Mamluk History Through Architecture: Monuments, Culture and Politics in Medieval Egypt and Syria 9780755697472, 9781845119645

Citation preview

LIST OF ILLUSTR ATIONS FIG. 1: Signature of Ibn Ghana’im on the muqarnas of the portal of the Mausoleum of Sultan al-Zahir Baybars in Damascus FIG. 2: Signature of Ibn Ahmad Zaghlish on the side of the portal of the Palace of Amir Qawsun in Cairo FIG. 3: Detail of the door of the Bimaristan (hospital) al-Nuri, Damascus FIG. 4: Non-architectural mosaic of the eastern corner of the Qubba al-Zahiriyya FIG. 5: Mosaic of the qibli (southern) wall of the Qubba al-Zahiriyya FIG. 6: Mosaic of the western wall of the Qubba al-Zahiriyya FIG. 7: Mosaic of the northern wall of the Qubba al-Zahiriyya FIG. 8: Detail of the mosaic of the northern wall of the Qubba al-Zahiriyya FIG. 9: Detail of the mosaic of the Cairo Citadel’s Qa‘a FIG. 10: The Citadel of Aleppo FIG. 11: Three-level inscriptions, main gate, the Citadel of Aleppo FIG. 12: The Serpent Gate (Bab al-Hayyat), the Citadel of Aleppo FIG. 13: The Talisman Gate in Baghdad (after Sarre & Herzfeld) FIG. 14: Carved scene of two lions, the Citadel of Aleppo FIG. 15: The Crying Lion, the Citadel of Aleppo Fig. 16: The Smiling Lion, the Citadel of Aleppo FIG. 17: Detail of the niche from the Gu’ Kummet at Sinjar, the Bunduqdar (courtesy of Yasser Tabbaa) FIG. 18: Rank of Amir Qawsun, carved on a stone slab, garden of the Mosque of Sultan Hasan, Cairo FIG. 19: The Madrasa of Sultan Hasan FIG. 20: The Madrasa of al-Ghuri FIG. 21: Aerial view of al-Azhar from the south (courtesy AKSE, Aga Khan Services, Egypt) FIG. 22: The minarets of al-Aqbughawiyya, of Qaytbay, and of Qansuh al-Ghuri FIG. 23: Bab al-Muzayinin (Gate of the Barbers) of ‘Abd al-Rahman Katkhuda FIG. 24: Main or south-eastern façade of al-Azhar Mosque’s haram FIG. 25: Roundel and niche stucco ornaments above the arches of al-Azhar’s courtyard FIG. 26: Plan of al-Azhar Mosque in the early twentieth century FIG. 27: Proposed reconstruction of the original plan of al-Azhar Mosque FIG. 28: Plan of Fatimid al-Qahira (after Ravaisse) FIG. 29: The four minarets of al-Azhar including the two Ottoman minarets of ‘Abd al-Rahman Katkhuda at Bab al-Muzayinin and al-Aqbughawiyya (drawn by Prisse D’Avennes, c.1870). FIG. 30: Portal of the Riwaq al-‘Abbasi FIG. 31: Façade of the Iwan Kisra: elevation (after Sarre and Herzfeld) and in a photograph taken before the flood of 1883 –vi–

38 39 41 49 50 51 52 53 54 61 64 65 66 67 68 68 69 69 70 70 74 74 75 76 76 77 79 85

93 95 105

L I ST OF I L LUST R AT IONS

FIG. 32: Reconstructed plan of the Palace of Abu Muslim al Khurasani in Merv (after K. A. C. Creswell) FIG. 33: Aerial view of the Madrasa of Sultan Hasan (courtesy of Gary Otte) FIG. 34: Plan of the Madrasa of Sultan Hasan (after Max Herz) FIG. 35: Interior view of the Madrasa of Sultan Hasan FIG. 36: Plan of the Madrasa of al-Salih Najm al-Din Ayyub (after K. A. C. Creswell) FIG. 37: Plan of the Madrasa of Sultan al-Zahir Baybars (after Michael Meinecke) FIG. 38: Plan of the Madrasa of al-Nasir Muhammad (after K. A. C. Creswell) FIG. 39: Plan of the Madrasa of Amir Sarghatmish (after Christel Kessler) FIG. 40: Schematic plan of the four palaces of al-Qasr al-Ablaq FIG. 41: General view of the Palace of Qawsun FIG. 42: Qa‘a of the Palace of Bashtak FIG. 43: General view of the Palace of Alin Aq from the street FIG. 44: Site of the Ablaq Palace at the Citadel FIG. 45: Vaults of the underground hall below the Ablaq Palace FIG. 46: Façade of the Palace of Bashtak (courtesy of Bernard O’Kane) FIG. 47: The Maq‘ad of Amir Mamay al-Sayfi FIG. 48: Plan and sections of the Great Iwan (Description de l’Égypt, Etat-Moderne, vol.1, pl. 76) FIG. 49: The plan of the Great Iwan (after the Description de l’Égypte) FIG. 50: Interior of the Great Iwan (after the Description de l’Égypte) FIG. 51: Plan of the Mashhad al-Juyushi (after K. A. C. Creswell) FIG. 52: Plan of a Fustat house (after Jamal Mihriz) FIG. 53: Plan of the hall of al-Salih Najm al-Din Ayyub (after K. A. C. Creswell) FIG. 54: Plan of the Madrasa and Mausoleum of Qalawun (after K. A. C. Creswell) FIG. 55: View of al-Mundhir Basilica in Rusafa (Sergiopolis) FIG. 56: The location of the Dar al-‘Adl in Damascus FIG. 57: The location of the Dar al-‘Adl in Aleppo FIG. 58: The location of the three Dur al-‘Adl in Cairo FIG. 59: The location of the Divan de Joseph in the Citadel of Cairo as reproduced in the Description de l’Égypte FIG. 60: Diagram of the sultan’s entry into the Great Iwan on the Dar al-‘Adl days FIG. 61: Diagram of the circle forming around the sultan on the Dar al-‘Adl days FIG. 62: Diagram of the circle around the sultan and the Amirs of Hundred FIG. 63: Diagram of the complete seating layout on the Dar al-‘Adl days FIG. 64: Plan of the Iwan with the seating layout of the Dar al-‘Adl superimposed on it FIG. 65: Façade of the mosque designed by Coste for the Citadel FIG. 66: Façade of the actual Mosque of Muhammad ‘Ali FIG. 67: Aerial view of the Mosques of Sultan Hasan and the Rifa‘i (courtesy of Gary Ott) FIG. 68: Southern façade of the Rifa‘i Mosque as designed by Hussein Pasha Fahmi FIG. 69: Southern façade of the Rifa‘i Mosque as built by Herz FIG. 70: Western façade of the Rifa‘i Mosque FIG. 71: View of the Egyptian National Library (Dar al-Kutub)

106 106 107 108 109 110 110 110 114 115 116 117 120 120 122 123 126 129 130 131 131 132 133 135 151 152 154 156 159 159 159 160 161 178 179 183 184 185 186 186

All plans and figures with no credits are the author’s. The term ‘after’ denotes the source from which the plan was digitised. All digitised plans were made by students of the author. –vii–

ACKNOWLEDGEMENTS

F

irst and foremost, I would like to extend my warmest thanks to the editors of the journals and books in which the chapters in this book originally appeared (as listed below). They were all very receptive to my requests to reprint the articles in this collection: ‘The Changing Concept of Mamluk in the Mamluk Sultanate in Egypt and Syria’, in Slave Elites in the Middle East and Africa: A Comparative Study, ed. Miura Toru And John Edward Philips (London and New York: Kegan Paul International, 2000), 81–98. ‘Representing the Mamluks in Mamluk Historical Writing’, in The Historiography of Islamic Egypt, c. 950–1800, ed. Hugh Kennedy (Leiden, E. J. Brill, 2000), 59–75.

‘The “Militarization” of Taste in Medieval Bilad al-Sham’, in Muslim Military Architecture in Greater Syria: From the Coming of Islam to the Ottoman Period, ed. Hugh Kennedy (Leiden: E. J. Brill, 2005), 84–105. ‘Al-Azhar Mosque: An Architectural Chronicle of Cairo’s History’, Muqarnas 13 (1996): 45–67. ‘Documenting Buildings in the Waqf System’, Thresholds 28, Concerto Barocco: Essays in Honor of Henry A. Millon (2005): 30–2. ‘The Iwans of the Madrasa of Sultan Hasan’, ARCE Newsletter 143–44 (1988/89): 5–8. ‘Mamluk Throne Halls: Qubba or Iwan’, Ars Orientalis 23 (1993): 201–18.

‘Perception of Architecture in Mamluk Sources’, Mamluk Studies Review 6 (2002): 155–76.

‘Writing the History of Islamic Architecture in Cairo’, Design Book Review 31 (1994): 48–51.

‘Architects and Artists in Mamluk Society: The Perspective of the Sources’, Journal of Architectural Education 52/1 (1998): 30–7.

‘The Ideological Significance of the Dar al-‘Adl in the Medieval Islamic Orient’, International Journal of Middle Eastern Studies 27/1 (1995): 3–28.

‘The Visual Milieu of the Counter-Crusade in Syria and Egypt’, in The Crusades: Other Experiences, Alternate Perspectives. Selected Proceedings from the 32d Annual CEMERS Conference, ed. K. I. Semaan (Binghamton, NY: Center for Medieval and Early Renaissance Studies, 2003), 71–81. ‘The Mosaics of the Qubba al-Zahiriyya in Damascus: A Classical Syrian Medium Acquires a Mamluk Signature’, Aram 9–10 (1997–1998 [1999]): 1–13.

‘ ‘Ajib and Gharib: Artistic Perception in Medieval Arabic Sources’, The Medieval History Journal 9/1 (2006): 99–113. ‘The Formation of the Neo-Mamluk Style in Modern Egypt’, in The Education of the Architect: Historiography, Urbanism and the Growth of Architectural Knowledge. Essays Presented to Stanford Anderson on his Sixty-Second Birthday, ed. Martha Pollak (Cambridge, MA: MIT Press, 1997), 363–86.

–viii–

ACK NOW L E DGE M E N T S

I would also like to express my gratitude to the Barakat Foundation, USA, and its two founders Hamida ‘Ali Reza and Rima Qabbani, for the generous financial support that allowed the production of this book. Friendship, support, and humorous teasing have marked our relationship for more than ten years. My sincere thanks go also to Gary Otte, Bernard O’Kane, Yasser Tabbaa, and the architectural team at AKES (Cairo) who gave me permissions to use their photos. My students Maryam Eskandari, Saeed Arida, Omar Rabie, Mohamed Elshahed, and Peter Christensen and AKPIA Administrative Assistant/graphic designer

José Luis Argüello rendered the digital figures in the book, giving many of its chapters a new look. My thanks go to all of them. Finally, many readers, especially Islamic art historians and Mamlukists of my generation, will undoubtedly perceive in a number of interpretations and explanations or in the judicious use of certain words and sentences the presence of two outstanding women, Laila ‘Ali Ibrahim and Margaret Bentley Sevcenko. For more than ten years, these very dear ladies guided and nurtured my study of Mamluk history and architecture. This collection is dedicated to their memory.

–i x–

PREFACE

T

he emergence of a distinct Mamluk culture in late medieval Egypt and Syria is a fascinating yet littlestudied phenomenon. The historical framework of the Mamluk sultanate is fairly well established and its chronological unfolding well documented by numerous annals, biographical dictionaries and other primary historical sources. So are the state’s institutions, its protocol and rituals, as well as the broad ethnic, ideological and political contours of its ruling elite. This was a onegeneration military aristocracy composed of young slaves (mamluks), all foreign and forcibly imported to Egypt and Syria, where they were trained and educated for at least ten years before being manumitted and conscripted in the army to rule and defend an extraordinary empire that lasted for almost two hundred and seventy years (1250–1517). The Mamluks, already separated by linguistic and ethnic barriers from the rest of society, created a regime that stressed exclusion as a means of control and manipulation of wealth. According to Ibn Khaldun and al-Maqrizi, both keen contemporary observers of the Mamluk system, the separation was thought to ensure the survival of the state’s structure, for it was assumed to preserve the intensity of the manufactured solidarity (‘asabiyya) and its concomitant fighting instinct among the Mamluks. But it also produced a strong sense of superiority over the native population that remained palpable until the end of the empire, even after the Mamluks began to slowly adopt some of the characteristics of the indigenous urban culture while, at the same time, some

of their more peculiar customs seeped into the local milieu. Despite a recent upsurge in Mamluk studies that has seen the publication of most primary sources, including minor ones, and the appearance of a specialised journal, Mamluk Studies Review, we still know little about the Mamluks’ manners, attitudes, beliefs, preferences and taste, all fundamental elements in the composition of culture. Only a few scholars have attempted to go beyond the analysis of Mamluk social structures – the army, slavery system, household, training practices, and so forth – or their ceremonies and rituals to try to understand their underlying cultural and mental makeup. This is so primarily because culture manifests itself in ways that are not easily captured in biographical or annalistic writing, which constitute our main primary sources, especially when the authors of these sources, mostly members of the religiously educated class, are either alien to the Mamluks and their culture or seem incapable of understanding, uninterested in, or downright antagonistic to it. Furthermore, even though the Mamluks shared the same religion, Islam, and the broad sense of identity it engendered with the majority of their subjects, the two groups occupied different cultural spheres and observed distinct codes of behaviour, with sometimes hostile views and practices. Mamluk culture was novel, almost experimental, and foreign in origin and design. It was presumably aimed at bolstering a common frame of reference between the disparate groups of young, imported –x–

P R E FAC E

Mamluks who, lacking a shared background, had to forge their norms from the bits and pieces of customs and traditions they brought with them from their countries of origin and the experiences they underwent and absorbed as Mamluks in their new abode. This syncretism defined most details of Mamluk culture, such as habits of socialisation, moral rules, etiquette, business dealings and marriage arrangements, as well as their image of themselves. It was also symbolically deployed in the particular nomenclature and insignia (rank) systems they used and the elaborate visual, dietary, musical, ceremonial and dress codes they introduced and guarded exclusively as theirs. Consequently, to begin to decode the culture of the Mamluks and reconstruct its twists and turns over time we need more than just the historical knowledge that allows us to identify the relevant texts for the task and the linguistic affinity that permits us to decipher and understand them. We also need more than the detective instinct that may guide us to catch the hints, innuendos and bits of data preserved in the written sources – sometimes inadvertently – that bear directly on the culture of the Mamluks or on the cultures of their lands of origin in Central Asia and the Caucasus or of their ethnic kin and adversaries, the Mongols. We actually need to cast a wider net to identify and employ all possible interpretable historical documents in which the expressions of culture have been preserved in order to compensate for the limitations and biases of the written sources. An entire domain of inquiry – art and architecture – has as yet not been fully or systematically explored as a source for social and cultural history in Mamluk studies. This is not to say that the study of Mamluk art and architecture is wanting; in fact, among the recognised periods of Islamic art history, the Mamluk period is one of the better covered. All Mamluk buildings and artefacts have been dated, described, classi-

fied and evaluated, frequently in comparison to other contemporary architectural and artistic traditions around the Mediterranean or in the larger Islamic world. Likewise, patrons of art and architecture, and to a much lesser extent artists and builders, have been identified, their biographies assembled and their motivations assessed following a narrative method that goes back to the fifteenth-century historian al-Maqrizi and his famous book al-Khitat (a study of cities’ topography and architecture). However, most of the studies done until fairly recently have not bothered with the sociocultural background or import of art or architecture. In fact, it is only in the last decade or so that Mamluk architecture in particular has begun to be treated as a cultural expression that embodies the tastes, views and mentalités of its patrons and builders, reflects their attitudes and reveals their ambitions. Promoting this kind of enquiry is the main purpose of this collection of essays on Mamluk architecture, culture and polity. The essays, written over a period of sixteen years (1989–2005) and published in various journals and edited books, move in two directions. The first explores the ways by which Mamluk written historical sources represent the Mamluk elite, their culture and architecture. It aims to extricate from the written sources their impressions, assumptions of, and projections onto the Mamluk culture, incomplete and dispersed as they may be, and to posit them in their historical and intellectual contexts. A subgroup of these essays goes further to analyse the scholarly strategies, biases and limitations that informed the shaping of the Mamluk image in these written sources. The second direction draws primarily on art and architecture, but also on epigraphy and numismatics, to evince manifestations and reflections of the mentalités and the ideological and cultural attitudes of the Mamluk elite. Within that framework, a number of essays identify and examine famous but problematic artistic media and techniques,

–xi–

P R E FAC E

and monuments and building types to highlight their multifarious, and sometimes even contradictory, visual, formal and symbolic lineage and significance. All the essays, however, aim to use art and architecture in conjunction with the insights gained from the critical examination of the written sources to formulate a better informed, more balanced and perhaps historically more complete analysis of Mamluk society and culture and the role of the Mamluks themselves in shaping that culture. Such enquiries are pertinent to two historical disciplines: art history and Islamic history. A socially and culturally attentive investigation offers a richer and more textured art-historical framework than the repeatedly criticised and inadequate taxonomic, typological or dynastic frameworks that have thus far been used for the study and interpretation of Mamluk art and architecture, and Islamic art in general. Using visual along with written documents in an examination of Mamluk culture brings a new dimension to interpretation that is lacking in conventional accounts. It also complements the evidence of other cultural expressions such as the ‘mirror-of-princes’ literature (nasa’ih al-muluk), religious texts, poetry and belles-lettres (adab) that a number of historians have begun to exploit in order to better understand Mamluk culture. Visual ‘documents’ may in fact be even more representative of collective views, opinions and beliefs than literature because they are more apt to reflect collective sentiments and attitudes and, being public, they need to be understood by all elements of the society, not just by the elite. The fifteen chapters of the book are arranged in four parts. The first part is essentially historiographical. Entitled ‘Unpacking Mamluk Sources’, it is meant to set the inquisitive tone of the entire project by delving into the written Mamluk sources to uncover their textual strategies and referential codes in order to elucidate how they influenced the representation of the Mamluk

regime and culture, and the Mamluk individuals themselves. The first chapter, ‘The Changing Concept of Mamluk in the Mamluk Sultanate in Egypt and Syria’, traces the development of the Mamluk concept from early Islamic times through the formative Mamluk period and explains the measures introduced by the heroic early Mamluk sultans to ensure the efficacy of the system and the longevity of the sultanate. The second chapter, ‘Representing the Mamluks in Mamluk Historical Writing’, analyses the scholarly strategies, biases and limitations that informed the shaping of the Mamluk image in Mamluk historiography and tries to understand whether this representation reflected the motives, thoughts, mindsets and beliefs of the Mamluk elite, or those of the chroniclers and interpreters who wrote about them. The third chapter, ‘Perception of Architecture in Mamluk Sources’, surveys the Mamluk sources’ interest in buildings, analyses their different approaches and textual techniques, and elucidates the conceptual ramifications of the textual evidence for the study of Mamluk architecture and culture. The last chapter in this part, ‘Architects and Artists in Mamluk Society: The Perspective of the Sources’, examines the social standing of Mamluk artists and architects and argues that the few who achieved social recognition had to transform themselves in order to move beyond the confines of small-time artisanal limitations. Exploring how architecture can and should be seen as a historical source, the second part, entitled ‘Architecture as History’, moves closer to the book’s main theme. It starts with the chapter ‘The Mosaics of the Qubba al-Zahiriyya in Damascus: A Classical Syrian Medium Acquires a Mamluk Signature’, which discusses the mosaics of the most famous Mamluk architectural example, the Mausoleum of al-Zahir Baybars in Damascus (completed in 1284), and posits that these mosaics were variations on the Umayyad

–xii–

P R E FAC E

representational themes adapted to Mamluk tastes and pictorial practices. Chapter 6, ‘The ‘Militarisation’ of Taste in Medieval Bilad al-Sham’, addresses the artistic and architectural ramifications of the coming of the Turks to the Middle East in the eleventh century and their establishment of a new polity that dominated the region for centuries. Its argument is that the art and architecture of the new Turkish dynasties underwent a process of militarisation that reflected their social and political structures and mentalities. The last chapter in this part, ‘al-Azhar Mosque: An Architectural Chronicle of Cairo’s History’, uses the changing circumstances of alAzhar Mosque as a barometer of the transformation of Cairo from a capital of the religiously defined Fatimid dynasty to a centre of the Mamluk military state, then to a pre-modern Ottoman provincial capital, and finally to a modern metropolis. The third part, ‘Architecture and Language’, delves even further into the book’s theme by focusing on the affinity between architecture and text as two interrelated forms of historical documentation. It begins with a short essay, ‘Documenting Buildings in the Waqf System’, that explains how documenting buildings in the waqf (endowment) deeds reflected both an interest in the purely functional and socioeconomic dimensions of architecture and a specific vision of the role of buildings in social and urban space in the medieval period. Next, ‘The Iwans of the Madrasa of Sultan Hasan’, argues that the iwans of the Madrasa of Sultan Hasan epitomised the semantic field of the word iwan in the Mamluk period. The following chapter, ‘Qasr: An Agent of Monumentality in Mamluk Architecture’, deals with a richly textured architectural term as well, qasr, generally understood to signify ‘palace.’ It reconstructs the architecture of the Mamluk palaces in Cairo to assert that qasr was used to designate a specific Cairene architectural unit in the Bahri Mamluk period (1250–1382)

adapted to the exigencies of the ruling class’s image. A fourth chapter dealing with a building type, ‘Mamluk Throne Halls: Qubba or Iwan’, traces the evolution of these structures, establishes their genealogy, and shows that the interplay between the two terms qubba and iwan reflects changes in the urban, commemorative and royal contexts of the Mamluk period. The last part of the book, ‘Architecture as Cultural Index’, brings culture into the discussion in a bid to historicise it on the one hand and to problematise the interplay of architecture and history by adding culture as a variable that changes at a rate different from that of architecture but still affects it directly and is affected by it nonetheless. ‘Writing the History of Islamic Architecture in Cairo’ is a concise survey of more than a century of writing on the history of Cairene architecture that challenges the discipline of architectural history in general and urges the Islamic architectural historians to move beyond set methodological boundaries. Next, ‘The Ideological Significance of the Dar al-‘Adl in the Medieval Islamic Orient’, postulates a relationship between the counter-Crusade during the Ayyubid and early Mamluk periods and the building of palaces of justice, and presents an ideological and historical explanation for this synchrony. The third essay, ‘‘Ajib and Gharib: Artistic Perception in Medieval Arabic Sources’, tries to uncover the attitudes of Mamluk historians towards the visual arts by examining the vocabulary and terminology they used to see and understand art, especially the two terms ‘ajib and gharib, and then relates these attitudes to their education and professional outlook. The final chapter, ‘The Formation of the Neo-Mamluk Style in Modern Egypt’, brings us almost to the present without losing the focus on things Mamluk. The chapter considers a case study of an East–West architectural exchange in nineteenth-century Egypt: the emergence of the neo-Mamluk style, which demonstrates the working

–xiii–

P R E FAC E

of multiple intellectual, architectural and political programmes in its development. I have tried to keep the chapters in their original form despite the fact that new research has appeared in the meantime that may alter some of my conclusions and despite my own views having changed in the last twenty years. Only minimal new editing was entered into the text: the few typos that escaped the initial checks were corrected and the language was anglicised at the request of my British publisher. Despite the ensuing

repetition, I have also kept the footnotes integral to every chapter so that each can be read as an independent piece without the need to verify titles and authors in other articles. I reduced the number of figures from their original number and digitised them. I also eliminated all transliteration except for the ‘ayn (‘) and hamza (’). This was also at the request of the publisher, who felt that they encumbered the non-specialist and were redundant to the specialist trained in Arabic or Persian; a viewpoint to which I concur.

