Boston Public Library Robert 1 lillenbrand was educated at the universities of Cambridge and Oxford. Since [97] he has taught .


?•.:’.« ..

;i I I e n b r a n d



Boston Public Library



Robert 1 lillenbrand

was educated at the universities

of Cambridge and Oxford. Since [97] he has taught .it the

University of Edinburgh, and in [989 was made Professoi of

Islamic Art there. He has travelled extensively throughout tl Islamic world from Morocco to Southeast \sia, and has held

visiting professorships at Princeton, 1 os Angeles. 1 Dartmouth

and Bamberg. He is the author of over a hundred publications, including books on Persian painting

and Islamic architecture.



This famous series

provides the widest available

range of illustrated books on art in all its aspe< ts

Ifyou would like to receive a complete list

of titles m print please write to: 1 HAM] s wo ill Dso\

30 Bloomsbury Street. I ondon w< IB 3QP

In the United States please write to: – f m . THAMES AJMJ^fll^ Avenue. New York, Ne* York 101 1.6 , *’:. New >

No le-

gale of !

the I







Robert Hillenbrand

Islamic Art

and Architecture 2jo illustrations, So in color




For Margaret and Ruthie,

with love and thanksfor years of encouragement

frontispiece Rustam lassoing Kamus. Firdausi. Shaknama, probably

Tabriz, c. 1505; attributable to Sultan Muhammad. Rustam. the hero of the Shahnama, the Persian national epic, is distinguished bv the tiger

skin that he wears over his body armour and his leapaid’s-head casque

crowned by a mighty seven-fold plume. Note the floating text panels

and concealed grotesques in the landscape.

ACKNOWLEDGMENT I owe a great deal to Dr Barbara Brend. who read the text with painstaking care and unstinnnglv provided main


suggestions for its improvement. The book is much the richer for her


Any copy of this book issued by the publisher as j is sold

subject to the condition that it shall not by way of trade or otherwise be lent, resold, hired out or otherwise circulated without the publisher’s

prior consent in any form of binding or cover other than that in which it

is published and without a similar condition including these words being

imposed on a subsequent purchaser.

© 1999 Thames and Hudson Ltd. London

First published in paperback in the United States ofAmerica U

by Thames and Hudson Inc.. SOO Firth Avenue. New York. New York I o 1 1

Library of Congress Catalog Card Number NA380 isbn 0-500-20305-9

. H52

All Rights Reserved No pan of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical,

including photocopy, recording or any other information storage and

retrieval system, without prior permission in writing from the publisher

Printed and bound in Slovenia


^A380 .H52 1999x





CHAPTER ONE The Birth of Islamic Art: the Umayyads 10




CHAPTER THREE The Fatimids 6l


CHAPTER FIVE The Age of the Atabegs: Syria, Iraq and Anatolia, i 100-1300 1 1 1

CHAPTER SIX The Mamluks [38

CHAPTER SEVEN The Muslim West [67

CHAPTER EIGHT The Ilkhanids and Timurids [96

CHAPTER NINE The Safavids 226

CHAPTER TEN The Ottomans 2$ 5








Bukhara * Samarqajul


vr > l'<y, Tinnidh

»Ti Sarakhs

M . , . • Mashhad C; A R J I S T A N

Sultaniy; ^amgnaj .^^ Nishapur- , Sangbast Qazvm Tehran •Simnan

• Zuzan Kabu] Raw~7J #X araniin

Qumm» Kashan*


Natanz •



Herat Ghazna






Any attempt to make sense of Islamic art and architecture as a whole while retaining a chronological framework runs the risk of distor-

tion. Bias of several different kinds is hard to avoid. It is simply not

possible to be equally well informed and equally interested in all

aspects of the subject. The need to consider in some detail the early centuries of Islamic art is made imperative by the major impact which work of this period had on later art. But a great deal of this early art has perished, and to do justice to what survives in the

context of its own time and of subsequent periods demands a closer and more detailed focus than is appropriate for the more numerous examples of later art. Some degree of over-balance is therefore inevitable.

Certain art forms such as calligraphy or textiles continued to be

produced in most parts ot the Islamic world from early times, but

they are not of equal significance in each area or period. Thus the absence of a discussion of, say, Tulunid woodwork, Maghribi pottery, Timurid textiles, Spanish metahvork or Ottoman Qur’ans should not be interpreted as a signal that they did not exist, have not sur-

vived or are of peripheral interest. It is simply that it seemed best to reserve a discussion of certain media for those periods m which pro- duction was of the most significant scale and quality Similarly, the art

of entire dynasties – Ghaznavid, Turcoman, the beyliks of Anatolia, the muluk al-tawa’if of medieval Spain – is virtually ignored. Such omissions are dictated by the rigorous word limit and the need to see the wood rather than the trees. In other words, the option of trying to say something, however little, about almost everything, and thus writing a rather bland and trivial text, was rejected. It seemed prefer- able to single out key objects and monuments for relatively detailed scrutiny, in the hope that they would provide a means of entry into the school or style that produced them. This book, then, is more a study of the peaks than of the valleys; its colours are intended to be

bold and primary.



A secondary aim has been to sot the various schools and types of Islamic art in a reasonably full historical context so that the images

are not, so to speak, trapped in limbo. Specialists will have to console

themselves with the thought that this hook waf not written with them in mind. It is truly no more than an introduction to .1 vast field.

Moreover, the very (act that .1 hook with the .ill inclusive tide of Islamic Art and Architecture can be written whereas the hooks on

western European art in the Work! of Art series are of a very much more specialized kind, and are often devoted to .1 single school, or

even artist – is .1 reminder that the volume of scholarship consecrated

to this field is tiny in comparison with that available foi I uropean

art. Basic guides to the territory therefore still have their function.

But it would be a serious mist. ike to assume from that disparity that there is any less ‘going on’ in Islamic than in European art. You just

have to dig rather deeper for it.



For the sake of simplicity and consistency, year dates are shown m accordance with the Gregorian calendar, but with occasional men tions of their equivalents in the Muslim calendar (based on the lunai

I y< le) in c on nee tion with spec it u ally dated buildings Or works of art.

Muslim years are < al< ulated Mom the date of the hijra the Prophet’s journey from Mecca to Medina in |uK 1



rrri ^

-•-V~: i

i The pivot of Islam. The Ka’ba in the Masjid al-Haram, Mecca: principal Islamic shrine and the goal of Muslim pilgrimage. Frequently restored, it contains the Black Stone, the directional focus for Muslim prayer, and is covered – like a bride’ according to medieval poets – with the kisica, a silken veil, now black but formerly in many colours.




The Birth of Islamic Art: the Umayyads

The genesis of Islamic art is customarily linked with, indeed often attributed to, the whirlwind military conquests of the Arabs follow-

ing the death of the Prophet Muhammad in ad 632. Such an idea is plausible enough. The creation of a world empire, the proclamation of a new faith, the formation of an art that bears its name – all seem to belong together. But do they? Is there a causal connection, and –

if so – what is the exact chronological sequence? Dazzling and excit- ing as the spectacle of the Arab conquests is, it in fact has relatively

little to do with the early years of Islamic art. Yet the formative

nature of those early years is plain. What, then, is the precise

connection between the seismic political events of the seventh

century and the earliest Islamic art?

The answer to such questions demands a refinement of the chronological and geographical focus. To view early Islamic art as even approximately representative of an empire that stretched from

the Atlantic to India and the borders of China is grossly to misunder-

stand its context. In the two generations which saw the Arabs flood

out of their desert homeland and overrun all of western Asia and North Africa there was, it seems, neither the desire nor the time to foster artistic expression. That was to be the achievement not of the

first conquerors themselves but of their grandchildren. At all events,

no major building or artefact survives from these early years. This

sluggish start may owe something to the fact that in this period the nascent Muslim state was being ruled from Arabia, an environment

in which the visual arts, though by no means absent – as recent

excavations at Qaryat al-Faw (frescoes of royal scenes) and elsewhere

(figural sculpture) have shown – nevertheless had no very significant role, though architecture flourished. Arabia certainly lagged far

behind the Levant. Similarly, there can be no question of a ‘universal’

Islamic art at this early stage. The horizons of that art were

effectively limited to Syria. The rest of the Islamic empire might as well scarcely have existed .it all, except insofar .is works oi art or

1 1



craftsmen from outside Syria were active within that province and

thus exerted external influence on the art produced there.

These remarks might lead one to expect a somewhat parochial

quality in the earliest Islamic art. and also a certain timidity or lack of

purpose. Yet this is not so. Such characteristics might well have

marked the verv first monuments which the Muslims erected, for example in Fustat. Basra and Kllfa – although there is no way o\

clinching this, for they have not survived But if Islamic art was slow

to start, it was quick to gather speed. Certainly the first major monu- ment to survive, the Dome of the Rock in Jerusalem, radiates assur- ance. A new art has arrived It established itself quickly and, for all that numerous experiments and changes of mind can he detected during the rule of the Umavvad dynast) 66] ~so). the pervasive confidence of the age remained undimmed.

This confidence, one of the most striking features of Umavvad art, was founded on several interrelated factors. Chief among them, perhaps, was the astonishing military si;, the Arabs in their

foreign campaigns. To their enemies they must have appeared to bear

charmed lives, their winning streak seeming unassailable for much o\ the Umayyad period. Decade after decade the borders of the dor al- islam steadily expanded, until m ~;^ exactly a century after the Prophet’s death – the Arab defeat at Poitiers m central France sig- nalled (though only with the hindsight of history) the end of sub-

stantial territorial gams for some centuries. But the splendid confidence ot the Umayyads was not based entircK on military success abroad; it was founded also on the ability of the new dynasty to survive numerous challenges from within. Such challenges were at their most dangerous m the first thim years ofUmayyad rule. It mav be no more than a coincidence that this same period was singularly barren so far as the production of works ofart was concerned. Yet it is probable that the outburst of building activity which followed the consolidation of Umavvad power and the dynasty’s triumph over its internal enemies should be seen at least partly in a political light – in this particular ease, as a celebration of” Umayyad dominance. This propaganda dimension was frequently to reappear in Islamic art, especially in architecture, although it tended to be ot secondary

rather than primarv significance.

Allied to the understandable confidence generated b\ spectacular

military successes at home and abroad was a confidence based on a sense of secure dynastic power. The Umayyads had abrogated the primordial Islamic notion o{ an elective succession to the caliphate



and replaced it by the dynastic principle. The internal political turmoil of the later seventh century was in large measure caused –

and maintained – by that action. Once victorious over their enemies, however, the Umayyads were able to indulge a heady consciousness of family power for which history can offer few parallels. For several caliphs – notably


Abd al-Malik’, al-Walid I and al-Wahd II – this sense of dynastic pride found its most public expression in ambitious

building campaigns. The caliphs Sulaiman and Hisham were not far behind, and other princes of the royal family, such as al-


Abbas b. al-

Walid and Ghamr b. Yazid, followed suit. Indeed, to judge by the quantities of religious and secular buildings erected in Syria between

690 and 750 under the direct patronage of the Umayyad royal house, architecture speedily became a family business. The immense financial resources of the Islamic state, whose exchequer was swollen by the accumulated booty of the Arab conquests and by the taxation

revenue which came pouring in thereafter, were at the disposal of the Umayyad builders. Thus


Abd al-Malik was able to set aside the tax revenues of Egypt for seven years to pay for the Dome of the Rock, while his son al-Walid I devoted the entire tax revenue of

Syria for seven years to the building and embellishment of the Great

Mosque of Damascus. There was thus both the will and the means to embark on grandiose building projects.

Enough has been said to account for the superb self-confidence which triggered and then fuelled the massive building programme ot the Umayyads. Yet the geographical location of these buildings also

requires explanation. Given that they are to be found, with very few

exceptions, exclusively in Syria, how was an undue parochialism, peculiarly inappropriate to a world empire, avoided? The answer is three-fold. First, Syria under the Umayyads was beyond compare the most favoured land in the Islamic empire. Its inhabitants enjoyed

privileges and concessions denied to those from other provinces. Its

principal city, Damascus, was from 661 the capital of the empire.

Here was established the Umayyad court and administration, when these were not to be found toiling in the wake ot semi-nomadic

caliphs. The massive caliphal investment 111 agricultural installations canals, dams, wells, gardens and so on, culminating in the planned

but abortive diversion of the RiverJordan itself- made Syria perhaps

even exceed Iraq .is the richest province m the empire. Thus abun- dant wealth complemented its politic. il prestige.

Parochialism in Umayyad art was further discouraged by the prac rice of conscripting labour and materials from other provinces. 1 his



custom ensured that Syrian material culture would be metropolitan.

The caliphs could dip at will into an extensive labour pool within their own domains, and could supplement this by importing still more craftsmen and materials from outside the Islamic world, notably from Byzantium. The chance survival of a cache of papyri from Aphrodito in Upper Egypt documents the workings of an Islamic corvee system – essentially the leiturgid practised by Rome and Byzantium – in the early eighth century. The local governor, one Qurra b. Sharik, was responsible for sending a specified number of men to work on the Damascus mosque, and he had to provide money to cover their living expenses too. Such documentary proof of the corvee system can be supplemented by literary references – for

example, al-Taban mentions the activity of Syrian and Coptic

workmen in the building of the mosque at Medina – And, above all, by the evidence of the buildings themselves. Stucco sculpture of

Persian type, Iraqi techniques of vault construction, mouldings from

south-eastern Anatolia, a hgural style closely paralleled in Coptic

sculpture — all furnish unmistakable evidence that the style and build- ing practice of Syria was enriched by ideas and traditions from much further afield. There was no danger that the local Syrian craftsmen

would cling to their own traditions and thus risk stagnation. Finally, the position of Syria, both geographically and politically,

militated against parochialism. The province was uniquely placed to

draw inspiration from the major cultures newly yoked together to

form the Islamic empire, lb die north, west and south west lav lands

in which Graeco-Roman culture was dominant and which were either Byzantine or, like Egypt and North Africa, had recently been

wrested from Byzantine rule, [b the south was Arabia, which at this

early stage in Islamic history was still by no means a spent force in religious, cultural or political terms. lb the east lav Mesopotamia and Persia, comprising the accumulated heritage of Assyria and Babylon, and of the Achaemenids, Parthians and Sasanians. Here the tradition

of world empires died hard, though the horizons of these Middle

Eastern states were appreciably narrower than those of the


Within the Umayyad empire, then, which stretched from France to the Indus, Syria was ideally placed to act as a central point from

which metropolitan influences radiated to the outlying provinces. No other region of the Islamic world combined such a deeply rooted Hellenism with an openness to the ancient cultures of the Near East. By virtue ot its geographical position and its political pre-eminence,




i Standard mosque types

[above) The enlarged house of

the Prophet Muhammad, Medina, 624: the inspiration for

much mosque architecture.

[right) The Great Mosque of

Kufa, original form, 638: note

the hypostyle sanctuary.

[above) The Friday Mosque of Zivaratgah. Herat, [482: a

standard Iranian 4-iuwN layout.

[right) The Uc Serefeli Mosque Edirne, 144″ an example of

the domical emphasis in

Ottoman mosque architecture.



Syria was a natural bridge between east and west, north and south. It

was only to be expected that under the Umayyads its art should reflect this unique situation. The tact that those same Umayyads were not a family of local Syrian notables but the representatives of

the greatest empire in the contemporary world gave their art a

mission of the utmost seriousness. It had a public, an imperial, role.

In the immediately pre-Islamic period Syria had perforce been con-

strained to yield centre stage to Constantinople and even Alexandria,

and was thus to a certain extent an eastern appendage of a

Mediterranean-centred empire. The emergence of Islam as a world power decisively changed all this and brought Syria its scant century

of glory. Umayyad art was the public expression of that glory. So far as the future of Islamic art was concerned, this was a

crucial century, in which the face of the Mediterranean world and

the Near East was permanently redrawn. This century established

the principle that Islamic art, tar from being intrinsically universal,

could have (as it certainly began by having) a well-defined regional

and dynastic character, a feature which it consistently retained in later centuries. The Umayyad period also ensured that the funis and ideas of classical art. which were much better understood m Syria than in the lands further to the east, would enter the blood-

stream of Islamic art. As a result. Islamic architecture tends to feel

familiar to a Western observer; it employs, after all. the familiar

vocabulary of column and capital, pointed arch and dome, rib and vault. It was under the Umayyads, too. that a distinct iconography of princely life, centring around the formal, ceremonial activities of

the monarch and his leisure pursuits, was developed and refined. This set of images was to become a leitmotif of secular art through- out the Islamic world. Similarly, the success of Umayyad solutions to many problems of religious and secular architecture ensured that the building types evolved during this period repeatedly recurred in one

guise or another in subsequent centuries. This readiness of Liter

generations to copy Umayyad prototypes was at least partly due to the unique glamour which invested this, the first and most powerful

of Islamic dynasties. As already noted, too. the Umayyads recog- nized the propaganda dimension inherent in splendid buildings and

symbolic images; this also was to remain a constant of later Islamic

art. Yet this same development was viewed with some mistrust .it first, and Mu’awiva. the first Umayyad caliph, when challenged about his taste for ostentation on the Byzantine model, defended himself by asserting that ‘we are at the frontier and I desire to




rival the enemy in martial pomp, so that he may be witness to the prestige of Islam’.

Finally, the Umayyads’ choice of Syria as their power base had tremendous consequences for later Islamic art, since the generative impact of Syria was greater than that of any potential rival among the other provinces in the Islamic empire. Islamic art would have developed in a very different fashion if the Umayyads had settled. tor example, in Arabia, in Spain or in India. At the same time, lest

too much be claimed tor the art of this period, it is worth remem- bering that some of the media which were later to become most typically Islamic, such as glazed pottery, metalwork, carpets, book painting and textiles, are either totally or virtually absent from art of

this period.

What, then, are the principal expressions of Islamic art under Umayyad dominion? The so-called ‘minor arts’ are quickly disposed of. the textile fragment which, if its attribution to Marwan II is correct, would be datable to c. 750, and whose arabesques and flgural style would readily suggest Coptic work but for the Arabic inscrip- tion; some ivories for which Coptic and Byzantine as well as Umayyad provenances have been suggested, a controversy which itself sheds much light on the intrinsic nature of Umayyad art; and a little metalwork, much of it also of disputed date and provenance. The so-called ‘Marwan ewer’ in Cairo may be late Umayyad or early c

Abbasid; but its date, and indeed its provenance, is less important

than its form, which is prophetic of much of later Islamic metalwork in that it typifies the preferred Islamic response to the sculpture of

living creatures. The body of the ewer is occupied principally by a continuous arcade enclosing rosettes and animals, all lightly incised.

A pair of dolphins in high relief support the handle; but the piece de resistance is the fully three-dimensional crowing cockerel, craning

forward eagerly with his beak open in full cry, who perches on (and then himself forms) the spout of the ewer. The utilitarian function of such sculpture may well have sufficed, from the standpoint of strict orthodoxy, to justify its otherwise impiously mimetic quality. Several

similar but less ornate pieces testify to the popularity of this model.

A new chapter in Umayyad metalwork was opened with the discov- ery in [985, at the ancient site of al-Fudain, of a square bronze

brazier on wheels. At each corner stands a naked girl, sculpted 111 the

round and holding .1 bird; along the only complete side is a set of

panels with erotic images and scenes of revelling. The piece bean close affinities to the sculptures ofKhirbat al Marjar (see p.



3 (lefty Engraved base metal ewer

aseribod to the caliph Marwan II, _

s . found in Egypt. Its blind

arcades (here with solar rosettes)

recur m sura dividers m early Qur’ans, ungjazed clay lamps and

jars, and Umayyad architecture; they may. like the cockerel, symbolize

light – and even boundaries or

protection. Sasanian and 1 lellenistH

elements combine with a

distinctively Islamic aesthetic of

all-over decoration.

4 (/>«7eiD Private taste. Bronze and

iron brazier from al-Fudain, Jordan,

before 750. Probably <. ast by the lost-

wax method Ceremonial braziers occur on Assyrian reliefs, but the

griffins with outstretched wings

which form the feet recall Sasanian

metarwork, while the frank sensuality

of these pneumath figures owes

much to Coptic art and the corner figures recall the Syrian goddess






s Economic interdependence. Money scarcely existed m the Arabia ofMuhammad. For tins reason, and to maintain economic stability, the Muslims long forbore to replace the existing coinage. They contented themselves with unobtrusive tine tuning, adding Arabic inscriptions and removing religious symbols.

Another significant expression of Umayyad art deserves brief mention here: the coinage of the period. To a quite remarkable degree this coinage mirrors and encapsulates the artistic tendencies

traceable in the much more complex field of architecture. Umayyad coins faithfully reflect the long fallow period which preceded the

serious involvement of Umayyad patrons with ambitious works of art. The significant innovations in coinage are almost exactly con- temporary with the Dome of the Rock (completed in 691). As in architecture, so in coins the evolutionary trend is clear: an initially

slavish dependence on classical models gives way to an increasing preference for themes and techniques inherited from the ancient

Near East, and the resultant period of experiment produces some unexpected reworkings of old ideas in new contexts. Finally an orig- inally and distinctively Islamic solution is fashioned from these

heterogeneous elements. This entire process of acculturation and

innovation was, it seems, telescoped into little more than a decide; perhaps the limited physical scope offered by coinage resulted m Islamic forms being introduced at an accelerated pace. The evolution of coins therefore epitomizes a process which 111 other media,

notably architecture, occurred much more slowly and tentatively. In Iraq, Persia and areas even further east, Sasanian silver coins

were copied with virtually no alteration. The favourite design tea- s tured on the obverse a portrait head of Khusrau II. one of the last rulers before the Islamic conquest of Persia, and on the

reverse a tire altar with attendants. Even the name ot the Sasanian ruler in the Persian Pahlavi characters was retained, as were the




Pahlavi mint marks, while the date was given successively in the two

Sasanian calendars and then in the Islamic or Hijra reckoning. When the Muslim governor’s name was given, it was also written in Pahlavi characters. The only distinctively Islamic feature was the addition of pious expressions in Kufic script, such as ‘in the name ot God’ or ‘praise be to God’. Thus presumably Persian die makers continued to work under the Muslims. These Arab-Sasanian coins, then, show the willingness of the Muslims to maintain the status quo.

In Syria, with the spectre of a weakened but unconquered Byzantine state just north of the border, the situation was different

Here the Arabs naturally encountered not Sasanian but Byzantine coinage. They were already familiar with this, since the words dinar and dirham (from denarius and drachma respectively occur in the

Qur’an. Despite the Greek derivation of its name, the dirham men- tioned in the Qur’an, in the chapter ofJoseph, is probably a Sasanian

coin since this was by tar the most widespread silver com in the Near East – the dollar of late antiquity. The Sasanian economy was based on silver iust as that ot Byzantium was based on gold. Under the



• W?IY( ^



6 A language of symbols. Lace yth-century coins. (a) Caliph (?) at prayer with attendants. Muslim

adaptation of a standard Sasanian type. Silver dirham

minted by Bishr b. Marwan. 692-3.

(b) Bar-less cross on altar steps. Syrian gold dinar,

c. 692. The cross was modified, by letters, as here (RI: Rex Ishmaelorum. “King of the Ismaelites’?), or by a circle, thus creating a Greek phi.

(c) Standing caliph with the Muslim creed around the rim. Syrian gold dinar minted by ‘Abel al-

Malik. 696—7. A response to Byzantine gold issues. (d) Aniconic epigraphic gold dinar, Syrian. 696-7.

The forerunner of almost all later Muslim coinage.

7 God’s caliph. Silver dirham, probably Damascus

c. 692. The niche, common in classical and Christian art. here prefigures the mihrab dnd

contains the Prophet’s lance. The caliph is also metaphorically present through his title.


caliph c

Umar I, for example, the Syrians paid their taxes in gold while the Iraqis paid theirs in silver. The Arabic copper coin, the fab, is the Greek fottis in disguise; here, too, the Byzantine designs were

copied. At first Byzantine types were used without any alteration;

this was sound economic sense, for the Arabs had long been familiar with these coins in commerce. Indeed, a Syriac chronicle records

that when in 661 the caliph Mu’awiya minted gold and silver ‘the populace did not accept it as there was no cross on it’. Several well- known Byzantine types were copied, some of single standing imper- ial figures, others showing the emperor Heraclius and his two sons.

Soon, however, tiny but momentous changes were introduced; on the reverse, the cross on a stepped podium lost its horizontal bar, the monogram denoting Christ was deprived of its initial letter and hence its meaning, and the crosses surmounting the imperial crowns

were removed. The intention behind these changes was clearly to de-Christianize the coins, but to do so as unobtrusively as possible.

retaining their Byzantine look.

By degrees the Muslims embarked on bolder innovations, replac- ing for example the Byzantine ruler with orb and sceptre by .1

recognizably Arab figure, bearded, wearing the traditional Bedouin

headdress, and clasping a sword – a pose evocative ot’ the caliph

delivering, as his office demanded, the khutba or bidding prayer at

the congregational mosque on Fridays. In other experimental issues

this ‘standing caliph’ was replaced by other images with ,111 even

more unmistakably Islamic religious significance, such .is the caliph

j 1



8 (below) Imagery of the afterlife. Dome of the Rock: polychrome and mother-of-pearl mosaic. Motifs of

secondary importance in Byzantine tradition are now greatly enlarged and promoted to centre stage. Jewelled vases and celestial plants glorify the Rock and create an other-worldly ambience, employing new symbols of power and Paradise.





[slam triumphant I )ome of the Rock. Jerusalem, completed in 691. limit on the platform formerly occupied bv Solomon’s Temple, it reinterprets .1

standard type ofByzantine centralized

commemorative building intended for pilgrimage Its central mi k evokes Arabian litholatrv. associations of the

Creation and the Last Judgment, and

Muhammad’s Night fourney t<> Heaven rilework replaced the external mosaics

between 1 s^s and 1

me «>t the- l<o. k three cutaway viev« revealin

underlying geometry <>i the plan



flanked by attendants and with his hands raised in prayer, or a mihrab

enclosing the Prophet’s lance. Finally, in a far-reaching currency

reform which extended to most of the Islamic world and was carried

out between 695 and 697, all figural images were expunged, to be

replaced by the quintessential Islamic icon: Qur’anic epigraphy. In

these coins, which were minted in their millions, inter-confessional

rivalries took on a new and explicit edge. A direct attack on the Christian doctrines of the Incarnation and the Trinity can be seen in

the words emblazoned on the field of these coins: ‘There is no god but God; Muhammad is the Messenger of God. He has no associate; He does not beget, nor was He begotten.’ Seldom in world history has the propaganda potential of coinage been so fully exploited.

Despite the unquestioned significance of Umayyad coins as histor- ical documents, and the curiosity value of the minor arts datable to this period, there can be no doubt that the intrinsic nature of

Umayyad art can be gauged only by means of the architecture of the time. Sadly, some of the finest Umayyad mosques have vanished, like the Mosque of the Prophet at Medina, constructed in 707 on the site of his house (see p. 15) as part of a far-sighted programme of major mosques at sites of key importance: in its time this must have rivalled

the very finest of Umayyad religious monuments. Others have been totally rebuilt, like the Great Mosque of Aleppo, built by the caliph Sulaiman, or the Aqsa mosque in Jerusalem, possibly founded by al- Walid I. Nevertheless, two supreme masterpieces of religious archi-

tecture do survive. They show that, while early Islamic art was still in the thrall of the Byzantine and classical heritage, the Muslims were

already developing their own visual language and were well able to use inherited forms for their own ends. These buildings confidently proclaimed that the new faith had come to stay in the formerly Christian strongholds of the Near East.

10 The Dome of the Rock in Jerusalem was completed after a turbu- lent decade in which the Umayyads briefly lost control of the Hijaz, and with it the holy cities of Mecca and Medina, and survived further serious challenges from religious opposition groups. This

particular historical background has prompted some scholars to explain it as a victory monument and even as a place of worldwide Muslim pilgrimage to supplement, if not to supplant, Mecca itself. Yet its site and its form also suggest other interpretations. It stood on what was incontestably the prime plot of real estate in all Jerusalem –

the vast high platform on which Solomon’s Temple had rested, shunned by Jew and Christian alike since the destruction of that




Temple by Titus in AD 70. It marked an enigmatic outcrop of rock traditionally associated with the Creation itself and with the near-

sacrifice of Isaac by Abraham, the prelude to Clod’s covenant with man. Later Muslim belief identified this as the place of the Prophet’s Ascent to Seven Heavens (his mi’raj) in the course of his miraculous

Night Journey. In form the building is a domed octagon with a double ambulatory encircling the rock; 111 essence, then, .i centralized

structure ot’ a type long familiar in Roman mausolea and Christian martyria. The choice ot’ form probably stems from a desire to upstage the nearby domed church of the Holy Sepulchre, perhaps the most sacred shrine of Christianity, also built over a rock; the

diameters of the two domes differ by only a centimetre. Nevertheless, the earlier building was confined within the urban

fabric ofJerusalem, while the Dome of the Rock enjoyed, as it still does, a matchlessly uncluttered and highly visible site. In much the same way, the qumtessentiallv Byzantine medium of wall mosaic was used to decorate the interior and exterior ot the Dome of the Rock on a scale unparalleled in any surviving earlier Byzantine church.

The pervasive motifs of jewelled plants, trees and chalices have been interpreted as references to Muslim victory, Solomon’s Temple and Paradise itself, while the earliest epigraphic programme 111 Islamic architecture comprises lengthy Qur’anic quotations exhorting

believers and attacking – as did contemporary coins – such Christian

doctrines as the Trinity and the Incarnation.

The Great Mosque ot Damascus (705-1 s) offers the natural 11-14 pendant to this great building – again. .1 royal foundation occupying

the most public and hallowed site in its city. Here too its topograph

ical dominance has clear political overtones. It too is ot’ impressive size and splendour, and uses Qur’anic inscriptions (now unfortu-

nately lost) for proselytizing purposes. The caliph al-Walid I pur- chased the entire site, comprising the walled enclosure of the temple

of Jupiter Damascenus and the Christian church of St. John the

Baptist within it. and forthwith demolished that church and ever)

other structure within the walls. The revered model of the Prophet’s house m Medina – the primordial mosque ot’ Islam as refined b\ slightly later mosques built 111 the garrison cities of Iraq and else-

where, seems to have inspired much of what now followed. An open courtyard tilled most of” the rectangle created by this wholesale

demolition, with the covered sanctuary of the mosque on its long

south side. Yet this arrangement is not entnvK Muslim. It boldk

recast the standard components ot .1 typical Christian basilica to



1 1 Great Mosque ofDamascus:

isometric view. The superstructure ot the minarets is Mamluk and later. An original entrance to the north is now blocked.

\2 Christian architecture [slamicized. Great Mosque of Damascus, completed in ^ i s by the caliph al-Wahd I In the foreground, the treasury I he

sanctuary facade and dome were rebuilt after the catastrophic fire of’ in<>?; the original arcade would have been much lighter.

SI i




si^ lijj mm it}W\ y


i. 3 5S”‘ 3 “*–

•:•.’ «>i” C5—..-9-— •.

a – n icMain

13, 14 (ofoiv) The new Rome- Great Mosque ofDamascus, mosaic on west wall. This fantasy architecture uses – no doubt for political purposes – a Roman, not a Byzantine, vocabulary, {beloir) A world transfigured. Drawing ot the landscape panorama in the Great Mosque, showing its hill context.



secure a new lateral emphasis in keeping with the needs of Islamic worship. The three, aisles remained, but the direction of prayer ran at right angles across them and was marked in elevation by a towering domed gable which clove through the pitched roof to form a central transept. Its facade was a free variation on the standard west front of Syrian churches. This T-shaped partition of the sanctuary was des-

tined to have a long posterity in the mosques of the western Islamic

world (see p. 186).

Carved marble window grilles with elaborate geometrical patterns loosely inspired by late antique wall mosaics presage the enduring

geometric bias of much Islamic ornament. Quartered marble, so cut that the veining oi the stone continues from one slab to the next,

formed dados in typical Byzantine fashion. Above them unfolded

the glory of the mosque: hundreds of square metres of wall mosaic in

the predominantly green and gold tonality already encountered in

the Dome of the Rock mosaics. The caliph seems to have obtained artists and materials from Byzantium itself for this great work; cer-

tainly the technical standard of the mosaics is beyond reproach.

Along the inner wall ot the ancient enclosure, above a continuous

golden vine-scroll (now lost) winch functioned like a religious cordon

sanitaire tor the entire mosque, is unveiled .1 vast panoramic land-

scape. Along the banks ot a river regularly punctuated by gigantic trees rises a fantasy architecture ofvillages and palaces m endless pro- fusion. The link with Roman wall-paintings of the type found at Pompeii is unmistakable; but here the idea is put to new and unex-

pected use, for it strikes the dominant note in a huge monument of religious architecture. Some of these multi-storey structures also evoke South Arabian vernacular architecture. Human and animal figures are conspicuously absent, indicating – as at the Dome of the Rock – that a distaste for hgural ornament m a religious context had already taken root. Despite the obvious success of these mosaics as

pure decoration, main meanings have been proposed for them: topographical references to Damascus or to Syria in general, wish-

tulhlling depictions of a world at peace under Islamic sway, or evoca-

tions of Paradise itself. Perhaps Mich ambiguity is intentional.

Clearly, these two buildings belong together as a considered Muslim response to the splendours of classical and Christian archi- tecture around them, and an assertion of the power and presence of the new faith. The same message ismics from the much more numer- ous desert establishments founded under royal patronage. The Umayyad princes – chafing under the moral and physical restraints of



city life, apprehensive o\ the plague which recurrently menaced those cities, and perhaps atavisticaHy drawn to desert life – moved restlessly from one of these desert residences to another. With a few- exceptions – among them a khan or travellers’ lodging place at Qasr al-Hair al-Gharbi and perhaps one at Qasr al-Hair al-Sharqi too, and

a miniature city at c

Anjar laid out on a Roman grid plan – these foundations fall into a well-defined category. Here, too, pre-Islamic

forms are pressed into service.

Yet much more than mere imitation is involved. Where the Dome of the Rock sedulously copied Christian martyria and the Damascus mosque reworked the Christian basilica, the desert residences radi- cally refashioned inherited forms. They combined two familiar building types whose origins are Roman, not Byzantine – and this association with the remoter but more prestigious imperium (rather than its still unconquered successor) is surely significant. The two building types in question – the villa mstica and the frontier fort – are

intrinsically unrelated and are thus quite naturally segregated in their

parent culture. This unawaited combination springs from the need.

peculiar to this group o\ patrons, to integrate two essentially dis-

similar functions. These residences served at once as the nerve centre

of a working agricultural estate, in which the caliph was – so to speak – lord of the manor, and as outward symbols o\ conspicuous

consumption and political power. The shell of the Roman frontier



1 5 Roman authority. Qasr al-Hair al-Sharqi, main gate of caravansarai or palace(?), c. 728. The projecting towers, arcuated lintel and alternation of stone and brick are all Roman; the central machicolation copies local models.



16 Rustic idyll. Qusair ‘Amra, hunting lodge and bathing establishment, early Nth

century. Vault fresco with human ,\nd animal figures in a lozenge pattern adapted from classical floor mosaics.

fort, complete with salient gateway, corner towers, battlements, and

even its favoured Roman dimensions, was retained. But now it was shorn of virtually all us functioning defensive devices, and con-

tained both luxury royal apartments and service quarters grouped

in two stories around a central courtyard. Qasr al-Hair al-Gharbi,

Usais and Khirbat al-Minya all attest this type.

A rather different kind of establishment is represented by a pair of 16 sites north-east of

£ Amman in the Jordanian desert – Qusair c

Amra and Hammam al-Sarakh. These also make free with a classical build- ing type — in this case, the bath. In approved Roman fashion, cold, warm and hot rooms, all variously vaulted, succeed each other. The novelty lies in adding a ceremonial vaulted hall, complete with royal

niche, to this humble ensemble and thereby exalting it to a new dignity. Qusair


Amra is especially notable for its matchless series of wall-paintings, the most extensive sequence of true frescoes to have

survived from the late antique and early medieval world. Shot

through with techniques and iconographical allusions of classical

origin, they celebrate the pleasures of wine, women and song – to




i” Concert hall? Khirbat al-Matjar. bath

hall. c. 740. The 21 vaulted spaces gave this chamber a magnificent acoustic; its

patron. al-Walid II. loved to hear

performances ofpoetry and music. 1 he

cross-in-squarc format and the vaulting

system are taken directly from Byzantine

church architecture; some parody may be intended.

is The Umayyad world view Khirbat al-Matjar. mosaic in diwan (retiring

room) ofpalace, (“. ~ao. A hunting scene linked to late antique rloor mosaics b\

theme and technique alike is

transformed into a powerful allegory oi

a world divided between Muslim and

infidel. Here presumably sat the caliph,

dispensing justice: reward on his right,

punishment on his left.

^”I “‘T l’i !<” , ‘ ” • —>

r —1




say nothing of the dance, the bath and the hunt – in a remarkably uninhibited idiom. Among several images in a more serious vein, some of them with Solomonic echoes as at Khirbat al-Mafjar (see below), a scene of six kings in submissive pose, identified by inscrip-

tions as the monarchs of the earth, is especially notable. It symbolizes

the entry of the Umayyads into the exclusive club of world leaders, and implies the dominant role of their dynasty in that club. The epi- curean lifestyle conjured up by the main body of frescoes has to be seen within the context of this overt bid for imperial status. Thus political concerns infiltrate even the carefree atmosphere of this

remote hunting lodge, to which the anonymous prince occasionally repaired for a few days of recreation – there was no provision for

him to live at this site permanently. At the very end of the Umayyad period, in response to the

increasingly extravagant ambitions of the playboy caliph al-Walid II,

17, 18 greatly enlarged multi-functional palaces were built. Khirbat al-

Mafjar (unfinished; before 743) is a five variation on the loosely

planned agglomeration of discrete units found m the Roman and Byzantine pal.ices of Tivoli, Piazza Armerina and Constantinople. Here, in the fertile valle\ of Jericho, and linked by little more than their proximity within an enclosing wall, are disposed a palace, a

mosque, an underground bath with shower, a courtyard with an

imposing central tholos (a circular colonnaded structure) over a foun-

tain, and finally the jewel of the site a huge domed and vaulted bath hall, a precocious forerunner of the Byzantine cross-in-square

church. A peerless array of thirty-nine adjoining panels together create the largest single floor mosaic to survive from the medieval or

indeed the ancient world, and provide a fitting match tor the spatial subtleties ot the elevation. Other amenities include a bathing pool, a plunge bath which held wine, a luxurious royal retiring-room

perhaps used for private audiences, for banqueting or as a tribunal,

and finally a splendidly appointed latrine designed to accommodate some thirty-three visitors at a time, Fresco and tempera paintings and, above all. stucco carving ot unexampled vigour and resource complemented the splendours ot” the architecture and floor mosaic. The sculptures of athletes and serving girls in particular seem to epitomize the joie de vivre which the entire establishment exudes.

19, 20 Mshatta is altogether more sober, not to say gloomy Its size –

144 111 (472 ft) per side – is unprecedented among Umayyad palaces and greatly accentuates its sombre, dominating impact. Though it was never finished, enough survives to reveal the basic principle of its




iy Totalitarian architecture.

Ground plan of the palace of

Mshatta, Jordan, c. 744. The side

tracts are a speculative


20 A petrified textile. Filigree ornament is at odds with the

fortified air of the Mshatta palace

facade. Solar rosettes – a Sasanian

theme – stand proud o\\\ thicket

of classical vine-scroll ornament

compartmentalized by a zigzag

moulding adapted from Christian

Syrian architecture.



layout — a sequential subdivision into three parts on an ever-dimin- ishing scale. An iron logic governs the working out of this scheme. While the caliph’s own quarters, at the far end of the central tract, were no doubt as lavishly appointed as their counterparts in other Umayyad palaces, they are not enough to explain the overwhelming scale of the ensemble; indeed, they are sufficiently small to underline

the fact that this was no mere pleasure palace. The key to the build- ing, therefore, must lie in the side tracts, which were scarcely begun

when work on the whole complex was abruptly stopped. Their huge size suggests that Mshatta, unlike the other Umayyad residences, was intended to accommodate large numbers of people – perhaps the entire Umayyad court complete with administration and bodyguard, or even pilgrims returning from the hajj, the pilgrimage to Mecca,

though this is less likely since it would happen only once a year. If Mshatta really was a palace city it would be the natural precursor to the Round City of Baghdad, built barely a generation later (see pp. 40—1). Whatever its function, there can be no doubt that Mshatta

draws inspiration from the tradition which produced Diocletian’s

palace at Split, itself no villa but the apotheosis of the castrum or

Roman military camp. Once again, then, the source is Roman rather than Byzantine. Yet Mshatta is no mere copy. Its tightly regi-

mented square design is subtly orchestrated to assert the absolute power of the monarch; the language of military architecture is made to serve the ends of political propaganda. Not even the celebrated carved facade which extends along the outer face of the central or

royal tract, and that tract only, can mask this grim political message.

What conclusions as to the nature of Umayyad art can be drawn from the material surveyed in this chapter- Three consistent

characteristics can be isolated: it is eclectic, experimental and propa-

gandist. The eclecticism is easily explained. The fact that Umayyad art developed in Syria meant that it was open to the influence not only of the local school of late antique art but also to the art of con-

temporary metropolitan Byzantium. Coptic Egypt and Armenia, and

of course imperial Rome, whose monuments were ubiquitous. Borrowings from the East – Mesopotamia. Sasanian Iran, Central Asia, even India – waxed as classical influences waned in response to the increasingly definitive alignment of the Umayyad state towards its eastern territories. Given the relatively primitive stage of artistic

expression which characterized much of the Arabian peninsula in pre-Islamic times, there was no question of the Umayyads importing their own ready-made indigenous Arabian art into Syria. Thus they




i\ Conspicuous consumption. al-Hair al-Gharbi, detail of floor fresco,

Regimented and abstracted floral rosettes frame three scenes: a beribboned prince hunting gazelles, using stirrups and a compound bow; a flautist and lutanist; and a groom in a game park (not shown). All this reflects Sasanian roek reliefs and silverware, with their iconography ofpleasure, and the increasingly Eastern orientation ofUmayyad art

had perforce to adopt the initially alien styles of the people they had

conquered. Their practice of conscripting labour from provinces

outside Syria ensured the meeting of widely divergent styles.

This helps to account tor the second hallmark of L’inawad an

its experimental nature. Virtually limitless funds were set aside for projects; and the speed with which they were com pleted shows large teams of workmen laboured side b\ side. Naturally they learned horn, and competed with, each Other. It i^



thus scarcely surprising that, in the heady atmosphere created by a

continuous building spree, and in response to the urgings of patrons

who delighted in all-over decoration, the sense of restraint integral to classical art and its descendants was soon thrown off. Experiment

became the watchword. It has its serious side, as shown m the austere geometric wall-paintings of Hisn Maslama. an Umayyad residence and settlement on the Euphrates in Syria. But in general one is

struck by the infectious gusto of Umayyad decorative art. especially its figural stucco and painting, where the effect is heightened by bold, even garish, colours. Unshackled by convention, open-

minded, endlessly inventive, artists delighted to turn old ideas to new account, equally ready to trivialize important motifs by dwarfing

them and to inflate essentially minor themes so as to lend them an unexpected significance. Umayyad artists were far less inhibited than their contemporary counterparts elsewhere in the Mediterranean

world. Hence they freely combined themes and media which tradi- tion had hitherto kept apart: at Mshatta, for example, brick vaults o\

Sasanian type are found a tew teet away from a classically-inspired

triple-arched entrance in cut stone. Transpositions are equally

common: cornice designs are used for plinths, epigraphy overruns both capital and shaft o\~ a column and patterns normally created by quartered marble are imitated in plaster. In this high-spirited and

often vulgar art. parody is never far away.

Yet alongside this robustness, this often wayward originality,

Umayyad art consistently strikes a more serious note. Virtually all the significant buildings to survive were the result of royal patronage, and

their political and proclamatory dimension cannot be ignored.

Sometimes, as in the references to Paradise in the lost inscriptions of

the Damascus mosaics, or in the frontal attacks on Christianity in the

inscriptions of the Dome ot the Rock and later Umayyad coinage, the message is religious. More often it is political, asserting – as in the ground plan of Mshatta – the lonely pre-eminence of the caliph, or as

21 in the floor frescoes ofQasr al-Hair al-Gharbi – Umayyad dominance over east and west alike. The apse mosaic m the diwan at Khirbat al- Mafjar goes further still in its unmistakable warning of the sudden death

which awaits the enemies of Islam. It is peculiarly fitting in this context that it should be Umayyad Syria, not Rome or Byzantium, that can claim the most extensive programme ot wall mosaics and the largest single floor mosaic to survive from ancient or medieval times. From c

Abd al-Malik onwards, the masters of the new Arab imperium needed no instruction in the prestige value ofsuch glamorous decoration.





4JL WUU %&jz4tw^$tm^d

W &’M \*J rfix^^’JlF ^

^Aj5^Ti 22 Types of Islamic writing (top to bottom): simple Kufic; foliated Kufic; floriated Kufic naskhi; thulth; and nasta’liq.

The historical and geographical sorting of Umayyad art made it inevitable that some of the directions it took turned out to be dead ends. Such classical or Byzantine borrowings as figural sculpture and

wall mosaic, for example, struck few chords in later Islamic crafts-

men. Yet it was the Umayyad period which integrated the classical tradition into Islamic art. which devised some of the basic types of mosque and palace destined to recur repeatedly in later generations, which established the sovereign importance of applied ornament

geometric, rloral and epigraphic in Islamic art, and finally which

showed that a distinctive new style could be welded together from the most disparate elements. In so doing it moulded the future development of Islamic art.




The Abbasids

The Umayyad ruling class had been a tiny Arab minority maintained in power only by its military strength and riven internally by reli-

gious and tribal disputes which hastened its downfall. Tolerance of

other religions and dependence on mawali, non-Arabs who had turned Muslim, were therefore political necessities. Victimized by

illegal taxation, reduced status in the army and the racialist scorn of the Arabs, the mawali manifested their social and economic griev-

ances by participating in a series of uprisings that m 749 culminated in a brilliantly orchestrated revolution that toppled Umayyad power and championed the cause of those descended from the Prophet’s uncle, al-


Abbas. The new ‘Abbasid dynasty vaunted these blood links with Muhammad and claimed to usher in the true Islam based on universal brotherhood irrespective of race.

Politically the change of dynasty marked the eclipse of Syria and a

consequent weakening of Creek Influence in the burgeoning Islamic

culture. It also signalled the end of purely Arab dominion. The foundation ofa new capital. Baghdad, at the eastern extremity of the

Arab-speaking world, epitomized this process. Its site near two major

rivers suitable for sea-going traffic – the Tigris and Euphrates – made Baghdad a much greater mart than 1 )amascus had ever been, and its huge volume o\ trade opened it to very diverse influences, from China to black Africa. Such trade benefited from the adoption of Arabic as a lingua franca throughout the empire. Nearby, there still

stood the palace of Ctesiphon, the Sasanian (ancient Persian) capital

whose legendary splendours were now arrogated to Baghdad. Persian costume became fashionable at the ‘Abbasid court, the Persian New Year was celebrated and Baghdad became an intellectual centre where the philosophical and scientific heritage of the ancient world was to be translated into Arabic, the prime language of culture as of

religion, and thence transmitted via Muslim Spain throughout Europe. Such features of Sasanian government as the court execu- tioner, the intelligence service and the formal periodic review of the




army were now introduced. The new Persianized administrative system hinged on the vizier, a post which was often hereditary and gradually came to erode the caliph’s power. But in the first century of ‘Abbasid rule that power was absolute, as the chilling anecdotes of contemporary chronicles testify. To the Western world, the figure of Harun al-Rashid – who sent his contemporary Charlemagne an ele- phant – has always symbolized the oriental potentate, and it is the

golden prime of eighth-century Baghdad that is celebrated in the Arabian Nights. There can be no doubt of the immense cultural superiority of the Muslim East over western Europe at this time. Court life attained an unequalled peak of sophistication and luxury m manners, costume, food and entertainment.

This gilded world was underpinned by a complex financial machine to which capital investment, liquidity and long-term credits

were familiar concepts. As late as the eleventh century a Saljuq vizier

could pay a boatman on the Oxus with a draft cashable in Damascus. Wars of conquest had now ceased, along with their attendant booty, but the resultant Pax Islamica allowed the collection of revenue and

the expansion of trade to proceed smoothly. Perhaps the major dis-

tinguishing feature of the early ‘Abbasid empire was thus the

immense wealth that it commanded. Btit this idyll was short-lived. Squabbles over the succession pinpointed much deeper rifts, for example between Arab and Persian, and between the various reli-

gious groupings. Gradually the extremities of the empire – in Spam,

North Africa, Central Asia and Afghanistan – gained autonomy. Iran

in particular saw a blossoming of national sentiment which found

expression in literary controversies with the Arabs, in heterodox reli-

gious movements and – under the Samanid dynasty in particular

(8 1 9— 1005) – in a revival of pre-Islamic Persian culture. Meanwhile,

m Baghdad the caliphs* increasing reliance on slave troops of Darkish stock caused so much local unrest that in 836 they moved then- capital northwards to Samarra. a move which led to their eventual domination by these Praetorian guards. This situation was formalized

m 945 when the Persian Buyid dynasty, whose Shi ite rulers func- tioned as mayors of the palace, dealt cahphal prestige a catastrophic

blow by assuming direct control of the state. Nevertheless, a cos-

mopolitan Islamic civilization had been made possible by a basic unity oflanguage, faith and religious institutions which exists in large

measure to this day, transcending ethnicity and diverse political

systems. It was only after 94s that the political divisions of the

Islamic world between east and west began to take final shape.




The shift in the centre of gravity- from Damascus to Baghdad involved not merely a geographical adjustment of five hundred miles.

It had potent repercussions in politics, culture and art. Baghdad became, in a way that Damascus had not, an Islamic Rome. It absorbed ideas, artefacts, and influences from the East – from the Iranian world, India, China and the Eurasian steppe, and then

exported them, transformed, throughout the Islamic world, stamped

with its own unique cachet and glamour. Nine-bay mosques in Afghanistan and Spain. Baghdadi textiles laboriously copied in

Andalusia, even down to the inscription identifying the piece as ‘made in Baghdad’, Iraqi stucco forms in Egypt and Central Asia –

all attest the unchallenged cultural dominance of Baghdad. The cumulative gravitational pull exerted by the eastern territories broke

the grip of Mediterranean culture, and specifically ot Graeco-

Roman classicism and its Byzantine Christian descendant, on Islamic art. Classical forms can still be dimly discerned on occasion – the

triumphal arch underlies the portals of c

Abbasid palaces, and all

three styles of Samarran stucco are foreshadowed in early Byzantine

art — but they have undergone a sea-change. New contexts and new functions transform them.

21 The caliph as cosmocrator. Round City of Baghdad. 762: reconstruction drawing. The 9th-century historian al-Ya



calls Iraq ‘the navel of the earth’

and Baghdad ‘the centre of Iraq’; at its heart was the

caliph’s palace.

In architecture, the process of change is exemplified in the Round City of Baghdad, founded in 762. This concentric circular design

was probably derived from such Sasanian models as Firuzabad,

Darabjird and Merv. Housing for the citizens occupied the outer perimeter while the caliph’s palace, oriented to the four points of the






24 From villa to palace-city. Fortified residence ofUkhaidir, Iraq, c. — 5 6, probably built by the governor ofKufa,


Isa b. Musa. Desert now surrounds it, but extensive traces of cultivation explain its name of’the little green one”. It betrays a typically


Abbasid obsession with security and ceremonial; its design looks

both to Syrian and Iranian traditions tor inspiration.

compass and dwarfing the Friday mosque beside it – Caesar took precedence to God here – was located at the dc.xd centre of the city and girdled by a Largely empty precinct. This powerful symbol of cosmic dominion and royal absolutism owed little to the Graeco- Roman world but had a long pedigree in the ancient Near East. All this splendour has left not a wrack behind.

For surviving ‘Abbasid architecture in Iraq one must turn to the

palace ofUkhaidir, generally dated c. 775-6. Flamboyandy isolated, it

evokes in equal measure the despotic and the pleasure loving charac-

ter of” the dynasty. Despite the palace’s gigantic size (175 • 169 m,

574 • 554 ft . its living quarters are cramped, therein perpetuating

Arab tradition: but its luxurious amenities and ceremonial aspect

are strongly Persian in flavour, notably in the interplay oi iwatts and




25, 26, 27 An Islamic aesthetic: all-over decoration. Samaria, stucco wall panels, yth century. Three styles occur contemporaneously, despite differences of conception and technique. The hrst (top) uses a broadly naturalistic classical vocabulary of five-lobed vine Leaves and tendrils arranged in rows or circles. The second (above left) flattens, abstracts and geometricizes this idiom. The third (above right), now moulded, not hand-carved, has a quilted and sculptural quality: its abstract, tactile forms are at once suggestive and

ambivalent. In all three, equal attention is given to precise rendering of detail and to the overall design.

large courtyards, and in the use ot ornamental brickwork, numerous small domes and ingenious vaults. Although this palace embodied such advanced military features as continuous machicolation and a

portcullis, the difficulties of supply and an inherent inefficiency of

design make it hard to imagine how it actually functioned. This concept of the palace-city was perpetuated in the following century

at Samarra, with its numerous sprawling official residences laid out in




ribbon development and galvanized by remorselessly axial planning,

tor example by the use of the familiar three-tract design borrowed from Umayyad palaces (see p. 34). Proportional ratios (often 3:2) and striet axiality hold these structures together. Interior building materi

als – principally mud-brick – are disguised by lavish revetments, and less important wall surface were covered at top speed with stUCCO.

lliese palaces were rendered independent of the outside world by

integrating gardens, domestic housing, military and administrative

quarters and royal compound within a single but vast walled enclosure. It was at Samana that Islamic art came ofage, and from that centre

it spread virtually throughout the entire Muslim world, also mtln

encing local Jewish and Christian art. The new aesthetic is perhaps

best expressed by the wall decoration most fashionable in Samarra in -s-27

palaces and houses alike: polychrome painted stucco, both carved and moulded. Three major styles have been isolated: their

chronological order is disputed, but their roots m the transformation ot classical naturalism and in the two-dimensionality o\ early

Byzantine art is plain. In the first, the surface is divided mto polygo- nal compartments, with borders of pearl roundels. Each compart

ment is filled with vine stems bearing lobed leaves or with fancifully

curved vegetal elements too stylized to equate with any actual pi. int.

In the seeond style, this tendency is accentuated to the point where

recognizably natural tonus disappear. The borders become plain and the compartments themselves more varied. The Chinese motifof yin and yung appears frequently. Finally, in the third style, the decoration

is not painstakingly carved by hand but is rapidly applied by moulds

in a rigorously abstract bevelled style capable (like wallpaper) ot

indefinite extension. The motifs themselves are more loosely and riowmglv arranged, and are more varied – spirals, lobed designs, bottle-shaped forms and other motifs no longer dependent on

vegetal life. This style established itself rapidly and was still full of life

five centuries later. The labour-saving properties ot the moulded bevelled style were ideally suited to the mushroom growth ot Samarra. and the humble mud-bnek of which even the palates were mostly built was cheaply and effectively disguised by this mass

produced decoration. Its abstraction and its even patterning fitted it

for any number of architectural contexts walls, columns, an lies,

window grilles and the ‘Samarran stvle*. especially 111 the bevelled

technique, soon penetrated the so c ailed ‘minor arts’ too

The Samarran palaces show how the secluded, relatively small scale splendours of the classically inspired I in.iv desert residem es




^ £

till! 28 Courtyard ofmosque of Ibn Tulun. Cairo, 876-9 (the foreground dome dates from .296) Its essence is Iraqi: outer enclosure, brick construction, piers with engaged columns, crenellations, stucco ornament and minaret. The pointed arch serves as a leitmotif. The mosque was connected by a broad read to its natron’s palace. ‘

raLed I 71 “V’f’V””‘* ‘””‘”, MoSqUe °fQainwan- Tum^- »* »*« «*. aerial view. The hW^rH it * ” ” *3 ‘””P ™™” 1 ™« J “J ‘he huge axial nnnaree derive



gave way to vast urban palaces, or rather palace-cities, conceived on the Perso-Sasanian model, where massive scale is the dominant

factor. Gigantic scale also characterizes many of the major mosques (Samarra – the largest mosque in the world – and Abu Dulafin Iraq; Ibn Tulun in Egypt: Tunis and Qairawan in western North Africa).

