28a. 68  Pavón Maldonado 1981. 69  For a summary of the most recent scholarship and a marshalling of the evidence in favour of a Spanish provenance, see A. Contadini in Venice 1993, pp. 126-131, especially p. 130. See also Contadini, Camber and Northover 2002, especially pp. 65-70. 70  The closest study of its epigraphy to date has been that of A. S. MelikianChirvani, Melikian-Chirvani 1968, pp. 78-84; he argues in favour of a probably Khurasani provenance, but is alone in so doing. Nevertheless, see note 56 above; there are clear affinities between Spanish and certain eastern Iranian inscriptions in the 10th to 12th centuries, and the sheer variety of hands in the latter area during this period can scarcely be exaggerated. Lévi-Provençal notes the close

THE SYRIAN CONNECTION

187

such inscriptions (compare also fig. 2c) lag well behind the evolution of epigraphy in other media in Muslim Spain. Thus the famous “Veil of Hisham II”, for example, datable between 976 and 1013, employs backward-curving rising tails, triply-foliated terminations, and horizontal 71 extension in certain letters (fig. 3a). The “foliated” style This more luxuriant epigraphic fashion can be found in several preTaifa ivories of the 10th and early 11th centuries. The fact that this richer epigraphic fashion was popular at the same time as the “severe” style indicates that craftsmen specializing in luxury arts were at liberty 72 to choose one or the other manner: the “foliated” style did not, as it were, grow out of the “severe” one. What were the characteristics of this “foliated” style? Two general points need to be made at the outset. First, the principal ornament of these inscription bands is indeed foliate, though it is used in more ways than one. Secondly, just because these bands are more decorative than those in the “severe” style – initially, perhaps, by virtue of their foliate elements – they do tend to use a variety of other devices for extra ornamental effect, though it has to be stressed that this is not enough to make these inscriptions positively luxurious. The casket of c. 962 in the Victoria and Albert Museum is a case in point (cat. no. 2a; see also figs. 2d-2g and 13-16). Here, certain letter forms, such as the otiose rising shaft in a terminal ha’, with or without a foliate ending, incline forwards gracefully (figs. 2d-2f); dal/dhal acquires a terminal vertical shaft with a languid backward curve (unrelated to the letter itself, though joined to it) in the form of a bifoliate leaf (cat. no. 8a) and sometimes a trifoliate one (figs. 2d and 2e). Many uprights thicken as they rise (fig. 2e). The slenderness and delicacy of the letter forms contrast strongly with the chunky quality favoured in the letters of the “severe” style. Sometimes they are of markedly uneven height, thereby varying the pace of the inscription. A bud pushing its way through two parted leaves sits on certain letters (mim in cat. no. 7a, kha’ in cat. no. 7b, waw in cat. nos. 8a and 73 12b, and the bearer of hamza in cat. no. 7c) or even between them. The links between the formulae used in Spanish tombstones and the formulae current in the Mashriq; Lévi-Provençal 1931a, p. xviii. 71  New York 1992, p. 225. 72  Or indeed a middle way between them, combining elements of both: see the next paragraph. 73  Kühnel 1971, pl. XV, 25d.

188

generally botanical quality of the inscription band as a whole fits neatly with the vegetal content of the other panels on this casket. The pyxis made by Khalaf now in New York and datable to c. 970 has sufficient foliation to be squeezed out of the “severe” category, but it plainly belongs between the two styles (cat. no. 14; fig. 3b). In such features as the extended horizontal backward tail of the ya’, the marked forward reach of the arm of the voiced and unvoiced ha’, the redundant triangle attached to the base of the rising tail of nun in the word ahsan (fig. 3b), the foliation added to terminal ra’ in the word manzurati (cat. no. 14a), and the right-facing spur attached to the redundant lower shaft of the letter dal (fig. 3b) one may see the workings of an urge to ornament. Some of these features are also intrinsically dynamic and, in conjunction 74 with the slight forward incline of the letters as a whole, they subtly propel the inscription onwards. The occasional overlapping of the spiked 75 terminations of the shafts also helps to increase momentum (cf. fig. 3c). The “foliated” style can be found at its most elaborate in two objects from Cuenca: the casket dated 417 / 1026-1027, now in Burgos (cat. no. 25b), and another casket dated 441 / 1049-1050, now in Madrid and formerly in the cathedral of Palencia (cat. no. 29a). It is of course perfectly possible that the elaboration of the “foliated” style was a local characteristic developed in that city; and indeed, the motif of foliation is pursued in these two objects with a single-mindedness that borders on monomania. The pyxis datable to c. 1030-1040 now in Narbonne, and stated in its inscription to have been made in Cuenca, at first looks somewhat plainer (cat. no. 28a), but the same obsession with burgeoning foliate endings can be seen in the treatment of the adjoining initial nun and kaf of the word 76 Cuenca; neither of these forms is a standard vehicle for ornament. The bifoliate leaf which crowns the letter nun goes on to generate an even more elaborate leaf, while the one which terminates the shaft of the letter kaf sprouts unequally to left and right. This motif develops into a leafy 77 canopy in three different places in this inscription. Overall, the effect of this botanical exuberance in the upper reaches of the epigraphic band  For example waw, a letter often treated this way in the epigraphy of these ivories (for example the Pamplona casket, cat. no. 20b and 20c); its back curves in marked fashion to the left (figs. 31a and b). 75  As in an 11th-century textile fragment (fig. 33c) from the Colls church in Puente de Montañana in Huesca; New York 1992, p. 227. 76  Kühnel 1971, pl. XXXVIII, 42c; Gómez-Moreno 1951, p. 309, fig. 370. 77  Over the lam of ‘amal (cat. no. 28a); the alif of quwad and the central alif of Isma‘il, cf. Kühnel 1971, pl. XXXVIII, 42b. 74

THE SYRIAN CONNECTION

189

is to embower the letters, thick and clunky though they be, in full-blown vegetation. Given that the main body of the pyxis is itself an abstract garden, a veritable hymn to fertility, it is entirely fitting that the letters themselves should become organic. One may recognise here an echo of the Great Mosque of Córdoba, with its distinctive contrast between the exuberance of the upper tier of the sanctuary, which as it were bursts into 78 flower, and the stark simplicity of endlessly repeated files of columns. In the caskets at Burgos and Madrid (cat. nos. 25b and 29a) this idea is taken so much further as to turn the inscription band into something akin to a herbaceous border. Its riotous growth threatens momentarily to spill over its allotted parameters. Just as with the Narbonne pyxis (cat. no. 28a), the most notable feature in the Burgos casket (fig. 33d) is the 79 key role played by the upright shafts of such letters as lam, alif, and dal (cat. no. 25b) – in other words, the very letters which remain plain in most foliated scripts, where letters with rising tails, whose serpentine form somehow predisposes them to break into leaf, are the preferred focus for extra decoration. Here, on the contrary, it is the letters with vertical shafts 80 that predominantly play this role; they are crowned by leafy flourishes as baroque as the limited space available permits. And they are placed back to back, thereby creating an almost continuous upper frieze, a kind of awning or pelmet. There is even a redundant alif in the word ‘ashara so that this particular word can play its appropriate part in the decorative ensemble (cat. no. 25b). The most elaborate of these foliate ramifications issues from the lam of ‘amala, the key word announcing the identity of the artist, which somehow contrives to be at the approximate centre of the band and thus crowns the central axis of this whole face of the casket (cat. no. 25b). 81 Such a detail speaks volumes for the artist’s sense of his own importance;  New York 1992, pp. 2 and 19, fig. 10, an effect repeated in the Salon Rico at Madinat al-Zahra (New York 1992, p. 32, figs. 5-6), in the pavilion at Malaga (New York 1992, p. 52, fig. 2), and in the palace of Aljafería in Saragossa (New York 1992, p. 59, fig. 9). 79  For both of these, see Kühnel 1971, pl. XXXIV, 40d. 80  There are indeed genuine foliated rising tails in this inscription, as in the final dal/nun of Ziyad/Zayyan or the dal of Muhammad, but not all are what they seem. Thus the tail of nun in ibn is stuck onto an already fully formed letter of standard simple form rather than being an integral part of it. The same goes for the serpentine foliation attached to the initial ‘ayn of a’azzahu (fig. 33d). 81  It is well to remember that the Pamplona casket (cat. no. 20) bears no less than thirteen signatures of craftsmen; R. Holod in New York 1992, pp. 198199. None of them hogs the limelight to the extent that Muhammad ibn Ziyad/ Zayyan does in the Burgos casket (cat. no. 25). 78

190

in an almost literal sense, his personality broods over the heart of the 82 design. The other hallmark of this inscription is the tendency for the bevelled endings of many letters to curve inwards slightly, which makes them look barbed. The cumulative impact of these characteristics is to create a rich, mobile epigraphic tapestry whose sense of pulsating life is now at the opposite pole to the “severe” style from which it is ultimately 83 derived. The last of these pieces from Cuenca is the casket formerly in Palencia, now in Madrid, made in 441 / 1049-1050 for the city’s governor, Husam al-Dawla (Isma‘il ibn Yahya al-Ma’mun) (cat. no. 29a; see also fig. 3e). Here, even more obviously than in the Burgos casket, the embellishment of the letters is banished to the very top of the band; the style should perhaps be termed “partially foliated”, for the basic form of the letters once again recognisably adheres to the “severe” style. In the remarkably lissom and rounded forms of this upper storey, where the extended foliations evoke banners streaming in the wind, the leaves tend to curl tightly so as to form little circles which echo those of the inscription below. These fluttering foliations are arranged to form affronted pairs, sometimes close together, but elsewhere some distance apart. This device of bilateral symmetry helps to integrate the inscription as a whole. So too does the triple repetition of dhi, with the flattened, elongated dhal and the long tail of the terminal 84 ya’ (fig. 3e). The suppleness of the leaves matches that of the letters themselves. Other objects that belong in the foliated category – the caskets now in Florence (cat. no. 18a; see also fig. 4a) and Pamplona (cat. no. 20a-20d, see also fig. 4a), and the pyxis now in Braga (cat. no. 21a; see also fig. 82  In the larger panel below the inscription band, this central point is marked by the intertwined necks of two peacocks (a theme also found in slightly different form in the controversial Alp Arslan salver; see Grube [1967], p. 75, pl. 41). Note too the six-times-repeated theme of the lion attacking the gazelle, whose classic Islamic version is the mosaic at Khirbat al-Mafjar and whose erotic significance has been deciphered by Doris Behrens-Abouseif (see note 10 above). Is this, then, another echo of Syrian Umayyad art in Spain? 83  There is no space here to do more than to draw attention to other characteristic features such as the triply-foliated tail of lam, the undulating termination of ra’ which ends in a point, the simplified fleur-de-lis perching on some letters, the curious rippling, rightward and sublinear extension of the medial ḥā’ in the word Muhammad, the curvaceous backward sweep of terminal ‘ayn or the bud between two leaves which hangs from some of the larger foliations. 84  This same flattening distinguishes the letter dad in al-Zafir and the kaf in Quwanka.

THE SYRIAN CONNECTION

191

4b) – display rather more muted versions of the foliated motif than do the Cuenca caskets. But they also have their own distinctive character. The Pamplona casket is the only Spanish Islamic ivory whose letters are studded with tiny squares, circles, or ovals in low relief. The resultant spotted effect, highlighted by means of its plain outline – and perhaps 85 originally by colour – is the most striking characteristic of this inscription. It adds an extra and highly unusual level of complexity to the object, and 86 takes into new directions the desire to see letters as vehicles of ornament. The casket in Florence also features the use of internal ornament for its inscription – this time an incised line which runs through the centre of each letter and mimics even its decorative terminations (fig. 4a). Thus a general tendency to conservatism did not exclude a range of original devices intended to enrich the visual impact of these inscriptions. Close examination reveals that in the Pamplona casket, too, most of the letters are executed in the “severe” style, notably in their triangular terminations (figs. 1a and 1b), which on occasion also sprout a lower 87 leaf – a curiously unsatisfying hybrid form, neither fully foliate nor fully severe (fig. 1a). Indeed, the foliated impression which this inscription presents at first sight is due mainly to the presence of foliated bits and pieces which hover in the upper reaches of the inscription band and swoop down at intervals to fill the interstices of the epigraphy. They are not actually attached to any letter, though they may perch just above 88 one or graze its crown (fig. 31a). These floating bio-ornaments, which 89 include symmetrical twin palmettes, cones, and even perforated circles 90 dangling – appropriately enough – above letters of identical form, are themselves packed with busy infill. What is interesting here is that the artist has declined the obvious challenge of transforming the epigraphy 85 

Cf. New York 1992, p. 203.

 12th-13th century animated inscriptions, in which ascenders ended in heads,

86

or the entire letter took the form of a human body, are another expression of the same urge; see Rice 1955, pp. 22-33 and figs. 19-31. So too is the post-medieval manipulation of letters to form figural images, buildings and other forms. For examples, see Schimmel 1970, pls. XVIIIa-c, XXVIIIb, XLIVb, and XLVI-XLVII. 87  As in the last two letters of baraka and ghibta in the opening words of the inscription (cat. no. 20a). 88  As in the second ra’ of surur and the word wa, respectively, in the second half of the first line (cat. no. 20a). 89  These may be filled with tiny pearl roundels, very much in a Samarran style, which is also used for the leaf held by the figure of authority (cat. no. 20d). 90  Cat. no. 20b, above the head of the central figure. Such mirror symmetry between epigraphic and vegetal forms is uncommon.

192

from within by incorporating these motifs into the writing itself; instead, he has limited their role to floating somewhat aimlessly around it. Other features of interest include the curled-up tails of terminal ‘ayn (fig. 1a) and terminal ha’ and the way that waw rests on the baseline rather than dipping below it. Here, too, the Florence casket offers a parallel in its use of bifurcated leaves and other leaf forms as interstitial or added ornaments to supplement the foliation which is an organic part of the letter itself (fig. 4a). So much, then, for the “foliated” style. And that epigraphy had a brighter future than the “severe” epigraphy which dominated the ivories made in the reign of al-Hakam II, for the ivories made later in the 11th century employ a noticeably richer, even luxurious epigraphic style. The nature of, and the reasons for, that later development deserve exploration in a separate paper. Links with Umayyad Syria Now that the distinctive epigraphy of these ivories has been discussed in some detail, it is time to consider its links with the inscriptions of Umayyad Syria. Since this means effectively the inscriptions of the Dome 91 92 of the Rock – as noted earlier, Umayyad coins and milestones do not seem to offer useful parallels – it is as well to open with a caveat. Monumental inscriptions in mosaic are subject to ground rules decisively different from those which apply to inscriptions in carved ivory on a tiny scale, even if both texts were originally worked out on a comparable scale on vellum. One may note in the Pamplona casket (cat. no. 20b and 20c; see also figs. 1a and 1b), as in the inscriptions on other ivories such as the pyxides in London (cat. nos. 5a, 5b; and 12a, 12b) and the one in Braga (cat. no. 21a and fig. 4b), some striking resemblances to the Dome of the Rock, which include the stumpy ascenders, the open ‘ayn, the forward 93 inclination of waw, and the backward horizontal tail to the terminal ya’. These points of contact are interesting enough, but they do not suffice to establish a close connection between the two traditions. A far more telling point of contact is a shared general tendency to favour the breadth rather than the height of individual letters. This tendency is, however, rather more marked, indeed pervasive, in the Dome  These are now best studied in Nuseibeh and Grabar 1996, which offers the fullest coverage in colour so far available. 92  See Walker 1956, pls. XII-XXII; Berchem 1922, pp. 17-29. 93  Nuseibeh and Grabar 1996, p. 104, top left and right. 91

THE SYRIAN CONNECTION

193

of the Rock. Here the height of medial sin and ‘alifs is in a ratio of 1:3, 95 while alif itself has a width to height ratio of roughly 1:5 (fig. 4c). Visually, the overall impression that such a disposition of the letters gives is that they have been squashed in the vertical axis. Archaic the ivory inscriptions may be, but that does not mean that epigraphic fashion stood still in the three centuries that separate the Dome of the Rock from the Cordovan ivories. For example, the background ceases to be plain, acquiring added 96 ornamental motifs. In particular, that sense of being squashed down disappears, and the letters gain extra height. Thus in the Pamplona casket (cat. no. 20c), the height of medial sin as opposed to alif has a ratio of 97 about 1:2 as against 1:3 in the Dome of the Rock, while alif has a width 98 to height ratio of 1:8, a major development from the 1:5 in the Dome of the Rock; but the horizontal emphasis is still unmistakable. It is achieved principally by the palpable gaps between both the letters and the words. As in many early Qur’anic manuscripts, though not for the same 99 reasons, these gaps slow down the pace of the epigraphy. They dissipate the nervous tension built up by the rapid succession of tightly massed shafts. By the 10th century, this horizontal emphasis had become a rather conservative trait in the Islamic world at large. The alifs in the St Josse silk, 100 made in Central Asia before 961, have a ratio of 1:11.5 (fig. 4e), as do some – though not all – of those in the Louvre pyxis (cat. no. 11a-11d; see also figs. 1c and 1d). If the individual letters can be identical in contemporary inscriptions so widely removed from each other in space, how is it that so many of the Spanish ivories have inscription bands in which horizontality is still such a major feature? The clue lies less in the proportions of the individual letters than in the spacing of those letters (and words) in the band as a whole. The fashion in Spain did not favour back-to-back ascenders, or for that matter minimal spaces between letters or words. But in other regions 94

 To be precise, 8:23.  To be precise, 2:11.

94 95

96  This tendency gathered momentum in the course of the 3rd / 9th century, and, in the case of Egypt at least, its stages can be pinpointed through the evidence of tombstones; Wiet 1936-1940; Strzygowski 1911. 97  To be precise, 18:41. 98  In the initial alif of the basmala (cat. no. 20a). 99  Those reasons may well have to do with deliberately slowing the reader down so as to maximize attention to the text; to take a random example of this kind of breaking up of words in defiance of sense, see Atil 1975, pp. 18-19, cat. no. 3. 100  Baker 1995, pp. 46-47.

194

it did, and this gave calligraphers wider scope to vary the rhythm and pace of their inscriptions, making aesthetic capital of irregular contraction and expansion. That liberal space between letters and words – usually comfortable (though of course there are exceptions) and sometimes very generous – can be seen as a legacy of the epigraphy of Umayyad Syria, and specifically of the Dome of the Rock. The overall effect of such ample spacing is apt to be overlooked if one focuses only on a given word. For example, in the Dome of the Rock inscriptions the three ascenders of the word “Allah” are broader than they are high (fig. 4c); the ratio is 4:3. In the Pamplona casket – as in the Louvre pyxis – the emphasis has shifted decisively to 1:2. In the St Josse silk, the comparable ratio for the self-same letters is about 1:2.3, so that they are 101 more than twice as high as they are broad. But it would be a cardinal error to deduce from this close similarity that the feel of the inscriptions on the Spanish casket and the Central Asian silk is comparable. It is not. In the course of a mere eleven words, including in that total the definite article, the St Josse band has no less than 13 examples of letters or words separated by only the thinnest of lines (fig. 4e). In a sequence of comparable length, the Louvre casket has barely two. In other words, while ascenders such as alif have shot up by leaps and bounds in purely absolute terms in both Spain and Central Asia, and to almost the same degree, the two inscriptions are very different because their conventions for spacing have very little in common. In the St Josse silk, the inscription as a whole now seems loftier still because the space between the letters has been reduced in such a dramatic fashion. They are jammed together. That new, exciting verticality transforms the whole character of the inscription. The Spanish inscriptions, by contrast, seem 102 altogether slower and more sedate. Another area of epigraphic fashion in which the Spanish ivories owe something to Umayyad Syria, but build on that inheritance, is the use of bevelling. The Dome of the Rock inscriptions sometimes have a slight 103 bevel of say 70°, though blunt or tapering points (fig. 4d) are also

 To be precise, 5:11.5.  It may be relevant to recall that Ibn Khaldun notes that in the Muslim West,

101 102

scribes were taught to write by copying complete words rather than individual letters; this would naturally lead to an innate sense of the proper spacing between words: Muqaddimah 1967, II, p. 378. 103  As in the two alifs of ila; Nuseibeh and Grabar 1996, p. 100.

THE SYRIAN CONNECTION

195

employed as letter terminations. These latter two characteristics are not found in Spanish ivories, whose version of the bevelled motif, as already mentioned, tends towards a 45° angle. Some ivories also have the lower ends of the bevel touching but the upper bevels recoiling from each other, thereby opening up a gulf of space between the letters above their baseline contact – yet another way of letting air into these epigraphic bands. Other features with a possibly Syrian origin could be cited. On the Pamplona casket, for example, an annulet at the base of a twin palmette near the beginning of the inscription, between the words wa and ghibta (fig. 1a), can be recognised as a device which recurs frequently in the depictions 105 of jewellery or bejewelled plants in the Dome of the Rock. In the use of polychromy, too, one may perhaps reckon with a memory of Umayyad Syria, for the inscriptions of the Dome of the Rock owe much of their effectiveness to the contrast between the gold of the epigraphy set against the blue and green of its background. Another way of defining the strength of the Syrian connection is to stress how consistently the ivories of the Cordovan Caliphate avoid what might be called the extremes of contemporary epigraphic fashion. Not for them the serried shafts of certain Buyid inscriptions, like that of Baha’ alDaula (fig. 2b), where even waw, ra’, za’, and terminal nun are pressed into service as ascenders and the result is as bold as a newspaper headline, with 106 every other letter so to speak shouting for attention. Nor did they seek to emulate the Yemeni ikats of pious content, in which motifs extraneous to the letters themselves, such as knots and triangles, are engulfed alongside loops and foliations into a continuous, impenetrable, writhing interlace 107 – earth-bound as distinct from elevated (fig. 34f). Instead they follow a sober middle way in which simplicity and legibility are at a premium – exactly as they were in the inscriptions of the Dome of the Rock (fig. 1e). 104

The role of the inscriptions What of the role of these inscriptions in the object as a whole? Here a restraint similar to that of the epigraphy itself reigns. There is no question in any of these ivory objects of the epigraphy taking on the dominant role, as sometimes occurs in textiles or pottery from elsewhere in the   Nuseibeh and Grabar 1996, p. 88 (terminal ya’ in salla) and p. 101 (terminal

104

‘ayn).

 Creswell 1969, I/1, p. 276, figs. 275-276.  Blair 1998, p. 171, fig. 12.74. 107  Blair 1989, II, pp. 329-330 and 332, fig. 2. 105 106

196

Islamic world at this time. Instead, inscription bands are kept thoroughly subordinate to the decorative panels; for example, their texts are never 108 presented in a format longer than a single line. That helps to explain what is missing in the content of these inscriptions, such as long titles and Quar’anic quotations. It also accounts for the conservative nature of the epigraphy; it is no accident that it was when inscriptions were promoted from a subsidiary to a paramount role that their style of writing so to speak took off. Given the reigning aesthetic, there was simply no room for long inscriptions on objects as small as this, at least while current fashion dictated that they should bear principally vegetal ornament, sometimes interspersed with figural designs. Vegetal Ornament It is now time to turn to the primacy of vegetal ornament. This feature was by no means rare in 10th-century Islamic art, but even so it would be hard to find another group of works of art at this period that so consistently privileged that theme. Once again, the obvious parallel is found a good two centuries earlier, in Umayyad Syria. Mshatta at once comes to mind – with the difference that the sheer scale of that monument makes its reliance on this one theme still more striking. There, too, the design is kept under control by the imposition of geometrical forms, though at Mshatta they leap straight to the eye rather than operating just below the surface (fig. 17). The Syrian connection Perhaps it would help to review briefly those Syrian Umayyad characteristics that are relevant to this discussion. They include the sense of crowding; the preference for filigree detail; the lavish use of the drill to secure extra depth and thereby produce the ajouré effect, so that the design 109 seems to float above the surface; the inhabited scroll; and the preference  See, however, the transposition of the word mi’at into the upper register in the pyxis of 359 / 969-970 (cat. no. 12a), and the same transposition of the final hu in ‘abduhu and a ‘azzahu in the Burgos casket of 417 / 1026 (cat. no. 25b). Cf. too the transposition of thalatha mi’at in the ex-Zamora pyxis (cat. no. 4 – see Kühnel 1971, pl. XII, 22b) and of the same phrase in the Pamplona casket (cat. no. 20c). In both cases, these words are written on a smaller scale than the rest of the date. Is this bad planning, careless execution or a tacit acknowledgment that the date is of secondary importance? 109  Cat. no. 25b and Beckwith 1960, pl. 33, both from Cuenca and of 11thcentury date. See also note 12. 108

THE SYRIAN CONNECTION

1. a, b Inscription on the Pamplona casket (cat. no. 20) c, d Inscription on the al-Mughira pyxis (cat. no. 11) e Inscription in the Dome of Rock. Jerusalem, 691-692. f Inscription on the casket in Paris. (cat. no. 9)

197

198

2. a Inscription on the Fitero casket (cat. no. 7) b Inscription on silk cloth, Iraq or western Iran, c. 1000. – Textile Museum, Washington, D.C., inv. no. 3.116. c Inscription on a silver box. Spain, 10th-11th century. – Museo Arqueológico Nacional, Madrid, inv. no. 50.889. d, e, f, g Inscription on the casket in the Victoria and Albert Museum, London (cat. no.2)

THE SYRIAN CONNECTION

199

3. a Inscription on the textile known as “The Veil of Hisham II”. Spain, c. 9761013. – Real Academia de la Historia, Madrid, inv. no. 292. b Inscription on the pyxis made by Khalaf (cat. no. 14) c Inscription on the textile from Colls church. Spain, 11th century. – Museo Episcopal y Capitular de Arqueologia Sagrada, Huseca. d Inscription on the Burgos casket (cat. no. 25). e Inscription on the Palencia casket (cat. no. 29).

200

4. a Inscription on the casket in Florence (cat. no. 18) . b Inscription on the pyxis in Braga (cat. no. 21) c Inscription in the Dome of Rock. Jerusalem, 691-692. d Inscription in the Dome of Rock. Jerusalem, 691-692. e. Inscription on St Josse silk, probably Bukhara, c.960 f Inscription on an ikat. Yemen, 10th century. – Textile Museum, Washington, D.C., inv. no. 73.377. .

THE SYRIAN CONNECTION

201

5. Qasr al-Hair West, claustrum. Syria, 728-729.

6. Pyxis, ivory. Syria, 7th-8th century. H: 14; Bottom diam: 8.5 cm. – Victoria and Albert Museum, London, inv. no. 136-1866.

202

7. Panel, bone. Syria or Egypt, 8th century. H: 19.5 cm. – Museum für Islamische Kunst, Berlin, inv. no. J3798.

8. Carved panel, marble, 953-956. – Salón Rico, Madinat al-Zahra.

THE SYRIAN CONNECTION

203

9. Carved panel, marble, with paired scrolls, from Madinat al-Zahra, 936-976. – Museo Arquelógico Provincial de Córdoba, inv. no. 487.

204

10. Panel, wood, with paired scrolls, 8th century. – Al-Aqsa Mosque, Jerusalem.

THE SYRIAN CONNECTION

205

11. Casket of Hisham II, wood covered with silver and decorated in silver, gilt, and niello. Spain, 976. H: 27; L: 38.5; D: 23.5 cm. – Treasury of the Cathederal of Gerona, inv. no. 64.

206

12. Signature on the casket of Hisham II on (fig. 41) on the back of the hinge.

13. Comparative view of the fronts of V&A caskets A.580-1910 (cat. no. 3 – left) and 301-1866 (cat. no. 2 – right)

THE SYRIAN CONNECTION

14. Comparative view of the left-hand flanks of V&A caskets A.580-1910 (cat. no. 3 – left) and 301-1866 (cat. no. 2 – right).

15. Comparative view of the backs of V&A caskets A.580-1910 (cat. no. 3 – left) and 301-1866 (cat. no. 2 – right).

16. Comparative view of the right-hand flanks of V&A caskets A.580-1910 (cat. no. 3 – left) and 301-1866 (cat. no. 2 – right).

207

208

17. Mshatta façade, detail. Jordan, c. 750. – Museum für Islamische Kunst, Berlin.

THE SYRIAN CONNECTION

209

18. Cylindrical box, carved ivory. Egypt or Syria, 7th-8th century. H: 9.5; Bottom diam: 9 cm. – Museum für Islamische Kunst, Berlin, inv. no. 2977.

210

19. Fragmentary plaque, carved ivory. Probably Egypt, 8th century. H: 11.5; W: 7.7 cm. – The David Collection, inv. no. 20/1978.

THE SYRIAN CONNECTION

211

for axial designs, usually with bilateral symmetry and sometimes with a central element – a trunk, a vase, an extended upright stem – to assert the vertical axis (fig. 5). Yet the pervasive sense of free, untrammelled, organic growth ensures that such space-dividers remain unobtrusive. Comparisons, we are told, are odious. In the present case, it is well to remember that one is by no means comparing like with like. The vegetal motifs that cover the bodies of the ivory pyxides and caskets – note the distinction between the seamless continuity appropriate to the former form and the closed, strictly finite format dictated by the latter – are of a miniature scale which demands an intense scrutiny and a corresponding fineness of execution. In contrast, many of the wall panels of related type and iconography carved in stone or stucco that survive in the palaces of Umayyad Syria were intended to be seen from a distance. Thus it is their overall effect rather than their fine detail that counts, even if on occasion (as at Mshatta, fig. 17, or at Qasr al-Hair West) there is a curious disjunction between their intricate detail and their function as architectural ornament. The design of the ivories corresponds much better to their physical size; in this small-scale context the vegetal motifs are successful both when they dominate the design and when they are relegated to the background. But in both cases it is vegetal ornament that is the vehicle for the overall covering of the field, an idea again present in Umayyad Syria, notably in the palace façades of Qasr al-Hair West and Mshatta (fig. 17) and in panels from Khirbat al-Mafjar. The well-nigh total subordination of surface to ornament in these ivories illustrates an aesthetic derived from Umayyad Syria and radically at variance with that of say southern Italian, Sicilian, or Byzantine ivories. In this medium it also suggests that the smooth texture of this precious material was of secondary interest and that it was the decoration that gave these pieces added value. Yet architectural decoration is by no means the only source on which these ivories might have drawn; indeed, the practical obstacle of distance would have ensured that such a connection would necessarily have been indirect. It is therefore appropriate to compare like with like and thus look at the evidence provided by the ivories of Umayyad Syria. Little 110 enough has survived, but the three pyxides in London (fig. 6), Kuwait, 111 and Berlin (fig. 18), the ivory fragments in Doha and Copenhagen (fig.   The al-Sabah Collection, Kuwait, inv. no. LNS 181; for a colour illustration, see Jenkins 1983, pl. 32. 111   Museum of Islamic Art, Doha, Qatar, inv. no. IV.15.99, cf. Doha 2004, pp. 12-13. 110

212

7), and the incomplete Berlin bone panel (fig. 7) offer enough material to work on, even if they are not all of top quality. At the very least, they indicate that some of the themes so characteristic of Spanish ivories were already part of an earlier Umayyad repertoire. The links with Spanish work on the Berlin fragment, for example, include the high relief, the very deep and widespread drilling in between the leaf and tendril forms, and the dense overall covering of the surface. The Doha pieces share these features but attest many more perforations (usually four per leaf) and parcel up the space much more obviously by means of circular scrolls. As for the London piece (fig. 6), it includes an architectural motif, in this case an arcade at its base; inhabited scrolls (though these inhabitants are represented by a grand total of two undersized birds marooned somewhat uncomfortably amidst a plethora of oversized leaves); clusters of fruit; a vase out of which these plants issue; and a painted background. In the Berlin pyxis (fig. 18), the subject matter is much the same, but for the absence of the birds and the presence of four instead of two tiers of vine scrolls, which results in a density approaching that of the Berlin bone fragment (fig. 7). In both pyxides, the shallowness of the carving is a striking feature, and it is undeniable that this lack of depth reduces not only the degree of naturalism but also the sheer presence and power of the piece. Neither this low relief nor the full-bellied form of the pyxis itself is found on the Spanish ivories. Nor are the distinctive serrated, arcaded, or three-leafed borders. In the Kuwait pyxis, it is the ajouré effect which is most immediately apparent, and which invests the ornament with a distinctive lightness. Even the fruit which hangs from the tendrils functions also as pinpoints of light. In a scarcely perceptible fashion, the languidly unfolding scrolls subdivide the surface in the vertical plane into a series of compartments; it is only the regularly spaced chalices which anchor these undulations. In all this Levantine material, to which one must add the ivories of 113 Coptic Egypt, there is a strong sense of natural growth despite the obviously geometric layout. In Spanish Muslim ivories, that sense of underlying geometry is taken still further. Indeed, most of the Cordovan ivories are regimented by comparison, though they employ the same formal 114 elements. But perhaps the most instructive difference is that whereas 112

 Museum für Islamische Kunst, Berlin, inv. no. I. 5893: Spuler and SourdelThomine 1973, p. 186 and pl. 67 (K. Brisch). 113  Creswell 1969, I/2, pp. 620-622, figs. 677-680 and 683-684. 114  The pyxis, cat. no. 16a, is a good example, as is Beckwith 1960, pl. 33. 112

THE SYRIAN CONNECTION

213

the ivory and bone material from Umayyad Syria is still firmly rooted, so 115 to speak, in the Classical and late antique tradition of the vine scroll, even though that form is flattened and simplified, the Spanish craftsmen typically replace that plant with one of their own devising: their distinctive 116 version of the arabesque. That switch from a natural to a cerebral form encapsulates what happened to Islamic art in the two centuries after the fall of the Syrian Umayyads. The “sculptural” style As with the epigraphy, the many variations undergone by vegetal ornament (the discussion here leaves out of account those ivories in which living figures play a significant role) can grosso modo be fitted into just two categories. One, represented largely by objects made in the 960s and typified by the London casket and casket lid (cat. nos. 2a and 6a), the 117 ex-Zamora pyxis (cat. no. 4a), the Fitero casket (cat. no. 7a), the Madrid 118 casket (cat. no. 8a), and the Hispanic Society pyxis (cat. no. 14a), uses deep undercutting and thereby has a sculptural, three-dimensional character. Every detail stands out with full clarity and depth against the strongly accentuated background. That background usually occupies a generous space, a factor which is one of the key defining characteristics of this first style. Extra plasticity is assured by such devices as the central groove or filament which runs down most tendrils, by the contrast between the fat, full-bottomed leaves and the spindly stems from which they sprout, and by the constant rise and fall of the leaves themselves. Those leaves, moreover, are packed with a bizarre inner life. This expresses itself in a nervous, obsessive hatching of much of the surface, in the plethora of dots and tiny roundels which festoon it, in the scalloped edges, in the constantly changing convexities and concavities which lend these leaves such languorous contours, in their tendency to bifurcate, and in their sharply pointed or hooked endings. Yet all this astonishingly precise  Stern 1954 and Stern 1963.  This is not to deny that the two traditions can come very close to each other,

115 116

as in the case of an ivory fragment in the Benaki Collection – probably Coptic, but perhaps already Islamic (Creswell 1969, I/2, fig. 683) – which displays that busy veining and perforations of the leaves which seem to announce the Spanish style. Yet these are vine leaves, and the scrolls generate fat bunches of grapes – and are thus distinctly different from the vegetal ornament of Spanish ivories. 117  Colour plate in Castejón and Arizala 1985, pl. 47. 118  Its latest representative is a mid-11th century panel from Cuenca in the Victoria and Albert Museum; Beckwith 1960, pl. 33.

214

and idiosyncratic detail is subordinated to an overall design which seems effortlessly to make room for each scroll, each tendril, each leaf to unfold as it pleases – while filling all the available space with no sense of strain. And one’s first and lasting impression is of the mobility and freedom of that design. What are the basic principles at work? The Hispanic Society pyxis of about 965-970 offers a convenient point of departure (cat. no. 14a). It is also perhaps the most developed example of the contemporary Spanish arabesque. Every leaf is manicured and every tendril enriched by a lightly engraved central line. Every botanical motif occupies its allotted place in strict obedience to the underlying symmetrical design. Knots, loops, scrolls, and leaves are disposed over the surface with an immaculate sense of interval and in accordance with the principle of mirror symmetry. Or so it seems. But as you look closer you recognise that, despite this regimentation, a kind of hyper-naturalism reigns. The unpredictability of nature is everywhere – in the uneven furling, perforation, hatching, or size of leaves, in the different sub-genera of the slender tall-stemmed plants which regularly sub-divide the surface along the vertical axis, or in a stray bud emerging from a tendril. Even the motifs marking the junctions of tendrils along the lowest register of the pyxis are constantly changing. 119 Sometimes there is an annulet clasping the stem, sometimes not. So while a rigorously abstract framework ensured the even covering of the main body of the pyxis below the inscription band, the artist executing that design reserved to himself the right to vary it in detail. And that is why it lives, even though its form is invented rather than observed. Its constant undulating movement, as its scrolls double back on each other and its branching tendrils weave in and out of the foliage, would have exerted a well-nigh hypnotic force as the pyxis was slowly turned, and its labyrinthine quality would readily have suggested infinity. This is one of the great masterpieces of Islamic ornament. The “tapestry” style The other style is much less well represented among surviving ivories. Its principal hallmarks are a tapestrylike even covering of the field; flatness rather than plasticity; and the ajouré effect which causes the decoration to float over a background that has been demoted to insignificance. The  The same outworking of a basic template in a sequence of subtly different variations characterises such Syrian work as the Kuwait pyxis. 119

THE SYRIAN CONNECTION

215

casket panel from New York (cat. no. 24a) and the Palencia casket (cat. no. 29a) of 1049-1050 in Madrid illustrate two rather different sub-sets of this manner. The Palencia piece edges closer perhaps than any other ivory to the Samarra aesthetic in its broad leaves, busy infill, and regular, repetitious covering of the surface. To that extent it is atypical. It was, however, this flat tapestry style – rare though it was in ivories, which were perhaps too small to permit the full flowering of this fashion – that was favoured for panels of vegetal ornament in the great Spanish Umayyad buildings of the 10th century. The expressions of this idea in architecture vary in scale, complexity, and intensity. But at their most ambitious, as can be seen in some quite 120 remarkable examples from the Great Mosque and from Madinat al121 Zahra, they carry vegetal ornament in Islamic art into hitherto unknown territory, particularly the large panels with a “Tree of Life” motif. In 122 a celebrated example from Madinat al-Zahra, tendencies that are somewhat more latent in the ivories – for example, the dense infill of the leaves with ornament which entirely covers their surface and is even rather than uneven, flat rather than rounded – are taken to new heights (fig. 8). Musical analogies come to mind: a fugue, perhaps, or a theme with endless variations. Within an overall general symmetry there are minor but complementary divergences, so that a tendril curls over its fellow on the right and under its fellow on the left. The sovereign control of the principal motif, a perpendicular tree-trunk with successive serpentine branches curling in and out of each other, as knotted as Medusa’s locks, allows the artist to develop the infill to his heart’s content. And it proliferates obligingly to fill every nook and cranny, with no apparent strain. The result is a surface contrast of plain branches and decorated leaves; but what leaves! From one and the same stem appears as if by magic an entire lexicon of plant forms: pine-cones and pistils, calices and cornucopiae, fronds, ferns, foliate sprays and fruit, and, above all, leaves 123 galore partitioned into anything from two to eighteen segments. It can  Goitia 1965, pl. 24.  Ewert 1996 tackles this material with typical thoroughness and insight. 122  Best colour plate in Barrucand and Bednorz 1992, p. 67, with valuable 120 121

comments on the impact of the techniques of vertical carving and hard outline and with the suggestion that this is a tree of life. The same idea is put forward by Finster 2005, pp. 156 and 158; cf. also pp. 49-50. Here again there are parallels in Umayyad Syria, for example at Qasr al-Hair West; Finster 2005, p. 158. 123  The contrast with the vegetal ornament of the ivory oliphants, with its somewhat sterile, much reduced vocabulary, is startling; Shalem 2004a.

216

soberly be said that no form occurs more than once, though some forms come close. And miraculously enough, each element has room to breathe. This is an eye-popping display of virtuosity; there was no further to go 124 along this road. Conclusion It will be clear by now that there is no intention here to suggest that Spanish Umayyad artists working in ivory drew directly on Syrian Umayyad art. Such of that art as survived – portable objects excluded, though very little is known about these – was to be found, after all, thousands of miles away in hostile territory. Rather is it plausible that these ivory carvers drew on the decorative repertoire developed and refined in the other arts of Muslim Spain, especially those of carved stone and stucco. And here the norms and forms of Umayyad Syria, only slightly modified by the passage of the centuries, held unchallenged sway. They were probably brought by those “hordes of Syrians” who settled in Andalusia both before and after the Syrian Umayyad prince ‘Abd al-Rahman I came to power there. This repertoire can be pieced together from the excavations of Madinat al-Zahra. That site, like the ivories themselves, served the Umayyad court. Here if anywhere – the obvious rival is the slightly earlier, but fundamentally very different, stuccowork of Samarra – one can experience the full impact of the celebrated Islamic horror vacui. And as at Samarra, it is not just the huge surface area covered with closely-knit ornament but also the absence of living creatures in the decorative repertoire that is so noteworthy. The large-scale use of small-scale ornament creates a very distinctive aesthetic. Some of the greatest works of caliphal Spain make no sense unless they are seen in the context of this Umayyad Syrian heritage. To take a single example, the set of paired scrolls filling a rectangular marble panel row by row at Madinat al-Zahra (fig. 9) takes up a theme familiar in various guises 125 in the woodwork of the Aqsa mosque (fig. 10). From the Aqsa, too, come panels whose architectural character is as strange and unexpected in

124  A broadly parallel course is struck by the ornament at the palace of Aljafería at Saragossa; see Ewert 1978-1980. 125  For the Spanish panel, which is vertically rectangular in form, see New York 1992, p. 242. This is the standard format used in the wooden panels from the Aqsa mosque: Hillenbrand 1999b, pp. 271-310, especially pls. 39, 41-42, and 53.

THE SYRIAN CONNECTION

217

that context (fig. 19), as are the horseshoe arches of the Braga pyxis (cat. no. 21a). In both cases the infill is vegetal. Yet not all the Syrian heritage was enthusiastically embraced. The Sasanian element so pronounced in later Umayyad art, indeed the whole style of carving generated by the medium of stucco – typified by the lunette fillings of Qasr al-Hair West (fig. 5) – arouses no answering interest in Spain. There – perhaps because the Classical and Byzantine heritage in Spain was uncontaminated by themes from Iraq and Iran – the preference is unambiguously for themes derived from the vine and the acanthus and executed in stone. Filigree carving, which requires virtuoso 127 technical display, dominates at Madinat al-Zahra as at related sites, just as it does in the ivories. Spanish work asserts itself by a preference for furling in the leaves, for punctuations all over the surface, whether or not they are botanically justified, and for stylisation and abstraction that seem to veer in the direction of the human face, or caricatures of it. Once the eye registers this biomorphism, ambiguous though it is, it turns out to be pervasive. It is noticeable, too, how the inhabited scroll so characteristic of Umayyad Syria is conflated with the heavily braided and often foliate roundel, which is clearly derived from textile design and has no specific botanical character. Even such a specific detail as the palmette goes its own 128 way in this style. Yet despite these changes – to which must be added the transformations undergone by Syrian epigraphic modes on Spanish soil – Spain continued doggedly to cleave to Syria while the rest of the Islamic world moved on. It is a moving testimony to perhaps misplaced loyalty. 126

126  Hillenbrand 1999b, pls. 58-62. For a Coptic parallel which takes the same diabolical liberties with the proprieties of architecture, see Talgam 2004, fig. 140. 127  Such as Cortijo del Alcaide, less than 10 km. away from it; see Ewert 1998 and pls. 41-58. 128  Vallejo Triano 2004.

218

Bibliography Almagro et al. 1975 M. Almagro, L. Caballero, J. Zozaya, and A. Almagro: Qusayr ‘Amra. Residencia y Baños omeyas en el Desierto de Jordania, Madrid 1975. Atıl 1975 Esin Atıl: Art of the Arab World, Washington, D.C. 1975. Baker 1995 Patricia L. Baker: Islamic Textiles, London 1995. Barrucand and Bednorz 1992 Marianne Barrucand and Achim Bednorz: Moorish Architecture in Andalusia, Cologne 1992. Beckwith 1960 John Beckwith: Caskets from Cordoba, London 1960. Behrens-Abouseif 1997 Doris Behrens-Abouseif: “The Lion-Gazelle Mosaic at Khirbat alMafjar,” Muqarnas 14, 1997, pp. 11-18. Berchem 1922 Max van Berchem: Matériaux pour un Corpus Inscriptionum Arabicarum. 2e Partie. — Syrie du Sud. Tome 1er, Jérusalem “Ville,” 1er Fascicule, Cairo 1922 (Mémoires publiés par les membres de l’Institut Français d’Archéologie Orientale du Caire, 43). Blair 1989 Sheila S. Blair: “Legibility Versus Decor in Islamic Epigraphy: The Case of Interlacing,” World Art: Themes of Unity in Diversity. Acts of the XXVIth International Congress of the History of Art, ed. I. Lavin, University Park 1989, pp. 229-231. Blair 1992b Sheila S. Blair: The monumental inscriptions from early Islamic Iran and Transoxiana, Leiden 1992 (Studies in Islamic art and architecture. Supplements to Muqarnas, 5).

THE SYRIAN CONNECTION

219

Blair 1998 Sheila S. Blair: Islamic Inscriptions, Edinburgh 1998. Bloom 1989b Jonathan M. Bloom: Minaret. Symbol of Islam, Oxford 1989 (Oxford Studies in Islamic Art, 7). Bothmer 1987 Hans-Caspar Graf von Bothmer: “Architekturbilder im Koran. Eine Prachthandschrift der Umayyadenzeit aus dem Yemen,” Pantheon 45, 1987, pp. 4-20. Brisch 1968 Klaus Brisch: Die Fenstergitter und verwandte Ornamente der Hauptmoschee von Córdoba, Berlin 1966 (Madrider Forschungen, 4). Burns 1992 Ross Burns: Monuments of Syria. An historical Guide, London 1992. Castejón and Arizala 1985 Rafael Castejón y Martínez de Arizala: Medina Azahara. The Palace City of the Caliphs of Cordova, Madrid 1985. Chalmeta 1992 Pedro Chalmeta: “An Approximate Picture of the Economy of alAndalus,” The Legacy of Muslim Spain, ed. S. K. Jayyusi, Leiden 1992, pp. 741-758 (Handbuch der Orientalistik, section 1, vol. 12). Contadini, Camber and Northover 2002 Anna Contadini, Richard Camber and Peter Northover: “Beasts that roared: the Pisa Griffin and the New York Lion,” Cairo to Kabul. Afghan and Islamic Studies presented to Ralph Pinder-Wilson, eds. Warwick Ball and Leonard W. Harrow, London 2002, pp. 65-83. Creswell 1969 K. A. C. Creswell: Early Muslim Architecture. Umayyads A.D. 622-750, I, 2nd ed., 2 vols., Oxford 1969. Creswell 1979

220

K. A. C. Creswell: Early Muslim Architecture. Early Abbasids … A.D. 751905, II, reprint, New York 1979. Diez 1918 Ernst Diez: Churasanische Baudenkmäler, Berlin 1918. Ettinghausen 1962 Richard Ettinghausen: Arab Painting, Geneva 1962. Ettinghausen and Grabar 1987 Richard Ettinghausen and Oleg Grabar: The Art and Architecture of Islam 650-1250, Harmondsworth 1987. Ewert 1978-1980 Christian Ewert: Spanisch-islamische Systeme sich kreuzender Bögen. III: Die Aljafería in Zaragoza. Berlin 1978-1980; 2 vols. + Beilagen (Madrider Forschungen, 12). Ewert 1996 Christian Ewert: Die Dekorelemente der Wandfelder im Reichen Saal von Madinat az-Zahra’. Eine Studie zum westumaiyadischen Bauschmuck des hohen 10. Jahrhunderts, Mainz 1996 (Madrider Beiträge, 23). Ewert 1998 Christian Ewert: “Die Dekorelemente des spätumaiyadischen Fundkomplexes aus dem Cortijo del Alcaide (Prov. Córdoba),” Madrider Mitteilungen 39, 1998, pp. 356-532. Fernández-Puertas 1992 Antonio Fernández-Puertas: “Calligraphy in al-Andalus,” The Legacy of Muslim Spain, ed. S. K. Jayyusi, Leiden 1992, pp. 639-676 (Handbuch der Orientalistik, section 1, vol. 12). Ferrandis 1935-1940 José Ferrandis: Marfiles Árabes de Occidente, 2 vols., Madrid 1935-1940. Finster 2005 Barbara Finster: “Vine Ornament and Pomegranates as Palace Decoration in ‘Anjar,” The Iconography of Islamic Art. Studies in

THE SYRIAN CONNECTION

221

Honour of Robert Hillenbrand, ed. Bernard O’Kane, Edinburgh 2005, pp. 143-158. Flury 1920 S. Flury: Islamische Schriftbänder Amida — Diarbekr, XI. Jahrhundert, Anhang: Kairuan, Mayyâfâriqîn, Tirmidh, Basel 1920. Flury 1925 S. Flury: “Le décor épigraphique des monuments de Ghazna,” Syria 6, 1925, pp. 62-90. Flury 1930 S. Flury: “La Mosquée de Nayin,” Syria 11, 1930, pp. 43-58. García Sánchez 1992 Expiración García Sánchez: “Agriculture in Muslim Spain,” The Legacy of Muslim Spain, ed. S. K. Jayyusi, Leiden 1992, pp. 987-999 (Handbuch der Orientalistik, section 1, vol. 12). Glick 1992 Thomas F. Glick: “Hydraulic Technology in al-Andalus,” The Legacy of Muslim Spain, ed. S. K. Jayyusi, Leiden 1992, pp. 974-986 (Handbuch der Orientalistik, section 1, vol. 12). Goitia 1965 Fernando Chueca Goitia: La Moschea di Cordova, Florence 1965. Gómez-Moreno 1951 Manuel Gómez-Moreno: El arte árabe español hasta los Almohades. Arte Mozárabe, Madrid [1951] (Ars Hispaniae, 3). Grabar 1992 Oleg Grabar: “Two Paradoxes in the Islamic Art of the Spanish Peninsula,” The Legacy of Muslim Spain, ed. S. K. Jayyusi, Leiden 1992, pp. 583-891 (Handbuch der Orientalistik, section 1, vol. 12). Grohmann 1957 A. Grohmann: “The Origin and Early Development of Floriated Kufic,” Ars Orientalis 2, 1957, pp. 183-213.

222

Grube 1967 Ernst J. Grube: The World of Islam, New York [1967]. Hamilton 1959 Robert W. Hamilton: Khirbat al Mafjar. An Arabian Mansion in the Jordan Valley, Oxford 1959. Hamilton 1969 Robert W. Hamilton: “Who built Khirbat al Mafjar?” Levant 1, 1969, pp. 61-72. Hamilton 1992 Robert W. Hamilton: “Once again the Aqsa,” Bayt al-Maqdis. ‘Abd alMalik’s Jerusalem. Part one, eds. Julian Raby and Jeremy Johns, Oxford 1992, pp. 141-144 (Oxford Studies in Islamic Art, 9:1). Hillenbrand 1982 Robert Hillenbrand: “La dolce vita in Early Islamic Syria: the Evidence of Later Umayyad Palaces,” Art History 5:1, 1982, pp. 1-35. Hillenbrand 1999a Carole Hillenbrand: The Crusades. Islamic Perspectives, Edinburgh 1999. Hillenbrand 1999b Robert Hillenbrand: “Umayyad Woodwork in the Aqsa Mosque,” Bayt al-Maqdis: Jerusalem and Early Islam, ed. Jeremy Johns, Oxford 1999, pp. 271-310 (Oxford Studies in Islamic Art, 9:2). Hillenbrand 2004 Robert Hillenbrand: “For God, Empire and Mammon: Some ArtHistorical Aspects of the Reformed Dinars of ‘Abd al-Malik,” Al-Andalus und Europa zwischen Orient und Okzident, eds. Martina Müller-Wiener … [et al.], Petersberg 2004, pp. 20-38. Hitti 1968 Philip K. Hitti: Makers of Arab History, New York 1968. Ibn Hauqal 1964 Configuration de la Terre (Kitab Surat al-Ard), Ibn Hauqal, introd. and trans.

THE SYRIAN CONNECTION

223

J. H. Kramers and G. Wiet, Beirut 1964, 2 vols. Jenkins 1983 Islamic Art in the Kuwait National Museum: The al-Sabah Collection, ed. Marilyn Jenkins[-Madina], London 1983. Kennedy 1996 Hugh Kennedy: Muslim Spain and Portugal. A Political History of al-Andalus, London 1996. King 1999 David A. King: World-Maps for Finding the Direction and Distance to Mecca. Innovation and Tradition in Islamic Science, Leiden 1999 (Islamic philosophy, theology and science. Texts and studies, 36). Kühnel 1971 Ernst Kühnel: Die islamischen Elfenbeinskulpturen, VIII-XIII. Jahrhundert, Berlin 1971. Lévi-Provençal 1931a É. Lévi-Provençal: Inscriptions Arabes d’Espagne, 2 vols., Leiden 1931. Lévi-Provençal 1932 É. Lévi-Provençal: L’Espagne musulmane au Xe Siecle: Institutions et vie sociale, Paris 1932. Lévi-Provençal 1948a É. Lévi-Provençal: La civilisation des arabes en Espagne. Vue générale, Paris 1948 (Islam d’hier et d’aujourd’hui, 1). Lévi-Provençal 1948b Études… É. Lévi-Provençal: “Un échange d’ambassades entre Cordue et Byzance au XIe siècle,” É. Lévi-Provençal: Islam d’Occident. Études d’histoire médiévale, Paris 1948. Lévi-Provençal 1950-1953 É. Lévi-Provençal: Histoire de l’Espagne Musulmane, 3 vols., Paris 19501953.

224

Melikian-Chirvani 1968 Asadullah S. Melikian-Chirvani: “Le griffon iranien de Pise. Matériaux pour un corpus de l’argenterie et du bronze iraniens. III,” Kunst des Orients V:2, 1968, pp. 68-86. Muqaddimah 1967 Ibn Khaldun: The Muqaddimah. An Introduction to History, trans. Franz Rosenthal, 2nd ed., Princeton 1967, 3 vols. Musil 1907 Kusejr ‘Amra, Alois Musil … [et al.], Vienna 1907, 2 vols. New York 1992 Al-Andalus. The Art of Islamic Spain, ed. Jerrilynn D. Dodds, exhibition catalogue, The Metropolitan Museum of Art, New York 1992. Nuseibeh and Grabar 1996 Saïd Nuseibeh and Oleg Grabar: The Dome of the Rock, London 1996. Ocaña Jiménez 1970 Manuel Ocaña Jiménez: El cúfico hispano y su evolución, Madrid 1970 (Cuadernos de Historia, Economía y Derecho HispanoMusulmán, I). Pavón Maldonado 1981 Basilio Pavón Maldonado: El Arte Hispano-Musulmán en su Decoración Floral, Madrid 1981. Prado-Vilar 1997 Francisco Prado-Vilar: “Circular Visions of Fertility and Punishment: Caliphal Ivory Caskets from al-Andalus,” Muqarnas 14, 1997, pp. 19-41. Rice 1955 D. S. Rice: The Wade Cup in the Cleveland Museum of Art, Paris 1955. Rodríguez and Souto 2000 M. J. Rodríguez and J. A. Souto: “De gliptografía omeya: signos lapidarios en la Mezquita Aljama de Córdoba. Situación e indice,”

THE SYRIAN CONNECTION

225

Actes du XIe Colloque International de Glyptographie de Palma de Majorque, 1998, Braine-le-Château 2000, pp. 359-392. Ruggles 2000 D. Fairchild Ruggles: Gardens, Landscape, and Vision in the Palaces of Islamic Spain, University Park 2000. Schimmel 1970 Annemarie Schimmel: Islamic Calligraphy, Leiden 1970. Schlumberger 1946-1948 Daniel Schlumberger: “Deux fresques omeyyades,” Syria XXV:1-2, 19461948, pp. 85-102. Schlumberger 1986 Daniel Schlumberger: Qasr el-Heir el Gharbi, Paris 1986 (Institut Français d’Archéologie du Proche-Orient, Bibliothèque Archéologique et Historique, 120). Shalem 2004a Avinoam Shalem: The Oliphant. Islamic Objects in Historical Context, Leiden 2004 (Islamic History and Civilization. Studies and Texts, 54). Souto 1998 Juan A. Souto: “Epigraphy and building in Umayyad al-Andalus: Genesis and prospects for a research project,” Philosophy and Arts in the Islamic World. Proceedings of the Eighteenth Congress of the UEAI, eds. U. Vermeulen and D. De Smet, Louvain 1998, pp. 303-324. Souto 2001 Juan A. Souto: “Glyptographie omeyyade: signes lapidaires à la Grande Mosquée de Cordoue. Documentation de noms propres,” Actes du XIIe Colloque International de Glyptographie de Saint-Christophe-enBrionnais, 2000, Braine-le-Château 2001. Souto 2004 Juan A. Souto: “De gliptografía omeya: Signos lapidarios en la Mezquita Aljama de Córdoba. Ejemplares fuera del oratorio: Situación y ensayo de cronología,” Al-Andalus und Europa zwischen Orient und

226

Okzident, eds. Martina Müller-Wiener … [et al.], Petersberg 2004, pp. 92-100. Sourdel-Thomine 1973 Die Kunst des Islam, eds. B. Spuler and J. Sourdel-Thomine, Berlin 1973 (Propyläen Kunstgeschichte, 4). Stern 1954 Henri Stern: “Quelques oeuvres sculptées en bois, os et ivoire de style omeyyade,” Ars Orientalis I, 1954, pp. 119-131. Stern 1963 Henri Stern: “The Ivories on the ambo of the Cathedral of Aix-laChapelle,” The Connoisseur 153:617, 1963, pp. 166-171. Strzygowski 1911 J. Strzygowski: “Ornamente altarabischer Grabsteine in Kairo,” Der Islam II, 1911, pp. 305-336. Talgam 2004 Rina Talgam: The Stylistic Origins of Umayyad Sculpture and Architectural Decoration, Wiesbaden 2004, 2 vols. Torres Balbás 1971 Leopoldo Torres Balbás: Ciudades Hispano-Musulmanas. Advertencia preliminar introd. y conclusión por Henri Terrasse, Madrid 1971. Venice 1993 Eredità dell’Islam. Arte islamica in Italia, ed. Giovanni Curatola, exhibition catalogue, Palazzo Ducale, Venice, Milan 1993. Viollet and Flury 1921 H. Viollet and S. Flury: “Un monument des premiers siècles de l’hégire en Perse,” Syria II, 1921, pp. 226-234, pls. 28-34. Volov 1966 Lisa Volov: “Plaited Kufic on Samanid Epigraphic Pottery,” Ars Orientalis VI, 1966, pp. 107-133. Walker 1941

THE SYRIAN CONNECTION

J. Walker: A Catalogue of the Muhammadan Coins in the British Museum. [1]: A Catalogue of the Arab-Sassanian Coins (Umaiyad governors in the east, Arab-Ephthalites, ‘Abbasid governors in Tabaristan and Bukhara, London 1941. Walker 1956 J. Walker: A Catalogue of the Muhammadan Coins in the British Museum. 2: A Catalogue of the Arab-Byzantine and Post-Reform Umaiyad Coins, London 1956. Weibel 1972 Adèle C. Weibel: Two Thousand Years of Textiles. The Figured Textiles of Europe and the Near East, reprint, New York 1972. Wiet 1936-1940 Stèles Gaston Wiet: Catalogue général du Musée arabe du Caire. Steles funéraires, II and IV-X, Cairo 1936-1942.

227

228

The catalogue numbers given from pages 228 to 263 refer to the catalogue included in the 2-volume book by K. von Folsach and J. Meyer: Journal of the David Collection. The Ivories of Muslim Spain. Papers from a Symposium held in Copenhagen from the 18th to the 20th of November 2003 [Volumes 1-2 of Journal of the David Collection: Davids Samling, 2005].

Cat. no. 2a CAT. NO. 2. CASKET. Victoria and Albert Museum, London, inv. no. 301-1866. H: 4.5; W: 9.5; D: 7 cm. After 961. Made for a daughter of ‘Abd al-Rahman III. Ferrandis 1935-1940, I, cat. no. 2; Beckwith 1960, pp. 6-7, pls. 2-4; Kühnel 1971, cat. no. 20. Illustrations: Cat. no. 2a; figs. 23, 32d, 32e, 32b 32g, 162, 163, 164, 165, 166.

THE SYRIAN CONNECTION

Cat. no. 3a. CAT. NO. 3. CASKET. Victoria and Albert Museum, London, inv. no. A.580-1910. H: 8; W: 12.5; D: 8 cm. After 961. Made for a daughter of ‘Abd al-Rahman III. Ferrandis 1935-1940, I, cat. no. 3; Beckwith 1960, p. 34, figs. 32-33; Kühnel 1971, cat. no. 21. Illustrations: Cat. no. 3a; figs. 162, 163, 164, 165, 166.

229

230

Cat. no. 4a. CAT. NO. 4. PYXIS (“THE ZAMORA PYXIS”). Museo Arqueológico Nacional, Madrid, inv. no. 2.113. H: 18; Diam: 10.5 cm. 353 / 964. Ordered by al-Hakam II, supervised by Durri al-Saghir, made for Subh. Ferrandis 1935-1940, I. cat. no. 4; Beckwith 1960, pp. 10-13. pl. 6, Kühnel 1971, cat. no. 22. Illustrations: Cat. no. 4a: fig. 61.

THE SYRIAN CONNECTION

Cat. no. 5a. CAT. NO. 5. PYXIS. Victoria and Albert Museum, London, inv. no. 217-1865. H: 7.7; Diam: 10 cm. C. 965. Ordered by al-Hakam II, supervised by Durri al-Saghir. Ferrandis 1935-1940, I, cat. no. 5; Beckwith 1960, p. 14, pls. 7-8; Kühnel 1971, cat. no. 27. Illustrations: Cat. no. 5a, b.

Cat. no. 5b

231

232

Cat. no. 6a. CAT. NO. 6. LID OF CASKET. Victoria and Albert Museum, London, inv. no. 1057-1855. H: 7.5; W: 13.5 cm. C. 965. Ferrandis 1935-1940, I, cat. no. 10; Beckwith p. 14, pl. 11; Kühnel 1971, cat. no. 26. Illustrations: Cat. no. 6a.

THE SYRIAN CONNECTION

233

Cat. no. 7a. CAT. NO. 7. CASKET (“THE FITERO CASKET”). Santa Maria La Real, Fitero. H: 8.9; W: 12.8; D: 8.3 cm. 355 / 965-966. Made in Madinat al-Zahra by Khalaf for Subh. Ferrandis 1935-1940, I, cat. no. 8; Beckwith 1960, p. 14, pl. 10; Kühnel 1971, cat. no. 23. Illustrations: Cat. no. 7a, b, c; figs. 32a, 69

234

Cat. no. 7b

THE SYRIAN CONNECTION

Cat. no. 7c. Right side

235

236

Cat. no. 8a CAT. NO. 8. CASKET. Instituto de Valencia de Don Juan, Madrid, inv. no. 4860. H: 4.3; W: 8.5; D: 5.8 cm. 355 / 965-966. Made in Madinat al-Zahra for Subh. Ferrandis 1935-1940, I, cat. no. 7; Beckwith 1960, p. 14, pl. 9; Kühnel 1971, cat. no. 24. Illustrations: Cat. no. 8a.

THE SYRIAN CONNECTION

Cat. no. 9a. CAT. NO. 9. CASKET. Musée des Arts Décoratifs, Paris, inv. no. 4417. H: 10.2; W: 20; D: 12.5 cm. 355 / 965-966. Ferrandis 1935-1940, I, cat. no. 11; Beckwith 1960, p. 16, pl. 13; Kühnel 1971, cat. no. 25. Illustrations: Cat. no. 9a, b; figs. 31f, 63, 64.

Cat. no. 9b

237

238

Cat. no. 10a. CAT NO. 10. CASKET. The David Collection, Copenhagen, inv. no. 9/2002. H: 9.9; W: 14-3; D: 9.3 cm. C. 966-968. Not in Ferrandis 1935-1940, Beckwith 1960, or Kühnel 1971. Illustrations: Cat. no. 10a, b, c, d, e, f; figs. 65, 66, 68, 70, 71, 72.

Cat. no. 10b

THE SYRIAN CONNECTION

Cat. no. 10c. Left side.

Cat. no. 10d

239

240

Cat. no. 10e. Right side.

Cat. no. 10f

THE SYRIAN CONNECTION

241

Cat. no. 11a. CAT. NO. 11. PYXIS (“THE AL-MUGHIRA PYXIS”). Musée du Louvre, Paris, inv. no. AO 4068. H: 17.6; Diam: 11.5 (top) / 11.2 (bottom) cm. Weight: 269 g (lid), 876 g (total). 357 / 967-968. Made for al-Mughira. Ferrandis 1935-1940, I, cat. no. 13; Beckwith 1960, pp. 16-20, pls. 14-17 and 22; Kühnel 1971, cat. no. 31. Illustrations: Cat. no. 11a, b, c, d, e; figs. 25, 31c, 31d, 74, 75, 77.

242

Cat. no. 11b

THE SYRIAN CONNECTION

Cat. no. 11c

243

244

Cat. no. 11d

THE SYRIAN CONNECTION

Cat. no. 11e

245

246

Cat. no. 12a. CAT. NO. 12. PYXIS. Victoria and Albert Museum, London, inv. no. 368-1880. H: 18.5; Diam: 11.5 cm. 359 / 969-970. Made for Ziyad ibn Aflah. Ferrandis 1935-1940, I, cat. no. 14; Beckwith 1960, pp. 20-21, pls. 18-22; Kühnel 1971, cat. no. 32. Illustrations: Cat. no. 12a, b; figs. 21, 171.

THE SYRIAN CONNECTION

Cat. no. 12b

247

248

Cat. no. 13a. CAT. NO. 13. PYXIS (LID MISSING). Musée du Louvre, Paris, inv. no. OA 2774. H: 10.3; Diam; 11 cm. C. 970. Ferrandis 1935-1940, I, cat. no. 15; not in Beckwith 1960; Kühnel 1971, cat. no. 33. Illustrations: Cat. no. 13a.

THE SYRIAN CONNECTION

249

Cat. no. 14a. CAT. NO. 14. PYXIS. Hispanic Society of America, New York, inv. no. D752. H: 16.5; Diam (bottom): 9.8 cm. C. 970. Made by Khalaf. Ferrandis 1935-1940, I, cat. no. 9; Beckwith 1960, p. 14, pl. 12; Kühnel 1971, cat. no. 28. Illustrations: Cat. no. 14a; figs. 19, 27, 33b, 59.

250

Cat. no. 15a CAT. NO. l5. PYXIS (LID MISSING). The Metropolitan Museum of Art, The Cloisters, New York, inv. no. 1970.234.5 (ex Kofler-Truniger Collection). H: 11.8; Diam: 10.5 (top)/10.49 (bottom) cm. Weight: 441g. Not in Ferrandis 1936-1940 or Beckwith 1960; Kühnel 1971, no. 29. Illustrations: Cat. no. 15a; figs. 22, 24, 28.

THE SYRIAN CONNECTION

251

Cat. no. 16a. CAT. NO. l6. PYXIS (LID MISSING). The Metropolitan Museum of Art, New York, inv. no. 30.95.175. H: 6.8; Diam: 8 cm. C. 970. Ferrandis 1935-1940, I, cat. no. 12; not in Beckwith 1960; Kühnel 1971, cat. no. 30. Illustrations: Cat. no. 16a.

252

Cat. no. 17a. CAT. NO. 17. LID OF PYXIS. The Ashmolean Museum, Oxford, inv. no. 1987.3. H: 4; Diam: 10.5 cm. Weight: 200 g. 998-999. Made for Sanchuelo. Not in Ferrandis 1935-1940, Beckwith 1960, or Kühnel 1971. Illustrations: Cat. no. 17a, b; fig. 161.

Cat. no. 18a. CAT. NO. l8. CASKET. Museo Nazionale del Bargello, Florence, inv. no. Bg C81. H: 6.9; W: 12.9; D: 9.3 cm. End of 10th century. Ferrandis 1935-1940, I, cat. no. 21; Beckwith 1960, p. 29, pl. 26; Kühnel 1971, cat. no. 34. Illustrations: Cat. no. 18a; fig. 34a.