–xiv–

1. THE CHANGING CONCEPT OF MAMLUK IN THE MAMLUK SULTANATE IN EGYPT AND SYRIA

T

mamluk is an adjective meaning ‘owned’; it can be applied to either a person or a thing. When it is applied to a person, it corresponds to the word ‘abd, the common term for slave. Mamluk, however, and its functional equivalent ghulam (youth), were used exclusively, at least from the time of the Abbasid caliph al-Mu‘tasim, to designate a particular kind of owned person – a young man serving in a military capacity.1 The Abbasids, and after them many Islamic states, imported huge numbers of young white men from pagan countries bordering on the Dar al-Islam (Central Asia, southern Europe, Russia, the Crimea and Caucasia) to serve in their armies.2 In time, most, but by no means all, Mamluks tended to come from the Turkic tribes of the Asiatic steppes, who were famous as fierce warriors, agile horsemen and archers.3 Mamluk and Ghulam regiments of mounted Turkish archers became a ubiquitous component of most Islamic armies from Egypt to Transoxania; they even served in nonIslamic armies such as the Georgian and the Byzantine.4 These regiments, however, were but few among many. They were not even the only slaves troops; many armies, especially in the Western Islamic world, had units composed entirely of black slaves

he rise of Mamluk elites to political power in the medieval and early modern Islamic world is generally seen as a distinctly Islamic phenomenon that has very few parallels elsewhere. While the reasons for the appearance of the Mamluk institution in the first place are still hotly debated, there is a general consensus that by the eleventh century Mamluk elites had pervaded all existing state structures in the Islamic world from the tribally based principalities to the vast caliphal and imperial domains. This indisputable historical fact has encouraged an uncritical acceptance of a general model of Mamluk that is applied everywhere Mamluk elites were prominent, despite clear differences in the circumstances of their formation, their accession to power and their structures. This chapter will try to counter this tendency by critically examining the historical particularities of one case, the Mamluk sultanate in Egypt and Syria (1250–1517), which is perhaps the most outstanding example of a Mamluk elite and the one most assiduously studied and analysed. First, however, we need to briefly consider the evolution of the Mamluk institution in Islamic history up to the eve of the Egyptian Mamluk coup in 1250. The word –3–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

named al-Sudan and of Saqaliba slaves (Slavs or perhaps Europeans in general).5 But the word mamluk was never used for black slaves, even when they served exclusively as warriors, as they did in the Fatimid period in Egypt. Thus, the word mamluk came to have geographic and ethnic connotations: it meant white, mostly Turkish or Turkicised young men, at least in the Seljuq and post-Seljuq Eastern Islamic world. The concept of mamluk underwent yet another decisive change by the middle of the thirteenth century. Many of the Ayyubid princes, who had ruled Egypt and Syria since the death of Salah al-Din as a constantly disputed family appanage had come to rely heavily on Mamluk units in their armies. This development coincided with the eruption of the Mongols across Central Asia under Chengiz Khan which drove many tribes – notably the group of tribes, or nation, of Qipchaq – west and north and reduced many to slavery.6 The Ayyubid princes seem to have taken advantage of the overflow of Qipchaq Mamluks onto the slave markets of Syria and Egypt. But alSalih Najm al-Din Ayyub (r. 1240–1249), an overly ambitious and particularly ruthless and cunning Ayyubid prince, by far outdid his peers and predecessors in the number of Qipchaq Mamluks he purchased, trained and enlisted. After a number of serious setbacks early in his career when he was deserted by some of his free troops, al-Salih realised that he could only depend on the loyality of his Mamluks. An elite unit made up almost completely of young Qipchaqs and named al-Bahriyya became his most trusted unit; the soldiers were used in the most delicate and important tasks. Al-Salih kept them around him at all times and showered them with all kinds of financial and political favours. They distinguished themselves against a particularly serious Frankish attack on Damietta in 1249, but al-Salih’s untimely death during that campaign must have left them fearful for their acquired privileges and mounting political

independence and military dominance, especially after they grasped the gravity of the threats posed by his son and successor Turan Shah. The new sultan began promoting his own Mamluks and – outrageously in the opinion of contemporary commentators – his black eunuchs against the Salihis (i.e. the Mamluks of al-Salih including the Bahriyya) and, particularly when drunk, proclaimed his intention of ridding himself of the Bahriyya. They struck first and assassinated Turan Shah in 1250.7 The Salihi Mamluks who killed the son of their master seemed not to have any clear plan for ruling or a candidate for the sultanate. They twice resorted to the tried and tested formula of installing a legitimising figurehead while they sorted out their differences behind the scenes. First they chose Shajar al-Durr, the wife and exMamluka of al-Salih Ayyub, who, extraordinarily for a woman, ruled as a sultana for three months in 1250. On the coins she struck and the decrees she issued, she styled herself Umm Khalil (Mother of Khalil), the by-then dead son she bore al-Salih; perhaps as a claim to an (obviously illusory) Ayyubid continuity. Internal and external – both Ayyubid and caliphal – hostility to the rule of a woman and a Mamluka soon forced her to abdicate. The various Mamluk factions of al-Salihiyya then elevated to the sultanate Aybak al-Turkmani, a Mamluk amir and a former jashinkir (taster) of al-Salih Ayyub, who married Shajar al-Durr, obviously in recognition of her enduring hold on political power in Egypt. Aybak’s election, however, did not satisfy all dissenters, especially the Bahriyya, who rightly felt that they were the real power brokers in the country. Less than a week later, they forced Aybak’s demotion to the position of atabek and installed a minor Ayyubid prince, al-Ashraf Musa, as a nominal sultan ostensibly to appease the Ayyubid clan in Syria. Aybak, lacking a real military and political counterbalance to the Bahriyya, acquiesced. It took him four years (1250–54) of manoeuvering to fend off the

–4–

T H E CH A NGI NG CONCEP T OF M A M LUK I N EGY P T A N D SY R I A

Ayyubids of Syria, stabilise the new regime in Egypt, obtain caliphal recognition from Baghdad and, most importantly, acquire a body of Mamluks of his own, the Mu‘izziya (from his regnal title al-Mu‘izz). He then felt strong enough to move against his opponents, the Bahriyya, eliminate their leader, and finally dismiss the Ayyubid titular ruler and install himself as sovereign sultan.8 This momentous decision ushered in a new regime which, although it continued to follow the patterns of organisation established by al-Salih Ayyub or inherited from the earlier systems of the Seljuqs, Zengids, Ayyubids or even Fatimids, was, in a more fundamental and conceptual sense, revolutionary. This was not merely because the Mamluks ruled without a mediator; that practice had a long history and deep roots in Egypt and elsewhere in Islamic lands.9 But this was the first, and arguably the only, time when a consciously perceived and carefully formulated Mamluk system became the structural backbone of a new and long-lived polity and political culture.10 This barely recorded and almost completely overlooked process was not immediately applied, nor was it even conceived, at least not by Aybak, the first Mamluk sultan or his entourage. We do not find any hint at it in the sources during his reign. Nor did his two immediate successors, his adolescent son al-Mansur ‘Ali or his Mamluk Qutuz, the hero of the battle of ‘Ayn Jalut, during their short and eventful rules have the vision or time to institute the transformation. It took the genius, farsightedness, perseverance and good luck of two outstanding rulers, Baybars and Qalawun, and their allies among the Bahriyya or their own Mamluks, to distill from the Mamluk principle and system, combined with some external sources including the Mongol regime, a whole new political and military regime. That they did it so successfully is evident from the long-lasting Mamluk rule, which, to any casual modern observer, appears over and over to have been fatally shaken, yet always managed to survive.11

Perhaps a more congenial measure of their success is that their contemporary chroniclers, who were otherwise so disdainful of the Mamluk class, appear to have adapted to the profound upheaval in the structure of their society with minimal fuss. Many simply carried on recording their annals or biographies with only the most perfunctory notice of the passing of the rule from the Ayyubids to the Mamluks.12 Some, especially among the fuqaha’ (jurists), were disturbed by the Mamluks’ usurpation of rule, but not by the mechanisms with which the Mamluks imposed and sustained their restructuring of the power apparatus. They only attempted to reconcile the end result, the Mamluk rule itself, with their established and almost sacred social and political order. They did so by either postulating a legally dubious doctrine, the bay‘a qahriyya (allegiance under duress), to sanction the Mamluk rule in the interest of the Islamic community or by linking it (unconvincingly at times) to early Islamic precedents.13 Only a few chroniclers tried to link the rise of the new Mamluk regime to the events of that tumultuous period. They sought an almost preordained cause for it in the need to counter the onerous threat to Islam and Muslims posed by the seemingly invincible Mongol invasion and the annihilation of the Abbasid caliphate in 1258. We find the earliest formulation of such a rationale in some of the panegyrics recited after the battle of ‘Ayn Jalut in September 1260. Only the Mamluk Turks, a couplet asserts, could have defeated the Mongols because they were kin, shared battle tactics and fighting skills and therefore were predisposed to know and exploit their weaknesses.14 This justification of the Mamluk takeover, anachronistic as it was (the Mamluk revolution occurred eight years before the Mongol invasion), was eagerly taken up in the official Mamluk discourse as a powerful tool for legitimisation. It was to frequently resurface, particularly in critical moments of change or succession. This is especially –5–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

evident in the khutba of investiture delivered in 1261 by the first Egyptian Abbasid caliph al-Mustansir II, installed by Baybars; in the second khutba of the second caliph, alHakim, installed again by Baybars in 1262 after the death of al-Mustansir in his illfated campaign against the Mongols; and by the fact that the same khutba was again delivered by al-Hakim, more than thirty years later, when al-Ashraf Khalil released him from house arrest in 1291 and ordered him publicly to praise his conquest of Acre.15 The same justification was also adopted, for obvious reasons, by the historian Baybars alMansuri, the closest representative we have of a Mamluk viewpoint. He revived the old topos of the Turks’ higher military prowess both as archers and horsemen and their glorious military record and presented it as the reason why God had willed the passing of the rule to the Turkish Mamluks to preserve Islam and resurrect the caliphate.16 Much later, Ibn Khaldun connected the idea to the role of ‘asabiyya (a polyvalent Khaldunian term which signifies solidarity in all its permutations) in his metahistorical cycle of the formation and decline of states. He perceptively noted that the Mamluk system of one-generation aristocracy ensured the survival of the state for it preserved the intensity of a primitive ‘asabiyya and its concomitant fighting instinct among the freshly uprooted Mamluks; a characteristic that would have been definitely lost among their more refined and decadent offspring.17 But what is really revealing is that, on the whole, the chroniclers fail to list or even to mention the sociopolitical changes in the Mamluk realm that had become glaringly obvious by the third quarter of the thirteenth century. The new Mamluk system, which was to endure in its basic tenets until the early sixteenth century, is nowhere described in the Mamluk sources.18 A new ruling class, new elite language, new behavioural codes, even new legal and organisational rules, which the chroniclers definitely witnessed and sometimes suffered from, are

passed over in silence. Only al-Maqrizi in the first quarter of the fifteenth century summarises in a quite convoluted and perhaps anachronistic description – as is usual in his Khitat – some of the innovations of Baybars, Qalawun and al-Ashraf Khalil.19 The other organising steps undertaken in the formative period have to be teased out of the chronicles and biographies of the Mamluk sultans and their great amirs, which renders the prospect of determining the system’s chronological unfolding eminently dim. As for noting the system’s cultural underpinnings and their historical premises, the sources are even more vague. The few Mamluk historians other than alMaqrizi who mention in passing some of Baybars’ or Qalawun’s ‘inventions’ in terms of the army, the administration or the Mamluk duties and privileges, offer only sporadic, and at times probably fanciful, explanations and contextualisations of these inventions. Among the most famous examples are two dubious part-legal, partmythical justifications for influential Mamluk actions which otherwise seemed incomprehensible to the chroniclers. The first is the ‘Law of the Turks’, cited to explain the mechanism for succession adopted after the killing of Qutuz, and attributed to Aybak al-Musta‘rib, the assassinated sultan’s atabek al-‘asakir (army chief).20 The second is the Chengisid legal code, known as the Yasa and alleged by al-Maqrizi to have been copied from some original in Baghdad and vituperatively invoked as the source of Mamluk judgments that did not conform to the Islamic shari‘a.21

STRUCTURING THE NEW M AMLUK SYSTEM The Mamluk revolution generated a set of complications that needed to be resolved if the Mamluks were to legally retain the rule among themselves. One of the most critical problems was their status under Islamic law:

–6–

T H E CH A NGI NG CONCEP T OF M A M LUK I N EGY P T A N D SY R I A

were they free men and therefore entitled to rule and pass judgment over other free Muslims or not?22 Although the general assumption in modern scholarship is that the manumission of Mamluks before their conscription in the army was a foregone conclusion, there is nothing in the sources that indicates that this was the case.23 Many Mamluks of the Abbasids, Ghaznavids and Seljuqs seem to have remained slaves even when they led their masters’ armies or occupied high positions in the court. Even in the Ayyubid period there is no conclusive proof that the custom of manumitting young Mamluks after they had completed their training was universally applied. In fact, many references suggest exactly the opposite, namely that not all Mamluks were regularly freed before they were appointed to the army and given an iqta‘ (land revenue). For instance, when Husn al-Din Tha‘lab and his Arab Bedouins revolted against Aybak at the beginning of his rule, they declared that the Mamluks were the ‘slaves of the khawarij (i. e. the Ayyubids, whom they considered outsiders to Islam).’24 A similar conclusion can be reached from a report about the famous qadi (judge) ‘Izz al-Din ibn ‘Abd alSalam (1182–1262) who is said to have defied the Mamluk authorities, probably under Aybak, by insisting on selling some of his amirs, including the na’ib al-saltana (vicegerent), who, the qadi discovered, were still in bondage.25 Whether the story, which is reported in only two sources among the many that relate the life and career of ibn ‘Abd al-Salam, really occurred or not is immaterial. What is relevant is that the fourteenth-century jurist al-Subki, who was probably the first to record the story, finds nothing surprising in a group of unmanumitted amirs serving in the government in the thirteenth century, when such a situation would have been unthinkable in his own time. The question of manumission could not have been left unresolved or casually handled in a Mamluk state. Sometime dur-

ing the reign of either Baybars or Qalawun, the two real architects of the Mamluk system, freeing Mamluks after the completion of their training must have become the standard practice, though the sources say nothing about it. Even al-Maqrizi, in his unique description of the Mamluk system, credits no one in particular with implementing the practice of manumission.26 From alMaqrizi’s text and various references in other sources, however, it becomes clear that the Mamluk career revolved around the ceremony of manumission, which marked the Mamluk’s readiness to enter the army as a fully fledged member of the military elite. This process appears to have applied to all Mamluks whether they were trained in the royal or amirial households. However, as our knowledge of the amirial households is even sketchier than that of the royal one, for the purpose of outlining the acquisition and training of the Mamluks, therefore, we will have to depend on the little information available on the Mamluks of the sultan and assume that the amirial households were only smaller and less well organised versions of the royal household. Young royal Mamluks, al-mamalik alsultaniyya, were acquired from a wide range of sources, some of which were only temporarily viable before they dried up. The majority were purchased from the slave markets of Cairo or, to lesser extent, from markets in Syria, especially Aleppo and Damascus. Others were brought specifically for the sultan by special slave-traders – the most famous of whom was al-Majd alSallami who served al-Nasir Muhammad27 – from their lands of origin (the sources mention the land of the Qipchaqs, the Crimea, the Caucasus, Anatolia, and even Tabriz and Baghdad, meaning the Ilkhanid realm). Particularly valued recruits were received as presents from various courts, primarily from the Ayyubids of Hama or the princes of Anatolian principalities, and later from the Mongols of the Golden Horde, known to the Mamluk sources as –7–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

‘the kingdom of Uzbek’, after the first Muslim ruler of the Golden Horde who engaged in friendly relations with the Mamluks.28 A far from negligible number were either inherited from deceased sultans and amirs or transferred to the ruling sultan when he had confiscated the property of an amir who had fallen from grace, or been captured in campaigns against the Crusaders, the Mongols or the Armenians.29 In very curious and clearly unusual instances (in contrast to the late Ottoman devshirme system), young boys were seized from local Christian populations as punishment for siding against the Mamluks or for vicious actions against the Muslims, as happened to the inhabitants of Qara, north of Damascus, in 1264.30 The last known category are the Mongol wafidiyya or newcomers, who despite their numerical importance constituted so brief a phenomenon that they had a random and perhaps ultimately ephemeral effect on the composition of the royal and amirial Mamluks.31 We know that large numbers of Mongol refugees with their families (who were not strictly ethnically Mongol but included Turkic and other groups as well) came into the Mamluk realm on at least two occasions during the reign of Baybars (1262–1264 and 1267) and once during the short reign of al-‘Adil Kitbugha (1295).32 The sultans always welcomed them and tried to integrate them into the ranks of the Mamluk army probably because of the high esteem in which the Mamluks held their formidable Mongol adversaries and the fact that the Mongols were after all kin to the Qipchaqs.33 But the waves of wafidiyya were also sudden, unpredictable and of suspect loyalty. Moreover, enlisting some of the wafidiyya or their sons into the Mamluk army caused serious problems of cultural dissimilarity, in addition to the legal difficulty inherent in inducting free individuals into an essentially slave company. Mongol refugees therefore contributed very few members to the royal or amirial mamluks.34

Most of them were dispatched to the auxiliary and less privileged corps of al-halaqa, about whose composition and training during the early Mamluk period we know very little, except that it included free warriors in its ranks.35 Beginning with the reign of Baybars, young royal Mamluks were lodged at the Citadel of Cairo. There, they appear to have been grouped according to their projected function in the army (silahdariyya, jamadariyya, etc.) – which was sometimes decided upon according to their ethnicity – and were put through several years of rigorous training under the supervision of a group of strict and exacting masters. Their indoctrination began with basic religious education and instruction in reading the Qur’an, which inculcated in them a knowledge and a respect for Islamic doctrines, although it definitely did not erase all traces of their cultural past, especially their native languages. This was followed by military training of gradually mounting complexity, at the end of which the Mamluks would have learned the principles of horsemanship, swordsmanship, archery, war and parade formations, and combat. Once they had completed their training the young royal Mamluks were formally and publicly freed and given a manumission license (frequently called ‘itaqa, or simply ‘manumittance’) and appointed as soldiers ( jund) in the Mamluk army with a salary, a horse and a sword.36

SOLIDARITY AND THE M AMLUK SYSTEM In the citadel, young Mamluks were quartered in the tibaqs, a term usually translated as ‘barracks’ but which in reality indicated a variety of building types such as towers and underground halls.37 A few lucky ones – apparently those endowed with a striking physical appearance (at least during the reign of al-Nasir Muhammad) or those sent as gifts from other royal courts – were selected to reside

–8–

T H E CH A NGI NG CONCEP T OF M A M LUK I N EGY P T A N D SY R I A

in the royal palaces and to be tutored with the sons of the sultans as a prelude to their ascent to the highest ranks. But for the majority, the tibaqs were to become their home for five to ten years or even more. Life in the tibaqs must have been paramount in shaping the character, outlook and behaviour of the recruits. Our sources are, unfortunately, uninformed about it. They might mention some peculiar social arrangements in the tibaqs in reference to specific incidents, such as when the Mamluks collected donations for the needy among them, in the context of reporting an uprising against Qawsun in 1342.38 Otherwise they pass over Mamluk social formation in silence. The sources, however, often use – and at times probably misuse – an evidently tibaqs-derived terminology to describe some of the relational and behavioural social patterns among the Mamluks. Terms such as ustadh, agha, ani and especially khushdash appear in the sources to describe the alternative relationships that are said to have been developed by and for the Mamluks and that seem to have governed and dictated their notions of kinship and loyalty.39 An ustadh is the last owner and the manumitter of a Mamluk and is presented as a father figure who commands loyalty and obedience. A khushdash is their companion in the tibaqs and in the manumission ceremony, and he is thus seen as the equivalent of a classmate or even a brother with whom to bond and to whom a Mamluk owes fraternal devotion and support. An ani is a younger Mamluk in the tibaqs, a sort of a younger brother in need of protection and guidance. The agha, finally, is the teacher and supervisor (there were many aghas in each tibaq), who was more often than not a eunuch; a normal condition in royal households and a necessary precaution in an all-boy environment. The agha’s figurative image, however, is a little ambiguous: he is part schoolmaster, part lala (the Mamluk term for governess) and part father.

