Central Islamic Lands

Richard Ettinghausen et al.

From The Art and Architecture of Islam: 650-1250
© 1987 Yale University Press
Reproduced by permission of Yale University Press

INTRODUCTION

For reasons provided in the Prologue to Part II of this volume, the presentation of the medieval arts in central Islamic lands has been divided into two sections.

The first section deals with the rule of the Fatimid dynasty, which began in Ifriqiya (present-day Tunisia) around 908, moved its capital to Egypt in 969 under the leadership of the brilliant caliph al-Mu'izz, and ruled from there an area of shifting frontiers which, at its time of greatest expanse, extended from central Algeria to northern Syria, the middle Euphrates valley, and the holy places of Arabia. Its very diminished authority, affected by internal dissensions and by the Crusades, was eliminated by Saladin in 1171. The dynasties dependent on them vanished from North Africa by 1159, while Sicily had been conquered by the Normans in 1071.

The second section focuses on the art of the whole area in the twelfth and thirteenth centuries (at least until 1260), but only on its eastern part, essentially the Mesopotamian valley, in the eleventh. Several interlocking dynasties were involved in struggles and competitions which were as constant as they are difficult to describe and to recall. The lands of Iraq, the Jazira, Syria, Anatolia, Palestine, Egypt, Arabia, and Yemen were a mosaic of feudal rules enriched by the overall prosperity of the area, much involved in the elimination of the Crusaders' states, and largely committed to the revival of Sunnism and the destruction of what they considered to be a Shi'ite heresy. Although ideological opponents of the Fatimids, these feudal rulers shared with them both taste and material culture, and the visual distinctions between the arts of the two realms is not always easy to demonstrate.



PART I

The Fatimids In Egypt, Palestine, and Syria

The arts of this period of some 250 years are difficult to define on account of regional differences and of the growing complexity of Fatimid contacts with the rest of the Muslim world, the Christian West, Byzantium, and even India and China. The Fatimid era is North African, Egyptian, Syrian, and Arabian; but it is also Mediterranean and pan-Islamic.1

Politically, and in many ways culturally and artistically, Fatimid power and wealth were at their highest before the middle of the eleventh century. Shortly after 1050, however, in the middle of the long reign of the caliph al-Mustansir (1036-94), financial difficulties, famines, droughts, and social unrest led to two decades of internal confusion out of which order was not re-established until the 1070s. At the same time, in North Africa, an attempt by local Berber dynasties to shake off Shi'ite allegiance led to a new invasion by Arab tribesmen and to a thorough change of economic and political structure2, as Tunisia and western Algeria lost much of their agricultural wealth and entered by the twelfth century into a western rather than eastern Islamic and Mediterranean cultural sphere. During the last century of their existence the Fatimids controlled hardly anything but Egypt. Whether the major changes in Islamic art which they had earlier set in motion were the result of their own, Mediterranean, contacts with the classical tradition or of the upheavals which, especially in the eleventh century, affected the whole eastern Muslim world remains an open question.



ARCHITECTURE AND ARCHITECTURAL DECORATION: NORTH AFRICA

The Fatimids founded their first capital at Mahdiyya on the eastern coast of Tunisia.4 Not much has remained of its superb walls and gates or its artificial harbour, but surveys and early descriptions have allowed the reconstruction of a magnificent gate decorated on both sides with lions, of parts of the harbour, and of a long hall or covered street similar to those already found at Baghdad, Ukhaydir, and even Mshatta.5 The parts of the palace so far excavated6 have yielded two features of interest [286]. First, there was a curious entrance complex, consisting of a triple gate, its centre set out within a large rectangular tower. As one proceeds inwards, however, this gate ends in a blank wall. Two narrow halls on each side of the central axis lead into the court; the side entrances, on the other hand, proceed directly into the interior. The purpose of this odd arrangement could hardly be defensive; perhaps the four entries were to accommodate some of the extensive processions which, at least in later times, characterized Fatimid court life.7 Second, we cannot determine whether the decoration of some of the rooms with geometric floor mosaics sprang from memories of Umayyad palaces or imitated the many pre-Islamic mosaics of Tunisia.

A much restored mosque also remains from Fatimid Mahdiyya [287, 288].8 It was initially a rectangular hypostyle with a covered hall of prayer consisting of nine naves at right angles to the qibla. An axial nave led to a dome in front of the mihrab, and a portico in front of the covered hall served as a transition between open and covered areas and as part of a court with four porticoes. But the most significant novelty is the monumental façade, involving the whole of the north western wall of the mosque. It consists of two massive salients at each corner, which emphasize and control the limits of the building, and three symmetrically arranged gates, the central one set within another salient decorated with niches. This earliest known instance of a composed mosque façade gives a sense of unity not only to the outer wall but also to the building as a whole. Its origins should probably be sought in royal palace architecture, where such compositions were known as early as the Umayyad period.

From the second capital built by the Fatimids in North Africa, Sabra-al-Mansuriyya near Qayrawan, we know so far only of a very remarkable throne room which combines the eastern iwan with the characteristic western Islamic unit of two long halls at right angles to each other.9

The last two major monuments from North Africa to be attributed to the Fatimid cultural sphere are (if we except certain minor utilitarian structures) rarities in that geographical area. The first is the palace of Ashir, in central Algeria, where, under Fatimid patronage, the Zirid dynasty founded a capital around 947.10 It is a rectangle (72 by 40 metres) with towers of varying sizes [290]. The single outer gate of the complex is transformed into two entrances into the palace proper. On one side of the court is a portico. The presumed throne-room complex comprises a long hall with three entrances and a squarish room extending beyond the outer line of the wall and no doubt dominating the landscape. On each side of the central official unit, lining a courtyard, are two residential buildings consisting mostly of long halls. This symmetrical organization of living quarters around official areas recalls Mshatta or Qasr al-Hayr rather than the sprawling royal cities of Samarra and Madinat al-Zahra. Moreover, the palace is remarkable for its great simplicity: limited design, no columns, probably simple vaults, and very limited applied decoration. Though but a pale reflection of the architecture created on the Tunisian coast, Ashir is nevertheless precious for the completeness and preservation of its plan.

The second monument is the Qal'a of the Beni Hammad [291] in central Algeria, begun around 1007 by a Berber dynasty related by blood to the Zirids and also under the cultural impact of the Fatimid centres of Tunisia.11 It was a whole city with an immense royal compound comprising a huge tower with pavilions at the top [292], a complex of buildings crowded around a large (67 by 47 metres) artificial pool in which nautical spectacles took place, a bath, a mosque with a superb maqsura, and a series of individual houses and palaces. Neither the chronology nor the ceremonial or symbolic meaning of these buildings is clear; typologically, however, the Qala belongs to the succession of Samarra's or Madinat al-Zahra's sprawling ensembles, but with the emphasis on a setting for leisure and pleasure. A celebrated poem describing a lost palace of the eleventh century in Bijaya (Bougie) in present-day Algeria elaborates on this luxury and describes an imagery charged with heavenly and secular topics.12 It was possibly this ideal of luxury that inspired the twelfth-century architecture of the Norman kings of Sicily, about which more will be said in Chapter 8.

One group of fragments of unusual importance from the Qal'a of the Beni Hammad is a series of long ceramic parallelepipeds with grooves at one end; they must have projected unevenly from a ceiling or a cornice, looking like stalactites of a particularly unusual kind [452]13 . Other plaster fragments were certainly more typical muqarnas transitions. The origin and inspiration of these features is still unresolved. They could have been local inventions or, a more likely hypothesis, local interpretations of forms and ideas from the east, and prepared the way for the later vaults of Morocco and of Norman Sicily, especially the large ceiling of Roger II's Cappella Palatina at Palermo.14 Nearly all these examples occur in secular architecture, suggesting that the muqarnas may have been transmitted through the medium of private dwellings. But this matter also is not resolved, and we shall return to these very same fragments in Chapter 7.



EGYPT UNTIL 1060

After their conquest of Egypt in 969, the Fatimids embarked on a truly grandiose program of building, some of which survives; much more has been recorded by later Egyptian compilers such as the invaluable al-Maqrizi.15 These accounts in turn led, already before the first World War, to a series of very important, although not complete, topographical surveys by members of the French School in Cairo16, supplemented by a study of epigraphical material by Max van Berchem and Gaston Wiet17 and Creswell's monumental work. Both archaeological and interpretive concerns are ongoing activities in Egypt18 and, thus, for once in the study of Islamic art, both original documents and scholarly studies are numerous.

The centre of the Fatimid world was the imperial and military city of Cairo (al-Qahira, 'the triumphant').19 Nothing has remained of the first foundation, inaugurated in great pomp in the presence of astrologers with the purpose of controlling the older Muslim town of Fustat and its communications with the east. Yet its size is known (almost a square, about 1000 by 1150 metres), as are its north-south, almost straight axial street (the present Mu'izz al-Din street), its two huge palaces more or less in the middle of each side of the central street, with a wide open space for parades between them. It was provided with eight irregularly set gates (two on each side). Even the sites and names of the quarters assigned to the military groups permitted to share the city with the caliph have been recorded, because so much of the later topography and toponymy of Cairo is based on that of the town built between 969 and 973. Thereafter, little by little, the whole area to the south and southwest was transformed so that by the year 1000 Cairo, with the old city of Fustat, had become one of the largest and most cosmopolitan urban complexes of the medieval world, with its markets, mosques, streets, gardens, multistoreyed apartments, and private houses. Fatimid urban developments elsewhere are less well known, except for Jerusalem, Mecca, and, to a smaller degree, Ascalon on the Palestinian coast. In most of these instances, religious considerations dictated new constructions, but it is probably justified to believe that the establishment of Fatimid authority included transformations in the urban fabric of all the cities controlled by the dynasty.

The buildings of the early Fatimid period can be divided into three groups. Of the first, the palaces, nothing has remained, but the lengthy compilation of al-Maqrizi and the on-the-spot descriptions of Nasir-i Khosrow (1047) and al-Muqaddasi (985), as well as archaeological data brought together by Herz20, Ravaisse21, and Pauty22, allow for a few remarks about these palaces. Most remarkable was the Great Eastern Palace, whose main Golden Gate opening on the central square was surmounted by a pavilion from which the caliph watched crowds and parades.23 Inside, a complex succession of long halls led to the throne room, an iwan containing the sidilla; this was 'a construction closed on three sides, open on the fourth and covered by three domes; on the open side there was a sort of fenced opening known as a shubbak'. Painted scenes, probably of royal pastimes since we know that they included hunting scenes, constituted much of the decoration.24 For all its brilliance, the Eastern Palace seems to have been comparatively rambling in planning; the Western Palace (c.975-96, rebuilt after 1055) was smaller but more regular, centred on a long court with halls and pavilions on both axes.

Fatimid secular architecture can be characterized by two further features. The first (an apparent novelty in Islamic palaces) consisted in the royal pavilions spread all over the city and its suburbs.25 To these, for amusement or for ritual purposes, the caliph repaired in the ceremonial processions which brought the practices of the Fatimid court so close to those of Byzantium26 . Their shape is unknown, but most seem to have been set in gardens, often with pools and fountains, very much like the remaining twelfth-century constructions of the Normans in Sicily. The second feature is the layout of a number of private houses excavated in Fustat [293]. They abut each other in very irregular ways, and the streets on which they are found are often both narrow and crooked. But the interior arrangement of the larger ones is often quite regular and symmetrical. In almost all instances, open iwans or else two long halls at right angles to each other, or even both, surround a central court.27 The forms themselves often recall palatial ones, and the quality of most construction and the sophistication of the civil engineering are at times quite amazing.

There also remain from the early Fatimid period in Cairo two large congregational mosques. The celebrated al-Azhar ('the splendid') was founded together with the city to serve as its main place for ritual gathering. Because it grew slowly into a great centre of religious learning, it has undergone frequent alterations (the court façade, for example [294, 295, 296], is later, although still Fatimid). The original mosque can be reconstructed as a simple hypostyle (85 by 69 metres), with a prayer hall of five aisles parallel to the qibla wall and porticoes.28 The hall of prayer was bisected by a wide axial nave leading to a superb mihrab decorated with stucco [296]; in front of the mihrab was a dome, probably with two other domes framing it.29 The remaining dome now in front of the axial nave, built between 1130 and 1149, recalls, by its position, the one introduced in the mosque of Qayrawan. Throughout the mosque the supports were columns, single or double, often spoils from older and abandoned buildings. A great deal of decoration - mostly stucco - remains in the spandrels of the axial nave, on the qibla wall, and elsewhere. To its themes we shall return later; its position seemed to emphasize the main directions and lines of the building. The Fatimid exterior has not been preserved. Al-Maqrizi relates that there were royal pavilions and that a number of official ceremonies took place which were probably reflected in architecture. Without these accessories, the first Azhar Mosque appears almost as simple as the first hypostyles with axial naves known in Islam [297].

The second early Fatimid mosque in Cairo, the mosque of al-Hakim, redone and inaugurated by that caliph in 1013, was begun by his father in 990. Its original purpose is not evident, for it was outside the city walls to the north, in a sparsely populated area. Clarification is provided by a long passage in al-Maqrizi.30 Until 1266 (when it returned to the Azhar), the first and most splendid of a cycle of long and elaborate ceremonies of caliphal prayer, including the khutba or allegiance to the sovereign, took place here, to be followed on successive Fridays in Cairo's other large three mosques (early mosque of Amr, Ibn Tulun, al-Azhar). We may, then, interpret this building as an imperial foundation, whose primary function was to emphasize the religious and secular presence of the caliph and to serve as a setting for the ceremonial pageantry of the dynasty.31 While no doubt related to earlier mosques such as those of Damascus, Baghdad, and Cordoba, in all of which the nearby presence of the ruler played a part, the al-Hakim Mosque had a more restricted purpose as a royal sanctuary some way away from the city proper, not far from the mausoleums of the Fatimid family, and illustrating the very complex nature of Fatimid kingship. The private oratory in one of the minarets, and a possible mystical explanation of some of the decorative motifs like stars and a pentogram found on the masonry,32 lend credence to the idea of the building's special character.

The mosque of al-Hakim was a large and slightly irregular rectangle (121 by 131 metres) [298, 299]. At the west and south, on the corners of the façade, are two minarets now partly enclosed in later constructions, a feature obviously related to Mahdiyya's mosque. The minarets are remarkable for their decoration and for being of different shapes, one cylindrical, the other square. Between them in the main façade [300] is the monumental (15 by 6 metres) entrance; four more doors with flat arches and a very classical moulding complete the composition, and there is a further gate on each side of the mosque. The interior hypostyle combines features from the mosque of Ibn Tulun [25-27] (five-aisled sanctuary parallel to the qibla wall, large brick piers with engaged columns, single arcade on the other three sides) with innovations from North Africa (higher central nave, dome in front of the mihrab with two corner domes on squinches and drums). Thus, compared to the Azhar Mosque, al-Hakim's is much more carefully thought out, blending several architectural traditions and drawing especially on its North African roots. But it is still in most aspects traditional, and its most expressive features are the domes, whose outer appearance (square, octagon, cupola) is one step removed from the inside (octagon, drum, cupola), and the façade, whose symmetry is so curiously broken by the different shapes of the minarets which frame it.

Both the Azhar and Hakim mosques are remarkable for their architectural decoration, although themes and style differ considerably. At the Azhar the stucco panels on the wall spaces provided by the arches and the qibla aim, like those at Samarra, to cover the whole surface. The comparison is all the more justified since - except for the epigraphical borders rarely found in Iraqi palaces - the shapes of the panels, the motif of constant interplay of leaves and flowers around symmetrically arranged rigid stems, and the techniques of outlining, notching, and dotting are all certainly related to the art of Samarra, probably through the impact of the latter on Tulunid Egypt.33 However, the floral element is more pronounced and more clearly recognizable, and the background again plays an important part. After the Samarra-inspired experiments, therefore, the Azhar stuccoes seem to indicate a preference for an earlier and more natural treatment of vegetal motifs. The inscriptions of the Azhar have been chosen to proclaim the ideological bases of the Fatimid dynasty.

The decoration of the al-Hakim Mosque is quite different. Flat ornamental panels are rarer; when they exist, as on certain niche-heads of the entrance or on the windows of the domes, they consist of symmetrical designs of stems and leaves or of more complex arabesques, always set off by a visible background. Most of the decoration is of stone and is concentrated in a series of horizontal and (more rarely) vertical bands which emphasize the minarets and the gateway block. The designs include vegetal as well as geometric and epigraphical motifs, almost always in relief leaving the stone background visible. As already mentioned, it may be that some of the devices, such as pentagrams or the heavily decorated medallions which occasionally replace the horizontal bands, had a symbolic significance.34 There is no doubt that the inscriptions of the mosque were meant to proclaim an ideological message of caliphal power, and the striking originality and novelty of the al-Hakim example is that this message occurs on the outside of the building in direct and immediate contact with all the inhabitants of the city rather than being restricted to those permitted to pray inside.35 The al-Hakim decoration as a whole, however, is most notable, especially when compared to the Azhar, for its sobriety. Both the sobriety and the complex composition recall North Africa rather than the East, although it is possible that the ubiquitous classical background of the Mediterranean was wilfully employed by the Fatimids both in the simplicity of ornamentation and in the revival of more naturalistic themes of design.

A last point about early Fatimid decoration is that it was not limited to stucco or stone. Wood was common, although little has remained in situ. Mosaics were also used, which we know mostly from texts and from the superb decoration of the large dome in front of the mihrab of the Aqsa Mosque in Jerusalem (early eleventh century). The mosaics of the drum [301] probably copy much earlier Umayyad work,36 but those of the triumphal arch and of the pendentives are original Fatimid compositions [302], and their technical quality indicates that the older traditions picked up by the Muslim world in Byzantium were not yet totally lost or that, especially in Jerusalem, the Fatimids were reviving Late Antique techniques they knew as Umayyad.

The last group of monuments datable to the first Fatimid decades in Egypt consists of mausoleums, whose erection is attributable both to the Fatimid sense of imperial pomp and to their Shi'ite veneration of the descendants of Ali.37 The earliest royal and religious mausoleums are known through texts only, but two major early groups remain, a small one near Cairo,38 and another sixty-odd-strong in the great Aswan cemetery in southern Egypt [303].39 None is dated, but the indirect evidence of texts and certain details of construction indicate that they probably belong to the early decades of the eleventh century. By then the mausoleum was no longer either a royal prerogative or a place of religious commemoration, but a widely available form of conspicuous consumption. The social and pietistic conditions of the time suggest that the new patrons of architecture in this field were women and the middle class of merchants and artisans. There is, for instance, the very curious case of the Qarafa Mosque, sponsored by two noble women in 976 in the southern cemetery of Cairo; it shows that, quite early in Fatimid times, the place of the dead became a site for the expression of piety by another patronage than that of rulers.40

The mausoleums are simple squares with openings on one, two, or three sides. Built in brick or stone in mortar, or in combinations of the two methods, most of them probably had whitewashed walls with little decoration. All were covered with domes on simple squinches with an octagonal drum whose purpose was to give greater height and more light. Some of the mausoleums had over twelve windows, which emphasized their openness, perhaps in order to indicate that they were not full-fledged buildings and therefore did not entirely controvert the religious opposition to the expression of wealth or power after death.41 Their ultimate origin undoubtedly lies in the ancient mausoleums and canopy tombs of Syria and Anatolia, but how this form, which was rarely used in Christian and early Islamic times, came to be revived here in the tenth century is still unclear.

The early Fatimid period saw, then, not only the creation and growth of the new city of Cairo with its great congregational mosques but also some spectacular, if no longer available, secular and memorial building. Egypt was asserting itself as one of the great artistic and cultural centres of the Islamic world and a new and varied patronage for architecture came into being.



EGYPT: 1060-1171

Following the crisis of the middle of the eleventh century there was a marked change in the functions and plans of religious buildings in Cairo. Congregational mosques are few. Instead, the common building for prayers was the masjid, a small oratory, usually privately built and endowed, often with a specific commemorative or philanthropic purpose42 and without the city-wide significance of the first Fatimid mosques. Only much later did some of them acquire congregational status when they were assigned a special khatib or preacher to represent the state. In plan, the two remaining examples, the Aqmar Mosque, (1125)43 and that of al-Salih Tala'i (1160),44 are remarkable for their modest dimensions, their location within the urban fabric, and their external shape. The Al-Aqmar Mosque [304, 305] has a curious façade which is not parallel to the qibla; the mosque of al-Salih could be reached only by bridges, since it was built over shops. In both instances, previously existing streets and monuments determined the shape of the building, for each is on or near the main north-south artery of Cairo, where property was expensive, and the religious monument had to adapt itself to the more consistent fixed order of the urban community. The internal arrangement was not very original, unlike that of the vanished Mosque of the Elephants built by al-Afdal in 1085-86. Its peculiar name derived from the nine domes over the sanctuary, which were surrounded by balustrades and from afar looked like the howdahs carried by elephants in caliphal processions.45 Nine-dome mosques are known elsewhere,46 and belong to a rare type whose significance remains unexplained.

Commemorative structures also changed. In addition to mausoleums, sanctuaries usually called mashhads (literally 'places of witnessing'), such as those of Sayyidah Ruqqayah (1133) and Aswan (c.1100),47 began to appear for purposes of prayer, pilgrimage, and private piety. They include the still somewhat mysterious so-called mosque of al-Juyushi [306, 307] on the Moqattam hills overlooking Cairo, dramatically restored in recent years. The dedicatory inscription, which dates the building to 1085, clearly calls it a mashhad, yet its function as such is unclear; it does not seem to have been associated with a tomb, and Badr al-Jamali built it during his own lifetime. Qur'anic inscriptions suggest that it was erected in commemoration of some event which has remained unrecorded and which could simply be the reestablishment of peace and order after decades of strife and turmoil.48 The plans of most of these buildings are closely related. An entrance complex (domed at Aswan, topped by a minaret at al-Juyushi) leads to a small courtyard; the sanctuary has a large dome in front of the mihrab [308], always with a vaulted room on either side and usually with halls or rooms between it and the court (at al-Juyushi, three vaulted halls, one of which opened on the court through an ill-composed triple arcade; elsewhere all the rooms were covered with domes). The domes, like the zones of transition, were of brick, covered with plaster, and almost always four-centred in section; their surfaces varied from plain to ribbed. We have no record of how this kind of building was used, but we can say that religious practices must have changed significantly to justify the growth of this new type.

The importance of domes in these mausoleums and martyria explains the second major novelty of the later Fatmid period: a new mode of transition from square to dome. At al-Juyushi, an octagonal drum with eight windows surmounted a classical squinch. But already at Aswan by the turn of the century, and then in most other mausoleums, a muqarnas [309] replaced the squinch. The muqarnas is tripartite. A central niche bridges the corner framed by two sections of vault; above it is a sort of squinch vault approximately equal to the two sections on the lower level. Unlike contemporary Iranian examples, no arch enclosed the composition. The outline of the motif became standard for windows, so that the openings of the late Fatimid mausoleums are remarkable for their variety and complexity.49 The Egyptian and Iranian motifs are not alike; yet in purpose and basic tripartite composition they are closely related. The Iranian examples, however, are probably earlier, and their function, ambiguous though it may sometimes appear, is more clearly structural than in the smaller Cairene mausoleums. These points suggest that the Egyptian muqarnas squinch was inspired by Iran, but not blindly taken over (although awkward imitations existed); rather, it was adapted and scaled to the needs and possibilities of the more modest Egyptian monuments.51 It is, in fact, in Egypt that a muqarnas niche in plaster was discovered, just as they existed50 in Nishapur in Iran, with painted representations of a youth with a cup [310]. It was found in the ruins of a bath house.

The most spectacular remaining features of Cairene secular architecture are three gates to the new and enlarged city whose walls were redone, according to tradition, by three Christian architects from Edessa working under Badr al-Jamali [311].52 The towers of the Bab al-Nasr [312]53 are square, those of the Bab al-Futuh and the Bab Zuwaylah massive and semicircular. In his masterly analysis Creswell not only pointed out their close connection with the military architecture of the northern Mesopotamian region but also showed that they introduce to Cairo the use of stone as the sole mode of construction, a new type of pendentive, the stone cross-vault, and the round arch, all features of pre-Islamic Mediterranean civilizations which continued to be used in the upper Jazira and Armenia.

The architectural decoration of the second Fatmid period is less impressive than that found in contemporary Iran or in the earlier Fatimid period. Much - especially in mihrabs - consisted of the wooden panels discussed later in this chapter, but older techniques such as stucco (especially remarkable at al-Juyushi) and stone-carving (especially on the gates) were still employed. The designs, almost always subordinated to architectural lines, were both geometric (the backbone of the design) and vegetal. One of the more interesting compositions is the façade of the al-Aqmar Mosque [305], recently restored and in effect redone, in which a Romanesque effect of forceful projection of a religious monument into the city is produced by the central gate framed by two rows of niches, the false gates on the side, the upper band of decorative epigraphy, and the symmetrical arrangement on the walls of rectangular, circular, and rhomboid panels. Most striking are the peculiarly effective transformation of the conch motif in niche-heads, and the use of muqarnas as a flat decorative design. Both features occur also in mihrabs, and an investigation based in large part on the Quranic quotations of the mosque has proposed that all these motifs carried a Shi'ite message.53

Altogether, the later Fatimid period witnessed an extension of architectural patronage reflected in the growth of smaller monuments, the development of the mashhad, the use of the muqarnas in architecture and decoration, and a partial return to stone. Whether or not these features are of local origin is often still a delicate problem; but most of them are also characteristic of Muslim architecture elsewhere, strongly suggesting that, despite its heterodoxy, the Fatimid world fully partook of the pan-Islamic changes of the eleventh and twelfth centuries. In some cases, it is even possible to compare Fatimid architecture, especially in its second phase, with that of contemporary, more particularly western, Christianity.