THE SYRIAN CONNECTION

Cat. no. 19a. 19b CAT. NO. l9. CASKET. (“The Doha box). Qatar Museum, Doha, inv. no. IV.04.98. H: 4.5; W: 36.7; D: 7.1 cm. Weight: 1159g. 395/1003-1004. Not in Ferrandis 1935-40, Beckwith 1960, or Kühnel 1971. Illustrations: Cat. no. 19a, b, c; figs. 83, 84, 85, 86.

253

254

Cat. no. 20a

Cat. no. 20b

THE SYRIAN CONNECTION

Cat. no. 20c. Right side. CAT. NO. 20. CASKET (“THE PAMPLONA CASKET”). Museo de Navarra, Pamplona, inv. no. 1360-B. H: 23.6; W: 38.4; D: 23.7 cm. 395 / 1004-1005. Ordered by ‘Abd al-Malik ibn Mansur, supervised by Zuhayr ibn Muhammad al-‘Amiri, made by Faraj, Misbah, Khayr, Rashid, and Sa‘ada. Ferrandis 1935-1940, I, cat. no. 19; Beckwith 1960, pp. 26 and 29, pls. 23-24; Kühnel 1971, cat. no. 35. Illustrations: Cat. no. 20a, b, c, d; figs. 31a, 31b, 170.

Cat. no. 20d. Detail on the front.

255

256

Cat. no. 21a. CAT. NO. 21. PYXIS. Cathedral Treasury, Braga. H: 20; Diam: 10.4 cm. C. 1005-1006. Ordered by ‘Abd al-Malik ibn Mansur, supervised by Zuhayr ibn Muhammad al-’Amiri. Ferrandis 1935-1940, I, cat. no. 20; Beckwith 1960, p. 29, pl. 25; Kühnel 1971, cat. no. 36. Illustrations: Cat. no. 21a; figs. 34b, 56.

THE SYRIAN CONNECTION

257

Cat. no. 22a

Cat. no. 22b. Right side. CAT. NO. 22. CASKET. Victoria and Albert Museum, London, inv. no. 10-1866. H: 21.5; W: 27; D: 16.5 cm. Beginning of 11th century. Ferrandis 1935-1940, I, cat. no. 22; Beckwith 1960, pp. 29-30, pls. 27-30; Kühnel 1971, cat. no. 37. Illustrations: Cat. no. 22a, b.

258

Cat. no. 23a. CAT. NO. 23. PYXIS (WITH LATER LID). The al-Sabah Collection, Kuwait, inv. no. LNS 19 I (ex Marquis de Ganay Collection). H: 9.9; Diam: 7.5 cm. Beginning of 11th century. Ferrandis 1935-1940, I, cat. no. 16; not in Beckwith 1960; Kühnel 1971, cat. no. 38. Illustrations: Cat. no. 23a.

Cat. no. 24a. CAT. NO. 24. PLAQUE. The Metropolitan Museum of Art, New York, inv. no. 13.141. H 10.8; W: 20.3 cm. 11th century or earlier. Ferrandis 1935-1940, I, cat. no. 6; Beckwith, p. 13, pl. 5; Kühnel 1971, cat. no. 39. Illustrations: Cat. no. 24a; figs. 26, 29.

THE SYRIAN CONNECTION

259

Cat. no. 25a

260

Cat. no. 25b. Right side. CAT. NO. 25. CASKET (“THE SILOS CASKET”). Museo dc Burgos, Burgos, inv. no. 198. H: 19; W: 34; D: 21 cm. 417 / 1026-1027. Made in Cuenca by Muhammad ibn Zayyan. Ferrandis 1935-1940, I, cat. no. 25; Beckwith 1960, p. 30, pl. 31; Kühnel 1971, cat. no. 40. Illustrations: Cat. no. 25a, b; fig. 33d.

THE SYRIAN CONNECTION

261

262

Cat. no. 26a

Cat. no. 28a. CAT. NO. 28. PYXIS. Treasury of the Cathedral of Saint-Just and Saint-Pasteur, Narbonne, inv. no. Cl. MH 13.06, 1906. H: 10; Diam: 7.3 cm. C. 1040-1050. Made in Cuenca for Isma‘il ibn Yahya al-Ma’mun. Ferrandis 1935-1940, I, cat. no. 30; Beckwith 1960, p. 32; Kühnel 1971, cat. no. 42. Illustrations: Cat. no. 28a; fig. 60.

THE SYRIAN CONNECTION

Cat. no. 29a. CAT. NO. 29. CASKET (“THE PALENCIA CASKET”). Museo Arqueológico Nacional, Madrid, inv. no. 7.371. H: 23; W: 34; D: 23.5 cm. 411 / 1049-1050. Made in Cuenca by ‘Abd al-Rahman ibn Zayyan for Isma‘il ibn Yahya alMa’mun. Ferrandis 1935-1940, I, cat. no. 27; Beckwith 1960, pp. 30, 32, pl. 32; Kühnel 1971, cat. no. 43. Illustrations: Cat. no. 29a; figs. 33e, 57.

263

XI Islamic Figural Sculpture

T

he Islamic world was remarkably united in an instinctive and rooted distaste for figural sculpture as architectural decoration. Even when figural sculpture was most popular, as in Syria in the first half of the 8th century or in Anatolia and the Jazira in the 12th and 13th, an appreciation of figural sculpture was an acquired and unusual taste. The theological prohibition of figural art in all contexts began to be formulated in a precise and exhaustive manner in the 8th century, but the unequivocal presence of figural sculpture on the palaces of the Umayyad caliphs, who were the temporal and spiritual heads of the Muslim community, shows either that the prohibition was not yet in effect or that there was a clear dividing line between the religious and secular spheres. Later an aversion to figural art operated primarily in the religious sphere, and the ban on sculpture was contravened only in specific contexts. In the public domain these included talismans and expressions of victory on bridges and city or castle gates; in a more private context such images were confined to bath and palace interiors. The taste for sculpture usually evinced a latitudinarianism, a broad-minded indifference to the finer points of theological detail, rather than a marked enthusiasm for such decoration. Pre-Islamic or non-Islamic figural sculpture was sometimes used in Islamic architecture for triumphal purposes. For example, pharaonic reliefs were reused as thresholds in the Islamic monuments of Cairo, such as the funerary madrasa of Barquq intra muros (1386), and Indian sculptures from Somnath were sent by Mahmud of Ghazna (reigned 998–1030) for the threshold of the ḥaram at Mecca. Less obviously religious in tenor were Classical statues placed at intervals along the city walls of Konya. In Iran the mosque at Qazvin was known as the Mosque of the Bull, presumably

ISLAMIC FIGURAL SCULPTURE

265

because Achaemenid bull-headed protomes were reused to support the roof. In the Yemen pre-Islamic figural carving was reused in the congregational mosques at San‘a’ and Dhamar. The unusual interest in figural sculpture during the later Umayyad period in Syria was triggered by the Romano-Byzantine art of the eastern Mediterranean region. In all probability such work was also produced further afield in Egypt and Iraq. This figural sculpture was employed exclusively in palatial residences and vividly expressed the sybaritic lifestyle followed there. Virtually all the surviving sculpture comes from three excavated sites, Qasr al-Hayr West, Khirbat al-Mafjar and Mshatta. All three were caliphal residences and had figural sculpture on their exterior façades in plain public view. Khirbat al-Mafjar and Mshatta are attributable to the patronage of al-Walid II (reigned 743–4), who even as heir-apparent delighted in flying in the face of public opinion, but Qasr al-Hayr is attributable to Hisham (reigned 724–43), a ruler who observed religious forms punctiliously. These sculptures show a wide range of subjects in unprecedented combinations. A cycle of images associated with the princely lifestyle had been developed by the end of the Umayyad period. This princely cycle is expressed most fully in the stucco sculptures from Qasr al-Hayr, which include enthroned rulers, in both the Byzantine and Sasanian manner, female attendants offering flowers and beverages, male servants carrying animals and dishes, bodyguards, reclining figures (perhaps intended to represent banqueters), strutting birds and pacing lions. They occupy the interstices of a continuous balustrade encircling the upper storey of the palace court, a location that enabled them to function as visual referents to the courtly activities pursued in the rooms behind. At Khirbat al-Mafjar the porch of the bath-hall is dominated by the image of a caliph with drawn sword supported by a pair of snarling lions. Traces of colour indicate that polychromy and gilding were essential parts of the decorative scheme. Figures of mountain sheep on the ledge below were presumably intended as royal symbols in the Persian manner. Within the porch at gallery level, a bevy of bare-breasted maidens offer drinks and flowers to the visitor. Among them was an incongruous figure garbed like a Roman centurion. Athletes in loincloths, whose poses parody atlantids, occupied the pendentives. The dome over the diwan, the private audience chamber within the bath-hall, was decorated with birds, winged horses and a striking arrangement of six heads emerging from a flower, all suggestive of apotheosis. The significance of other themes, such as the heads emerging

266

from knotted medallions in the vault of the entrance vestibule, remains enigmatic. On the façade at Mshatta, almost all the triangular panels to the left of the entrance are inhabited by real and imaginary animals, whereas those to the right, behind the palace mosque, have no living beings. The western Islamic world, including Spain, was generally hostile to figural decoration. Few examples of architectural sculpture survive, and the scattered references in literary sources indicate that figural decoration was primarily used in palaces. According to al-Maqqari (d 1632), the Umayyad caliph ‘Abd al-Rahman III al-Nasir (reigned 912–61) placed a statue of his favourite wife al-Zahra’ over the gate of Madinat al-Zahra’, the palace-city near Córdoba named after her, and excavations there have yielded a fragmentary sculpture of a draped figure executed in the Roman tradition. Lions formerly adorned the city gate at Mahdia, the Fatimid capital in Tunisia inaugurated in 921, and a marble relief of a cross-legged ruler listening to a flautist was unearthed there; the remains of lead in the crevices suggest that the surface was enlivened by pieces of stone in variegated colours. Stucco sculptures representing various creatures, including a parrot with fruit in its beak and lions, were found in the ruins of Sabra al-Mansuriyya, the next Fatimid capital in Tunisia. According to the poet Ibn Hamdis (d 1132–3), lion sculptures adorned the gate of the Zirid palace at Bougie (Algeria) and acted as waterspouts in fountains there. A grey marble lion found at Qal‘at Bani Hammad in Algeria apparently adjoined the springing of the arch of a palace gate, and the same place on the Gate of the Udayas at Rabat (c. 1195–9) was decorated with a sculpted fish. A column in the zāwiya of Sidi ‘Abid alGharyani at Qairawan in Tunisia shows two confronted and highly stylized birds drinking from a chalice. The principal example of later architectural sculpture from Spain is a stucco slab depicting interlaced arches weirdly interspersed with human heads (Berlin, Museum für Islamische Kunst ). It was probably made for the Nasrids (1230-1492), and the hairstyles suggest the impact of western European models. Blocklike lions whose stylized ferocity excludes naturalism decorate fountains in the Patio de los Leones and the Palacio del Partal at the Alhambra in Granada. The most significant body of figural sculpture to survive from the western Islamic world is a series of carved basins and tanks or troughs, perhaps intended for ritual ablution. The earliest dated piece is the tank made in 987–8 at Madinat al-Zahra’ for the chief minister Ibn Abi‘Amir, known as al-Mansur (Madrid, Museo Arqueologico Nacional). Its propagandistic purpose is clear from its decoration, which juxtaposes

ISLAMIC FIGURAL SCULPTURE

267

the patron’s name and titles with lions crouching on the wings of eagles grasping deer. A similar but larger (700×1550×820 mm) tank now in the Ben Yusuf Madrasa in Marrakesh was made for al-Mansur’s son ‘Abd alMalik, probably at Madinat al-Zahra’ between 1002 and 1007. The bestknown piece is the Pila of Játiva (Játiva, Museo Munincipal.), which was probably made for a rich commoner in the 11th century. Its repertory of images from the princely cycle is the fullest yet encountered on a single work of art from Muslim Spain. Figural sculpture in relief or in the round enjoyed a certain vogue in palace decoration in the Iranian world between the 11th and the 13th century. A large number of almost life-size stucco reliefs and sculptures of princes, attendants and scenes of court life are said to have come from Rayy in Iran, but their authenticity has been questioned. The largest quantity of figural sculpture in stucco scientifically excavated in the Iranian world was found in the ruined palace at Termez on the Oxus River. Figural panels were set above a dado of geometric, floral and epigraphic motifs. Arched panels containing animals were flanked by others with strapwork patterns or with curvilinear motifs in a debased bevelled style. Two of the four panels depict fearsome pairs of lions emerging from a single head: the head above the rampant pair has an extended tongue and Medusa-like mane. Two other damaged panels depict a pair of addorsed crouching felines and a lion bringing down a bull, and fragments were found of two griffins sculpted in the round. The subject-matter and decoration with rosettes and solar discs suggest a zodiacal or apotropaic meaning. Similar motifs characterize the contemporary stucco-work at Khulbuk in Tajikistan and the marble sculpture found at Ghazna in Afghanistan, capital of the Ghaznavid dynasty (977–1186), where an octagonal water basin is bordered with a variety of real and imaginary animals. A series of waterspouts in the form of lions with gaping jaws seems to have been part of the irrigation system in the palace garden. A panel depicting a leaping monkey and a man carrying a basket is undoubtedly derived from Indian models, as are the lotus shape of the basin and creatures sharing a single head. The extensive campaigns waged by the Ghaznavids in north India and the fabulous booty they amassed there easily explain this transfer of motifs. An unusually rich tradition of figural sculpture emerged during the 12th and 13th centuries in parts of Turkey, Syria, Iraq and Iran, which were then under the control of several Turkish dynasties, including the Saljuks of Anatolia and the Artuqids. This tradition waned rapidly

268

in the 14th century when the area came under the domination of the Ilkhanid dynasty. Most surviving examples were carved in stone, but excavations of the Saljuq palaces at Kubadabad near Beyşehir and at Konya have revealed significant quantities of figural decoration in stucco. Figural sculptures decorated a variety of building types, including castles, palaces, caravanserais, bridges, mausolea, madrasas and even mosques. Nevertheless, this taste was clearly exceptional, for figural decoration is found on only eight of the hundred surviving caravanserais and only three mausolea, but its presence on six mosques and three madrasas shows that it was not an aberration. The repertory includes real and imaginary animals, birds and people. Lions were most commonly represented, followed by single- and double-headed eagles; leopards, bulls, panthers, dogs, horses, hares, deer, peacocks, serpents, dragons, sphinxes and harpies were also shown. These creatures were depicted singly or in confronted or addorsed pairs. Representations of the human figure include rulers, court officials, servants, angels and sirens. One of the most famous depictions was on the Talisman Gate at Baghdad (1221; destroyed in 1917), where a seated monarch, presumably the ‘Abbasid caliph al-Nasir, held two dragons by their tongues. Animals and figures were occasionally shown in association with fruiting trees, as at the Çifte Minareli Madrasa at Erzurum (1243) and the Döner Kümbed at Kayseri (late 13th century), or with signs of the zodiac, as in the combination of the lion and sun at the bridge at Cizre (1164). The most complete zodiacal cycle survives at the octagonal mausoleum at Khachen Dorbatly in Nagorno-Karabagh (1314), but the dodecagonal tomb of the Ilkhanid sultan Ghazan Khan at Tabriz (1304) probably displayed the full zodiacal cycle in a more appropriate setting. These figural representations were usually distributed somewhat haphazardly over the exterior of buildings, but they were also used as gargoyles or applied to such significant locations as window and door surrounds and the spandrels of arches or gateways. The meaning of these images has been hotly disputed. Some were heraldic emblems, such as the panther found at Lydda (Israel), which is associated with Baybars I (reigned 1260–77), the Mamluk sultan of Egypt. The eagles on the walls of Diyarbakir are Artuqid emblems. Many were references to royalty. Lions and eagles, quintessential images of sovereignty, often decorate buildings erected under royal patronage, and others, such as the lions at the castle of Mayyafariqin (1156) or the double-headed lion at Alay Han (c. 1215), are more generalized symbols of royalty. Some are personal crests, while others referred to ownership, as in

ISLAMIC FIGURAL SCULPTURE

269

the simplified images of animals included in the tamghas or brands, used by Turkish nomads. Other images express shamanistic, totemic or magic beliefs. Many representations were probably apotropaic in function, but the role of others, such as the menageries at Ak Han (1253–4) and Karatay Han (completed 1240–41) or the series of animal heads at the Sunqur Bey Mosque at Niğde (1335; destroyed.) and the Gök Madrasa at Sivas (1271), is unclear. Images of the sun and moon at the hospital of Kayka’us at Sivas (1217–18), the minaret of Abu’l-Muzaffar at Mayyafariqin (1212) and the mosque of Alaeddin at Niğde (1223) may have been references to spiritual illumination or a survival of the age-old veneration of heavenly bodies by the Turkic tribes of Central Asia, Mongolia and Siberia. The combination of the lion and the sun on the Incir Han and the bridge at Cizre was presumably intended in a zodiacal sense, as the planet Sun under the influence of Leo. Some images may also have been intended to refer to the afterlife, for ‘He became a falcon’ is a common Turkish expression to describe death, and the tree of life surmounted by an eagle is often used as a symbol of paradise. It is hard to reconcile a single directed iconography with the often haphazard nature and placement of these images. Many of these representations may have been inspired by the Armenian tradition of stone sculpture, for they share high relief, haphazard distribution over the exterior and emblematic significance. The unique architectural example of zoomorphic script, found around the main keep of the fortress at Baku on the Caspian Sea (1234–5), may also have been derived from Armenian manuscripts, although it was commonly used on Islamic metalwork.

Chronological Bibliography S. M. Flury: Die Ornamente der Hakim- und Ashar-Moschee (Heidelberg, 1912) R. M. Riefstahl: ‘Persian Islamic Stucco Sculpture’, Art Bulletin, xiii (1931), pp. 439–63 K. A. C. Creswell: Early Muslim Architecture (2nd. ed., Oxford, 1969). Uneyyads A.D. 622-750 M. Dimand: ‘Studies in Islamic Ornament, i: Some Aspects of Omaiyad and Early Abbasid Ornament’, Ars Islamicaiv (1937), pp. 293–337 B. P. Denike: Arkhitekturnii ornament srednei Azii [The architectural

270

ornament of Central Asia] (Moscow, 1939) M. Dimand: ‘Studies in Islamic Ornament, ii’, Archaeologica Orientalia in Memoriam Ernst Herzfeld, ed. G.C Miles (Locust Valley, NY, 1952), pp. 62–8 R. Ettinghausen: ‘The “Beveled Style” in the Post-Samarra Period’, Archaeologica Orientalia in Memoriam Ernst Herzfeld, ed. G.C Miles (Locust Valley, NY, 1952), pp. 72–83 L. I. Rempel’: Arkhitekturni ornament uzbekistana [The architectural ornament of Uzbekistan] (Tashkent, 1961) D. Hill and O. Grabar: Islamic Architecture and its Decoration, A.D. 800– 1500 (Chicago, 1964) A. U. Pope and P. Ackerman, eds: A Survey of Persian Art from Prehistoric Times to the Present (London and New York, 1938–9, 2/1964–7), iii, pp. 1258–364 E. Baer: ‘The “Pila” of Játiva: A Document of Secular Urban Art in Western Islam’, Kunst des Orients, vii (1970–71), pp. 144–66 D. Hill and L. Golvin: Islamic Architecture in North Africa (London, 1976) C. Kessler: The Carved Masonry Domes of Mediaeval Cairo (London, 1976) M. S. Bulatov: Geometricheskaya garmonizatsiya v arkhitekture srednei Azii, IX–XV vv. [Geometric harmonization in the architecture of Central Asia, 9th–15th centuries] (Moscow, 1978) G. Öney: ‘Architectural Decoration and the Minor Arts’, in Art and Architecture of Turkey, ed. E. Akurgal (New York, 1980), pp. 170–74 C. Ewert and J.-P. Wisshak: Forschungen zur almohadischen Moschee, i (Mainz, 1981) E. Whelan: ‘Representations of the Khāssakīyah and the Origins of Mamluk Emblems’, in Content and Context of Visual Arts in the Islamic World, ed. P. P. Soucek (New York, 1988), pp. 219–53

XII Islamic Art: ‘Infinite Riches in a Little Room’?

T

here can be little doubt that, within the discipline of art history in general, Islamic art history has Cinderella status. Its concerns are at best tangential to most art historians, and it is therefore prevented — like the art history of other non-European cultures — from establishing continuous dialogue with the rest of the profession. As a result its focus is turned inwards. Secure behind their barricades of unpronounceable names bristling with macrons and diacritical points, Islamic art historians tend (with a few honourable exceptions) to write for the narrow constituency of their fellow-specialists. Stagnation and the obsessive pursuit of trivia are among the dangers of such a course. It may legitimately be asked why Islamic art history is so isolated and isolating a field. Part of the reason, of course, lies in its very strangeness and unfamiliarity. Most of those who work and publish in it have a Western training, but that training normally begins only at the postgraduate level, Thus the nascent scholar of Islamic art finds that much of his earlier education is of only very general value with marginal relevance to his new field. He has to lay foundations while his colleague in Western art history is already on the first floor. He must grapple not only with European languages, in which like other art historians he may already have some training, but also with the formidable languages of the Near East. He must above all familiarise himself with an alien culture and its history. These complex problems create a series of obstacles to the free interchange of ideas and information between specialists in Islamic and Western art. The

272

resultant divorce between them is effectively symbolised by the fact that they have in the past rarely published in the same journals. Reasons of a still more practical nature have ensured that Islamic art history has developed at a different speed and in a different direction from those of its Western counterpart. Broadly speaking, Islamic art has no large-scale sculpture or — even more important — easel painting. Moreover, it is intrinsically not only an impersonal art but one which purposely sets its face against the cult of the individual. Names are indeed recorded, and in plenty. But there is no archival material, no Vasari, no letters or court transcripts to bring the bare names to life. The Islamic artist had no daemon and it was only very rarely indeed that he rose above the status of a craftsman. If one adds to such immanent characteristics of the subject the fact that so much Islamic art and architecture is physically remote from the Western world and difficult of access, the reasons for its neglect will be sufficiently clear. One final compounding factor may be noted — the dearth of university posts in the subject. This has had a domino effect. Few students are trained, fewer still are appointed to permanent posts, and the majority of these posts are themselves in museums and therefore involve no regular teaching. Small wonder, then, that Islamic art history lags somewhat behind that of Western Europe. The uneven development of the field was highlighted in 1976, that annus mirabilis for Islamic art, when the World of Islam Festival generated scores of books, films, lectures and exhibitions of which disappointingly little was of permanent value. The World of Islam Festival was, however, a landmark in public awareness of Islamic art. One has only to set the publications associated with the 1976 Festival beside those of the great Persian Art exhibition of 1931 to realise what an information explosion had occurred in the intervening years. Publication in Islamic art has continued to increase in an almost geometrical progression and this seems a convenient moment to review some of the productions of the last few years. Shortage of space makes it impossible to provide a detailed review of each of the numerous books dealt with here, and they will therefore be discussed according to the categories to which they belong. Easily the largest category is that of the catalogue. This is symptomatic of publications on Islamic art. Such catalogues are of many different types. Only a small minority, unfortunately, are intended as detailed records of a permanent collection; most commemorate an exhibition or are associated with an auction sale. Indeed, to this day much work published on the

INFINITE RICHES

273

minor arts of Islam caters for the collector. This was even more the case in the early decades of the century, and is still largely true of publications on carpets. Among the catalogues reviewed here, three stand out. Two concern book painting: the catalogue of the Persian illustrated manuscripts in the John Rylands Library, Manchester, by Basil Robinson, and that of Turkish painting in the British Museum by Norah Titley. The Rylands catalogue brings to a triumphant conclusion a trilogy which has been in the making for almost half a century and began with Mr Robinson’s B. Litt. thesis on the Persian paintings in the Bodleian. A greatly expanded version of that thesis appeared in 1958 and was followed by an exhaustive catalogue of the Persian paintings in the India Office Library. The John Rylands Library collection is the only other one in the provinces which is of major importance. With these three books Mr Robinson has therefore bequeathed to scholarship a meticulous account of the major British holdings of Timurid and Safavid painting, complete with lists of articles, scribes, illuminators, subject matter and manuscripts for comparison. Lavish numbers of illustrations fittingly complement his analyses of the Rylands miniatures. Miss Titley’s volume also crowns a lifetime’s labour in this field and it too needs to be evaluated in the context of similar earlier work, namely her book on Persian paintings in the British Library and the British Museum. Together these two books are a monument of selfless dedication. They render accessible some 13,000 Islamic paintings, perhaps the world’s greatest collection of such material. Each painting has a separate entry: bibliographical references are cited for each manuscript where appropriate: and exhaustive, carefully planned indices make it easy to track down the information required, be it a scribe, a given subject, a style or a manuscript. Even in this illustrious company, however, pride of place must go to the publishing sensation of the last few years so far as Islamic art is concerned — the catalogue of the Persian metalwork in the Victoria and Albert Museum by Dr. Assadullah Souren Melikian-Chirvani. Seldom has a work of this kind been so eagerly awaited: and it can safely be said that it puts all previous publications on Islamic metalwork in the shade. Here at last is a great catalogue for a great collection — an achievement which even the most distinguished of earlier specialists in Islamic metalwork, David Storm Rice, was unable to realise. To a daunting degree it is a solo performance — the detailed exploitation of obscure medieval literary sources, the decipherment of hundreds of inscriptions often bearing enigmatic poems

274

and hitherto unrecorded personal names, the close-up photography and of course the full command of the relevant specialist literature in half a dozen languages. A wide gamut of topics from technology to iconography is explored with flair and assurance. At last the reader can form a reliable idea of Persian Islamic metalwork as a whole, seen as a continuous coherent development spanning well over a millennium. This represents a crucially important progression beyond the stage of detailed but scattered analyses of individual objects, schools or problems. Other recent catalogues are necessarily on a smaller scale because they deal with exhibitions or with much smaller collections. Stuart Cary Welch and Anthony Welch elegantly present the cream of the Islamic paintings from the collection of Prince Sadruddin Khan exhibited in 1982-3 at the Asia Society, New York, and elsewhere. The full collection has been catalogued recently by Anthony Welch in four sumptuous volumes; indeed both authors have extensive experience in this kind of writing. The Asia Society catalogue, then, represents a fortunate conjunction of two of the most seasoned and productive scholars of Islamic painting, who have also opened up new areas of research. Stuart Cary Welch has developed new perspectives in Safavid painting by identifying the Turkman element in the court painting of the early sixteenth century and by his bold attempts to link unsigned paintings with specific named artists. Anthony Welch in his turn has shed much light on the nature of patronage in the high Safavid period, focusing among other topics on the roles and personalities of the various court painters and the degree of independence they were permitted. Both authors share a full-blooded zest for the paintings they describe and an ability to communicate that enthusiasm — and their own idiosyncratic insights — in prose that is by turns stylish, pungent and allusive. The reader need never fear a dull moment; moreover, the items are so selected as to provide a conspectus of Persian and Indian painting, though the range of Arab and Turkish material is too limited to serve a similar purpose. Finally, brief mention must be made of two catalogues from the prolific pen of Géza Fehérvári. One, entitled 1400 Years of Islamic Art and co-authored by Yasin Safadi, documents the resplendent Khalili collection, especially strong in early Qur’ans, medieval metalwork and the epigraphic ware of Samarqand and Nishapur in the tenth and eleventh centuries. The other records an unusually choice exhibition of seventeen pieces of thirteenth-century Iranian lustreware at the Mansour Gallery. In both catalogues the objects are meticulously recorded, and hitherto undeciphered inscriptions published; the Persian inscriptions on the

INFINITE RICHES

275

Mansour Gallery pieces were read by Manijeh Bayani. Good colour plates enhance the value of both catalogues as permanent works of reference. It can only be a matter of deep regret that such full and richly illustrated catalogues are so rarely produced for permanent public collections, whose holdings are sometimes not even recorded in a handlist. With one exception, the remaining titles covered in this review all belong in the category of monographs. The exception may, however, conveniently be discussed at this stage. It is a notable example of teamwork; under the editorship of Basil Gray some dozen contributors have written chapters on aspects of book production and illustration in Central Asia during the fourteenth and fifteenth centuries. It is an indication of the progress in the field of Islamic painting that multiple authorship should be required for such a volume. Although it is somewhat invidious to single out individual chapters, the sections on calligraphy, on illumination and on the school of Bukhara to 1550 all deserve special praise. All three cover subjects which, for all their intrinsic importance, have hitherto been accorded only transitory attention. The editor’s coup d’oeil over the critical fourteenth century is equally valuable as the most mature and comprehensive account of the subject published to date, based as it is on an unrivalled experience of these paintings extending over some fifty years. The achievement of the book in toto is radically to reappraise an apparently well-known body of material and in the process to present many unexpected insights. Multiple interpretations, which themselves change from one generation to the next, are the stock-in-trade of European art history, but in Islamic art history the conservative nature of the discipline has until recently inhibited their free development. This gradual process of emancipation is encouraged in many of the volumes under review — but preeminently in that on Central Asian painting — by a careful choice of plates which avoids the hackneyed. In the latter case most of the eighty-odd colour plates are published here for the first time, which serves not only to emphasise once again the protean variety of this branch of Islamic art but also to challenge received opinions on such subjects as the relationship between the Demotte and other Shahnamas later in the fourteenth century, the nature of Turkman painting, and the individuality of the Shiraz school around the first half of the fifteenth century. The remaining books to be reviewed here fall naturally into the category of monographs, except Mamluk Painting, which is a useful illustrated handlist, and Paradise as a Garden, an uneven but sparkling general account of Western Asian gardens from Achaemenid to late Islamic times. It is

276

beautifully presented and illustrated and is at pains to link the great gardens with entertaining anecdotal detail about their creators. The other volumes are all devoted to architecture. Certainly the most massive among them is the 3-volume treatment of the Aljafería, in Zaragoza. It conforms to a trend noticeable for a decade and more in German scholarship on Islamic architecture — the publication of massively detailed, definitive treatments of strictly limited areas or monuments, as in the works of Brisch, Ewert, Meinecke, Schmid and Kröger. This is cause for rejoicing. The days when a Creswell could spend most of a lifetime producing a final statement of the totality of early Islamic or Egyptian Islamic architecture are gone. The need for accuracy is as great as ever, but the sheer number of monuments in other periods and countries easily defeats such attempts at an extensive synthesis. In such a situation the solution adopted by the German scholars mentioned above is entirely sensible. It does however carry with it the dangers of missing the wood for the trees or of publishing a monument in more detail than its intrinsic importance warrants. Happily the volume on the Aljafería is to be welcomed on several counts. It establishes the increasing commitment of German art historians to Western Islamic architecture, a field which has traditionally been the preserve of French and Spanish scholars. It restores to its rightful place what is perhaps, despite its unjust neglect, the seminal building of Moorish architecture between the mosque of Cordoba and the Alhambra. Finally, its point of departure is not so much the building in toto but the place of the Aljafería in the development of that most typical Andalusian feature: the system of interlaced arches. The book is therefore primarily the intensive study of an architectural motif, and should be seen in the context of the author’s earlier studies of this motif at Cordoba, Toledo, Málaga and elsewhere. Such monographic treatment of a single architectural motif is virtually unknown in the field of Islamic art — there are, for example, no fulllength studies of the Islamic engaged column, capital, dome, squinch zone or muqarnas vault — and it is therefore heartening that this pioneering study is so successful. The volumes on Lashkari-i Bazar present the much-delayed final report of the French excavations carried out on this twelfth-century site in southwest Afghanistan from 1949-52. Mme Sourdel-Thomine and her colleagues deserve the warmest congratulations for their act of piety in rescuing the unfinished manuscript from potential oblivion in the Nachlass of Daniel Schlumberger and fleshing out its intermittently skeletal structure. Their labour has made possible a new evaluation of the art of the Ghurids, a

INFINITE RICHES

277

subject hitherto rendered obscure for lack of material — for such splendid monuments as the minaret of Jam and the madrasa of Shah-i Mashhad are merely isolated survivals. Lashkar-i Bazar, however, with its companion site of Bust nearby, fortunately comprises not a single monument but a royal city, with a very varied range of buildings — mosques, palaces, bazaar, tomb, citadel, pavilion, dovecote and the residences of wealthy notables. The sites have yielded an astonishing wealth of information on vaulting techniques, water installations, and systems of architectural ornament including carved stucco and terracotta, epigraphy and even figural wall paintings. As the only site of this period so far excavated in Iran or Afghanistan, Lashkar-i Bazar crucially enriches the understanding of eastern Iranian art at a key moment of its evolution. This book therefore allows the Ghurid art of Afghanistan to take its rightful place beside the related traditions of the late Saljuqs in Iran proper and the Khwarizmshahs in Central Asia. Coincidentally, the medieval architecture of south-west Afghanistan is also the subject of Manfred Klinkott’s book; but the architecture in question is very different from that of Lashkar-i Bazar. Its wide distribution throughout the ancient province of Sistan, and its almost exclusive use of mud brick instead of the much more expensive material of baked brick, is sufficient indication that this cannot be the product of royal patronage. The climatic peculiarities of the region have ensured the survival of scores of large buildings which re-assemble in unexpected ways such familiar concepts of Iranian architecture as domes, iwans, courtyards surrounded by arcades and façades articulated by niches. Here if anywhere in the Iranian world may be found that dimension of domestic function so signally absent from histories of Iranian architecture, which for lack of evidence must perforce focus on public buildings. For all the inherent interest of this material, however, it resists attempts to fit it into a wider context. The almost total absence of epigraphy and applied decoration in stucco or brick is remarkable enough in itself, but has serious repercussions: the usual dating controls are not available. Since there have been no full-scale excavations associated with this architecture, nor any systematic sharding for the area, this whole body of material cannot be closely dated. The Mongol invasions from 1220 onwards, which brought in their train the swift desiccation of the Helmand basin by the destruction of the irrigation systems of the area, do however afford a rough terminus. Even if the historical context of these buildings remains somewhat nebulous, Dr Klinkott must be warmly congratulated for the

278

major work of synthesis which he has performed; at last, after almost three decades of archaeological activity in Sistan, the important typologies of plan and structure have been clearly delineated. The remarks made earlier in this review concerning premature synthesis apply in full measure to the book on Saljuq architecture by Brandenburg and Brüsehoff. A comprehensive study of this field to set beside Wilber’s work on the architecture of the thirteenth and fourteenth centuries has long been needed; but this potpourri of information is not it. Familiar photographs of the standard monuments and well-worn arguments about stylistic development are set in an exiguously delineated historical context. The generalisations are apt to be excessively bold and straight factual errors abound; worst of all, the material is arranged in so eccentric a fashion that it is unreasonably hard to track down information. In short, the work bears every mark of having been written in a hurry. It does, however, contain some material not easy to find elsewhere: the discussion of Central Asian architecture and particularly of the city of Merv. These passages suggest that the authors might profitably embark on a study in depth of the early medieval architecture of this area, which remains far too little known because the relevant publications are virtually all in Russian. If the foregoing remarks have highlighted some of the difficulties inherent in the study of Islamic art, this review will have served its purpose. But the sheer variety of the books reviewed here should also have made it plain that Islamic art history is an unusually rewarding field of study precisely because it is still at a significantly earlier stage of its evolution than its Western counterpart. There is quite simply far more scope for original work here; in many areas of the subject the terrain has not even been mapped out properly. For any budding art historian with the instincts of the explorer, the challenge is loud and clear. Given, then, that Islamic art history has been late in coming of age, the books reviewed above attest the speedy pace of progress in the last few years. Gaps remain, and in plenty. There is still no full-scale monograph on any individual Muslim painter, not even Bihzad or Rida-i ‘Abbasi. A comprehensive survey of the activity of European antiquarians and travellers — an expanded version of Aris’ delightful book on Bhutan — is urgently needed. Finally, and most seriously, the study of Islamic iconography is barely beyond its infancy. Even so, if the information explosion of the last two decades continues, there is every reason to hope that such gaps will soon be filled.

INFINITE RICHES

279

Persian Paintings in the John Rylands Library by Basil W. Robinson, London: Sotheby Parke Bernet, 1980, 365 pp., 315 ills, 16 colour pls. Miniatures from Turkish Manuscripts. A Catalogue and Subject Index of Paintings in the British Library and British Museum by Norah M. Titley, London: The British Library, 1981, 144 pp. 54 ills. Islamic Metalwork from the Iranian World. 8-18th centuries by Assadullah Souren Melikian-Chirvani, London: HMSO, 1982, 443 pp., 425 ills, 1 colour pl. Arts of the Islamic Book. The Collection of Prince Sadruddin Aga Khan (New York, The Asia Society) by Anthony Welch and Stuart Cary Welch, Ithaca and London: Cornell University Press, for The Asia Society, 1982, 240 pp., 80 ills, 24 colour pls, £42 cloth, paper 1400 Years of Islamic Art. A Descriptive Catalogue (London, Khalili Gallery) by Géza Fehérvári and Yasin H. Safadi, London, 1981, 247 pp., 165 colour pls. Exhibition of Islamic Art. Iranian Lustreware of the Thirteenth Century (London, Mansour Gallery) by Géza Fehérvári, London, 1979, 61 pp., 17 colour pls. The Arts of the Book in Central Asia, general editor Basil Gray, London and Paris: Serindia Publications and UNESCO, 1979, 314 pp., 171 ills, 78 colour pls. Mamluk Painting by Duncan Haldane, Warminster: Aris & Phillips, 1978, 107 pp., 72 ills, 1 colour pl. Paradise as a Garden in Persia and Mughal India by Elizabeth B. Moynihan, London: Scolar Press, 1980, 168 pp., 110 ills. Spanisch-Islamiche Systeme sich kreuzender Bögen. III. Die Aljafería in Zaragoza (3 vols) by Christian Ewert, Berlin: Walter de Gruyter, 1978-80, 303 pp., 4 tables, 59 figures, 92 plates, 42 fold-out diagrams. Lashkari Bazar. Une résidence royale ghaznévide et ghoride (3 vols) 1A: l’architecture by Daniel Schlumberger; 1B: Le décor non figuratif et les inscriptions by Janine Sourdel-Thomine, Paris: de Boccard, 1978, 110 and 77 pp., 152 pls

280 Islamische Baukunst in Afghanisch-Sīstān. Mit einem geschichtlichem Überblick von Alexander dem Grossen bis zur Zeit der Safawiden-Dynastie by Manfred Klinkott, Berlin: Dietrich Reimer, 1982, 295 pp., 172 line drawings and 7 fold-out diagrams. Die Seldschuken. Baukunst des Islam in Persien und Turkmenien by Dietrich Brandenburg and Kurt Brüsehoff, Graz: Akademische Druck- und Verlagsanstalt, 1980, 93 pp., 34 figures, 72 ills, 11 colour pls. Views of Medieval Bhutan. The Diary and Drawings of Samuel Davis 1783 by Michael Aris, London: Serindia Publications and Washington, DC: Smithsonian Institute Press, 1982, 124 pp., 66 ills, 13 colour pls.

XIII What Happened to the Sasanian Hunt in Islamic Art?

T

he 400 years and more between the Arab conquest of Iran in 637 AD and the coming of the Seljuqs in 1055 is a quite remarkably dark age in Iranian art. In part, the grand political movements of this era were responsible. The centre of gravity in the Islamic world was never Iran in this period. It was Syria under the Umayyads, Iraq under the ‘Abbasids, and then Egypt and the lands to the west under the Fatimids. And the arts seem to tell the same story, notwithstanding the peerless epigraphic wares of Khorasan and the figural pottery of Nishapur. Part of the problem is that too little survives to permit a consistent pattern of production, style and iconography to emerge. Every new important piece that turns up is a surprise, and each one changes the 1 2 picture. Particularly in the case of precious metalwork and textiles, the problems of patchy survival are exacerbated by disputed provenance, date and — thanks to illegal excavations — authenticity. Yet it is highly unlikely that such a sparse and dubious picture of this key period of almost half a 3 millennium is correct. The papers published in this book have demonstrated, in many different 4 ways, the strength of Iranian national sentiment and sense of corporate identity in this seemingly dark age. The Shāhnāma underlines the rooted affection for Iran’s ancient heritage, not least that of the Sasanians. Nor  Atıl, Chase and Jett 1985, Appendix III.  Blair, Bloom and Wardwell 1992. 3  It is telling evidence of this scholarly neglect that the last attempt at an overview of Buyid art, and that a pretty short one, was published over fifty years ago in Kühnel 1956. 4  Stern 1971 represents an early attempt to tackle this problem. 1 2

282

were those feelings confined to eastern Iran. Indeed, it could be argued, at least on the basis of surviving monuments, that, especially in Parthian and Sasanian times, western Iran was culturally some way ahead of eastern 6 Iran. The impact of Buddhism also gave eastern Iran a cultural colouring 7 not replicated to the west. It was only from later Umayyad times, when Khorasan became the focus for disaffection with Syrian Arab rule, and indeed the seed-bed for the Abbasid revolution, that the ethnic, political and to some degree even religious opposition to the Islamic polity that had 8 developed far to the west took shape. Under the Samanids in particular, and specifically between the late ninth and the early twelfth centuries, a new Perso-Islamic culture took shape in the east; in the visual arts this 9 process can most easily be traced in architecture. Its other expressions are sufficiently well known, such as the growth of Persian national sentiment under the Tahirids, Saffarids and Samanids; the development of New Persian as its preferred literary vehicle; the galaxy of poets who flourished in the tenth and eleventh centuries in eastern Iran and whose preferred language was again New Persian; and the dawn of a golden age of scientific 10 achievement, best personified by the Khwarazmian al-Biruni. To the west, under the Buyids and lesser dynasties like the Bawandids, Kakuyids and Ziyarids, the immediately pre-Islamic past was also celebrated 11 12 13 14 — from titles to coins, from crowns to thrones, from literary revivals 5

5  The whole issue of eastern versus western Iran as the source for major cultural achievements was hotly debated in the earlier part of the last century. See Bartol’d 1922 conveniently summarized in Minorsky 1964: 61-63. 6  Herzfeld 1921: 119, 147, and 152, where he notes that the evidence of historical sources, surviving monuments, and references to vanished monuments all concur in suggesting that eastern Iran was culturally far behind western Iran in Parthian and Sasanian times. 7  Herzfeld 1921: 146; Melikian-Chirvani 1974; Melikian-Chirvani 1975. 8 Herzfeld 1921: 120. 9  Key monuments include a quartet of mausolea, every one of them a masterpiece — the tomb of the Samanids in Bukhara, the ‘Arab-Ata mausoleum in Tim, the Gunbad-i Qabus and the tomb tower of Radkan West, all probably built within a single century and each very different from the next. 10  Herzfeld 1921: 120-121, 173-174. See also Chelkowski 1975: 1-168. 11  Madelung 1969; Richter-Bernburg 1980. 12  Such as the titles Shāhanshāh and its Arabic equivalent Malik al-mulūk, or hybrid titles like al-Malik Shāhanshāh, al-Amīr Shāhanshāh or Shāhanshāh al-a’zam. See Treadwell 2001: xxx-xxxii and xxxiv. 13  Particularly Bahrami 1952: pl. 1/1a, 2a, 2b and 4a. On the medal depicted in pl. 1/1ab, see Atıl, Chase and Jett 1985: 267, where it is listed among works that “require further research to establish their provenance and date”. 14  See Spuler 1952: 346 for the case of the golden throne made in direct imitation of Sasanian work by Mardawij b. Ziyar.

283

THE SASANIAN HUNT IN ISLAMIC ART

like Vis u Ramin to the use of the Yazdigirdi solar calendar, as well as 17 in the long survival of Sasanian typologies in the coins of Tabaristan, in the use of Pahlavi inscriptions alongside Arabic ones in the Caspian 18 region, and in the celebrations of such traditional Iranian festivals as Noruz and Mihragan. Nor should one forget that the espousal of Shi‘ism could also be interpreted, at least in part, as an assertion of difference from the Sunni orthodoxy of the Abbasid caliphate. So while Islam was in the ascendant in many ways, the pre-Islamic tradition was still full of sap. And that certainly applies to the visual arts. The passage of time brought a dulling of the memory; old forms acquired new meanings, as in the paradigmatic case of Taq-i Bustan, which was 19 brought to romantic life by stories of Farhad and Shirin. Provincialism, too, could distort forms and themes that were originally metropolitan. Moreover, the decay or the reworking of Sasanian forms co-existed with the creation of brand new ones. All this is to say, then, that early Islamic Iran witnessed many different accommodations between the old and the new. Architecture is perhaps the only field in the visual arts where it is possible to piece together a reasonably connected story — even though there are 20 still many gaps. Take the way that a simple countryside fire temple or chahar taq, of which the one in Firuzabad is a standard example, could be turned into a mosque simply by blocking off the opening nearest the qibla so as to make a full wall out of it, as the case of the chahar taq at Yazd-i 21 Khast shows. It is equally instructive to see how that small-scale, slimmeddown formula of a dome set on a square chamber with four axial openings could be transformed, by little more than a process of enlargement, into one of the great monumental Seljuq domed chambers like that of the Gulpaygan mosque. The process could equally well work in reverse. Thus the archetypal, iconic Sasanian palatial structure — the arch of Khusrau’s palace at Ctesiphon, the Taq-i Kisra — could be reduced in scale, shorn of its lateral embellishments, with their uncertain use of Graeco-Roman classical vocabulary, and re-used in a new and now religious context as the 15

16

 Gurgani 1972 Van Berchem 1918: 104-106. 17  Walker 1941: 130-161 and pls. xxiii-xxvii. 18  Herzfeld 1932; Godard 1936: 112-114, 120; Herzfeld 1936: 78-81. 19  Soucek 1974. 20  The most up-to-date detailed survey of this material is to be found in Anisi 2007. For a brief conspectus of this material, see Hillenbrand 1985. 21  Smith 1940; Siroux 1947. 15

16 

284

great qibla aivan in so many of the Seljuq congregational mosques, as at 22 Isfahan. These examples — and they could easily be multiplied — show how Sasanian forms provided a launching pad for creating buildings of a totally different purpose and, eventually, of significantly different form. All this is by way of preamble. The Sasanian hunt is depicted at its grandest and most ceremonious on one flank of the great grotto at Taq-i 23 Bustan. It is set in a huge game preserve (which helps to explain the amount and variety of game in Sasanian hunting images and their Islamic descendants), madly stampeding animals, an orchestra of musicians, serried rows of court ladies to watch the fun, grooms and elephants to clean up the carnage, and of course the star of the show himself, the shah, shown twice for good measure and larger than anyone else so that you cannot miss him. This scene, carved and painted in the late sixth century AD, takes place on a huge canvas — an entire wall of a mighty man-made grotto. But this same theme was enthusiastically developed by the Sasanians on 24 a much smaller scale, especially in metalwork. In this medium court artists created an arresting, propagandistic and increasingly streamlined, indeed iconic, image of the serene, dauntless, ever-victorious monarch, dressed in full fig and taking on the most ferocious beasts single-handed and at close quarters. As frequent finds from sites far to the north of Iran, in the lands of the steppe nomads, show, these dishes travelled as silent ambassadors and reminders of the supreme power of the Sasanian kings well beyond the Iranian sphere. But then came the Arabs and the humiliatingly rapid collapse of this apparently invulnerable monarchy. And political collapse ushered in a certain diminution, even dissolution, of the triumphant visual vocabulary of the royal hunter. This seems to be the historical context for the textile which is at the heart of this paper. It is a fragmentary silk now in the church of Sant’ Ambrogio in Milan which is part of a sequence of four very closely connected but not identical versions of the same theme. In its complete form, as preserved in Cologne, it has a repeat pattern of two affronted 25 horsemen shooting their prey. None of the four versions is dated, though Grabar 1990: 91.  For the fullest account of all the reliefs see Fukai 1972 and 1984. 24  Erdmann 1936; Grabar 1967: 47-55; and Harper 1981: 40-98 and pls. 8-32, 37-38. Also see von Gall 1990 for the latest detailed survey of the theme. 25  Von Falke 1913: I, 70-74 and Abb. 69; Pierce and Tyler 1936 (the Prague silk); Volbach 1969: 101, pl. 46 for a colour illustration of the Milan fragment. 22  23

THE SASANIAN HUNT IN ISLAMIC ART

285

the presence of the Milan piece on an altar in Sant’ Ambrogio gives a terminus post quem of 835 AD. The textile was used to cover the inner faces of the gold altar made by a certain Volvinus in that church. Such a privileged setting clearly indicates that this textile was highly prized despite its fragmentary state. Similarly, the almost identical textile in Prague forms part of the binding of a ninth-century Gospel book. Above all, though, in a field which is notoriously devoid of all but a very few fixed points in the matter of absolute dating — a field, it is worth repeating, in which dating and provenance are alike insecure, and where scores of pieces (especially transitional proto-Sasanian or early Islamic ones) are removed from the scholarly discussion because their authenticity is suspect — this proof that the Milan textile dates before 835 CE is especially precious. Its provenance, however, remains unknown. The costume details have suggested Syrian workmanship to some, but with the reservation that the workshop itself 26 was Iranian. This is certainly a working hypothesis, but it could be regarded as fudging the issue. Of course it is perfectly possible that a craftsman from one country — more, from one culture — should work in another. But such a theory sidesteps the real problem, which is how to explain the co-existence in one object of different figural, iconographic and even religious traditions. Better, surely, to interpret such pieces — especially since they are numerous — as documents of a transitional period when artists were modifying inherited traditions by injecting new ideas, motifs, meanings and ways of seeing. In the case of the Milan textile, there can be no doubt that the scene illustrates a debased Sasanian iconography, and that is what constitutes its interest so far as this paper is concerned. The Milan fragment shows two right halves (not matching halves) of the silk, and in its present state this half-image corresponds with tolerable exactness to a familiar type of Sasanian hunting image, usually encountered on silverware, in which the princely archer slews round in the saddle to 27 loose a shaft at a rearing lion which menaces him. The Parthian shot, as 28 this pose is traditionally called, thus has a powerful inbuilt tension, with the horse galloping in one direction, away from danger, while the rider, twisted in the saddle, courageously confronts that danger. The pose is the 26 Ghirshman 1962: 235. The same reasoning would presumably apply to the other versions at Cologne, Prague and St Calais (Sarthe). Ettinghausen 1972: 38 believes that the Milan piece is probably Umayyad. 27  Harper 1981: pls. 14 and 37; Hayashi 1975: pl. 84. 28  Rostovtzeff 1943.

286

essence of drama. It also reflects the actual practice of the hunt — and of war. As a piece of design, too, such a combination of Parthian shot and menacing beast integrates the two opposing directions in an admirably self-contained unit ideally suited to the roundel form. The fact that versions of this particular design, with widely differing colour schemes, survive in Prague, Cologne, Milan and St Calais (between Tours and Chartres) make it one of the most widespread compositions of the early Middle Ages, comparable in popularity, say, to the charioteer or 29 quadriga silks. It is now time to take a more general look at what is happening in this silk textile. This is a textile, we must remember, that survives only in a severely cut-down form. It is therefore important to try to reconstruct its original impact, and the dark blue version in Cologne, being more fully preserved, provides a better basis for this to be done, helped by the reconstruction 30 published by Ierusalimskaja and Borkopp. If one imagines a textile of substantially larger size where this same design is repeated many times, it will become plain that its multiple roundels are disposed in serried horizontal, vertical and diagonal rows, depending on how one looks at them, and they themselves assert continually confusing and contradictory directions. Nor is this all. It is not merely a matter of overall impact, but of the effect at closer range, that is to say when one focuses on a single roundel. This particular textile — whether it was originally intended to serve as a curtain, a wall hanging or a garment — was so overloaded with detail that the main theme had to fight hard, and rather unavailingly, in its attempt to establish visual dominance. The mounted archer is merely one detail among many. There is simply too much to be taken in at first or even second glance. One detail after another, important or trivial, clamours for attention. It is indeed a classic case of the celebrated Islamic horror vacui. And on the micro as on the macro scale it is the sudden changes of direction that bewilder the viewer. This would be most true if the textile were part of a garment. When the wearer of such a robe moved, this decoration would quickly become a mere blur. Yet this textile in Milan, and especially its Sasanian antecedents, cannot be fully understood without a closer look at how the theme of the hunt was interpreted in this very same medium of textiles, first to the west of  Volbach 1969: 88. It is worth noting Volbach’s view that these four versions of the mounted hunter, like the quadriga fabric in Aachen, are “Byzantine work…more stylized and predominantly influenced by Persia”. 30  Ierusalimskaja and Borkopp 1996: 13. 29

THE SASANIAN HUNT IN ISLAMIC ART

1. Taq-i Bustan: hunting scene (after Ghirshman 1962: pl.236).

2. Silk from Sant’ Ambrogio, Milan (after Vollbach 1969: pl. 46).

287

288

3. Reconstruction of the Milan silk (after Ierusalimskaja and Borkopp 1996: pl. on page 13).

4. Coptic decorative roundel with two horsemen, 6th century AD (after Vollbach 1969: pl. 34).

THE SASANIAN HUNT IN ISLAMIC ART

289

5. Medallion with an emperor on a lion-hunt, perhaps 8th century AD, Mozac (after Vollbach 1969: pl. 55).

6. Silk medallion with two horsemen, Maastricht (after Vollbach 1969: pl. 47).

290

7. So-called Yazdigird silk, Berlin (after von Falke 1913: pl. 105).

8. Bahram Gur silk (after Bréhier 1936: pl. LXXXVI).

THE SASANIAN HUNT IN ISLAMIC ART

9. Japanese silk of Sasanian inspiration (after Ghirshman 1962: pl. 445).

10. Japanese silk of Sasanian inspiration (after Ghirshman 1962: pl. 444).

291

292

11. Japanese silk of Sasanian inspiration (after Hayashi 1975: pl. 82).

THE SASANIAN HUNT IN ISLAMIC ART

293

Iran, then in the Iranian sphere itself, and finally to the east. This should establish the basic visual language on which the Milan textile depends, no matter how far it departs from these inherited norms. First, then, the west, which in this context does not mean western Europe but the lands of Byzantium, Syria and Egypt. Roundels from sixth-century Coptic Egypt clearly betray a vulgarisation of more complex models with their indifference to scale, their caricatured doll-like figures, 31 frozen gestures and jerky rhythms. In such textiles, which are not of silk, but of wool on linen, and therefore aimed at a less wealthy clientele, the notion of an actual hunt has almost disappeared — indeed, some of the animals are literally only half there. The plethora of detail, some of it only marginally relevant, also takes away the urgency of the event, and it is hard to take seriously the toy weapons that the riders brandish. Sometimes the hunting scene is still further diminished by being demoted to a surround for another scene 32 altogether, such as Samson grappling with a lion. Much more serious competition comes from a solemn hieratic Byzantine 33 image of two emperors on a lion hunt. Movement has slowed to a standstill and while hunting dogs frisk about nuzzling their hind legs the lions are, it seems, obligingly trotting up to swallow the spears pointed with casual negligence in their direction by the gorgeously apparelled emperor. These are again a pair, and they stare out at us with a pointed disengagement from the business at hand. But for all the lifelessness of the scene — it makes a solemn ritual of the hunt — the artist has taken extreme care to dispose his figures to maximum advantage, spotlighting them against a plain dark ground and sparing no effort in decorating emperors, horses and lions like Christmas trees. Here we see the quintessential Byzantine severity and formal ceremony in its most heightened form. This textile has the added advantage of at least a clear terminus post quem dating of 761 when it was donated to St Calmin in Mozac by Pippin the Short. Other Byzantine silks continue this theme of matched pairs of hunters, whether mounted or on foot, affronted or addorsed. All show a careful attention to symmetry, to reasonably accurate proportional relationships, and to a due

 Volbach 1969: 57, and, for a colour illustration, pl. 34.  Ibid. pl. 35 for a colour illustration. 33  Bréhier 1936: 101 and pl. xc. He expresses doubts about the connection with Pippin the Short but gives no reason for doing so. Volbach 1969: 113-114, 118 regards this tradition as genuine. For a colour illustration, see Volbach 1969: pl. 55. Cf. von Falke 1913: II, 4-5. 31 32

294

concentration on the key theme of hunting. In most cases the theme is set in a roundel whose border comprises a series of vegetal or heart-shaped motifs. The link to late antique, early Christian and Byzantine wall paintings and floor mosaics is patent. Sometimes the design features a pair of hunters on foot in the upper level, each spearing a ferocious beast, and two addorsed hunters similarly occupied below. This simple counter-change is enough to ward off monotony, and indeed to add visual interest. In all these cases, despite the expensive material of the textile itself, the dress of the hunters — usually a short tunic — and the 35 frequent lack of headdress leaves no doubt that they are not of princely rank. So much for the Christian Near East. What of the Iranian sphere itself ? Here a couple of silks destroyed in World War II were widely accepted by those who had studied them as Iranian work, though their date remained uncertain. There is little doubt that if they turned up on the antiquities market today they would be incontinently dismissed as gross forgeries. This is a reminder of how little we still know of the norms and forms of early Islamic textiles in the Iranian world. In the case of these two textiles their ancient connections with church treasuries gave them an unimpeachable pedigree, so it is strange that they have been relatively neglected in recent scholarly literature, all the more since they are of major importance iconographically. As types they are far removed from any Byzantine silks, and their subject matter (as well as numerous details of their execution) are unmistakably Iranian. But their gross simplification of Sasanian crowns places them both firmly in the post-Sasanian era. That said, they draw on rather different sources for their iconography. An obvious marker of a post-Sasanian date is the pillbox hat with an aerial worn by the hook-nosed and savagely determined king on the so36 called Yazdigird silk — that identification is merely a name-tag. This silk is associated with a papal gift of the eighth century to the monastery of Gerresheim in Germany, yet another reminder of how far from home such silks could travel, as we have already seen in the Mozac and Milan silks. In the so-called Yazdigird silk, the emphasis lies on elements derived, through an incalculable number of intermediate stages, from Achaemenid and even more ancient Near Eastern iconography. Hybrid monsters play a dominant role here, and it is the business of the king, like Darius and Xerxes before him, to subdue them. So, cheered on by the homunculus 34

 E.g. the Maastricht silk, Balty et al. 1993: 115 to be supplemented by Volbach 1969: pl. 47. 35  E.g. Volbach 1969: pls. 50 and 59. 36  Von Falke 1913: 83-85 and pl. 105. 34

THE SASANIAN HUNT IN ISLAMIC ART

295

in a nearby tree, he bridles the eagle-beaked griffin on which he rides and — turning in the saddle — holds in an iron grip the rearing horned and clawed monster that threatens him. In other words, he masters not only the natural but also the supernatural world. Again, the parallels with Achaemenid 37 sculpture are instructive. Such an image could scarcely be further removed from an Islamic world-view. These references to another world inhabited by mythical and symbolic creatures proliferate in Sasanian and early Islamic textiles alike. Lions couchant and regardant, and moufflon passant — these heraldic terms seem to fit their solemn, measured visual tone — fill the background in layered strips, a compositional device that recalls the friezes 38 of pacing (and usually ferocious) animals so common in ancient Iranian art and is so far removed from the standard medallion image of early medieval Iranian and Byzantine textiles. Some borrowing from another medium might explain this. Even the distinctive modelling of the animals’ bodies by patches of colour seems to be a textile-based response to the strongly plastic internal modelling of animals in, say, Achaemenid sculpture. All in all, then, this silk wields with remarkable aplomb the visual language of ancient Iran even though it was probably designed at least a millennium later. The hunt has taken on mythical and religious overtones. As Ettinghausen has shown, 39 this awareness of the distant past has parallels in early Islamic art in Iran, and of course the Sasanians themselves had shown the way here in their numerous references to Achaemenid art and iconography and their pointed re-use of Achaemenid sites. The Yazdigird silk owes very little to the classical world besides the minor solecism of the clavi or strips of embroidered cloth on the ruler’s chest. The other clearly post-Sasanian silk, however, positively glories in such classical and Byzantine allusions, even though they are often woefully inaccurate, and this has led some to regard it as the work of artists in 40 Constantinople. They range from the use of an inhabited vine scroll for the border of this large medallion (87 cm in diameter) to the rigid frontality of the riders’ poses, from the turriform element in their crowns, which recalls crowns worn by female classical personifications of cities, to their bare feet, and from their gallant but seriously flawed attempt to replicate imperial Byzantine costume to their long flowing hair, which is in marked 37  Curtis and Tallis 2005: pl. 42; 82, cat. nos. 41-42; 92-94, cat. nos. 66-74; 158-159, cat. no. 202; and 160, cat. no. 208. Cf. Ghirshman 1964: pls. 250-253. 38  As at Ziwiye, for example Diba 1965: figs 69-73b and 75-76. 39  Ettinghausen 1969. 40  Bréhier 1936: 99 and pl. lxxxvi; cf. von Falke 1913: I, 85-86.

296

contrast to the bunched hairstyle of Sasanian monarchs. Yet for all these borrowings from an alien tradition, this silk retains an essentially Sasanian flavour. Numerous details substantiate this, for example the affronted cockerels at 12 and 6 o’clock, the moufflon in the spandrels (both of them favoured creatures in Sasanian textiles), the pearled collars worn by the winged horses, and the diminutive lions bringing down horned ungulates immediately beneath the hindquarters of the winged horses. These are echoed by the larger lions each pouncing on their stag below. Each king grasps a branch of the central tree on which perches a bird of prey, perhaps an allusion to falconry. This again would point to a post-Sasanian date — the earliest visual references to the sport occur on Umayyad bronze 41 coins — as would the plethora of animals (no less than 14) that take up the lower half of the composition. Nevertheless, the designer has left just enough empty space, and has maintained a reasonably consistent scale, for the image to have room to breathe and to make sense. The principal Sasanian feature is of course the subject matter itself. This depicts Bahram Gur snatching a lion cub from its enraged mother — five teats are carefully delineated — which leaps at his horse’s hooves. It is beyond question that the doubling of the riders is in this case fatally inappropriate and destroys the credibility of the narrative. All this leads naturally to the specific connection between the Milan textile and Bahram Gur, this most celebrated of Sasanian hunters, whether as prince or shah. As it happens, the key surviving literary texts which 42 discuss the hunting feats of Bahram Gur all postdate this silk, though it is entirely possible that the lost pre-Islamic Khuday Namak, a narrative of the Sasanian kings, dealt with this theme. The earliest of the surviving accounts is that of Ibn Qutaiba, writing in the tenth century, which deals in detail with the story of Bahram Gur and his slave girl Azada. She challenges him to perform three near-impossible feats: making a buck a doe, making a doe a buck, and transfixing a deer’s foot and ear with a single arrow. He duly performs these tasks, whereupon she berates him for having the soul of a demon, not a man. He then pushes her off the camel they are riding, and tramples her to death. That is the last time he goes hunting with a woman. It is certainly the last time that she goes hunting with him. This became the defining episode illustrating Bahram’s hunting skills, and as such it had a millennial lineage in Persian metalwork,  Oddy 1991; cf. Allen 1980: 127-128.  See Ettinghausen 1979 for what follows.

41 42

297

THE SASANIAN HUNT IN ISLAMIC ART

pottery, miniature painting, carpets and other media. The story is retold in Firdausi’s Shāhnāma, itself explicitly based on stories drawn from prelslamic sources, and in the Haft Paikar, or Seven Princesses, of Nizami, written around the year 1200. Yet these are not the only hunting feats of Bahram Gur. He was also a famed dragon-slayer; he snatched the royal crown from a throne to which two ferocious lions were tethered and which he first had to kill; we have seen how he taunted a furious lioness by stealing her cub; and he killed a lion and a gazelle with a single arrow. It is this latter feat that seems to be illustrated in the Milan textile. The embarrassment is that Bahram Gur seems to have had nothing to do with it; his attention has been deflected and is firmly engaged elsewhere. Iconographically, of course, this is a major solecism; and such solecisms 43

are frequent in these post-Sasanian treatments of Sasanian themes.

The theme of Bahram Gur and Azada turns up several times in the Sasanian period itself, in intaglios, in stucco and — most memorably — 44 on a large silver-gilt dish. This is sufficient evidence that a pre-Islamic iconographic tradition focusing on at least some of Bahram Gur’s feats 45 was well established, and this in turn makes it likely enough that there were further, now lost, works of Sasanian art which depicted these other feats and could thus have served as a model for our textile. Moreover, a 46 post-Sasanian textile in Augsburg also depicts Bahram Gur and Azada. But, as we shall shortly see, the Milan textile contains ample proof that the image of the mounted hunter that it bears has already travelled a considerable distance from any putative Sasanian source. It is now time to return to the Milan silk which is the core of this paper. It is a remarkable fact that this silk, whenever and wherever it was made, contrives in comprehensive fashion to squander the particular advantages of the Sasanian model on which it is based. It does so in various ways: duplication, omission, displacement and over-crowding. These are, as will soon appear, characteristic of non-Sasanian interpretations of Sasanian themes. It will be worth looking at each of them briefly in turn. First, duplication. The presence of two horsemen instead of one and the fact that every detail of the design is duplicated in a symmetrical mirror image completely alters the impact of the composition. A single  Hillenbrand 2006.  Harper 1978: 48-50. 45  See also the ivory casket formerly in the Stoclet Collection in Brussels; Shalem 2004: 125, fig. 9.10. 46  Shalem 2004. 43 44

298

rider occupying the main portion of the design naturally exudes the dominance associated with royalty. Such a design is effective in symbolic terms and there are many examples on Sasanian textiles — though not of a horseman. That entire dimension is lost when the image is doubled. A speaking image is demoted to a decorative motif. Doubling was, of course, 47 merely one option. A version of this theme in the Shoso-in at Nara, datable to the rule of Shomu in the mid-eighth century at the latest and executed in a Japanese idiom (note the Chinese pictograph), is so designed that the hunter faces left and right in successive roundels. Moreover, the roundels are staggered in rows so that the same alternation operates on the vertical as well as the horizontal axis. This mirror effect is common in Sasanian textiles. One might add that, when the textile is seen as a whole, the effect of these successive subtle shifts in focus is to create a kind of continuous mobility and low-key variation which is very pleasing to the 48 eye. In another Japanese version of this Sasanian theme the medallion contains not one but four such horsemen, the top pair affronted though each is turned away from the other to shoot his prey, while the lower pair echoes this action but (absurdly enough) upside down. This theme of extra enrichment extends to the border itself, where the standard pearl roundel so typical of Sasanian textiles is itself surrounded by a further floral scrolling border. In the other Japanese textile previously mentioned the border is the simple pearl roundel alone, but there too an extra enrichment has been introduced, in that the field of the roundel border alternates regularly from green to blue. In a third Japanese version 49 of the Sasanian hunt, datable c. 750 AD, where the Chinese characters on the horses’ flanks read “mountain” and ‘happiness’, four horsemen in superposed pairs, plus four rearing lions and a tree defining the central axis, are crammed into a single medallion with a mainly pearl roundel border. Thus the options open to craftsmen wishing to develop the theme of the Sasanian horseman were indeed wide. The best evidence of omission, our second category, lies in the area behind the horses. The Parthian shot makes sense if the game is ferocious. Yet here the space behind the horses is taken up by nothing more formidable than a tree flanked by leaping hares, and each hunter’s arrow points straight at one of these hares, at such close range moreover that he would be hard put to miss. No feats of skill here — and yet the whole  See Ghirshman 1962: pl. 445 for a colour illustration.  Hayashi 1975: pl. 13. For a very similar piece from the Horiyushi temple see von Falke 1913: I, fig. 111. 49  Hayashi 1975: pl. 82. 47 48

THE SASANIAN HUNT IN ISLAMIC ART

299

purpose of the original iconography was to vaunt Bahram Gur’s uncanny skill as an archer. The pose has therefore lost most of its purpose; no Sasanian shāhanshāh would deign to be depicted hunting game as small 50 as this. As for the costume of the rider and the treatment of the horse, the paraphernalia of royalty have been drastically reduced. The knotted tail, the acorn-shaped tassels, the ribbons streaming from the belt, the magnificent crown - all these have been filtered out, and with them much of the heraldic and symbolic power of the image drains away. Misplacement, our third category, is seen in the tableau beneath each of the riders. It depicts an onager being brought down by a lion in a pose which was already ancient in the Near East in Sasanian times, and indeed 51 accurately reproduces the lion attacking the bull at Persepolis. Yet this ancient astrological image, emblematic also of royal victory, is here put to an unexpected use as part of a narrative. An arrow lodged in each lion’s shoulder identifies the scene as one of the hunting feats of the Sasanian king Bahram Bahram V (421-439) later shāhanshāh) later celebrated as Bahram Gur. He spied a lion savaging an onager and promptly shot an 52 arrow which transfixed the pair of them and drove deep into the ground — a detail overlooked here. Moreover, the rider is about to loosen a shaft in the opposite direction, so that there is no obvious connection between him and the wounded lion. This is not the only example of misplacement. Other animals fan out in front of the horse in approved Sasanian fashion; but, since the rider is looking in precisely the opposite direction, they too are so misplaced as to lose much of their meaning in this scene. One might add that the combination of several types of quarry is unusual 53 though not unknown in Sasanian hunting images. It is the norm for the ruler to concentrate on one type of quarry at a time. Yet another slight solecism is the visual connection between the animals below the horse and the defeated enemy trampled by the shah in many a Sasanian investiture 54 or triumph. Finally, the impact of the traditional Sasanian pearl roundel 55 56 motif, commonly used as the sole border design in Sasanian textiles, is  Harper 1981: passim; Harper 1978: passim.  Ghirshman 1964: pl. 240. 52  Tabari 1999: 85. This incident does not figure in the discussion of the iconography of Bahram Gur as given in Ettinghausen 1979. 53  Ghirshman 1962: fig. 247 — a plate which shows two gazelles, a moufflon and two bears. 54  Herrmann 1977: 90-93, 96 (plates illustrating six examples in all). 55  Meister 1970. 56  E.g. the boar’s head textile from Astana in Ghirshman 1962: fig. 281. It was standard practice for such borders to include a different motif, such as a rosette or a square, at the 50 51

300

much reduced by the way that the pearl roundel has been swallowed up by a much more complex border design, and indeed has taken on some of the character of a bead-and-reel motif. Thus conflation is the essence of this design. The last change to be considered is the addition of superfluous motifs. These are so plentiful that the design becomes overcrowded. The central tree is a case in point. Its principal function is to divide the picture space down the central axis. One may well doubt whether it has any deeper meaning than this. Around the riders a positive menagerie of creatures 57 disport themselves — eagles, other birds, stags, hares and dogs. Such little empty space as remains is filled with severed flowering branches, plants or even leaves. The Sasanian artists who created the great hunting scenes on the silver plates took good care that the background was left plain and thus did not allow it to compete with the principal image. To summarise, then, the scene was in theory entirely suitable for a roundel composition. But the arrangement of the design here ensures that the ostensibly main subject has been relegated to a side-show, while the fact that the scene is shown twice means that it cannot really be interpreted as a historical record of Bahram’s feat. The artist, as noted above, has signally failed to place the rider, lion and onager in rational rapport with each other. In this respect too, then, a potentially powerful theme has been drained of meaning. A specific theme has become a generic one. The influx of otiose detail has the same effect. So too has the absence of the identifying attributes of royalty — for example, the majestic Sasanian crown has given way to a winged cap with fluttering ribbons. Most striking of all, a dominant single figure has been replaced by a crowded, busy scene. All that is on the debit side; on the credit side, however, is a degree of rhythm, symmetry and decorative pattern that the lost original probably never possessed. What, in conclusion, does the Milan textile reveal about how a Sasanian aesthetic turned into an Islamic one? To put it briefly, the sense of awe has evaporated. Craftsmen had lost an easy familiarity with the tout ensemble of Sasanian royal iconography; they saw it as in a glass darkly. This was only to be expected once the major court or metropolitan manufactories, the four cardinal points; also, Balty et al. 1993: 136, fig. 131; 274; and 276. 57  The presence of dogs strengthens the case for a post-Sasanian dating. The frescoes of Qusair ‘Amra in the Jordanian desert show a pack of hunting dogs. See Almagro et al. 1975: Láms. xxviiib and xxixa, b. By way of contrast, Sasanian hunting plates on the whole do not depict hunting dogs, though a solitary specimen occurs on the Ufa plate; see Fajans 1957: 65 and fig. 16.