These and other correlations explicitly made between the Mamluk relational terms and the usual terms for blood kinships suggest a Mamluk structure in the tibaqs that is analogous to, and perhaps even surrogate for, normal family life of which the mamluks were effectively deprived.40 However, it is very difficult to confirm that this was indeed how the Mamluks themselves regarded their lot, or that their chroniclers had to fashion a structure fathomable to themselves to explain the alien criteria that governed Mamluk relationships. The relevance of these presumed relational structures in accounting for the shifting loyalties among the Mamluks after their manumission is at best mixed and is often disappointing. The number of incidents reported by the chroniclers in which a khushdash came to the assistance of his khushdash or an ustadh was unquestionably backed by his Mamluks are only exceeded by the incidents in which the exact opposite occurred.41 The Mamluk sources often mention another obvious (and in fact most problematic and troublesome) bond: the ethnic ‘asabiyya, which did not come to the forefront until after the reign of Qalawun. The majority of the Salihi Mamluks, and certainly all of the Bahriyya, appear to have been Qipchaqs. So were most of Baybars’ Mamluks, although the Mongol wafidiyya began to occupy a strong second place in the Mamluk army. Qalawun, in his fervour to augment the numbers of his Mamluks and his desire to mitigate the slackening supply of Qipchaqs, introduced new ethnic groups into the more or less homogeneous corps of the royal Mamluks, despite the fact that he was Qipchaq himself. His mamalik al-burjiyya (‘mamluks of the towers’, so named because they were quartered in the towers of the citadel) were made up completely of nonQipchaqs: most were Circassian, a few were either Armenian or Greek from Anatolia or Abkhazian.42 Not long after Qalawun’s death, rising ethnic ‘asabiyya seem to have competed with –9–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

the tibaq family allegiances to ustadh and khushdash whenever crucial questions of survival and succession arose. In two instances at least, Circassian and Mongol ‘asabiyyas dictated the shuffling of alliances during the turbulent first two reigns of al-Nasir Muhammad between 1293 and 1310. Two successful usurpers among his father’s amirs, Kitbugha and Baybars al-Jashankir (r. 1308–1310) appear to have depended on their ethnic ‘asabiyyas to ascend the throne. Kitbugha appealed to his Mongol kin, whereas Baybars al-Jashankir depended on the Circassian element which constituted the majority among the Burjiyya.43 The only one among them, Sanjar al-Shuja‘i, who tried to rely on the royal Mamluks of the tibaqs to advance his candidature lost his life in 1294 in the attempt, although he probably was the most cunning of the Mansuri amirs.44 Other strong bonds were certainly at work, bonds that softened, transcended or incorporated the effects of the tibaq family allegiances or the ethnic ‘asabiyya. Otherwise, the modicum of class cohesion that allowed the Mamluks to remain in power for two and a half centuries would be very difficult to understand. Some ties were appropriated through the Mamluk training system and inculcated in the young trainees to counter the residual disruptive effects of their past loyalties, not very successfully it must be noted, or to strengthen their integrative ones. Two bonds in particular – the foreignness of the Mamluks and their shared experience of slavery – were clearly and consciously used to distinguish them and to reinforce their separation from the local population, although these were possibly less effective in restraining their interclass rivalries. First was a remembered homeland outside the Mamluk realm and alien to it, alsharq, the east, or al-bilad, the country, in the parlance of the period – the steppes, the lands of the Qipchaqs during the early period, to which was added the Caucasus at

a later stage. We do not know how the Mamluks themselves regarded the vast differences between their various countries of origin but our sources rarely distinguish between them. They even present al-sharq as the material foundation of a generally fictitious yet apparently strongly believed common origin – the nomadic Turkic one – at least in the formative period, with longidealised notions of horsmanship, military prowess, pride, tenacious attachment to freedom and obedience to one’s leader. Even those Mamluks brought from Anatolia and Armenia or captured from the Crusaders and the Mongols had to adopt Turkish names once they joined the Mamluks’ ranks (the exception seems to have been the Mongol wafidiyya, who not only kept their names, but seem to have been permitted to retain their un-Islamic customs).45 Second was the slavery bond: a real and transforming experience which was probably institutionalised from the beginning and made a prerequisite for inclusion within the Mamluk class primarily because of its effect in creating a sense of belonging and of social cohesion. Obviously the sources, mostly written by local ulama and awlad al-nas (free-born sons of the Mamluks), could not fully appreciate either bond; they either ignore them or try to homogenise or overstate their signs in their writing. The measures introduced by the early Mamluk sultans thus marked a new development in the meaning of the word mamluk, at least in medieval Egypt and Syria. No longer was a Mamluk a warrior-slave subjugated all his life to his master: he was now destined to be freed and conscripted into the army to begin his ascent through its ranks, sometimes all the way to the top – that is, to the position of sultan. Once he reached the rank of an amir, a Mamluk was also expected to begin to purchase, train and free his own Mamluks. Thus a purely Mamluk caste, that is a caste made solely of people of diverse ethnic origins, all foreign and forcibly imported to Egypt and Syria,

–10 –

T H E CH A NGI NG CONCEP T OF M A M LUK I N EGY P T A N D SY R I A

who had gone through the Mamluk experience of enslavement, conversion to Islam, systematic and formal military training and finally manumission could be created and perpetuated. This Mamluk caste remained overtly foreign in inspiration and outlook at least until the middle of the fourteenth century. After nearly seventy years of almost continuous warfare, the Mamluk state had managed to eliminate the dangers threatening its hold on Egypt and Syria. The stabilisation of Mamluk rule and its acceptance by most subject groups led to the relaxation of the Mamluk insular lifestyle during the long, relatively stable and highly autocratic rule of al-Nasir Muhammad.46 The Mamluk ruling elite began slowly to adopt some of the characteristics of the indigenous urban culture while, at the same time, some of the

Mamluks’ more peculiar customs seeped into the local milieu. This process continued after al-Nasir Muhammad’s reign, and seems to have been accelerated by the softening of discipline in Mamluk training which occurred in the last quarter of the fourteenth century. New Mamluks were recruited at a fairly advanced age, after their character had already been formed, and were no longer required to undergo an extensive religious and military education before their manumission. By the beginning of the fifteenth century the Mamluks’ acculturation was evident in their social attitudes, tastes and preferences, and also in the lack of interest they showed in their military role as defenders of Islam or their pro-religion stances in their political image-making.47 But this is a different topic that requires a separate treatment.

– 11 –

2. REPRESENTING THE MAMLUKS IN MAMLUK HISTORICAL WRITING

M

amluk historiography is among the richest and most varied of Islamic historical traditions. Scores of annals, biographical compendia, manuals for the chancery, geographical treatises (masalik) and topographical tracts (khitat) span the period between the thirteenth and sixteenth centuries.1 The majority of these sources have been edited and published, many in the last quarter-century, which has added considerably to our knowledge of the history of the period and to our understanding of the parameters, structures and limitations of this extraordinary historical output.2 Yet scholarly evaluations and interpretations of Mamluk historiography are few.3 Even fewer are those studies that go beyond primarily historiographical questions, such as the analysis of methods, genres, periodisation and scopes, to theorise on the concepts, uses and import of history and history writing in the Mamluk society.4 In this chapter, I will focus on a crucial but rarely discussed problem in Mamluk historiography: the representation of the Mamluk ruling class in contemporary chronicles and biographies. I will try to answer two questions. The first is, how are the Mamluks represented? The second is,

does this representation reflect the Mamluks’ motives, thoughts, mindsets and beliefs, or those of the chroniclers and interpreters who wrote about them? My argument will be that, more than most other analogous historical categories of class, race or sect – which are all in essence discursively constructed – the Mamluk image in history is a construct in which the Mamluks themselves played a negligible part. From the perspective of Mamluk historical sources, Mamluk urban society can be divided grosso modo into three basic groups. First are the ruling Mamluks, a hodgepodge of imported slave-warriors of various ethnicities, but mostly Turkish or Turkicised, who formed an exclusive military elite. Second are the learned people or those whose professions involved reading and writing, who, for simplicity’s sake, will be referred to as the literati; a heterogeneous assemblage which includes the ulama, the kuttab, and most of the awlad al-nas – the Muslim-born sons of the Mamluks who were usually encouraged to pursue careers in the administration or the religious establishment.5 Finally, there are the common people, encompassing several socioeconomic layers from rich international

–12–

R EPR ESEN T I NG T H E M A M LUK S I N M A M LUK H ISTOR ICA L W R I T I NG

merchants to poor artisans, with a few fluid fringe elements such as the Sufis and the ahdath (urban youth). These three groups did not constitute clearly differentiated social formations or classes in the modern sense, although they could be classified according to a loose economic stratification with the Mamluks way up above all the rest, followed by some international merchants who were much richer than the literati, who constituted a kind of middle professional class, and finally the commoners.6 Nor were these classifications clearly demarcated or totally impermeable, although movement in and out of the Mamluk caste was very restricted, and those who managed it were exceptional cases that have to be explained on an individual basis.7 The ranks of the literati, on the other hand, were in principle open to all people, military and common alike, although ulama and kuttab families strove to preserve their status by manipulating salaried offices and trying to pass them on to their sons, often with at least short-lived success.8 The three categories are neither equally nor comparably represented in the primary sources. The ruling Mamluks and the commoners in all their diversity are either stereotypically or caricaturally represented or misrepresented, though to varying degrees and in different ways. The Mamluks occupy centre stage in most historical narratives, but they appear literally as actors given lines to read, or worse still, they are spoken (or written) for and about, but rarely speak themselves. They are, in other words, truly re-presented. The commoners, on the other hand, are seldom referred to either individually or collectively. The only exception are rich merchants who are singled out for their learned status or when their wealth is publicly confiscated by the authorities. Otherwise, when they are mentioned at all, the commoners are almost always represented in general terms, such as al-nas (people), al-

‘amma (the commoners) or ahl al-balad (the inhabitant of the place), or as accessories or ornamental devices to complete a scene or to accentuate an event. The literati, finally, make up an articulate and highly visible group between the Mamluks and the commoners. They controlled all historical writing, dominated the narratives, and represented themselves and others through their own perceptions, views, prejudices and frameworks of interpretation. The majority of the authors and compilers of historical books were members either of the ulama or the kuttab classes (the two greatly overlapped in the Mamluk period). A sizable minority consisted of awlad al-nas, who should be intellectually, socially and ideologically classified with the ulama and kuttab despite their Mamluk lineage, Mamluk privileges and knowledge of the Turkish language. Only very few of the known authors belonged to the Mamluk caste itself. One was Baybars al-Mansuri (d. 1324/25), a great amir who held several high positions under al-Nasir Muhammad. Another was the unknown author of the chronicle edited by K. V. Zetterstéen, who was perhaps a simple jundi (soldier) in the service of a Nasiri amir. Still another was the author known as Shams al-Din alShuja‘i, tentatively identified as Aqsunqur al-Muzaffari Amir Jandar, who played an important though very brief role during the brutal reign of al-Muzaffar Hajji (1346–1347).9 They should, theoretically, provide us with different perspectives on the Mamluk culture as lived by their class, but aside from a scattering of scholastic remarks on the Arabic meaning of Turkish terms used by the Mamluks and a palpable sense of pride in the Mamluks’ role in defending a besieged Islam in the thirteenth and early fourteenth centuries, neither the awlad al-nas nor the Mamluk historians provide any unusual insight into the Mamluk mentality. The exceptions are Ibn al-Dawadari in the fourteenth century and Ibn Taghri-Birdi in the fifteenth, who

–13 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

occasionally drop a hint on the significance of some Mamluk custom or action, and the peculiar and uncommonly revelatory but unfortunately incomplete chronicle of alYusufi who apparently was a low-ranking yet highly opinionated local recruit in the halaqa in the entourage of al-Nasir Muhammad.10 Apart from these, no identifiable Mamluk perspective emerges from the texts written by other Mamluks and sons of Mamluks. Why is unclear. One reason might be the desire of the two groups, who were literary newcomers, to identify with their local scholarly masters by adopting their dominant strategies of interpretation and by underplaying or omitting their Mamluk outlook from their writing.11 The awlad al-nas’s attempt to join the literati’s fellowship, nonetheless, was not always well received by the local literati. Ibn TaghriBardi in particular, probably because of his pomposity and propensity to correct his peers’ interpretation of Turkish words, was recorded by many chroniclers as being both an ‘ammi (commoner) and ignorant as well as prejudiced to the Turks and, an even worse accusation, to the Copts.12 Misrepresenting, or even ignoring, the common people is to be expected. It is typical of pre-modern historiography everywhere, when only a privileged minority was literate and learned enough to engage in historical writing and interpretation. Medieval historians focused on dynasties, kings and literati – which in the case of Muslims meant mainly the ulama of all specialisations – to the exclusion of the lower classes who were seen as static, undifferentiated and largely uninteresting. The few cases where a historian pays more than passing attention to the common people in general or a commoner in particular, such as in the extraordinary book by Abu Shama, al-Dhayl ‘ala al-Rawdatayn, are surprising, isolated and unfortunately rare.13 But a medieval historiography in which the ruling class from the monarch down has

little control over whether and how it is represented, although not a uniquely Mamluk phenomenon, is unusual. It stems from the circumstances that governed the relationship between the military Mamluks who made history and the literati who recorded it.14

REPRESENTING THE M AMLUKS IN M AMLUK HISTORIOGRAPHY In the Abbasid period, the Mamluks and the literati were distinctly separate groups although over time they had developed a symbiosis from which they both benefited. In later regimes that used Mamluks in their armies – especially the Seljuqs, Zengids and Ayyubids – both Mamluks and the literati served the ruling dynasty or clan, and had equal access to power. Several of the literati, in fact, reached high positions, such as al-Qadi al-Fadil (1131–1199), Salah al-Din’s vizier, who helped shape the policies of his sultan and stirred him towards the jihad against the Crusaders.15 Some among the important literati even owned their own Mamluks. The most notable case is that of Nizam alMulk al-Tusi (r. 1063–1092), the great vizier of Alp Arslan and Malikshah. In his famous treatise on politics, Siyasat-Namah, he stressed the importance of employing Mamluks in what had up to then been an army of mostly free-born horsemen. He also surrounded himself with an army of Turkish Mamluks that numbered in the thousands. They were probably the reason why he kept his position for as long as he did in the treacherous Seljuq court.16 When the Mamluk regime took over in Egypt and Syria, however, it imposed new restrictions on the civilian elite that redefined the relationship between the Mamluks and literati. It effectively barred the literati from attaining high political positions.17 The army and, through it, all political power and its attendant financial

– 14 –

R EPR ESEN T I NG T H E M A M LUK S I N M A M LUK H ISTOR ICA L W R I T I NG

and landholding prerogatives became the exclusive domain of the Mamluk class. This military aristocracy was closed to all but the Mamluks (and, after the formative period, their sons or, in few cases, exceptional local recruits). Their clannish mentality, reinforced by linguistic and ethnic separation from the rest of the society, erected a regime that stressed exclusion and segregation as means of control. It was inscribed in particular nomenclature and insignia (rank) systems and in elaborate vestmental, dietary, musical and ceremonial codes.18 In the Ayyubid and Mamluk periods, the literati appear to be a loosely defined and relatively open intermediate social group. They included, in addition to the traditional categories of ulama and kuttab, who constituted by far their majority, three subcategories that were either new or particularly Egyptian or Mamluk in origin. The first were the ulama/kuttab who straddled both vocations and were a product of the officially sponsored madrasa system first introduced into Egypt under Salah alDin. The second were the Christian Coptic kuttab and converts. They played an extremely important role in the administration of the Mamluk state, but from the standpoint of history writing their contribution was minimal. The third subgroup, the awlad al-nas, was essentially a byproduct of the Mamluk caste’s structure as a onegeneration military aristocracy. The awlad al-nas were rarely recruited into the Mamluk army, at least until the reign of alNasir Muhammad.19 Even after the changes instituted by al-Nasir allowed their promotion to high amirial levels, most awlad al-nas still belonged to the literati and chose, or were compelled to accept, a career in a religious or administrative field. The Mamluk regime permitted the literati little political power, but they dominated the production and transmission of knowledge in addition to the traditional religious functions held by the ulama, which

kept them in touch with the people and in a position to affect public opinion. They also controlled the judicial, administrative, educational and waqf services through which they wielded tremendous influence as agents of mediation and arbitration.20 The Mamluks were thus compelled to patronise and employ the literati to manage the sultanate and maintain its religious, social and fiscal systems. However, they kept them under constant check, enforced with the real threat of confiscation, arrest and sometimes exceedingly brutal punishment. This paradoxical situation affected the ways in which the literati saw their relationship with the Mamluk ruling class, from whose ranks they were banished but to which they were financially, politically and even existentially beholden. It translated into an attitude of uneasy acquiescence laced with jealousy and an affected haughtiness, that found their way into all genres of writing of the time, but especially historical/biographical texts. The Mamluks were the rulers to fear and the patrons to please: they had to be presented in a positive light. The praise often comes across as transparently perfunctory, however, especially in the official, commissioned annals, such as the three encomia for Baybars, Qalawun and alAshraf Khalil, written by Ibn ‘Abd al-Zahir (1223–1292), who held high chancery positions under all three.21 More sincere and heartfelt were the representations of the Mamluks, particularly the founders from Qutuz to al-Ashraf Khalil, as the heroes of jihad who defended Islam against its enemies: the Crusaders, the Mongols, the Armenians and some domestic splinter groups.22 This loudly and justly celebrated achievement, however, did not obfuscate the frequent though prudently subdued discontent with the Mamluks as a ruling class. Mamluks were the manipulators of financial rewards and public office coveted by many literati and, in the final account,

–15 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

the outsiders who had seized and maintained power by force. Thus, from the early days of the regime, the Mamluks’ military and equestrian qualities were seen as both the source of their success and the tools of their coercion, and therefore grounds for both admiration and apprehension. The castigation became louder and more explicit in the fifteenth century, when the Mamluks began, for various reasons, to neglect their political and religious duties. Contemporary observers no longer saw them as the deserving leaders they had once been, skilfully and thoughtfully managing a great empire and fighting for the cause of Islam.23 Another set of mostly disparaging remarks turn up in the sources whenever the non-military qualities of the Mamluks are discussed. Religious learning and literary and artistic talents – all trademarks of the literati – were also displayed by at least some of the Mamluk amirs and even simple soldiers. But when they were, the chroniclers either explained them away or diminished their value. A Mamluk might be mentioned as a composer of poetry in Arabic, for instance, but then he is either put down as ‘ignorant of the intricacies of the language’, or described as ‘unusual among members of his race’.24 Phrases like these reflect the underlying belief that the Mamluks – because of their race and ‘natural’ aversion to learn – were incapable of mastering the Arabic language or other intellectually demanding sciences. This view may also have represented an attempt to disqualify the Mamluks, and even the awlad al-nas, from attaining high rank in either religious sciences (‘ilm) or literature (adab). This would automatically ensure that they could not encroach on the sacrosanct domain of the literati. Having been made to relinquish their political power to the Mamluks, the literati were left with their religious, literary and intellectual advantages. They could not entertain, let alone fully admit, even the slight possibility

that the Mamluks could also share in these privileges, which they were intent on protecting from any infiltration. This sociopolitical gap was widened by another source of separation between the Mamluks and their historians. This was language. The Mamluks spoke Turkish and sometimes colloquial Arabic; other ethnic groups among them, such as the Mongols, spoke their own native language as well.25 When they did write, at first the Mamluks wrote mainly in Arabic and Qipchaq Turkish, later on they wrote in Arabic and Oghuz Turkish (called also Turkoman and later Ottoman).26 Local literati spoke colloquial Arabic, perhaps in a more refined form than that used by the Mamluks or the common people, and wrote only in classical Arabic, albeit in a prose that reveals the degree to which some vernacular expressions (such as aysh [what]) had penetrated even the highest literary and judicial language in the medieval period. Two subgroups of the literati were native speakers of Turkish: the awlad al-nas, especially those whose mothers were Turkish, and the Turkish Anatolian, Iraqi and Azerbaijani literati who came to the Mamluk sultanate in the Burji period in search of employment and patronage.27 Arab literati rarely tried to learn Turkish, possibly as a sign of alienation or contempt. Whenever an individual among them showed fluency in both Arabic and Turkish, the chroniclers seem to have been utterly surprised. Qadi Sulayman ibn Ibrahim, known as Ibn Katib Sunqur, is said by alSafadi (himself a member of the awlad alnas) to ‘have [had] the gift of composing rhymed [Arabic] poetry almost effortlessly, although he speaks fluent Qipchaq Turkish’,28 seemingly implying that the two skills are somehow incompatible. The polyglot qadi Ibrahim Jamal al-Kufat, who served in the administration from the end of al-Nasir Muhammad’s reign until his execution by al-Salih Isma‘il in 1344, was put in a class of his own because he spoke,