THE ART OF THE OBJECT

The Fatimid period is of singular importance as the era when Egypt reached an outstanding position in the Muslim world, not only as the focal point of vast trading activities extending as far as Spain in the west and India in the east (as well as outside the Islamic regions) but also as a great manufacturing centre. The arts and crafts were so highly specialized during that epoch that it has been possible to establish no fewer than 210 different categories of artisans, compared to 150 in ancient Rome.54 Production for the lower and middle class was on a very large scale.55

Our most vivid and also most sumptuous picture of this period is provided by historical accounts, both contemporary and later, reporting on an event during the reign of al-Mustansir (r. 1036-94). In 1067-68 the great treasury of the Fatimids was ransacked when the troops rebelled and demanded to be paid. The stories of this plundering mention not only great quantities of pearls and jewels, crowns, swords, and other imperial accoutrements but also many objects in rare materials and of enormous size.56 Eighteen thousand pieces of rock crystal and cut glass were swiftly looted from the palace, and twice as many jewelled objects; also large numbers of gold and silver knives all richly set with jewels; valuable chess and backgammon pieces; various types of hand mirrors, skilfully decorated; six thousand perfume bottles in gilded silver; and so on. More specifically, we learn of enormous pieces of rock crystal inscribed with caliphal names; of gold animals encrusted with jewels and enamels; of a large golden palm tree; and even of a whole garden partially gilded and decorated with niello. There was also an immensely rich treasury of furniture, carpets, curtains, and wall coverings, many embroidered in gold, often with designs incorporating birds and quadrupeds, kings and their notables, and even a whole range of geographical vistas.57 Relatively few of these objects have survived,58 most of them very small; but the finest are impressive enough to lend substance to the vivid picture painted in the historical accounts of this vanished world of luxury.

There is no exception to the pattern we have been following now in our investigation, namely the cycle of adoption, adaptation and innovation, as regards the objects created for the Fatimids after their conquest of Egypt in 969. At first the artists working under the aegis of this dynasty seem to have continued to explore the possibilities inherent in forms long current in Egypt or more recently imported from the East. Only gradually do they seem to have introduced new decorative elements which had begun evolving in the western Islamic lands during the previous, Early Islamic, period under Umayyad, Abbasid, and indigenous influences. Once this innovative phase began, artistic problems were approached in an entirely new spirit.

As regards wood, treasured in Egypt because of its scarcity, early in the Fatimid period we can witness the continued popularity of the bevelled style first encountered in the Abbasid heartland and later in Tulunid Egypt [99]. The carved decoration on a tie-beam in the mosque of al-Hakim, dated 1003, is still based on the true Samarra Style C but it is also illustrative of a further development of that style in that the lines delineating the rather restricted number of motifs are wider, thus giving quite a different impression. Unlike the prototype, here the distinction between pattern and interstitial spaces is clearly defined.59 This feature is even more pronounced on the panels of a wooden door dated 1010, also inscribed to the caliph al-Hakim [313],60 where the individual bevelled patterns stand out clearly from a dark background. The major design elements are themselves decorated with small-scale surface patterns. The resulting textures, along with the contrast between light and dark, produce more varied, lively, and accented compositions than earlier on.61

By the third quarter of the eleventh century, however, a further evolution is discernible. The bevelled elements are reduced to thin, spiralling stems against a deeply carved background, and figural and animal designs begin to come to the fore. The early stages of this innovative trend are well illustrated by the panel [314]. Although the vegetal and figural designs can here be interpreted as being given equal treatment, the former motifs are beginning to be relegated to the background, and pride of place is moving toward the zoomorphized split palmette. Instead of starkly abstract, static, and purely sculptural qualities, there is now a dramatic interplay between abstract and more realistic parts, between elements conceived three-dimensionally and purely linear ones, and between light and shadow. In addition there is a new sense of movement.

This panel and another in the Museum of Islamic Art, Cairo, are closely related to those comprising a fragmentary door believed to have come from the Western Fatimid Palace, built by the caliph al-'Aziz and completely renovated by al-Mustansir in 1058.62 Destroyed by the Ayyubid conqueror Salah al-Din (Saladin) in the late twelfth century, it was bought in 1283 by the Mamluk sultan Qala'un, who then proceeded to build his great complex consisting of a school, mausoleum, and hospital. These were completed in fifteen months, and in this hasty project the amir in charge of the project took advantage of the already existing woodwork and other material on the site. Were it not for this medieval recycling, this beautiful panel and many others - some of which will be discussed presently - would probably have been lost to posterity.

Particularly important among these is a series of horizontally oriented carved wooden boards - some with decoration organized in interlaced cartouches containing designs of animals and human figures all carved against a background of formalized vine scrolls in lower relief [315] and others with a symmetrically arranged animal decoration [316].63 The horse protomes seen on the contemporary door panel discussed earlier, because their outline was made to conform to that of a split palmette, appear very stiff when compared to the liveliness of the varied motifs on these friezes and the realism conveyed by them. Human figures predominate now, and the rich repertory of subjects includes a number of male and female dancers portrayed in animated postures. In keeping with the new taste for scenes from everyday life, a woman peers out through the open curtains of a palanquin on the back of a camel, which is escorted by a man. In another compartment a drinking party is in progress. Two turbaned figures grasp goblets, one of them pouring from a bottle. From one side a servant approaches carrying a large vessel, presumably in order to replenish the bottle. Although the roughness of execution means that details are not as clear as those on similar representations in other media, traces of red and blue pigment suggest that specifics of facial features, costume, implements, etc. may have been precisely delineated in paint. Similar wood-carvings, more refined in workmanship but reflecting even more strongly the late Fatimid taste for observation from life, are to be found in a Christian context in Cairo, in the Coptic convent of Dayr al-Banat.64

Probably dating from about the same time is the fragmentary panel [317], decorated with a bird of prey attacking a hare, which must have originally functioned as one of the sides of a chest.65 This is a simpler but equally beautiful example of the technique known as marquetry that had a long life in Egypt [98]. Both the latter, more intricate, version of this art and the particular adaptation seen here were to be adopted later by the Almohads in al-Andalus and the Maghrib.66

Although the earliest extant datable woodwork with figural decoration from Muslim Egypt is from the third quarter of the eleventh century, architectural elements with such ornamentation were being utilized in Fatimid Ifriqiya more than one hundred years earlier. The fact that the capital Sabra al-Mansuriyya, founded in 947, contained buildings adorned with carved wood decorated with birds and stucco sculpture in human, bird, and animal form may indicate that early Fatimid structures in Cairo which no longer survive were similarly decorated.67 Thus, the vogue for carved wooden architectural elements with figural decoration may have been concurrent with that for the vegetal decoration that was evolving from the bevelled style.

Ivory carvings attributed to Fatimid craftsmen show close parallels in style and iconography to wood, but here the workmanship demonstrates the greater refinement appropriate to so expensive a material. The openwork plaque [318] that apparently once sheathed a casket or other small object repeats a long-common motif: the scarf dancer, skipping, her draperies swirling about her twisted body, her arms gesturing sinuously.68 Particularly noteworthy on this panel is the grace of the performer, her weight convincingly distributed, her headdress precisely knotted. The naturalism of all the figures on such plaques is heightened by the refined technique. Although there are two main levels of relief, as on the wooden boards, here the frames and figures are so delicately modelled that they appear fully rounded as if they were actually emerging from the vegetal scrolls that constitute the background.69 A device that contributes to this three-dimensional effect is undercutting. Furthermore, the leaves project forward from the vine scrolls in the background so that the two planes of the carving seem interconnected. These ivories are distinguished by the care lavished on detail, for example in the rendering of textile patterns, the texturing of animal fur and bird feathers, and the veining of leaves.

Because of their highly developed style, these ivories and comparable pieces in Berlin and Paris have been dated to the late eleventh or early twelfth century. However there appears to be no reason why they could not be contemporary in date with the carved wooden panels from the Western Fatimid Palace [315, 316], i.e. datable to c.1050. When the paint was intact on the latter decorative elements, these panels could have been as highly developed in detail as the ornament on the ivories.

The stylistic development we have been able to follow in the carved decoration of wood during the Fatimid period can be observed also in the ornamentation of lustre-painted pottery. Early in the period the designs adorning ceramic objects are often based on the bevelled style, but the motifs that took on sculptural qualities in carved wood had here to be rendered two-dimensionally. The earliest datable lustre-painted object so decorated is a fragmentary dish [319] bearing the name of a commander-in-chief of the caliph al-Hakim who held the title for only two years - November 1011 to November 1013.70 At this time as well, we know that this particular type of vegetal decoration was sometimes combined with figural designs. A fine example of such a transitional work is the bowl [320] signed by Muslim ibn al-Dahhan, a very productive artisan whose period of artistic output is known to us by means of a fragmentary dish in the Benaki Museum bearing an inscription stating that it was made by the above-named ceramist for a courtier of al-Hakim (r. 996-102171 ). The winged griffin in the centre of the bowl illustrated here is rendered basically in silhouette, but parts of its body structure are clarified and emphasized by keeping certain interior articulations free of the overglaze lustre paint. This attention to naturalistic detail represents a further departure from the caricature-like quality of the animals and human figures on the Iraqi monochrome lustre-painted pottery vessels of the Early Islamic period [107], as do the greater grace and innate movement in the griffin's body.

Although there are numerous lustre-painted bowls from Fatimid Egypt that bear figural designs as their principal decoration, there is only one such vessel known to us which can be securely dated [321], employed as a bacino in the Church of San Sisto in Pisa, Italy, dating to the last quarter of the eleventh century. We can be certain that the style of decoration exhibited on this bowl was current at this time since it must be assumed that this and the many other bowls from Egypt and elsewhere that once graced or still adorn the façades of Romanesque churches and/or campaniles in Italy were installed at the time of the construction of these buildings. Two other bacini, in one instance adorned with an animal and in another with calligraphic designs, can be used to date another group of lustre-painted pottery [322]. On the basis of the evidence provided by these two bacini, this group, which has long been associated with Fatimid Syria, more specifically with Tell Minis - a village in the central part of the country - and dated to the middle of the twelfth century, must now be placed in the second half of the eleventh.73

The firmly datable bowl [321], however, appears to exhibit a somewhat different figural style from that found on the majority of the extant lustre-painted bowls with such decoration from Fatimid Egypt which are ornamented in a style closer to that found earlier at Samarra and in Ifriqiya. A dated or datable example of this latter category would be necessary before we could ascertain whether this type was contemporary with that exemplified by the bacino, or whether it preceded or succeeded that production.

Although the rendering of most of the faces and the coiffures on these so far not clearly datable vessels betrays descent from the Abbasid figural style - especially the large, round face, the staring eyes and small mouth, as well as the side curls - the animation of the body and exaggerated gestures of the limbs are illustrative, however, of an approach quite different from the frozen monumentality of even the most active figures in the wall paintings from Samarra [84]. As was the case vis-a-vis Fatimid woodwork from Egypt, the influence of the artists working under the aegis of this dynasty in Ifriqiya must be seen as an important inspiration for the new trends that can be documented in Egyptian pottery, foremost among them being an intensified interest in naturalistic representation of the human figure, which was always greater in the areas bordering the Mediterranean than in the eastern parts of the Islamic world.74

Whenever this innovation occurred on Egyptian ceramics, the craftsmen of this undated and so far not datable pottery group managed, by means of a number of devices, to achieve naturalistic effects quite far removed from the two-dimenslonal stylization of Mesopotamian lustre-painted designs, and even from the rather static vegetal and animal motifs [319, 320]. Among these devices were the use of an energetic line, off-balance poses, and dramatic gestures to convey a sense of movement and animation. In addition [323], greater attention was devoted to realistic details of costume, jewellery, and vessels. Furthermore, the ceramist of this bowl managed to accentuate the fullness of the arms, the grip of the fingers, and even the dissipation of the eyes.75 This group also explored the episodic nature of a theme, a convention we have already seen in the tile from Sabra al-Mansuriyya [141]; instead of human figures and animals presented singly or serially, some bowls in this category illustrate groups engaged in particular activities [324]. Here a lady with two female attendants reclines on a couch and the main protagonist seems to be taking up her lute or relinquishing it to the servant. In contrast to the ceremonious quality of courtly scenes on Spanish ivory boxes [145], the Fatimid pottery examples have the informal flavour of an event observed from daily life.76

Although the pottery decorated with lustre-painting was the most luxurious of the kiln production of Fatimid Egypt, it was not the only ceramic type manufactured during this period. The bowl [325] belongs to a type of pottery known as champlevé that until very recently was generally dated to the late twelfth or early thirteenth century and usually attributed to Iran. However, following the excavation of ten vessels of this type from a shipwreck in Serce Limani, a small natural harbour on the southern Turkish coast just opposite Rhodes, we can now confidently date this category to c.1025 and place its production either in Fustat, Egypt, or in a manufacturing centre somewhere in the Fatimids' Syrian province.77 The decoration of such wares was created by first applying a slip of light-coloured clay to the interior and part of the exterior surfaces. When dry, the slip was partially carved away to leave the desired design in relief. Details were then incised in the slip and the vessel was finally covered with a transparent, clear or coloured, lead glaze.78

The second type of glazed pottery found in this shipwreck was a variety of splash-decorated ware. It, too, was previously vaguely dated - in this case as early as the ninth to as late as the twelfth century. Thanks to this chance find, at least wares in this category with similar designs and shapes can now be given a secure time frame as well as place in the history of Islamic pottery.79 The vase [326] seems not only to be a variant of this splash-painted type but also to be representative of the Egyptian version of a category that was so popular in the western Islamic lands during the Early Islamic period. This was the type that imitated the opaque white-glazed group manufactured in Basra, Iraq, under the Abbasids [141, 142].80

The potters working in Egypt and Syria during this period also produced monochrome glazed carved and/or incised ceramics, decorative techniques previously met with on pottery produced during the Early Islamic period [102]. The Egyptian version [327] is much closer to its boldly deorated antecedent than are representatives of the group made in Syria and associated with Tell Minis.81 The latter appears to exhibit for the most part a more delicate decorative style which seems to lead directly into that produced slightly later in Iran [265].82 The Tell Minis carved and/or incised category is datable to the middle of the eleventh century by means of a bacino.83 Since the so-called lakabi ('glazed') type of pottery [328] shares not only principal decorative motifs but body profiles as well as a peculiar rim design with the Tell Minis variety, this group must be attributed to the same period in Syria as well.84

Near the end of the first millennium or at the very beginning of the second, Islamic glassmakers in the central Islamic lands85 inaugurated a second period of innovation86 that brought them increasingly further from Roman imperial glass and culminated in superb relief-cut vessels. Without question, the most beautiful Islamic object of this type extant is the cameo-glass ewer [329]. The artisan initially formed a clear-colourless glass blank around which was blown a gather (or viscous and extremely ductile melted batch of ingredients) of turquoise-blue glass. The surface was then selectively cut away with a wheel, leaving the design in relief, with the highest point of the decoration representing the original, in this case turquoise-blue, surface. Considering the difficulty of working with such thin and brittle material, high-relief glass cutting is a remarkable achievement and often - as here - a real tour de force87 . Unlike its highly creative technique, a commonly found earlier shape was adopted for this vessel [92].

Another fine and famous example of the relief-cut technique is the bowl [330] executed in opaque turquoise-coloured glass and obviously meant to imitate a bowl carved from a mineral88, further supporting the hypothesis that the flowering of the craft of cut glass - especially relief-cut glass - was an offshoot of the technique for working precious or semi-precious stones, be they turquoise, emerald, or rock crystal. The close relationship between cut glass and cut stone, especially rock crystal, had been fully understood by medieval Muslims, for they are repeatedly listed together in reports of the Fatimid treasures in Cairo; and, we are told by a medieval Iranian author that Syria and the Maghrib were known for a type of green glass used to imitate emeralds.89

The ewer [331], bearing the name of the early Fatimid caliph al-'Aziz (r. 975-96) and exemplifying the finest quality of workmanship possible at the time, belongs to a group of highly important rock-crystal objects, several of which are firmly datable90 . This vessel and several others in the group share not only the same traditional shape with the cameoglass vessel just discussed but also many iconographic and stylistic features. However, whether the relief-cut glass objects led up to, were contemporary with, or were made in imitation of the rock-crystal ones is yet to be determined91 . In addition to this tour de force, the five other very similar extant pouring vessels92, and a small group of objects of comparable calibre in other shapes, there are a good many non-epigraphic and non-figural rock-crystal objects with decoration in the bevelled style. Were these also made near the turn of the millennium for a more conservative clientele? Or was there an extended earlier development in this difficult medium leading to the accomplished style of the inscribed works - a hypothesis supported by some pieces decorated in the Samarra Style C, which reached European church treasuries in the period from 973 to 982?93 An overlapping of styles is possible, but it is more likely that a slow, unilateral growth led up to the climax of this art in the eleventh century, a development following fairly closely that which we have already outlined in this chapter vis-a-vis the decoration of wood, ivory, and lustre-painted ceramics.94

Very shortly after the technique of wheel cutting reached its Islamic zenith in the relief-cut glass just discussed, its gradual simplification began. It was not long before a totally bevel-cut decoration often with no foreground or background had evolved, a stage beautifully exemplified by the vessel [332], decorated with finely executed lions, which was found in the Serçe Limani shipwreck previously mentioned in connection with the champlevé bowl [325] and is, therefore, firmly datable to the first half of the eleventh century.

The simplification of this lapidary technique as applied to glass was to reach its logical conclusion in totally plain but beautiful vessels with bodies faceted like gemstones.96

In addition to experimenting with wheel-cutting techniques, which could be employed after the glass had cooled, the glassmakers in the Medieval period continued to adapt techniques applicable only to a hot gather (viscous and extremely ductile portion of molten glass 'gathered' for use in glass-blowing) or parison (glass bubble): mould blowing and thread or coil trailing. The latter decorative device is beautifully exemplified in the Fatimid period by the cup [333] also excavated at Serçe Limani. The less time-consuming technique of thread trailing employed here in a boldly contrasting colour to set off the rim of the drinking vessel was often used at this time as well to imitate relief-cut designs.97

Glass products and glassmakers themselves moved from country to country. Documents from the Cairo Geniza mention that in the eleventh and twelfth centuries glassmakers from Greater Syria, fleeing the almost permanent state of war there, came to Egypt in such masses that they were competing with local artisans. Further, in a document dated 1011 it is noted that thirty-seven bales of glass were sent from Tyre, presumably to Egypt. Such emigrations and importations make precise attributions risky and international styles more likely. However, it is generally assumed that the ill-fated ship that sank in Serçe Limani took on its cargo at a port at the eastern end of the Mediterranean, thus indicating a Greater Syrian provenance for its contents.

The vessel [334] continues the tradition of lustre-painting on glass that we first met with in Egypt during the early Abbasid period [110]98 . While considerably simpler than the decoration found on Fatimid lustre-painted pottery, the style of the rinceau and the convention of setting off the ornamented bands with double (or single) plain lustre fillets, not to mention the shape of the vessel itself, are all familiar elements in the repertoire of the period.

Ample evidence for the importance of textiles during the Fatimid period is provided by the detailed descriptions of the dispersal of the imperial treasury in 1067-68 as well as by reports of contemporary geographers. These invaluable texts inform us as to the quantities and diverse origins of the numerous types of textiles being stored in various areas of the palace at the time of the catastrophe and the different types of textiles being woven in various parts of the Fatimid realm. We learn not only that this dynasty imported stuffs of many different kinds from al-Andalus, Mesopotamia, and Persia as well as from Byzantium, but that locally made products were also very highly valued both within and outside Egypt99 . The tiraz [335] was probably the back of an over-garment similar to modern 'abaya and belongs to a rare and deluxe group of Fatimid textiles datable to the reign of the caliph al-Mustali (r. 1094-1101) and to the factories of Damietta in the Delta100 . The decorative bands and ornamental roundels are tapestry woven in coloured silks and gold file (silk core wrapped with a gold wire) on a fine linen. This group has been associated with a type of textile called qasab described in 1047, by the medieval Persian traveller Nasir-i Khosrow, as being woven in Tinnis and Damietta for the sole use of the ruler.101 While the gazelles and prancing sphinxes reflect a figural style with which we have become familiar on other objects of this period executed in many different media, the layout as well as the style of the garment itself were adopted from the fashion of the Copts in pre-Islamic Egypt.

For the most part earlier and considerably more plentiful than the deluxe group just discussed were the group of Fatimid textiles adorned solely with epigraphic and narrow decorative bands [336]. This veil is particularly sumptuous and not only bears the name of the caliph al-'Aziz and the date of 373/983-84 but also informs us that it was made in the private tiraz in Tinnis. Its epigraphic ornamentation shows a continuation of the style begun under the last Abbasid ruler of Egypt, al-Muti' (r. 946-74), in which the hastae (vertical stems) of the letters of the large silk tapestry woven inscription bands end in very graceful half palmettes102 . These tirazes, like the group from the period of al-Musta'li just discussed, closely followed not only the layout and content of the decoration found on the textiles produced in pre-Fatimid Egypt but the styles of the earlier garments as well.

Growing out of a long tradition established during the Umayyad and Abbasid periods [100, 150], the vogue for small and large copper-alloy animal sculpture persisted in Egypt and the Maghrib at least until the end of the eleventh century. Representing griffins, stags, gazelles, lions, rabbits, eagles, and other types of birds, they were used as aquamaniles, incense burners, fountain spouts, padlocks, and possibly vessel supports, and they share not only a high degree of stylization, which, however, never impairs effective recognition of the subject, but also such secondary features as frequent all-over decoration and zoomorphic handles.103 The most famous as well as the most beautiful and monumental example of this tradition in the central Islamic lands is undoubtedly the celebrated so-called Pisa griffin [337], the immediate precursor of which is a quadruped from Ifriqiya.104 On this copper-alloy object (the orginal function of which is unknown) everything is formalized: not only the body and its parts but also the engraved decoration, which consists of roundels, inscriptions, and designs of small animals - none of them detracting, however, from the grandiose impression that this object, more than one metre high, makes on the beholder. Made under the Fatimid aegis most probably during the eleventh century, this object could very well have been part of the large booty taken by the Pisans after their successful invasion of the Zirid capital, Mahdia, in the summer of 1087.105

It is not yet possible to assign precise dates to these sculptures, and therefore it is not known whether or not the small, more realistic, copper-alloy hare [338] - which most probably served as a fountain spout - was contemporary with or made before or after the griffin.106 Whatever the style, the Fatimid works are impressive as animal sculptures. Furthermore, they seem to have served as prototypes for Romanesque pieces.107

Although we are informed that the Fatimid treasury contained silver articles with niello decoration, until recently we were at a loss as to the appearance of any of these items as none of them seemed to have survived. The box [339], therefore, bearing the name of a vizier of al-Mustansir who served only for three years - 1044-47 fills an important gap. As Geniza documents support the idea that large quantities of silver vessels were exported to the Maghrib and India from Egypt in the Medieval period, we can assume that this small container - most probably used as a box to keep jewels - was made in that country.108

Not only did Fatimid craftsmen excel in the making of objects of fine silver, as can be judged from contemporary sources and the above-mentioned object, but their goldsmiths' work was of the highest quality as well. The elements comprising the necklace [340], especially the biconical and two spherical beads near the centre of the ensemble that are totally fabricated from gold wire and decorated with granulation, were of a type known to have been executed during the first half of the eleventh century in either Greater Syria or Egypt and may very well have been of the variety described by the eleventh-century author Ibn Zubayr as 'unusual, very beautifully fashioned gold jewellery' that was sent to the Byzantine king Romanos I Diogenes in 1071.109

It has been suggested that the new imagery with its animation and fully realized observation of the details of everyday living that we have seen especially in the ornamentation of wood, ivory, and lustre-painted ceramic objects during the Fatimid period reflected developments that occurred first in painting.110 Unfortunately, only a few fragments of wall and ceiling painting and not many more drawings on paper survive from this era, none of which is dated or datable [341-3]. Therefore, it is impossible at this juncture of our knowledge to prove or disprove this suggestion.111



THE ART OF THE BOOK

As to the drawings on paper from this period, the tattooed female figure [341] is perhaps the most accomplished picture that has come down to us. It shares with the undated and so far undatable lustre-painted pottery group, discussed above [323, 324], the new Fatimid imagery, exhibiting greater animation and interest in the naturalistic representation of the human figure. However, the rendering of the face and the coiffure still betrays a dependence on the figural style at the temporary Abbasid capital at Samarra.

Also exhibiting the new trends is the drawing [342]. This bears very close comparison to the decoration on tiles from Sabra al-Mansuriyya [141]. Such similarity provides proof of the important inspiration for the new imagery to be found in the output of the artists working under the aegis of the Fatimid dynasty in Ifriqiya.