THE SASANIAN HUNT IN ISLAMIC ART

301

natural repositories of centuries of experience and hence the guardians of what was fitting, had ceased to function. There was a random quality to 58 what was remembered, and to how accurate the memories were. Artists took what they could from the past, and tried to integrate it with their own vision. So there is a good deal of stuttering. But these were also images of royal majesty and power which proved surprisingly resilient, and which could triumph over numerous mistakes of detail and of balance. Something else is at work here. As was only to be expected, these postSasanian pictorial textiles, being transitional, often have a foot in two camps. They look back to a world of isolated large-scale images which owe much of their impact to their absolute size and simplicity. But they also look forward to a world in which small-scale pattern played a much greater role, and where ornament was perceived as a busy overall covering of a surface. That is the world of late Umayyad carved stone and plaster, and of Samarra stucco. Figural art, in textiles as in other media, had to make some compromises to fit into that new aesthetic. The Milan textile, which is perhaps of Umayyad date, marks a moment when pattern was beginning to swallow up meaning but was itself still not yet under full control. Nor do inscriptions yet play a part in its design. Gradually over the next couple of centuries pattern, and with it the role of inscriptions, asserted itself with increasing force, and figural elements had to share the available space with these competitors. The St. Josse silk is a good example of this next stage of evolution. But it was not until the Seljuq period that a comfortable balance between abstract, vegetal or geometric ornament, epigraphy and figural elements was attained. The time is ripe for a re-assessment of early Iranian Islamic textiles of pictorial character. The doubts cast on the authenticity of so much of this material by recent scholarship should not be allowed to obscure the fact that in European church treasuries there still exists a mass of pictorial post-Sasanian textiles whose authenticity is unimpeachable, with pedigrees stretching back sometimes a thousand years and more. Controlled excavations are gradually adding to this total. The textual references so meticulously assembled by Serjeant two generations ago provide ample literary evidence to indicate just how important and pervasive costly textiles were as markers of rank and wealth. These various groups of data now need to be re-assessed together with a view to developing an up-to-date synthesis that will restore these early Islamic Iranian textiles to their rightful position in the history of Iranian art. 58  An extreme case is a silk depicting a cavalier, which was found in the region of the Urals and is now in the museum at Tcheliabinsk. See Balty et al. 1993: 117, fig. 106.

XIV The Major Minor Arts of Islam

A

s this planet shrinks, Eurocentricity of any kind becomes harder and harder to justify. We drive Japanese cars, we consume Arab oil; we happily adopt Indian and Chinese cuisine; famines in Africa fill our TV screens. Ethnic minorities are at last finding their voice in many Western societies. The attractions of Eastern religions are evoking an increasingly positive response among young people in the West. On the political front, too, not a day passes without the West feeling the repercussions of events in the non-Western world. It is time that our educational system, even including art history, took account of these sea changes in our society. It is no doubt reprehensible that Western culture should have been so content with its own company for so long. Yet there are understandable reasons for this state of affairs. For some, it may take a determined mental effort to break free of the comfortable, familiar cocoon of Western culture and begin to explore the world outside. It is hardly fair to blame the vast majority who, for one reason or another, never make that effort, let alone those that do, only quickly to retire from such a sortie, discouraged by the apparent absence of reference points, the essential unfamiliarity of the landscape, or a daunting sense of its vast extent. For those that persist, much of course has to be learned — new languages, new religions, huge new tracts of history, and new social systems whose rhythms and norms are very different from those of the Western world. There is no readymade anchor for Western students embarking on such studies; people have to fashion their own. More than all this, however, much must also be unlearned — and it is this aspect, being so much less obvious than the need to amass information

THE MAJOR MINOR ARTS OF ISLAM

303

on the new field of study, that has tended to be overlooked and will, therefore, be investigated here. The scope of this article dictates that the inquiry will be pursued within the confines of Islamic art; but of course some of the same factors, as well as several new ones, would apply in the context of other traditions of non-Western art. In the specific case of Islamic art, then, what has to be unlearned by someone trained in Western art history who wants to investigate this new field? Perhaps the single key step is to relinquish the generally accepted hierarchy of media in Western art. To put it bluntly, Islamic art has no pictures in the sense that people in the West understand that term. For all practical purposes, Islam does not employ the easel or panel painting, ignores print-making and related arts, and never seriously developed fresco painting. It is worth taking stock of such a statement. What are its implications? At the most obvious level, it means that the art form which has furnished the very foundation of Western art history throughout the past six centuries is simply not there in the Islamic world. There can be no gallery of Islamic art along the lines of the great picture galleries of the Western world. This is not to say that there was no painting at all in Islamic art. In an Islamic context, however, painting connoted almost exclusively book painting. One has only to look at Western illustrated manuscripts to be aware of the inherent limitations of that medium. The need to conform, however broadly, to a pre-ordained text; the relatively small size; the essentially private nature of the image; the implication that the person looking at the book is literate, wealthy and cultivated — all this is selfevident, and is as true in the East as in the West. Art historians for whom the word ‘painting’ connotes a work on wood, on canvas or on a wall, will, therefore, have to develop a different (and not, incidentally, a lower) set of expectations when they encounter Islamic painting. Significant social and economic implications emerge from this consistent emphasis on book painting, to the virtual exclusion of other types of painting, in Islamic culture. There is no place for a popular element in this art. The artist, being unable to sell his work on an open market, is constrained to follow the taste of his patron. His scope for originality, especially in the choice of subject-matter, is correspondingly reduced. Only slightly less fundamental than this is the virtual absence of largescale sculpture in the Islamic world. That the sculptural impulse was there, and in abundance, is sufficiently plain from a cursory glance at Islamic decoration in stone or stucco. But custom, perhaps inspired originally by

304

religious feelings related to the Second Commandment, denied the wouldbe sculptor in Muslim society the opportunity to realise his talent in a fully three-dimensional way. With a restriction in scope came even more serious restrictions in meaning and function. Architecture has always occupied a similarly honoured position in both Islam and the West, but the absence in the Islamic world of the other two basic forms of artistic expression in the West, namely large-scale painting and sculpture, means that there was a yawning gap to be filled in Islamic art. It was the ‘minor’ arts that filled it, and thereby achieved major status. The next body of information to be unlearned is that which embraces Christian iconography. At the risk of repeating the blindingly obvious, it has to be remembered that while much of Islamic art is not religious, it was, nevertheless, produced in a society which followed the Muslim faith and therefore had no room for Christian images. Moreover, Islam had no significant religious iconography of its own with which to replace such images, and had no equivalent of the Christian clergy to provide the necessary patronage. Since most Islamic art dates from the medieval period, and thus from the very centuries which in Western art were dominated by Christian iconography, this is indeed a momentous difference between the two cultures. It is not only the detailed knowledge of Christian iconography that has to be unlearned, but also much of the support system within which medieval Christian art functioned. That system had no medieval Islamic equivalent. Classical mythology also struck no answering chord in the Muslim mind and is, therefore, an encumbrance which the art historian would do well to shed when coming to grips with Islamic art. Thus yet another foundation of Western art history is demolished in the move from one culture to the other. Even the language of classical architecture played no significant role in Islamic buildings beyond the first century or so after the death of Muhammad. There is no Islamic analogue for the gradual Western rediscovery of the antique, let alone for the academic accuracy with which classical themes and motifs were used. No Islamic architect looks nervously over his shoulder to check whether Vitruvius is watching him. It would be possible to continue at some length in this vein, identifying still more aspects of Western art which are either unknown or raise only the most distant echo in the world of Islam. They include the gradual development of the notion that the artist is a man of genius, a personality worth exploring for his own sake as well as for the insights that his

THE MAJOR MINOR ARTS OF ISLAM

305

biography might yield to help the understanding of his work. Islamic art knows no such cult of personality, even though, paradoxically, lists of Islamic artists and craftsmen known from inscriptions and signatures run into thousands. Equally foreign to Islam is the notion that man is the measure of all things, with the consequent emphasis on the human body as a means of expression and as an intrinsically worthy subject for the artist. For the most part, Islamic artists were not attracted by the ideal of naturalism, and, perhaps as a result, they never developed such devices as modelling or perspective until triggered by contact with European art. They were capable enough of observing the natural world, but preferred to transfigure or transmute its elements into something distinctively other, something perhaps more akin to a Platonic ideal. It never rains in Islamic painting. Such a detail may seem trivial, but it explains a good deal. In Islam, the artist’s eye functions not as a camera lens, faithfully reproducing the clutter of everyday life, but as the obedient tool of an organising, abstracting mind. This is not the place to follow up the implications of such remarks, but some awareness of these basic differences in focus between Western and Islamic art is crucial for a proper evaluation of where the ‘minor’ arts belong in the canon of Islamic art. Accordingly, the comments made so far are addressed principally to Western art historians, and are intended to underline the fact that the hierarchy of art forms in East and West is very different. An artistic medium — like pottery, for example — may deservedly be relegated to minor status in the West, but its status in the Islamic East (and in the Far East too, for that matter) may be very different. It is the cavalier transference of assumptions and values from one culture to another that is so dangerous and distorting. Specialists in Islamic art are of course perfectly well aware, in one sense, of the central importance of the ‘minor’ arts in their discipline, and of how much of the very warp and weft of Western art is absent from its Islamic counterpart. But it is one thing to be aware of an issue passively and quite another to register it actively. To accept that the ‘minor’ arts together constitute the greater part of Islamic art (architecture always excepted) in turn imposes the realisation that if pictorial art is to say anything in the Islamic domain, its message must be transmitted through these media. It follows that, insofar as Islamic art is not merely decorative, they are central rather than peripheral. It is worth underlining these points since, in the scant century of its existence, Islamic art history has developed from the largely uncontested

306

assumption that the art of the Muslim world was essentially decorative rather than meaningful. That assumption, or rather prejudice, insensibly infected the expectations of scholars as they investigated new material. Not surprisingly, what they did not look for they did not find. If one were to highlight a single significant development in this field during the last two decades, it would be the recognition that Islam too has an art of images, of meaning. Once that fact has fully percolated into what one might call the vast collective unconscious of the art-historical community, the way will be clear to understand Islamic art better. 1 So much for generalities. What of the books under review, and how do they bear on the concerns discussed above? They fall into several categories — catalogues, collected articles, monographs, iconographical studies — and thus testify to the vitality of Islamic art history nowadays. Indeed, they reflect the exponential growth of the field in the last twenty years, and also — it might be noted — the marked concentration on the ‘minor’ arts in this period. Scholarly production in this field far outweighs the rate of publication in Islamic architecture. The books under review here also show that a wide variety of approaches can successfully be adopted   Lustre Pottery. Technique, Tradition and Innovation in Islam and the Western World by Alan Caiger-Smith, London: Faber and Faber, 1985, 246 pp., 33 colour plates, 118 b. and w. illus. Persian Lustre Ware by Oliver Watson, London: Faber and Faber, 1985, 209 pp., 16 colour plates, 148 b. and w. illus. Syria and Iran. Three Studies in Medieval Ceramics edited by James Allan and Caroline Roberts, Oxford: Oxford University Press, 1987, 248 pp., 231 b. and w. illus. Studies in Islamic Art by Ralph Pinder-Wilson, London: The Pindar Press, 1985, 326 pp., 241 b. and w. illus., The Illustrations of the Maqamat by Oleg Grabar, Chicago and London: The University of Chicago Press, 1984, 195 pp., 723 b. and w. illus. on 9 fiches. Shah Jahan and Orpheus. The Pietre Dure Decoration and the Programme of the Throne in the Hall of Public Audiences at the Red Fort of Delhi by Ebba Koch, Graz: Akademische Druck- und Verlagsanstalt, 1988, 56 pp., 33 colour plates, 31 b. and w. illus. The Topkapi Saray Museum. Costumes, Embroideries and Other Textiles by Hülye Tezcan and Selma Delibaş, translated, expanded and edited by J.M. Rogers, London: Thames and Hudson, 1986, 216 pp., 136 colour plates. The Topkapi Saray Museum. Carpets by Hülye Tezcan, translated, expanded and edited by J.M. Rogers, London: Thames and Hudson, 1987, 248 pp., 98 colour plates. The Topkapi Saray Museum. The Treasury by Cengiz Köseoğlu, translated, expanded and edited by J.M. Rogers, London: Thames and Hudson, 1987, 215 pp., 124 colour plates. The Topkapi Saray Museum. The Albums and Illustrated Manuscripts by Filiz Çağman and Zeren Tanindi, translated, expanded and edited by J.M. Rogers, London: Thames and Hudson, 1986, 280 pp., 181 colour plates. 1

THE MAJOR MINOR ARTS OF ISLAM

307

to deal with this material. More than that, they show how significantly the standard of publication has risen in the last generation, and with it the scope and ambitions of scholarship in Islamic art history. New types of problem have been identified in paintings illustrating certain literary genres. The data underpinning monographic surveys of pottery, for example (and the same could be said for metalwork) have a fullness and precision not hitherto encountered in that field. Museum collections of the first rank whose importance has been undervalued in the past because too little has been written about them are now being published in extenso. Each of these advances is illustrated in one or other of the books under review here. Above all, perhaps, a new awareness of the subtleties of Islamic iconography informs these various endeavours. In sum, it is clear that Islamic art history has outgrown its earlier positivist and taxonomic bent and is embarking on new challenges. High time too. It is of course possible to argue back and forth interminably about the reasons why Islamic art turned its back on panel and easel painting and on sculpture. It is much more to the point, however, to recognise that the energy which, in the West, is directed towards large-scale painting and sculpture is here channelled into the applied arts. No other culture of the medieval world — not Byzantium, not India, not even China — put this degree of emphasis on the applied arts or developed such a wide range of them. Almost inevitably it follows that, insofar as the Islamic artist had anything to say, he would naturally gravitate towards these media in order to say it. If this argument can be accepted, how is it possible that the role of the Islamic ‘minor’ arts as vectors of meaning has been so thoroughly underestimated for so long? Here again, the fact that most Islamic art historians have been, and still are, Westerners — whose first encounter with the visual arts has been with European work, where painting and sculpture reign supreme — has exercised a malign influence. That influence can be traced right back to the semantics of the subject, which in turn reflect negative value-judgements. The phrases coined by Western art historians to describe such material within their own culture — ‘objects of vertu’, ‘applied arts’, ‘decorative arts’ and, most tellingly, ‘minor arts’ (or, in an even more patronising vein, the German term Kleinkunst) — imply in some obscure way a reduced intellectual input on the part of the artist, especially vis-à-vis the ‘major’ arts of painting and sculpture. Many would naturally take issue with so global a generalisation, and protest that in this or that case a work of the ‘minor’ arts like the Cellini salt-cellar is packed with meaning; but the portmanteau terms used to describe these arts speak

308

for themselves. They suggest the trivial and frivolous. University courses on art history, too, marginalise those arts that fall outside the rubric of architecture, painting and sculpture, and thus perpetuate the prejudice. So, too, do galleries, and even — though to a lesser extent — the sale rooms. Only in the museum world are these attitudes rectified. Thus in the discipline of art history at large the second-class status of the ‘minor’ arts seems set to continue. Islamic art has been a casualty of this process. Yet none of this accurately reflects the position of these arts in the field of Islamic art history. Their major status in that field is assured first of all by their role in Islamic society, where the products of weavers, metalworkers, potters, calligraphers and craftsmen in wood, ivory and glass had no whit less status than those of architects and painters. Nature, we are told, abhors a vacuum; and, as noted above, it is the vacuum left by painting and sculpture that these arts fill. They were allotted not merely the private and semi-public role which they had in the West, but also an unashamedly public, ceremonial and frequently symbolic significance. This public dimension made itself felt in all kinds of different ways. Islamic glazed pottery may serve as a paradigm of this proposition. Much of it had a display function, and half a dozen pointers in this direction readily come to mind. Excavations in the Buyid and Saljuq capital of Rayy have revealed a wall in a rich merchant’s house with specially shaped cavities to accommodate his collection of pottery; and a later version of this idea can still be seen in royal Şafavid buildings at Ardabil and Isfahan. These Islamic versions of the Welsh dresser imply not merely the accumulation of pottery, but purpose-built cabinets specifically designed for pieces of particular shapes, each piece nestling within its own specially created space like an object on display in a museum. The overglaze painting of mina’i ware would disappear almost immediately such pieces were used; and pottery denied a functional use must have been intended for display. Some of the double-shell wares of Saljuq times had a pierced outer body of such filigree delicacy that they could not have stood up to frequent use. Yet another proof of the display function of the finer wares is provided by fourteenth- and especially fifteenth-century book painting. Many court scenes have their foregrounds filled by small tables loaded with fine pottery and metalwork. Given that such pictures, with their narrow vertical format, are meant to be read from bottom to top, it is these vessels which meet the eye first and identify the scene as a courtly one. In medieval Islamic iconography, a standard visual reference to the King showed him seated and holding a small drinking cup — an image that spelt out the

THE MAJOR MINOR ARTS OF ISLAM

309

connection between ceramics and royalty. Finally, if still further evidence were needed for the display function of pottery, one might cite the case of Samanid epigraphic ware with its deliberately complex script spelling out sententious maxims not in the local language of Persian but in the high language of Arabic. Like the use of French on a menu outside France, this detail may possibly betray snobbery, or at any rate a desire to claim higher social status. In sum, then, these very varied clues all tend to the conclusion that in certain kinds of Islamic pottery the obvious utilitarian function was secondary, if, indeed, it was there at all. Display is merely one aspect of Islamic pottery which elevates its status. Another is its use of inscriptions. Islamic society venerated the written word; after all, in the Qur’an God Himself swears by the Pen. Often they are in Kufic, a script whose formal associations (it was the preferred script for building inscriptions) lent it particular authority. In others, the texts comprise Persian poetry, written in a hurried everyday hand; and here the implication is that only a cultivated person would be able to appreciate the work. Moreover, in both cases the inscriptions, far from being confined to a few words, were lengthy and so laid out as to be a major feature of the composition. In neither case, then, could such pottery — despite its humble material — be regarded as popular art. Truly humble pottery, of course, was not glazed and accounts for most of the ceramic finds in excavation and survey work, as the Sirjan material published in Syria and Iran shows. Another pointer to the high status of at least certain types of glazed pottery in Islamic society is the frequent use of royal iconography on it. Sometimes, it is true, the images are too simplified to be taken seriously as a deliberate exaltation of royal power; but often enough the images are densely composed and full of accurate and significant detail. They show monarchs enthroned, with crowns suspended above them, attended by pages and genii, and flanked by sword- or spear-bearers. The conventions of centrality, increased size, frontality and the like, as encountered in more obviously formal arts like book painting, are dutifully observed. Above all, a mass of relevant detail is faithfully reproduced on a plunging curved surface that actively discourages painting of such delicacy and precision. Moreover, such pieces were frequently executed in the difficult and expensive lustre technique, where the failure rate was higher than in any other type of pottery — in other words, every successful piece had to be priced so as to incorporate the cost of the failures — and were on occasion signed and dated. The implication seems to be that such ceramics were either a direct

310

instrument of royal propaganda or were intended to evoke a courtly ethos. These comments lead naturally to a realisation that ornament is an index of the major status enjoyed by pottery in the Islamic world. Once again, this factor is relevant not only in the case of pottery but applies to nearly all the other ‘minor’ arts as well. It is one of the most useful barometers of their standing. To note that so much Islamic art is decorative is merely to beg the question. The point at issue is why such decoration was applied so lavishly, since it is this lavishness which goes far to condition its cost and thus its status. The most likely answer is that certain types of ceramic were widely perceived as luxury items, and decoration was one way — perhaps the best way — of asserting this for all to see. Decoration is of course intimately related to technique, since so much decoration depends on the technique employed. ‘Abbasid blue and white ware, for instance, derives the distinctive blotting-paper quality of its ornament from the technique of applying the blue design to a still-wet slip. In sgraffiato ware the fact that the design has to be incised with a sharp instrument makes finesse impossible and puts a premium on boldly simplified motifs. Similar considerations apply to champlevé and lakabi wares, or to the imitations of T’ang splashed wares. Thus the most complex and ambitious designs were reserved for pottery that had a smooth surface, and in particular for lustre and mina’i ware. Both these latter types employed techniques invented in the Islamic world and peculiar to it. It is a particular strength of the books by Watson and Caiger-Smith that, for all their awareness of the limitations imposed on the potter by the very difficult technique of lustre ware, and their desire to chart its development as a pottery type, both authors are sensitive to the extra dimension introduced by the kind of decoration peculiar to such pottery. Caiger-Smith in particular waxes eloquent on the religious and metaphysical associations of lustre ware in Islamic society, especially its links with the symbolism of light, while Watson assembles a formidable mass of evidence, based principally on the original location of the material and its inscriptions, to suggest the close connection between Shi‘ite shrines and lustre tiles. Following quite a different line of research, Williamson (whose fieldwork, published posthumously, fills most of the volume entitled Syria and Iran) found that polychrome-glazed slip-painted wares, lustre ware and porcelain, alone of all the varied types of glazed pottery used in medieval Iran, were exported far from their place of manufacture. Thus in three separate books the evidence of iconography, site, epigraphy and archaeology fleshes out in new detail the role of lustre pottery in medieval Iranian society.

THE MAJOR MINOR ARTS OF ISLAM

311

Lustre pottery, then, for all its intrinsic interest, can also be seen as a paradigm of the central importance of the minor arts in medieval Islam. Evidence of a very different kind, but pointing in the same direction, is provided by the four sumptuous volumes devoted to the treasures of the Topkapi Saray Museum, all produced in a reworked English version by Michael Rogers. His colleagues owe him a most substantial debt for his self-effacing labour in mediating (and often quietly revising) current Turkish scholarship on this incomparable collection of artefacts. Of no royal European collection could it be said that the material has been amassed continuously since 1453 without being decimated by the normal hazards of war and acute problems of cash flow. Moreover, the fact that the Ottomans controlled much of the Islamic world — from Algeria to Iraq — made them over the centuries the natural recipients of gifts, tribute and plunder beyond ready computation. In particular, the accumulated riches of Cairo, Baghdad and Tabriz — the capitals or major cities of the rival Islamic powers whom the Ottoman sultans defeated — fell to their share. The volumes under review present some of the highlights of this unique collection, from which sculpture and easel or panel paintings are virtually excluded. It is above all the range of this imperial collection that is most diagnostic of the role of the ‘minor’ arts. The arts of the book, for example, comprise not only complete illustrated manuscripts but also individual leaves, sketches, illuminations, doublures (whether of painted leather or of paper cut-out work), fine bindings of leather, silk and lacquer, paper marbled, dyed, gilded and sprinkled, chinoiseries and calligraphies of all kinds. The contents of the Treasury extend well beyond the personal jewellery and regalia that one might expect to find, for they include thrones and other types of furniture, with lavish use of ebony, sandalwood, motherof-pearl, ivory and the like; arms and armour, including helmets, shields, quivers, swords and daggers, often ornamented far beyond the norm of their European equivalents; hardstones such as jades, rock crystal, jasper and agate; cutlery, gaming pieces, mirrors, boxes, flasks, vases and various domestic utensils, many of them of gold and richly encrusted with jewels. The carpets and textiles alone are not always fully representative of the best available work, for they were especially vulnerable to wear and tear. The pieces in the best condition owe their survival to the practice of bundling up much of the clothing, and occasionally even the carpets, of a sultan or vizier after his death and putting them into indefinite storage. Most of the carpets are of the nineteenth century and thus fall outside the

312

purview of this discussion. Not so the textiles. Their range of uses tells its own tale. Almost no function was too mundane to be exalted by highquality textiles of silk, velvet, brocade, satin and — usually for the sultan only — cloth of gold. Thus, in addition to the gowns, kaftans, robes and cloaks that one would expect, the collection contains tablecloths, sashes (kamarband, whence our ‘cummerbund’), towel-ends, handkerchiefs and napkins, bolster-cases, barbers’ towels, bags to hold banners, turbancovers, bedspreads, head-bands, and covers for such objects as mugs, books, boxes, vases, quivers and bow-cases. The sumptuousness of such fabrics has to be seen to be fully appreciated, but the quality of the colour plates in this volume, as in the others of the series, is a most acceptable substitute for the real thing. This visual spendour is in large part due to the use of intrinsically precious materials; and even when the materials themselves are not valuable they are made to seem so, whether by their mimetic quality — as in the case of the sheen of lustre ware, which evokes gold and silver plate — or by the complexity of ornament which they bear, as inlaid metalwork testifies. One has only to consult the selection of articles by Ralph PinderWilson, which contains the distilled wisdom of a lifetime’s work with Islamic objects in the British Museum, to be struck both by the range of materials in which the Islamic craftsman expressed himself — glass, ivory, stucco, jade, paper, rock crystal, silver, bronze and glazed ceramics all figure in these articles — and by the wealth of meanings attached to the works of art themselves. Objects made of the materials just cited bore the titles of rulers or high officials; scenes of royal iconography; astronomical and astrological images; Arabic, Persian and even Chinese inscriptions, including poetic and Qur’anic texts; and animals which carried various symbolic meanings: ibex, lion and tiger, or fabulous beasts such as the dragon, hippocampus, harpy and griffin. In some areas, obviously enough, Islamic art lends itself to methods of study long familiar to Western art historians. Oleg Grabar’s book on the maqamat, the picaresque romances which so fascinated thirteenth-century Arab audiences, is a case in point. The brief catalogue of extant manuscripts, the analysis of the interplay between text and image, the discussion of sources for these images, the identification of major themes and characteristics shared by most illustrated manuscripts of this text — all these are familiar exercises for the historian of Western and Byzantine book painting. Yet they are relatively new in the study of Islamic painting and are made all the more difficult to carry out because, on the whole, they all have to be done by a

THE MAJOR MINOR ARTS OF ISLAM

313

single person instead of a team or even generations of scholars. Thus it is truly a major achievement to have reduced to such clear order the 723 illustrations (the vast majority of them never previously discussed in print) taken from eleven manuscripts. Moreover, Professor Grabar’s success in relating the pictures to the quirks of the society which produced them sets new standards in the study of Islamic painting. It is no wonder that the book was over twenty years in gestation. But it has acquired in the maturing a remarkable richness of texture and suggestion; a very great deal of matter — information, analysis, speculation, comparisons, generalisations — is compressed into less than 200 pages. It is for this reason that the book should be read not only by specialists in Islamic painting, but by all Islamic art historians. The text is informed throughout by a wonderfully clear perception of where the maqamat images and their details belong, not only in the context of thirteenth-century Iraqi art, but within the slower and changing rhythms of medieval Islamic art in general. It is this awareness of wider perspectives transcending the barriers of time and space and media that lends Professor Grabar’s findings an authority currently unique in contemporary Islamic art history. One can only be grateful that he has brought to fruition his long preoccupation with this illustrative cycle, and that in so doing he has blazed a trail for future studies of Islamic painting — and, as it were incidentally, of the Islamic ‘minor’ arts. Ebba Koch’s book, a tour de force of erudition, forensic skill and disciplined imagination, is at the opposite pole to Oleg Grabar’s study of the maqamat. Its focus is microscopic, not macroscopic. Yet it, too, explores, in a manner still much too rarely attempted, the encounter between East and West in the world of symbols. The stage on which this interplay is acted out is the Mughal court of the seventeenth century, and specifically the jharoka or architectural setting for the enthronement of Shah Jahan in Delhi. Here, the only image of a human being in the entire ensemble is the figure of Orpheus, presiding over the mise-en-scène from on high. He was equated with Solomon and thence, by an easy transition, with the Mughal Emperor himself. Related minglings of Eastern and Western ideas have been studied recently by Julian Raby in the context of the early Ottoman state, while Ettinghausen’s analysis of the bath hall at Khirbat al-Mafjar shows that the process was well under way as early as the Umayyad period. No doubt a determined search would unearth further examples. Dr Koch’s study is a salutary warning that one should never be misled by the glamour of ‘mere’ decoration to underestimate the intellectual content of Islamic art.

XV Recent Work on Islamic Iconography

A

Festschrift for Richard Ettinghausen was long overdue, as one of the 1 contributors to this volume notes. Though it is the first, let us hope that it is not the last; for Professor Ettinghausen is the acknowledged doyen of Islamic art studies and his career is still in full flood. Moreover, of the fifteen articles in the book (excluding a most valuable bibliography by June Taboroff), a mere four are by art historians proper; the remaining eleven are principally by specialists in literature. While the links between 2 art and literature are thus considered in extenso, there is still ample scope for others of Professor Ettinghausen’s more immediate colleagues to honour him in a future Festschrift. It seems sensible, given the remit of Oriental Art, to concentrate attention on the four art historical papers, even though this will result in the literary contributions receiving comparatively short shrift. Their contents will, however, be summarised. Kathleen R. F. Burrill writes on “The Farhād and Shïrïn Story and its Further Development from Persian into Turkish Literature”. She mentions in passing how the legend of Farhād’s Chinese origins grew; veneration for the Chinese as craftsmen par excellence can be noted in the 3 early ‘Abbāsid period and quickly became a topos of Islamic literature. It is  Studies in Art and Literature of the Near East in honor of Richard Ettinghausen, ed. P. J. Chelkowski (New York, 1974). 2  The article by Professor Cammann is the only one which does not have a substantial discussion of the literary elements of the work of art under discussion. 3  See for example al-Tha‘ālibī (The Book of Curious and Entertaining Information. The Latā’if al-ma‘ārif of Tha‘ālibī, tr. C. E. Bosworth [Edinburgh, 1968], 141 ; Arabic text inaccessible to me); and al-Mas‘ūdī (Murūj al-dhahab wa ma‘ādin al-jauhar ed. C. Barbier de Meynard and Pavet de Courteille [Paris, 1861], I, 322, 1. 10–324, 1

RECENT WORK ON ISLAMIC ICONOGRAPHY

315

of a piece with this legend that Farhād should be portrayed as a universal man in the range of his artistic abilities, a kind of Islamic Leonardo. Professor Burrill also contributes a substantial paper of some fifty pages tracing the role of literature and art as barometers of change in Turkish society—”From Gazi State to Republic: A Changing Scene for Turkish Artists and Men of Letters”. In its mastery of material, its unobtrusive compression and its sure grasp of essentials, this article is an impressive achievement, contriving as it does to guide the reader through six hundred years of complicated political and social history. Ehsan Yarshater demonstrates in a wide-ranging paper modestly entitled “Affinities between Persian Poetry and Music” how the rhythm, ornament and mood in Persian poetry all have their analogues in Persian music. His comments on the dependence of folk poetry on music dovetail remarkably with recent work by Professor Pierre Cachia on Arabic popular literature. This suggests that the origins of the zajal and the muwashshaḥ lie mainly in Arabic folk poetry which was often intended to be sung, although it also seems that in the Arab world different prosodic standards were followed by practitioners of these arts within the same region and at the same time. By way of preface Professor Yarshater gives a useful summary of the current state of knowledge about Middle Persian poetry and speculates on the prime role played by music in this tradition. Despite its importance, however, music was a rather elusive element of poetry, and in this sense a parallel can be drawn with the strict spatial organisation of 15th and 16th century Persian miniatures, an organisation which is also 4 not immediately apparent. The theme of William Hanaway’s somewhat inconclusive article on “Interart Correspondences” is the difficulty of establishing the existence of links between the arts. Lois A. Giffen attempts, in her article “In the Artist’s Mind (?); Some Prevalent Ideas about Animals from the Qur’an and Tradition Literature”, to discover a religious dimension in the depiction of animals in Islamic painting. But despite the wealth of quotations she amasses from the Qur’ān and ḥadīth the picture she draws is, as she herself recognises, one-sided, for it ignores such major secular compilations as Zakariyā Qazvīnī’s ‘Ajā’ib al-makhlūqāt wa gharā’ib al-maujūdāt or the numerous bestiaries. Moreover, it remains to be proved that the artists of the Maqāmāt in particular were not 1. 4; tr. A. Sprenger: El-Mas’údí’s historical encyclopaedia, entitled “Meadows of Gold and Mines of Gems”; [London, 1841], 340–1. 4  See G. D. Guest, Shiraz painting in the sixteenth century (Washington, 1948), 25–32, figs. 1–9.

316

simply portraying the animals familiar to them in their daily life. Detailed, even loving, observation could account for the artists’ mastery of animal types without the need to invoke conscious, let alone unconscious, religious motives. In a later article—”The Lion and the Panther, the Ibex and the Beaver: Remarks on the Durability of Some Zoological Knowledge”—Professor Giffen demonstrates beyond the possibility of counter-argument the dependence on Aristotle of certain material in both Western and Islamic bestiaries. Her findings are particularly useful in the light of the illustrated Manāfi’ al-Hayawān of Ibn Bakhtīshū’ in the Pierpont Morgan Library. They show how ample the scope is for further detailed iconographic studies in this field, to supplement Professor Ettinghausen’s 5 own work; it is a pity that the material which the late D. S. Rice accumulated for his proposed study on the cheetah has never appeared. Marcia E. Maguire, in “The Shāhnāmah and the Persian Miniaturist”, gives new point to the familar caveat that there was no standard text of the Shāhnāma by discussing a case where the inscription below a picture is at variance with the received text. She notes several implications of this incongruity. Perhaps, too, the desire for variety or for greater dramatic impact could on occasion drive the artists to ignore, falsify or just flatly contradict the text. In the case under discussion, where the hero is killing 6 an animal identified by the inscription as a rhinoceros, though the text calls for the slaying of a dragon here, one could well imagine that the artist had had his fill of illustrating the dragon-killings that punctuate the 7 text and was ready for a change even if the text would not oblige him. Her remark that the Shāhnāma was valued for its descriptive capacity by artists as well as littérateurs goes far to explain the bustling activity of Shāhnāma illustrations in the Tïmūrid and Şafavid periods. In an art form which had by and large lost that sense of raw drama and high seriousness which marks much of Mongol painting, it was natural to turn instead to  See items 122, 123 and 126 in the bibliography which prefaces this Festschrift; cf. also J. Sourdel-Thomine, “Animals in art”, s.v. Ḥayawān in The Encyclopaedia of Islam, 2nd ed., II, 309–11, which incorporates a useful bibliography. 6   Though as Professor Ettinghausen shows in his study of the unicorn, this same word could be used to describe several kinds of animal (Studies in Muslim Iconography I: The Unicorn [Washington, 1950] 6–10). 7  The surviving portions of the “Demotte” Shāhnāma, from which this illustration comes, contain two scenes featuring dragons (D. Brian, “A reconstruction of the miniature cycle in the Demotte Shah Namah”, Ars Islamica VI/2 [1939], nos. 5 and 49). Moreover, it is highly likely that the artist, in previously illustrating the Shāhnāma and other epic or heroic poetry, would already have had ample opportunity to depict dragon-slayings. 5

RECENT WORK ON ISLAMIC ICONOGRAPHY

317

the conscientious and even pedestrian depiction of detail. The text of the Shāhnāma gave ample warrant for such an approach. In her subsequent article Marcia Maguire, comparing “The Haft Khvān of Rustam and Isfandiyār”, draws an important distinction between the poet’s treatment of Rustam—the simple, upright man of action—and his treatment of Isfandiyār as a hero whose deeds are motivated by religious (that is, Zoroastrian) fervour as well as by the desire to do good. She notes that it is typical of the non-intellectual approach of the Persian miniature painter that he does not seek to follow the poet in distinguishing these two very different types of hero. She realises that the main problem besetting the miniature painter was that he had to produce varied illustrations for a text whose content was basically repetitive, and this gives particular value to her discussion of the relationship between text and picture. Minoo S. Southgate, writing on “Fate in Firdawsï’s “Rustam va Suhrāb’”, stresses that a sense of fate’s immutability is central to the poem, while Seyyed Hossein Nasr generalises on the role of “Sacred Art in Persian Culture”. Finally, Khosrow Mostofi provides a scrupulously fair assessment of “Idries Shah, a Latter-day Sufi,” a paper which includes a lucid and concise account of Şūfism and its origins. It will be apparent that most of these articles contain valuable insights on the relationship between literature and art in the Islamic world. But for art historians the principal interest of the volume lies in the four major studies mentioned earlier: one on the interplay between Sasanian and Islamic iconography, two on miniature painting and one on carpets. In an ambitious and superbly documented study entitled “Farhād and Ţāq-i Būstān: The Growth of a Legend”, Priscilla P. Soucek defines the grotto as a victory monument erected by the newly enthroned Khusrau II (Fig. 1). She then explores how legend and myth gradually infiltrated the later interpretations of the monument. Madame Gautier-van Berchem’s damning inventory of the progressively inflated Arab accounts of 8 Byzantine involvement in the Damascus mosaics springs to mind. The material presented by Professor Soucek implies a paradoxical conclusion, namely that, while the inspiration of the artists dealing with the story of Khusrau, Farhād and Shïrïn was primarily Sasanian, virtually no attempt was made to check the details of the story against the extant monument itself, which was easily accessible. The topographical element which was  See M. Gautier-van Berchem’s contribution to K.A.C. Creswell, Early Muslim Architecture (revised ed., Oxford, 1969), I/2, 231–42. 8

318

to be so notable a feature of Ottoman painting was poorly represented in Iran. It was perhaps inevitable therefore that the original meaning of the monument should have become overlaid with legends which had a more immediate appeal. One may wonder how these romantic confections could have maintained credibility in face of an examination of the monument in situ. The only possible conclusion seems to be that the artists usually saw what they wanted to see and ignored the rest. Yet two of the illustrations of Niẓāmī’s poem which are reproduced in the article show that the artist had accurate information about the site and preferred to depict that than to represent visually the inaccuracies of Niẓāmī’s text. A similar independence of the text, the ostensible subject-matter of the illustrations, is noted elsewhere in this Festschrift in the context of the Shāhnāma and the Maqāmāt. It would be useful to draw together these and other scattered references and attempt to define the various types of response of the artist to his text. The work of Marcia Maguire and Dr. Swietochowski has identified some of the approaches, but their conclusions would no doubt be modified and expanded in the light of other types of illustrated texts. The major issue of the paper, however, concerns the attitude of the medieval Iranian artist to his Sasanian heritage. Here Professor Soucek breaks new ground, and with signal success; for although T. W. Arnold devoted a lecture to this topic which outlined some of the basic problems, 9 his work has until now remained almost alone in its area. If the single theme discussed here can be treated in such detail, one can confidently say that an entirely new field has been opened for future research. No doubt other and equally striking examples will be found to testify how reverence for the Sasanian past could co-exist with gross misunderstandings of it. It is remarkable how quickly a historical figure like Khusrau II could recede into the mists of legend and how the stories of Firdausï could seem more immediate to the medieval Muslim than the history of three or 10 four centuries before. The entry of Farhād onto the stage at Ṭāq-i Būstān is of a piece with the folk identification of Persepolis as the throne of 9  T. W. Arnold, Survivals of Sasanian and Manichaean art in Persian painting (Oxford, 1924), esp. 9–10. 10  Another folk memory of his activity in this area is the name given to the huge smoothed surface at Bīsutūn, which forms a panel measuring some 200 m. x 55 m.—the Tarāsh-i Farhād. In the context of Professor Soucek’s study, it is interesting to note that Dr. L. Trümpelmann has argued that this site was intended for the tomb of Khusrau II (in his paper “Die Gräber der Sasanidenkönige”, deliverd at the VIIth International Congress of Iranian Art and Archaeology in Munich in 1976 [now published as an abstract in W. Kleiss, ed., Akten des VII.

RECENT WORK ON ISLAMIC ICONOGRAPHY

319

Jamshīd or the Sasanian bas-reliefs nearby as pictures of Rustam. The difference is that in the case of Ṭāq-i Būstān, the monument’s historical association with Khusrau would have made the transition to the legend of Khusrau and Shïrïn quite natural and appropriate. Dr. Melikian Chirvani has recently added another dimension to this falsification of the past by analysing the element of religious veneration in the popular attitude to 11 pre-Islamic sites in Iran. The Islamic attitude, which at times amounts to a rewriting of history, is all the more puzzling in view of the remnants of 12 Sasanian tradition preserved by later geographers, historians and poets. These features include themes such as ritual libations, royal legitimacy, 13 homage scenes, banquets and drinking bouts. But the Muslims either ignored or had lost the clues that would make sense of these isolated scraps of the past and, once committed to a romantic misconstruction of the reliefs, they could not at the same time keep in the back of their minds the possibility of other interpretations. Thus Anahita became Shīrīn, and elements which would not clarify the romantic theme—e.g. the winged figures, the hunting scenes and the equestrian statue—were simply ignored. We may note that even the earlier Islamic interpretations, which 14 represented a half-way stage between factual history and later myth, were swept away once the idée fixe of the Khusrau and Shīrīn story had gained ascendancy. Despite the wealth of evidence adduced here, Professor Soucek’s conclusion that the grotto at Ṭāq-i Bīstān is in part a victory monument could be challenged. Virtually all the material she cites indicates the dominance of a double theme: investiture and the power of the king. Internationalen Kongresses für Iranische Kunst und Archäologie München 7.–10. September 1976 (Berlin, 1979), abstract on 370–1]. 11  “Le royaume de Salomon. Les inscriptions perses de sites achéménides”, Le Monde iranien et l’Islam I (1971), 1–41. 12  The evidence of Tha‘ālibīs Ghurar akhbār mulūk al-Furs wa siyaruhum (ed. and tr. H. Zotenberg [Paris, 1900]) shows that detailed and factual accounts of the Sasanian period were available at the very time that the legendary material began to win general acceptance. See A. Christensen, L’Iran sous les Sassanides (2nd ed., Copenhagen, 1944), 59–74. 13  Professor Soucek gives examples referring to Ṭāq-i Būstān (op. cit., 40–41; other passages with a more general relevance have been collected by O. Grabar in “The painting of the six kings at Quṣayr ‘Amrah”, Ars Orientalis I (1954), 186. E. Herzfeld has shown that this misunderstanding of Sasanian tradition was already current in early Islamic times (“’Die Könige der Erde’ zu ‘Der Islam’ XIV, 402– 6”, Archäologische Mitteilungen aus Iran V [1933], 151). 14  It is easy to see how the 10th-century text of Bal‘amī could give rise to the popular legends of the later middle ages (Soucek, op. cit., 44).

320

Specific pointers to the victory theme are conspicuously absent, apart from the winged figures in the spandrels of the façade (Fig. 2). It is true that the location, form, gestures and attributes of these figures all make 16 the classical and Byzantine personification of victory the obvious and natural parallel. But it is certainly worth bearing in mind, as Professor Soucek herself does, the possibility of other interpretations. Such winged 17 figures occur very rarely in Sasanian iconography, a fact which is all the more surprising since victory is such a leitmotif of Sasanian official art. Each figure, moreover, seems to issue from a cornucopia, which would 18 strengthen the association with the themes of fertility and prosperity. A further reason against identifying these winged figures as symbols of victory is the stylised plant decoration immediately beneath them (Fig. 3). This motif inevitably recalls one of the most ancient motifs in Near 19 Eastern art, the tree of life. It cannot easily be made to fit into a victory iconography, yet its presence has to be explained, since its conspicuous location on the façade, when seen in conjunction with the highly organised iconography of the monument as a whole, argues that it is not intended to be a neutral element in the decorative scheme. It may in fact provide a clue towards a different interpretation of the façade, and one that is touched on by Professor Soucek herself when she suggests that the winged figures 20 may be fravashīs. Could the façade be intended as a kind of auspicious prologue to the imperial themes developed within the īwān? Its images, at once cultic and apotropaic, could very easily be taken to refer to the good fortune which would attend the pious ruler on earth and in the hereafter. This would explain, for example, the bowl of fruit which the 15

15  Professor Soucek interprets the boar hunt in this light (op. cit., 31–2), but leaves the stag hunt—an exact pendant to that scene—out of account. Yet both scenes are given equal prominence in the grotto. It seems preferable, therefore, to interpret both of them as expressions of royal might, or even as depictions of court life, rather than to see specific victory overtones in only one section of these reliefs and not in the other. 16  For a representative series of parallels, all taken from the spandrels of arches, see D. E. Strong, Roman Imperial Sculpture (London, 1961), 104 and pl. 141 a–e. 17  An example may be seen at Bīshāpūr, in the relief of Shapur I and Valerian (E. Herzfeld, Archaeological History of Iran [London, 1935], pl. XI, bottom). 18  Mentioned en passant by Professor Soucek (op. cit., 31). 19  Cf. G. Lechler, “The Tree of Life in Indo-European and Islamic Cultures”, AI IV (1937), 369–419; E. Baer, Sphinxes and harpies in medieval Islamic art. An iconographical study, (Jerusalem, 1965), 56–7, 65; and G. Öney, “Das Lebensbaum Motiv in der seldschukischen Kunst in Anatolien”, Belleten XXXII, No. 125 (1968), 37–50. 20  Soucek, op. cit., 38.

RECENT WORK ON ISLAMIC ICONOGRAPHY

321

winged figures hold; the association of the cup and of certain types of fruit with themes of cosmic power and paradise maintained itself in later 21 Islamic iconography. The other object which the winged figures hold is the beribboned circlet, the emblem of sovereignty; other Sasanian reliefs show that its prime use was in scenes of investiture, in which victory appeared as a secondary theme, if at all. Lastly, the crescent moon at the apex of the arch, as the emblem of Anahita, goddess of the waters, fits in with the site of the monument by a spring, and with the depiction of the goddess on the back wall of the grotto. It is wholly fitting that the place of honour should be reserved for the emblem of the deity of the place. But this detail, too, does not readily lend itself to being interpreted as a symbol of victory. Similarly, the equestrian portrait or statue in the late antique and Byzantine 22 world did not in itself connote victory. It was essentially an expression of majesty and power, and it was the addition of extra figures such as tyches or 23 winged victories which gave it the extra gloss of conquest. Other variants of the victory theme in early Byzantine art—the most relevant parallels for the present purpose—usually employed unmistakable references to conquest: processions of barbarians offering tribute, defeated chieftains holding the imperial spear, and personifications of the earth holding the 24 rider’s foot Other Sasanian reliefs show that similar images were used in contemporary Iran; the most famous example is perhaps the relief 21  K. Otto-Dorn, L’art de l’Islam, tr. J.-P. Simon (Paris, 1968), 85; eadem, “Türkisch-islamisches Bildgut in den Figurenreliefs von Achthamar”, Anatolia VI (1961), 2 and pl. II. Incidentally, although a wall painting from the ḥarīm of the Jausaq al-Khaqānī published by Herzfeld seems to constitute the ideal missing link between Ṭāq-i Būstān and later Islamic iconography, the key part of the painting is a reconstruction (Die Malereien von Samarra [Berlin, 1927], 22, Abb. 7 and Taf. XII-XIII). 22 Cf. for example the statue of Marcus Aurelius (D. E. Strong, Roman Art, prepared for press by J. M. C. Tonybee [Harmondsworth, 1976], 113 and pl. 151) or that of Theodosius I (N. Q. King, There’s such Divinity doth hedge a King [Edinburgh, 1960], 15) or those of Justinian (Averil Cameron, “Corippus’ poem on Justin II: a terminus of antique art?”, Annali della Scuola Normale Superiore di Pisa, Classe di Lettere e Filosofia, Serie III, V/1 [1975], 161). I am grateful to Dr. Robin Cormack for drawing my attention to this article. 23  E.g. the silk in the Cathedral Treasury at Bamberg showing a mounted emperor receiving a crown and a diadem from personifications of cities (A. Grabar, L’empereur dans l’art byzantin [Strasburg, 1936], pl. VII/1). 24  Cf. the panel from a diptych datable c. 500 in the Museum of the Castello Sforzesco, Milan; and the so-called “Barberini” diptych of roughly the same date, in the Louvre (D. T. Rice, The Art of Byzantium [London, 1959], pls. 18 and 19). For further types of submission images cf. Cameron, op. cit., 144–5.

322

of Valerian kneeling before Shapur II at Naqsh-i Rustam, while other 26 27 reliefs depict the emperor standing or riding on his prostrate opponent, 28 frontally enthroned with rows of tribute bearers approaching him, and 29 bringing down his enemy like a medieval knight at a tournament. The lack of any comparable image in the Ṭāq-i Būstān grotto, which at the same time offers perhaps the richest repertory of any group of Sasanian reliefs, gravely weakens the interpretation suggested by Professor Soucek. This conclusion also has repercussions on the date which she proposes for the monument, since it is a lynchpin of her argument that the victory element in the iconography is a reference to Khusrau II regaining his ancestral throne. By that reckoning the grotto would date to c. 591. Even if this date needs to be revised, however, there does not seem to be any compelling reason for doubting that the monument was erected by Khusrau II. If the themes of investiture and power are more closely examined it will be seen that the very features which for Professor Soucek connote victory are also susceptible to these other interpretations. Winged figures preside over numerous Byzantine consular diptychs of the 5th and 6th centuries, and the purpose of these diptychs was to announce the investiture of the 30 consul in his new office. Similarly, the hunting scenes can equally well be interpreted as expressions of the kingliness of the ruler, who alone in the 31 land can undertake these ambitious ceremonial hunts, as of victory. The presence of musicians in a boat suggests that the king’s daily life is being depicted (Fig. 4); there would be no place for them in a victory image. Significantly enough, they recur in Byzantine images of royal or official 32 power, and the association of the hunt and music persisted into Umayyad 25

 K. Erdmann, Die Kunst Irans zur Zeit der Sasaniden (2nd ed., Mainz, 1969), pl.

25

21.

 Ibid., pl. 28.  R. Ghirshman, Iran. Parthians and Sassanians, tr. S. Gilbert and J. Emmons

26 27

(London, 1962), pls. 168, 196, 202 and 206. 28  Ibid., pl. 225. 29  Ibid., pl. 220. 30  Indeed, he is commonly shown holding the mappa circensis, which he dropped to inaugurate the games held to celebrate his taking office (E. Capps, Jr., “The style of the consular diptychs”, Art Bulletin X [1927], 62 and figs. 2, 4, 7–8, 21–2, 27 and 29-32). 31   Cf. note 14 above. 32   E.g. at the base of the Theodosian obelisk, usually dated to c. 400 (A. Grabar, op. cit., 66 and pl. XI) and in the frescoes at St. Sophia at Kiev, datable c. 1037 (ibid., 72 and fig. 2). In both cases the context is the ceremonial of the Hippodrome, in which the imperial victory was an important theme, but there appears to be no precise connotation of victory in these depictions of music-making.

RECENT WORK ON ISLAMIC ICONOGRAPHY

323

times. Finally, the equestrian statue here, as indicated above, is robbed of associations with victory by the very lack of winged attendants or of 34 a defeated opponent. But as a solid, threatening mass this huge hulking figure, dehumanised by the helmet which masks his face, is royal power 35 incarnate (Fig. 5). The multiple appearances of the king in the sculpture of the grotto 36 make best sense in the context of Sasanian silverware. There too several facets of the king’s life are depicted. The reliefs within the Ṭāq-i Būstān grotto are uniquely full and detailed and thus it is not surprising that the particular combination of royal images which they present is not repeated elsewhere. But the individual themes are quite within the spirit of Sasanian royal iconography: investiture, hunting and music. To show the king arrayed in the full panoply of war puts the finishing touch to this 37 précis of royal life. In the scenes of court life at Pandjikent, which are roughly contemporary with the sculpture in the grotto at Ṭāq-i Būstān, formidable equestrian figures are also juxtaposed with scenes of hunting 38 and banqueting. It was perhaps not wholly naïve of medieval Islamic writers to interpret the great horse as Shabdīz—the favourite horse of Khusrau Aparvez— and the grotto as his monument, “the stall of Shabdīz”. They record, after 33

33  Notably in a floor painting at Qaṣr al-Hair al-Gharbī (D. Schlumberger, “Deux fresques omeyyades”, Syria XXV [1946], 97–8 and pl. B). 34   The sculptor must have been aware, too, of the investiture relief of Ardashir II only a few yards outside the grotto. Here again the motif of conquest is unmistakable (Erdmann, op. cit., pl. 28). 35  This equestrian figure has often been likened to the fearsome cataphract of the Persian armies (cf. the wall painting from Dura Europos illustrated in Ghirshman, op. cit., pl. 63b). Part of the terror which these cavalry troops inspired stemmed from the extraordinary size of their horses. The image at Ṭāq-i Būstān may therefore be an attempt to communicate as directly as possible the frightening aspects of royal power. Herzfeld analysed a similar form of visual shorthand in the frontal posture of the enthroned king, his knees akimbo and his sword planted between them. He stigmatized this favoured Sasanian theme as barbarous in its directness (Archaeological History, 87). 36  This is abundantly clear as one leafs through the plates in O. Grabar and M. Carter, Sasanian Silver. Late Antique and Early Mediaeval Arts of Luxury from Iran (Michigan, 1967). 37  Compare the many different iconographic references to Christ found, say, at S. Vitale in Ravenna; on this general point see A. Grabar, “Mosaïques à Ravenne”, Cahiers Archéologiques VIII (1956), 252. 38  See M. Bussagli, Painting of Central Asia, tr. L. Small (Geneva, 1963), pls. on 44–6; A. Belenitsky, The ancient civilisation of Central Asia, tr. J. Hogarth (London, 1969), pls. 134, 139–42, 144 and 145.

324

all, that Bahrām Gūr built a mausoleum (nā’ūs) at Qaṣr-i Shīrīn intended 39 jointly for his paramour and for the gazelle he had killed, and they may even have been influenced (if only indirectly) by the fabled devotion which Alexander the Great cherished for his horse Bucephalus, in whose memory he built the city that bore its name. The hybrid style of the Ṭāq-i Būstān reliefs has one further Western parallel which may be worth noting. Resemblances to Coptic art can be noted particularly in the winged figures occupying the spandrels of the façade. The pneumatic quality of these figures, their pronounced voluptuousness, is quite alien to Byzantine art, even though their stylistic peculiarities attest close links with sixth-century sculpture from Constantinople. The spirit which animates these superficially Byzantine forms is readily matched in Coptic work. To be sure, Coptic craftsmen quickly developed a distinctive, terse and essentially profane idiom which dispensed with the laborious imitation of classical forms displayed by the spandrel figures at Ṭāq-i 40 Būstān. Nevertheless, there do exist some Coptic sculptures, generally dated in the early part of this development, which are directly relevant to 41 the present discussion. Their function is official rather than popular, and their style is ponderous rather than light-hearted. While they retain that sensual flavour which is the hallmark of the Coptic style, they contrive to combine it with competent classicising detail, and thus provide close analogues in both form and spirit to the winged figures at Ṭāq-i Būstān. Similarly, Coptic art provides the most convincing available parallels 42 for the wig-like coiffure of the spandrel figures, and for the treatment of their hemlines, where the drapery recalls the “trumpet folds” of ivories 43 attributed to Alexandria. These analogies need not necessarily mean that the Ṭāq-i Būstān reliefs were directly influenced by Christian Egypt; they could be explained by the common ancestry of Coptic and classicising Sasanian work in Hellenistic and thence late antique art. But it is worth 39  See Ibn al-Faqīh, Kitāb al-Buldān, ed. M. J. de Goeje (Leiden, 1885), 255, l.17 – 256, l.17; tr. P. Schwarz, Iran im Mittelalter nach den arabischen Geographen (repr. Hildesheim and New York, 1969), 547–8. See also Yāqūt, Mu‘jam al-Buldān, Cairo edition (1885), VII, 100, 11. 7–14 and VIII, 241, 1. 9–242, 1. 2; partial translation and paraphrase in Schwarz, op. cit., 547, nn. 8–9 and 548, n. 1. 40  Cf. J. Beckwith, Coptic Sculpture 300–1300 (London, 1963), pls. 30, 38, 60 and 61. 41  Ibid., pls. 19 and 103. 42  E. Kitzinger, “Notes on Early Coptic Sculpture”, Archaeologia 87 (1937), pls. LXVII/1 and LXXIV/2. 43  Beckwith, op, cit., pls. 39 and 95.

RECENT WORK ON ISLAMIC ICONOGRAPHY

1. The grotto at Ṭāq-i Būstān (after A Survey of Persian Art).

2. Ṭāq-i Būstān: winged figure in spandrel of main īwān (after A Survey of Persian Art).

325

326

3. Ṭāq-i Būstān: plant motif from façade of main īwān (after A Survey of Persian Art).

4. Ṭāq-i Būstān: musicians.

RECENT WORK ON ISLAMIC ICONOGRAPHY

327

5. Ṭāq-i Būstān: equestrian figure (after A Survey of Persian Art).

6. al-Ḥarīrī, Maqāmāt, dated 1237. Bibliothèque Nationale, Paris, ms. arabe 5847, fol. 101: 32nd maqāma (after R. Ettinghausen, Arab Painting) .

328

7. Firdausī, Shāhnāma, dated 1429, Gūlistān Palace Library, Teheran: Isfandiyār kills Arjāsp in the Brazen Castle and rescues his sisters (after A Survey of Persian Art).

RECENT WORK ON ISLAMIC ICONOGRAPHY

329

8. Niẓāmī, Khusrau u Shīrīn, c. 1540, Royal Scottish Museum, Edinburgh: battle between Khusrau and Bahrām Chūbīna and their armies (after Robinson, Persian Miniature Painting).

330

9. The Sanguszko Carpet, Metropolitan Museum of Art, New York (after A Survey of Persian Art).

10. Shrine of Turbat-i Shaikh Jām, restored under Shāh ‘Abbās I: detail showing dragons in spandrels (photo Bernard O’Kane).

11. Imāmzāda ‘Abbās, near Āmul: wall-painting of ‘Alī and other figures.

RECENT WORK ON ISLAMIC ICONOGRAPHY

331

12. Imāmzāda ‘Abbās, near Āmul: wall-painting showing equestrian figures of Shī‘ite imāms.

332

13. Imāmzāda ‘Abbās, near Āmul: wall-painting depicting dragon, mosque lamp and “Birds of Paradise”

RECENT WORK ON ISLAMIC ICONOGRAPHY

333

334

remembering that surprisingly direct Sasanian influence can be detected in 44 45 certain Coptic ivories and textiles. Such direct quotations from Sasanian 46 iconography are notably absent in Byzantine art proper. This can only strengthen the case for contacts between Iran and Egypt. In “Pictures or Commentaries: The Illustrations of the Maqāmāt of al-Ḥarīrī”, Oleg Grabar re-opens a subject which has already long engaged his attention. In a paper which abounds in acute insights and unexpected conclusions it may be invidious to single out the odd remark as questionable; but Professor Grabar’s statement that “each manuscript must be seen as a creation independent of any other known one” is nothing short of startling. True, he qualifies it by admitting that among some 1100 Maqāmāt illustrations certain iconographic details do recur, but he maintains that “there is not a single instance in which one could prove that any one known manuscript derives from any other known manuscript or from its immediate model”. Given the general tendency of Islamic 47 painters to copy the work of their predecessors, and the bestseller status 48 of al-Ḥarīrī’s Maqāmāt, this is indeed a surprising conclusion. Moreover, it comes close to being disproved by the evidence cited by Professor Grabar himself. He rightly characterises the detail of the woman driving camels in the 32nd maqāma as “a very minor” element in a very long maqāma, and even the presence of the hero and the narrator in the same picture or in an adjoining one does not redeem its modest status (Fig. 6). Why then should this same incident be illustrated in at least three Maqāmāt  E.g. a 7th-century ivory at Baltimore depicting an enthroned monarch in Sasanian style (Ghirshman, op. cit., pl. 402). 45  W. F. Volbach, Early Decorative Textiles, tr. Y. Gabriel (Feltham, 1969), pl. 26. 46  Nevertheless, Sasanian iconography did penetrate westwards, a phenomenon analysed in general terms by Ghirshman (op. cit., 283–315) and, in the particular case of 10th-century Spain, by A. Grabar (“Eléments sassanides et islamiques dans les enluminures des manuscrits espagnols du haut moyen Âge”, Atti del IIo Convegno per lo studio dell’arte dell’alto medio evo tenuto presso l’Università di Pavia nel settembre 1950, ed. E. Arslan [Turin, 1952], 312–19 and tav. CXCIV–CXCVIII). 47  See most recently a paper by N. Titley: “Persian Miniature Painting: the Repetition of Compositions during the Fifteenth Century”, to be published in the proceedings of the VIIth International Congress of Iranian Art and Archaeology. It demonstrates, with a wealth of evidence, that certain miniatures from the Kulliyāt of Khwājū Kirmānī (1396) and from the miscellany executed for Iskandar Sulṭān in 1410–11 were the ancestors of a whole series of similar scenes throughout the 15th century. [Now published in Kleiss, Akten, 471–91.] 48  A very different conclusion—in the case of the three London Ḥarīrī manuscripts—is reached by H. Buchthal in “Three Illustrated Ḥarīrī Manuscripts in the British Museum”, Burlington Magazine LXXVII (1940), 148; he believes that each of them repeatedly copied the Schefer Ḥarīrī or a similar manuscript. 44

335

RECENT WORK ON ISLAMIC ICONOGRAPHY

manuscripts? This is surely copying rather than coincidence. Whether the original manuscript from which the others derive is lost or is one of those still extant is of secondary importance. As for the reason why such an incident was chosen for illustration, the evidence in the case of alWāsiṭī, the artist who illustrated the Schefer Ḥarīrī, suggests that he was drawn to this scene because it enabled him to exhibit his particular talent for arranging groups of people or animals in a tightly massed composition. The undulating silhouettes of a drove of camels afforded him unusual 51 scope in this direction and he employs them to even better effect in the celebrated miniature in the same manuscript depicting the departure of 52 the pilgrim caravan for Mecca. It may be significant that this picture illustrates the immediately preceding maqāma. His choice of subject in the 32nd maqāma makes best sense in the context of his own rather than his putative patron’s personal preferences and thus implies a large measure of independence on his part. No doubt this had much to do with the nature of the text; had its content been religious, the artist would no doubt have been subject to severe restraints. Personal preferences may also have motivated his choice of the double-page spread for his compositions. The visual separation of one part of the story from another results in the reader being pulled up short and having to focus more carefully on the illustration. The link between the two pictures has to be made in his own mind; it is not made for him in full by the artist. It may even be a deliberately witty conceit. In most traditions of book illustration the 49

50

49  I.e. those at Paris (Bibliothèque Nationale, arabe 5847), London (British Library, Or. 1200) and Leningrad (Asiatic Museum 523). 50   It may be that Professor Grabar’s argument is really that no one manuscript is wholly a copy of another. That argument would admit of no contradiction, but it is surely more important to accept that copying did take place than to stress that it was never wholly slavish. It is not clear whether the “Instances where the same iconographic details or interpretations are found” are regarded by Professor Grabar as coincidence or copies or whether he regards them as belonging to some other category (e.g. similar “manners of illustration”). But even if one accepts the validity of the latter category, the fact that several manuscripts illustrate the same minor episode from a given maqāma remains unexplained unless one accepts that copying did take place. 51  He was not the first artist to respond to and exploit the rhythmic qualities of a file of camels; an equally famous example occurs in the 6th-century Vienna Genesis (E. Wellesz, The Vienna Genesis [London, 1960], pl. 3). 52  See R. Ettinghausen, Arab Painting (Geneva, 1962), pl. on 119. It is interesting to note that al-Wāsiṭī makes camels a major element in a closely linked group of pictures—those illustrating the 31st, 32nd and 33rd maqāmāt (ibid., pls. on 116, 117 and 119). It is as if he had suddenly struck a lucky vein.

336

artists have been conscious of the pitfalls of making the pictures selfcontained and thus potentially remote from the text. In Persian miniatures a favourite remedy for this disadvantage in the case of framed miniatures is of course to break the frame by an element of the pictorial design— in the case of battle scenes, for instance, lances commonly project into the text and in other cases it was standard practice to introduce panels of the text into the picture space. It is not surprising that a painter as original as al-Wāsiṭī should have exploited an unusual, humorous and 53 essentially visual method of linking text and illustration. More than that, he has achieved something of a visual pun—very much as if one figure 54 were pointing out to the other what was happening on the next page. Before leaving the subject of the 32nd maqāma one might perhaps add another possible explanation to those proposed by Professor Grabar for the choice of subject. He himself acknowledges that perhaps the prime function of the illustrations was to render the book more attractive. If one can accept this, and if one bears in mind that the 32nd maqäma is, as already stated, very long, it seems sensible for al-Wāsiṭī to have garnished it with illustration at some point. It is not clear from the lacunary published accounts of the Schefer Ḥarīrī whether this manuscript has any other pictures accompanying the 32nd maqāma or indeed how the miniatures for the 31st, 32nd and 33rd maqämāt are distributed. Thus the question of why al-Wāsiṭī should have placed an illustration at this particular point cannot be answered completely here. As suggested above, the text’s reference to a drove of camels may have caught his fancy, but it is just as likely that the layout of the manuscript made it desirable to place a picture—any picture—at approximately this point. The text did not help him much but he made the best of a bad job, and chose the episode in the story which lent itself most satisfactorily to illustration, even if it was incidental to the story proper. Presumably it was no more desirable in coffee-table books of the 13th than of the 20th century to allow a great many pages to pass by without an illustration. Pictures that were both numerous and 53  A parallel for this curiously literal approach to narrative is provided by the staircase reliefs leading to the Tripylon at Persepolis, where the reliefs flanking the staircase show courtiers themselves mounting the same steps as the visitor (A. T. Olmstead, History of the Persian Empire [Chicago, 1948], 180 and pl. XXVII). 54  A more pedestrian type of unity is achieved in a Mamlūk Maqāmāt manuscript in the British Library (Or. 9718): the progress of a caravan is shown successively on facing pages at ff. 119v and 120r (J. D. Haldane, An illustrated Ḥarīrī manuscript and its relevance to later Mamlūk painting [unpublished Ph.D. thesis, Edinburgh, 1973], 127 and 319, figs. 91–2).