–16 –

R EPR ESEN T I NG T H E M A M LUK S I N M A M LUK H ISTOR ICA L W R I T I NG

beside flawless Turkish, two African languages, Takruri (of Takrur, present-day Tchad and Mali) and Nubian.29 In the later, or Burji period, the historian Badr al-Din al-‘Ayni was noted for his fluency in both Arabic and Turkish, which he may have learned in his youth in ‘Ayntab (Gaziantep in Turkey today), which had a sizable Turkoman community in the fourteenth century.30 He is reported to have read his chronicle, ‘Iqd al-Juman, aloud to his patron, Sultan Barsbay, and translated it into Turkish as he went along. Still, in the dignified custom of the time he published it only in Arabic, and apparently never recorded his Turkish translation despite the broadened audience such a translation would have provided, not to mention rewards from powerful Mamluk patrons. Only one late-Mamluk historian, to my knowledge, composed a history in Turkish. This was the qadi al-jaysh (qadi of the army) Shams al-Din Muhammad ibn Aja (1417–1476/7), who wrote a version of Futuh al-Sham of al-Waqidi in Turkish, probably for his patron the great amir Yashbak min Mahdi, who was himself a learned man of diverse scholarly interests.31 The language barrier must have caused communication problems between the Mamluks and the literati, although the chroniclers never speak of any.32 They do, however, note that many key figures among the Mamluks, including Sultan Qalawun, barely spoke Arabic.33 They even use a disparaging term, ghutumi (inarticulate or dumb), to describe those Mamluks who failed to learn proper Arabic,34 although they do not tell us how they communicated with them and reported what they said. This raises serious questions about the veracity and authenticity of the numerous conversations between members of the Mamluk ruling class reported in the sources,35 for they must have been originally spoken in Turkish (I doubt that any extensive dialogue was in colloquial Arabic.)36 This means that they had to be

translated for the chroniclers who did not understand Turkish before they could be recorded in the form we have today. The rephrasing would surely have distorted their content and tone. Furthermore, the nuances, images and figures of speech used by the Mamluks would also have been lost to us through translation, so that we will never be able to glimpse their world through their own words. The problem, however, goes beyond the inability of most Mamluk historians to convey the content, tone and nuances of dialogues of the Mamluks’ original Turkish accurately to what I think is deliberate misrepresentation. Many of the conversations between Mamluks reported in colloquial Arabic in the sources appear to be not translated, but rather fabricated, or at least edited and embellished.37 This is not the common form of falsification where a historian twists, omits or exaggerates a fact either to serve partisan ends or to please a patron or ward off his anger. Such a practice was current wherever historians had to work under the patronage of powerful politicians and noblemen, and it is not confined to Mamluk historians. Nor did the Mamluk historians invent or ornament these dialogues only to make their accounts more entertaining and attract a wider audience in a period noted for the profusion of historical writing with a popular bent.38 The reason, I think, has primarily to do with the image of the Mamluks that our historians, as the representatives of the literati class, wished to project; an image, it has to be added, that was far from complimentary. A glaring example of what I am describing can be found in a discussion that ostensibly took place between Bashtak and Qawsun, the two competing great amirs and close confidants of Sultan al-Nasir Muhammad, who were with him when he died.39 The report transmitted by al-Safadi begins by saying that Qawsun went to the window with its view overlooking Cairo and

– 17 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

called Bashtak to him for a confidential conversation. Qawsun claimed and ultimately managed to convince Bashtak that since neither of them had a real chance to become sultan they had better stop plotting against each other. His argument was that neither of them was a real Mamluk: they had both started as petty peddlers, were noticed by al-Nasir primarily for their physical attractiveness, and were bought by him from one another when neither legally owned the other. The reasons offered by Qawsun to dissuade Bashtak from seeking the throne were both logical and convincing. I had even previously considered them of paramount importance in explaining the events that followed the death of the long-reigning and formidable al-Nasir Muhammad, when neither Bashtak nor Qawsun – after he managed to eliminate Bashtak – dared usurp the sultanate despite their power and cunning.40 But now I think that the discussion was altogether fabricated, because of the circumstances of its occurrence, its content and language. First, it is difficult, though not entirely impossible, to explain how it could have been transmitted to al-Safadi when it had allegedly taken place in the sultan’s chamber by his deathbed. Second, Qawsun’s statement that al-Nasir bought him from Bashtak contradicts what we know about how he became a Mamluk. AlSafadi himself in his biography of Qawsun in the same book, al-Wafi bi-l-Wafiyyat, specifies that al-Nasir sent Qawsun’s price to his brother Susun back in the Crimea.41 Other sources report the unusual procedure by which the sultan bought Qawsun from himself.42 However, the strongest refutation of the authenticity of this discussion is the language in which it is presented: a vernacular Arabic of the kind that was probably spoken among the lower classes in Cairo. Qawsun and Bashtak would have spoken Turkish, since neither of them was really at home in Arabic, having learned it at a rela-

tively late age.43 Furthermore, the street dialect is interspersed now and then with incongruously high literary expressions, such as ma yasa‘ana illa imtithali amrihi (we cannot but obey his order), that were not a typical part of that vernacular dialect. Inserting dialogues in historical narrative is a known technique for lending an account an appearance of authenticity and immediacy. The more plausible the language and expressions used in these conversations, the more believable the account itself.44 In the case of the Qawsun/Bashtak dialogue, Turkish would have been the most authentic. Al-Safadi could have explained first that the exchange was originally in Turkish, since he himself spoke it fluently, then proceeded to translate it into the kind of classicising Arabic he used in his writing. Instead, he chose to render it in an affected vernacular (presumably because he could not produce an exact imitation) in a way that suggests that these were the exact words of Qawsun and Bashtak. The puzzling question is why did alSafadi use this street vernacular in reporting a conversation between the two amirs when his standard conversation language was a more classicising one? The answer, in my opinion, is that he wanted the dialect to carry more than the impression of familiarity with the protagonists and closeness to the event. He intended it to signify the uncouth and uncultivated Mamluks, something akin to the way Hollywood films have German, Asian or Arab villains speak in broken English to show their villainy rather than their unfamiliarity with the language (which they obviously would not have spoken to each other in the first place). Al-Safadi is not the only Mamluk historian to have used this literary device in his narrative. Others used it as well to reflect the contempt in which the literati held the Mamluks.45 We can sometimes even sense the relish with which they twisted the tongues of their Mamluk protagonists and made them utter vulgar

–18 –

R EPR ESEN T I NG T H E M A M LUK S I N M A M LUK H ISTOR ICA L W R I T I NG

expressions.46 This way, the literati could mock the Mamluks or put them down without running the risk of being detected and punished. Conversely, the spoken words of the literati or of ‘pure’ Arabs, such as ‘Issa ibn Muhanna, the famous Bedouin chief, are reported in grammatically correct, classicising Arabic, even by chroniclers who otherwise record their own words in vernacular, such as al-Yusufi.47 The hostile relationship between the Mamluks and the literati, therefore, is not only expressed in the representations of the Mamluks made by the literati in the sources. It is also often inscribed in their words which are in turn encoded with opposing values. Dialogues in classical Arabic seem to imply culture and learning or to designate a native Arabic speaker, both seen as positive values; speech in the vernacular denotes either an uneducated or a foreign speaker, and conveys a sense of derision and insolence.

Thus, the binary image of the Mamluks as valiant defenders of Islam or rapacious exploiters of their subjects, often encountered in historical texts and emphasised in modern histories, represents only two facets of a complex portrayal by the literati. These two aspects were dampened and occasionally undermined by sometimes equally essentialising, sometimes subtler, representations, distortions and even downright omissions. The whole tells us more about the literati and their mixed and shifting feelings towards the Mamluks than it does about the Mamluks, their culture or their views of themselves. This conclusion, elementary as it may seem from a post-structuralist perspective, has rarely been discussed or taken into account in contemporary Mamluk studies, even by those concerned with cultural and social questions.48

–19 –

3. PERCEPTION OF ARCHITECTURE IN MAMLUK SOURCES

M

amluk architecture is one of the most extensively though unevenly studied categories in the field of Islamic architectural history today. Several surveys, varying in scope, numerous articles and monographs on individual monuments, and a few comparative studies of regional variations in architectural style exist. Many more are being published at an unprecedented rate as the field of Mamluk studies gains more students and researchers, and now has a journal of its own, Mamluk Studies Review.1 Even a few preliminary theoretical discussions have been held on some of the formal, symbolic and sociocultural attributes of this architectural tradition, and a number of historiographic essays have attempted to understand it in the context of Mamluk and Islamic cultural and social history, something that is generally lacking for other medieval Islamic architectural traditions.2 This scholarly attention should not be surprising to anyone familiar with the sheer number and variety of Mamluk buildings still standing in Egyptian, Palestinian, Jordanian, Lebanese and Syrian cities – and they constitute only a fraction of the total that can be computed from the sources. For 267 years, scores of projects of

all types – small and large, private and public, pious and commercial, pompous and poised, purposeful and frivolous – were sponsored by sultans, amirs and members of the local elite in practically every corner of the sultanate, particularly in Cairo, Damascus, Aleppo, Tripoli and Jerusalem, but also in smaller towns and villages.3 Mamluk written sources, chronicles and biographical dictionaries, but especially encyclopedic manuals, geographical treatises (masalik) and topographical tracts (khitat), in their capacity as records of their time reflect both the profusion of buildings and the interest in architecture that Mamluk culture manifested. They all pay more than passing attention to buildings and land reclamation projects sponsored by sultans, amirs and lesser notables. Some, like Ibn Shaddad, al-Maqrizi and Abu Hamid al-Qudsi, show a genuine interest in buildings and cities and sometimes even an expert and appreciative handling of their particular qualities in the descriptions they provide of them – in fact, each of them makes buildings the backbone of one key book in his historical oeuvre. Many biographers, especially those directly commissioned by some sultan or grandee, even wax lyrical on the projects sponsored by

–20 –

PERCEP T ION OF A RCH I T ECT U R E I N M A M LU K SOU RCE S

their patrons. Sometimes they exaggerate their numbers, costs and sizes. At other times, they emphasise their grandeur and rhetorically compare them with paradigmatic monuments known from literature or from the past. The veracity, intensity and enthusiasm of their coverage, or lack thereof, however, were neither constant nor uniform. They fluctuated over time, following both the shifting investment in architecture among the Mamluk patrons, sometimes from one reign to the next, and the inclination of the individual reporters to notice and discuss it which may or may not have been affected by the importance placed on building by the Mamluk patrons. Yet, over the entire Mamluk period, there is a marked progression in the reports towards a more informed and involved discussion of buildings and projects, and even a growing interest in their architectural, historical and sociocultural qualities. This evolving attitude seems to have transcended the individual inclination of a particular author; it affects every genre of historical writing, even annals and biographies, aside from its more concrete consequence of animating special types with architectural focus such as the masalik and the khitat. It is discernible in the texts of the most architecturally reticent among the late-Mamluk authors such as al-Suyuti and al-Sakhawi, who could not help but reflect the more sophisticated handling of architecture achieved by their literary peers.4 The trend appears to have peaked in the late fourteenth/early fifteenth century, at a time when the culture in general was coming to terms with the magnificent architectural endowments of the previous Bahri period which changed the face of many Mamluk cities, especially Cairo. This is the moment when Ibn Khaldun came to Cairo and declared it to be the centre of Islam and the epitome of ‘imran, a concept encompassing both civilisation and urbanisation, which he was busy theorising about

at the same time.5 He was soon followed by his brilliant student, Ahmad ibn ‘Ali al-Maqrizi, who devoted a tremendous amount of time and effort to producing the first encyclopedic work on the history, development and architectural monuments of a city in Islam, al Mawa‘iz wa-l-l‘tibar biDhikr al-Khitat wa-l-Athar, which is essentially a paean to Cairo. Other contemporary scholars like Ibn Duqmaq, Ibn al-Furat, alQalqashandi, Badr al-Din al-‘Ayni, Ibn Hajar al-‘Asqalani, Ibn ‘Arab Shah and Ibn Taghri-Birdi, though not as architecturally articulate as al-Maqrizi was or as theoretically astute as Ibn Khaldun, still show in their different ways a maturing sensitivity to the role of architecture in the life of the city and the reputation of patrons. However, this cultural interest in buildings and urban projects was not without its immediate political agenda: Mamluk authors for a variety of reasons disapproved of the Burji sultans, comparing them unfavourably to the great sultans of the late thirteenth and early fourteenth centuries. One of the main arguments they used to disparage their contemporary sultans was that they could not maintain the urban and architectural momentum generated by their illustrious predecessors, and they thus lacked their drive, commitment, good management and generosity. Many Mamluk authors harp on this point, even some who belonged to the Mamluk ruling class, such as Ibn Taghri Birdi.6 This vocal criticism, however, may indicate not so much a general and popular disapproval of the Mamluks’ performance as rulers as it did a growing divergence between the ruling Mamluks and the educated classes who controlled all historical writing and represented themselves and others through their own views, prejudices and frameworks of interpretation.7 Historicising the Mamluk interest in buildings, identifying its various proponents among the historians and analysing their different approaches and textual techniques,

– 21 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

and elucidating its conceptual ramifications for the study of Mamluk architecture and culture form the subject of this chapter. My approach and interpretations have been greatly influenced by the ideas of the late Ulrich Haarmann on the writing of Mamluk history, especially as he began to articulate them in his latest contributions before his enormously regretted and untimely death. This chapter is but a small token to his brilliant and original scholarship.

THE PHILOLOGICAL AND LITERARY CONTEXT OF WRITING ON ARCHITECTURE By its complex nature and exigencies, architecture can be neither a solitary nor modest activity. As Ibn Khaldun noted, architectural projects, whether monuments or entire cities, required huge outlays of time, money and manpower that could only be supplied by strong, stable, wealthy and – most importantly – urban patrons.8 Architectural projects fulfilled social and pietistic functions and went a long way towards enhancing the reputation of their founders and patrons, propagating their claims and embellishing their images. This made architecture, especially when it came to monumental buildings, primarily a royal or elite pursuit, and, as such, grist for the mill of chroniclers and biographers who wrote on the lives and deeds of the ruling class. The interest shown in architecture by the chroniclers of the Mamluk period, however, is not new in Islamic historiography. Biographers from earlier times recorded royal architectural projects and noted some of their peculiarities when they summarised the deeds of their founders. But Mamluk authors paid considerably more attention to architecture than their predecessors had done both in scope and depth. Their references were more numerous, comprehensive and detailed than those of earlier historians, although like their pred-

ecessors and their successors until the nineteenth century, they never used graphic illustrations to convey their impressions of the buildings they described. They did, however, make a great effort to emphasise urban, political, social, economic and cultural contexts, though rarely to consider formal, artistic or symbolic significance. Aside from mentioning how large or tall or strange a building was, or listing particularly expensive materials in its construction, or indicating that a certain surface was ornamented using a certain complicated technique, formal or spatial qualities of the buildings were passed over in silence. This unaesthetic tendency is apparent in the language used by Mamluk authors as well. When they write about architecture, they use primarily mundane and functional terms and rarely treat any spatial, artistic or conceptual point. Buildings are hardly ever qualified as beautiful (hasan, jamil), proportionate or harmonious (mutanasib, mu’talif, muntazam), or pleasing (bahij),9 all terms associated with aesthetic concepts that had a venerable history in philosophical and adab treatises of an earlier period. Although they appear most frequently in literary and abstract discussions, they may initially have been introduced to express formal or visual appreciation of human types, of objects or of engineering projects.10 But they do not seem to have made their way into architectural description. On the other hand, they had been absorbed into literary criticism and were frequently used to express aesthetic judgment of prose or poetic style. Many of the Mamluk authors who write about architecture in fact show a certain ease with the denotative intricacies of these aesthetic terms when applied to literary analysis, which they all practiced and proudly displayed in their soberer books, though they seem not to have been able, or perhaps had no interest in, making the leap from literature to architecture.

–22–

PERCEP T ION OF A RCH I T ECT U R E I N M A M LU K SOU RCE S

Formal and architectural investigations seem to have been outside the intellectual curiosity or scholarly training of Mamluk authors.11 Because of that handicap, they do not seem to have developed the techniques and terminology to carry out such examinations, and it shows in their texts. They apparently never attempted to transpose familiar aesthetic concepts from the literary domains, with which they were thoroughly familiar, or the less practiced disciplines of philosophy, geometry, music and the like to the unfamiliar field of architecture.12 Nor does it seem to have occurred to them to adapt the professional vocabulary that might have been used by the builders to describe the buildings because of the sharp social division that separated them from these craftsmen and artisans and that consequently hindered communication between the two social groups.13 Philosophical ideas, however truncated or distracted, sometimes did seep into Mamluk texts, but virtually no professional architectural or constructional terms at all found their way into them.14 The very few and significant exceptions, such as Ibn Shaddad, al-Maqrizi and Abu Hamid al-Qudsi, who at times reveal a certain affinity with professional terminology, may have developed their interest in the building crafts after having been exposed to them in some official capacity, such as serving as muhtasib (city inspector) as in the case of al-Maqrizi, or shadd (building supervisor), or some other similar function. It should be stressed, however, that the predisposition of these three to buildings and to the ways they are apprehended by craftsmen was peculiar to them: not all who wrote and also served at some point as building supervisors show either interest or a comparable mastery of professional terminology and modes of description.15 When buildings are at all noticed in the Mamluk sources for their visual qualities, they are generally described as unusual or marvellous (‘ajib and gharib) and

never further elaborated on; or they are mentioned for their monumentality and display of wealth, usually expressed in terms such as kabir, ‘azim or fakhir. Although monumentality is primarily considered an aesthetic and spatial quality in today’s architectural discourse, the few references to monumentality in Mamluk sources seem to have been less aesthetically construed and more politically, or at least ideologically, driven. They often bore a competitive edge: the authors mention buildings as comparable in massiveness to those of their patrons that had been built in the realm of the Ilkhanids, the Mamluks’ main Islamic rivals, or other less important Islamic powers such as the North African Marinids or the smaller Anatolian principalities.16 Praising the monumentality of their patrons’ buildings was at times coupled with downplaying the monumentality of those of other sovereigns. That writers served this propagandistic purpose may have been induced by the patrons themselves, especially during the early Mamluk period when the Mongol Ilkhanid threat was real and the propaganda war between the two sides fierce and multifaceted.17 The emphasis on monumentality may also have reflected a heightened historical awareness among the Mamluk authors, which was expressed in the comparisons encountered in the sources between the Mamluk buildings and famous monuments of both the mythical and historical past including the pre-Islamic period. This too is a pre-Mamluk phenomenon, but it found formal expression in the Mamluk period with the development of a more or less fixed list of venerated ancient monuments that constituted a monumental category in medieval Arabic literature and are often mentioned when the achievements of past nations, a favourite topic in adab, are discussed.18 Many descriptions of contemporary Mamluk monuments refer the reader to one or another of these structures, most frequently to the Iwan-i Kisra in al-Mada’in