The manner in which animals are depicted in this medium is no exception to the new sylistic trends we have been following from this period not only in the art of the book but also in that of the object. The hare [343] with its heavy body, long and large ears, snub nose, and short tail closely resembles those seen earlier in this chapter executed in the marquetry technique and in metal [317, 338]. However, this illustration shows an animal drawn even more realistically - owing, perhaps, to the medium checking his flank before hopping along. The verso of this folio is adorned with a lion.112

Other than the few drawings on paper, virtually no arts of the book produced in Egypt during the Fatimid period have up to now been identified. Because of the heightened interest in the human figure during this period to which so many of the decorative arts bear witness, it is difficult to imagine that the art of miniature painting was not highly developed. However, apparently no illustrated manuscript or fragment of one has survived. In fact, none has even been attributed to this period. We learn from the eleventh-century report by Ibn Zubayr of the dispersal of the Fatimid treasury during the reign of al-Mustansir that

the number of libraries (within the palace) was forty, including 18,000 books on ancient sciences. The books included also 2,400 complete copies of the Qur'an [kept] in Qur'an boxes. They were written in well-proportioned calligraphy of the highest beauty, and adorned with gold, silver, and other [colours]. This was besides [the books] kept in the libraries of Dar al-'Ilm in Cairo.113

None of these appears to be extant.114 The meagre knowledge we have of the arts of the book in the Fatimid realm, other than that found in texts, is that gleaned from those manuscripts produced under the aegis of this dynasty in Sicily [154] or under that of their governors in Ifriqiya [471]. This total lack of Fatimid Egyptian manuscripts has never been satisfactorily explained. The fact that none has been positively, or even seriously, identified after more than eight hundred years might indicate that all of them, even the Qur'an manuscripts, were methodically destroyed in the Sunni revival that followed the fall of this heterodox dynasty - the fulfilling of a duty to extirpate heresies and reinstate true orthodoxy and thus part of the systematic attempt at reeducation undertaken by the Sunnis. The solution to this puzzle has so far not presented itself.



CONCLUSION

The most striking feature of the arts under Fatimid rule was the establishment of Egypt, and more particularly of the newly created city of Cairo, as a major centre for artistic activities. The latter involved the construction of many buildings, their decoration in many techniques, the establishment of a brilliant art of lustre-painting ceramics and glass, carving ivory, rock crystal or wood, chasing and engraving (but apparently not inlaying) metalwork, and an elaborate art of textile weaving. Cairo also became a consumer for foreign goods, silks and ceramics from China, gifts from Christian rulers farther north. Expensive curios from many places were brought to the city as parts of an extremely active international trade in items that must have been considered works of art. All these things were available to a wealthy middle class best known through its Jewish component which left so many documents of private and professional life.115 Or else they were kept in an imperial treasury whose variety is demonstrated by the lists made after the looting in the middle of the eleventh century.116 Some of these sources even imply the existence of an art market, where new and old objects, sometimes outright frauds like the saddles attributed to Alexander the Great, were peddled by otherwise unknown dealers.117 Cairo became a major employer of artisans and technicians, and it is, for instance, to the importation of stone-cutters from Armenia and northern Mesopotamia that some of the novelties in late eleventh-century construction techniques have been attributed.

But beyond these economic and technical considerations, the detailed evolution of which still needs investigation, a more important and particularly original feature of the arts under the Fatimids is the blurring of the boundaries between public and private art. Many of the new artistic developments, especially the buildings in the city of Cairo but also the lustre ceramics, were made to publicize and to display power, ideology, wealth, taste, or whatever else a patron or an owner wished to proclaim. This novelty is particularly visible in the importance assumed by inscriptions, the 'public text' identified by I. Bierman,118 on the outside of buildings, by individualized images on ceramics, and in the colourful restoration of great sanctuaries like those of Jerusalem.119 Nasir-i Khosrow, the Persian traveller and propagandist for the Fatimids, was allowed to visit the imperial palaces in Cairo and described at great length their elaborate decoration.120

While it is easy enough to demonstrate the artistic vitality of Fatimid Cairo and some of the social and ideological functions of individual monuments or objects, it is much more difficult to identify and explain the characteristics of the art itself. Three of these may help to define the paradox of Fatimid art.121 One is the possibility of demonstrating a progression away from the dry and severe formalism of ninth-century vegetal decoration, as in the stucco ornament of the mosque of Ibn Tulun in Cairo and in many related pieces of woodwork, to a much more lively arabesque with highly naturalistic features in the eleventh century, and, eventually, in the twelfth, to an elaborate geometry with its own formalism. Whether an evolution which is apparent in woodwork and stucco ornament is true for all media remains to be seen. A second characteristic is the frequent appearance of representations of people and animals in almost all media. Sometimes hidden in vegetal ornament, animals and personages also appear as the motifs decorating muqarnas niches and lustre-painted ceramics. The motifs represented in the latter are quite varied both in style and in originality, but the essential point is that their range goes from traditional scenes of royal pastimes to very lively images of daily life. Stretching a point slightly, R. Ettinghausen even talked of 'realism' in this Fatimid art.122 It is unfortunate that we are not yet able to date accurately the appearance of these representations, but there seems to be little doubt that it preceded by almost a century the same phenomenon in Iraq, in Syria, and especially in the eastern Islamic provinces. It could be connected to a renewed awareness of Late Antique art and, in general, to the artistic explosion of the whole Mediterranean area in the eleventh century rather than to some uniquely Islamic developments, but the matter still requires further reflection.123 And, finally, the art of the Fatimids reflected and satisfied the needs of a stratified society. It is, at times, difficult to decide whether a given object, or even a building, should be attributed to a royal, aristocratic, or middle-class patron or user, or whether he or she was a Christian, a Muslim, or a Jew. But all these possibilities are open.

Thus, and therein lies the paradox, the art of the Fatimids is more difficult to explain than to describe or to define. Should it indeed be considered a Mediterranean art which may have picked up certain features from eastern Islamic lands, but which developed largely independently within a different context of civilization? Or was it the precursor and even possibly the inventor of changes, like the appearance of representations or the growth of a public art, which were soon to become common? There are as yet no answers to these questions which illustrate the second paradox of Fatimid art within the broad context of medieval Islamic art. It exhibited an aesthetic vitality which seems absent from the rest of the Islamic world during the same period. Is it an accident? Does it have something to do with the doctrines of Isma'ili Shi'ism and the ways in which they were applied to the rich and complex society of Egypt and of the provinces, like Ifriqiya or Syria and Palestine, under its domination in the eleventh century? Or, perhaps, Fatimid art and culture were an original phenomenon hatched in tenth-century Ifriqiya by a brilliant leadership around the caliph al-Mu'izz, which would have created its own synthesis of Islamic doctrines and practices with Mediterranean art and culture.124



PART II

The Saljuqs, Artuqids, Zangids, and Ayyubids in Iraq, Anatolia, Syria, Palestine, and Egypt: Architecture and Architectural Decoration

In contrast to early Islamic times, it is not possible, at the present stage of research and interpretation, to provide a single, continuous, chronological account of medieval (eleventh to thirteenth centuries) architecture in the central Islamic lands which were not under Fatimid rule. Two approaches could be proposed. One is dynastic and political; it would identify monuments and architectural characteristics according to definable areas of shared power and culture. Seljuq rule in Anatolia or Ayyubid control of Egypt and the Levant led to an architecture with recognizable forms of its own. But it is difficult to identify an independent set of forms associated with the Zangids of the Jazira and Syria, the Artuqids in northern Jazira, or Abbasid rule in Baghdad. Therefore, we have preferred a second approach, which is to present these lands in terms of four geographical regions with partly interlocking dynastic histories: Iraq; Jazira; Syria with Palestine and Egypt as well as a brief foray into Yemen; Anatolia. Chronological sequence will suffer no doubt, but it is possible to argue that, during a politically complicated period such as this one, architectural consistency lies in lands rather than in rulers.



IRAQ

The history and development of Iraq, defined here in the medieval sense as the lower part of the Tigris and Euphrates valleys, were somewhat overshadowed by the momentous events taking place in Anatolia, Palestine, and Syria. Nevertheless, the orthodox Abbasid caliph resided in Baghdad and in spiritual and intellectual power the city was as great as ever. Nizam al-Mulk, the celebrated vizier and ideological guide of Saljuq rulers, founded his most important madrasa there. The tombs of Ali and Husayn, the tragic heroes of Shi'ism, and of the great founders of Sunni schools of jurisprudence were in Baghdad, Kufa, Kerbela, and Najaf; they all became centres of large religious establishments for pilgrims and other wayfarers; their impact was as wide as the Islamic world.125 The port city of Basra in the extreme south was still one of the major Muslim gates to the Indian Ocean, and travellers such as the Muslim Ibn Jubayr and the Jew Benjamin of Tudela continued to be struck by the wealth and importance of most Iraqi cities. Iraq's political significance, on the other hand, had shrunk, to revive briefly in the first decades of the thirteenth century, when the caliphs al-Nasir and al-Mustansir asserted themselves as more than figureheads.126 In 1258 the last caliph was killed by the Mongols and the city sacked.

Few monuments survive from this period. Of those mentioned in texts, mostly chronicles,127 many were reconstructions of, or comparatively minor additions to, older buildings. The vast complex of the Mustansiriya in Baghdad, inaugurated in 1233 by the caliph al-Mustansir,128 stands out both ideologically and architecturally [344, 345, 410]. The first recorded madrasa built for all four Islamic schools of jurisprudence, it reflected the idea of the caliphate as the sponsor of an ecumenical Sunnism, a strong feature of the new guidance that the later Abbasids attempted to provide. Built along the Tigris, the Mustansiriya is a huge rectangle (106 by 48 metres) with a large central court (62 by 26 metres). It had three iwans opening on the court, one of which served as an oratory. Between iwans and oratory lay long halls at right angles to the court, and various other halls and rooms extended to the north and south, probably equally divided between the four official schools of Islamic law, according to Sunni practice.

Extensive reconstructions and long use of the Mustansiriya as a customs house have greatly altered the internal aspect of the building, but two points about it are of some significance. First, although, with its two superposed arches in rows symmetrically arranged on either side of larger single arches, it was clearly influenced by the Iranian court with four iwans, it differs in that one of the iwans was transformed into an oratory whose function separated it from the rest of the building. This function was emphasized by a triple entrance on the qibla side of the court [346], balanced by a totally artificial triple façade on the opposite (entrance) side. Thus the openings on the court do not correspond with the same clarity as in Iran to the purposes and forms of the covered parts behind them; from being a meaningful screen, the court façade has become a mask or a veil, perhaps aiming to provide an eastern Islamic effect to a Baghdadi building and being, thus, the expression of social taste. Second, the ratio between length and width, the multiplication of long vaulted halls, and the peculiar separateness of the oratory are all anomalous features, at least from the point of view of an Iranian model. They could be due to the location of the Mustansiriya in a bazaar area where previous constructions and all urban rhythm of life imposed peculiar forms on the new buildings; on the other hand they may derive from earlier developments in Iraq in the eleventh and twelfth centuries of which as yet we have no knowledge. In any case the originality of the ecumenical function of the Mustansiriya is indubitable.

A second Baghdadi monument, probably of the twelfth century,129 the minaret in the Suq al-Ghazi [347], uses the brick technique and decoration of Iranian minarets. Stucco decoration on one side shows that it was once attached to some larger construction.

The mausoleums erected in Iraq during this period are particularly novel and original among the funerary buildings seen so far. Many of the most celebrated, such as those of al-Hanafi and the Shi'ite ones in Kufa, Najaf, and Kerbela, either have been altered beyond recognition of their early appearance or have never been available for scientific analysis. Two less holy ones which have been studied are the Imam Dur (or Dawr), near Samarra, datable around 1085 [348, 349], and the so-called tomb of Zubayda in Baghdad, datable around 1152, and very much restored in recent times.130 In both instances the individuals originally commemorated in these mausoleums are unknown. Their plan is quite simple: a polygon or a square covered with a dome. The curious development occurs in the dome: over an octagon, five (at Imam Dur) or nine (in Baghdad) rows of niches lead up to a very small cupola. The dome has been transformed into a sort of muqarnas cone. In detail the two monuments vary considerably: the Baghdad one is drier and more obviously logical in its construction than Imam Dur, where there is a much more complex combination of geometric forms, particularly inside. In both domes, however, great height was achieved through a geometrically conceived multiplication of single three-dimensional units of architectural origin. This type remained quite popular in the Persian Gulf,131 and its impact elsewhere has been well documented. Although the full documentation of the point is difficult to make, recent and forthcoming works are very suggestive in proposing that the type identifies the Baghdadi version, if not invention, of the muqarnas as an ideological statement of Sunnism.

Secular architecture is represented by a few fragments from thirteenth-century palaces in Baghdad (now much restored and transformed into museums),132 with iwans and porticoed courts and rich stucco decoration covering most of the walls, and by two gates to the city. The more interesting, the Gate of the Talisman, was blown up in 1918.133 Built by the caliph al-Nasir in 1221, it was remarkable for the sculpted figure of a small personage pulling the tongue of two dragons, possibly a symbol of the caliph destroying the heresies threatening the empire, or perhaps a more general, apotropaic talisman, as was fairly common in the popular culture of the Fertile Crescent at that time.



THE JAZIRA (NORTHERN MESOPOTAMIA)

The middle and upper parts of the Tigris and Euphrates valleys and the mountains and semi-deserts between the two rivers and their affluents, known as the Jazira ('island') in medieval times, consisted of three parts: the Diyar Mudar, essentially the middle Euphrates valley, more or less coinciding with present eastern Syria; the Diyar Rabi'a, the middle Tigris valley, corresponding to the present northern Iraq; and the Diyar Bakr, including the more mountainous regions of the upper Tigris and Euphrates, now almost totally in Turkey. Great mountains - the Tauric chains, the Armenian knot, the Zagros and the Kurdish ranges - surround the Jazira on the east, north, and northwest. To the west and southwest lies the Syrian desert; to the south, Iraq. For centuries the main battleground between Mediterranean and Iranian empires, northern Mesopotamia was conquered by the Muslims in the first years of their expansion. For several hundred years thereafter it remained an area of transition, a passageway from Baghdad to Syria through Raqqa and Aleppo for trade and armies guarding the Anatolian frontier against the Byzantines. At least in the Euphrates valley, an area of major agricultural settlements had developed in the shadow of fortified towns such as Raqqa, Harran,134 and Diyarbakr.135 Prosperity declined considerably in the tenth century, as nomadic incursions threatened trade and weakened agriculture.

In the twelfth and thirteenth centuries enormous social and political changes took place. The conquest of Anatolia, the Crusades, Turkish and Kurdish population movements, the necessity of providing for large armies marching against Christians and Fatimids led to the transformation of the Jazira into one of the liveliest regions of the Muslim world. Old towns were revived, small villages transformed into major centres.136 As the danger from the nomads in the desert was checked, agriculture developed around some of the more important settlements. From impregnable fortresses enterprising feudal rulers or robber barons exacted taxes and tribute from passing caravans and armies. The cities of Mosul, Sinjar, Diyarbakr, Mayyafariqin (modern Silvan), Mardin, Hisn Kayfa (modern Hasankeyf), Jazira ibn Umar (modern Cizre), Harran, and many others suddenly hummed with power and activity. Armenian, Nestorian, and Jacobite Christians fully participated in the wealth and growth of northern Mesopotamia, and the building of new churches and monasteries is almost as remarkable as that of forts and mosques.

Prosperity did not last long, however; the Mongol invasions came and, as the destinies of Iran, Anatolia, and Syria moved in different directions in the following centuries, the Jazira reverted for the most part to an impoverished and largely deserted region of a few strongholds separated by menacing wastelands. Such it remained until the twentieth century. This fate, as well as its modern divisions between remote regions of three different countries, explains why its numerous monuments are still very little known and, with exceptions in present-day Turkey, unrecorded and little studied. Yet the interest and significance of the Jazira in the twelfth and thirteenth centuries, both for Syria and for Anatolia, cannot be overestimated. The Zengids and the Ayyubids, future rulers of Syria and Egypt, came from this area and, in the middle of the twelfth century under Nur al-Din and in the second quarter of the thirteenth under Badr al-Din Lu'lu' in Mosul, the Jazira was one of the truly great centres of Islamic economic and political life. Builders were busy, as a list of Nur al-Din's constructions proves,137 but limited investigation so far allows only for the identification of some of the more significant monuments and a suggestion of their importance.

New congregational mosques were constructed and older ones rebuilt. In Raqqa the old Abbasid mosque was redecorated and largely rebuilt in 1146-47, 1158, and especially in 1165-66.138 In Mosul almost all of the mosque built in 1148 under Nur al-Din (rebuilt in 1170-72) to replace an early Islamic shrine has disappeared or been redone; Herzfeld reconstructed it as a hypostyle with vaults in the Iranian manner.139 All that remains from the early construction, a superb minaret [350], cylindrical on a square base and curiously leaning, shows the impact of Iran in both construction and decoration. It may not date from the time of Nur al-Din, for another minaret certainly sponsored by him at Raqqa140 is a simple round structure, hardly showing an Iranian impact; or possibly the western part of the Jazira was slower to adapt new fashions than the eastern, for the earliest minaret in the middle Euphrates area clearly to show such brick influence Is the one erected in 1210-11 at Balis (modern Meskene).141

One of the most remarkable congregational mosques of the period, begun in 1204, is at Dunaysir (modern Kochisar) [351].142 All that remains is the prayer hall, a rectangle 63 by 16 metres divided, like the mosque of Damascus, into three naves parallel to the qibla - a Syrian-Umayyad plan to which was added a feature of undoubted Iranian origin: a huge dome in front of the mihrab which takes up two of the aisles. Also Iranian in origin is the squinch arch filled with muqarnas and the decoration of the spandrels of the squinches [352]; but the superb stone piers and brick vaults are in the pure classical tradition of Late Antiquity and of Byzantium. Equally classical is the traditionally moulded lintel gate, but the luxurious and monumental mihrab with its complex geometric, floral, and epigraphic designs reflects oriental influence [353], while the rather strange interlace motif of the façade recalls Armenian or Georgian themes and hardly fits with the decorative imagery of the Islamic Near East. The minaret was square, just like the one of 1211-13 farther west at Edessa (modern Urfa).143

The mosques of Malatya (1247-48, restored 1273-74),144 Mayyafariqin (1157-1227),145 Kharput (1165),146 Mardin,147 and a number of other cities of the area, though by no means yet thoroughly studied, plainly share the stylistic feature of characteristics drawn from various sources. The muqarnas squinch at Malatya is an almost perfect copy of a central Iranian type; indeed inscriptions confirm that there were Persian builders there. All exhibit a fascinating variety of decorative themes, from the 'brick style' and incrustation in the Persian tradition to portals with half-muqarnas domes of an Iraqi type here translated into stone, writing carved on an arabesque background, and rude but striking geometric themes also carved from stone. At Harran, even classical ornament was literally copied on capitals and friezes. Nowhere is this relation to a pre-Islamic world more apparent than in the mosque of Diyarbakr (ancient Amida). Quite close to Damascus in plan and proportions, its most remarkable feature is its court façades [354], at first glance an extraordinary jumble of antique and medieval elements. Undoubtedly the mosque was, in its main parts, erected in the twelfth century,148 adding new decorative motifs to elements of construction from older ruins, so that Late Antique vine rinceaux appear next to Islamic arabesques and Arabic writing. The result is less appealing aesthetically than it is fascinating as one of the most remarkable instances of the catholicity of taste which characterized the period and the area.

Besides congregational mosques, the cities of the Jazira boasted many smaller religious buildings. madrasa - none of which remains in its original form - are known from Mosul (seventeen of them),149 Diyarbakr,15 and most other cities. Some were attached to the tomb of the founder - the first instance of the combination of the mausoleum with some endowed public function which later became so popular in Syria and Egypt. Still standing in and around Mosul are a considerable number of sanctuaries dedicated to saints, prophets, and holy men,151 including Jonah and St George as well as medieval Muslims, indicating that ancient holy places were often taken over by the predominant faith. Their central feature was always a domed room, often conical or pyramidal on the outside, and the more elaborate ones frequently had an inner muqarnas dome (often in stucco, as in the mashhad of Awn al-Din [355], dated 1248-49) in complex polyhedral shapes related to Iraqi types, and handsomely carved mihrabs [356]. Christian churches took this form as well.152 In mosques the entrance proper is framed by interlaced polylobed niches filled with decorative designs, in churches by figures. The few known mausoleums and sanctuaries in and around Mosul are no indication of the numbers erected in the Jazira: a guidebook to places of pilgrimage written in the late twelfth century lists many more.153 Several are visible on the cliffs which border the Euphrates in Syria, and others could probably be found along the roads of the upper valleys of the two rivers.

The great sanctuaries of Edessa and Harran remain uninvestigated. These sanctuaries differ from known Iranian and Iraqi buildings in two ways. First, instead of being only tombs, they are usually associated with constructions dedicated to some cultic, philanthropic, or ceremonial purpose. Second, the architectural qualities found in Iranian mausoleums are not as consistently displayed in the Jazira. This may be because some of the best examples have disappeared, although it is more likely to be a reflection of a wider social basis among patrons and users in the Arab countries of northern Mesopotamia: the long and complex history of Arab cities - with their many religious, economic, and tribal components - might easily have led to greater differentiation in patronage and function than was likely in the constantly shifting and more ephemeral cities of Iran.

The secular architecture of the Jazira is equally varied and even less well known. The area became studded with castles, fortresses, and citadels. They occur along the Euphrates, as at Qal'a Jabar,154 Qa'la Najm.155 At Diyarbakr [357] the striking black basalt wall and massive round towers built over older foundations and often decorated with curious examples of animal sculpture.156 At Harran, strong walls and towers with long vaulted halls and impressive gates are still standing.157 And the celebrated Baghdad Gate at Raqqa [358] with its intricate decoration of brickwork clearly belongs to this period, as has recently been demonstrated. Other remains have not yet been systematically studied. Only small fragments remain of palaces. At the Qara Saray in Mosul,159 generally identified with the thirteenth-century residence of Badr al-Din Lu'lu', only a few mud-brick walls remain from what must have been a great pavilion on the Tigris; the only interest of the building now is its stucco decoration of interlaced niches with figures, as already encountered in religious architecture. The most significant feature of the single remaining caravanserai, al-Khan near Sinjar,160 is its two façade sculptures of a bearded man transfixing a snakelike dragon [359]. Of several surviving bridges, most of them ruined, the most interesting is at Jazira ibn Umar (modern Cizre), with its astronomical sculptures on the piles.161 Further explorations and occasional excavations will certainly bring to light other bridges, as well as standard monuments of secular architecture like caravanserais and bazaars so far known only from texts.



SYRIA, PALESTINE, AND EGYPT

The Zangid and Ayyubid princes who assumed control in Muslim Syria from various petty local dynasts first succeeded in ejecting the Crusaders from Edessa in the Jazira (1146), then took over Egypt (1171), and finally pushed the Crusaders back until, by the time of the Mongol invasion in 1258-60, only a few fortresses remained in Christian hands in Syria and Palestine, and a constantly diminishing Armenian kingdom barely subsisted in Cilicia (now south central Turkey). The changes in Fatimid Egypt after the middle of the eleventh century have already been discussed; this section concentrates on Syria and Palestine under Seljuq, Zangid, and Ayyubid rulers, and on Egypt after its conquest by the Ayyubid Saladin. Brief mention will be made of Yemen, remote and isolated from the main stream of central Islamic lands, but where a branch of the Ayyubid family established itself after the end of Fatimid rule.

These were memorable centuries for Islamic architecture in Syria. The two old cities of Aleppo and Damascus were totally revitalized,162 and small and at times almost abandoned towns and villages were transformed into major centres.163 It was a period of intense architectural activity which is finally drawing the attention of scholars and, most interestingly, of architects and urbanists involved in the rehabilitation of old cities and the restoration of their monuments. Enough material exists to justify, as was the case with eastern Islamic lands, a presentation of monuments separately from observations and considerations on techniques of architecture.



The monuments

Few congregational mosques were built, since most towns had had them since the first Muslim century, when Syria was the centre of power. But the old establishments in Aleppo,164 Damascus,165 Busra,166 and Jerusalem after the reconquest,167 were refurbished or repaired, increased or modified. At Aleppo, for example, while the plan and setting of the mosque are Umayyad, the porticoes (1146-47), courtyard, and minaret (1090) are from the Medieval period. But, as in late Fatimid Egypt, large institutions are rarer than smaller masjids or less ambitious congregational mosques serving either one of the many suburbs which sprang up at the time or some precise social or symbolic purpose. Such are the Hanbalite mosque in the Salihiya suburb of Damascus (before 1215-16),168 the Mosque of Repentance in a formerly ill-famed part of the same city,169 and various similar institutions in Aleppo known from texts or inscriptions.170 They are usually traditional hypostyles based more or less directly on the early model of the mosque of Damascus.

Several mosques were built or rebuilt in Yemen at this time. Some, like the mosque of al-Abbas at Asnaf (1126) or that of Sarha (thirteenth century) are closed chambers without windows, with a single entrance and, often, with beautifully decorated carved and painted ceilings [360]. Others are hypostyle buildings, like the mosque, founded by a woman, of Arwa bint Ahmad in Jibla (1088-89) with a courtyard and an axial nave reminiscent of Fatimid architecture in Cairo [361]. Monumental minarets and portals were added in the twelfth century to the mosques of Zabid and San'a.171

Of greater interest and importance are the institutions of Islamic learning sponsored by the new masters of Syria and Egypt. Most were madrasa for one or, more rarely, two of the four Sunnite schools of jurisprudence. At the Salihiya in Cairo,172 built in 1242, however, as at the Mustansiriya in Baghdad and probably under its influence, all four rites were united. In addition to madrasa there were several dar al-hadith for the expounding of Traditions;173 in many cases these also included the tomb of the founder or of a member of his family. The number of these schools is quite staggering. Later texts record the construction of forty-seven in Aleppo, eighty-two in Damascus, nine in Jerusalem, and nineteen in Cairo around the twelfth and thirteenth centuries. They were the most popular form of piety at the time, fulfilling more than a simple teaching function. While undoubtedly their systematic construction by great leaders such as Nur al-Din and Saladin reveals political, ideological, and religious intentions,174 many madrasa, especially in the thirteenth century, with large endowed properties attached to them, were also examples of conspicuous consumption and a way of restricting private fortunes to the same families.