Powerful bastions militarize the mosque, which can even be inter-

preted (as in the case of Qairawan) as an emblem ofjihad. Nor is this the only symbolism at work. Recent research has revealed that

certain columns looted from predominantly Christian buildings and

reused in the Qairawan mosque were colour-coded and so pi. iced that the red and blue columns respectively outlined in simplified

form the ground plans of the Dome of the Rock and the Aqs.i mosque, the major religious sanctuaries of Greater Syria.

Thus in the Tunisian capital worshippers could make a regular sym-

bolic pilgrimage to some of the holiest spots m the Islamic world.


30 The interplay between metropolis and province. Hajji Piyada Mosque. Balkh, Afghanistan, probably yth century. This diminutive nine-bayed multi-domed mosque without .1 courtyard may reflect .1 lost Iraqi prototype: certainly us abundant stucco decoration faithfully mirrors the idiom of Samarra. I he stumpy

piers have Sasanian antecedents.






In some instances, these c

Abbasid mosques are surrounded by further enclosure* which serve to mediate between sacred and profane space. Typically, they were built on the sites of new Islamic towns and thus catered for the whole population – hence their great size, which often brings monotony and repetition in its train. Monumental minarets proclaim the Islamic presence but they often assert an axial and qibla emphasis and serve as reminders of royal

power. The forms of these minarets have a complex heritage; some derive from Graeco-Roman lighthouses, others (e.g. Harran) from Christian campaniles, and yet others from ancient Mesopotamian

ziggurats or temple-towers. They demonstrate both the absorptive and the creative transforming power ot


Abbasid art. These building

projects were huge; the historian al-Ya c

qubi notes that over 100,000

men were recruited for the construction of Baghdad, and the city of Ja’fariya near Samarra. whose rums cover [7 square kilometres (6.5 square miles), was completed in a single year (AD 859). Schemes of this magnitude could only have been organized by a corvee system

(see p. 14). This system had a significant by-product: native craftsmen

learnt the traditions of their imported fellow -workmen. Forms of

varied foreign origin were at first juxtaposed and then, within the

course of one or two generations, blended. This blend was in turn

exported by the new generation throughout the Islamic world.

30 Hence the basic similarity of Style which underlies provincial varia- tions in early Islamic art.

31 The figural iconography of Samarran palaces such as [ausaq al- Khaqani attests the gradual consolidation and refinement of .1 cycle

of princely pleasures – music, banqueting, hunting, wrestling,

dancing and the like. These are to be interpreted not literally but as a

sequence of coded references to a luxurious royal lifestyle that was

summarized by the eleventh-century Persian poet Manuchihri in the rhyming jingle sharab u rabab u kabab – “wine and music and meat’. This cycle was assiduously copied by


Abbasid successor states or

rival polities from Spain (Cordoba and Jam. i< and Sicily (Cappella Palatina, Palermo, see pp. 68-72) to Armenia (the palace chapel at Aght


amar) and Afghanistan (the palaces o\~ Lashkar-i Bazar). It

occurs on marble troughs and ivory boxes, on brass or bronze buckets and ceremonial silks, on the exteriors and interiors of

Christian churches, and ot course in numerous palaces. The figural type popular in these paintings – characterized by pop-eyes, over- large heads, curling love locks, scalloped fringes and minuscule feet —

had an equally wide dissemination.




ii Th<le courtly ethos. Restored wall painting from thejausaq al-Khaqani palace, Samarra,

I he early ‘Abbasid period saw the apogee of wine poetry (khamriyya); many such poems praise the cup-bearer. The kiss-curl, the scalloped fringe and the agitated hem find parallels m Central Asian art possibly brought to Iraq by Turkish slave troops.

The immense financial resources of the early ‘Abbasid empire generated luxury arts galore. Rock crystal workshops flourished in Basra. Gold and silver vessels with figural decoration including hunting scenes and dancing girls .ire described m the Bacchic poetry of the court laureate Abu Nuwas. Surviving wares, mostly m the form of plates, dishes, jugs and ewers – mainly of base metal alloy

such as brass or bronze but sometimes silver, and occasionally even

gold – display .1 somewhat degenerate Sasaman iconography of tabu lous beasts, royal diwan scenes and prmccK hunters. Some important bronze sculptures (serving tor example as aquamaniles or incense

burners; depict birds and beasts of prey Medieval texts mention



22 (left) The princely cycle. Silver-gilt dish, Iran, perhaps yth

century. Details such as the piled-

up cushions, bench throne,

musicians, putto, ribbons, and the

courtier with hands crossed on his chest, his face masked so that his breath does not pollute the royal

presence, derive from Sasanian art

but are already coarsened.

}} (below) The word as benediction. Ikiit cotton cloth with

applied gilt decoration, Yemen, ioth century. The text reads ‘Glory is from God. (?) In the name of God. And the blessing ofGod be upon Muhammad”. Such striped cottons (bitniil) were a Yemeni speciality: the Persian i ith-century

traveller Nasir-i Khusrau wrote

ofSanV that ‘her striped coats, stuffs of silk and embroideries

have the greatest reputation”.

presentation gold medals of prodigious size minted by the Buyids,

but the surviving pieces are much smaller. Their iconography, however, is significant: it includes images of princes seated cross-

legged and entertained by musicians, portrait busts of rulers wearing

crowns of pseudo-Sasanian type, mounted horsemen and the ancient royal motif of the lion bringing down a bull. Some bear Pahlavi



inscriptions and use the ancient Persian title Shahanshah, ‘King of

Kmp’. All this indicates a radical departure from the aniconic norms of Muslim numismatics.

But the art form par excellence was textiles. Byzantine ambassadors j j marvelled at the 38,000 precious hangings displayed to them in a

caliphal palace. Such textiles played a key role in architecture, tor

e word as official

livery. Part of the St Josse silk.

Khurasan, before 961. The inscription wishes ‘glory and

prosperity to the qa’id Abu Marour Bukhtegin, may Chk\ prolong I I favours to him?)’. I he two small

ns crouching at the feet of the

elephants evoke China, the camels

inian ribbons, and the

patron is .1 lurk: a remarkable

mixture of sources

I / ;utr vulgarisation. Slip-

painted bowl from Nishapur, Iran,

man \ilverv. favoured hunting scenes; that

theme is much reduced here, and barely makes sense, thanks to the

• leopard, the fal< on ami the

riptions. |i

nimals that till the

Bui a distant




they were used not only as wall decoration which could be regularly changed and so transform the spaces thus hung, but also to partition

rooms, to curtain off private spaces, and to bedeck key areas like

entrances. They formed a crucial element in public ceremonies and parades. Above all, they were a form of liquidity thanks to their port- ability and their sometimes prodigious cash value. Copious literary

references testify to the hundreds of different centres throughout the

length and breadth of the Islamic world which specialized in given types of textiles and indicate that this was the most prestigious art

form of the time.

Palace and other government-run workshops known as tiraz pro- duced textiles (also called tiraz) bearing laudatory or benedictory

inscriptions with the name of the ruling caliph, making the courtiers who wore them walking advertisements for their monarch – an Islamic form of livery. Other silks were pictorial, like the so-called

34 St Josse silk woven in Khurasan before 960 for the Samanid amir Abu Mansur Bukhtegin. Affronted elephants whose aberrant form betrays Chinese rather than Indian influence take up the field, while

Bactrian camels and cockerels pace the borders, supplemented by a

benedictory Kufic inscription in lapidary style. This silk is typical of

many formal pictorial silks from Sasanian and Islamic Iran, Iraq, Syria and Byzantium which found their way westward and were preserved in church treasuries because they were used to wrap relics. Foremost

among the themes of such Islamic pieces were heraldic images m roundels, among which lions and eagles took pride of place. Moulded cameo glass with relief inscriptions and lustre painting typ- ified the technical advances achieved by Islamic craftsmen. Nearly all

the objects in precious materials such as ebony, ivory and alabaster

described in medieval texts have vanished, but they must be home in mind in reconstructing the ambience of ‘Abbasid art. Thus it is all the more regrettable that the fullest sequence of any imperial c

Abbasid art form should survive in the humblest material of .ill –

35 pottery – which thereby, fault de mieux, takes on a defining role in modern perceptions of ‘Abbasid art. This assuredly leads to a grossly distorted view of what courtly


Abbasid art was really like, yet this

material does provide a paradigm o\ the radical innovation which characterized this period.

Indeed, the ninth century sees the beginning of the long and dis-

tinguished tradition of Muslim ceramics. Strangely enough, there is no teeling ot hesitation 111 these early styles; the technique and decoration are equally assured, and several major varieties of ceramics




are encountered in this first century. This immediate maturity is puz-

zling. It is true that the rather earlier Nabatean painted pottery of the

Levant does have some striking connections with c

Abbasid wares (as

in the use of the ‘peacock’s eye’ motif), and that lead-glazed wares

had already been made in Egypt for a millennium. But the virtual absence of tine Umayyad pottery, together with the fact that glazed pottery – which accounts for most quality medieval ware – though known in ancient Egypt and Parthia, did not achieve the status of a tine art in the ancient world, underlines the lack ofimmediate prece-

dents tor these wares. The earliest Arab pottery, being simply for domestic use. continued this utilitarian bias and was sparsely deco-

rated with simple incised or relief designs.

Under the c

Abbasids, pottery was suddenly promoted to an art

form. Why? The impact of Chinese ceramics seems to have been the galvanizing factor. Ample literary references testify that pottery was imported in quantity from China, both overland through Persia – by the celebrated Silk Road – and by the sea route via India; and imported Chinese wares have been found in nearly all excavations on

Islamic sites. In the early centuries of Islam, Chinese art had a pecu-

liar cachet: Severus ibn al-MuqanV wrote The Chinese are a nation of artists but they have no other merits’, while al-Baihaqi reports

the governor of Khurasan in eastern Iran sent the caliph Harun al- Rashid ‘twenty pieces of Chinese imperial porcelain, the like of

which had never been seen in a caliph’s court before’, together with

two thousand other pieces of porcelain. The latter were no doubt the product of the Chinese export industry; as is usual in China, the

finest pieces are the ones made for home consumption. In the field of ceramics, then, China was held to be supreme. There alone

pottery had been cultivated for many centuries as a fine art. Given the prestige attached to Chinese wares, it would be natural tor the

‘Abbasids to supplement the always insufficient imports o( choice

pottery by establishing a local industry. Hence, perhaps, the sudden

explosion of the ceramics industry in the ninth century. Theological

prohibitions might also have contributed in slight measure, for

various hadiths (sayings of the Prophet) condemn the use of gold and silver vessels. The development of pottery with a sheen imitating precious metals lends some credence to this view. Finally, the advent nt~ [slam led to a much-reduced output in certain well established

media – notably sculpture – which depended on figural motifs.

Perhaps the burgeoning quality-ceramic tradition was an attempt,

conscious or not, to develop an alternative means o! expression for

s i



36 (above) The lore of the stars. Lustre bowl found in Samarra, 9th century. It depicts Cygnus (the swan), a fixed star from the constellations of the northern

hemisphere. The subject-matter implies a cultivated patron. The bird has been transformed into a vegetal design; the busy hatched and squiggly background is typical of lustreware.

37 (lift) Images of light. Great Mosque of Qairawan, c. 836: four of the 139 surviving monochrome and polychrome lustre tiles decorating the mihrab. Literary

evidence indicates that a craftsman

from Baghdad was partially responsible for them; perhaps the remainder were made locally. The mihrab itself and the minbarweve also Baghdadi imports.



this type of subject matter. In all c

Abbasid pottery – whose secular bias requires emphasis – the intention of the potter is clearly to

devise colourful and stimulating surface decoration. 1 le was able to

use hgural motifs, often with a pronounced courtly flavour, as well as

geometric designs, epigraphy and a whole range o\’ vegetal orna-

ment. With this embarras de choix in the field of decoration, it is not surprising that his interest is not focused on technical refinements of

body or glaze or on the shape of the pottery itself”.

Among apparent imitations of Chinese ware, the most common perhaps because it was also the cheapest is the so-called splashed

ware that recalls the mottled decoration of certain Chinese ceramics

of the contemporary T’ang period and also Liao wares (907 [125).

The connection is. however, uncertain because fang mottled wares seem to have been reserved for funerary use. One must therefore reckon with the possibility of an independent invention on the part

of Muslim potters, even though the parallels with Chinese pieees

seem to be too close for coincidence. This lead-glazed ware is also

known as ‘egg and spinach’ after its predominant colours; sometimes it was lightly incised. Chinese celadon, much prized because it was thought to shatter when poisoned food was placed in it, was also widely imitated. Hut Islamic potters were spurred above all to

emulate white Chinese porcelain. Lack o\~ suitable raw material

;^ Poetry on pottery. Glazed lustre

relict” dish from Hira. Iraq, mid-oth

century The Kufk inscription is .1 couplet by Muhammad b. Bashir al-khanji (d. Nam: \)o not

abandon the hope, long though the

quest may endure. That you will find ease of heart, it” but to patience

you ding.’



locally meant that the c

Abbasid potters could not reproduce its

stone-hard body, but they copied this much-admired and coveted monochrome ware by applying an opaque white glaze to ordinary earthenware. Typically, they did not rest content with this, but began to decorate such tin-glazed ware, which was painted and glazed in one firing. In this technique, the colour is absorbed into the glaze and spreads like ink in blotting paper. This running of the glaze

betrays a lack of technical expertise, a deficiency here turned to good account. But the potters were soon able to devise glazes that would not run and so allowed a controlled precision in the application of

paint. Much more complex designs were therefore made possible. The Chinese emphasis on form. body, touch – even the sound a piece made when struck – was replaced, at least in part, by applied decoration not encountered in the prototype. This change of

emphasis lays bare the profoundly different priorities of Muslim taste. The major technical breakthrough in this period is the develop-

ment ofa difficult technique entirely new to ceramics (and to glass) —

that of lustre. A fragment of Egyptian lustred glass datable as early as 772 suggests that the technique may even have been known in Umayyad times. In such pottery, sulphur and metallic oxides are combined with ochre and vinegar and the mixture is painted on to an already glazed vessel. It is then lightly fired in a reducing kiln in

which the metal oxides dimmish to an iridescent metallic sheen on

the surface, reminiscent of the splendour of precious metals. Such

democratization or vulgarization of more expensive art forms and materials became characteristic ot Islamic art. The lustre process was difficult: the vessels were liable to overnre, undernre, or crack during

the second firing. c

Ahbasid lustre has been found as far afield as

Samarqand. Sind, Egypt, Tunisia (where over one hundred lustre tiles

37 decorate the mihrab of the Great Mosque of Qairawan) and Spain; presumably it was usually the pottery that was exported rather than

the craftsmen. The commonest colours are brown and yellow, and at first decoration is extremely simple, consisting mainly of spots,

squares and dashes. But after about 900, animal and human figures with a dotted background enclosing the central design become popular. These figures are often grotesquely, almost fnghteningly,

distorted; often they employ royal or magical themes.

Probably the outstanding achievement of Iranian potters at this

time is the Samanid ware associated with Samarqand and Nishapur,

though similar wares have been found at numerous other sites in

Central Asia. Iran and Afghanistan. The hallmark of this slip-painted




19 Restraint. Pish covered

with white slip and painted

with brown kiitic

inscription; Samarqand,

9th-ioth century. In the

centre, the Chinese tai-kt

motif. The text is in Arabic,

not Persian (the language ot~

dailv life) and reads:

ledge: its taste is

bitter at first, but in the end

sweeter than honey. C^^A health (to the owner].’

ware is its stylish, often virtuoso epigraphy, which unfolds in majestic rhythm around the surface of the dishes. The inscriptions are all in Kufic, and this choice ot~ hand itself imparts a certain formality to

these pieces, implying that they were intended to be displayed as

serious works ot art. The numerous varieties of script encountered often point unambiguously to profession.d calligraphers. lint the urge to decorate is at war with the desire to inform. 1 hese inscrip-

tions share an almost wilful complexity, as if they were meant to

elude ready decipherment. The oracular, gnomic quality of” the aphorisms that they express is thus entirely appropriate, though man\

are of a Shfite tenor. As decorative ensembles, these wares are

remarkable m their appreciation of void space as a positive factoi of the design. Human figures are never found, and birds and animals occur only m severely stylized form. A comparable austerity usually restricts the colour range to cream and dark brown, purple or red.

thereby heightening the starkness of the inscriptions. A clue to the

origin of this decoration may be sought in Chin imics and m contemporary Qur’ans. I hese dishes apparently offer th examples in Islamic art of Arabi< scrip! being used .is die n




element in surface decoration, if one excepts coins, where the epig-

raphy has a mainly utilitarian function. In the stark simplicity of

these inscriptions one may recognize at once a minimalist aesthetic and beauty of a highly intellectual order.

Other contemporary work at Nishapur did not share this cerebral quality. Of outstanding interest is a group of wares distinguished by sprawling, cluttered compositions and violent colour contrasts,

which usually glory in a bright mustard-yellow. Here the designs are simplified almost to the limit of recognition, but they maintain the

directness and vitality of an unsophisticated folk art. Birds, rosettes

and scattered Arabic inscriptions that seem to call down a hail of blessings on the owner are all used as space-fillers. Sometimes the design is a bastard survival of the Sasanian royal iconography of the

banqueting scene or the hunt, and astrological themes are also found.

Such pottery belongs to the so-called ‘ceramic underworld of Islam’, a category represented by wares from numerous provincial centres. Thus, Sari may have been the centre of production for a type ofware closely akin to folk art in the primitive vigour and garish colouring

of its stylized animal drawing. But the commonest category of pro- vincial wares is the sgraffito type, so called after the technique of

incising the design into the body before or after glazing. It is found

40 The word as icon. Qur’an leaf with sura heading m gold; parchment, perhaps yth-ccntur\ Iraq. Rc<.\ dots indicate vowelling, thin black strokes (made with another pen’) diacritical marks. Spacing between individual letters, sequences and whole words can be very wide and therein privilege certain syllables.





widely distributed throughout north-west Iran. Its decoration fre-

quently apes metalwork, even to the use ot the incised lines to

prevent colours from running. A particular class of champlevi ware, in which the white slip is gouged away to form the design, is associated

with the Garrus area in Kurdistan. These varied provincial schools

were independent of influences from the court and from abroad,

though reminiscences of Sasanian iconography were common. Their subject matter favours single figures of animals and monsters or bold

abstract designs.

The other art form which has survived in substantial quantity is calligraphy. It is exercised above all in Qur’ans – the major illustrated secular manuscript of the period is a copy of al-SufYs astronomical

treatise (Bodleian Library, Oxford), dated 1009 and probably pro-

duced in Baghdad, with drawings of constellation images in mixed

Central Asian and Samarran style. Under ‘Abbasid patronage the somewhat haphazard penmanship of the early Hijazi Qur’ans, expressed in irregular letter forms, skewed lines of text, spasmodic illumination and a general indifference to visual effect, was replaced

by a solemn discipline appropriate to holy writ and redolent of epig-

raphy on paper. Horizontal parchment sheets often accommodated 40 no more than four lines of text, thus leading to prodigally expensive

Qur’ans of thirty or even sixty volumes. The script would be so spaced, and with letter forms subject to such extremes of Stylization, as to slow- down recognition ot’ the words themselves: an objective correlative to the awesome enigmas found in the text itself. A supple. flexible system of extension and contraction allowed calligraphers to

balance words on a page with the utmost finesse and thus to create striking visual harmonies. Symmetries and asymmetries, echoes,

41,42 repetitions, and a seemingly endless variety of patterns and rhythms abound. Clearly, therefore, the scribes had ample licence to experi- ment and were not constrained to limit themselves to a text block characterized by regular, even spacing.

A major benchmark of new developments is the Qur’an on paper which, according to its probably reliable colophon, was copied by Ibn al-Bawwab, the most renowned contemporary master, and written in the naskhi (cursive) script which he allegedly invented, in Baghdad in 1000-1 (Chester Beatty Library. Dublin). Its diminutive size cannot mar the well-nigh endlessly varied splendour of its orna- mental palmettes. its frontispieces and finispieces conceived like the leaves of doors and structured with a ponderous, recondite rhythm around a theme of interlaced semicircles, perhaps intended to have




i^^— <m* » ^1 L ,— ^J I -i_J L

jO-i^^JI >0^>ll fllll jOOLJ


4; “In the name of God, the Merciful, the Compassionate’. Almost every sura in the Qur’an begins with this phrase, known as the bismillah from its opening three words. Often displayed by calligraphers, in popular belief it has special power as an amulet. Here it is executed 111 some major Qur’ank hand

from top) early Kufic. square Kufic. eastern Kufic, iluilih: (right,jrom top) naskhi, muhaqqaq, rihani, taliq.

an apotropaic effect. It should be noted that in manuscripts the so-

called ‘Kufic’ types of script (named after the town of Kuta m Iraq) were restricted to Qur’ans, although they could be used for head-

ings, captions and the like in other manuscripts. This style spread, it

seems, throughout the ‘Abbasid dominions with only minor local

variations. It thus typifies the prestige and paramount authority

enjoyed by the art of Baghdad: a fact of life epitomized by the

courtier Ziryab, who imported the lifestyle of the Iraqi capital 111 food, language, clothing and art to far-off Cordoba, the capital of

Umayyad Spam, in the tenth century (see p. [75 . In the


Abbasid period Eastern – including Central Asian, Turkish

and Chinese – motifs, techniques and themes begin to infiltrate

Islamic art. The political and economic background tor this is the shifting of the capital from Damascus to Baghdad, which brought

with it a rush of Iranian ideas; the importation of furkish soldiers

who gradually usurped supreme power; and the rapid growth ol




long-distance trade with lands to the east, both overland and by sea.

Islamic art now largely severed its connections with the classical world, and turned its back on the Mediterranean. In architecture,

Sasanian forms were dominant for city plans, palaces and mausolea.

Baked brick, mud-brick and even stamped earth often replaced

stone. Classical ornament of foliate inspiration became ever more

abstract and this abstraction – which led, among other motifs, to the arabesque in its final form – became the basis of much later Islamic art. Such classical materials as carved stone and mosaic were largely

rejected in favour of stucco, which was to become the decoration par excellence of eastern Islam. The unusually yielding quality of stucco made it an excellent testing ground for new techniques and designs. In certain fields such as Kufic Qur’ans and epigraph ic pottery – and

perhaps also lustreware — the achievements ot the ‘Abbasid period were to remain unrivalled; but still more important was the full elaboration of the thematic cycle of court life begun under the

Umayyads and destined to be eagerly taken up by later Islamic dynasties. Thus from the point of view of materials, techniques and subject matter,


Abbasid art was to offer a much richer quarry for later generations than Umayyad art. It is also to be found over an incomparably wider geographical area. Moreover, it ^till enjoyed the

same advantage of a corvee system which, by making craftsmen mobile, disseminated the latest developments over a wide area. Within the empire, there were no frontiers, a fact which can be explained by a basic unity of faith and political institutions. The divi-

sion of the Islamic world between East and West was not to become definitive until the Saljuq period.




( MAIM 1 R I UK! 1

The Fatimids

From the death ot the Prophet onwards, a body of Muslim opinion held unswervingly – though with main internal divergences of opinion – that supreme power m the Islamic state could be vested solely in a member of the Prophet’s own family. The first and obvious such candidate was Muhammad’s cousin


Ali, who by mar- rying the Prophets daughter Fatima also became his son-in-law. c

All’s claims to the caliphate were pressed by the so-called ‘party of

‘AH’ {shi’at ‘AH – whence the term ShTite), but after his assassination

in 66] the caliphate passed to the Umayyad family. Thereafter, despite frequent and bloody Shfite insurrections (of which the most

significant, historically speaking, was that of the Prophet’s grandson

al-Husam, who was killed at Karbala in 6So). a pattern th.u Listed for centuries was established: no ShTite ruler wielded enough power to

disturb the political status quo. Shi’ite principalities in the Yemen,

the Caspian region and elsewhere were protected but also impris-

oned by their remoteness.

All this changed with the advent of the Fatimids, who took their name (and claimed descent) from the Prophet’s daughter, and who held the belief that the authentic line of imams or rightful rulers had

ceased with the death o\” Isma’il, the seventh Imam. This belief

caused them to be dubbed [sma’ilis. The dynasty was founded by a certain ‘Ubaidallah who proclaimed himself the Chosen One (.//- mahdi) and from obscure beginnings m eastern Algeria took over the central Maghrib within a few years. In 921 he set the seal on his con-

quests by founding a city on the Tunisian coast which he named al-

Mahdiya after himself. His successors consolidated their hold on the

eastern Maghrib before turning their eyes further afield to Egypt.

Eventually they conquered its capital, Fustat, and founded their own capital – al-Qahira. ‘the victorious’ nearby, in 969. In so doing they

hastened the dismemberment of the ‘Abbasid state which had begun

with the loss ofSpain to the Umayyads ot Cordoba in But the Fatimids were interested in more than merely winning



their independence from Baghdad; they sought to supplant the c

Abbasids altogether and to establish a Shi’ite hegemony in the

Islamic world. It was this grandiose aim which differentiated them so

markedly from the numerous other (and minor) successor states

which broke off allegiance to the caliphate of Baghdad in the course

of the ninth and tenth centuries. The Fatimids sought to achieve their goal of pan-Islamic domination under the banner of [sma’ili

Shi’ism not only by military expansion, especially into Syria and

Arabia, but also on the ideological plane. Hence their creation of a

corps of missionaries (da’is) who were sent throughout the orthodox Islamic world to preach their doctrines in secret. These I sma’ili doc-

trines offered hope to the disenfranchised and to political dissidents,

while their mystical and esoteric flavour exerted a widespread appeal.

Not surprisingly, some of the religious zeal of the Fatimids can be detected in their art, especially their architecture, and it has left its

unmistakable imprint in over four thousand tombstones of the ninth

and tenth centuries whose cumulative evidence suggests that Fatimid missionaries were active in the cemetery area of Fustat and that many of their sympathizers and converts were women (see p. 77). The same zeal found expression, but unfortunately only m an imperma- nent form, in the royal ceremonies and processions so characteristic

of Fatimid court life. These celebrated, for example, the tour Fridays

of Ramadan, numerous Shi’ite holidays (especially the birthdays of the Prophet’s family), the New Year, which coincided with the high- water mark of the Nile, the circumcisions of the royal children, and the Breaking of the Dike, which symbolized the beginning of the agricultural year.

From the art-historical point of view the importance of the Fatimids is due both to geography and chronology – for the art of this dynasty forms a bridge in time and space between east and west in the Muslim world, between the pervasive influence of first Umayyad and then


Abbasid art and the rather different art of the

eastern Islamic world which developed in the wake of the Saljuq invasions of the eleventh century. It was the Fatimids who dominated the southern Mediterranean world, with its millennial heritage of

Hellenism, and whose contact with the Christian powers to the north brought fresh ideas into Islamic art. The metropolitan status of Cairo, probably the major Muslim city of the eleventh century, can only have accentuated this internationalism.

Discussion of the evolution and even to some extent the nature of Fatimid art – though not the architecture of the period – has been




bedevilled by the extreme scarcity of datable (not to mention actually dated) objects. Only two ceramics, three rock crystals and two woodcarvings are securely datable. The great exception is the tiraz textile production of the period, but these pieces are principally of

interest for the history of epigraphy Happily the ceiling of the

Cappella Palatina in Palermo, Sicily, which is of immense value .is a guide to Fatimid art. can be dated securely between 1 140 and 1 [50; nevertheless, its paintings contain significant European elements

which have yet to be sufficiently disengaged from their othcrw ise Islamic context. The essential danger in dealing with Fatimid art, therefore, is the absence of dated objects to act as landmarks.

In 106S the Fatimid palace treasury was pillaged by troops whose arrears of pay had driven them to mutiny, an event which brought the contents into the public eye and – even more to the point – the open market. While recent research has raised some doubts about earlier theories which proposed that the dispersal of the objets d’art in the Fatimid treasury triggered a democratization of the minor arts, so

that an affluent bourgeoisie began to adopt (and, by degrees, to

debase) courtly themes, it cannot be denied that the events of 106N

brought into the public domain thousands of objects which had hitherto been kept secluded in the royal collections. Medieval

accounts of the tabled contents of this treasury prove beyond doubt

that the surviving legacy of Fatimid art is a very pale shadow of its

original splendour and multiplicity. But this information has to be

used with care. It so happens that the description by al-Maqrizi is

more detailed and circumstantial than any other medieval Islamic account of objects of virtu in royal collections. That is why it is so frequently quoted. But it should not be interpreted to mean that such collections were confined to the Fatimids. On the contrary, dis- jointed snippets of information indicate that the ‘AbbasuK of

Baghdad and the Umayyads of Spain – to name only two dynasties also amassed staggeringly rich collections of precious works of art.

Similarly, the preference for courtly themes among patrons appar- ently unconnected with the court is by no means confined to

Fatimid art; this tendency is, for example, equally characteristic of

the Saljuq period.

Other aspects of Fatimid art, too, require rc-cv. dilation. The

much-vaunted realism of Fatimid lustreware, tar from being an

innovation of this period, has been shown to have much deeper roots

in the arts of the Copts and even in the Hellenized late antique

world of the eastern Mediterranean than previous scholarship had



recognized. Finally (thanks to some crucial re-attributions), a much larger body of Fatimid metalwork is now available for study than was previously the case. Pride of place must go to the numerous small- scale animal sculptures. They include camels, lions, cats, gazelles, rabbits, ibexes, goats and even parrots – the latter functioning as ornaments for hanging lamps. Many such pieces were cast and were therefore presumably made for a mass market. From the time of the Muslim conquest onwards, a series of inde-

pendent settlements had been built at intervals of about a century

around the site of modern Cairo — Babylon, al- c

Askar, Fustat and al-

Qita’i. Al-Qahira supplanted all of these and became the nucleus first of medieval and then eventually of modern Cairo. Its foundation should be seen, like that of Baghdad, al-Mutawakkiliya, al-Mahdiya

or Madinat al-Zahra, as an expression of renal aspiration and pomp; an action to be expected of a powerful ruler. That same political dominance was reflected in the location of the court within a ten-

44 The Gate of Victories in the City of Victory. Bab al-Futuh. Cairo, [087. One of three Fatimid gates built by three Armenian brothers from Edessa, and incorporating the very latest defensive devices, this fortification belongs to a Ja/iran tradition best illustrated

by the walls of Diyarbakr. It was one ofmany gates m the palace-city wall.



4.s Recreation .is ceremony.

Carved ivory, Egypt,

nth-1 2th century. Perhaps a Fatimid book (.over.

ultimately derived from a

Byzantine five-part ivory

>.’o\cr. The unbroken continuity of these images

aptly suggests the formalized

ritual ofcourt life,

encompassing both business

and pleasure. This is a visual

equivalent of the “delight of

days and nights, without

surcease or change’

mentioned in contemporary textile epigraphy.

gated enclosure walled orT from the rest o\ the city; the palace area

inside was cordonned off by chains and still further distanced by a

huge cleared space constantly patrolled by guards. The two palaces themselves were a byword for ostentatious splendour, with then- gilded marble cloisters, their gardens prinked out with artificial trees

of precious metal on which perched clockwork singing birds, then-

chambers crammed with luxury textiles and above all the great golden filigree screen behind which the caliph sat to enjoy court fes-

tivals. C)t all these splendours only some fragments of woodcarving remain. They depict scenes of revelry and are in the standard Islamic idiom favoured for such subject-matter (see p. 68 ,

Textual sources alone preserve the memory of some of the most spectacular Fatimid treasures, such .is a world map woven of blue silk with every feature identified in gold, silver and silk writing. Mecca and Medina, the ultimate goals o\~ fatimid ambition, were given

special prominence. Made in 964, it allegedly cost 22,000 dinars. I he surviving and similarly coloured Star Mantle of the ( )ttonian

emperor Henry II. made in the early eleventh century in a south

j 1




Italian milieu saturated with Islamic influence, perhaps gives some

idea of this tour deforce. Another vanished masterpiece is the shamsa

(solar rosette) made for the Ka’ba at the order of the caliph al-Mu’izz in 973. It was a sun-shaped object stuffed with powdered musk, con-

taining openwork golden balls which themselves each held fifty pearls the size of doves’ eggs — and the whole object was appropri- ately surrounded by verses from the Surat al-Hajj executed 111 eme- ralds with the interstitial spaces in the writing ‘filled with pearls as

big as could be’. These treasures, which often had a curiosity as well

as a purely monetary value, could serve a political role as diplomatic

gifts and as instruments for the display of royal power. In Byzantium,

too, precious objects were sometimes put to the same uses – for example, the marriage contract of the princess Theophanou, who was despatched from Constantinople in 970 to marry the German emperor, was drawn up in gold lettering on purple-tinted vellum. From much the same time dates the celebrated though now dismembered ‘Blue Qur’an’, probably made c. 1020 in Qairawan –




46, 4~ A religion of the book. Leaves from the ‘Blue Qur’an’; parchment dyed indigo; perhaps Qairawan, i. 1020. Qur’an folios dyed saffron, salmon-pink and pale yellow are also known, and presumably derive ultimately from Byzantine 6th-eentury purple codices. The absolute control of the text block, and the overall symmetry to which individual variations ofscript are subordinated, bespeak the professional scribe.

though Cordoba has also been suggested – whose gold Kufic script unfolds against a background of indigo-dyed vellum. It is the only such Qur’an known. An almost musical sensibility controls the expansions and contractions of the letters, an aesthetic device here

carried to its highest point. Whether these visual rhythms were intended to correspond to the way the text was recited or chanted is

a matter tor future research. Unfortunately, as yet there is no wa\ of

placing the Blue Qur’an securely m its contemporary and probably Fatinnd context, given the near-tot. il absence of Qur ans dated or datable to the Fatinnd period.

In painting, as with so much else m Fatimid art. the ruh literary sources underline how incompletely the tew survivals reflect either

the splendour or the variety ofcontemporary produ< tion I he ^ aliph

al- c

Ainir, for instance, had a belvedere bedecked with paintings o(

6, r



notable poets, each portrait accompanied by quotations from that

poet’s works. And a competition organized by the vizier of the caliph al-Mustansir in the 1050s between a local artist and an Iraqi rival

hinged on their mastery of illusionistic techniques dependent on

strong colour rather than three-dimensional space. Both had to

depict a dancing girl, but the local man (who won) was to show her entering a niche while the Iraqi tried to show her coming out of it.

Al-Maqrizi’s account of this incident suggests that one master relied

on line while the other preferred to exploit blocks of colour – evi-

dence, perhaps, that at least two quite distinctive styles of painting

flourished in Fatimid Egypt. Painted lustre pottery tells the same

story. A similar mastery ot lllusionism is implied by al-Maqrizi’s description of a fountain which decorated a mosque.

None of the paintings which graced the Fatimid palaces has sur- vived, but by a curious freak of chance a wonderfully complete cycle – almost a thousand pictures – of royal images. 111 all probability basi- cally of Fatimid origin, though overlaid at times by influences from

southern and western Europe, is preserved on the ceiling of the

Cappella Palatina in Palermo. The subject matter ranges widely and is somewhat unfocused. It gives the impression o\~ an originally coherent cycle of images rendered incoherent by repetition, and

diluted by the introduction o\ non-courtly and even non-Islamic

subjects. Nevertheless, the core ot the Iconographic programme is clear enough: it is the fullest rendition extant of the cycle of courtly

pursuits that is known in the medieval Islamic world – though the intention may have been to enrich that general theme by others with paradisal and astrological overtones. Wresders, dancers, seated rulers

with and without attendants, grooms carrying game or birds, nobles playing chess or backgammon, exotic animals or birds, hunting and jousting scenes, processions, animal combats, men wrestling with animals, races, drinking, music-making – all are promiscuously inter- mingled. Other scenes plainly do not fit into this courtly category but have apparently zodiacal significance (such as the scene of Aquarius at the well) or are clearly mythological m content, such as the image of the eagle with outspread wings bearing a human figure aloft, or the numerous fantastic beasts: sphinxes, gnrhns, harpies and animals with two bodies but only one face. Such images draw on a millennial Middle Eastern heritage. Scenes which seem to derive

48, 49 from western sources include one of a man grappling with a lion (Samson), an old man holding a bird (Noah?) and a Norman knight engaged in close combat with a Muslim cavalier. Apparently, then,




the artists ransacked the entire repertoire ofimages available to them,

relevant and not so relevant, and used it to cobble together a pro-

gramme of sorts. The inconsistencies already noted thus fall into place as a necessary consequence of a visual cycle forced to expand

tar beyond its normal confines.

The honeycomb vaulting of the ceiling’s muqarnas form ensures a constant variation m the size, shape and angle of the painted surface. But what renders the entire scheme so distinctively Islamic is the way that every inch of the ceiling is painted, to the extent that there is no

toil of plain framing bands or of empty Banking niches or spandrels

for the larger designs. Instead, the images are crammed into the avail- able surfaces irrespective of whether they fit naturally into them or

not. In this respect, the images on the Cappella Palatina roof are

treated in a dramatically different way from those of the normal princely cycle, in which each image is given the same emphasis as

the next. The form ot the ceiling decisively forbids this and imposes on the painters an iconographically uneven handling of their

material. Thus on occasion minor elements may be accorded more space than major ones. Even the narrow bands which separate the stars at the apex of the ceiling are made part of the decorative scheme: they bear elaborate Kufic inscriptions invoking benedictions

– presumably on the Norman king. Roger If for whom the chapel was built between 1 132 and 1 140. Just as at the Armenian church of

Aghfamar two centuries earlier, and in a comparable context of a Christian monarchy trying to come to terms with a much more powerful Muslim neighbour, a palace chapel is the setting tor a

diplomatic visual acknowledgment of the Muslim presence. In both

cases the Muslim cycle is set well above the Christian images at the

outer limit of visibility, and in both cases it is emphatically secular m tone. This suggests, perhaps, that the prestige of Muslim court lite

was such that it became the natural target for emulation even by

non-Muslim princes who were geographically within the orbit of Muslim power. Much the same process can be traced continuously tor centuries m Spam. What of the style of the paintings m Palermo’ Its hallmark is its

unbroken fluency. The smooth line is complemented by a capacity to reduce torm into rounded masses which rlow easily into each

other. Simplification is the key to this way of seeing. Often it results

in mirror symmetry, but zigzag or diagonally emphasized composi

tions are also common, features such as sealloped fringes, large eyes, kiss-CUrls by the ear all point to the influence <>t unpen. il ‘Abbasid




Vt) A ( “ story retold W.ill painting from nave roof, ( !appella I’al.itm.i.

Palermo, Sicily, mid-12th century. The equestrian dragon-slayer probably derives

from Byzantine or Coptic images of Saints

George or rheodore, though Saljuq Turkish

sculpture and even the Shahnama offer further

parallels, sometimes with cosmic or

astrological implications, .is the dragon’s

Coiled tail suggests.

49 {left) The Byzantine saint secularized. Man with beaker, Cappella Palatina. The haggard,

cadaverous figure framed by his halo clearly

derives from a Byzantine source, and the

seated posture is Western. Despite religious

prohibitions, ritual drinking on the Sasanian

Persian model often marked medieval Islamic

court life. The space-filling jug and pot accord with the subject-matter.



50 (right) A pluralistic society. Section of nave roof. Cappella

Palauna. This, the most highly

evolved surviving Islamic wooden

roof of its time, in unmistakably

Muslim despite its

setting – perhaps, together with

the Byzantine mosaics below .1

metaphor for religious tolerance.

Its trilohed. keel-shaped and

tegmental arches echo Fatimid

and Saljuq architecture.

51 (below left) Courtliness

commercialized. Wall painting

from a hath outside Cairo, pre-

1 10 s”. depicting a youth seated in

Islamic pose- (unlike 49) with a

beaker Nearby are depicted

affronted birds and a scarf dancer,

whose damaged state may reflect the edict of al- Hakim in 1 o 1 3 — 1


itch out depictions of

women in public baths.

52 (below right) Soldiers of the

Prophet. Outline o\\\ Fatimid

design (woodblock print?). 1 ith

century The tree with birds derives from textile design. An inscription (half-complete?)

wishes ‘glory and ^oo<.\ fortune to

the leader Abu Man[sur]’- blessings partially repeated on the

I Note the chain mail, two-

horned helmet and accoutrements

of rank hanging from the belt.

. m (5.5 x s.s in).


‘V- ‘ ‘


v. v-V»



art as developed at Samarra. A textbook case of this influence is pro- vided by the seated cross-legged monarch whose pose – even to his tiny feet – seems” to owe much to images ot the Buddha. That a secular king should be depicted by Islamic artists in a Far Eastern

religious manner in a Christian church at the very centre of the Mediterranean highlights in the most telling way the international

quality of medieval Islamic art.

It must be conceded that virtually no echo of this rich cycle of

images has been found in Fatimid Egypt proper, with the minor

exception of a hammam excavated at Fustat, whose painted decora- 51 tion included a seated youth and affronted birds – scenes roughly

comparable with those on the ceiling of the Cappella Palatina. It is

true that many single leaves have been found, notably one now m the British Museum showing a fortress being besieged, and a wood-

52 block print depicting a foot soldier: but for the most part these

works are of coarse quality indeed, some are mere scrawls) and suggest that manuscript painting was still relatively underdeveloped at

this period. But a few of these leaves, of which the largest group is

that m the private Keir Collection in Surrey, England, illustrate not so much an


Abbasid manner as a familiarity with Byzantine art, which during the Macedonian Renaissance (especially m the tenth century) had developed an extraordinarily fresh and natural remter-

pretation ot the techniques ot Graeco-Roman iUusionism by means of modelling And sketchy line.

Too little Fatimid religious architecture is preserved to permit reli- able generalizations about it. and the mosques built b\ the Sunni majority in this period have almost entirely disappeared. But the tew

surviving major monuments ot the dynasty do encourage some interesting speculations, lor example, the absence of minarets in the mosques is noteworthy, the major exception being the I lakim

53 mosque, remarkable also for its elaborately articulated facade, in which broad stairways played a significant part, and even more for the enormous size of its courtyard. Perhaps this absence reflects a dis- taste for the minaret as a culpable innovation, and the converse emphasis on large-scale projecting portals (as at al-Mahdiya) may be linked with a desire to give the call to prayer from them, in a deliber-

ate return to the primordial Islamic practice established by the

Prophet himself. On the other hand, a distinctive genre o\~ brick minarets (possibly reflecting Hijazi prototypes) attained popularity in

Upper Egvpt during this period eg. at Luxor and Esnaj. It is plain enough from the major differences between the few




Fatimid mosques which do survive that it is no Longer possible to identify the nature of the typical mosque of this period. Some strik- ing innovations may be noted. These include the use of towers as corner salients on the facade of the Hakim mosque – a device which gave them a new and crucial articulating function – and the high-

lighting of the area in front of the mihrab by placing domes at the two corners of the qibla wall. Such corner domes serve .is pendants to the more familiar dome over the mihrab (Azhar and Hakim mosques). 54

we) A fortress tor the faith. The mosque ofal-Hakim, Cairo, yyo-1013. develops themes

encountered earlier m the mosques of Samaria and oflbn Tulun. Cairo

nil. 28). The degree of emphasis on the mam facade is. however, new. Oenellations and towers lend it a

military flavour. The triple-arched portal copies palace architecture.

‘ The world’s oldest

university. The much rebuilt And constantly enlarged A/har mosque.

Cairo, founded in 070-2 and

intended also as .1 centre of learning

and of Isma’ili propaganda. Note

such trademarks of Fatimid

architecture as radiating roundels

and niches, keel-shaped arches

and clustered columns.



S5i 5<5

This idea is closely related to the creation of a T-plan in mosque sanctuaries – that is. using .1 central axial nave which leads up to the

mihrab at its tar end. and similarly picking out the transverse aisle

along and immediately in trout of the qibla – a device which was 10

have a long history in the Maghrib. Sikh features show a bold and intriguing readiness to conceive of .1 mosque 111 modular terms. 1 ike the triple entrance to the mosque perhaps .1 feature which the Fatimids derived from palace architecture these ideas suggest that

the use ot the mosque tor royal ceremonies, a custom well docu mented in the literary sources tor the Fatimid period, encouraged .1

fundamental rethinking ot its layout. Hence, presumably, the intro-

duction of features already long familiar in palaces.

To judge b\ surviving public structures, mausolea occupied an

especially honoured place in Fatimid architecture, rhey certainly

survive in much greater abundance than any other building type. In Cairo alone, fourteen funerary structures ot the period survive, .is

against five mosques. Of Still greater significance numerically and structurally – though not from a religious or political viewpoint – are

the fifty-odd mausolea probably of eleventh—twelfth century date .it

Aswan in Upper Egypt. They form an entire necropolis. .1 custom of course already established in Egypt for millennia. Since this city was

a major departure point for pilgrim caravans, it is likely enough the original tenants of” these mausolea were pilgrims, though some could aKo have fallen in holy war. since Aswan was on the Nubian border .md therefore faced infidel lands. Although – or perhaps

because – main of these structures are of mud-brick, they demon-




: -4 *^ •-

53 (above) Tombs for martyrs, i ith-

century necropolis at Aswan. Upper I he Prophet forbade all

ostentation in funerary ceremonies;

nevertheless, mausolea quickly became

>nable. Among the various justifications devised for them was – as

here – the desire to honour ghazis or mujahidin (warriors tor the faith).

$6 [right) Legal or illegal? Interior of

an i ith-century tomb. Aswan. The orthodox ban on mausolea might be

circumvented if the burial spot was

open to the wind and the ram. Hence the popularity of lavishly fenestrated

chambers like this one.

strate a degree of fantasy and playfulness quite unusual in Islamic

mausolea. and this was intensified by their plaster rendering, which

lent an extra sharpness to mouldings and other details. The contrast between flat and curved planes: between the heavy cubic mass of the

lower chamber and the diminutive, almost frolicsome, domed aedicule perched a trifle incongruously upon it; between die pro-

nounced, eccentric rhythms of the transition /ones, with then-

swooping concave volumes, and the austere simplicity of the areas

above and below – all this makes these monuments consistend) appealing, a satisfying exercise in solid geometry. Their interiors have

a curious dimension of surprise: domes spangled with star shaped



openings, the muscular contraction and expansion ot their squinch

zones, and above all the innate sculptural sense expressed in the

strong articulation of the wall surface, scooped out as it were by a

giant hand. Such devices ensure continual variation from one mau-

soleum to the next. Some of the detailing – e.g. the ribbed domes, the busy articulation of the drum, the serried pilasters in the transi-

tion zone — recall Coptic or even mid-Byzantine architecture. The chronology of these buildings is now hard to retrieve because the

dated tombstones which they once contained have been removed,

mainly at the end of the last century, without any record being made of their exact provenance in the necropolis. Aswan provides the earli-

est example of such clusters of mausolea in the medieval Islamic

world but others survive in Cairo, Fez. Raw (near modern Tehran), Samarqand and Delhi.

Some basic types can be identified. They include ‘canopy’ tombs. open-plan with an entrance on each side: the same, but with one

wall closed to house a mihrab; tombs with all tour sides closed and a

single entrance; tombs with added courtyard or sanctuary; and

adjoining tombs which create a continuous vaulted space. All these

types are continually overlaid and obscured by variations which,

however minor in themselves, still change the essential aspect of these buildings. In both Aswan and Cairo, a type of mausoleum not found in Egypt after the Fatimid period makes its appearance: the

mashhad (literally ‘place ofmartyrdom’). It comprises a domed square encompassed on three sides b\ an ambulatory. Important examples of the type are the mashhads ofYahya al-Shabih and Qasim Abu fayyib. Another type of mashhad (such as that ofal-Guyushi in Cairo and another in Aswan) comprises a triple-bayed sanctuary with a dome over the central bay, therein creating a compact cluster ofbuildings. The Fatimid emphasis on mausolea was a major innovation in the

Arab world and it is surely no coincidence that a similar emphasis

can also be detected in contemporary Buyid architecture in Iran.

Shi c

ism seems to be the connecting thread in this development. I he Fatimid caliphs themselves continued the traditional custom of house

burial, though in a dynastic tomb situated in the Eastern Palace later dubbed ‘The Saffron Tomb’ (turbai al-zajaran) because of the custom of anointing it regularly with that perfume. Perhaps this building was intended to rival the ‘Abbasid family tomb .it Samarra, the Qubbat al-Sulaibiya. The great majority of Fatimid tombs in Cairo, however, which date from the later Fatimid period, were erected to commemorate Shfite saints and martyrs, perhaps m an




attempt to create a funerary cult that would support the caliphal family, which of course claimed descent from


Ali. The tombs in the Qarafa cemetery especially became centres for weekend outings, with the faithful spending the night in their vicinity, circumambulat-

ing them and hoping for answers to prayers through the intercession of these personages, and. more generally, to profit from the baraka or spiritual power associated with these tombs. The forms ofsuch m.ui lolea are very much more modest than those of contemporary funer- ary structures in the eastern Islamic world, for the most part they

comprise small square chambers, largely plain not only externally

except for the occasional fluted dome) but also inside, apart from their monumental mihrabs, the latter sometimes disposed in groups of five (Sayyida Ruqayya) or three (Ikhwat Yusuf). Most were dedi cated to members of the Prophet’s family; several honour martyrs for the Shi’ite cause; and the prominence ofwomen among their tenants is quite remarkable – perhaps a reflection of the key role in popular piety played by women and expressed also in the area of the Qarafa cemetery (that centre of female piety) by the unusually high propor-

tion of tombstones honouring Shi c

ite women. Perhaps the most striking feature of these mihrabs is their use of

ribs radiating outwards from a central boss – the very image of light. It was a tenet of Isma’ili belief that the Holy Family existed m the form of light even before the world was created. The Qur’anic inscriptions in these tombs frequently contain references to the sun.

the moon and the stars, and these were interpreted to connote the Shi’ite Holy Family. Any name on the central roundel (significantly termed shamsa in Arabic, from the word shams, ‘sun’) would most

naturally be interpreted as the source of light, and it is exclusively

‘Allah’, ‘Muhammad’ or ‘ c

Ali’. The latter two names are again found at the centre of a star in the dome of the mashhad of al-Guyushi, sur- rounded by a Qur’anic quotation which includes the words ‘It is 1 le

who appointed you viceroys in the earth”. In the highly charged Fatimid context, light had the specific extra connotation of ttOSS (the

explicit statement whereby legitimacy was conferred on an imam,

often interpreted as divine light). Hence the repeated literary associa- tions between, for example. Husain – the son of ‘Ah – and light. Not

for nothing, perhaps, was one of the names for the Hakim mosque

‘The Mosque ofLights’ (jami c al-anwar), jus\ as a contemporary tomb

was known as the Shrine ofl ight. And the Aqinar mosque, whose very name means ‘The Moonlit’, has its facade festooned with radi atmg designs, with

‘ c

Ah’ at the centre of the largest, direcd) over the




L!v.^lMfe^’H^ ,A^!a^’^3>tiliJ

57 A religious and political manifesto. The Aqinar mosque m Cairo was erected in i 12s beside the eastern Fatimid caliphal palace, and was perhaps intended as a court oratory, teaJung institution and tomb tor al- Husain, the Prophet’s grandson. Its inscriptions implore God to give the caliph ‘victory over all Infidels


(presumably the Crusaders and schismatics) and exalt the family of ‘Ah by quoting Qur’an 33:33.

entrance — presumably a reference to the well-known saying that c


is the Gate of the City ofKnowledge; ‘let those who want to acquire knowledge approach it by its proper gate*. The entire facade of the

57 Aqmar mosque functions .is ,1 gigantic qibla articulated by mihrabs large and small – an allusion that would have been tar clearer in Fatimid times, when the standard mihrab had precisely the form of the blind niches on that facade. Indeed, the use of three great mihrabs

on the original facade (though that facade was not orientated to the

qibla, being skewed in relation to the rest o\~ the mosque) seems to

1 have been a deliberate echo of the layout of the contemporary kiswa, the cloth draped over the Ka’ba in Mecca. Pierced window designs, stars and carved representations of lamps are further references to

light (and specifically as it is described in Sura [Qur’anic chapter]

24:35) in this iconographically dense facade, in which – as in other Fatimid monuments – numerical symbolism hinging on the numbers

3, 5 and especially 7 (Isma’ilis being Sevener Shmtes) is also at work.




Fatimid pottery is dominated by lustrewares. Their origin is dis puted; some argue that the earlier lustrewares of Samaria provided the direct inspiration, as indeed similarities of graphic style suggest,

while others seek its source in glass (see p. 54). Hut the key point is

not the source of the technique but the fact that the importance o\~

Fatimid lustrewares extends tar beyond Egypt, since they usher in the rust consistent attempt in the medieval Islamic world to make luxury ceramics say something in visual terms. Not that they are the firs! Islamic wares to carry images – at Samaria and Nishapur. to name only the two most important centres of production, wares in various techniques had already used hgural designs. Hut the number and variety of such themes and motifs expanded dramatically 111 1 atimid

lustreware – although most pieces bore abstract and vegetal designs.

half-palmettes being especially popular. Not surprisingly – perhaps because this was indeed a time o\ experimentation – many o\’ the themes are new to pottery, and some even to Islamic art. A groom leads a giraffe, an elderly man tends a cheetah, men squat on their haunches absorbed in a cockfight, porters carry heavy loads, a lad\ n^

scrutinizes herself in a mirror while a maidservant hovers nearby,

wrestlers strain against each other, men tight with sticks, a C “optic priest swings a censer, falconers or mailed cavaliers ride richly S9»6° caparisoned horses – there is even an image of Christ in true

Byzantine manner. Animals abound: whether commonplace (hares, dogs, ibexes, fish, birds), exotic (lions and elephants) or fantastic

(griffins and harpies). Some scenes involve energetic action, such as ,1 dog biting the leg of another animal. A courtly ambience pervades scenes of banquets, dancers, seated drinkers and lutanists (both male

and female) who serenade them. The various hgural images can indeed be taken at face value as

genre or courtly scenes, but they could equally well (and simultane-

ously) operate on a deeper level as symbols (e.g. of love or tempta-

tion, harmony or fellowship) or as deliberate copies of classical

themes – for of course Egypt had been one of the major centres of

Hellenistic art, and was a country where the sense of an immemorial

p. ist was stronger and more pervasive than elsewhere in the Near

last. Moreover, through its (“optic minority which had preserved

much of the Hellenistic heritage – Egypt was more open to these themes than other areas of the Islamic world. A tew examples will

put these remarks into perspective. Some of the banqueting figures, whose setting could be interpreted .is a garden, may refer to I’. u.ulise

and may derive from classical images of heroization. Several of the



58 Fatimid lustre bowl. uth-i2th century, with cock-fighting scene. Hie tradition of such images whuh typically contrast a bearded older man and a youth, stretched back to ancient Greece .md carried assoria! tions of warlike spirit, virility and homosexuality. Diam. 2 59 The survival of other faiths. Early 12th-century lustre bowl. ^n.-A bv the Muslim potter Sa’d, whuh shows a Coptic priest swinging a censer; a gigantic ankh hieroglyph denotes ‘life’. I Ham. 2 1 .9 < 111 (8.6 in).