–23 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

(Ctesiphon) in Iraq, the epitome of monumentality which is sometimes brazenly claimed to have been matched or surpassed by the building under discussion.19 This practice is more than a literary trope despite its poetic and literary origins and its frequent usage in the sources. The veracity of the comparison itself is much less important to both authors and readers than the historic connection and the contest across time implied. The monumental category is a way of reclaiming the golden age inscribed in the Mamluk collective memory at a time when the Mamluk state was showing signs of its ability to recoup some of the glories of that golden age. It had very swiftly defeated the Crusaders and Mongols, asserted its rule over all the Syro-Egyptian territories and devised a new caliphal legitimacy with the installation of an Abbasid caliph in Cairo after the annihilation of the Baghdadi caliphate by the Mongols in 1258. The culture reacted to these Mamluk victories with renewed hope of recapturing the glorious past and reviving the true caliphate after two centuries of uncertainty, a feeling which lasted well into the fifteenth century. This was reflected in the reorientation of Mamluk historical writing towards a pan-Islamic outlook reminiscent of the writing of eighth- and ninth-century historians who lived under an, at least nominally, unified Islamic world.20 Thus, an entire generation of Mamluk historians – including al-‘Umari and al-Nuwayri in Cairo and Ibn Kathir and al-Zahabi in Damascus – adopted a universal and upbeat approach and covered the entire Islamic world in their writing. A similar historic emphasis is expressed in visual references to the venerated monuments of the early Islamic period which dot the early Mamluk architecture built in the time of Baybars and Qalawun and his sons.21 Both references to the monuments of the past in the sources and to the past in early Mamluk architecture embody and reinforce the rekindled Mamluk sense of

historical continuity and represent a conscious effort to give it shape: one in space, the other in words. If Mamluk sources lacked a developed aesthetic or architectural language, they did have another specialised language at their disposal, and that was the legal language of the waqf (endowment) documents with which they seem to have been thoroughly familiar. The institution of the waqf, an old and venerable Islamic legal-fiscal system for organising charity, social services, and the management and inheritance of real estate and agricultural land, had by the Mamluk period developed a language and a procedure for documenting buildings that satisfied contractual and legal requirements and reflected both an interest in the purely functional and socioeconomic dimensions of architecture and a specific vision of the role of buildings in social and urban spaces. 22 The description of a building in a waqf usually begins with recording its surroundings, which delimit its urban boundaries. Then came the description of every individual space in the building as seen by a person walking through it. The description started at the entrance and then enumerated the various features of each space in the building sequentially. More attention was paid to circulation, especially the location of doors, and to the specific functions of parts of the spaces, than to their appearance. The verbal description of architecture was, nonetheless, highly developed. Every legalistically structured expression carried precise connotations that captured what was valuable in the structure being described. These codified expressions found their way into the descriptions of buildings in the sources. Like the waqfs, the historical texts placed very little emphasis on the status of buildings as aesthetic objects. They even often ignored it. They cared more about the building’s contextual effects, experiential qualities or functional capacity. A building, moreover, was never

– 24 –

PERCEP T ION OF A RCH I T ECT U R E I N M A M LU K SOU RCE S

seen as a separate, stand-alone object: it could only make sense as a component in an urban context or in the landscape. This view was probably a reflection of the prevalent forms of dense layout in the city and the scattered pavilion arrangements in the gardens. The only exceptions were citadels and isolated caravanserais or khanqahs in the countryside which usually elicited brief comments on their exterior walls and mass, mainly on their strength and solidity, as would be expected.23 Otherwise, hardly any description of an urban façade can be found in Mamluk sources or in waqf documents. Only the location of entrances and position of minarets were noted, emphasising the link between the public space of the street and the building proper. Even interior spaces were seen in the context of their connectivity and functionality and never in an abstracted way as arranged spaces or volumes. Their architectural characteristics were never noticed except to indicate how they were accessed and whether or not they had built-in usable spaces, such as recesses, niches and alcoves. Mamluk historical sources and waqf documents alike were most concerned with what can be termed the socioeconomic aspects of buildings. The writers spent much of their energy discussing patrons, cost, intended functions, capacity for services and the abundance or inadequacy of the waqfs attached to buildings for their upkeep and to support their designated users, and very little on anything else. The form and structure of the source descriptions resembled those found in the waqf documents themselves, not only because the two types stemmed from similar literary and legal traditions, but also because many of their authors were also legal experts and may have been personally involved in the redaction of waqf documents. The language of one form flowed into the other as authors themselves moved between the two. Many authors even incorporated parts of the waqf documents that

they had access to, thanks to their position in the administration or the judicial system, into their historical texts describing major buildings, probably because the waqf texts already contained in an authoritative style the information they wished to present. This practice in fact has preserved some of the waqf texts that otherwise would have disappeared from the record.24 Yet there is a small, though structurally and formally significant, difference between the two forms. The chronicles obviously did not have to carry the legal responsibility the waqf texts did. They thus were able to adopt a less rigid and formulaic structure and to allow literary tropes and storytelling techniques to permeate their waqf-inspired texts and imbue them with informative and entertaining anecdotal, historical and comparative details.25 The use of the ‘monumental category’ mentioned earlier is one of these techniques. So are the abundant poetic quotations, which may have been part literary bravado, part expressive tool. The same applies to historical references, mythically based comparisons, and reported conversations which seem often to have been totally fabricated.26 This historical reporting laced with adab techniques and tropes has to be seen within the larger framework of the profusion of ‘literarised’ history with a popular bent, observed by some contemporary students of Mamluk history and recently problematised by Ulrich Haarmann as a testing ground for the post-structuralist challenge to the conventional historiographical binary opposition of ‘narrativity versus facticity’.27 ‘Narrativity’, Haarmann observed, had always enlivened Arabic history writing with its close ties to adab, but its treatment by historians varied even during the same time period. Some, like Abu Hamid alQudsi, may have intentionally ornamented their accounts to make them more novelistic and enticing, whereas others, like alMaqrizi, preferred a more serious, solemn

–25 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

and learned outlook. The ‘literarised’ modes seem to have expanded in the fourteenth and fifteenth centuries when history was apparently a popular subject with a wide readership, judging from the large number of compilations and abridgments of earlier works and of new compositions produced during the Mamluk period.28

THE SOCIAL AND POLITICAL CONTEXT OF WRITING ON A RCHITECTURE The language of the sources dealing with buildings can be summed up as financially concerned, conservatively driven, legally and literary based, and visually inexperienced; qualities that distinguished the groups to which most of the Mamluk authors belonged. These groups comprised the ulama and the kuttab (the two greatly overlapped in the Mamluk period) and the awlad al-nas, the literarily inclined sons of Mamluk amirs and soldiers who should be intellectually, socially and ideologically classified with the ulama and kuttab despite their Mamluk lineage, Mamluk privileges and knowledge of the Turkish language.29 Members of these groups, who can loosely be termed the literati, also formed the reading public for the same sources, which reinforced the development of a closed discourse and facilitated the formation of an endogenous and insular school of history, in which every member was linked in more than one way to the others, and every member’s work was inevitably and immediately measured against the works of others, who practically covered the same terrain.30 The literati also shared the same general ethos. Their sense of themselves was grounded primarily in educational background, scholarly or chancery specialisation, or jurisprudential affiliation (madhhab). They depended on the Mamluk military elite for their livelihood; the Mamluks patronised and employed them to administer the religious, social and fiscal

systems of the sultanate because they were the most educated groups. They were, however, excluded from any political decisionmaking and kept under constant check. The Mamluk sources treat architecture, therefore, in a way that reflects not only the personal inclination of an author or the collective social and intellectual structures or even the expectations, tastes and preferences of the readership. Authors were powerfully beholden to the wishes and interests of their Mamluk patrons and their desire to have their work documented, celebrated and memorialised. There were certainly sultans, such as Baybars, al-Nasir Muhammad and Qaytbay, whose interest in building was pronounced to the point that it affected their rule and how their amirs and notables handled their wealth and expressed their positions in society.31 Each in his own way and for his own particular set of reasons and preferences endowed the cities of the realm with large numbers of religious, charitable, commercial, military and palatial monuments. Each also is reported to have been directly involved in the projects he commissioned, sometimes interfering in the planning stages, sometimes dictating the design and decoration of a specific building, and at other times even working on the construction. Commentators on their reigns did not fail to notice this prodigious production and personal involvement and to be impressed by both.32 Whole sections of the biographies they dedicated to these sultans read like building rolls, recording every project they sponsored in every city of the sultanate. This practice, routine and trivial as it may seem to us, was a historical novelty in the early Mamluk period. It was the first time that an effort had been made to list all the building projects and systematically register them in a separate section of a sultan’s biography so they could be inserted into an established set of subjects considered essential to the biography of a grandee, especially a sultan: his personal

–26 –

PERCEP T ION OF A RCH I T ECT U R E I N M A M LU K SOU RCE S

qualities and virtues, his military campaigns, his embassies, his main associates and functionaries, and his buildings and other projects. This arrangement seems first to have been introduced into the annals of the reign of Baybars I, an indefatigable builder and the first true organiser of the Mamluk state and system. The individual who can be credited with this biographical innovation is ‘Izz al-Din Muhammad b. ‘Ali, Ibn Shaddad (1217–1285), an Aleppine scholar and katib who began his career in his native city in the administration of its last Ayyubid ruler, al-Nasir Salah al-Din Yusuf. After the Mongol invasion of Syria in 1260–61, he fled to Cairo and as a distinguished refugee was soon serving in Baybars’ administration, in the entourage of the famous vizier Baha’ al-Din Ibn Hinna.33 Ibn Shaddad’s annals of Baybars’ reign, of which only the last third survive, were recently published. Though probably not officially commissioned, they appear nonetheless to have been approved by the sultan and perhaps even compiled from conversations with him. They were, however, completed after Baybars’ death, during the reign of his son Baraka Khan (1276–1279) as is clear from the last part. At the end of the annals, Ibn Shaddad affixes an extended and eulogistic biography of Baybars. In it, he provides an exhaustive list of the numerous structures Baybars built all over his sultanate, structure by structure and city by city beginning with Cairo and moving onto all the SyroPalestininan cities in which Baybars sponsored building projects.34 For the royal structures in or around the Citadel of Cairo – that is, where Ibn Shaddad lived and worked – he sometimes even goes a step further and provides measurements or supplies superlatives to convey the quality of particular structures. He also enumerates the architectural components of every palace and qa‘a Baybars built for himself, his son Baraka Khan and his favourite amirs.35

In itself, Ibn Shaddad’s list is unusual, but more remarkable is the attention he devotes to space organisation and architectural terminology, certainly rare among medieval historians (the only comparable historians are al-Maqrizi and Abu Hamid al-Qudsi, both of whom lived more than a century later).36 Two possible explanations can be advanced for this special treatment. Firstly, the list could simply have resulted from the importance Baybars placed on architecture; he might have ordered these detailed descriptions of his most important projects to be included in his inventory of achievements. However, this explanation is weakened by the fact that Ibn ‘Abd al-Zahir, Baybars’ official biographer, does not include anything comparable in his otherwise extensive encomium, al-Rawd al-Zahir fi Tarikh al-Malik al-Zahir. Alternatively, Ibn Shaddad’s list could have been inspired by his own expertise and interest in architecture, inducing him to dedicate a disproportionate amount of space to the reporting of building projects. His expertise is apparent from the precise and assured language, attention to detail and professional terminology displayed in describing his patron’s structures. Nor was this interest new: it is already discernible in his important compendium on the history and topography of Syria and Jazira, al-A‘laq al-Khatira fi-Zikr Umara’ al-Sham wa-l-Jazira, commissioned by Baybars and written during his reign, probably in recognition of Ibn Shaddad’s knowledge of the various principalities in those two regions and in preparation for their ultimate annexation to the Mamluk sultanate. Al-A‘laq, a pioneering work that anticipated al-Maqrizi’s Khitat in its orientation, structure and appreciation of architecture, is divided into sections on Aleppo and its environs, Damascus and its surrounding regions (including Lebanon and Palestine), and the Jaziran cities. It includes a systematic list of the major buildings – citadel, main mosque, madrasas, khanqahs and caravanserais – for each

–27–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

city and, in the case of the major cities such as Damascus and Aleppo, the history of each structure in detail as well.37 The main difference between al-A‘laq and Ibn Shaddad’s biography of Baybars is that the list of buildings in the latter is presented as the final category of Baybars’ achievements and qualities and is meant to complement and perhaps to illustrate or concretise them. It is an innovative modification to the usual structure of eulogistic biographies where the list of architectural projects undertaken by the subject, in addition to providing a record of the patron’s architectural accomplishments, is invested with propagandistic and political import. Ibn Shaddad’s biography inaugurated a new convention in Mamluk royal and princely biographies: it aimed at comprehensiveness and avoided the usual exaggerated and lyrical invocation of key monuments. His successors all began to record all the building projects of their subjects, not just the highlights, although no one reached the same degree of detail that Ibn Shaddad achieved. Later chroniclers, such as Ibn Shakir alKutubi (1282–1363) in his biographical dictionary Fawat al-Wafiyyat and Ibn TaghriBirdi (1410–1470), the fifteenth-century chronicler and son of a Mamluk amir, in his Nujum give shorter lists of Baybars’ structures with slight differences from Ibn Shaddad’s, but they both eliminate the description of the citadel’s palaces and qa‘as.38 The details of the buildings that Ibn Shaddad so relished adding to his essential list thus seem to have been a personal quirk, perhaps a sign of some architectural expertise that was not recorded in his biography. They did not reappear in any later account of building projects by Mamluk patrons.

THE ARCHITECTURALLY CONSCIOUS GENRES By the end of the fourteenth century, a significant development can be detected in all

the sources dealing with architecture, including the usual annals, biographies and encyclopedias: they begin to show more interest in the sociocultural, symbolic and expressive import of buildings. No comprehensive explanation of this shift has ever been offered, but several modern historians, notably Oleg Grabar and R. Stephen Humphreys, have tried to connect it to the sheer number of art objects that were being produced for both the upper and middle classes, including the wealthier kuttab and ulama, and the monuments that were crowding urban space and influencing how people viewed and experienced their cities or used their public areas.39 Both Humphreys and Grabar then used this observation to move in a direction that serves their own aims. Humphreys, in a thirty-year-old study that is still quoted today by Islamic architectural historians, used architecture in the city to propose an interpretation of the social dynamics that developed between the Mamluk military elite and their indigenous subjects. He ascribed to the Mamluks, especially in Cairo, a heightened awareness of the role buildings can play in enhancing the reputation of their patrons and in assuring their position in the public eye. He saw in the endless rows of monuments whose façades competed along the streets of Cairo ample proof of that understanding, which he called ‘the expressive intent’ of Mamluk architecture. He also detected a ‘tension’ between the ostentation and striving for visibility of these monuments and their ostensibly pious and charitable functions, and read it as signifying the merger of the political agenda of the Mamluk military elite and the religious expectations and needs of the Muslim population, at least as it was articulated by the literati whose writing constitutes our main source of information. Humphreys singled out other sociopolitical measures introduced by the Mamluks, such as the reorganisation of the court system under Baybars and the tight-

–28 –

PERCEP T ION OF A RCH I T ECT U R E I N M A M LU K SOU RCE S

ening of state control over the ulama class, as other manifestations of the same tension he saw in the architecture between the political and the social and religious forces in Mamluk society. Grabar’s purpose was very different. He was seeking to classify and understand Mamluk art and architecture and to highlight the sources useful for their study. In an earlier essay, he had noted a correlation between the level of artistic and architectural production all over the Islamic world in the fourteenth century, but especially in the Mamluk sultanate, and the appearance in historical treatises of interpretations that linked the degree of cultural sophistication to sponsorship of art and architecture and interest in city life.40 He identified Ibn Khaldun and his distinguished student al-Maqrizi as the two most prominent protagonists of this correlation, and hailed their two famous works, the Muqaddima of the former and the Khitat of the latter, as its main illustrations. This new awareness of the sociological significance of architecture makes its impact mostly in the language and orientation of the masalik wa-l mamalik and khitat books, two interrelated literary genres whose resurgence in the fourteenth century is tied in more than one way to the concurrent interest in architecture and urban development.41 Traditionally, however, neither masalik nor khitat was primarily concerned with the buildings themselves, their forms and functions, and their intended or perceived messages. Al-masalik wa-l mamalik was essentially a loosely defined adab type that was developed out of the combination of several scholarly, literary and administrative genres, including futuh (chronicles of the conquests), travel and ziyarat (pilgrimage) literature, chancery and kharaj (taxation) manuals, and surat al-ard (cartography).42 Its framework was geographic, bordering on the cosmographic, with a universalistic Islamic scope that rarely ventured outside the frontiers of the Islamic

world. Its heyday was the ninth and tenth centuries when a number of outstanding geographer travellers criss-crossed the Islamic world, compiling their depictions of one Islamic world, after its political unity held together by the Abbasid caliphate had passed. Buildings figured in it primarily as unusual and distinguishing features of a region or city. They would be noted in passing in a fashion akin to the way the natural and supernatural ‘aja’ib of a place, including unusual or ancient monuments, were often mentioned. The startling early victories of the Mamluks against the Crusaders and the Mongols in the thirteenth century reinvigorated the literati and renewed their trust in Islamic political and territorial unity. The masalik’s orientation moved towards the geopolitical, a shift exemplified by the seminal work of Ibn Fadl-Allah al-‘Umari (1301–1349), Masalik al-Absar fi Mamalik al-Amsar,43 compiled in the late 1330s when the author was serving as a high administrator at the court of al-Nasir Muhammad (including a stint as the sultan’s private secretary (katib al-sirr) between 1329 and 1332). In addition to geographical surveys of the countries of Islam and their immediate neighbours, al‘Umari provides topographic descriptions of important Islamic cities and holy sites and firsthand information on the ceremonies and duties of their rulers and lists of the ranks, functions, and protocols of their officials and caretakers. Buildings in his text are presented in their sociopolitical context as expressions of dynastic and royal pride and splendour and as positive architectural achievements functionally and spatially distinguishing their urban setting. In other words, they are seen as cultural artefacts. The khitat form is an almost exclusively Egyptian and significantly more localised genre than the masalik wa-l mamalik. Its cosmocentric focus is often linked to a deep-rooted affinity with Egypt as a

–29 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

homeland which persists in the writing of Egyptian historians from the early Islamic period onwards.44 These feelings intensified in medieval times, especially after the establishment of the Fatimid caliphate, which created in Egypt a new and vigorous authority independent of Baghdad.45 In the khitat books, they are thought to have found expression in careful and meticulous descriptions of Egypt’s topography, history and monuments; particularly Cairo as Egypt’s capital and major political, economic and cultural centre. Within this framework, buildings most often appear as urban landmarks examined in the context of their streets and neighbourhoods. Their patrons, costs and circumstances are also noted and their historical significance weighed. The khitat genre reached its apogee around the middle of the fifteenth century with Taqiyy al-Din Ahmad al-Maqrizi’s al Mawa‘iz wa-l-I‘tibar bi-Dhikr al-Khitat wa-lAthar. Composed between 1415 and 1442, this magisterial compendium offers the most elaborate and spirited testimony we have of Islamic Egypt’s urban history.46 In his introduction, al-Maqrizi describes his book as a ‘summary of the history of the monuments of Egypt from the earliest times, and of the surviving structures in Fustat, and the palaces, buildings, and quarters of al-Qahira with short biographies of their patrons and sponsors’.47 This is the most straightforward definition of a khitat book we have and a rather truthful and precise description of the scope of the book, which briefly covers Egyptian cities other than the two capitals, expands its range when it deals with Fustat, but reserves the most detailed treatment for Fatimid al-Qahira and its Ayyubid and Mamluk extensions. Al-Maqrizi also presents a concise statement of the reasons behind the writing of the book, the most prominent of which is his filial affection towards his country, his city and even his hara (neighbourhood), Harat al-Burjuwan

in the heart of Fatimid al-Qahira, which had prompted him since his youth to collect every bit of information on its history he came across. Misr (in this context probably meaning both the country and the city) was, according to him, the ‘place of my birth, playground of my mates, nexus of my society and clan, home to my family and public, the bosom where I acquired my wings, and the niche I seek and yearn for’.48 Al-Maqrizi’s method was influenced by the sociohistorical theories of his revered teacher, the great Ibn Khaldun, with whom he studied for a long time.49 The overarching cycle of the rise and fall of dynasties that formed the basis of Ibn Khaldun’s hermeneutical framework in explaining historical process seems also to have informed al-Maqrizi’s thinking and structuring of his Khitat, albeit indirectly.50 Al-Maqrizi seems to have subsumed the Khaldunian structure in his text as a way of classifying and understanding the vast amount of historical, topographic and architectural material he had collected over the years. He seems to have devised an analogous cycle of prosperity and urban expansion followed by decay and urban contraction to frame his exposition of the fate of Cairo under the successive dynasties that governed Egypt in the Islamic era: the Tulunids, Ikhshidids, Fatimids, Ayyubids, and Qalawunid and Circassian Mamluks. The political fortune of each of these dynasties or families is plotted against the fluctuations of the urban and architectural prosperity of Cairo in a way that echoes the Khaldunian cyclical view of human history.51 In this recursive scheme, architecture constituted the visual, palpable and measurable signifier of every stage in the historical cycle of the rise and fall of Cairo. Buildings, streets, the entire city and the whole country were analysed and meticulously described by al-Maqrizi, not only because they embodied the obviously idealised past but also because they narrated

–30 –

PERCEP T ION OF A RCH I T ECT U R E I N M A M LU K SOU RCE S

through their particular architectural and urban forms the history of Egypt under its various rulers. Al-Maqrizi’s work, under the combined impact of his passionate attachment to his city and the theoretical framework he absorbed from his teacher, is an idiosyncratically melancholy and culturally oriented architectural and urban history which introduces a new role for architecture as the agency of both personal memories and collective aspirations. Such a powerful evocation of the meaning of architecture would not again be articulated as purposefully as al-Maqrizi did until Victor Hugo wrote the celebrated chapter, ‘Ceci Tuera Cela’, for his medieval novel, Notre Dâme de Paris, published in 1832, to convey the role of architecture as the carrier of meaning for historical cultures.52 With al-Maqrizi’s Khitat, we reach the most elaborate exploration into history writing through the chronicling of buildings and topography that remains an exception in Mamluk historiography. Although the book was copied and abridged numerous times by later Mamluk and Ottoman historians, as evidenced by its more than 185 extant manuscripts, no later Mamluk historian seems to have managed to absorb the method adopted by al-Maqrizi from Ibn Khaldun or to capture the mood and intensity displayed in al-Maqrizi’s text. Mamluk historians continued to produce books on urban and architectural history, such as alSuyuti’s Husn al-Muhadara fi Tarikh Misr wa-lQahira, Abu Hamid al-Qudsi’s (not Ibn Zahira as the published book asserts)53 alFada’il al-Bahira fi Mahasin Misr wa-l-Qahira and Ibn Shahin al-Zahiri’s Zubdat Kashf al-Mamlik wa-Bayan al-Turuq wa-l-Masalik. However, although they all show an understanding of the sociocultural significance of architecture, they all revert to older methods or frameworks, such as that of fada’il, or the masalik format, or the classificatory listings of Ibn ‘Abd al-Zahir’s and Ibn Duqmaq’s khitat books, with no underlying historical or cultural interpretations.