In contrast to those in Iran, these institutions were usually small, especially in Damascus, squeezed between other buildings in older parts of cities, often with only a narrow façade to the street but spreading out at the back. This apparent constriction, at times avoided by building in the suburbs, arose from the power of the landowning bourgeoisie in Arab countries,175 which made urban sites far more expensive in Syria than they were further east.

In spite of considerable variations in plan, and of differences both within one city and from one city to another, almost all these buildings are related, as can be seen by an analysis of six of them: the madrasa in Busra of 1135 [362], the earliest known in Syria; Nur al-Din's dar al-hadith (1171-72) [363] and madrasa (1167-68) [364] in Damascus; the Adiliya (1123) [365, 366] in the same city; and two of the greater Aleppo madrasa, the Zahiriya (1219) [367] and the Firdaws (1235-36) [368, 369]. All are rectangular structures around a central court often with a pool; at Busra, however, a curious corbel seems to suggest that the court was not open but vaulted. The entrance is usually in the middle of one of the narrow sides [370], although a significant number of side entrances exist. Around the courtyard there is always at least one iwan, and sometimes three or four; when there are four, one is usually small and connected with the entrance. The oratory is generally a long hall occupying one of the sides of the court - not necessarily that facing the entrance, since proper orientation is often precluded by the exigencies of the site. In a few instances the iwan facing the entrance is also the oratory with a mihrab, as in the Sahibiya in the Damascus suburb of Salihiya.176 A simple triple (or, in the madrasa of Nur al-Din and al-Adil in Damascus, quintuple) arcade led from the court to the oratory. Elsewhere, vaulted halls occupied the space between main iwan entrance and oratory. In the great buildings of Aleppo all elements of design were larger and more monumental than in Damascus, and the courtyard was generally surrounded by a portico.

The origin of the plan of the Syrian madrasa has been the subject of much controversy.177 There is general agreement that it was imported from the east, as the madrasa evolved there earlier, as the iwan was hardly known in Syria, and as a frequent awkwardness in planning, construction, and decoration can best be explained through new influences. Yet it is remarkable how rapidly the Syrian madrasa became a type of its own with a number of variables which could be used for other functions. For this reason, and in the absence of earlier Iraqi examples, the hypothesis of a primarily Syrian and Zangid creation, with no doubt some impact from the east, is the most likely one.

When we turn to mausoleums, most of the extant freestanding ones are of archeological interest only. An exception is the spectacular (and often restored) tomb of al-Shafi'i in Cairo (1217) [371], on a simple and traditional plan, but superbly decorated, with one of the largest domes (15 metres wide) of the time.178 Also in Cairo is the smaller so-called mausoleum of the Abbasids, dated around 1240.179 In Aleppo and Damascus are found a number of mashhads and khangahs (houses for Sufi orders) on madrasa-like plans.

Much of Zangid and Ayyubid secular architecture is gone: of the more than three hundred public baths recorded in Aleppo and Damascus, only a handful remain.180 The quality of construction and decoration of the caravanserais still standing on the main roads of Syria181 is not nearly so high as in Iran or Anatolia; nevertheless, together with the great markets, such as the one in Aleppo planned in medieval times although later in its present shape, they testify to the Saljuq and Ayyubid princes' interest in commerce, which is borne out by an account of a military leader buying a palace in Aleppo and transforming it into a warehouse and oil press.182 Hospitals were the most common philanthropic foundations; Nur al-Din's [372, 373], built in Damascus in 1154 on the ubiquitous four-iwan plan, still stands.183 It is one of the most harmoniously composed masterpieces of twelfth-century architecture, with a particularly elegant façade combining the geometry of a muqarnas half-dome with a classical lintel below.

The most spectacular secular architecture is military. The walls of Aleppo, Damascus, Jerusalem, and Cairo all remain in part. Some (as in Jerusalem) were reconstructions or repairs of older walls, but more often a new enceinte was heeded to correspond to the growth of the city. Many of the new gates, generally with several turns for better defence, are still major landmarks of cities.184 But even more striking were the citadels known as qal'as. It was the residence and symbol of the sultan, usually overlooking the city and often set across its walls for combined control of the city and independence from it. In Aleppo work done on the citadel as early as the tenth century, at the time of Byzantine attacks, was continued under the Midrasids (1025-79) and the Zenguids, who built one of the sanctuaries inside. The magnificent construction now towering over the city [374], in spite of many later repairs and additions, goes back to the early thirteenth century and the sultan al-Zahir Ghazi, who was responsible for the spectacular glacis, the triple entrance, most of the towers, the great water tanks and food stores of the interior, and the mosque. Significant parts of the palatial ensembles have been cleared in recent years. The citadel of Damascus is not as striking. Saladin's brother entirely rebuilt it on the remains of an older and more primitive construction. It included private quarters, offensive and defensive gates, and an oratory.186 In Jerusalem the Crusaders and Saladin had transformed the ancient Herodian, and even earlier, citadel. The spectacular but often-repaired qal'a on a hill overlooking Cairo has been throughly analysed by K. A. C. Creswell.187 The one at Busra grew up round an ancient Roman theatre and thus succeeded in creating one of the most stunning contrasts in architectural design, as the sombre, vaulted, frightening halls of a basalt-built fortress lead to the brilliantly lit trabeated marble of the theatre.188

The interiors of these citadels, later rearranged, were probably rather monotonous, as in most military architecture, with long halls, narrow openings, various devices for defence, courts, stables, and originally austere living quarters. Yet some of the details from the citadel in Aleppo show considerable care given to details and a sober but effective masonry decoration. The gates were the most impressive feature, at times bearing figurative symbolic sculptures [375], always with magnificent inscriptions which were symbols both of possession and of the power and prestige of the individual sultan. It is unlikely that the often ephemeral rule of constantly warring princes gave rise at that time to any significant ceremonies inside the citadels, nor even to any elaborate cultural life, as happened at the same time in the Muslim West (see below, Chapter 7). There does not seem to have been much of an architecture of pleasure and comfort in most of them. But, since the later quite luxurious baths and halls in the citadels of Aleppo and Cairo were built over earlier palaces about which we know a little from texts,189 careful archeological investigations may yield many surprises.



Construction and decoration

Much work has been done on the methods of construction and decoration in Egypt and Syria. Consequently quite fine distinctions can be drawn between the individual architectural idioms of Cairo, Aleppo, and Damascus. At the same time, constant influences and movements of craftsmen and ideas from one city to another contributed many variations.190

Materials are traditional: stone in Syria, with brick fairly common for vaults in Damascus, basalt in the Hauran, brick and stone in Egypt; wood throughout for limited purposes. Unexpected techniques that appeared occasionally, such as the use of wood in the dar al-hadith of Nur al-Din in Damascus (between courses of stone, a feature common in brick, but in stone serving only to weaken the wall),191 indicate a new dependence on northern Mesopotamia or Iraq. But, in general, the masonry is simple, except on certain façades and arches where joggled voussoirs and the ablaq technique of alternating stones of different colours are used. Rubble in mortar - both inexpensive and quick - was fairly common in vaulting in caravanserais and citadels.

The supports consisted of traditional columns with capitals (sometimes utilizing new muqarnas-based designs) and of piers carrying arches. But, more and more, heavy walls, often pierced by bays, appear in new buildings, as they did in Iran. This is largely due to the spread of vaulting, which came about partly through the penetration of themes from the north and east, partly because often wood could not be used (particularly in military architecture) for fear of fire. A few masjids and the oratories of some madrasa have old-fashioned wooden ceilings, but barrel-vaults, often of simple semicircular section, as well as cross-vaults are usual on rectangular spaces and are especially typical of the long galleries of military architecture. Flat arches, usually in combination with relieving ones, are also occasionally revived.192

Domes and zones of transition are of almost unbelievable variety. The large wooden dome and the muqarnas zone of transition of the mausoleum of al-Shafi'i in Cairo date from the fifteenth century. Elsewhere in Egypt, as in the tomb of the Abbasids, the Ayyubid models simply transformed the Fatimid muqarnas squinch into a composition covering the whole zone of transition. The citadel of Cairo and most Syrian monuments use the squinch and pendentive alone or combined with muqarnas. The mosque of Busra may have had a corbelled zone of transition, in line with the corbelled roofing of the pre-Islamic Hauran, but it is still unclear whether a dome covered the centre of the madrasa. The Iraqi and northern Mesopotamian technique of high domes on rows of muqarnas did not reach Egypt in Ayyubid times, but became fully acclimatized in Syria with the first Zengid monuments. Translated into stone, it provided some of the most effective domes over tombs and entrances and half-domes on façades, probably endowing them at the same time with a rather cold and dry mathematical quality.

Decoration in Syria and Egypt was on the whole remarkable for its sobriety and simplicity. It was limited to gates, where single sculpted panels were often put on the walls around the entrance; to plaques and bands of writing, using Qur'anic quotations or established formulas to point up the purpose of the building and the glory of its founder; to the elaborate stone or stucco grilles of windows and oculi; and to mihrabs in wood, stone, stucco, or the peculiarly characteristic new technique of marble incrustation. Themes were traditional, including arcades (as in the minaret of the Great Mosque in Aleppo), classical and early Christian motifs reused from older buildings, or further developments on the Fatimid geometry based on star patterns [ ]. Three newer features are particularly significant. The first is a motif of interlacing heavy lines, varying in the complexity of their geometry and in the relationship between right angles and curves. It occurs most commonly in mihrabs [376] - more specifically, in the rectangle framing the nichehead - and also in gates, creating a strong and immediate visual effect. The motif reflects a simpler and ruder tradition and taste than the minute arabesques of Fatimid times, but its influence was to be quite strong in Anatolia and in Mamluk Egypt. The second characteristic theme is writing, often used in conjunction with floral motifs. Like contemporary objects, architecture bore both angular, somewhat artificially archaizing inscriptions and the more common cursive ones. Like contemporary sculpture in western cathedrals, the epigraphy both illustrates the purpose of the building and emphasizes its main axes and lines, fulfilling the function of a moulding in architecture as well as reflecting the expressive value and meaning of a monument. A most striking example occurs in the Firdows Mosque in Aleppo [377], where the mystical imagery of the inscriptions sets the tone for the peaceful and otherworldly atmosphere of the building.193

The third motif involves the windows and medallions used on qibla walls,194 domes, and façades, geometric in Syria, but often incorporating magnificent floral arabesques of leaves and stems. Related though they are to Fatimid or

Iranian themes, the main quality of these complex designs - as can be seen for instance in the Abbasid mausoleum in Cairo195 and the Maydaniya in Damascus196 - is their remarkable clarity, which enables the eye to catch the major lines of the movement without being bored with endless repetition. Such arabesques do not have the wealth of their Iranian or Iraqi counterparts, but they make up for the consistent simplicity of their designs by their elegance and restraint.

Two more original techniques are those of mosaic and of representational sculpture. Mosaics occur in some mihrab niches in Egypt197 and in Saladin's reconstructions in Jerusalem, in particular in the Aqsa Mosque. Saladin probably used mosaics in a conscious attempt to revive the decorative methods of the first conquest of Jerusalem by the Muslims in early Islamic times. The actual quality of the workmanship is not very high, but its presence attests to the major task of rehabilitating the Haram al-Sharif.

The second technique, representational sculpture, was applied chiefly to secular architecture, most interestingly at Aleppo, where intertwined dragons and lions guard each of the three gates to the citadel. Their iconography and their simple but effective style relate them to similar images in Iraq and northern Mesopotamia, and their prophylactic aim is confirmed by several texts,198 but their origin and their application to contemporary Aleppo are unclear.

Zangid and Ayyubid Syria was the second of the Muslim regions after Iran to evolve a great medieval architecture. Although the citadels of Aleppo and Cairo are the only monuments to rival some of those farther east in size and in the complexity of their history, Syria must nevertheless be singled out for the variety of its constructions, the growth of military architecture, the incorporation of motifs and techniques from the east and from the north, the importance of cities in determining the sizes and types of buildings, and the transformations given to the muqarnas. Many of these features reflect the religious and cultural needs of the time and illustrate phenomena wider than either Syria or the Arab world, most particularly that great Sunni revival which became the mission of many of the region's rulers. The simplicity and clarity of construction, excellence of workmanship, successful use of stone, the sobriety of decoration, fondness for geometric lines and for clear surfaces all reflect, wilfully or accidentally, some of the qualities of the rich heritage of Late Antiquity. Some scholars have even talked about a classical revival.199



ANATOLlA

The battle of Manzikert opened Anatolia (known in medieval Islamic sources as al-Rum) to Islam in 1071, but it is not until the turn of the thirteenth century that the Saljuqs of Rum, a few minor dynasties related to them, and many relatives of ruling princes or government officials were sufficiently established to engage in major building activities. Only indirectly affected by the Mongol conquests, except for the refugees from Iran and Iraq who poured into Anatolia, the Saljuqs of Rum did not disappear from the scene until the beginning of the fourteenth century, when internal dissensions gave rise to a number of more or less independent principalities. Thus Anatolian Muslim architecture developed mostly after the main features of Iranian and Syrian medieval architecture had been established. A further peculiarity of Saljuq Rum was its cultural, social, and ethnic make-up. As a newly conquered Islamic province, it counted many non-Muslims and recent converts, with the twin consequences of eclecticism and of a wide range of cultural components, especially from the Christian Caucasus. As a frontier area it attracted Muslim militants, from ghazi (militant) warriors to the adherents of mystical Sufi orders. Just as in Syria, the large number of preserved monuments and the absence of monographic studies devoted to any one of them justifies a presentation which separates comments on the buildings from the processes of construction and idiosyncrasies of styles.



The monuments

Being in control of a newly Muslim area, the Saljuqs of Anatolia had the task of erecting all the buildings which had by then become characteristic of Islamic civilization. The most important was the congregational mosque. An early one at Mayyafariqin (perhaps of the eleventh century) was a simple rectangle (65 by 61 metres) with a court and a hall of prayer eleven naves at right angles to the qibla; pillars and arches carried a flat wooden roof.200 The mosque of Ala al-Din in Konya [378] (built between 1156 and 1235 with later additions) is historically more complicated because it was not built at one time and because it was included within the palace area and also served as a place of burial for princes. In spite of this, its last addition in 1235 was a simple hypostyle in an early Islamic tradition even to the point of using columns and capitals from older buildings.201 A number of other such simple hypostyles, for example at Beysehir and Afyon, lacked courts, and several were almost entirely of wood, reflecting both its availability in the mountains of Anatolia and, perhaps, the impact of Central Asian tradition.

More original plans occur in the much restored Ulu Cami at Kayseri,202 and in the mosques attached to philanthropic and religious institutions, such as the Khuand Khatun complex of mosque, madrasa, and tomb at Kayseri (1237-38), and the mosque and hospital at Divrik (1228-29). The courts have all shrunk to simple central squares. The naves of the Ulu Cami are at right angles to the minuscule court (later domed). In front of the mihrab is a large Persian-style dome. The mosque of Khuand Khatun [379, 380] is divided into square bays, plus a sort of axial nave of two wide bays and a large dome. At Divrik (Divrigi) [381-383] it is a five-aisled basilical hall with a wider central aisle, the naves consisting of rectangular bays except for the square one in front of the mihrab. All three-mosques have three entrances, one on each side other than the qibla, symmetrically arranged only in the Ulu Cami. As can be expected in a newly conquered area with an old history, aberrant types exist as well, for example the three-aisled mosque of the castle at Sivas (1180-81),204 and the Iplikci Mosque (1182-1202) with its three rows of seven square bays with the qibla on one of the long sides and three domes leading from the front door to the mihrab.205

Madrasas were also common. They are of two types. The first, exemplified by the Saraj al-Din (1238-39), Khuand Khatun, and Sahibiya (1267) madrasas at Kayseri,206 the Gök (1271) at Sivas,207 and the Sircali (1242-43) at Konya,208 is closely related to the Syrian and farther eastern types. On a court with porticoes open varying numbers of iwans, of which one is always connected with the entrance. The tomb of the founder is usually by the entrance or on the side opposite it. The interior consists of long halls at right angles to the court. Different from Syrian prototypes are the protruding iwan-like entrances, sometimes framed, as in the Gök madrasa, by two high minarets. The most monumental and remarkable variant of this type, a transformation of an Iranian tradition, is the Cifte Minareli madrasa at Erzerum (1253) [384-386]. Here is one of the earliest instances of afaçade with two minarets. The circular mausoleum is on the axis of the building at the back of a long iwan, and the iwans have two-storey arcades.209

The main centre of the second group, which is more peculiar to Anatolia, is Konya, the capital of the Saljuqs of Rum. There, in the Karatay (1252) [387-388] and Ince Minareli (1258) madrasas [389-391],210 the single iwan-like feature, the long halls, the domed rooms on either side of the iwan, and the magnificent façades are clearly connected with earlier traditions. However the court has been replaced by a dome and the buildings are understandably smaller, a development related of course to the similar abandonment or diminution of the court in congregational mosques, without any major modification of the rest of the building. This change, generally explained as a consequence of the rigorous climate on the Anatolian plateau, had a far-reaching formal significance, especially for the madrasa, for the characteristically Iranian monumental inner court façade based on the iwan was replaced by a building with a large outer façade, planned around a central dome. Probably, beyond climatic reasons, the Christian architecture of Armenia and Byzantium, which consisted wholly of such centrally planned buildings, affected Muslim architects.211

Just as in Iran and Azerbayjan, the single mausoleum, generally known in Anatolia as a türbe, was much more common than in Syria. A few were square212, but the vast majority were polygonal or circular, on high bases, usually with a crypt and a domed interior, with pyramidal or conical roofs, and richly decorated façades. At the curious Mama Hatun mausoleum at Tercan, a circular enclosure surrounded the türbe like an ancient temenos. In central Anatolia there also existed a so-called iwan-tomb with a prayer chamber open at one of its ends and with vaults covering both crypt and main chamber; such are the tombs of Haci Cikinik at Niksar (1183) and of Sayid Ghazi in Eskisehir (1207-08). These tombs are most closely related to those of Azerbayjan, but local traditions may have been involved as well. Oddly enough, funeral architecture was influenced primarily from the Iranian world, whereas mosques and madrasas apparently often arrived through Syria and the Jazira.

As to secular architecture, remains exist of hospitals, for example the one at Divrik whose plan is so similar to that of a madrasa; there were others, for instance the four-iwans one of Gevher Nesibe Hatun in Kayseri and the recently excavated one of Izzedin Keykavus in Sivas; Saljuq Anatolia was known for its great medical schools. Many of the hospitals were attached to mosques, mausoleums, madrasas, and other socially pious buildings. Anatolia in the thirteenth century witnessed the appearance of complexes containing both pious and useful functions, often supported by an endowment, a kind of development which will find a spectacular expansion in later Ottoman architecture.213

Of the great palaces of the Saljuqs, whose wonders are related by the chroniclers,214 only a few walls have remained in Konya, although previous travellers saw more.215 A huge palatial complex at Kubadabad, near Lake Beysehir, has been partly excavated and belongs to the grand tradition of early Islamic palaces. K. Erdmann identified as Saljuq a widespread number of small structures which he called Seraibauten216 - perhaps hunting lodges or bases for agricultural exploitation, as had existed in Central Asia and Umayyad Syria. Of numerous remains of Saljuq fortifications many were destroyed in the twentieth century to make way for new towns.217

By far the most spectacular constructions of Saljuq Anatolia are the superb caravanserais, nearly two hundred and fifty of them from the thirteenth century.218 There are three basic plans. The first is comparatively rare and, like the Syrian examples mentioned earlier, consists of a rectangular or even square building with a central court from which open halls of varying sizes. The second, exemplified by Zivrik Han [392]219, is square or rectangular and lacks a court. The best examples have a central nave abutted by others at right angles, often with a central dome for light and air. The third plan - that of the two Sultan Hans, one near Kayseri220 - is the most remarkable [393, 394]. A court with halls at right angles to its sides precedes a long covered building with a central nave and others at right angles. Superb portals led often to an oratory (at times a separate pavilion in the middle of the court) and to a bath. While their monumentality and construction are peculiarly Anatolian, the last two plans are related to earlier ones in Iran and Central Asia and may well be pan-Islamic. But it is probably more appropriate to consider them as reflections of functional objectives and practical uses in detail which cannot, at this stage of investigation, be reconstructed or even imagined.

Why did chains of caravanserais of such quality emerge so suddenly? Sponsored by the ruling princes themselves, they are in all likelihood a rare attempt to capture international trade at a time of shifting directions for commerce and of constantly moving populations; but how they fit within the economic policies and activities of the time remains to be investigated.



Construction and decoration

By far the most common material for monumental buildings was stone, though brick was used for secular vaults and also in cities such as Konya. Wood also was sometimes employed, occasionally for entire buildings. Rubble in mortar was common for simpler vaults and walls.

Most structures were vaulted. Supports might consist of long and solid walls, especially in the madrasas, where the Iranian iwan had imposed its plan and elevation. A more common and original type, however, used piers and columns, ranging from borrowed older columns to new ones with muqarnas capitals, from polygonal, round, or even cross-shaped low piers carrying high and wide arches to the superb high piers and arches of the caravanserais. The arches are usually carefully outlined, even when bonded with the masonry. Like the piers and columns, they are an interesting revival of Late Antique and early Christian practices in the Near East.

Vaults display an equally fascinating variety. The most common system of roofing a long space, for instance in the khans, was by means of tunnel-vaults, often divided by transverse arches. Small rectangular areas, as at Divrik, show endless variations on the simple theme of the crossvault, with a multiplication of decorative rather than structural ribs. Domes were usually on pendentives, at times with muqarnas although squinches are also known. At Konya an original mode of transition is the 'Turkish triangle' [391], a rationalization of the pendentive into simple geometric forms. Sometimes a combination of several long triangles gave greater, but still very angular, movement to the passage from square to dome.

Seljuq architectural decoration reflects the same multiple influences as building construction and design.221 The portals of the Ala al-Din mosque in Konya and of the Karatay madrasa are typically Syrian in style, while other mosques and caravanserais use the muqarnas half-dome of Syria and Iraq. At Divrik, Kayseri, and in the Sultan Hans, stone carvings on façades and along the major lines of the architecture reflect the brick decoration of Iran. Apparently more original is the use of mosaics of glazed tiles, not merely as an element of emphasis but completely to cover large wall surfaces. The best-preserved examples are at Konya, in the iwan of the Sircali madrasa [417] and in the zone of transition of the Karatay madrasa. The technique originated in Iran, but it first occurs independent of other decorative devices in Anatolia.

However, the most spectacular results were achieved in stone-carving and on façades. Almost every monument warrants a detailed study, for each presents peculiar problems. Of the two groups which define the most striking characteristics of this decoration, the first comprises certain monuments of Konya. The impact of Syria is obvious, but, on the portal of the Ince Minareli madrasa [389-390], the reserve of the Karatay façade [387] has given way to an odd composition of columns, recesses, and mouldings. Architectural elements transformed the elevation into a non-architectural combination of thick interlacing epigraphic bands and geometric or floral designs in both very low and very high relief. The arches of the portal are absurdly composed, and the architectural elements do not lead into an architectural composition. In addition the constant opposition between kinds of relief and the lack of appropriate proportion between such diverse elements as a column and an epigraphic band contribute to the fascination of the façade, but also to the general impression it gives of being a sort of collage.

The second and much more spectacular group includes the façades of the main buildings of Sivas [395] and Divrik [382].222 The whole wall is involved in the composition. Highly developed corner towers frame it, while two tall thin minarets emphasize a huge central portal. The portal at Divrik is even splayed. The decoration includes both the traditional Islamic epigraphy and geometric or floral arabesques and fantastic combinations of vegetal and even animal forms which, in their tortured violence, recall Celtic miniatures and Romanesque façades. An actually Romanesque origin can possibly be proposed for a portal in the mosque of Sivas. Even the geometric designs, like the ones on the Sultan Hans [396], are not always of the Islamic symmetrical and organized type but recall the endless meandering of northern, so-called barbarian, ornament. In now disappeared secular buildings, figural sculpture was often used:223 dragons, lions, elephants, fantastic animals, astronomical figures, princes and court and attendants. At times crude, this sculpture seems to reflect a visual awareness of the artistic wealth of the Anatolian past and perhaps the memory of ancient pagan beliefs from Central Asia.

Medieval, so-called Saljuq, architecture of Anatolia was a highly original achievement. The last Islamic province to develop in the Middle Ages, it took its inspiration from Iran, Syria, and the Jazira, drawing also on the strong indigenous Christian and even earlier traditions of Anatolia and conscious of the grand architecture of Christian Europe. The great achievement of this architecture was that it drew together so many disparate elements to create monuments which may at times seem awkward and strangely composed, but which always express the powerful spirit of the conquerors and a passionate need for expression through buildings. This complexity in the process of architectural creation is demonstrated, among many other arguments, by the presence of many signatures of builders on the monuments.224 The relationship between patrons and builders became in Anatolia much more elaborate and much better documented than elsewhere in the medieval Islamic world.