60 ‘Good health and

complete joy to the

owner, may he be saved

from evil’ runs the

inscription around this

lustre plate depicting

.i mounted falconer. The iconography is international, but this

piece ofso-called Fustat

ware is inscribed ‘well

made in Egypt’. H. 7 cm (2.8 m». diam. 38.3 cm (is in).

animals depicted on Fatimid lustreware have been interpreted as

lunar, solar or astrological symbols. Some pieces depict dragons or snakes, perhaps a reference to Jawzahar, the eclipse monster regarded

by medieval Muslims as the antagonist of the sun and the moon. A link with the Tree of Life and with water is suggested by trees or fish

on some of these pieces, though apotropaic or mystical overtones could also be present. Sometimes, as m the plate depicting a giraffe being led by a groom, a reference to contemporary royal processions

may well be intended – a medieval equivalent to the Coronation or other souvenir mug.

Polychrome lustrewares are usually decorated with geometric or vegetal motifs, though a few depict birds and – as noted above

dragons or snakes. Some pieces are entirely covered with cross banded inscriptions. The considerable differences in style and execu- tion between the major groups o\~ Fatimid lustreware especial!)

between the monochrome and the polychrome types drive home the tact fatimid Egypt was open to ideas and influences trom all

over the Mediterranean and beyond, and that one should not there

fore expect an integrated style. Fatimid painting tells the same sn>i\.




It is in the Fatimid period above all that potters’ signatures come

into their own. Signed wares are admittedly encountered earlier – for

example, in c

Abbasid blue-and-white pottery, in glazed relief wares

and in yet others found at Madmat al-Zahra in Spain – but they are rare. Over seventy Fatimid pieces are signed, and the names of some

twenty-one potters have been recorded. Chief among them is a certain potter called Muslim, whose name appears on a score of sur-

viving pieces. Many of the signatures include the information that the ceramic was made in Misr (probably meaning Cairo, though pos-

sibly Egypt). Specialization was well advanced in Fatimid times, as

the Geniza records – a cache ot medieval Jewish commercial docu-

ments – show: thus mention is made of the qaddar who produced pots, the kuzi who made spoutless water jugs and the ghaza’iri who made translucent porcellaneous dishes. Perhaps the painters o\’ ceramics were equallv specialized and were distinct from those who produced the pottery shapes themselves. Epigraphic evidence on two

lustre pieces indicates that they were made to order – one for a high official, the other conceivably for the caliph himself- and thus prove

the existence of high patronage tor at least some pottery produced in

this period. Frequently these wares carry benedictory inscriptions;

forty-six of them are inscribed >\m/. ‘good luck”, though this could

also be the name ot the potter. The colours o\ the monochrome lustrewares vary from golden-

yellow through copper to dark brown and the design is usually painted in silhouette, though occasionally it is reserved in white

against a lustre ground. The body is coarse and sandy. Shapes vary considerably, though bowls, dishes and cups predominate. Other wares of the period include glazed relief coloured glazed and incised

types – the latter apparently owing something to ( Ihinese celadons. Indeed many lustre pieces derive their forms directly or indirectly from Chinese prototypes. The floruit of Fatimid lustre seems to have been the late tenth and early eleventh centuries, though production continued in Egypt (and possibly Syria) until the fill of the dynasty in 1 171. It seems possible that at least some Fatimid lustrew.ire specialists thereupon migrated to Iran, where powerful Shi’ite minorities offered a degree o\ security conspicuously absent under

the aggressive orthodoxy of the Ayyubids. The presence ot many Fatimid ceramics walled as ornaments into

the exteriors of churches in Italy (e.g. at Pisa, Rome, Ravenna and Ravallo) and even in France and Greece has been explained as the

result of Crusaders bringing back souvenirs from the East. Whether




this or just plain trade was the source of such wares, they give a valu-

able indication ot the type of luxury pottery produced in the late

Fatimid period. The style of these pieces is so varied that it seems reasonable to look for several quite different sources of inspiration.

Sometimes the trick of painting the face, or the exterior of the piece,

is so like that o\ Iraqi lustre that a close connection is incontestable.

But other pieces have the vivacity and freedom o\’ 1 [ellenistic art .is

developed at Alexandria, and yet other images recall Coptic textiles.

The figures often have a marked illusionism in the way that their movement is suggested by cross-hatching. But the unpredictable play of light which is so integral to the whole effect of lustre is inimical to such nascent realism, since it abstracts form rather than defining it.

Thus hue and colour are at war.

Most specialists attribute the great majority of the i So-odd surviv-

ing medieval Islamic rock crystal objects to Fatimid Egypt and to the

period before c. 1060. The raw material apparently came from the Maghrib and latterly the Red Sea, though the Basra school was sup- plied from Madagascar, Kashmir and the Maldives, and perhaps these

areas also supplied Egypt. Although expertise in hardstone carving

was required to work rock crystal satisfactorily, the artists obviously learned much from wheel-cut carving on glass. Al-Maqrizi cites detailed evewitness accounts describing the rock crystal objects held

6l Frozen light. Rock crystal ewer naming the Fatimid caliph

al-‘Aziz bi’llah (975-06),

venerated as a source ot divine

light. The ibex perched by the run embodies the conceit that

the animal is drinking from the

ewer, a theme rooted in pre-

Islamic religious rituals and

common in metarwork. Rock crystal was believed to be a form

of ice, but also to concentrate the

sun’s rays; the goblets of paradise

are from this material (Qur’an – II [8 cm – . 1 in).



<)2 Cut-price luxury. (ilass

beaker with cut relief ornament,

resembling precious stone;

probably Egypt or Syria, [2th

century. Veste Coburg, Germany;

formerly owned by Martin Luther. Sixteen such objects,

known .is Hedwig glasses (from the tradition that St Hedwig changed water into wine in one

ofthem), survive. I he thi< k

walled, smoky topaz glass bears highly abstract motifs ultimately

ot Samarran origin II. c. 12.7 cm

in the Fatimid treasuries in the 1060s – even their size and market value were noted, and a group ot pieces bearing the name ot the

61 caliph al- c

Aziz were singled out for special attention. I he total

number of rock crystal vessels in this treasury alone was then i,Xoo. The only three inscribed rock crystals bearing the names ot notables were clearly made 111 Egypt; they mention the caliphs al-‘Aziz and al-Zahir, and al-Hakim’s generalissimo, Husain b. Jauhar. The luxury

nature of the craft is emphasized by the technique itself; the ewers,

for example, were made by patiently hollowing out a solid block ot crystal until the walls had been reduced to extreme thinness. The themes of the six closer) related ewers which illustrate Fatimid roek

crystals at their best are uniformly large palmettes (perhaps intended

to represent the Tree o\ Life) flanked by various animals: lions,

ostriches, hawks, moufflons and gazelles. Some of these animals also 62 appear on the heavy monumental ‘Hedwig’ beakers which mimic in

glass the technique of rock crystals. None of these have turned up in the Islamic world, which has siiL^ested to some scholars that they are of European (perhaps South Italian or Sicilian; provenance; but they

could also have been made in Eiwpt or Syria for a European market.





6j From Muslim robe ofhonour to

church treasury. Linen cloth made in Damietta, Egypt, and bearing tapestry

ornament depicting birds, addorsed

iphinxes and other animals. Known as ‘the veil of St Anne”, it has inscriptions

dated 1096-7 mentioning the Fatimid

caliph al-Musta c

li bi’llah and his vizier

al-At’dal. Apt. Vaucluse. France;

probably booty 60m the First Crusade. Detail of copy made in 1850. 310 x 150 cm (122 x 59 in).

) oil


SI it <*j 11

it ^ 1

Their presence in church treasuries is an argument in favour of an Islamic origin, for this is where most of the finest medieval Islamic textiles and rock crystals have been preserved, usually because they

had fallaciously acquired some sacred Christian association. Many rock crystals were used as containers for relics believed to be drops of

Christ’s blood or associated with the Last Supper or the Crucifixion.

Hence the elaborate European ecclesiastical mountings in which so

many of them are set. The surviving Fatimid textiles do not measure up to those known

from literary sources – for example, the hundreds of textiles men- tioned by al-Maqrizi with pictures of rulers and other celebrities, .ill

identified by name and accompanied by some commentary. Most of the surviving pieces are tiraz products whose sole ornament is their

epigraphy. Sometimes they are very close imitations of ‘Abbasid te\

tiles from Baghdad – for example, a tiraz of al-Mu’i// which copies one of the ‘Abbasid caliph al-Mutf. A few are of silk but most are of humbler materials such as linen and cotton. Their epigraphy is far

more varied than that of Fatimid inscriptions in other matt-rials;

some may ape Styles of pen-made writing confined to chain er\ use.





The Saljuqs

Arab dominion of the eastern Islamic world came to an end in 945 when the caliphs were forced to surrender their temporal authority to their army commanders, who belonged to the Persian Buyid family. Henceforth the caliphs preserved only the forms and not the sub-

stance of power. For the next century, political control of this huge

area passed to various dynasties, principally of Persian origin, among which the Buyid family was pre-eminent. One dynasty alone broke this mould: the Ghaznavids, who controlled Afghanistan, much of the Punjab and parts of eastern Iran. They had begun as Turkish military slaves but had assimilated Perso-Islamic ways. This Turkish hegemony became definitive under the Great Saljuqs, whose followers – known as Turcomans – were Turkish nomads and marauders who had recently converted to Islam. The Saljuqs dispossessed the Ghaznavids and Buyids alike, took over Baghdad m loss and thereafter began a fundamental reshaping of the body politic.

For the first time since the seventh century, nomads ruled the Middle East – for the Saljuq lurks expanded westwards to the shores of the Mediterranean, controlling Anatolia, Iraq and parts of Syria as

well as the Iranian world, from obscure pagan beginnings in their Central Asian homeland on the fringes of the Islamic world they had risen in three generations to become the greatest Muslim power of the day. No contemporary written Turkish sources describe this process, which can therefore be studied only through the medium of much later Arab and Persian historians, whose perception of events is essentially Muslim. It is clear, however, that m their rise to power the Saljuqs had preserved intact their ethnic and tribal identity, and with

it their military strength. Henceforth many traditions of steppe society infiltrated the Muslim world. Among these was the principle of clan ownership, with no clearly defined hereditary succession. Territory was often partitioned among a ruler’s male relations. Another custom decreed the appointment of a guardian or atabeg for a prince in his minority – and such atabegs tended to be military




commanders who often usurped power. Turkish traditions like these clashed with Muslim norms and destabilized Islamic society.

Yet this Turkish element was counterbalanced by more ancient ones. Guides to good government (‘Mirrors tor Princes’) were written for the Saljuq rulers, in which the Sasanian tradition of the divine right of kings was modified by the principle that the monarch must obey the law as defined by Muslim jurists. I ike the Ghaznavids, the Saljuqs acknowledged the caliph’s sphere of influence and gener-

ally operated within the existing political framework, tor example by

having their names mentioned alongside the caliph’s m the Friday sermon at Baghdad and on coins. Even their regnal titles stressed the word din (‘religion’). These various strands are symbolized in the name of the greatest Saljuq ruler. Sultan Malikshah, which blends the contemporary royal titles ofArab, Persian and Turk.

Saljuq administration struck a similar balance between Turkish and

Islamic ways. A tripartite system developed m which the Turkish military aristocracy was supported by Persian high officials and a

Persian or Arab religious class. Moreover, the revival o\’ orthodoxy

which had begun in Baghdad as a politico-religious response to the

Shi’ism of the Buyids was consolidated under the Great Saljuqs.

Through their high officials the Saljuqs gave vigorous impetus to the building of madrasas – colleges where the orthodox Islamic sciences

were taught and the administrators of the regime were educated.

They favoured Sufism or Islamic mysticism – indeed, some o\’ the sultans and their officials adopted notable Sufis as their private

mentors, and encouraged the movement to become part o( official orthodox Islam, with organized fraternities. It was under the Saljuqs

that the pivotal figure of al-Ghazali, the leading Muslim intellectual

and theologian of the Middle Ages, formulated his synthesis o\~

Sufism and Sunnism, thereby introducing a moderate mystical

element into orthodoxy. The Saljuqs also took vigorous measures against the extreme Shfites. The greatest of Saljuq viziers, Nizam al- Mulk. in his work ‘The Book of Government’ (c. 1090). advised his master the sultan not to employ them, and they were cursed from

the pulpits. Shfite mosques, madrasas and libraries were pillaged.

This repression was in part prompted by .1 powerful resurgence of the

extreme branch of the Shfites. the Isma’ilis. who terrorized the Middle East by the weapon of assassination indeed, they are better known m the West .is the Assassins.

Yet the apogee of the Great Saljuq state was short lived. Indeed,

after the death of Sultan Muhammad in 111^ tin- Saljuq empire



split; the long reign of Sultan Sanjar (d. 1 1 57) ensured stability in

the east, but the western territories were riven with discord. By

degrees landed property became so devalued that the entire

landowning class, the dihqans – who had survived nearly five cen- turies of Islamic rule – was wiped out by the early thirteenth

century. The bureaucracy was top-heavy; offices were bought and sold, nepotism flourished (the twelve sons of the great vizier Nizam

al-Mulk were honoured as if they were religious leaders), and

officials shared the fate of their disgraced masters. The fissiparous system of family and clan ownership, the institution of the atabeg

and the uncertain succession all combined to create periodic crises

in the ruling house. A corrupt system of tax-tanning led to a loss of central control as the military class gamed power at the expense of

the state. At a lower level, the nomadic element in Saljuq society

was profoundly destructive. The Turcomans were resentful of the ‘Persianization’ of their chiefs: their prune aim was plunder and they

resisted settlement. New waves of nomads, notably the (muzz tribes which in the 1150s captured Sultan Sanjar himself and his eonsort,

created further havoc. The last Saljuq sultans followed each other in quick succession and ruled steadily diminishing territories until

Tughril II, the last of the line, was killed in battle m 1 104. While remarkably little in the way of the visual arts has survived

from pre-Saljuq Iran, under the Saljuqs this situation is dramatically

reversed, and for the first time 111 Islamic Iran the flavour of a period

can be captured adequately by studying a mass of its artefacts. I his

period, like that of the Umayyads, witnessed a prodigious expansion

in the forms, techniques and ideas ofthe visual arts.

The heritage of the Saljuqs political, religious and cultural – can scarcely be exaggerated. The contrast between the pre-Saljuq and the post-Saljuq periods is striking. The tenth and eleventh centuries

had seen minor Persian and Arab dynasties throughout the eastern Islamic world flourish at the expense of the enfeebled caliphate. The unity of the faith had disintegrated, although Arabic was still the

predominant language. By the late twelfth century the situation had changed decisively: orthodox Islam was now much stronger, having absorbed some heterodoxies and defeated others. This was princi- pally due to the Turkish dynasties of the Ghaznavids and the ( Saljuqs. The caliph had regained his theoretical power by allying himself with the sultan, and was about to recover actual political strength too. The Turks now dominated the Middle East; certain of the territories which they controlled, such as Anatolia, north-west




Iran and Central Asia, have remained Turkish-speaking ever since.

Theirs was in some senses a disruptive influence; they represented a pastoral economy immemorially opposed to agriculture. One con- temporary historian remarked wryly that tax-farming was ‘the only

way to interest Turks in agriculture’. They constituted a recurrent political threat because certain tribes could with impunity flout the

authority of the sultan. Plunder was the only aim of many of the tribesmen and this could not always be channelled into holy wars.

But the Saljuq leaders quickly adapted themselves to the Persian \\,i\

of life. Under their aegis Persian became widespread throughout the empire and Iran itself became an artistic centre of the first impor- tance. Above all. the centre o\~ gravity in the Islamic world had shifted from the Arab territories to Anatolia and Iran. The tradi- tional centres o\ Islamic power in the Middle East – 1 )amascus and Baghdad – had now to some extent been supplanted by such Saljuq capitals as Merv, Nishapur, Rayy and Isfahan – every one of them in the Iranian world. This dominance of eastern Islam, together with the rule of the Shi’ite Fatimids in Egypt and sometimes Syria, made final that break between the eastern and western parts of the Islamic Near East which has endured virtually ever since.

The area within which Great Saljuq art flourished is often Loosely taken to be that of modern Iran, but more of it was m fact outside those political boundaries than within them. Modern scholarship has not progressed far enough to identify the full range of local schools inside the Iranian world with confidence, though it is clear that the

arts of Syria and Anatolia had their own distinctive character. Similarly, the chronology of ‘Saljuq’ art is hard to correlate with

political events: the rhythms of stylistic development are not those ol

dynasties. Typically Saljuq work is found in the early eleventh as m the early thirteenth century, and thus outside the time-span of Saljuq

political power. And extremely similar work in various fields notably architecture – was practised under the dynasties that coex- isted with or succeeded the Saljuqs proper. Nonetheless, this chapter

will confine itself to the output of the Saljuq period m Iran, tor that was the centre of Great Saljuq power; and while some Saljuq rulers

extended their authority far to the west, and to the north-east, then-

hold on this territory was more tenuous. Moreover, the visual arts in

Syria and Iraq in this period followed their own path, in which local traditions played a major role.

The importance of Saljuq art within the broader context of

Islamic art as a whole lies m the way that it established the dominant




position of Iran; one may compare the pivotal role of late-medieval Italy in European art. It also determined the future development of

art in the Iranian world for centuries. In its own time its impact was felt, either through the agency of the Saljuqs themselves or through

their successor states, from Syria to northern India. The period 1 000-1220 set benchmarks in various media, from pottery and met-

alwork to the arts of the book and architecture. However, the overlap between Saljuq art and that of the Buyids, Ghaznavids, Ghurids,

Qarakhanids and Khwarizmshahs — to name only some of the major stylistic groupings of the time — is such that these dynastic labels are often unhelpful if not downright misleading. The basic fact to bear in mind is the existence of an artistic koine in the eastern Islamic world between iooo and 1220. That dialect, moreover, was at its

most vigorous in the years of Saljuq decline and after the fall of the

dynasty in 1194, and it owed much to the political unity imposed by the Saljuqs on eastern and western Iran. It is to this later period that

the major technical advances of Saljuq art can be attributed, though

in the fields of architecture and Qur’anic manuscripts consummate masterpieces were produced long before then. Still, the trend is clear.

66 The second half of the twelfth century (the liobrinski bucket of 1 163 in the Hermitage provides a convenient point o\ departure) saw an

unprecedented expansion of hgural decoration, whether in the form

of narrative scenes (taken for example from the Shahnama of

Firdausi), pictures of courtiers, animals, zodiacal themes, and images

from the princely cycle featuring hunting, banqueting, music-

making and the like. Long benedictory inscriptions m Arabic 64 become the norm in the portable arts. Sculpture in stucco, ceramic

and metal now takes on a new importance. The sheer productivity of these centuries m the visual arts repre-

sents, in comparison with the output of earlier centuries, a quantum leap forward. With this increased quantity – which is helped by a standardization of shapes – comes an expansion 111 patronage, which now not only operates at court level but also has a new popular dimension, perhaps an expression of widespread urban wealth deriv-

ing from a buoyant economy. This art, then, reveals a cross-section of contemporary society and its tastes: luxury and utility Qur’ans; large royal and small provincial mosques; expensive lustre or minai (overglaze-painted) pottery and coarse glazed ware reminiscent of

folk art; elaborately inlaid metalwork and virtually plain cast pieces. One can identify numerous local schools, for example in architec- ture and ceramics. A natural by-product of this intense activity was a




<< i


64 Pomp and circumstance, Turkish style. Stucco relief from Raw. late 12th century It depicts the enthroned Saljuq sultan Tughril (II?; d. 1 [94) surrounded by his officers. Direcdy beneath his feet is written

“the victorious, just king’ and in the panel above are his titles, interrupted atypically by the sultans personal

name placed directly over his head.

wide range of technical and stylistic innovations. It must be remem- bered, however, that the picture is skewed, especially in the fields of

pottery and metahvork, by the massive scale of illegal excavations m Iran over the past hundred years. In other countries of the Islamic

world most of the comparable material is still in the ground. And the paucity of detailed monographic studies of key objects and

buildings means that much basic information is still either unavail- able or inadequately contextuahzed.

Thus the originality of Saljuq art is apt to be exaggerated. In many cases the artists of the Saljuq period (it is misleading to speak of ‘the

Saljuqs’ in this connection) consolidated, and indeed at times per-

fected, forms and ideas that had long been known. In architecture

one may cite the cruciform \-iwan plan, the domed s.mctuarv pavil ion in the mosque, and the tomb tower; in Qur’ank calligraphy, the apotheosis of what has been termed the ‘New Style’ oi Kulic. now

integrated with lavish illumination; in metalwork, the technique of

inlay using several metals; and in painting, the development of the

frontispiece. Above all. there is surprisingly little tor which .1 source

right outside the Iranian world can be posited. Although the Saljuqs

themselves were lurks, it is hard to point to .inv specifically Turkish



elements in the art of Iran and its eastern provinces in the period

under review, with the possible exception of the moon, or Buddha, face in figural depictions. This seems to point to the dominance of Iranian artisans in the visual arts. Parenthetically one may note that the picture in Anatolia, where people of Turkish extraction formed a larger proportion of the population, is distinctively different; there,

references to pagan Turkish religious beliefs, funerary customs and

royal ceremonial are frequently encountered.

What of patronage? Only two pieces of Saljuq pottery made for persons of high rank, one an amir, the other a vizier, are known, and the situation is little better in the case of metalwork. The over- whelmingly rich and varied production in these fields ought presum- ably, therefore, to be attributed to patronage exercised at a lower

level of society, such as merchants, members of the learned class and professional people. Most likely, much of it was made for the market, though this would not exclude its use by those o\ high rank. Architecture, involving as it did much larger sums o( money, is a different story altogether. Inscriptions m mosques and mausolea mention the Saljuq sultans themselves, tor example Mahkshah and Muhammad, or viziers such as Nizam al-Mulk or Taj al-Mulk. Turkish chieftains are named in the tomb towers of khanaqan, army commanders at Urmiya. and numerous amirs, tor example at Maragha, Mihmandust. Qazvin and Abarquh.

Problems of provenance have bedevilled the study of the decora- tive arts in the Saljuq period. These problems have been exacerbated by the fact that most o\’ the known material has not been scientifically excavated and lacks inscriptions yielding solid informa- tion on provenance. Confusing and contradictory information on this topic proliferates. The very tew securely provenanced items per- force act as a peg on which to hang all manner of other pieces, and their evidential value is simply not enough to justify this practice. It is now generally accepted that virtually all lustre and mina’i wares –

65 the most expensive ceramics of the period – were made m Kashan (though the distinctive heavy red body ot” lustre tiles found in the Kirman area suggests local production there), and sherding studies suggest that this luxury ware was widely traded. Conversely many other slightly less luxurious but still tine wares cannot be securely associated with any one city or area, and they might therefore have been produced in several places independently (like the Samanid epi- graphic ware ot the tenth century which was produced in both Samarqand and Nishapur, and apparently in Merv too).

65 Ceramic sculpture. Lustre mihrab re-used in the Masjid^i Maidan. Kashan. It is dated 1226. inscribed with the names of the Twelve Shi ite [mams and signed by al-Hasan lbn


Arabshah. Its many components were separately fired and fitted together. The contrast ofbuff and blue mimics the palette of contemporary architecture. H. 2.84 m (9.3 ft).



IS^m^^y f%


B£ mi



66 Similarly, the fact that the celebrated Bobrinski bucket and the jug

of 1 182 now in.Tiflis both bear an inscription stating that they were made in Herat indicates that fine inlaid metalwork was produced in that city. The craftsmen’s names, which are traditionally supple- mented by their place of origin (nisba), indicate Khurasani cities –

Herat, Merv, Nishapur – and thus confirm the important role of this province in metalwork. But it is not enough to justify the wholesale

attribution to Herat of wares that merely share some of the features found on work from that city. This is particularly unlikely for metal- work that is technically simpler than the inlaid pieces, since the demand for such cheaper work must have been too widespread to be catered for by a single production centre. But exactly where these other Iranian workshops were located must be determined by future

research. The astonishing range of forms encountered in Saljuq metalwork (including many derived from architectural forms) also points to numerous centres of production. It seems likely that some of the best craftsmen travelled widely to execute commissions, and

that fine pieces (e.g. of Kashan tilework) were shipped over long dis-

tances. There is evidence too of a division of labour in metalwork

*~’-‘\ 1

<><> I he ultimate pilgrimage

accessor) ‘ I he ‘Bobrinski

bucket*, technicolour t .1st and

inlaid bronze; possibly for

ablution during the hajj. Six

long inscriptions mention die

makers (caster and decorator,

ofequal rank); provenance

and dai II rat, Muharram

559 I )ecember 1163); and me inflated titles of its patron, .1

pious merchant from far-off

Zanjan. Note the human- headed letters (the earliest dated

examples known) and the scenes

ofleisure pursuits H 18 cm (7.1 in).



and Lustreware that ensured a higher level of quality overall. But the

key question remains: scholarship has not yet established whether the

pockets of intense activity in a limited geographical area have a wider

significance tor pan-Iranian production or whether they reflect a

well-developed specialization confined to a given area.

Laboratory examination has yet to be used in a systematic way on

Saljuq metalwork; the evidence that it would provide on alloys, for instance, could then be correlated with other factors shape, tech

nique. decoration – to create a more nuanced picture of the various known types. In the current state of knowledge it is sate to say that wares constructed from sheet metal were made of brass while most

others were of a four-part alloy; true bronzes are uncommon. A twenty per cent tin bronze was also used but traditional low-tm

bronzes are unknown.

The very tew pieces ot Saljuq metalwork in silver point to a serious shortage o\ that metal which became more critical as the

eleventh century advanced. It was perhaps in part .1 result o\~ the

practice followed by the Viking traders travelling along the great

Russian rivers, who hoarded the Islamic silver coins with which they


67 Latent iconophobia?

Signed open-work

roomorphic incense-

burner, bronze inlaid with

silver. North-east Iran.

1 1 th century. The btheness and ferocity of

this creature are much exaggerated, the body

itself dematerialized and

reduced to an inscribed

and decorated surface.

Thus the artist avoids

usurping God’s prerogative

ot creating lite. Compare Ibn ‘Abbas advising a

painter: “You must

decapitate animals so that

1 not seem to be alive .md trv to make them look like flowers*.




68 (right) Luxury tableware. Silver rose-

water sprinkler with cap: repousshind

chased, with niello decoration and gilding.

Rose-water was used to scent the beard

before eating, for washing hands, perfunung

clothes and carpets and flavouring food.

The rose was the favourite Muslim flower and figured largely in Persian love and

mystical poetry. H. 24.9 cm (9.8 in), body diam. 12 cm (4.7 in).

69 (below) The courtly ethos. Polychrome painted minai bowl with confronted

horsemen and peacocks (symbols of Paradise), a design derived from textiles.

Iran, probably early 13th century. The androgynous figures follow contemporary fashion with their Turkish caps and long

plaits. The fit palmettes echo Qur’anic illumination. Diam. 21.5 cm (8.5 m).



were paid for slaves, firrs and amber and who thus took the coins out of circulation. The gradual cessation of the minting of silver coins m the Iranian world and Anatolia m this period, and their replacement by copper dirhams, provides incontrovertible evidence of this trend,

anecdotal evidence of the survival or use of individual silver objects

notwithstanding. Base metal had perforce to fill the £A\\ but its value

was greatly enhanced by the practice of inlaying it with copper.

Silver, gold and a bituminous black substance, the whole giving an

effect ofpolychrome splendour. 71ms tine craftsmanship did dut\ for precious metal. This technique, with its plethora of detail, lent itself

to the creation of elaborate figural scenes; even inscriptions took on

human and animal form. These inlaid objects survive in large quanti- ties, probably because their metal content (unlike that of silver and

gold objects) was not sufficiently valuable to be worth melting them

down, whereas the intrinsic value of their top-quality craftsmanship

was obvious.

In ceramics, the earliest dated underglaze-pamted. lustre and

mina’i wares are respectively placed by their inscriptions to the years

i [66, i 179 and 1 1 So, and therefore all postdate the death of the last

Great Saljuq ruler. Sultan Sanjar, in 1157. Conversely, in metalwork

there are several dated pieces between 1063 and 1 14N – i.e., from in

the Saljuq period. The frequency of dated ceramics (and many are signed) argues a higher status for fine pottery than had previously


70 A clue to lost Saljuq book painting. Moulded lustre plate made in 12 10 by Sayyid Shams al-Din

al-Hasani tor a military

commander. A royal groom sleeps by a pool, oblivious

of the monarch’s entourage,

and dreams of a water-

sprite. The fish, water. woman and horse all relate to Sufi mystical metaphors.

H. 3.7 cm .5 mi. di.un. .Vs. 2 cm 1 [3.9 no.




obtained. A new light body known as stone-paste or fritware was devised, though whether this was a Saljuq or a Fatimid invention

remains unclear. It was made largely from ground quartz, with small quantities of ground glass and fine clay, presumably an attempt by Islamic potters to imitate the body of Chinese porcelain. (The dis- covery of Ding and Ch’ing-pei wares at Gulf coast sites provides some of the necessary evidence ot trade with China.) Such pieces were mostly moulded. Others belonged to categories known as sil- houette and double-shell wares. In these, as m sgraffito wares, much of the decoration was incised with a knife or a pointed object. In the

silhouette type the design was often scraped through a black slip

under a turquoise glaze. Such incised wares continued a fashion w ell- established before the Saljuq period. Underglaze painting in blue and

black was also popular, as was a type of translucent white ware, often

pierced for greater effect. Main ot the more expensive wares bear hurried cursive inscriptions or Persian love poetry, mostly indifferent

71 (below left) A wedding present’ Overglaze-painted beaker in mina’i or hafi rang (seven-colour) technique. Iran, early 13th century. The narrative, in strip cartoon format, recounts the love storv ii\ Bizhari and Manizha, a highlight of Firdausi’s Shahnama. This beaker long predates manuscript illustrations of the storv

and proves that a well-developed Shahnama iconography existed by c. 1200. H. 12 cm (4.7 in).

72 (below right) The arts were interdependent. Spouted jug, painted black under a transparent turquoise glaze. Kashan, early 13 th century. The elaborate perforated shell transfers to the fragile medium of ceramics a technique first developed in metalwork. and better suited to that material.



73 The world ofmagic. Peep green bowl, in champlevi technique (i.e. with large areas of slip cut away). Western Iran, i ith century. The sphinx has apotropaic, paradisal and astro logical associations, and is also often deputed with the griffin, thereby symbolizing the sun’s journey through the heavens. The rosettes echo this theme. The facial features conform to the contemporary ideal ofbeauty Diam. 25 cm (9.8 in).

m quality, and praise the maker of the piece. Scientific analysis of pottery has successfully differentiated between the original ceramic

and modern repairs to body and decoration alike, a crucial distinc- tion since, apart from a cache of lustre wares found at Gurgan, virtu-

ally no medieval pieces have remained intact. A close connection existed between the most elaborate Saljuq

ceramics and book painting, including – in the case of abstract orna-

ment – Qur’anic illumination, as shown by figural types, narrative strips and numerous stylistic features, while many details of the shape and decoration of Saljuq ceramics handles, stepped teet. Imitation

chains, incising, gilding, fluting derive from metalwork. Similarly,




the ornamental sheen and decorative motifs of Saljuq metalwork

reveal close familiarity with manuscript illumination. All this points

both to the interdependence of the arts in this period and to the

existence of hierarchies within the visual arts, since the cheaper arts

copy the more expensive ones — never vice versa. The recent demonstration that the majority of textiles once

thought to be Buyid or Saljuq are in fact of modern manufacture has made it imperative to submit all so-called Saljuq silks to scientific tests, and proper renders premature any art-historical

enquiry into them.

The principal centre of book painting in the twelfth and thir- teenth centuries was Iraq, which was then under the control of the newly renascent caliphate (see pp. 125-32). But this painting often

has marked Iranian features, suggesting the existence of an earlier

pan-Saljuq school of painting in which distinctions between Iraq and Iran were perhaps not very significant. The most likely candidate to represent the largely vanished art of Saljuq book painting is the verse romance Varqa va Gulshah (‘Varqa and Gulshah’), written in Persian

by the poet c

Ayyuqi and signed by the painter c

Abd al-Mu nun al- Khuyi. This suggests a provenance in north-west Iran, but Anatolia is

a distinct possibility too. The manuscript (in the Topkapi Saray 74 Library in Istanbul) has seventy brightly coloured illustrations in strip

format against a plain coloured or patterned ground, with figural

types of the kind familiar in minax pottery. The paintings have a strong narrative drive enriched by a complex iconography in which the animals which figure in many of the pictures take on symbolic meaning, connoting for example watchfulness, fidelity, treachery and

courage. A fragment of al-Suffs Kitdh Suwar al-Kawakib al-Thabita (‘Treatise on the Fixed Stars’) in the Bodleian Library, Oxford (Or.

133), undated and unprovenanced but probably of the thirteenth

century, might be of Iranian origin. But for all the paucity of the

surviving material, the clear dependence of both fine ceramics and

fine metalwork on manuscript painting and illumination shows clearly enough the high profile which the arts of the book enjoyed in the Saljuq period.

Several fine Saljuq Qur’ans have survived. They include dated examples in Mashhad (466, i.e. 1073 ni tne Christian calendar), Tehran (485/1092 and 606-8/ 1 209-1 1), Philadelphia (559/1164; produced in Hamadan) and London (582/1186), as well as examples which slightly predate the advent of the Saljuqs (London, 427/1036 and Dublin, 428/1037). There are also numerous undated but proba-




bly Saljuq examples in Dublin. Paris, Istanbul. Tehran and London,

to say nothing of parts of Qur’ans or individual leaves in dozens of collections throughout the world. Saljuq Qur’ans are notable for their magnificent full-page or double-page frontispieces and

colophon pages, often ot pronounced geometric character, with script in panels taking a prime role. They are known both in naskhi 76 and in ‘New Style’ (or ‘East Persian’) Kufic. There is .1 substantial 75 variation in scale – from small one-volume Qur’ans measuring only [2 > 10 cm (4.” X 4 in) to large ones of 41 • 28 cm (16 > 11 in), and in some the limited amount of text per page resulted in Qur’ans o( thirty or sixty parts, large and small, each part with its own frontispiece. The diversity of size and layout extends to the number of lines per page, which varies from two to twenty, and to the scale.

quantity and placing of illumination. The task o\~ establishing dates and provenances tor this ample material, and devising working cate

gories for it. has only just begun.

In architecture even more than in other fields the dividing line, so

far as style is concerned, between what is defmably Saljuq and what precedes that period is very hard to draw, though the Mongol inva- sion and the architectural vacuum that followed mean that there is a distinct break in continuity after c. 1220. A few examples will make this clear. The characteristic minarets of Saljuq type – lofty, cylindri- cal, set on a polygonal plinth and garnished with inscription bands

and geometric brick patterning – are known from at least as early as the 1 020s (for example at Damghan and Simnan). Of the two Stan- 77 dard types of Saljuq mausoleum, the tomb tower perhaps reached its apogee 111 the Gunbad-i Qabus, dated 1006-7, while the other type, 78 the domed square, had already been brought to a pitch of perfection in the so-called ‘Tomb of the Samanids’ in Bukhara, datable before -<>

943. That building also exhibits a highly developed style ofbrick and

terracotta ornament. Similarly, such standard features of Saljuq archi-

tecture as the trilobed squinch and the pishtaq or monumental portal

are already to be encountered in the tenth century, for example in

the mausoleum of Arab-ata at Tim. The same phenomenon can be detected in other art forms, for example in sgraffito potter) ^v the

continuity o\~ ring and dot decoration from pre Saljuq to Saljuq

metalwork; and while the quantity and range of architectural tile

work is indisputably a ‘Saljuq’ phenomenon, its roots 111 Islamic monuments he as far back as Abbasid Samaria. The distinctive Saljuq contribution lies rather in tin- final

establishment of several of the J.issu fin ins ot Iranian architecture



^p^^-^1 \\JV> J +>

‘\£}gj&±W&> J^ V

-4 {above) symbolism.

Varqa, recuperating m bed from his travails, asks the maul where his beloved Gulshah is. Vbrqa w Gulshah, perhaps Anatolia, inid-

13th century. The clog with his

catch is a metaphor for Varqa’s

persistent and successful son h tor Gulshah. I he narrative moves

from left to right.

Saljuq Kufu Qur’an, 1 i j6 Ran hment Qur’ans were

typically oblong, paper ones

vertical. Note the suprahnear

and sublinear flourishes which

add an expressive dynamism to the calligraphy. In general, red

dots Add vowelhng; in some

manuscripts, green dots indicate

primary variant readings, while

yellow and blue ones represent

specific orthographic elements or

sounds, or secondary and tertiary

variants in the text. 33.5 x 23.H

cm (13.2 x 9.4 in).



, Saljuqnos/WnQur’an Copied and illuminated by Mahmud b ,1 I lu,„., ..I I-

u’itc, in Hamadan in .164. it also has u nent inscription

headings arc in Kufic; note the interlin, , ‘

out: it has five lines of text, not the usual




A tire temple in Islamic dress. Bukhara, ‘tomb

of the Samanids’, before 943. A precocious masterpiei c in brick, integrating compa t monumentality with refined

all-over geometric ornament (derived from basketwork?).

The pre-Islamic open-plan domed square is enlivened by engaged columns, gallery

and corner domes.



— (opposite left) Damghan, minaret of Tan Khana mosque, built at the order of the

chamberlain (hajib) Abu Harb Bakhtiyar m 1026. Typically built of baked brick, many

such minarets survived the mud-brick

mosques which they adjoined. This minaret,

now 26 m (85.3 ft) high, probably had an upper gallery for the muezzin,

posite right) Syncretism. Gunbad-i

Qabus, [006—7. This tower crowns an

artificial mound and dominates the surrounding countryside; it is > 1 m | if>~ .; ft) high. Ten knife-edge flanges girdle the

central cylinder in a strikingly modern and minimalist design. The glass coffin was

suspended under the roof oriented towards

the rising sun. a non-Islamic burial practice

with Zoroastnan associations.

So {right) Marital devotion. Mausoleum o\~ the princess Mu’mina Khatun. ‘Chastity ot~ Islam and the Muslims”. Nakhchivan. 1 1 so.

built by her husband, the amir Ildegiz: a

landmark for the use ofglazed ornament in Iranian architecture.


w\ • II


and in the capacity o\ Saljuq architects to draw out the utmost

variety from these types. Mosques with one. two. three or tour iwans are known, and the \-iu\m plan receives its classic formulation m association with an open courtyard and a monumental domed chamber; a hierarchy ot size distinguished major iwans from minor ones. The Friday Mosques of Zavara, Ardistan and above .ill Isfahan are outstanding examples of this trend. Saljuq domed chambers are characterized by external simplicity, with a frank emphasis on the

exterior zone of transition, now reduced to powerful contrasting geometric planes. The interior of the dome chamber is dominated by a highly elaborate transition zone (in the Isfahan area this made a leitmotif of the trilobed arch) whose depth, energy and rhythmical movement has as its foil the austere, low-relief articulation ot the lower walls and the inner dome itself Hut other Saljuq mosque types, such as the free-standing domed chamber or the arcaded hall.

:lso known. In mausolea, the pishtaq was developed from a simple salient porch screen which conferred a grandiose facade on the building

behind it. as at Tus and Sarakhs. I he originall) simple formula ot the

domed square underwent other major changes too, not.ihk 111 the

s I



8i {above and right) Community centre. Friday Mosque, Isfahan, 10th century onwards. Successive genera- tions embellished, repaired and extended this mosque, which was engulfed in the city’s bazaars and served many functions. The two Saljuq dome chambers mark the principal axis, {right} Beneath the myriad small domes resembling molehills which encircle the courtyard lies an endlessly varied scries of vaults, many of them unique. The best of them date to loth—I2th centuries.



tsite) Tents tor the afterlife. Tomb towers built in open country at Kharraqan,

western Iran, by the same architect in 1067

and 10S6 for Turkish chieftains. Many details ofStructure and ornament evoke the yurt or

tent of the Turkic nomads. The later tower. a torn it force of decorative brickwork, has

almost 70 different patterns.

>ht) A lance aimed at the infidel. Minaret ofJam. Afghanistan, 1 190. Built by

Sultan Muhammad of Ghur, hammer of the Indians and the local pagans, it bears

quotations from the Qur’anic sura ofVictory

and the whole of the sura of Maryam. traditionally used as an instrument of religious

conversion. Its three tiers rise to a height ot

c. 60 m (197 ft)-



development of a gallery zone (Sangbast), engaged corner columns

(Takistan and Hamadan), and double dome (the mausoleum of Sultan Sanjar at Merv). Lofty tomb towers proliferated across north-

82 ern Iran, many of them built as secular memorials for amirs and others of high rank, though some have mihrabs and therefore served at least in part a religious purpose. Their form varied: some were square, cylindrical or flanged, but most had 7. 8, 10 or 12 sides, with

inner domes crowned by conical or polyhedral roofs. Their form was well suited to the development of brick ornament, for it ensured a

constant change of plane and therefore much variety in the play of 80 shadow. Here, too, are found some of the earliest surviving examples

of glazed tilework.

The impressive sequence of some forty Saljuq minarets com- prises all manner o\ structural variations, including single or double staircases with or without a central column. Baring corbelled bal-

conies, three-tier elevations, shafts articulated by Manges and

engaged columns, and – an innovation destined to have a long

history in Iranian architecture – the double minaret Hanking a

portal, whether this was the entrance to a building or the qibla

iwan. Many have a symbolic rather than a strictly liturgical role. 83 Minarets also occur as highly visible free-standing monuments

unrelated to Other buildings, and 111 such cases seem to have func-

tioned as land-locked lighthouses.

No Saljuq palaces survive in good condition, though excavations have revealed the ground plan of the 4-iwan palace at Merv and the palatial kiosk ofQaTa-yi Dukhtar in Azerbaijan still stands despite its

ruined state. Hut the palaces of Tirnndh. Ghazna and Lashkar-i Bazar, all yielding abundant decoration, belong to much the same cultural sphere even though they are linked to Samanid and

Ghaznavid rulers respectively. The same situation applies in the case ot the madrasa, a particularly serious deficiency given the unambigu-

ous testimony of the literary sources to the effect that such buildings

were erected throughout the Saljuq empire. Controversial rem.11ns

surviving at Khargird, Tabas, Raw. Samarqand, and near Sayot in Tajikistan (Khwaja Mashhad) permit no clear statement .is to the form of the madrasa in Saljuq tunes. The luxuriously embellished and largely ruined Shah-i Mashhad ot’ 1 175-76 in Garjistan, identified by its inscription as a madrasa, is a Ghurid foundation, while the build- ing at Zuzan, dated 615 [218—19 2Uid also identified epigraphically as

a madrasa, was erected by a governor of the Khwanzmshahs. Taken together, their awesome scale and magnificence suggest that the




M Massive monumentality Central Asian caravansarais of the tith and [2th centuries, though built for trade, owe much to earlier local fortified manor houses. The Rabat i Malik of 1078, in Uzbekistan, now mostly destroyed, has a stark power and sense of volume which is strangely modem in feeling.

madrasas of the Iranian world in this period far outshone those from other Islamic territories.

Several caravansarais (lodging places for travellers and their ani in. lis 1 datable to Saljuq times are known; four of them – K1b.1t 1 Malik. Daya Khatun, Ribat-i Main and Ribat-i Sharaf bear lavish decoration. Indeed. Ribat-i Sharaf (probably built 1114 is. repaired

1154—55), Wltn lts huge double courtyard plan (repeated .it AJccha Qal a m Turkmenistan) is a museum of contemporary decorative techniques. Such splendour, when linked to its location astride the main from Mere to Nish.ipur. 111. ikes it plausible this build ing served as .1 royal stopover Must Saljuq caravansarais, however, are built tor use rather than display, with rubble masonry, strong




fortifications and minimal comfort. In many of these buildings the prescriptive power of the \-iwan plan made itself felt. The Great Saljuqs, then, ushered in the last major period ot

ferment in medieval Islamic art. The innovative power ot this era in virtually all media cannot be gainsaid – though too little is known ot the immediately preceding centuries to allow the Saljuqs to be hailed

as the absolute inventors o\\\ given feature or technique. The role of the Saljuq ruling class was in any case that ot’ a catalyst rather than

85 that of an originator, though influences from steppe society can fre-

quently be detected in Saljuq art. and helped to fashion its distinctive

character. That character had a decisive impact on the art not only of

Iran but of the numerous Saljuq successor states. These close connec- tions, and thus the full canonical power ofSaljuq art. are obscured by the tendency ot modern scholars to think in terms of watertight chronological and geographical entities. The key concept here is that the Saljuq synthesis left its mark on all later Islamic medieval art from Egypt eastwards.

‘Ai’> y *5

85 The royal hum of the sun. Stone reli lestan. 12th century Hunting here takes on cosmic overtones, tor the animals forming .1 wheel design (itself found on ancient Mesopotamian cylinder seals, and especially popular in the J.i/ira. the Caucasus and north-west Iran at this time) are a solar symbol, representing the full astral cycle across the heavens.





The Age of the Atabegs: Syria, Iraq and Anatolia, 1100-1300

Between the eleventh and the early thirteenth centuries Anatolia and

the Levant experienced something of a power vacuum. Neither the ailing


Abbasid caliphate nor the Fatimids were able to extend then-

writ to all these areas, some of them remote from their own home base, and even tor the Great Saljuqs these regions were peripheral. Thus, while from time to time


Abbasid, Fatimid or Saljuq hege-

mony was recognized in some of these territories during the period under discussion, the norm was for power to be wielded over a limited area by a local warlord. The political orientation of such rulers was decisively to the east, however, the focus of orthodoxy and

the Sunni revival, rather than to the ShTite state of Fatimid Egypt.

They tended, for example, to be more involved in the power politics of Iraq, where the gradually resurgent ‘Abbasid caliphate, shorn of its

pan-Islamic power, was attempting to assert itself territorially. This

situation became still more pronounced after about I 100. when the power of the Great Saljuqs waned just as the Crusaders arrived in the Near East. The contemporary power vacuum made it much easier tor the

Crusaders to establish first a bridgehead and then several fully-

fledged independent states. It became the steadfast aim of the more

important Great Saljuq successor states in the area of Syria, Anatolia

and the Jazira (northern Mesopotamia) – Artuqids, Zangids, Saljuqs of Rum (Anatolia), Ayyubids – to crush the Crusaders, and the gradual build-up of Muslim religious fervour to this end. culminat-

ing in full-scale jihad (holy war), can be traced throughout the

twelfth century. But they were scarcely less keen to tight each other,

and the boundaries of their mini-states were m continual flux, espe- cially .is they practised the ruinous system of divided inheritance, as

had the Great Saljuqs before them. Thus the Rum Saljuq sultan Kilij partitioned his empire between his eleven sons, bequeathing

them a legacy of envy and strife. The situation was complicated still

further by the rise to power of Still lesser dynasties usually based o\\ a




single town, such as, in Anatolia, the Shah-i Annan at Ahlat and the Mengjukids at. Divrigi. Ethnically, these new rulers were neither Arabs nor Persians but Turks or Kurds, and this added a new element to the political complexion of the time.

De facto power in the Levant, Anatolia and the Jazira was now vested in warlike Turcomans whose tribal and nomadic heritage inevitably placed them at loggerheads with the peasantry and the urban populations whom they ruled. They had entered the Islamic world as the shock troops of the Great Saljuqs, and, proving difficult

for the sultans to control, had been despatched to the outskirts of

Saljuq territory. They had wrested much ot Anatolia from Armenian and Byzantine hands and had also infiltrated the long-Islamized

territories of northern Iraq and Syria. In the process, they had en-

countered not merely urban and rural Muslims but also Christ-

ians of various confessional allegiances – Orthodox, Armenian and

Jacobite. Similarly, the Ayyubids in the Levant ruled a large popu-

lation of Christians, mostly oriental but some from western Europe. All this made for a pluralistic, multilingual, multi-ethnic society in- stinctively hostile to the imposition of orthodoxy. Nevertheless, there

are frequent references in the sources to measures taken against the

indigenous Christians: tor example, destroying churches, refusing

permission to have them rebuilt, or converting them to mosques.

Much of the interest ot~ the ait o\ the Atabeg polities derives pre- cisely from the wined accommodations which they fashioned with

non-Muslim traditions. They frequently employed non-Muslim artists. Equally interesting is the undertow o\~ Persian modes m Anatolia and of Arab ones in Syria and the Jazira. though a Turkish

military elite was dominant m all three areas. The Turcomans’ version of [slam seems to have had a distinctive

character, involving as it Jul animistic and folk elements absent from

orthodox interpretations of the faith. Hut this did not prevent rulers

of Turkish stock from parading themselves as paragons of orthodoxy.

The political gams of such a stance in the long-drawn-out wars against the Crusaders were obvious. In the case of the later Zangids

and early Ayyubids. moreover – especially Nur al-Din and Saladin –

the personal piety ot certain rulers is harder to doubt and must have

imparted an extra charge of energy to their prosecution ofjihad and

their hostility to the Isma’ilis. The titulature of these rulers reflected precisely these concerns. Many of these rulers also treated members ol the


ulama (religious classes) with special and public marks of dis-

tinction. They were also great builders of religious foundations.

I 12



Others lavished honour on Sufis and founded special establishments (khanqahs) for them.

It is extraordinarily difficult to define a significant degree of

homogeneity – politically, ethnically or culturally – in these areas during the twelfth and thirteenth centuries, even though they were

contiguous geographically and experienced similar tonus of govern-

ment. In some ways the period constitutes an interlude’ sandwiched between epochs that were dominated by more powerful dynasties. Moreover, in the visual arts the inherited tradition differed markedly

from one region to the next, and this alone effectively forbade the

creation of a single style. But the vigour of contemporary art gives

no clue to the uncertain political complexion of the times; .is so

often, the rhythms of art and polities do not synchronize. Not surprisingly. Ayyubid art continues Fatimid modes. Nurid art m northern Iraq is clearly a province of the Saljuq art of (ire. iter Iran.

while the Saljuqs of Rum owed much to Armenian architecture. Hence the difficulty of understanding the place of each of these very different styles within the composite picture o\~ contemporary art.

Accordingly, to lump them together m a single chapter may smack of manipulating historical realities. What do Ayyubid Syria and Saljuq Anatolia have in common? Perhaps what most links them is the fact that all of them could be described as Saljuq successor sutes – and

that not only because they followed the Saljuqs chronologically but

also because they maintained similar traditions o\~ Sunni orthodoxy

and governmental practice, and were shot through with Turkish

customs and habits of thought.

Saljuq Anatolia was the most long-lived of these polities.

Culturally speaking, it was in many respects a province of Iran. This situation had its roots in political realities, for the area was originally

overrun by Turcomans from Iran, and the ruling dynasty from i i sc>

was related by blood to the Great Saljuqs of Iran. Soon after the line

of” the Great Saljuqs had been extinguished, the Mongol invasion of

Iran brought in its tram not only unprecedented carnage but also a

new state, major Upheavals in religious practice, and new cultural

priorities. These factors caused a stream of Iranian refugees to seek

asylum in the safer, more familiar and congenial atmosphere o\

Saljuq Anatolia. Persian poets, mystics and men of letters like Kuim and Nasir al-1 )m Tusi received a warm welcome m court circles, where the language of cultural interchange was Persian, where

viziers of Persian origin Hike the Pcrvanc. Bunijirdi and the Jmaini

brothers; wielded power and the court ( hromclei’s dike Ibn Ihbi a\\A

i i I



Aqsarayi) wrote in Persian. It was the same story in the visual arts. Persian architects and tileworkers left their names and those of their home towns on Anatolian buildings from Konya to Divrigi. Much Anatolian luxury pottery favoured and continued the themes and

techniques of Saljuq Iran (as at Qubadabad), as did the local metal-

work (e.g. the basin known as the Nisan Tasi). Even Anatolian mau- solea replicated, though on a smaller scale and in stone rather than in

brick, the forms of Iranian Saljuq tomb towers. A Saljuq Anatolian provenance has been proposed for the earliest

surviving illustrated Persian manuscript, the romantic epic Varqa va

Gulshah (see p. ioo). By far the closest parallels to the style and format of its cartoon-strip pictures are provided by the so-called

‘small Shahnamas 1

(‘Books of Kinds’ i. whose provenance has long been disputed. This is because they do not tit easily into the evolu-

tion of book painting in Iran proper during the fourteenth century, in which the influence of ideas derived from China is ubiquitous.

True, these Shahnamas contain occasional references to Mongol costume and armour, but that is entirely appropriate in view of the

imposition of Mongol rule in Anatolia after the battle of Kose Dagh in 1243. Given the close kinship of the hgural types of the small

Shahnamas with those of lustre and mina’i pottery, given too the

Iranophile and Iranophone nature of the court culture of the Rum Saljuqs, and finally given the remarkable fondness for archaic

Shahnama names evinced by successive Rum Saljuq rulers Kaikhusrau. Kaika’us, Kaiqubad – it does seem at least tenable to see

these small Shahnamas as a kind of refugee .in and thus to consider .is

a possible provenance later thirteenth-century Anatolia, with its per-

vasive fashion tor all things Iranian, even though – or perhaps

because – the area was .1 Mongol protectorate. The choice o( text would then evince a st.umch patriotic commitment to the home country – temporarily down, but emphatically not out. That said, equally strong arguments could also be marshalled for a provenance

in Iran itself.

The absence ot a major Islamic power in the region ot Syria, Anatolia and the Jazira meant that each dynasty tended to establish its

own court, and ot’ course these local “courts’, if is the right word, varied in size and sophistication. It was the Saljuqs of Rum whose lifestyle was the most ambitious and lavish of all; they were at pains to model their court ceremonies on those of the ( Saljuqs, and their chronicler Ibn Bibi provides detailed descriptions of the

protocol followed at public audiences, banqueting, the hunt and




16 The classical afterglow. Alexander the Great – a powerful and mythic totem tor last and West alike ascending to heaven. Detail ofbronze plate inlaid with 7-colour cloisonne enamel hearing the name of the Turkish Artuqid prince Da’ud (reigned 1114-42). This object is thoroughly international. It bean nisirip

dons in Arahic and Persian: its technique is Byzantine; and its iconography has Byzantine, Georgian and

Islamic connections.

Other activities. Whether it is strictly accurate to speak in terms of an Artuqid court, on the other hand, is another matter. These rulers

were campaigning tor most of the year and were therefore constantly

on the mow. and – unlike the situation under the Great Saljuqs they did not have an established bureaucracy to back up political and

military control with administrative authority. They had their palatial

residences m their citadels – examples have survived m Mardin, Mosul. Diyarbakr, Aleppo and smaller castles hke Sahyun and they

would at tunes commission works of art such .is doors and automata.

1 1 s



illustrated manuscripts and enamelled metalwork. It is even possible

that by the end of the twelfth century palace workshops were main-

tained on a regular basis.

But by far the major expression of royal patronage was religious

architecture. In this respect the ruling class was conforming to an

ancient Islamic ideal which dictated that the ruler should build widely for the public good. Thus it was standard practice for amirs to build madrasas, usually with their own tombs attached, as soon as they had the means to do so – and it was this custom above all, more than any government-sponsored building programme, that ensured

the rapid spread of these institutions of learning throughout the Near

East in the twelfth and thirteenth centuries. Large cities like

87 Damascus and Aleppo had scores of such monuments, while in

Saljuq Anatolia the tally ran to several hundred in the thirteenth

century alone. Indeed, it is here that the early architectural history of

the madrasa institution can best be traced. The function did not require any particular form. Domed madrasas focused on a single large chamber with adjoining cells and occasionally a courtyard may perhaps have catered for a reduced clientele of students, while the

more ambitious 2-iwatl or \-iwan madrasas, which were often graced 88 with imposing 2-minaret facades as at Kayseri, Sivas or Erzurum)

could house substantially larger numbers. In some larger cities competition between viziers seems to have generated increasingly

elaborate buildings.