This chapter has attempted to articulate and contextualise the perception of architecture as gleaned from Mamluk historical sources. For modern historians and architectural historians dealing with these sources, the findings presented here raise a number of methodological and historiographical questions.54 The formidable architectural production of the Mamluk period suggests that architecture played a substantial role in the display, articulation, assertion, transfer and symbolising of wealth, social status and perhaps other values as well. Most historians and architectural historians normally begin their analysis with this observation and ‘read’ the architecture itself – and more readily its inscriptions – for clues about its significance to its society. They then scout the sources to confirm or further their formulations, glossing over the elementary fact that these sources do not necessarily represent common attitudes towards architecture in the Mamluk society at large. This oversight has led to a variety of sometimes conflicting, sometimes impressionistic interpretations, many of which rest on thin historical conjecture, which has prompted some observers to question the validity of the entire exercise of searching for architectural meaning.55 Before using the sources for interpreting the meaning of architecture for Mamluk society, one must first understand their peculiarities and commonalities. For, aside from individual quirks, these sources essentially reflect the collective background, education and social manipulations of their authors and, to a lesser extent, their readers, both of whom were almost certainly restricted to members of the educated classes. What they really and clearly tell us is that, for this influential and vocal group in the Mamluk society, architecture was mainly thought of as a tool of political and personal propaganda and of legal and financial gain, as a source of complaint and employment, and perhaps of entertaining

– 31 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

anecdotes. But it was puzzling aesthetically and almost meaningless symbolically. The few exceptional observers – ‘Abd al-Latif al-Baghdadi, Ibn Shaddad, Ibn Fadl-Allah al-‘Umari, al-Maqrizi and Abu Hamid alQudsi – added primarily sociocultural and historical dimensions to the meaning of architecture, but their dealing with it remained essentially textual, literal and unarchitectural. Al-Maqrizi is obviously a special case. Although he too did not profer an ‘architectural’ understanding of architecture as we conceive of it today, his ingenious induction of the elements of the built environment as historical indices in the service of his overall interpretation of the history of Egypt put him in a class by himself. However, this methodological innovation is not why he is usually consulted by modern historians. The exceptional historians otherwise did not really break rank with their social support group, the literati, either intellectually or politically, and therefore cannot be seen as representing a fundamentally different take on the meaning of architecture as seen from their vantage point. This condition colours all modern explanations of Mamluk architecture, which perforce have had to go through the prism of the sources before reaching their conclusions. Thus, we know practically nothing about the views of the architects (or master builders), or the general popu-

lation for that matter, simply because their voices are never heard in the sources.56 Conversely, the patrons – either members of the ruling Mamluk class or to a lesser degree, wealthy merchants and ulama, appear to have played a major or defining role in the conception of architecture and its eventual signification and appreciation. They are not only said to have contributed to the design and decoration of the buildings they commissioned,57 but they are also presented as the ones whose tastes, attitudes and preferences habitually gave architecture its extra-artistic and extrafunctional significance. Therefore, one could argue that the widespread and accepted scholarly assumption of today that Mamluk architecture should be understood primarily through the roles, aspirations and circumstances of its patrons is predicated on the peculiar structures and limitations of the sources, as well as on the complex relationships that their authors, as individuals but primarily as social groups, had with the Mamluk elite. Elementary as it might seem, this conclusion helps us keep in mind that, although our views on the signification of Mamluk architecture are tilted towards a large role for the Mamluk ruling class, this bias is intrinsically sustained and probably exaggerated by none other than their sometimes satisfied, sometimes disgruntled interpreters, the Mamluk historians.58

– 32–

4. ARCHITECTS AND ARTISTS IN MAMLUK SOCIETY: THE PERSPECTIVE OF THE SOURCES

I

n the field of Islamic art and architectural history, the Mamluk period in Egypt and Syria (1250–1517) is perhaps the most thoroughly studied in all the Islamic middle ages. This scholarly attention should not be surprising to anyone familiar with the sheer number of Mamluk monuments still standing in Egyptian, Palestinian and Syrian cities, particularly Cairo, Damascus, Aleppo and Jerusalem, and the dazzling array of Mamluk artefacts, especially metalworks, pottery and textiles, held in numerous museums and private collections. Furthermore, our knowledge of this multitude of buildings and artefacts is enhanced by the existence of an unusually rich variety of historical sources, especially chronicles and biographical dictionaries but also encyclopedic manuals for the chancery, geographical treatises (masalik) and topographical tracts (khitat) that cover practically every one of the 250-plus years of Mamluk rule, often from different perspectives and by more than one observer. Consequently, we now have several surveys of Mamluk art and architecture, monographs on various cities, monuments, types and celebrated objects, and some studies that compare Mamluk art and architecture to other Islamic and non-Islamic traditions.

We even have a number of theoretical discussions of some of the formal, symbolic and sociocultural attributes of this art and architecture, something that is generally lacking for other pre-modern Islamic artistic and architectural traditions. Yet even this impressive body of knowledge suffers from fundamental lacunae in both basic and contextual data that render any conceptual, interpretive or definitive statement on the subject of Mamluk art and architecture highly speculative. One of the most challenging of these gaps – especially to any contemporary student accustomed to viewing artistic output in the light of the artist’s life, background and ideas – is the fact that we know very little about the builders and artists who conceived of and created Mamluk art and architecture. Not only have they left us no written testimonies about themselves or their work – a practice that was rather alien to the medieval mentality generally – but they very rarely, and then only cursorily, appear in the copious Mamluk biographical dictionaries, which otherwise accorded countless people from various walks of life their place in history. This neglect of artists and building craftsmen is all the more perplexing since the Mamluk period witnessed a formidable outburst of artistic and

– 33 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

architectural production and since it is clear from all the historical clues we possess that art and architecture played a crucial role in the financial, cultural and social lives of Mamluk cities and communities, and in the exhibition, articulation, assertion, transfer and symbolisation of wealth and social status. Meagre as it is, however, the information in the Mamluk sources can still shed some light on several issues related to Mamluk building and art professionals. Among the most important of these issues is where artists and architects in general and specific classes among them in particular stood in the Mamluk social hierarchy – at least in the view of the chroniclers and biographers. Were they regarded as members of a separate group, or as an integral social group, or even as part of a larger group such as artisans (arbab al-hiraf wa-al sana’i‘)? Were they seen together as craftsmen, and therefore of rather modest standing? Or were they seen as educated professionals, more or less equal to others, including ulama (religious scholars), literati or kuttabs (scribes or administrators)? Was their social position predetermined, static and fixed, or were they mobile social agents able to improve their lot? If the latter was the case, what were the possible routes for them as individuals or groups to rise in the social order? And what do these routes tell us both about the Mamluk society and about its views on men in building and the arts? Some of these questions have been tackled before, particularly in regard to architecture. Leo Mayer, after carefully ferreting out the primary sources available to him in the 1940s, came to the frustrated conclusion that his work could only slightly lift one corner of the anonymity that covers the history of Islamic architecture, with the partial exception of Ottoman Turkey. He also conceded that building craftsmen seem to have occupied a rather lowly position in the hierarchy in all Islamic societies.1 Doris Behrens-Abouseif, who had many more

sources at her disposal and who concentrated on Mamluk texts in particular, was able to sharpen Mayer’s conclusions. She distinguished between building craftsmen, who, with very few exceptions, remained anonymous, and building supervisors and construction managers who are well documented in the sources. Among the latter, she noted that the majority in the first half of the Mamluk period, commonly called the ‘Bahri’ period (1250–1382), were members of the ruling Mamluk or the powerful administrative classes. In the second, or ‘Burji’, period (1382–1517), a new group, building managers with an artisanal background, seem to have risen to share with the amirs and administrators the decisionmaking authority in building projects. They and their patrons appear to have been responsible for the ‘rough’ design of the buildings they supervised; the design of the architectural, structural and decorative details and their actual execution were left to the craftsmen.2 In revisiting the question of the social standing of Mamluk building craftsmen and artists, I will address a number of issues. Firstly, I will evaluate the language the medieval sources use to describe them. Secondly, I will tease from those sources what they can tell us about the artists’ presence and influence in society and the attitude of the authors of the sources – all members of either the ruling or the intellectual and religious classes – towards them. Thirdly, I will analyse a number of biographies of individual artists to uncover the strategies they adopted to break away from their limited station in life and rise in the social hierarchy, at least high enough to earn a place in the biographical dictionaries. Given the nature and limitations of the sources and the tiny number of biographies I was able to collect, however, no firm conclusions can be provided. Instead, I will only present a number of observations that will further qualify the conclusions reached by Mayer and Behrens-Abouseif and stimulate

–34 –

A RCH I T ECT S A N D A RT I ST S I N M A M LU K SOCI ET Y

further investigation into this important yet still little-known and little-appreciated subject.

THE QUESTION OF TERMINOLOGY When reading references to either artisans and artists or to building craftsmen one is immediately struck by the ambivalent terminology the sources use to designate them. In the building crafts, the most frequently encountered terms are mu‘alim, mi‘mar, muhandis and shadd. In the arts, most terms, on the face of it, seem to designate specific vocations or skills such as muzawwiq, naqqash, musawwir, muzammik, while some appear to be generic such as mu‘alim and sani‘. Scholars who have studied these terms have been unable to determine whether they were consistently used to indicate qualified artisans with particular skills or were vaguely and randomly applied. Furthermore, the apparent interchangeability of some of these terms betrays a certain flexibility, or even confusion, about their significance, at least among the authors of the sources if not in general. S. D. Goitein, the eminent student of the medieval Jewish community in Egypt, for example, could not decide on the difference between a dahhan, a muzawwiq and a muzayyin as they appear in the Geniza documents from the ninth to the early twelfth century, that is, just before the Mamluk period.3 Dahhan today designates a house painter, but it may then have indicated a specialised kind of painter, perhaps one who used an oil-based paint. A muzawwiq was also a painter, most probably one who used zawwuq, an amalgam of gold and mercurybased paint in a complicated painting, or rather gilding, process. Another possible meaning of muzawwiq is mosaicist, for Goitein pointed out a Fatimid inscription in the al-Aqsa Mosque, dated 1035, in which the word appears to mean exactly that, perhaps in reference to the dominant colour of

the tesserae, that is, gold.4 Muzayyin is a decorator in today’s usage, but may have meant someone who painted figures and patterns in medieval times. Similarly, of the eleven individuals from the Mamluk period in a list of people involved in the painting and drawing crafts throughout medieval Arabic history compiled by Ahmad Taymur Pasha, three are identified primarily as rassam (draftsman in today’s usage), three as naqqash (carver), two as dahhan, two as mudhahhib (painter in gold) and one as muzammik (illuminator or encruster or simply colourist).5 However, it is not clear what the difference was between these specialities in the overall painting profession. Some are even interchangeably identified as both rassam and naqqash, or dahhan and rassam. Likewise, tadhhib (gilding) and tazmik are usually listed as independent professions associated with calligraphy, but they seem to overlap too much to warrant drawing a distinction between them. They both appear to involve the laying on of colours, including gilt, and they sometimes seem to indicate sequential rather than separate tasks.6 The same ambiguity turns up in the use of the term mu‘allim, a relatively new term in the fourteenth century, which is casually translated as ‘master’. J. W. Allan has recently argued that the term is far from clear in its connotation, especially as it appears in relation to craft industries such as metalwork, weaving or even building. He, furthermore, contends that craftsmen in the medieval period were not confined to one craft; they usually mastered the entire range of skills related to a basic art such as metalworking. Allan has demonstrated, for instance, that a skilled artist in brass inlaying such as the mu‘allim Muhammad ibn alZain, who was active in late thirteenth/early fourteenth century and is famous today for his basin known as the Baptistère de Saint Louis (Louvre, LP 16) and his bowl (Louvre, MAO 331), was also an accomplished blacksmith who made window grilles and other

– 35 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

objects of iron; a very different craft from the delicate and expensive inlaying of brass with silver.7 In the building crafts, there seems to have been no single word to encompass the meaning of ‘designer’ or ‘architect’ as we understand them today. The term mi‘mar, used today in most Islamic languages to mean architect, appears in the Mamluk sources only in the sense of mason. Muhandis (more correctly muhandiz from the Persian etymology hundaz meaning ‘measurement’) seems to be the closest to our ‘architect’.8 It is the only term that indicates a professional craftsman with a wide range of technical aptitudes and theoretical knowledge that we associate today with designer/engineer. Basically, a muhandis was a surveyor with a primary training in geometry and perhaps hydrography which he may have acquired through a combination of apprenticeship and formal education. In Fatimid and Mamluk Egypt (and elsewhere in the Islamic world), a muhandis was primarily responsible for the building of bridges, canals, aqueducts and the like, somewhat like the French ponts et chaussées engineer. His architect-like role derived from his engineering background and function. In an urban context, his expertise was called upon to check boundaries between properties, to estimate values of real state, to assess the structural integrity of buildings, and in very few instances to ‘design’.9 The problem in architecture, however, is less one of defining each profession clearly than of deciding between the designoriented functions and the more administrative or managerial or financial roles that are not really art-related. The main reason for the indeterminate nature of these crafts is that royal building projects – about which we know the most – were usually entrusted to and supervised by non-designers, sometimes even military officers who carried the title of shadd (superintendent [of the project]).10 This practice goes back at least to the early years of the reign of Salah al-

Din, when in 1171 he appointed his lieutenant, the eunuch amir Qaraqush to oversee the building of Cairo’s fortifications, and later its citadel, which he partially completed in 1176.11 The practice was, like many other aspects of construction, formally regulated by the Mamluk sultan al-Nasir Muhammad who displayed a keen interest in controlling and directing the prodigious urban expansion of Cairo he initiated during his third reign (1310–1340). Chroniclers of the period attribute to him the establishment of a special department for building (diwan al-‘ama’ir) to coordinate the multitude of land reclamation projects and social and religious/charitable structures in which he and his amirs were engaged.12 It was then that the shadd al-‘ama’ir (superintendent of buildings) became a regular title of an important position held by a first-rank amir at the Mamluk court, although it lapsed back to a secondary position soon after al-Nasir Muhammad’s death. The shadd al-‘ama’ir was the equivalent of a director of public works; each royal project had its special shadd, who was usually a second-rank amir who reported directly to the autarchic sultan rather than to a nominal superior.13 The shadd in the Bahri Mamluk period not only supervised the project entrusted to him, he also controlled the budget and coordinated the work of the different craftsmen involved in it. This, and probably the fact that he was above all a Mamluk amir, qualified him to be almost always identified in the sources as the main figure associated with the project to the detriment of other individuals, the muhandis in particular, who might have had a more direct role in the creative process of building. It is not clear from the sources, however, whether the shadd played any part in that same process, in addition to conveying the desire of the sultan to the builders. In some instances though, the shadd, especially if he was experienced in construction projects, may have assumed a more decisive function in the

–36 –

A RCH I T ECT S A N D A RT I ST S I N M A M LU K SOCI ET Y

design process. This was apparently the case of Amir ‘Alam al-Din al-Shuja‘i who supervised the construction of the huge civic complex of Sultan Qalawun in Cairo (known as al-Bimaristan al-Mansuri and comprising a hospital, a mosque/madrasa and a mausoleum) after having successfully built the Mausoleum of Qalawun’s wife, Fatima Khawand Umm al-Salih (1284). He went on to complete many important structures in Damascus for Qalawun’s son, Sultan alAshraf Khalil, when he served there as governor (1291–1293).14 The title shadd rarely appears in the Burji or Circassian period. The position of shadd seems to have been replaced by that of mu‘alim, or the more grandiose mu‘alim almu‘alimin, or chief of all mu‘alims. This shift in nomenclature underlines a very important change in terms of the civilian penetration of some of the lower ranks of the military hierarchy in the late Mamluk period, but one that does not fundamentally modify the relationship of the muhandis or any other possible designer to the contractor/ supervisor of royal architectural projects. Doris Behrens-Abouseif rightly stresses the civilian background of many mu‘alims who appear to have risen in the class hierarchy of Mamluk society primarily because of their close contact with the sultan and great amirs.15 Most striking is the case of the Ibn al-Tuluni family, of whose members at least six intermittently occupied the position of royal mu‘alim al-mu‘alimin from the beginning of Barquq’s reign (1377) to the end of alGhuri’s (1517). The last one, Ahmad ibn alTuluni, was in fact among those notables expatriated to Istanbul by the Ottoman sultan Selim I in 1517 after he conquered Cairo and eliminated the Mamluk sultanate. The family emerged initially from the professional classes. Shams al-Din Muhammad Ibn al-Tuluni in the late fourteenth century was indeed a muhandis in addition to having occupied the taqdima (headship) of stonecarvers and builders in Cairo. So was his son, Shihab al-Din

Ahmad, whose biography found its way into the literature of the time essentially because, through his dealing with Sultan Barquq as his mu‘alim al-mu‘alimin, he managed to marry his daughter to the sultan, not only because he was responsible for his madrasa in the main thoroughfare in Cairo, among other buildings. His son, Shihab alDin Ahmad (the younger), owes his place in the biographical dictionaries solely to his high standing in Barquq’s court. He is identified as an amir and is distinguished for having dressed à la Turque, that is like a member of the military ruling class. The sultan divorced the muhandis’s sister and married his daughter instead, and on the occasion gave him an amirate among the khassakiyya, the most privileged corps in the Mamluk army, which constituted the sultan’s immediate entourage.16 Only individuals like the Tulunis have their names preserved in the sources in connection with building and architecture. Designers, builders and craftsmen in all their variety who might have been involved in the actual design and execution of the buildings and their decoration are very rarely identified or commemorated. If they are mentioned at all, it is almost always indirectly and in passing. Abjij, an otherwise unknown muhandis, for example, appears only once in al-Maqrizi’s Khitat. He was sent, we are told, by Sultan al-Salih Isma‘il in 1345 to Hama with Amir Aqbugha, who was the royal shadd, and a group of craftsmen to inspect a palace built by its ruler al-Malik al-Mu’ayyad so that he could build one like it at the Citadel in Cairo, a responsibility that would seem to warrant more than just passing mention.17 Other, more revealing cases are those of the ulama who are introduced in the biographical dictionaries with the epithet ibn al-muhandis (the son of the muhandis), presumably because one of their ancestors was a well-known muhandis. The irony is that the biographers in most cases did not preserve the name of the muhandis in question.18 Only

– 37–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

one of these muhandisin who had sons among the ulama is known to us: Ibrahim ibn Ghana’im. He was the designer of al-Zahir Baybars’s famous Qasr al-Ablaq in Damascus, constructed in 1264, and his mausoleum constructed between 1277 and 1281 where Ibrahim ibn Ghana’im’s name is still inscribed in a miniature niche on the left side of the portal’s muqarnas conch (fig. 1).19 Yet, despite his apparent connection to Sultan Baybars, Ibrahim ibn Ghana’im himself never made it to the prosopographies of his time. Only his two sons, Ahmad and Muhammad, and his grandson ‘Abdallah, have their biographies included, but clearly because they distinguished themselves as ulama and not because they were the sons of a great royal muhandis.20 A few other builders managed to put their signatures in some corner of their buildings, but they too remain otherwise unknown to us, for, beside a name and a FIG. 1. Signature of Ibn Ghana’im on the muqarnas of the portal of the Mausoleum of Sultan al-Zahir Baybars in Damascus

possible place of origin revealed in a toponymical nisba (surname), we have no further information about them. A well-known case is Muhammad ibn Ahmad Zaghlish al-Shami (the Syrian or the Damascene) whose name is inscribed above the portal of the Palace of Qawsun al-Nasiri (1337) in Cairo.21 Aside from his nisba and the possibility that his father, whose name is Ahmad Zaghlish, may have been a Turk, we know nothing about this possible designer of the most spectacular of Mamluk palaces (fig. 2). The exasperation felt by an earlier generation of scholars is best exemplified by S. D. Goitein when he poignantly remarked that ‘although one of the glories of Islam was its architecture, “an Islamic roll of architects” had to be created by a scholar of the twentieth century (L. A. Mayer, Islamic Architects and Their Works)’.22