RECENT WORK ON ISLAMIC ICONOGRAPHY

337

well distributed throughout the book could not be expected to maintain 56 a uniformly high degree of relevance to the text. Further investigation along these lines would be greatly aided by the publication of detailed catalogues of the existing illustrated Ḥarīrī manuscripts; a beginning has 57 been made with the Mamlūk Ḥarīrī Or. 9718 in the British Museum. Only with such information could one gain a satisfactory overview of Maqämät illustrations. It would then be possible to ascertain whether there was a pattern, either within manuscripts or within the whole genre, to the scenes selected for illustration, and whether the artists betrayed set reactions to certain themes, e.g conversation, carousal, crowd scenes and so on. More information is still needed, too, on how such a manuscript was planned— whether the text was written first, whether the artist was ever the scribe as well, and what factors governed the size of an illustration. Such topics can most fruitfully be discussed in the context of all the illustrated Maqāmāt texts; let us hope that the ambitious synthesis which Professor Grabar describes as largely complete will soon be published. The core of Professor Grabar’s paper is his identification of several different attitudes towards the text: literal, descriptive, interpretative and 58 visual. This is very valuable, and his discussion of each mode of seeing is challenging, especially his analysis of the interpretative element in alWāsiṭī, who emerges as one of the earliest satirists in Islamic painting. One might however take issue with the argument that the literal type of illustration “is more likely to arise as a result of already existing images than as a spontaneous generation from the text.” This is a rather circular theory: if the images are indeed literal, would one not expect their models to be literal too, all the way back to the original image which was 55

55  Information on how regularly the pictures are distributed throughout the text in the various Ḥarīrī manuscripts would be of great help in formulating precise conclusions as to the motivation of the artists in the choice of illustrations. Equally valuable, and for the same reasons, would be statistics about which maqāmāt had more than one illustration and—by extension—which were the most popular. 56  Cf. L. Golombek, “Toward a Classification of Islamic Painting”, in Islamic Art in The Metropolitan Museum of Art, ed. R. Ettinghausen (New York, 1972), 23. 57  See Haldane, op. cit., 127–70. See also ibid., 196–203, 211–13 and 239–42 for brief lists of the subjects in the Ḥarīrī manuscripts in London, Oxford and Vienna. Unfortunately, however, these subject lists are not correlated with the individual maqāmāt to which they belong. Perhaps this deficiency has been made good in the documentation assembled long since at the University of Michigan (O. Grabar, “A Newly Discovered Illustrated Manuscript of the Maqāmāt of Ḥarīrī”,AO V [1963], 97). 58  In his preamble he mentions five, but the last one is not given.

338

presumably inspired by the text? Moreover, one would expect the initial pictures accompanying a hitherto unillustrated text to be very thoroughly dependent on that text. It is only on the basis of such tentative, simplified and literal illustration that later artists can develop their more sophisticated modes. Professor Grabar is at his most persuasive in proposing that the Maqāmāt were popular not only for their literary virtuosity but because they were regarded as commentaries on contemporary society, and that it is this aspect which best explains their illustrations, aside of course from the function of the miniatures as embellishment for the book. These concluding remarks show his method at its best. Marie Lukens Swietochowski, in “Some Aspects of the Persian Miniature Painter in Relation to His Texts”, documents the movement from reliance on the text in the early 14th-century to a marginal regard for it in the later 14th, 15th and 16th centuries. But the gist of her argument is that the change was caused by the love of extraneous detail rather than by a re-interpretation of the text. It is only natural that, in breaking almost entirely new ground, she should raise as many questions as she answers. For example, it is certainly a fair generalisation that the subject matter of early 14th century paintings is pruned of superfluous detail. But why should later painters treat the text in cavalier fashion even though their own compositions were frequently simplified? Or again, in cases where the text describes two given rulers in very different terms, one must ask why the artist should portray them as alike. It is not a satisfactory explanation that the artist assumed that his readers would be familiar with 59 the text and for that reason refrained from becoming very particular. In view of the penchant for ornamental detail displayed throughout Persian miniature painting one would rather expect the basic story line to be embroidered visually. The sameness of, for example, enthronement scenes in the early 14th century can perhaps be explained most naturally by the profound conservatism of the artists. It was they themsleves, if anyone, who ignored the text they were illustrating. But Mrs. Swietochowski’s concept of “minimum iconography” is also valuable, since it offers a novel explanation of why certain parts of the text were illustrated repeatedly. It would be useful if the operation of this “minimum iconography” could be clarified by the publication of statistical tables, whether arranged 59  The fact that his patron might well be illiterate (Sanjar and Tīmūr spring to mind) ought rather to have encouraged the artist to embellish his work with extraneous detail.

RECENT WORK ON ISLAMIC ICONOGRAPHY

339

according to types of book or given periods of time. It would be even more valuable to investigate a series of versions of some passage of text in order to establish the fidelity of the artist to the written word. It is surely right of Mrs. Swietochowski to stress how certain scenes were treated in repetitive fashion, but it would be useful to investigate why it should have been these scenes in particular that were regularly depicted. The text of the epic would allow a far greater variety of scenes than most artists provide. Should the reason for this be sought in the preferences of the patron, in the innate conservatism of the artists or, as has been suggested above in the case of the illustrated Ḥarīrī manuscripts, in the need to space the pictures fairly regularly throughout the book, which would restrict the artist’s freedom of choice? It may even have been that the artists were drawn to certain scenes because they had developed a welltried formula for them which would ensure speedy completion. In the case of the mediocre artist, the availability of a traditional composition saved him the trouble of devising a new one—a conclusion which is relevant to Maqāmāt illustrations too. It is highly significant that, when in later centuries the “minimum iconography” was expanded, the result was actually to belittle the importance of the central scene. A violent death, the ostensible subject of the picture, may take up only a minor portion of the composition, and the rest of the area will be full of unrelated or neutral details (Fig. 7). It was not the practice for the artist to enlist such features in the turmoil of 60 the main scene or to make them serve as a commentary to that scene. Yet in many cases the text gave the artist ample scope to widen the emotional range of his picture by including other figures or by choosing to depict the moment immediately before or after a catastrophe. Mrs. Swietochowski cites certain exceptions to this widespread and rather pedestrian approach, notably in a deliberate parallelism between events far separated in time, so that they become visually related—a device used especially in cycles of murder and vengeance. It is a pity that such deliberate echoes are not

60   This generalisation does not apply with equal force to 14th-century painting. In some scenes depicting Rustam shooting Shaghād, for example, the tree itself seems to partake of the agony of death (e.g. the Jāmi‘ al-Tawārīkh in the Edinburgh University Library—see D. T. Rice, The illustrations to the ‘World History’” of Rashïd alDīn, ed. B. Gray [Edinburgh, 1976], 77 and pl. 19—and the “Demotte” Shāhnāma, illustrated in D. Barrett, Persian Painting of the Fourteenth Century [London, 1952], 14 and pl. 6. Cf. also Golombek, op. cit., 24–7.

340

commoner; they show the artist responding to the underlying themes of 61 the text rather than to its surface action alone. Finally, Mrs. Swietochowski convincingly postulates wall paintings as a source for the action-packed battle scenes of later Persian painting (Fig. 8); certain ceramics no doubt have a similar source. Such scenes appear to have been much more popular than symbolic scenes of single combat. 62 Depictions of fights between individual named heroes are not rare, and indeed the text often calls for them; but the single combat as a paradigm of war itself, as epitomised so frequently in the symbolic jousts of Sasanian reliefs, was not an image which appealed to Islamic painters. The neglect of this particular native tradition is all the more surprising in view of the wholesale survival of Sasanian themes in later Islamic iconography. “Cosmic Symbolism on Carpets in the Sanguszko Group”, by Schuyler van R. Cammann, catapults us far beyond the conventional domain of Islamic art historical enquiry. Perhaps by that very fact many of his readers will tend uncritically towards outright acceptance or outright rejection of his ideas. His basic thesis is that certain Şafavid carpets, and specifically the so-called Sangusko carpet (Fig. 9), express a highly complex symbolism, at once cosmological and religious, and that the elements of this symbolism are Chinese in origin. If his conclusions are even partially valid, the current approach to Şafavid carpets in general—not merely the Sangusko group—will require radical revision. Clearly there is room for discussion in Professor Cammann’s interpretation of details, but it may prove more helpful to begin by approaching the problem in general terms. First, is it justified to regard rugs as vehicles for religious symbolism; and second, can the particular group of motifs assembled on the Sanguszko carpet rank as specifically religious rather than multivalent? The paucity of preŞafavid rugs makes it impossible to ascertain whether earlier Islamic rugs ever bore religious symbols, but it is perhaps justifiable to doubt it. The few surviving early rugs seem to show no trace of such symbolism. Where religious motifs are found on rugs they are of a very straightforward kind—one has only to recall the countless prayer rugs showing miḥrābs, 63 mosque lamps and Qur’ānic quotations. Thus there was no reluctance 61  For an earlier example of this trend, cf. the scene from the “Demotte” Shāhnāma depicting Zāl and Bahmān embracing—a dramatic irony in view of the later feud between their families (see the exhibition catalogue, The Arts of Islam, ed. D. Jones and G. Michell [London, 1976], 329). 62 E.g. ibid., 341. 63  For typical examples, see K. Erdmann, Seven Hundred Years of Oriental Carpets, ed. H. Erdmann, tr. M. H. Beattie and H. Herzog (London, 1970), colour pl. XI

RECENT WORK ON ISLAMIC ICONOGRAPHY

341

on the part of the designers to use religious themes on rugs—the early 64 65 Christian, and perhaps even Umayyad, disinclination to permit sacred or imperial subjects to be trodden upon and thus symbolically dishonoured seems to have lapsed in later Islamic times. If, then, there was no Islamic prejudice against the use of religious themes on rugs, it remains only to consider whether the motifs on the Sanguszko carpet can be regarded as such. Corroborative material is sorely needed. Yet religious symbols of the kind discussed by Professor Cammann are not found in the very medium where one would expect them—book painting. While religious 66 themes were an established if rather limited genre in Islamic painting, they were normally treated in a straightforward, even literal, way rather than symbolically. The numerous versions of Muhammad’s night journey (mi‘rāj), a theme which might have been expected to lend itself to symbolic 67 treatment, illustrate this clearly. So do the pictures in a late 15th-century 68 Khawarnāma detailing the exploits of ‘Alī, one of the rare examples of an illustrated religious cycle, and the same can be said of the frequent 69 depictions of the Ka‘ba and of pilgrimage. The nearest approach to religious symbolism can be seen in the tendency to make figural designs out of calligraphic elements, a fashion which was particularly marked in 70 71 Ottoman and Qājār art. But even here the words, once decoded from the 72 design which they form, spell out sacred names and phrases, so that there can be no question of true religious symbolism in such cases. The obvious and pls. 19, 87, 89, 91, 115, 187 and 190. 64  E. Kitzinger, Israeli mosaics of the Byzantine period (New York 1965), 6–7. 65  E.g. a mosaic at Khirbat al-Mafjar, as analysed by R. Ettinghausen in his book From Byzantium to Sasanian Iran and the Islamic World (Leiden, 1972), 24. 66  See T. W. Arnold, Painting in Islam (Oxford, 1928), passim but especially 91– 122; and, more recently, J. M. Rogers, “The genesis of Şafawid religious painting”, Iran VIII (1970), 125–39. 67  This is all the more surprising in that this subject inspired extensive literary treatment. Yet no illustrated Mi‘rājnāma exploits abstract symbolism. 68  For illustrations cf. B. Gray, Persian Painting (Geneva, 1961), plates on 105–7; and B. W. Robinson et al., Persian and Mughal Art, a catalogue of an exhibition held at Colnaghi’s (London, 1976), 24 and plate on 106. 69  R. Ettinghausen, “Die bildliche Darstellung der Ka‘ba im islamischen Kulturkreis”, Zeitschrift der Deutschen Morgenländischen Gesellschaft N.F. 12 (Band 87) [1934], 111–37; see especially Abb. 7–8. 70  A. Schimmel, Islamic Calligraphy (Leiden, 1970), 31 and pl. XLVIIa. 71  Ibid., pls. XLVI and XLVIIb; Y. Ghulam, Introduction to the art of Arabic calligraphy in Iran (Shiraz, n.d.; c. 1971), 95–7. 72  These deliberately obscure inscriptions deserve more detailed analysis, especially as regards the functions which they served. It would be interesting to know, for example, how often puns occur in such inscriptions or how often the

342

conclusion is that Islamic painting did not develop a pictorial vocabulary for symbolically expressing religious ideas. This conclusion is important in view of the dependence of Safavid carpet designs on the themes and pictorial types of miniature painting. The contrast with Islamic literature, in which religious symbolism abounds, could scarcely be more marked. But it is precisely the fact that religious symbolism is so widespread, in Persian literature especially, that should make it possible to produce the indispensable literary parallels for Professor Cammann’s theories. In default of such written evidence one is forced to conclude that if the Sanguszko carpet and related rugs do indeed present a religious iconography, it was one which was developed purely for carpets, and which renounced the images of love, wine and natural phenomena favoured by Ṣūfī poets. This is hard to credit. Not only that, but one has to believe that the iconography used on carpets relied very largely on Chinese—that is, non-Islamic— concepts and symbols to convey Islamic notions of paradise. Yet this theory too runs into difficulties, for traditional exegesis in Islam painted a very specific picture of paradise, basing itself on the details given in the Qur’ān and ḥadīth and stressing such features as gardens, springs, palaces, 73 houris and music. Moreover, numerous carpets exist which contain both figural and literary references to paradise—references which are couched 74 in unambiguously Islamic terms. These remarks suggest that the evidence of the carpet could, on the contrary, be interpreted more easily to suggest that Chinese images were being used to propound an essentially Chinese view of the hereafter, but in that case the great questions would be how such a carpet could have been produced under court or even royal patronage, and what it could have meant to its patrons. To assume that the designer manipulated these images in a neutral, uncommitted way while at the same time being conversant with their deeper resonances—so that the whole carpet is a kind of rebus—might be even closer to the truth, and would raise fewer problems. That paradise should be the theme of such carpets is not in itself unlikely—the old Persian equation of the garden and, in general, 75 the fertile landscape, with paradise would make this an unexceptionable letters forming an image do in fact spell out the name of that image. On this whole question see E. Kühnel, Islamische Schriftkunst (2nd ed., Graz, 1972), 77–9. 73  See L. Gardet’s article Djanna in EI2, II, 448–9. 74  Erdmann, op. cit., 72–3, 164; pl. III; and figs. 78, 81 and 206. 75  See D. N. Wilber, Persian Gardens and Garden Pavilions (Rutland, 1962), 19, 39, 42, 44, 46–7 and 51.

RECENT WORK ON ISLAMIC ICONOGRAPHY

343

interpretation. But in his desire to make the multiple images of the carpet consistent Professor Cammann is led to propose complex meanings for them all, even for those which turn up on countless other carpets of the period as stock decorative motifs (e.g. the hunters, the ass, the phoenix and the dragon). Moreover, such elements abound in earlier Persian art, and in these earlier works there can be no question of Islamic religious symbolism. Numerous Sasanian dishes show the Shāh shooting onagers or 76 similar creatures; Bahrām Gūr’s very name is a reminder of his prowess at hunting the wild ass and for centuries Persian miniature paintings had shown him thus; the phoenix and the dragon were downgraded to serve 77 as a repetitive dado decoration in Abāqā’s palace at Takht-i Sulaimān; and conflicts between dragons or between dragons and phoenixes were 78 a favoured cliché of 16th-century painting and for that matter earlier 79 Islamic art. It seems intrinsically unlikely that motifs of this kind, which had long been neutral, should suddenly acquire portentous symbolic overtones. It would be no disgrace to the Sanguszko carpet if some of the elements of its design had a purely ornamental purpose. But in Professor Cammann’s exhaustive analysis even the t’ao-t’ieh masks of the border, whose original function recalls that of Ragnarok, the apocalyptic world-devouring wolf of Norse mythology, are seen as apotropaic deities 80 rather than decorative clichés. For him the whole carpet is a manifesto of Ṣūfī ideas, but whether concepts about the Sun Gate, the Three Worlds, the primordial Sunbird and the outer Sky Door really loomed large with Islamic mystics remains to be proved. It seems preferable to interpret the undoubtedly Chinese features in the decoration as elements which have lost their original significance in the process of being transferred 76  Grabar and Carter, Sasanian Silver, pl. 3; A Survey of Peṛsian Art from Prehistoric Times to the Present, eds. A. U. Pope and P. Ackerman (London and New York, 1939), pls. 214 and 229a. 77  E. and R. Naumann, “Ein Köşk im Sommerpalast des Abaqa Chan auf dem Tacht-i Sulaiman und seine Dekoration”, Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann, eds. O. Aslanapa and R. Naumann (Istanbul, 1969), 54–8, colour plate and Abb. 3–4, 11–12. 78  E. J. Grube, “Miniatures in Istanbul Libraries. I. A Group of Miniatures in the Albums Hazine 2147, 2153 and 2162 in the Top Kapi Saray Collection and some related material”, Pantheon XX (1962), 213–26 and pls. 1–9, 14. 79  See the celebrated Anatolian rug in Berlin, datable c. 1400 (Erdmann, op. cit., 17, colour plate on cover and pl. 1). 80  E. Newman-Perper has recently re-interpreted this motif in a paper entitled “A New Approach to the ‘T’ao-Tieh Problem’”, delivered at the VIIth International Congress of Iranian Art and Archaeology in 1976.

344

from one culture to another. This neutralisation of significant motifs has 81 characterised Islamic art from the time of the Umayyads onwards and Persian examples of this trend abound; one need only consider the Chinese creatures, flowers and shorthand symbols for landscape that invaded the Iranian repertoire under the Īlkhāns. Their indiscriminate use is the surest indication that their original significance was quickly lost in the Islamic milieu. This precedent makes it doubly unlikely that the Chinese motifs which dominate the Sanguszko carpet should have managed to retain their former symbolic associations after a lapse of centuries. To take a single example, the white circular void placed at the very centre of the carpet is interpreted by Professor Cammann as the Sun Gate, and thence as a reference to Chinese cosmology. But similar motifs occur earlier in Islamic art, also with symbolic intent, and their context rules out the possibility of Chinese influence. Thus at a late Umayyad palace near Jericho, Khirbat al-Mafjar, the largest of the floor mosaics depicts a whirling spiral, with the central spot left blank—the visual equivalent of the point of rest in 82 the centre of a whirlpool. Since this spot is directly below the apex of the main dome of the building, and since the Umayyads were familiar with the 83 symbolic equation of the dome with the heavens, it is logical to regard this central spot, the omphalos of the whole building, as the equivalent of the crown of the dome, a kind of oculus where inner and outer worlds meet. Buildings like the Pantheon and the Ravenna baptisteries embody similar ideas. The corresponding feature of the Sanguszko carpet can likewise be interpreted as a reference to the heavens; this would not conflict with the dominant paradisal theme. There is, in short, no need to introduce Chinese beliefs in order to establish the symbolic connotations of the motif. Professor Cammann’s brief comments on the birds usually mis84 identified as “peacocks” are very valuable and it is a pity that he did not  The process has been traced in illuminating detail in the geometric decoration of Iraqi minarets. See ‘Abdal-Raḥmān al-Gailānī, The origins of Islamic art and the role of China (unpublished Ph.D. thesis, Edinburgh, 1973), chapters II–IX. 82  Illustrated in R. W. Hamilton, Khirbat al Mafjar. An Arabian mansion in the Jordan valley (Oxford, 1959), pls. LXXXV and XCVIIIa. 83  As the calidarium fresco at Quṣair ‘Amra proves; see Creswell, op. cit. I/2, 399–400, 424–40, figs. 477–9 and pl. 76a. 84  Possibly because of their important role as a symbol of immortality in Christian iconography. But it is unlikely that there is any connection between the Christian and the Islamic traditions in this particular case, for in Islam, to quote G. M. Meredith-Owens, “the peacock is a symbol of cupidity and man’s desire for worldly life” (Persian Illustrated Manuscripts [London, 1965], 28, n. 12). 81

RECENT WORK ON ISLAMIC ICONOGRAPHY

345

expatiate on this topic. He is right to draw attention to the composite nature of these birds, and the alternative term he suggests—“Bird of Paradise”—is an apt and elegant coinage. As he says, these birds occur in Muslim shrines. “Persian shrines” would be more accurate, though it would indeed be strange if they were not also found outside Iran. This whole topic merits detailed investigation. It is, first of all, noteworthy that in later medieval times the Iranian world above all should use such 85 motifs in a religious context. They raise in the most pointed way the whole issue of iconoclasm in later Islamic art. Clearly the attitudes to this subject which were tolerated if not prevalent in the Iranian world in 86 Timūrid and Şafavid times were not those of early Islam. It was not a matter of furtively transgressing theological orthodoxy, for these animals and birds occur at the entrances to some of the principal mosques and 87 shrines of the land—the dynastic shrine at Ardabīl and the Masjid-i Shāh 88 at Isfahān among them. Moreover, the practice seems not to have been a narrowly sectarian one, for while the two sites just mentioned are Shī‘ite, 89 90 others such as the mosque at Anau, the Shīr Dār madrasa at Samarqand 91 and the shrine at Turbat-i Shaikh Jām (Fig. 10) were built for Sunnīs. It is significant that in all the cases mentioned the creatures adorn the entrance rather than the interior of the mosque or shrine; this strongly suggests an apotropaic function. Similar decoration is found on buildings in Anatolia  One might suggest that this motif could be a reminiscence of the homa, the sacred bird of pre-Islamic Iran, and that it could be related also to the sīmurgh, which became a vehicle for mystical symbolism in medieval Persian literature (e.g. ‘Aṭṭār’s Manṭiq al-Ṭair). 86  See the bibliography on this subject assembled by K.A.C. Creswell in A Bibliography of the Architecture, Arts and Crafts of Islam to 1st Jan. 1960 (Cairo, 1961), cols. 979–82; and idem, A Bibliography of the Architecture, Arts and Crafts of Islam. Supplement, Jan. 1960 to Jan. 1972 (Cairo, 1973), cot. 293. 87  F. Sarre, Denkmäler Persischer Baukunst (Berlin, 1901-10), Taf. XXXIII–XXXIV. 88  S. P. Seher-Thoss, Design and Color in Islamic Architecture (Washington, 1968), pl. 81. 89  E. Cohn-Wiener, Turan. Islamische Baukunst in Mittelasien (Berlin, 1930), pl. LXXIII. For an example from a Ṣafavid madrasa, cf. the dragons in the spandrels of the portal arch of the Madrasa-yi Khān at Shirāz (H. Khoubnazar and W. Kleiss, “Die Madrasa-yi Khān in Schiras”, AMI N.F. VIII [1975], 273 and Taf. 57 .2). 90 D. Brandenburg, Samarkand. Studien zur islamischen Baukunst in Uzbekistan (Zentralasien) (Berlin, 1972), 170–2 and Taf. 98. 91  Though the portal īwān at Turbat-i Shaikh Jäm was redecorated by Shah ‘Abbās, who was of course Shī‘ite . Legend avers that he only did so when evidence came to light that the eponymous saint was not a heretic (L. Golombek, “The Chronology of Turbat-i Shaikh Jäm”, Iran IX [1971], 35). 85

346

and further west in the 12th and 13th centuries. In Anatolia, these motifs 92 appear on both religious and secular buildings; but in the Arab world the 93 context is almost always secular. The favourite site for such motifs was 94 the city gate. In all these buildings the function of the creatures seems to have been to ward off evil. The choice of guardian creature must also be significant, for the architectural portals bear the very birds and animals which occur on the Sanguszko carpet: “Birds of Paradise”, dragons, 95 and sun-faces attached—again significantly—to tiger bodies. The latter hybrid began to acquire almost heraldic status in the later Şafavid and Qājār periods, finally becoming so much a national emblem that it is found on early Persian postage stamps. The precise meaning which was attributed to these symbolic creatures can only be established on the basis of detailed literary research. At this stage it would be premature to deduce, from the appearance of the same motifs on religious architecture and on carpets of the Sanguszko type, that such carpets were intended to have a primarily and specifically religious function. Professor Cammann never hazards an opinion about the purpose which the Sanguszko carpet was intended to serve, but it seems that—by implication at any rate—he does not think that it was made for a place of worship. For among the carpets he discusses he attributes this function to only one piece, the Anhalt carpet, and it is vital to note that this rug is devoid of figural motifs. In this connection it may be important to remember that in Iran the creatures found in sacred architecture are relegated to the exterior alone, though this generalisation does not apply   Cf. a series of articles by G. Öney, conveniently tabulated in Creswell, Supplement, cols. 189 and 323–4; cf. also eadem, “Anadolu Selçuk Mimarisinde Avcı Kuşlar, Tek ve Çift Başlı Kartal,” Malazgirt Armağanı (Ankara, 1972), 139–72. 93   Cf. however, the miḥrāb of the mosque of Nūr al-Dīn at Ḥamā; this part of the mosque is datable between 626/1229 and 642/1244 (E. Herzfeld, “Mshattâ, Ḥîra und Bâdiya. Die Mitteländer des Islam und ihre Baukunst”, Jahrbuch der Preuszischen Kunstsammlungen 42 [1921], 135 and Abb. 9). Another exception may be the building at Sinjar (G. Reitlinger, “Mediaeval antiquities west of Mosul”, Iraq V [1938], 151–3 and figs. 14–17; see also the extensive discussion of this monument in T. J. al-Janābī, Studies in mediaeval Iraqi architecture [unpublished Ph.D. thesis, Edinburgh, 1975], I, 308–11 and II, pl. 164). 94   Notable examples are the gates of ‘Amādiya, al-Khān near Mosul, Baghdād and Aleppo. The fashion even penetrated to the far west of the Islamic world, as the gate of the Kasba of the Udayyas at Rabāṭ shows. 95   The sun with a human face is not in itself a rare motif; it is found commonly in Western medieval art (especially in scenes of the Crucifixion) and in Islam too (e.g. in Mamlūk painting). It is the addition of the tiger’s body that makes the motif distinctively Iranian. 92

RECENT WORK ON ISLAMIC ICONOGRAPHY

347

to lustre tiles. Could the themes found on carpets and portals alike have been part of an iconography which was neither specifically terrestrial nor specifically geared to Islam as a religion, but which expressed an ancient and popular cosmology? If so, it would be an extension of that fascination with astrology and kindred subjects which has characterised the people of the Middle East for millennia. The images on mosque or shrine portals, and on carpets, would in that case command widespread acceptance even among the religious classes because they reflected popular beliefs and preoccupations. But they would not be regarded as appropriate affirmations of Islamic orthodoxy, and hence would not be allowed within 97 the mosque or shrine. The depiction of certain specific creatures might have been permitted on the portals of reigious buildings partly because they warded off evil but principally because they connoted paradise. This association with the hereafter, coupled with their location outside the building, alone allowed them to escape the iconoclastic ban. These reflections have a direct bearing on the function of the Sanguszko carpet and others of its type. Since its images are couched in the idiom of rugs rather than of book painting or literature, and since they appear to lack satisfactory parallels with literary or religious texts, it is possible that the carpet is no more than a tour de force. As such it displays exquisite refinement and subtle calculation, and is worthy to challenge the penetration of a connoisseur. Such a man would be familiar with the iconography employed; he would catch its nuances and appreciate the hierarchical order in which the various elements were disposed. But one need not assume that he felt a sense of commitment to the system of beliefs expressed by these images. He would not need to be a devotee of Ṣūfism to understand them and, as has been argued, he would be unlikely to see Ṣūfï meanings in them. Thus there is no need to assume highly unorthodox patronage for such carpets. Their splendour makes it likely enough that their patrons were royal, and there was nothing in their content that need have occasioned embarrassment to such patrons. The 96

96  O. Watson, “The Masjid-i ‘Alī, Quhrūd: an architectural and epigraphic survey”, Iran XIII (1975), 67 and n. 40. 97  In buildings of minor and largely local importance this rule would often be stretched, as the Imāmzāda Panj Tan at Lāhijān shows (Survey, pl. 509C; the caption mistakenly identifies the building as the town’s Masjid-i Jāmi‘). It is noticeable, however, that such figural paintings usually depict Shī‘ite themes and thus were presumably not regarded as offensive. Typical examples of such paintings may be found in the undated but probably post-Safavid Imāmzāda ‘Abbās near Āmul (figs. 11–13).

348

fact that such carpets ceased to be made after the reign of Shāh ‘Abbās can be explained not only as a vagary of fashion but also as an expression of that progressive decline which afflicted Ṣafavid art from the early decades of the 17th century. It is clear enough from the discussion so far that there is room for argument in Professor Cammann’s interpretation of the motifs on the Sanguszko carpet. But it must also be emphasised that this article is, with the possible exception of Professor Soucek’s, the most stimulating 98 in the whole volume. Professor Cammann’s other publications show how resolutely he has been breaking new ground—and in the context of Islamic art in general, the role of China is one of the last central topics which is still virtually unexplored. It is certain that future studies will reveal a hitherto unsuspected dependence of Islamic art on China in the realm 99 of iconography in particular. In every case the crucial question will be how much of the meaning was transmitted with the form. No-one could doubt, after considering Professor Camman’s painstaking analysis, that the Sanguszko carpet derives the majority of its images from Chinese sources. To have demonstrated that so decisively is itself a noteworthy achievement. But opinions will no doubt continue to differ as to how far the Islamic artist understood the significance of the Chinese material he used. Only when an approximate answer has been provided for that question can one begin to distinguish between the copying, the adaptation and the re-interpretation of Chinese motifs in Islamic art. Professor Cammann’s excursus into the cosmic symbolism which he sees expressed on a Persian carpet exemplifies the catholic scope of this Festschrift. But this breadth of interest is only part of the book’s special appeal; for its particular focus is the role of literary evidence in recovering the context of a work of art. Thus one writer after another has managed to clarify the artist’s use of his sources. There are lessons here for every Islamic art historian.

 A selection of these are listed in footnotes 1, 7, 10, 13, 25 and 58 of the article under discussion. 99  For a detailed demonstration of this see the Ph.D. thesis of Dr. al-Gailānī cited above, n. 79. 98

XVI The Islamic Reworking of the Sasanian Heritage: Two Case Studies

F

or the best part of a century now, scholars of Islamic art have busied themselves with the fascinating problem of how the Muslims came 1 to fashion their own distinctive visual vocabulary. The problem is indeed an absorbing one, not only from the viewpoint of the art historian as a detective seeking to trace the origins of this or that motif, but also as an intellectual exercise. It poses in a searching way the question of what constitutes Muslim identity, and — as in similar surveys of Islamic law or the religion of Islam — its findings therefore have a relevance to Islamic society which extends far beyond the original scope of the enquiry. As it happens, this investigation of the origins of Islamic art has nearly 2 always been undertaken in the context of Umayyad art. Yet while this is sensible enough, since early Islamic art is broadly speaking co-extensive with the period of Umayyad power (i.e. 661–750), it does have its  See for example Ettinghausen, Byzantium to Sasanian; O. Grabar, The Formation of Islamic Art, London and New Haven, (revised and enlarged ed.) 1987, especially p. 4. 2  The classical instance, in that it progresses beyond the mere assembling of data to confront the patterns of borrowing, is the seminal article of E. Herzfeld, ‘Die Genesis der islamischen Kunst und das Mshatta-Problem’, Der Islam I, 1910, pp. 27–63 and 105–44. This is now available in English: ‘The Genesis of Islamic Art and the Problem of Mshattā’, tr. F. K. M. Hillenbrand with assistance from Jonathan M. Bloom, in Jonathan M. Bloom (ed.), Early Islamic Art and Architecture in L. I. Conrad (general editor), The Formation of the Classical Islamic World, XXIII, Aldershot and Burlington, Vt., 2002, pp. 7–86. For an up-to-date assessment of this problem of origins, see Talgam, Origins, I, pp. 47–120. 1

350

disadvantages. It tends to imply – wrongly – that Islamic art entered a new and maturer stage of development with the accession of the ‘Abbasids in 750. In actual fact, of course, the process was a seamless one and absolute dates are only useful as general indicators. Thus material which postdates 750 can be entirely relevant to an analysis of the character of early Islamic art. More seriously, close focus on the Umayyad period has brought in its wake an inevitable distortion in the modern view of early Islamic art. This is because virtually all surviving Umayyad art is to be found in Syria, a land whose dominant culture at the time of the Islamic conquest was Christian with Graeco-Roman roots. It follows that the classical origins of early Islamic art have tended to monopolise the attention of scholars, to 3 the detriment of ideas derived from the ancient Near East. The present paper is a modest attempt to remedy this deficiency. That there were significant Persian elements in the earliest Islamic art 4 has long been recognised. They are particularly in evidence in the closing 5 years of the Umayyad caliphate. Yet it was inevitable that Persian ideas imported into Syria, and thus far from their native soil, could not flourish as they could at home. They were not fully understood, and in any event had to compete with the very well established and classically-inspired art of Syria, rooted as it was in the Mediterranean world. Umayyad art is therefore incapable of giving a true picture of how Sasanian art adapted itself to Islamic modes. That process must be traced in the art of Iran itself.  Indeed, Jean Sauvaget went so far as to propose that Sasanian art, far from being a key creative influence on Islam, was itself largely derived from Hellenistic art (‘Remarques sur l’art sassanide. Questions de méthode à propos d’une exposition’, Revue des Etudes Islamiques, XII, [1938], especially pp. 125–6 and 130–1). One may set against this view the remarks of A. Northedge, Studies on Roman and Islamic ‘Amman. Volume I: History, Site and Architecture. British Academy Monographs in Archaeology, III, Oxford, 1992, pp. 100–4; M. Rosen-Ayalon, ‘Themes of Sasanian Origin in Islamic Art’, JSAI IV, 1984, pp. 69–80; and Talgam, Origins, I, 48–68. The topic of course extends still further back in time: M. C. Root, ‘From the Heart: Powerful Persianisms in the Art of the Western Empire’, in H. SancisiWeerdenburg and A. Kuhrt (eds.), Asia Minor and Egypt: Old Cultures in a New Empire. Achaemenid History, VI, Leiden, 1991, pp. 1–29. 4  For example, in the mosaics of the Dome of the Rock: O. Grabar, ‘The Umayyad Dome of the Rock in Jerusalem’, AO, III, 1959, pp. 48, 50–52. 5  D. Schlumberger, ‘Deux fresques omeyyades’, Syria XXV/1–2, 1946–48, pp. 88–93, 97–8; Ettinghausen, Byzantium to Sasanian Iran, pp. 28–33, 36–43; R. Hillenbrand, ‘Islamic Art at the Crossroads: East Versus West at Mshatta’, in A. Daneshvari (ed.), Essays in Honor of Katharina Otto-Dorn, Malibu, 1981, pp. 63–86 (repr. in R. Hillenbrand, Studies in Medieval Islamic Architecture I, London, 2001, pp. 114–59). 3

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

351

A first question, and a very obvious one, is why it was felt necessary to make any changes at all. After all, the images in question had served the Sasanian monarchs well, and consistent re-use over a period of centuries had streamlined them to such a degree that they were scarcely susceptible of further improvement in that context. Form and function dovetailed 6 7 perfectly. Hunting (fig. 1) and enthronement are the classic examples. Naturally there was no question of overtly religious themes maintaining their currency under the new dispensation, but such themes in any case formed only a small minority of those in regular use. In the matter of the various depictions of royalty, including its symbols and its pursuits, which formed the staple of Sasanian iconography, the Muslims after a brief initial hesitation enthusiastically embraced the ethos which underpinned these visual images, and thus had no difficulty in accepting the images themselves. In much the same way, of course, they took on board many Sasanian practices and ideas in fields as diverse as administration, court 8 protocol, military affairs, costume and literature. If the religion of Islam made relatively little impact on this inherited legacy of symbols, why then were any changes made in that legacy? The answer lies in the passage of time. As the Sasanian world receded from remembered history into myth, it was inevitable that the physical details of Sasanian court life – the very details which imbued Sasanian iconography with life and meaning – should gradually be forgotten. Their outward form might be preserved for a space, but their raison d’être would be lost; and, once that was lost, it was only a matter of time before the form itself would undergo change, becoming ever less recognisable in the process. It was equally inevitable that new meanings would attach themselves to forms which were still obscurely felt to be in some way significant. The rock-cut grotto of Taq-i Bustan in Western Iran is a perfect case in point; the Sasanian connection was preserved, but the monument’s actual association with the return to power of Khusrau II in the 590s  K. Erdmann, ‘Die sasanidische Jagdschalen’, Jahrbuch der preuszischen Kunstsammlungen LVII 1936, pp. 193–231; Grabar, Sasanian Silver, pp. 47–55; and Harper (with a technical study by P. Meyers), Silver Vessels, 1981, pp. 40–98 and pls. 8–32, 37–38. 7  H. von Gall, ‘Entwicklung und Gestalt des Thrones im vorislamischen Iran’, AMI, N.F. IV, 1971, pp. 207–35; and P. O. Harper, ‘Thrones and Enthronement Scenes in Sasanian Art’ Iran, XVII, 1979, pp. 49–64. 8  S. Shaked, ‘From Iran to Islam. Notes on some themes in transmission’, JSAI, IV 1984, pp. 31–67; and idem, ‘A facetious recipe and the two wisdoms: Iranian themes in Muslim garb’, JSAI IX, 1987, pp. 24–35. 6

352

was overlaid by the romantic confections of Khusrau, Shirin and Farhad 9 as recounted by Nizami more than half a millennium later. Myth proved more potent than fact. Common sense alone is sufficient assurance that a transition from Sasanian to Islamic modes must have taken place in the visual arts; but, 10 for some reason, its details have rarely been plotted. This is indeed a Dark Age in Islamic art history. The objects themselves are sufficiently well known in the sense that they have been described and published; but their 11 context and chronology remain somewhat nebulous. Thus the successive stages in the process of transition are not easily identified. Chronology is a particularly thorny problem. Virtually none of the buildings, ceramics, textiles and metalware objects which are conventionally attributed to the Iranian world between the Islamic conquest and the beginning of the 11th century are closely datable – say to a given decade. Most of them cannot even be dated with confidence to a given century. It is little wonder that, from the art-historical point of view, they still exist in a kind of limbo. 12 13 Textiles and metalwork in particular compound these difficulties by the 9  P. P. Soucek, ‘Farhad and Taq-i Bustan: The Growth of a Legend’, in P.J. Chelkowski (ed.), Studies in Art and Literature of the Near East in honor of Richard Ettinghausen, New York and Salt Lake City, 1974, pp. 27–52. 10  There are of course a few exceptions. Apart from Soucek’s article (see previous note), one might cite Sir Thomas W. Arnold, ‘The Survival of Sasanian Motifs in Persian Painting’, in Studien zur Kunst des Ostens (Strzygowski Festschrift), Vienna and Hallerau, 1923, pp. 95–7; idem, Survivals of Sasanian & Manichaean Art in Persian Painting, Oxford, 1924; F. Sarre, ‘Die Tradition in der islamischen Kunst’ in Mémoires: IIIe Congrès International d’Art et Archéologie Iraniens, Moscow and Leningrad, 1939, pp. 227–8; and Ettinghausen, ‘Traditionalism’, pp. 88–110. Cf. also R. Hillenbrand, ‘Notes on the Symbolism of the Rayed Nimbus in Early Islamic Art’, Cosmos II, 1988, pp. 1–52. 11  An isolated attempt to assess the art of this period was made two generations ago by E. Kühnel, ‘Die Kunst Irans zur Zeit der Buyiden’, Zeitschrift der deutschen morgenländischen Gesellschaft CVI, (N.F. 31), 1956, pp. 78–92. 12  Compare, for example, D. Shepherd, ‘The authenticity of the Rayy silks’, Bulletin de Liaison du Centre International d’étude des textiles anciens XXXIX–XL, 1974, with S. S. Blair, J. M. Bloom and A. E. Wardwell, ‘Reevaluating the Date of the ‘Buyid’ Silks by Epigraphic and Radiocarbon Analysis’, AO XXII, 1992, pp. 1–43. 13  It is interesting to note, for example, the celebrated objects of precious metal silently excluded from Boris Marshak’s survey of early medieval silver in the Iranian world (Marshak, Silberschätze). Cf. the items banished to Appendix III of E. Atil, W. T. Chase and P. Jett, Islamic Metalwork in the Freer Gallery of Art, Washington, DC, 1985 – a ‘salon des refusés’ of some former stars of Islamic metalwork – to which should be added the ostensibly pre-Islamic objects (nos. 43–7) which figure at the end of the catalogue in A. C. Gunter and P. Jett, Ancient Iranian Metalwork. In the Arthur M. Sackler Gallery and the Freer Gallery of Art,

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

353

still unresolved problems of provenance and even authenticity which they pose, and the same is true for many types of ceramics. The pace of change is wholly unpredictable and varies dramatically from one medium to the next: work in an unimpeachably Sasanian idiom may postdate that dynasty by three or four centuries, while broad parodies of Sasanian art may be 14 contemporary with the real thing. Such echoes or outright caricatures are 15 to be found across broad gulfs of space and time. There is no easy way out of these difficulties, but one method recommends itself if only because it has not often been systematically applied to this material. It is to investigate the iconography of the images depicted on these various objects with a view to identifying the diagnostic changes which are introduced. Of course this method is not fool-proof. The ability and knowledge of the artist, to say nothing of his originality, will affect what he makes of the material he has inherited. Nevertheless it ought to be possible to date a given work of art early or late in the process of transition, and to assess whether it is still psychologically dependent on pre-Islamic art in most respects or whether it contains pointers to the more fully Islamic styles which had come into being in most media by the 11th century. Even so, it must be conceded that an image could become a remote reflection of its original source not simply by virtue of the passage of time but equally well because it was produced in a provincial atelier, or indeed in Byzantine territory – a provenance suggested for many textiles 16 with Sasanian-related themes. But whether it is chronology or provenance

Washington, DC, 1992. On the problem of forgeries, see ibid., pp. 20 and 22, n.28 (for a recent brief bibliography of relevant scientific studies). 14  Such as a Coptic ivory that brutally (but memorably) simplifies the frontal pose of the Sasanian seated monarch, or an Egyptian textile replete with partially digested and mildly comic versions of such familiar Sasanian themes as the royal crown, the seated monarch or the so-called Parthian shot – see Ghirshman, Parthians and Sassanians, pls. 402 and 289 respectively. 15  For representative examples see Y. I. Smirnov, Vostochnoe Serebro. Atlas Drevnei Serebryanoi i Zolotoi Posudi Vostochnago Proiskhodzhdeniya, Naidennoi Preimushchestvenno v Predelakh Rossiiskoi Imperii, St. Petersburg, 1909, nos. CLV–CLXI; for a conspectus of the Sasanian legacy in unexpected places, see A. Grabar, ‘Eléments sassanides et islamiques dans les enluminures des manuscripts espagnols du haut Moyen Age’ in E. Arslan (ed.), Arte del Primo Millennio. Atti del IIo Convegno per lo studio dell’ arte dell’ alto medio evo tenuto presso l’università di Pavia nel Settembre 1950, Viglongo, n.d., pp. 312–19 and Tav. CLXFV–CLXXVI (see especially p. 314 and pl. CXVI), and Ghirshman, Parthians and Sassanians, pp. 283–336 16  Such attributions are commonplace in O. von Falke, Decorative Silks, London, 1936.

354

that distorts the metropolitan model, the key point is that distortion does occur, and it is instructive to investigate the form that it takes. One further caveat is worth making at this stage. No ingrained belief in the necessity of ‘progress’ or of ‘decline’ (however those terms are to be defined) underlies the comments that follow. When the art of early Islamic Iran is studied as a whole, it becomes clear that two processes are at work independently of each other: the gradual decay and transformation of Sasanian modes, and the evolution of new forms of art with no Sasanian antecedents. The latter category, which includes art of Turkic and 17 steppe origin, as well as works of art which seem to be entirely original 18 inventions, does not figure in the discussion which follows. But that is not to downgrade its importance. In order to drive home the point that the process of transition operated throughout the visual arts, the examples examined in this paper will be taken from two different media – ceramics and metalwork. It is no accident that these are the two media to which Géza Fehérvári has devoted most of his career and publications, and I hope that he will forgive me for 19 trespassing on his turf. The examples chosen will also each illustrate a separate iconographic theme. It would be simple enough to include architecture as well – the mosques of Damghan, Fahraj, Na’in or Niriz would lend themselves admirably to such an investigation – but there is little need to do so. Scholars have been alert to the Sasanian heritage in the early Islamic architecture of Iran ever since the 1930s and the subject has 20 had a good airing. It is interesting, however, in the context of this paper, to note that such architecture tells the same story as the other arts, in that old forms are given new functions and therefore new meaning. Yet the very familiarity of those forms can disguise the changes that have taken place. Thus the non-directional fire-temple or chahar taq is turned into a mosque sanctuary by the simple expedient of blocking up the side most 17  Cf. the material assembled by Otto-Dorn, ‘Bildgut’, pp. 1–69; and eadem, L’art de l’Islam, pp. 77–8, 81–90. 18  E.g. the Gunbad-i Qabus. 19  I was both honoured and delighted to receive the invitation to contribute to this Festschrift. For nearly forty years Géza has been a model of approachability and kindness – I well remember how he encouraged me at an early stage of my doctorate – and his openness and generosity of spirit has inspired a quite special affection in his students and colleagues alike. This paper is offered in fond recollection of many happy hours spent in his company. 20  For the most recent extended discussion of this theme, see R. Hillenbrand, ‘’Abbasid mosques in Iran’, Rivista degli Studi Orientali LIX, 1985, pp. 175–212.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

355

nearly aligned to Mecca and placing a mihrab in the centre. Outwardly there is very little change; but in actual fact the building has undergone a total transformation. Similar remarks could be made à propos the dome, the iwan, the squinch and even applied ornament. Architecture too has its iconography, although Islamic art historians have been tardy in recognising 22 the fact; and that iconography responded to the seismic changes which occurred as Iran shifted not merely its political and religious allegiance but also its social structure from Sasanian patterns to Islamic ones. So much for theory. How far do the objects themselves corroborate it? In the first example to be discussed the medium is pottery with glazed figurative decoration from the eastern Iranian world. In this geographical region, glazed pottery produced on this copious scale and with such varied decoration is unprecedented – one major relevant and local pre-Islamic exception notwithstanding. This same exception is no doubt an accident of survival, but was presumably one of many in its own time. It is perhaps significant that this piece, a gigantic vase or amphora generally attributed to 6th-century Merv (fig. 2), should bear scenes of courtly life – a cavalier, 23 a reclining prince, musicians and female attendants. Such scenes were to be the staple of the figurative pottery of Samanid Nishapur. Yet while it is plausible enough to suggest that the almost man-sized vase from Merv was made for a courtly milieu – its ambitious scale and accurate iconography point in this direction – the buff ware of Nishapur is a different case altogether. It survives not in a few isolated examples but in its hundreds. This fact, when taken together with the humble material of which it is 24 made, points not to courtly patronage but to a much wider public. The fact that the major centres of Samanid power were Bukhara and Samarqand, 21

21 The textbook case is the mosque of Yazd-i Khwast. See M. B. Smith, ‘Three Monuments at Yazd-i Khwast’, Ars Islamica, VII, 1940, pp. 104–6 and M. Siroux, ‘La Mosquée Djum’a de Yezd-i-Khast’, Bulletin de l’Institut français d’Archéologie orientale XLIV, 1947, pp. 101–18. 22  A turning point in this respect is furnished by an article by Oleg Grabar: ‘The Iconography of Islamic Architecture’ in P. P. Soucek (ed.), Content and Context of Visual Arts in the Islamic World. Papers from A Colloquium in Memory of Richard Ettinghausen, Institute of Fine Arts, New York University, 2–4 April 1980. Planned and Organized by Carol Manson Bier, University Park and London, 1988, pp. 51–65. 23  Ainy, Avicenna Epoch, colour pls. 129–34; G. A. Pugachenkova and L. I. Rempel’, Ocherki Iskusstva Srednei Azii, Moscow, 1982, pl. on p. 110. 24  T. F. Fitzherbert, Themes and Images on the Animate Buff Ware of Medieval Nishapur, unpublished M.Phil. thesis, University of Oxford, 1983, pp. 29–33; for a summary of her findings, see Grube, Cobalt and Lustre, p. 51. See also Ettinghausen, ‘Traditionalism’, pp.100, 109–10.

356

not Nishapur, adds extra weight to this conclusion. If this assessment is accurate, the use of motifs derived from Sasanian iconography in this ware has very different implications from the use of such themes in more expensive and obviously royal media like silver and silk. It means that these themes were rooted in the popular imagination. Perhaps they were valued for their nostalgic evocation of past glories. Perhaps, too, they were 25 more a trigger for somewhat unspecific but uplifting memories than a deliberate attempt to copy Sasanian themes accurately. Even if this is going too far, the mere fact that they were beamed towards a lower social level guaranteed that the themes themselves would be less fully understood and therefore less accurately reproduced than they would be in work made for the court. In other words, some degree of vulgarisation is only to be 26 expected. That word is used deliberately here to distinguish this particular type of Sasanian-related art from the other kinds of reshaping to which Sasanian themes were subjected at the hands of Islamic artists. Only in pottery among surviving artefacts can one catch a glimpse of the impact of Sasanian visual ideas at a popular level. In Western Iran, Garrus and 27 Yasskand/Yasukand ceramics are the vehicle for this process. To assess the degree of this alleged vulgarisation, it will be convenient to compare like with like; and, taking the theme of the cavalier, to see how this motif was treated in both the relatively high and the relatively low art of eastern Iran in the centuries closely following the Arab invasion. Two wall paintings from the residences of princes and nobles from Afrasiyab 28 29 (fig. 3) and Nishapur (fig. 4) respectively can act as a benchmark. These imposing murals are almost life-size and they contain a remarkable range of detail which seems to be accurately delineated. Although the cavaliers are by no means identical in costume, accoutrements and horse trappings, they are still more clearly differentiated from the standard Sasanian royal  Ettinghausen, ‘Traditionalism’, pp. 98, 107–9.  This is shown too by the simplified renditions of images of the fixed stars

25 26

that so consistently appear on these wares. Cf. J. Zick-Nissen, ‘”Ghabri-Ware” und “Nischapur-Keramik”’, Keramos LXIV, 1974, pp. 35–45; eadem, ‘Figuren auf mittelalterlich-orientalischen Keramikschalen und die “Sphaera Barbarica”’, AMI, N.F. VIII, 1975, pp. 217–40. 27 Cf. the Garrus plate which depicts the tyrant Zahhak with snakes issuing from his shoulders, a theme familiar from the Shahnama (colour pl. in C. du Ruy, Art of Islam, tr. A. Brown, New York, 1970, p. 55). 28 Belenizki, Kunst der Sogden, p.11. For an up-to-date context for Sogdian painting, see B. I. Marshak and V. A. Livshits, Legends, Tales and Fables in the Art of Sogdiana (Biennial Ehsan Yarshater Lecture Series, No. 1), New York, 2002. 29  Wilkinson, Nishapur: Buildings, pp. 205–14; Otto-Dorn, L’Art de l’Islam, p. 98.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

357

cavalier as found on numerous silver plates. These murals prove, therefore, that an alternative iconographic tradition for depicting a rider of rank had developed by the post-Sasanian period. How closely it conformed to non-royal Sasanian originals must for the moment remain a matter for speculation. At all events, the scale and quality of these murals establish that the motif of the cavalier had a certain cachet in noble and well-to-do circles in the immediately post-Sasanian period. What, then, was the fate of this motif when demoted to the humble medium of ceramics? The key example here is the celebrated rider dish 30 now in the Iran Bastan Museum in Tehran ( (fig. 5). Here the changes in the august model are legion. They can be summed up in the observation that the Sasanian image has regressed dramatically, but in the process has acquired all the vitality and directness of authentic folk art, qualities which are also to the fore in other versions of this theme in Nishapur pottery 31 (fig. 6). The artist has gravitated naturally to a bold and eye-catching idiom, sometimes alarmingly simplified to the point that the image is more a pictograph than a representation of the ostensible subject. The familiar Sasanian motifs take on a new and vigorous life in this guise. Symptomatic of the whole approach is the treatment of the eye as a huge horizontal almond shape which takes up most of the face, and the disappearance of the mouth. One counts; the other does not. In some versions of this visual idiom, even the nose has gone. An uninhibited relish in curvilinear design accounts for the baroque exuberance of the ballooning thighs tapering into tiny legs and miniscule feet, a detail also found in other

 For a colour illustration of the Tehran piece, see Bagherzadeh et al., Oriental Ceramics, pl. 23. The best description of it is by Wilkinson, Nishapur: Pottery, pp. 20–2. 31  Other examples of the rider motif in Nishapur ceramics, not all of which have undergone searching technical scrutiny to determine the degree of their authenticity, include the bowl in the Tenri Museum (M. Yoshida, In search of Persian pottery, tr. J. M. Shields, New York and Tokyo, 1972, colour pl. 11), another bowl in the Mahboubian Collection (Mahboubian, Treasures, no. 21) – both very curious pieces – a bowl formerly in the Foroughi Collection and now in the Iran Bastan Museum (D. Talbot Rice, Islamic Art, London, 1965, pl. 43 – see fig. 5 above); one formerly in the Khalili Gallery (G. Fehérvári and Y. H. Safadi, 1400 years of Islamic art: A Descriptive Catalogue, London, 1981, p. 154 and pl. 93); and another in the Tareq Rajab Collection (G. Fehérvári, Ceramics of the Islamic World in the Tareq Rajah Museum, London and New York, 2000, pp. 34–35 and 50). For references to still further examples, see E. J. Grube, Islamic Pottery of the Eighth to the Fifteenth Century in the Keir Collection, London, 1976, p. 70, n.5. 30

358

types of Nishapur figural ware. The sub-division of the main figure into various boldly delineated compartments only loosely related to the actual volumes and surfaces they represent also reflects a preference for line over modelling. In this respect it is moving decisively away from the somewhat laboured attempts at realism which characterise such subject-matter in 33 Sasanian metalwork. Later Iranian pottery, notably of the Saljuq period, 34 maintained this emphasis. A complicated system of hatching and dotting does duty for modelling and serves to distinguish one compartment from the next. This technique, typical as it is of folk art, has the property of reducing meaningful detail to mere pattern. A particular kind of belt, which 35 may perhaps denote a given rank, is reduced to a chequerboard design. A specific type of headdress, perhaps even personalised, is simplified into a rectangular box. Minor sartorial details which, taken all together, may identify the wearer, or at least his rank, are wiped out to make way for an overall and meaningless pattern. To treat potentially significant parts of the composition as if they were so much wallpaper not only robs them of meaning; it also flattens the entire design. This conforms to the patternmaking urge which so often characterises folk art. It could be argued, then, on the basis of these and similar details, that mere design has crowded out meaning in this image. But there can be no cavilling at the result – a vivid evocation of stately pride and jingling harness. Two other factors play an important role in this image: the apparently random elements that float around the central figure; and, closely related to this, the various supplementary creatures which encircle him. First, the random elements. These are so copious that one might almost say there is standing room only on the plate. A farrago of disjointed elements, all jostling for space, clamours for attention. This is typical of such Nishapur 36 ware. Numerous inscriptions, better described perhaps as pseudo32

 E.g. Mahboubian, Treasures, no. 25; C. K. Wilkinson, Iranian Ceramics, New York, 1963, pl. 26; idem, Nishapur: Pottery, nos. 64 and 66; Grube, Cobalt and Lustre, nos. 36 and 57. 33  Harper, Silver Vessels, pls. 8, 10, 15, 28 and 38. 34  O. Watson, Persian Lustre Ware, London, 1985, colour pls. A, B, D and J. 35  Otto-Dorn, ‘Bildgut’, pp. 9–15; R. Ghirshman, ‘La ceinture en Iran’, Iranica Antiqua XIV, 1979, pp. 167–96; cf. E. Esin, ‘Quelques aspects des influences de l’art des anciens nomads eurasiens et de l’art du Turkestan pré-Islamique sur les arts plastiques et picturaux turcs’, First International Congress of Turkish Arts, Ankara 19th–24th October, 1959. Communications presented to the Congress, Ankara, 1961, pp. 114–15. 36  E. g. Wilkinson, Nishapur: Pottery, colour pl. 2 and nos. 74, 81, 86 and 88. 32

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

359

inscriptions or epigraphic bits and pieces, punctuate the rim of the plate. They lend the design an instantly recognisable Islamic flavour; but the 37 key point is that they mean nothing. Are they intended to confer prestige on the image? There is no way of telling, but it is clear that the obvious and primary purpose of writing, namely to inform, is not operative here. Perhaps the public for which such plates were made was illiterate. In that event the writing might well have had symbolic, quasi-magical associations. Other epigraphical fragments intrude into the central portion of the design – for example, between the horseman and the crouching feline, or under his elbow. The latter word, which is repeated several times elsewhere on the plate, seems to be a crudely inaccurate approximation of the word baraka which is frequently found on the epigraphic plates made 38 at the same time and in the same place as this dish. Other random motifs include posies, leaves, flowers and tiny ovals. And so to the extra creatures. The animals surrounding the rider are indeed a weird assortment. The negligent pose of the cavalier, with his (sheathed?) sword resting easily on his shoulder, suggests a casual outing rather than the chase and discourages the supposition that any of them represent game for the hunt. The obvious candidate for that role, namely the feline valiantly trying to attract the cavalier’s attention while scrabbling to stay on the horse’s back, is probably intended to represent a cheetah. The practice of hunting with cheetahs and training them to sit on the croup of the hunter’s horse is first recorded in the Middle East, it seems, 39 under the Umayyads. Like the stirrup, therefore, which is also not found in Sasanian art, it provides a dating clue which points to the post-Sasanian period. The transitional nature of the image is epitomised by the cheetah’s pose. Rather than crouching, curled up, on the horse’s croup, as the practice of the day indicated, it seems rather to be leaping on to the horse’s back. In other words, the pose is that of the menacing feline so often depicted on

Cf. the remarks by R. Ettinghausen in an important article: ‘Arabic Epigraphy: Communication or Symbolic Affirmation’ in D. K. Kouymjian (ed.), Near Eastern Numismatics, Iconography, Epigraphy and History: Studies in Honor of George C. Miles, Beirut, 1974, pp. 297–317. 38  S. Flury, ‘Une formule épigraphique de la céramique archaïque de l’Islam’, Syria V, 1924, pp. 53–66. 39  F. Viré, EI2, s.v. ‘Fahd’. Cf. I. Goldziher, Muslim Studies, ed. S. M. Stern, tr. C. R. Barber and S. M.Stern, London, 1967, I, p. 143. 37

360

1. Silver plate with monarch hunting various animals, 6th–7th century. Bibliothèque nationale de France, Paris. After SPA, VII, pl. 214.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

361

2. Vase from Merv, 6th century (?). State Museum of the History of Turkmenia, Ashkhabad. After Ainy, Avicenna Epoch, pl. 132.

362

3. Drawing of wall painting of a mounted aristocrat, Afrasiyab, 7th century. After Belenizki, Kunst der Sogden, p. 11.

4. Drawing of wall painting of a mounted aristocrat, Nishapur, 10th century (?). Iran Bastan Museum, Tehran. After Wilkinson, Nishapur: Buildings, p. 207, fig. 2.40

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

363

5. Buff-ware bowl with rider, Nishapur, 10th century (?). Inv. no. 3909, Iran Bastan Musuem, Tehran. After Robert Charlston, Masterpieces of Western and Near Eastern Ceramics, IV, Tokyo, no .24.

6. Buff-ware bowl depicting a falconer, Nishapur, 10th century (?). Ex-Foroughi Collection; Iran Bastan Museum, Tehran. After David Talbot Rice, Islamic Art, 1965, pl. 43.

364

7. Drawing of left-hand side of double frontispiece, Shahnama, probably Shiraz, 733/1333. 2a, ex-Dorn 329, State Public Library, St Petersburg. After A. T. Adamova and L. T. Giuzalian, Miniatiury rukopisi poemy “Shakhname” 1333 goda, Leningrad, 1985, p. 42.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

365

8. Drawings of enthroned rulers as depicted on coin (after Herzfeld). (reading from top left to bottom): Phraates III, Orodes II, Artaban II, and Volagases II; the Klimova plate; the Baltimore plate; Kushan coin.

366

9. Silver dish with a royal couple on a takht, 6th–7th century (?). Inv. 57.709, The Walters Art Gallery, Baltimore. After Splendeur des Sassanides, p. 211, no. 65.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

10. Silver plate with enthroned monarch with courtiers, from Strelka, 6th century (?). Acc. no. S250, State Hermitage Museum, St Petersburg. After Splendeur des Sassanides, p. 207, no. 61.

367

368

11. Silver plate with ruler hunting lions, post-Sasanian. State Hermitage Museum, St Petersburg. After SPA, VII, pl. 218.

12. Silver gilt dish with ruler on a takht surrounded by courtiers and musicians, post- Sasanian. Inv. S–47, State Hermitage Museum, St Petersburg. After Splendeur des Sassanides, p. 210, no. 64.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

369

Sasanian (and even some post-Sasanian ) hunting plates – with the crucial difference that the cavalier, far from facing the animal, has his back to it. And the animal is now tame! Thus the familiar pose serves a new purpose, but does so awkwardly. It is in such awkwardnesses that the transitional 41 nature of the image betrays itself. It is noteworthy that most of the other creatures are peacocks, birds 42 with strong royal associations; the only exception consists of two moufflons whose stylishly elongated horns describe a graceful arc over their backs in a style markedly reminiscent of the earliest painted pottery 43 of Iran, namely the 4th millennium B.C. wares of Susa. Yet their scale is too diminutive to make them convincing as the quarry of the cavalier. In general, then, the force of this mass of extra detail has been to replace the 44 idea of the hunt by that of an outing. Finally, what is the evidence of metalwork? The example selected for analysis here is a small gilded silver plate in the Hermitage in St. Petersburg which shows a prince reclining on a couch and surrounded by courtiers 45 and entertainers (pl. XIV, b). It has been variously attributed to very early Islamic times and to the Buyid period, i.e. 945–1055, and this range is 46 itself a clear reminder of the uncertain chronology of such pieces. The 40

40  E.g. the Pur-i Vahman plate (colour plate in Splendeur des Sassanides, p. 197); for a discussion of it, and of the contentious term ‘post-Sasanian’, see Harper, Silver Vessels, pp. 139–41. 41  For a telling example of another and quite different transitional version of the Sasanian hunt motif, this time conflated with a human-headed beast, compare the jug from the Nagyszentmiklós treasure (G. László and I. Rácz, The Treasure of Nagyszentmiklós, tr. H. Tarnoy [n.p., 1984], picture 9, p. 47). 42  For a brief survey of the subject, see the discussion in A. Daneshvari, Medieval Tomb Towers of Iran: An Iconographical Study, Lexington, Ky., 1986, p. 54; its association with the sun (ibid., pp. 56—8) was of course also susceptible to such an interpretation. Cf. also idem, ‘A preliminary study of the iconography of the peacock in medieval Islam’ in R. Hillenbrand (ed.), The Art of the Saljuqs in Iran and Anatolia, Costa Mesa, Cal., 1994, pp. 192–200. 43  Cf. H. M.-J. Steve, ‘Susa. Die Akropolis: Letzte archäologische Entdeckungen’, Archaeologia Viva I/1, 1968, p. 142, figs. 185–6. Similar wares are known from Tepe Sialk and Tepe Hisar (Bagherzadeh et al., Oriental Ceramics, pls. 21–2). 44  For a discussion of the possible patrons of such wares, see Grube, Cobalt and Lustre, pp. 47–48 and 51, where he draws on the theories of Richard Bulliet and Teresa Fitzherbert. 45  For related scenes on Sasanian and post-Sasanian metalwork, see Ward, IM, pp. 44–5, pls. 29–30. 46  For some of the proposed datings, see SPA, VII, pl. 230B, caption; Bahrami, ‘Gold Medal’, p. 13 (but see n. 13 above); K. Erdmann, Die Kunst Irans zur Zeit der Sasaniden, Mainz, 1969, p. 104; V. G. Lukonin, Archaeologia Mundi: Persia II, tr. J. Hogarth, Geneva, Paris and Munich, 1967, p. 222; Marshak, Silberschätze, pp.

370

Sasanian flavour of the scene as a whole – it is, after all, the traditional bazm (banquet), a festival of wine, women and song, which was as much 47 a part of royal life as razm (battle ) – is instantly recognisable, and in this respect the image conforms to the pattern set by the other examples discussed earlier in this paper. It is almost as if the principal aim of the artist had been to evoke a Sasanian ambience, and that he was not fussy about how he did so. It argues a sure instinct for Sasanian modes, if not a solid familiarity with them, for the artist to be able to conjure up an authentic pre-Islamic atmosphere by means of non-authentic detail. Individually, to be sure, some of the details are accurate enough. But they are too big or too small, too prominent or too insignificant, to work as they should. Or they do not belong together in the same image. It is above all a transitional piece. The notion of the monarch at the centre of the composition, with attendant figures grouped in a circle around him like satellites wheeling around their parent planet, was to be a characteristic of Islamic royal iconography. Yet in such fully realised Islamic images the ruler would be seated cross-legged or with one knee raised, usually on a throne, in a fully 48 frontal position (fig. 7). Here he is neither seated on a throne nor is he fully frontal. In both these aspects the plate is a transitional piece. His face and trunk are frontal, but the rest of his body is in profile. He is seated half-upright, but half reclining on a couch. In this context it is worth remembering Herzfeld’s analysis of the standard Parthian and Sasanian motif of the monarch on a couch, and his magisterial demonstration of how this formal image gradually disintegrated with the passage of time 276–8; Splendeur des Sassanides, p. 210; A. Soudavar, The Aura of Kings: Legitimacy and Divine Sanction in Iranian Kingship, Costa Mesa, 2003, pp. 35–6 and 160 (he suggests a connection with Anahita). 47  A. S. Melikian-Chirvani, ‘The Iranian Bazm in early Persian Sources’ in R. Gyselen (ed.), Banquets d’Orient (Res Orientales, IV), Bures-sur-Yvette, 1992, pp. 95–119; R. Ghirshman, ‘Notes iraniennes V: Scènes de banquet sur l’argenterie sassanide’, Artibus Asiae XVI, 1953, pp. 51–76; M. L. Carter, ‘Royal Festal Themes in Sasanian Silverwork and Their Central Asian Parallels’, Acta Iranica, I (Commémoration Cyrus), Leiden, 1974, pp. 171–202; A. C. Gunter, ‘The Art of Eating and Drinking in Ancient Iran’, Asian Art, I/2, 1988, pp. 7–52. For the role of wine in particular, see A. S. Melikian-Chirvani, ‘The Wine-Bull and the Magian Master’, Societas Iranologica Europea. Recurrent Patterns in Iranian Religions from Mazdaism to Sufism. Proceedings of the Round Table held in Bamberg (30th September–4th October 1991) (Cahiers de Studia Iranica, 11), Paris, 1992, pp. 101–34. 48  For a typical example of this iconography in lustre ceramics, see Bagherzadeh et al., Oriental Ceramics, colour pl. 35. For an example in book painting, from the Schefer Hariri, see B. Farès, Le Livre de la Thériaque, Cairo, 1953, p. 16, fig. 3.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

371

(fig. 8). From reclining gracefully the monarch began to loll and then to sprawl until the image had lost all its original decorum and the ruler was falling as it were drunkenly off his perch. The latter detail was to acquire extra verisimilitude when the ruler was depicted holding a cup of wine, as in this case. It was in this form that the image was to enjoy its greatest popularity, travelling at least as far east as India, where it is found 50 in the 6th–century caves at Ajanta. What is significant in this gradual deterioration of an image of majesty is that it was not Sasanian but foreign (for example, Kushan) or post-Sasanian craftsmen who were responsible for the various deformations to which it was subjected. There is still a good deal more to be teased out of this image, especially the origins and meaning of the royal figure, poised so uncomfortably, it seems, between official pomp and unofficial revelry, an ambiguity repeated in comparable silver dishes (fig. 9). The reason for such uncertainty of tone lies in the origin of this iconographic type. A whole series of Parthian coins depicts the king in profile seated on a throne, posed to show both their legs in a thoroughly graceful yet relaxed fashion (fig. 8, top four 51 images with captions). One hand is usually stretched out either holding 52 an emblem of dominion or victory, perhaps even divinity, or receiving some insignia or emblem from a figure such as a nike or a tyche. These are representations of investiture, which was of course to become one of the most favoured of all Sasanian themes. The link with the Hermitage plate is clear, but certain changes have been introduced and these break the continuity of meaning. These changes are worth noting in detail. Gone is the extra figure; gone too is the profile view which allowed the king to communicate with it. Instead, the king’s face is both frontal and preternaturally enlarged. The result is a conflation of two iconographic types normally kept distinct: the king as reveller, absorbed in the pleasures of wine and music, and thus turned away from the viewer; and the king as a formal image of power, with a corresponding emphasis on frontality so that the monarch is staring 49

49  E. Herzfeld, Archaeological History of Iran, Oxford, 1935, pp. 70–71 and 72, fig. 9. Herzfeld does not identify the individual images in his fig. 9, but they are of Kushan type: see J. M. Rosenfield, The Dynastic Arts of the Kushans, Berkeley, Cal., 1967, pls. II–V and VII. 50   Ghirshman, Parthians and Sassanians, pl. 441. 51  E. Herzfeld, Iran in the Ancient East, London and New York, 1941, pp. 295 and 296, fig. 388. 52  As on the Parthian relief at Tang-i Sarwak (Harper, Silver Vessels, pp. 105–6 and pl. 30).