Mausolea proliferated even more, becoming the favoured means of conspicuous consumption m architecture. In Syria the standard form was a domed square, relatively plain inside and out, with the all- important identification of the tenant, complete with genealogy and

titulature. on a panel over the door. Much of the appeal of such buildings lies in their stonework. Egypt preferred something bigger

and grander, with a much more elaborate /one of transition incorpo- rating multiple tiers of squinches, as m the tomb of al-Shaffi. In northern Iraq rather squat square tombs with pyramidal roofs and

facades m decorated and glazed brick were the rule ‘for example those of Imam ‘Awn al-Dm and Imam Yahya b. al-Qasim). Further south the muqamas dome reigned supreme, with a sculpted many- tiered sugarloaf on the exterior matched by an inner dome like a fit- fully illuminated honeycomb (as in the tombs of Sitt Zubaida and e

Umar al-Suhrawardi in Baghdad); the fashion spread briefly to Damascus as can be seen in the funerary madrasa and maristan of Nur al-Dm). But it was in Anatolia that the mausoleum genre was

SS {opposite) College tor the religious elite. Erzurum. Qiftc Minare Medrese, before 1242. This, the largest

madrasa in medieval Anatolia, accommodated perhaps a hundred studentv A mausoleum was added c. 1284, perhaps by Padishah Khatun. the wife or” two Mongol khans. The building is of local volcanic stone, except for the two Iranian-style fluted brick towers with dazed ornament.



B7 [above) The teaching of the law Al-Firdaus Friday Mosque and madrasa, Aleppo. 123 s f>. limit by I >.ufa Khatun. the wife of the local ruler. Sultan al-Malik al-Zahir – it was a tunc of lavish female patronage of

religious buildings – it illustrates a newly fashionable type, the funerary complex. The marble floor with geometric designs is a local speciality.


I — -M





89, 9° The pervasive Middle Eastern fashion tor m.uisolca dates from the i ith century onwards and took many forms. At Gevas {left) in eastern Turkey the tomb tower of Hahma Khatun. 1335, has clear Armenian connections in material and design; the mausoleum ot Sitt Zubaida {right), built c. 1200, has a standard Iranian lower half, decorated with geometric brickwork, but is crowned by a conical muqamas dome whose external form has ancient Mesopotamia!! associations.



explored with the greatest energy and ingenuity The preferred type was the tomb tower, as it had been m Iran tor the preceding two centuries: a circular or polygonal lower chamber (often with an underground crypt for the body) crowned by .1 conical or polyhedral

roof. Yet now. probably under the impetus ^\ ideas drawn from

Armenian church architecture, the articulation of the facade by medallions, animal sculpture, blind arcading and multiple mouldings — all executed in stone – transformed the Iranian model. Brick mau- solea of authentic Iranian type were common in eastern Anatolia, while to the west and south forms cognate with the Syrian tradition

flourished. Iranian modes, expressed in cylindrical brick minarets

(Raqqa and QaTat Ja’bar) and decorative brickwork (Raqqa Gate)

were also briefly fashionable in the later twelfth century in eastern

91 (opposite) The mosque as dynastic memorial. Congregational mosque of ‘Ala al-Din, built intermittently between ns> and 1220. in Konya. the Rum Saliuq capital. Its hypostyle or Arab plan, with re-used columns, includes a dome chamber in the Damascus manner, and two royal mausolea. It shows the use of carpets not only for the comfort of the worshippers but also to inject colour into the interior.




«$ \


92 Entrepot for the slave trade. Dock at Alanya ‘Ala’ivva. a city founded m I22fl by the Saljuq sultan ‘Ala al-Din Kaiqubad and exceptionally named after him. It wax a summer resort and a fortified seaport of strategic importance, with an artificial harbour. In this rare example of Islamic naval architecture, with five

brick-vaulted galleries some 40 m (131 ft) deep, large ships could be built safely and secretly.

Syria, while Damascus and Aleppo saw a brief classical revival which

manifested itself in astonishingly accurate renditions ot classical

mouldings, capitals and the like.

More than any other contemporary dynasty, the Sal]iic)s of Rum concentrated their patronage ofthe arts into the medium ofarchitec- ture. And it was not only in the genre o\~ mausolea that they com- bined their own ideas with others taken from Syria and Iran to

91 fashion a new style. The architecture of the mosque, tor example, experienced an absolute transformation. The gable type popularized

by the Damascus mosque, the ancient Arabian hypostyle, the Iranian

\-iwan schema – all found a place in the Anatolian world, but all were subtly changed by a new emphasis cm an integrated domed space. It may well be that the germ ofOttoman religious architecture is to be sought here. As in the case of madrasas and mausolea, the

sheer number of mosques is as remarkable as their variety. The cumulative impression is unmistakable: there was no building tradi-




93 Castles of commerce. Caravansaiai at Tercan, eastern Anatolia, early i.uli century I Ik- plan is of a madrasa with two iwans (hays) m its living accommodation, hut it has additional lateral halls, which served as stahles. Such buildings, usually financed by the state, punctuated the major overland routes .it intervals o\\\ day’s journey.

tion m the entire Near East to rival that of the Saljuqs of Rum in the thirteenth century. This was partly due to the geographical position

of their territories, which made them open to ideas from east, west and south; but a consistent commitment to architecture by the ruling elite made it possible for local schools to flourish mightily, so that even minor Anatolian towns can often boast major monuments in this period – most of them constructed in finely dressed stone.

The consistent state involvement in architecture is seen to best

advantage in the network o\~ kluius or caravansarais which criss-

crossed the country. Many of them have fortifications on a scale better befitting a castle than a stopover tor the caravan trade. I he\

could often hold scores of travellers and hundreds of animals. Ornate

and stately portals give on to an open, arcaded courtyard, rrequend)

furnished with a central raised kiosk to serve as .1 mosque. Cells tor

travellers lie behind the arcades. On the axis of the entrance there may be .1 lofty three-aisled extension oi aspect. I his




94 (kft) The click of castanets. Dancer on a gilded glass (wine?) bowl with the title of the Atabeg Zangi (reigned 1127-46). Cliches

of Islamic (and Byzantine) images

of dancers include the raised right

foot indicating a dance step, the

three-quarter depiction of the

head, the frontal torso, and the

tree. Dancers, like musicians, were

regarded as children of Venus and

thus under the influence of that


9> {opposite) The Turkish image of power. Group of cross-and-star tiles, palace of Qubadabad. 13th century. The disposition of tiles and the firing technique are

Iranian, but the iconography draws

on pre-Islainu Turkish snuru

tombstones) from Inner Asia I he

cobalt used for the blue colour was

worth its weight in gold and was

doled out carefully to selected

potters from the royal treasury

space was used tor stables. Chains of Saljuq caravansarais serviced

particular trade routes. Thus the slave trade from the Black Sea ports

used a north—south artery departing from Sinop and terminating at

Alauva. where the ample docking facilities helped in the trans-

shipment ot the slaves to Mamluk Egypt. Net other routes hooked up with the long-haul traffic from Iran and points east, and also with

Syria and Iraq. About a hundred ofthese caravans, irais have survived, perhaps a quarter ot” the total originally built – telling testimony to

the administrative efficiency ofthe Rum Saljuq state. This Saljuq architecture, built tor the most part of well-dressed

stone and sometimes employing ornamental marble inlays which

suggest Syrian workmanship, has a wonderfully varied decorative

repertoire. The traditionally Muslim predilection tor geometric and vegetal themes is enriched by Armenian interpretations ot such

motifs – for in the largely Christian hinterland of eastern Anatolia, Armenian architecture was m full flower. Figural carving abounds, m flat defiance of standard Muslim practice, and is found on religious

and secular monuments alike. Its themes sometimes reflect Armenian influence, but the dominant impression is of a pagan Turkish




thought-world rather than an Islamic one. especially m the images of lions, eagles and bulls. L3ut elements from the ancient Near East isolar images, fish motifs, the lion/bull combat) and the East Asian

animal calendar also appear, as do the astrological and planetary

images so popular throughout the Middle East in this period Glazed

and plain brick ornament of Iranian type, often executed by Iranians, was also popular; in fact glazed tilework, perhaps developed with the

help of Iranian craftsmen fleeing from the Mongols, reaches a level of

design and technique unequalled in Iran for another century or so.

Figural tiles – mma\ lustre and above all underglaze-painted in blue and white – were used lavishly in the Saljuq royal palaces at Konya, Qubadiye and Qubadabad.

Iraqi architecture of this period is uneven in quality, though its

decoration – whether in carved stucco, terracotta or polychrome inlaid marble – is often splendid. In the area of Mosul there flour- ished, from the later twelfth century onwards, a school of figural

carving which decorated khans (Sinjar), palaces (Sinjar and Mosul).

mausolea (Mosul), bridges (Jazirat ibn c

Umar) and city gates .it Mosul

and ‘Amadiya. Sometimes the themes .ire ceremonial .it Sinjar, for

example, tiers of the ruler’s mamluks, each bearing an emblem of office, flank a throne niche – but animal themes dominate, including

griffins, lions, serpents and dragons, .ill of which seem to have served

an apotropaic function, as on the appropriately named Halisman Gate at Baghdad. Similar designs, augmented by heraldic and astrological

motifs often refer in punning fashion to the ruler himself, appear



1 2




96 (left) The sultan’s slaves. Royal niche from Sinjar. northern Iraq, before 1240. A small baldachin above the apex of the arch marks the axis

ofsovereignty; in the remaining niches plant motifs

alternate with images of beardless figures in military

dreNs holding emblems of office. These are specially selected slaves destined for high office.

97 (below) Sympathetic magic. Apotropaic relief

at the apex of the now destroyed Talisman Gate, Baghdad. 1221-2. Like its accompanying magic-

saturated inscription, it underlines the role of the

caliph al-Nasir as sole divinely appointed head of

the Muslim community; it symbolizes his victory over two internal enemies: the Grand Master of the V- tssins and the Khwarizmshah Muhammad, both of whom died around this time.

c language ofpower. Mustansiriya

madrasA, Baghdad, Inscription dated 63 i_.;_ 3 commemorating the building of the ‘noble madrasa tor the students of wisdom, which brings happiness

to 1 alb creation and us) an illuminated path in the

eye ofGod’ by the ruling caliph and calling down God*S blessings on him. The text is replete with

Our’anic echoes

on Atabegid coins. The same visual language is to be found in con- temporary Syriac churches in the Mosul area (Mar Behnam and Mar Shem’un), and in the Yazidi shrine at


Ain Sifiii nearby.

The Friday Mosque ofNur al-Din at Mosul, completed in 1172, is notable for a major dome over the mihrab surrounded by a battery of vaults in the sanctuary. Several contemporary madrasas are known in Mosul and Wasit, but these pale before two masterpieces of” the

genre in Baghdad: the Sharabiya and the Mustansiriya of 1232. The latter was built by the eponymous reigning caliph on a massive scale as an instrument of politico-religious propaganda, as shown by its



huge external riverside inscription extolling the caliph in letters

almost a metre high. It was designed to serve all four major schools

of Islamic law – a major innovation which turned it into a symbol of Muslim unity and resurgent caliphal power. No standard plan was followed in these structures, though all those in central Iraq are on

two floors and feature monumental portals and rooms opening off a central courtyard, itself often furnished with iwans. The Sharabiya

(formerly known as the ‘Abbasid palace) is notable for its narrow corridors crowned by multiple tiers of steeply stilted muqamas vault-

ing arranged in diminishing perspective towards a distant vanishing

point. These vaults exploit illusionistic devices m an entirely novel way They are covered in lacy terracotta carving of remarkable preci- sion and intricacy whose closest parallels lie. intriguingly enough, in

Central Asia.

The thirteenth century witnessed the first golden age of Islamic book painting, most of it produced m Syria and Iraq, where the major centres were Mosul and Baghdad. The reasons for this sudden flowering are obscure. That it soon spread very widely – to Iran.

Egypt, even Spam – is beyond doubt. To be sure, earlier Islamic painting on paper exists, but the rubbish-heaps ofFustat have yielded

only individual sheets with illustrations ofmosdy indifferent quality, while high-grade illumination was confined to Qur’ans. 1 itcrar\

references prove conclusively that illustrated manuscripts were pro

duced .it the ‘Abbasid and Samanid courts, and that examples of

Sasaman book painting were carefully preserved in southern Iran in

the tenth century. Moreover, astronomical texts had been illustrated

for centuries – the earliest surviving version ol al Sufi’s ‘ I realise on



99 The old gods die hard. Apotropaic frontispiece to 77i<- Treatise on Snakebite, 1199. I his bod written during an eclipse, which was perhaps thought to nuke its recipes tor snakebite more effi< a. ious. In popular belief, eclipses happened when the monster Jawzahr swallowed the sun or moon; hence the personification ofthe moon (ofBabylonian origin) is within the serpents’ stomachs. Note the echoes of Buddhist iconography. 2 1 \ 1 4




i Mark ol nw nership I rontispiei

.1 book from i 2< ‘ volume set ol the

Kitab al-Agham (‘Book ol Songs’) m.

fbi the >iimi oi Mosul. Badi .il I )in

I ii’ln’, [217 [9 I he patron, whose

name appears on Ins .h tnbands, is

resplendent in moiri silk and sable in

.mil. ii) .m unusual adaptation oi stan

enthronement u onography, towers i

Ins .nit like i oui tiers. I lying angels (

Yu ioi ies) hold an honorifi< canopy his head. E< hoes ol Byzantine at l


e if


> hat





the Fixed Stars’ was produced .it Baghdad in [009. No claim can therefore In- entertained tor the chronological primacy ol thirteenth

century painting. But it docs seem likely hook Illustration was onU sparsely practised in earlier centuries in tin- [slami< world, .mil this was .it least m part due to the severely practical and didactic function to which it was confined. The notion illustrations to .i text could be inn seems to have dawned on [slami< .utists only during the thirteenth i entury.

Sunt- few ol the extant illustrated manuscripts are dated, and fai

fewei still .m- provenanced, the detailed history ol tins school oi

painting has occasioned lively debate. I he < m rem scholarly nuistMi mis favours Iraq .is the prin< ipal ( entre ol production, with ateliers .it

Mosul and Baghdad, and .i secondary Syrian school based it

Damascus. Production srnns to have tailed oil dramatically after the

s.u k ot Baghdad by the Mongols iii 1258. The debut of this vigorous and inventive school oi painting is

cK-c idedly low key. Byzantine influent es .in- dominant both in the

choice ot texts, the- subject in.ntei oi whuh is largely botanical and




pharmaceutical, and in the didactic and diagrammatic style favoured

for the illustrations. In this tradition, which continued that of the

classical world, the picture was the handmaiden of the text, although

in some cases – such as the very popular Automata manuscripts, o\ which fifteen copies are known – the pictures were needed to make sense of the text. There was no question of giving them an elaborate

background or frame – indeed, the plain colour of the paper serves as

the background – or of allotting a full page to An illustration.

One important exception to this rule, however, must be noted. In accordance with classical and Byzantine precedent, the frontispiece

used a full-page painting to honour either the author or the patron

of the manuscript – or even both. This practice had classical roots,

themselves reworked in the Byzantine evangelist portraits which may have been the immediate source of the Islamic version. Hence the omission of the author’s muse: hence, too, the gold background and

the white highlights on the drapery. The Islamic contribution is .it first limited to details of costume and architecture. By degrees the

range of options widened to include visual references to the content

99 of the work (Paris Kitab al-Diryaq, “Treatise on Snakebite’, 1199; Rasa’il Ikhwan al-Safa, ‘The Epistles of the Sincere Brethren’. 12N7;

MukliLir al-Hikam wa Mahasin al-Kalim, ‘Choicest Maxims and Best Sayings’) or to the activities o( court life (Vienna Kitab al-Diryaq\

100 Kitab al-Aghani, ‘Book ofSongs 1


Byzantine influence makes itself felt even in the layout of some o\ these frontispieces, which mimic the division of space used in ivoi\

polyptychs. Lavish application of blue and gold, imitating Byzantine

enamelwork and chrysography, to say nothing of iconographic motifs such as angels or victories and the symbolic use of drapery, lend

some ot these frontispieces an unmistakably Byzantine flavour. So

too do drapery conventions and details of costume. Nor is this sur- prising, given that the Islamic paintings associated with northern

Mesopotamia were produced in a predominantly Christian milieu, and that Jacobite painting had a strong impact on them. Where the manuscript was rounded off with a fmispiece, the design would be identical to that ofthe frontispiece (Kitab al-Diryaq, 1 199; Mukhtat al-

Hikam u\i Mahasin al-Kalim), perhaps following the example of Qur’anic design. The layout of all these frontispieces quickly became the vehicle for quite complex messages which had to do with sub- jects as diverse as talismanic protection, the royal lifestyle or scholarly

activity, whether co-operative or confrontational. Yet for .ill their Byzantine flavour, these paintings also looked to the East. Some




facial types, for example, with their slant eyes and heavy jowls, are

familiar from Saljuq lustre pottery, while other figures are best paral-

leled in mina’i work.

Alongside the practical treatises, which maintained their popular-

ity throughout the thirteenth century and even later to some extent, and which gave Arab painting such a potentially wide range, con-

temporary taste also favoured works of literature or belles-lettres whose entertainment value was paramount. Two works in particular enjoyed widespread popularity: the Maqamat (‘Assemblies’) ot al-Hariri (d. 1 120) of which more than a dozen illustrated thirteenth-century ver- sions survive, and the collection of animal fables known as Kalila wa Dimna (‘Kalila and Dimna’). The text of the Maqamat consists of fifty episodes m the career of a con-man, one Abu Zaid, whose trickeries depend on his surpassing mastery ot the Arabic language. Al- Hariri

was a wordsmith who fashioned each maqama so as to exploit the full resources ot the language. Virtuoso linguistic display is thus the

keynote ot the text, which is peppered with quotations, allusions,

puns and obscure vocabulary. It is essentially a thesaurus oi curious

and recondite terms in which action, narrative and drama are of dis-

tinctly secondary importance, and so it gave minimum encourage- ment to the artist. In a sense, text and image are at cross purposes.

But the picaresque framework presented illustrators with the chance

to produce some remarkably varied settings, often with a strongly realistic flavour but occasionally of a fantastic nature too, such as the

images illustrating the adventures of Abu Zaid in the Eastern Isles. The crowning masterpiece of the school is the Maqamat copied

and illustrated by Yahya b. Mahmud al-Wasiti in 1237, now in Pans. 101 Its illustrations throb with vitality Husbands and wives bicker,

plaintiffs harangue judges, drunkards carouse in taverns to the strum-

ming of lutes. The artist favours scenes ot intrigue, fraud, disputa- 102 tion: he loves to group his many figures in tight bunches and is at his best in depicting processions. He relies on precise draughtsmanship and bright colour rather than on modelling or an elaborate land-

scape. The action is always crammed into the frontal plane while the background is the neutral colour o\~ the paper itself. Architectural

settings are rendered with a notable precision although without any

attempt at perspective. Indeed, laborious spatial devices .ire consis-

tently avoided. A few fleshy plants do duty tor .1 landscape and the sky is rarely indicated. The scene shifts from the slave market to a village, from a Bedouin encampment to formal parades or pilgrims

departing tor Mecca.

1 .:<;



– ;

indiloquence rewarded.

This is die message ofdie

Maqamat, whether the speaker

performs in a secular context or

in .1 religious one (as here, m a Baghdadi manuscript of

Here, however, the usual

oratorical posturing; of the

disreputable hero. Aim Zaid, atypicalry serve piety rather

than self-interesl Spotlighted

on a hill, he harangues an

audience of pilgrims, which (in

satirical vein?) even includes

camels Mu< h depends on gesture and the glance m the

mtexl is deftly delineated

\ well worn landscape props Note the mahmal or

palanquin, typically used by

noble ladies tor the Pilgrimage

Such illustrations Jo not depend on any earlier pictorial tradition, and among the main hundred Maqamat pictures of this period no consistent iconographic cycle can he recognized. All this suggests

that the artists held up a mirror to daily lite and found out ot then-

own resources an appropriate visual equivalent tor what the\ s.iw though they might haw been inspired, tor example in their choice <>t

103 the silhouette mode, by the contemporary shadow theatre. Hut older-

Near Eastern traditions also make themselves felt in the Strong outline drawing, the exaggeratedly large eyes and the interest 111

surface patterning which here expresses itself in the technique ot

rendering drapery in convoluted scrolling folds. Given the atrocious

difficulty of the text couched in rhyming prose) one may wonder whether the t\ood of illustrations (about a hundred in some manu- scripts), which resulted in a picture for every two or three pages on




102 Literacy begins early. This

Maqamat image (possibly

Damascus; the tablet held by

the boy in front states that the

manuscript was executed in

1222—3) shows Abu Zaid in the guise of a schoolteacher. He holds a split cane, the

traditional instrument of

correction. Undeterred, his

motley class crowds around

him. with most of the boys

holding a tabula ansata (the

form of which replicates the

writing slate of classical times)

covered with Kufic writing, not

the cursive script one would expect. The teacher’s authority- is suggested by his greater size,

his stately turban and the

honorific arch underneath he

alone sits.

average, was the result of a specific trend in patronage. Clearly it was

fashionable to possess a copy ot this most popular of contemporary texts m the field of light literature; and those patrons whose literary accomplishments were too slight to profit from the text itself could

nevertheless derive enjoyment from the pictures. These, then, arc

some of the earliest coffee-table books. Much the same could be said of the illustrated versions of the

Kalila wa Dimna, a text of Sanskrit origin which had already been

translated frequently into the languages of Europe and western Asia.

Here the purpose of the text is only incidentally entertainment. Its

real purpose is to provide a ‘Mirror for Princes’ through the medium of animal stones whose anthropomorphic quality is only thinly dis-

guised. As with the Maqamat, then, the attractiveness of such manu- scripts was two-fold. Ancient Indian, ‘Abbasid, Sanunid and south

1 \ 1





103 Popular culture. Egyptian river boat ofpainted leather. Mamluk Egypt, 1 sth century. Prop tor a shadow play; the styie recalls MaqanuA painting?. Hie articulated figures were

backlit against a white wall. The contemporary historian Ibn Khaldun noted of the lurks

that they place their archers ‘into three lines, one placed behind the other I h

shoot from a squatting or kneeling position.’

Italian illustrated cycles of these stones cither survive or arc recorded

in literary sources and it is therefore no surprise that in the Syrian and Iraqi versions, too, specific iconographic cycles can be recog-

nized. Bright colours, strong, dramatic profile poses, simple symmet- rical compositions all combine to push the narrative along. 1 ater Mesopotamian painting petered out in a stale imitation ofthe style in vogue c. 1240. though court painting in the Persian manner was

occasionally practised in Baghdad.

Speculation has abounded as to the patrons who called this school of painting into being. The short and unhelpful answer is that none

of these manuscripts specifically names the patron who ordered it, though the ruler who figures m the frontispieces of the surviving volumes o( a luxury Kitab al-Aghani “hook of Songs*) wears tiraz bands with the name of” Hadr al-Din L.u’lu’. The popular nature of the most trequentlv illustrated texts perhaps encourages the notion

that they were produced either directly for members of a well-to-do middle class – that very class whose life is mirrored so accurately in the Maqamal manuscripts – or that they were intended for sale in the open market. 111 the certain knowledge that a steady demand for such




works existed. Such bourgeois patronage would contrast sharply with the courtly milieu in which almost all the host Persian painting was to be produced (see pp. 205-12).

In no area of the visual arts is the flux of cultures represented 111

thejazira and neighbouring areas in the twelfth-thirteenth centuries

more apparent than in coinage. Here a decisive break was made with the long-established Muslim tradition that coins should bear inscrip-

tions only, and not images. The Artuqid and Zangid rulers minted literally scores ot different figural types drawn from a bewildering

farrago of sources. This phenomenon remains basically unexplained. The aberrant issues were confined to large copper coins and were thus intended for local circulation; gold and silver denominations

the latter comparatively rare) would have travelled further afield and

thus perhaps remained strictly orthodox in design and content.

Since the copper coins were of substantial size (up to 36 mm [1.4 in] in diameter), they could accommodate quite elaborate designs. Their reverses customarily bore confidently executed Arabic inscrip-

tions. Figural themes include more or less maladroit copies of the busts of dozens of specific Greek, Seleucid, Byzantine, Sasanian and

contemporary Turkish rulers, standing or enthroned figures of Christ

or the Virgin, and planetary and astrological images such as Libra,

Virgo, Jupiter in Sagittarius, Mars in Aries, figures seated 0.1 a lion

(Mars in Leo?) or a serpent (the constellation Serpens?), the lion and

the sun, and a seated figure holding a crescent, a representation of

the moon. Such images referring to the heavenly bodies are

common in the other arts of the Jazira and were accessible not just to Muslims but also to people of other faiths. This pervasive fascination.

bordering on obsession, with astrological imagery may well reflect, as


104 Malignant planets. Artuqid

coin, Mardin, 1 199-1200,

depicting Mars. Detailed familiarity

with astrological concepts was

expected of a man of culture and is a stock-in-trade of medieval Islamic

literature, especially Persian poetry.

Reporting on the arrival of the

Crusaders on the Levantine coast

in 1096. al-‘A/uni adds laconically

‘Saturn was in Virgo’: in other

words, disaster was imminent.



recent research suggests, the abnormal frequency of eclipses and

other celestial phenomena in this area and period, which must have struck dread into the hearts of those who experienced them. Other themes include horsemen; double-headed eagles; angels and Victo-

ries; and affronted or addorsed heads, sometimes crowned with

eagles. On occasion Roman and Byzantine models turn up on the same coin, or a borrowed type is altered for no clear reason – thus

Herachus has his beard shaved off. Some of the prototypes were well over a millennium old. which armies some antiquarian interest on the part of the mint-master: others were recent Byzantine issues.

They share an indifference to the accurate rendering ot the model; hence their frequently grotesque proportions. Other designs rnajj have been copied from seals or other small objects, and would thus have had to be reworked for use on a com.

What characterizes this body of coins above all is the random, jackdaw interest m the disjecta membra of the past a past which had been inherited by virtue of conquest. Sometimes the design refers in punning fashion to the ruler himself In that sense these unns could be interpreted as a formal pi \\\A. ancient non Islamic coins provided a ready-made numismatic source for royal iconogra-

phy – so perhaps the Turcoman rulers adopted these ancient imperial busts as self-poitraitS. The Christian themes of SO man\ of these coins can be explained by the tact that the population in these terri-

tories was largely Christian and had long been accustomed to exclu- sively Christian copper coinage. As with funerary and architectural

sculpture of figural type, the lesson of these coins is that m a Iurkish context the hand of Islamic orthodox) la) rather lightly, and that artists were accustomed to look very far afield for inspiration.

Medieval twelfth- and thirteenth-centur\ Syrian ceramic wares

include fine specimens of lustre and of undergla/e. the latter typically m turquoise and black ‘ :ns has yielded the most distinctive group, and several glazed apothecaries’ jars can be associated with

Damascus, but the exact localization of the numerous no i ailed ‘Raqqa’ wares, with their trademark silhouette style, is still disputed.

Unglazed barbotine ware in this period, seen to best advantage in the habb or storage jar. often draws on a remarkably tenacious repertoire ot pre-Islamic mystical and apotropaic images. Syria also produced quantities ot glazed three-dimensional ceramic sculpture in this

period – animals, horsemen and even nursing women. Metalwork reached new heights of technical sophistication in the

thirteenth century. It is now generally agreed that some of the artists




105 /eic de vivre. Brass basin, inlaid with silver, used for washing the hands; made for the libertine Avvubid sultan al-Malik al-‘Adil; datable [238-40 and signed by Ahmad al-Dhaki al-Mausili. The upper images, reading left to right, depict a striding falconer, dancing monkeys, acrobatic dancers (one a nude female. unique in Islamic metalwork); man killing lion; below, man spearing winged quadruped, two moufflons affronted, bull attacked by winged lion; man fighting bear. H. 19 cm (7.5 in), diam. 4-. 2 cm (18.6 in).

who fled westward from Khurasan during the Mongol invasions settled in Iraq and that they included metalworkers. The principal centre of production was Mosul in northern Iraq, but the industry also flourished at Damascus and Siirt (Is


ird) in eastern Anatolia.

Mosul was tamed throughout Muslim lands for its inlay work in red copper, silver and even gold, though the technique had been

employed in the Iranian world since the first half of the twelfth

century, as had many of the favoured Mosul themes. The names of several craftsmen have survived in a sequence of some thirty signed or dated pieces which can be attributed to the city and which extend

over the entire thirteenth century. Foremost among them was one Ahmad al-Dhaki who flourished in the 1220s. A hallmark of Mosul work is the intricate background of interlocking T-shapes, while ani-

mated scripts in both Kufic and naskhi scripts proliferated in narrow

hands which compartmentalized the densely worked surface. Other

bands were crammed with figural cycles depicting scones from court lite – music, banqueting, the hunt, mounted combats, enthrone- ments; or similar motifs occupied lobed medallions set against geo-

metrical ornament or interlace, as m the Blacas ewer in 1 ondon.


1 ;s



Bowls, vases, ewers and candlesticks predominated in Mosul metal-

work. The coming of the Mongols enfeebled and eventually killed the industry in its home city: but local craftsmen took their skills elsewhere, notably to Damascus and Cairo, to such effect that it is

sometimes hard to distinguish between authentic Mosul production

and that of these other centres.

One short-lived specialization associated especially with Damascus is represented by a group of eighteen surviving inlaid brasses of

106 extremely elaborate workmanship depicting New Testament narra- tive scenes alongside the standard cycle of princely amusements. This

suggests less the activity ofCrusader patrons though is also pos-

sible) than a new readiness to make such decoration reflect the con- temporary culture m .ill its diversity – a culture in which native Christians had acclimatized themselves and m which refugees from the East were .inning m ever greater numbers. Such mixtures of Muslim and Christian images are also known in Armenia and Sicily, areas where Muslim and Christian lived in close proximity. Certainly many Latin Christians developed, like the Normans of Sicily before them, a taste tor the luxuries of local life; as a certain lonelier

remarked. ‘We who are occidentals have now become orientals’. It would be wrong to interpret such ‘Christian


metalwork as intended

purely tor Christian patrons, partly because of the strong Islamic tenor ensured by their Arabic inscriptions and bv scenes from the

Klainic princely cycle, and partly because of the very way that the

Christian themes are treated. The) contain numerous iconographic solecisms and their layout, with its frequent emphasis on sing) paired figures m arcades, suggests that visual symmetry counted tor more than meaning. Thus the Christian themes were rendered more decorative and less meaningful and their iconographic charge was

defused. The absence of images of the Crucifixion and the Resurrection is a pointer in the same direction These ke\ event! were not part of Muslim belief and their absence helped to make such objects acceptable to the Muslim majority. I his metalwork may even have connoted the subject status of the Christians under

Muslim rule, as the details of hierarchical placing suggest. Such works ot art had something to say to Crusaders, local Christians and Muslims alike: and. although three of the most splendid were dearly made tor Avvuhid rulers or dignitaries, the majority were perhaps produced not to order but tor the market.

Damascus, along with Aleppo, was also a major centre tor the rel.i- tivelv new and technically demanding craft of enamelled glassware.

[ 36



106 Pluralism. Brass hanging canteen inlaid with silver and black organic material. The form derives from pilgrim flasks; its design features units of three. Muslim ornament and inscriptions alternate with Christian themes: the central Madonna and Child, and the Nativity, Presentation and Entry into Jerusalem in the outer /one. Their iconography

reflects Christian Syriac manuscript illustration. Diam. 36.9 cm ( 1 4

. s in).

Here too, alongside the inscribed mosque Lamps produced in large quantities tor the Muslim market, beakers and goblets – and occa- sionally larger pieces like the Cavour vase, midnight blue in hue, and the Corning Museum candlestick – found their way into European ownership. Their decoration of hunting and battle scenes, sometimes

enlivened by themes, all executed 111 .1 style akin to the

Irani.m mina’i wares, ensured them an success.


‘ I




The Mamluks

In a narrow political sense the Mamluk dynasty began in i:s direct continuation of Ayyubid rule m Egypt and the Levant, with the significant difference that power now passed into the hands of the slave soldiers employed by the Ayyubid sultans the Arabic word

mamluk means ‘owned’ . Yet while the notion of loyalty to one’s

master rather than to one’s family was theoretically fundamental to

Mamluk society, the reality often contradicted this, and occasional short-lived family dynasties wielded power in the Mamluk state. Court life revolved around the sovereign, who like his officers –

had begun his career as the military slave of some powerful amir. Fresh supplies or such slaves were procured at regular intervals,

mostly from the Eurasian steppe via the great markets .it the Black

Sea ports. The majority ol these slaves were ethnic lurks. I hus for two and a half centuries the central Arab lands • and the

Levant were under non-Arab control.



ioy {opposite) Nascent orientalism?

Anonymous Venetian ‘Reception qj the Ambassadors’ , after 14SS. set in

Damascus. Tins symbol of East-West

interface is crammed with accurate architectural, heraldic and sartorial

detail: the lure of exoticism makes

itself felt in the emphasis on

extravagant headgear (note the sultan’s

spoked turban) and on non-European

animals like the monkey, camels and


10S [right) “Seek the bounty of your

Lord by trading’ (Qur’an 2:199) ~ the Prophet Muhammad was himself a merchant. Lithograph by David

Roberts. [849. Bazaar of the silk

merchants in Cairo, sandwiched

beween the towering funerary

complex of Sultan Qansuh al-Ghuri, 1”. I 504. High windows leave ample room for stalls below. Such foundations, deliberately over-

endowed for investment purposes (see ill. 111). effectively legalized


On the broader political scene the Mamluks achieved spectacular successes. Above all, it was they who turned back the seemingly irre- sistible tide of Mongol conquest and decisively prevented a Mongol takeover of the entire Near East. Beginning with the first major defeat inflicted on the Mongols by an Islamic army – at


Ain Jalut (Goliath’s Spring) in 1260 – they maintained a steady and successful defence for more than half a century against repeated Mongol incur- sions in Syria and the Hijaz. At the same time they flushed out the

Crusaders and the Armenians of Cilicia from their remaining strong-

holds and thereby established a grip on the Levant which was not to be broken until the Ottoman conquest in 1517. Thus they were able to present themselves to the rest of the Muslim world as the succes-

sors of S.iladin and the upholders of Islamic orthodoxy. In the process,

they confirmed and extended the Egyptian dominance of the Near




East begun in the Fatimid period. In the field ofart and architecture,

this entrenched primacy of Egypt was Mich as to inhibit the develop-

ment of separate, individual styles in the Levant, and to eclipse Iraq. Indeed, when the Mongol invasions created a widespread refugee problem, Cairo was the obvious haven for the displaced craftsmen

from Iraq and Iran. Hence, it seems, the sudden flowering of metal-

work in Egypt and the introduction of glazed tilework, high drums and ribbed domes into Cairene architecture. Only in the far west o\

the Islamic world, sundered from Egypt by thousands of miles of sea

and sand, did an Arab art independent ofEgypt continue to flourish.

The key to Mamluk art is the city ot Cairo. This was quite simply the greatest Islamic metropolis of the Middle Ages, and reduced to

provincial status even such renowned cities as Damascus. Jerusalem and Aleppo. It had inherited not only the prestige o\ HA^hddd. but

also – because it now housed the ‘Abbasid caliph, even though he was no more than a puppet – the religio-pohtical authority that was

inseparably linked to the institution of the caliphate. In that respect

Mamluk Cairo was able to play the central role tor orthodox Muslims which had been denied to it as the capital ot’ the Shfite

Fatimids. Cairo was better placed geographically than was Baghdad

to be the pivot of the Arab world, but by the same token it was o\~

course further removed from the Iranian sphere and the lands still

further east. Nevertheless, its favoured location meant that its hori-

zons opened on to the Mediterranean and thus the Christian cultures

to the north, as well as on to the \KcA Sea and so the trade with India

and South-East \sia Mamluk trading interests embraced the Italian city-states, notably Florence; southern Russia and the Eurasian

steppe, the source ot the slaves on which the M.unluk elite depended; and the trade in spues and other luxuries with India and

points east. For all these products Cairo was the natural mart. This

far-flung international trade helped to make Cairo the most cos- mopolitan Muslim city of its tune. As such it provided the perfect setting tor the tales of the Thousand and One Nights, a text which took shape in the Mamluk period, although its milieu purports to be that of early ‘Abbasid Baghdad. It says much tor the glamour of late medieval Cairo that it could take on the mantle of the most pres-

tigious o\~ Islamic cities.

That glamour was in large measure created and sustained by the

public pomp and circumstance which distinguished the Mamluk court. An anonymous Venetian artist working in the late fifteenth century captured the pageantry ot a Mamluk procession in a painting




(now in the Louvre) entitled The Reception of the Ambassadors. [07 Gorgeous turbans, heraldic blazons, festive hangings and stately

steeds all play their part in a carefully stage-managed spectacle. That

same theatricality permeates Mamluk architecture, which of course provided the setting for so much of public life. The emphasis on facades is especially telling in this respect. Paradoxically, the sheer

quantity of surviving Mamluk buildings seems to haw deterred scholars (with the exception ot Meinecke) from tackling the archi-

tecture of this period as a whole. Individual monuments have attracted close study, but a few detailed monographs are a poor exchange for a full general survey. The last two centuries of Mamluk 1 m rule in particular – in other words, most of the period in question –

have been especially neglected. This means that no general picture

has yet emerged of how Mamluk architecture changed in the course of almost three centuries.

One major factor in these changes was the sheer density of urban development in medieval Cairo. Given that space was at a premium (and this was equally true in, say, Jerusalem), architects not only had

to think in terms of gap sites, with all the shifts and compromises

which that entailed, but also were almost forced to develop their

buildings vertically rather than horizontally. The canyons of modern Manhattan were foreshadowed in the great thoroughfares of

medieval Cairo, especially in Bain al-Qasrain, effectively an extensive

open-air gallery where the buildings ot one royal patron after another vied for space. The resultant emphasis on facades often con- flicts with the basic need for a qibia orientation. Solutions to this

problem became a deliberate aim pursued with increasing sophistica-

tion in a wide range of buildings; and often the aim is to ensure not

that the building should proclaim its Mecca orientation from afar but rather that the facade should blend smoothly into context, leaving

accurate qibla orientation for the interior alone, safely out of sight.

The same emphasis on outward appearance dictates that minarets are subtly designed so as to yield their best only at root level and

above. Open loggias or belvederes break out at second-storey level. In dome chambers, bulks-eye windows – single, double or in groups – lighten the zone below the cupola; powerfully sculptural roll

mouldings accentuate the chamfer marks the extern. il /one ot

transition. Narrow portals streak upwards to explode as in the

Sultan Hasan madrasa – some 25 111 (So ft) up in a vault of sunburst design. Homes are set on high drums and develop a vocabulary all ot their own. full fantasy: a high stilt, rippling ribs like

1 \\



fefcjjll military men Small wonder that their domes resembh

helmets. The larger doi ( liro, fittingly symbolizes his wealth and status. Its interlocking networks ofstellai and arabesque design

synchronize to perfect* in three dimensioi

organ pipes, networks or arabesques or of geometric designs. I hey

became a natural focus of attention and the architects consistently

exploited the fact that they were visible from .ill skies, lower walls are kept deliberately plain so that they function visually as sheer dirts.

If they are given articulation, the effect is again to emphasize their

height – whether by minor salients, pilasters or mouldings. Although the almost continuous building boom, and its

concentration on three or tour building types, imposed a certain

* w-H



sameness on the architectural forms themselves, the architects had plenty of room for manoeuvre in the choice, the type and the placing of ornament. Certain architectural features in particular

became the focus for decoration. F : lat lintels with shallow relieving

arches above proved consistently popular, with the individual stones

or voussoirs of either lintel or arch, and occasionally both, taking

ever more complex and baroque forms. Eventually these interlocking

voussoirs were executed in marble of different colours. The motif of the radiating shell niche, originally a Fatimid device, was definitively

removed from its earlier preferred context of the mihrab, though the

keel arch continued in use for some time. Instead, during the Bahri period (1250—1382) the shell motif was widely employed all over the

exterior of a building and now developed into a multi-framed

flattened muqarnas composition.

Fenestration takes on a new importance. Mamluk architects favoured long narrow windows with gridded screens to provide external articulation and also to reduce and modulate the play of

light within. Windows animated, for example, the base of a dome and the zone of transition. In a single building – such as the itinerary

khanqah of Baibars al-Gashankir – they can vary continuously

110 ‘The best protection tor the community’s money is the community itself’ said the Prophet. I lence the small domed treasuries in some mosques, as here in the Great Mosque at llama. Syria, originally a church (8th—14th centuries). This example – probably Umayyad – employs re-used Byzantine columns. The square minaret derives from local pre-IsLmiic Christian bell-towers.



m l I l {left) The sultan’s turrets. Funerary madrasa of Salar and Sanjai

al-Jauli, Cairo, 1303-4. The funerary complex, a charitable endowment, served to keep wealth in the family;

thus relatives tilled key posts and

shared in surplus revenues. The Mamluk obsession with rank and status resulted in domed private mausolea hijacking the public

madrasas which were ostensibly the purpose of such foundations.

posite above) Luxury flats tor

medieval merchants. The wakala or

khan of the Mamluk sultan Qansuh al-Ghuri, Cairo, IS04.-S. Its revenues

serviced his funerary madrasa next

.ivH>r 1 he lower two stories served as

warehouses and workshops, with a

gallery for easy circulation; the upper

part consisted of apartments tor rent,

each on three fk»

posite below) Antiquarianism

with an agenda Mausoleum of Sultan Qala’un, Cairo, 1284 s I his

interior, opulent even by Mamluk standards, echoes the porychromy

and varied media and textures of the id monuments of I >amas< us

and Jerusalem, and quotes from them

the octagonal plan, vine scroll baud

and arcaded mihrab Is Qala’un

claiming something *>t their san< titv

and thereby legitimizing his dynasty?

1 13

in both scale and shape. Some o( their tonus – such .is the double Lancet window with .1 crowning oculus – may evolve from contem- porary European st\les. Aside from their function of providing light

and ventilation, they operate .is black voids in .1 blank external wall,

and thus animate otherwise dc.u\ space. This contrast between a plain

expanse and some form of articulation is .1 favoured device of Mamluk architects. It can best be appreciated in the densely carved openwork medallions which so frequently garnish Mamluk facades, such as that of the funerary madrasa ofSunqur Sa’di. New forms of window grille, usihl: the time-honoured vegetal and geometric modes, made their appearance as in the mosque ^\ Baibars or the Qala’un complex), often framed by inscription bands. Similar

epigraphic friezes girdle the bases of domes and even of columns.




ll Ml ‘III]



In the Burji period (1382-15 17), however, it became common practice to employ the wall as a neutral surface for panels of elaborate

ornament like pictures in a gallery. Since much of the colour in Mamluk interiors was provided by polychrome marble inlay and not principally by more perishable materials, its effect can be measured to

115 this day. Integral to the overall impact was the use of ablaq (literally

‘piebald’): marble used in bands of contrasting colours, for example

horizontally along a wall or vertically in the sorhts or undersides of an

arch. It was handled with consummate virtuosity in mihrabs to produce explosive radiating designs, and was ideally suited to parade

the complexities of interlocking voussoirs. Fresco, mosaic, enamel

and stained glass widened still further the range of colour and texture.

Mamluk architectural decoration is distinguished not merely by its strong sense of colour but also by its pervasive sculptural quality This

finds expression, for example, in multiple mouldings with intricate

and profuse detailing, but most of all in the enthusiastic application

of honeycomb muqarnas vaulting to surfaces suitable and not so suit- able: portal domes, niche hoods, squinches, mihrabs, and tier upon

tier of cornices on minarets. Whereas m other Islamic traditions, tor

114 A mosque in miniature. Minbar in die mosque ofQijmas il Ishaqi, Cairo. 1479 si rhe doors serve ai

entrance portal, the steps as die

principal qibU axis, the arch at the

top a miniature mtlimb surmounted

by an elaborate muqamas domical

construction I Ik- imagery o( Light,

suggesting spiritual illumination, is

pervasive ivory, gilding, steUai and

sunburst designs, and lamps galore.

1 1 > I hieves stealing from thieves’

was how the i sth-i/enturv historian 1/1 des< ribed 1 ontemporar)

( lairene patrons ofan hitecture; the

dearth of line materials forced them

to earlier buildings. Here,

the salient facade ofthe nunlhhii.

mosque, mausoleum and khanqah of

Sultan Barquq, 1386, employs two-

tone marble veneer and a bronze

oculus grille.



:#***** p m p> m •’ » * •



n6 In the shadow of the pyramids. The Egyptian obsession with death resurfaces in new guise under the Mamluks. The funerary complex (mosque, mausoleum, nuuhusa, khanqah) of Sultan Inal. 1 4. s 1 — < « . in the so- called City of the Dead in Cairo, flouts the Prophet’s insistence on modest burial.

example in Iranian nlework or Iraqi carved terracotta, the applied

ornament within the individual muqamas cells in of major visual

importance, in Mamluk buildings their sculptural role, intensified by stark contrasts oflight and shade, is paramount.

The presence of so main major buildings in Mich a small space is not confined to Bam al-Qasrain. It is repeated in the area of the Eastern Cemetery and. on a lesser scale, in other parts of Cairo. More to the point, perhaps, it recurs m Tripoli. Damascus, Jerusalem and Aleppo. The sheer quantity of Mamluk architecture tells its own story. Buildings tend to make their impact en masse rather than indi- vidually. Each draws on its neighbours – in tact they frequently blend

into each other. Altogether more than a thousand buildings of Mamluk date survive in the Near Fast, and they pose insistently the question ‘why?’ For all that such buildings perform a notional func-

tion – prayer, burial, teaching, accommodation – the fact that time and again they cluster close together in areas where they can be assured ot the greatest public exposure (notably the area of the

Haram al-Sharif and its surrounding streets 111 Jerusalem) betrays quite another motivation. This building activity was fundamentally

competitive. Of course it also had an economic function, in that it allowed an amir to sink his money in an enterprise whose charitable status protected it from confiscation or a later takeover, but which




could nevertheless benefit his descendants as well as serving the

wider public. Above all, though, the building of such monuments was expected of a member of the Mamluk elite once he had reached a certain position. By so doing he joined the club. And he was also playing his part in ensuring that the dominance of the Mamluk elite was well understood by the average citizen.

Hence, perhaps, the emphasis in Mamluk architecture on those individual elements of the design that have the most direct impact:

domes, portals and minarets. Again and again it is these elements that

dictate the entire aspect o\\\ building, as if the rest of the monument were of merely secondary importance. This may help to explain the modular nature of so much Mamluk architecture and why the more interesting monuments are those specifically designed for an unusual, often prestigious site or purpose, or those m which the architect has had to grapple with an unfavourable setting – say a gap site – or has tried to accommodate in one structure the divergent axes o\~ the street and the qibla.

The obsession with hierarchy and status was carried further in Mamluk society than anywhere in the medieval Islamic world. Presumably it had much to do with the thorough militarization of the ruling elite. Whatever the reason for it, the result was to per-

meate the art of the period with references to official rank. At one

level this was achieved by epigraphy Mamluk calligraphers found an ideal objective correlative to the class-conscious and rank-dominated

court in the mannered script which they developed for official inscriptions. Form matches content to perfection. The inscriptions which unfold so majestically and rhetorically across the surfaces of”

hundreds of metal bowls and dishes, ceramic vessels, glass lamps and.

ot course, buildings, often list a lengthy protocol of titles held by

even quite minor officials. Their impact is intentionally cumulative, and this again is as true visually as it is of their meaning. Formality

and discipline are of the essence, and are indeed taken to extremes.

These inscriptions seem so designed that their massed uprights stand

at attention like soldiers on parade – no mean teat w hen one consid- ers the uneven distribution of vertical letters in a given piece of

Arabic prose. It is a tribute to the flexibility and resource of’ the

Arabic alphabet it cm express such rhythmic power. The effect could be likened to a drum roll or a fanfare of trumpets. Moreover, the nouns and epithets which constitute these titles .ire carefull) bal

anced not just for their meaning but also tor rhyme and rhythm,

assonance and alliteration. A typical sequence might read and it

i [Q



helps to utter the words in Arabic, so as to transmit at least some flavour of their grandiloquence:


\zz li-maulana al-sultan al-Malik al-

Nasir, al- c

alim, al- e

amil, al-mujahid, al-murabit, al-muthaghir, Nasir al-

Dunya wa’l-Din (‘Glory to our Lord the Sultan, the victorious king,

the learned, the diligent, the holy warrior, the warrior on the fron- tier, the guardian of the frontiers, the protector of the world and of

the faith’). Such inscriptions are of course intended not merely to

117 inform. They boast. They assert ownership. They advertise power, and often – if the object they decorate has a religious purpose – piety as well. And they naturally function as ornament too. Not surpris- ingly, such inscriptions were widely copied – in Nasrid Spain, in Central Asia, southern Iran and even Sultanate India.

Rank was expressed in Mamluk times not only by inscriptions – as had long been standard practice in the Islamic world – but also by a new device: the blazon. Like the distinctively-styled official Mamluk epigraphy, the blazon functioned as a logo of possession and

identification, and was almost as pervasive. Indeed, within half a

century of the appearance of the first such blazon in Egypt, its stan-

dard form – a circular medallion with a thick horizontal strip at the centre — had already been adapted to carry epigraphic messages. The commonest of these was ‘Glory to our Lord the Sultan’, but it soon became common practice to tit into this same format an abbreviated version of the sultan’s titles or a reference to him by name. Thus was developed the epigraphic blazon, perhaps the single most defining

characteristic of Mamluk art The blazon, then, tunctioned as a kind ot livery and was encoun-

tered very widely m the Mamluk domains, and even beyond. Thus the Mamluk amir Qarasunqur. who suffered political disgrace and had to seek asylum in Iran, nevertheless saw to it that his mausoleum in Maragha bore the emblem of his long-defunct rank as polo- master. Not all of the symbols employed have been fully explained, but there is general agreement on the meaning ot most ot’ them —

not, as it happens, because ot detailed explanations in literary

sources, but because a given blazon is often accompanied by an

inscription identifying the official in question. The remarkably rich and detailed historical sources covering the Mamluk period make it possible to put together quite a full biography of many high officials, and both to trace and to date the various promotions of their careers.

Coins are a very useful check to such sources, since they are strictly

contemporary documents, and many of them bear blazons. Considerations ot ready legibility, easy reproduction and symbolic




expressiveness ensured that the designs of these blazons were kept

simple. It is instructive to note that these blazons postdate the first

European coats of arms, which the Muslims may well have encoun- tered as early as the First Crusade (109s onwards). By the early twelfth century European powers were using .is emblems the lion, the fleur-de-lys and the eagle. Certain similarities of design, such as

the round or shield-shaped cartouche which enclosed the emblem proper and made it a blazon, and the division of the field into separ- ate segments, seem to support such a connection, as does the use o\’

certain animals, for example the lion that was the personal totem oi

Sultan Baibars, or the double-headed eagle associated with Nasir al-

Din Muhammad. Similarly, a rosette was used for two centuries as the dynastic emblem of the Rasulids in the Yemen. A high-stemmed cup indicated the butler, a napkin the jamdar or master of the

wardrobe, paired polo sticks the polo-master, a bow the bunduqdar or bowman, a sword the silahdar or sword-bearer, a fesse (a plain three- fielded shield) the courier, a crescent or horseshoe the stable-master,

a ewer the quartermaster, a round table the royal taster and a pen-

box the secretary. Other devices included the mace, the banner and the drum, all connoting specific offices.

These logos are almost exclusively the preserve of the nobility; the

sultan himself used an inscribed roundel or shield as his emblem.

However, neat as these definitions seem, they should not be taken .it

face value, for the evidence of Mamluk copper coins and ceramics demonstrates beyond question how indiscriminately these images were employed – whether as emblems of authority (e.g. the eagle with wings displayed), as specific blazons, or as mere decorative

motifs. The random way in which they are combined points to the same conclusion. A similar debasement can be traced in this period in the use of titles, so much so that the lengthier and more high- sounding the title is, the lowlier the rank of the person who claims it. What is new in these devices is that blazons were used as emblems and identification tags of official rank – the office rather than the

man. As such it was appropriate for that blazon to be used for every-

body and everything within the household of the amir in question.

Hence the sheer ubiquity of blazons in Mamluk art – and their effect was no doubt intensified by the importance allotted to colour in

their design (indeed, the Arabic term for blazon is rank, meaning

‘colour’). Sometimes the enclosing shield is subdivided and holds

Several emblems, thus functioning as a composite blazon. Sometimes

such blazons were used collectively by all tin- slaves or mamluks ot .1

I si



sultan. These various distinctions all reveal a society obsessively con-

cerned with rank and status.

Aside from architecture the major art form in the Mamluk period was unquestionably metalwork. Many hundreds of pieces are known; probably the social system of the Mamluk military elite, which favoured a complete service of objects as part of the appropriate

ambience of an amir, offered a powerful impetus for their produc-

tion, and ensured a steady demand. Hence, no doubt, the predomi-

nance of pieces bearing lengthy official titulatures. The sudden efflorescence of elaborately executed metalwork m a region which appears earlier to have lagged well behind the best work of the age, as exemplified principally by the schools of Herat and Mosul, sug-

gests that strong influences from abroad revolutionized the local

situation. To what extent this change was wrought through imported

techniques or through an influx of large numbers of actual craftsmen

from Iraq and Iran is a matter ofsome dispute. When Mamluk metalwork is considered as a whole, the immedi-

ate impression is one of mass production, and thus inevitably of a

decline in quality in comparison with earlier metalwork in the

eastern Islamic world. This impression is based on a scries of inter-

connected changes, embracing shape, material and technique, as well

as the sheer quantity of surviving pieces. In comparison with

twelfth-century metalwork. the range of shapes is now drastically reduced. This is not to deny that a wide variety of tonus can be

found m Mamluk metalwork. but the overwhelming majority of pieces falls into a tew well-defined categories: lamps, basins, candle-

sticks, dishes, h now became common practice for the more popular pieces to be cast. The material changed too. From the thirteenth century, brass began to replace bronze. In itself”, this might not be

regarded as a significant innovation, but a concomitant veering of

fashion away from inlay work resulted in a much reduced chromatic range. In place ot multicoloured inlay work, metalworkers turned to

engraving and thus produced monochrome metalwork. It seems probable that short cuts were taken in technical matters

too. The preference tor huge inscriptions as the principal decorative accent, their relatively uniform style, and the tact that most of them comprise formulaic sequences, probably encouraged the use of tem- plates, whether in thin metal, paper, leather or other materials. Thus quite elaborate objects could be executed m a relatively short time. But it was also inevitable that the metahvorking industry would succumb to staleness and repetition, especially at the lower end of

i s:



the market. Here thin-walled single-metal wares prevailed, then-

vegetal or geometric engraved ornament ofsomewhat restricted type setting off the dominant inscription band. Sometimes this ornament adopted Far Eastern motifs like the lotus or the peony, a reminder

that the Mamluk domains provided a ready haven tor refugees from the East. But these were little more than cosmetic changes; they did not herald a thorough sinicization of Mamluk decorative vocabulary The best Mamluk metalwork is ot course an entirely different

matter and is a worthy continuation of” the Mosul school. Such

pieces as the incense-burner ot Muhammad b. Qala’un, the pen-box of Mahmud b. Sunqur, the mirror made for Amir Altunbugha, and [19 above all the three works signed by or attributed to the craftsman

Muhammad b. al-Zain – notably the Baptistere de St Louis – invite 11.x comparison with the very best of Islamic metalwork. Such pieces,

made for sultans or high amirs, display dazzling technical skill, fre- quently innovative shapes and above all a capacity to cram the

worked surface with all manner of designs. In the works of Muhammad b. al-Zain, the technique of inlay is pushed further than ever before and placed at the service ot a wonderfully fluid and

ingenious pictorial composition. Hunters, grooms, animals and

vegetal scrolls intermingle and overlap with no sense of strain.

Surfaces are carefully differentiated by modelling and hatching –

birds’ feathers, animal tun the scroll-folds ot a tunic. Preternaturally

elongated salukis, leopards and other animals prowl along narrow

borders. Facial features are rendered in sufficient detail to allow

ethnic distinctions to be made, and figures adopt a variety of poses:

they turn to speak to each other, bend their backs to shoulder a

burden, look up or down – and all this in the context of music- making, banqueting or taking part in a ceremonial procession.