THE ART OF THE OBJECT

Although during most of this period the once omnipotent Abbasid caliphate was only a shadow of its former self and, for all intents and purposes, virtually powerless, the textiles of Baghdad were both highly esteemed in medieval Europe and also much in demand there; indeed, European languages have been enriched by terms drawn from their names. From Baldacco, the Italian designation for Baghdad, is derived the term baldacchino for luxury textiles, especially those used for canopies; in England the fabrics of Baghdad were called baudekin, or baldachin. Matthew Paris, the English monk and historian (d. 1259), mentioned that Henry III wore a robe de preciosissimo Baldekino at an investiture at Westminster Abbey in 1247. Inventories of St Paul's Cathedral from 1245 and 1295 indicate that these baudekins were patterned with roundels containing griffins' heads and diminutive lions, double-headed birds with wings displayed, elephants, men on horseback, archers, and 'Samson the Strong'.225 Most of these textiles we know only through literary references, which are, however, not precise enough to permit definite identification with preserved fabrics.226

Happily, however, there are two textiles that, though manufactured in Spain, do give some clue as to the appearance of these costly fabrics, namely a fragment in Leon and the silk [458].227 Although, in the inscriptions on both, Baghdad is claimed as the city of origin, technical features as well as paleographic details reveal that the silks are actually Spanish copies of Iraqi originals. Their compositions, based on a series of roundels framing pairs of animals or fantastic creatures in symmetrical arrangements, represent medieval paraphrases of Sasanian textile designs and are thus closely related to contemporary Iranian silks [251]. Such circumstantial evidence is corroborated by the statements of the twelfth-century Arab geographer al-Idrisi and the early seventeenth-century Maghribi author al-Maqqari mentioning that 'attabi fabrics (named after the 'Attabiyya quarter of Baghdad) were made at Almeria in Spain.228 There are many other such references to the copying of textile patterns in distant parts of the medieval Muslim world, and this testimony helps to explain the difficulty of distinguishing between the products of different regions.

In the thirteenth century, Mosul in the Jazira was also an important centre of textile manufacture,229 as is again attested by words still current in European languages: muslin, mousseline, muselina, and mussolina. It has not yet, however, been possible to isolate this type of fabric from among those that have come down to us.

Among the textiles being produced in contemporary Syria under the rule of the Ayyubids, one [397] - the decoration on which bears close comparison with the figural and vegetal ornament filling the interstitial area on the fragment [252] - illustrates clearly how the tremendous displacement of artisans during the Mongol era resulting in an enforced migration from east to west contributed greatly to an eventual blending of styles. The same phenomenon is witnessed in the textile [398], probably belonging to the Rum Saljuqid sultan Kayqubad ibn Kaykhusraw (r. 1219-37). In gold on a crimson ground, two addorsed lions in roundels form the main design with the ever-popular arabesques filling the space within and between the circular frames. Once again the general layout, juxtaposed animals, and interstitial configurations betray distant Sasanian origins, but a new elegance and lightness permeate this design, which can also be found on a few extant and approximately contemporary fabrics woven in eastern Iran.230

The dearth of surviving metalwork from the later twelfth century in the central Islamic lands is partly made up for by a wealth of objects from the first half of the thirteenth century. The rise of artistic patronage under various Turkic groups including the Artuqids, Zangids and Rum Saljuqs as well as the Kurdish Ayyubids caused a sudden flowering of the local artistic tradition. This burgeoning seems to have been advanced, just as in the textile industry, by an influx of refugee metalworkers from Iran, whose presence can be deduced from the nisba of one of them,231 as well as from stylistic evidence.

Proof that Artuqid metalworking centres specialized in casting is abundant from the material extant. A large copper alloy talismanic mirror made in the mid-thirteenth century for Artuq Shah, a member of that dynasty, is adorned on its flat back with an heraldic bird in high relief in the centre and a long princely inscription framing its outer edge [399]. The intermediary space bears two interconnected decorative bands, one with twelve contiguous interlaced roundels each ornamented with a zodiacal sign with its planetary lord, and the other with a second inscription interrupted by 'classical' busts of the seven planetary gods. This dependence upon classical or Byzantine prototypes and emphasis on propitious heavenly bodies is reflected also in contemporary copper coins issued in the Jazira and helps to define the iconographic preferences in that area.232

Another category of cast metal objects for which the Jazira is known during this period is that of copper alloy door fittings. Although purely utilitarian in nature, both their bold yet intricate arabesque designs [400] - often on many levels and thus a tour de force of the caster's art - and the sinuous yet realistic fantastic animal forms are not only paradigms of Jaziran style and iconography but also illustrative of how that area in general and Mosul with its great wealth in particular served as a bridge between Iraq and Syria and thence to Anatolia under the Saljuqs of Rum.233

Not only did the metalworking centres under the artistic patronage of various Turkic dynasties cast outstanding works in copper alloys, they worked in other metals as well [401]. This mirror is cast in steel and inlaid with gold. It incorporates decorative motifs and stylistic conventions we encountered earlier in this section and shall continue to see on many other media produced in this area.234

The metalworkers in this region also lavished great care on the musical instruments they crafted such as the large drum [402]. This was probably part of the issue of a military band accompanying an Artuqid ruler into battle or on ceremonial occasions. The principal decoration on this rare survival consists of a playful animated angular inscription that incorporates both human and dragon heads, the latter very similar in style to those on the door handles from the Ulu Cami in Jazira ibn Umar (modern Cizre) referred to above.235

On a totally different scale are the gilded silver belt fittings [403]. These elements originally adorned a type of flexible belt from which were suspended short straps, also bearing fittings, that had been popular in pre-Islamic Iran and continued in use in early and medieval Islamic times until it was largely superseded around 1400 by a new type that remained in vogue for centuries to come. Such ornaments and another, slightly later and datable, group give us a tantalizing glimpse of the splendour of personal adornment in the Jazira and Syria during this period.236

Belonging to the tradition that created these fine groups of fittings is the particularly splendid gilded copper alloy mosque lamp known from its inscription to have been made at Konya in 1280-81 [404].237 On its body are graceful arabesques in repoussé, a technique rare on copper alloy objects of this period. The principal decoration on the neck is a Quranic passage from the Sura of Light, 24:35238 referring to the ineffable presence of the deity himself, in the form of a glass lamp suspended in a niche or arch. The entire surface of the Konya lamp is pierced to allow light to shine forth from a glass container within, casting intricate and beautiful shadows. Three bulls' heads serve to attach the (now lost) suspension chains.

As regards this area's important metal objects inlaid with silver, the earliest dated example from the Jazira is a miniature box of 1220. Production of such pieces, however, must have begun around the turn of the century, for the maker of the dated container, Isma'il ibn Ward, was a pupil of an already practising master called Ibrahim ibn Mawaliya.239 Both artisans ended their signatures with the generic 'al-Mawsili', 'of Mosul' - a designation used on at least twenty-eight objects240 (one from as late as 1321) by twenty metalworkers - thus implying metal production in that city; but only one craftsman, Shuja' Ibn Man'a, stated specifically that he made his object, a ewer, in Mosul in 1232 [405]; a second named Damascus, and five others Cairo. In other instances names of the owners suggest that Mosul itself was not the city of origin. Indeed, only six pieces, besides the one by Shuja', can be said with certainty to have come from Mosul, for they were ordered by the local ruler Badr al-Din Lu'lu' (r. 1237-59) or by members of his court.241 Many Mawsili artists worked in styles quite different from those attested by these six pieces, and in the work of one single artist there are stylistic differences that may imply various locales.242 This pattern reveals how difficult it is to make attributions of metal objects from this period when historical inscriptions are lacking.

There are, however, certain features that do distinguish the metalwork of the Jazira, including that of Mosul, from contemporary Iranian production. First, representations of princes, still rather rare in Iran, are frequent, which is only natural in view of the high percentage of pieces known to have been ordered by royal patrons. Indeed, the princely theme is the keynote of these works, and in certain cases it is developed into an all-encompassing royal ambiance. On the other hand, genre scenes extend well beyond the stock Iranian motifs of revellers, hunters, and polo players, and their composition is more sophisticated. Their variety is astonishing as well: gardeners with spades and mattocks, peasants ploughing with pairs of oxen, a flute-playing shepherd in the shade of a tree surrounded by his flock and faithful dog, boys shooting at birds with blowpipes, a relaxed youth reclining on a couch with his cupbearer and sommelier in attendance, a noble lady admiring herself in a mirror while a servant girl stands by with a box of toiletries, and so on [406, 407].243 Even the more formal royal images are set outdoors and are connected with merrymaking and hunting. Like the ceramic decorators of Iran, the rich repertory of the Mosul metalworkers often shows an indebtedness to the inventiveness of manuscript illustrators. The fundamental importance of the painting styles of Iraq, the Jazira, and Syria in this period as sources of imagery for the art of the object can be observed again and again, even though it is not always possible to distinguish the style of one specific area from that of another (see discussion of manuscript painting, below). Finally, the shapes themselves show greater variety.

More surprising still is the novel organization of themes. Whereas parallel bands, unfolding without interruption, and differently shaped spaces filled with closely packed motifs of equal importance, are typical of Iranian work, in the Jazira they were replaced by sequences of vignettes, some of them relatively large, some small. Avoiding a sampler-like display of individual motifs, the craftsmen organized vessel surfaces by connecting the frames around the images and by using smaller ornaments as links. In addition the vessels are encircled at various levels by bands that tie the compositions together. Furthermore, to prevent monotony, the even flow of main elements is punctuated at intervals by secondary motifs. Even the spaces between the vignettes have an artistic function. Instead of being left undecorated and thus neutral, as on Iranian works, they are covered with delicate webs of arabesques or interlocking fret patterns based on swastikas or Ts,244 which lend tension to the surfaces and serve as foils for the main features. By these means the monophonic co-ordination of equal parts has been replaced by a polyphonic form, of graded subordination, in which the many different parts of a complex composition are made to interact and interrelate. As a result both aesthetic and intellectual requirements are fully satisfied. No wonder, then, that metalworkers of Mosul origin were in demand in the highest places everywhere: their works 'were exported to kings', according to a contemporary Spanish Muslim,245 and that is why they signed their products with the name of the town from which they hailed.

We have seen above how Mawsili artists migrated to other major cities outside the Jazira including Damascus. Thus, it is not surprising that the distinction between Jaziran and Syrian inlaid pieces is often hard to draw.246 The stylistic differences are more subtle and the borrowings more wholesale than, for example, those between Persia and Jaziran work. The complex 'polyphonic style' of the latter continued, including the fretted backgrounds, but in Syria the work was in general drier and more meticulous; representations such as throne scenes became more formal and the arabesque spirals in the background more pronounced. Nevertheless, the short Ayyubid period (1171-1250) is commemorated by some remarkable and varied pieces of inlaid metalwork, often with novel features. Foremost among them is the use of Christian motifs, including New Testament scenes and, within arcading, figures of ecclesiastics and saints testifying to the relationship between the Latin states and the Ayyubids or to the presence of Christians in high positions at the Sunni Muslim court. One example is the so-called Arenberg Basin made for the Ayyubid Sultan of Cairo and Damascus al-Malik al-Salih Ayyub (r. 1239-49) [408].247 Another fine example of the type is the anonymous canteen in the same collection which, like the basin, combines Christian subjects with scenes depicting mounted horsemen including Crusader knights.248 Another feature that makes its debut in Syria is gold inlay, which was used on a basin made in 1250 by another Mosul metalworker, Dawud ibn Salama.249

An even better aid to attribution than stylistic and iconographic clues, however, are inscriptions, which give the dates and original - usually royal - owners of many distinguished objects. The earliest inlaid piece with an Ayyubid association, a ewer made in 1232 by Qasim ibn 'Ali of Mawsili origin,250 has a unique decorative scheme that would be difficult to place without the evidence of the inscription. Both body and neck are covered, not with the usual figural designs, but exclusively with fine arabesques enclosed in a network of ovoid compartments - the first occurrence of the all-over, purely abstract vegetal patterns on metalwork which was to be more common in subsequent periods.

Mosul, like Sinjar and Takrit, was renowned at this time also for its large unglazed water jars called habbs, the ornamentation of which attained a particularly high artistic level in response to the demands of affluent consumers. Unglazed pottery household vessels account for a high percentage of the total ceramic output of the Islamic world during the period covered by this volume. However, most of this large production was not as ambitious as, and considerably less refined than, the profusely and finely decorated ewer [90] and the habbs being discussed here. Because of their basically utilitarian function - liquids stored in them were kept cool by the evaporation that occurred through their porous walls - these storage jars were popular for centuries, their antecedents predating the arrival of Islam in the area and their descendants continuing in use down to the modern period. All their surfaces except for the rounded bottoms (which were set into the ground or placed on stands) are covered with relief decoration, adroit combinations of moulded, incised, carved, pierced, and barbotine work - a technique in which rolled strips and circles of clay were applied to the surface and, sometimes, decorated [409].241 The motifs constitute a fascinating potpourri of ancient gods and their sacred animals juxtaposed with the latest images of princes, musicians, revellers, and court officials. The popularity of such designs throughout the whole central Islamic area during the medieval period is witnessed by the fact that they can be found on so many different media in Iraq, the Jazira, Syria, and Anatolia at this time - exhibiting the very same style as on the unglazed habbs. The now destroyed Gate of the Talisman in Baghdad, stone architectural elements from the Jazira [426], a gateway in the Aleppo citadel [375], and wooden doors from Anatolia [424] - to mention only a few examples - testify to the veracity of this statement.252

The beautiful arabesque decoration on the cut brick from the Mustansiriya in Baghdad [410], founded in 1233 and discussed earlier in this chapter, is a testament to the continuation in Iraq of this striking style up to the end of the Abbasid period. As was the case with the figural and animal designs, such vegetal motifs also seem to have had universal appeal in the central Islamic lands at this time and can be seen in the Jazira [400] and Ayyubid Syria [411] as well as in contemporary Saljuqid Anatolia [423].

Turning now to the glazed pottery produced in these areas during this period, it appears that we have two parallel and interconnected ceramic traditions. One was located in an as yet unknown Syrian production centre253 and the other most probably in or near the Anatolian city of Konya. As regards Ayyubid Syria, the production of lustre-painted ceramics was most probably a continuation of and a further development upon the earlier so-called Tell Minis ware [322] - the first pottery produced in that country to be decorated in this technique. It is generally assumed that the art of lustre-painting on pottery entered the repertoire of the Syrian potters from Egypt, brought by migrating ceramists during the decline of the Fatimid dynasty in the latter half of its hegemony. The basin [411] is a paradigm of Ayyubid Syrian lustre-painted pottery, bearing as it does the characteristic chocolate-brown lustre combined with underglaze-painted blue, the organization (with its abundant metal prototypes) of the various calligraphic, geometric, and vegetal designs into a series of concentric bands interrupted by medallions; and a background of tightly coiled spirals reminiscent of engraved or chased scrolls on contemporary metalwork as well. The shape of this particular vessel and of other lustre-painted objects from this centre also echoes those in metal.254 Another, considerably less common, variety of Syrian pottery in this technique is represented here by the jar [412]. Unlike the other various Ayyubid pottery types discussed whose centre of production has not yet been isolated, both of the cursive inscriptions on this storage vessel bear witness to the fact that Damascus was producing exquisite lustre-painted ware in the thirteenth century, stating that it was made for Asad al-Iskandarani by a certain Yusuf in that city. The principal decoration on this vessel is a bold angular calligraphic design in lustre on a deep cobalt-blue ground.255

Such a design was also popular on carved and monochrome glazed or glazed and lustre-painted vases from Syria.256 The decorative technique employed on the latter ware must have developed from the incised so-called Tell Minis ware, which, in turn, owed a great debt to the monochrome sgraffiato ware so popular in Egypt during the Fatimid period [327]. The same Syrian kilns must also have produced the moulded and monochrome glazed objects among which we find shapes borrowed from other media, such as the low triangular, square, rectangular, and octagonal tables with relief decoration or ornamentation derived from turned wooden originals. Other examples are pierced mosque lamps imitating metal prototypes.257

Another technique employed in Syria at this time was that of underglaze painting, either in black under a clear colourless or turquoise-blue glaze [413] or a variety in which red, black, and blue designs were painted under a clear colourless glaze.258 Derived from calligraphy, the decoration on this bowl, so popular at the time, bears close comparison to that on the lustre-painted vase made in Damascus [412]. Lustre-painted as well as monochrome and polychrome underglaze-painted pottery was likewise produced in or near Konya. As was the case with their architecture, the Saljuqs of Rum also emulated their Syrian neighbours vis-à-vis their ceramic production and, consequently, some of it exhibits the strong Ayyubid influence.

Structures that stand to this day in central Anatolia as well as those revealed during excavations attest to the Saljuq fondness for covering the walls of their buildings with tiles arranged in geometric patterns. The hexagonal grouping [414] which probably came from the palace of Qilich Arslan II (r. 1156-92) at Konya exhibits the technical and iconographical influence of Syrian objects in the star-shaped underglaze-painted tile with the sphinx and the technical influence of Persian mina'i ware on both types of the four-sided tiles [272]. The tiles [415] belong to those from a somewhat later royal residence, Kubadabad on Lake Beysehir, founded by 'Ala' al-Din Kayqubad I in 1227. Both types of lustre-painted tiles comprising the panel betray their dependence on the so-called Tell Minis variety of Syrian pottery [322] vis-à-vis the iconography and style employed as well as the convention of incising details through the lustre and lustre-painting on to a coloured ground. As regards the underglaze painted tiles from this complex seen comprising the panel [416], on the other hand, both the examples painted in black under a transparent turquoise glaze and those polychrome-painted under a clear colourless glaze are definitely related to those Syrian productions represented here [413].259 All of these examples point to the fact that the building tiles in Anatolia were produced with the help of imported or migrant craftsmen.

Tiles were not the only objects produced in these central Anatolian kilns; bowls, and perhaps other objects as well, were manufactured in both underglaze-painted varieties and with lustre-painted decoration.260

It would seem, therefore, that while the unknown centre or centres in northern Syria mentioned above were producing pottery that developed to a large extent from that manufactured in Egypt during the Fatimid period, the main influences on the ceramic production of contemporary Anatolia appear to have come both from Iran and from Syria. The Mongol invasion forced artisans working in the former country to seek new patrons. It is well documented that some of them found their way to Anatolia, and thus the possibility of Persian potters influencing Anatolian production in general is quite plausible. For example, we know - concerning ceramic architectural decoration - that a master potter from Tus in Khurasan was working in Konya. The building providing this information is the Sircali madrasa, in the central Anatolian Saljuq capital, founded in 1242; it is the earliest surviving dated example in the Islamic world of a total surface decorated with glazed tiling [417]. It may therefore be assumed that such colourful assemblages were already known in eastern Iran.261 As regards the influence from Syria on Anatolian production, the seeming interdependence discernible in so many instances has not yet been fully explored. Although there was most definitely an international vogue for certain types of pottery in this period, for the most part that produced in the various areas can be easily differentiated by such elements as style, iconography, and profile.

The Fatimid tradition of lustre-painting on glass exemplified by the diminutive vessel [334]262 appears to have led directly into gilded and/or enamel-painted decoration on the surface of the glass vessel. The earliest datable gilded object is the fragmentary vessel [418] bearing an inscription containing the laqab of 'Imad al-Din Zangi, Atabeg of Mosul and Aleppo (r. 1127-46).263 Its figural ornamentation bridges the gap between that found on objects of the Fatimid and that of the late Ayyubid and early Mamluk periods. This flask was probably made in Syria, as were the three earliest datable enamel-painted glass objects, a beaker bearing the name of sultan Sanjar Shah (r. 1180-1209),266 the flask in the name of the last Ayyubid ruler of Damascus and Aleppo, al-Malik al-Nasir II Salah al-Din Yusuf (r. 1237-59), and the dish [419] with the names and titles of Ghiyath al-Din Kaykhusraw II (r. 1237-46), the son of the imperial founder of Kubadabad in central Anatolia. The design on the outer wall of the dish in particular clearly shows an indebtedness to the decoration on lustre-painted objects, as it is drawn from the same repertoire.

The gilded and enamelled tazza [420] exhibits both techniques combined in a masterful, yet tentative, manner bearing an abundance of gilding and the almost experimental application of enamel colours in a highly varied palette (red, blue, yellow, green, white, and black). The style, scale, and rich decorative vocabulary of designs (including entertainers, geometric patterns, arabesques, virtual bestiaries of both real and fantastic animals, and secular inscriptions) displayed in the horizontal bands of varying widths are typically found on various media from the first half of the thirteenth century. However, it is in metalwork that the closest parallels are encountered.

Tapering beakers with outward-curving rims were the most popular shape for enamelled and gilded glass at this time. However, other characteristic forms include rosewater sprinklers (Arabic qumqum) with tall tapering necks, straight-sided mugs, and basins. The production of enamelled and gilded mosque lamps with high, wide, flaring necks also began during the period under discussion here, reaching its peak in the fourteenth century.

The popularity of enamelled glass among the Frankish invaders of the Holy Land is attested not only by fragmentary vessels found in the ruins of their chateaux and by objects brought back for deposit in European churches but also by the fact that this deluxe technique is one the Franks seem to have copied from the Muslims while they were in Syria: an example of this is the beaker in the British Museum signed by Magister Aldrevandin.267

Unquestionably such glass products also greatly influenced those of Venice, the foremost European glass-manufacturing centre - especially those from the formative years of its industry. The mystique of these vessels continued to enthrall the West long after the Middle Ages, as is evident from Ludwig Uhland's poem 'Das Glück von Edenhall' and Henry Wadsworth Longfellow's English version of it.268 The 'Luck of Edenhall' was shattered; but fortunately the enamelled Syrian beaker that inspired the poems exists today in perfect condition.269

Another decorative technique, known as marvering and combing, which appears to have been very popular for ornamenting glass especially during the Ayyubid period, has a long pre-Islamic history in the area; its ultimate origins lie in Egyptian core-formed vessels of the Eighteenth Dynasty. This type of decoration was executed by winding a thread of contrasting colour around the object and subsequently marvering, or pressing the thread into the surface by rolling the vessel on a flat stone slab. A comblike tool was then utilized to create the feather-like design. The beautiful lidded bowl [421], executed in red glass with opaque white threads, is unique for the type, being the only container so ornamented that still retains its original lid.270 We know from the Cairo Geniza documents that red (manganese-coloured) glass was a specialty of Beirut. As the provenance of the Metropolitan Museum object is reported to have been nearby Sidon, Greater Syria (and perhaps even Beirut itself) can be suggested as the place of manufacture. Its shape is echoed in both the underglaze-painted and the underglaze- and lustre-painted pottery associated with this area as well.271

Many outstanding carved wooden objects are extant or known from the central Islamic lands during this period.

One of the chief production centres seems to have been Aleppo, where several masters signed their works, and we know of at least one son following his father in making mosque furniture - a parallel to the family of potters producing mihrab tiles in Kashan.272

Using dated examples as our guideposts, we can follow the craft from its archaizing phase - exemplified by a maqsura dated 1104 - to an austere, almost abstract style found on a minbar of the Zangid ruler Nur al-Din Mahmud (r. 1146-74), dated 1163 and partially preserved in the Great Mosque at Hama, Syria. The latter is an appropriate reflection of the age of Islamic scholasticism, dominated by this puritan, even ascetic ruler, who devoted his life to waging 'war against the enemies of his faith', as he claimed in his minbar inscription, and to leading the jihad against the Crusaders.

The dated maqsura, or screen, probably intended as an enclosure for a tomb in a cemetery at Damascus,273 clearly reflects the pivotal position of Syria. By incorporating both certain principles of the abstract Style C of Samarra with elements met with on Fatimid wooden pieces, Syria's function as a bridge between the East (particularly Iraq and the Jazira), on the one hand, and Egypt and the western Islamic lands, on the other, can be clearly discerned.274 The minbar of Nur al-Din, which was crafted sixty years later than the screen, was also executed to some extent in the purely linear Style C; other sections, however, were deeply carved to produce a carefully planned lacework of spiralling, bifurcating, and intersecting stems, all of the same width. In their clarity and formality these coiling stems, which produce few leaves or flowers, are remote from natural forms. They cross and recross the arched configuration of the fillet 'which is no longer a boundary but a melody running through a fugue'. This apt description emphasizes the innate musical quality of the design.275 The closest stylistic parallels are the carvings on a stone mihrab in Mosul executed by a Baghdadi artist during the reign in al-Jazira of Nur al-Din's brother Sayf al-Din Ghazi (r. 1146-49). A more developed version of the same type can be seen on a wooden door of 1209 donated by the reigning caliph al-Nasir (r. 1180-1225) to a sanctuary in Samarra.276

The most important example of the wood-carvers' art crafted in the central Islamic lands during this time, however, was the (now destroyed) minbar Nur al-Din ordered forthe Aqsa Mosque in Jerusalem, begun in 1168 but not finished until 1174, after the donor's death [422]. The minbar was to be a thank-offering for the reconquest of the holy city, which Nur al-Din did not live to see; in the meantime it was kept in the Great Mosque at Aleppo, the city where three carvers had produced it under the guidance of a master 'unequalled in the perfection of his art', in the words of the historian Abu Shama (1203-68). After Jerusalem was finally taken in 1187, Nur al-Din's successor as leader of the jihad, the famous Salah al-Din (Saladin), placed the minbar in its destined home.277

It is clear that this style could hardly have been developed further. A pair of wooden doors dated 1219 in the citadel of Aleppo reflects a different tendency, namely towards geometric configurations related to those developed in Egypt after the middle of the twelfth century. Nevertheless, the Syrian craftsman's mastery is still amazing, however, and the composition of eleven-pointed stars interlaced with twelve- and ten-pointed stars has been described as an 'almost unsolvable problem' and the 'most complicated design ever produced by that branch of art'.278

The wood-carvers' art is also beautifully represented in Saljuq Anatolia. An outstanding example is the folding wooden Qur'an stand (Turkish rahle) made, according to its elegant cursive inscriptions, in 1279 for the tomb of the mystic poet and saint Jalal al-Din Rumi in Konya [423].279 The four outer surfaces, all originally painted and gilded, are each carved with a rhythmic arabesque composition which exhibits an indebtedness to earlier Artuqid vegetal designs [400]. On the two upper surfaces is a twice-repeated design - executed in gold, black, red, and blue of a doubleheaded bird of prey on a field of arabesque scrolls inhabited by fourteen lions. On three similarly decorated doors - one from the Haci Hasan Mosque in Ankara,280 another from the public soup kitchen (Turkish imaret) of Ibrahim Bey in Karaman and the third now in the Museum für Islamische Kunst in Berlin from an unknown building [424],281 the basic repertory of arabesques and inscriptions is enriched by an infinite repeat pattern in the large central medallion; in addition, despite the religious and public settings of the doors, there are bold representations of confronted lions and addorsed griffins, as well as of paired dragons and frontally oriented human figures that are also highly reminiscent of decorative motifs seen or mentioned earlier in both Iraq and the Jazira.