372

directly outwards. Other post-Sasanian images, to say nothing of later ones – as in the Aghani frontispieces of the early 13th century – took this conflation a stage further and created a new combination of ideas: the seated frontal prince surrounded by references to his leisure pursuits 54 (fig. 7). The Hermitage plate marks a stage in that evolution. The transitional nature of this plate (fig. 12) is betrayed by a host of other minor inaccuracies. A solitary winged genius hovers to the left of the prince’s head. He carries not a crown, not some trophy of victory, not a cup, not even a sash, but a crescent hanging from a chain – an attribute 55 for which no parallel in such a context suggests itself. Moreover, there is no second genius on the other side of the ruler’s head. The same lopsided quality can be recognised in the two attendant courtiers. Both have a mask strapped over their mouths so that their breath cannot pollute the royal 56 presence – an authentic echo of Sasanian court protocol. Yet only one of them is in the obviously correct place, namely at one side of the ruler’s couch. The other, ousted by two musicians, and significantly reduced in size because the composition compels him to occupy a much smaller space, has wandered somewhat aimlessly into the foreground, where he glumly contemplates a ewer. His arms are crossed over his chest, like those of his counterpart, but with the difference that his hands are tucked away almost out of sight, while the other attendant has his hands splayed on his chest. In a Sasanian royal context the former pose is usually associated 57 with courtiers attending the king (fig. 10), and that model is accurately reproduced here. Yet the hands splayed on the chest strike a discordant 58 note and do not correspond to any known Sasanian precedent. Such iconographical inconsistencies or downright solecisms abound. The errant placing of this figure, marooned in the exergue of the plate, is itself enough to reduce his significance. He seems to have been demoted 53

53  For comparable pieces see Grabar, Sasanian Silver, pp. 58–9 and 102–3, nos. 14 and 15. It is worth bearing in mind here the doubts concerning the authenticity of some of the silver objects of Sasanian inspiration that suddenly appeared on the art market in the 1960s. 54  SPA, VII, pl. 208a; for an excellent drawing of the central part of this key piece, see Bahrami, ‘Gold Medal’, p. 14, fig. g. 55  Though the chain with a pendant and numinous object does bring to mind the famous throne apse at Khirbat al-Mafjar (Ettinghausen, Byzantium to Sasanian Iran, pp. 28–30). 56  Cf. the Qazvin plate (Ghirshman, Parthians and Sassanians, fig. 246). 57  Cf. the Strelka bowl (ibid., fig. 245). 58  Significantly enough, however, there is indeed a close parallel in post-Sasanian metalwork (e.g. Ward, IM, pl. 30).

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

373

to overseeing the picnic suggested by the vessel hanging from a tripod over an open fire. This latter feature, when taken in conjunction with the ewers on either side of it, may be interpreted as a somewhat inept early version of that ostentatious array of containers and food which often takes up the foreground of court scenes in late 14th and early 15th-century 59 Persian miniatures. Sometimes attendants in such paintings are busying 60 themselves with preparing food, while in one still earlier book painting, namely the frontispiece of the Kitab al-Diryaq in Vienna, datable to the early 13th century, an attendant seated below the monarch is cooking 61 kebabs over a grille. The rather primitive frontality of the monarch on this plate is of a piece with that found on other metalware conventionally dubbed ‘postSasanian’, such as the plate showing a rider in the act of spearing a lion 62 63 (fig. 11), another similar piece in which the rider shoots a lion and a third dish with a pacing lion whose face is wrenched round to a frontal 64 pose. In Sasanian hunting dishes such frontality is precluded by the rider’s concentration on the business in hand, sometimes almost breaking his neck in the process. In the plate under discussion, as already noted, it jars in a scene whose entire layout implies a narrative. Other features which help to create an acceptably Sasanian atmosphere but are still flawed in detail include the crown, which lacks the usual domical cap and a crest (indeed, the crest is misplaced in the centre of the crown’s base); the unrecognisable animal depicted in the medallion on the ruler’s tunic; the flower he is sniffing, which is rather larger than the norm in Sasanian art; the ruler’s hair, in which, conversely, the normal ballooning flounces are reduced to near-insignificance; and the similarly reduced ribbon which trails from his head. Yet details such as his tiny feet and the cushions on which he reclines are faithfully observed, just as still others – such as the cup he is holding, for which a relevant, perhaps slightly earlier, parallel is 65 a bowl in the Hermitage with scenes from Syleus – may already betray a  SPA, VII, pl. 870; B. Gray, Persian Painting, Geneva, 1961, pp. 54, 103–4; idem, Persian Painting, London, 1947, pl. 4. 60  E.g. the 1429 Kalila wa Dimna in the Topkapi Sarayi Library (R.1022), Istanbul; see T. W. Lentz and G. D. Lowry, Timur and the Princely Vision. Persian Art and Culture in the Fifteenth Century, Los Angeles and Washington, DC, 1989, pp. 110–11. 61  Colour illustration in R. Ettinghausen, Arab Painting, Geneva, 1962, p. 91. 62  SPA, VII, pl. 218. 63  Ibid., pl. 208B. 64  Ibid., pl. 232B. 65  Ghirshman, Parthians and Sassanians, pl. 359. 59

374

date in the Islamic period. Indeed, such a cup, decorated with images of female musicians and inscribed as the property of Windad Ohrmazd of 66 the Karens, has been attributed to 8th–century Mazandaran. One further detail, again diagnostic of the gaucheries of post-Sasanian metalwork, and encountered earlier in the Nishapur rider bowl in Tehran, is perhaps worth noting. This is the general sense of clutter which pervades the composition. Too much is going on to be comfortably fitted 67 into the space available. The rosette, for example, is a royal and divine 68 symbol of great antiquity in Near Eastern art, and thus its location here, hovering over the prince’s head, is potentially of great symbolic force. Yet its impact is fatally weakened by another (smaller) rosette and no less than three heart-shaped motifs which are scattered broadcast throughout the composition. Such devices have their proper place in Sasanian textiles, even if they are sometimes only space-fillers; but here, divorced from their normal function, they are just litter. Worse, they obscure and vitiate motifs that do count. The same defective sense of composition, incidentally, is betrayed in the way that the design is heavily over-balanced on one side. What conclusions can be reached on the basis of this evidence? The two works of art presented in detail in this paper have, perhaps, shed some light on how, in the field of the visual arts, Sasanian themes became Islamic ones. It must be admitted that the results were rarely felicitous. Much was lost, and the time was not yet ripe for a fully mature Islamic style to come into its own. The serious attempt to evoke a pre-Islamic Persian ambience was successful in broad outline, but it was badly flawed in detail. It would certainly not stand up to detailed scrutiny. Conversely, the addition of novel and significant detail – such as the use of new attributes of royalty like the bow, the cheetah and the saddle-cloth – was not smoothly integrated into the image as a whole. It is a tribute to the immanent power of Sasanian images, and their remarkable tenacity, that they survived this process of transplantation, with all its attendant indignities, so well. Perhaps the single overriding characteristic of the four 66   See M. L. Carter in P. O. Harper, The Royal Hunter: Art of the Sasanian Empire, New York, 1978, pp. 77–8. For a fuller publication of this piece, and a different reading of its inscription, see R. Ghirshman, ‘Argenterie d’un seigneur sassanide’, AO II, 1957, pp. 77–82 67   It might be argued that some earlier, and as it were fully Sasanian, plates are also crowded (e.g. pl. 1); but that crowding is integral to the narrative – in this case, the animals are stampeding in terror of the pursuing monarch, and that idea would be lost if only one or two of the quarry were depicted. 68  Ettinghausen, From Byzantium to Sasanian Iran, pp. 36–9.

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

375

centuries between the Muslim conquest and the arrival of the Saljuqs was that the Sasanian heritage, while theoretically revered, was in actual fact gradually belittled and plundered to become a casual quarry for ideas – ideas often expressed with an inadequate sense of fitness. The proliferation of trivial detail tended to impoverish rather than to enrich these inherited images, and sometimes the effect is (no doubt unintentionally) downright 69 absurd. Changes in the relative size or positioning of the key image, or for that matter repetitions and omissions, could have the same debasing results. Sasanian craftsmen had already refined many of their standard images to a high finish, and subsequent tinkering with the model was apt to weaken it. The cumulative effect of all these changes, many of them minor in themselves, was to stress decoration at the expense of meaning. It was the achievement of the subsequent period, the age of the Saljuqs, to streamline the art so painfully and erratically forged in these centuries and to dovetail meaning and decoration in a new and spectacularly successful synthesis.

Abbreviations Ainy, Avicenna Epoch L. Ainy, The Central Asian Art of Avicenna Epoch, tr. L. A. Schwartz, Dushanbe, 1980. AMI Archaeologische Mitteilungen aus Iran AO Ars Orientalis Bagherzadeh et al., Oriental Ceramics F. Bagherzadeh, A. Saurat and M. Sato, The World’s Great Collections: Oriental Ceramics. Vol. 4. Iran Bastan Museum, Teheran, Tokyo, New York and San Francisco, 1981. Bahrami, ‘Gold Medal’ M. Bahrami, ‘A Gold Medal in the Freer Gallery of Art’, in G. C. Miles (ed.), Archaeologia Orientalia In Memoriam Ernst Herzfeld, Locust Valley, NY, 1953, pp. 5–20.   E.g. in the textile from Antinoë, now in Lyons, already mentioned above (n. 14), which anarchically juxtaposes a putto, a black captive, a caricature of a Sasanian ruler enthroned and a bare-breasted Amazon menacing a mounted rider executing a Parthian shot (Ghirshman, Parthians and Sassanians, pl. 289). 69

376

Belenizki, Kunst der Sogden A. M. Belenizki, Mittelasien: Kunst der Sogden, tr. L. Schirmer, Leipzig, 1980. Ettinghausen, Byzantium to Sasanian Iran R. Ettinghausen, From Byzantium to Sasanian Iran and the Islamic World, Three Modes of Artistic Influence, Leiden, 1972. Ettinghausen, ‘Traditionalism’ R. Ettinghausen, ‘A Case of Traditionalism in Iranian Art’ in O. Aslanapa and R. Naumann (eds.), Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann, Istanbul, 1969, pp. 88–110. Ghirshman, Parthians and Sassanians R. Ghirshman, Parthians and Sassanians, tr. S. Gilbert and J. Emmons, London, 1963. Grabar, Sasanian Silver O. Grabar, Sasanian Silver: Late Antique and Early Mediaeval Arts of Luxury From Iran, Ann Arbor, 1967. Grube, Cobalt and Lustre E. J. Grube, Cobalt and Lustre: The first centuries of Islamic pottery. The Nasser D. Khalili Collection of Islamic Art, General Editor: J. Raby, IX, London and Oxford, 1994. Harper, Silver Vessels P. O. Harper, Silver Vessels of the Sasanian Period. Volume One: Royal Imagery, New York, 1981. JSAI Jerusalem Studies in Arabic and Islam Mahboubian, Treasures M. Mahboubian, Treasures of Persian Art after Islam: The Mahboubian Collection, New York, 1970. Marshak, Silberschätze B. Marshak, Silberschätze des Orients, tr. L. Schirmer, Leipzig, 1986. Otto-Dorn, ‘Bildgut’ K. Otto-Dorn, ‘Türkisch-islamisches Bildgut in den Figurenreliefs von

THE ISLAMIC RE-WORKING OF THE SASANIAN HERITAGE

377

Achthamar’, Anatolia VI, 1961–62, pp. 1–69. Otto-Dorn, L’art de l’Islam K. Otto-Dorn, L’art de l’Islam, tr. J.-P. Simon, Paris, 1967. SPA, VII A Survey of Persian Art from Prehistoric Times to the Present, VII, eds. A. U. Pope and P. Ackerman, London and New York, 1938–39. Splendeur des Sassanides Splendeur des Sassanides (Hofkunst van de Sassanieden), exhibition catalogue, Brussels (Koninklijke Musea voor Kunst en Geschiedenis), 1993. Talgam, Origins R. Talgam, The Stylistic Origins of Umayyad Sculpture and Architectural Decoration, 2 vols., Wiesbaden, 2004. Ward, IM R. Ward, Islamic Metalwork, London, 1993. Wilkinson, Nishapur: Pottery C. K. Wilkinson, Nishapur: Pottery of the Early Islamic Period, New York, n.d. [1973] Wilkinson, Nishapur: Buildings C. K. Wilkinson, Nishapur: Some Early Islamic Buildings and Their Decoration, New York, 1986

XVII A Leaf from my Rose Garden: Themes in Persian Art

T

Introduction

his choice collection of the Islamic art of Iran manages to achieve several aims simultaneously. It tells the story of Persian art over a period of more than a thousand years, from Umayyad times until the onset of the modern period (661–1750). It covers the major themes of Persian art in that millennium. It does so through a wide range of media – the arts of the book, ceramics, textiles, carpets, metalwork, glass, ivory and woodwork. And it springs many a surprise on the way, introducing to the wider world a host of pieces of the first quality and often of signal importance. The limited space available here suggests that the most productive approach might be to view this diverse material through the lens of a series of major themes – writing, people, fauna and flora, and geometry. These themes consistently overlap with each other, so it is futile to draw up hard and fast categories. Other themes might be proposed – religion, symbolism, colour, magic and the supernatural, among others – but to deal with these as well would be to say too little about too much. Writing Islam is a religion in which the word is paramount. Thousands of objects bear inscriptions of one kind or another, which obviously implies that literacy was widespread and by no means confined to the court elite. In Iran, moreover, the vast majority of those inscriptions are not in Persian but in Arabic, the lingua franca of the entire medieval Islamic world. Inscriptions by definition are meant to be read, but even this apparently

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

379

simple proposition seems open to doubt when one considers the sheer élan, the bravura rhythms and the baroque complexities of the texts that are inscribed on many an object or building. Clearly the urge to communicate is at cross purposes, if not at war, with the urge to embellish a surface. To read such texts requires the skills of the cryptographer or the crosswordpuzzle addict. And often enough the inscription has to assert itself against the counter-claims of the other ornament that shares the field (pl. 4). Ceramics – a medium in which this collection is especially strong – are an excellent prism through which to view this millennial evolution. The shifts from major to minor, from stately solemnity to liquid fluency, from the angular to the curvilinear, compel admiration even while they hinder comprehension. In practically every case the writing enhances the beauty of the piece. The content of these inscriptions, which in the portable arts are found mostly on pottery and metalwork, draws on a somewhat limited repertoire. In some types of Kashan ceramics and tiles of late Saljuq and Mongol date the texts consist of Persian lyric love poetry in a rather sentimental vein, though some verses have Sufi undertones. But material in Arabic dominates, and while there are Qur’anic or historical texts and others which fall into the category of proverbs and moral axioms (pl. 11) the majority express good wishes to the anonymous owner (pl. 7). They do so, however, in a very particular way. The good things called down upon the owner may take the form of long lists of desiderata – ‘dominion and glory and happiness and peace and prosperity’. Or they form symmetrical pairs like ‘perpetual glory and lasting peace’. Yet other combinations are known. All share a measured cadence (sometimes the components even rhyme) which has a ritual, incantatory flavour like a priestly blessing. This is entirely appropriate, for of course in Muslim belief it is only God that can bestow such blessings. In this sense, then, even apparently banal formulae contain elements of the sacred and thus sanctify the object that they decorate. Just as many Persian intellectuals, both scientists and literary men, from Abbasid times onwards, surpassed many Arabs in their use of the Arabic language, so too many Persian calligraphers and artists produced Arabic texts and inscriptions of the highest quality. This skill found expression in an astonishing variety of Kufic scripts created in all media in the period up to the Mongol invasion. Two basic divisions operate, admittedly with some overlap. The first type is based on pen-made writing, in which thin strokes and curvilinear emphases play an important role. The second

380

features thick letters, often given decorative internal modifications or carrying extra ornamental devices, and imbued with a rhythm that is at once stiff and stately. The principal focus of fine writing in the arts of the book was the Qur’an, which was of course in Arabic. Such manuscripts indeed became the forcing ground for all kinds of calligraphic experiment. The collection has excellent examples of most of the major scripts. The earliest are traditionally associated with Arabia itself, and the collection has a fine example of this primeval ductus (pl. 1), a seventh-century 23-line Hijazi leaf characterized by a marked leftward lean and a tendency to extend certain letters in their terminal form into a dramatic flourish. In this script, which still preserves echoes of Nabataean Aramaic inscriptions – of which hundreds survive in Arabia – many letters are not yet fully developed, vocalization is rudimentary and spacing uncertain. A zigzag ornamental panel in red and black fills the space between the end of the sura and the end of the line. There is clearly no attempt to impose any kind of pattern on the page; the text is not fully justified to either left or right. It was not long before artists, not content with experimenting boldly with letter forms, some of them involving dramatic extensions and contractions, transformed the look of the book by means of ornament. Sometimes the text is encased by no fewer than seven decorative bands much in the manner of the closing page of the luxury Umayyad Qur’an from San‘a’. These are held together by a muscular knotted design in the central band. The multiple frames serve to enhance the importance of the text block, and pairs of corner rosettes, axial escutcheons and an outer row of merlons ranged like battlements provide further grace-notes. From such a visually intense composition it was only a short step to an entire page devoted to ornament. Whereas broadly contemporary Hiberno-Saxon Gospel books, which also contain such ‘carpet’ pages, are already medieval in spirit, their Qur’anic counterparts still owe an obvious debt to Roman and Byzantine floor mosaics and thus remain recognizably within the fold of Mediterranean classical culture, despite the infiltration of the arabesque. A bifolium from the 15-line Blue Qur’an (pl. 2), which has recently been re-attributed by Alain George to Baghdad about 800, shows how quickly Qur’anic calligraphy became an art subject to the most rigorous professionalism, with a text block strictly justified to left and right and a tight, invisible linear grid to ensure pitch-perfect spacing between lines. Yet this severe discipline left ample room for the scribe to experiment with

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

381

the extreme elongation (mashq) of some letters. Thus he could achieve some recondite felicities in the rhythm of the page as a whole, contrasting elongated letters, tightly packed groups of letters and isolated singletons in constantly changing patterns. An Eastern or ‘broken’ Kufic page limited to five short lines (pl. 5) exalts the sacred text in quite a different way, for example by gathering together clumps of Z-shaped letters or ending four consecutive lines with a similar extended sub-linear flourish, thereby suggesting that these lines rhyme. The ever-changing alternation between upright and diagonal strokes, the latter extending even to the marks indicating vocalization, imparts a restless dynamism to every page. In later centuries fashion favoured much more legible cursive hands, from a majestic muhaqqaq to a dense, tight naskh. In a fine example of the first (pl. 13), most of the page is used and the text is totally dominant, sometimes with no added ornament at all, but at other times enhanced by an illuminated roundel or two. The calligrapher is confident enough to allow his letters to lean slightly at times, even though he is clearly capable of rendering them with absolute verticality. Terminations are lengthened for aesthetic reasons and some come to a sharp point, while individual letters or groups of letters undulate above or below a notional baseline. Such devices add an individual touch and rescue the text block from any suggestion of monotony. Large thick letters executed in gold and outlined in black evoke the sacred nature of the text. Conversely, in another and near-contemporary Qur’an (pl. 15), it is the ornamental framing of the text that proclaims its unique sanctity. Stately Kufic ansate panels, in gold on a ground of ultramarine, are placed above and below the text block, and its golden frame is punctuated by square blue studs. Finally the pink ground of the text bears decorative cartouches at top and bottom, and is sprinkled with gold roundels. The text itself is in neat small letters riding on white clouds – truly a message from heaven. These very varied manifestations of Qur’anic calligraphy indicate that there were practically no limits placed on the originality of the calligrapher even in this most conservative of arts. People It is almost an article of faith among those unfamiliar with Islamic art that Islamic tradition forbids the representation of the human form and indeed of living creatures in general. A few choice examples of Timurid, Turkman, Safavid and Mughal book painting in this collection should suffice to dispel this canard. Yet there is a grain of truth in this misconception, in that

382

such subject-matter is discouraged in most religious contexts, and beyond that one quickly recognizes a fundamental lack of interest in such aspects of figural art as naturalism, portraiture and genre. Thus the emphasis throughout is on types rather than individuals, and convention trumps patient observation of people, movement, expression and personality. People, like landscapes or objects of all kinds, are presented so as to yield maximum information and visual effect. Thus, in depicting a Timurid palace interior, the painter conjures up a world of improbably slender wasp-waisted courtiers and ladies, their heads nodding like narcissi in a setting of the utmost refinement and splendour (pl. 19). The figures are as precious – in both senses – as their surroundings. Gestures, poses and groupings, complementary and contrasting colours and textures, are calculated with mathematical precision. There is, in short, a built-in contradiction between patient rendering of surface detail and the lack of engagement with individuality. The same applies to battle scenes, whether realised in broadly simplified single combats, as in the ‘Big-Head’ Shahnama (1.MS. 4022; pl. 20), or as found in the peerless page from the Shahnama of Shah Tahmasp (pl. 23) depicting a nocturnal skirmish between the forces of Iran and Turan, in which the unhorsing of the Turanian champion takes centre stage. Here some sixty warriors are depicted with no sense of strain or crush, and there is an easy interplay of plot and sub-plot, with each figure, small or large, slotting comfortably into its allotted place. The faces, young or old, though all but expressionless, are treated with the same care as is devoted to weapons and horse armour. Yet the painting as a whole is alive with action: trumpets blare, drums roll, bowstrings twang, banners flutter and cressets flare in a saturated turquoise gloom. Even the horses are mostly frozen in a flying gallop. The Safavid rise to power spelled, at least for the time being, the eclipse of Herat as the most important school of Persian book-painting. Some of its finest practitioners were relocated to the royal capital of Tabriz, and the most celebrated Herati painter of all, Bihzad, was put in charge of the court atelier there. Safavid painting was the issue of a shotgun marriage of the Herati and royal Turkman styles. But some Herati painters sought refuge at the Shaibanid court in Bukhara and practised their traditional manner there. A leaf datable to the 1540s (pl. 24) exploits in a distinctively Herati way the contrast between high and low life to point a moral and adorn a tale. Hatim al-Ta’i, the traditional exemplar of generosity, is shown in the foreground forsaking the tents of the great and the good to spend time with a humble woodcutter. He, hunched beneath his load, offers

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

383

a mute contrast to the idle rich in conclave above. A gold background, luxury carpets and an ornate awning with a royal inscription distinguish their world from the bleak environment allotted to the labouring man, just as the convivial party of fourteen contrasts with the loneliness of the outcast woodcutter. The social satire here is unmistakable. While the intricacies of modelling and movement were of no interest to these artists, the Mughal painters of northern India, exposed as they were to prolonged contact with European art, found a middle way between the demands of realism and those of an essentially idealized world, as can be seen both in the thickly populated court scenes in which these painters specialized and in their numerous portraits of splendidly attired courtiers. Such images, precursors of the official photograph, are a standby of Mughal painting. They obey a set convention whereby the subject is depicted standing in profile, as if at attention, his costume and accessories – in this case shield, stick, turban and sword – rendered in meticulous detail. Yet the personality behind this finery, and behind that grave expression, eludes us. The faces of the seven men disposed around the margin are equally unreadable. In the main image the surface allotted to the face itself is too small to challenge the painter’s powers of perception and insight. Thus what purports to be a portrait is more a record of a cog in the Mughal court hierarchy. That same combination of close-focus observation and sketchy background with no pretensions to naturalism can be seen in Iran, too, right up to Qajar times, for example in a pencase where the poses and modelling of the figures betray a close study of European modes (pl. 31). Outside book painting, figures occur in many media, but usually not in a narrative context. Thus a Saljuq silver bottle credibly depicts a hunter galloping along with his falcon and his saluki (pl. 10) a theme that retained its popularity in Ilkhanid metalwork. Luxury ceramics in particular often focus on figures. Thus fashionably attired lovers commune by a stream at the foot of a cypress, emblem of eternity (pl. 9). A wind player and a female harpist discreetly incline towards each other with a wine jug strategically placed between them (pl. 7). Similar musicians would provide the accompaniment for an intricate all-male dance, whose reciprocal movements, interlinked clasped hands and raised knees recall the rising and falling rhythms, the advance and retreat, and the unfolding pattern of a Scottish reel. To suggest all this with such economy of means is quite a feat.

384

Nishapur produced a genre of ceramics that enjoyed just as much popularity as the epigraphic wares. Some scholars have seen in the jerky, expressive figural style of these wares the directness and unselfconscious vigour of folk art, and have interpreted them as purely decorative but random ensembles of highly abstract birds, animals and people, as seen in the image of a rider and his hunting dogs. Others propose that these images refer to the ceremonies and rituals of Iranian festivals, or that they reflect at several removes and in inferior materials a dim memory of Sasanian court art. Either way they express, in however clumsy and hesitant a fashion, some degree of national sentiment. One may contrast this with the delicate volumes and beautifully judged low relief of the ivory panel (pl. 14) which depicts a Turco-Mongol archer wearing a caftan with dovetailed lapels, short boots and a jaunty rounded hat with upturned brim. Apart from his bow and arrows he holds a weapon in his right hand. Two long oiled locks frame his expressionless, fashionably round face. His pose is immobile but somehow it radiates contained power. An eagle perches above; it was one of the favoured totems of such warriors, on whom Islam sat lightly. Sometimes elements of caricature enter, as in a lustre dish depicting a seated man, his robe stamped with chintamani designs and his grossly prehensile fingers clasping his knee and holding a fruit, a tour de force of dexterous calligraphic draughtsmanship. His beady eye, wispy beard and ambivalent smile instantly ring alarm bells; here is a real personality caught in a few strokes. But this kind of individuality is rare. Artists preferred to depict safely beautiful people impeccably attired in the height of fashion, complete with kiss-curl and perky hat as in a Safavid silk, or small groups enjoying an exclusive boating party (pl. 22). Such scenes evoke a privileged courtly lifestyle. Animals and birds There is a long tradition in Islamic art of depicting animals and birds. The theme announces itself from the earliest years. Excavations at Moshchevaya Balka in the Caucasus have unearthed Sogdian figured silken robes and other garments of the eighth to ninth centuries whose design repeats a roundel with an image of a bird or an animal, for example the senmurv. Numerous fragments of such silks are known from Sasanian times onward. A remarkably well-preserved caftan (pl. 4), perhaps of Saljuq date, is a variation on that same theme, and makes its wearer a walking menagerie. The roundels enclose a pair of affronted pacing birds, perhaps geese, on either side of a stylized plant with a bulbous base.

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

385

Four crouching horses, again in pairs, disposed on either side of a hybrid plant, with one pair depicted upside down, fill the interstices between the roundels. A Kufic benedictory inscription encircles each roundel. The use of animals and birds is particularly marked in ceramics, and Persian potters were adept at devising poses for birds and animals that would match the surfaces to be covered. With no apparent sense of strain a square space accommodates a stork with outstretched wings on a laqabi pot (pl. 6), or a wild goose in flight has a wing-span exactly matching the diameter of a round ‘Bujnurd’ bowl (pl. 17), while a strutting cockerel with extended tail feathers and curved crest also fits to perfection the allotted roundel of the plate he decorates. In all these cases bright colours or bold patterning elbow out observation from life, but nevertheless the essential character of the creature is evident at first glance. The background is often very busy but it is the bird that dominates the composition. A Safavid image of a bird (pl. 29), clearly a study from nature, recalls the closely focused studies of animals and birds executed under the patronage of the emperor Jahangir. Here one may note an intriguing contrast between the patient fidelity with which the details of the bird’s head, plumage and feet are rendered and the blithe disregard of accurate proportion, texture and colour in the depiction of the rocky wooded environment in which the bird is set. On a lajvardina bottle golden Chinese phoenixes dive and swoop through the branches of a tree in a composition of artful asymmetry (pl. 11). In another quite different approach to filling the space available, a dragon unfolds its sinuous bulk along the pot-bellied curve of a blue and white bottle (pl. 18). Here the impact of China can be felt in both theme and technique. A readiness to borrow from China can be detected in Persian art from at least the ninth century, though Sino-Persian trade flourished along the Silk Road for more than a millennium before this. Landscape and nature In none of the images of human beings, or for that matter animals and birds, do the artists evoke a credibly three-dimensional setting. Landscape in itself never interested artists as a serious theme for painting. The outdoor setting of most scenes comprises an array of well-worn conventional components rather than a specific, fully realised landscape. This perfunctory approach to landscape continues as late as Safavid times, though by now the idiom derives from European models. In classical Persian painting artists made do with symbolic notations that did duty for

386

landscape, shrubs so abstracted that their species is unrecognizable, and indeed they bear impossibly different kinds of leaves – agitated curlicues to suggest water (pl. 9), and jumbled-up heaps of multicoloured rock to suggest mountains. Flowers proliferate, but again they are not botanically accurate. They are almost always in bloom and are thus shown at their best, just as trees are in perpetual leaf. Clouds take on ornamental writhing shapes and otherworldly colours like gold, and may end in pointed streamers of Chinese derivation. In the lajvardina bottle mentioned earlier (pl. 11), behind the convincingly random disposition of branches and leaves another kind of nature makes itself felt – the abstracted curled leafy scrolls known as arabesques, executed for greater clarity in white against the midnight blue ground. Further versions of the same theme serve as decorative bands around the neck of the vase. In much the same way they provide a continuous backdrop for a Qur’anic inscription in deep blue on a lustre tile (pl. 12). Coiled as tight as springs, these scrolls sprout leaves and buds at regular intervals. Similar scrolls, but now producing huge fleshy leaves, hooked, curled and multi-lobed, fill the space around a seated man on a lustre plate already noted (pl. 8); the scrolls are large or small according to the space available, and so they are squashed to a greater or lesser degree. The capacity of arabesque filler ornament to expand or shrink at will, to acquire or shed extra vegetal elements as space dictates, is the secret of its abiding popularity. Despite this endless flexibility, it always contrived to suggest nature, even when it flouted nature’s rules (pl. 21). It could work as well on a large scale as on a small one. It could produce a satisfyingly even coverage of a large field, as in the dense and disciplined scrolling openwork thicket against which the huge letters of a monumental cursive inscription on a Safavid steel panel create banner headlines (pl. 25). Yet on a Safavid animal carpet fragment (pl. 26) the network of tendrils against which various animals or birds disport themselves or savage their prey produces sudden bursts of blossom, sprays or floral cartouches, often of Chinese inspiration and apparently loose and random. The arabesque can even pierce the bodies of birds. It overlies further black tendrils festooned with white blossoms and is in its turn overlaid by slender writhing Chinese cloud forms. Thus multiple palimpsests, mostly of a vegetal nature, fill the space not taken up by animals and birds. The entire carpet pullulates with life, celebrating the abundance of creation, as constantly shifting and competing graphic and chromatic accents entice the eye. The resultant confusion, as genres mix, merge and transform themselves, is deliberate;

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

387

nothing stays still for long. It is a classic case of the horror vacui traditionally associated with Islamic art, something rather different from the even covering of the field, and with the added complication of contrasting levels to hinder an easy reading of the design. Geometry Some forms of arabesque comprise scrolls drawn with such mathematical precision that one suspects a compass was used. This is a good example of the way categories blend and overlap. The circular form of so many bowls and dishes encourages the use of curvilinear geometric forms, again compass-based, even when they ostensibly represent a complex blossom executed in white on blue (pl. 27). Rows of matching ovals (pl. 28) or medallions with chrysanthemum blossoms of Chinese derivation marking their points of contact or intersection create a surface of geometric regularity for many Safavid silks. Angular geometry, for example in the form of interlocking meanders or key patterns, acts as a space-filler in many metalwork objects, as a candlestick shows (pl. 16). Yet geometry had its practical aspects too, as shown in astrolabes, which were used among other things to determine latitude or the direction of the qibla (pl. 30). A very common use of geometry was to suggest infinity, a concept which of course had religious implications. Hence many complex polygonal networks are cut off on all sides by the frame, so that the attentive viewer is left in no doubt that the design continues outside the field in all directions. But to make that discovery it is often necessary to study the pattern with some care. Such study can encourage a clearing of the mind, and hence contemplation at a deeper level. This, too, can of course have a religious dimension. These processes are encouraged by the fact that so much Islamic ornament is easily visible, whether by virtue of its location or of its size. Altogether, the role of Islamic decoration as a means of exercising the mind – whether to decipher a deliberately complex inscription or to understand how a given pattern is generated – is easily underestimated by Westerners, who, accustomed as they are to seeing figural art as the key bearer of meaning, are apt to see ornament as mere wallpaper.

388

Conclusion The discussion above has suggested some possible approaches to the study of this important collection, which is still unfamiliar to the public at large. It has focused on a few outstanding pieces. But those pieces transmit many other messages, and when the entire collection is brought into the picture it will prove to be large and rich enough to sustain all kinds of investigations. It is matter for celebration that a valuable new source for an understanding of medieval Persian art has become available, and these works of art are certain to stimulate much significant scholarship in the years to come.

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

1. Qur’an leaf in Hijazi script

389

2a, 2b. Bifolium from the Blue Qur’an

390

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

3. Samanid plate with inscription

391

392

4. Saljuq silk robe

393

5. Qur’an bifolium in Eastern Kufic script

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

394

6. Laqabi albarello

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

7. Turquoise ewer, Saljuq Kashan

395

396

8. Lustre dish, Saljuq Kashan

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

9. Mina’i bowl, Saljuq Kashan

397

398

10. Saljuq silver vessel

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

11.Ilkhanid lajvardina bottle

399

400

12. Kashan lustre tile with thuluth inscription

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

13. Juz’ of a Qur’an in muhaqqaq script, Mosul, 1306-7

401

402

14. Ilkhanid ivory plaque

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

15. Qur’an in naskh script, Tabriz, 1325

403

404

16. Inlaid candlestick with horsemen, Iran, 14th century

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

17. Bowl with flying bird, Juvain, eastern Iran, 14th century

405

406

18. Bottle with dragon, Iran, 15th century

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

19. Farhad visits Shirin in her chamber, Iran, early 15th century

407

408

20. Bizhan slays Nastihan, so-called “Big-Head” Shahnama, Lahijan, 1493-4

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

21. Gold and niello ring with jade seal, Iran, early 16th century

409

410

22. Safavid textile fragment with boating scenes

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

23. Qaran slays Barman, Shahnama-yi Shahi, early 16th century

411

412

24. The feast of Hatim al-Ta’i, Bukhara, c.1525-50

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

25. Steel door-plaque, probably Isfahan, 1574-5

413

414

26. Animal carpet, Iran, c.1550

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

27. Dish with split palmettes, Kirman, 17th century

415

416

28. Textile panel with ogival design, Iran, first half of the 17th century

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

29. A bird on a rock, painted by Mu‘in Musavvir for his son, Isfahan, 8th June 1686

417

418

30. Astrolabe, perhaps Isfahan, late 17th – early 18th century

31. Lacquerpen-case (qalamdan) by Muhammad ‘Ali ibn Muhammad Zaman, 1717

A LEAF FROM MY ROSE GARDEN: THEMES IN PERSIAN ART

419

XVIII Oleg Grabar, Distinguished Historian of Islamic Art

T

he passing of Oleg Grabar marks a sea-change in the history of Islamic art, for he was its last renaissance man. He did more than anyone else in post-war times to secure widespread recognition for this field in the community of art historians, with its stubborn, indefensible but unrepentant bias in favor of western art. He did so by lectures aimed at general as well as specialist audiences, by charismatic teaching, by a continuous flow of publications over a period of almost sixty years and by sheer force of personality. And he had plenty of charm; he knew how to get along with all sorts of people and find common ground, so that it was a pleasure to be in his company. His range of contacts outside his chosen field was formidable and mirrored the catholic scope of his intellectual interests. His personality was expansive and generous; he was above all gregarious, and collegial to young and old alike. Not for him the proconsular manner. His conversation fizzed with ideas, many of them brilliant, profound and mind-expanding; and if some of them fell flat, there were plenty more to come. You always knew when he was in the room. His particular genius was for teaching. Please read those six words again, or you are liable to miss the essence of the man. For his achievements as a teacher are liable to be unfairly overshadowed by his prodigious productivity as a publishing academic from the age of 23 onwards. He was a natural performer, perhaps not so much as a lecturer—though few scholars could more easily fill an auditorium—but rather in the free and easy atmosphere of a seminar or a general discussion, when he could turn

OLEG GRABAR, DISTINGUISHED HISTORIAN OF ISLAMIC ART

1. Oleg Grabar

421

422

on a sixpence and lead the debate into a totally unexpected direction. He had a well-developed sense of mischief, and was adept at deflating pomposity by a humorous aside. His giggle was infectious; there was often something of the little boy about him. He loved to bounce ideas back and forth and had a way of making students feel that they were equal partners in an intellectual adventure, and that what they had to say was worth hearing. The destination was not important; it was the journey itself that mattered. Every university student, one feels, should have a taste of this intoxicating intellectual excitement, and yet it is a depressingly rare experience. It was Grabar’s playfulness, his intellectual agility and his generosity of spirit that captivated generations of students, many of whom took a course in Islamic art on the off chance and suddenly found themselves hooked. It was as if, almost before they knew what they were doing, they were launched on a doctorate—and Grabar supervised over sixty PhD theses. This is yet another achievement that is hard to parallel, and it is worth asking how he did it. One after another of his doctoral students has reported in private or in public—most recently on the internet— what a pleasure and privilege it was to be taught by him. And since the relationship between supervisor and student is so often, it seems, made not in heaven but in hell, that is quite an achievement. Other supervisors might subject every sentence of a thesis to close scrutiny, red pen poised to skewer a mistake. Not Grabar. His forte was to inspire his students, to give them the sense of embarking on an exciting intellectual adventure. No spoon-feeding; he did them the honor of treating them as grownups. He could sketch the entire outline of a research project in a few visionary and beguiling sentences, spotlighting possibilities galore for future research and perhaps passing on, without fanfare, his own notes and references on the subject. Small wonder that many a student left the room walking on air. He had the priceless gift that is common to all great teachers, that of imparting self-belief, and hence self-confidence, to his students. That is a profoundly unegotistical achievement, and its impact can last a lifetime. People who have had such teachers—and many of us have had that experience at least once in the course of our schooling— never forget them. Naturally enough, his students responded to him in kind. On occasion they expressed their appreciation for a memorable seminar by producing a one-off souvenir. Thus his seminar on Sasanian silver in the mid-1960s yielded a hunting plate with Oleg’s face superimposed on that of the royal

OLEG GRABAR, DISTINGUISHED HISTORIAN OF ISLAMIC ART

423

horseman, while the beasts he was chasing bore the faces of the students themselves. Similarly, his legendary seminar on the Great Mongol Shahnama in the mid-1970s yielded a mock-up of one of the most celebrated images in the manuscript, Iskandar (Alexander the Great) before the Talking Tree, whose fruit was talking heads. The faces of the students in this seminar replaced these heads; Oleg’s own face was superimposed on that of Iskandar; and a serendipitous metrical similarity between his name and that of the world conqueror allowed the text of the page to be adapted so that each reference to “Iskandar” was made to read “Ograbar.” Later still, well into the 1980s, an Ottoman seminar resulted in the image of an Ottoman medal depicting a well-fleshed sultan being unobtrusively doctored so that on closer inspection it turned out to bear the familiar Grabar profile. These touching, light-hearted leg-pulls tell their own story of the affectionate interplay between Oleg and generations of his students. For him, knowledge was a gift to be shared, not a possession to be hoarded. And that is how his vast fund of experience, his international network of contacts and the fruits of a lifetime of reading and research could in an instant be put at the disposal of a neophyte student—and thus change the whole trajectory of his or her professional life. And this was yet another area in which his success rate was phenomenal. Over the years he had the satisfaction of seeing dozens of those students finding employment in universities, museums and galleries across the world, particularly in the United States. No other scholar of Islamic art has come close to exerting this degree of influence on the development of the field, and doing so in such a personal way. And he nourished these relationships. He was brought up by his father to answer his correspondence by return, and he kept that impressive habit to the very end. Although Grabar spent almost his entire professional life in the United States, taking degrees at Harvard and Princeton, he was born in Strasbourg, in northern France (3 November 1929) and his schooling was in that country, with two qualifications from the University of Paris in 1948 and 1950. Indeed, he never lost his slightly exotic accent. His writing bore the marks of the severe French educational system, with a typically tripartite arrangement of the argument, topped and tailed by introduction and conclusion: thèse, antithèse, synthèse. As a native speaker of French, moreover, he had effortless access to French academic circles, and indeed was as much a European as an American. He also spoke Russian easily, and happily mixed with Russian scholars. His father André had fostered this gift of tongues—Oleg described once how the languages spoken at the

424

dinner table at home switched from one day to the next. Such anecdotes reveal that he was almost destined by birthright to be an art historian, though they also suggest that it was not plain sailing to have the doyen of Byzantine art historians for a father. At all events, the apple fell not far from the tree, and Oleg Grabar retained all his life an instinctive feel for Byzantine art. Indeed, that intuitive familiarity may have played its part in his choice of the Umayyad period, and hence the birth of Islamic art in the Near East, as the subject for his PhD at Princeton (completed in 1954 and surprisingly enough never published). It was a fateful choice that would determine the principal long-term trajectory of his scholarly career, from his first great article in 1959 on the Umayyad Dome of the Rock—a precocious masterpiece that made his reputation—to a trio of books on the same subject published in the course of his last fifteen years. He matured early. Already by 1972 he was an absolute star. His lecture that year on the Kharraqan tomb towers (which he never published) at the Oxford Congress of Persian Art was a sell-out, with scholars falling over themselves for a piece of him. When one considers his career as a whole, it is the variety of his achievements that compels admiration. His was a restless spirit. Not content with an exceptionally full life as a teaching and publishing academic, he eagerly embraced unusual challenges that came his way, to all of which he brought his customary efficiency and people skills. These tasks included heading a college house at Harvard for a while, editing the two major journals in his field for many years, running an Umayyad dig in Syria for seven years, making films, mounting an ambitious exhibition of pre-Mongol Persian art, editing the work of other scholars or ensuring that their work was published, and serving on the panel of the Aga Khan Award for Architecture. This last responsibility catapulted him into a world remote from academe, and he reveled in the opportunity it gave him to learn about modern Islamic architecture and to see how it flourished in the most out-of-the-way corners of the Muslim world. He valued the interaction with architects and planners, bankers and conservationists that the annual cycles of the Award brought. His most lasting legacy is of course his published work, and it is worth reflecting on how it was that he achieved so much across such a wide range—a range that made him unique among Islamic art historians. That field has seen such exponential growth over the last thirty years that to keep up with its literature is now too much to expect of anyone. But the resultant increase in specialization inevitably entails the loss of

OLEG GRABAR, DISTINGUISHED HISTORIAN OF ISLAMIC ART

425

those wide horizons over which Grabar presided so comfortably, indeed apparently without effort. He had an eagle eye for a promising but still under-researched subject, and it was his practice to publish his findings as soon as possible after he had done the necessary work. This can be seen for example in his book on the coinage of the Tulunids, the fruit of a busy summer at the American Numismatic Society in New York. He often chose for a graduate seminar a topic that promised well for a major publication; examples include the Friday mosque of Isfahan, Sasanian silver, the Maqamat and the Great Mongol Shahnama. Again and again he picked up a subject which might have seemed sufficiently familiar at the time—such as the Alhambra or the whole issue of ornament—and trained a searchlight on it that quite transformed its character and revealed unsuspected riches. It seems that he never tried to write the definitive book on a topic, but instead typically summarized current scholarship, noted its weaknesses and took the subject a quantum leap further; and the way he did so left plenty of room for others to come after him. In contemplating such an output, one can only heave a sigh of relief that he never seriously pursued his ambition to write a book on the causes of the First World War. In retrospect, the masterpiece among his books is perhaps The Formation of Islamic Art, a difficult book that takes no prisoners and is more suitable for a graduate than an undergraduate readership. Supple and sophisticated, dense in argument and rich in allusion and reference, it is more a work of cultural than art history. It springs from a deeper familiarity with the thought-world of early medieval Islam than any of today’s Islamic art historians possess, and it should be required reading for Islamic historians. Its foundations were laid not only in the work for his doctorate but at the outset of his teaching career, when the University of Michigan, making a fabulously far-sighted investment, gave him three years of very light duties so that he could immerse himself in an ambitious program of academic reading. He said later that those years laid the foundations for his entire scholarly career. He himself had a particularly soft spot for his book on the thirteenth-century illustrated manuscripts of the Maqamat, that masterpiece of salty and picaresque medieval Arabic literature. This monograph was the culmination of three decades and more of thinking, research and teaching on this remarkable body of material to which the legendary Kurt Weitzmann had directed him while he was still a graduate student. Grabar wrote several pioneering articles on this material, but his book of 1984 takes the discussion a good deal further and opens wide

426

horizons. Unfortunately it was produced in an uninspiring format and a small print run by Chicago University Press, and (even worse) was distinctly decaffeinated on the visual side, in that its illustrations were on microfilm only, in black and white. The look of the book did no favors to these paintings, raucous and colorful evocations of street life in medieval Baghdad. He was justifiably disappointed that it caused so little stir among specialists in the field. But then Grabar, for all that he wrote on aesthetics, for example in the two volumes on book painting that he produced in his seventies, was more interested in ideas and context than in the close-focus study of surface detail, or indeed in the objects themselves as works of art. The same could be said of most of the eighty-three articles that he chose to reprint as the cream of his oeuvre in that form. They fill four volumes and will arguably constitute his most lasting legacy. They show him at his most distinctive: as a creative thinker who consistently anchored his speculations in a specific historical context, and who instinctively knew just how high to fly his kites. He tended to use lectures as a vehicle for floating ideas that he might well reject at a later stage, and he was careful not to commit his more baroque ideas to print until he had assembled clinching evidence. By contrast, he used articles to bring the thinking on a topic further forward. In no field is that more true than in that of the Umayyad palaces, perhaps a topic that came his way in his earliest encounters with Islamic art and architecture in France before he was 22. He also wrote happily on demand, especially a number of op-ed pieces on the nature of Islamic art, for he was well aware of the need to reach a wider public than merely his colleagues in the field. Yet he refused to reprint such pieces in his collected articles, which as a result represent less than half of his published output in the form of articles or short occasional pieces. Grabar was married to Terry, herself a professor of English literature, for fifty-nine years. They had two children: Nicolas and Anne-Louise. The lingering death of their daughter in 1988 was an agonizing experience and it kept Grabar away from what he termed “gatherings of the tribe” for a long time. But he assuredly found some solace in his work. After fifteen years at the University of Michigan (1954–69) he moved to Harvard, and when he retired from that university in 1990 he settled at the Institute of Advanced Study at Princeton, where he remained until his second retirement in 1998. But age could not wither him, nor custom stale his infinite variety. His schedule remained as packed as ever, and he was producing books and articles to the very end, which came in Princeton

OLEG GRABAR, DISTINGUISHED HISTORIAN OF ISLAMIC ART

427

on 8 January 2011. His library, possibly the best private collection in the world, went to the Getty; one may hope that this priceless resource will inspire future generations of Islamic art historians on the west coast. Two compendious Festschriften, dedicated to him by his students, were published in 1993 and 2008 respectively; that long goodbye bears collective testimony to the deep loyalty and affection that he inspired. And so the curtain falls on an almost unimaginably rich and productive professional life.

XIX Oleg Grabar: the scholarly legacy Valediction

I

t is safe to say that the flood of reminiscences, obituaries, and various kinds of public necrologies that have marked the death of Oleg Grabar 2 are quite without parallel in the history of Islamic art history. They complement the numerous appreciations of him that were published in 1

1  For a selection of these, see William Grimes, The New York Times, 13 January 2011, A23; Anon., New York Times, 10 January 2011; Dictionary of Art Historians, s.v. ‘Oleg Grabar’ [http://www.dictionaryofarthistorians.org/grabaro.htm accessed 06.04.2012]; Linda Stein, The Trenton Times, 12 January 2011; Anon., The National, 15 January 2011; Anon. [Robert Hillenbrand], The Times, 17 February 2011; Anon., The Daily Telegraph, 14 March 2011; Anon., ‘Oleg Grabar 1929-2011’ – a notably full and even-handed account from the website of The Institute for Advanced Study, Princeton [http://www.ias.edu/ news/press-releases/2011/01/10/grabar accessed 06.04.2012]; Ülkü Ü. Bates, International Journal of Middle Eastern Studies, 43(4), 2011, 777-778; Robert Hillenbrand, ‘Oleg Grabar, Distinguished Historian of Islamic Art’, Iranian Studies, 45(1), 2012, 139-44; and Gülru Necipoğlu, ‘Editor’s Foreword. In Memoriam: Oleg Grabar’, Muqarnas, 28, 2011, vii-xiii. Further anonymous obituaries can be found on the internet. 2   I should like to express my heartfelt thanks to Marianna Shreve Simpson for her swift, frequent and comprehensive help with this paper, which involved her in a lot of work. In particular, she pointed me in numerous directions that I would never have discovered on my own. It is not often that one encounters a colleague ready to provide such extensive, imaginative and selfless help. It is a pleasure to acknowledge how much I owe to Terry Grabar for tactfully putting me right on several details, and – a great encouragement, this – for her whole-hearted support in this sometimes delicate enterprise. I shall also long treasure a wonderfully extended and open discussion with Alexis Grabar, Oleg’s nephew, about his much-loved poncle. And this is the place to say a big ‘thank you’ to the numerous students and colleagues of Oleg who responded to my plea for their views. The sheer volume of that response is a tribute all the more impressive for being offered in private.

429

OLEG GRABAR: THE SCHOLARLY LEGACY

his lifetime, and indeed his own reflections on his career. In the months following his death in January 2011 a series of meetings was convened 5 at which scholars spoke about his work, and the anniversary of his death 3

4

Such as Milo Cleveland Beach’s ‘Introduction’, Eleventh Presentation of the Charles Lang Freer Medal, 5 April 2001, 4-7, and the introductions to the two Festschriften published in Muqarnas, 10, 1993, vii-ix (‘Introduction’) and Muqarnas, 25, 2008, vii (‘In Tribute to Oleg Grabar’), or John S. Bowman, The Cambridge Dictionary of American Biography, New York: Cambridge University Press, 1995, 285. 4  The first place to look is the set of prefaces and introductions to each of the four volumes of his reprinted articles, since these are crammed with illuminating anecdotes and virtually add up to an autobiography: Oleg Grabar, Early Islamic Art, 650-1100: Constructing the Study of Islamic Art volume I, Aldershot: Ashgate, 2005, xiii-xvii and xxi-xxviii; Islamic Visual Culture, 1100-1800: Constructing the Study of Islamic Art volume II, Aldershot: Ashgate, 2006, xxv-xxxvi; Islamic Art and Beyond: Constructing the Study of Islamic Art volume III, Aldershot: Ashgate, 2006, xix-xxv; and Jerusalem: Constructing the Study of Islamic Art volume IV, Aldershot: Ashgate, 2005, xix-xxxi. See also ‘The Practice of Islamic Art History’, interview conducted at the Institute for Advanced Study, Princeton, by Richard Candida Smith, transcribed by the J. Paul Getty Trust, 2000 (Research Library, J. Paul Getty Research Institute, accession no. 940109, box I/45) as part of the ongoing project Interviews with Art Historians, 1991-2002; ‘From the Museum to the University and Back’, Eleventh Presentation of the Charles Lang Freer Medal, 5 April 2001, 9-30; ‘Sixty Years of Scholarship in the History of Art’, The Institute Letter, Princeton: Institute for Advanced Study, Fall 2010, 7; and the presentation volume published by the Aga Khan Foundation on the occasion of his receipt of the 2010 Chairman’s Award for lifetime achievement, including a collection of photographs illustrating the chapter devoted to his life’s work. Of particular interest is a long and reflective account of how the field of Islamic art has changed in his lifetime, and how it is set to change still further, in the speech he made on that occasion, at the Aga Khan Award for Architecture 2010 Award Presentation Ceremony (at Doha, Qatar) on 24 November 2010, less than two months before his death. It can be read as an extended valediction. An energetic, visionary, optimistic openness to the future, and to the opportunities brought by new technologies, an embracing of the relevance of other disciplines, permeates this speech, but it is also shot through with darker undertones. Its sober warnings about the dangers of ignorance and nationalism – one wonders what the impact of his early adolescence in German-occupied France had on the later man – show him in unusually prophetic mode (‘I preach’ comes strangely from him). But this serious tone is appropriate, since his theme was knowledge itself and how to use, cherish and nurture it. The text can be found online [http://www.akdn.org/Content/1037 accessed 06.04.2012]. 5 It will be convenient to tabulate these in chronological order: (a) A panel at the College Art Association meeting in New York on 10 February 2011, entitled ‘The Other Middle Ages: The Medieval Mediterranean as Theater of the Arts’, had been due to feature concluding remarks by Grabar. Its members (organizers William Tronzo and Caroline Bruzelius, and panellists Elisa Foster, Amity Law, Ruggero Longo, Kathryn Moore and Tara Tohme) announced that [we] ‘will dedicate the session to him and all that he accomplished in his lifetime…to honor the memory of a great scholar and a wonderful mentor, colleague and friend’. (b) Presentations at his memorial service, held at the Memorial Church at Harvard on 23 April 2011, by William Graham, Gülru Necipoğlu, Lisa Golombek, Renata Holod, Nasser Rabbat, Nicolas Grabar, Neil Levine, Marianna Shreve Simpson, Giles Constable, 3

430

was marked by a symposium in Istanbul to celebrate his contributions with Massumeh Farhad reading excerpts from the Shahnama in English translation. Thomas Lentz added his tribute at the reception that followed, where Professor Richard Frye – so Shreve Simpson tells me – also held forth extemporaneously and with great verve about his old friend. These are the place to start: deeply felt, carefully considered valedictory reminiscences that, taken together, give much of the measure of this multi-facetted man. The texts have been compiled and presented online at the website for the Aga Khan programme for Islamic architecture at Harvard: ‘In memoriam Oleg Grabar, November 3, 1929 – January 8, 2011’ [http://agakhan.fas.harvard.edu/fs/docs/icb.topic888988.files/ Oleg_Grabar_Remembrances_04-23-2011_.pdf accessed 06.04.2012]. (c) A gathering held under the auspices of the Historians of Islamic Art Association (HIAA) took place on 2 December 2011 at the Middle East Studies Association conference in Washington DC, at which, under the rubric ‘Remembering Oleg Grabar’, ‘friends and colleagues, led by Jere Bacharach and Jacob Lassner’, were ‘invited to celebrate the life and contributions of MESA Founding Member and Honorary Fellow, Oleg Grabar’. (d) A full-scale session at the MESA conference on 3 December 2011, also organized by HIAA and entitled ‘Oleg Grabar’s contributions to the cultural history of the Near and Middle East’ (chair: Renata Holod; discussant: Nasser Rabbat). The speakers were: Prudence Harper, ‘“Brilliant Impressions”: Sasanian Silver Revisited’; Richard Bulliet, ‘Perspectives on Grabar’s Formation of Islamic Art’; Benjamin Z. Kedar, ‘Oleg Grabar’s Vision on the Past, Present and Future of Jerusalem’s Old City’; and Stephen Humphreys, ‘Grabar and the Socio-Political Milieu of Art Production’. (e) A special session at the CAA meeting on 23 February 2012 in Los Angeles, again organized by HIAA and entitled ‘Oleg Grabar’s impact on the practice and history of art’. Noting that Grabar’s death ‘inevitably leads to reflections on his immense intellectual legacy, including his contributions to and impact on both the practice of art and the study of art history beyond his own areas of specialization’, this session was planned to examine ‘selected aspects of Professor Grabar’s myriad contributions to the discipline writ large, with presentations by a prominent artist and three leading art historians influenced by his ideas and scholarship’. Chaired by Marianna Shreve Simpson, with Larry Silver as discussant, the panel featured papers by Larry Nees, ‘Border Problems: Oleg Grabar and Medieval Art in the Western Mediterranean’; Nancy S. Steinhardt, ‘Domes of Heaven Reconsidered’; and Margaret Olin, ‘Oleg Grabar in Conversation about Ornament with Alois Riegl, E.H. Gombrich and Me’. A fourth projected paper, by Philip Taaffe, ‘Oleg Grabar and Contemporary Art: A Painter’s Perspective’, was not delivered; instead Simpson and Silver ‘performed’ excerpts of an interview between Grabar and Taaffe published in 1994. (f) A specially convened session of the American Oriental Society held on 17 March 2012, organized and chaired by Renata Holod and sponsored by HIAA, with presentations by Judith A. Lerner, ‘Oleg Grabar and the Lure of Sasanian art’; Deborah Klimburg-Salter, ‘Zones of Transition: Reconsidering Early Islamic Art in Afghanistan’; and Jacob Lassner, ‘Oleg Grabar and the legitimization of Islamic art and architecture in the American Academy’. It is also relevant to mention that a campaign, inaugurated by an HIAA committee (comprising Massumeh Farhad, Renata Holod and Marianna Shreve Simpson), has been launched to raise funds for the Oleg Grabar Memorial Fund to support grants and fellowships. So far it has raised over $150,000; further information can be found online [http://www.historiansofislamicart.org/Home/Oleg-Grabar-Memorial-Fund.aspx accessed 06.04.2012].

OLEG GRABAR: THE SCHOLARLY LEGACY

431

to the understanding of Turkish Islamic art. Other great figures in the field of Islamic art have had their full meed of honour, with memorial 7 8 services and colloquia, and tributes from the great and the good, as well as obituaries not only in academic journals, where one would expect to 9 find them, but also in broadsheets. But the reaction to Oleg Grabar’s death has been at once more widespread and more profound than this. The sense that an era has ended runs through many of the comments made in both public and private. The obvious question – ‘why?’ – does not have a single obvious answer. It has several, and at times they may seem to contradict each other. Most of his younger colleagues have emphasized above all the unforgettable impact of his colourful and multi-sided personality, and more than one of them has noted that it was hard, on reflection, to disentangle his 10 personality from his output. For while he expressed himself with equal ease in both the spoken and the written word, it was talking that was essentially his instrument of suasion. That was what made his lectures 11 so memorable; but perhaps it worked best in free-ranging one-on-one conversations. For the period of that conversation you knew you had his 6

6 A seminar entitled ‘Oleg Grabar’s Contributions to the Study of Turkish Islamic Art and Architectural History’ was held in Istanbul on 8 January 2012, the first anniversary of his death. According to its organizers, who drew attention to the fact that Grabar had served on the Advisory Board of Sakıp Sabancı Museum when the museum was being founded, it featured ‘lectures by leading academicians and museum curators who are his former students’, namely Esin Atıl, Ulkü Ü. Bates, Ayda Arel, Gülru Necipoğlu, Tülay Artan, Scott Redford, Oya Pancaroglu, Çigdem Kafesçioglu and Barry Wood. 7 For example, the memorial services held for Richard Ettinghausen in Princeton in May 1979 or for Ernst Grube in London in October 2011. 8 Such as that held in memory of Ettinghausen at the Institute of Fine Arts, New York University, on 2-4 April 1980, planned and organized by Carol Bier and later published: Priscilla Soucek, ed., Content and Context of Visual Arts in the Islamic World, University Park and London: Pennsylvania State University Press, 1988. 9 For example, Ernst Grube (The Times, 13 July 2011) and Edmund de Unger (The Daily Telegraph, 16 February 2011). 10 As Eva Hoffman noted: ‘You ask about his scholarship “not just his personality”. In Oleg’s case, I am not sure if the two are divisible. His intellectual excitement, his ability to give students the benefit of the doubt and the breathing space to be creative and adventurous – these are the tangible, lasting models of “scholarship” that he imparted’ (personal message). 11 Ed Keall gives an impression of Grabar’s teaching in 1967: ‘Sometimes he was difficult to follow because his thoughts were going so fast that his words could hardly keep up with the ideas he wanted to express. Above all I had never encountered anyone before who wanted to challenge long entrenched ideas and so went back to the basics, starting with the textual evidence’ (personal message).

432

unflagging attention. And you left the meeting with a full tank. That is 12 quite some gift for a teacher to possess. His scholarly output: general reflections But of course most Islamic art historians alive today did not know Grabar 13 personally, and it is here that what he wrote ought in theory to come to the 14 fore. The purpose of the present paper is precisely to try to assess more what he wrote than what he was – though the sheer weight of personal reminiscence from those who knew him has made that a difficult task. There is, moreover, a further obstacle to that apparently simple project thanks to the many listings of his achievements: the honours, awards and prizes heaped on him, the films he made, the exhibitions he curated, his membership of prestigious academic societies across the world, his 15 key work as editor of the major journals in his field, the positions of responsibility that he held, the institutions or enterprises that he directed and – above all else in its direct human impact – the students he taught and the practical ways in which he encouraged so many of them to stay in the field. While it is thoroughly appropriate to celebrate these multifarious successes, they can easily have the unintended effect of casting his publications into the shade. And yet for decades past, and especially outside 12 Thomas Lentz has brought out very well his distinctive qualities in this respect: ‘As a teacher, he was fair, honest, demanding and a possessor of a powerful work ethic. He was, above all, a person of absolute transcendent curiosity—he was seemingly interested in everything… He had the remarkable ability to seek out the individual strengths and interests of each of us and to nurture them. I never once saw or heard him be dismissive of other students; as someone genuinely interested in cultivating young minds and critical faculties, he produced a staggering array of students with interests as wide and varied as the Islamic world itself ’ (footnote 5[b] above). 13 But he remained a magnet for colleagues young and old until the end, not least in his office at the Institute for Advanced Study in Princeton, and his indefatigable travels to what he himself termed ‘the gatherings of the tribe’, coupled with his natural gregariousness and affability, led to informal meetings beyond count. At the HIAA 2010 symposium at the Freer/Sackler, for example, he was constantly surrounded by young people, including M.A. students. 14 As Eric Broug, who has written at length on geometry in Islamic art, wrote: ‘he helped me make a lifetime commitment to Islamic art and architecture, for which I will be forever grateful, and indebted to him’. Remembering Oleg Grabar (1929-2011) [memoryog. tumblr.com accessed 06.04.2012]. 15 He was in fact the founding editor of Muqarnas, currently the principal journal dedicated exclusively to Islamic art. And Volumes II-VII of Ars Orientalis, which are overwhelmingly Islamic in content, and were produced under his editorship in the Michigan years, are a monument not just to his industry but also to his ambition for the field to develop and to his ability in attracting scholars of stature to contribute important articles.

OLEG GRABAR: THE SCHOLARLY LEGACY

433

the United States, it has been his written work that has propelled him to pole position on the international stage among Islamic art historians. In his own words, set down a few months before his death, that written work 16 comprised, apart from material that had not yet appeared, ‘some twenty books, several of which were translated into at least seven languages, and 17 over one hundred and twenty more or less significant articles’. That latter number, incidentally, represents a very severe judgment on his part of the long-term value of a good deal of his published articles, chapters, and occasional pieces. And since the material he selected for republication in the four volumes of his ‘collected’ – it should be ‘selected’ – articles comprises only eighty-three items, it follows that he rejected a third even 18 of those that he himself regarded as ‘more or less significant’. He was thus a harsh critic of his own work, particularly of his shorter and general 19 papers. Some of his books, like some of his articles, were admittedly of 20 uneven quality and depth, especially in his later years. But they had an 16 He noted then that ‘there are still now two or three studies in the process of being printed or ready to appear on the Internet’ (Oleg Grabar, ‘Sixty Years of Scholarship in the History of Art’, The Institute Letter, Princeton: Institute for Advanced Study, Fall 2010, 7). Something very close indeed to a full bibliography of his work can be constructed by putting together the bibliographies found in the two Festschriften produced in his honour: Muqarnas, 10, 1993, ix-xiii and Muqarnas, 25, 2008, viii-x, supplemented by Muqarnas, 28, 2011, xv. A significant missing item here is ‘Al-Mushatta, Baghdad and Wasit’, in R. Bayly Winder, ed., The World of Islam (Studies in honour of P.K.Hitti), London: Macmillan, 1959, 99-108. 17 Grabar, ‘Sixty Years of Scholarship’, 7. 18 The raw count of his articles and contributions to books, excluding one-page introductions and book reviews (some of which were substantial essays in their own right), is 199. Material which appeared after his death or is still in press brings that total to over 200. So his ‘collected’ articles represent significantly less than half of his oeuvre in that format. It almost goes without saying that some much-admired pieces do not make it into the final cut.  19 Renata Holod called him ‘fiercely critical’ and Thomas Lentz praised ‘his fundamental honesty’ and rigour (footnote 5[b] above), a reminder that Grabar’s judgment on his own work should be final. 20 A student of mine, Katherine Rose, whom it is a pleasure to thank for her help with this paper, wrote a fifty-one-page Edinburgh University undergraduate dissertation in 1999 entitled Ornament or Art? An Examination of Oleg Grabar’s Challenge to the Boundaries of Islamic Art History, and sent a copy to Grabar. In a letter to her dated 23 December 1999 (the day before Christmas Eve!), Grabar noted ‘My last two major books (Mediation and Shape) are at the same time the most flawed and the most exciting’ and wrote of Sasanian Silver ‘There are few books I enjoyed putting together as much as that one’. (Grabar, The Mediation of Ornament (The A.W. Mellon Lectures in the Fine Arts, 1989), Bollingen Series XXXV 38, Princeton: Princeton University Press, 1992; The Shape of the Holy: Early Islamic Jerusalem, Princeton: Princeton University Press, 1996; and Oleg Grabar and Martha Carter,

434

impressive reach. Nor should one underestimate their cumulative impact. The translations of The Formation of Islamic Art and of The Alhambra in 21 particular gave those works a truly international readership, while in other cases he either published a work first in French and then reworked it in 22 English, or vice versa. This practice showed, as it were, his dual intellectual 23 citizenship. It seems that certain trains of thought, and idioms, came to him more easily in French. But if ever there was a truly international scholar, it was he. It is telling that he described himself as a man without 24 a country, except for academia. As in what he said, so in what he wrote one may discern a wide range of registers. It seems that he was as happy writing op-ed pieces about 25 the nature of Islamic art, or popular surveys, or quasi-philosophical tours Sasanian Silver: late antique and early mediaeval arts of luxury from Iran, Ann Arbor: University of Michigan Museum of Art, 1967.)  21 Oleg Grabar, The Formation of Islamic Art, New Haven: Yale University Press, 1973, 1987; and The Alhambra, Cambridge, MA: Harvard University Press, 1978. The absence of translations into Arabic is noteworthy. It suggests that regrettably there still does not exist a significant arabophone readership for his work – and by extension, that of the wider community of Islamic art historians. In this respect Iran and Turkey are well ahead of the Arab world. 22 For example Oleg Grabar, Peinture Persane: une introduction (Paris: Presses Universitaires de France, 1999) and Mostly Miniatures. An Introduction to Persian Painting (Princeton: Princeton University Press, 2000; see the judicious review article by Bernard O’Kane: ‘Lifting the Veil from the Face of of Persian Painting’, Oriental Art, 48(2), 2002, 55-60); or the complex interrelationship between The Mediation of Ornament (1992), L’Ornement: Formes et Fonctions dans l’Art Islamique (Paris: Flammarion, 1996), and Penser l’Art Islamique. Une Esthétique de l’Ornement (Paris: Institut du monde arabe, 1996). He himself described the latter book in a letter (footnote 20 above) as ‘an introduction with a twist’. An article in German is also relevant here: ‘Das Ornament in der islamischen Kunst’, Zeitschrift der Deutschen Morgenländischen Gesellschaft, supplement 3(1), 1977, xli-liv. 23 For some examples, see Marianna Shreve Simpson (footnote 5[b] above). Scott Redford adds another: Grabar – who was, he thought, ‘worn down by the many teaching and other demands on his time’ – confessed to him in 1982 that he was ‘tournant en rond’ (going round in circles) at the time (personal message).  24 Personal communication from Terry Grabar. Note too the astute observation of his son Nicolas, who terms him ‘this citizen of everywhere and nowhere’: ‘His lifelong modus operandi was to be an outsider, finding a vantage point outside any specific culture, or beliefs, or allegiances. So as a child among the French he was a Russian, and as an adult among the Americans he was a European, and of course his professional life was devoted to a culture that was fundamentally not his own. His only patrie or fatherland was the academy, and I’m quite sure the only uniform he ever put on—literally, or figuratively— was cap, gown and doctoral regalia’ (footnote 5[b] above). 25 His compact survey of the entirety of Islamic art (‘Islamic Peoples, Arts of,’ Encyclopedia Britannica, 15th ed., 1974, 952-1011), has regrettably attracted very little notice; but this piece has particular value as his principal attempt – undertaken early on in his very full teaching career – to distil that teaching experience and to highlight the core

OLEG GRABAR: THE SCHOLARLY LEGACY

435

d’horizon, as he was with the close-focus treatment of a source, an event 26 and its implications, an object, or a monument. Whatever the register, he 27 delighted in stretching his readers, and to that end questions galore filled 28 the pages of his books and articles. Grabar himself suggested rather different categories for his writings, and characteristically divided them, like Caesar’s Gaul, into three parts, a division that was also, as he saw it, chronological. The latter is a crucial observation. First came ‘traditional research based on the publication of documents, the excavation of new documents, and the significance of 29 these documents within relatively strict chronological and spatial limits’. He acknowledged that such work had a limited audience. For many scholars, of course (though emphatically not for him), that is a matter of no concern whatever. Second came work that ‘seeks a larger public, as it tries to interpret works or periods of Islamic art as historical, aesthetic, social, or cultural models whose meanings extend beyond their specific characteristics of Islamic art. Interestingly enough, in the very same year he published a shorter digest of one aspect of the same subject: ‘Architecture’, in Joseph Schacht and Clifford Edmund Bosworth, eds, The Legacy of Islam, Oxford: Oxford University Press, 1974, 244-73. It is possible that parts of these two contributions were deliberate drafts for Grabar’s section (namely architecture) of the Pelican volume commissioned by Nikolaus Pevsner. This may suggest that it was Ettinghausen rather than Grabar who was responsible for the long delays in the latter project. The article for the Encyclopaedia Britannica is worth closer study than the one for The Legacy of Islam not only because it is very much longer but also because the brief to which he worked was clearly to cover the ground methodically. That closed off the option of omitting areas that did not interest him, or doing a ‘think piece’. Here, then, are his views on how Islamic art as a whole developed. 26  Such as the sack of the Fatimid treasuries in 1069 (‘Imperial and Urban Art in Islam: The Subject-Matter of Fatimid Art’, in André Raymond, J.M. Rogers and Magdi Wahba, eds, Colloque International sur l’Histoire du Caire, Leipzig: Gräfenhainichen, 1973, 183-186), or the mutilation of a Christian image by an early Muslim iconoclast (‘Islamic Art and Byzantium’, Dumbarton Oaks Papers, 18, 1964, 69). 27 As Jane Jakeman put it: ‘he always made one jump forward trying to keep up!’ (personal message). 28 Glenn Lowry reflects on ‘his unique ability to frame a question, to probe an idea, and ultimately to posit an interpretation and reveal a narrative that was not immediately apparent – a methodology that was not circumscribed by his own areas of expertise. It is not surprising to me that many of his students ended up outside the academy because the way Oleg approached a problem taught us all how to think through questions regardless of subject’ (personal message). Cf. Stephennie Mulder’s comment: ‘he was much more interested in the question and in provoking a reaction than in finding any definitive answers’ [memoryog.tumblr.com accessed 06.04.2012]. 29 This kind of research, he wrote, ‘dominates the first half of my creative years’ but the output of such works ‘has clearly diminished with time, even though their scientific quality (or weakness) tends to remain steady over the years’ (Grabar, ‘Sixty Years of Scholarship’, 7). 