In such pieces, metalwork begins to take on the lineaments ot”

painting. The purpose of the design is apparently to capture the spirit of court life rather than to tell a particular story; as such, it replaces

the benedictory inscriptions so often contained within similar bands.

Whether a deliberate pun is intended – whether the serried upright

figures acting out their privileged lifestyle are intended to evoke the

rhythmical sequence of upright letters m inscriptions referring to that same lifestyle – must remain a matter of speculation. It is perhaps

more likely that the manipulation of epigraphy to create .1 succession

of tightly massed uprights deliberately conjured up the image oi a

protective hedge surrounding the property of an amis or a sultan, tor

all the world like .1 verbal bodyguard. Earlier animated inscriptions



117 (kfi) Epigraphic overload. Brass

hexagonal table inlaid with silver, dated

1328 and made (presumably in Cairo) by Muhammad al-Sankan (?). As full of official inscriptions as ill. 1 18 is empty of them, it is a tour deforce of epigraphy, with

the words ‘Glory to our Lord the Sultan,

al-Malik al-Nasir Muhammad’ repeated DO less than $4 times.

1 18 (below) The supreme masterpiece of Mamie metalwork? Syrian?, c. 1300; brass inlaid with silver and gold. Proudly signed

six times by its maker Muhammad b. al- Zain, the ‘Baptistere de St Louis’ was

employed until 1856 to baptize infants of the French royal family. With remarkable narrative flair, the roundels show rulers hunting or righting, while Hanking panels

show the aftermath of each activity. Soon

inscriptions replaced these witty and lively

figural scenes.

• posih I he ruler as cosmocrator.

Bronze mirror with gold and silver inlay

made by Muhammad al-Waziri for ‘Ala’ al-Din AJtunbugha (d. 1 J42), viceroy of

Syria and cup-bearer to Sultan Nasir al-

Din Muhammad It symbolizes the universe note the planetary and zodiacal

signs ruled over by the sultan, w hose-

central epigraphic ‘image* and radiating

inscription have solar associations, like the

rosettes and outermost r.i\ I



indicate that the visual connections between animal or human bodies and the letters of the alphabet had been thoroughly exploited a hill

century earlier. Such puns would not be the only imaginative use of inscriptions in this period. In later Mamluk times, tor example, the severity of the tightly-packed thulth inscriptions was offset and light-

ened by changes in the upper epigraphic storey, for example by

devising pincer-shaped terminations for the shafts, or intercalating a

second inscription half-way up the forest of shafts. The radiating inscriptions found on some of the best Mamluk mctalwork irre- sistibly evoke the image of the sun, an ide.i driven home by the use ot gold inlay for the letters. And the phrase ‘Glory to our I ord the Sultan


at the centre of the sunburst harnesses such solar imagery to

the glorification of the ruler.

I ss



Yet other pieces employ a well-worn visual vocabulary of solar,

lunar, astral, planetary and astrological images of talismanic intent.

The lavish use ot gold and silver inlay not only renders such pieces more precious but is also singularly appropriate for such themes. It is characteristic of these more ambitious pieces that they operate on several different levels both visually and intellectually, with for

example both Luge and small inscriptions on the same piece, or

major and minor themes.

The Mamluk metalworking industry by no means followed a con- sistent development. On the contrary, it had its full share of ups and downs. One may cite the sudden fall in the production ot brasses between c. 13S0 and c. 1450, or conversely the lavish output under

Sultan Qa’itbay (1468—96). Economic factors may have played a significant role here – shortages ot’ the more expensive metals, a ruthless quarrying ot precious materials from earlier buildings, a new

ingenuity in making a little metal ^o a long way (for example,

designing doors so that they contain roundels and other isolated ele-

ments in expensive metals which are attached to cheaper materials).

The increasing scarcity ot copper and silver meant that objects and accoutrements in these metals and ot” course gold) disappeared

almost entirely, and the metalworking industry also suffered exten-

sively as a result of continuous inflation which peaked between c.

1394 and c 141C); its effects were exacerbated by armed conflicts

between rival Mamluk groups, [amines, plagues and disastrous tires. and by the Mamluk government’s insatiable demand tor silver and copper to mint the coins needed to buy new mamluks when war was decimating their supplies ofmanpower. Mamluk glass is closely related to contemporary metalwork in the

vocabulary of its decoration: it favours the same heraldic motifs and

epigraphic style. At the top end of the market, namely enamelled

glass, most of the known production was ot mosque lamps – with the odd exceptions provided by beakers, perfume bottles, bowls or

candlestick bases. Here Qur’anic inscriptions – especially 9:18, ‘Only

they shall enter (iod*s sanctuaries who observe the poor due’, the verse which occurs more frequently than any other in Islamic archi- tecture – supplement the usual parade of titles. Once again, piety and

120 propaganda converge; tor when lit, these lamps would have blazoned forth not only the word of God but also the names and titles of the great and the good. In this medium too. patronage was effectively confined to the ranks ot sultans and amirs. In form and technique

Mamluk glass follows traditions established in the Ayyubid period. A




particular speciality was dark blue or purple glass with trailing designs

in brilliant white.

Easily the largest body of late medieval Islamic textiles are those

from Mainluk Egypt. This need not necessarily reflect the actual rate

of production – for example, the historian Abu’1-Fida mentions that

Abu Sa c

id, the Mongol ruler of Iran, sent the Mamluk sultan a) Nasir Muhammad seven hundred precious textiles in 1323, which argues a well-established luxury textile industry 111 Iran at that time, as indeed

contemporary literary sources confirm. The Mamluk textile industry was under constant pressure from abroad. Until the late fourteenth

century, it dominated the Mediterranean market, but a mere fift\

years later, undercut by the products of Spanish. Italian and ( hinese

weavers, it had suffered irreversible decline. Yet the fact that these

weavers copied Mamluk textiles so closely is a incisure o\~ the high international status ot the Mamluk textile industry In Europe, for instance, textiles with Arabic inscriptions – no matter what those

inscriptions actually said – were honorific objects. As such they are

frequently encountered in early Renaissance paintings; thus the robe

of the Virgin Mary or the haloes of saints bear official Arabic titula- ture. This is also why so many ot the finest Mamluk textiles have been preserved in Western cathedral treasuries, and were used tor the

shrouds ot European monarchs and for ecclesiastical vestments.

Europe also provided a ready market for Mamluk damasks, probably made (as the name indicates) at Damascus. Such terms as fustian, cashmere, mohair, organdy, taffeta, tabby and muslin also point to the

Islamic (sometimes Mamluk) origins of these fabrics. The single overriding problem in the study of Mamluk textiles is

that of determining whether a given piece was actually made 111 the Mainluk domains, and if so whether .it Cairo or at other major

centres of production such as Asyut, Alexandria or Damascus. Yuan

textiles made for the Mamluk market have irreproachably accurate and appropriate Arabic inscriptions but may reveal their Chinese origin by motifs like chi’lins, dragons, phoenixes and turtles.

Otherwise they can be recognized as Chinese only by their different

style and technique. They may even be signed by craftsmen (such as

a certain c

Abd al- c

Aziz) with Muslim names. line textiles were

widely used as a 111. irk of rank or office – for example, a different

textile was hung behind the seat of each member ot the council ot state – and were often used to drape objects m other materials, and to decorate or partition architecture and its spaces. Promotion within

the Mamluk hierarchy was often rewarded b\ the gifi o( .1 set ot




iii ‘Patience is the blessing of al-Nasir; everything has its appointed end’ proclaims this block-printed linen. Egypt. 14th century. Its decorative repertoire of epigraphy, whirling rosettes and stars is all derived

from more prestigious metakvork.

garments, and elaborate ceremonies involving a complete change of

wardrobe, literally from head to toe (for both silk caps and silk slip-

pers with royal titles have survived), marked the beginning of spring and of autumn.

The range of patterns and motifs was very wide, but certain types recur with such frequency that they can be taken as a trademark of

the period – for example, ogival or otherwise curvilinear lattice

designs, sometimes made up of inscriptions, multicoloured sequences of narrow horizontal bands or vertical stripes containing inscriptions,

floral motifs, or roundels with animals and repeated tear-shaped blos-

soms often alternating with lotuses, peonies or other Far Eastern

flowers. Blazons are often separately sewn on to otherwise finished textiles. A remarkably high number of of fine Mamluk textiles bear inscriptions mentioning sultans or amirs, often woven separately in thin strips and then applied to garments such as sashes or turbans.

The migration of courtly themes to humbler milieux is illustrated b\ the popularity of block-printed linens and cottons (a technique 1^1

perhaps borrowed from India) decorated with the ihuhli Inscriptions

that were so popular elsewhere 111 Mamluk art. for the most part

I lis I lght is as a im he wherein is a Ian : reads the ins, ription on the ne» k ol

this glass lamp with enamelled decoration mad quzrimur al-Han of Sultan Muhammad b Qala’un Hisbla im as the cu|



these inscriptions refer to rulers or high officials, or are benedictory,

but sometimes they spell out proverbs (‘Patience is blessed with

success and everything is rewarded’) or apostrophize the viewer (To whoever looks, I am the moon’). Similar sentiments are encountered in contemporary Nasrid textiles in Spain. Indeed, it seems very pos-

sible that textiles were the source for certain design conventions

found in other media. Mich as the division of the field into bands

as found in metalwork. glass and ceramics. The repeat textile patterns 137 found on metal animal sculpture such as the Pisa Griffin exemplify

this. Here, then, is further evidence tor the primacy of the textile

industry in medieval Islamic art.

Finally, Mamluk Egypt is tamed tor a unique type of rug whose production can be documented only for the very end of the period and may have been introduced by refugees from the Qaraqoyyunlu court in Iran after i4/>~. Their colouring of crimson, lime green and

pale blue is unmistakable, as are their designs, usually dominated by a

123 central radiating stellar form within an octagon, with further

octagons wheeling around the periphery. The kinship of such compositions with the astral character o\ the frontispieces to Mamluk Qur’ans or mosque doors leaps to the eye. The binders are usually taken up by linked circular or oval cartouches, a disposition familiar

in Iranian Qur’ans. Two of the known Mamluk carpets bear blazons, but none are inscribed, a minor curiosity given the dominance of

epigraphy in Mamluk art generally. Their manufacture continued long after the Ottoman conquest. Mamluk pottery has remarkably little in common with its Fatimid

predecessors. I ustre becomes much rarer, though there was a strong demand tor it in western Europe until production ceased around 1400. Instead, contemporary fashion favoured underglaze wares of

predominantly blue and white tonality, broadly derived from Chinese

.on domestic pottery

reflected the obsession with st.itus.

This coarse, heavy red-bodied ware

was covered with white engobe,

incised with bold, somewhat

playful tludtli inscriptions, painted

in coloured slips And finally glazed

– here, m yellow. An Egyptian Speciality, it copied metalwork

shapes and often bore annral

blazons and titles.



• firmament on the floor Mamluk carpet, late 15th century The lustrous silk and sheep

I, the dominant intense red palette and the radiating solar and stellar designs which find parallels in

Mamluk marble fl< ns and stoi md also Buddhist mandalas ch

carpets generally



porcelain, as in the many Burji Mamluk hexagonal tiles also made in Syria. Frequently the wares are divided into radiating segments, while

vertically striped or multitoil designs are also common. Creatures borrowed from the Chinese repertory, such as geese and ducks, also make their appearance, and Yuan celadons were widely copied in the Bahri period. The impact of Chinese Ming ceramics waxed ever stronger in the later Mamluk period. A popular category of glazed yellow and brown sgraffito wares, presumably mimicking the more

122 prestigious metalwork of the time, was mass-produced for the amiral

market and displayed blazons and inscriptions giving official titles.

These seem to have been made m sets. The Fatimid penchant for signing ceramics continued apace; some thirty signatures have been found on Mamluk wares. I >amascus, it seems, bade fair to rival Cairo as a centre of production, especially tor underglaze wares. The work- shops there, to judge b\ style and the references to Tabrizi craftsmen,

seem to have derived inspiration from Iran, as did the masters respon-

sible for the tilework in early Ottoman Bursa at the same tune. In no medium of Mamluk art is the evidence tor the strength and consis- tency of the trade links with the Far East clearer than it is in ceramics.

The art of the book m Mamluk times presents a fascinating paradox. Secular book illustration languished. No single text cap- tured the imagination ot contemporary patrons in the way that the

Maqamat had done in thirteenth-century Iraq see pp. \z^ ;i or that the poems ot Firdausi and Ni/ann were to do m [ran ‘see p. 224). Mamluk painting b\ and large is the neglected handmaiden of the didactic text which it accompanies, whether that text is eon

cerned with military exercises fumsiya), annual lore or automata.

Admittedly, some Maqamat texts were illustrated in Mamluk tunes, but they are cleark at the tail end of the artistic ferment which had

earlier generated the Iraqi school of painting, and the same goes tor

ne, women and song. Islamic u ene (defying the Islamii

prohibition ofalcohol) from the Maqamat

probably made ‘ types reflect the

fashion tor Far Eastern ideals ofbeauty.

tocrati< leisure.

Frontispiece to the Maqamat ofal-Hariri,

1337 Produced tor one N.isir al-Din

Taranta’i, probably in Egypt, its

iconography is of Persian derivation

(compare the 1307 Kalila wa Dimna double frontispiece), and implies .1 (lost) facing

page with an enthroned monarch. Note the tamed cheetah leaping from the hunter’s horse before chasing the prey.

Female musicians perform below.




illustrated versions of the Kalila wa Dimna. A peculiar stiffness invests Mamluk figural painting and even the depiction of annuals betrays a quality of rote not entirely disguised by bright, cheerful colours. In

secular painting, then, the Mamluk realms can fairly be described as a backwater; this work significantly lacks that Far Eastern element which energizes Liter Mamluk pottery and tnetalwork and which was of course so consistent an Inspiration for Iranian painters.




Yet side by side with this uninspired, run-of-the-mill work was produced the most consistently superlative sequence of illuminated

Qur’ans in the history of Islamic art. This contrast speaks volumes

about the nature ot patronage m the Mamluk period. The manu- scripts containing secular paintings are for the most part anonymous.

The Qur’ans, on the other hand, very frequently bear the names o\ 26 sultans and high amirs. Like so much of Mamluk art, then, this

patronage had a public dimension – for such Qur’ans were com- monly donated to mosques where they could be displayed. Indeed, such was their size – often more than a metre in height – that they could only be read when displayed on a lectern. It was customary, moreover, for a patron to endow a religious foundation with an appropriately splendid Qur’an, and given the building boom in Mamluk times this ensured a steady demand for such luxury copies of the sacred text. Like so much Mamluk architecture, they served to proclaim the patron’s piety. Some sultans gamed still further renown by copying out the Qur’an m their own hand. While most Qur’ans were transcribed m a single volume or m two-volume sets, it was common practice at this time tor the text of the grandest Qur’ans to



126 opposite) Sultan ‘Sha’ban . . .

has bequeathed all this Noble

Qur’an as a legal true bequest to

find favour with his Lord’ reads the

colophon to this single-volume

Qur’an. Illuminated by Ibrahim al-

Anndi. Cairo, \,~2. Subtly

distorted axes ensure that this

portion of an apparently infinite

pattern fits harmoniously into the

available space. The overall effect recalls contemporary enamelled


ight) The Opening. In the

1372 Qur’an. the first double-page

spread of text incorporates all of

Sura i . the Fatiha, written roughly

twice as large as the rest of the text.

There are three lines to the page.

in black muhaqqaq jali script

outlined in razor sharpness with

gold, and gold roundels as verse





– -Ml


be transcribed in thirty volumes; often each page of text contained

no more than three lines. These would be enclosed by cloud-like tonus which themselves were sandwiched between vegetal scrolls;

and this entire field would then be enclosed by a continuous braided gold band, with Kufic captions in elaborate cartouches at the top and

bottom of the page. Additional roundels or extra borders animated

the outer edge. All this elaborate ornament would be laid upon an

otherwise empty page; indeed, sometimes more than sixty per cent

of a given page would be empty. Thus the ornament, with its pre-

ponderance of gold and blue, would gam maximum eclat from the dull ivory of its setting.

These Qur’ans, however, also epitomize the art of the- period in

their formality and conservatism, and in the way that they echo other




crafts and techniques. Their great frontispieces and finispieces, for

example, are less carpet pages than great doors which swing open to reveal the sacred text and which solemnly close the book. Their

design is often essentially identical to that of mosque doors in metal- work or inlaid wood. But the predominant theme of these pages is almost always a geometric framework of centrifugal designs which

explode with astonishing energy from the central figure. It is hard

not to read such designs as references to the heavenly bodies, espe-

cially in view of the prevalence of gold and lapis lazuli, and the sacred

nature of the text. That text itself frequently employs a stately thulth

127 or muhaqqaq script developed m a manner akin to the inscriptions on metalwork, while Far Eastern flower motifs proliferate, as the) do on

fourteenth-century Mamluk metalwork. Occasionally the colours mimic the tones and the effect ofcontemporary enamelled glass.

Since such steady, long-term patronage was available from the

highest in the land, it is not surprising that several schools developed

and that the careers of certain craftsmen, such as Sandal, can be

traced in detail. The first such school was that created by the patron- age of Sultan Nasir al-Din Muhammad, who ruled intermittently from [293 to [341. Seven calligraphers signed major Qur’ans pro-

duced between [304 and 1372, and the signatures of three illumina-

tors are known from the same period. This suggests the prestige attached to working on Qur’ans and incidentally indicates that the

calligraphers enjoyed a higher status than the illuminators. Some o\~ these artists also turned occasionally to the illumination of (“optic

Gospel books, winch are thoroughly in the Mamluk idiom. By degrees a standard format evolved, with a frontispiece comprising a

double-page spread followed by a double page of illumination pre-

ceding the text itself The same arrangement operated m reverse at the end of the volume.

Sometimes a patron would donate several Qur’ans to a single

foundation; thus in 142s Sultan Barsbay endowed his madrasa with single-volume, double-volume and thirty-volume Qur’ans. Just as

generations of Mamluk amirs vied with each other to erect buildings. so too did the competitive spirit – a reflection on the artistic plane o\~

the endless and ruthless jockeying for power m court life – extend to the production and embellishment of Qur’ans. And this same emula- tion may itself account for the co-existence of several different styles of illumination.





The Muslim West

The art of the Muslim world west of Egypt (the Maghrib) was conditioned to a remarkable extent by its geography. Sundered from

the rest of the Islamic world by the extensive deserts of Libya and

western Egypt, its maritime communications with the East frequendy

threatened by hostile Christian powers, the Maghrib was compelled as early as the eighth century to turn its focus inwards. This process

was sealed after 1050 with the invasion of the Banu Hilal and other nomadic Bedouin tribes who, travelling westwards from Egypt.

overran the eastern Maghrib like locusts. Their flocks devastated good agricultural land, bringing in their wake economic ruin and permanently destroying the ecological balance of the whole terri-

tory. Sedentary lite contracted; towns and villages, deprived of then-

agricultural hinterland, decayed. The break between eastern and western Islam proved irreparable. Not surprisingly, this physical isola tion entailed a gradual but destructive intellectual and cultural isola-

tion. In most fields of Islamic learning, Maghribi scholars were either

unproductive or lagged far behind their colleagues to the east. It is

true that many of them travelled eastwards in search of knowledge; but this was one-way traffic. By the later Middle Ages, an occasional luminary like the philosopher-historian Ibn Khaldun stands out by

his very rarity.

For some three centuries Muslim Spam, whose history begins

with the Arab invasion in 711, constituted an honourable exception

to this trend. Yet the gradual erosion of its territory as a result of

unremitting pressure from the Christians to the north put an increas-

ingly forseeable term to this intellectual flowering. Even so. certain

cities of Muslim Spam, notably Cordoba and Toledo, were important

centres of scholarship in the secular sciences, such as medicine,

astronomy and mathematics. In Toledo, after the Christian recon-

cjuest and under the rule of Alfonso VI (from mSs) and some of his

successors, these works were translated into 1 atm and thence made

their way throughout Europe. Thus Muslim Spam served as the conduit for scholarship to travel tmin east to west.






But for good reason the Muslims of Spam could not match the self-confidence of their co-religionists to the east. They lacked the lands and the wealth, and – less tangibly – the security that came from being surrounded by a vast and successful community of the faithful. It is no accident that Spam is more richly endowed in medieval Islamic castles and fortified settlements – from cities to

villages – than any territory of comparable size in the Muslim world. For most of its existence – an existence eked out to the threshold of the modern age, for Granada tell in 1492, the year that Columbus discovered America – Muslim Spain had a beleaguered mentality, especially after the tall of the Umayyad caliphate of Cordoba in 1031 mk\ the consequent proliferation of some forty minor dynasties (the so-called taifas). Even at the height ot its power it was still only a small principality sandwiched between the aggressively encroaching

Christian presence to the north and the Berber-dominated territory

to the south, whose restrictive cultural and religious orthodoxies were in many ways profoundly at variance with those of Andalusia. In this relationship. North Africa, for all its vastly greater physical

extent, was undoubtedly the junior partner in most respects, contin-

ually looking northwards for inspiration. This is not to deny that the

lands now subsumed 111 the modern states of Morocco, Algeria and Tunisia did indeed produce great empires, notably those of the

Almoravids (1 054-1 147) and the Almohads (1 130-1269), which even ruled briefly in southern Spain. But these states were short-lived and

the activity of their rulers as patrons of art was limited to religious

architecture. This lack of a dynamic or even useful hinterland threw

Muslim Spain back on its own resources. Small wonder that it looked ever more obsessively back to a golden past.

For indeed the art of the Muslim West cannot be understood

without reference to the Umayyads of Syria. The tantalizing memory of that dynasty, of its fabulous wealth and power, is the con- stant subtext of the art of the Muslim West, particularly of Spain, and

of the rest of its culture too. The sole Umayyad prince to escape the massacre of his ruling house by the


Abbasids in 750 had. after many adventures, made his way to Spam, where he had set about recreating the lost glories of Umayyad Syria in an alien kind. Hence, tor example, the deliberate rejection of that imperial ‘Abbasid art whose

various manifestations infiltrated the rest of the Islamic world, surfac-

ing as far west .is Oairawan in Tunisia. Conceivably this rejection

brought in its tram a corresponding reluctance to develop such major c

Abbasid art forms as pottery, metalwork and the arts of the book

1 6q



fields which (illuminated Qur’ans apart) are represented in only mar- ginal form in the Muslim West. Or one could attribute this reluc- tance to the fact that the Umayyads of Syria did not seriously practise these arts. Yet another possibility would be to see the key factor as religious, specifically the dominance of the ultra-conservative Maliki

madhhab, a school of Islamic law whose strong puritanical streak made it hostile to the arts m general. None of these explanations is entirely satisfactory, though the fact that they can all be entertained

might suggest that all these factors played their part. But the absence

of several key art forms practised enthusiastically by Muslims further

east had the effect of concentrating attention on those fields of artis-

tic endeavour that did enjoy popularity, such as textiles and ivories.

In such cases it is Muslim Spam, not North Africa, that produced work of quality. Similarly, in architecture, the absence of building types which were commonplace further east, notably mausolea and khans (the latter absent perhaps because most land-based trade was

with the sub-Saharail regions, and followed routes for which kham were impracticable 1 would have treed the energies of the medieval Maghribi builder (and the resources of patrons and communities)

mainly for religious architecture. Net cumulatively these various

lacunae would have resulted m an environment in which the visual arts could nut have had the same impact as they had in the Muslim

East. In some respects, one might even say, the Maghrib might have struck a visitor from the East as provincial.

The well-nigh crippling mental dependence of” the art of Umayyad Spam on that of L’mavvad Syria ruled out the constant replenishment of” traditional forms b\ others from outside that magic

circle. Thus there is no parallel in the western Islamic world for the major change of direction between. s.i\. Saljuq and Ottoman archi- tecture in Anatolia (in which Byzantium was the alien factor) or the

Chinese element in the painting and minor arts of the medieval Mashriq – the eastern Islamic lands – or the constant and fruitful

interplay between the various media in Iranian art. It is true that

Islamic ideas and motifs transformed certain aspects of medieval and

even later art in Spam and the Spanish dominions overseas, but there was no comparably strong reciprocal current of Christian influence in the medieval art of western Islam. Tradition exerted a vice-like

grip on the visual arts which, despite their distinctive character and some significant innovations, favoured the mutation of earlier modes, expressed tor example m a preference for archaizing compositions made up of confronted or addorsed figures or animals.




The high-water mark of western Islamic art is synonymous with Cordoba, already .\n ancient city when it was selected as the Umayyad capital in 756. Andalusia was quickly colonized by Syrian refugees who brought their dialect, their tribal rivalries, their place names and even their plants with them. It was therefore natural that

the memory o\ the Umayyad architecture of Syria should also be kept alive. This memory expressed itself in various ways. Tradition asserts that the Great Mosque of Cordoba, like its predecessor 111 Damascus, was built on the site ot\\ Christian church bought and

then demolished by the Muslims. For centuries, the model of the

Damascus mosque was copied more faithfully in the Maghrib than

anywhere else in the Muslim world. It was the Muslim West and no other area that adopted as canonical the lofty square minaret tradi-

tionally associated with Syria. The Cordoba mosque used a vocabu-

lary of horseshoe-shaped arches and two-tiered arcades first found at

Damascus, and there too Byzantine craftsmen were called in to

execute mosaic decoration.

128 christian vandals. The Great Mosque. Cordoba. The minaret, like the Giralda in Seville, be.irs an elaborate Christian Baroque superstructure. But it was the intrusive chapel inserted into the heart of the

structure in 1523 (tor triumphalist motives?) that really disfigured the mosque Gothic vertuahtv versus

Islamic horizontally – and called down the wrath of the Emperor Charles V upon the local clergy.



1 2v. 130, [31 Royal precinct

Great Mosque. Cordoba: maqsura,

96] % White columns and arches, like the .>><> oil lamps, would only partially have reduced the

claustrophobic gloom in the vast low sanctuary, often uncomfortably

dose and warm. Amidst this

relative darkness the lavishly

fenestrated and vaulted royal

enclosure around the mihrab stood

out. 1 ight flickered off its golden

mosaics, a metaphor of spiritual

illumination to which the caliph,

as the prayer leader according to

Islamic law, subtle staked his own claim as he stood framed in the

rayed mihrab on Fridays. I he mihrab

was the work of a t raftsman

specially brought with his

materials – from Constantinople, a

tribute to the pan-Mediterranean

culture of the time.

Yet these Syrian echoes which could be multiplied) do not give

the measure of Islamic art in Spam – often termed Moorish or Hispano-Moresque. The Great Mosque of Cordoba in particular developed over the centuries its own distinctive architectural and decorative vocabulary. In the course o\~ tour expansions in less than

three centuries – a vivid illustration of the additive nature of the

mosque in general – it grew to become one of the largest mosques of all. This gigantic size seems to have encouraged its architects to

explore subtleties of lighting, repetition and rhythm to a degree rare





m mosque architecture. They repeatedly employed the vanishing point to suggest infinity, and a concentration of ornament to exalt the area around the mihrab. Yet all these visual effects were, so to speak,

incidental to a major structural innovation called forth by the need to

roof a vast area even though only short columns were available. Extra

height was imperative and was secured by building broad block-like

piers resting on these columns and braced by strainer arches. These

arches were illusionistically lightened by the use of alternating red and

white voussoirs. but the contrast between airy, freely circulating space

in the lower elevation and a relatively dense thicket above was unmis-

takable. In the area around the mihrab, as rebuilt from 96] onwards,

the notion of a forest – an analogy already suggested by the files o\

living trees planted in the courtyard, which would have merged

smoothly with the sanctuary arcades – is intensified. A roval enclo- sure defined b\ a network ot interlacing multifoil arches with

arabesque decoration creates a blooming petrified garden in which

honorific and paradisal undertones mingle – and from which the

congregation at large was excluded. A remarkable sequence ofribbed domes ot variety and complexity technologically far beyond anything known in the rest ot Europe at this time) provides a fitting culmination for these splendours, and indeed intensifies their impact

by a nexus of interrelated solar and celestial references. I his .uv.i of

the mosque abutted dircctlv on the royal palace and provided a fitting

environment for a monarchy which had only recendy 928 claimed the numinous title ofcaliph. The princes o( ( ordoha thus challenged

the c

Abhasids .is the divinely-ordained rulers of the Muslim world,

and their mosque was part of that challenge. So too. perhaps, is the palace-cit\ of” Madmat ab/ahra outside

Cordoba for it is highly likely that rumours of the fabled luxury of

Samarra soon filtered back to Spain and lost nothing in the telling.

Its marvels included ponds of quicksilver and jasper floors, while the

statue of a Roman Venus over the main entrance was .1 reminder o\~ the Mediterranean heritage o\~ Isfnn. Most of this material was

looted within a ccntur\ ofthe city’s foundation, but excavations haw-

revealed audience halls with columns ofmany colours and. above all, abundant carved stu oration with arabesque ornament so deli-

cate that it is almost finicky. This is the authentic idiom of the

Mshatta facade see p. ;;> . but somewhat reduced in scale so that it belongs with the minor arts rather than with architecture. Indeed, the striking absence o\ major buildings at Madmat al-Zahra tells its own story. Main of the standard hallmarks of royal pomp .ire duly



there – the ceremonial triple-arched portal, the lavish use of water, the use of axial emphasis to exalt the monarch – but the vital dimen- sion of scale was missing. Madinat al-Zahra was large enough in all conscience, but it was apparently innocent of any integrated overall

plan. Its buildings ramble, and none of them is of substantial size. The will to power expressed so unmistakably in the architecture of Samarra is just not there. Here if anywhere m the Islamic world was a pleasure capital.

And this is entirely appropriate, for Cordoba in its prime had no peer in Europe for the amenities of civilized life. Its houses were bountifully supplied with hot and cold running water, its streets were

lit at night, its royal library – if one may trust the chroniclers – had 400.000 volumes at a time when the major libraries m western Europe scarcely reached a thousand. In this metropolis, moreover,

Muslim, Christian and Jew lived together with a degree of harmony rare in the Middle Ages, while Berbers, negroes and Slavs formed

the caliphal bodyguard. Cordoba owed much of its sophistication to this multicultural and multi-confessional environment.

Yet the rapturous reception accorded to a flashy visitor from

Baghdad, the singer Ziryab. who quickly became the arbiter elegan- tiarum on matters of taste, costume and etiquette, suggests that main

Cordohans were uneasily aware of their isolation from the rest of the

Islamic world. Indeed, there survives a medieval Spanish silk whose

inscription fraudulently claims that it was made in Baghdad. A similar dependence on the art of Iraq can be detected in the only known illustrated manuscript of non-scientific character from Muslim Spam,

HaMrh Bayad wa Riyad (‘The Story of Bayad and Riyacf). a tale of

courtly love. The manuscript is generally attributed to the thirteenth [32 century. The poses and gestures of its figures, often in silhouette. have a studied intensity which seems to owe something to the art of mime, as do contemporary illustrated Maqamat (‘Assemblies’) manu- scripts from Iraq. Both schools also share a love for the antithetical

placing of figures in the interests of dramatic storytelling. Yet both

the courtly atmosphere of the Spanish manuscript – so at variance

with the robust, rapacious, street-wise world of the Maqamat paint-

ings – and its ambience of gardens, watcrw heels, polylobed arches,

square towers and luxury pavilions faithfully evoke the spirit oi

medieval Andalusia, the land of the lute and the lyric, rather than

Iraq. The prominence given to women here, which reflects their roles as scribes, musicians, librarians and poets 111 Muslim Spam, is

also foreign to the male-dominated society pictured in the Maqamat.




^.’yi : J,jj i -j* : u^=^>}i;a^»«*l$c<$>

132 Courtly love. The storj of Bayad and Ri\ s . 1 ;th century [“his type of romance originates in the


Udhri poetry of ancient Arabia: idealized, melancholic and unfulfilled. Here the machinating go-

between tries to inject some backbone into the lovelorn youth.

A distinctive style of Qur’anic illumination developed in the Maghrib and Muslim Spam. The bonks are small, usually about 20 cm (8 in) square The script itself was decidedly different from all the styles that prevailed further east, tor instance in the tonus of certain

letters, the placing of some diacritical marks and the preference for the horizontal rather than the diagonal in vocalization. Hut the

prime trademark of this style was the extravagant looping of terminal

flourishes, creating a tangled thicket of lines. Sura or chapter head-

ings are in archaic Kufic which is unmistakably of Syrian Umayyad origin, even though this was many centuries out of date. The letters are usually in gold and are set within ornamental panels. I his was ,1

profoundly conservative style and it continued with relatively little

change almost until modern times. Equally conservative was the

135 preference tor vellum right up to the fourteenth century, at a time

when the rest of the Islamic world had long switched to paper tor illuminated Qur’ans. As with Qur’ans from Egypt and points east,

there are examples of pages with only tour lines of text, and the




words so widely spaced that there is room for no more than a dozen of them; equally, there are very closely written Qur’ans with over

twenty lines per page. Extreme elongation (mashq) is used here, ax in Qur’ans produced elsewhere in the Muslim world, for visual, rhvth

line and perhaps spiritual effect. Little progress has been made with the identification of the individual schools in this tradition. One major centre for this art. however, was Valencia in the later twelfth

and early thirteenth centuries, where the craftsmen specialized in

full-page square polygonal designs of dynamic, indeed explosive,

character, cunningly interspersed with inscriptions. These composi-

tions, placed at the beginning of a volume and mirrored in the

binding itself, may have had apotropaic intent. In the minor arts, the glory of the Cordoban caliphate is its ivory

carving. The intense it short-lived concentration on working this costlv material finds a parallel in the rock crystals made for the eon- temporary Fatimid court in Egypt. As befits the nature of ivory, the

objects themselves are small, principally caskets, cosmetic cases,

pyxides and the like, which held perfumes, unguents or jewels. Their inscriptions mention high-ranking personages of the court such as

the princess Subh, the prince al-Mughira or the chief o\ police.

Ziyad b. Arlah. Recent research has revealed that main such items

were intended as presents to ladies who had given birth to an heir- apparent, and thus their splendour had a major politic. ll dimension.

Cross-references with architecture may sometimes be detected, as in

i.u- n>

[33 leaf from multi-volume

Qur’an in gold on vellum.

Probably Spam. 1 ith century.

The reed pen is cut to produce letters of even thickness, unlike

the dramatic alternations o\~

thick and thin found in eastern

Kufic hands. Maghribi scribes

were taught, says Ibn Khaldun.

by writing complete words, not

individual letters. i~ x 22 cm (10.6 x 8.7 mi.



cylindrical boxes with domical lids, and even the celestial associations

of such forms can be exploited, for example by depicting on the lid

eagles whose outstretched wings bear a raved solar rosette. On the bodies of such objects, affronted creatures such as camels, deer, goats.

and more often those of royal symbolism – usually griffins, elephants,

lions, peacocks and eagles – parade against a backcloth ot densely

carved arabesque which again has close affinities with Umayyad

work in Svria. Many o\ these animals may have been intended as references to the royal game park. The grander dimensions thus

implied lend an impressive monumentality to these little objects.

Other images include ruler figures seated in majesty on elephants,

hunting, jousting or banqueting scenes, musicians and wrestlers – in

other words, images derived from the princely cycle first established

in Islamic art by the Umayyads of Syria see p. ;>2 The amount of

detail crammed into such a diminutive space is quite remarkable.

This intricate carving, formerly embellished by bright colours, is

especially appropriate for such precious material and subtly under-

lines the association with jewelry.

musk, .. imphor and

ambcTy;- is the inscription

M i>v|iu- mother

‘ the caliph .il I lakam II

Arch.t include the cpigraphu

A the vir ill

• hunt 1 dated


,;l i»t huntei itta< Iced by

• l.



hove) Worn by St Thomas-a- Becket? Made in 1 1 16 in Almena. the centre of the Spanish textile

industry, this is the earliest Islamic

. embroidery inscribed with both date

and provenance. It was refashioned

into a chasuble and, tradition says,

owned by the saint, a sharp dresser. Its light-blue silk is embroidered in

gold thread with some 40 roundels depicting animals, birds, mythical

creatures and courtly scenes.

137 {right) “Talisman. The Pisa Gnrfm. Probably Spanish. 1 ith

century; the largest Islamic bronze

figure, though much smaller than some classical, ancient Near Eastern.

medieval Italian or Chinese bronzes.

Essentially a fantasy assembled from

bits and pieces of many creatures, its exaggeration betrays very little

feeling tor sculpture. Perhaps one of a pair, displayed with apotropaic

intent on a pedestal by a fountain or

gateway. H. 107 cm (42.1 in), 1. 87 I in;.



Most of these scenes are enclosed within heavily outlined pearled roundels or rosettes which are joined by knots. This is clearly a bor- rowing from textiles, and indeed such textiles survive in abundance

from slightly later periods. Tiraz factories operated in Seville and

Cordoba; later, Almena and Malaga became celebrated for their silks, and indeed at Malaga mulberry groves provided the raw material on the spot. Fabulous beasts Mich as griffins, double-headed eagles,

basilisks, harpies and sphinxes, plus exotic creatures like lions, ante-

136 lopes, and birds galore – strutting peacocks, eagles and ducks –

formed the repertoire tor such silks. Their rarity or other-worldliness

made them apt ideological symbols of majesty. Many themes, however, expressed the theme of royal power more explicitly, whether by inscriptions which gave royal titles or by symbols such as

the lion-strangjer or the eagle seizing its prey Some of these themes 137 turn up 111 the stone sculptures of the Cordoban caliphate, for

example in the troughs now atjativa, Granada and Marrakesh. Others turn up in metalwork. The bronze griffin in Pisa, triumphantly dis-

played for centuries on the facade of the towns cathedral, was proba- bly captured as booty m one ot several campaigns against the Muslims around the turn ofthe eleventh century. At 1.07 m (3.5 ft) in height, it is the largest piece of animal sculpture in Islamic metalwork, and

represents a formidable technical achievement. Net its hybrid form

and overall patterning, both markedly anti-naturalistic, betray the

residual unease which Muslim artists felt tor representational sculp- ture. It may originally have guarded, perhaps as one ot” a pair, the entrance to a palace or throne room. Its back bears textile patterning

which, like the style ofits epigraphy, suggests a Spanish source.

[] ten

connections, [bledo,

Mosque >>t Bab Mardum, I -.r. uted in

humble bri< k and rubble, and later re used as a

( Ihristian c hurch, it

illustrates a type of nine-

bayed mosque, probably

derived from a Baghdad]

prototype, thai bail spread

throughout much of the Islamic world by this time.

Here mterlac ed and

horseshoe arches lend it

Spanish character.



139 Islamic rococo. Aljaferia,

Zaragoza; southern portico, after

ioso. This palace of the Banu Hud dynasty, completed in ioSo. now

s the autonomous Aragonese parliament. Note the tiny columns, stripped of structural function.

marooned in a thicket of interlace

ornament. Partly mathematical

theorem, partly geometry as

contemplation, this is made deliberately hard to read because

only a small section ot a much larger design is shown.

The authority- and prestige of the art of Cordoba was such that it imposed itself on the various lesser Muslim principalities of Iberia, and continued to do so long after the tall of the Cordoban caliphate, when the whole country became split into numerous warring states. Thus the mosque of Bab Mardum at Toledo, dated 999, is in some respects a miniaturized version of the Great Mosque at Cordoba, complete with a facade of interlaced arches (presumably a toned-

down version ot the sanctuary facade at Cordoba) and a set of nine patterned ribbed vaults, each one different. Interlaced arches are

again a leitmotif at the palace of the Banu Hud at Zaragoza, known as Aljaferia (10S0), but they are now earned to dizzy heights ot complexity, especially in the area of the audience hall and the

oratory. The latter building, a kind of pocket Venus in architectural and decorative terms, is a lineal descendant of the mihrab ba\ at

Cordoba. As for the palace itself, which acquires extra importance

because it documents the art of a period from which very tew

significant remains survive in Spam, and thus provides a link between

Madinat al-Zahra and the AJhambra, it demonstrates yet again how

thoroughly Islamic Spam was m the thrall of Umayyad Syria. I he multi-bastioned exterior with its single centrall) placed monumental






140 Islamic baroque. Marrakoh. Qubbal al-Harudivvm. between \l<><> and 1142, inte-

rior of dome. The heritage ofCoidoban vaulting (cf. 1//. 130) in enriched by burgeoning ornament and a new lightness and aspiration derived from the open -plan design.

gateway quite clearly derives from the desert residences of the eighth

century. Numerous other Spanish Muslim castles echoed this form; but they did so principally with military intent, where. is in North

Africa (as at Ashir or Raqqada the association with palatial architec- ture persisted. More generally, the architectural vocabulary of Andalusia – horseshoe arches, roll mouldings, rib vaults, interlacing arcades – infiltrated the Christian architecture of the north and even crossed the Pyrenees, leaving its mark on the Romanesque churches of south-western France in particular, as .it L.e Puy, where even the

cathedral door bears debased Kuflcizing inscriptions.

The political vacuum left by the demise of the Cordoban caliphate was soon filled, and from an unexpected quarter. I )eep in the newly

Islamized territory of western Africa, in what is now Senegal, .1 puritanical movement was gathering momentum. Its members were

141 The Islamic counter-offensive. Rabat. Mosque of Hassan, tower; 1199. Unfinished, but probably planned to reach 85 m (279 ft) including its lantern. This gigantic size was a symbolic response to the advancing Christian reamquista of Spain. The exterior is of stone; the interior, with its succession of six vaulted chambers, is of brick.



#> :-

TSm ms




dedicated to the ideals of jihad (holy war), which they waged from fortified camps (ribats) along the frontier. Hence their dynasty bore the name Almoravid (al-murabitun – the men ot the ribat). Their brand of ferocious piety appealed to the Berbers of North Africa, whose overwhelming support enabled them to storm into Spain and, for a space, recover most of the long-lost Muslim territory in the peninsula. Yet it is precisely their puritanical fervour which helps to explain why they produced so little in the visual arts. Their religious foundations in Spam have vanished completely, and the seriousness of that loss can be gauged by the sparse surviving evidence of their

work elsewhere in the Maghrib, comprising principally the Great Mosques of Algiers (1097 onwards) and Fez (mainly 113s) and a

140 kiosk (presumably a fountain) once apparently part of the Great

Mosque ofMarrakesh. The last two monuments m particular testify to the remarkable

vigour and imagination with which Almoravid architects trans-

formed the time-honoured motifs ot interlaced arch and rib vault

which they had inherited from Andalusia. Whereas in Spam itself the heritage of Cordoba was dissipated m increasingly finicky A\id small- scale ornament in which spatial values played a diminishing role (as

instanced by the Aljaferia palace at Zaragoza), the buildings in Fez

and Marrakesh. instinct with a formidable energy, are triumphantly

three-dimensional. This dynamic manipulation of space combined themes earlier kept apart – ornamental arch tonus and decorative

vaults – to produce a distinctively Maghribi version of the muqamas. The hallmark of these niHi]artLi> systems – m contrast to contempo- rary Iraqi and Syrian versions of such themes – is that the exterior

yields no hint of the internal configuration. At Fez, moreover – as in

the almost contemporary Cappella Palatina at Palermo and the QaTa or the Banu Hammad in Algeria the muqamas vault is built up over a rectangular space and is therefore not interpreted as a dome.

Instead, it works visually like a suspended ceiling. It is significantly

more evolved in design and complexity than contemporary examples in the eastern Islamic world. In the Mosque of the Dead attached to the Friday Mosque o\~ Fez. as in the Marrakesh fountain, variously lobed and crinkled arch forms executed with truly sculptural verve

are used to build up powerful contrasting rhythms. Thus architec- tural forms, not applied decoration, animate the building.

The coming of the Almohads brought a new impetus to religious 141 architecture, as shown by the huge congregational mosques which

they built in Seville, Rabat and Marrakesh. Their keynote is a




14- The long shadow of Spain.

Tlemcen, Great Mosque, mihrab,

mntlhir and dome, 1135—6. Dwarf arcades; cusped, interlaced and

borseshoe arches; ribbed dome; and

two-tone masonry all underline the

defining role of the Cordoba mosque m Maghribi religious architecture; the patron. ‘Ah b. Yusufb. Tashfin,

brought craftsmen from Andalusia to

work in Tlemcen. However, cheap

painted stucco now replaces

expensive glass mosaics.

puritanical simplicity of design, unexpectedly relieved on the qibla

axis by voluptuously lobed arch profiles. These mosques tend to

enlarge the covered sanctuary at the expense of the courtyard, which

shrinks dramatically. They all make much of the measured repetition ot a single unit such as a bay or an arcade to create directed vistas and

a sense of illimitable space; while their plain exteriors, articulated by

little more than niches, exude a comparable austerity. All the more remarkable, then, is the contrast provided by the towering square

minarets of these three mosques, all decorated with overwhelming richness, all dated to the last decides of the twelfth century, and all of

them lllusionisticallv rendered more gigantic still by the contrast with

the low expanse of the mosque itself. In all three cases, each face of the tower is elaborately fenestrated (their interiors comprise a scries

of superposed vaulted chambers linked by a ramp) and bears a differ

cut set ot latticed designs in raised stonework.




143 Dynastic mosque. In Tmnial. the Almohad capital, the self-appointed ‘caliph 1

‘Aba” al-Mu’mm built .1 bijou version of the Cordoba mosque in 1 1 $3. It measures just 4* X 43 m (157.5 141

But the masterpiece among these Almohad mosques, and indeed 143 of later Maghribi architecture, is the mosque of Tinmal, the

Almohad capital near Marrakesh. Here, too, the juxtaposition o\’ plain and ornate is breathtaking; the eye moves from huge expanses

of plain wall to intricately carved capitals – indeed, the Almohad Kutubiya mosque 111 Marrakesh has perhaps the most complex capital in all of Islamic architecture. Among other noteworthy features at Tinmal is the location of the minaret behind the mihrab, strongly salient from the back wall of the mosque ‘a major innovation, soon to be copied at the mosques ot Algiers and Sale), with two sets of

three projecting lateral portals and the three domed bays marking the corners and the centre of the qibla wall. This is part of the character-

istically Maghribi T-shape, which is created by the junction of the central nave (wider than all the others) with the transverse aisle abut-

ting the qibla wall. The vaults o\~ this aisle run parallel to the qibla wall, not perpendicular to it like those elsewhere in the sanctuary.

The area constituting the T is singled out from the rest of the mosque by a sudden quickening of the decorative tempo, evident in applied ornament, arch profiles and vaulting.





144 ‘Orient pearls at random rung’. Fez. Madrasat al- ‘ Attar in. 1325. The wall (including the tiled dado

below) in conceived as a kind

ofpicture gallery, a series of

individual panels of changing

scales, media, designs,

textures, tones, colours and

depth: little care has been

taken to integrate them into a single composition.

& ‘ <r ‘v. « /|S V-. ‘is •

D. %ff /•>. >•’ #\ •*** ‘1

;,.,.; ,-»j;,’/.- , *!,


:iJ/,> toll.

The demise of the Almohads left the way open for the creation of smaller principalities, of which three dominated the later Middle Ages m the Maghrib – the Marinids of Morocco (1217-1465). the Zayyanids of western Algeria (1236-1555) with their capital at

Tlemcen. and the \ [afsids ofeastern Algeria and Tunisia ( 1 229 All three dynasties concentrated their patronage on architecture, and

most other arts – such as pottery, textiles and mctalwork h.ireK rose above the artisanaJ level in this period. The madrasas of the Marinids deserve particular attention: these buildings reflect the orthodox

Mam. propagated by a conservative religious elite, thai flourished m cities like Fez, Taza and Marrakesh, rather than the populai Islam

1 11




of the countryside which expressed itself through the veneration of saints, religious brotherhoods and descendants of the Prophet,

and through countless modest shrines and mausolea of indeterminate

date. Most of the madrasas, by contrast, are firmly dated and are extravagantly embellished with glazed tilework, carved wood and

145 intricate stucco ornament. The standard design featured an enclosed courtyard with a pool and two-storey facades. The lower floor served for lecture rooms and a prayer hall; while in the upper storey, quite removed from all this visual splendour, lay bare, comfortless and overcrowded cells for the theological students themselves.

From the sixteenth century onwards, most o\’ the Maghrib fell under Ottoman rule and lost much of its local character. Nevertheless the grand public Ottoman mosques like the Fishermen’s Mosque in Algiers were complemented by domestic architecture of an altogether lighter atmosphere, with fountains;

painted and fretted woodwork screens, doors and fretted wood (mashtab

i yd) windows; and garish tilework.

The later medieval history of Muslim Spam, and of its art, is rather different. The Christian victory at the battle of Las Navas de Tolosa

1 ;v

tllege life. Fez, Madrasat al-

‘Attarin: founded, started and

endowed m 1 J2J by the Marinid

sultan Abu Sa’id, who was present when the foundations were laid. The origins ofmany madn tsas in the teacher’s own house are reflected here m the intimate domestic flavour and small stale I he contrast

between the imposing and glamorous

public face and the austerity of the

students” quarters, safely out of sight,

is striking.




14′- Hie afterlife of Samarran ornament. Marble capital from the unfinished Marinid palace at al-Mansuriya, Algeria; early 14th century Such capitals of expensive stone echo

the bevelled style (cf. ill. 27) and Almohad modes in then lavish carving and whimsical curvilinearity This recalls the Maghribi fascination with exotic arch tonus.

in 1 2 12. with the consequent loss of Cordoba and Seville, was a catastrophe for the future of Islam in Spam. Yet Muslims continued

to practise their arts and crafts under Christian patronage, as shown by the Alcazar at Seville (rebuilt by Pedro the Cruel in [364), in

many ways a reduced version ot~ the Alhambra, and by the lustre pottery of Manises. Against all the odds, moreover, one Muslim

principality managed to survive the general collapse by dint of the humiliating accommodations it reached with its Christian neigh

hours: the Nasrids of Granada (1232-1492). They maintained Islam and its civilization, together with the Arabic language. 111 Spain for

over two and a half centuries. A\)d with such panache and success that, according to an Arabic saying. “Paradise is that part oi the

heavens which is above Granada’. This emirate, despite its small size and constandy threatened posi

tion. took in refugees from elsewhere in Muslim Spain and its court

welcomed historians and scientists, philologists and geographers, and above all poets. Some of the Islamic city’s streets, baths, inns. \ ill. is. madrasas, markets – even converted mosques and minarets still

survive. The wealth of the realm rested p.utK on its agriculture and




147 (left) Symbol ofjihad. ‘In God I find refuge from Satan’ proclaims this

Almohad banner. 1212-50. The inscriptions contain Qur’an 61:10-12,

with its promises of Paradise to the

faithful in holy war. The iconography features a central talismanic star and

other celestial motifs, much like contemporary Qur’anic frontispieces.

\i<uc) A Muslim monopoly under Christian control. Valencian lustre

dish with abbreviated Arabic inscriptions

repeated ‘good health’), early 14th

century. Royal, ducal and other

inventories, itemized orders and

contemporary paintings indicate how coveted this ware was throughout

Europe. The Muslim craftsmen worked predominantly m Christian territory; thus in 14SS .1 certain Juan Murci was

executing orders for 200.000 tiles to be

supplied to rulers in Naples and

Valencia. The shields ret all the armorial

function of much V.ileiu 1.111 ware.

partly on such products as textiles and lustre pottery. The range ot Granadine art. which includes several objects and techniques not

found elsewhere in medieval Islamic art, is quite remarkable. It

includes writing desks, doors and caskets of marquetry work in wood and ivory, carved ivory and silver pyxides, jewelry, bridles, parade

helmets, scabbards and swords decorated with cloisonne enamel,

metal openwork mosque lamps and a spectacular array of textiles:

147 hangings, pillow-covers, capes, rugs, banners, mantles and curtains.

These provide abundant proof that the textile industry of Muslim

Spain maintained its vitality and its capacity to innovate right to the

end, seemingly unaffected by the loss of almost all its earlier centres

of production. The motifs tend to emphasize not the heraldic beasts that were the stock-in-trade of luxury textile designs across the

Mediterranean world in previous centuries but instead defiantly pro-

claim their Islamic origins: interlocking motifs suggesting infinite




patterns, crenellations, and Arabic inscriptions with mottoes like

‘Power belongs to God’ and ‘Glory to our Lord the Sultan’, as well as more light-hearted messages, like ‘I am made for pleasure, for pleasure am [\ The wide distribution of these luxury textiles can be deduced from the frequency of their appearance in medieval Spanish and Portuguese paintings of Christian content. To the very end. then. Islamic textiles maintained their ancient associations with

wealth and authority.

Spanish lustreware had similar associations. It was made, often for

Christian patrons, with western coats of arms and Christian Inscrip

tions such as ‘Ave Maria’, at various centres including Manises,

Paterna and Malaga (hence ‘majolica’), and was widely exported,

especially in the form of great dishes and apothecaries’jars (albarellos).

Pride of place in this technique, however, must ^) to .1 speciality

apparently unique to ( .ranada the so called Alhambr.i v.iscs. huge

1 ifl




amphora-like containers more than 1.20 in (4 ft) high, decorated in lustre or lustre and blue, and furnished with decorative handles.

Some bear apotropaic motifs Mich as the khams or sacred hand, sug- gesting that their contents (wine? oil? perfumed water?) required

149 protection. For the potter, their size alone posed a daunting technical

challenge; the firing of lustrewares was a tricky operation at the best

of times, and a kiln capacious enough to hold pieces this big would not generate constant heat. Hence the traces of uneven firing which they bear. Well do some of them bear the single word ‘power’ (.//- mulk – also the title or” a Qur’anic sura or chapter) repeated like a mantra in a closed circuit around their bodies. Other vases are

inscribed with wishes • : health, good fortune, prosperity, pleasure, or with the motto “Power belongs to God’. C V they may apostrophize themselves in flattering terms, a popular convention in

later Moorish art. There is a certain irons and pathos in the tact that

these, by tar the largest and most spectacular examples of all Islamic

luxury wares, were produced in a tiny principality situated on the

outer fringes of the Muslim world and doomed to rapid extinction.

! he .trust’s hand has embroidered

me like 1 r»>K- of silk’ Wing-handled ”

! ik luu\ tun- I he gold

and blue palette typical of Spanish lustre

tided’ .nut ‘silvered’ m i he Ar.ibu


cions, >>t different scale, type .nut


‘ ;/’ ,1111111

tttune nut prosperity’

r I ile below the

affronted animals lise.

M 135 en




Much the same can be said of the Alhambra itself, the only large- scale medieval Islamic palace to survive, admittedly much repaired over the centuries. It is not so much a palace as a torn tied royal city set athwart a mountainous outcrop dominating Granada, with the

River Darro below and remote vistas of the Sierra Nevada beyond. The complex originally contained six palaces, of which five remain, plus numerous subsidiary buildings and gardens. Like much of Nasrid art it is delicate, even over-refined, and of fastidiously choice


execution. The literary sources speak of Moorish landscape archi- tects; this is their handiwork. The Alhambra brings the forces of nature into play at every turn: water in movement – trickling, running, cascading, spurting – or still, in tranquil expanses; carefull) harbered trees and bushes; sunken flower beds; sudden glimpses of

mountains or gardens framed in a casement, or, more ambitiously, miradors and belvederes cunningly placed to exploit sight lines over

an entire landscape; and, above all, light. The Alhambra studiously manipulates contrasts of light and dark, with bent entrances, shafts of

sunlight angled into shadowy interiors, dim passageways suddenly opening into a courtyard open to the blazing sun, and light reflected from placid ponds or walls clad in glistening tiles.