The Early Islamic vogue for adorning the interior walls of palaces or private homes with either moulded or carved stucco decoration which we have discussed as regards Abbasid Samarra and Samanid Nishapur continued in the subsequent period and seems to have been particularly popular in Anatolia during the reign of the Saljuqs of Rum. The example of this art [425] was excavated in Kubadabad on Lake Beysehir, where it formed part of a wall of recessed cupboards not unlike that found in Samarra [83]. The buildings in this complex (the summer palace of Ala al-Din Kayqubad - r. 1219-37) when newly completed must have been quite dazzling with their minutely decorated stucco panels and colourful lustre- and underglaze-painted tile assemblages [415, 416]. The style exhibited here, especially of the peacocks, as well as on other stucco panels from this residence and from contemporary palaces and pavilions in Konya, is related to that used for the animals, birds, and human figures depicted on the unglazed ceramic water storage jars (habbs) made in the Jazira [409] - a direct line of influence we have been able to follow here on objects in many media other than stucco.282

The fashion for ornamenting buildings in Iraq and the Jazira with figural decoration has already been mentioned, and the stone niche from the latter area [426] is a striking example of this vogue. Although it incorporates the shape as well as the design layout that is typical for mihrabs from this area, the array of courtiers depicted in many of the interconnected niches of this architectural element preclude its use as the focal point of a qibla wall and another function, perhaps that of a fountain or a throne niche,283 must be sought. It is the tradition of figural architectural adornment not only in the Jazira but also in Iraq that helped give rise to the Rum Saljuqid proclivity for ornamenting the walls of their palaces and private residences with such decoration in various media. The stucco cupboard and the wooden doors discussed above and the stone relief [427], with a winged and crowned figure moving spiritedly to its left - one of many such reliefs in the citadel in Konya itself - are only a few examples of this prevalent vogue.284



THE ART OF THE BOOK

The carpet pages [428] comprise the initial double-page illumination from the twenty-eighth juz' of a Qur'an made for the library of a Zangid prince who ruled Sinjar, Khabur, and Nisibin in the Jazira from 1198 to 1219.285 This section is one of five extant parts of the only Zangid Qur'an from the Jazira known to have survived; and, it is the only juz' providing information about the provenance and date of this manuscript. The calligraphic, vegetal, and geometric decoration lavishly executed in gold is closely related to that found adorning a number of the objects that were produced in the same area or in realms with close ties to this branch of the Zangid dynasty. The symmetrical arabesque designs on this frontispiece, for example, are particularly reminiscent of those on contemporary metal, ceramic and glass objects as well as those on textiles and wood [398, 400, 411, 415, 416, 420, 423].

Thanks to the identification of five sections of the Qutb al-Din Qur'an, it is possible to begin to follow the evolution of the art of book illumination in the Jazira. Since each extant juz' is fully illuminated, the decoration within these five codices can be used like a pattern book for manuscript illumination around the year 1200 in the Jazira, about which we previously knew almost nothing.286 The ornamentation of these folios helps us to understand better the tradition that was soon to give rise to the superb illumination found in the earliest surviving copy of the Masnavi of Jalal al-DinRumi most probably produced in Konya and now in the Mevlana Museum in that city. Consisting of six splendidly bound volumes comprising a total of 613 folios, this manuscript is dated 1268-69 and was calligraphed by 'Abd Allah ibn Muhammad al-Konyali and illuminated by Mukhlis ibn 'Abd Allah al-Hindi. Each volume opens with a double-page composition illuminated in gold of such exquisite execution that the manuscript has been called 'one of the finest - if not the finest - illuminated Islamic manuscript of the thirteenth century' [429].287

The illumination of the so-called Qarmatian Qur'an [430] - its original volumes are still in Istanbul but many of its leaves are scattered around the world - shares certain features with the ornamentation of the two Qur'an manuscripts just discussed as well as with the decorative motifs employed on other media produced in the central Islamic lands at this time.288 To cite just two examples, the bold arabesque design gracing the lower border of the leaf illustrated here is strikingly similar to that filling the illumination from the Masnavi [429]; and the vegetal rinceau and its background of tightly coiled spirals which characteristically completely fills the spaces between the lines and letters of the folios of the so-called Qarmatian Qur'an is very commonly found on the group of Syrian lustre-painted pottery exemplified by the basin [411] as is the outlining of the script itself.

As regards the art of manuscript illustration in the central Islamic lands at this time, that practised in Iraq, the Jazira and Syria was so closely related in theme, style, and iconography that it seems best to treat it under a single heading. Thematically the material falls into two or perhaps three major groups: illustrations of technical and scientific subject matter which served as visual aids to ensure proper identification and to facilitate explanation; illustrations accompanying works of belles-lettres; and possibly, as a third group, illustrations for philosophical treatises.

The first category consists mainly of self-contained pictures accompanying works by authors such as al-Sufi, Ibn al-Ahnaf, Ibn Bukhtishu, al-Zahrawi, and al-Jazari, as well as the anonymous writers known as Pseudo-Aristotle and Pseudo-Galen. Arabic translations of works by Dioskorides and Heron of Alexandria were also copied and illustrated.289 The subjects depicted range from personifications of constellations to animal representations, illustrations of veterinary procedures, medicinal plants, surgical instruments, and automata. Episodic action is relatively rare and reflects the influence of the next thematic category.

This second group consists primarily of illustrated copies of two books that were very popular during the medieval period. The first, Kalila wa-Dimna, was a compendium of fables named after the two main characters, a pair of jackals. Allegedly composed by the wise Brahmin Bidba (or Bidpai), it belongs to the literary genre of 'mirrors for princes' which embodies precepts for rulers regarding good governance - here made more palatable by the animal guises of the characters. These stories originated in the Indian Panchatantra, of which a Middle Persian version had been translated into Arabic under the title Kalila wa-Dimna in the eighth century.290 The second 'best seller' belongs to the literary genre called maqamat ('assemblies' or 'entertaining dialogues'), an indigenous Muslim creation. The first of its kind was written in Arabic by the Iranian al-Hamadhani (968-1007). However, the one authored by the Iraqi al-Hariri (1054-1122) was the most popular of the type, and it was his Maqamat that was most frequently copied and lavishly illustrated during this period.

The text of al-Hariri's Maqamat consists of fifty picaresque tales narrated by al-Harith ibn Hammam, each set in a different part of the Muslim world. In every story a group of people is so overwhelmed by the astounding eloquence and erudition of an aged stranger, Abu Zayd of Sarub that in the end they amply reward him with money, which he as often as not spends improperly. The real purpose of the book is to demonstrate the most elaborate linguistic fireworks; and, therefore, only the barest indications of action and setting are given. Nevertheless, several copies are enriched by a series of imaginative compositions in appropriate, often quite detailed, settings in which the characters express by attitude and gesture the liveliest interest and even active participation in the events depicted. The wealth and variety of scenes - often depicting several episodes in one story - is astonishing, and varies from manuscript to manuscript.292 There are episodes on land and sea; in towns, villages, and deserts; indoors and outdoors; involving human beings, animals, or both. Scenes in mosques and palaces occur, but those of everyday urban life constitute the characteristic settings. Keen insight into the psychology of situations and personality types is the hallmark of this art. Also noticeable is a tendency towards satire, directed against the Turkic ruling class,293 which reveals sentiments apparently shared at that time by much of the Arab population of the area. Indeed, these miniatures provide a unique mirror of contemporary civilization.

The panoramic view of a village [433] is representative of the close attention paid in the Maqamat paintings to details of quotidian existence in the multifaceted Arab mercantile society, a characteristic that makes these illustrations highly reflective of this specific milieu. The unusually detailed vignettes punctuating several extant copies of this manuscript inform us better than those in any other medium about contemporary daily life in the Arab world. As has been stated earlier in this work, the inventiveness of the illustrators of manuscripts such as these influenced the rich repertory of Jaziran metalworkers.

A special kind of painting, common to both thematic groups of miniatures, is the frontispiece. There seems to have been no hard and fast rule, but scientific treatises are generally introduced by 'author portraits', while works of belles-lettres frequently include idealized 'portraits' of rulers, sometimes of the patrons of the books.294

The not well represented and therefore vaguely defined third group - illustrations for philosophical treatises or wisdom literature - exhibits most features of the second group in addition to the introductory 'author portraits' of the first.295

Three major stylistic categories can be established in the art of manuscript illustration at this time, which to some extent cut across the boundaries of the thematic groups and occasionally, it would seem, reflect regional origins. At present, however, the number of manuscripts and the historical data they contain are too limited to permit further, more precise classification.

In the first stylistic category there is heavy reliance on Byzantine prototypes such as scientific works and bibles, gospels, and lives of saints.296 Although the human figures are shown in turbans and caftan-like garments, quite often their postures and even groupings derive from Greek manuscripts. Identical copies of whole compositions, feature by feature, are rare; usually specific elements have been adapted and rearranged, and motifs from other sources as well as contemporary additions are incorporated. Most important from an artistic point of view is the frequent 'humanization' of the figural scenes: personal relations exist between figures, usually between a speaker and a listener; and, compared to the Greek models, the action has far greater immediacy and relevance.297 On the other hand, vegetal forms are more stylized, and the treatment of animals more varied: sometimes they are conventionalized, but often they take on human traits [431]. The influence of Greek originals is especially clear in the three-dimensional modelling of figures by means of shading and the relatively natural fall of garment folds.

In this category are to be found many of the illustrated scientific texts that have survived from this period, of which the finest are two manuscripts of Dioskorides' work on the pharmacological properties of plants, De materia medica - one dated 1224, the other 1229 [432].298 Of several al-Sufi manuscripts, the most spirited are in Istanbul, Paris, and London, the first dated 1130, the last two undated but of the thirteenth century.299 Some literary works, primarily a 1222-23 copy of al-Hariri's Maqamat, also belong to this group.300 The Dioskorides copy of 1224 and a Pseudo-Galen of 1199301 also incorporate features from the second stylistic category to be discussed below, pointing up the cross-fertilization that occurred in much of the art of this period. The colophon of the 1229 Dioskorides contains expressions in Syriac, perhaps indicating an origin in Syria or the Jazira; furthermore, some of the architectural details in al-Hariri's Maqamat of 1222-23 and two related copies of Kalila wa-Dimna are characteristic of the area of Aleppo.302

The second stylistic category, apparently without a single firm tradition, is basically original to this period and area.303 This group is best represented by three fine manuscripts of al-Hariri's Maqamat: an undated one in St Petersburg, probably executed between 1225 and 1235; a second painted by Yahya ibn Mahmud al-Wasiti in 1237 [433]; and, possibly the most elaborate but unfortunately also the most desecrated by iconophobes, a copy made during the reign of the last Abbasid caliph, al-Mu'tasim (1242-58).304

The emphasis here is on action and on realistic detail. Instead of moulding the body, garments swirl around it under the impetus of rapid motion, and this effect is underscored by energetic gesture and lively facial expression. Despite the originality of this 'Iraqi action style', medieval artists in the Muslim world, as in Europe, customarily worked from earlier models, so that related or parallel sources probably furnished catalysts, if not prototypes; the most likely are the brightly coloured figures and evocative scenery from the shadow plays, to which there are many references in contemporary Arabic and Persian literature.305 Whatever their sources, these Maqamat miniatures must be regarded as outstanding pictorial creations of the period and the finest ever produced in the Arabic-speaking world. Indeed, the Iraqi action style had so much vitality that it survived the sack of Baghdad by the Mongols in 1258 and the collapse of the ruling caliphate. A double frontispiece with 'author portraits' in the same animated manner appears at the beginning of a copy of Rasail Ikhwan al-Safa (The Epistles of the Brethren of Sincerity) of 1284.306 Even as late as 1297 or 1299 the same style enlivened eleven miniatures depicting mammals in the first part of a Persian copy of Manafi al-Hayawan (Beneficial Uses of Animals) by Ibn Bukhtishu, painted in Maragha in northwestern Iran.307 After this the style disappeared.

A third, stylistic category, instead of characterizing all of the illustrations in a given manuscript, comprises two specific types of miniatures. The first is the 'princely frontispiece', where the enthroned, frontally rendered, ruler is flanked by attendants standing stiffly at attention or ready to serve - the latter being depicted in a smaller scale than the potentate himself. These compositions are akin to the royal representations on Sasanian rock reliefs, which were also adapted to various early Islamic scenes [194]. They occur particularly in manuscripts attributable to Iraq, including several volumes of one copy of Kitab al-Aghani (The Book of Songs) probably made for a prince of Mosul in the second decade of the thirteenth century [434] and the Maqamat of al-Hariri dated 1237 and discussed previously.308 The second type is the antithetical scene, especially as found in Kalila wa-Dimna illustrations, in which pairs of animals flank central trees or plants.309 This device, too, is known from Sasanian and early Islamic silver, stuccoes, and silks. The fact that these two types of miniatures occur in manuscripts attributable to several different regions of the Muslim world attests to the thoroughness with which earlier traditions had been absorbed into the arts of the medieval Islamic world.

Let us turn, finally, to the illustrations of an undated copy of the romantic Persian poem Warqa wa-Gulshah,310 consisting of scenes in a kind of ribbon format, wider than it is high, with the figures usually extending over much of the height between the lower and upper edges of the picture band [435]. The cultural climate in which this manuscript was created was not unlike that in which so many of the objects seen in this section were produced. Consequently, it is not surprising to see the effect in this medium as well of the tremendous displacement of artisans at this point in the history of the medieval Islamic world. We have already discussed how an enforced migration from east to west contributed greatly to an eventual blending of styles in the art of the weaver, the potter, and the metalworker. That of the miniaturist was no exception.

Thus, although certain Persian influences are discernible in this manuscript, the coloured backgrounds, ribbon format of the scenes, type of vegetation, and figural style are also all quite closely related to depictions in manuscripts probably produced in the Jazira in the middle of the thirteenth century.311 The particular type of arabesque filling the background on the miniature illustrated here and on others in the codex is to be found not only on early thirteenth-century Kashan pottery [280] but also decorating the draperies, thrones, tents, pillows and garments in two manuscripts of al-Hariri's Maqamat, one dated 1237 and the other datable to some time between 1225 and 1235 as well as in the Paris Pseudo-Galen of 1199 probably copied in the Jazira (all of which were mentioned earlier) and in Anatolian Qur'an illumination.312 One encounters the same figural style in miniature painting from the Jazira as well as in the polychrome overglaze- and underglaze-painted ceramics discussed earlier [272-275, 414, 416] from both Anatolia and Iran. Furthermore, stylistic comparisons can be made with inlaid metalwork from the Jazira and northern Syria.

Because of the blending of styles seen here, the provenance of this unique codex has long been debated. However, several representations and biographical information on the painter seem to tip the balance in favour of the central Islamic lands as the place of origin - a general provenance reinforced by a number of the comparisons discussed above. Not only is the pre-Islamic ruler depicted as a Turkic military leader but Crusader foot-soldiers, armed with a type of weapon common in medieval Europe, and Christian knights are represented in some of the paintings.313 Since one of the pages is signed in large letters by the painter 'Abd al-Mu'min ibn Muhammad, whose family originated from Khoy, Azerbaijan, and settled in Kastamonu north of Ankara, and since we know that the painter witnessed the deed of endowment for the Karatay madrasa in Konya in 1252-53, perhaps we can be even more specific. It might be safe to assume that he was living and working in the capital at that time and to suggest further that he illustrated the manuscript there some time during the middle decades of the thirteenth century.314

The number of precisely dated or datable leather bindings extant from the period covered by this volume is extremely small. The two previously discussed examples of this art [120, 155], dating from the end of the ninth and end of the tenth centuries respectively, both exhibit the horizontal format common during the early Islamic period. The example [436], datable to 1182 or slightly earlier, and thus approximately two hundred years later than the binding [155], exhibits several new characteristics which were to dominate the art of bookbinding in the Islamic world for centuries. The first of these is the three-part construction of the binding, consisting of an upper cover (missing here), lower cover, and, attached to the fore-edge of the latter, a pentagonal envelope flap.315 Thought to have made its first appearance in the eleventh century, this classic type was to remain an intrinsic feature of Islamic bindings at least until the eighteenth century, when the influence of those from Europe brought about a slow disappearance of the traditional fore-edge flap. Another new characteristic seen here is the vertical format that was to be so universally popular from the medieval period onwards. That this is an early example of the new orientation is seen in the fact that the only clue to the vertical format on the binding itself is the lack of a central border at the sides of the back cover, thus rendering the design higher than it is wide. Finally, we see here an early example of the use of triangular corner designs in the central rectangle, a convention that was to remain popular not only for Islamic bindings but for those of the Renaissance as well.316 This tooled binding can be attributed to Damascus on the basis of two notations in the binding's manuscript.317



CONCLUSION

The medieval arts of Iraq, the Jazira, Syria, and Anatolia do not lend themselves to simple and easy generalizations. Because of its historical and ideological associations, the patronage of the Abbasid caliphs in Baghdad still had wide repercussions all over the Islamic world, but the rest of the area was governed by the twin powers of feudal military rulers loosely organized into dynastic families and by an urban middle class of merchants and landowners. Both groups invested heavily in the building of what may be called Sunni 'Islamic' cities, with many mosques, madrasas, and other establishments reflecting the social piety of the times. Non-Muslims were part of the picture, especially Christian communities, which witnessed a considerable artistic revival (to be sketched in Chapter 8). Princes also built citadels, usually better preserved than palaces, while rulers as well as the middle class profited from a network of fancy khans or caravanserais for local and international trade.

Both groups of patrons were served by an inlaid metalwork with nearly identical iconographic programmes, while the most original book illustrations of the Maqamat were restricted to the literate Arab middle class. In objects and manuscripts as well as in architecture, there was a fair amount of ostentation. Frontispieces in books or ornate portals in buildings are both instances of a concern for display, possibly illustrating rivalries between patrons and artisans. Families and even dynasties of artisans, best known among metalworkers, probably travelled from city to city or court to court, wherever there was an opportunity for lucrative commissions. In general, while, thanks in part to archeological work, it is possible to identify some of the regional differences in the arts of the object, it is the similarities that seem to overwhelm, especially in the art of inlaid metalwork and in the art of the book. But these preliminary conclusions must be tested against further research. In short, there was a distinctive patronage within central Islamic lands and a shared supply of craftsmanship as well as a common vision of a structured urban environment and of the implements needed for a satisfying life.

When we move to forms, matters become more complicated. Syria (including Palestine in the thirteenth century) and Anatolia are two well-defined artistic spaces, comparable to but different from each other. The comparison is particularly striking in architecture. The same pious or secular functions are translated into buildings, primarily in stone, utilizing the same vocabulary of structural and decorative forms (portals, iwan, court, portico, etc.). Syria exhibits an almost classical sobriety and purity in the treatment of stone, with a sharply defined geometry of decoration, and a preference for elegant but restrained ornament and writing. Anatolia, on the other hand, shows much more variety, more inventiveness in ways to build and decorate, and fewer inscriptions. Some of the portals exhibit a baroque virtuosity and probably illustrate individual experiments or reflections of some unique circumstances. Brick occurs as well as stone, and Syrian forms cohabit with Iranian ones. Anatolian peculiarities can be explained by the fact that it was a newly Islamized province at a major frontier between Islam and the Christian world, with many non-Muslim or recently converted groups, with a fluid sociey of immigrants from many parts of the rest of the Muslim world, even with sectarian communities at the edge of Muslim orthodoxy. Most of the Anatolian development is also later by a generation or two than the Syrian one and depended in part on the latter's achievements. Some scholars have also sought to explain Anatolian art through the introduction of practices and ideas brought by Turks from Central Asia. Less clearly delineated than their Syrian counterparts, objects made in Anatolia illustrate most of the techniques found elsewhere, with a possible tendency to greater complexity in design.

The two remaining provinces comprising the central lands - Iraq and the Jazira - remain, with one exception, less clearly focused. Their architectural monuments are not well preserved, and whatever remains can easily be related to those in Syria or Iran. The spectacular 'Mosul' school of metalwork is a relatively late phenomenon and could be interpreted as reflecting the needs of a social class - the feudal rulers and the wealthy patricians - more readily than the practices of a region. Only the art of book illustration seems to have appeared in these provinces much more frequently than elsewhere, for reasons which are not really clear. It is possible that, just as geography has divided the Jazira into many discrete independent zones, so the arts of this area will eventually be defined through the study of smaller and physically separate regions.

But there is yet another way to look at the arts of the central Islamic lands and of defining their character and their evolution. Following a pattern begun by Turkish scholars dealing with Anatolia and, to a smaller degree, by other scholars involved with Syria, one can argue in primarily dynastic terms. One could thus identify an Artuqid art (primarily in the first half of the twelfth century) centred in the northern Jazira, with features borrowed from many surrounding areas and with relatively less formal cohesion in architecture than in other arts. A late Abbasid flowering occurred around Baghdad in the thirteenth century, identifiable simultaneously by the Mustansiriyya, the calligraphical systematization attributed to Yaqut al-Mustasimi, and the illustrations of the Maqamat. Zangid art could be seen in the area of Mosul in the middle of the twelfth century and then in Syria at the time of Nur al-Din, an art that led directly into that of the Ayyubids, with many major accomplishments in Palestine, Arabia, and Egypt after the defeat of the Fatimids and of the Crusaders. The rich memory of Late Antiquity in Syria was adapted to new functions and new tastes. Finally, the Saljuqs of Anatolia, as well as a few secondary Anatolian dynasties established in the area, created an unusually original art in a newly Muslim area, bringing together functions and forms from Syria, Iraq, or Iran and mixing them with the rich heritage of Byzantium, Armenia, and Georgia, not to speak of an Anatolian Late Antique.

There was, of course, much that all these dynasties shared, ideologically as well as in terms of taste, and all of them profited from the accrued wealth of urban trade and manufacture, but there were many differences between them. The latter are most visible in architecture, because architecture has been better studied than other arts, but it should be apparent as well in ceramics after the many archaeological enterprises have put their information together, and in metalwork or the art of the book. It should be added that Muslim patronage was never alone. There was a significant Christian component within the Muslim empire, but, more importantly, this was a time of constant contacts with the Christian worlds of the Crusaders and of Byzantium or of other eastern Christian realms. It was a time when the Christian and Jewish population of Syrian and Anatolian cities profited from the general prosperity. Altogether, just as with the Fatimids (although perhaps less obviously), it may well be a common Zeitgeist which inspired the astounding creativity of the times.

Finally, it is worthwhile to ponder about the art of these centuries throughout the Muslim world between Central Asia and India to the east and present-day central Algeria to the west. This vast area, united by comparable social, political, linguistic, cultural, ideological, and political changes, was also connected by the functions and forms of its arts. However blurred the distinctions between them may be, a patronage of cities coexisted throughout with that of princes. Everywhere, but at different rhythms, representations appeared on objects and in books; an iconographic language came into being to illustrate or to adorn ambitions at many levels of society. The Arabic language still dominated, but Persian became a major vehicle for expression, and the writing of both acquired a sophistication of form which was hitherto unknown. But perhaps the most important achievement of these centuries was to have created social and economic conditions as well as a creative energy which led to the invention and spread of technical means to produce a true art of objects in all media and accessible to large segments of the population.



ENDNOTES

1. S. D. Goitein, A Mediterranean Society, 3 vols (Berkeley, 1967-77), among many studies by this scholar, and essays by B. Lewis, G. von Grunebaum, O. Grabar, and others in A. Raymond, M. Rogers, and M. Wahba, eds, Colloque International sur L'histoire du Caire (Leipzig, 1973). See now the exhibition catalogue Trésors fatimides du Caire (Institut du Monde Arabe, Paris, 1998) or its Viennese version Schätze der Kalifen (Vienna, 1998) and the forthcoming volume edited by M. Barrucand of the congress on the Fatimids held in Paris in 1998.