SOCIAL MOBILITY AND BIOGRAPHICAL DICTIONARIES No Mamluk artist or architect (or any other medieval craftsman for that matter) seems to have left us any writing about his profession, his individual work, his general concepts of art and architecture, or craft schools he might have belonged to, in a manner comparable to the familiar way in which other cultural agents did, especially the ulama and literati in their assorted array of professions. Nor were artists of any category granted a voice of their own either collectively or individually in a separate volume or a special section in any one in the multitude of chronicles and biographical dictionaries covering that period (with the unverifiable exception of Tabaqat alMuzawwiqin [the Classes of Painters] or Daw’ al-Nibras wa Anas al-Jullas fi Akhbar alMuzawwiqin min al-Nas, cited once by alMaqrizi).23 They even rarely appear in general biographical dictionaries. The two exceptions are the calligraphers, who probably show up because in a certain sense they

–38 –

A RCH I T ECT S A N D A RT I ST S I N M A M LU K SOCI ET Y

FIG. 2. Signature of Ibn Ahmad Zaghlish on the side of the portal of the Palace of Amir Qawsun in Cairo

were considered members of the intellectual class since they dealt with texts and writing, and the mu‘alims al-mu‘alimin who were highly placed administrators and not real architects or builders. There are, however, a few artists, in the broad sense of the word, who were neither calligraphers nor building supervisors but who nonetheless made it onto the pages of the biographical dictionaries. So few are they, in fact, that they ought to be studied individually in order to find out what, in each and every instance, warranted their inclusion in the texts when they are so obviously exceptions to the practices of the time. Can we speak of a predictable trajectory for them? Were they viewed differently from other artists because of their skills and connections, a combination of both, or mere coincidence? Was the pattern of inclusion consistent from one case to another? Here we will examine

the biographical notes on four individuals. Three of them come from the well-known yet little-consulted compendium of Salah al-Din Khalil ibn Aybak al-Safadi, al-Wafi bi-lWafiyyat, the most comprehensive biographical lexicon of the early Mamluk period.24 The last one is lifted from Ibn Abi Usaybi‘a’s ‘Uyun al-Anba fi Tabaqat al-Atibba, the largest medieval collection of bibliographical material on physicians. The first biography is that of a polymath of the late Ayyubid period, al-ra’is ‘Alam al-Din Qaysar al-Katib al-Hanafi known by the nickname Ta‘asif (?) (1179–1251).25 He was a katib (serving in the diwan in Cairo), a faqih (jurist, though not a refined one), a geometrician, mathematician, astronomer and a muhandis, which seems here to mean a civil and military engineer. He is credited with building a mill on the Orontes for the Ayyubid king of

– 39 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

Hama, al-Muzzafar Mahmud (d. 1244), and an unspecified number of towers around the city, for which he invented a number of engineering devices, or designs (hiyal handasiyya). He is also said to have built a wooden globe for the same ruler on which he marked all the stars in the celestial sphere (preserved at the Museo Nazionale in Naples). His chief claim to fame, however, is that he was the one chosen by Sultan alKamil to solve the geometric riddles sent to him as a challenge by the emperor Frederick II, a task that he appeared to have succeeded in accomplishing.26 The second biography is that of the amir ‘Ala’ al-Din Aydaghdi al-Rukni (d. 1294) who appears to have caught the attention of biographers not only because he was an amir of some standing and a patron and a builder of numerous structures in Jerusalem, al-Khalil (Hebron) and Medina, but because he was also blind.27 Aydaghdi is said to have been a divine sign (’aya), because, despite his blindness, he was an accomplished muhandis. Using his cane as a yardstick, he even discovered a mistake in the measurement of a hall in his ribat (monastery) in Jerusalem that went unnoticed by his clear-sighted muhandisin and their assistants. He was also an avid horse breeder, and was able to recognise his horses by their trot or their smell. Aydaghdi represents a special case in the bibliographical dictionaries. He is a clear example of the occasional reversal of the normality/marginality division that seems to have defined the Mamluk society’s attitude towards the blind, or any other group of handicapped people for that matter.28 As a literary technique, the reversal aimed at challenging the dominance of that social division by pointing to those cases where the marginal turns out to be superior or extraordinary. The third biography is that of ‘Izz alDin Jawad ibn Amir al-Gharb (in Lebanon) (1305–1355), who was much more than a highly placed amateur metalworker, as proposed by Mayer.29 He was another prodi-

gious jack of all trades like Ta‘asif, though one apparently more inclined to crafts and hands-on experience and less scientifically minded. His biography reads like a lexicon of contemporary arts and crafts. He distinguished himself in a whole array of crafts, some artistic such as calligraphy, lacquering and inlaying, others militaristic such as arrow and armour making, veterinary practice and steel making, and some plain domestic crafts such as tailoring, embroidery and locksmithing. He seems to have been in demand at the court of Tankiz, the powerful governor of Damascus during most of al-Nasir Muhammad’s reign (1312–1340), for the wonderous objects he made, such as writing several verses of the Qur’an on a grain of rice (one is still preserved in the Military Museum in Damascus). We are told in fact that Tankiz wanted to make him a zardakash (supervisor of the armoury) and gave him an iqta‘ (land tenure) in the halaqa (a by-then minor Mamluk army division). Our biographers also do not fail to note, as if they needed to seal the biography with their stamp of approval, that Amir ‘Izz alDin has also memorised the Qur’an and learned some fiqh (jurisprudence) and Arabic grammar, in addition to his prowess in the martial arts.30 The fourth biography is that of Muhammad ibn ‘Abd al-Karim al-Harithy known as al-muhandis (d. 1204).31 He was originally a carpenter and a stone carver who distinguished himself with his delicate woodwork, especially on the doors he reportedly made for the bimaristan (hospital) of al‘Adil Nur al-Din ibn Zengi (built in 1154, but the doors may have been added later) (fig. 3). After he had reached the highest possible position in his craft, he aspired to improve his ability – and probably his social standing as well – by studying Euclid’s Elements of Geometry, which had been translated into Arabic as early as the reign of the  FIG. 3 Detail of the door of the Bimaristan (hospital) al-Nuri, Damascus

–40–

A RCH I T ECT S A N D A RT I ST S I N M A M LU K SOCI ET Y

– 41 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

famous Abbasid caliph Harun al-Rashid (786–809).32 This led him to concentrate on geometry and to become a geometrician first, then a medical scholar and a clock maker, and even later on an adib (litterateur) who composed poetry. With the exception of Aydaghdi alRukni who represents a special case, the three others present some commonalities that might explain why they, as opposed to other artists and builders, were deemed worthy of being included in the dictionaries. The first is that none of the three individuals were limited to a single speciality; they all boasted a combination of outstanding skills in a variety of artistic and architectural crafts. They also seem to have either combined theoretical and practical knowledge, as in the case of Ta‘asif, or acquired the theoretical knowledge to supplement their practical expertise, indicating perhaps a bias towards theoretical studies in medieval society. This impression is corroborated by the fact that they all seem to have dabbled in some literary or religious fields, such as writing poetry, studying fiqh and the like, which supplanted their professional architectural or artisanal identity and brought them closer to the mentality and occupation of the authors of the dictionaries. Finally, they all took the well-trodden path to renown and fortune, and possibly to immortalisation through the biographical dictionaries, by attaching themselves to a powerful amir or sultan and climbing the social ladder in his wake. Al-Harithy made his name by working for Nur al-Din. ‘Alam al-Din Qaysar had both Sultan al-Kamil and alMuzzafar Mahmud of Hama as patrons. ‘Izz al-Din Jawad worked almost exclusively for Amir Tankiz and his entourage. But, as interesting as these exceptional cases might be, their number is unfortunately too small to allow us to suggest specific patterns of selection and inclusion in the dictionaries for artists in general. The evidence at hand, however, indicates that being only a good muhandis or a talented

muzawwiq or naqqash, was not enough to merit social recognition. Artists and architects had to transform themselves intellectually and socially to move beyond the confines of small-time artisanal limitations. They had, in fact, to become something else in addition to being artists and builders – preferably men of the pen – before they could be noticed by the biographers, themselves obviously scholars and literatteurs who were predisposed to favour their intellectual kin. Otherwise, the overwhelming majority of artists and builders remained unacknowledged, even during their lifetime, given the fact that their names are very rarely noted when the completion of the monuments they built or the artefacts they made is reported. So we come back to the question of the social standing of Mamluk artists in general. The majority who remained anonymous clearly had a rather modest status not so very different from other craftsmen. This, however, does not constitute a particularly Mamluk cultural trait. Hardly any medieval culture elevated its builders and art makers to the level of other thinkers and professionals, or warriors and administrators for that matter. In fact, our historical view is distorted by the unprecedented phenomenon of Renaissance Italy when architects and artists became the model humanists and played a large role in defining the culture of their time. What they achieved was, in fact, to radically redefine their professions. An artist no longer had to step outside the boundaries of his profession to garner social recognition; the professions themselves were transformed into highly celebrated and intellectually intense pursuits. Art and architecture acquired conceptual frameworks and the requisite body of theoretical and historical knowledge. In the process, professional artists and architects became introspective and began to reflect on their professions. They also rose in the social esteem of their contemporaries and had biographers and critics to eval-

– 42–

A RCH I T ECT S A N D A RT I ST S I N M A M LU K SOCI ET Y

uate their work. From then on, the expectations of historians led them to search for similar roles for the artists and architects of earlier periods, when it is evident that they could not have played these

roles both because of the hierarchised sociocultural and intellectual contexts in which they functioned and because of their own and their society’s views of themselves and their professions.

– 43 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

–44–

5. THE MOSAICS OF THE QUBBA AL-ZAHIRIYYA IN DAMASCUS: A CLASSICAL SYRIAN MEDIUM ACQUIRES A MAMLUK SIGNATURE

G

lass mosaic was a luxurious medium of decoration around the Mediterranean in regions that either belonged to or were influenced by Byzantine artistic traditions. It also played a major role in the ornamentation of Umayyad architecture of the seventh and eighth centuries. It was profusely applied on the walls of commemorative monuments in all major urban centres and many palatial retreats in the countryside. After the fall of the Umayyads in 750, mosaics seem to have been slowly abandoned in favour of other decorative techniques, save for a few consciously historicising examples, such as those of the Cordoba Mosque in Umayyad Spain (c. 961), and the routine repairs of the remaining Umayyad monuments in Syria and Palestine. Then, sometime during the thirteenth century, the medium made a forceful reappearance in Mamluk architecture before it totally disappeared for unknown reasons by the middle of the fourteenth century. During that short period, at least seventeen buildings that had been adorned with mosaic have been recorded: in Cairo (8), Damascus (3), Jerusalem (3), Hebron (1) and Tripoli (2). The Umayyad mosaics of the Dome of the Rock in Jerusalem and the Great

Mosque of Damascus have generated numerous interpretations, and, something rare in the study of Islamic art, provoked scholarly debate about their artistic origin, the identity of their makers and their intended meanings. However, unlike their Umayyad predecessors, Mamluk mosaics have never been analysed nor interpreted, perhaps because most of them consist of simple vegetal and floriated motifs set in conchas and spandrels of arched windows.1 One exception, which is probably the earliest, is the Mausoleum of al-Zahir Baybars (the Qubba al-Zahiriyya) in Damascus. Its mosaic panels are distinguished both by the relatively large surface area they occupy and by their themes, which include architectural representations that are similar, but not identical, to those in the nearby Umayyad Mosque. These architectural representations were seen as either an aberration or a single attempt to adopt the old Umayyad scenes outside of their own context, which was deemed a failure by the patrons. However, in 1985, a Mamluk qa‘a was discovered at the Citadel of Cairo with fragments of mosaic displaying architectural representations. This new find clearly demonstrates that the Qubba al-Zahiriyya’s scenes were not an isolated experiment and

– 47 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

provides new clues to the uses and range of themes of mosaic decoration under the Mamluks.2 In this chapter I will analyse the architectural scenes in the Qubba al-Zahiriyya in Damascus and the qa‘a at the Citadel of Cairo. My argument is that these scenes are not mere copies or alterations from admired models in the Umayyad mosaics or from some other source, but rather variations on available representational themes that were used to convey ideological messages which stem from the context of the early Mamluk period.

THE MOSAICS OF THE QUBBA AL-ZAHIRIYYA Al-Zahir Baybars died in Damascus in 1276, and was buried there in a mausoleum (qubba) hastily built in the outer room of a house bought by order of his son and successor Baraka Khan in 1277, even though Baybars had prepared a mausoleum for himself in Cairo.3 The qubba was apparently not completed until 1281, after the death of Baraka Khan and his entombment with his father by order of Qalawun, the new sultan who usurped the throne from the descendants of Baybars.4 The qubba is a square room covered with a dome on a high octagonal drum supported on four simple squinches. Its four walls are profusely decorated to a height of about five metres (fifteen feet) with a marble and marble-mosaic dado, followed by a narrow band of intricately carved stucco scroll, intersected in certain places by a narrower band of elaborate marble mosaic, and topped by a frieze of glass mosaic – recently renovated – that runs around the four walls, and in the tympanums and soffits of the windows. Three of the frieze’s four sides depict architectural representations. The fourth side, the four corners, and the tympanums and soffits of the six windows consist of natural and stylised vegetal motifs coming out of vases, trees and cornucopia,

and leafy scrolls reminiscent of the more sophisticated Umayyad examples (fig. 4).5 The three architectural ensembles depicted in the Qubba al-Zahiriyya are symmetrically arranged around the centres of their walls. The most elaborate scene is set on the south wall, or the qibla wall, facing the entrance to the dome (fig. 5). Its central field is occupied by two superimposed compositions. The lower one consists of a four-column arcade followed by five freestanding columns crowned with natural leaves instead of the usual Corinthian capitols.6 The upper group consists of two identical long structures flanking a central tower with a spire. The two buildings seem to be depicted in perspective as their side walls run at an angle with the middle tower. Their fronts are made of colonnades surmounted by galleries, which are in turn roofed with conical gables. Both buildings are topped by what looks like central domes over high drums with pointed profiles rising behind the gabled roofs. The space between the two buildings and the central tower is occupied by wavy lines of white and dark blue tesserae, possibly suggesting water, and probably borrowed from the famous ‘Barada’ scene of the Umayyad Mosque. The scene is flanked by naturalistic representation and surrounded by a guilloche frame. The western wall has two groups of structures which are similar in composition to the ones in the south wall but are typologically different from them (fig. 6). The frontal three-column arcade is stocky and short. Its light brown arches are covered with curving vegetal motifs, as if they were trellises carrying plants. Their spans are closed with curtains tied near their lower end and wavering in the wind, in a manner reminiscent of the curtains in the representation of the Palace of Theodoric in St Apollinare Nuovo at Ravenna, or even closer to the few representations of curtains at the Umayyad Mosque, of which one is attributed to the renovations ordered by Baybars himself.7 The rear group is symmet-

–48–

T H E MO S A IC S OF T H E QU BBA A L -Z A H I R I Y YA I N DA M A SCUS

rically organised around a central spired tower. The two side structures, represented at an angle, are long and gabled, each with two round finials at the two extremities of their gables. The two side corners of the composition are occupied by two domed towers, frontally represented and divided as if to mark two storied structures. The architectural scene is flanked by two huge bowls with simplistic arrangements reminiscent of the Dome of the Rock’s vases, but rougher in execution.8 The two ends of the field above the curves of the windows’ arches are taken up with two more architectural groups which are mirror images of one another. Each is made of two long gabled structures and a domed tower behind the front building.

The third architectural scene appears on the northern wall above the entrance (fig. 7). Its centre is taken up by a huge tree whose branches and foliage are executed in the same method of gradual shading found in the Umayyad representations of trees. Two identical groups of buildings, both oriented towards the east so as not to give the impression of being reflections of one another, flank the tree. Both groups are displayed atop what appears to be a crude rendition of a green hill, lush with plants and trees. The architectural compositions use the same formal basic elements as in the other two groups. A huge domed tower occupies the centre, surrounded by two long gabled buildings, behind each of which stands another tower similar to the central

FIG. 4. Non-architectural mosaic of the eastern corner of the Qubba al-Zahiriyya

– 49 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E FIG. 5. Mosaic of the qibli (southern) wall of the Qubba al-Zahiriyya

one but smaller. A strange, flat and slanted element appears midway along the height of the central tower. If this does not represent some unknown convention of roof, it may be depicting the slanted roof of a wind-catcher (badhahanj), which was a familiar element in the Mamluk city’s skyline (fig. 8). Two remarks may be drawn from this brief description. Firstly, clearly the Mamluk mosaicists have not preserved the technical tradition that was initially passed on from Byzantine Syrian masters to early Islamic ones when the Umayyad mosaics were executed. Their trees and vegetal motifs rely on thick outlines and large-sized elements to fill their fields. Their columns are distorted

and confused reproductions of ancient types which might have been obscure to them, but still carried the allure of antiquity and the weight of convention. Secondly, the Mamluk mosaicists apparently derived some of their architectural compositions from real examples, probably to lend their scenes an air of authenticity and contemporaneity, but also because those were the only ones they understood enough to be able to render them in mosaic. First among these elements are the domes and domed towers, which appear here in prominent places as opposed to their secondary role in the Umayyad panels. They might refer to minarets or to fortified towers and qubbas, or domed pala-

–50 –

T H E MO S A IC S OF T H E QU BBA A L -Z A H I R I Y YA I N DA M A SCUS

FIG. 6. Mosaic of the western wall of the Qubba al-Zahiriyya

tial and commemorative structures which dotted the Mamluk cities.9 Second is the badhahanj, which has been previously used to identify a Mamluk city’s skyline.10 Third, the Mamluk mosaic scenes manage to create the effect of varied architectural compositions with a restricted repertoire. This is a characteristic of Mamluk architecture itself, which manipulated a small number of elements to produce variations suitable for multiple functions. It is possible that the

Mamluk mosaicists were expressing this feature of the architecture they were depicting by restricting their own repertoire of basic forms. This concordance between the architectural representation in mosaics, within the limits of the medium and its conventions, and the actual Mamluk architecture appears to have been further developed in the scenes of the qa‘a in the Citadel of Cairo, to which I shall now turn briefly.

– 51 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 7. Mosaic of the northern wall of the Qubba al-Zahiriyya

THE MOSAICS OF THE CITADEL OF CAIRO’S QA‘A Unfortunately, two fragments are all that remains from the large mosaic frieze which once ran around the inner walls of the excavated qa‘a above a high marble dado and another continuous frieze of small trilobed niches. The first is so effaced that no description of its composition is possible, although it undoubtedly shows some type of structure on its left corner. The second displays a building flanked by trees in an almost conventional façade style with no attempt to create a sense

of depth. The workmanship is different from that of the Qubba al-Zahiriyya, and the size of the tesserae is much larger, but the representation is nonetheless expressive. The structure in the middle appears to be a light, tripartite and domed garden pavilion with wide openings. Its first level consists of an arcaded porch, made of what appears to be a double-arched opening in the middle, flanked by closed bays, and another set of arched openings on the sides. The centre of the second level is occupied by a two-storied structure surmounted by a pear-shaped lantern whose base is embel-

– 52–

T H E MO S A IC S OF T H E QU BBA A L -Z A H I R I Y YA I N DA M A SCUS

FIG. 8. Detail of the mosaic of the northern wall of the Qubba al-Zahiriyya

lished by round tesserae of mother-of-pearl. It is flanked by two light structures whose tops are unfortunately deleted, but which must have been domed lanterns, because the ubiquitous pointed representations of drums, suggestive of a dome on top, are still intact (fig. 9). Trees with unnatural, large leaves appear from behind these two structures as if to suggest a sense of depth in what is otherwise a totally flat architectural representation. Like the scenes of the Qubba alZahiriyya, the architectural scene in the qa‘a seems to have been related to contemporary Mamluk architecture. It recalls in its general character the type of structures named manzara in contemporary sources. A manzara was primarily a light pavilion in a pleasance garden with numerous openings. The etymology of the word alludes to the structure’s

basic function as the place whence one looks out, perhaps the equivalent of a belvedere. However, unfortunately, we do not have any contemporary description of the actual layout and form of a manzara.