Frequent destruction and new buildings (which have come and gone as architectural fashions change) have wrecked the original

carefully planned processional sequence which channelled petitioners

to the royal quarters. The principal elements are largely datable between c. 1333 and c. 1391. Two of these stand out. First in date is the Comares palace. Its entrance facade evokes a vast petrified Nasrid

hanging, its staggeringly complex yet still two-dimensional ornament

articulated by doors to left and right and multiple windows. Blank

walls on either side focused attention on the centre, where the sultan sat on his throne, itself raised on three steps, to deliver public judg-

ment. To the east lay the Hall of the Ambassadors, a private audience

chamber whose lacy insubstantial architecture is paradoxically

encased in a huge bastion, its unadorned walls an unmistakable meta-

phor of brute strength. The ceiling within represents m schematic form the seven heavens oi the Islamic cosmos, above which is the

throne of God Himself, whose protection thus extended to the ruler. God’s earthly deputy, enthroned below, (lose by this hall were the

private quarters of the sultan, plus four apartments tor the tour wives

permitted by Islamic law; each apartment was two-storey, with suites

for summer and winter. Hut the largest and most scenic feature ot the

Comares is the Court of the Myrdes, focusing on a sheet of




s If-advertiscment

‘Incomparable is this

basin! Allah, the Exalted

t >ne, desired I it

should surpass everything

in wonderful beaut)

( lourt nt the Lions,

Alhambra, r. 1375 I he

lions are taken from an

1 ith-century e; they

spouted water, .1 restful

sound in this secluded spot In the rising and

tailing rhythms of these

delicate penc il like ringed

columns, an hitecture

melts into nuisii

water bordered by aromatic shrubs. It lends an axial emphasis to the

complex while its water serves both as a cooling device and to reflect the surrounding structures.

The most famous element of the Alhambra is the Court of the 150 Lions, set at right angles to the Court of the Myrtles. It is essentially

a classical villa tor the private recreation o\’ the monarch, with a

shaded portico and a garden courtyard. The latter is divided into four quadrants, perhaps to suggest the seasons, perhaps to suggest in mini-

ature the world itself: at the centre stands the Fountain of the Lions.






151 ‘It surpasses the stars in the heavens’ wrote Ihn Zamrak of this muqarnas vault, the Dome of the I\\n Sisters in the Alhambra; c. 1380. The 23 stanzas ofhis poem inscribed here evoke a rotating heavenly dome which mirrors the changing constellations and the eternal cycle ofday and night. They also touch on other interrelated themes – fertility, gardens, money, victory, service, textiles, jewels and divine protection

thereby revealing contemporary attitudes to the building.

a theme with resonances of the Solomonic Temple. Four halls, again axially grouped, served for private entertainment such as music –

hence their acoustic ceilings – and official functions. The master- piece here is the Dome of the Two Sisters, a muqarnas composition involving over five thousand cells; the structure bears an inscription

by the local poet Ibn Zamrak linking it to the constellations. This conceit is taken up by the multitude of reflections caused by the play

ot light on this honeycombed surface, which evokes the revolving heavens mentioned in the inscription. The emigration ot artisans

from Muslim Spain ensured that echoes ot the Alhambra and its

decoration persisted in North Africa for centuries, though none of

its many descendants rivalled it as .1 machine for gracious living.

1 si





The 1 1 khan ids andTimurids

The robbery and murder ofa caravan ofmerchants from Mongolia in 1 21 7 by Muslim customs officials in Central Asia unleashed a deadly retribution. From 1219 the Iranian world was devastated by repeated invasions of Mongol hordes, originally commanded by Genghis Khan and later by members of his family. These campaigns culmi- nated m the sack of Baghdad and the extinction of the ‘Abbasid caliphate 111 1:0. The whirlwind nature of the Mongol conquests compounded the sheer terror caused by their apparent invincibility, by their awesome cruelty and by their essentially alien nature – they were, after all. a people who were not Muslims, spoke no Islamic lan- guage and for whom the norms of Islamic culture and society were foreign. All this left an indelible mark on their victims. Moreover, this was not mere conquest. The Mongols waged total war and inflicted on much ot Iran an eco-catastrophe from which it never recovered. Canals were destroyed, orchards felled, wells blocked up. fields sown with salt: the \er\ cats and dogs were killed. After the s.ick of the

great cn\ ot Merv. die Mongols forced a muezzin to give the call to

prayer and then slaughtered the survivors as they crept out of their

hiding places. Altogether, the loss of life amounted to genocide. Only artisans were spared, to be sent back to Mongolia where no trace of

then- activity remains. Vast areas, especially in north east Iran and

Central Asia, were depopulated, and refugees streamed westwards. In

the held of the visual arts, this brought to Egypt and the I evant ideas

and techniques from the area between Iraq and Afghanis!. m.

Gradually a new political order arose from these rums. At first it

had little in common with the past. Iran was relegated to a mere province in a pan-Asiatic empire which at its height comprised much ot the Eurasian land mass from Korea to east Germany – .111 empire whose continuous extent is unparalleled m world history. Karakorum m Mongolia, and later Peking, where the Great Khan resided, was the new centre of power, and the Golden Horde (covering much of Russia^ and the Qkhanid realm comprising Anatolia. Iraq, Iran and




Central Asia) were its sub-states. The Pax Mongolica imposed on this vast tract of land, and maintained by fearsome penalties inflicted

On wrongdoers, dramatically facilitated communication between East and West. For the first time m history, it was safe to travel overland from Europe to China. Merchants like Marco Polo took full advan- tage of the opportunity, and Christian missionaries trekked to the

Great Khan’s court bearing messages from the Pope, leaving vivid

accounts of their journeys.

The Christian powers, who at first equated the Mongol ruler with the legendary Prester John, a Christian monarch believed to rule in the heart of Asia, saw the new order as .1 heaven-sent opportunity to give the death-blow to Islam, and initiated diplomatic moves to this

end. For their part, the Mongol khans, especially in [ran, maintained an active correspondence with European rulers, principally with an

alliance against the Ayyubids and Mamluks, who ruled Egypt and Syria, in mind. They saw these European monarchs as suppliants and the language oi their letters is arrogant and high-handed. Western

hopes ot converting the Mongols were in any event ill-founded. In religious matters they showed a tolerance remarkable in the medieval

period; they themselves practised by turns their ancestral shamanism.

Buddhism. Christianity and eventually Islam, of both the Sunni and

the Shi’ite persuasion. This tolerance created a golden age for

Christians and Jews, who repeatedly rose to high administrative office. There can be no doubt that this openness to other cultures

and beliefs was a key element in the formation of Ilkhanid art. Hut

the Mongols also imposed their own civil code, the yasa first pro- mulgated by Genghis Khan, which had little in common with the shari’a code that had traditionally governed Muslim lite.

The centre of Ilkhanid power was in north-western Iran, whose fertile uplands were a potent attraction to the still-nomadic Mongol elite. Here Abaqa Khan (1265-82) built a palace on the rums of a

spot already sacred to the Sasanians, Takht-i Sulaiman (* The Throne

of Solomon’), centred on a perpetual lake 111 the crater ot an extinct

volcano. These ancient resonances were deliberately exploited – and

enriched by theological undertones – by some ot the leading Persian

intellectuals of the time when the site was refurbished. Lengthy and carefully chosen quotations from the Persian national cp\c. the

Shahnama (‘Book of Kings’), are incorporated into an iconography ot

markedly Chinese character, in which animal and landscape themes

of Far Eastern origin – trees, clouds, grass and fabulous beasts like

phoenixes and dragons predominate. Yet lakht 1 Sulaiman. tor .ill




its size and complexity, its free-standing kiosks and its technical

innovations in muqamas vaulting, is still exceptional in its time; it took some eighty years before the arts of Iran had recovered from the destruction wrought by the Mongol invasions.

That they did recover was in large part due to the greatest of the Ilkhanid rulers, Ghazan Khan (ruled 1 295-1304). He established his capital at Tabriz, which briefly became perhaps the major inter- national metropolis of the rime, a magnet for ambassadors, mer-

chants and artists from most of the known world, where Persians and Mongols mingled with Arabs, Turks, Chinese, Armenians, Byzantines and western Europeans. The Italian republics, notably Venice, Genoa and Pisa, maintained a\\ especially high commercial profile. It was Ghazan. an energetic and far-sighted ruler, who took the momentous decision to embrace Islam and thereby to anchor his own people in Iranian life and culture. This conversion, in which the Mongol ruling class participated, only confirmed the growing power of the Iranian bureaucracy and triggered an explosive expansion o\

output in the visual arts, m which a revival of national sentiment and of Islamic piety arc unmistakable. Hie balance of power had shifted.

Ghazan founded a suburb m Tabriz named after himself where, in the shadow of his own gigantic tomb tower, institutions of learning proliferated. He commissioned his vizier, Rashid al-Din, a physician ofjewish extraction, to write though perhaps ‘editing’ would better

describe the actual process) a history of the Mongols in the context

of a much larger history of the world. Parts ol this great enterprise



pposite) Sheer mass. Mosque of the vizie

‘Ah Shah. Tabriz, r. 1315; the qibk wall. Its

monumentaliry. echoed in other imperial

Dkhanid buildings, stimulated rival Mamluk

buildings in Cairo. The vanished entrance

portal was built to eclipse the Sasanian palace

arch at Ctesiphon near Baghdad – evidence

that this symbol of pre-lslamic majesty

remained a touchstone eight centuries Liter.

Iran’s Taj Mahal Mausoleum of

Oljeitu, Sultaniya, [304-15. Hiis complex

building, echoing both the Dome of the Kock

in its vast empty precinct) and nomadic ‘

ngol tents, tits into a long line of

royal mausolea in the eastern Islamic world.

• :n A numinous space. Interior ot

mausoleum of Oljeitu. This overwhelmingly

volumetric interior soars to <; m i~- inscriptions are ot politico religious ini

• ingthe Abrahamk Ka’ba and the co

ind more indirecdy M lerusalem, the three holiest cities <»t




155 Stucco as sculpture. Mihrab in the Friday Mosque, Isfahan, [310. Its inscriptions

stand proud of floral motifs which recall an undulating bed of water-lilies ]’he\ are

signed by Haidar, a pupil of the celebrated Yaqut al-Musta’simi, and mention the twelve

Shi’ite imams, thus reflecting Oljeitu’s conversion to Shi’ism.



have survived, perhaps because Rashid al-I )in ordered multiple illus-

trated copies of his work to be distributed at regular intervals to the major cities of the Ukhanid realm. Simultaneously, Ghazan reformed

the yasa to bring it closer to the shari’a and embarked on an ambi-

tious building programme, which was designed to provide every

village in the country with its own mosque, financed by the revenues of the bath (hammam).

His brother and successor. Oljeitu, was an equally lavish patron of

architecture, even founding a new city, named Sultaniya (‘The Royal’), to act as capital. Its cynosure was the mausoleum of Oljeitu himself, still one of the finest buildings in Asia. Structurally it was at

the forefront of building technology, with its vast pointed dome rising, it seems, directly from the octagonal chamber below but with the intermediary ot\\ spectacular vaulted gallery. Recent research has

highlighted its similarity to Brunelleschfs Duomo in Florence. It can be seen as one ofa long line of Islamic reinterpretations of the essen-

tial schema of the Dome of the Rock and as another indication of




the desire ot the later Dkhanids to identify themselves with then-

adopted culture. Craftsmen from many areas of the [lkhanid domains were conscripted to contribute to this \ast project, which invoked a

tuge precinct with numerous subsidiary buildings, and they dis- seminated the latest fashions and techniques on their return home.

A veritable building boom, at its height between [300 and 1 |

especially in such central Iranian cities as Qumm, Isfahan, Ya/d and Abarquh. but continuing tor decades thereafter, was triggered by

these vast imperial projects at Tabriz and Sultaniya. Existing mosques were expanded and enriched, as shown by the prayer hall, mihrab and madrasa .\ddcd to the Friday Mosque of Isfahan. As tor new mosques, 155 the Mongols favoured established types, such as the \-iwan plan, .is .it Varamin and Hafshuya; the covered sanctuary, as at the Masjid-i


Ali, [56

Quhrud and the recentlv destroyed Friday Mosque at Barzuk; and the isolated dome chamber, as at Kaj, Dashti and A/iran. The mosque of


Ali Shah at Tabriz is atypical, tor it consists only of a 152

courtyard and a qibla iwan. Like the tombs ofGhazan and Oljeitu, it illustrates the huge scale ot some Mongol foundations. The Mongol contribution lay principally in a refining and attenuation of Saljuq

tonus: one may compare the relation ot Gothic to Romanesque. Twin-minaret portals best expressed this new trend. The elaborate articulation of Saljuq transition zones was toned down, while 111

decoration the emphasis gradually shitted from brick patterning to

glazed tilework. which brought a dramatic infusion ot colour to

Iranian architecture. New accents such as saffron and green enriched the palette ot” the craftsmen, and tile mosaic with floral, geometric

and epigraphic decoration became widespread. So too did wall

painting, especially in mausolea; the designs were often stencilled and

rfarcrical Iranian mosque. Friday Mosque, Varamin, [322 6 Generously proportioned (according

to the 2:5 ratio familiar since ‘Abbasid tunes, and based on a grid of equilateral triangles), this traditional

4-nrjM structure stresses the longitudinal axis by a deep portal and a mighty dome at opposite ends


in d



157 ‘There is no God but He. the Mighty, the Wise (Quran 3:6) proclaims the text on these tour huge flawlessly executed lustre tiles. Kashan. c 1300. Each tile 57 \ 4- cm (22.4 x is.s in).


owed much to Qur’anic illumination, .is a group o( mud-brick buildings in Yazd and Abarquh show.

The Mongol period saw a major shift of emphasis in the building of mausole.i. with most tomb towers being built to serve religious (specifically Shi’ite) purposes rather than secular ones, and often dee-

orated internal}) with lustre tiles. The concomitant popularity o(

religious shrines, often of Sufi character, as the focus tor local as well

as imperial patronage can be seen m such sites as Natanz, Bastam, Linjan and Ardabil, with their pronounced welfare function. These

shrines, continually added to over the centuries, both depended on

and fostered local piety, and they also had an important role in the

local economy through the land they owned. Mongol ceramics are dominated by lustre tiles, which were pro-

duced in great quantities – especially in Kashan. where, as inscrip-

tions show, the tradition was sometimes handed down from father to son for generations. Several entire mihrabs composed of these rectangular tiles have survived. The production o\ such tiles seems

actually to have increased during the Mongol period, though even in this area a hiatus of about twenty-five years (c. 1230—55) is apparent

and was presumably caused by the chaotic aftermath of the Mongol invasions. Lustre tiles of star or cross shapes, often with interlocking

monochrome glazed tiles to act as a foil, created huge shimmering dados in religious buildings and palaces. Very often the individual

tiles would each bear a Persian inscription rendered in a hurried




;r) The fashion for the Far Fast. Pottery

bowl with lobed sides. Sultanabad type, early

14th century. The roundel depicts two Mongols

of high rank. The panels feature a running fox, arabesques and foliage of Chinese inspiration.

159 (below) Textiles: portable wealth, portable

propaganda. Silk and gold-thread slit tapestry

roundel, early 14th century, perhaps a royal table

Cover. Images such as the tortoise and crane by a

fishpond and the lotus scroll, and the overall

aesthetic link it to tapestry weaving (kesi) from

Chinese Turkestan and thus to Uighur culture.

But most of the iconography is Islamic, as are the

Benedictory inscriptions.

scrawl all along its outer edge. Floral or animal motifs are the staple

decoration of these tiles. Some Qur’anic inscriptions on these tiles, in defiance ot orthodox Islamic practice, have a background in

which birds feature among arabesque scrolls. In the larger rectangular tiles m which living creatures dominate the design, an effective combination of relief and lustre painting was devised.



While many other wares ot Saljuq type continued to be made under the Ilkhanids, two major new types of pottery appear. Lajvardina is a ‘simplified successor of the mina’i technique. The courtly scenes of the earlier ware are replaced by geometric and epi-

graphic themes and by Far Eastern mythical creatures. Gold over- painting set against a deep, royal blue glaze makes lajvardina ware one

of the most spectacular ever produced in Persia. The other new Ilkhanid ware is dowdy by comparison. Traditionally associated with the Sultanabad region, it is heavily potted and makes frequent use of

a grey slip with thick outlines, while another type displays black

painting under a turquoise glaze. The drawing is of indifferent quality but the ware as a whole has a special interest as a classic

158 example of the way Chinese motifs invaded the Persian ceramic tradition. Earlier, Chinese techniques and shapes alone had inspired

the Persian potter: but from now on his iconographic repertoire drew widely on Chinese sources. Dragons, phoenixes, mandarin

ducks, cloud bands, peonies and lotuses are all standard Ilkhanid

themes, and they are treated with a new naturalism also inspired by

China. Such Chinese elements are equally marked in the relatively

159 few surviving Ilkhanid textiles.

In metalwork as m ceramic production, the Mongol invasion fatally disrupted a flourishing industry. The ravaged province of Khurasan, in particular, never again supported a major mctakvorkmg industry. After a ^\p in production ofalmost a century, which can be paralleled

closely 111 architecture and painting, the industry revived – but in new

centres. One was m Central Asia; another was in Azerbaijan, the principal centre ot Mongol culture; but it was southern Iran really came to the tore. This area had been spared Mongol devasta-

160 tion. but was. of course, open [o Mongol Stylistic features. Hence there appear in the metalwork ot Fan such features as the peony, the lotus, flying ducks, ju-i heads a tri-lobed motif) and Chinese

phoenixes. The figures, slim and narrow-waisted, have something of the elegance that characterizes the figures of late fourteenth-century

Persian painting, a link that extends also to their costume. A general readiness 10 adopt alien fashions would explain the presence of geo-

metrical patterning of Mosul type and of the bold elongated thulth inscriptions that were the hallmark of contemporary Mamluk metal- work. But this same school seems to haw popularized the use of Persian poetry on metalwork. and its epigraphic formulae celebrating Solomon and Alexander are rooted deep in the Persian tradition.

In the field of the arts, apart from architecture, pride of place in




160 Women m authority. Detail ofa candlestick made tor the

Inju ruler of Fars, Abu Ishaq (reigned (341-56). This, one

offour decorative roundels with

enthronement scenes derived

from contemporary manuscript

illustrations, presumably depicts

hiN queen, who wears .1 version of the baqtaq, the headdress worn

bv married Mongol women; her presence on two of the tour

roundels corresponds to the

conspicuous honour paid to royal women in Mongol iconography

the Ilkhanid period goes to the art of the book. Ilkhanid senbes and illuminators, especially those of Mosul and Baghdad, rivalled the best

Mamluk work and may indeed have laid the foundations for it. Characteristic of this school is the tise of very large sheets (up to

72 x 50 cm, 2(S x 20 in) of Baghdad paper and correspondingly large scale scripts, especially muhaqqaq. The vaults m the gallery ofOljeitu’s tomb owe much to the designs in the frontispieces and carpet pages of these Qur’ans. of which a good two dozen have survived. Among the various traditions of Islamic manuscript painting

Arab. Persian. Turkish and Indian – Persian painting, which

effectively begins with the Ilkhamds and attained its classic style

tinder the Timurids. must take precedence on several scores. For

diversity it is without parallel in Islam; for sheer output it rivals even

India; and while the Arab world can boast slightly earlier work, it

cannot match the continuity of the Persian tradition. I he origins of

that tradition are probably destined to remain obscure, though

textual evidence establishes the continuity ot this art from the

Sasanian period. Iranian book painting tor the first five centuries of

Islam is thus an almost total blank adA must be reconstructed with



161 The Garden – and die tire I ho Chronology ofAncient Nations by the nth-century polymath al-Biruni discusses all the calendrical systems known to him, often with an Islamic slant. But the Zoroastrian storv of the evil spirit Ahrmian tempting Misha and Mishyana. the first man and woman, reflects Genesis iconography Perhaps Habriz, 1307.

the help of painted pottery, and a tew wall paintings from Samanid

Nishapur and Ghaznavid Lashkar-i Bazar, as well as the probably

Saljuq manuscript of Varqa va Gulshah (‘Varqa and Gulshah’; see

p. 100). The first really useful clues are provided b\ the late Saljuq painting practised in Iraq see pp. i_s -32). This probably reflects

contemporary Persian work, to judge by the painted Persian pottery

161 of the time, just as the paintings of al-Biruni’s al-Athar al-Baqiya

(‘Chronology ofAncient Nations’), dated 1307, echo the style of the

Baghdad school. From the eleventh to the fourteenth century, both Iran and Iraq were frequently part of the same political unit, so these

close links are to be expected; and indeed, both Tabriz and Mosul

have been suggested as the provenance tor this manuscript. Its

emphasis on calendrical systems, which caters to the same interests as

Hulegu Khan’s great observatory in Maragha (125S), testifies to the

Mongol interest 111 science, also manifest in illustrated bestiaries and




eligious pragmatism. The images ofRashid al-Din’s World History are multi-confes- sional; f. 1 5 b illustrates Qur\m 2:26] (cf. The Valley of the I )ry Bones, Ezekiel 37:1-14), which tells how Cod causes .1 man to die and be resurrected, along with his donkey and

1 century later. Note the Chinese conventions tor tree and stream. Tabriz, I] 14.

encyclopaedias such as the Mu’nis al-Ahrat fi Daqa’iq al-Ash’ar (‘The

Free Men’s Companion to the Subtleties of Poems’) by al-Jajarmi. The Biruni manuscript is a fortunate survival, for it documents the

invasion of the established pictorial idiom of cistern Islam by totally

alien influences, especially from the Far East. Hut the resultant

degree of Mux in fourteenth-century painting is. nevertheless, sur-

prising. Several distinct styles flourished, some of them owing little to each other and quite remote in spirit and m style from the pre- Mongol traditions. The Biruni manuscript, with its emphasis on non-Islamic heresies and faiths, especially Christianity, its astrological

content, and its choice of key ShTite themes, is ,1 case in point. I he

Manafi-i hayavan (‘On the Usefulness of Animals’, produced in the

1 290s m Maragha), essentially a bestiary, and the manuscripts illus- trating Rashid al-Din’s Jam?al-Tawarikh (‘World History*) share the

stress on Biblical themes. In content, ‘On the Usefulness of Animals’



[6j Chinese art Islamized. On the Usefulness ofAnimals, Maragha, 1290s.

“The Simurgh, found in inaccessible islands … is fearless beyond all other animals. He can carry off exceedingly large animals like the elephant and

the rhinoceros. . . .’ Muslims, like

Europeans, had their ‘Marvels ofthe Hast* literature, part fantasy, part

reality. This gaudy bird echoes the

Chinese phoenix.

– of which several other contemporary versions arc known – belongs firmly within the orbit of those practical treatises long popular in

Mesopotamia and issuing from an ancient Byzantine and classical

tradition. But this bestiary inhabits a different world from that of its

Arab equivalents. The clue lies in the artists’ partiality for drama.

They invest essentially undramatic subjects with a portentous power wholly at variance with the stirt”. woodenly articulated animals o\

Arab bestiaries. Some of the painters obey the formulae o\ Mesopotamian painting tor details of plants, landscape, drapery and

facial features. Other miniatures are infused with a new Chinese spirit expressed in the treatment of landscape details and especially 111

the overlapping planes that lend depth to a composition. Hut, .is in

164 ‘They came, they sapped, they burnt, they slew, they plundered, they departed * I he Mongols at work, as described m the World History ofRashid al-Din.




later times. Persian painters were never fully attuned to the artistic conventions that underlie Chinese painting. They preferred to borrow, and frequently to use out of context, eye-catching individual motifs such as exotic creatures (the phoenix [often representing the mmurgh, the bird oi Persian legend], giraffes, elephants), plants

ipeonies. lotuses), blossoming trees, and the conventions for render-

ing water, fire and clouds.

The manuscripts of the “World History’ of Rashid al-Din are on an altogether different scale. Their provenance in the cosmopolitan

cit\ of Tabriz guaranteed the paintings a remarkably mixed ancestry

in which Chinese, Byzantine and Uighur (Eastern Turkish) elements mingle with Persian and Arab strains. The large oblong format usually employed for these paintings allowed the artists ample scope

for scenes expressing the savage lust o\~ battle as well as for solemn

tableaux of enthroned monarchs. Their ferocious battle scenes, full of [64 authentic Mongol military detail, mirror the invasions that had trau- matized the Persian psyche a century earlier and whose memory was dearly still green. Conversely, their scenes from the Old and New- Testaments, the Buddha cycle and – for the first time in Islamic art the life ot Muhammad reflect the Mongol curiosity about religion. One manuscript, in London, is prefaced by dozens o\~ stereotyped royal portraits that are pastiches of Chinese models even to details of

dress and pose. The same obedience to formula governs the many court scenes in a contemporary codex o\~ the Diwan (‘Collected

Poems’) of Mu’izzi. But whether the scenes depicted are inventive or

merely routine, the hybrid style associated with the atelier ofRashid

al-Din is instantly recognizable. So united are the tones and so domi-

nant the role of line that many paintings resemble tinted drawings. Some impressive court scenes were produced 111 this style, and it

lingered for several decades. By 1330, the fashion for illustrated

Shahnamas bulked large in southern Iran, tinder the Inju dynasty. \ lere

national sentiment was fostered, perhaps because the area was not

tinder direct Mongol rule. The dating and provenance of the so-called “small’ Shahnamas pose a different set of problems, linn data have 165

proved hard to establish and ii^ood cases have been made for attributing these manuscripts to Baghdad and Tabriz; even Anatolia is a possibil

iry. At all events it is a Shahnama is the undisputed masterpiec e ot

the List years of Mongol rule. This is the incomplete Great Mongol

Shahnama, whose scale may reflect the growing commitment ot the

Mongols to the land they were governing. It is presumably a royal

manuscript made tor the last Ilkhamd ruler. Abu Sa’id l}l6



^&&^A6WMJJ* %&^k »&&U^ ^ter&i ctet^->ioi>

^9^^ tsa^rJ^j fefrfeW ^^^ i^tiN^H ^^fe^u’ J0^u^/. p*5^^ ^ihP^h !$J*^«itti >M*Jfc? B&H&tfi

^grKK** d*^iH*6r ‘$i&£fo ^^W&> ^^(f&bd r-S*tM^ U*0^>i^ k,^*^;.– frfoffifc’S* j^^yy,* y^it^t!^ iS*B»*feS6:

*-%&$ &-*£*&&> f^Xatjr/JW ^53ft – ^#*W* f^tfetfeU^- xU^U4 vtti&s^ >**AV-»&L *<&&}& 3^l ‘&&(&*$

x$»j> . ‘/


4hfii£^ ^ifc^yftte ^W^f ^w^ – v /L>>->^6^

16s (above) Pre-nuptial festivil ‘•’ ins riding an elephant celebrate, according to the caption, ‘the

physical union ofZal with Rudabah*. ‘Small’ Shaimama, provenance uncertain, c. 1300 Firdausi’s I the event in Kabulistan (modern Afghanistan), where there was indeed an 1 [th-century elephant park

too” (opposite above) The resurgence of Iranian myths. In the (ireat Mongol Shahnama (Tabriz, c. 1 – tale of Bahrain (mr. legendary lover and hunter, in reinterpreted as a moral evolution from irresponsible

playboy to just ruler. When, at his lover’s request, his arrow pins together a gazelles head, ear and hind leg, she reproaches him for his demonic spirit h\k\ he tramples her to death.

pposite below) The M ry of death. The bier of Iskandar. from the (ireat Mongol Shahnatm The unrestrained expression of grief flouts Muslim norms but mirrors Mongol practice and European art. The lavish appointments match literary accounts of the mausoleum of the Ukhan ( iharan.




Fifty-eigfit of its original two hundred or so illustrations are known. In certain painting? ambition and execution do not fully coincide; this is a style in flux. It is notable especially tor the dense design and spatial

complexity of many ofits paintings, their subtle gamut ofcolours, and a marked predilection for drama and violent action. Full of elements

from China and western Europe alike, alive with contemporary allu-

sions, and of an emotional power never recaptured in later versions o\ this text, it expresses at every turn millennial Iranian ideas of royal

1 66 legitimacy, and is therefore an apt metaphor for the resurgence of

Iranian national sentiment in later Ilkhanid tunes. So by the end oi

this period, as in the case ofGreece and Rome many centuries before, captive Iran had made captive its conqueror. The Great Mongol

\<- Shahnama is thus an appropriate envoi to the whole period, ushering

in the dissolution o\ the Mongol state and the re-establishment of native Iranian centres of power in much of the couir

After the collaps entral authority in i ; j6, Iran disin-

tegrated into several independent political entities, each with Us own geographical centre of gravity. Of these the most important were the Jala’irids who ruled western Iraq and Iran and the Muzarfarids who controlled central and southern Irai Both devoted their m patronage to architecture and to painting l.ila’ind architecture is best

[68 represented bv the madnsa and khan of Amir Mirjan in Baghdad, both built in the Lite the style ofintri >>tt.i

ornament and bold transverse vaulting first developed in late

>f their spatial subtlet\ is the triple

domed prayer hall in th< i the full) operational set of rooms in the u; I the khat

illery, a design echoed in .» Maaamai illustratioi In book painting tin- Jala irids took over the mande <>r Mongol

imperial pan Timenting both with new messages tor

familiar texts and with new relationships between text and mi.

The later fourteenth century saw the incubation of the rimurid ‘classical* style which lominate Persian painting for over two centuries. The ids built energetically in Isfahan where

several maus* I minarets datable i

in Yazd. which develope I painted ornament closely allied to Qur’anic illuminati the Mums!

ries ofmausolea . and in Kirman, as in the portals of the Friday Mosque and the Masjid-i r. which display a transitional style betweei M both in structure and in such decorative details as i ible mouldings ind tile mo



bon ups s rto ~ – m iw I h cvenucs i i


The last two decades ol the fourteenth centur) say ilu^ rist power ol -i conqueroi scarcer) less fearsome than lus anccsto Genghis Khan. Hmui the Lame, whose target than lite mentor) ‘Tamburlaine’) galvanized tlu N imagination ol Renaissance Europe,

was an illiterate tribal chieftain who gradual!) built up » contedcra tion ol riirco Mongol tribes from Centra) \m.» and beyond, launch ing them on a series ol victorious campaigns that lasted until lus

death in 1405 rhese frenetu conquests proved transitory; theit

memory was theii heritage rhe Iranian world* India, Anatolia, Syri%t .ill fell to him, with horrendous destruction and loss ol lite evi

though he posed as .» pious Muslim Craftsmen were spared from

these massacres and transported to Ins capital Samarkand, which the)

beautified vs ith spe< ta< ulai buildings, in< luding now \ inished pal u cs

with wall paintings depicting I imur*s victories I he) ilso worked in

1 ;



other Central Asian cities like Kish and Yasi. This empire tell apart

on his death as the standard Turkish practice of dividing the patri- mony among the various sons asserted itself. None of his successors had his military genius, but they applied

themselves with equal fervour to the arts oi peace, and rapidly

acquired a taste for Persian culture, as shown by the pocket antholo- gies of classic Persian poetic texts which they ordered for their per- sonal use. Samarqand under Ulugh Beg became renowned as a scientific centre; the gigantic quadrant ot his observatory still sur-

vives and the astronomical tables he drew up were m use at the University o\ Oxford as late as K>r>s. Shahrukh and his son Baisunqur, ruling at Herat, devoted themselves to literature and

painting respectively; the 1420s saw an ambitious attempt to com- plete the ‘World History


of Rashid al-I)in by bringing it into con-

temporary times, and histories ofTimur were written and illustrated. Painting also flourished under Iskandar Sultan in Isfahan and Shiraz,

and later under his cousin Sultan Ibrahim in the latter city. Each o{

these princes established his own court and kept an eye on what his relatives were doing in the cultural as well as the political sphere.

Thus emulation was .1 constant spur to achievement. Yet the cultural values espoused b\ these Timurid princes were not exclusively Persian; Chaghata) or Eastern Turkish was spoken at court, poetry

was composed in that tongue and the famous Mi’rajttama (‘hook o\ the Ascension*) ot [436 deputing the Prophet’s journeys to Heaven

and Hell has its primary text in Chaghata) with abbreviated cribs in

Persian and Arabu

The fifteenth century s.iw the Timurids consistently losing ground to their enemies, notably the Aqqoyunlu or ‘White Sheep’ Turcomans who dominated central and western Iran and eastern Anatolia from the u;><>s. while in the last years of the dynasty the

Uzbek Khans encroached on their Central Asian territories. Eventually the Timurid ’empire’ had shrunk to Herat and its sur-

roundings, and the much reduced revenues available to royal and aristocratic patrons sufficiently explain why there are tew paintings in the great manuscripts and win the decoration of the major buildings is concentrated at a few points only. Yet at the last moment this self- consciously exquisite civilization produced vet .mother adventurer,

Babur, whose dreams of empire were destined to be realized in India, where he founded the Mughal (i.e. Mongol) dynasty in 1526.

As with the Dkhanids, so with the Timurids architecture and the

arts of the book take pride of place over other art forms. With the

169 Here lies ‘the Scourge of God*. Gur-i Amir, the tomb ofTimur, Samarqand. from 1404. The high drum, stilted and fluted melon dome and glazed bricks spelling out sacred messages (such as ‘God is Eternal”) and the names of Allah and Muhammad all serve to transform the familiar schema of the domed square. By degrees this became a dynastic mausoleum.




irruption ot Timur and his hordes into Iran at the end of the four- teenth century, the desolation ot the Mongol conquest repeated itself. Craftsmen were once again transported, but this time to Transoxiana. where the signatures ot men from Isfahan and Tabriz have survived in architecture and metalwork.

Imperial Timurid architecture reflects political realities m that its centre of gravity is squarely within the north-eastern Iranian world.

Thus the areas ot Khurasan and Transoxiana replace north-western

and central Iran as the source ot’ major innovations in design as in

structural and decorative techniques. Yet the continuity ot Ukhanid

and Muzarfand modes is manifest, helped no doubt by the voluntary or enforced migration ofcraftsmen from central and southern Iran to

the centres ot Timurid power, and by the tact that Timurs cam- paigns were less destructive m these areas than elsewhere.

The Timurid period marks the apogee ot colour in Iranian archi- tecture, both in sheer technical expertise and in the astonishing

variety of designs and textures. Colour transforms exteriors and

interiors alike, yet it is not allowed to run amok. Most I linund

buildings project .1 solid sense ot the structural skeleton itself, which

the use ot’ colour enhances but does not overwhelm. This delicate

balance was apt to be lost in subsequent centuries. The equilibrium

between structure and decoration meant that brick, the basic build-

ing material, was available to serve as a toil, both in colour and in

texture, to the applied ornament. The designs themselves, tor

example medallions and arabesques, can often be matched in other

media and many ma\ well have originated in the royal ateliers, to be

circulated later tor use in bookbindings, carpets, textiles, manuscript

illumination, potter) and woodwork as well as architectural orna- ment. Hence, perhaps, the tendency tor general decorative schemes

-applied on an architectural facade, say, or a vault to be subdivided

into individual and seemingly unrelated panels si) that the overall

144 effect is that ot a picture gallery with paintings ot varying size hung at different levels. Arabic and Persian poetry, felicitously described as

‘orient pearls at random strung’, suggests itself as an intriguing parallel. Sometimes the decorative scheme is strongly sculptural, as in

the use of glazed medallions standing proud from a glazed surface, or

the multiple levels of Timurid vaults. More common, however, is a contrast ot texture, for example glazed terracotta juxtaposed with

smooth glazed tiles, or carved stucco set against painted plaster, or contrasts of marble and glazed tilework, wood and ivory. Yet the use ot snow-white muqamas domes or bottle-green dados reveals




Timurid craftsmen experimenting with the potential of a single colour to dominate an interior.

Colossal size is the defining feature of some of the most chanu istic imperial Timurid buildings – the Rigistan and Gur-j Amir in Samarqand and Timurs own palace at Shahr-i Sabz, whose iwan apparently soared to 40 m (131 ft). Indeed, the portal now takes on major significance as the cynosure of a facade, often dwarfing the

actual building behind it. as at Anau. At the shrine of Gazur Gah near Herat, it may symbolically suggest entry to the hereafter. Flanking minarets placed behind the tomb at the far end o\~ the building make some of these iwans illusionistically still more lofty Many such portals function as huge screens or hoardings inscribed with religious or political messages, but others are proudly salient

and of spatially adventurous design. Perimeter walls, too. take on a

new importance which is reflected in the overall brick and glazed ornament which they bear. These are buildings meant to be experi- enced in the round, not designed with a single viewpoint 111 mind.

The popularity of ribbed domes, high drums and multiple minarets [69 (as in the Bibi Khanum mosque 111 Samarqand and the madrasas o\~ Herat) again reveals Timurid architects to haw been fully alert to the scenic dimension of their buildings.

Fifteenth- and sixteenth-century drawings found 111 Istanbul and

Tashkent contain, among much other material, detailed notations for the layout of ground plans

and the construction of muqarnas vaults.

Their use of gridded paper and modular units provides independent

documentary evidence for what could be deduced from the monu- ments themselves — that a mastery of geometrical concepts and of proportional relationships was needed to control these vast spaces and

to order them into harmonious, symmetrical designs. It is size above all that empowers such factors as axiahty. rhythm, repetition.

anticipation and echo to yield their full effect. Thus in the \-iwan

courtyard madrasa ofUlugh Beg in Samarqand (141 7). the compo- nent parts are all interdependent and logically related to each other,

while at the Shah-i Zinda – a necropolis largely intended, it seems. 171, [72 for Timurid princesses – the individual mausolea are not sited hap-

hazardly but operate m concert, forming a processional way towards the tomb of the eponymous saint. A long monumental staircase creates a suitable air of expectancy and ensures from the outset

pilgrims are channelled towards the tomb along the desired route. It

is ,1 textbook case of the capacity of Timurid architects to think big

and to exploit space to the full. The whole site seems to have been



deliberately designed as an open-air gallery displaying the latest dec-

orative techniques. Perhaps there was a certain competitive edge too.

It is no surprise” that major Timurid artists such as Qawwam al-Din Shirazi were figures of consequence at court.

A fascination with the expressive potential of vaulting can be sensed in the more experimental Timurid buildings. Gone now is the Saljuq and Ilkhamd preoccupation with the tripartite elevation of

a dome chamber – base, transition zone, dome – and in its place there reigns a much more fluid transition from one plane to the next. Typically a network ofsmall vaults, often ofrhomboidal form, cloaks

the upper reaches o\\\ building (with simpler, sturdier vaults behind

them doing the actual work); these give way eventually to the dome itself, which rises serenely above the apparent contusion below. The polychrome vaults ofKhargird and Herat, instinct with dynamic ten-

sion, create a vortex of frantic pyrotechnic energy which constantly

teeters on the brink of chaos. Vet this explosive power is of course

entirely controlled, as can be understood at once from a plan of such

a vault. The heaven!) associations oi domical vaults, complete with fixed and shooting st.irs and a central solar motif, are unmistakable.

Relatively little high-quality metahvork has survived, though

miniatures of the period whose obsessive detail makes them an

excellent guide to contemporary luxury objects) show that ewers

with long curved spouts were developed at this time. A few spectac- ular but isolated survivals give a clue to this largely vanished industry.

Thev include a candlestick base formed by knotted dragon heads, tall




170 (opposite) The holy man as

a source of authority’. The

shnnc complex of Shaikh

|anul al-Din, Anau. [455-6.

The buildings apparently

comprised a mosque, madrasa

and convent (khanqah), all

domed, plus accommodation

tor visitors, dervishes and

pilgrims around the courtyard;

a loth portal featured spandrels

hearing images ot dragons,

presumably as talismans.

1-1 {right) Gateway to death

and life. Entrance to the Shah-i

Zinda complex. Samarqand,

mainly c. 1350-1450; dedicated

to the cult of the saint Qutham h. al-


Ahbas. a cousin of the

Prophet. He allegedly met Khizr (guardian of the Water of

I ife) and continued to live m splendour under his own tomb.

Mir) City of the i.k\\d.

General view of the complex of

Shah-i Zinda, Samarqand. The desire to develop an expressive

skvlme explains the emphasis

on hill-top sites and

exaggeratedly high drums.

tint] > * “, ;,.’.: ?rft TTPI * tmTPlTnfrrinnniiin*



tubular candlesticks with a succession of bold annular mouldings, and

173 a pair of massive bronze cauldrons, now in St Petersburg and Herat respectively, both made in the 1390s. The closest analogies for these impressive vessels, which are virtually undecorated apart from their inscriptions, are in the metalwork ofDaghestan in the Caucasus. In most Timurid metalwork. the Saljuq and Mongol motif of a figural scene within a cartouche seems to have been definitively superseded

by closely knit floral designs. A new style heavily dependent on manuscript illumination is found on Khurasani inlaid metalwork in

the later fifteenth century, while the same province generated

simultaneously a style ot~ engraved metalwork that leads without a

break into the Safavid period. Persian poetry is a staple feature o\

the decoration o\ much Timurid metalwork; it often has Sufi over- tones. The inlaid brass jugs that were a speciality o\ Herat shortly before 1500 illustrate these features.

Recent research has demonstrated that the talk’ of surviving

Timurid ceramics is not nearly .is meagre as it was once thought to

be. The vogue for chinoiserie continued unabated. Indeed, side In- side with such traditional techniques as lustre, the quality of which

was appreciably lower than in earlier centuries, the Persian potter

173 Amassing credit tor the hereafter. Bronze basin intended to serve water to pilgrim*

visiting the shrine of the Sufi shaikh Ahmad Yasavi. Inscriptions on the basin state that Timur ordered it in 1399 for this shrine; they also quote the particularly relevant Sura

9:19 and the Prophetic hadith ‘He who builds .1 place for drinking for hol\ purpos will build tor him a pool in Heaven’ I )iam 2 ;



now produced blue-and-white wares inspired by imported porcelain of Ming type. The ultimate origin of these wares is disputed, but reciprocal influences between China and Persia are certain. Timurid copies of this Chinese porcelain body are also known. Quite difTer- ent in style is a category of pottery made in northern Iran from at least the 1460s until the seventeenth century. These ceramics were all

found in the Caucasian village of Kubachi, a famed metalworking centre; presumably they had been exchanged for local metalwork. The earlier pieces of this school are painted in black under a bluish- green glaze and eschew figural designs in favour of floral cartouches

or scrollwork and epigraphic motifs.

The origins ofTimurid panning are mysterious. The exact nature oi the debt it owes to the early Jala’irids, whose activity in this held was mentioned above (see p. 212). is still a matter of lively dispute, and the potted history of earlier Persian painting with which the librarian I Hist

Muhammad prefaced the muraqqa 1

(album) o\~ the Safavid prince

Bahrain Mirza in 1 S44 is tantahzmgly incomplete and obscure in its

treatment of the fourteenth century. The backbone o\ his account is not chronology but a sequence of masters and pupils, and it is no easy

matter to match their reported output with the fragments that now survive in Istanbul and Berlin. But there is no doubt that .1 whole series

of key decisions had been made by the time that the Mathnavis (poems m rhyming couplets) ofKhwaju Kirmani was painted for Sultan Ahmad fala’ir at Baghdad in 1396 by a certain Junaid, the first Persian painter oi

the new age whose signature survives on his work. Hence the sheer assurance and the dazzling virtuosity of this masterpiece. The size ofthe book has decreased; paintings are much fewer 111 number: text is some- times whittled down to a brief two-line panel placed at will in the picture space: and the full-page illustration has now come to stay. \ high rectangular format with a correspondingly high horizon permits

an uncluttered arrangement of numerous spatially distinct figures, hue.

the emotional range has been toned down and the sense of drama evap- orated in a dreamlike fantasy world. But the technical skill ofthe artist is

now staggering. It embraces the preparation of the surface, the appli< 1 tion ofpaint, the purity of colour, the balancing of hues, the effects ot

crescendo and diminuendo in the composition and in the distribution

of colour, and the pinpoint accuracy of the smallest detail. Moreover,

while the paintings are often physically no smaller than the greatest

Dkhanid paintings, they contain \er\ much more. I his is truK mini

ature painting. Everything is calculated; these an- images that demand a deal of the viewer, and they <\” not yield up then sc< rets lighdy.



us*Cf} !ls> >JP^Si ijrjlfisjl *j£f\3 >/

r s

m <.>-

174 A minor tor princes. Dimna visited in prison by Kalila. from Kalila wo Dimna, a collection of animal tables with moral and political applications. Herat. 1429 (f. 5 6a). The calculated technical perfection so characteristic of Baisunqur’s atelier led to a certain stiffness in the treatment of figures and landscape, though new accommodations are forged between written and painted surface.



175 Hunting: the royal pastime. This double frontispiece of’.. 1470 may depict the then ruler of western Iran. Uzun Hasan. It shows a grand battue, or mass hunt, in which the game is driven tor many days until trapped m the constricted killing fields. Plunging vistas and the semi-circle ofspectators draw thi- eve to the frenzied slaughter at the heart oi the painting. Leering grotesques people the to*, ks

Persian painting had now found itself and henceforth generations of artists strove to achieve perfection in this style. Important centres

60m c. 1390 to c. 1420 included Baghdad and Shiraz, the Litter espe- cially under the rule of lskandar Sultan, but this style unquestionably

reached its peak in the work of the academy founded .it Herat by the

celebrated bibliophile Prince Baisunqur b. Shahrukh (1397-1433), who in the intervals of the dissipation that plunged him into an early grave found time to oversee the production oi fastidiously choice

illustrated copies of the classics of Persian literature. Some fort) artists were active m his atelier, including not only painters but also scribes, illuminators, gilders, tent-makers, designers, bookbinders,

leatherworkers and sculptors; sour- were proficient at several crafts.

Moreover, their status had risen to the extent that some were boon

companions of the prince himself The great masterpieces ot this



174 Herat school that survive include two copies of the Kalila wa Dimna, an Anthology, a Gulistan (‘Rose Garden’) of Sa


di (1426) and

at least one Shahnama (1429), for which Baisunqur – who had commissioned a new edition of Firdausfs masterpiece – himself wrote a preface. This had a tenth of the illustrations of the Great

Mongol Shahnama (see pp. 21 1-12). and that difference alone betrays the new aesthetic. Imperial Timurid painting was produced for con- noisseurs, so the painters took care to load every rift with ore.

Several other quite distinct styles developed in the fifteenth

century. One was a simplified recreation of the manner (including the format) of the Rashid al-Din manuscripts, which was used for

primarily historical texts produced at Herat for Timur’s son

Shahrukh (ruled 1405-47). Another was a vigorous, bold, minimalist

style whose keynote is action, featuring huge indexible figures

woodenly disposed within a ruthlessly simplified landscape: this was

developed under the patronage o{ Ibrahim Sultan, a brother o\~

Baisunqur, at Shiraz. The same city was later a major centre for a third style commonly dubbed Turcoman after the ruling dynasty of

17s western and southern Iran. Commercial production predominated;

its hallmarks were a revelling m picturesque it fanciful landscape detail in an astonishing range of greens, and squat, jowly figures with

rosy cheeks. These features were heightened and enriched in court

Turcoman painting after about 14V*. especially under Sultan Ya’qub in Tabriz, in several fragmentary Nizami and Shahnama manuscripts. These attain a psychedelic exuberance of colour allied to complex,

multi-planar compositions often with a sense of illimitable distance

not hitherto encountered in Persian painting. The scale and ambi-

tion of these pictures explains clearly enough why the projects of which they were part remained unfinished (see frontispiece). The patronage of” the last Iimund prince. Sultan Husain b.

Mansur b. Baiqara reigned 1468—1506), was on a scale to rival that of his contemporary Lorenzo de’ Medici in Florence. Herat flour-

ished as never before and main believe that here Persian painting reached its apogee. In these decades the names and achievements of

painters – such as Qasim b.’Ali. Aqa Mirak and. above all, Bihzad –

begin to attract the notice of” chroniclers. Yet these .ire perhaps not

the competing geniuses of” Renaissance Italy but might rather be seen

as colleagues and courtiers in a royal atelier, pooling their talents,

developing an increasingly accomplished and seamless house style

and maybe even working together on a picture. Often, therefore, Western scholars bent on attributing this or that painting to a given



Ij6 The ruler rebuked. In this scene from Sa’di’s Bustan, copied in Hor.u

in [488 for Sultan Husain Baiqara,

the prince, who has lost his way while hunting, comes across his own horsemaster and. not recognizing

him, strings his bow to shoot. His unthinking aggression disrupts the

peaceful, ordered scene. I’he

horsemaster knows his own charges by sight and fearlessly tells the prince

to follow his example.

artist seem to be chasing a will-o’-the-wisp. Moreover, the nature of Timurid connoisseurship is difficult to pin down, for the critical

vocabulary it uses has no unambiguous set of western equivalents. What defines this late Herat school? Once again, the pictures are few in number in any one manuscript, and thus each one matters. The singular intensity of colour and vibrant chromatic contrasts owes

much to the Baisunquri academy, but the delight in spatial complex- ities, variegated poses and individual types is new. None of this amounts to realism; these painters, like their predecessors, customar-

ily favoured the general at the expense o\~ the particular, and their

paintings testify as much to intellectual abstraction as to patient observation. Figures are sharply differentiated, but the sense ot .1

living, unique personality is generally absent. Quite apart from these

factors, however, is a new spirit reflected both 111 what is chosen for illustration and in what is omitted, as well .is m the manner ot stor\ telling. It seems that a democratic spirit was 111 the air, a quietly sub-

versive set of values that exalted the common man and his daily t.isks and made him a mirror for the selfishness of his ‘betters’. Heme an unprecedented emphasis on daily lite on a building site, in a bath.

at pasture, in a cemetery – with people buying and selling, cooking,

digging, cutting wood and ploughing. Often these humble scenes are .1 metaphor for the spiritual realities ot the Sufi path I hus form and

content combine in .i profoundly satisfying synthesis.





The Safavids

The dynasty that made Iran Shfite venerated as its ancestor and took its name from a saintly but not Shi’ite shaikh, Safi al-Din (d. 1334), the head of a sectarian Sufi order based at Ardabil in north-western

Iran. The masters ot this order exercised gradually increasing reli- gious and political authority, until by the late fifteenth century they

were a force to be reckoned with in eastern Anatolia and north-

western Iran, with several charismatic leaders. F : or almost two cen- turies after the fall of the Dkhanids m [336, apart from the meteoric career ofTimur (see pp. 213 14 . Iran had lacked cohesiveness or rel- atively fixed boundaries. This period was also one o\ religious

ferment which witnessed a flowering t»t folk Islam, Sufism and

extreme Shi’ism. I he Safavids combined these three elements,

turning the order at Ardabil into a revolutionary and at times mes-

sianic Shfite movement originally dominated b\ Turkish tribesmen (the qizilbash or ‘red-heads* from eastern Anatolia and Azerbaijan,

and creating a successful political system which from 1 soi was

quickly imposed b\ Bat and terror on the country as a whole. A Turcoman military aristocracy held power under the authority of the shah himself”, though their internal clan feuds, their rebellions and

their antipathy to the Persian bureaucracy continually shook the

body politic. Yet the shah needed both the ‘Men of the Sword” and the ‘Men ot~ the Pen”; and the second Safavid shah. Iahmasp

(1524—76) eventually created a third force at court, a corps ofghulatttt

(slaves) comprising Georgian, Armenian and Circassian converts to Islam, to stabilize the situation.

The Safavids taught that their legitimacy depended on their descent from the family ot” the Prophet Muhammad, on their authority as masters of’ the order at Ardabil and on the divinely

ordained office of shah as the shadow of God on earth. These ideas were propagated by a hastily assembled body of theologians, many of them from the Levant, Iraq and Bahrain, who with Iranian converts

177 The Masjid-i Shah (now Masjid-i Imam), Isfahan, largely 1612-30. This spectacular hlue-uled mosque epitomizes one era. one style and one man – Shah ‘Abbas, the greatest Safavid ruler. Though its portal ter- minates the long axis of the great mmdan or piazza, the rest of the mosque is set at an angle from it so as to be correctly oriented towards the qibla. Thus sacred and secular geometry diverge.




% … s ^



from Sunnism created a new religious establishment. It was for such teachers of religious law (mujtahids) to exercise their personal judg-

ment (ijtihad) until the ultimate return of the Mahdi, the Hidden (and twelfth) Imam. In later Safavid times their power at court grew significantly. The regime was thus thoroughly theocratic. By making Shi’ism the official religion the Safavids forged an ideology that not

only strengthened the state but also helped to create a new sense of national identity and so enabled Iran to escape being absorbed into

the empires of the neighbouring superpowers – although its borders were frequently contested and the shah had to fight a war on two

fronts. It was the Safavids who made Iran (with the old Shi’ite centres ot Iraq) the spiritual bastion of the Shi’a against the

onslaughts of orthodox Sunni Islam, and the repository o\ Persian

cultural traditions and self-awareness. They largely lVrsiamzed a country whose Turkish. Arab and Kurdish elements had hitherto been stronger, and to some extent they ruptured the cultural as well as religious ties that had earlier bound Iran to the Islamic common- wealth. At long last Shfism had round a ‘national



Yet for the indigenous Iranian population Twelver Shi’ism was at

first alien, and powerful Opposition manifested itself”, especially in the

east of the country In much the same way the rule of the Ottomans in Anatolia, the Near East and North Africa, and perhaps even of the

Uzbeks m Central Asia and the Mughals m the Indian subcontinent, can be interpreted as an imposition of religious conformity which

coincided with a hardening of political, national and even religious

boundaries. This. then, was the age of the Islamic superpowers; and,

significantly, all ot them shared the same Turco-Pcrsian rather than Arab culture. 1 bus Islam, like Europe, emancipated itself from its

medieval heritage by creating larger political groupings. The rulers n\~ these superpowers were keenly competitive, alert to match claims

(e.g. to the caliphate with counter-claims. Their horizons were

wide. Isnu c

il. the first Safavid shah, bore the title of Persian Emperor

(Padishah-i Iran) with its implicit notion of an Iranian state stretching

from Afghanistan as far as the Euphrates, and from the Oxus to the Persian Gulf.

The long reign o\~ [sma’iTs son Tahmasp helped to establish Iran’s role vis-a-vis its neighbours, to tone down religious extremism and to control the power of the clergy. Hut it fell to Tahmasp’s grandson, Shah “Abbas 1 ruled [587- [629 . to set the country on the road to

greatness by creating au efficient standing army (in which the role of

the qizilbash was much reduced and that ofthe ghulams strengthened)




and a centralized administration, and therein- to lay the foundations of the modern Iranian state in its political, religious and geographical aspects. And that state he regarded in some sense as his personal property, which he governed (and milked) through the hierarchical administrative apparatus of his court, headed by the Grand Vizier and the Intendant (nazir), the latter functioning in effect as treasurer.

It is under the Safavids that one can trace more clearly than ever before the stirrings of a national sentiment that would eventually become, centuries later, a fully-fledged nationalism, and tor which

the territorial integrity established by the Safavids was a necessar)


The Safavids continued the attempts of the Ilkhanids to foster closer diplomatic ties with the European powers, as evidenced by the frequent exchange of embassies with the various courts of Europe in

order to cement alliances against the Ottomans. Similarly, they were alert to the political and economic implications of the opening of the sea route to the Far East in [496, which diverted Ottoman pressure away from Iran to the Red Sea and the Indian Ocean; the Dutch, the English and the Portuguese were permitted to establish trading

posts on the Persian Gulf, where Indian merchants also settled. For

the Iranians, this meant revenue from customs dues, while for the European powers such posts were essential if they were to control the

increasingly lucrative East India trade. The inevitable clash of inter- ests, however, resulted in frequent hostilities, especially with the

Portuguese. Attempts were also made to avoid Ottoman customs dues by relocating the silk and spice routes to the north across

Russia. Much Safavid silk reached Europe, especially the I labsburg domains and Scandinavia, in this way (see p. 250). Indeed, some tex-

tiles and carpets were made specifically for the West, and bear (not

always accurately) the arms of royal and noble houses. Conversely,

Europe exported muskets, mail shirts, clocks, Italian paintings.

Chinese porcelain, Japanese screens and even plants, fruits and veg-

etables unknown in Iran, and Shah ‘Abbas had several Europe his permanent service. The dramatic increase 111 commercial diplomatic relations with the European powers was fostered by

tolerant and multi-racial society 111 Iran. Colonies of Armenians.