2. G. Maçais, La Berbèrie musulmane et l'Orient (Paris, 1946, repr. Casablanca, 1991).

3. The main book on Fatimid architecture is K. A. C. Creswell, Muslim Architecture of Egypt (Oxford, 1952). More recent studies are mentioned with the monuments or issues they discuss.

4. Creswell, MAE, 1-10; Marçais, Architecture, 78 ff.; S. M. Zbiss, 'Mahdia et Sabra-Mansouriya', Journal Asiatique 244 (1956), 78 ff.; Alexandre Lézine, Mahdiya (Paris, 1965).

5. For two different views on the interpretation of the remains see Creswell, MAE, 3-5, and Marçais, Architecture, 90-92; mostly superseded by Lézine, 17ff.

6. Marçais, Architecture, 78-79; Zbiss, 'Mahdia', 79 ff.

7. M. Canard, 'Le Cérémonial fatimite et le cérémonial byzantin', Byzantion 21 (1951). See now P. Saunders, Ritual, Politics, and the City in Fatimid Cairo (Albany 1994).

8. Creswell, MAE, 5-9; Marçais, Architecture, 69-70; Lézine, Mahdiya, 65 ff., with many improvements in interpretation.

9. Zbiss, 'Mahdia ', on the whole theme see G. Marçais, 'Salle, Antisalle', AIEO 10 (1952), 274 ff.

10. L. Golvin, Le Maghreb Central à l'époque des Zirides (Paris, 1957).

11. Latest statement by Lucien Golvin, Recherches archéologiques à la Qala des Banu Hammad (Paris, 1965).

12. F. Gabrieli, 'Il palazzo hammadita di Bigaya', in R. Ettinghausen, ed., Festschrift für Ernst Kühnel (Berlin, 1959).

13. Golvin, Recherches, 123 ff.

14. Ugo Monneret de Villard, Le pitture musulmane al soffitto della Cappella Palatina in Palermo (Rome, 1950).

15. Maqrizi, Khitat, 2 vols (Cairo, A.H. 1270).

16. Published in MMAT and MIFAO.

17. M. van Berchem in MMAF 19 (1903), with supplement by G. Wiet in MIFAO 52 (1929).

18. For Fustat exacavations see the numerous reports by G. Scanlon, especially in the Journal of the American Research Center in Egypt (ARCE); for recent summaries R.-P. Gayraud, ed., Colloque International d'Archéologie Islamique (Cairo, 1998) and Annales Islamologiques 29 (1995).

19. The origin of the word is discussed by Creswell, MAE, 21-22.

20. M. Herz, Die Baugruppe von Qalaun (Hamburg, 1919).

21. In MMAF I (1887-89).

22. E. Pauty, Les Palais et les maisons d'époque musulmane au Caire, MIFAO 63 (1933).

23. Maqrizi, Khitat 1, 432-33.

24. Nasir-i Khosrow, Sefer-Nameh, trans. C. Schefer (Paris, 1881), 127 ff.

25. Ten are enumerated by Maqrizi, Khitat I, 465 ff.

26. M. Canard, 'La Procession du nouvel an', AIEO 13 (1955), based on Inostrantzev's great work on the subject; see also Saunders, Ritual.

27. Creswell, MAE, 129 ff.; Aly Baghat and A. Gabriel, Les Fouilles de Foustat (Paris, 1932). Scanlon in particular has commented on many such houses.

28. Creswell, MAE, 59, fig. 23.

29. Whether this was so hinges on the interpretation of Maqrizi, Khitat 2, 273; 2, 21-27; Creswell, MAE, 36, believed that there were domes at the corners of the hall of prayer, although the text mentions a dome 'in the first arcade to the right of the mihrab'.

30. Maqrizi, Khitat, 280-81.

31. For a new interpretation of the Hakim mosque see Jonathan Bloom, 'The Mosque of el-Hakim in Cairo', Muqarnas I (1982).

32. Creswell, MAE, 94 ff.

33. S. Flury, Die Ornamente der Hakim- and Ashar-Moschee (Heidelberg, 1912), whose comparative material is, however, much out of date.

34. Creswell, MAE, 104.

35. I. Bierman, Writing Signs, the Fatimid Public Text (Berkeley, 1998).

36. H. Stern, in Ars Orientalis 5 (1963); O. Grabar, The Shape of the Holy (Princeton, 1996), 135-69.

37. O. Grabar, 'The Earlier Islamic Commemorative Structures', Ars Orientalis 6 (1966), 7-46. For a different view see Youssouf Ragheb, 'Les Premiers Monuments funéraires de l'Islam', Annales Islamologiques 9 (1970), 21-36. One should note the peculiar earlier occurrence of the Tabataba mausoleum in Cairo, if this is what it was; Creswell, MAE 2, 16 ff., and Grabar, II; see also C. Taylor, 'Reconstructing the Shi'i Role', Muqarnas 9 (1992).

38. Creswell, MAE, 107 ff.

39. Creswell, MAE, 131 ff., with a full account of the disaster which led to the disappearance of the inscriptions; A. M. Abd al-Tawab, Stèles islamiques de la Nécropole d'Aswan, rev. Solange Ory (Cairo, 1977).

40. J. Bloom, 'The Mosque of the Qarafa in Cairo', Muqarnas 4 (1987).

41. A very interesting group of tombs with important finds of textiles has recently been uncovered by Gayraud.

42. Maqrizi, Khitat, 2, 298.

43. Creswell, MAE I, 241 ff.

44. Creswell, MAE I, 275 ff.

45. Maqrizi, Khitat 2, 289.

46. Creswell, MAE I, II; above, pp. 29, 87.

47. Creswell, MAE I, 222 ff. and 247 ff.

48. Creswell, MAE I, 155 ff.; Max van Berchem, 'Une mosquée du temps des Fatimides au Caire', Bulletin de l'Institut d'Egypte 3 (1899), 605 ff.; Grabar, Commemorative, 27-29.

49. Creswell, MAE I, 110 ff.

50. Such is my interpretation of an example in a Coptic church which was taken by Creswell to be an experimental one (Creswell, MAE I, fig. 131).

51. For a more sophisticated explanation see J. Bloom, 'The Muqarnas in Egypt', Muqarnas 5 (1988).

52. Bloom, 'Muqarnas', 162-63.

53. Caroline Williams, 'The Cult of Alid Saints in the Fatimid Monuments of Cairo', Muqarnas 1 (1983).

54. S. D. Goitein, 'The Main Industries of the Mediterranean Area as Reflected in the Records of the Cairo Geniza', Journal of Economic and Social History of the Orient 4 (1961), 168-69.

55. For documentation on textile production, for example, see R. B. Serjeant, 'Material for a History of Islamic Textiles up to the Mongol Conquest', Ars Islamica 13-14 (1948), 88-113.

56. These data have been recorded by al-Qadi al-Rashid ibn al-Zubayr, Kitab al-Dhakhair wa-l-Tuhaf, ed. M. Hamidullah (Kuwait, 1959), trans. and annot. G. al-Qaddumi. Book of Gifts and Rarities, (Cambridge, MA, 1996) paragraphs 370-414 and by Maqrizi in Kitab al-Mawa'iz wa'l-I'tibar bi-dhikr al-khitat wa'l-'Athar I (Bulaq, 1854), 414-16, trans. P. Kahle, 'Die Schatze der Fatimiden', Zeitschrift der Deutschen Morgenländischen Gesellschaft, N.F. 14 (1935), 338-61. They form the basis of Zaky M. Hassan, Kunuz al-Fatimiyin (The Treasures of the Fatimids) (Cairo, 1937).

57. Serjeant, 'Material', 111-13, again quoting Maqrizi.

58 . In spite of the enormous numbers mentioned by Maqrizi, only about 181 carved rock-crystal pieces have been discovered so far; see K. Erdmann, 'Neue islamische Bergkristalle', Ars Orientalis 3 (1959), 201.

59. R. Ettinghausen, 'The "Beveled Style" in the Post-Samarra Period', in G. C. Miles, ed., Archaeologica Orientalia in Memoriam Ernst Herzfeld (Locust Valley, N.Y., 1952), 75, pl. X no. 3.

60. E. Pauty, Catalogue général du Musée Arabe du Caire (Cairo, 1931), pls XXIII-XXV.

61. The vogue for similarly decorated doors was to continue in Egypt for more than a hundred years. The panelled doors depicted on the stone façade of the Mosque of al-Aqmar (1125) are quite close in design to those illustrated here and also resemble the wooden doors in the mosque itself. See Figure 305; D. Behrens-Abouseif, 'The Façade of the Aqmar Mosque in the Context of Fatimid Ceremonial', Muqarnas (1992) 9, fig. 4, p. 34; and M. Jenkins, 'An Eleventh-Century Woodcarving from a Cairo Nunnery', in R. Ettinghausen, ed., Islamic Art in the Metropolitan Museum of Art (New York, 1972), fig. 20, 238. Eight similar door panels found in the area of Raqqa (Ebla to Damascus: Art and Archaeology of Ancient Syria, exh. cat., Washington, D.C., 1985, cat. no. 271, pp. 522-23) testify to the popularity of this type of architectural ornament in Greater Syria at this time as well. Its echoes can be seen as late as c. 1143 in the doors of La Martorana in Palermo, Sicily.

That such decoration continued for coffered wooden ceilings and corbels can be seen in the representations of such architectural elements in stone on Bab al-Futuh (1087); cf. Creswell, MAE I (Oxford, 1952), pls 65c, d and 66a, b.

62. See Pauty, Catalogue, pls XLI and XLII.

63. Pauty, Catalogue, pls XLVI-LVIII. Some of the themes have been analysed by G. Marçais, 'Les Figures d'hommes et bêtes dans les bois sculptés d'êpoque fatimite musulmane', Mélanges Maspero, III: Orient islamique (Cairo, 1940), 241-57. See also Jenkins, 'Eleventh-century Woodcarving' 227-30; Pauty, Catalogue, pl. XXXVIIIab; and Treasures of Islam, exh. cat. Musée d'Art et d'Histoire, Geneva, 1985, cat. no. 357, p. 343. For excellent illustrations of the series illustrated [315] and many other objects produced under Fatimid aegis and presently housed in the Museum of Islamic Art, Cairo, see Schätze der Kalifen: Islamische Kunst zur Fatimidenzeit, exh. cat., Vienna, 16 November 1998-21 February 1999.

64. See R. Ettinghausen, 'Early Realism in Islamic Art', Studi orientalistici in onore di Giorgio Levi della Vida I (Rome, 1956), 259-62. Echoes of this stylistic trend can be seen in the fragmentary wooden ceiling of the first half of the twelfth century from the Cappella Palatina, now in the Galleria Regionale della Sicilia, Palermo; cf. G. Curatola, Eredità dell'Islam: arte islamica in Italia, exh. cat., Palazzo Ducale, Venice, 30 October 1993-30 April 1994, cat. no. 86, pp. 197-98. Both of these groups of wood-carvings will be discussed in Chapter 8.

65. Islamic Art in Egypt, 969-1517, exh. cat., Cairo, April 1969, cat. no. 234, p. 246, fig. 40.

66. J. Sourdel-Thomine and B. Spuler, Die Kunst des Islam (Berlin, 1973), fig. 240, and Chapter 7, below [457].

67. See 30 ans au service du patrimoine, Institut National d'Archéologie et d'Art, Tunis, 1986, nos IV.51-54, 56, pp. 257-9.

68. E . Kuhnel, Die islamischen Elfenbeinskulpturen VIII-XIII Fahrhunderts (Berlin, 1971) pl. XCIX. For illustrations of other similar plaques in Berlin and the Louvre see also pls XCVII-XCVIII.

69. We know that this illusion of three dimensions was highly appreciated in eleventh-century Egypt from al-Maqrizi's report of an artistic competition in which an Iraqi painted a dancing girl in such a way that she seemed to be emerging from a niche, but was outdone by his Egyptian rival, who painted a dancing girl as if she were entering a niche. For references to this passage anda discussion of its significance see R. Ettinghausen, 'Painting in the Fatimid Period: A Reconstruction', Ars Islamica 9 (1942), 112-13.

70. M. Jenkins, 'The Palmette Tree: A Study of the Iconography of Egyptian Lustre Painted Pottery', Journal of the American Research Center in Egypt 7 (1968), 119-26, and R. Pinder-Wilson, 'An Early Fatimid Bowl Decorated in Lustre', in R. Ettinghausen, ed., Aus des Welt der islamischen Kunst: Festschrift fur Ernst Kühnel (Berlin, 1959), 139-43. Another large single fragment from this dish was seen by Marilyn Jenkins-Madina in 1974 in the International Ceramics Museum, Faenza, Italy, acc. no. AB1231.

71. M. Jenkins, 'Muslim: An Early Fatimid Ceramist', The Metropolitan Museum of Art Bulletin (May 1968), 359-69. Contrary to A. Contadini, Fatimid Art at the Victoria and Albert Museum (London, 1998), 80, the motif (or motifs) decorating the centre of this dish is impossible to ascertain.

72. G. Berti and L. Tongiorgi, Ceramici medievali delle Chiese di Pisa (Rome, 1981), 56 and tav. CLXXXIX top.

73. F. Aguzzi, 'I bacini della Torre Civica', Sibrium (1973-75), fig. 1, p. 191, and fig. 4, p. 192. M. Jenkins-Madina's dating for such ware supersedes that suggested by V. Porter and O. Watson in their "'Tell Minis" Wares', in Syria and Iran: Three Studies in Medieval Ceramics (Oxford Studies in Islamic Art, 4) (1987), 175-248. See also M. Jenkins, 'Early Medieval Islamic Pottery: The Eleventh Century Reconsidered', Muqarnas 9 (1992), 56-66.

74. As with the wood, it is not at all surprising to find an increase of human and animal motifs in Fatimid Egyptian ceramic objects given the traditions of their predecessors, the Aghlabids, and of their forbears in Ifriqiya - 'it was simply a part of their tradition and it couId have come into Egypt as early as they did. As to how it happened, there are many possible answers, some of which include the fact that the conquering Fatimids could have brought potters or pottery or both with them from Sabra al-Mansuriyya; trade with the Ifriqiyan capital and/or with Spain and al-Qal'a; potters emmigrating from a ruined Madinat al-Zahra or Sabra al-Mansuriyya after the bedouin invasion - the devastation of Ifriqiya more or less occurring during the time of the liquidation of the Fatimid imperial treasury.' M. Jenkins, 'Western Islamic Influences on Fatimid Egyptian Iconography', Kunst des Orients 10 (1976), 105, in response to O. Grabar, 'Imperial and Urban Art in Islam: The Subject Matter of Fatimid Art', Colloque International sur l'Histoire du Caire (Cairo, 1972), 173-89.

75. Acc. no. 14987. M. Mostafa, 'Fatimid Lustred Ceramics', Egypt Travel Magazine 2 (1954), fig. 10.

76. Museum of Islamic Art, Cairo, no. 13080. See also E. J. Grube, Islamic Pottery of the Eighth to the Fifteenth Century in the Keir Collection (London, 1976), 138-42, pl. facing 136, top. For a general discussion of ceramics see M. Jenkins, 'Islamic Pottery', Bull. MMA, 40:4 (1983). Pieces with clearly identifiable Christian subject matter fall into this undatable lustre-painted group as well. Reflecting the important role that Coptic Christians played in medieval Egyptian society, clerics, for example, are represented, and on one fragment Christ is depicted making the gesture of blessing; cf. exh. cat., Schätze der Kalifen, cat. no. 126, p.159. See also a dish with an image of a priest swinging a censer, now in the Victoria and Albert Museum, London, in A. Lane, Early Islamic Pottery: Mesopotamia, Egypt and Persia, London, 1947, pl. 26A, and see also M. Jenkins-Madina entry in The Glory of Byzantium: Art and Culture of the Middle Byzantine Era, A.D. 843-1261, exh. cat., New York, 1997, cat. no. 273, p. 417. For a discussion of the inscriptions on this bowl see M. Jenkins, 'Sad: Content and Context', in Priscilla P. Soucek, ed., Content and Context of Visual Arts in the Islamic World (University Park, PA, 1988), pp. 67-89.

77. M. Jenkins, 'Early Medieval Islamic Pottery'; also R. B. Mason, R. M. Farquhar, and P. E. Smith, 'Lead-Isotope Analysis of Islamic Glazes: An Exploratory Study', Muqarnas 9 (1992), 67-71. This category was previously thought to have come from the Garrus district of Iran.

78. The compelling parallels between the ornamentation on wares of this type and those with lustre-painted decoration are just beginning to be explored. Perhaps this find will eventually help scholars to answer the questions raised above about the date of the luster group with figural designs. One is also struck by the similarity between the decoration on these ceramic objects and that on a type of metalwork (see R. Ettinghausen and O. Grabar, Art and Architecture of Islam, 650-1250 A.D. (London, 1987), fig. 252, which has been variously dated and attributed. Perhaps this wreck can help to answer questions about the metal group as well.

79. M. Jenkins-Madina, 'Glazed Pottery', in G. F. Bass, S. Matthews, J. R. Steffy, and F. H. vanDoorninck, Jr, Serce Limani: An Eleventh-Century Shipwreck, Volume II: The Cargo (in press - Texas A & M University Press). For discussion of earlier such ware, see above, Chapter 4, p. 118 and [185].

80. Jenkins, 'Western Islamic Influences' and Scanlon 'Slip-painted Early Lead-glazed Wares from Fustat: A Dilemma of Nomenclature', Colloque International d'archéologie Islamique, Cairo, 3-7 fevrier, 1993. Besides Ifriqiya, southern Spain, and Egypt, such ware was also made in what is now Portugal cf. C. Torres, Cerâmica islâmica portuguesa, exh. cat., Fundação Calouste Gulbenkian, Lisbon, 16-27 November 1987, no. 79. It is clear from this accidental find, however, that the theory that within a century after the height of the 'Samarra revolution' sgraffiato ware 'dominated to the exclusion of many of the earlier wares' (see E. J. Grube, Cobalt and Lustre: The First Centuries of Islamic Pottery (London, 1994), 34) should not be construed to cover splash-decorated ware. Bacini in the Abbazia in Pomposa further confirm a date for this vase in the first half of the eleventh century; cf. G. Ballardini, 'Pompose e i suoi Bacini', Faenza 24 (1936), 121-28, tavola XXX. This category was previously known as 'Fayyumi'.

81. V. Porter and O. Watson, '"Tell Minis" Wares,' Syria and Iran: Three Studies in Medieval Ceramics, (Oxford Studies in Islamic Art), eds. J. Allan and C. Roberts, IV, Oxford, 1987, p. 238, figs B8a, B9; pp. 242-43, figs B24-B29, and p. 245, fig. C8. Pottery of a very similar type was also produced in Ifriqiya and Portugal e.g. G. Vitelli, Islamic Carthage: The Archaeological, Historical and Ceramic Evidence, Institut National d'Archéologie et d'Art de Tunisie, CEDAC, Dossier 2, 1981, 112, no. 1.13; and C. Torres, Ceramica, nos 5-56.

82. See Jenkins, 'Early Medieval Islamic Pottery,' fig. 19, p. 63.

83. Berti and Tongiorgi, Ceramici medievali, no. 81, pl. CLXXXVI.

84. The technique employed in executing the decoration on both of these ceramic types is identical to that used on the champlevé group except that on the former groups it is the body itself that is carved away and not simply the slip. Jenkins, 'Early Medieval Islamic Pottery,' 64.

85. It has been assumed, solely on the basis of ten fragmentary relief-cut glass vessels with calligraphic, vegetal, or figural decoration excavated at Nishapur, that such sophisticated table ware was produced in that city; cf. J. Kroger, Nishapur: Glass of the Early Islamic Period, (New York, 1995), 20-21, 137-46. Opinions are divided, but Marilyn Jenkins-Madina, for the reasons outlined below, feels much more comfortable placing the manufacture of the highly refined relief-cut glass vessels and those of rock crystal in Egypt or Iraq. Not only was the number of the finest relief-cut glass objects recovered at Nishapur exceedingly small but the fact that no kilns for any type of glass-making were discovered during the excavations at that site and no contemporary texts sing the praises of any type of Nishapur glass production makes one question, at this juncture in our study of this medium, whether glass was produced in Nishapur at this time at all. However, as our study of glass production in the Muslim world is in its infancy, this statement may have to be modified at a later date.

86. The first was that during which the earliest lustre-painted glass vessels were produced.

87. Perhaps the only other artisans to apply this exacting and difficult lapidary technique to glass with such perfect skill were the fashioners of the Late Antique so-called diatreta cups; cf. D. B. Harden, H. Hellenkemper, K. Painter, and D. Whitehouse, Glass of the Caesars (Milan, 1987), cat. nos 134-39, pp. 238-49.

88. Its mounting is not original to the object. The engraved gold, interior, mounting is dated to the tenth century and the gilded silver exterior mounting with Byzantine enamel and filigree decorated plaques set with semi-precious stones is dated to the twelfth or thirteenth centuries; cf. Avinoam Shalem, Islam Christianized: Islamic Portable Objects in the Medieval Church Treasuries of the Latin West (Frankfurt, 1996), no. 77, p. 227 and fig. 16. See also H. R. Hahnloser, ed., Il tesoro de San Marco (Florence, 1971), 2 vols, cat. no. 117. Although Marilyn Jenkins-Madina does not yet have a satisfactory reading for the Arabic word in angular script carved on the base of this bowl, the likelihood of its reading 'Khurasan' is extremely slight, as this is the name not of a city or town but of a province. This particular type of metal, which is exceedingly rare for three-dimensional objects, was not used for coin weights until the reign of the Fatimid caliph al-Aziz (r. 975-96). Two other complete objects in this metal are the carafe in the Corning Museum of Glass, Corning, N.Y.: (Islam and the Medieval West, exh. cat., State University of New York at Binghamton, 6 April-4 May 1975, cat. no. G9), and a lustre-painted object in the Metropolitan Museum of Art (see M. Jenkins, 'Islamic Glass: A Brief History', Bull. MMA, (Fall 1986), 23, no. 21). Threads of this metal are also to be seen on glass objects from the Arab world, e.g. The Madina Collection, New York, 110. G0060.

89. See Kahle, 'Schätze', 329-62, esp. items 8, 10, 27, and Priscilla Soucek, 'Mina'i, Encyclopedia of Islam, N.S. 7, 73. The green relief-cut glass vessel in Venice (cf. The Treasury of San Marco Venice, exh. cat., New York, Metropolitan Museum of Art, 1984, cat. no. 27, p. 100, and Shalem, Islam Christianized, 59 and 228, fig. 53) is, most probably, one such vessel made in imitation of an emerald receptacle. See also M. Jenkins-Madina, 'Fatimid Decorative Arts: The Picture the Sources Paint', in L'Egypte fatimide: son art et son histoire (Paris, 28, 29, and 30 mai 1998, (Paris, 1999)).

90. One of these is a closely related, and roughly contemporary (1000-08), rock-crystal ewer in the Pitti Palace, Florence (see Ettinghausen and Grabar, Art and Architecture, 194, fig. 179), which in 1998 was accidentally broken (see 'Oops, It Slipped', Art Newspaper (January 1999). The third datable object in this medium is a crescent in the name of al-Zahir (r. 1021-36) in the Germanisches Nationalmuseum, Nuremberg (Ettinghansen and Grabar, 193, fig. 178).

91. The similarly shaped glass ewer (without, however, the relief-cut decoration) found in the Northern Pagoda of Chaoyang and datable to the Chongxi reign (1032-51) of the Liao dynasty may indicate that the glass versions imitated those in rock-crystal, see An Jiayao, 'Dated Islamic Glass in China', Bulletin of the Asia Institute, N.S. 5 (1991), fig. 17. The glass ewer discussed in Chapter 5 [281] with its datable parallel unearthed in China supports a similar conclusion.

92. The Treasury of San Marco, nos 31, 32, pp. 216-27.

93. C. J. Lamm, Mittelalterliche Gläser and Steinschnittarbeiten aus dem Nahen Osten (Berlin, 1930), 1, 211, nos 10, 11 pl. 75 no. 10. For a list of other early arrivals in Europe see K. Erdmann, 'Fatimid Rock Crystals', Oriental Art 3 (1951), 142. The first section of Shalem, Islam Christianized, 17-125, is a good discussion of why Islamic objects are to be found in European church treasuries and how they got there.

94. In view of the close technical and stylistic relationship between relief-cut glass and carved rock crystal, the latter mainly worked in Egypt, and of the great number of pieces of this type of glass reported in contemporary accounts, as well as of vessels and fragments of this variety actually found there, we must conclude that such glass was manufactured in Egypt during the early Fatimid period and even before. Although no indisputable evidence of glass manufacture has been excavated in Fustat, the Geniza documents provide conclusive proof that glass (of unspecified types except for the moulded variety) was being produced in Egypt during the Medieval Islamic period; cf. Goitein, A Mediterranean Society, Vol. I, 94, 110, 363, and 365. It also seems highly probable that it was made in Iraq where, in the ninth century, Basra had a reputation as an outstanding glass producing centre (Lamm, Mittelalterliche Gläser 2, 496-98, esp. nos 76, 81-83, 91, 92). The only examples of glass dating from the ninth or tenth century to provide epigraphic evidence of an Iraqi origin, however, are lustre-painted pieces associated with Basra and the mould-blown vessels from Baghdad discussed in Chapter 2, above (R. Ettinghausen, 'An Early Islamic Glass-Making Center', Record of the Museum of Historic Art, Princeton University I (1942), 4-7; D. S. Rice, 'Early Signed Islamic Glass', Journal of the Royal Asiatic Society (April, 1958), 12-16, fig. 4, pls IV-VI). As it did in Egypt, the cutting of semi-precious stones in Iraq probably affected the glass industry, for, according to the famous Iranian scientist at al-Biruni, Basra was the outstanding centre for the carving of rock crystal (P. Kahle, 'Bergkristall, Glas and Glasflusse nach dem Steinbuch von el-Beruni', Zeitschrift der Deutschen Morgenländischen Gesellschaft, N.F. 15 (1936), 332). It must have been a well-organized craft, for al-Biruni speaks of highly paid designers who found the most suitable shape for each rock and of carvers who executed the work. Unfortunately, however, with one possible exception (a small flacon found during the excavations of Wasit, the first important town to the north of Basra on the Tigris route; Erdmann, 'Bergkristalle', 202, text figure A and fig. 4), none of the existing carved rock crystals can be definitely attributed to Basra. Sheila Blair, 'An Inscribed Rock Crystal from 10th-century Iran or Iraq', Riggisberger Berichte 6 (1998), 345-53.