436

context; it deals with issues like new forms, ornament, and aesthetics.’ The third category, largely published in the course of a remarkably productive Indian summer that lasted from his retirement from Harvard until his death, embraced ‘reactions to requests for introductions or conclusions to the publications of others, or remarks about the work of others reflecting the current fad of colloquia and multi-authored publications’, and he freely admitted that these pieces were of variable quality. On the other hand, he noted – with regret, but not, I think, waspishly – ‘they do fit with a contemporary mood that is less concerned with exploring new things than with stating appropriately old ideas and 31 recalling well-known monuments of art’. And out of this third category developed, again in his own words, an interest in both ‘historiography… the importance of past scholarship and the often strange characters of past 32 scholars’, and, following the need to impose order on his personal papers, ‘personal historiography’ which ‘has now led me to study the history of my 33 family and ancestry’. His astonishing range of contacts of course made him the obvious person to undertake a serious study of the historiography of Islamic art, but he never did so and indeed consistently adopted a notable discretion in discussing colleagues of his own generation and the 34 one immediately preceding it. This could be exquisitely frustrating: he knew, and had opinions about, far more than he would divulge in print. 30

30 Grabar wrote that this ‘is a smaller group than the first one and does not really flourish until the seventies of last century, but it attracted the attention of historians of art and other scholars in many different areas and helped to strengthen the notion that, when one deals with the arts, approaches and conclusions can be extended from one culture to other traditions’. He noted that the resultant globalization of Islamic art had attracted some criticism (Grabar, ‘Sixty Years of Scholarship’, 7).  31 This trend had ‘dominated the past twenty years’, i.e. 1990 to 2010 (Grabar, ‘Sixty Years of Scholarship’, 7); it may be significant that it begins with his retirement from active teaching.  32 His published writings on this subject are all too rare. For a few examples, see his contribution to the memorial service for Richard Ettinghausen held at Princeton in May 1979 (privately printed) and the obituary cited in footnote 76 below; ‘K.A.C. Creswell and his Work’, Muqarnas, 8, 1991, 1-3; and his obituary of Robert Hamilton (The Independent, 6 October 1995); ‘Michael Meinecke and his Last Book’, Muqarnas, 13, 1996, 1-7; and ‘Jean Sauvaget’, in Eric M. Meyers, ed., The Oxford Encyclopedia of Archaeology in the Near East, 4, 1997, 495-496.  33 See the comments of his son Nicolas at the memorial service (see footnote 5[b] above).  34 Perhaps that was just my own experience; apparently he spoke more freely in interviews taped by Betsy Sears and Marianna Shreve Simpson in 2005 and 2007. And of course he loved to gossip. Giles Constable hints as much: ‘He was also a marvellous

OLEG GRABAR: THE SCHOLARLY LEGACY

437

These, then, are the broad divisions of his scholarly output. And, no matter how one chooses to categorize his written work, from now on it is plainly his writings, not his larger-than-life personality, that will constitute his major heritage. Grabar’s range and skills But neither his personality nor his writings tell the whole story or are quite enough to explain the sense of loss, the sense of an era that has ended, that I mentioned at the outset of this paper. And indeed several scholars have highlighted the indefatigable intellectual curiosity that drove him to investigate the highways and byways of a world which to him was one and in which all kinds of aspects were potentially of interest. Nihil islamicum 35 mihi alienum puto, he might justifiably have said. His range was astonishing, and perhaps it is not strange that so few scholars have sought to follow that example. Of course he was formidably equipped for his scholarly role by birth, by upbringing, by his severe French education, by his languages and – perhaps more intangibly – by his air of European sophistication. And his intuitions were very sharp. He knew how far to go. So he was a hard act to follow. That range meant that it was no surprise that he was interested in the interplay between the Islamic world and other cultures, from Europe to China; that he had a taste for the latest intellectual fashion or theory; that he embraced the impact of new technologies on the field of Islamic art; that he threw himself into the work of the Aga Khan Award, which catapulted Islamic architecture into the here and now and where he relished the interaction between academics of various stripes and architects, planners, bankers, administrators and heritage specialists. For his colleagues and his students there was something exhilarating and inspiring in being able to connect with such broad sympathy, which could act like an electrical charge in the instant intellectual energy that it generated. The habit of ‘thinking outside the box’ came to him as naturally 36 as breathing; he tried to teach it to his students, many of whom have 37 put it to exemplary use in their roles as, for example, museum directors. conversationalist and raconteur, though not all his stories would bear repetition here’ (footnote 5[b] above). 35 A paraphrase of Terence: nihil humanum mihi alienum puto (‘I consider nothing that is human alien to me’).  36  This is a phrase that so many people have used of him that it must be regarded as a kind of trademark. 37 Thomas Lentz, a museum director himself, quotes a true gem from the treasury of Grabar humour: ‘we used to send all the dummies into museum work, but that might be

438

So it is the sense that an eager, questing intellect – nourished by many decades of thinking, study and debate, and ready to turn its power onto any number of issues and problems – is now no more that has left so many people forlorn. It is worth stressing that even among his teachers, scholars who for the most part had been born in the nineteenth century, when the field of Islamic art scarcely existed as an intellectual construct, that kind of wide-ranging sympathy was not the norm. Look at Creswell, for example, for whom the doors of Islamic architecture shut with a clang at the IraqiIranian border. Or Marçais, king of Maghribi architecture but of nowhere else. Or Gabriel, the grand old man of Turkish architecture alone. Or Migeon, who wrote on the decorative arts but for whom architecture was invisible. Or Gray, of whom the same could be said. Even the greatest of Grabar’s immediate predecessors, Sauvaget and Herzfeld, devoted most of their mental capital to architecture, and pre-eminently to the Near East, and the occasional excursus that they made into the so-called minor 38 arts, though illuminating, remains teasingly isolated in their oeuvre. The foremost historians of Spanish Islamic art remained both physically and mentally immured in al-Andalus, with a consequent loss of perspective that is incalculable. Specialists in the architecture of India and of Central Asia offer further examples of this melancholy phenomenon. The great exception, of course, was Richard Ettinghausen, of whom more anon. And it is an unfortunate but ineluctable by-product of such specialization that it leads to a jealous consciousness of turf, and thence to vicious practices. This reaction is peculiarly inappropriate in a field as young as Islamic art, where there is more than enough research material for all, not only in the present generation but even far into the foreseeable future. Grabar managed to suggest that there was a huge banqueting table laid out before the budding scholar: an intoxicating and liberating prospect. Will that attitude continue to thrive? Not if turf wins. Where turf is guarded, the instinct is to repulse interlopers rather than to welcome visitors. One of the many charms of Oleg Grabar was that he was blessedly uninfected with this disease. So he gave the best of examples to his students and colleagues alike. The way that he encouraged others to disagree with him about the Dome of the Rock is an example of this easy affability and changing’ (footnote 5[b] above). 38 Jean Sauvaget, ‘Remarques sur l’art sassanide: questions de méthode à propos d’une exposition’, Revue des Études Islamiques, 12, 1938, 113-31; Ernst Herzfeld, ‘A Bronze Pencase’, Ars Islamica, 3, 1936, 35-43. 

OLEG GRABAR: THE SCHOLARLY LEGACY

439

open-mindedness. He welcomed outsiders. He did not take it ill when 41 students challenged him. That lack of a proprietorial instinct meant in turn that he did not invest undue emotional resources in this or that theory or body of material. This practice created in him a certain tranquillity or equability that robbed even harsh criticism of its full bite. It is instructive to note that only his first major book, The Formation of Islamic Art, triggered a certain degree of opposition (see below). Nor is that surprising. In 1973, at 43 years old, he was still the coming man, and could thus expect to take his fair share of negative criticism. With the passing of the years, however, his position as doyen of the field gradually became impregnable, and this certainly shielded him to some extent. Such immunity is well understood as one of the perks of power. But I can testify from personal experience that he really was able to take negative criticism in good part. His particular skill – and one that practically never failed him, since he was constantly honing it on new material – was to remain undaunted before the sheer, smoothly polished cliff-face of conventional wisdom and to detect tiny fissures, cracks and hand-holds in it. He was thus able almost literally to climb up and over this kind of obstacle to original thinking and then to suggest hitherto untried approaches even to longfamiliar material. He was equally capable of ignoring conventional wisdom altogether and casually proposing some radically new approach. He had 39

40

Larry Nees tells me that when his book on early Islamic art in Jerusalem is finished, he will dedicate it to his wife, first and foremost, and to Grabar ‘even though he knew I would be introducing material about the Dome of the Rock that he had simply overlooked, and ultimately suggesting interpretations rather different than his own. He was not just accepting of this situation, he was actually excited about it!’ (personal message).  40 Larry Nees (footnote 5[e] above): ‘He liked having an outsider’s perspective, and even asked me to review the chapters on early Islamic art in the manuscript of the revised Pelican History of Art volume. He was unfailingly encouraging to me, and I think to other non-Islamicists who for one reason or another came to teach the material so close to his heart. He was never one to patrol borders, and was as conscious of the dangers of specialization as he was…of too-easily adumbrated comparisons or linkages.’  41 Scott Redford, for example, found that some of Grabar’s ‘theories that were spun out of very little often took on the weight of truth once they had entered the literature. So, I became the more pedantic scholar that I am. When writing my dissertation, he and I several times had cordial disagreements about this, with Oleg always insisting on the need for “trajectory” of an argument or another of mine that he found too timid. I responded on one occasion that an argument is like a chain, only as strong as its weakest link’ (personal message). It is worth remembering that a good French dressing needs vinegar as well as oil; in the same message Scott Redford pays tribute to how Grabar’s ‘teaching and scholarship taught me the value of ideas, and the need to communicate to a larger audience what was exciting about the field, and I strive for this, but try to document and substantiate more firmly my points’.  39

440

the poet’s capacity to spotlight the links between apparently unconnected, 42 disparate ideas. It was continuously exciting to see him exercise this faculty – he did it smoothly, without apparent effort but also not in a facile manner. And above all, such unexpected felicities, insights or angles of approach arose naturally out of the discussion that was taking place. These were not carefully polished aperçus prepared well in advance. Among his books, one may mention The Alhambra and The Great Mosque of Isfahan as examples of how his broad sympathies and his profound erudition in cognate, collateral fields could bring these over-exposed masterpieces to life in unexpected ways, for he saw them so to speak through the other 43 end of the telescope. That position revealed insights denied to scholars 44 who knew these buildings far more intimately than he did, but could 45 not contextualize them so richly. No wonder that his lectures were packed by ‘embryonic scholars from many disciplines listening with rapt 46 attention’. His messages were beamed at wider audiences than the tightlyknit community of historians of Islamic art, even if his impact outside

Cf. the comment of Larry Silver: ‘Grabar’s own ability to perceive relationships between the specific and the general, between the creation and the creating society, endured as the hallmark of his own work’ (footnote 5[e] above). 43 Oleg Grabar, The Alhambra, 1978; and The Great Mosque of Isfahan, London: I.B. Tauris, 1990. Giles Constable recalls that Grabar was well aware ‘that his ideas on western Islamic art might seem to some readers too tentative and unclear, “but they will have fulfilled their purpose (he wrote) if they inspire or irritate others to come up with alternative ones”. This is vintage Oleg—to inspire or irritate—and expresses the essence of his qualities as a conversationalist, and also, no doubt, as a scholar and a teacher’ (footnote 5[b] above).  44 As Sheila Canby notes: ‘Grabar concentrated more on determining whether the monument and its decoration were typical of palace architecture across the Islamic world or if the monument is unique. In the process of treating this problem, he provides a description of the monument and interpretations of its form and decoration. Neither here nor in his other publications does he claim to present the definitive answer to all the questions about the subject at hand, in this case the Alhambra. Yet, by discussing its inscriptions and placing them and the rooms in which they appear in the broader context of Islamic architecture, he leads us to logical conclusions about the uses and meaning of this palace.’.(Personal message.)  45 Larry Silver compares him to Otto von Simson on Gothic cathedrals, or Rudolf Wittkower on Renaissance Italian architectural principles, in his ability ‘to situate wider cultural resonances of architectural symbolism in Islamic legends of power, going back to biblical King Solomon himself ’ (footnote 5[e] above). 46 As Larry Silver wrote; he goes on to explain that Grabar stood high on the ‘list of must-reads’, for ‘He…refashioned—and greatly expanded—the range of questions and methods we all could employ one day as tools, regardless of period, regardless of field’ (footnote 5[e] above).  42

OLEG GRABAR: THE SCHOLARLY LEGACY

441

that field was not as powerful as he deserved. But within it he well knew 48 where to concentrate his fire-power and what to avoid. The classic case of his rare capacity to transcend conventional wisdom is his precociously magisterial article on the Dome of the Rock, published 49 in 1959. It is worth remembering that his dissertation was not about that building, so this was fresh research. It was a piece that became legendary, and beyond doubt it has been the most widely cited article in the entire forty-three volumes of Ars Orientalis. Who would have thought that there was so much still to say about the Dome of the Rock, which in 1959 was the most widely published single building in all of Islamic architecture? Yet here was a big new article about a big old building. How revealing that Creswell, whose bibliographical mania was a byword, managed to find no room for this seminal article in his exhaustive bibliography of scholarship on the Dome of the Rock, published ten years after Grabar’s article and 50 running to 334 titles ending in 1964. A back-handed compliment indeed. And Grabar went on mining this mother-lode for the rest of his career, with articles galore; appropriately enough, his very last single-authored 47

47 Cf. Larry Nees, (footnote 5[e] above): ‘Oleg Grabar’s impact on the development of scholarship about medieval art in western Europe has been more indirect, and more difficult to trace and assess, than might be expected…Probably most important has been the influence of his teaching, not only upon those who would continue to study Islamic art but also upon the many others who studied with him but concentrated in other areas, notably the arts of medieval Spain and southern Italy’. 48 Cf. Nancy Steinhardt’s comment (footnote 5[e] above): ‘one of Oleg’s countless strengths was that he chose research problems that were conceptually and intellectually worthy of him’. As to what he avoided, he was (justifiably, as it turned out) cagey about the Andarz Nama, a faked illustrated manuscript that purported to date from the eleventh century: ‘Remarks’, in Arthur U. Pope and Phyllis Ackerman, eds, A Survey of Persian Art from Prehistoric Times to the Present. Volume XIII – Fascicle. Addendum A – The Andarz Nama. Proceedings of the IVth International Congress of Iranian Art and Archaeology, April 24 – May 3, 1960, Tehran, London, New York and Tokyo: Asia Institute of Pahlavi University, 1968, A/64-A/65 (another item not to be found in his published bibliography). Similarly, he avoided any detailed engagement with the Buyid silks and the Alp Arslan salver. He devotes a few sentences to the silks (‘whose authenticity …seems to me to have been in recent years proved for the majority if not all of the known fragments’: see ‘The Visual Arts’ in R.N. Frye, ed., The Cambridge History of Iran. Volume 4. The Period from the Arab Invasion to the Saljuqs, Cambridge: Cambridge University Press, 1975, 359), and even less to the salver, which he uses to illustrate the proposition that in the history of Iranian art the position of securely dated pieces produced between 1050 and 1150 is quite unclear: ‘The Visual Arts’, in J.A. Boyle, ed, The Cambridge History of Iran. Volume 5. The Saljuq and Mongol Periods, Cambridge: Cambridge University Press, 1968, 643-4. 49 Oleg Grabar, ‘The Umayyad Dome of the Rock’, Ars Orientalis, 3, 1959, 33-62.  50 K.A.C. Creswell, Early Muslim Architecture, 2nd ed., 1(1), Oxford: Clarendon Press, 1969, 124-9. 

442

book was devoted to that same building, and it is the centrepiece of his 52 last edited work. 51

Areas of lesser interest to Grabar What of the inevitable gaps? While no one could dispute the sheer range and depth of Grabar’s expertise, or doubt that his restless curiosity led him down all manner of byways, some of which were inevitably blind alleys, it is worth reflecting on the areas that he tended to leave relatively unexplored. The word ‘relatively’ is important in this context, for the many occasions on which he wrote in general terms about the nature of Islamic art ensured that he made some comment or other even about those areas of the field that had never captured his enthusiasm. Carpets, for example, seem to have left him cold, and perhaps he never fully engaged with Iznik 53 ware or Mughal painting. You can’t like it all. But perhaps the major perspective which, while not altogether missing, is nevertheless markedly underdeveloped in his work is the impact of 54 religion. This may sound like a paradoxical comment to make about a scholar who wrote so widely and so often, even in book form, on 55 mosques and on buildings whose very raison d’être was largely religious, 56 like the Dome of the Rock or, more generally, funerary architecture. It is paradoxical also in a more personal sense, for several tributes to him

Oleg Grabar, The Dome of the Rock, Cambridge, MA: Harvard University Press, 2008.    See Oleg Grabar and Benjamin Z. Kedar, eds, Where Heaven and Earth Meet. Jerusalem’s Sacred Esplanade, Austin: University of Texas Press, 2009. 53 But see his paper ‘About two Mughal Miniatures’, Damaszener Mitteilungen, 11, 1999, 179-83.  54 I am grateful to Yasser Tabbaa for some penetrating remarks on this topic. He notes, as Grabar himself did, that few of his students came from Islamic studies, and goes on: ‘This is not surprising, for throughout his career Grabar favored secular or humanistic interpretations over ones emanating from Islamic beliefs and ritual practices. His early formulations of “court ceremonial,” “symbolic appropriation,” “ideological warfare” dealt far more effectively with questions of secular power and authority than with concepts of faith and theology. Not surprisingly, he had very few Muslim students early on’ (personal message). Cf. also Barbara Brend’s review of Mediation, Bulletin of the School of Oriental and African Studies, 58, 1995, 362: ‘Grabar is chary of allowing too great an influence to Revelation’. 55   Grabar, Formation, chapter 5; and The Great Mosque of Isfahan, ‘The Mosque in Islamic Society Today’ in Martin Frishman and Hasan-Uddin Khan, eds, The Mosque: History, Architectural Development and Regional Diversity, London: Thames & Hudson, 1994, 242~245. 56 ‘The Earliest Islamic Commemorative Structures, Notes and Documents’, Ars Orientalis, 6, 1966, 7-46.  51 52

443

OLEG GRABAR: THE SCHOLARLY LEGACY

have noted that he did possess a profound religious side, found solace there, and was particularly drawn to the rituals of the Orthodox Christian 58 faith. He did not fail to acknowledge the religious impulse behind so much Islamic art; and he was capable of sensitive analyses of the thinking behind the use of specific Qur’anic verses in specific locations within a 59 building, as is repeatedly evident in his book on the Isfahan jami‘. But such insights, revealing as they are, do not invalidate one’s basic impression that he did not make it a high priority to probe in depth the religious impulse in 60 Islamic art. It is striking that the list of his publications does not contain 61 a single article or book devoted exclusively to a Qur’anic manuscript, or to a general study of religious inscriptions, or to extended reflections on the nature of Shi‘ite art. And the only extensive treatment of calligraphy (a field which is dominated by Qur’ans) in his entire oeuvre is a general 62 chapter in The Mediation of Ornament. And there is still more to be said on this issue. Given that his father wrote a monograph definitive in its 63 own time on the theme of iconoclasm, it is not strange that the son 64 should briefly have taken up this very subject a few years later. But this did not lead him to produce a nuanced assessment of the vexed question 57

Of those in the public domain, the comments of his son – ‘I think my father came to care very much about belonging… late in life, to his church—a development that certainly took me by surprise’ – and of Renata Holod (see footnote 5[b] above) are important here; and I am grateful to Marianna Shreve Simpson for further insights on this matter.  58 Giles Constable has noted that he ‘was one of the few people I have known who appreciated the form more that the content of religious rites, though the content was also important to him and helped to shape the form’ (see footnote 5[b] above). 59  Grabar, The Great Mosque of Isfahan, 32-33, 39-40, 48 and 50. 60 Yasser Tabbaa argues that the difference between himself and Grabar ‘rests squarely on the impact of religion on architecture, not as an essentialist force but as a social construction that interacts with social and historical factors. Grabar would not—perhaps could not—see it that way, holding true to the very end to his secularist-humanist ideals as a corrective against the excesses of religious dogmatism. Even his later publications on Jerusalem, which replace “ideological warfare” with the more irenic and universal views of secularization and accommodation, still attempt to neutralize religion’ (personal message).  61 Nevertheless, his review of David Storm Rice’s short but meaty study of the epochmaking Ibn al-Bawwab Qur’an (David Storm Rice, The Unique Ibn al-Bawwab Manuscript in the Chester Beatty Library, Dublin: Emery Walker, 1955; reviewed in The Art Bulletin, 39, 1957, 240-242) is well worth reading. See also Grabar, ‘The Qur’an as a source of artistic inspiration’, in Fahmida Suleman, ed., Word of God, Art of Man: The Qur’an and its Creative Expressions, Oxford: Oxford University Press, 2007, 27-39.  62 Yet he taught at least one seminar in this subject.  63 André Grabar, L’iconoclasme byzantine, Paris: College de France, 1957.  64 Oleg Grabar, ‘Islamic Art and Byzantium’, 69-70; and ‘Islam and Iconoclasm’, in Anthony Bryer and Judith Herrin, eds, Iconoclasm, Birmingham: University of Birmingham, 1977, 45-52. See also Grabar, ‘From the Icon to Aniconism: Islam and the Image’, Museum 57

444

of the Islamic attitude to images, a theme for which his peerless range 65 of expertise would have fitted him particularly well. More generally, one senses that his particular configuration of interests led him to downplay the overwhelmingly religious motivation of so much Islamic art, especially Qur’anic manuscripts, sacred architecture and the impact of waqf. The advantage of this innate tendency was that he explored further than any scholar before him the secular drive of certain areas of Islamic art, and developed new frameworks for their study. Chief among these was the so-called ‘princely cycle’, essentially a Grabar construct although he never laid it out formally as such, or indeed attempted a history of this recurrent group of courtly images. But above all, he never grappled in full detail with the many-layered impact of the Qur’an on Islamic art and the people who produced it. Grabar’s travels In the course of his career, and especially after his involvement with the Aga Khan Award for Architecture developed from the late 1970s onwards, Grabar’s travels took him into some of the remotest parts of the Islamic world. He was the modern version of the medieval wandering scholar. With the partial exception of some African countries, he acquired first-hand experience of almost every Muslim country from Morocco to Indonesia. And he visited many of them repeatedly. These travels, many of them exciting adventures, created an ever-expanding reservoir of visual impressions and insights which fed his research in often unexpected ways. On a personal level they deepened his understanding of, and sympathy for, Muslim culture in the widest sense and of course for the people of those countries he visited. And they responded in kind. But even so, there can be little doubt that it was the Arab world that first captured his heart and for which he felt the deepest affection. And of all the cities in that world he had the closest affinity with Jerusalem, the still centre of his turning world, so much so that one of the four volumes of the articles he chose to republish bears the name of that city. The original spur for this preference is plain enough: in 1953-54, in the course of his graduate studies, he received a one-year fellowship from the American School of Oriental Research, and he chose Jerusalem as his base. He came back there as Director in 1960-1. Not surprisingly, the city cast its spell over him, International, 218, 2003, 46-55. A rounded if generalized treatment of the subject can also be found in Formation, chapter 4, 2nd ed., 72-98.  65 A major and comprehensive book by Finbarr (Barry) Flood on this topic is imminent. 

445

OLEG GRABAR: THE SCHOLARLY LEGACY

and he wrote movingly about these formative experiences. Moreover, the relatively relaxed political situation at the time allowed him to travel freely to neighbouring countries, so these Wanderjahre laid the foundations for his lifelong commitment to the art of the Umayyads. They also set the pattern for his later journeys. Thus his famous never-to-be-completed 67 project on Khurasan involved a long trip, part of which he made in the company of Janine Sourdel-Thomine and Oliver Watson (in 1971 or 1972). 68 As the latter tells the story, he sat in the car and listened to the excited conversation of Oleg and Mme Sourdel, likening the theory-building that took place between the two of them to building houses of cards. One would propose a theory, only to have the other object, flattening the house of cards, but, undaunted, either Oleg or Mme Sourdel would immediately begin to construct another over-arching theory, another house of cards. Then it was her husband ‘Dominique who, otherwise silent, would bring 69 them crashing down with a few quiet and pithy comments’. 66

Mentors and models Any assessment of Grabar’s written work requires some discussion of the scholars, including close colleagues, who exerted an influence of one kind or another upon him. The key person was of course his own father. André Grabar set the gold standard for his son from an early age: his field was cognate with that of his son, and he showed the way in rescuing a neglected field of art-historical scholarship and bringing it into fashion. 66   Grabar, Jerusalem, xix-xxxv. See too Renata Holod’s affecting reminiscences on this theme (footnote 5[b] above). 67 See Grabar, Early Islamic Art, 650-1100, xxvi-xxvii. Scott Redford recalls another such never-to-be realized project, ‘a book idea he had (one he never worked on, alas, as far as I know) in which he wanted to compare various border societies – Norman Sicily, medieval Anatolia and the Caucasus, Spain – in the medieval period’ (personal message). 68 Scott Redford comments that the story ‘wonderfully encapsulates the hunger he [Grabar] had for interesting ideas’. Oliver Watson fleshes out the story by recalling how the Iranian driver (distracted by the intellectual high jinks?) ‘narrowly avoided hitting a herd of camels, but drove the car into a ditch, thereby wrecking its suspension. He then departed to Mashhad in search of another car. Leaving us in the dark on our own. With great aplomb, Oleg strides back to the stricken car, opens the boot, and within that his suitcase, and produces a bottle of whiskey and some small glasses. He insists we all need a shot to calm our nerves. Oleg, Janine and I toast our luck…Oleg was always very generous with the young. He really wanted company on his visit to Iran then, so invited me - in my early postgrad BIPS fellowship - to accompany him through Khurasan, and later into Azerbaijan, and chatted and discussed things with me as though I had something useful to say.’ (personal message). 69  Oliver Watson (personal message).

446

I have written elsewhere of other aspects of this relationship, and this is in any event a delicate subject best left to those much better informed 71 72 about it. But his father’s prodigious productivity was surely a spur. And Jane Jakeman is right, when reviewing the book of Oleg’s 1989 Mellon lectures (which was dedicated to his father), to recall that his father gave the Mellon lectures in 1960 and to salute the pair of them: ‘The Grabars, father and son, belong to that rarest group of art historians, the bridge73 builders’. Next in line is Richard Ettinghausen, Grabar’s most important predecessor as an Islamic art historian in the United States and his principal mentor there. Ettinghausen previously held the Michigan job that Grabar made so much his own, and was also the editor of the only journal dedicated to Islamic art history, Ars Islamica. So the parallels were plain, although Ettinghausen, twenty-three years Grabar’s senior, safely belonged to an older generation. His skills and qualities were complementary to those of Grabar – for example, Ettinghausen almost never wrote about architecture, and his absorption in objects, and his capacity to analyze them minutely in search of their meanings, was different in kind from Grabar’s approach, though the latter certainly learned from Ettinghausen’s example here. But there can be little doubt that the older man presented – by force of personality, by his long-established reputation, and by the range, depth and number of his publications – a challenging role model for his younger colleague, and that there were complex aspects to their relationship. As things turned out, Ettinghausen’s later career lay firmly in the museum world, in Washington D.C. and New York, while Grabar’s was fixed on that of the university, at Ann Arbor and at Harvard. But they maintained 74 more than a merely collegial relationship. Grabar fondly recalled how Ettinghausen kept a watching brief on his career, which involved not only regular phone calls and letters but also periodical progress reports sent 70

See the obituaries listed in footnote 1 above.  For example, Grabar expressed to me his reservations about my assessment of the impact of exile on Ettinghausen (Robert Hillenbrand, ‘Richard Ettinghausen and the iconography of Islamic art’, in Stephen Vernoit, ed., Discovering Islamic Art: Scholars, Collectors and Collections, London: I.B. Tauris, 2000, 171-181).  72 Neil Levine notes ‘I will never forget his remark that one should publish a book every three years and three articles a year. But I also remember that he never applied that yardstick to anyone but himself. Thank God, and thank Oleg’ (footnote 5[b] above).  73   Grabar, Mediation of Ornament, reviewed by Jane Jakeman, Ars Orientalis, 24, 1994, 152. 74  Among the written comments he made on the draft of a paper I had prepared on Ettinghausen (see note 71 above) was the simple statement ‘I was fond of R.E.’ 70 71

OLEG GRABAR: THE SCHOLARLY LEGACY

447

to Grabar’s father. From 1958 to 1969 Grabar also held an honorary curatorship at the Freer Gallery of Art, where Ettinghausen was in charge of the Islamic collection; and they collaborated closely on the projected Pelican handbook on Islamic art. Grabar delivered a generous appreciation of the older master at Ettinghausen’s memorial service, with 76 a fuller obituary published in the same year. These two men, then, were probably the two living scholars who exerted the major formative influences on his scholarship, and they set 77 the bar high. But one should also not forget Philip Hitti and the cadre of Princeton Arabists like Bayly Winder who gave Grabar his grounding in Arabic. However, for all his facility with languages he was not an old78 style Orientalist philologist. He never engaged seriously with Persian or 79 Turkish, and despite occasional excursions into Arabic textual studies, he never again used Arabic as intensively as he had done in his doctoral 80 thesis, and more and more over the years he turned himself into a cultural historian rather than a text-based one. It was not these men who stimulated him intellectually at Princeton, but rather the art historians, for example Kurt Weitzmann, who started him on the road that led eventually to his Maqamat book, and Baldwin Smith, with his late antique and medieval 81 background and his fascination with architectural iconography. 75

The Michigan and Harvard years Grabar’s first academic post was at the University of Michigan in Ann Arbor, where he began as an instructor in 1954 and where he spent fifteen Grabar, Islamic Visual Culture, xxxi-xxxii.  Grabar, ‘Richard Ettinghausen’, Artibus Asiae, 41, 1979, 281-284. 77 Nor did Grabar: ‘Philip Khuri Hitti’, in Patricia Marks, ed., Luminaries: Princeton Faculty Remembered, Princeton: Princeton University Press, 1996, 119-124. 78 ‘His father André had fostered this gift of tongues — Oleg described once how the languages spoken at the dinner table at home switched from one day to the next’ (see my obituary in The Times, footnote 1 above). Reflecting in 2004 on what he had learned in his years of interaction with the varied cast of characters at the Aga Khan Award for Architecture, he reflected mischievously that he remained ‘regretfully convinced that a juicy footnote with quotes in six languages is a greater contribution to knowledge than a fancy meeting to discuss environmental development. Or is it?’ (Grabar, Islamic Art and Beyond, xxv). The humorous self-mockery and the gleeful deflating of pomposity in that quotation epitomizes his charm. 79 Notably ‘Upon Reading al-Azraqi’, Muqarnas, 3, 1985, 1-7.  80 Note, however, that among his many and varied seminar offerings at Harvard was one on inscriptions and epigraphy.  81 In the roll-call of eminent scholars whom Grabar cites as role models (Early Islamic Art, xxi) not one is a card-carrying Orientalist or philologist. 75 76

448

happy and fruitful years. At Michigan, Grabar benefitted in many ways, as he himself generously acknowledged, from the vision and imaginative support of the chairman of the Art History Department, George H. Forsyth Jr., who besides helping to fund Grabar’s expensive field trips to the Near East also contrived to give this junior scholar three free semesters 83 in his first four years of teaching. Grabar later said that this freedom to read widely at such an early stage was foundational for his entire career. He was free to teach Islamic art in whatever way he wanted and also to present it to large audiences of undergraduates in the context of a general survey of art history. He responded to that freedom with gusto, and was rewarded with a fine cadre of students from both America and abroad in those years, many of whom went on to become serious scholars in the field. But when he moved to Harvard in 1969, the situation changed. New responsibilities came his way, new resources – principally through the newly-founded Aga Khan Program for Islamic Architecture, whose first professor he became – and many new opportunities of all kinds. Naturally enough, the reach of his influence widened. And in teaching, too, things were not quite the same. Islamic book painting had long been taught there, admittedly on an ad hoc basis – for there was no established post in that field – by Eric Schroeder, a scholar of serious stature with important things to say about architecture as well as painting, but who unfortunately died two years after Grabar’s arrival, and, since 1956 (as honorary assistant 84 keeper of Islamic Art at the Fogg Museum of Art), by Stuart Cary Welch. 85 Welch was as much a collector and a connoisseur as a scholar and teacher, and eventually also took on further responsibilities as special consultant (1979-87) in charge of the Department of Islamic Art at the Metropolitan Museum of Art in New York. While Welch specialized in later Persian manuscript painting and in the arts of Islamic India, fields in which Grabar 82

Scott Redford suggests perceptively that Grabar viewed ‘the high point of his teaching as a series of seminars he held when he was still teaching at the University of Michigan’ (personal message).  83 For Grabar’s reflections on his Michigan years, see Islamic Visual Culture 1100-1800, xxxiii-xxxv.  84 See the anonymous obituary of 8 September 2008 on the College Art Association webpage [http://www.collegeart.org/obituaries/stuartcarywelchjr accessed 06.04.2012], and M. Fox, ‘Stuart Cary Welch, Scholar and Collector of Islamic and Indian Art, Dies at 80’, The New York Times, 10 September 2008. He retired from Harvard in 1995, five years later than Grabar.  85 The CAA obituary notes that in 1960 he taught the first course in Near Eastern art at Harvard. 82

OLEG GRABAR: THE SCHOLARLY LEGACY

449

never showed a significant research interest, there was no absolutely tidy division of teaching between them; indeed, they regularly co-supervised doctoral students. So it was not the case that Grabar left the teaching of Persian book painting to Welch; it was at Harvard in the mid-1970s that Grabar taught his legendary seminar on the Great Mongol Shahnama. 86 Grabar focussed on architecture, archaeology and almost the entire gamut of the so-called ‘minor’ arts. But Welch taught notable scholars in his special areas of expertise: Anthony Welch, Glenn Lowry, Thomas Lentz, Michael Brand, Sheila Canby and John Seyller among others. In their personalities, in the range of their interests and in their approach to 87 scholarship Grabar and Welch could scarcely have been more different. Not surprisingly, to a certain extent two camps developed, and students who elected to study with both of them could sometimes feel caught in the middle. How the field changed from 1953 to 2011 A good way to assess Grabar’s contribution to the scholarship on Islamic art is to consider what the field looked like when he entered it with his first publication in 1953, and what it looked like at his death almost sixty years 88 later. In those two generations momentous things happened, and Grabar 89 himself played nothing less than the key role in those changes. First of all, the centre of gravity in the field shifted from Europe to the United States. As the key European figures died – Herzfeld and Sauvaget slightly earlier, in 1948 and 1950 respectively, Diez in 1961, Kühnel and Erdmann in 1964, Marçais in 1966, Gabriel in 1972, and Creswell in 1974 – their places were not quickly filled. The German presence shrank, with no senior university post to act as a magnet for younger scholars. There As Nancy Steinhardt writes (footnote 5[e] above), ‘Monuments that he could intellectualize are, I think, what drove him’.  87 Walter Denny tells me ‘I’ve always maintained that if you put them both in a pot and boiled them down to make one art historian, you would have had the perfect historian of Islamic art’.  88 Nasser Rabbat succinctly summarizes the process of change: ‘a field that had not been totally weaned from its antiquarian, orientalist, and archaeological wet-nurses when Oleg came on the scene’ which Grabar ‘set… on solid foundations that he cemented by his teaching and supervising of several generations of Islamic art historians who today occupy important positions in universities and museums around the world’ (footnote 5[b] above). 89 For an assessment of Grabar’s role in a wide perspective, see Sheila Blair and Jonathan Bloom, ‘The Mirage of Islamic Art: Reflections on the Study of an Unwieldy Field’, The Art Bulletin 85(1), 2003, 156 and 172. This is perhaps the fullest historiographical survey of the field. 86

450

was no chair at an English university. With the loss of French political control in the Maghrib and the Levant, the opportunities for fieldwork and employment diminished. And the Islamic world itself was not yet producing a sufficient cohort of major scholars. Above all, no established framework was in place anywhere in Europe in the 1950s and 1960s for the next generation of scholars to receive a thorough grounding in the field, with government or university funding for fellowships, fieldwork and doctorates. Islamic art had to be studied in an ad hoc way. It was America, with its rigorous training programmes and substantial funding 90 and fellowship possibilities, that gradually showed the way forward. It was here that a cohort of doctoral students began to form from the early 1960s and where these students, once they had gained their doctorates, 91 began to find employment. Grabar was at the very centre of this process, 92 first in Michigan, then in Harvard, and by the end he had supervised a staggering number of doctoral students, about seventy by his own count, 93 a record that would be very hard to beat. That bald statistic highlights 94 a cardinal fact: that his students were central to his life. Many of them sensed that – hence the affectionate nicknames they bestowed upon him,

90 The context is sketched with sure strokes by Jacob Lassner in an unpublished lecture, in which he writes ‘young Oleg, without a degree in art history [italics mine], shouldered almost the entire burden of developing the field of Islamic art’ (footnote 5[f] above). I have some fellow-feeling here, for I too do not possess such a degree.  91 Lisa Golombek remembers (footnote 5[b] above) that in 1962, when she came to Michigan to study Islamic art with Grabar, ‘no one else in North America was teaching this subject full-time’. In 2012 it takes an effort of the imagination to picture that situation. The word that recurs in her reminiscences is ‘adventure’. It is the mot juste. He made his field sound like an adventure, and adventurous spirits leaped into it at his urging, or maybe just by the force of his example.  92 I quote Jacob Lassner again: ‘…in Ann Arbor Oleg was one of the princes of the university and as such had immediate access to the power brokers in the administration. He exercised that influence time and again on behalf of his students and colleagues. He was absolutely fearless when it came to knocking on doors’ (footnote 5[f] above).  93  I am grateful to Terry Grabar for this information. 94 Lisa Golombek put it memorably (footnote 5[b] above): ‘We, his students, colleagues and friends, are his enormous legacy that will carry on in his spirit’. A similar thought was expressed by Shaikha Hussah Sabah al-Salem al-Sabah: ‘he’s left a legacy that will grow far into the future: young scholars that share his vision and his passion’ [memoryog.tumblr. com]. Or as Esin Atıl put it to me: ‘we, the students – past and present – formed a bond of collegiality and shared research materials…Oleg’s greatest contribution was to individually nourish each and every one of his students…We still feel that we are a part of the same “medrese”. 

OLEG GRABAR: THE SCHOLARLY LEGACY

451

like ‘Poppa Bear’ – and eloquently gave thanks for it, not least in the prefaces to their dissertations and books. Next, the languages of continental Europe gradually but decisively lost their privileged status in the field of Islamic art history. English steadily supplanted them. To have undertaken the specialized study of Islamic art before the 1939-45 war without a good command of German would have been professional suicide. That was the language of most of the 96 handbooks, of key monographs on Islamic cities, Islamic building 97 98 99 types, schools of architecture, and the other major media. It was in that language, too, that the key theoretical debates about the origins of Islamic art were conducted before the 1914-18 war. French was not far behind. Most of the early scholarship on Islamic painting was in French; with the establishment of the French mandate in Syria and Lebanon, the scholarship on Islamic art and architecture in that area, as in the Maghrib, was largely francophone, as were the journals that served that scholarship, notably Syria and a clutch of outlets for the study of the Maghrib. And the first serious journal to cover the Islamic art of the Mashriq, Athar-e Iran, was in French. Van Berchem, who for some thirty years had the field of 95

95 Jack Renard kindly told me this. It brings to mind a reminiscence of Jaclynne Kerner from 10 December 2010: ‘After lunch, Oleg brought me to his office at the IAS. He had some offprints and career advice to give me. As we said our goodbyes, he told me the story of a mismatched trio of teddy bears sitting on his credenza. The largest bear symbolized Oleg, while the smaller pair represented the two students who had given him these gifts. Technically, Oleg explained, the students were not his. Both attended MIT, but as they presented the bears to Oleg, they told him that they considered themselves his students, too’ [memoryog.tumblr.com]. 96 Karl Wulzinger and Carl Watzinger, Damaskus. Die islamische Stadt, Berlin and Leipzig: Wissenschaftliche Veröffentlichungen des deutsch-türkischen DenkmalschutzKommandos 5, 1924; Max van Berchem and Josef Strzygowski (with a contribution by Gertrude L. Bell), Amida, Heidelberg and Paris: C. Winter, 1910.  97 Hermann Thiersch, Pharos, Antike, Islam und Occident. Ein Beitrag zur Architekturgeschichte, Leipzig and Berlin: B.G. Teubner, 1909; Friedrich Wetzel, Islamische Grabbauten in Indien aus der Zeit der Soldatenkaiser, 1320-1540, Leipzig: J.C. Hinrichs, 1918.  98 Friedrich Sarre and Ernst Herzfeld, Archäologische Reise im Euphrat- und Tigrisgebiet, 4 vols, Berlin: Reimer, 1911-20; Friedrich Sarre, Denkmäler Persischer Baukunst, Berlin: E. Wasmuth, 1910.  99 Ernst Kühnel, Miniaturmalerei im islamischen Orient, Berlin: B. Cassirer, 1922; Philipp Walter Schulz, Die persisch-islamische Miniaturmalerei, Leipzig: K.W. Hiersemann, 1914; Friedrich Sarre, Sammlung F. Sarre. Erzeugnisse islamischer Kunst. Bearbeitet von Friedrich Sarre mit epigraphischen Beiträgen von Eugen Mittwoch. Teil 1 – Metall, Berlin: 1906; Friedrich Sarre and Hermann Trenkwald, Old Oriental Carpets, tr. A.F. Kendrick, Leipzig: A. Schroll 1926-1929 (a translation of two fundamental surveys in German that had appeared in 1892 and 1908); and Otto von Falke, Kunstgeschichte der Seidenweberei, 2 vols, Berlin: Wasmuth, 1913.

452

Islamic epigraphy almost to himself, wrote largely in German and French. Most of the literature on the art of Central Asia was in Russian. Now, seventy years later, the landscape of scholarship looks very different. Conferences in Germany whose participants are largely German are conducted in English, and this is symptomatic of a trend that has swept Europe in the last couple of decades. Nowadays, to write in German or French (let alone Italian, Spanish, Russian or any of the languages of the Islamic world) is to risk losing an audience. And paradoxically enough, in the scholarship of Islamic art the time of the Islamic languages – Arabic, Persian, Turkish, Urdu – has, on the whole, still not come as yet except for native speakers of those languages. That said, this situation is changing fast, with the volume of scholarship on Turkish and Iranian art, in that order, growing at a dizzy rate and absolutely demanding international attention. Nevertheless, the major historians of Islamic art from those countries still turn more naturally to English for their key publications. For Grabar’s entire career, the principal audience for his work was the Western world, and English has been, and for the immediately foreseeable future will remain, the preferred medium of instruction. A further significant change was the shift from an amateur to a professional mindset. As Constable remarked, ‘a self-taught artist is one taught by a very ignorant man’. Much of the groundwork in Islamic art has been carried out, from the nineteenth century onwards, by dedicated amateurs. Many of them had quite catholic sympathies – Pascal Coste, for example, spent significant time recording the monuments not only of Iran but also of Cairo. Others, like Creswell, lacked the security of a permanent position and perforce had to train themselves. But with the establishment of graduate training programmes, first in the United States and then elsewhere, prospective specialists in Islamic art have had access to training in art history, in the relevant languages, and in Islamic studies as a prelude to undertaking a doctoral dissertation. As the number of such professionally trained specialists has grown, so more and more hitherto neglected areas of the field have been revealed. The cumulative growth of understanding and expertise has been remarkable. Finally, the field of Islamic art as a whole has developed a greater theoretical sophistication. This is directly related to the much greater number of scholars in that field, which makes for a corresponding increase in competition, since neither the number of posts nor the number of outlets for publication has kept pace with that exponential growth in the number of participants. Description and the accumulation of data are no

OLEG GRABAR: THE SCHOLARLY LEGACY

453

longer enough. Grabar himself pointed the way here to some extent, for 100 example, inter alia, in his attention to semiotics, structuralism, linguistic 101 theory and the importance of Bakhtin, and he played a large part in changing accordingly the direction of the Harvard department that he 102 chaired for some years. He also frequently spoke of the need to respond to technological advances in the processing of information. But at base his forte was speculation rather than theory, and he was consistently drawn to explore the meaning of works of art in their own time. In the field of architecture in particular, this led him far beyond much of earlier scholarship. Nevertheless, his major books and articles, while dense with information and analysis, and brimful of ideas, are cast in fairly traditional 103 mould. Grabar, ‘Are Pictures Signs Yet?’, Semiotica, 25, 1979, 185-188.  Yasser Tabbaa comments:  ‘Oleg had a nimble mind, and he knew before others when a methodology had run aground and could no longer support or advance his expanding horizons in Islamic art. The switch to aesthetics and semiotics—already at hand in The Alhambra (1978) and fully central in The Mediation of Ornament (1992)—was not simply “fashionable,” as some had proposed then, but was motivated, I believe, by the increasing aridity of studies of patronage and the social history of art and by Grabar’s increased involvement with architectural practice. This methodological transformation attracted many students to Grabar, more theoretical and less archeological than his earlier students, and included a higher proportion of students of Arab or Middle Eastern descent.’(Personal message.) 102 And these interests had a much wider field of operation, for they impacted directly on how art history was taught at Harvard, as Neil Levine notes: ‘a strengthening of the department’s own coherence as a faculty devoted to a wide range of historical areas including African, Islamic, ancient Near Eastern, and Southeast Asian as well as a focus on critical theory and new and diverse methodological approaches, which Oleg’s broad interests and wide-ranging enthusiasms had inspired in us’ (footnote 5[b] above).  103 Barry Flood kindly responded as follows to my invitation to define what the distinctive contributions of Grabar’s scholarship were. His response was so powerful that I make no apology for citing it in extenso: ‘Oleg’s 1959 Ars Orientalis article on the Dome of the Rock… illustrated the way in which a single powerful intervention can shape a discourse, determining its parameters for decades to come, and establishing scholarly orthodoxies…[it] reminded me of how important Oleg’s lifelong interest in the Umayyads was to many of our intellectual formations…[he was able] to relocate the Umayyads within an extended late antiquity rather than outside its upper limits, calling into question the very idea of the Umayyad period as a watershed… A second thread running through his scholarship… is his insistence on looking beyond the obvious for comparative material, especially during periods of social or political change that foster enhanced levels of transregional mobility…[this] alerted me to the need to think “outside the box”… The third aspect of Oleg’s scholarship that has left an enduring mark is his engagement with experimental methodologies and openness to theoretical frames drawn from other fields within the discipline of art history, or even outside… his ability to move from a coin, fragment or shard to an entire universe of dynamic ideas was dizzyingly exciting…[he had] an excellent “nose” for topics with the ability to engage 100 101

454

But for all these manifold changes, and despite Grabar’s frequent reiteration of his belief that new technologies and new ways of seeing would transform the field of Islamic art from within, it remains the case that the views of individual scholars, well thought out and well expressed, lie – as they have done for many decades past – at the centre of Islamic art history today, as is the case with the humanities in general. The computer and the internet are still ancillary to the action of the mind. And Grabar himself is the best example of that principle. Early work and The Formation of Islamic Art It is time to turn to a closer scrutiny of his books. As it happens, it is useful that among the very few somewhat challenging reviews that his books received, two of the longest and most serious assessed what to 104 many people is still his major work: The Formation of Islamic Art (1973). This is the book that made his reputation in a wider circle, beyond the and excite the discipline of art history as a whole… In some ways his ability to anticipate trends within the broader academic world was uncanny - especially so, when one recalls that he published two key articles on images of the Prophet only a year or two before the subject erupted onto the global scene in such an unfortunate way. The final, and certainly the most important, lesson learned from Oleg’s scholarship was the absolute need to forge dialogues across the discipline of art history, not only in a spirit of collegiality or intellectual exploration, but to ensure the very survival of the field itself. In short, he realized the need to resist attempts to allot the study of Islamic art its corner as an exotic sub-field of the discipline, and the need for historians of Islamic art to be actively involved in the ongoing process of shaping the discipline…he did this in three ways: by speaking and writing on topics well-chosen for their intrinsic interest or their ability to raise major questions of an interpretive or theoretical nature that resonated beyond the study of Islamic art; by his adoption of and experimentation with theoretical frames that enabled the “translation” of specific problems across the discipline, engaging colleagues working on the art of wildly divergent times and places; by his ability to frame even rather abstract questions in ways that made their import readily comprehensible to an audience well beyond a small clique of Islamic art historians.’ (Personal message.) 104 These are by Erica Dodd and Michael Rogers. The first review (in The Art Bulletin 57[2], 1975, 267-70) contains mildly couched criticisms that focus on her perception that Grabar downplays the impact of the built environment of pre-Islamic Syria and of the Oriental Hellenistic heritage in particular, and that he has a less than sure touch on matters to do with religion, notably the mosque and the mihrab. The second is essentially a fourteen-page review article (in Kunst des Orients 9(1-2), 1973-74, 153-66). Recognizing that the book is based on a series of public lectures, Rogers follows that schema in his chapter-by-chapter response to it. He acknowledges many of the strengths of the book whilst pointing out areas where Grabar could have gone further or is, in his view, mistaken. The tone is vigorous and the range of comment is wide-ranging. But the review is not really a sustained attempt to engage with the broader horizons of the material. Instead it can be seen as a series of penetrating insights or counter-arguments (often interesting and suggestive in themselves) that are marshalled against this or that passage in the book. 

OLEG GRABAR: THE SCHOLARLY LEGACY

455

narrow constituency of Islamic art historians. It is by now well established as the book of choice to be cited by scholars who, from the specific viewpoints of their various disciplines, engage with a notoriously complex 105 phenomenon – the transition from late antiquity to early medieval times – and who wish to draw attention to the Islamic dimension of that 106 process. This book is a testament to his early maturity. For many of his own students, it had the force of revelation, a breath of fresh air after the 107 stodgy handbooks to which students had previously been consigned. And it demanded a re-appraisal of what was meant by the term ‘Islamic 108 109 art’. In all sorts of ways, it set new benchmarks in the field.

Gülru Necipoğlu pinpoints how this book ‘made Islamic art appear wide open to hugely exciting questions of cultural history, captivated my imagination and was the single most important factor that triggered my conversion to this newly budding field. Indeed, Oleg had a very special talent for making Islamic Art seductive and appealing to nonspecialists; thereby vastly broadening its recognition within the two disciplines of Art History and Islamic Studies’ (see footnote 5[b] above). 106 As Sheila Canby says, this text ‘not only poses questions about the essential nature of Islamic art but also explores the context in which it was produced. In fact, the societies of the central Islamic lands are the starting point for Grabar’s discussion, rather than the monuments or works of art.’ (Personal message).  107 For example, Thomas Lentz: ‘It was undoubtedly the analytical and contextual framework he managed to place over an array of fundamental issues and questions that was his lasting legacy. No work better epitomizes that approach for me than The Formation of Islamic Art – and I say that as someone whose primary interests stood far away and much later than his focus in that book…it is in many ways more important for its questions than answers, but it instilled in the reader a deconstructive mindset that was then carried to other subjects and issues, and at the time struck me as quite different than other approaches to Islamic art (at least to my young and untutored mind). For me, it was both revelatory and liberating to have a complement to the descriptive and taxonomical approaches that made up so much of the field at that time.’ (Personal message.)  108 Glenn Lowry concludes: ‘For me The Formation of Islamic Art remains the quintessential book because Oleg was able to ask a series of questions that led to a theory of Islamic art that was exhilarating even if it was not necessarily always right’ (personal message). Sheila Canby adds: ‘Grabar broached the conundrum of how art historians could call the art of places separated by thousands of miles and several centuries “Islamic” without defining the qualities that unify such works and also differentiate them. Questions such as this continue to dog the field today, but the discourse on this subject would be far more incoherent had Grabar not confronted it in 1973’ (personal message).  109 As Yasser Tabbaa notes: ‘Reading Formation was especially significant for me, for it led me to think of questions of social history and patronage for my dissertation on Nur alDin…Formation remains his most important book, marking as it does a decisive shift in the field from product to process and from focusing on the object to examining its contextual relations’ (personal message). 105

456

It is in fact his first real book, for the slim volume on Tulunid coinage published in 1957, the fruit of an intensive but brief course of study at the American Numismatic Society in New York, is an apprentice work of carefully limited scope cast in the familiar numismatic mould of the ANS mini-monograph. It ticks all the correct technical boxes, it does what it says on the tin, it is an early signal of his virtuosity; and one would think that its subject matter forbids just those flights of creative, wellgrounded speculation that were to become Grabar’s trademark. Wrong. No less an authority than Sir Hamilton Gibb, doyen of British Arabists in his generation, praised the way that this book supplemented the existing literary evidence with that drawn from archaeology (or what would nowadays be called ‘material culture’). That latter evidence, he notes, was largely neglected by historians; but ‘more than once in this study, it raises problems to which neither it nor the literary evidence supplies a clear 111 answer’. The book on Tulunid coinage is significant in a wider sense in that it shows how early Grabar formed the habit of publishing his results as soon as possible after he had garnered the necessary data. Not for him the endless neurotic polishing of a text in a doomed bid for perfection, where the law of diminishing returns dictates a steadily decreasing yield for each hour of effort. Nor did he spend valuable time in trying to accumulate all the available data before committing himself. It was as if he had instinctively realised, when he was still only in his twenties, that the very best is the enemy of the very good. Of course, all this is closely linked to personality; what perfectionist could ever consider the opportunity cost of hunting down that very last footnote, or question that only perfection will do? Grabar’s spacious attitude to footnotes was proverbial; he was quite frank about it, and unrepentant when the need for foonotes was pointed out to him, even going so far as to suggest that they were a waste of time. Rank heresy or common sense? And in this respect he did on occasion practise what he preached, in that sometimes he made no serious effort in his own work to ensure that they were consistently full or even accurate. 110

110   Neither the Michigan exhibition catalogue of 1959 (Oleg Grabar, Persian Art Before and After the Mongol Conquest, Ann Arbor: University of Michigan Museum of Art, 1959) nor Sasanian Silver (1967) can be regarded as single-authored monographs, the first because it was a catalogue and the second because it was co-authored (with Martha Carter). 111 Oleg Grabar, The Coinage of the Tulunids, New York: American Numismatic Society, 1957; review by Hamilton Gibbs in Bulletin of the School of Oriental and African Studies, 21, 1958, 631.

OLEG GRABAR: THE SCHOLARLY LEGACY

457

Let us return to The Formation of Islamic Art. This is emphatically not the book-of-the-thesis; indeed, Grabar never published his 330-page thesis in book form, nor even any of its chapters as discrete articles. But it was not put aside or forgotten; indeed, it worked like yeast on his imagination. Its looming presence can be sensed in many places in The Formation of Islamic Art, especially chapter six, and the hard-won evidence of the thesis of 1954 lends bulk, authority and conviction to the book of 1973. The best part of two decades lay between those two achievements, and in that time he had grown exponentially in stature as a scholar. He had also left the alluring problems of Umayyad art, not just for Tulunid coinage but 112 also for Sasanian metalwork and the Seljuq art of Iran, and so he came back to those problems thoroughly refreshed by his excursions into other fields. This trajectory enabled him to sidestep the principal danger which the modern academic climate poses for the young scholar: to publish the thesis as soon as possible, without the luxury of taking a long break from it so as to come back to it with a changed and much less pressured mindset and examine it from a different perspective. And nowadays if such young scholars do indeed take that break, it puts their tenure applications into jeopardy, while if they continue to work on the thesis without that interruption so as to make a book out of it, they run the risk of becoming stale and narrow-minded just at the age when they should spread their wings. Grabar came to maturity much more naturally, in a more spacious and less policed age. So The Formation of Islamic Art provides solid intellectual fare; it is durchkomponiert, with an enviable suppleness and a width of reference which bespeaks Grabar’s solid grounding in Islamic studies and which, to speak frankly, is simply beyond the grasp of any of today’s historians of Islamic 113 art. In that sense, he was the last of his kind. This work is anything but a picture book, and the illustrations, all in black and white, have low production values with no pretensions to glamour. It is as if the subliminal message were that this book is not about art, but more widely about culture as seen through the prism of art. And this was a new approach, a 112  Prudence Harper writes: ‘His real contribution to Sasanian studies was his ability to articulate significant questions and propose hypothetical answers some of which he stuck with others of which he discarded…but the gift was the incredible skill in sifting through the data and coming up with thoughtful points that needed investigation… The result was that he inspired everyone to stretch their minds a bit and reach for the broader understanding’ (personal message). 113 His facility with European languages had much to do with this. 

458

much more ambitious one than that followed in earlier handbooks. With its publication Grabar consolidated his emergence from the ranks of the jeunesse dorée and became definitively a force to be reckoned with, a scholar who had momentous things to say about momentous issues. Nor is this the whole story. Since Islamic art history had first announced itself as a discipline at the end of the nineteenth century, the key question that had fascinated scholars and was the main bone of contention concerned the origins of Islamic art. Orient oder Rom? thundered Strzygowski, a born polemicist. That question was typically posed in the general context of late antique art rather than the particular one of Islamic art, although 115 Islamic art was quickly commandeered into the discussion, and those who debated that question before the 1914-18 war shared a background in which Greece and Rome took centre stage as if by right. Trained from early boyhood (they were all men) in the classical languages, the scholars of the time naturally saw things from that perspective. Yet it blinkered them. In assessing Byzantine art, that was no great disadvantage. But the tools furnished by an intensive classical education were inadequate to explain the earliest Islamic art. Indeed, they were liable to misrepresent it. The detailed knowledge which such scholars possessed of the Semitic and Iranian components of Western Asian culture dating from the first millennium CE lagged almost absurdly far behind their knowledge of the 116 Graeco-Roman world. The Arabian background, especially the impact of the Qur’an, scarcely entered the picture. Grabar’s great book is the long-delayed counterblast to the feverish pre-1914 theorizing about the emergence of a recognizably Islamic art. And its title is carefully chosen, echoing as it does the single epoch-making contribution of that pre-1914 generation, Ernst Herzfeld’s precocious masterpiece on Mshatta and the 117 origins of Islamic art. How is Grabar’s riposte configured? Paradoxically (given the title of the book) the argument is not framed in principally art-historical terms. Instead, the reader is treated to a panorama of early Islamic civilisation. It is these beguilingly wide perspectives that give the book its muscle: 114

 Such as Dimand, Talbot Rice, Kühnel, Diez, Marçais – to name only those handbooks published around the 1960s and which were, therefore, the ones to beat. 115  Especially in the context of Qusayr ‘Amra; see Robert Hillenbrand, ‘Creswell and Central European Scholarship’, Muqarnas, 8, 1991, 25. 116 An important exception must be noted here: those scholars with a Jewish background who had an early training in Hebrew and Judaica.  117 Ernst Herzfeld, ‘Die Genesis der islamischen Kunst und das Mshatta-Problem’, Der Islam, 1, 1910, 27-63 and 105-44. 114

OLEG GRABAR: THE SCHOLARLY LEGACY

459

the visual material is firmly integrated into a fully realised society. In this respect the book is a world away from the perfunctory summaries of political history which preface the chapters of the handbooks available at the time. In such books one can sense the author’s impatience to get to what he regards as the real meat of the matter – the art itself – not the history, the society, the faith, the entire thought-world, that encompasses the art and out of which the art grows. Grabar did not see art as something self-contained in that way. For him it was just one of the many ways that a given society expressed itself, its concerns, its beliefs, its view of the world. It is worth remembering that his first degrees, at the universities of 118 119 Paris and Harvard respectively, were in ancient, medieval and modern history. And a historian he remained to the end. Art historians often find their métier by a knight’s move from disciplines as varied as philosophy, philology, classical studies, literature, architecture or the social sciences, among others; Grabar’s route, first and last, was history. From Shahnama to Maqamat History is at the very heart of a short but major book that came out of the Harvard seminar on the Great Mongol (or “Demotte”) Shahnama 120 121 held in Spring 1975. Grabar co-authored this with Sheila Blair, one of 122 the participants in the seminar. Its title proclaims its audacious scope: Epic Images and Contemporary History. The Illustrations of the Great Mongol Shahnama. It is a perfect example of how a single key book can trigger 118   Certificat de licence, Ancient History, University of Paris (1948); Certificats de licence, Medieval History and Modern History, University of Paris (1950). 119  B.A. (magna cum laude), Harvard University, Medieval History (1950); he gained his M.A. and Ph.D. at Princeton University in 1953 and 1955 respectively, in Oriental Languages and Literatures and History of Art. 120 He had held two earlier seminars on the manuscript at the University of Michigan in 1958 and 1959 (Oleg Grabar and Sheila Blair, Epic Images and Contemporary History. The Illustrations of the Great Mongol Shahnama, Chicago and London: Chicago University Press, 1980, ix).  121 Apart from books of collaborative authorship or editorship, Grabar published articles with other former students of his (Renata Holod, on Iranian architecture) or younger colleagues (Mika Natif, on images of Muhammad; she also did a great deal of work on preparing the four volumes of his collected articles). And he collaborated less formally with many more (see the comments of Muhammad al-Asad [memoryog.tumblr. com]). 122 The first chapter – a tour de force of patient but imaginative detective work in reconstructing the many vicissitudes that the manuscript had undergone – is evidently hers, the second his; the exact degree of collaboration between them in the later part of the book remains unclear. 

460

an avalanche of publications, as if indeed there had been a pent-up body of information and speculation that had found no previous outlet. The book is packed with protein. In retrospect, it is clear that almost all earlier publications on this acknowledged masterpiece of early Persian painting 123 had signally failed to take its measure. So this is a quantum leap. Grabar himself had published a dry run for part of this book eleven years 124 previously in article form. That outline, somewhat further developed, remains the core of the book’s second chapter. But that is merely an hors d’oeuvre for the third and fourth chapters, which take the discussion onto a new plane of sophistication and subtlety for which no precedent in the scholarship on Islamic book painting springs to mind. The sense of intellectual excitement here is palpable; you feel you are in the engine room itself. At long last the historical context is not perfunctory, not mere window-dressing, but rather integral to the argument. The dovetailing of separate nuggets of historical information with specific images justifies a truly radical proposal: that these paintings gave the ancient Shahnama stories a topical Mongol slant. Sixteen years later, Abolala Soudavar gave 125 chapter and verse in support of this theory. This book, then, opened a new window in the scholarship of Persian manuscript-painting. And it was 126 a team effort, with Grabar as the conductor of the orchestra. As one of the seminar members describes it, Grabar’s intellectual energy was the glue that held this communal enquiry together; he knew the right place to 127 put each individual contribution so as to make best sense of it. Grabar’s interest in the Maqamat of al-Hariri began in his salad days at Princeton, when he took classes on medieval manuscripts with Kurt Weitzmann. He returned to the theme at intervals over the next thirty years, producing a series of teasing, thought-provoking investigations of this exceptional corpus of illustrated Arabic manuscripts, almost all of 123   With the exception of a magisterial article by Eric Schroeder, ‘Ahmed Musa and Shams al-Din: A Review of Fourteenth-Century Painting’, Ars Islamica, 6, 1939, 113-142. 124 Grabar, ‘Notes on the Iconography of the “Demotte” Shahnama’, in Ralph PinderWilson, ed., Paintings from Islamic Lands, Oriental Studies 4, Oxford: Bruno Cassirer, 1969, 32-47.  125 Abolala Soudavar, ‘The Saga of Abu Sa’id Bahador Khan. The Abu Sa’idnamé’, in Julian Raby and Teresa Fitzherbert, eds, The Court of the Il-khans 1290-1340, Oxford Studies in Islamic Art, 12, Oxford: Oxford University Press, 1996, 95-211.  126 The names of the participants are duly listed and handsomely acknowledged in Epic Images, xiv. Marianna Shreve Simpson recalls: ‘Years later Oleg described this group as “an extraordinary team working together”. All of us still remember this as one of the great moments of our lives’ (footnote 5[b] above).  127 My thanks to Jack Renard for this.