Georgians and Hindus were settled in villages or key towns, while

western orders such as the Augustinians. Carmelites and Capuchins

rounded religious houses m Isfahan and other major centres as part oi .1 world-wide missionary campaign which also embraced China.

Japan and the Americas at this time.

ins 111






178 ‘Isfahan is half the world’ runs the Persian proverb. This enormous open rectangle the nerve centre of a capital to which visitors flocked from Fast and West

the iinuJdii was

179 An Islamic invention. Portal arch. Masjid-i Imam. Isfahan. The muqamas or honeycomb vault has many functions in Islamic architecture: it articulates a curved space, dissolves surfaces, bridges contrasting spaces,

and creates a frame for related but discrete motifs .see 1// 50). I Erectly over the door is the Shah’s name.

The principal achievements of the Safavids were architectural. Pride of place goes to the expansion of Isfahan masterminded by

Shah c

Abbas I from [598 onwards; it is one of the most ambitions

and novel sehemes of town planning 111 [slamic history. This resulted

178 in the famous maidan, which, with its measurements of 512 x [59 m (i6(So x 523 ft), is perhaps the largest piazza in the world; the Chahar

Bagh esplanade and royal quarter linking the maidan with the river

(the Zayandarud); and the huge covered bazaar. It may be no coin- cidence that a mere two decades earlier the Mughal emperor Akbar had built a sumptuous new capital from scratch at Fathpur Sikri.

The Masjid-i Imam, the Masjid-i Shaikh Lutfallah and the inter- dependent complex ofmadrasa, khan and bazaar built by Shah Sultan Husain are the finest public buildings of the time.

Yet it would be a mistake to regard any of them as fundamentally original buildings. The Lutfallah mosque (1602-19) and the Masjid-i







Shah (1612—30), each repeat a familiar schema – the domed square chamber and the+4-iwan plan respectively. Their exceptional size and

179 splendid decoration make it easy to overlook their essential conser- vatism. In the Masjid-i Shah (now renamed the Masjid-i Imam) this huge scale allows the incorporation of dome chambers behind the subsidiary iwans, a rectangular pool which serves as the focal point of the courtyard, ample facilities for ablution, a winter prayer hall and

madrasas flanking the main prayer chamber. Yet visually all is sub-

ordinated to the sheer bulk of the portal and qibla iwan and the prin-

cipal dome chamber. As tor the Lutfallah mosque, the structural complexities of earlier domed squares have been toned down dramatically, leaving a vast and minimally articulated inner space

which is organized mainly by its ornament – for example, the bright

blue cable mouldings which define the pendentives or the two-

dimensional honeycomb ornament of the inner surface of the dome. Exterior and interior alike accord a major role to plain unglazed

brick, which serves as a background to sparingly applied glazed tile-

work in floral and geometric patterns; the interplay between these

two accents seems illusionistically to confer a glazed sheen on the

plain brick. The facades of both mosques open on to the great maidan, which was the centre of the new city, and both have bent entrances, so that the mosques themselves are correctly orientated

but do not compromise the regularity of the facades defining the

square. Those facades are kept low so that the major buildings which

punctuate them stand proud of their surroundings. Here, as in so

much Safavid architecture, one may detect an innate sense of theatre and a delight m the grand scale. The high officials of the Safavid court, often in response to direct

pressure from the shah, built widely m Isfahan, including several mosques (e.g. the Masjid-i Hakim . but as a group these do not display any marked originality. The complex of Shah Sultan Husain

on the Chahar Bagh [706—15) is a resounding coda for Safavid

Isfahan: deeply traditional in its core plan – this is one of the largest

Iranian madrasas – spacious throughout, its key are. is (and those

alone) richly decorated with glazed rilework, and the whole creating

an agreeable sense ot~ rus in urbe with its pools, trees and gardens.

Moreover, the complex worked in economic as well as aesthetic

terms, for it comprised a huge caravansaraj and bazaar adjoining the

madrasa, and the revenues of the secular establishments funded the

religious foundation. Indeed, such complexes flourished throughout

Iran in Safavid times, whether these were shrines (Ardabil, Qumm,





1S0 The hub of commerce. Caravansarai built by Shah Sultan Husain with adjoining madrasa and bazaar; Isfahan. 1706-15. Spacious and austerely practical, it could accommodate hundreds of animals and theii loads, while the two-storied arcades held living chambers.

Mashhad and Mahan – many of them comprehensively remodelled in this period) or institutions that were both religious and secular (for

example the foundation of Ganj c

Ali Khan at Kirman). Isfahan is notable for its secular architecture too, especially the royal

palaces like the c

Ali Qapu, Chihil Suttm and Hasht Bihisht. Embowered m gardens and embellished with verandahs, wall paint- ings and fanciful muqamas vaults, they expressed to perfection the

luxurious lifestyle of the court. The c

Ali Qapu (‘Sublime Porte’) is an arched portico crowned by a flat-roofed balcony with wooden columns, from which the shah and his entourage could watch specta-

cles 111 the maidan below. Like several other Safavid palaces, it was

designed to be seen frontally or at an angle, not from behind. The formal gardens and watercourses into which it leads were once scat-

tered with courts, two-storey open-plan kiosks, and pavilions, of

which one. the Chihil Sutun (‘Forty Columns’), has a Bat-roofed

portico with wooden columns, following pre-Islamic Persian practice. This precedes the shah’s throne room. The sculpture in the surround

ing garden may also be intended to evoke ancient imperial memories. The Chahar Bagh (Tour Gardens’), an avenue lined with trees.

streams and the palaces of the nobility (evocatively named after roses,

mulberries, nightingales and Paradise itself’) gave access to the capital

from the south – a curtain-raiser nearly a mile long Massive bridges,

such as the Pul-i Allahvardi Khan with its thirty three arches, or the

Pul-i Khwaju featuring not only sluice gates but also pavilions tor the





royal party which served as vantage points from which to watch regat- tas and other water sports, linked Isfahan with some of its suburbs.

In secular architecture the network of khans, or caravansarais erected across the country by Shah ‘Abbas I deserves special note. Most of them follow a \-iwan plan, with the space in the corners serving for stables while the entrance and domed vestibule define the major axis. Features common to many caravansarais include a single massive portal, with an entrance high enough to admit a loaded camel; kitchens in the corners; and adjoining enclosures for tether-

ing animals. Caravansarais in the open country were often built at intervals of a day’s journev – about 2$ km (16 miles) – along the major trade routes. Those in the towns (often very numerous; Isfahan, for example, had nearly two thousand in the seventeenth

century) served not only to house and teed travellers, but also as

warehouses and centres tor a particular trade or group of merchants.

However. Safavid art has tended in modern scholarship to be dominated by the undeniable glamour of the Isfahan of Shah



to the detriment ot the Iranian art produced m the previous century under less charismatic rulers. The penalty tor this over-exposure has

been a seriously distorted perspective ot early Safavid art, in which it

181 A Sasanian tradition revived. The Allahvardi Khan bridge, Isfahan, stretches 300 in (984 ft) Designed to carry traffic and regulate flooding with its massive buttresses; u also has pleasure pavilions.



S indarwood and ivory sarcophagus of the dreaded Shah Isma’il I. founder of the Shi’ite Safavid state ‘on seeing him. outsiders would prefer to turn to stone’ at the shrine of Shaikh Sari, Ardabil, .. i sj i

is underestimated and seen merely as a curtain-raiser tor what was to

come. The surpassing quality of book painting and of carpets in the early sixteenth century suggests that such an attitude is a cardinal

mistake, and that it would be worth investigating other media in an attempt to redress such an injustice. The lack of substantial art- historical research on the reign of Shah Isma’il has been a major

barrier in this respect. So too has the poor survival rate of the build-

ings produced under his patronage, and the fact that Axdabil, the first

major centre of Safavid power and art, has been unaccountably oxer-

looked by art historians – except tor its architecture. The shrine has, of course, been comprehensively pillaged over the last two centuries

and the process continues to this day. What survives comprises not only architecture but metalwork (including several standards or

Warns), carving in wood and stone, and tilework. The constellation of talent seen displayed in the decorative arts of Ardabil makes .1 re

assessment of early Safavid work imperative. It is worth noting that.

despite Shah [sma’il’s disastrous defeat b\ the C >tttnn.u is .it ( haldiran

in is 14 and the consequent loss of eastern Anatolia, he did mana



to bring under his control all of Iran including much of Afghanistan and some of the -neighbouring territories north of the Oxus and the Araxes rivers. This gave him a far more extensive financial base than had been enjoyed either by the Timunds or by the Turcoman dynas- ties, and it therefore allowed him correspondingly greater latitude in his patronage of the arts.

The Ardabil material in architecture and the minor arts is also of vital importance in highlighting the continuity between late Timurid

and early Safavid work; if that continuity is not recognized, the

transition to the mature Safavid style under Shah c

Abbas I is lost, and

instead one is faced with an abrupt and perplexing juxtaposition of two essentially unrelated styles. Finally Ardabil is important in that, for the most part, it is a kind of time capsule in which the arts of

early Safavid Iran are displayed side by side and medium by medium, to create an ensemble in which, for once, the decorative arts cm be seen in context, enhancing each other and almost bandying themes

across the space of the shrine.

182 The sarcophagus of Shah Imii.i hi 1 has the traditional rectangular form, but is of quite special splendour. Timurid imperial woodwork, as demonstrated by the doors m the Gur-i Amir and the shrine of Khwaja Ahmad Yasavi, had effectively marked the technical limits of fine, filigree carving in wood. At Ardabil the carvers struck out in

another direction, drawing inspiration from work on a much smaller scale, such as boxes inlaid with ebony and ivory. The geometric strapwork which forms the skeletal structure of the decoration is

familiar enough: but its interstices are ornamented with grace notes

applied almost parsimoniously at key points, in black-and-white

marquetry. Tiny spots of green probably stained bone or ivory) add

an extra touch of luxury. The borders are marked by further clusters of marquetry in a sprightly dancing rhythm.

Many doors in the shrine are entirely silver-plated, and some are 183 gilded too. A common design is a curvilinear lattice work whose

slight rises and dips m plane create an undulating quilted effect. Multiple mouldings constitute the only added ornament, so that

these doors project a powerful sense of harmony and serenity – an excellent foil for the more assertive accents of tilework, carpets and carved wood.

Ardabil also sets the scene tor subsequent developments in the

visual arts, which continued to cluster around the capital city as the

court removed first to Qazvin and finally to Isfahan, ever further

distant from the sensitive frontier with the Ottomans. In this respect




the Safavids continued and consolidated a process which can alread) be detected under the Timurids, wherein most art of top flight

quality in the principal media was produced in the immediate orbit

of the court. In practice this spelled the disappearance of.provincial

ateliers of the first rank, though it did not exclude occasional lavish

patronage on the part ot a provincial governor, tor example at

Mashhad or Kirman. The Safavid capture of Herat in i so~ meant that the I lmund

library and its craftsmen, including Bihzad, tell into Safavid hands

and were eventually transported to the new capital ot Iabn/. under

the patronage ot Shah Isma’il I. His successor. Shah Tahmasp, himself

a painter, even expanded the royal atelier. Early Safavid painting combined the traditions of Timund Herat and Turcoman labia/ to reach a peak ot technical excellence and ot emotional expressiveness

which for many is the finest hour oi Persian painting. The master- piece of the age is the Shahnama-yi Shahi (“The Kind’s Hook ot

Kings*, formerly known as the Houghton Shahnama) which, with its

258 paintings, was the most lavishly illustrated Shahnama recorded in

all of Persian history and which monopolized the resources ot the


I So

1 s 3 a ) Opener of Doors!’ is .111

inscription commonly associated with gateways and doors m the 1 s th and [6th centuries, rhe reference is to God; and

the notion ot\i saintly person .is .1 gate was

commonplace – ‘Ali. tor example, was

called the Gate to the City ofKnowledge.

Such associations enrich these and other

silver-faced and gilded doors at the

Ardahil shrine. 161 1-12. Each door leaf

has .1 carpet-inspired ogival lattice design

of 1 14 blossoms, one tor each sura of the





m/ K-r/



±£* iSj Pastoral idyll. Everyday rustic occupations rendered bv Muhammadi, 1578, in .1 minimalist aesthetic employing the unusual technique of tinted drawing and the play oflight and dark accents

ix yields to love. Royal Turcoman Shahtuma, Tabriz. 600-year life of the Shahtamas greatest hero, Rustam.

v s

I he single romantic episode in the

royal atelier tor a generation. Its pictures lay special stress on the war

between Iran and Turan (Central Asia) and may thus reflect contem- porary political concerns. It has justly been termed ‘a portable art

gallery’ because all the most illustrious painters of the time contrib-

uted to it. Its lissom, eternally youthful figures are apparently an orig-

inal creation. Perfection proved hard to sustain, and before long artists

were overreaching themselves, lor in some works of this school the Khamsa or ‘Quintet’ ofNizami in the British I ibrary, dated 1 >;<>

and 1543) st> much detail is crammed into the composition that its fastidious precision tails to make its full effect. Similarly, the colour

range may be so kaleidoscopic that the very richness confuses th

A dream world –

natural dantv with an other

erousry lured to h



It could be argued that the manuscript page had too limited a scope to accommodate the increasing complexity of these compositions.

In his middle years. Shah Tahmasp became a religious extremist, which resulted – among many other, more significant, changes – in his losing interest in painting and disbanding the royal atelier. The court style associated with Qazvin, which became the capital m 1548, is marked, despite certain exceptions (like the Haft Aurang, the

‘Seven Thrones’, of Jami in Washington), by a palpable decline in

quality. Compared with the best Tabriz work, landscape becomes simpler, with large areas given up to a single colour (as in the

Shahnama of Isma c

il II, 1576). Figures tend to increase in size and

they exhibit a curious stiffness. Yet m courtly tableaux, youths and maidens are rendered with a consistently suave line. No trace remains of the vigorously differentiated types of the school ofBihzad, and the

earlier obsession with detail gradually disappears. In the later six-

teenth century, the enforced change of patronage, which meant that

the day of the luxury book was effectively over, led the best artists

(such as Muhammadi, who specialized in figure studies and tinted drawings of carefully understated scenes c>\ peasant life) to produce

single Leaves that were eagerly collected by connoisseurs and bound into albums. Figure studies of pages, prisoners and princes, among others – were a popular subject tor such leaves, which became the forte o\’ the Qazvin school. Their subjects came to include genre scenes of the utmost delicacy, with pastel shades enhancing the

composition. The sequence of drawings, paintings, ornament and specimens of calligraphy is carefully calculated so that facing pages

contrast with, complement or mutually enrich each other, A\)d it is possible to recognize the development of themes. It is therefore mis-

leading to regard such an album (muraqqa*) as a scrapbook. There is

nothing random about it. This sea-change brought artists out of the court and into the public market, a process which accelerated the

break with anonymity and the rise o\ the artist – for

example Sadiqi Beg or Siyavush – as a personality. It is hard to account for the radical change in taste and style in

seventeenth-century painting. Technically it is easy enough to point

to the fashion both tor single-leaf paintings and for tinted drawings

in the later sixteenth century; but while presaging the divorce of

painting from book illustration, neither of these developments fully implied the shape of things to come. Some time around the end of the sixteenth century – and the timing accords too well with the change of capital to Isfahan in 1597 to be coincidental – a massive




and apparently unofficial deregulation of the traditional codes of practice governing book painting took place. It is not clear whether this was market-led – precipitated perhaps b\ a turn for the worse in official patronage for illustrated books or whether the pressure for change came from the painters themselves, or .it least from a tew strong-willed radicals among them. Hut the results are perfectly plain to see. They govern execution, subject matter, output, patronage and expense. As with all such revolutions m taste, the pace ot change was uneven, with some able painters stubbornly hanging on to traditional ways. Hut the logic of the new style was inexorable. It brooked no rivals. Hook painting as it had been under stood tor the previous thousand years was now finished, tor the most part relegated to the bazaar.

For centuries, paintings had served to illustrate and explain the

great classics of Persian poetry. Now. that increasingly elaborate, refined symbiosis had been shattered, and Persian painting lost its t.ip

root. And once the tradition had been broken it could not simply be reinstated. Hence some ot the best Iranian paintings ot” the later Safavid period and thereafter that are still produced in the context of

the luxury book remain stubbornly divorced from a continuous

text; they serve as frontispieces or are inserted into earlier volumes

like the Hntish Library Nizami o\ [543. Some painters turned to [87 other media, experimenting with book-covers or (under European

influence) with full-length oil -paintings. Lacquerware, too. devel-

oped significantly in the Safavid period, being used principally for

secular book covers, and drawing on the lyric and epic themes

created by book painters. Hut lacquer doors and boxes are also

known and these have a wider repertoire, including tor example audience scenes, (dearly, later Safavid painting will not tit neatK

into pigeon-holes.

The principal change in pamtmg on single leaves concerns subject matter. That sense of an intellectual construction, a hard-won and

hard-edged abstraction of reality, that stamps earlier Persian painting

and prevents it from disintegrating into a self-indulgent dream world,

has gone. The subject-matter on which it had traditionally been

exercised is no more. Perhaps .1 new style had ot necessity to be fash

ioned to match the new subject matter. Hitherto the Persian painter

had come to terms with the world around him by miniaturizing it.

by viewing it .is it were from the wrong end o( a telescope, and then

Decomposing selected elements of what he saw 111 his minds

Now he holds a distorting mirror 10 reality Sometimes he ma\



187 The past as myth. Nizami, Khamsa, Tabriz. 1539-43. In this icon of imperial majesty drawing on pre-Islamic Persian legend. Khusrau and his consort Shirin spend the evening

listening to stories told by her maids. The scene evokes Scheherezade and the 1,001 nights.



[88 A taste nf Ecstasy- Using techniques recalling marbled paper, this hallucinatory image ol a mystical journey evokes the swirling rhythms of a cosmic dance. It celebrates the unity ofbeing (wahdai al-wujud), a concept elaborated by Ibn al-‘Arabi and variously described .is monism or pantheism. Thus created tonus dissolve into each other. The composition suggests infinity, for the frame cannot confine it.

capture the exaltation and ecstasy of a mystical vision. More often, tar from keeping his own distance, and forcing us to keep ours, he thrusts his discoveries right under our noses. More than a hint of the fairground freak can be sensed in some of the images served up by seventeenth-century painters. Studies of single figures dominate, and

they are decidedly unheroic. Seedy dervishes, Sufi shaikhs, bagpipers,

grooms, beggars, ageing painters, ne’er-do-wells, merchants – a whole gamut of portraits of the middle and lower reaches oi society

(for the grandees of the court are not depicted) meets the eye. Their

likenesses are seized with formidable accuracy and speed, a notable

achievement for an art form that, minor exceptions apart, had tor

centuries disdained portraiture. Yet despite the searching realism of

some of these portraits, perhaps a juster term would often be carica-

tures – for, while serious portrait studies oi profound psychological insight. o\ intimacy and tenderness, are not lacking, the main body

of this material has an altogether minister tone, and is frequently

downright coarse. Satire and knockabout humour are the driving forces behind most of the images. Hence the unsparing focus on

drunken, pot-bellied holy men, or on dervishes built like storks,

smoking pipes of opium and wearing fur h.its tilted at rakish angles. 1 fence too the unHattenngK exaggerated length of nose or chin.




No-one could gainsay the bite and individuality of these mainly low-life character studies. Yet they are only half the story. Some of the self-same artists lent their talents to an altogether different genre

of painting, at once vulgar and louche: the high-gloss pin-up. Pert

189 androgynous pages with inviting smiles and thighs of awesome girth subside gracefully onto clumps of grass, proffering cups of wine. Or they pose smirking for their portraits, dressed in the height of

European fashion, from the saucy feather in their hats to the jewelled

buckles on their kid boots, and holding some newfangled import from the West, such as a turnip. Their poses are often of a coy and

studied awkwardness. Their faces – full moons with rosebud mouths, as contemporary taste would have seen them – exhibit perhaps the final degradation of the Buddha image in Persian painting. All con- crete sense of the individual human form is dissipated in these spine- less, indeed altogether boneless, forms, mere confections of elegant

line. More rarely the theme is definitively a girl, or a pair of lovers, sometimes depicted in explicitly erotic poses, with wraps of filmy

lawn incompetently veiling their nakedness. Thus cloying sweetness alternates with startling grossness. This again marks a new departure in Persian painting.

The execution ot these leaves is as novel as their themes, and is intimately related to the scale o\~ their output. The same coldly efficient, mannered, economically linear style is employed through-

out. Once again, and here in the most unexpected context, the Islamic capacity to geometricize forms finds expression. The very large quantity of these pin-ups to survive has done much to skew modern understanding of Safavid art. and it should caution against evaluating them very highly. It proves, of course, that they were

extremely popular – but no more. Indeed, close inspection suggests

that they were produced almost mechanically. Colouring is flat and

simplified, laid on in broad washes. Expressions are of a uniform sim-

pering blandness. The background setting for the figures is as exigu- ous as it well could be. a clear indication that it was regarded as being

of no importance. And the virtuoso mastery of line cannot disguise that the lines themselves are remarkably few. Here, then, was a way to make a lot of money with minimum effort. The attraction is on the surface: what you see is what you buy. The conservative nature of public taste in this genre explains why it is difficult to tell the work of the various masters apart; but that same similarity reveals also howT much each of them subordinated his own individuality to the dictates of fashion. Many such leaves were inscribed with the




f ^_: – j •





. wir*

I he androgynous centrefold. Note, for example, the disproportionately tiny feel

‘.7i slv humour. Inscribed ‘He [is God]. Portrait ofNashmi the Archer Completed on ruesday, j Rabi’ II 103 1 [10 January [6j | Work of the humble Rida’ yi ‘Abbasi’ I Ins dishevelled characti his cadaverous race and one stocking .it half-mast, h.ts dearly fallen on evil times.

name of the artist, the place and date of execution and the identity of the sitter; and the works of certain masters Sadiqi Beg, Rida’ yi

‘Abbasi, Minn Musawwir, Muhammad Qasim, Habiballah were eagerly sought after. Contemporary gossip pounced on their person

alities and eccentricities. Sometimes the) worked at home. What of the patrons who bought these leaves, and the prices they

paid tor them- An anecdote about Sadiqi Beg reveals him using two of his drawings in part payment for .1 poem written and rented in Ins honour, and asserting that each work could be sold to merchants foi



3 tumans (worth £7 in English seventeenth-century money — a very considerable sum). It appears from this story that there was a market for his work as far afield as India – and indeed many Persian special- ists in the various arts of the book found a good living both at the Ottoman and the Mughal courts.

Carpet fragments of thirteenth-century date from Saljuq Anatolia, and even earlier survivals from Islamic Egypt, show that this art was well developed in some parts of the medieval Muslim world, and literary sources amply corroborate this, mentioning dozens of carpet

types and production centres of which no tangible evidence survives. Persian paintings of the fourteenth and fifteenth centuries, with their

close focus on detail, flesh out this picture and reveal the remarkable range of contemporary carpet types. But the Safavid period is the

earliest from which a critical mass of physical evidence has survived –

enough to allow the history of this art form in Iran from 1 500 to 1700 to be written. The reason, perhaps, is that the Safavids turned a cottage industry into a national one. Shah


Abbas I founded carpet

factories at Isfahan and Kashan. The weavers, who sometimes sign their work (e.g. Maqsud Kashani or Ghiyath al-Din Jami), display consummate technical mastery across a wide spectrum of materials –

silk, wool, gold and silver thread – and techniques, from flat weaves

(zilus and kilims) to pile carpets knotted coarsely or so closely as to

total eight hundred knots per square inch. Red, white, yellow and

blue are by fir the most popular colours, irrespective of the design.

Gigantic carpets were produced; 111 [539—40, the Ardabil shrine was

graced, at the behest of Shah Tahmasp, with a pair of carpets, the

191 larger and better preserved one comprising some thirty-three million knots and measuring c. 1 1 x 5.40 m (36 x 18 ft).

As in the Timurid period, no significant barriers operated

between most media in the matter of design, and thus many of the motifs and themes encountered in carpets can be paralleled in con-

temporary tilework, wall painting, lacquer, metalwork, manuscript

illumination and illustration, and so on. So close are the analogies

with contemporary Safavid book painting, in particular, that some of the most complex carpets can be regarded as paintings executed in a different medium. This has its disadvantages, as can be seen with modern rugs bearing portraits of statesmen – for the language and range of expression natural to a carpet is being ignored, indeed sup-

pressed. What, then, made the design of carpets distinctive? Perhaps

191 A woven Paradise. The larger Ardabil carpet (1539-40), of superlative technical quality, is saturated with the imagery of heavenly light. At its dead centre is a pond with floating lotus blossoms – perhaps the Qur’anic Pool of Kauthar.The carpet bears verse- by the poet Hafiz and a signature: ‘Except for thy haven

there is no refuge for me in this world;/Other than here, there is no place for my head./Work of a servant of the court, Maqsud of Kashan. 946’.






H ,/:•<



more than in any other art form, carpet design manipulates multiple levels of pattern, .sometimes five at a time, using colour as the princi-

pal means of distinguishing the different schemes. The designs are often predicated on constantly shifting viewpoints, which add further complexities and lend the ensemble extra dynamism. As with many panels of tilework, moreover, the design is deliberately not complete but is only a portion of an unimaginably large but thor-

oughly disciplined composition. Hence the viewer receives intima- tions ofinfinity, even eternity all the more affecting because they are not explicit.

Various major categories of design may be distinguished, though they continually overlap – for example, borders comprising car- touches with animal scenes enrich otherwise abstract designs. Garden

carpets perpetuate a type known as far back as the Sasanian period, when the rug known as ‘The Springtime ot Khusrau’ was a national treasure. This was divided into tour plots, each representing one of

the seasons and depicting appropriate flora, all executed in gold and

silver thread and precious jewels. Safavid garden carpets seem to

reflect simultaneously a side view and a bird’s-eye view, though dras-

tically schematized, ot water-channels stocked with fish, flower-beds,

terracing, pavilions, ponds and fountains. The poems inscribed on them compare them to roses and tulips. Sometimes peacocks and

lesser birds, lions, leopards, hares and deer can be glimpsed in the

foliage: their colours as in Safavid tilework) may bear no relation to nature. The other-worldly associations ot the garden theme are always subliminally present. A second category consists of hunting scenes conceived as a sequence ofloosely linked vignettes, with the

border sometimes depicting an alfresco royal reception, lutanists and

even angels (major examples survive in museums in Boston and Milan, the latter dated [522—23). The so-called animal rugs, in which creatures of various species frolic or attack each other, can be classified alongside the hunting rugs. A third type can be described rather inadequately as a medallion carpet because its centrepiece is

usually a huge circular or oval medallion with numerous smaller

medallions orbiting around it. The Ardabil carpets are the most dis-

tinguished examples of this variety, and display the extra refinement

of a mosque lamp hanging from the inner circle of smaller satellite medallions – all silhouetted against a ground ot deep indigo which makes the carpet like a window on to the Milky Way. Placed on the floor directly beneath the dome, it evoked that dome and thus the heavens. Further themes m these two carpets contain references to




Paradise as described in Islamic tradition; other carpets depict angels

or houris. Quite another type is the vase carpet, in which vases o( various sizes form the leitmotif of the composition. A last major category is represented by the floral rugs, in which multiple flowering sprays of various sizes, linked by thin tendrils, spill across the field.

These carpets, though essentially conn mi. were actually made in numerous manufactories throughout the Safavid realm, though opin ions differ over allocating types of rugs to specific places, labia/. Qazvin, Kashan, Isfahan. Herat and Kirman were all major centres. though it was common for a single centre to produce several different types of rug. The finer carpets represented a considerable investment in time and money, and it is therefore not surprising that they bear quite full inscriptions; the London Ardabil carpet is signed by one Maqsud Kashani, who styles himself Servant of the court”, and bears the date [539—40. A carpet 111 the Najaf shrine was donated, s.ivs the inscription, “by the dog of this shrine, ‘Abbas’ (Shah ‘Abbas I). I he strongly pictorial character of so main Safavid carpets plainly owes much to Safavid book pamtmg. a borrowing which extends even to the concept ot pictorial space, as m the adoption of the high horizon and stepped planes and the plethora of small-scale detail; but it has .1

much more marked anti-naturalistic quality (for example in the choice ot colour), which suggests that even figural images were seen

in some sense as abstract motifs, and the roster of Chinese motifs far exceeds the norms ofcontemporary painting. Cloud collars, undulat-

ing cloud bands, cranes, phoenixes, dragons, lotuses, peonies and

numerous fabulous creatures (as in the Sanguszko carpet in New York) – all are grist to the mill, though their context owes very little to

C )hina. Nor is it certain whether they retain their C Ihinese significance or whether they have acquired new symbolic meanings (the

dragon phoenix combat, for example, emblematic ot happiness in

China, may now refer to dualist beliefs). This obsession is still imper fectly explained in current scholarship; not even Safavid pottery is so

saturated in Chinese motifs. On the other hand, many ot the finest carpets were made as presents tor foreign potentates or tor commercial export to the West. The horizons ofthe Safavid carpet industry were

therefore very wide. Hut while it did not ignore the immemorial past

of carpet production in the Ne.n East the abstract, usually geometri<

designs of tribal weavers – its supreme masterpieces bear .1 very differ

ent kind oficonography. As .1 result, they had no root in the medium itself; for all their splendour and majestic scale, they were an intrusion.



Architecture, painting and carpets may fairly claim to represent the principal achievements of the Safavids in the visual arts, but this was

a productive period in many other media too. Textiles were pro- duced in various materials – printed cotton, silk, shorn velvet, reversible brocades in gold and silver thread, and embroidery – not only in Isfahan but also in Yazd and Kashan. Similarly, textile fac-

tories were established by royal command all over the realm, from Shirvan to Isfahan, Yazd, Kashan, Mashhad and Rinnan, each with orders to ‘weave in its own manner’. Here too much of the produc- tion was for export, and notable painters like Rida’-yi


Abbasi were

co-opted to provide designs. Velvets, brocades and block-printed

cottons were made in huge quantities. ‘They last forever’, as Chardin said. Hence they survive in great quantity, and they too were exported — an entire room in Rosenborg Castle in Denmark was hung with them. Their designs bear once again the unmistakable imprint of book painting, with themes like dallying lovers, winsome pages, picnics, horsemen leading prisoners of Turkish stock, hunts-

men on foot or mounted, animal combats, Moral patterns and the familiar episodes of Nizami and Firdausi. The absence o\^ serious

themes of religious or political iconography is noticeable. Reds and yellows are especially favoured colours. While some of these textiles were intended as hangings and tent decoration, most were garments

naturally intended for the wealthy, and advertised the luxury o\

the Iranian court. They attracted admiring notice when worn by Iranian ambassadors abroad, and were often sent as gifts to European


The last decades of the sixteenth century saw a vigorous revival of the pottery industry in Iran, which despite occasional fine – even dated – pieces had been m the doldrums. Safavid potters developed new types of Chinese-inspired blue-and-white wares, due perhaps to

the influence of the three hundred Chinese potters and their families

settled m Iran by Shah ‘Abbas I. and indeed some of the vases made in Kirman and depicting swirling dragons are .1 passable pastiche of Chinese work, a comment that applies equally to their semi- porcelain body. Much of this ‘chinoiserie’ pottery was produced in response to European demand. More typically Iranian designs, however, also appear on such pieces. Subtler references to Chinese

originals include the so-called Gombroon ware, which depends for its effect on distinction of form and on its white translucent body, to

which celadon slip may be added. Lustre enjoyed a revival under Shah


Abbas and was produced in great quantity, but it has a rather




11m V m





‘v’ •-. ftfll

. He 16th « eniury 1 1″‘ I K1U h

sk£j^s^£=s= “” wi



brassy sheen which, combined with an emphasis on underglaze blue, results in pieces, inferior to earlier lustre in aesthetic quality. The decoration is restricted to vegetal motifs. Polychrome ware was pro- duced in great quantities in Kirman. Figural designs on such Safavid pottery as eschews chinoisene echo the mannered style of the painter Rida’-yi


Abbasi, whose idiom in fact dominates all later Safavid figural art; they also recall, more generally. Safavid carpets and textiles. Later Kubachi ware has a much wider and brighter

193 colour range and often has a central medallion enclosing an engaging

portrait bust executed with rapid strokes in typical late Safavid style.

The Safavid pictorial manner lingered long after the tall of the dynasty and. in an enfeebled state, remained the staple of Qaiar

potters. This staleness, combined with the widespread popularity of cheap European ceramics from the eighteenth century onwards,

brought about the final demise ot fine wares in Iran, though

Mashhad and Kirman. the major centres ot Safavid production (neither city being noted for its pottery m earlier times) continued to

194 produce blue-and-w hue ware with black outlines into the nine-

teenth century. Only in architectural tilework did the Safavid tradi-

tion continue with undiminished vigour after the fall of the dynasty

with new colour schemes favouring yellow and pink, an unprece-

dented emphasis on relief and themes of European origin.




Ait thexB-«M



In metalwork, the engraved technique developed in Khurasan in the fifteenth century retained its popularity well into Safavid times,

and indeed that province, now supplemented by Azerbaijan, contin- ued to be a major centre for this medium. It is curious that Safavid metalwork has been so long neglected even though it produced significant innovations in form, design and technique. They include a type of tall octagonal torch-holder on a circular plinth, a new type of ewer of Chinese inspiration, and the almost total disappearance of Arabic inscriptions in favour of those containing Persian poetry,

often by Hafiz and Sa c

di. The content of these verses is frequently religious and is apt to have a strong Sufi tinge. Dense arabesques and floral designs were more to contemporary taste than figural motifs, even though such motifs dominated the other visual arts of the

period. Perhaps they would have coexisted somewhat uneasily with the huge fields of epigraphy which are the major visual accent on

these pieces, and which find their ideal foil in the low-key vegetal

background which expands effortlessly to till the remaining available

space. Inscriptions .ire now allotted a greater surface than ever before, in bold zigzags and cartouches as well as the more familiar encircling bands. A few pieces commissioned by Armenian patrons juxtapose lines from Persian mystical poets with Armenian inscriptions.

Armenian architecture and pottery produced m Iran proper uses Iranian modes m just the same unselfconscious manner even though their iconographic message is unmistakably Christian: a silent testi-

mony to the religious tolerance practised by the Safavid state. Safavid brasses were apparently often tinned to simulate silver,

195 though the most luxurious metalwork, of which only a few pieces

are known, was inlaid with gold and incrusted with jewels. Other

lost types ofSafavid metalwork can be reconstructed with the help of

ceramic copies. Hut some o\~ the best Islamic armour extant –

stirrups, shields, battle-axes fashioned in iron and especially steel,

was produced h\ Safavid smiths. Important pieces include a group of

scimitars {shamshirt) from Khurasan, as well as steel banners inlaid

with Twelver Shi’ite invocations. ‘Damascening 1

(watering and gold

overlay or gold inlay and openwork were the most common decora- tive techniques used for such objects, which often bear lengthy

inscriptions. Safavid metalwork, like so main of the other visual arts, remained the standard for subsequent artists, and Zand and Qajar work perpetuates its shapes and decorative conventions, though the execution tends to lose itself in a meaningless intricacy.




( II API I \t I I \

The Ottomans

The Ottoman empire began modestly, .is a principality at the western vanguard of the Turkish campaign to [slamize Anatolia and bring down the Byzantine empire. A melee of small and often mutually antagonistic emirates (beyliks) had arisen to till the political vacuum in Anatolia after the final fall of the Saljuqs of Rum in 1308 and the decline of Mongol authority soon afterwards. The Ottomans gradu- ally consolidated their position in western Anatolia and expanded

their territories at the expense both of Byzantium and their Muslim neighbours, changing their capital in the process from [znik right

on the Byzantine frontier- to Bursa, further south. A scries ofambi tious and able rulers in the later fourteenth and early fifteenth cen-

turies enabled them to survive a catastrophic defeat by I miiir in 1402 and to expand into Europe, first via Greece and then the Balkans.

The noose around Constantinople tightened inexorably and in retro spect it is remarkable how long the city was able to survive. It fell to the charismatic young sultan Mehmed II, after .1 prolonged and gallant defence, in 1453. The name Qustantmiva continued on the coinage, but Istanbul now claimed much of the thousand-year her- itage of the Byzantine empire – a message driven home by the sil houettes of the mosques which quickly dominated the city’s skyline.

The psychological boost provided by this victory can be said to haw launched the Ottomans on the road to a world empire; in the course of the next century they had established themselves not just as

the principal Islamic- state but .is a superpower – the only medieval

polity both to achieve this distinction and carry it into the modern

age. Neither their Nafavid nor their Mughal contemporaries could

match them tor power or territory or perhaps even wealth, for two

centuries and more after the fall of ( Constantinople this was an empire on the move, transforming much ofthe Mediterranean into a [iirkish lake, its persistent encroachment into Europe fmalK turned back as

late as [683 .it the Siege of Vienna. So powerful was its grip on (In-

lands it had a< quired that even the long dec line- ot the eighteenth and



early nineteenth centuries saw only minor losses of territory. Even later, when Turkey began to be described as ‘the sick man of Europe’, Ottoman possessions in Asia remained firmly within the empire. What was the secret of this success? Several answers can be pro-

posed. First, this was a thoroughly militarized state; yet, far more than its Islamic predecessors, it was open to technological advance, a true gunpowder empire. Second, it presented itself as the consum- mate Islamic state, fully integrating government and Islamic law (the

shari’a). The Ottomans took over from the Mamluks the mantle of the leadership ofSunni [slam and the Ottoman sultan subsumed into his own titles those of the caliph, thus ending the shadow ‘Abbasid caliphate that had survived m Cairo until [517. Yet the Islam propa- gated by the Ottoman state was not only that of the ‘ulama, the men of theological learning, which was apt to become and and legalistic; for the sultans also sponsored certain Sufi orders and were on intim-

ate terms with their leaders. Thus they had a tap-root to popular

piety, and the flourishing of the dervish lodge (tchhc) in their domin- ions is clear proof of this. Thirdly, this was not only a militarized but

also a quintessential^ bureaucratic state, with well-established hierar-

chies in all the affairs of government. The Ottoman millet system formalized and protected the position of non-Muslim minorities in the body politic; the devshirme policy systematically uprooted young

Christian boys from their native lands and brought them to the

capital for service in the palace or in crack regiments. The Ottoman archives, which survive in abundance, document the workings of this bureaucracy, tor example the way that the Iznik potteries were run

and financed, or how building campaigns were managed. Fourthly, the Ottomans were Staggeringly wealthy. Their empire stretched

from Hungary to the Yemen, from Algeria to Iraq. With such taxa-

tion revenues to draw on. it is Little wonder that the Ottomans did not – though to their ultimate cost – evince serious interest in the

Americas. India, the Far East or the burgeoning of

seaborne trade. Finally, they wielded the weapon ofpropaganda most

effectively. They were thoroughly international, as witnessed, say, by the contents of the Topkapi Saray (see p. 280) or by their welcome of Western experts in art. the sciences and military techniques. Yet they

also represented the culmination of five centuries of Turkish rule in

11 the Islamic world, and that heritage manifested itself culturally,

socially and politically in numerous ways. Brit they also stood for the

whole ot the Islamic world, for example as guardians of the Holy

Places of Mecca. Medina And Jerusalem. They were the very image




of the infidel so far as Europeans were concerned, and their pomp and ceremony was recorded with a sense of awe In European an sadors and travellers.

Ottoman art is in a category of its own within the wider world of Islamic art. It certainly has us own distinctive character in the major media such as architecture, ceramics, book painting and textiles, and its products hear witness to the massive financial resources of the most powerful empire of its time. Net the remarkable uniformity o( much of the Ottoman visual arts gives one pause. On the technical side, much that was produced attained the highest standard of excel lence. But this superlative execution is liable to be offset at times b\ a

cold, rational formalism that drams the lite out of the work. I he perfection itself has a somewhat forbidding, hard-edged quality. t’ascs in point are the austerity of so much architectural decoration or the combination of high-gloss surface and a rather restricted dec

orative repertoire in Izmk wares. It seems permissible to suggest, on

the basis both of the relatively static nature ofmuch Ottoman art and of its high technical quality, that government control had a consistent and decisive impact on the art of the period. On the credit side, this kept very large numbers of artists occupied and ensured that their

work met the most exacting standards. Hut there was a price to \\\\. most evidently in architecture, where the existence ot centrally pro

duced blueprints is well documented, but plainly in other fields too – in a word, standardization. Top-flight artists paradoxically had less

freedom of manoeuvre than less able ones; they were palace emplo\

ees rather than subject either to their own preferences or to t he- demands of the open market. Ottoman architecture is unique in the Islamic world tor its

unswerving fidelity to a single central idea – that ot the domed square unit. This basic theme announces itself on a tiny scale at the

wry beginning of the Ottoman period, in the mosque ot Ertughrul at Soghut, datable to the early fourteenth century Over the next five

centuries it was developed with a singular intensity ot purpose. It is.

so to speak, the spinal cord running through the body ot Ottoman

architecture: it controls every major development ot that style .\n<\ its

influence can be felt even m peripheral areas. I he intrinsic simplicity of the domed square unit as a structural form made it ideally suited

to function on any scale, large or small, without sacrificing clarity 01

monumentality. It readily accommodated, too, numerous appurte-

nances – domed buttresses, semi-domes, porticoes, domed cloi

courtyards and minarets. Possession of this solid structural



196 The seed ofgreatness. Haci Ozbek mosque. Iznik, 1333. This is the earliest Ottoman mosque dated by inscription. Ottoman Islamic power began in the tar west of Anatolia and this mosque is an early symbol ofthat

sovereignty, appropriating former

Byzantine territory. The cloisonne masonry (stone trained

by brick) is a technique nt~

Byzantine origin.

enabled Ottoman architecture to retain its quintessential character intact, independent of superficial local detailing m materials or decoration, throughout the many provinces of the empire. From Algeria to Iraq, from Syria to the Yemen, a distinctively Turkish

Ottoman architecture effortlessly imposes its presence. This feat is all the more impressive when one recalls that the

domed square unit had already had a long history in Islamic architec- ture even before the coming of the Ottomans. But in, for example, the Iranian tradition between, say, Soo and 1700, for all that the

domed square recurs repeatedly for many centuries and in most major building types, it never acquired the exclusive status which it enjoyed in Ottoman times. The same could be said of Syria, Egypt, the Maghrib or India. Instead, the domed square is merely one of several equal and mutually dependent architectural forms: the portal,

the iwan, the hypostyle hall, the two-tier facade and so on. Ottoman architecture, however, somehow suppresses these other forms, though without discarding them completely, and in so doing elevates the domed square to premier rank. It fashions a new balance of com- ponent parts, arranged according to a much more explicit hierarchy




than before. And the domed square tends to invade areas from which it was formerly excluded, and swamp them. Thus high courtyard facades give way to shallow arcades covered by a succession o( domes, becoming effectively a sequence of domed square units; the same is true ofhypostyle halls.

This consistency does haw its drawbacks. It can lead to a certain inflexibility, a tendency to apply a rote solution, devised at the drawing board, irrespective of the peculiarities of the site. I he degree of centralization necessitated by so vast an empire encouraged

a Strong bureaucratic input into architectural design, .1 process

formalized by the creation o\ a special department ofst.n

with preparing blueprints for repeated use. I his swem effectively tied the hands of provincial architects and imposed a perhaps unde

sirable uniformity on their work. They could choose it was only practical that they should do so – whether to use naked brick or to

coat it with plaster, whether to have their masonry plainly dressed or

striped. They had a comparable freedom with applied decoration. But in the crucial matter of managing space – which of course lies .it the heart of all architecture – they clearly had to work within para- meters imposed from outside. Thus the vigorous local traditions which had characterized the Islamic lands bordering the

Mediterranean, and beyond, tor almost a millennium, suffered a

serious decline. Pre-Ottoman traditions in most of these areas had

positively benefited from the absence o\~ central governmental

control, in that individual provinces, and often even individual

towns, had managed to develop their own distinctive styles of archi- tecture. No doubt this phenomenon had much to do with political independence. But whatever caused it. the fact of this diversity is

undeniable; strong local roots nurtured it; and its vitality had ensured

continuous change and evolution across a broad front encompassing

form, structure and ornament. Regrettably, the Ottoman conquest put a damper on this process. More than that, it brought Ottoman architectural forms to places where they simply did not beloi

building like the Fishermen’s Mosque in Algiers stands out like a

sore thumb, a metropolitan Turkish import into a solidly Maghribi

landscape. The political implications .ire just as inescapable as those

attached to the architecture of the British Kaj in India. I his was cul

tural imperialism at work.

Ottoman architecture could fairly In- termed mosque driven. Hie

unchallenged prestige- of the mosque made it the natural foCUS o(

royal patronage. I h us it became the- most public showcase foi





innovation, ensuring that new ideas were disseminated quickly and that they had the. imprimatur of the most venerated building type of all. Moreover, virtually all the significant stages in the evolution of mosque design were accomplished in the capital — first Iznik, then Bursa, finally Istanbul – which naturally conferred on them a metro- politan glamour. And the extremely large size of so many of the imperial Ottoman mosques, with their courtyards, made them perhaps unexpectedly useful models for quite different building types

such as caravansarais. madrasas and elements of kulliyes, foundations

centring on a mosque but comprising multiple buildings. Indeed, the core forms of Ottoman architecture, domes and courtyards, are basi- cally interchangeable; a domed square with a vaulted two-bay porch can as easily be a mausoleum as a mosque. But it is the dome that dominates. Hence the various illusionistic devices adopted to magnify the size ot the dome – such as the low roorline of the adjoining cloister, porch or sanctuary, or the siting o\” the climactic

dome at the highest point ot a sloping site.

197 Calligraphy writ large. Interior of Ulu Cami, Bursa, [396 9. Huge calligraphic panels used as wall decoration are an Ottoman speciality, as is mirror writing. Here the central panel reads ‘The Guide’ (<//-

hadi), one of the 99 names of God. Note the multi-domed interior.



The concentration of major mosques in Istanbul can be explained in political terms as part of a sustained attempt to transform the visual aspect of the ancient Christian city of Constantinople au<\ to give it a new Islamic identity as Istanbul. In the rough!) contempo rary Safavid and Mughal empires, too. the capitals Isfahan, Delhi – were given massive face-lifts by means oi ambitious building programmes, and the resultant architecture had an unmistakably scenic purpose. Hut the well-nigh obsessional focus on huge mosques which characterizes Istanbul tor the century or so after the conquest of the city – it is even recorded that a detailed model ot the Suleymaniye was on a show in a public procession held in

1 582 is

noticeably absent in these lands. Of course major mosques were erected, but palaces, bazaars, mausolea and piazzas account tor much of the new building in India and Iran. And the demographic pies suies caused by an influx of Muslims into a previously Christian cit\ can only partially explain this frenzied building campaign.

Competition between successive sultans must also be taken into account. And given that the first great mosque erected by Mehmed the Conqueror in Istanbul was itself a very substantial building, the sultans who succeeded him had little option but to follow him down the same path. For that simple hum.111 reason the imperial mosques

o\~ Istanbul are unreasonably large and plentiful. Presumably the

expansion ot the city beyond its Byzantine walls meant that there was

less premium on space and that architects were able to spread them- selves. Moreover, since there was plainly no overriding liturgical

need for so main large mosques, their architects designed them to

serve additional functions. Hence the kiilliye or composite found.


tion saw an unprecedented expansion in this period.

Several other factors also encouraged the rapid evolution of a

dynamic innovatory style in sixteenth-century Istanbul, for example,

architects paid close attention to the work of their rivals, identifying

the weak spots in their designs and improving on them in their own work. The increasingly compact and concentrated silhouette ot

mature Ottoman mosques owes much to the intelligent observation of failure. That visually satisfying sense of interdependence between

component parts, that carefully staggered sequence of supporting

elements, which lend these mosques their .111 of confident, tour

square stability, is a prune example of this pragmatic approach

n\’ course each new variation on the familiar theme could be

convenicntlv examined at Jose quarters, because virtuall) .ill the

major mosques were in Istanbul, \nother contributor) factoi was



IS* “..,




the creation of a government department, a kind o( Ministry of Works, to oversee the continuous building campaigns bunched by the sultans. This created a pool of administrative expertise which greatly facilitated the rapid completion of these complex projects. One should remember, too, that the Ottomans constituted the largest and most dreaded power m Europe and western Asia, control ling as they did an area of almost a million square miles. It was only fitting that their architecture should reflect tins fact, Foreign embassies Hocked to Istanbul and reported back admiringly on the scale and magnificence of the new buildings there. Thus the propa ganda dimension ot these mosques was an integral part of their con temporary context; and the colossal financial resources of the

Ottoman empire allowed the sultans to exercise patronage on a scale perhaps unmatched since the high days of the ‘Abbasid caliphate. The increasing size of these mosques brought daunting technical

problems m its train. Hence the introduction o\~ ever more refined Systems of buttressing, experiments with semi clonics, more lavish fenestration with a consequent reinforcement of the adjoining masonry, and so on. Since the self-same process lud taken place m early Christian and Byzantine architecture it is not surprising that

Ottoman architects happened independently on many of the same discoveries – though for obvious religious reasons their buildings did not incorporate the cruciform element so important to Christian

architects. Yet the catalyst was undoubtedly the conquest ot

Constantinople, which brought Ottoman architects fate to face with the greatest of Byzantine monuments, Haglna Sophia, and the

various churches related to or developed from it. This encounter,

moreover, took place at a cusp ot history when the Ottoman state

enjoyed boundless self-confidence and was expanding its boundaries,

when revenues were ample and when the new masters ot Bwantium were laying claim to its ancient heritage. The tune was. in short,

perfect for propaganda gestures, and architecture was the obvious

medium. It was entirely natural for Turkish architects to study

Haghia Sophia intensively, learn from it and resolve to outdo it. Thus

all major Ottoman mosques in Istanbul were built in tull awareness ot

the challenge posed by Haghia Sophia, and they were in some sense

m its shadow. Yet for .ill the indisputable similarities between I laghia Sophia and

the great Ottoman mosques from the- Lmh (ami i \ onwards

ie apogee of Ottoman powei Interioi <>t Sul ‘• the dome, \n!

(26 <> n half its height, • ‘ Allah,

Muhammad and tin- tour Righth, < i r.iptu-r ot the time I he mosqi

s ;nm Warn

tor\ ot die sultanate and the 1 alipl



tiered elevation, grouped windows, cascading volumes, prominent semi-domes and ‘domed turret-like buttresses – the differences are manifest; and it must also be recalled that such features also occur in earlier Ottoman buildings, though in a less developed form. Where Haghia Sophia exploited mystery and ambiguity, the Ottoman style put a premium on clarity and logic. While the exterior of the great Byzantine church is treated largely as a shell for the interior,

Ottoman mosques maintain a scrupulous balance between the two.

199 Where Haghia Sophia visually plays down the mechanics whereby the great dome is supported, partly by dim lighting and partly by

198 decoration, such mosques as the Sehzade, the Suleymaniye and the Bayezid Cami are flooded with light and glory in their great free- standing piers which carry the central dome; the spatial divisions are muscular and clear-cut. The decoration of the vaults, in which designs familiar from manuscripts and carpets (but now vastly inflated) predominate, further emphasizes these divisions by virtue of

being concentrated in selected spots. A totally different aesthetic governs the way that the billowing volumes of Haghia Sophia flow into each other with little perceptible transition. Where the Byzantine church suggests, its Ottoman successors display. The central figure in sixteenth-century Ottoman architecture is

Sinan (c. 1491-1588), probably a Greek convert to Islam; he rose

to become Chief Architect (effectively Master of Works) and was responsible for something over one hundred buildings in his excep-

tionally long working life. He is the most famous of Islamic archi- 200 tects, both for his own work – crowned by the Selimiye mosque in

Edirne, finished in 1575 when he was over eighty – and for the plans generated by his office, which were exported throughout the

Ottoman empire and were duly executed by his subordinates or by local builders using local detailing. Here, if anywhere, is an Islamic

equivalent to Sir Christopher Wren; and, like Wren, Sinan placed his stamp on an entire city. He it was who fleshed out the rather stark and unadorned external elevation of Ottoman mosques with a whole battery of articulating devices: single- or double-entrance porticoes, porches, gates, fountains, windows, turnform buttresses,

grilles, variegated fenestration, and, above all, the interlocking

volumes of the domes and semi-domes clustering around the great central cupola, with a new type of slender, pencil-shaped minaret of great height (up to 70 m or 230 ft) defining the key points of the

199 The challenge. Interior of Haghia Sophia. Constantinople, founded 537. Following a common Muslim practice, Mehmed II commemorated his capture of the city by turning its great church into a mosque. The 15th-century historian Tursun Beg specifically notes that this same sultan ‘built a Great Mosque based on the design of Haghia Sophia, which not only encompassed all the arts of Haghia Sophia, but moreover incorporated modern features constituting a fresh new idiom unequalled in beauty’.



«. *:




200 Triumph – in Europe. Selimiye mosque, fdirnc. 1569-75. This, the masterpiece ofSinan, the greatest Ottoman architect, has the first Muslim dome whose diameter equals that of Haghia Sophia.


perimeter. His finest mosques are great grey mountains of masonry, but as complex and coordinated as a fugue. This is truly architects’

architecture. The details of mouldings, two-tone voussoirs, embra- sures, muqamas hoods, door-frames and the like reveal a fastidious attention to detail, but they are placed parsimoniously and scarcely

counteract the prevailing tone of austerity. The knife-edge sharpness of the stonework is its own ornament. The interiors are more colourful, with floral and geometric designs painted red, blue and

yellow, calligraphic roundels, stained glass and Iznik tiles, the latter




201 The end of the toad. Complex of Sultan Ahmed I (the Blue Mosque), Istanbul, i 609 17 Nm.m’s stu dents worked in Ins idiom but could not match his vision; this mosque is noted tor its 21, … l/mk tiles

sometimes used, .is at the Rustem Pasha mosque, to cover most of the interior. This balance between structure and decoration was

gradually lost after the Sultan Ahmed complex of [609 17; thereafter Turkish architecture entered .1 baroque phase- which emphasized

loftiness, profuse detail and a restless curvilinearity

The sterling craftsmanship, the confidence in marshalling huge spaces, the long experience in working with modular designs easil)

adaptable to different functions, the- structural know how accumu lated m the course of building campaigns throughout the- empire





all this gives Ottoman architecture its distinctive stamp. It is manifest in the chains of c-aravansarais built along the major trade and pilgrim

routes, in joint foundations and above all in the great kulliyes for which the vast revenues of the Ottoman state were available. Some of these complexes are on a scale without parallel in Islamic architec- ture, like the Bayezid II complex at Edirne or the Suleymaniye, where the mosque is merely the centrepiece of a congeries of build- ings which include an asylum, medical college, boys’ school, kitchen, hostel, cistern, hammam, two mausolea and four madrasas. The largest complex of all is mainly secular in character, namely

the Topkapi Saray – more a royal city than a palace and thus a dis- tinctive Turkish version of a building type also known in Nasrid Granada, Safavid Isfahan and the Mughal capitals of Delhi, Agra and Lahore. Hadrian’s palace at Tivoli shows how ancient this tradition was in the Mediterranean world. The scores of buildings scattered over the many courts of the Topkapi Saray were erected over a period of some four centuries, and many were repaired again and

202 Welfare institution. Complex ofBayezid II. Edirne, begun 14N4. It included a hospi- tal with attached medical college and a separate section for the insane; the Islamic

medical tradition led the world m the Middle Aces.



– j Nerve-centre of the empire. Bird’s-eye view of the tbpkapi palace, Istanbul, i s i h 19th cen tunes. Despite its huge si/c. much of it an intimate, domestic atmosphere. It borrows from the Byzantine imperial palace the notion m~ .1 city within .1 city, [“he informal layout gives little him of the extreme luxury of its interiors.

again. They housed an army of servants and retainers; the royal harem and the various corps of eunuchs; and of course soldiers and

civil servants. The absence of grand structures of imposing scale, on the lines of contemporary European palaces, is noteworthy. Indeed.

many of the buildings arc diminutive and are not articulated into an owrall design of any pronounced a\ialn\ or symmetry. I he atmos

phere is by turns formal and domestic, which is entirely appropriate

since the Tbpkapi Sara) ministered to hod) the public and the private

aspects of the sultan’s lite.