95. For a carafe with a very similar decoration found at Sabra al-Mansuriyya, see G. Marçais and L. Poinssot, Objets kairouanais, IXe au XIIIe siecle: reliures, verreries, cuivres et bronzes, bijoux, Notes & Documents, XI - Fasc. 2 (Tunis, 1952), 379-82, LV, LVIII. The bodies of the animals depicted on both the illustrated beaker and the vessel found in Ifriqiya are ornamented with hatched lines, a convention very popular on pottery produced in Ifriqiya and at Qal'at Bani Hammad in present-day Algeria.

96. Jenkins, Islamic Glass, no. 31, pp. 30-31.

97. Jenkins, Islamic Glass, no. 41, p. 34.

98. Lamm, Mittelalterliche Gläser I, 109, no. 3.

99. R. B. Serjeant, Islamic Textiles, Material for a History up to the Mongol Conquest (Beirut, 1972), 135-60. See also J. A. Sokoly, 'Towards a Model of Early Islamic Textile Institutions in Egypt' and Y. K. Stillman, 'Textiles and Patterns Come to Life Through the Cairo Geniza', both Riggisberger Berichte 5 (1996), 115-22 and 35-52, respectively.

100. Trésors fatimides du Caire, exh. cat., 'Institut du Monde Arabe, 28 April-30 August 1998, cat. no. 209, pp. 232-33; one other piece in the Metropolitan Museum of Art, New York, acc. no. 29.136.4, unpublished, and that in Tissus d'Egypte témoins du monde arabe VIII-XV siecles, Collection Bouvier, exh. cat., Musée d'Art et d'Histoire, Geneva, 1993-94, cat. no. 134. The most important of these is, unquestionably, the Veil of Saint Anne, which is complete and dated and bears the name of the ruler and his vizier - thus providing the date for the few extant fragments of the same type. However, it is difficult to reproduce and, when shown in detail, the individual motifs are not rendered as beautifully as those on the New York fragment illustrated here.

101. Serjeant, Islamic Textiles, 142-43. In addition to that illustrated here and that mentioned in the previous note, the Metropolitan Museum of Art possesses two simpler fragments of this textile category. Employing a minimal amount of silk, these may have been knock-offs for the hoi polloi of the epitome in royal fashion at the turn of the eleventh century (acc. nos 1974.112.14a and 1974.113.14b).

102. Ernst Kuhnel, 'Four Remarkable Tiraz Textiles', in G. C. Miles, ed., Archeologica Orientalia in Memoriam Ernst Herzfeld (Locust Valley, N.Y., 1952), 144-49.

103. The largest published group of such animals of copper alloy is to be found in G. Migeon, Manuel d'art musulman (Paris, 1927), figs 182-91; the only more critical evaluation is by K. Erdmann, 'Islamische Giessgefasse des II Jahrhunderts', Pantheon 22 (1938), 251-54. See also E. C. Dodd, 'On the Origins of Medieval Dinanderie: The Equestrian Statue in Islam', Art Bulletin 51 (1969), 220-32; and 'On a Bronze Rabbit from Fatimid Egypt', Kunst des Orients 8 (1972), 60-76.

104. See The Art of Medieval Spain<, A.D. 500-1200 (exh. cat., New York, 1993), illustration p. 81.

105. M. Jenkins, 'New Evidence for the Possible Provenance and Fate of the So-Called Pisa Griffin', Islamic Archaeological Studies I (1978) (Cairo, 1982), 79-85. For two different views see A. S. Melikian-Chirvani, 'Le Griffon iranien de Pise: matériaux pour un corpus de l'argenterie et du bronze iraniens, III', Kunst des Orients 5 (1968), 68-86, and Al-Andalus: The Art of Islamic Spain, exh. cat., New York, 1992, cat. no. 115, pp. 216-18. A large metal lion sculpture recently sold at auction and published in Curatola, Eredità dell'Islam, pp. 128-29, Figures 43a and 43b, shares a number of striking technical and stylistic parallels with the so-called Pisa griffin. Since scientific and art historical research currently being undertaken on the lion, as this book goes to press, seems to be pointing to a European and not an Islamic provenance for that object, all prior attributions for the griffin - including that presented here -must be considered as under review at this time.

106. Judging from the decoration on these two sculptures, however, it seems plausible to assume that from the small copper-alloy figure of a long-eared deer in the Museum of Islamic Art, Cairo, no. 15062 (see E. C. Dodd, 'On a Bronze Rabbit from Fatimid Egypt', Kunst des Orients 8 (1972), fig. 13), the development progressed into the type of copper-alloy sculpture represented by the rabbit which, in turn, points the way to the griffin.

107. E. Meyer, 'Romanische Bronzen und ihre islamischen Vorbilder', in R. Ettinghausen, ed., Aus der Welt der islamischen Kunst, 317-22.

108. Art of Medieval Spain, no. 47, pp. 99-100. Dr Carboni's reading of the inscription on this object not only confirmed the sources regarding silver objects with niello decoration but provided us with proof of the beauty of their execution. See also S. D. Goitein, A Mediterranean Society, Vol. IV (Berkeley, Los Angeles, and London, 1983), 223 and note 533, p. 429. In October 1998 a massive hoard of about six hundred copper-alloy objects was excavated in Tiberias under the auspices of the Archeological Institute of The Hebrew University. Found in conjunction with coins, this large cache of candelabra, bowls, trays, jugs, oil lamps, incense burners, numerous receptacles and house-hold items such as handles and legs of furniture promises to revolutionize our understanding of the metalworking industry during the Fatimid period. In spite of numerous requests, the authors were unable to secure any photographs of this material or to ascertain the dates of the coins. Another large cache, of more than two hundred - principally metal - objects, was found in 1995 during the ongoing excavations in Caesarea conducted by the combined Caesarea Expedition and the Israeli Antiquities Authority. See Ayala Lester, Y. D. Arnon and Rachel Polak, 'The Fatimid Hoard from Caesarea: A Preliminary Report,' in L'Egypte fatimide son art et son histoire (ed. M. Barrucand), Paris, 1999, pp. 233-48. Among the metal finds are candlesticks, basins, jugs, bowls, trays, and braziers. Unlike the Tiberias hoard, however, no coins were found in the Caesarea cache.

109. Glory of Byzantium, cat. nos. 274-78, pp. 418-21.

110. For example, see Ettinghausen, Early Realism in Islamic Art, 267-69; Arab Painting (Geneva, 1962), 54-6; and E. J. Grube, The World of Islam (London, 1966), 67. See also O. Grabar, 'Fatimid Art, Precursor or Culmination', in S. H. Nasr, ed., Isma'ili Contributions to Islamic Culture (Tehran, 1977), esp. 218.

111. Since any direct Fatimid influence on the style or iconography of the wooden ceiling of the Cappella Palatina in the Norman royal residence in Palermo, Sicily (built for the Christian king Roger II in the 1140s), is impossible to prove (there being no documents or inscriptions to indicate the nationality of the craftsmen responsible) and up to now has only been conjectured, this monument will be discussed not here but in Chapter 8.

112. D. S. Rice, 'A Drawing of the Fatimid Period', Bulletin of the School of Oriental and African Studies 21 (1958),31-39, and Dalu Jones, 'Notes on a Tattooed Musician: A Drawing of the Fatimid Period', Art and Archaeology Researth Papers 7 (1975), 1-14. B. Gray, 'A Fatimid Drawing', British Museum Quarterly 12 (1938), 91-96. E. J. Grube, 'Three Miniatures from Fustat in the Metropolitan Museum of Art in New York', Ars Orientalis 5 (1963), 89-95, pls 1-6. Contrary to the view expressed by the latter author, the relationship of text to illustration on both the verso and the recto of the folio illustrated [343] is not clear and it is doubtful that Kab al-Ahbar himself is the author of the manuscript. For the most recent bibliography on such drawings, in general, cf. Schatze der Kalifen, cat. nos 20-25, 28-32, 36, 37, 41, 121, pp. 84-93, 95-96, 99, and 154-55.

113. Ibn al-Zubayr, trans. Gh. H. Qaddumi, op.cit., paragraph 413.

114. Contadini, Fatimid Art, 11, 12, figs 7, 8 notwithstanding. Other than the date of the manuscript in the Chester Beatty Library cited by her (a date that falls during the rule of a number of other Muslim dynasties besides the Fatimid), she gives no cogent and supportable reason for that particular Qur'an to qualify as the only Fatimid Egyptian Qur'an to have survived to this day.

115. These are the Geniza documents whose data has been analysed but not exhausted by S. D. Goitein, A Mediterranean Society, 6 vols (Berkeley, 1967-93).

116. The most accessible document is the Kitab al-Dhakha'ir, trans. G. Qaddumi, Book of Gifts (Cambridge, 1996).

117. These episodes have been integrated in a broad fresco of artistic collecting by Joseph Alsop, The Rare Art Traditions (Princeton, 1981), 156.

118. I. Bierman, Writing Signs.

119. O. Grabar, The Shape of the Holy (Princeton, 1996), 135 ff.

120. Nasir-i Khosrow, Sefer-nameh (tr. Albany, 1986) pp. 42-57.

121. See another form of these conclusions in O. Grabar, 'Le Problème de l'art fatimide', in M. Barrucand, ed., L'Egypte fatimide son art et son histoire (Paris, 1999).

122. R. Ettinghausen, 'Early Realism in Islamic Art', in Collected Papers, 158.

123. Some preliminary thoughts were presented by O. Grabar in 'Imperial and Urban Art in Islam', with a response by M. Jenkins in 'Western Islamic Influences.'

124. J. Bloom, 'The Origins of Fatimid Art', Muqarnas 3 (1985), pp. 20-38 hinted in this direction in a persuasive and provocative paper.

125. Topography and bibliography of Baghdad in G. Makdisi, 'The Topography of Eleventh Century Baghdad', Arabica 6 (1959); S. A. Ali, Baghdad, madinat al-salam (Baghdad, 1985), for a general survey of the city. Introductions to other cities can be found in the Encyclopedia of Islam.

126. A. Hartmann, An-Nasir li-Din Allah (Wiesbaden, 1975).

127. For the eleventh century and earlier the main source is al-Khatib al-Baghdadi, ed. and trans. G. Salmon, L'Introduction topographique (Paris, 1904); then Ibn al-Jawzi, al-Muntazam (Hyadarabad, 1938 and ff.). See also J. Lassner, The Topography of Baghdad in the Early Middle Ages (Detroit, 1970).

128. Creswell, MAE 2, 124 ff.; H. Schmid, 'Die Madrasa al-Mustansiriyya in Baghdad', Architectura 9 (1979). Hillenbrand, Architecture, 223-24.

129. L. Massignon, Mission en Mesopotamie 2 (Cairo, 1912), 41 ff.; see also F. Sarre and E. Herzfeld, Archäologische Reise im Euphrat- und Tigris-Gebiet (Forschungen zur islamischen Kunst I) (Berlin, 1911), I, 44-45.

130. E. Herzfeld, 'Damascus', Ars Islamica 9, 13-14 (1942), 18 ff.; also M. Jawad, 'Al-Imarat al islamiyah', Sumer 3 (1947), 38 ff. Y. Tabbaa, 'The Muqarnas Dome', Muqarnas 3 (1985), to be revised and put in a wider context in his forthcoming Transformations in Islamic Architecture during the Sunni Revival.

131. Herzfeld, Reise 1, 151 ff.

132. M. Awad, 'Al-Qasr al-Abbasi', Sumer I (1945).

133. Sarre and Herzfeld, Reise 2, 151 ff.; Massignon, Mission 2, 47 ff.

134. Creswell, EMA 1, 644 ff.; D. S. Rice, 'Medieval Harran', Anatolian Studies 2 (1952).

135. See Max van Berchem and J. Strzygowski, Amida (Heidelberg, 1910) I; A. Gabriel, Voyages archéologiques (Paris, 1940) I, 85 ff.; M. Sözen, Diyarbakir'da Turk Mimarisi, (Istanbul, 1971); also a general attempt at characterizing Ortoqid architecture by A. Altun, Anadoluda Devri Türk Mimarisinin Gelismesi (Istanbul, 1978).

136. Archeological investigations, especially rescue operations surrounding the building of dams on the Euphrates, are slowly bringing to light interesting new documents on the material culture of the area; see, for example, Scott Redford, 'Excavations at Gritille', Anatolian Studies 36 (1986).

137. N. Elisséev, 'Les Monuments de Nur al-Din', Bulletin des Etudes Orientales 13 (1949-50). Y. Tabbaa's thesis The Architectural Patronage of Nur al-Din, New York University, 1982.

138. Excellent introduction by M. Meinecke, 'Rakka', in Encyclopedia of Islam, 2nd ed.

139. Sarre and Herzfeld, Reise 2, 215 ff.

140. Elisséev, 'Les Monuments', 37-38.

141. Sarre and Herzfeld, Reise I, 123 ff.; D. Sourdel and J. Sourdel-Thomine, 'Notes d'épigraphie et de topographie', Annales Archéologiques de Syrie 3 (1953). The minaret has, since then, been relocated; A. Raymond and others, Balis II: Histoire de Balis (Damascus, 1995) for an introduction to the site.

142. A. Gabriel, 'Dunaysir', Ars Islamica 4 (1936).

143. Gabriel, Voyages Archéologiques, 227 ff.

144. Gabriel, Voyages archéologiques, 263 ff.

145. Gabrlel, Voyages archéologiques, 221 ff.; see also Sauvaget in AIEO 4 (1938), 82 ff.

146. Gabriel, Voyages archéologiques, 255 ff.

147. Gabriel, Voyages archéologiques, 3 ff.; see also Ara Altun, Mardinde Türk devri mimarsi (Istanbul, 1971).

148. This had been established by Gabriel and Sauvaget in Voyages, 184 ff. For interpretations see T. Allen, A Classical Revival in Islamic Architecture (Wiesbaden, 1986), and T. Sinclair, 'Early Artukid Mosque Architecture', in J. Raby, ed., The Art of Syria and the Jazira 1100-1250 (Oxford, 1985).

149. S. al-Diwahji, 'Madaris al-Mausil', Sumer 13 (1957).

150. Gabriel, Voyages archéologiques, 195 ff.

151. Sarre and Herzfeld, Reise 2, 234 ff. More recently some archaeological work has been accomplished in some of these shrines; Said al-Diwahji in Sumer 10 (1954). It should be added that the plans published by Sarre and Herzfeld are far from reliable.

152. For these see mostly C. Preusser, Nordmesopotamische Baudenkmäler (Leipzig, 1911), 2 ff. For related Christian monuments see J. M. Fiey, Assyrie chrétienne (Beirut, 1965) and Mossul chrétienne (Beirut, 1959).

153. Al-Harawi, Guide des lieux de pélerinage, trans. J. Sourdel-Thomine (Damascus, 1957), 135-59.

154. G. Bell, Amurath to Aurath (London, 1924), 48-51; Elisséev, Monuments, 36; the site has recently been investigated by A. R. Zaqzuq, 'Fouilles de la citadelle de Jabar', Syria 62 (1985); Cristina Tonghini, Qalat Jabbar Pottery (Oxford, 1998).

155. Elisséev, Monuments, 36-37.

156. Gabriel, Voyages archéologiques.

157. Rice, 'Medieval Harran'.

158. J. Warren, Art and Archaeology Papers 13 (1978); C. Hillenbrand in Raby, ed., Syria and the Jazira.

159. Sarre and Herzfeld, Reise 2, 239.

160. Sarre and Herzfeld, Reise 2, 11.

161. W. Hartner, 'The Pseudoplanetary Nodes of the Moon's Orbit', Ars Islamica 5 (1937), has explained them and provides sources for its original publication by C. Preusser.

162. For Aleppo the classical study is by J. Sauvaget, Alep (Paris, 1941); it has been much revised in recent years; E. Wirth and H. Gaube, Aleppo: historische und geographische Beiträge (Wiesbaden, 1984), present a very different view of the city. For Damascus, Sauvaget, 'Esquisse d'une histoire de la ville de Damas', Revue des Etudes Islamiques 8 (1934), 421-80, and now Dorothee Sack, Damaskus: Entwicklung und Struktur einer orientalisch-islamischen Stadt (Mainz am Rhein, 1989).

163. See e.g. J. Sourdel-Thomine, 'Le Peuplement de la région des "villes mortes"', Arabica I (1954); D. Sourdel, 'Ruhin, lieu de pelerinage musulman', Syria 30 (1953).

164. E. Herzfeld, Matériaux pour un Corpus Inscriptionum Arabicarum: Syrie du Nord: Alep (Cairo, 1954), 143 ff.; Sauvaget, Alep, and 'Inventaire des monuments musulman de la ville d'Alep', Revue des Etudes Islamiques 5 (1931), 73; Wirth and Gaube, Aleppo; and now Y. Tabbaa, Construction of Power and Piety in Medieval Aleppo (University Park, PA 1997).

165. J. Sauvaget, Les Monuments historiques de Damas (Beirut, 1932), 16; Sack, Damaskus.

166. J. Sauvaget, 'Les Inscriptions arabes de la mosquée de Bosra', Syria 22 (1941); M. Meinecke, Patterns of Stylistic Change in Islamic Architecture (New York, 1995), 31 ff.

167. Best account in Max van Berchem, Matériaux pour un Corpus Inscriptionum Arabicarum. Deuxième partie; Syrie du Sud: Jérusalem (MIFAO 42-45), 2 vols (Cairo, 1920-27). A thesis on the subject of Ayyubid monuments in Jerusalem is being completed by M. Harawy at Oxford University; in the meantime see S. Jarrar, 'Suq al-Marifa: An Ayyubid Hanbalite Shrine in al-Haram al-Sharif', Muqarnas 15 (1998).

168. Sauvaget, Monuments, 95-96; E. Herzfeld, 'Damascus: Studies in Architecture - IV', Ars Islamica 13-14 (1948), 1118 ff.

169. Sauvaget, Monuments, 64; Herzfeld, 'Damascus', 123 ff.

170. Sauvaget, 'Inventaire', 82 and 86 for a few remains; D. Sourdel, ed., La Description d'Alep d'Ibn Shaddad (Damascus, 1953), 42 ff.

171. R. Lewcock in W. Daum, ed., Yemen, (Innsbruck, 1987); B. Finster, 'An Outline of the History of Islamic Religious Architecture in Yemen', Muqarnas 9 (1992), for an introduction to the subject with bibliographies.

172. Creswell, MAE 2, 94 ff.

173. J. Sauvaget el al., Monuments ayyoubides de Damas, 4 vols (Paris, 1938 ff.) 1, 120.

174. On these issues see now Tabbaa, Medieval Aleppo and his forthcoming Sunni Revival.

175. E. Herzfeld, 'Damascus: Studies in Architecture - II', Ars Islamica 11-12 (1946), 32-38.

176. Sauvaget, Monuments, 100-02.

177. Summarized in Creswell, MAE 2, 104 ff.

178. Creswell, MAE, 64 ff.

179. Creswell, MAE, 88 ff.

180. J. Sauvaget et al., Monuments ayyoubides 3, 92 ff.; M. Ecochard and C. LeCocur, Les Bains de Damas, 2 vols (Beirut, 1942-43); for Aleppo see Sauvaget, 'Inventaire'.

181. J. Sauvaget, 'Caravanserails syrens', Ars Islamica 6 (1939).

182. Sourdel, ed., La Description d'Alep d'Ibn Shaddad, 2.

183. Sauvaget, Monuments Ayyoubides, 46; E. Herzfeld, 'Damascus: Studies in Architecture - I', Ars Islamica 9 (1942), 1-53.

184. T. Allen, Five Essays on Islamic Art (Sebastopol, CA, 1988).

185. S. Saouaf, The Citadel of Aleppo (Aleppo, 1958); Herzfeld, Matériaux 77ff.; Rogers, The Spread of Islam (Oxford, 1976), 43 ff.; Tabbaa, Aleppo, 53 ff.

186. D. J. Cathcart King, 'The Defenses of the Citadel of Damascus', Archaeologia 94 (1951). The building has now been cleared and is in the process of being investigated.

187. Creswell, MAE 2, 1-63; N. Rabbat, The Citadel of Cairo (Leiden,1995) which, however, deals primarily with its later phases.

188. A. Abel, La Citadelle ayyubite de Bosra', Annales Archéologiques de Syrie 6 (1956).

189. Sourdel, ed., La Description d'Alep d'Ibn Shaddad, 24; Herzfeld, Alep, 134-45.

190. J. Sauvaget, 'L'Architecture musulmane en Syrie', Revue des Arts Asiatiques 3 (1934); further remarks throughout other studies by Sauvaget and Herzfeld, as well as in specialized articles such as J. Lauffray, 'Une madrasa ayyoubide de la Syrie du Nord', Annales Archéologiques de Syrie 3 (1953) and especially maurice Ecochard, Filiation de monuments grecs, byzantins et islamiques (Paris, 1977), summarizing some of his earlier works on architectural forms and their creation.

191. Monuments ayyoubides I, 21-23. However, it was already used in the eighth century at Qasr al-Hayr West, where stone and brick were used together.

192. Creswell, MAE 2, pl. 19; many instances exist in Syria.

193. Besides Herzfeld's work and J. Sourdel-Thomine's contribution in Monuments ayyoubides 4, see the classic article by Max van Berchem, 'Inscriptions arabes de Syrie', Mémoires de l'Institut Egyptien 3 (1900) and Tabbaa, Aleppo, 99 ff.

194. Monuments ayyoubides 2, pl. XV.

195. Creswell, MAE 2, 84 ff.

196. Monuments ayyoubides 3, 121 ff.

197. Creswell, MAE 2, p. 138, note 5.

198. For instance J. Sauvaget, Les Perles Choisies d'Ibn ach-Chihna (Beirut, 1933), 136.

199. Tabbaa, 'Survivals and Archaisms in the Architecture of Northern Syria', Muqarnas 10 (1993); Terry Allen, A Classical Revival.

200. For Mayyafariqin see A. Gabriel, Monuments turcs d'Anatolie (Paris, 1931-34) 2, 143.

201. F. Sarre, Reise in Kleinasien (Berlin, 1986), 47-48; Sarre, Konia (Berlin, 1921); I. H. Konyali, Konya Tarihi (Konya, 1964), esp. 293 ff; Scott Redford, 'The Aleddin Mosque', Artibus Asiae, 51 (1991); T. Baykara, Türkiye Selcuklurlari Devrinde Konya (Ankara, 1985).

202. Gabriel, Monuments, 32 ff.

203. Gabriel, Monuments 2, 39 ff and 174 ff.

204. Gabriel, Monuments 2, 173 ff.

205. Gabriel, Monuments 2, 176; B. Ünsal, Turkish Islamic Architecture (London, 1959), 17. The mosque is attributed to a vizier of Kilicarslan II in the late twelfth century; for justification see E. Diez and O. Aslanapa, Türk Sanati (Istanbul, 1955), 55.

206. Gabriel, Monuments 1, 62 ff and 46 ff.

207. Gabriel, Monuments 2, 155 ff.

208. Sarre, Reise, 51-54; Konyali, Konya Tarihi, 4 523 ff; Baykara, Konya.

209. Ünsal, Architecture, 36-38; S. K. Yetkin, Turkish Architecture (London, 1966), 22 ff; for the problem of the building's date see M. J. Rogers, 'The Date of the Cifte Minare Madrasa', Kunst des Orients 8 (1974).

210. Sarre, Reise, 48-51; Yetkin, Turkish Architecture, 28 ff; Konyali, Konya Tarihi, 950, 1049.

211. D. Kuban, Anadolu-Türk Mimarisinin Kaynak ve Sorunlari (Istanbul, 1965), argues for the convergence between Islamic needs and local practices; see S. Redford, 'The Seljugs of Rum and the Antique', Muqarnas 10 (1993).

212. Thus Amasya in Ünsal, Architecture, 45 and Yetkin, Turkish Architecture, 35 ff.

213. For the waqfs or endowment deeds in Anatolia see A. Durukan, 'Anadolu Selçukler Sanati Açinsinlari Vakfieyeleri Önemi', Vakiflar Dergisi 26 (1997).

214. H. W. Duda, Die Seldschukengeschichte des Ibn Bibi (Copenhagen, 1959), 146-48.






Website © 2002-19 National Institute for Technology and Liberal Education.

This website is compliant with the XHTML 1.0 standard as defined by the W3C.

Valid XHTML1.0!