OLEG GRABAR: THE SCHOLARLY LEGACY

461

thirteenth-century date and of Iraqi or Syrian provenance. He responded with gusto to the salty, raucous atmosphere which one gifted artist in particular (Yahya al-Wasiti) conjured up from these picaresque narratives, with their cast of streetwise, disputatious characters often on the wrong side of the law. Indeed, Grabar had a particular fondness for the Paris manuscript of 1237 AD, with its unsparing exposure of the seamy side of Islamic urban society, its vignettes of rural life and its unashamed flights of fancy. Eventually he wrote a long introduction to a facsimile edition of 128 this masterpiece. In 1984 he published what was for him the definitive study of the entire genre, which tackled head-on some of the basic problems that it posed: the reasons for the brief but intense fashion for illustrated versions of the Maqamat, and the interrelationships between the manuscripts and the sources of the illustrations. In this text he surveys the manuscripts in detail one by one, while the longest chapter deals with the illustrations themselves, and the messages that they carry. The framework for this discussion is the close study of the illustrations for each of the forty-eight maqamat in turn. But once again, as with Epic Images, he was let down by the poor production values of Chicago University Press. The decision to publish all 789 images on microfiche – which seems to have entailed a corresponding absence of any printed illustrations in the book – condemned interested readers to hunting out laboriously, in front of humming projectors, the images that especially interested them. So the crucial visual dimension gets lost. At all events, Grabar was disappointed by what he perceived as a lack of response to a book he had pondered and 129 worked towards for decades. The Mediation of Ornament 130 Of all his books, it is The Mediation of Ornament – an enigmatic title, this – that has attracted the widest response, much of it from outside the charmed circle of Islamic art historians, as shown by the many reviews it

Grabar, ‘Introduction’, Maqamat Al-Hariri: Illustrated by Y. Al-Wasiti (facsimile edition of Bibliothèque National de France, ms. ar. 5847), London: Touchart, 2003, 1-62.  129 Note, for example, the lukewarm review by Nabil Safwat in Bulletin of the School of Oriental and African Studies, 50(1), 1988, 351-352. 130 See the review by Brend in Bulletin of the School of Oriental and African Studies, 58(2), 1995, 361. The longer, teasing title of the Mellon lectures on which the book is based (Intermediary Demons, Towards a Theory of Ornament) is briefly discussed by Sylvia J. Auld in a review in Art History, 17(1), 1994, 138.  128

462

received. The reason is not far to seek. Ornament is part of the universal language of art. Most art historians have to come to grips with it to some extent. So it marks an area of overlap where it is indeed possible to learn something from scholars far outside one’s own field. It is possible that Grabar was led to think about ornament across the whole spectrum of Islamic art by Gombrich’s famous book on the subject, with its emphasis on naturalism and its adoption of a perspective which, though notionally global, is really confined to Europe, leaving non-Western traditions, 132 including the Islamic one, with no more than a bit part to play. So Mediation, whose inherently cross-cultural scope won many plaudits, may 133 have been Grabar’s response to a perceived challenge. At all events, this is a deeply considered examination of the very heart of Islamic art, and it is clearly the work of a scholar at the very height of his powers, capable of approaching this complex subject from any number of angles. That said, it has an impish quality well illustrated by its tongue-in-cheek (and tonguetwisting) neologisms (calliphoric, terpnopoietic, chronotopic, monoptic and optisemic), perhaps created as a way of signalling that the author is entering uncharted territory. The theme is a grand and difficult one, but Grabar rises to it and goes well beyond earlier treatments of geometry, writing and architecture; even the much-studied arabesque takes on new 134 dimensions and reveals new subtleties under his scrutiny. His profound 131

n addition to the reviews cited in the previous footnote, see Margaret Olin in The Art Bulletin 75(4), 1995, 728-731; Rudolf Arnheim in The Journal of Aesthetics and Art Criticism, 53(2), 1995, 218-219; Catherine Kapikian in Interpretation: A Journal of Bible and Theology, April 1994, 224; John Rhodes in Speculum, 69(4), 1994, 178-79; Carol Bier, ‘Ornament and Islamic Art’, Middle East Studies Association Bulletin, 28(1), 1994, 28-30 – the excellent closing paragraph of this review cites Grabar’s own assessment of the book: ‘a conscious and deliberate departure from traditional scholarship in the study of Islamic art’ (Mediation, xxii); Jane Jakeman, Ars Orientalis, 24, 1994, 151-152 (which says a great deal in a few words); Rose, Ornament or Art?, 30-43; and Marianna Shreve Simpson, ‘Oleg Grabar, Intermediary Demons: Toward a Theory of Ornament’, in The A. W. Mellon Lectures in the Fine Arts, Fifty Years, Washington DC: 2002.  132 Ernst H. Gombrich, The Sense of Order. A Study in the Psychology of Decorative Art, London and Ithaca, New York: Phaidon Press, 1979.  133 Sussan Babaie hails it as ‘a key intellectual resource for…exploring transcultural notions in art’ (personal message).  134 See Alois Riegl, Problems of Style: Foundations for a History of Ornament, tr. Evelyn Kain, with annotations and introduction by David Castriota, Princeton: Princeton University Press, 1992; Ernst Herzfeld, ‘Arabesque’, Encyclopaedia of Islam (1st ed.), s.v., 363a-367b; Ernst Kühnel, Die Arabeske, Wiesbaden: Dieterich’sche Verlagsbuchhandlung, 1949, tr. Richard Ettinghausen, The Arabesque: Meaning and Transformation of an Ornament, Graz: Verlag für Sammler, 1977; and James Trilling (another beneficiary of Grabar’s support), ‘“Meaning” and Meanings in Ornament: In Search of Universals’, Raritan, 12, 1993, 52–69.  131

OLEG GRABAR: THE SCHOLARLY LEGACY

463

and nuanced treatment shows up the stale and threadbare clichés that characterise the perfunctory discussions of these topics in much earlier scholarship. Above all, he takes on the challenge of explaining how it is that these abstract categories become vehicles for expressing not only abstract concepts but also emotion and, most contentiously of all, an Islamic view of the world. But it is still, as many reviewers noted, a difficult book with 135 some inherent contradictions: to map out so much new territory took 136 its toll. The book’s forty pages of notes bear testimony to the range and 137 depth of his enquiries. The Shape of the Holy The Shape of the Holy was, of all his books, the one that Grabar was born to write. The idea of it had probably been at the back of his mind for well over forty years. Again and again in that time Jerusalem had called him, and the preface to the book explains why. It is a walk down memory lane. A palpable nostalgia, a sense of ubi sunt?, runs like a powerful current through the whole book, creating a Jerusalem of the mind that takes shape alongside the physical Jerusalem of Umayyad and Fatimid times that this study seeks to recreate. This is the long-delayed but natural sequel to all his early work on the Dome of the Rock in that it attempts to provide a detailed physical as well as historical context for that remarkable building. That involved scrutinising the Aqsa Mosque, the Dome of the Chain, the Haram and its various gates and subterranean structures – in short, the whole project of the Umayyad refashioning of the heart of the city. The explorations and adventures of his youth, when he had clambered over and under and through these buildings, sometimes at risk of life and limb, had given him not only hands-on experience but, more intangibly, a feel for them. And in true Grabar fashion, he seized the opportunity of exploiting up-to-date technology, in this case computer-aided design programmes overseen by Mohammad al-Asad and Abeer Audeh, to create a simulacrum of how the city might have looked at various times in the period of Muslim rule. These diagrams, supplemented by superlative colour photographs taken by Said Nuseibeh, made it possible to present the pre-Ayyubid city (and in particular its major monuments) in excitingly Particularly Arnheim, Jakeman, Olin and Brend.  The lectures and book, moreover, took shape at a time of devastating personal bereavements: his daughter, Anne-Louise, and his father.  137 Though this wealth of annotation, which was not typical of his later work, may have something to do with the research assistants mentioned in Mediation, xx-xxi.  135 136

464

new ways. But the core of the book remains its text, in which Grabar nimbly deploys decades of research and thinking to reveal, bit by bit, the grandiose Umayyad vision of a Jerusalem in which (as in the Qur’an itself) a final revelation – this time taking physical, architectural form – set the seal on the earlier incomplete ones of Judaism and Christianity. Above all, this book has authority. He had an intuitive grasp of the Umayyad period which he had built up the hard, old-fashioned way. That, incidentally, explains why the best parts of Formation are the Umayyad, not the ‘Abbasid 138 ones. Grabar’s articles Grabar produced many other books, but this is not the place to assess them one by one. Rather is it time to consider, however briefly, his articles. A few preliminary remarks are perhaps in order. It is of course far too early to assess the full range and impact of Grabar’s legacy. Only time will tell which of his works will remain most fresh for future generations. But it would not be appropriate to end this survey of a wonderfully productive academic career without giving some thought to the example which that career sets, and here the articles are every bit as crucial as the books. For Grabar’s legacy is a good deal more than the sum total of his publications, or indeed of the vibrant personal memories that he has bequeathed to hundreds of students and colleagues. Of their very nature, such memories are bound to fade. But his example is something else again. It is built into the fabric of what he wrote. And the message is clear: do not imprison 139 yourself in a narrow field of specialization. If you do, each trip to the well will mean a little less water in your bucket. The danger is that we know more and more about less and less, and it stunts the mind. The endless subsets of Islamic art – and, for that matter, of art history as a whole – should not be kept in hermetically sealed compartments. They impact on each other and are thereby enriched. And what is true of the art itself is equally true of those who study it. This helps to explain an otherwise puzzling fact – that so few of his students went on to specialize in the Umayyad period. Maybe they hesitated to trespass on his turf, even though a less turf-conscious scholar is hard to imagine. But turf works both ways. And Grabar’s command of this period was truly formidable. 139 Cf. the remarks of Thomas Lentz: ‘I saw him then [the late 1970s] and still do as a kind of exotic, straddling older European scholarship while embracing new methodologies, one of the last generalists in a domain soon to be dominated by specialists’ (footnote 5[b] above).  138

OLEG GRABAR: THE SCHOLARLY LEGACY

465

It may seem perverse to insist on the value of the articles when books such as The Formation of Islamic Art and The Mediation of Ornament clamour for attention. Grabar’s finest books show his capacity to tackle a big subject in the round. They also showcase his versatility. But the articles do so to an even greater degree. They often show him flying kites, true; but oftener still they show his capacity to absorb detail and then triumphantly invigorate that detail by giving it an unexpected context or application, thus enabling it to play its part in a wider narrative. Taken as a whole, it is they, rather than the books, that give the full measure of the man. The choice of articles in the four volumes published in 2005 and 2006 was 140 carefully, even painfully, made and thus repays close scrutiny. Eightythree articles, many of them of significant length, is an impressive total in itself, independent of the astonishing range of topics covered in them. They take their place as of right beside the similar monuments to the life’s 141 work of his two principal mentors: his father André Grabar, and Richard 142 Ettinghausen. It would take far too long to cherrypick the key articles in this selection, but it is worth noting that while ‘The Umayyad Dome of the Rock’, whose title conjoins two of the fixed points of his scholarly life, is cited by many as the epitome of his scholarship, others have highlighted a much shorter 143 piece: ‘An Art of the Object’, which was triggered by the revamping of 144 the Islamic galleries at the Metropolitan Museum of Art. The impact 140 See footnote 4 above. For the sake of completeness, it is worth mentioning that a preliminary collection of fourteen of his articles appeared much earlier under the title Studies in Medieval Islamic Art, London: Variorum, 1976. 141 André Grabar, L’Art de la Fin de l’Antiquité et du Moyen Age, 3 vols, Paris: Collège de France, 1968 (ninety-four articles).  142   Myriam Rosen-Ayalon, ed., Richard Ettinghausen: Islamic Art and Archaeology. Collected Papers, Berlin: Gebr. Mann Verlag, 1984 (fifty-eight articles). 143 Such as Thomas Lentz and Marianna Shreve Simpson. The former has commented: ‘He was not, for example, a “museum person,” yet his 1976 Artforum article (‘The Art of the Object’) was a profoundly insightful meditation on the conundrums presented by the display of Islamic art in museum spaces. It remains a touchstone for the field’ (footnote 5[b] above). The latter describes how she developed the theme for a round-table discussion at HIAA’s Second Biennial Symposium in Washington in 2010: ‘Oleg sat in the back of the Freer auditorium and, in typical fashion, whispered comments to his neighbors all through the session’. One might add that his whispers tended to be echoing stage whispers and to create an alternative focus of interest in the room.  144 Oleg Grabar, ‘An Art of the Object’, Art Forum, 14, 1976, 36-43. Gülru Necipoğlu recounts how, when she was undecided about where to undertake graduate work, Grabar ‘helped me make up my mind with just a few words encapsulating the differences of methodological approach between himself and Professor Ettinghausen, whom he always admired…: If you wish to start with ideas and then choose relevant objects, come here; but

466

of this article gives the lie to those who maintained that Grabar was not 145 primarily interested in objects qua objects. The truth was more complex. How his students and colleagues view his legacy Most of this paper has tried to answer an apparently simple question: what is likely to matter most among Grabar’s writings now that the man himself is gone? The comments made above on that issue are of course merely one person’s view – and not just that, but the view of someone outside that North American scene which Grabar bestrode 146 like a colossus. So it seemed sensible to canvass opinions among those 147 who knew him at first hand as a teacher and then as a colleague. But the result of a fairly extensive straw poll among his former students and some of his colleagues, designed to elicit answers to that same apparently simple question, is intriguing. What emerges is virtually a hagiography. The majority of those who replied continue to stress the impact of his 148 personality, and it is really quite extraordinary, and very heartening, to realise how the human face of Islamic art history reveals itself as they 149 do so. For some, the encounter was life-changing. They remember his 150 helping hand at a crucial time, for example at career crossroads; the way you should go to the Institute if you prefer to move from objects to ideas…Interestingly, later in his career, these differences in approach diminished, as Oleg increasingly became enamored of aesthetics and the visual pleasure of objects’ (see footnote 5[b] above).  145   Glenn Lowry remarks: ‘Some of his peers accused him of not liking objects but that was wrong – he was fascinated by them, but not because of their aesthetic qualities but because of the stories they told’ (personal message). But see the previous note. 146   In the luminous phrase of Jacob Lassner, he was ‘the Godfather of Islamic art history in this country’. See footnote 147 below. 147 I should particularly like to thank Jacob Lassner, Larry Nees and Nancy Steinhardt for letting me see, in advance of publication, their own extended reflections on Grabar. I have tried (though not with complete success) to resist the temptation to plunder these fascinating texts here, but together they provide an indispensable context for my own assessment, and in so doing have filled in numerous gaps in my knowledge. I should add that I deliberately left reading them to the very last moment before submitting this article to the editors, for I wanted to write it as much as possible from my own more distant perspective, as one whose career was on the other side of the pond and who had never studied or worked with Grabar.  148 Thus Esin Atıl writes: ‘It was his charismatic lecturing style and enthusiastic approach to presenting Islamic civilizations that attracted me’ (personal message).  149 For example Valerie Gonzales [memoryog.tumblr.com] or Gülru Necipoğlu: ‘he touched all stages of my adult life’ (see footnote 5[b] above).  150  From Cynthia Robinson: ‘thanks to you, I stuck with academia’; she then details one example after another of his readiness to offer help and crucial advice [memoryog.tumblr. com].

467

OLEG GRABAR: THE SCHOLARLY LEGACY

he remembered them after decades; and his challenging questions: ‘What 153 do you think the person, for whom this was made, thought of it?’ They have kept personal letters which reflect his sympathy at a bereavement or other crises, or his readiness to give career advice. They stress the 154 155 impact of his teaching: generous, opening wide perspectives, utterly 156 157 158 captivating, mesmerizing, mentoring and mind-expanding, and – a 159 frequent memory – they loved being treated as intellectual equals. They 151

152

 Ed Keall, who studied with Grabar in 1967-69, looks back: ‘when there was a celebratory event in Harvard for him (retirement maybe?), I opted to fly in to see him. When he saw me enter the Faculty Club to the over-packed event he came over and gave me a huge bear hug and said “Good god, I never expected to see you at an event like this.” I am so glad I went’ and – though emphatically an archaeologist, not an art historian – he acknowledges ‘my eventual own style of teaching and research was immensely influenced by his style’ (personal message). 152 Sheila Canby notes: ‘His writing challenges the reader to find answers that are rarely offered in full. Instead Grabar presented a range of possibilities. Even when arguing forcefully for a particular interpretation, Grabar rarely rubbished other points of view. He simply marshalled the literary, historical and inscriptional evidence to support his points. This methodology, when practiced carefully and thoroughly, is one of the greatest of Oleg Grabar’s contributions to scholarship’ (personal message).  153 Lisa Golombek, on being shown a mina’i plate depicting a seated prince (footnote 5[b] above). Cf. Sussan Babaie, who remembers him saying that the most fascinating question about the Isfahan murals was not what they were about but that they were there at all! (personal message).  154 Ülkü Bates: ‘I must underline “generous” because that was the outstanding quality of his teaching’ [memoryog.tumblr.com].  155 Lisa Golombek: ‘I found the seminars more interesting than the books… I feel that his greatest contribution was to open doors…rather than …the counting of stitches’ (personal message).  156 Barry Flood, who heard him teaching in French (personal message).  157 Perween Hasan: ‘I, like all the others was totally mesmerized by his lectures’ (personal message).  158 From Perween Hasan: He encouraged wobbly-kneed students such as myself to be bold and assertive, float new ideas that you automatically learned to defend. That was how Oleg influenced me most, to look at art history in the broader context of cultural history, where continuities as well as new ideas must be understood… He…brought out a rickshaw painting that had been a quirky retirement gift from me, reiterating that he loved it. It was of a kurta-pajama clad Oleg sitting on a lotus in front of the Taj Mahal with his hands in the ‘abhaya’ (fear not) mudra. I believe that was his most amazing gift to his students: not to be afraid to air new ideas. (Personal message.) 151

Cf. Jaclynne Kerner’s description of Grabar’s delight in this image, one ‘of the treasured objects sharing Oleg’s office space’ [memoryog.tumblr.com].  159 ‘In the classroom Oleg encouraged his students to speak up, throw wonderful new ideas and argue. I wondered if this amused him, letting us ‘pretend’ that we were on the

468

relished his sense of humour, and recall his giggle (or gurgle) with affection. What a gift – to make learning fun! They celebrate his love of the good 160 things of life. But for all this outpouring of happy memories, many of 161 them did not draw attention to a single piece of his written work. The implication is inescapable: even after his death he is being remembered vividly, by those who knew him, not for what he wrote but for what he was. For many of his students, he was a father figure or a favourite uncle. And there is little doubt that this emphasis will continue for a long time to come. It will therefore have its impact on a new generation of Islamic art historians who will never have encountered him personally but for whom he will come alive through the memories, the anecdotes and the example of their teachers. In a longer perspective, no doubt, his written work, winnowed by the impartial hand of time, will eventually come into its own. But there is something profoundly satisfying about the thought that, for now at least, it is the heart and spirit of the man, rather than the written products of his intellect, that remain a cherished memory. It is a reminder, and a celebration, of the crucial human dimension of what we as scholars do.

same intellectual plane’ (Perween Hasan; personal message).  160 One after another remember working lunches made memorable by his ‘expansive personality, generosity of spirit, love of food and conviviality’ (Gülru Necipoğlu; see footnote 5[b] above). Cf. Stephennie Mulder: ‘He loved life, elegance, good food and good company, but most of all he loved to be in the middle of something, stirring things up. He could be alternately brilliant and infuriating, one moment full of earnest intellectual urgency, the next, his eyes sparkling with some academic mischief ’ [memoryog.tumblr. com]. 161 Though a leitmotif of the responses has been an idea expressed succinctly by Judith Lemer: ‘his writing also influenced us and not only its content’ (personal message). 

XX Wine in Islamic art and society , disdaining the nuances of a complicated csubject, insists that Islam prohibits the consumption of alcohol. And onventional wisdom

there can be no doubt that in this case conventional wisdom is correct, both as regards the medieval period and in the Islamic world today, as epitomised by the frequently repeated saying that alcohol is ‘the mother 1 of all evils’. Yet the very definition of what constitutes alcohol has been 2 hotly contested. A typical case is that of a mild fermented liquid called nabidh, usually made from raisins or dates; its consumption was legal if it was no more than two days old, though thereafter it became stronger – and 3 illegal. It seems that orthodox attitudes to alcohol hardened over time. A Qur’anic verse (5:90–1) forbids prayer by anyone who is intoxicated, while Qur’an 2:219 brackets strong drink with games of chance as something to be avoided. A very clear distinction is traditionally drawn between earthly wine that causes drunkenness and the wine of Paradise, which does not 4 and is delicious. Anecdotes reflecting orthodox Muslim distaste for wine are frequently found in the literary sources. Thus the traveller Ibn Jubayr, when working as a secretary for the governor of Granada in 1182, was forced to drink seven cups of wine. The penitent governor thereupon rewarded him with seven cups filled with gold dinars, money which he used 5 to perform the hajj. And when the army of the Mongol Il-Khan Hülegü 1 Karic 2002, p. 557; cf. Fulton 1948 on al-Firuzabadi’s list of the ills connected with wine and his compilation of almost 450 names for wines. 2   Ahsan 1979, p. 111, with copious references. 3 On nabidh see Wensinck 1978, p. 996a; Sadan 1977, p. 134; Jahiz/Pellat 1969, pp. 52–53; McAuliffe 1984, pp. 164–65. 4 McAuliffe 1984, pp. 160–63, 166 and 168–70. 5 Ibn Jubayr/Broadhurst 1952, p. 16.

470

entered Damascus, the triumphant Christians sprinkled the Muslims with 6 wine. But such stories need to be balanced against an overwhelming body of evidence indicating that, from Umayyad times onwards, caliphs, nobles 7 and courtiers habitually drank alcohol, often to excess. Among Umayyad c rulers Abd al-Malik, Yazid I, al-Walid I, Sulaiman and Hisham drank 8 wine regularly, the latter directly after the Friday prayer; and al-Walid II 9 surpassed them all. And most of the cAbbasid caliphs also drank wine, 10 some coming back to it after having forsworn it as a vice; indeed, the 11 stock reason for deposing a ruler was that he was addicted to wine, a vice often associated with an equally reprehensible love of music. Some of the practices associated with wine-drinking in early Islamic times may have Sasanian origins, for a complex ceremonial surrounded the consumption of wine at the Sasanian court. It was to be perfumed, consumed with dried fruit, a sandwich or candied myrobalan, taken with a meal (a practice not 12 13 followed in Islamic times) or used as a dip for oily pastry. Conventional wisdom of a more specialised kind, but equally unconcerned with the finer details of the matter, insists that scenes of wine-drinking are common in medieval Islamic art. It therefore comes as something of a surprise to discover that no general survey of the pictorial evidence for this assertion has been published, although scholars have 14 investigated in close detail wine-drinking in the art of Samarra and of 15 the early Turks. Those two themes converge in the many reports that the Turks of Samarra were notorious for their drunkenness. The literary references to wine-drinking from early Islamic times onwards are copious, as is the scholarship on wine poetry (khamriyyat), a distinct specialised genre. Clearly, then, there is a significant information gap between the written sources execrating wine-drinking and the pictorial evidence to the Spuler 1985, p. 173.  For a learned survey of the whole subject, see Sadan 1977. 8 R. Hillenbrand 1982, pp. 13 and 28. But the evidence for Hisham is contradictory; see Hamilton 1988, p. 111. 9  Derenk 1974, pp. 33–35, 59–62; R. Hillenbrand 1982, pp. 11–14, 18 and 28–29; Hamilton 1988, pp. 35–37, 88–89, 91–3, 118–19, 122, 124–25, 149 and 164; and Rotter 2004, pp. 105, 109, 115, 117, 122, 125 and 130. 10 E.g. al-Qahir, al-Radi and al-Mustakfi (Mez 1937, pp. 397–98, quoting respectively Miskawayh, al-Suli and al-Mas‘udi). 11 As in the case of the deposition of the caliph al-Rashid in 1136 (C. Hillenbrand 1990, p. 78, quoting Ibn al-Azraq). 12 Mez 1937, p. 396; Ahsan 1979, p. 164.  13   Azarnouche 2013, p. 57. 14   D.S. Rice 1958, with detailed descriptions of particularly select wines. 15   Esin 1969, pp. 241–57. 6 7

WINE IN ISLAMIC ART AND SOCIETY

471

contrary. But scholars agree that the aristocracy drank freely from early Islamic times onwards; indeed, a treatise on good manners specifies what a gentleman should know about the types and vintages of wine; and in the same spirit, the caliph al-Mustakfi convened a symposium on cooking at which the court poets vied with each other to describe the most recondite 16 dishes in appropriately extravagant terms. A drinking party where the host (the vizier Hasan b. Makhlad) drank many pints of wine before passing out and keeling over backwards on to his cushion is recorded. There is even a reference to taxes on wine in Diyar Rabi‘a in the early tenth century. The most sustained evidence for this widespread drinking of alcohol 17 lies in wine poetry. Yet one of its most constant themes, alongside love and obscenity, hedonism and blasphemy, is repentance for indulgence in alcohol. The undisputed master of this genre was the courtier and libertine Abu Nuwas (d. 815), who greatly extended its range of expression by means of the dialogue form, by evocations of the setting for serious drinking (descriptions of cup-bearers and singers, flowers and gardens, and even drinking vessels), and by the introduction of bold similes (fire, jewels, light and perfume). And wine figures in many other types of poetry too: thus the cAbbasid caliph al-Mahdi wrote: “In song and wine is my 18 felicity; And perfumed girls, music and gaiety”. A favourite trope in erotic 19 poetry was to liken the saliva of the beloved to wine, and the colour of wine to blood. Mystical khamriyyat poetry abounds and uses wine as a metaphor denoting the intoxication felt by the Sufis as they achieve closer proximity to God. The imagery can be extravagant, as in the “wine20 coloured” or “drunken” eyes of the beloved. Often it is liberally sprinkled with paradoxes, as in the lines by Bayazid Bastami (d. c. 874): “I am the 21 wine drinker and the wine and the cupbearer”. Or, as al-Hallaj put it: “When in my thirst I stooped my face to wine, Dark in the cup I saw a 22 shadow. Thine!” But yet other quotations make it clear that wine could also serve as an entirely religious metaphor. Thus Ibn al-Farid (d. 1235) wrote in the  Arberry 1967, pp. 156–64 (a translation of al-Mas‘udi, Muruj al-dhahhab). Bencheikh 1978; Kennedy 1997; Noorani 2004, pp. 345–46, explores, in the context of the wine poetry of Abu Nuwas, the notion that “the elite excesses of enjoyment … are rooted in the same moral order that allies political dominance with virtue”. 18  Tr. Schroeder 1955, p. 281. 19 Hamilton 1988, p. 105. 20 Nurbakhsh 1980, p. 80. 21  Rumi/Barks 1999, p. 288. 22  Tr. Schroeder 1955, p. 525. 16 17

472

tawil metre: “We drank upon the remembrance of the Beloved a wine wherewith we were drunken before ever the vine was created. The full 23 moon was a cup for it, itself being the sun.” Drunkenness is frequently recorded, from al-Walid II with his wine 24 pool at Khirbat al-Mafjar to the Il-Khan Abaqa, who succumbed to 25 delirium tremens, and many a prince in Mughal India died of drink. The Seljuq sultan Alp Arslan threw a tent-peg at his venerable vizier Nizam al-Mulk in a drunken rage (“I bumped against a tent-pole”, said the vizier 26 tactfully when his master enquired about his bruise the next morning). The Turkish warlord Il-Ghazi celebrated his victory over the Crusaders at the Field of Blood by retiring to his tent for a drunken debauch that 27 28 lasted a week, while the fearsome Zangi was assassinated in his cups. The standard issue of wine to the Mongol army, following a decree by Chinggis Khan, led to widespread alcohol abuse. Al-Jahiz, a person of encyclopaedic knowledge who was widely regarded as the finest of all classical Arabic stylists, wrote a scintillating disquisition on the very varied behaviour of drunken people, with no hint of disapproval; its climax is the account of a litigator who never appeared before a judge until he had drunk ten bottles, and then his brilliant oratory would infallibly win the day. 29 How, then, is the theme of drinking presented in medieval Islamic art? Even a rapid and superficial overview of the visual material that falls most easily to hand is enough to reveal some interesting anomalies. For example, surviving images, it seems, offer no clue as to the nature of the liquid in question; it could as easily be water, sherbet (a word of Arabic derivation) or fruit juice as wine or spirits. One looks in vain for diminutive bottles of the kind that might hold spirits, although the frontispiece to a manuscript of Sacdi’s Bustan from the Herat school, dated 1488, depicts a working 30 still. Instead, the standard containers are large jugs or ewers, most of them with a capacity of several litres, though wine-skins were also used. Amphorae were used for storing larger quantities of wine. Vessels for Arberry 1965, pp. 126–27. Hamilton 1959, pp. 53–55. 25  Spuler 1985, p. 369. 26  C. Hillenbrand 1995, p. 240. 27 C. Hillenbrand 1981, pp. 277 and 289, with copious references to the largely Arabic sources. 28  C. Hillenbrand 2001, p. 120. 29 For a rapid inventory of examples of this theme, see Sadan 1977, pp. 137–38. 30  Washington 1989, p. 261. 23 24

WINE IN ISLAMIC ART AND SOCIETY

473

holding alcohol of lesser strength, such as beer and fermented mare’s milk (kumis), much favoured by the Mongols and others of Turkic origin, were of different shape. According to William of Rubruck, kumis was made and 32 stored in skins with a mouthpiece tied with string. It was served in cups called kasa and there was often a strong ceremonial element in the use of 33 34 these at court. Jugs were also used. As for the drinking vessels used for wine, they too were very varied, and might be of glass, ceramic, brass or precious metal. They ranged from delicate beakers or shallow bowls or 35 36 ewers to hefty goblets or even the skulls of enemies. A special category concerns the ritual drinking of wine in crescent-shaped cups evoking the merging of the sun and the moon at sunset, and giving rise to the image of a boat sailing in a sea of wine. This concept, which can be traced back to Achaemenid times, has been thought to have acquired mystical significance in Sufi thought and to have found visual expression in the boat-shaped beggar’s bowl (kashkul) used by wandering dervishes into 37 modern times. While rhytons of Greek and thence Parthian form appear to have fallen into disuse with the coming of Islam, a closely related and still more ancient type – the wine leg – is recorded, and survived into the 38 Islamic era. Bull-shaped containers for drinking wine also have a long 39 tradition behind them and had Zoroastrian associations; other forms of 40 41 vessel included horns and birds, and wine-drinking played a key part in the celebration of Nowruz, the Iranian New Year, from pre-Islamic times 42 onwards. 31

 Ghouchani and Adle 1992.  Dawson 1966, pp. 96 and 98. 33  Spuler 1985, p. 367; Dawson 1966, p. 96. For an impressive list of citations to the ceremonial use of kasa see Rashid al-Din/Quatremère 1836, p. 354, n. 155 34 Spuler 1985, p. 146. 35   Ghouchani 1999 deals with the content of the wine poems that they bear. 36 It is recorded that in 1108 the Seljuq warlord Tughtegin drank from the skull of his scalped enemy, the Crusader knight Gervase de Basoches (C. Hillenbrand 2011, p. 463; see pp. 465 and 467 for further examples of the practice among the Scythians, Tibetans and Safavids, and cf. Dawson 1966, p. 142 for more details of Tibetan practice).  37 Melikian-Chirvani 1991 explores these themes with an impressively wide range of reference, drawing extensively on the evidence provided by medieval Persian poetry. 38 See Melikian-Chirvani 1997, pp. 82–84, for an example from Afrasiyab. 39  Melikian-Chirvani 1991 and 1992. 40 Melikian-Chirvani 1996; cf. also Melikian-Chirvani 1991. 41  Melikian-Chirvani 1995. The wine leg, used for both beakers and decanters, and sometimes spouted, is an even more remarkable receptacle for wine-drinking than a boat or a bird. For its survival into the medieval period in Iran, see Melikian-Chirvani 1997b, especially pp. 82–87. 42 Melikian-Chirvani 1996, pp. 115–24. 31 32

474

The absence of scenes of conviviality in which many people are shown drinking is noteworthy, nor do scenes of competitive drinking or wine tasting survive. Instead, a strict formality is maintained, signified for example by an isolated figure of authority sitting cross-legged in a rigidly frontal pose. Indeed, scenes of actual drinking – as distinct from images of a person of rank holding a goblet or beaker – are virtually unknown. There is too little sense of how far the custom of wine-drinking spread to the wider world outside the court. Sometimes a specific sense of context enriches images of the royal drinker, featuring such elements as the recitation of poetry, music-making, dancing or acrobatic displays. The conclusion that gradually imposes itself is that wine-drinking as depicted in medieval Islamic art is not an independent iconographic theme that is developed in its own right. It quickly stabilised as a visual cliché – clearly the fact that wine was forbidden in orthodox Islam caused artists no embarrassment. On the other hand, it is an indispensable component of the visual panorama of what might be termed the good life – the pictorial equivalent of the formulaic blessings called down on the owner of an object by its benedictory inscriptions. On the Courtauld bag, leaving aside the major narrative band in which drinking plays perhaps the key role, no less than six independent roundels depict people holding wine beakers (fig. 1). sometimes a wine jug too or a beaker in each hand. This is counterbalanced by an astonishing nine roundels depicting musicmaking. Riders account for only two roundels. No other piece of medieval Islamic metalwork comes close to this sustained emphasis on wine and music, to the virtual exclusion of other amusements. It reflects the close interdependence of these two themes and is a clear signal that these were the principal entertainments of at least one contemporary court. The accounts by al-Mascudi, al-Tabari, al-Isfahani and others of high life in c Abbasid Baghdad flesh out this statement with abundant corroborative detail, and it is this gilded ambience which set the gold standard for subsequent courts. It is time to consider how the theme of wine is treated in the art of the twelfth to fourteenth centuries, the period which provides a broad context for the Courtauld bag. Here the choice of medium is decisive. This is precisely the time from which the earliest Islamic (mainly Arab) illustrated manuscripts survive, and by a fortunate coincidence it also saw the floruit of Islamic inlaid metalwork. These are the two major sources for images of drinking in the medieval Islamic world, closely followed by luxury ceramics, in which lovers may be shown in intimate conversation,

WINE IN ISLAMIC ART AND SOCIETY

475

one plucking a harp and the other holding a goblet of wine, or a monarch holding a goblet may be presented amidst serried ranks of courtiers. Similar themes turn up in the ivory carving of Umayyad Spain and Fatimid Egypt – in the latter case (a long strip of animated figures in constantly varied poses), a seated figure serenaded by a quartet of female musicians pours himself a drink from a ewer, while further along a youth flanked by 43 female lute-players and holding a beaker reclines languidly on cushions. The close connection already noted between drinking and music could not be clearer. 44 And so to the evidence of manuscripts. Obviously enough, painting on paper allows far more latitude for the depiction of detail. Thus a copy of al-Hariri’s Maqamat dated 643/1237, whose 99 pictures hold up an unsparing mirror to life in the raw in thirteenth-century Baghdad, the foremost metropolis of the Islamic world, contains its full share of images of feasting, merry-making – and the pleasures of the tavern. In one such image the confidence trickster Abu Zayd occupies the place of honour, enthroned in a broad parody of a standard contemporary ruler 45 image (fig. 2). Like the well-bred gentleman that he is not, he sports a mandil, the napkin that was a must-have accessory for any drinking bout. Beside him a musician strums his lute. Other nattily dressed drinkers occupy the upper storey, and large wine storage jars are stacked at both levels. A topless slave-girl standing in a trough crushes the fruit with her feet; the juice runs into a bowl, while next to her a boy pours the liquid from a storage jar into a straining cloth over an amphora suspended by a tripod, with the purified red wine dripping into a bigger bowl. Meanwhile a servant leans over the balcony to hand a colleague a small two-handled wine jar. Despite its simplified format, this image presents several stages in the operation of a wine shop, from the making of the product to its consumption, though there is no reference to the spices that were added to finer wines – cardamom, turmeric, ginger, musk and so on. Nor is 46 there any sign here of the Christians (and Jews) who were traditionally associated with this industry. In fact Christian monasteries were, from

Spuler and Sourdel-Thomine 1973, p. 261, pl. 191 (entry by J. Zick-Nissen). For a tally of some significant examples of the theme, see Sadan 1977, pp. 132–33 and 138. 45 Stewart 1967, repr. p. 95.  46   Rotter 2004, pp. 118 and 212. 43 44

476

Umayyad times onwards, a favoured discreet location for Muslims intent 47 on heavy drinking. This Maqamat image has an unmistakable flavour of low life. On the other hand, two frontispiece paintings – the prestigious location is important, for it shows that the patron felt no need to hide his drinking – for books of science or light entertainment vividly evoke the aristocratic context in which wine-drinking was normal. Here they must do duty for a number of broadly similar painted images of the same period. The first, probably produced in Mosul or Northern Iraq c. 1250, illustrates a manuscript dealing with antidotes to snakebite (fig. 3). Here an entire 48 court is on display, with the central band depicting at each side the principal officials bearing the attributes of their posts, while an oblong panel above shows a hunting scene and a matching predella has court ladies and their attendants. At centre stage sits the ruler, holding a glass beaker full of red liquid. Meat is being roasted on a barbecue, while a table in the background holds a ewer and a wine-jug. The second prefaces a copy of al-Hariri’s Maqamat probably produced in Damascus in 734/1334 and its atmosphere is noticeably more formal; this is largely due to the presence of winged angels holding a cloth over the ruler’s head, mimicking a baldachin (fig. 4). An acrobat contorts himself into an improbable pose on a bowl directly at the foot of the throne, with musicians playing on either side. The ruler himself, again enthroned cross-legged, holds a mandil 49 and a beaker. Even closer in date and spirit to the Courtauld bag is the image from the Compendium of Chronicles of Rashid al-Din (714/1314–15) depicting the Seljuq Sultan Malikshah holding a wine cup just presented to him on a matching salver by a kneeling servant, while a seated female 50 harpist plays to the assembled company (fig. 5). The best parallel of all, however, comes from a series of scenes in the Diez and Istanbul albums depicting Mongol court receptions. In several of these the ruler shares his throne with his consort and both are drinking or are about to do so 51 (cat. 16, 17). Several Il-Khanid Shahnamas feature, as components of a royal image, a low table in front of the royal throne bearing a pair of long-necked ewers, clear evidence that the consumption of alcohol was an established element in court ceremonial (cat. 2).  Ahsan 1979, p. 271, quoting Shabustari.  Ettinghausen 1962, pp. 91–92 49 Ettinghausen 1962, pp. 147–49. 50 D.T. Rice 1976 51  New York 2002, pp. 80 and 82. 47 48

WINE IN ISLAMIC ART AND SOCIETY

477

A drinking vessel, whether cup or beaker, held against the chest 52 sometimes had a symbolic function, for from pre-Islamic times a Turkic icon of royal authority was the cup, as shown by the stone figures (balbals) distributed widely over the inner Asian steppe. A famous image from the painted ceiling of the Cappella Palatina in Palermo (after 1143) shows a seated ruler holding before him a large beaker filled with red liquid; this is at the centre of the image, and is clearly intended as an attribute of sovereignty since an attendant is pouring him a much smaller beaker 53 of wine from a pitcher. Thus in images of the royal wine-drinker one must be alert to a possible ceremonial dimension in its meaning. And the cup has yet further symbolic significance as a symbol of the sky itself. A related issue (to which Hafiz refers in his mathnawi beginning “Biya Saqi 54 …”) is the notion of the magical world-revealing cup of the mythical ruler of ancient Iran, Jamshid, who himself is credited with inventing 55 wine. The difficulty here is that there is no standard format for either a drinking vessel or a ceremonial cup. The other medium in which wine-drinking is a favoured theme is inlaid metalwork of the twelfth to the fourteenth century. Several standard scenes (excluding the ruler and his consort enthroned while drinking) occur – the drinker reclining on a couch, served by an attendant with a pouring vessel; a seated ruler turning to an attendant to have his outstretched beaker replenished; and, most common of all, the frontally depicted figure of authority seated cross-legged with a drinking cup or similar receptacle held at chest level (see fig. 39). This last formula is sometimes enriched by the addition of flanking attendants and musicians below. None of these compositions are new, and indeed many can be parallelled in Sasanian or post-Sasanian silverwork. What distinguishes the later versions of these themes is above all their visual context, which is one of merrymaking – not only drinking but eating, music-making, dancing, hunting and so on. These interconnected themes are, moreover, explicitly linked by the layout of these metalwares, which comprise a series of roundels, medallions or cartouches each devoted to one aspect of the good life. As the Ghaznavid poet Manuchihri Damghani writes: “We’re men of drinking, feasting and singing. Hurray for the rebab, kebab 56 and wine.”  Barry 2010, p. 151.  Ettinghausen 1962, pp. 45 and 47. 54  Arberry 1962, pp. 129–30; Hafiz 1941, p. 356. 55  Simpson 2013, pp. 353–54. 56  Clinton 1972, p. 12 (translation slightly adapted). 52 53

478

1. The Courtauld bag: detail of roundels with revellers

WINE IN ISLAMIC ART AND SOCIETY

479

2. ‘Tavern scene’, from the Schefer Maqamat, illustrated by Yahya ibn Mahmud al-Wasiti, Baghdad, 634/1237 Ink, colours and gold on paper, H: 20 cm, W: 24 cm (image) Paris, Bibliothèque nationale, Ms. Arabe 5847. fol. 33

480

3. Frontispiece from the Book of Antidotes (Kitab al-Diryaq), probably Mosul, mid-13th century. Ink, colours and gold on paper, H: 32 cm, W: 25 cm (folio) Vienna, Nationalbibliothek, a.f.10, fol. 1r

WINE IN ISLAMIC ART AND SOCIETY

4. Frontispiece from the Vienna Maqamat, probably Damascus, 734/1334. Ink, colours and gold on paper, H: 19.2 cm, W: 17.5 cm (image) Vienna, Nationalbibliothek, Cod. a.f.9, fol. 1r

481

5. ‘The court of Sultan Sanjar ibn Malik Shah’, from Rashid al-Din’s Compendium of Chronicles, Tabriz, 714/1314–15 Ink, colours and gold on paper. Edinburgh University Library, Ms. Ar. 20, fol. 142r

482

WINE IN ISLAMIC ART AND SOCIETY

483

Consolidated bibliography for chapters XIII and XX Ahsan 1979 M. M. Ahsan, Social Life under the Abbasids, London and New York. Allen 1980 M. Allen, Falconry in Arabia, London. Almagro et al. 1975 M. Almagro, L. Caballero, J. Zozaya, and A. Almagro, Qusayr ‘Amra. Residencia y baños omeyas en el desierto de Jordania, Madrid. Anisi 2007 A. Anisi, Early Islamic Architecture in Iran (637-1058). Unpublished PhD thesis, University of Edinburgh. Arberry 1962 A.J. Arberry, Fifty Poems of Hafiz, Cambridge. Arberry 1965 A.J. Arberry, Arabic Poetry. A Primer for Students, Cambridge. Arberry 1967 A.J. Arberry, Aspects of Islamic Civilization as Depicted in the Original Texts, Ann Arbor. Atıl et al. 1985 E. Atıl, W.T. Chase and P. Jett, Islamic Metalwork in the Freer Gallery of Art, Washington DC. Azarnouche 2013 S. Azarnouche (ed. and tr.), Husraw i Kawadan ud Redag-é Khosraw fils de Kawad et un Page, Cahiers de Studia Iranica 49, Paris. Bahrami 1952 M. Bahrami, ‘A Gold Medal in the Freer Gallery of Art’, in G.C. Miles (ed.), Archaeologia Orientalia in Memoriam Ernst Herzfeld, Locust Valley, 5-20.

484

Barry 2010 M. Barry, “The Persianised East and the Royal Tradition”, in V. Daiber and B.

Junod (eds), Treasures of the Aga Khan Museum. Masterpieces of Islamic Art, Berlin, 149-52.

Bartol’d 1922 V.V. Bartol’d, ‘Vostochno-iranskii vopros’, Izvestiya Materialnoi Kulturi II, 361-84. Bencheikh 1978 J.E. Bencheikh, “Khamriyya”, in Encyclopaedia of Islam, 2nd ed., IV, 998a-1009a. Berchem, van 1918 M. van Berchem, “Die Inschriften der Grabtürme”, in E. Diez, Churasanische Baudenkmäler, Berlin, 87-116. Blair et al. 1992 S.S. Blair, J.M. Bloom, and A.E. Wardwell, ‘Reevaluating the Date of the ‘Buyid’ Silks by Epigraphic and Radiocarbon Analysis’, Ars Orientalis XXII, 1- 41. Bréhier 1936 L. Bréhier, La sculpture et les arts mineurs byzantins, Paris. Chelkowski 1975 P.J. Chelkowski, The Scholar and the Saint: Studies in Commemoration of Abu Ί-Rayhan al-Biruni and Jalal al-Din al-Rumi, New York. Clinton 1972 J.W. Clinton, The Diwan of Manuchihri Damghani. A Critical Study, Minneapolis. Curtis and Tallis 2005 J. Curtis and N. Tallis, Forgotten Empire: The World of Ancient Persia, London. Dawson 1966

WINE IN ISLAMIC ART AND SOCIETY

485

C. Dawson, Mission to Asia, New York. Derenk 1974 D. Derenk, Leben und Dichtung des Omaiyadenkalifen al-Walid ihn Yazid - ein

quellenkritischer Beitrag, Frankfurt.

Diba 1965 P. Diba, Les trésors de Than et le vase en or des Mannéens, Paris. Erdmann 1936 K. Erdmann, ‘Die sasanidische Jagdschalen’, Jahrbuch der preusszischen Kunstsammlungen LVII, 193-231. Esin 1969 E. Esin, “‘AND’. The cup rites in Inner-Asian and Turkish art”, in O. Aslanapa and R. Naumann (eds), Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann. 9. September 1901-30. September 1964, Istanbul, 224-61. Ettinghausen 1962 R. Ettinghausen, Arab Painting, Geneva. Ettinghausen 1969 R. Ettinghausen, ‘A Case of Traditionalism in Iranian Art’, in O. Aslanapa and R. Naumann (eds), Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann. 9. September 1901-30. September 1964, Istanbul, 88-110. Ettinghausen 1972 R. Ettinghausen, From Byzantium to Sasanian Iran and the Islamic World. Three Modes of Artistic Influence, Leiden. Ettinghausen 1979 R.Ettinghausen, ‘Bahram Gur’s Hunting Feats or the Problem of Identification’, Iran XVII, 25-31. Fajans 1957 S. Fajans, ‘Recent Russian Literature on Newly Found Middle Eastern Metal Vessels’, Ars Orientalis II, 55-76.

486

Falke, von 1913 O. von Falke, O., Kunstgeschichte der Seidenweberei, I-II, Berlin. Fukai et al. 1972 S. Fukai et al., Taq-i Bustan II, Tokyo. Fukai et al. 1984 S. Fukai et al., Taq-i Bustan IV, Tokyo. Fulton 1948 A.S. Fulton, “Firuzabadi’s “wine-list””, Bulletin of the School of Oriental and African Studies

12/3-4, 579-85.

Gall, von 1990 H. von Gall, Das Reiterkampfbild in der iranischen und iranisch beeinflußten Kunst parthischer und sasanidischer Zeit, Teheraner Forschungen VI, Berlin. Ghirshman 1962 R. Ghrshman, Iran: Parthians and Sassanians, tr. S. Gilbert and J. Emmons, London. Ghirshman 1964 R. Ghirshman, The Art of Ancient Iran from its Origins to the Time of Alexander the Great, tr. S. Gilbert and J. Emmons, New York. Ghouchani 1999 A. Ghouchani, “Some 12th Century Iranian Wine Ewers and Their Poems”, Bulletin of the Asia Institute,

New Series 13, 141-50.

Ghouchani and Adle 1992 A. Ghouchani and C. Adle, “A Sphero-Conical Vessel known as Fuqqa‘a, or a Gourd for “Beer””, Muqarnas 9, 1992, 72-92 Godard 1936

WINE IN ISLAMIC ART AND SOCIETY

487

A. Godard, “Les tours de Ladjim et de Resget”, Athar-è Iran 1/1, 109-21. Grabar and Carter 1967 O. Grabar and M.L. Carter, Sasanian Silver: Late Antique and Early Mediaeval Arts of Luxury from Iran, Ann Arbor. Grabar 1990 O. Grabar, The Great Mosque of Isfahan, New York. Gurgani 1972 Gurgani, tr. G. Morrison, Vis and Ramin, Translated from the Persian of Fakhr ud-Din Gurgani, New York and London. Hafiz 1320/1941 Hafiz, Divan-i Hafiz, eds M. Qazvini and Q. Ghani, Tehran. Hamilton 1988 R.W. Hamilton, Walid and his Friends. An Umayyad Tragedy, Oxford. Harper 1978 P. Harper, The Royal Hunter: Art of the Sasanian Empire, New York. Harper 1981 P. Harper (with a technical study by P. Meyers), Silver Vessels of the Sasanian Period I, New York. Hayashi 1975 R. Hayashi, tr. R. Ricketts, The Silk Road and the Shoso-in, New York and Tokyo. Herrmann 1977 G. Herrmann The Iranian Revival, Oxford. Herzfeld 1921 E. Herzfeld, “Khorasan: Denkmalsgeographische Studien zur Kulturgeschichte des Islam in Iran”, Der Islam 11, 107-74. Herzfeld 1932 E. Herzfeld, “Postsasanidische Inschriften I — Mil i Radkan”, Archaeologische Mitteilungen aus Iran 1, 140-7.

488

Herzfeld 1936 E. Herzfeld, “Arabische Inschriften aus Iran und Syrien”, Archaeologische Mitteilungen aus Iran 8, 78-102. C. Hillenbrand 1981 C. Hillenbrand, “The career of Najm al-Din Il-Ghazi”, Der Islam 58/2, 250-92. C. Hillenbrand 1990 C. Hillenbrand, A Muslim Principality in Crusader Times. The Early Artuqid State, Istanbul. C. Hillenbrand 1995 C. Hillenbrand, “Ibn al-‘Adim’s biography of the Seljuq sultan Alp Arslan”, in C.V. de Bonito and M.A.M. Rodriguez (eds.), Actos XVI Congreso UAEl, Salamanca, 237-42. C.Hillenbrand 2001 C. Hillenbrand, “The career of Zengi”, in J. Phillips and M. Hoch (eds), The Second Crusade: Scope and Consequences, Manchester, 111-32. C. Hillenbrand 2011 C. Hillenbrand, “What’s in a Name? Tughtegin - “The Minister of the Antichrist”?” in O.A. de Unzaga (ed.), Fortresses of the Intellect. Ismaili and other Islamic Studies in Honour of Farhad Daftary, London and New York, 459-71. R. Hillenbrand 1982 R. Hillenbrand, “La dolce vita in early Islamic Syria: the evidence of later Umayyad palaces”, Art History V/1, 1-35. R. Hillenbrand 2006 R. Hllenbrand, “The Islamic Re-working of the Sasanian Heritage: Two Case Studies”, in P. Baker and B. Brend (eds), Studies in Honour of Géza Fehérvári, London, 1-14. Ibn Jubayr/Broadhurst 1952 Muhammad ibn Ahmad ibn Jubayr, The Travels of Ibn Jubayr, tr. R.J.C.

WINE IN ISLAMIC ART AND SOCIETY

489

Broadhurst, London. Jahiz /Pellat 1969 ‘Amr ibn Bahr Jahiz, The Life and Works of Jahiz. tr. C. Pellat; tr. from the French by D.M. Hawke, London. Ierusalimskaja and Borkopp 1996 A.A. Ierusalimskaja and B. Borkopp, Von China nach Byzanz: frühmittelalterliche Seiden aus der Staatlichen Ermitage St. Petersburg, Munich. Karic 2002 Enes Karic, “Intoxicants”, in Encyclopaedia of the Qur’an, ed. Jane D. McAuliffe, Leiden and Boston, II, pp. 555b-57b. Kennedy 1997 P.F. Kennedy, The Wine Song in Classical Arabic Poetry. Abu Nuwas and the Literary Tradition, Oxford. Kühnel 1956 E. Kühnel, “Die Kunst Persiens unter den Buyiden”, Zeitschrift der deutschen morgenländischen Gesellschaft 106, 78-92. McAuliffe 1984 J.D. McAuliffe, “The Wines of Earth and Paradise. Qur’anic Proscriptions and Promises”, in R.M. Savory and D.A. Agius (eds), Papers in medieval studies. 6. Logos Islamikos. Studia Islamica in Honorem Georgii Michaelis Wickens, Toronto, 159-74. Madelung 1969 W. Madelung, “The Assumption of the Title Shahanshah by the Büyids and ‘The Reign of the Daylam (Dawlat al-Daylam)’”, Journal of Near Eastern Studies 28, 84-108, 169-83. Meister 1970 M.W. Meister, “The Pearl Roundel in Chinese Textiles”, Ars Orientalis VIII, 255- 67. Melikian-Chirvani 1974

490

A.S. Melikian-Chirvani, “L’évocation littéraire du bouddhisme dans l’Iran musulman”, Le monde iranien et l’Islam: sociétés et cultures II, 1-72. Melikian-Chirvani 1975 A.S. Melikian-Chirvani, “Recherches sur l’architecture de l’Iran bouddhique I: Essai sur les origins et le symbolisme du stupa iranien”, Le monde iranien et l’Islam: sociétés et cultures III, 1-61. Melikian-Chirvani 1991 A.S. Melikian-Chirvani, “Les taureaux à vin et les cornes à boire de l’Iran islamique”, in P. Bernard and F. Grenet (eds.). Histoire et cultes de l’Asie Centrale préislamique, Paris, 101-25. Melikian-Chirvani 1992 A.S. Melikian-Chirvani, “The Wine-Bull and the Magian Master”, Studia Iranica (Cahier 11, Recurrent Patterns in Iranian Religions), Paris, 10134. Melikian-Chirvani 1995 A.S. Melikian-Chirvani, “The Wine Birds of Iran from Pre-Achaemenid to Islamic Times”, Bulletin of the Asia Institute, New Series 9, 41-97. Melikian-Chirvani 1996 A.S. Melikian-Chirvani, “The Iranian Wine Horn from Pre-Achaemenid Antiquity to the Safavid Age”, Bulletin of the Asia Institute, New Series 10, 85-139. Melikian-Chirvani 1997 A.S. Melikian-Chirvani, “The Iranian Wine Leg from Prehistory to Mongol Times”, Bulletin of the Asia Institute, New Series 11, 65-91. Mez 1937 A. Mez, The Renaissance of Islam, tr. S.K. Bakhsh and D.S. Margoliouth, Patna. Minorsky 1932 V. Minorsky, La domination des Daïlamites, Paris, 1932. (Reprinted in Iranica: Twenty Articles [Bist maqla-yi Mînurski, Tehran 1964, 12-30.) New York 2002

WINE IN ISLAMIC ART AND SOCIETY

491

L. Komaroff and S. Carboni (eds), The Legacy of Genghis Khan: Courtly Art and Culture in Western Asia, 1256-1353, New York, New Haven and London. Noorani 2004 Y. Noorani, “Heterotopia and the wine poem in early Islamic culture”, International Journal of Middle East Studies 36, 345-66. Nurbakhsh 1980 Nurbakhsh, What the Sufis Say, New York. Oddy 1991 W.A. Oddy, “Arab Imagery on Early Islamic Coins in Syria and Palestine: Evidence for Falconry”, Numismatic Circular 151, 59-66. Pierce and Tyler 1936 H. Pierce and R. Tyler, “The Prague Rider-Silk and the PersianByzantine Problem”, The Burlington Magazine LXV, 213-20. Rashid al-Din/Quatremère 1836 Rashid al-Din, Histoire des Mongols de la Perse, ed. and tr. É. Quatremère, Paris. Rice (David Storm) Rice 1958 D.S. Rice, “Deacon or drink? Some paintings from Samarra reexamined”, Arabica 5, 15-33. Rice (David Talbot) 1976 D.T. Rice, The Illustrations to the ‘World History’ of Rashid al-Din, ed. B. Gray, Edinburgh. Richter-Bernburg 1980 L. Richter-Bernburg, “Amir-Malik-Shahanshah: ‘Adud ad-Daula’s Titulature Reexamined”, Iran 18, 83-102. Rostovtzeff, M. I. (1943) M.I. Rostovtzeff, “The Parthian Shot”, American Journal of Archaeology 47/2, 174-87. Rotter 2004

492

G. Rotter, Abu’l Faradsch. Und der Kalif beschenkte ihn reichlich, Lenningen. Rumi/Barks 1999 Jalal al-Din Rumi, The Essential Rumi, tr. C. Barks, London. Sadan 1977 J. Sadan, “Vin - fait du civilisation”, in M. Rosen-Ayalon (ed.), Studies in Memory of Gaston Wiet, Jerusalem, 129-60. Schroeder 1955 E. Schroeder, Muhammad’s People, Portland, Maine. Shalem 2004 A. Shalem, “Bahram Gur Woven with Gold: A Silk Fragment in the Diocesan Museum of St. Afra in Augsburg and the Modes of Rendition of a Popular Theme”, in R. Hillenbrand (ed.), Shahnama: the Visual Language of the Persian Book of Kings, Aldershot, 117-27. Simpson 2013 M.S. Simpson, “A medieval representation of Kay Khusraw’s jam-i giti namay”, in R. Hillenbrand, A.C.S. Peacock and F. Abdullaeva (eds), Ferdowsi, the Mongols and the History of Iran. Art, Literature and Culture from Early Islam to Qajar Persia: Studies in Honour of Charles Melville, London and New York, 351-8. Siroux 1947 M. Siroux, “La mosquée djum‘a de Yezd-i-Khast”, Bulletin de l’Institut Français d’Archéologie Orientale XLIV, 101-18. Smith 1940 M.B. Smith, “Three Monuments at Yazd-i Khwast”, Ars Islamica VII, 104-6. Soucek 1974 P. Soucek, “Farhad and Taq-i Bustan: The Growth of a Legend”, in P.J. Chelkowski (ed.), Studies in Art and Literature of the Near East in Honor of Richard Ettinghausen, New York and Salt Lake City, 27-52. Spuler 1952

WINE IN ISLAMIC ART AND SOCIETY

493

B. Spuler, Iran in früh-islamischer Zeit: Politik, Kultur, Verwaltung und öffentliches Leben zwischen der arabischen und der seldschukischen Eroberung 633 bis 1055, Wiesbaden. Spuler 1985 B. Spuler, Die Mongolen in Iran. Politik, Verwaltung und Kultur der llchanzeit 1220-1350, 4th revised and expanded edition, Leiden. Spuler and Sourdel-Thomine 1973 B. Spuler and J. Sourdel-Thomine (eds), Die Kunst des Islam, Berlin. Stern 1971 S.M. Stern, “Ya‘qub the Coppersmith and Persian National Sentiment’, in C.E. Bosworth (ed.), Iran and Islam: In Memory of the Late Vladimir Minorsky, Edinburgh, 535-55. Stewart 1967 D. Stewart, Early Islam, Weert, 1967 Treadwell 2003 L. Treadwell, “Shahanshah and al-Malik al-Mu’ayyad: The Legitimation of Power in Samanid and Buyid Iran”, in F. Daftary and J. Meri (eds), Culture and Memory in Medieval Islam: Essays in Honour of Wilferd Madelung, London and New York, 318-37. Volbach 1969 W.F. Volbach, tr. Y. Gabriel, Early Decorative Textiles, Feltham. Walker 1941 J. Walker, A Catalogue of the Arab-Sassanian Coins in the British Museum (Umaiyad Governors in the East, Arab-Ephthalites, ‘Abbasid Governors in Tabaristan and Bukhara), London. Washington 1989 T.W.Lentz and G.D. Lowry, Timur and the Princely Vision, Washington. DC. (and the Los Angeles County Museum of Art). Wensinck 1978 A.J. Wensinck, “Khamr”, in Encyclopaedia of Islam, 2nd ed. IV, 994b -997a.