It is ironic the most famous and populai <>i .ill Ottoman art

forms – Iznik potter) and tilework was executed in .1 cheap and



204 (fc/f) Ceramic innovation.

Izmk dish, later 1 6th century. The factories at Iznik rose to prominence just before 1 soo when the Ottoman court began to oversee their management, and steadily increased the range, quality

and quantity of their output over

the next century. The palette was consistently modified, notably with

the introduction o\\\ brilliant

tomato red after I 560. Roses.

carnations, tulips and hyacinths


humble material. This is the best-known and most coveted of all Muslim ceramic wares. I /ink wares owe their name not only to their intrinsically high aesthetic and technical quality, but also to the fact

that very great numbers of these wares, both tiles and individual

pieces, survive, often in excellent condition. Moreover, if a compari-

son is made, saw with the Iranian world between 1000 and 1200 – a period which saw at least a score of different types of glazed wares

being produced – the difference clearly lies in the way that Ottoman potters concentrated their efforts m a single direction. Here the dominating influence o\\\ government-sponsored and government-

financed industry can be recognized immediately. Iznik pottery was

an official enterprise whose gigantic output required much bureau- cratic supervision and financial control. Production quotas were

enforced, salaries were pegged at levels judged appropriate to the

particular skills of the workmen, designs created by the palace studio in Istanbul were sent to the potters’ workshops for transfer on to

pottery and tiles. The wares – a royal monopoly – were exported throughout Europe, and although the heyday of Iznik was the six-

teenth century, the style, and its various provincial derivatives in

Syria and elsewhere, lasted at least another century. As in the case of

public architecture, this strong government interest tended to

standardize production and to reduce variety while maximizing

205 Divine illumination. Mihrab of mosque of the vizier Sokollu Mehmed Pasha, built by Sinan in 1 571


The white marble, the hanging lamp, candles, sunburst inscriptions and battery of windows above the mihrab all harp on the theme of light. Indeed, the Sufis developed around the Light Verse (Qur’an 24:35), traditionally associated with mihrabs, an intricate language for describing mystical experience.




output. And while Iznik became the main centre for high-quality glazed pottery, ot-her provincial production centres declined or were,

indeed, completely eclipsed. But of course the concentration of ceramic production in a single centre was nothing new in Islamic art.

Recent research has pinpointed the evolution ot the characteristic Iznik motifs and compositions and has developed a basic chronology.

Although some pieces bear double- or triple-decker Qur’anic inscriptions, representations of ships, birds and snakes, as well as geo-

metric or chinoisene designs, the favoured subject-matter for Iznik

wares, especially tiles, was floral motifs and the distinctive feathery

204 saz scrolls. Carnations, hyacinths, tulips and other flowers recur in

endless combinations. Set 111 apses m a qibla wall, they turn a mihrab into a paradise garden. The palette is limited and extremely dis- tinctive: white, light and dark blue, purple and a vivid tomato or

oxblood red. Most ot these tones are used tor foreground and back-

ground alike. The best Iznik ware owes its reputation to the purity and strength of these colours, and the decline of the industry can be

detected in the gradual tailing ofl ot colour rather than design.

The arts of the book in Ottoman times drew their initial inspira- tion predominantly from those of Iran under the Timurids and their

rivals, the Turcoman Aqqoyunlu, m the later fifteenth century. In some of the associated specialities – for example, in most Qur’anic illumination and in the illustration of verse romances – they did not progress significandy beyond this heritage. But for the most part they

struck out on their own. In the field ot book painting, for instance,

entire cycles ot” religious images were devised, almost for the first

time in Islamic art. This involved the creation ot literally hundreds ot

new images. The Anbiyaname Hook of the Prophets’) of 155N is an early attempt at this genre, but pride ot place naturally goes to the

six-volume illustrated life ot” the Prophet Muhammad, the Siyar-i Nebi (1594), whose images of the founder of Islam, with a white veil and a gigantic flame halo, sometimes attain a visionary intensity.

Illustrated guides to Jerusalem and to the Holy Cities of Arabia, enti-

tled Futuh al-Haramain, became popular during the sixteenth century. Despite their strongly abstract nature, expressed for example in the

simultaneous adoption ot’ multiple points ot’ view and in their

indifference to accurate spatial relationships, they have the bright

colour and lively directness ot a cartoon strip.

In secular painting, too, which accounts for most of the illustrated

manuscripts o\ the period, there were important developments,

inspired perhaps by a desire for realism – a quality pursued more



consistently than ever before in Islamic art. I he thirteenth-centun illustrations of the Maqamat ofal-Hariri had held up a satirical mi to daily life; the Timurid Zafamama (‘Book of Victories’ and perhaps the lost Qkhanid Chingiznama (*Book of Genghis Khan


) had chronicled military exploits in a loosel) epic style. But Ottoman painters, when they reverted to tins subject matter, approached it in the spirit ot\\ diarist or journalist, and inflated it to a m.nor artistic genre. The process began modestly enough with the versified Selimname o\\\ [525, but by c. [558 it reached its apogee with the Suleymanname of ‘Arifi. This history in verse has more in common with a newsreel as it patiently chronicles the ups and (occasionally)

the downs ot the sultan’s campaigns. Its great set pieces of sieges and 206 battles often extend over two pages. Sometimes an illustrated manu- script was devoted to a detailed prose account of a single campaign, such as that ofSzigetvar (written [568—69) or the one in Iraq and western Iran between [534 and [536, the subject of a manuscript produced in [537 by Matrakci Nasuh and containing 128 largely topographical paintings – essentially a traveller’s guide. The same painter recorded Sulcvmans Hungarian campaign and that of his

admiral Barbaras, both occurring in [543; here again there are tie

quent representations of cities. These works may owe something to the widespread contemporary European fashion for topographical woodcuts of major cities. Hut they are also remarkably up-to-date:

elaborate and richly illustrated accounts o\~ the imperial campaigns

were produced as soon as possible after the fighting had ended.

Illustrated prose histories of entire reigns, such as those of Have/id

II. Suleyman the Magnificent. Selnn and Murad 111. were also pro duced; so too were illustrated portmanteau histories of the Ottoman dynasty. In much the same vein were the books describing the various public festivals which enlivened Ottoman society, such as the Surname (1582). Yet for all the colour and detail of Ottoman paint

ing, and the often panoramic scope of individual pictures, something

is missing. The earnestly literal bent o\ Ottoman painters (which may explain the popularity of elaborate maps m Ottoman painting) was fundamentally .it odds with the conventions within which they

were operating. Constant!) refined through long use in Persian

painting, those conventions were designed to keep the real world .it .1

distance, and to transform nature mt<> art. ttius Persian painters

appealed to the imagination to decode their images Most >>f these

linages were transferred wholesale into Ottoman painting, but in the

proeess they lost much of the inner relationship between the figures,



206 (left) The Ottoman threat to Europe. The imperial army besieging Vienna in 1529. Loqrnan, Htinername, 1588 (£2570). The architecture, with its fanciful palette, reflects

at one remove tot) many the influence ot European topographica] engravings, but the

portable pomp ofthe great scarlet tent, complete with its flimsy battlements, is

vigorously rendered

207 (right) The sultan relaxes. Ahmed HI watches darners .iiul comedians at the

\ uppodrome at Istanbul. Sumame-yi I \hb\,

[720 32, illustrated by I evni. The festivities

to celebrate the cir<. unu lsion of the sultan’s tour sons m 1720 lasted fifteen days ami nights, and included guild processions,

firework displays, regattas, mock battles, magic shows and acrobatic performances

including tightrope walkers. I he variations

111 the scale of figures reflect social hierarchies.

or between figures and landscape or buildings, which had given the

original its depth and nuance. The densely populated images of

Ottoman painting describe the surface of things, and they cover a great deal of ground. But they are scarcely exploratory.

Compositions manipulate groups of buildings or people or even

colours en bloc, and these busy scenes, tor all their overwhelming detail, lack finesse and attest a bask inflexibility. There is a sense of

painting by the yard. Yet at its best, Ottoman painting – profiting, it seems, from the example of Italian Renaissance artists such as

Costanzo da Ferrara and Gentile Bellini, who painted a memorable likeness of Mehmed the Conqueror as a brooding, self-contained intellectual – produced a series of excellent portraits which owe vir- tually nothing to the Persian tradition. Here a mastery of line and a

series of happy inspirations in pose and colour evoke both the public

and the private faces of royalty.





208 Thirst for learning. Back

doublure ot Commentaries on the

Maqasid ofal-Taftazani, made for Prince Bayezid. Amasya,

1477. Several Ottoman sultans were bibliophiles; it was a

tradition tor members of the elite to have libraries, and

respect for knowledge found

expression in luxuriously

appointed manuscripts with

fine bindings. Ottoman scholars wrote copiously if not

originally on many topics, especially historiography and

the religious sciences.

Turkish bindings, and specifically those datable between c. 1430 and c. 1 5 10, are distinguished from most of their Arab and even

Persian counterparts, both contemporary and earlier, by their visual

aesthetic, which is dominated by an intense appreciation of colour

and its possibilities. The imperial Ottoman binders in this period definitively forsook the all-over geometrical and vegetal designs of

earlier tradition in favour of a field comprising a central medallion

and cornerpieces. While the size of the field is allowed to vary

dramatically according to the scale of these elements and of the

borders, the dominant feature of many bindings is a serene expanse of empty space. Its strong colour ensures that this space is never neutral. Set into it at precisely determined points are the cartouches

or medallions and corner-pieces which animate it. All is controlled,

as in Ottoman architecture, by a matchless sense of interval, the fruit of long observation and experiment. Perhaps this instinctively satis-

fying placement o\ decorative features is the result of mathematical

calculation – it may have been governed by proportional ratios of the kind seen in so much Islamic architecture and book painting.

While some ot the basic stock-in-trade of these Ottoman binders came from Iran – the use of coloured leather, filigree, cartouches –

they managed by dint of extreme concentration, and by setting up an imperial scriptorium on Iranian lines, to fashion their own style from




these elements. Perhaps, like the Fatih mosque in [stanbu unprecedented scale by Ottoman standards, and the earliest Ushak rugs, whose complexity represents a quantum leap forward in Turkish carpet design, the suceess of this ateltei owed something to the euphoria generated by the tall of( Constantinople, Some ninety manuscripts dedicated to Mehmed II survive, .1 record total an Islamic rulers, and this remarkable number provides .1 context tor the rapid evolution ofa distinctive imperial style, fhese original dated or datable bindings of the period 14 so 1 500 also offer precious clues t.. other apparently contemporary artefacts, especially in view of the total dearth dfOttoman textiles, ceramics or metalwork that period. The spirit and structure of” the decoration found m these bindings, as well as their detail, is therefore a benchmark tor the definition of Ottoman art in the key generation after [453. I he classic balance, the seemingly effortless harmony, the chaste detailing (so curiously prophetic, despite the different idiom, of’ the spirit of

Robert Adam) and the subdued luxury of contrasting colours and textures confer a unique eclat on these bindings.

Ottoman textiles of the sixteenth and seventeenth centuries survive in considerable numbers and. what is more, in the form of complete items of” clothing. While Islamic textiles of’ the pre Ottoman period do survive m reasonable quantity, they are nearly .ill incomplete. Thus, while they offer plentiful data on textile design

and technique, their evidential value tor the history of costume is

very limited: the wider context o\~ these fragments has gone. I his

depressing picture changes dramatically in the imperial Ottoman era. thanks to the holdings of the Topkapi Sarav. Even entire tents have

survived. The sultan’s silk robes were used only once and then stored

in the treasury, and a wardrobe-master was specifically charged with

this task. Official protocol dictated frequent changes of* apparel,

thereby ensuring that these clothes did not wear out. so that the

principal threat to their survival lay in the conditions of

Naturally, museum standards of conservation did not prevail, but the range of clothing still preserved is gratitvingh tull though very

clearly not all items were stored to begin with, and the collection

would require frequent sifting. Over ^.S”<» textiles remain.

1,000 of them kaftans. These were woven not just tor rovil use but

also, following an ancient Islamic custom, for distribution

foreign dignitaries and potentates or is .1 means of honouring high

officials. Hut the Topkapi collections also contain a wonderfully

varied assortment <>t embroidery work applied to obje< ts as disp



as leather boots, sjiarkskin boxes, bookbindings and tankards, hand- kerchiefs galore, headbands and kerchiefs for the royal ladies, gloves, sashes, purses, quilt covers, portfolios for documents or Qur’ans, cushions, and military or hunting equipment such as saddlecloths,

shields, quivers and bowcases. The more expensive embroidery employing gold thread (zcrduz) was made by specialist teams working in the palace; other types of embroidery were often produced by

women in their own homes. Floor coverings comprised not merely rugs, traditionally capable of withstanding heavy wear and tear, but

also brocaded satins and velvets, and sometimes silk was spread along

a prince’s route for him to ride over — a true ‘red carpet treatment’. Thrones, sofas, litters and the like also featured elaborate textiles.

Such was the scale of the textile industry that from early Ottoman times, while the capital was still at Bursa, it required close govern-

ment supervision. This extended well beyond the control of import, production and pricing of silk, for fine Ottoman textiles were exported in large quantities to Europe and were thus of importance

to the Ottoman economy and state. Quality control was therefore vital. Tailors, weavers, silk-spinners, textile designers, producers of

metallic thread and similar specialists were organized in guilds and

their salaries were fixed by the bureaucracy. Nor was it only the court workshops that were subject to imperial control. Punishments

were inflicted on those found guilty of cheating: in 1564 an imperial edict decreed that two-thirds of the looms then functioning in

Istanbul for the production o\ textiles using gold and silver thread

were to be shut down tor that very reason. Lists of palace expenses

209 A tented palace. Ottoman tent. 17th century. Textile architecture, with its immemorial nomadic asso- ciations, played a vital role in Ottoman lite, as can be gauged from its frequent appearance m contemporary painting. The fictive arches here themselves evoke textiles – m this ease, prayer rugs.




: Portable mosque. Silk

prayer doth, i~th century It

depicts .1 triple mihrab with

stylized finials in the form

of tulips and date palms.

openwork mosque lamps and outlines tor the worshipper’s

feet – which contain further

palms, a theme taken up again,

but in lusher form, in the

border: it may have paradisal asstK iations.

specify how much craftsmen were paid in cash and kind, and even the costs of the materials they used .ire recorded, rhese were not

independent craftsmen or even self-constituted teams operating a

COttage industry: artists serving these gigantU State enterprises were

in effect civil servants. And of course the\ had to conform ( leark there was .1 house Style to he followed, and while it developed

certain motifs to a pitch of elegant assurance the hatayi (Cathayan,

i.e. ( Chinese) or ./ ‘reed pen 1

or ‘red flower’ style with its serrated

leaves, rumh ‘Greek 1

spiralling arabesques), chintamam motifs i

combination of stripes and three dots>, multiple ogives it dis

couraged experimentation <>n .1 broad front.



Textiles, moreover, were only one element among many specialist luxury crafts. Every last accessory of the royal wardrobe displayed the

same love for expensive, often outre materials and for fastidiously del-

icate detail. It is entirely possible that the Ottoman artefacts of this kind, far from expressing a distinctively Ottoman taste for such objects, were typical of earlier Islamic dynasties too and that it is the

uniquely privileged status of Istanbul as an imperial Islamic capital

that was never sacked which accounts for the sheer quantity of sur- vivals – because, of course, such objects would be the very first to be looted. The lack of parallels, not only in the narrow chronological and stylistic sense, but in the most general typological way makes it very hard to evaluate this Ottoman Prachtkunst. How can one tell – other than by stylistic analysis – what is distinctively Ottoman about the mother-of-pearl belts, turban ornaments, Qur’an boxes, leather

canteens, quivers, wickerwork shields, gold-inlaid steel mirrors,

bejewelled gold maces, ivory buckles, jade tankards and a host of

other luxury objects, from masterpieces to elegant knick-knacks?

Yet this is not the main issue. The wider significance of this material, representing both Ottoman production and centuries’ worth of gifts from other Islamic states, is that it provides an unequalled visual

context for the life and ceremonial o\\\ Muslim court.

211 Royal logo, liighra of Suleyman the Magnificent. The form is a standardized emblem of the ruler, and was used for centuries, but is personalized in that it accommo- dates his own name. Its origins have been variously interpreted as an inky handprint or a bow with the ruler’s name written beneath it. It authenticated chancery documents and appeared on coins and banners; its misuse was a capital offence.



Select Bibliography

Reference Works

Ah. “AbdaDah YuMit

and Commentary, repr Brentwood, Maryland

Bacharach, Jere, 1 \ tie and I ondon, i>j-r>

Bosworth, c ‘MB

A Chronological and < . linburgh, 1996 Bosworth, Clifford I- and Joseph Schachi

77i t – L |

I Ktord. i’j~4

CresweO, Keppt

and (


S s ment, C ‘.urn. 1984

un, I ciden, 1961 onwards;

B vols to date

Endress, Gerhard tr.m^ Carole Hillenbrand), In Introduction

to Islam, Edinburgh

Hodgson, Marshal] G.S., The Ventun

ils . Chicago,

Lewis, Bernard, 1 ondon, 1958; many reprints

Lewis, Bernard ed . JTu World ofIslam Faith, P Culture, London and New York, i’j~<>. paperback ed. 1992

1 ewis, Bernard, Ann K s I ambton and Petei M I I mbridgc Histor) of Islam, 2 vols., Cambridgi

Rahman. Fazlur, Islam, C In

Turner, Jane S ed . The Dictionary ofArt, London, 1996


Brend, Barbara, Islami Irt, I ondon and Cambridg


Bloom, Jonathan M . and Sheila S. Blair, I I ondon. 1997

Bloom, Jonathan M . and Sheila s Blair, Th In and and I ondon.


Emnghausen, Richard, and Oleg Grabar,

fork and

Harmondsworth, 1987; paperback N Haven,

Cirabar. O en and

Grabe, I

Irwin, R


Allen !

Blair, Sh

I ondoi

Kuhnel ;

Kuhnel I

1 rhrmai

I Mil h icl I

I ondoi

\n liiw-i tun-

l Ireswell, Kepp

( Ireswell, Kepj

I /> 62

c Ireswell, Kep|

I lillenbrand, Robert, I

Hoag |ohn D |airazbho) . Rarique \

Bombay \lli lull. ( rCOl

I ondon and New York, 19 ed

I In- \rts ol tin- Hook

Arnold. I nomas W.I







Lings, Martin, The Qur’ank art of Calligraphy, London,


Pedersen, Johannes (translated by Geoffrey French,

edited by Robert HiDenbrand), Tfie Arabic Book.

Princeton, 1984

Safadi, Yasin H., Islamic Calligraphy. London. iv~ s

Schimmel, Annemane, Islamic Calligraphy. Leiden, 1970 Schimmel, Annemarie, Calligraphy and Islamic Culture, New-

York and London, 1984 Welch, Anthony, Calligraphy in the Arts of the Muslim World.

Austin, 1979

Welch, Stuart C, Royal Persian Manuscripts. London and New York, 1976

Ziauddin, M., A Monograph on Moslem Calligraphy. Calcutta, 1936

The Arts of the Loom

Baker, Patricia. L., Islamic Textiles. London, 1995 Bier, Carol (ed.), Wovenfrom the Soul. Spun from the Heart

Textile Arts of Safavid and Qatar Iran i6tli-ioth Centuries,

Washington, D.C., 1987

Black, David (ed.), Tlie Macmillan Adas ofRugs G Carpets, New York, 1985

Edwards, A. Cecil, The Persian Carpet, I ondon, 1953 Erdmann, Kurt (trans. May H. Beattie and Hildegard Herzog, ed. Hanna Erdmann). Seven Hundred ) Oriental Carpets. London, 1970

Folsach, Kjeld von, and Anne-Mane Keblow Benwed. Woven Treasures— Textiles from the World ot Islam.

Copenhagen, 1993 Kuhnel, Ernst, and Louise Bellinger. Catalogue ot Dated Tirol

Fabrics: Umayyad, Abhasid. Fatimid, Washington, I I (


Mackie, Louise W., Tlie Splendor of Turkish Weaving: an exhibition ofsilks and carpets of the ijth-lSth centime?.

Washington, D.C., 1973 May, Florence L., Silk Textiles ofSpain, eighth to fifteenth

centur)>, New York, 1957 Serjeant, Robert B., Islamic Textiles: Material for a History

up to the Mongol Conquest, Beirut. n>-i

Ceramics and Glass

Allan, James W., Islamic Ceramics. Oxford. 1991 Atasoy, Nurfaan, and Julian Raby (ed. Yanni Petsopoulos), hmk: Tlie Pottery of Ottoman Turkey. London. 1989

Aril. Esin. Ceramics from the World of Islam. Washington,

D.C., igrji

Caiger-Snuth. Alan. Lustre Potter): Technique, Tradition and

Innovation in the Islamic and the Western World, London.


C’arswell. John, Blue and White: Chinese Porcelain and its Impact on the Western World. Chicago. 1985

Grube, ErnstJ., Islamic Pottery of the Eighth to the Fifteenth Century m the Ken Collection. London, 1976

Jenkins, Marilyn, ‘Islamic Glass: A Brief History’, The Metropolitan Museum of An Bulletin. 44:2, 1986 (complete issue

Lane. Arthur, liarly Islamic Potter). I ondon. [947 Lane. Arthur. Liter Islamic Potter) (revised ed.), London, 1971 Ptnlon, Helen. B<naki Museum Athens, liarly blamk Ceramia Ninth to Late Twelfth Centuries, 1 ondon, 1980

Porter, Ycnctia. blame Ides. 1 ondon. 1993 Watson. Oliver. Persian Lustn Want, London, 1985 Watson, Oliver. ‘Ceramics*, in Tteasun 1 of Islam (ed. I oby

Raft), 1 ondon, lot


Atil. Esin, Renaissance of Islam; . In of the Mamhues,

\\ ashington, I > ( ,1981

Dodds, Jerriryni lalus The Art of Islamu Spam,

V ork. 199a

Hillenbrand. Robert (ed . Th \n of the Saljuqs m Iran and Anatok 1994

Petsopoulos, Yanni (ed I, Tulips, Arabesques and Turban-,

1 ondon, 198a

Arthur U., and Phyllis Ackermai I I Survey of m Prehislorii Times to tin Piesent, I ondon and

I Michael, and Rachel M Ward, Sukyman the Magnificent, I ondoi





Abd servant, slave (used in mam Muslim names in combination with one ol tin- names ol God, as in ‘Abd il Malik. ‘Servant oi the Kuil:”

ablaq literally ‘piebald’; used especially oftwo tone marble decoration

Abu father (of); used in mam Muslim names in combination with tlu- name ofthe first-bora son

amir commander, prince arabesque geometricized vegetal omamenl atabeg guardian ofa prince; often .1 governor

b. son of (Arabic ‘ibn*, ‘bin’]

biib gate . .iIno used in .1 spiritual sense

baraka ‘blessing’; .1 quant) ol divine grace dispensed In ( .>>d

and by those whom He selects bismillah ‘in the name ofGod’ caliph the name given to leaden ofthe Muslim communit)

(from Arabic khalifa, ‘successor’ to Muhammad end Frida) Mosque rurkish; the Arabic form is jam! caravansarai lodging place for travellers, men hants and

their goods; often fortified and situated on .1 trade route

chinoiserie decorative motifs derived trom Chinese sources

dinar cold coin

dirham silver coin later copper

diujn government office or ministry; royal reception

chamber; collection ofpoems Fatiha opening chapter ofthe Qur an ghazi warrior tor the faith

gunbad dome hadiths collective body oftraditions relating to Muhammad and his Companions; they constitute one ofthe sources of

guidance for Muslims

hajj the pilgrimage to Mec ( .1

hammam steam baths; bathing establishment tor the public hijra (hegira) Muhammad’s emigration trom Mecca to Medina in <>zz. the date which marks the beginning ol the

Muslim calendar hypostyle having a roofsupported by multiple columns a

standard type ofmosque ibn see b.

Ilkhan a Mongol mler. subordinate to the Great Mongol Khan

imam spiritual leader, prayer leader, descendant ol Muhammad’s son-in-law . leader oi the Shi ite 1 ommunit\

Islam submitting oneself to the will oi Allah

Isma ilis or Seveners member- ofa Shi’ite group who bebeve that the legitimate su Imamt ncludes

the seventh imam Ismail, son oi the inum Ja’far al-Sadiq

{MM or Hat-rooted hat! open at one end; often used to mean simpU a covered hall 111

i.imi the great mosque in which the communal I lied


khamsa prim i prim e

khanjjh Suti htional

tunerar\ ftu


MuJ book kuti.

kullty, 1 typically Ottoman foundanon comprising multiple buildings, . entred on 1 mosqui !”•;

edu< ational Ol Welfare IhiiI

m*4hh*b sc ho.. I ,.t law tttaditta . IIS

mmmm in institution tor the stud) oi the ortl s. hih es

Maghrib tin Muslim in. lulling I uiiisia

fri.iiJ./M open puhlii square >>i plaza; central m.imluk i often aho used of manumitted HMfian 1 private enclosure in the mi


mathkai mausoleum ol 1 1

mafhrabiya window grilll Ol with geonieliu al or iiiterl.u ed designs

masjid mosque literal!) ‘plac< ofprostrarj masjid-i jami Persian term tor the I rida) mosqui

mihrab arched niche, usuaO) concave but sometin

indicating the direction ofM qikta :iiuis..t pra\ er

ffiiru 1 potter) m which Colours are applied both u over the gla/e

mm a ret the lower oi a mosque trom w hi. h the faithful are 1 ailed to prayer

minbar stepped pulpit in a mosque, used lor the

pronouncement >>t the khuib.i

mi raj the as. ent ol the Prophet Muhammad into h mosque a pl.u e where Muslims worship mukaqfuq a majestic Our .mi. cursive script muqanias honeycomb or stalactite vaulting mule up .<t

individual . elk or small ar. lies

mur.j././.i m album ot pictures Muslim 1 person who follows the n MM 1 book ot writii P ttttfhhi ursive style oi \rabi< script; 1 scnbal hand

masta Kq

loops P

phhtaq loft) irch framii fa*** monumental

portal uall) m matters ot 1 i\ il law /.j-r



in M .juhh.i lomc (hir .111


S»\«iurs Isma ills


thtikh SuB



Shi’a (hence Shi’ite) generic term for a series of sects not

regarded as part of orthodox Islam: they all re’cognize ‘All

(cousin and son-in-law of the Prophet as the first

legitimate caliph

simurgh mythical bird like a phoenix

squinch an arch spanning the corners of a square chamber

and acting as a support for the dome Sufi Islamic mystic

sultan ruler, king

Sunni orthodox Muslim (see Shi’a)

sura chapter of the Qur’an

thulth a formalized and elongated version ot naskhi


tiraz inscribed fabrics made in state workshop and often presented by the ruler to those he wished to honour

tughra monogram ot the sultan (Turkish) Twelvers the most numerous branch of Shi’ites in modern

tinier they believe that the legitimate succession of imams ended with the twelfth imam Muhammad al-Mahdi

ulama those who possess knowledge’: scholars of Islamic theology and law : clerics; the learned class

vizier minister

Sources of Illustrations

Department of Antiquities. Amman 4. [6; Courtes) Cathedral Treasury, Apt 63; Al Arabv Magazine I; James

Allan 160, 182, 1 S3; Aga Khan Visual Archives, Ml

Denny (1984) 81b; Collection Prince Sadruddin Aga Khan

189; Press Broadcasting and Tourist Department. Ankara 92:

Architectural Review Photograph by Sheridan Cantacuzino 87; Azimut S.a.s. 136; Directorate General of Antiquities.

Baghdad 90, 96, 168; RoloffBenv National Archi Canada/PA-i9S936. by permission ofNickle Arts Museum. University of Calgary 82; Staadiche Museen zu Berlin –

PreuBicher Kulturbesitz Museum fur Islamische Kunst, C bpk 20, 25, 26, 27. 31, 36, 4.5, 65, 103: Sheila Blair and

Jonathan Bloom 140: Helen <x Alice Colbura Fund. Courtesy Museum of Fine Arts. Boston 33; Boudot-Lamottc 200; Barbara Brend 12; Courtesy ot the Byzantine Institute

199; DAI, Cairo 113; Picture by kind permission ofthe Egyptian Publishing Company – 1 ongman. Cairo 3. 1 i~. Museum of Islamic Art, Cairo si. 121; National Library, Cairo 126, 127, 176; Courtesy of the Arthur M Sackler Museum, Harvard Universm An Museums, Cambridge. MA 166 (gift of Edward W. Forbes 1 . boh Strong, Francis H. Burr and Friends of the Fogg An Museum Funds), 190 (Bequest of the Estate of Abbs Aldnch Rockefeller);

J. Carswell 1 10: Kunstammlungen der Veste

Coburg, Gennany 62; Photo AC Cooper 73; 1 he David Collection, Copenhagen, photo Ole Woldbyc 159 State Collection of Art, Cracow 209; after K.A C CresweD Early Muslim Architecture, vol. I (1969) 2 top 14; Abbas

Daneshvan 74; J. E. Dayton 153. 169: Photo Jean Dieuzade 197; Reproduced by kind permissionof the Tnistces ofthe Chester Beatty Library, Dublin 40. 41. 75, 133: Edinburgh University Library 161. 162. 164; Olga Ford 91. [j Godfrey Goodwin 195; Alhambr.i. Granada 149: 1″ Grube The World ofIslam 1196-” 53; Soma Halhdas Photography 8 (Photo by Jane Taylor


). 9. 203: after Robert Hillenbrand Islamic Architecture. Fonn. Function and Meaning

(1994) 2 (except top left); Keir Collection. Ham. England 58; after R.W. Hamilton Levant I. 1969 17; after R.W Hamilton Khirbat .1/ Mqfiar (1959) 1 8; R. Hillenbrand Islamic Architecture, Form, Function and Meaning 1,1994) 81; Robert


Hillenbrand 1S9; Tiroler Landesmuseum Ferdinandeum, Innsbruck 86; Museum ot I urkisli and Islamic Art. Istanbul 208; I opkapi Sarav Palace Museum. Istanbul 1 19. 206, 207,

1 >alu Jones 78; Ilie Nelson-Atkins Museum of Art, Kansas City, Missouri (Purchase Nelson [“rust) 19s; A. F.

Kersring 28, 44- s4. 109, 114. 116; K.irtav Madrasa Museum.

Konya 45; Jonas 1 ehrman 14s. Grassimuseum, Leipzig Museum fiir Kunsthandwerk: frontispiece. [86; Lawrence Lockhardt 178, 181; B\ permission of the British

Library, London 187; Copyright C British Museum. I ondon s. 6 ngl $7; ( onw.n I lbr.iry .

Courtauld Institute of Art. University ofl ondon 1 ss. I rorn ban Group bbrary. First published in The Qur’an

and Calligraphy, I ondon. Bernard Quariu li I td, catalogue

. ; 17; V eV A Picture I ibrary, London :- 8,191, IV2, 193, 194. -04; Mllseo

Arqueoi Madrid 1 54. Mas 128, 131. 1 j8, 139. 14a, 147; Viktoria Meinecke-Berg in, ii2;(.

Mott 130; American Numismack Society, New York 7; Courtesy, rrustecs ot the Pierponl Morgan Library, New York 163; Bernard O’Kane 13. jo, ss. 56, S7. 77. ns. 129.

141. 151, is”. i”4. 1 — , 198; Ashmolean Museum, Oxford 6 bottom left); I he Bodleian I ibrary, University of Oxford

.sell Archive. Ashmolean Museum, Oxford 1 s, 24.

98; Museo de Navarra, Pamplona 135; C Bibhotheque Nadonale de France. Pans 99. 101. 102. Musee du louvre.

Pans 34. 39; Philadelphia Museum of Art f>4; Umversit\ ot Pennsylvania Museum. Philadelphia 76; Josephine Powell 79,

93, 143. 144. 1-2. 1 79, 201; Photo ‘ RMN 105, 107, Il8; National I lbrarv of Russia, courtes) B N Robinson 175; Photograph by J. Rock and A J. Sutherland 4*. so: The State Hermitage Museum. St Petersburg 32, 94.

73; I ) Talbot Rice 1 S4: Mine. Th. Ullens de

Schooten 1S2; Foto Bibhoteca Vatican? 132; Treasury of St

Mark’s. Venice 61; Osterreichisches Museum fur angew mdte Kunst. Vienna 1 23; Osterreichische

Nationalbibhothek. Vienna 124; Photo Roger-Viollet 170;

Courtesy of the Freer Gallery of An. Smithsonian

Institution. Washington. D.C. 60, 6X, 70, 71, 106, 165, 167;

Edouard Widmer 202; Roger Wood/Corbis 29. 37.




Numbers in italics tefei to


Abarquh >^. io\-z

‘Abbas 1 -_f>. 228-30, 234,

236, -4″- -4

‘Abbasids 1 7, 38-9, 41


46-7, jo-i, $3 4. $9.

61-3, f»;. 72, 76, v 101. in, 125, 131, .

[69, 1-4. IV’. 201, »I2,

256, 263

‘AM al-Mu’min isf» Abraham :>. i<>>;

Abu Ishaq :ih, 160 Abu Said 1 57, ; Abu Zaid 129-31 Afghanistan ;<j 4 ,4s 6,

54, B6, i< v ~.




Aght’amar 40. <n,

Agra 261, 268

Ahmad lala’ir. Sultan 11 1 Ahmed III 274, -v- Ala al-Dm Kayqubad 1 1 Et,

Ala al-Dm Altunbugha

Alanva 120, 1::. jj

Aleppo ^4. 1 1 5-7, 1 2

140. 1 .

Alexander the Great n s.

Alexandria \<>. ($3, 157

Algeria 61, 169, [84, > s ~-

Algiers 184, i8t

Alhambra vases 1 «v 1 – — – ‘4^ Ah J7.6I, 77 •

Ah b Yusufb lashtiu 185 Aljaferia 181-2, 1 B4, 1 <^

Allah 77, iv4- -14 263

Allah vardi Khai

Abneri 1

Ahnohads 169

Abnoravids iti

97, 100, 102, III ;


Anau –




•\ra\rs rivei

Ardabil 2

235 f, 246 \i


Anti 273

Armenia 46, ”4. ‘>;

ti8, 1 j j. 1 ;r<. 1 39


armour 114, 254 \rtuqicb m, 1 1 5, i Assassins 8

*W\ na 14.1 s

astrology 56, 68, 7 . s 1 . <><>.

[23, 1 13, 1 $6

astronomy 125, 167

Aswan •\v\UI I 57

S III-2, \ 12. I 24

automata lis. 1.^. [62

•\\\ubkb 82, III |, 135-6,

1 ;v [97 ‘Ayyuq Azerbaijan 204, 226,

‘A/i/ bi’llah. al ^ ; 4

Hah al-1 Utuh f>4. 44 Bab Mardum mosque

1 \8

Baby Ion 14. 1 2<>

Baghd .’. $2, v- 9, 62 4 s 7,

s v. 1 if’.

175, 1 80, 196, 199

2 1 2, 22 1 , 22



Baibars 144- 151

Baibars al-Gashankir 14


Bain al Qasrain 14 1

Baisunqur b Shahrukh 214,

Balkh 4-

bannen 151, 1

Banu Hud 1 –

Bapnsterc de Saini I <>uin


Barquq 14”. 1








Bobrinski bu •


bru k onumi 1 •


Buddh Buddhism 1 j’>. 161

Bukhai 1

Bukhtegin, Abu Mansui



Burujirdi 1






Byzandum m

1 lire .

I 4′>. 1 }V 1 i


[64, 2

candlesti< ks 13

1 aravansarai 29


i arpets 1 7, 96, 118, •

216, 229

-•<. 191

castanets \n.

( au< asiu

1 eladon $3, 8;

( entral \sia .


( hah U !

1 >4



– mbul


( teupl


I )



I Kiln

I >haki al Mausili, Al

.• . 114





122, 125, 132, I3 8 -40,

148, ISO, 157, 159-60,

162, 167, 176-7, 196-7,

246, 258

Erzurum 116-7, 88

Euphrates 36, 38, 228

Europe 137, 144, IS*. ’57-

160, 174-5, r 90, 197-8,

208, 212-3, 229, 241, 244,

250, 252, 255-7, 263, 266,

269-70, 273-4

ewer 17, 18, 47, 83-4,

135-6, 151,218,253-4

Fars 204-5

Fatima 61

Fatimids 61-5, 67-8, 71-2,

74, 76-85, 89, 98, in,

113, 140, 143, 162, 177

Fez, 76, 184, 187, 144-5

Firdausi 90, 98, 162, 210,

224, 250

fire altar 19, 21

Firuzabad 40

folk art 56, 90, 187

fountains 32, 68, 179, 184,

188, 194-5, 248, 264

Fudain 17-18

Fustat 12, 61-2, 64,72, 81,


Garjistan 108

Genghis Khan 196-7, 213 Georgia 115, 226, 229

Gevas 118, 8g Ghazali, al- 87

Ghazan Khan 198, 200, 201 Ghaznavids 8, 86-8, 90,

108, 206

Ghiyath al-DinJami 246

Ghur 107 Ghunds 90, 108 glass 50, 54, 79, 83-4, 122,

136-7, 149, 156-7,

159-60, 165-6, 94, 120

Gothic art 171, 201

Granada 169, 180, 181,

189-91, 193, 268, 150-1

Great Khan 196-7 Greece 21, 38, 80, 82, 133,

212, 255, 264, 279 Gunbad-i Qabus 101, 105,


Gur-i Amir 214, 217, 236, i6g

Gurgan 99 Guyushi, al- 76-7

hadith 51, 220

Hafiz 246, 254 Hafsids 187

hajj 34, 94 Hakam II, al- 178

Hakim, al- 71, 84; mosque

72-3, 77, 53*

Hahma Khatun 1 1


Hama 143, 110 Hamadan 100, 103, 108 hammam 72, 200, 268 Haram al-Sharif 148 Hariri, al- 129, 162, 273

Harran 46

Harun al-Rashid 39, 51 Hedwig glass 84-5, 62 helmets 142, 190

Herachus 21, 134 Herat 94- >>2. 214. 217-8,

220, 222-5. 2 37. 249. 2

hieroglyphu S Bo

Hijaz 24, 58, 72. 1 39

Hira 53. 38

holy war ^4. >•’;. 111: see

also jihad

Hulegu Khan 206

Hunemame 2-4. 206 Hungary 256. 273 hunting ji-2, 15, 46-749,

56. 90, 110. 114. 135,

137, I 53-4. 162. 1 ft

223. ::v 250, 278, 1 $5, 135, 175

hypostyle 120, 2

Ibn ‘Abb.:s 95

Ibn al-‘Arabi 243

Ibn al-Bawwab $7 9 Ibn Bibi 113. 114

Ibn Khaldun 1 32. 167, 177 Ibr.ilnm Suit. in 224

Qkhanids 196-201, 204-5,

2cx;. 212. j 14. 216, 2 1 v 221. 226. 229, 273

illumination 90- 1 . 96,

99-101. 103. [25, 164-6,

170, 176. 202. 205. 212.

216. 220. 223. 246, 270.

40-I, 4-. \2t

Inal 14S. 1 [6

incense burners 4-. <;s. h;.


India 11. 1-. 34, 40. 50-1,

90, 107, 131. 140. 1 50,

159. 20s. 213-4. — –

246. 256, 258-9, 261

Inju 205. 209

Iraq 13-14. 19. 21. 25, 40-1.

44- v 4-. SO, 56. S9> 68, 83, 86, 89, 100,

111-3. 1 16. 122-5. ‘2~.

132, 135. 140. 1S2. 162.

175, 184, I96, 206. 2 12.

226-7, 256, 258, 273

Isfahan 89, 200- 1 , 212. 214.

216, 226, 229-30. 232-4.

236, 240, 246, 249-50.

261, 268; Friday Mosque

105-6, 200-1, 81, 155;

Masjid-i Imam (Shah) 226. 230, 177, 179

Iskandar Sultan 214. 223

Ismail I, Shah 235, 237, 182

Isma’U II, Shah 228. 240 Isma’ihs 61-2, 73, 77-8, 87,

1 12

Istanbul 100. 21-. 221


255, 260-1, 263-70. 2-4.

277-80, 198-9, 201.

203, 205, 20-; see ol<o


Italy 66. S4, 90. 132, 140,

157. 179. 198, 210. 224.

229, 2-4

ivory 17. 46. 50, 57. 65,

1 28, 146. 1-0. 177-8. 190,

2 16. ^ ;N-r,. 281 . 4J, / ,•-/-_>

ium 91. 105, 108-9, 1 16, 120-1, 125. 201. 21-. 232.

234, 258

I/mk 255-S. 260. !<•<

-:. 1 jo. 204-3

‘ 46

Jala’irids 212. 221

|.im 1

Jativa v

|aus iq il Khaqani 4”

jawzabi Bi, 1 1<<

|j/ir,i 64, 1102, 114. 133

Jericho \i

Jerusalem 1 2.

144. 148, 199. *56. 272

jewels :n. 66, 177-8, 190.

195. -44. -;•

Jews 24. 43, B2, 175, 197-8

jihad 4s. 111–:. 1 S4. 19c

Jordan 1 B, 50. 33 ni-i he 1

Junaid 221

K.rb 1 99, /

Kabuhstan 210

Kaiqubad 114. 120

Kahla urn Duma 129. 131. [62-3, 222. 224, 1-4 61

k.ishan 92. 94. V s . 202, 246.

249-51; lustreware 202-3,

1 1


Klhim>j 239. 242

khan 29. 121. 123. 144, 170,

212. 230, 234 Khan Mirjan 213, 168 khanqjlt 113. 143. 146. 1 4H,


Khargird 108, 218

Kharraqan 92, 107, 82

Khirbat al-Mafjar 17, 31-2,

36. 17-18

Khurasan 49-51,94, 135, 204, 216, 220, 254

khutba 2 1

Khwanzmshahs 90, 108, 97 Kirman 92, 212, 233, 237, 249-50, 252

Konya 91, 1 14, 118, 123 Kose Dagh 1 14 Kubachi 221. 252-3, ioj Kut’.i 12, 41, 57, 59, 2

Kufic 20, 37, 50, 53, 55, 59-60, 67, 69, 91, 101-3,

131, 135, 165, 176-7,22,


Kurds i 12. 22N

lacquer 241 . 246

lamps 18, 64, 78, 137, 146,

149, 152. 156, 172, 190,

24S, 270, 279, 120, 210

1 Bazar 46. 108,


leather 132, 152, 223, 280

light symbolism 52, 57, 77-8, 146, [56, 172. 19s.


lustreware 50, 52-4, 60, 63,

68, 79-83, 90, 92, 95, 97,

99. 114. 123. 129, 134,

1 60, 189-92, 202-3, 220,

250, 252, J6-7, 58-60, 65,

-,\ 148-9, 157

Madinat al-Zahr.i 64, 82,

174-v l8l

madrasa 87, 108, 1 16-7,

I20-I, 124, 144. 166.

187-9, 201, 212-3, 217,

2 19, 230, 2}2-i, 260, 268

Maghrib 8, 44, 61, 74, 83, 167, 170, 176-7, 184-9,

magic 54, 99, 124, 156, 274,


M.luli 228

M.iluliv.i, tl- 61 , 64, 72

maidan 226, 230, 232-3, 178

Malaga 180, 191 -2

M.dik .il-Adil, al- 135

Malik al-Nasir Muhammad, al- 154, 157

Mahksh.ih 87, 92

mamluk 123, 138, 151, 1 56

Mamluks 26, 122, 132, 138-

44, 146, 148-53, 155-7,

159-64, 166, 197, 199,

204-5, 25C)

Manises 1 89, 191

Mansur, al- 178

Mansunya, al- 189, 146 maps 65, 273 Maqamat 129-32, 162, 175,

212, 273, 101-2, 124-5




Maqnzi. .il- 63, 68, 83


Maqsud Kashani 24(1, -49 nttqmn 1 72 Marag^ia v2. 1 5

marble 28, j6, 46, 65, 1 17,

122-3. ’43. ‘4 f ‘. l6l, 1 ^v.

2i(>. .

Maidin 115, 1 .;.;

Marinids 1 s ~\ 1 B9

Mairakeao 180, 18a, 185,

1 s-. Great Mosque 1 B4

Marwan II 17, 18, ; Mashhad 100, 1

ISO, 2S2

HMSfttj 1 —

Matrakci Nasufa 273

mausolea 2s. ~4 s 92, 101


105. 10S. 1 14. 1 if. B

123. 144. 1 jo, [70, i s v •

1 a, 210. 212. j 14.

217, 2iv. 260- 1 . 268

: 14, 65, 75,

78, 12V. 141. IVV. 2So. I

Medina 14. -4 s – 65, 199,

256, -‘

Mehmed II .: s s . :m. 2^4. 274, 277

Men 40. s.;. t)i. V4 196

Mesopotamia 14. 34. 4f>.

1 10-1. ii^. 128, 1

metalwork B, 17-18, 47, 4v.

SI. jJ . • 1 >-2,

I, 114. 134-7. I4O,

14V. I S2-f>. 1 –

if>2-3. 166, 169, 1

1 87, lOO, 2<> 4 . 216, 21 B,

220-1. 23s. 2 4 r>. 25; . 2– ‘. ioy

t. 117-9, if.’. 1

195, 195

mihrjh 21 . 24. s.


I46, i-2. 1-4. l8l

200-2, 270, 272, 275

2*5, J is

mtrui: K),

114. 12;. i2v. 1



105. 107 V IIV 1; 1 4<V UV. 171, I s

2 12. J

trnnhat 21 . s2. 1 4′..

Ming 162, 221

• .-214

mirror ;io.

1 !•/.

1 J7, [96 •>. 1 1 .212.

214. 210. 22.

moon 77, S 1 , .;j, 1 :(, 1 ; ;_ 1 s<>. 160, 244.

Mora 1 o 169, 1 s • mosaics 21-3

|6 7, 71, u fi . 1 M 8, 12-14, •*• ‘ ”•’ ‘

Mosul 1 1 s. 1 2 ; j 1 is f>. 1 nJ




Mu’awiya k>. 21

Mu’izz, il <<<>. ^\

llHKV/lll |Os. [96

Mughal 214 2ss. 26l, 268

Muhammad (Prophet n-12. 19, 21.2; j 4V M. “I

– 1 |9, 143, 18!

214. 22o. 2-2. 2

Muhammad, Sultan S Muhammad b il Zain is;.


Muhammad b. Qala’un 153, is-;

Muhammadj 2 ;v 40, i<Tj muhaqqaq sv. k><>. 20s. ./ ,-.


uniforms f>v. 116, 1 in. i 2s.

143. ’46. 148. |S

t- 195.

ivV 2 K. 7, 219, .’.”

233, 263

Murad III 273 musk jo- 1 .


122, I2v. 1 is.

142. 1 $3, 162, i-s. 17X.

I ‘. 248, -‘ f ‘s

Must.nisir. il ‘>*

Muzaftarids 212. i\(<

mystic ism 62, Bi, B7, 96 7,


N.ikln hivan 1

Nasii il I )in Muhammad I SI. I s4. 166

ruskh J7, $7, –

Nizam il-Mull N1/.11111


Nurida 1 1 ;


c Njeitu, tomb ol j


.11 ;

169, IS

1 taomam B, •

5 61,

l >\u

Pahl i\



189 9

T fHj. 10/

Parthia 14, 51


Pisa Ba, 108; Pisa < rriffin

pishtaq 101, 103

planets 122 \, ij

I s’>. /.’./. I 10

poetry to, |i, |

it >$, 189, IVs

214 | U ‘<

Poitien 12

pon eJain si



I 2 s. 141. U”- 149

1 86, 9 1


Portugal 191

potter] –

mi il I >m Mi



Qljmai il Kli.i.ji 1 4′

Qubadabad Qubadr)

Qubbal il Barudiy) /’

Qunun 1 Qui an

ss 6

1 J9 I :



Qusaii •\1111


Rabat is.v 184, 1*1

Rabat 1 Mahl K 1111 62 K.ii|i|.i

Rashid il I

Rasulka 1 si

Ray) ‘

Pvenaa 1

K10.1 m M’l 1 1


rock a


S.ili il I ‘



Saljuq 39, 60, 62-3, 70-1,

86-92, 94-5, 97. 99-103,

105-6, 108-15, 129, 170,

201, 204, 206, 218, 220,


Saljuqs ofRum 1 1 1 , 11 3-4, 1 16, 1 18, 120-3, 2 55

Samanids 39, 50, 54, 92,

108, 125, 131, 206

Samarqand 54-5, 76, 92.

108, 213-4, 217, 219

Samarra 39-40, 42-3, 45~7.

52, 58, 72, 76, 79, 101,

174-5, 189,25-7,5;

San’a’ 48

Sanjar 87, 97, 108

Sanskrit 131

Sasanians 14, 18-21, 34-6.

38, 45, 47-50, 56, 58, 60.

70, 87, 125, 133, 197.

199, 205, 234, 248

Sayyida Ruqayya 77 sculpture u, 14, 17, 32, 37,

47, 51, 64. 70, 90, 92.

118, 122-3. 134, 160,

179-80, 223, 233

Selimname 273

Senegal 182

Severus b. Muqaffa’ 5 1

Seville 171, 180. 1K4. 1 89

sgraffito 56, 98, 101, 162

shadow theatre 130. 132.


Shah Sultan Husain 230,

232-3, 180

Shah-i Zinda 217. 219. ; _ i-


Shahnama 70, 90. 98, r 14.

197, 209-10, 224. 23-.

239-40. 1, 165, 166-7, lAfi


Shahrukh 214. 224

Shaikh Safi 235

shamsa 66, 77 shari’a 197, 200, 256

Shi’ites 39, 55, 61-2. 76 B,

82, 87. 89, 92, 103, 111.

140, 197. 200. 202. 20-.

226, 235 Shiraz 214. 223-4. 242

Shirvan 250

silk 49-50. 34 Simnan 101

simurgh 20S-9

Sinan 264, 266-7, 270 Sinjar 123-4. 96

Sitt Zuhaida 116. 90 1 16

Siyar-i Nebi 272

slaves 39, 4-. 86. 120. 122.

124. 129. 13S. 140. 151.

2 26

Soghut 25-


Solomon 22. 24-5. 32, 195. [97.204 *

Spain B, 1-. 3 s -40- 4<‘>- 54.

59, 61. 63, 69. >>2. 125.

150, 157, 160. 167,

169-70. 1-2. 1-4–.

1-‘/ 18-93

sphinx 68, B5, 99, 180

stan —v 1 59, 190. 195, 202. 2 1 S: stellar designs

142. 146. 1 60-


steel 253-4. 2 No. ig>

Suhh 1 — Sublime Porta .

Sufi, al- 58, 100. 125

SufisDl s_ . 9″. I 13. -02.

225-6, 243. -53-4- 256,

270: see also uiysticimi

Sulaiman 13. 24

Suleyman the Magnificent

263. 27

Suleymaniye complex 261.

1 98

Suleymaniuune i~\

Sultan .1:

Sultamva 199-201.

Sunnism -2. B7, 111. 113. 19-. 22

Surname 2

Syria 1 1-14. 16-17, –

34. J6, ;v 41. 4^

B4, v ” BO-OO, 111-2. 114.

1 16. us. rao, 1:2. 125, 12-. [32, 139, I<4. 1″2.

169-72, 1-6. i-v 181,

[84, iv~.


Syria* 21 . 124. 1 37

Tabriz [62, [98

206–. 209, 216. jj .

Tahm.ivp. Shall 22<‘

Taj al-Mulk 92

Taj Mahal 199


Takht-i Sulaiman i<;~

ta liq 59, 4


talisman 1 56, 1 79, [9

Tashkent 21″

Taza 1 s ~

Tehran -ft. 100

textile • 4 6.

6, 100. [57, i –

190-3. [95, U ;-4. 216. 229. :v -2, 257, –

33-4, 63, 121.

139. 192, 200.-10

dntlth 3-. 59, 155. 1 v

166. 204. 22. 4}

Tirlis 94

tilework 101. 105, [08, 1 14.

1 16-7, 122-3. UO. U s – 162. 187-8, 190, 193. 201.

212. 214. 216–. 226. 232.

236. 246. 24^. 2S2. 265.

Timur 213-4. 216–. 220. 224. 226. 255

Timund B, 205. 212. 214. 216-v 220-1. 224-j

246, 2-2-3

Tinmal j

lira: 50. 63. N5. 132. 1 So

Tlemcen 185, [87, 142 Toledo 1 ft-. 1 No- 1. 138 Tomb of the Samamds 101.

tomb towers 91-2. 101.

tombstones 62, 76-7, 122

ropkapi Sara.

;. Library 100

rransoxiana 216

Tughnl I


Tulumd B Tunis, dreat Mosque 4s Tunisia 44. S4. 6|.

I ur.u

I UICOI 12-?.


Tursun 1

Lighur :

Ukhaidir 4 1 . 24 iilatita 1 i ]

Uhigh Beg 214. 217 ‘Umax I 21

Umayyad 12-14. 16- 1 7, 19.

Uzbek 2 . L’/un \ \

Valencia 1 77, I

Varamio 201, 15/t

1 14. –


69, 71

141, 14

216-S. If: », 264


in 143. 146. !

wall paintings 1 1. 28. 30,

32. 35-6,47. 67-72. U6. 199. 201. 206. 2 12-3. 233.

246. 26s. It. 31, 48-Sl

women 32, 4-. 62. 68, – i


77, 79. 97- 102. 105. 1 1-.

130. 134-5. 162. [75, 1—


193. 205, 240, 244. 278,

4. 19,31-2, 45, -0-1. -4.

94, goj, 105-6, 125, 1 >’-‘•

ito-i. 166-7, ‘.s’e-~. igi

woodblock prints -1-2. 52 woodwork B, 57, 63, ft5,

166. 1 SS. 190. 2 16.

235-6, 42, 182

World History 207-9, 214.

162, 164

wrestling 46, 79, [78

writing. Islamic 40-1. 43,

4″. 75-6, 1 1-. 1 19. 12-.

133, [69, [73.179. ‘ s —

205, 210, 22

Ya’qub, Sultan 224

Ya’qubi, al- 4

yasa 197, 200

Yazd 201 -2. 2 1


Yemen 4 s . 61 . 1 s 1 . 256,

Yuan 1 J7, 162


/.ingi 122. >)4

/.ingids 1 1 1 -2. [33

Zanjan 94


Zavara, I Mosque 105 /.iw.inids 187

/iggur.its 46

Ziryab $9 Zryaratgah 1

zodiac 68, 90, 1 S4

/oro.istn.inism 21.1

Wahd II, al- 3. 24-6




•^ WORLD OF ART Islamic Art and Architecture

Robert Hillenbrand. 270 illustrations, 80 in color

Embracing a thousand years of history and an area

stretching from the Atlantic to the borders of India and

China, Robert Hillenbrand – a world authority on

Islamic art and architecture – has written an unrivaled

new synthesis of the arts of Islamic civilization. From the

death of the Prophet Muhammad to the survival of the

Ottoman empire well into the modern age, Hillenbrand

traces the evolution of an extraordinary range of art forms,

including architecture, calligraphy book illumination,

painting, ceramics, glassware, textiles and metalwork.

Complete with maps and glossary, this is au accessible and

definitive guide to the arts o\\\ vastly accomplished


Thames and Hudson

On the cover: Detail from a late

i-th-eenrurv panel

of tiles in the harem.

Topkapi Sara\

Museum. Istanbul.

Printed m Slovenia


ISBN 0-500-20305-9

9 780500″203057′


Is this part of your assignment? Get trusted writers to serve you on on your task
Our experts will take care of your task no matter the deadline!
Use the following coupon

Order Now