Index Aachen 286 n29 Abāqā Khan 472 palace 343 Abbāsid lustreware 64, 65, 67, 83, 84 Abbāsids 144,167,180, 268, 281, 282, 283, 310, 314, 350, 354 & n20, 370,464, 470, 471, 474 ‘Abd al-Malik 134–136, 19, 156, 255–256, 267, 470 ‘Abd al-Rahman I 178, 179 & n39, 216 ‘Abd al-Rahman III 177 n23, 179, 180, 184, 228, 229, 266 ‘Abd al-Raḥmān al-Gailānī 67 n8, 344 n81, 348 n99 ‘Abd al-Rahman ibn Zayyan 263 Abrahams, S. 44 n12 Abu’l-’Abbas 123 Abu’l-Khattar 178 Abu’l-Muzaffar Mosque, Mayyafariqin 269 Abu Nuwas 471 &n17 Abu’l-Qāsim 95 Abu Zaid 27, 95, 475 Achaemenid period 265, 275, 294–5, 473 Ackerman, P. 29 n142, 43 n9, 55 n46, 105, 343 n76, 441 n48 Adamova, A. T. 62 n71, 364 Ādharbāijān 138 & n35 Adle, Chahryar 473 n31 Afghanistan 73, 267, 276, 277 Afrasiyab 356, 362 Fig. 3, 473 n38 Aght’amar 6 n21, 86

Ahmad Mustafa 115 Ahsan, Imran Khan Nyazee 469 n2, 470 n12, 476 n47 Ainy, A. E. 355 n23, 361 ‘Ajā’ib al-makhlūqāt wa gharā’ib al-maujūdāt 315 Ajanta 371 Ak Han 268 al-Andalus 143 n53, 150 Fig. 4, 155 n70, 158, 161 n98, 167, 168 n127, 179 n40, 182 n54, 438 Andalusia 58, 117, 178 n29, 179, 181, 216, 276 al-Aqsa Mosque, Jerusalem 177 n24, 216 & n125, 463 wood panel 204 Fig.10 Al-athar al-baqiya 27 n131 Alaeddin Mosque, Niğde 269 Alay Han 268 Alhambra 126, 128, 173, 266, 276, 425, 434 & n21, 440 & n43, n44, 453 n101 ‘Ali 9, 27, 47, 116, 117, 331 Fig. 11, 341 ‘Alī b. Muḥammad b. abī Ṭāhir 96 Aljaferia palace, Saragossa 189 n78, 216 n124,276 Allah 47, 53, 58, 117, 156, 157, 159, 160, 162, 164, 194 Allan, J. W.4 n14, 15 n88, n92, 38 n2, 46 n22, 54 & n42, 55, 56 n50, 306 n1 Aleppo 346 n93 Allen,T 84 n28

496 Almagro, M. 175 n10, 176 n14–18, n20, n21, 300 n57 Alp Arslan, sultan 472 salver 190 n82, 441 n48 Altdorfer, Albrecht 130 Amādiya 346 n94 Amul 57, 331 Fig. 11, 332 Fig.12, 333 Fig.13, 347 n97 Anahita 321, 370 n46 Anatolia 23, 57 & n52, 62, 65, 79, 93, 264, 267, 343 n79, 345, 346, 445 n67 Anau mosque 345 ‘anāza 166 n123 Anhalt carpet 346 Anisi, Alireza 283 n20 Antinoë textile, Lyons 353 n14, 375 n69 Āqa Muhammad Khān Qājār 40 n4 Arab-Ata mausoleum, Tim 282 n9 Arberry, A. J. 471 n16, 471 n22, 476 n54 Arjāsp 328 Aris, Michael 278 Arminya 138 Ardabīl shrine 38–62, 38 & n1, 39, 40 n4 41 n5, 44 & n11, 45 n17, 46, 48– 52 Figs, 1–5, 56 n50, n51, 58, 59 n62, 61 n69, n70, 62 & n14, 308, 345 Ardashir II 323 n33 Arel, Ayda 431 n6 Armenia 115, 269 Arne-Petersen, K. 128 Arnheim, Rudolf 462 n131, 463 n135 Arnold, Thomas W. 318 & n9, 341 n66, 352 n10 Arslan, E. 334 n46, 353 n15 Artaban II drawing from a coin 365 Fig. 8 Artan, Tülay 431 n6 Artsruni, T. 6 n21 Artuqids 267, 268 Ashkhabad 361

Ashmolean Museum, Oxford lustreware dish 1, 4, 5 n19, 6–9, 14–17, 18 Fig. 1, 21 Fig. 11, 22 Fig. 13, 24 n107, 25–27, 29 Barlow Gift 18 Fig.1 pyxis 252 Asia Society, New York 274 Asiatic Museum, St. Petersburg 335 n49 Aslanapa, O. 343 n77 Astana 300 n55 ‘Ata Allah b. Muhammad al-Tabrizi 116 Atil, E. 13 n70, 16 n96, n99, 23 n103, 193 n99, 281 n1, 282 n13, 352 n13, 431 n6, 450 n94, 466 n148 Aṭṭār 345 n85 Audeh, Abeer 463 Auld, Sylvia J. 461 n130 Azada 296, 297 Azarnouche, Samra 470 n13 Avery, P. 40 n4 Azerbayjan 40 n4, 445 n68 Babaie, Sussan 463 n133, 467 n153 Bacharach, Jere 430 n5c Baer, E. 14 n83, 28 n137, 320 n19 Baghdad 268, 311, 346 n94, 380, 426, 433 n16, 474, 475, 479 Bagherzadeh, F. 19, 25 n115, 27 & n113, 29 n142, 357 n30, 369 n43, 370 n48 Baha al-Daula 195, 198 Fig. 2b Bahmān 340 n61 Bahrām Chūbīna 329 Bahram Gur 324, 343 silk 290 Fig. 8, 296, 299 & n52 Bahrami, M. 2 n8, 3 n13, 14 n81, 15 n85, 29 n143, 282 n13, 369 n46, 372 n54 Baker, Patricia L. 175 n7, 193 n100 Bakhit, M. A. 133 n4 Baku 93, 269 Bal’amī 319 n14 Balog, P. 18 n127

INDEX

Baltimore plate 365 Balty, J. 294 n34, 300 n56, 301 n58 Bamberg, Cathedral Treasury 321 n23 Barbaro, J. 59 Barber, R. 359 n39 Barcelona 128 Barman 411 Fig. 23 Barquq, Sultan 264 Barrett, D. 339 n60 Barry, M. 477 n52 Barrucand, Marianne 215 n122 Basra 138 n35, 169 n111 Batalha 124 Bates, Michael 132 & n4, 133 n5, n7, 136 n25, n27, 137 n30, n33, 141 n44, 142 n47, 143 n53, 145 n60, n61, 155 n70, 156 n73, 159 n89, n90, 163 n105, 165 n117, 167 n126 Bates, Ülku Ü. 428 n1, 431 n6 Barks, Coleman 471 n21 Bartol’d, Vasilii V 282 n5 Bawandids 282 Bayani, Manijeh 275 Bayazid Bastami 471 Baybars, I, 268 Beach, Milo Cleveland 429 n3 Beato, Felice 130 Beattie, M. H. 340 n63 Beauharnais, Palais 127 Beckwith, John 196 n109, 212, 213, 228–233, 236–238, 241, 246, 248–253, 255–256, 258–259, 261–263, 324 n40, n43 Bedford, James 130 Bednorz, Achim 215 n122 Behrens-Abouseif, D. 175 n10,190 n82 Belenitsky, A. 323 n38, 356 n28, 362 Ben Ysuf Madrasa, Marrakesh 266 Benaki Collection, Athens 213 Bencheik, Jamel Eddine 471 n17 Berchem, Max van 135n16, 182 n53, 192 n92, 283 n16, 317 & n8, 451 & n96

497 Belem 124 Bellini, Vincenzo. 129 Berlin 128, 290, 343 n79 Berkyaruq, Sultan 7 Bernie, V. 21 Beyşehir 268 Bhutan 278 Bibliothéque Nationale, Paris al-Ḥarīrī, Maqāmāt 327 Fig. 6 Schefer Maqamat 479 Fig.2 silver plate 335 n49, 360 Fig. 1, Bier, Carol 431 n8, 462 n131 Bihzad, 278, 382 Bishapur 10 n52,166 n123, 320 n17 bismillah 137, 143 & n53, 158–160, 162 Bīsutūn 318 n10 Bizhan 408 Fig. 20 Björkman, W. 14 n80 Blair, Sheila. 2 n6, 135 n17, 181, 49, 183 n61, n62, 195 n106, n107, 281 n2, 352 n12, 449 n89, 459 & n120 Bloom, Jonathan M. 176 n23, 281 n2, 349 n2, 352 n12, 449 n89 Blunt, Wilfred Scawen 126 Bobrinski ‘bucket’ 98 & n4, 106, 115 Bode, Wilhelm von 97, 99 Bodleian Library, Oxford thesis on Persian painting 273 Bonaparte, Napoleon 130 Bone, H. 168 n128 Bonfils, Felix 130 Boor, C. de 136 n25 Borodin, A. 125 Borkopp, Birgitt 286 & n30, 288 Boston hunting carpet 109 & n25 Bosworth, C. E. 40 n4, 314 n3, 435 n25 Bothmer, Hans-Caspar Graf von 136 n21 n22, 156 n77, 177 n23 Bougie, Algeria 266 Bowman, John S. 429 n3 Boyle, J.A. 108 n22, 441 n48 Brand, Michael 449

498 Brandenburg, Dietrich 277, 345 n90 Braga pyxis 190, 192, 200 Fig. 4b, 256 Fig. Brangwyn plate 23 n106, 90 Breckenridge, J. D. 136 & n25, n27 Breeskin, A. D. 97 n3 Bréhier, Emile 290, 293 n33, 295 n40 Brend, Barbara, 27 n132, 443 n54, 461 n130, 463 n135 Brian, D. 316 n7 Brighton 127 Brisch, Klaus 178 n25, 212 n112, 276 British Library, London 335 n49, 336 n54 British Museum, London Mamluk Hariri 337 Turkish painting catalogued 273 Broadhurst, J. C. 469 n5 Bronstein, L. 57 n56, n57, 60 n67 Brosh, N. 10 n49 Brown, A. 356 n27 Broug, Eric 431 n14 Brühl 127 Brüsehoff, Kurt 278 Bruzelius, Caroline 429 n5a Buchthal, Hugo 101, 334 n48 Buddhist art 26, 27 & n130, 171, 282 mudras 11 Bujnurd bowl 385, 405 Fig. 17 Bukhara 275, 282 n9, 355, 382, 412 Bulliet, Richard 369 n44, 430 n5d Burgos casket 188, 189 & n80, n81, 190 ,196 n108, n109, 199 Fig. 3d, 261 Fig see also: Silos casket Burton, Sir Richard 126 Burrill, Kathleen R. F. 314, 315 Bussagli, M. 323 n38 Bustan of Sa‘di 472 Buyid 8, 31, 195, 281 n3, 282, 195, 308, 352 n11, n12, 369, 441 n48 Byzantine art 12, 217, 265, 286 n29, 293, 320–322, 324, 334, 353,424,

458 book painting 312 ceramics 68, 84 coinage 136, 137 & n32, 138, 139 & n37, 140, 141,143, 144, 170 emperor 136, 180, 293 mosaic 84, 317, 380 ivory 4, 5n18, 211 silks 293–296 silverware 4, 26, 66 n2 wall painting 294 Byzantium 66, 110, 137, 169, 293, 307 Cachia, Pierre 315 Ca’ d’Oro, Venice 126 Çağman, Filiz 306 n1 Caiger-Smith, A. 2 n7, 8 n40, 15 n89, 17 n101, 23 n104, 24 n109, 110, 25 n111, n112, 28 n137, 29 n144, 95, 306 n1, 310 Cairo 264 calligraphy 47, 68, 69, 71, 111–121, 378–381 Camber, Richard 186 n69 Cameron, Averil 321 n22, n24 Camman, Schuyler van R. 314, 340–344, 346, 348 Camp, du Maxine 130 Canby, S. R. 38 n1, 46 n20, n21, n22, 440 n44, 455 n106, n108, 449, 467 n152 Canard, M 87 n32 Cappella Palatina, Palermo 10 n50, 86, 477 Carande 180 n43 Cardiff 128 Carpaccio, V. 129 Carter, M. L. 323 n36, 343 n74, 370 n47, 374 n66, 433 n20, 456 n110 Carthage 125 Capps, E. 322 n30 Caspian 283 Castello Sforzesco, museum 321 n24 Castejón, Rafael 113 n117 Castriota, David 462 n134

INDEX

Cellini 307 chahar taq 283, 354 Chaldiran 54 Chalmeta, P. 180 n43 Chambers, W. 126 Chardin, Sir John 124 Charlemagne 123 Charlston, Robert 363 Chase, W. T. 281 n1, 282 n13, 352 n13 Chelkowski, P. J. 105 n16, 282 n10, 314 n1, 352 n Chester Beatty, A. 64 Chester Beatty Library, Dublin 113 Chinese influences 11, 67, 68, 69, 82, 112, 127, 298, 302, 312, 314, 340, 342–344, 348, 385, 386, 387 Christensen, A. 319 n12 Christians and Christian art 4, 10, 86, 87 n2,115, 125, 135, 136, 143, 144, 171, 172, 175 n12, 178, 180, 294, 304, 324, 341, 344 n84, 350, 435, n26, 443, 464, 470, 475 Çifte Minarelli Madrasa, Erzurum 268 Cizre 268, 269 Clermont, A. P. de 155 n72 Cleveland Museum of Art 109 n23, 115 Clinton, J 99 n9 Clive, Robert (of India) 130 Cockerell, Sidney. J. 127 Cohn-Wiener, E. 345 n89 Conrad, L. I. 349 n2 Coryate, Thomas 126 Cohen, M. 135 n16 Colls church 188 n75, textile 199 Fig 3c Cologne 128 silk 284, 285 & n26, 286 Cook, W. 103 Constable, Giles 429 n5b, 436 n34, 440 n43, 443 n58, 452 Constantinople 139 n36, 179, 295, 324 Contadini, Anna 186 n69

499 Copenhagen 128 Coptic art 83, 87 & n32, 212, 213 n116, 217 n126, 288 Fig.4, 293, 324, 334, 353 n14 Cordoba 158, 266 ivories 175 n9, n10, 176 n16–18, 179–180 & n43, 182 n52, 185, 203, 266 Great Mosque 177 & n23, n24, 178 & n26, n28, n29, 180, 184 n64 ,189, 276 Cormack, Robin 321 n22 Courtauld bag 474, 476, 478 Fig. 1 Cortijo del Alcaide 217 n127 Courteille, Pavet de 314 n3 Cramb, John 130 Creswell, K.A.C. 66 n3, n4, n5,135 n19, 168 n30, 174–175 n3, 174 n4, n5, n8, n9, n12, 177 & n22–25, 178 & n26, n28, n29–32, 179 n34–36, 195 n105, n113, 213 n116, 276, 317 n8, 344 n83, 345 n86, 346 n92, 436 n32, 438, 441 & n50, 449, 452, 458 Ctesiphon 283 Cuenca 188, 190, 191, 196 n109, 213 n118, Figs. 261, 262, 263 Curtis, John 295 n37 Cup of Solomon 5 Czechoslovakia 127 Damascus and coinage 133 & n7, 134–135, 139, 146 n67, 159 n89, 160 n91, 167, 169 n132 Great Mosque 168,177 n23, 24, 178 n25, n28, 180, 184 n64 mosaics 317 maqamat 470, 476, 481 Fig.4 Dāmghān, mosque 96, 354, 477 Daneshvari, A. 350 n5, 369 n42 Daniell,T. 127 Daniell, W. 127 Darius 294 Darley-Doran, R. 138 n35, 144 n55 Dauphin, Claudine M. 175 n12

500 David collection, Copenhagen 57 ivory casket 238 Fig. ivory plaque 210 Fig 19, 211, plates 4 n16, 26 Dawlatshah 41 n7 Dawson, Christopher 473 n32, n33, n36 Debenham Collections 1, 7, 10–13, 15–16, 20 Fig.5, 28 Deccan 46 Delacroix, E. 129 Delibaş, Selma 306 n1 Demotte Shahnama 14, 316 n7, 339 n60, 340 n61, 423, 425, 449, 459, 460 Denike 57 & n58, 62 n72 Denny, Walter 449 n87 Derenk, D. 470 n9 DeShazo, A. S. 133 n5 Dhamar 265 Diba, Pouran 295 n38 Dibaj, I. 57 n54, 62 n74 Diez album 76 Diez, E. 182 n56, 449, 458 n114, 476 Dickson, M. B. 62 n73 Dimand, M. S. 57 n57, 458 n114 dīnār, reformed coinage 132–170, 147–154 Figs 1–17 see also: Umayyads Dioscorides Bodleian ms. 7 n27 Istanbul ms. 13 dirham 84 n28, 132, 133, 134 n8, 137 n33, 138 n35,139 n36, 140 & n39, 144 n56, 145 & n61, 146 n66, n67,157 n80, 165, 166, 168 & n127, n127, 169 & n131 Diyarbakir, Anatolia 80, 268 Diyar Rab’ia 471 Dodd, E.C. 53 n35, 454 n104 Dodds, J. D. 58 n60 Doha 211 & n111, 212, 253 Fig, 429 n4 Dome of the Rock, Jerusalem 84, 135 & n17, 138, 139 n37, 143,

156, 168, 178 n27, 180, 181, 185, 192–195, 204 Fig. 10, 350 n4, 424, 438, 439 n39, 441, 442, 453 n103, 463, 465 inscription 197 Fig.1e, 200 Fig.4c, 4d Döner Kümbed, Kayseri 268 Donovan, D. 11 n62 Doughty, C. M. 26 Dresden 128 Dura Europos 323 n35 Durri al-Saghir 230, 231 Egypt 53 n37, 58, 64, 71, 73, 82, 85, 87, 89, 105, 123, 126, 127, 10, 68 n129, 183 n57,193 n96, 212, 265, 268, 276, 281, 293, 334, 353 n14, 475 Ellis, C. G. 55 n 46 Emamy, M. 57 n54, 62 n74 Emmons, J. 322 n27 Epistles of the Sincere Brethren 17 n78 epigraphy in ivories 174–263 Erdmann, H. 7 n32, 340 n63 Erdmann, K. 13 n73, 98, 284 n24, 322 n25, n26, 323 n34, 340 n63, 342 n74, 343 n77, n79, 351 n6, 369 n46, 449 Erzurum, Anatolia 268 Esin, E. 10 n34, 13 n71, 16 n94, 17 n100, 358 n35, 431 n6, 450 n94, 466 n148, 470 n15 Ettinghausen, Richard 2 n4 n5, n6, 4 n17, 5 n18, 6 n25, 7 n31, 9 n44, n48, 10 n50, n51, 11 n58, n59, n60, 13 n74, n75, n77, 14 n78, n79,15 n93, 23 n103, 25 n121, 28 n137, n139, n140, n141, 29 n142, 67 n6, 79 n21, 84 n29, 97 n1–3, 97–110, 98 n4, 99 n7, 102 Fig.1, 105 n13, n14, n16, n17, 106 n20, 108 n21, n22, 109 n23–25,110 n27, 180 n44, 184 n63, 285 n26, 295 & n39, 296 n42, 299 n52, 313, 314, 316 & n6, 327, 335

INDEX

n52, 337 n56, 341 n65, n69, 349 n1, 350 n5, 355 n24, 356 n25, 359 n37, 372 n35, 373 n61, n63, n64, 374 n68, 431 n7, n8, 435 n25, 436 n32, 438, 446 & n71, n74, 447 n76, 462 n134, 465 n144, 476 n48, n49, 477 n53 Ettinghausen, Stephen. E. 97 n3 Ewert, Christian 170, 178 n26, 215 n121, 216 n124, 217 n127, 276 Eumorphopoulos, George 64 Fabriano, G. da 129 Fahraj, mosque 255, 354 Fajans, Salomea 300 n57 Falke, Otto von, 155 n71, 157 n79, 162 n101, 284 n25, 290, 293 n33, 294 n36, 295 n40, 298 n48, 353 n16, 451 n99 Farès, B. 11 n60, 12 n63, 13 n70, n76, 15 n93, 17 n100, 370 n48 Farhad 283, 314, 315, 317, 318, 352 & n9, 407 Fig. 19 Farhad, Massumeh 430 & n5f Faridun 14 Fatimid drawing 10 n49 capital cities 266 Egypt 475 lustre ware 63, 64, 71, 85–89 Mahdia 266 painting 98 & n4, 106 Sabra al-Mansuriyya 266 unglazed pottery 84 n28, 86 Fatimids 85, 87 n32, 88, 281, 435 n26, 463 Fehérvári, Géza 79 n21, 274, 354 & n19, 357 n31 Fernández-Puertas, A. 181 n48 Ferrandis, J. 181 n48, 228–233, 236–238, 241, 246, 248–253, 255–259, 261–263 Finster, Barbara 215 n122 Firdausi 14, 297, 318, 328 Firuzabad, Fars 283 al-Firuzabadi 469 n1

501 Fitero casket 186, 198 Fig. 2a, 213, 233 Fig. Fitzherbert, Teresa F. 24 n108, 355 n24, 369 n44, 460 n 125 Flaubert, G. 125 Flood, Finbar (Barry) 444 n65, 453 n103, 467 n156 Florence 190–192, 200 Fig. 4a, 252 Flury, S 182 n56, 183 n58–61, 359 n38 Folsach, Kjeld von 41 n7, 42 n8, 56 n49, 57 n53, n55 Foroughi Collection, Iran Bastan Museum, Tehran 357 n31 Foster, Elisa 429 n5a Fox, M. 448 n84 Frankfurt 128 Fraser, J. B. 45 n17 Freer Gallery of Art, Washington, DC 23 n103, 81, 88, 95, 98, 106 n20, 108, 352 n13, 432, 447, 465 n143 Freer medal 429 n3, n4 Freer plate 23 n103 Frishman, Martin 442 n55 Frith, Francis 130 Frye, R. N. 40 n4, 430 n5b, 441 n48 Fukai, 284 n23 Fulton, A. S. 469 n1 Gabriel, Albert 438 Gabriel, Y 334 n45, 449 Gagik, King of Vaspurakan 6 n21 Gall, H. von 12 n67, 284 n24, 351 n7 Galland, Antoine 124 Garcia Sánchez, E. 180 n44 Gardet, L. 342 n73 Garrus 356 & n27 Geneva 157 Gengis Khan 472 George, Alain 380 Germany 103, 125, 127, 294, 452 Gerona casket 186, 205 Fig. 11 Treasury of the Cathedral 205

502 Gerresheim, monastery 294 Gervase de Basoches 473 n36 Ghadanfar Abu Taghlib ibn Hasan ibn ‘Abdallah 179 Ghazan Khan, tomb of 268 Ghazna/Ghazni, Afghanistan 73, 75 Fig. 4, 79, 267 Ghaznavids 6 n21, 267 Ghirshman, R. 16 n97, 285 n26, 287, 291, 295 n37, 298 n47, 299 n51, n53, 300 n56, 322 n27–29, 323 n35, 334 n44, 46, 353 n14, n15, 358 n35, 370 n47, 371 n50, 372 n56, 373 n65, 374 n66, 375 n69 Ghouchani, Abdullah 473 n31, n35 Ghubayrā 73 Ghulam, Y. 341 n71 Ghurar akhbār mulūk al-Furs 319 n12 Ghūrid period 73, 276, 277 Gibb, Sir Hamilton A. R. 136 & n23, n25, 456 & n111 Giffen, Lois A. 315, 316 Gilbert, S. 322 n27 Gilmour, B. 56 n50 Giuzal’ian, L. T. 364 Glick, T. F. 180 n44 Gluck, Christoph W. composer 125 Gluck, J. 41 n7 Gluck, S. H. 4 n7, 125 Godard, André 283 n18 Goethe, Johann W. von 125 Goeje, M. J. de 324 n39 Goitein, S. D. 97 n3, 106 Goitia, F. C. 178 n 26, 215 n120 Gök Madrasa, Sivas 269 Goldziher, I. 359 n39 Golombek, Lisa 337 n56, 339 n60, 345 n91, 429 n5b, 450 n91, n94, 467 n153, n155 Goloubew, Victor 64 Gombrich, Sir Ernst. 101, 130, 430 n5e, 462 & n132 Gomez-Moreno, M. 178 n26, 188 n76

Gonzales, Valerie 466 n149 Grabar, Alexis 428 n2 Grabar, André 12 n68, 321 n22, 322 n32, 323 n37, 334 n46, 353 n15, 443 n63, 445, 447 n78, 465 & n141 Grabar, Nicolas 429 n5b, 436 n33 Grabar, Oleg 2 n5, n6, 9 n48, 28 n140, 97 n1, n3, 99 n9, 106 n19, 108 n22, 109 & n23, 135 n18, 136 n21, 143 n52, 156 n76, 178 n27, 181 n51, 184, n63 n64, 192 n91 n93, 194 n103, 195 n104, 284 n22, n24, 306 n1, 312, 313, 319 n19, 323 n36, 334, 335 n50, 336, 337 & n57, 338, 343 n76, 349 n1, 350 n4, 351 n6, 355 n22, 372 n53, 420–427, 421 Fig.1, 428–467, 428 n1, 429 n3, n4, n5a, 430 n5b–5f, 471 n6, n11, 433 n6,n17, n19, n20, 434 n21, n22, n23, 435 n25, n29,436 n30, n31, 437 n37, 439 n39, n41, 440 n42–44, n46, 441 n47, n49, 442 n51–52, n54–55, 443 n59–60, n64, 445 n66–68, 446 n71, n73, 447 n75–78, n81, 448 n92, 449 n88, n89, 450 n92, 453 n100–101, n103, 454 n104, 455 n106–107, 456 n110–111, 459 n120–121, 460 n124, 461 n128, 462 n131, n134, 464 n138, 465 n144, 466 n147, 467 n151–152, n158 Grabar, Terry 426, 428 n2, 434 n24, 450 n93 Granada 79, 173,180, 266, 469 Graves, Margaret 174 n1 Gray, Basil 62 n71, 275, 339 n60, 341 n68, 373 n59, 438 Greek 113, 134, 139 140 &n40, 141, 143, 475 Grierson, P. 133 n5 Grimes, William 428 n1 Grohmann, A. 127, 181 n50 Grube, Ernst. J. 7 n27, 25 n113, 109

INDEX

n26, 343 n78, 355 n24, 357 n31, 358 n32, 369 n44, 431 n7, n9 discussion of the Keir Collection 63–89 & n24, n30, n31, n32, 190 n82 Gruliow, Basilia 105 n17 Guest, G. D. 25 n121, 28 n137, 29 n142, 108 n21, 315 n4 Gulistān Palace Library, Teheran 328 Gunbad-i Qabus 84 n26, 182 n56, 282 n9, 354 n18 Gunter, A.C. 352 n13, 370 n47 Gūr-i Amir 43 n9, 59 Gurgan ware 3 n13, 93, 96, 281 n15 Gurgani, Fakhr ud-Din 283 n15 Hafez/Hafiz 118, 119, 121, 476 Haft Paikar 297 Hagedorn, A. 8 n38 Haidar 116 al-Hājjāj 135, 142 n66, 166 al-Hakam II 180, 186, 192, 230, 231 Haldane, J. D. 336 n54, 337 n57 Ḥamā 180, 346 n93 Hambly, G. 40 n4 Hamdanid dynasty 179 Hamilton, Robert W. 175, n6, n8, n10, n11, 176 n12, 177 n24, n25, 178 n26, 344 n82, 436 n32, 470 n8, n9, 471 n19, 472 n24 Hanaway, William 315 al-Hariri 11 & n60, 13 n74, 327, 334 & n48, 337 n57, 460, 475, 476 Harper, Prudence O. 284 n24, 285 n27, 297 n44, 299 n50, 351 n6, n7, 358 n33, 369 n40, 371 n52, 374 n66, 430 n5d, 457 n112 Hartmann, A. 25 n116 Hartner, W. 175 n10 Harun al-Rashid 123 Harran 177 n23 Hasan 117 Hasan, Perween 467 n157, n158 Hatim al-Ta’i 382, 412 Fig.24 Havermeyer, H. see: Metropolitan

503 Museum of Art, New York Hawting, G. R. 134 n10, n12, 136 n26, 161 n95 Hayes, J. R. 2 n4 Hayashi, Ryoichi 285 n27, 292, 298 n48, n49 Heidelberg 127 Helmand 277 Herat 62, 115, 116, 382, 472 Hermitage Museum, St. Petersburg 115, 369 silver plates 367 Fig 10, 368 Fig. 11, Fig. 12, 371, 372 see also: Strelka Herrmann, G. 10 n52, 16 n94, 299 n54 Herzfeld, E. 69 n13, 98 & n5, 103, 177 n23, 282 n6,7,8,10, 283 n18, 319 n13, 320 n17, 321 n21, 323 n35, 346 n93, 349 n2, 370, n371 n49, n51, 438 & n38, 449, 451 n98, 458 & n117, 462 n134 Herzog H. 340 n63 Hijazi script 380, 389 Fig. 1 Hillenbrand, C. 16 n98, 25 n117, 470 n11, 472 n26, n27, n28, 473 n36 Hillenbrand, F. K.M. 349 n2 Hillenbrand, R. 7 n30, 14 n81, n83, 15 n92, 58 n59, 99 n6, 176 n20, 182 n54, 184 n64, 216 n125, 217 n126, 283 n20, 297 n43, 349 n2, 350 n5, 352 n10, 354 n20, 369 n42, 428 n1, 446 n71, 458 n115, 470 n8, n9 Hims 179 Hinrichs, J. C. 451 n97 Hisham II 265, 470 & n8 casket 205 Fig.11, 206 Fig.12 Veil of 187, 199 Fig. 3a Hispanic Society of America 183, 213, 214, 249 Hitti, P. K. 134 n10, n13, 14, 135 n20, 179 n39, 433 n16, 447 & n77 Hodges, William 127

504 Hodgson, M. G. S. 135 n15 Hogarth, J. 323 n38, 369 n46 Hoffman, Eva 431 n10 Holman Hunt, William 129 Holod, Renata 189 n81, 429 n5b, 430 n5d, n5f, 433 n19, 443 n57, 445 n66, 459 n121 Honarfar, L. 41 n6, n7 Horiyushi temple 298 Huesca 188 n75 Hülegü, Khan 93, 469 Humāyūn 44, 45, 46 Humphreys, Stephen 430 n5d Husain 117 Husam al-Dawla 190 Ibn Abi ‘Amir see al-Mansur Ibn al-Bawwab 113, 114, 443 n61 Ibn al-Faqīh, 324 n39 Ibn al-Farid 471 Ibn Bakhtīshū 316 Ibn Battuta 100 Ibn Hamdis 266 Ibn Jubayr 25, 469 n5 IbnKhaldun 194 n102 Ibn Muqlah 113 Ibn Qutaiba 296 Idries Shah 317 Ierusalimskaja, Anna A. 286 &n30, 288 Ifrīqiya 143 n53, 155 n70, 158, 167 Il-Ghazi 472 Ilkhanid book painting 39 ceramics 92 ivory plaque 402 Fig.14 lajvardina bottle 399 Fig 11 metalwork 383 script 69 tiles 73, 82 tomb 268 Ilisch, L. 138 n35 Imāmzāda ‘Abbās 331, 332, 333, 347 n97 Imāmzāda Panj Tan, Lāhijān 347 n97

India Office Library, London Persian painting catalogue 273 Instituto de Valencia de Don Juan, Madrid 236 Ipşiroğlu, M. S. 8 n33, n34, 15 n93, 16 Incir Han 269 India and its influence 44 n11, 45 & n17, 46, 99, 100, 116, 118, 123, 124, 127,130, 264, 267, 273, 274, 302, 307, 371, 383, 438, 448 & n84, 472 Iran Bastan Museum, Tehran buff ware bowl 363 Fig. 5, Fig.6 lustreware dish 1, 9, 11, 12n 69, 17, 19 Fig.3, 24 n107, 25, 27, 29, 30, 357 & n31 sarcophagus 53 & n36 wall painting 362 Fig.4 Iraqi art 85, 313, 344 n81, 438, 461 the poet 9 n47 Isfahan 79, 284, 308, 345, 413, 417 , 418 , 425, 440, 467 n153 al-Isfahani 474 Isfandiyār 316, 328 Iskandar 334 n47 Isma‘il ibn Yahya al-Ma’mun 190, 262, 263 Ismā‘il, Shah sarcophagus of 38–62, 48–52, Figs. 1–5 aesthetics of 38, 54–58 calligraphy 47 general description 41–44 religious inscriptions 46–47, 53 woodwork 38, 58–62 Istanbul Albums 476 Ivories caskets 197 Figs.1a, 1b, 198 Figs. 2a, 2d–g, 199 Figs. 3d, 3e, 200 Fig. 4a, 205 Fig. 11, 206 Figs. 12, 13, 207 Figs. 14–16, Figs. 228, 229, 232, 233, 236–238, 252, 253, 255, 257, 261, 263 plaque 402 Fig. 14

INDEX

pyxis 197 Fig. 1c, 1d, 199 Fig. 3b, 200 Fig. 4b, 201 Fig. 6, Figs. 230, 231, 241, 246, 248–252, 256, 258, 262, 297 n45 Syrian/Spanish tradition174–263 Iznik ware 63, 82, 442 Jahangir 385 al-Jahiz, 469 n3, 472 Jakeman, Jane 435 n27, 446 & n73, 462 n131, 462 n135 Jām 73, 82, 277 Jāmi’ al-Tawārīkh 339 n60 James, D 43 n9 Jamshid 476 al-Janābī, T. A. 346 n93 Japanese silks 291 Figs. 9, 10, 292 Fig. 11, 298 Játiva 267 Jausaq al-Khaqānī 321 n21 Jayy 138 al-Jazīra 133 n7, 264 Jenkins, M. 97 n3, 211 n110 Jericho, Umayyad palace 344 Jerusalem 126, 139, 177 n24, 178, 197, 200, 204 Dome of the Chain 463 Haram 463 see also: Dome of the Rock Jesus 9, 143 Jett, P. 281 n1, 282 n13, 352 n13 Jiménez, O. 181 n48 John Rylands Library, Manchester Persian manuscripts catalogued 273 Johns, J. 135 n17 Jones, D. 10 n50, 53 n36, 67 n9, 340 n61 Jones, Owen 128 Jones, Sir William 130 Jordan 27, 179, 208, 300 n57 Justinian II 136 & n28, 321 n22 Juvain bowl 405 Fig.17 al-Juzjani 6 n21 Ka‘ba 341

505 Kafesçioglu, Çigdem 431 n6 Kakuyids 282 Kaikhusrau II 84 n28 Kain, Evelyn 462 n131 Kaiser-Friedrich Musem, see Berlin Kalila wa Dimna, Topkapi Saray Library, Istanbul 373 n60 Kapkian, Catherine 462 n131 Karatay Han 269 Karens 374 Karic, Ernes 469 Karl, Theodor, Elector 127 Kasbah of the Udayas, Rabat 266, 346 n94 Kashan lustreware 1–29, 64, 91, 108 n21, 379, 395–397 Figs. 7–9, 400 Fig 12 Kayka’us, Sivas 269 Kayseri, Anatolia 268 Kedar, Benjamin Z 430 n 5d, 442 n52 Keir Collection 63–89 Keall, Ed. 431 n11, 467 n151 Keene, H. 126 Kennedy H. 145 n61, 168, 169 n131, 179 n42 Kennedy, Philip H. 471 n17 Kensington Palace gardens, London 128 Kerner, Jaclynne 451 n95, 467 n158 Kew, Richmond 126 Khachen Dorbatly mausoleum, Nagorno-Karabagh 268 Khairallah, 53 n35 Khalaf 188, 199 Fig.3b, 233, 249 Khalili Collection, London 274, 357 n31 al-Khamis, U. 9 n41 khamriyyat poetry 470, 471 Khamseh 39, 62 n71 Khan, Hasan-Uddin 442 n55 al-Khān, near Mosul 346 n93 khatamkari design 41 & n54 Khawarnāma 341 Khayr 255 Khirbat al-Mafjar 28 n139, 66, 67,

506 108 n22, 174 n2, n3, 175 n5, n10, n11, 178 n26, 190 n82, 211, 265, 313, 341 n65, 344 & n82, 372 n55, 472 Khoubnazar, H. 345 n89 Khuday Namak 296 Khulbuk, Tajikistan 267 Khusrau 283, 317, 318 & n10, 319, 322, 323, 329 Fig. 8, 351, 352 Khurasan 62,68, 96, 186 n70, 281, 282, 445 & n68 Khwājū Kirmānī 334 n47 Khwarizmshāhs 7, 73, 277 Kiani, M.Y. 93 Kiev 322 n32 kitabkhāna 55 Kitab al-Aghani frontispieces 4 n17, 7 n26, 9 n44, 10 n51, 11 n58, 13 n76, 17 Kitab al-Diryaq 480 Fig. 3 Kitab al-Kharaj 168 n129 Kitāb Mu’jam al-buldān 324 n39 King, D. A. 178 n.29 King, N. Q. 26 n124 Kinglake, A. 126 Kirman dish 415 Fig. 27 Kitzinger, E. 103, 324 n42, 341 n64 Kleiss, W. 318 n10, 334 n47, 345 n89 Klimburg-Salter, Deborah 430 n5f Klimova plate 365 Klinkott, Manfred 277 Koch, Ebba 306 n1, 313 Kofler-Truniger Collection 250 Kokand 57 Konya 264, 268 Köseoğlu, Cengiz 38, 54, 55 n43, 56 n47, 306 n1 Kouymjian, D. K. 359 n37 Knights of St. John 125 Krafft, H. 127 Krautheimer, R. 103 Kröger, Jens 276 Kubadābād 80, 268 Kūfa 65, 138 & n35 Kufic script 47, 67, 69, 71, 117, 181

n48, n50, 309, 379, 381, 385 393 Fig.5 Kühnel, Ernst 97–99, 187 n73, 188 n76, n77, 189 n79, 196 n108, 228–233, 236–238, 241, 246, 248–253, 255–259, 261–263, 281 n3, 342 n72, 352 n11, 449, 451 n99, 458 n114, 462 n134 Kūhpāya 83 Kuhrt, A. 350 n3 Kulliyāt 334 n47 Kushan coins 365 Fig. 8 Lahijan 408 Lahore 46 lajvardina ware 82 &n24, 385, 386, 399 Fig.11 Lane, A. 28 n141, 70 n14 Lane Poole, S. E. 134 n8, 142 n48, 146 n67 Lapidus, I. M. 9 n43 laqabi albarello 385, 394 Fig. 6 dish 8 n38, 385 Lashkar-i Bāzār 8 n35, 81, 277 Lassner, Jacob 430 n5c, n5f, 450 n90, n92, 466 n146, n147 László, G. 369 n41 Latā’if al-ma‘ārif 314 n3 Law, Amity 429 n5a Lawrence, T. E. 126 Lechler, G. 320 n19 Leeds 128 Lentz, T. W. 9 n42, 29 n145, 59 n62, n63, n65, n66, 60 n68, 373 n.60, 430 n5b, 432 n12, 433 n19, 437 n37, 449, 455 n107, 464 n139, 465 n143 Lerner, Judith A. 430 n5f, 468 n161 Lessing, Gotthold Ephraim 124, 125 Levant 129, 143 n53, 175 n12, 212, 450 Levine, Neil 429 n5b, 446 n72, 453 n102 Lévi-Provençal, E. 178 n33, 179 n35, n40, n43, 180 n46, n47, 182

INDEX

n52, 186–187 n70 Linderhof 128 Livshits, V.A. 356 n28 Longo, Ruggero 429 n5a L’Orange, H. P. 28 n139 Lowry, G. D. 9 n42, 10 n49, 14 n82, 29 n145, 59 n62, n63, n65, n66, 60 n68, 373 n60, 435 n28, 449, 455 n108, 466 n145 Lukonin, V. G. 369 n46 lustreware 308–311 see also Kashan Lydda, Israel 268 Ma’arrat al-Nu’man, Great Mosque184 n64 Maastricht medallion 289 Fig. 6, 294 n34 McAuliffe, Jane D. 469 n3, n4 Madelung, Wilferd 282 n11 Ma’din Amīr al-Mu’minīn bi’l-Ḥijāz 150, 155 n70, 158, 159 n90, 160 Madinat al-Zahra 178 n26, 180, 189 n78, 215–217, 233, 236, 266, 267 marble panel 202 Fig. 8, 203 Fig. 9 Madrasa-yi Khān 345 n89 Madrid 190 Maghrib 73, 88, 438, 450, 451 Maguire, Marcia E. 316, 317, 318 Mahboubian Collection, London 357 n31, 358 n32 al-Mahdi 471 Mahdia, Tunisia 266 Mahdavi, Shireen. 132 n1 Mahmud of Ghazna 264 Makhlad, Hasan b. 471 Malaga 189 n78, 276 Malamud, B. 103 & n11 Maldonado, B. Pavón 186 n68 Malikshah, Sultan 476 Malta 125 Mamluks 15 n92, 43 n9, 64, 129, 268 painting 275, 336 n54, 337, 346 n95 metalwork 73 pottery 64, 72

507 al-Ma’mun, Isma‘il ibn Yahya 177 n23, 190, 162, 163 Manāfi’ al-Hayawān 316 Manchester 128 Mandeville, Sir John de 123 Mangutay 44 Mansour Gallery, London 274, 275 al-Mansur 177 n24, 266, 267 Manṭiq al-Ṭair 345 n85 Manuchihri Damghani 476 Maqamat 11 n59, n60, n61, 13 n74, 14 n79, 306 n1, 312, 315, 318, 327, 334 & n48, 339, 335 &n52, 336 n54, 337 n55, n57, 370 n48, 425, 447, 459–461, 461 n128, 475–476, 479 Fig 2, 481 Fig. 4 al-Maqqari 266 Maqṣūd ‘Alī 46, 62 & n74 Maragha 53 n35,93 Marҫais, G. 438, 449, 458 n114 Marco Polo 123 Marcus Aurelius 321 n22 Marks, John H. 97 Marks, Patricia 447 n77 Marrakesh 266 Marshak, Boris 8 n38, 352 n13, 356 n28, 369 n46 Mashhad 3, 80, 95, 96, 445 n68 Mashhad-i Misriyan 73 Mashriq 187 n70 Masjid-i Jāmi‘ 347 n97 Masjid-i Shāh, Isfahan 345 al-Mas’ūdī 314 n3, 470 n10, 471 n16, 474 Maulana Mahmud Chapnivis 116 Maulana Jalal al-Din Rumi 117 Mayyafariqin castle 268 mosque 269 Mazandaran 374 n66 Mecca 125, 264, 335, 355 Meinecke, Michael 276, 436 n32 Meister, Michael 300 n55 Melikian-Chirvani, A.S. 2 n4, 6 n22, 10 n54, 13 n70, 15 n87, n91, 23 n105, 25 n118, n120, 27 n130, 29

508 n146, 93, 186 n70, 273, 282 n7, 319, 370 n47, 473 n37–42 Melville, C 40 n4 Metropolitan Museum of Art, New York 57, 58, 79 n21, 83, 224, 250. 252, 259, 337 n56, 448 Anhalt carpet 346 Havemeyer bowl 1, 4, 7, 9, 10 n56, 12 n69, 13, 14, 16 n95,18 Fig. 2, 20 Fig.6, 21 Fig 7, 22 Fig. 12, 29, 32 pyxis 250, 251 Sanguszko carpet 330 Fig. 9 star tile 28 n140 Meredith-Owens, G. M. 344 n84 Merida 178 n28 Merv 278, 355, 361 vase 361 Fig. 2 Meyers, Eric M. 436 n32 Meyers, P. 351 n6 Meynard, C. Barbier de 314 n3 Mez, Adam 470 n10, n12 Michell, G. 10 n50, 53 n36, 67 n9, 340 n61 Mielich, Alphonse L. 176 n15 Migeon, Gaston 438 mihrab 70 & n21, 83, 92, 93, 95, 96, 117, 166 n23, 178 n26, 340, 346, 355, 454 n104 Mihragan 283 Milan 284 diptych 321 n24 silk 284, 285 & n26, 287, 288 Fig. 3, 293, 294, 296, 297, 299, 301, milaresia 139 & n36 Miles, George C. 133 n7, 136 n28, 139 n36, 144 n58, 146 n67, 158 n85, n88, 159 n90, 160 n92, n93, n94, 161 n96 n97, 166 &n123, 183 n57, 359 n37 Milstein, R. 10 n49 Minasian, P. D. 97 n3 mina’i ware 14 n81, 27, 71, 79, 81, 91, 96, 308, 310, 397 Fig.9, 467 minbar design 41, 83 Minorsky, Vladimir F. 282 n5

Mi‘rājnāma 341 n67 Mīr Qavvām al-Dīn al Mar’ashī 57 misbah 255 Mochiri, 140 n39 Mongols 8, 11, 15, 27 n130, 73, 79, 92, 95, 269, 277, 316, 379, 381,384, 424, 460, 469, 472, 473, 476 see also: Demotte Shahnama Montagu, Lady Mary 124 Montebello, Philippe de 97 n3 Montesquieu 124 Moore, Kathryn 429 n5a Moorish 127, 128, 129, 171 Spain 99, 276 Morier, J. 45 n17, 57 n54 Morony, M. G. 140 n38 Morton, A.H. 44 n116, 45 n17, 61 n69, n70 Mosque-Madrassa of Barquq, Cairo 264 Mosque of the Bull, Qazvin 264 Mostofi, Khosrow, 317 Mosul 346 n94, 401, 476, 480 Mozac silk 289 Fig.5, 293, 294 Mozart, Wolfgang Amadeus 125 Mshattā 66, 174–175 n3, 175 n4, n8, n9, n12, 178 n26, 196, 211, 265, 266, 346 n93, 349 n2, 350 n5, 458 & n117 facade 208 Fig 17 Mughals 44, 46, 54, 63, 127, 313, 341 n68, 381, 383, 442 & n53, 472 al-Mughira pyxis 184, 186, 197 Fig.1c, 1d, 241 Fig. Muhammad 58, 158 &n85, 189 190 Muhammad/Mohammad al-Asad 459 n.121, 463 Muhammad, the Prophet 47, 116, 117, 142, 143, 304, 341, 459 n121 Muhammad ‘Ali ibn Muhammad Zaman 419

509

INDEX

Muhammad b. Qala’un 15 n92 Muhammad ibn Zayyan 261 muhaqqaq script 381, 401 Fig. 13 Muharram 58 bowls 25 n115 plate 27 Mu’in Musavvir 417 Muir, Sir William 130 Mulder, Stephennie 435 n28, 468 n160 Müntz, J.H. 126 Munyat al-Rusafa 179 Muqaddimah 194 n102 Musée des Arts Décoratifs, Paris 237 Musée du Louvre, Paris 129 Barberini diptych 321 n24 casket 194 pyxis 184, 193–4, 241 Fig., 248 Fig. Museo Arqueológico Provincial de Córdoba marble panel 203 Fig. 9 Museo Arqueológico Nacional, Madrid carved tank 266 silver box 198 Fig. 2c zamorra pyxis 230 Fig. Museo de Burgos, Burgos 261 Museo Episcopal y Capitular de Arqueológico Sagrada, Huesca see: Colls Church Museo de Navarra, Pamplona 255 Museo Municipal, Játiva 267 Museo Nazionale del Bargello, Florence 252 Museum für angewandte Kunst, Vienna 54 n42 Museum für Islamische Kunst, Berlin box 209 Fig, 18, 211, 212 façade 208 Fig 17 figural sculpture 266 panel 202 Fig. 7, 211, 212 Museum of Fine Arts, Boston 109 Museum of Islamic Art, Doha, Qatar 211 n111, 253 Musil, A. 175 n10, 176 n15, n21

al-Mustakfi 470 n10, 471 al-Mustansir 25 Nabataean Aramaic 380 nabidh 469 & n3 Nagyszentmiklós 369 n41 Na’in, mosque 354 Najaf 96 Nara 29 Narbonne 188 Treasury of the Cathedral of SaintJust and Saint-Pasteur 262 Nagorno-Karabagh 268 Naqsh-i Rustam 322 Nash, J. 127 al-Nasir 25, 268 naskh script 381, 403 Fig. 15 Nasr, Seyyed Hossein 317 Nasrids 266 nasta’liq 118–121 Nastihan 408 Fig.20 Natanson, J. 4 n15, 5 n18, 26 n 125 Naṭanz 93 Natif, Mika 459 n122 Nationalbiblothek, Vienna 480, 481 Naumann, R, and E. 3 n9, 343 n77 Necipoğlu, Gülru 428 n1, 429 n5b, 431 n6, 455 n105, 465 n144, 466 n149, 468 n160 Nees, Larry 430 n5d, 439 n39, n40, 441 n47, 466 n147 Nemazee, S.10 n49, 14 n82 Nerval, Gerard de 125 Newman-Perper, E. 343 n80 Niğde 269 Niriz, mosque 354 Nishapur pottery 24, 67, 69, 73, 74 Fig.2, 79, 274, 281, 355, 356, 357 & n31, 358, 362 Fig. 4, 363 Figs. 5, 6, 374, 384 Nizam al-Mulk 472 Nizami 62 n71, 125, 297, 318, 329, 352 Noah’s Ark 117 Nöldeke, T. 135 n20 Noorani, Yaseen 47 n17

510 Northedge, A. 350 n3 Northover, Peter 186 n69 Noruz 283 Nür al-Dïn, mosque 346 n93, 455 n109 Nurbakhsh, J. 9 n45, n46, n47, 10 n54, 471 n20 Nuseibeh, Said 156 n76, 178 n27, 181 n51,192 n91, n93, 194 n103, 195 n104, 463 Oddy, Andrew 296 n41 Oikonomides, N. 140 n40 O’Kane, Bernard 330, 434 n22 Olearius, A. 127 Olin, Margaret 430 n5e, 462 n131, 463 n135 Olmstead, A.T. 336 n53 Önder, M. 10 n49 Öney, G. 3 n10, 320 n19, 346 n92 Otto-Dorn, K. 8 n38, 12 n64, 15 n93, 28 n136, 321 n21, 354 n17, 356 n29, 358 n35 Ottomans 40 n4, 54, 55, 56 n48, n51, 63, 82, 116, 124, 125, 129, 311, 313, 318, 341, 423 Orientalism 122–131 Orodes II drawing from a coin 365 Fig. 8 Pächt, O. 101 Pahlavi 55 n46, 134, 140& n39, 156, 158 & n82, 283, 441 Paikand 169 Palencia Casket 188, 190, 199 Fig 3e, 215, 263 Fig Palestine 179 Palgrave, William G.126 Pamplona Casket 181 n48, 188 n74, 189 n81, 190–196, 197 Figs.1a,1b, 255 Fig Pancaroğlu, Oya 431 n6 Pandjikent 323 Panofsky, E. 103 & n10, 105 n15 Pantheon, Rome 344 Paris 128

Parthians 282 & n6, 370, 371 &n52, 473 coins 365 Fig. 8 Parthian shot 285, 286, 298, 353 n14, 375 n69 Peacock, Jila 111, 115, 118, 119, 121 Pellat, Charles 469 n3 Persepolis 299, 318, 336 n53 Persian lustreware 90–96 painting 9 n42, 13 n71, 92, 105, 113, 173, 273, 340, 385, 434 n22, 460 Pevsner, Nikolaus 101, 435 n25 Phraates III, drawing from a coin 365 Fig. 8 Pickthall, M 47 n28, 53 n29, n30, n31 n33 Pierce, H. 284 n25 Pierpont Morgan Library, New York 316 Pigage, N. de 127 Pila of Játiva 267 Pinder-Wilson, Ralph 306 n1, 312, 460 n124 Pinner, R. 57 n52 Pippin the Short 293 Pīr-i Bakrān 82, 93 Pisa, Griffin of 186 Poland 127 Poole, R. 146 n67 Pope, A. U. 1 n1, 2, 3 n13, 5 n20, 6 n 23, n24, 7 n28, n29, 8 n35, n36, n38, 14 n81, 15 n88, 19, 21, 22, 23 n102, 25 n115, 29 n142, 43 n9, 55 n46, 105 & n17, 343 n76, 441 n48 Porden, William. 127 Portugal 123, 124 Potsdam 128 Prado-Vilar, F. 184 n65 Prague silk 284, 285 n26 Puccini, G. 125 Puente de Montañana, Huesca 188 n75 Pugachenkova, G. A. 355 n23

INDEX

Pur-i Vahman plate 369 n40 Pushkin, Alexander 125 pyxis see: ivories Qadi Ahmad 116 al-Qahir 470 n10 Qairawān, Tunisia 158, 266 Qajar period 40 n4, 80, 82, 116, 341, 346, 383 Qal’at Bani Hammad, Algeria 266 Qala’un, Sultan Muhammad b 15 n92 qandil 79 Qaran 411 Fig. 23 Qaṣr al-Hair al-Gharbī 323 n33 Qaṣr al-Hair West 174 n2, n3, 175 n3, 176 & n16, n18–21, 178 n26, 201, 11, 215 n122, 217, 265 Qaṣr-i Shīrīn 324 Qaṣr al-Tūba 66, 174 n3, 175 n4 Qazvin 264, 272 n56 Qazvīnī, Zakariyā 315 Qinnasrin 179 Qubadabad 3, 23 Quhrūd 80, 94, 347 n96 Qumm 3, 95, 96 Qur’an 28, 43 n9, 46 & n21,47 & n27, n28, 53 & n29, n30, n31, n32, n34, n35, 55 & n45, 59, 69, 70, 79, 93, 105 & n14, 111, 113, 117, 118, 119, 127, 135, 138, 142, 143, 156 & n27, 150 & n90, 166, 170, 172, 173, 184 n64, 274, 309, 312, 315, 340, 342, 379, 380, 381, 386, 389 Fig. 1, 390 Fig. 2, 393 Fig. 5, 401 Fig. 13, 403 Fig.15, 443, 444, 458, 464, 469 Qusayr ‘Amra 27, 175 n10, 176 & n14, n18, n20, n21, 300 n57, 319 n13, 344 n85, 458 n115 Rabat 266, 346 n94 Rabbat, Nasser 429 n5b, 430 n5d, 449 n88 Rabenou collection 1, 19 Fig. 4, 7–11, 13, 16, 25, 29

511 Rabi‘ b. Zaid 180 Raby, J. 135 n17, 313, 460 n125 Rácz, I. 369 n41 al-Radi 470 n10 Radkan West tomb tower 182 n56, 282 n9 Ragnarok 343 Rasgat 53 n35 Rashad, M. 138 n35 al-Rashid 470 n11 Rashid 255 Rashid al-Din 7, 37, 339, 473, 476, 482 Ravenna baptisteries 344 San Vitale 323 n37 Ray, S. 45 n17 Raymond, André 435 n26 Rayy 308 lustreware 64 sculpture 267 Real Academia de la Historia, Madrid see: Veil of Hasham II Redford, Scott 431 n6, 434 n23, 439 n41, 445 n67, n68, 448 n82 Reitlinger, G. 346 n93 Rembrandt van Rijn 129 Rempel’, L. I. 355 n23 Renard, Jack 451 n95, 460 n127 Repton, Humphrey 127 Rhie, M. M. 26 n126, n128, 129 Rhodes, John 462 n131 Rice, David Storm 10 n49, 191 n86, 273, 316, 443 n61, 470 n14 Rice, D. T. 321 n24, 339 n60, 476 n50 Richter-Bernburg, Lutz 282 n11 Rida-i ‘Abbasi 278 Riegl, Alois 430 n5e, 462 n134 Rimsky-Korsakov, Nikolai 125 riq’a inscriptions 46, 56 Robinson, Basil 273, 329, 341 n68 Robinson, Cynthia 466 n150 Roberts, Caroline 306 n1 Rogers, J. M. 8 n33, 54 n38, 55 n43, 56 n47, 306 n1, 311, 341 n66,

512 435 n26, 454 n104 Romanesque 129 Root, M. C. 350 n3 Rose, Katherine 433 n2 Rosen-Ayalon, M. 97 n1, 350 n3, 465 n142 Rosenthal, F 11 n59 Rostovtzeff, Michael 285 n28 Rowson, E. K. 137 n33 Rossini, Gioachino A. 125 Rotter, Gernot 475 n46 Roux, J.-P. 11 n57, n58 royal iconography 1–29 Royal Pavilion, Brighton 127 Royal Scottish Museum 82 n34, 329 Rubruck, William of 473 Rudenstine, A. Z. 97 n3 Rumi, Jalāl al-Dīn Muhammad 471 n21 Ruggles, D. F. 179 n37, n38 Rustam 80 n22, 93, 317, 319, 322, 339 n60 Haft Khvān 317 Russian intervention 39, 40 Sa‘ada 255 Sa‘di 472 al-Sabah Collection, Kuwait 211 n110, 214 n119, 258 al Saba, Shaikha Hussah Sabah alSalem450 n94 Sabra al-Mansuriyya, Tunisia 266 Sadan, J. 12 n65, n66, 469 n3, 470 n7, 472 n29, 475 n44 Safadi, Yasin H. 274, 357 n31 Safavids 38 & n1, n2, 39, 40, 42,44, 47 n20, 54 & n41, 55 & n45, 56 & n48, 59, 60, 62, 63, 308, 316, 345 & n89, 346, 347 & n97, 348, 473 n36 book painting 39, 273, 274, 381–386 carpets 39, 340, 342 doors, silver 39 n2, 40 sarcophagus 38–62 textile 410 Fig 22

tiles 80, 92 Saffarids 282 Safwat, Nabil 461 n129 Said, Edward 99 & n8, 130, 131 St. Calais, Sarthe silk 285 n36, 286 St. Calmin, Mozac 293 St. Josse silk 193, 194, 301 St. Lupicinus diptych 5 n18 St. Mark 126 St. Sophia, Kiev 322 n32 Saljuq period 281, 283, 301, 375, 457, 472, 473 n36, 476 architecture 277, 278 bricks 73 coinage 14 lustreware 1, 2, 10, 12–16, 23 n103, 24, 64–65, 71, 73, 79, 88, 92, 95, 358, 395 Fig. 7, 308, 358, 395 Fig. 7, 396 Fig. 8, 397 Fig. 9 Qur’ans 105 silk robe 384, 392 Fig. 4 silver 383, 398 Fig.10 sculpture 267, 268 tile 379 table 57 thrones 12 n64 wall painting 81 Sāmānids 6, 68, 69–71, 73, 282 & n9, 309, 355, 391 Fig. 3 Samarqand 67, 82 n24, 274, 345, 355 Samarra 28, 81, 83 n25, 84 n26, 180, 191 n89, 215, 216, 301, 470 San‘a’ 128, Great Mosque 177 n23, 265 pottery 71 n17 Qur’an 380 Sanchuelo 252 Sancisi-Weerdenburg, H. 350 n3 Sanguszko carpet 330 Fig. 9, 340–348 Sanjar ibn Malik Shah 338 n59, 482 Fig.5 Sanskrit 130

INDEX

Santa Maria La Real, Fitero 233 Sant’Ambrogio, Milan see: Milan silk San Vitale, Ravenna 323 n37 Sarāb 44 Saragossa 189 n78, 276 Sarikhani Collection 378–419 figural painting 381–385 geometry 387 landscape 385–387 writing 378–381 Sarre, F. 44 n11, 46 n19, 97, 98, 345 n87, 352 n10, 451 n98, n99 sarcophagus see: Isma‘il, Shah, Safavids Sasanian 85, 217, 265 arts 249–377 ceramics 5, 7, 8 n38,10, 12, 13 16, 23, 24, 32, 66, 67 coinage 137–139, 141–144, 166, 169 283 hunting imagery 279–301, 281 282 n6, n14, 283, 286, 287–22 Figs. 1–11, 369 & n41 silver 66 n2, 368 Figs. 11, 12 wine drinking 470 Sauvaget, J. 106 n19, 350 n3, 436 n32, 438 & n38, 449 Savory, R. M. 61 n69 Sawhill, J. C. 97 n3 Saxl, F. 101 Sayf al-Dīn Bakharzī 62 n72 Sāyin (Gādīkī) 44 Schacht, Joseph 435 n25 Schefer see: maqamat Schick R. 133 n4 Schimmel, A. 15 n91, 29 n142, 191 n86, 341 n70, n71 Schimmel, Norbert 97 n3 Schleswig 127 Schloss Castell, Tägerwilen 128 Schloss Gottorf 127 Schlumberger, Daniel. 6 n2, 8 n35, 174 n1, n2, 175 n3, 176 n13, n16, n18–21, 177 n25, 178 n26, 276, 323 n33, 350 n5

513 Schmid, Hansjörg 276 Schnyder, R. 15 n90 Schroeder, E. 13 n71, 448, 460 n123, 471 n18, n22 Schulz, Philipp Walter 451 n99 Schwarz, P. 324 n39 Schwetzingen 127 Scott, SirWalter 125 sculpture, figural 264–270 Sears, Betsy 436 n34 Sears, S.D. 140 n39, 158 n82 Segovia 178 n28 Seher-Thoss, S. P. 345 n88 Selim, Sultan 54 Setubal 124 Seyller, John 449 Sezincote 127 Shabustari 476 n47 Shaghād 339 n60 shahāda 117, 137, 140, 141, 144, 166 & n123 Shah ‘Abbās 330, 345 n91, 348 Shahbazi S. 13 n72 Shah-i Mashhad 277 Shah-i Zinda mosque 57 Shah Jahan 313 Shahnama 14, 15 n91, 39, 62, 80, 91, 93, 275, 281, 297, 316 &n7, 317, 318, 319, 328 Fig.7, 339 n60, 340 n61, 356 n67, 364 Fig.7, 382, 408 Fig. 20, 411 Fig.23, 423, 425, 430 n5b, 449, 459 & n120, 460 & n123, 476 Shahnama-yi Shahi 411 Fig. 23 Shaibanid 382 Shaked,S. 351 n8 Shalem, Avinoam 215 n123, 297 n45, n46 Shapur I relief 320 n17, 322 see also Freer plate Shaqq al-Taimara 138 Shaw, S. 40 n4 Shaykh Haydar 59 Shaykh Ibrahim 59 Shaykh Ṣadr al-Din Mūsā 59, 61 &

n69 Shaykh Ṣafī 41, 44, 46, 58, 59, 61 & n69 Shepherd, D. 352 n12 Sherley, Sir Robert 124 Shihāb al-Dīn Qāsim Ṭarāz 93 Shi‘ites 9, 93, 94, 332 Fig 12, 345 n91, 347 n97 Shiraz 118, 275, 345 n89, 364 Fig.7 Shīr Dār madrasa 345 Shirin 283, 314, 317, 319, 329 Fig. 8, 352, 407 Fig.19 Shoso-in, Nara 298 Siberia 269 Sidi ‘Abid al-Gharyani, Qairawān 266 Silos casket 261 Fig Silver, Larry 430 n5e, 440 n42, n45, n46 Simon, J.-P. 321 n21 Simpson, Marianne Shreve 428 n2, 430 n5b, n5e, 5f, 434 n23, 436 n34, 443 n57, 460 n126, 462 n131, 465 n143, 477 n58 Sims, E. G. 13 n71,29 n145 Simson, Otto von 440 n45 Sinjar 346 n93 Sīrāf 65 & n1 Sīrjān 73, 309 Siroux, M. 283 n21, 355 n21 Sīstān 141 n42, 277, 278 Sivas 269 Skelton, R. 57 n54 Smirnov, Y. I. 353 n15 Smith, Baldwin 447 Smith, M. B. 283 n21, 355 n21 Smith, Richard Candida 429 n4 Smyth, C. H. 97 n3 Sogdians 356 n28, 384 Somnath Temple, Gujarat 264 Soret, Frédéric 146 n67 Sotheby’s 56 n50, n51, 59, 137 n31, 138 n55, 143 n55, 155 n71, 156 n13, 157 n79, 158 n85, 161 n99, n100, 167 n125 Soucek, P. P. 27 n131, 97 n3, 283

n19, 317, 318 & n10, 319 & n13, n14, 320 & n15, n18, n20, 322, 348, 352 n9, n10, 355 n22, 431 n8 Soudavar, A. 60 n68, 370 n46, 460 & n125 Southgate, Minoo S. 317 Sourdel-Thomine, Janine 6 n21, 8 n35, 15, 28 n135, 84 n28, 166 n123, 175 n10, 176 n15, n17, n21, 212 n112, 276, 316 n5, 445 & n43, 475 n43 Souto, J. A. 182 n52 Spain 99, 100, 109, 123, 161 n98, 171, ivories 174–263 and Umayyad Syria 174–263 Sprenger, A 315 n4 Spuler, B. 15, 28 n135, 84 n28, 166 n123, 175 n10, 176 n15, n17, n21, 212 n112, 282 n14, 470 n6, 472 n25, 473 n33, n34, 475 n43 standing caliph 136, 137 & n32 State Museum of the History of Turkmenia, Ashkhabad vase 361 Fig 2 State Public Library, St. Petersburg Shahnameh, frontispiece 364 Fig. 7 Stein, Linda 428 n1 Steinhardt, Nancy S. 430 n5e, 441 n48, 49 n86, 466 n147 Stern, Henri 213 n115 Stern, S. M. 281 n4, 359 n39 Steve, H. M.-J. 369 n43 Stewart, Desmond 475 n45 Strelka 367, 372 n57 Strong, D. E. 320 n16, 321 n22 Strzygowski, J. 98, 193 n96, 451 n96, 458 Stuttgart 128 Subh 230, 233, 236 Sufis and Sufism 9 n42, n45, n46, n47, 11, 15 n91, 25, 27, 47 n24, 93, 172, 317, 343, 347, 370 n47, 379, 471, 473 Suhrāb 316

INDEX

al-Suhrawardi 15 Sulaiman, Caliph 470 Suleman, Fahmida 443 n61 Suleimanova, F. 60 n68 al-Suli 470 n10 Sulṭāniya 82 Sunni 9, 283, 345 Sunqur Bey Mosque, Niğde 269 sura 28 n141, 47, 53, 70 & n15, 114, 142, 143 & n51,159, 161, 380 Susa 84, 369 & n43 Swietochowski, Marie L. 318, 338, 339, 340 Syria 25, 65, 134, 143, 168 & n129, 264–265, 267, 281–281, 285, 293, 309–310, 350, 424, 451, 454 n104, 461 influence on Spanish ivories 174–263 Taaffe, Philip 430 n5e al-Tabari 137 n33, 299 n52, 474 al-Tha’ālibīs 314 n3, 319 n12 Tabaristan 283 Tabasi, poet 9 n47 Tabbaa, Yasser 442 n54, 443 n60, 453 n101, 455 n109 Taboroff, J. H.105 n16, 314 Tabrīz 55, 60, 62 n74, 116, 268, 311, 382, 403, 482 al-Tabrizi, Ata Allah b. Muhammad 116 Tafel, Emil Otto, 128 Tägerwilen 128 Tahirids 282 Tahmāsp, Shah 44 & n11, 46 n50, 56 n50, 62, 116, 382 Takht-i Sulaiman 3, 23, 80, 82, 91, 94, 343 Talbot-Rice, D. T. 7 n26, n32, 10 n49, 357 n31, 363, 458 n114 Talisman Gate, Baghdad 268 Tallis, Nigel 295 n37 Tang-i Sarwak 371 Tanindi, Zeren 306 n1

515 Tanovoli Collection 5 n50 Taq-i Bustan 283, 284, 287 Fig.1, 317, 318, 319 & n13, 321 n21, 322, 323 &n35, 324, 325 Figs, 1, 2, 326 Figs. 3, 4, 327 Fig. 5, 351 & n9 Taq-i Kisra 283 Tarāsh-i Farhād 318 n10 Tarnoy, H. 369 n41 Tatar 127 Tavernier, Jean-Baptiste 124 Tcheliabinsk silk 301 n58 Tenri University Sankōkan Museum, Japan 357 n31 Tepe Sialk, Kashan 369 n43 Tepe Hisar, Dāmghān 369 n43 Termez, palace 267 Terrasse, H. 180 n43 Textile Museum, Washington, D.C. ikat inscription 200 Fig. 4e silk cloth 198 Fig. 2b Tezcan, Hülye 306 n1 Theodosius 321 n22 obelisk 322 n32 missorium 26 Theophanes 136 n25 Thiersch, Hermann 451 n97 Throne Verse 117 Thurman, R.A. F. 26 n126, n128, n129 Tibet 26 Tiles Ghazni 73, 75 Fig. 4, 79 Ilkhanid 73, 82 Iran 76 Fig 5, 77 Fig 6 Iznik 82 Kashan 28, 400 Fig 12 Kūhpāya 83 mina’i 79 Qubadabad 3, 23 Sāmarrā 81 Safavid 80, 92 Saljuq 379 star tile 28 n140 Takht-i Sulaiman 3, 23, 35, 91 Timurid 82

516 in Victoria & Albert Museum, London 28 n141 Tiraz bands 13 n74 Timur 338 n59 Timurid iconography 345 painting 29, 39, 59 n62 n63, 60 & n68, 61, 62, 273, 316, 381, 382 pottery 64, 82,88 Titley, Norah 273, 334 n47 Tivoli, Copenhagen 128 Tohme,Tara 429 n5a Toledo 276 Tomar 124 Topkapı Sarayı Museum, Istanbul 54 & n38, 55, 56, 116, 173, 306 n1, 311, 373 n 60 Torres Balbás, L. 180 n43 Toynbee, J. M. C. 321 n22 Transoxiana 68 Treadwell, Luke 132 n2, 282 n12 Treasury of the Cathedral of Gerona see: Hisham II Treasury of the Cathedral of SaintJust and Saint-Pasteur, Narbonne 262 Treasury of the Cathedral, Bamberg Trenkwald, Hermann 451 n99 Trilling, James 462 n134 Trimingham, J.S. 25 n119 tripylon, Persepolis 336 n53 Tronzo, William 429 n 5a Trümpelmann, L 318 n10 Tughluqids 6 n21 Tughtegin 473 n36 Ṭūlūnid 71, 425, 456, 457 Turan 382 Turbat-i Shaikh Jam 82, 330 Fig. 10, 345 & n91 Turkman (Turcoman) book painting 62, 274, 275, 381, 382 Turkish society 16 Twain, Mark 104 & n12 Tyler 284 n25

Ufa plate 300 n57 Uighur 6 Ukhaiḍir 69 Ulūgh Beg 59, 62 University of Edinburgh Library 339 n60 University of Michigan 337 n57 Unger, Edmund de and the Keir Collection 63, 83, 431 n9 Umayyad 27, 139, 168, 174, 281, 349, 350, 359, 445, 453 n103 ceramics 65, 66, 67 architecture 66 coinage 67, 132 & n4, n5, 133 & n5, 140, 141 & n44, 144 & n56, n57, 145 n61, n62, 146 n67, 155, 157 & n80, 158, 159 & n90, 161, 164 & n113, n116, 165 & n117, 166, 167,169 & n131, 192, 296 ivories 174–263 mosaic 84, 344 silverware 66 & n2 textile 301 Syria 174–263 The Unicorn 106 n20, 109, 31 n6 Ushak carpets 57 n52 ‘Uthman, caliph 184 n64 Ūzgand 73 Valentino, Rudolph 126 Valerian 320 n17, 322 Vallejo Triano, Antonio 217 n128 Varamīn 96 Varqa va Gulshah 6 n22, 8 n34 Veil of Hisham II 199 Fig. 3a Venice 126, 186 n69 Vernoit, S. 105 n17, 109 n26, 446 n71 Veronese, P. 129 Victoria and Albert Museum, London 28, 126, ivory casket 187, 198 Figs 2d–g, 206 Fig. 13, 207 Figs, 14–16, 211, Figs. 228, 229, 232, 257 Persian metalwork catalogue 273 pyxis 201 Fig. 6, 213 n118, 231

517

INDEX

Fig. 246 Fig Vienna 128 Vienna Genesis, 335 n51 see also: Nationalbibliothek, Vienna Vienna Galen frontispiece 5 n18, 13 n77 Viré, F. 359 n39 Volbach, W. F. 284 n25, 286 n29, 287, 288, 289, 293 n31, n32, n33, 294 n34, n35, 334 n45, 482 n34 Vologases II, drawing from a coin 365 Fig. 8 Volov, L. 71, 183 n58 Volvinus 285 Wade Cup 109 & n23, 115 Wahba, Magdi 435 n26 al-Walid I 177 n24, 178, 470 al-Walid II 265, 470, 472 Walker, J. 137 n29, 142 n50,143 n53, 155 n71, 158 n86, n87, 160 n93, 161 n98, n100, 162 n101, 164 n110, n114, n116, 167 n124, 168 n127,182 n54, 192 n92, 283 n17, 443 n61 Walters Art Gallery, Baltimore 47 n26, 53 & n37, 334 n44, 365 silver dish 366 Fig. 9 Ward, R. 369 n45, 372 n58 Wardwell, A. E. 281 n2, 352 n12 Wāsiṭ 65, 165, 166, 167, 168 n127, 335 & n52, 433 n16 al-Wāsiti, Yahya 335 n52, 336, 337, 461 & n128, 479 Fig. 2 Wasmuth, E. 44 n11, 451 n98, 99 Watson, Oliver. 2 n7, 3 n11, n12, 11 n57, 14 n107, 25 n114, 27 n134, 72 n19, 90 & n1, 91–95, 306 n1, 310 347 n96, 358 n34, 445 & n68, n69 Watzinger, Carl 451 n96 Weaver, M. E. 41 n6, 44 n11, 45 n15, 46 n19, 59 n61, 62 n74 Weibel, A. C. 175 n9 Weil-Garris Brandt, K. 103 n10

Weir, M. G. 134 n10, n11 Weitzmann, K. 4 n16, 26 n123, 97 n3, 103, 425, 447, 460 Welch, Anthony 274, 449 Welch, Stuart Cary 62 & n73, 274, 448 & n84, 449 Wellesz, E. 335 n51 Wellhausen, J. 134 n10, n11, 142 n45, 168 n129 Wensinck, A. J. 469 n3 Wetzel, Friedrich 451 n97 Whelan, E. 10 n53 Whitehouse, D. 65 & n1 Wiesbaden 128 Wiet, G. 193 n96 Wilber, D. N. 278, 342 n75 Wilds, A. H. 127 Wilhelma Park, Stuttgart 128 Wilkie, Sir David 129 Wilkinson, C. K. 356 n29, 357 n30, 358 n32, n36, 362 Williamson, A. 64, 310 Windad Ohrmazd of the Karens 374 Winder, R. Bayly 97 n3, 433, 447 Wine drinking 469–482 Wittkower, Rudolf 101, 440 n45 Wood, Barry 431 n6 Wulff, H. E. 41 n7 Wulzinge, Karl 451 n96 Wyatt, M. D. 128 Xerxes 294 Yaḥyā ibn Ādam 168 n129 Yāqūt al-Hamawi 324 n39 Yarshater, Ehsan 315 Yasskand/Yasukand 356 Yazd 93 Yazdigird silk 290 Fig.7, 294, 295 Yazdigirdi calendar 283 Yazd-i Khast 283, 355 n21 Yemen 128, 195, 200 Yoshida, M. 357 n31 Yusuf-Ali, A. 53 n32 Yusuf-Jamali, M. K. 44 n13, n14, 45 n16

518

Zahhak 356 n27 al-Zahra 266 Zāl 341 n60 Zamora 181 n48, 86, pyxis 196 n108, 213, 330 Fig Zangi 472 Zanth, L. 128 Zayn, Dzul Haimi b. Muhammad 55 n45 Zick-Nissen, J. 356 n26, 475 n43 Zirid dynasty 266 Ziwiye treasure 295 n38

Ziyad ibn Aflah 246 Ziyad/Zayyan, Muhammad ibn 189 n80, n81, 261 see: Burgos Casket Ziyar, Mardawij b. 282 n14 Ziyarids 282 Zoroastrian 473 Zotenberg, H 319 n12 Zuhayr ibn Muhammad al-‘Amiri 255, 256 Zygulski, Z. 14 n80 Zwalf, W 26 n 127, n128

Acknowledgements The articles reprinted in this volume first appeared in the following works I

Islamic Art in the Ashmolean Museum, Part I, ed. J.W. Allan (Oxford Studies in Islamic Art, X, Oxford, 1995), 167–98

II

Society and Culture in the Early Modern Middle East. Studies on Iran in the Safavid Period, ed. A.J. Newman (Leiden, 2003), 165–92

III

Persica, Vol. 10 (1982), 119–147

IV

Oriental Art, Vol. 34, (1988–89), 282–283

V

Discovering Islamic Art. Scholars, Collectors and Collections, ed S. Vernoit (London, 2000), 171–81

VI

Ten Poems from Hafez, ed. Jila Peacock (Lewes, 2006), 9–17

VII

Journal of the Scottish Society for Art History, Vol. 6 (2001), 6–10

VIII

Al-Andalus und Europa. Zwischen Orient und Okzident, ed. Martina Müller-Wiener, Christiane Kothe (Petersberg, 2004), 20–38

IX

National Art Collections Fund Quarterly, Spring 2004, 48–51

X

The Ivories of Muslim Spain. Papers from a Symposium held in Copenhagen from the 18th to the 20th November, 2003, ed Kjeld von Folsach and Joachim Meyer (Copenhagen, 2005), 49–74

520

XI

The Dictionary of Art, ed J.S. Turner (London, 1996), 245–247

XII

Art History. Journal of the Association of Art Historians, Vol. 8, 2 (1985), 265–270

XIII

The Rise of Islam (The Idea of Iran. Volume IV), eds. V.S. Curtis and S. Stewart (London, 2009), 84–101.

XIV

Art History. Journal of the Association of Art Historians, Vol. 12, 1 (1989), 109–115

XV

Oriental Art, Vol. 24 (1978), 201–213

XVI

Sifting Sands, Reading Signs: Studies in Honour of Professor Géza Fehérvári, ed. Patricia L. Baker and Barbara Brend (London, 2006), 215–25

XVII

The Sarikhani Collection. An Introduction, ed. Ina Sandmann (London, 2011), 13-21

XVIII Iranian Studies Vol. 44/1 (2011), pp. 1–6 XIX

Journal of Art Historiography, Vol. 6 (June 2012), 1–35

XX

Court and Craft: A Masterpiece from Northern Iraq, ed. Rachel Ward (London 2014), 38–45