THE ICONOGRAPHY OF THE MOSAIC ARCHITECTURAL SCENES At this point, I will consider the mosaics of both examples in their artistic and iconographic contexts. The first point to elucidate is the reason behind the revival of glass mosaic after it had been confined to respected Umayyad monuments for at least five centuries. The sources do not offer any clue, but the simplest answer is to surmise that the rarity and opulence of the medium as it appears at the Umayyad Mosque and

–53 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 9. Detail of the mosaic of the Cairo Citadel’s Qa‘a

the Dome of the Rock had prompted this revival. This proposition, however, depends on the presence of interested patrons, who must have desired the reintroduction of mosaic, and who decided at a certain moment to ask the mosaicists that were usually called in only to repair some damaged scenes in some Umayyad monument to apply their medium in new structures. Michael Meinecke postulates such a moment. He argues that the revival of mosaic techniques was the result of an effort

by Baybars who sponsored an atelier, formed most probably of Damascene craftsmen, to restore the mosaics of the Umayyad monuments in Damascus and Jerusalem.11 This is indirectly revealed by Ibn Fadl-Allah al‘Umari who says that new mosaic tesserae were being made for the repair works of the Umayyad Mosque in his time, and that tesserae from the same stock had been used in the mosque of the governor of Damascus, Tankiz (built in 1317), whose building he witnessed.12

–54–

T H E MO S A IC S OF T H E QU BBA A L -Z A H I R I Y YA I N DA M A SCUS

The second point to delineate is the relationship between the mosaics of the Qubba al-Zahiriyya and those of the Citadel’s qa‘a. Given the fact that the former precede the latter by at least twenty years, and maybe forty years, depending on the correct dates of both structures, the two examples may be seen as two stages, derived from different pictorial sources, in the adaptation of the mosaic medium to representational purposes in the Mamluk period. From the previous analysis of the motifs and details of both examples, it becomes clear that there was an interest to represent recognisable and existent architecture in these scenes. In the qubba the craftsmen were still confined to the Umayyad models, although they were seemingly trying their best to infuse the new scenes with some degree of contextuality. By the time the qa‘a in the Citadel of Cairo was decorated, it seems that the mosaicists, or perhaps their patrons who should have been the ones most concerned about the intelligibility of the scenes, had already realised that a new mode of representation was needed. The source of infusion appears to have been sought in another artistic medium, miniature painting, which was similarly concerned with developing pictorial conventions. It is very difficult to substantiate this interpretation, but the striking resemblance between the scene at the citadel’s qa‘a and several almost-contemporaneous Mamluk and Mesopotamian manuscripts where a whole range of analogous tripartite, light and domed structures with arched openings rendered in façade form supports it.13 The closest of these examples, a miniature in a manuscript of the book Kashf al-Asrar, is in fact Mamluk from either Syria or Egypt and is dated mid fourteenth century.14 It is composed of a central structure flanked by two smaller rooms topped with ribbed, slightly bulbous domes and is open to the outside through numerous grilled, large windows. This hypothesis leads to the question of the reason behind the usage of mosaics with

architectural representations in the first place. The answer, it seems to me, has to do with the referential dimension of these scenes rather than with their artistic value or their elaborate execution. In other words, the architectural scenes were used because the patrons grasped the expressive potential of their architectural iconography. Mamluk sultans such as Baybars and Qalawun seem to have had ample opportunities to admire the mosaic scenes in the Umayyad monuments and to notice their anagogic effects as illustrated by the traditions that were current among the Damascenes.15 They also realised through the repair works they ordered in the same monuments that an acceptable degree of expertise was available to use the same techniques in new structures. The sultans, however, were interested in another kind of message related to their own status. When they considered the mosaic decoration with architectural scenes, the range of possible meanings they could read in them was clearly delimited by the subjects depicted: buildings and their natural surroundings, and settlements, large and small. But what was the meaning, or meanings, which were read into the mosaic architectural representations and which went above and beyond their architectural iconography? It was, I believe, pride in the territorial conquests of the early Mamluk sultans and the extent of their dominion. Al-Zahir Baybars, al-Mansur Qalawun and al-Ashraf Khalil had distinguished themselves in fighting the Crusaders and the Mongols and had managed to recover the whole of natural Syria from them and secure it for the Mamluk sultanate. They naturally would have wanted to celebrate these conquests, and through them to assert their sovereignty and to exalt their own persons in clear and lasting ways. Thus, probably, the idea of reading into the stylised architectural representations on the walls messages about the geographic extent of the sultan’s dominion was developed.16 Moreover,

–55 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

simplifying the architectural imagery was not only the result of the mosaicists’ limited capacity to invent new compositions.17 It may have also been intentional to a certain degree and aimed at making the message clearer and more intelligible to contemporary viewers so that the sultans can boast visually of their conquests and domains in one space of representation. This interpretation is very hard to substantiate through the written sources. The available texts are in fact silent in relation to the scenes of the Damascene mausoleum and the Cairene qa‘a. They do not even report their existence in the first place. However, a single reference furnishes an explicit indication of the type of messages assigned to architectural representations in a contemporary palatial structure. Ibn ‘Abd al-Zahir, who was a semiofficial court historian for both Baybars and Qalawun, relates in the events of the year 1284 that when Qalawun built a new throne hall (called qubba as well) in the court of the citadel, he ordered depicted on the walls of its riwaqs the likeness (sifat) of each of his castles (husun, pl. husn) and citadels (qila‘, pl. qal‘a) surrounded by mountains, valleys, rivers and seas.18 Clearly, what Ibn ‘Abd al-Zahir is expressing in this sentence is the official reading of the scenes. The verisimilitude, as we understand it today, between the architectural representations and Qalawun’s actual castles was of secondary importance, if at all, to the plausibility of their identification as such. What counted was that these scenes provided probable and variegated images that could be read as representing the real castles and fortresses. Slight differences in compositions, similar to the ones encountered between the scenes in the Qubba al-Zahiriyya, would have been sufficient to justify their interpretation as different localities by contemporary viewers. The patron need only supply the intended messages for the pictorial vehicle that broadly corresponds to them to be accepted by his

audience.19 At the citadel, Qalawun’s major objective was most probably to exploit these identifications to boast of his own military achievements and the extent of his dominion in a single, prominent space: his own throne hall. Ibn ‘Abd al-Zahir does not tell us in what medium the castles were rendered, but their themes and compositions recall the scenes in the Qubba al-Zahiriyya and the qa‘a at the Citadel of Cairo. Since these scenes were executed in mosaic, and since we do not know of any other medium that was used at that time for architectural scenes on walls, it is plausible that those in Qalawun’s hall were done in mosaic as well. Michael Meinecke has shown that Qalawun was responsible for the completion of the Qubba al-Zahiriyya in Damascus. He may also have brought Syrian craftsmen to Cairo to execute the decoration in his bimaristan complex, which was built shortly afterwards (1284), and which has a mosaic conch in its madrasa’s mihrab.20 The same Syrian mosaicists who worked on the scenes at the Qubba al-Zahiriyya in Damascus may then have created the representations of castles and fortresses in Qalawun’s new audience hall at the Citadel of Cairo. The question remains: how, if at all, could a medieval viewer identify specific sites in the mosaic scenes we see today as generic and idealised. To try and answer this question we will have to turn to the Umayyad Mosque in Damascus which has achieved such a hallowed position in the imagination of medieval Muslims that many considered it among the wonders of the world. Consequently, its mosaics ellicited a series of descriptions from the tenth century to the present that allow us to check if their meaning changed over time. On the surface, this does not seem to have been the case. In fact, the descriptions seem to copy one another and to repeat the same motifs. But a careful reading hints at a perceptual development in the descriptions that occurred in the thirteenth century, that is

–56 –

T H E MO S A IC S OF T H E QU BBA A L -Z A H I R I Y YA I N DA M A SCUS

when the mosaic medium was readopted in contemporary Mamluk structures. The earliest report we have comes from the geographer al-Muqqadasi (fl. 966–1000), who tells us in passing that the mosaics depict cities and villages, but does not identify any of them.21 Ibn ‘Asakir (1105–1176), the foremost medieval historian of Damascus, collects together all the previous reports on the acquisition and cost of the mosaics in his comprehensive compendium, Tarikh Madinat Dimashq. He also copies a poem composed after a fire in 1069 devastated the mosque in which the variety and enchanted nature of the trees represented are extolled, but the architectural scenes are not once mentioned.22 The encyclopedic geographer, Yaqut al-Hamawi (1179–1229), reports that the porticoes of the mosque contain representations of every city and every kind of tree in the world.23 He attributes this report to the famous essayist al-Jahiz (d. 869), but when he himself describes the mosaics he only praises the intricate depictions of plants and trees. The famous cosmographer of the thirteenth century, Zakariyya al-Qazwini (1203–1283), repeats the words of alMuqaddasi almost verbatim.24 Another famous cosmographer, Sheikh al-Rabwa al-Dimashqi (1256–1327), who was Damascene by birth, widens the scope of the scenes by adding castles and seas to al-Muqaddasi’s list and gives them an Islamic stamp of approval by remarking that they contain no representation of forbidden subjects (i.e. human figures).25 Reports from the second half of the fourteenth century begin to mention the architectural scenes with a new emphasis on their possible worldly models. The first to do so is the Damascene biographer Ibn Shakir al-Kutubi (d. 1362) who states that the mosaics ‘depict the Ka‘ba’s likeness [sifat], set above the mihrab, and the images of cities and villages, each represented with all that it produced of trees remarkable for their fruits or their flowers or other

objects’.26 His text was reused in several later Damascene sources such as by al‘Ulmawi in his description of Damascus (written 1566) and al-Busrawi in his Tuhfat al-Anam fi Fada’il al-Sham (c. 1595).27 Al-Kutubi’s description differs from those of his predecessors in two critical respects.28 He is the first to specify that the central part of the scenery depicts the holiest Islamic centre, the Ka‘ba, which spatially and ritually represented to the Muslims the centre of the world. He also implies a certain pictorial differentiation between the illustrations of cities based on what he terms their distinct products of fruits, flowers or other objects, which indeed are elaborately diverse in the mosaic compositions. A little-known Egyptian chronicler, Abu al-Baqa’ al-Badari, writing in 1494, develops this idea further by claiming that the cities and villages are each represented with the wonders (‘aja’ib) that distinguish it.29 He thus relegates all the unexplainable architectural details, mysterious landscapes and fantastic plants with jewel-like fruits we see today in these scenes, and which we cannot comprehend, to the category of ‘aja’ib and reads them as characterisations of their locales. For him, the accuracy of representation does not seem to have been important as long as the scenes suggested different cities, which could be distinguished by legible marvellous attributes, and which are arranged in a certain order around the Ka‘ba. Thus, it seems that Ibn Shakir and his successors saw in the mosaics an iconographic, global depiction of the countries of Islam, probably disposed in some hierarchical order on the two sides of the Ka‘ba’s scene along the porticoes of the mosque. This interpretation is new and it brings the scenery down from an eschatological realm to a political one. Al-Kutubi’s account, though based on the older historians, undoubtedly reflects current Mamluk interpretations of this type of scene. Such a logical, mundane and down-to-earth interpretation may have

– 57 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

originated in the Mamluk period after the methods of representation and themes of the Umayyad Mosque’s mosaics were taken over by the Mamluk sultans and applied in their structures to convey another kind of message. This message was in turn reapplied by al-Kutubi and his contemporaries and successors to the Umayyad Mosque’s scenes, which therefore acquired a new Mamluk meaning. This reciprocity of form

and meaning was possible only because by the thirteenth century the Umayyad imagery had become cryptic and was thus open to interpretation while, at the same time, the new Mamluk scenes had been composed and put to service in the expressive vocabulary of Mamluk art, thus introducing a novel dimension of significance to the mosaic architectural scenes, both new and old.

–58 –

6. THE MILITARISATION OF TASTE IN MEDIEVAL BILAD AL-SHAM

T

he immense impact of the encounter between European Christendom and Bilad al-Sham during the Crusades has somewhat overshadowed another pivotal event that began shortly before in the same area but had greater and more lasting effects. This was the coming of the Turks from Central Asia via Iran and their gradual settlement in the region between the middle of the eleventh and the beginning of the thirteenth century. The Turks, or the Ghuzz as medieval Muslims knew them after one of their main tribal confederations, had begun to enter the Islamic world as individuals, either free or as Mamluks, around the end of the ninth century.1 Shortly afterwards, some tribal groups crossed the Oxus River (Amu Daria today) and infiltrated the north-eastern frontiers of the Islamic world under pressure from competing Turkic clans. The most formidable among them, the Seljuqs, served the Ghaznavids before clashing with them and defeating them in the battle of Dandiqan in 1040. They then spread their authority across Iran, Iraq, and later Anatolia and Bilad al-Sham, to establish a loosely defined sultanate, which nonetheless ushered a new era in Islamic history. The Seljuqs restored the Sunni Abbasid

caliphate to religious prominence, opened Anatolia to Islamisation and Turkification after the decisive battle of Malazkurd (Manzikurt) (1071), and reinvigorated the flow of ideas and people between the Islamic Mediterranean, Iran and further east. But the most fundamental consequence of their short-lived sultanate was the introduction of a new polity that was to dominate the entire region through the many post-Seljuqid independent principalities that sprang up in Anatolia, Bilad al-Sham, Iraq and Egypt.2 Many of the founders of these principalities were Turkish-speaking amirs who belonged to the recently Islamised and staunchly Sunni military class brought up by the Seljuqs. They led small but agile armies made up mainly of Turkish and Kurdish free and manumitted cavalry, expanded their principalities through war and conquest, and distinguished themselves in the struggle against a host of enemies: the Byzantines, Crusaders, Armenians, various Shi‘ite states or insurgencies, Bedouin tribes and, later, the Mongols.3 Several attributes characterised the new ruling class and distinguished it from those of earlier Islamic societies. First was a strict military hierarchy, expressed in clearly stratified ranks of amirs – amirs of ten,

– 59 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

twenty, forty, 100 – and various tiers of soldiers divided by legal status (Mamluk v. free) or ethnic group (Turkic, Circassian, Kurdish, Armenian or Mongol). This military order represented the only path to political power and its attendant financial and landholding prerogatives, and was closed to all but few exceptional local recruits. It defined the new ruling class and distinguished it from other social groups such as merchants, ulama, artisans and peasants, which, despite their periodic disapproval of particularly corrupt rulers, seem to have rarely challenged the system itself. Bedouin tribes, organised city riffraff and other fringe groups that could potentially pose a security risk were either heavily suppressed through routine military campaigns or bribed by attaching them to the army as auxiliary regiments, especially in times of external danger. A second characteristic was the fortress mentality displayed and expressed by the new ruling class towards their subjects. This was almost always a noticeable trait, despite the internecine fights that otherwise marked the rulers’ interrelations, especially during succession contests. The fortress mentality was initially engendered by the elite’s linguistic and ethnic differences from subject populations, and ultimately became embedded in a total system that stressed exclusion and segregation as means of control.4 As such, separation defined most details of communal life, like habits of socialisation, business dealings and marriage. It was also inscribed in the spaces of the city and its suburbs, and expressed in particular and distinctive nomenclature and insignia (rank) systems and intricate dress, dietary, musical and ceremonial codes. The Mamluk sultanate in Egypt and Bilad al-Sham was the last and perhaps the most elaborate example of this exclusive polity, which nonetheless managed to penetrate the entire social spectrum and to modify it profoundly and in lasting ways.5

The third quality, which might be termed propagandist patronage, was almost a corollary of the second, though opposite to it in aim and effect. Despite their segregation, haughtiness and heavy dependence on their military absolutist power, the new rulers still sought a sociopolitical platform around which popular approval and support could coalesce. To this end they strove to fulfil the duties of rule as promulgated by Islamic political theories. They attempted to appeal to their subjects by presenting themselves as the defenders of the religion, the pursuers of Jihad, the upholders of justice, the supporters of ‘ilm and adab, and the benefactors of the poor and destitute. They allied themselves with, employed and patronised the religious class, whose members mediated between them and the people and provided them with legal and religious endorsement when needed.6 The rulers also endowed religious structures: mosques for the faithful, madrasas to educate a new class of scholars, ribats and khanqahs to lodge the Sufis, and mausolea to commemorate themselves and to aggrandise their deeds. They also built bimaristans, caravanserais, palaces of justice and other civic projects to prove their public commitment and support for social services.7 They widely publicised their enactment of strict religious rules against social transgressions, and relied heavily on elaborate titles that stressed their glorifying actions and qualifications, their dedication to Islamic causes and their symbolic links with the legendary heroes of the past. They inscribed these titles on their buildings and objects, and listed them in all of their correspondence.8 One concomitant transformation brought about by the new ruling class was the prevalence of military expression in the art and architecture they sponsored.9 This shift can be observed on so many levels from the largest monuments to the smallest decorative detail or everyday objects. For their residences, all the new rulers chose to live in citadels that stood on the edge of the city

–60–

T H E M I L I T A R I S A T I O N O F T A S T E I N M E D I E VA L B I L A D A L - S H A M

and enclosed their palaces, their audience halls, barracks for their armies and stables for their horses, while they had their hippodromes (mayadin) for military parades and polo games established nearby outside the walls.10 Between the eleventh and thirteenth centuries, more than 100 citadels were built or refurbished all over Bilad al-Sham and the surrounding regions, and only a fraction of them can be ascribed to the protracted attrition war between the Crusaders and Muslims. In today’s Syria we have citadels in Damascus, Aleppo, Homs, Hama, Palmyra, Bosra, Ja‘bar and many smaller towns. In northern Iraq we have the citadels of Mosul, Irbil, Takrit and Sinjar. In the Jazira, the main citadels are those of Diyarbakr (Amid), Mayyafariqin and Hisn Kifa, and in Anatolia, those of Kayseri, Sivas, Divrigi, Alanya and Konya. The largest, most awesome, most comprehensive and the last one constructed is the Citadel of the Mountain in Cairo, which was began by Salah al-Din in 1176. These citadels her-

alded the new regimes whose roots were foreign and preferences were military. They were built to be a refuge against attack, a residence for the military elite, a barrier against the ruled and a symbol of the rulers’ valour as the defenders of Islam. To my knowledge, the earliest of these citadels to embody the change in taste and function was the Citadel of Aleppo (fig. 10). Erected on a hill with many building levels that go back to at least the second millenium BC, the citadel lay deserted in the early Islamic period. According to ‘Izz al-Din ibn Shaddad, the chronicler of the cities of Syria and Jazira in the thirteenth century, the citadel was first refurbished by the Hamadanid Sayf al-Dawla after the Byzantine attack on Aleppo in 962 proved its impregnability and strategic potential.11 But it was not until the Merdasids seized Aleppo in 1023 that it became the seat of government. The Seljuqs and Zengids also lived there, and Nur alDin Mahmud is the first amir whose specific work there is reported by the sources. The

FIG. 10. The Citadel of Aleppo

– 61 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

real builder of the citadel, however, was alZahir Ghazi (1186–1216), Salah al-Din’s son, who monumentalised it and redefined its relation to the city. He reconstructed the fortifications, added the impressive glacis and massive entrance tower and bridge, and built the mosque with its tall minaret. He then added the palace named Dar al-‘Izz (which was rebuilt by his son and grandson), a few houses, qa‘as, hammams and a garden, and renovated the Maqam Ibrahim, a shrine dedicated to the prophet Ibrahim, popularily believed to be a sort of patron saint of Aleppo. Al-Zahir Ghazi also built a royal audience hall, known as the Dar al-‘Adl (palace of justice) south of the citadel, in what may be considered the transitional zone between the city and the citadel where the ruler and the ruled met. He enclosed the Dar al-‘Adl and the surrounding area, which contained the maydan (training field) built earlier by Nur al-Din, between the old walls of the city and new ones he had added especially for that purpose further to the east. The exclusivity of this new royal enclosure was conveyed through the control of its access and the establishment of a private passage leading to it from the royal palaces in the citadel.12 The second oldest urban citadel is that of Mayafariqin (Silvan in today’s Turkey), which was first built by Nasr al-Dawla Muhammad ibn Marwan, the founder of the Kurdish Marwanid dynasty who ruled between 1011 and 1061. A very revealing remark about the choice of location for an urban citadel is ascribed to a certain Abu al-Qasim, a vizier of Nasr al-Dawla. The remark also points to a possible terminus a quo for the change in perception about the dual function of physical separation and visual display desired of the ruler’s residence. When Nasr began building a palace in Mayafariqin where the old palace of the Hamadanids had stood, Abu al-Qasim advised him instead to build it at the highest point in the city. He also suggested that Nasr appropriated a preexisting tower,

known as Burj al-Mulk, and a nearby gate, and annexed them to the new royal enclosure. This way, Abu al-Qasim reasoned, the new palace would dominate the city and be highly defendable. Nasr not only adopted his vizier’s advice, but he went on to build a palace, a garden, a complicated waterworks system and, most importantly, a huge manzara or belvedere that overlooked the city from high atop the palace walls and symbolised the relationship between the ruler and the ruled. Even if the reconstruction of the citadels in Aleppo and Mayafariqin predates the Seljuqs – a point that may weaken the historical background of my interpretation – the militarisation of taste, of which building urban citadels was a key manifestation, was only strengthened by the arrival of the Turks. This was so because the Turkification of rule – a momentous phenomenon that deeply transformed the cultural, social and ethnic makeup of Bilad al-Sham and beyond – was primarily a consequence of its militarisation. By militarisation of the rule, I mean the reliance on military power to justify political authority without recourse to religious, ethnic or class-based legitimisation schemes traditionally upheld by monarchical or theocratic regimes. This was the case of the tribally controlled small principalities that sprang up in Bilad al-Sham in the late tenth and early eleventh centuries, such the Hamadanids and Marwanids. But what began as local, mostly Bedouin, responses to the disintegration of the centralised Fatimid and Abbasid caliphates soon gave way to stronger pressures from the north and east. First the Seljuq Turks, and then the various Turkic, post-Seljuqid groups took advantage of the power vacuum and consolidated their rule over the entire area.13 Like the local Bedouin chiefs they replaced, Turkic amirs depended on military hierarchies to protect and justify their tenure. They rose to dominance precisely because their stratification system was more thorough than the local one, and perhaps

– 62–

T H E M I L I T A R I S A T I O N O F T A S T E I N M E D I E VA L B I L A D A L - S H A M

also – if we were to trust popular beliefs – because their fighting individual was more disciplined and more agile than the Bedouin warrior.14 Hence linking the two processes of militarisation and Turkification of rule is predicated by the historical circumstances of the eleventh century. Yet the militarisation drive did not reach its full capacity until the Turks or Turkified warriors were completely ensconced as the rulers of the Islamic eastern Mediterranean in the twelfth century. Although I have not been able to locate a text that unequivocally establishes the precedence of the Citadels of Aleppo and Mayafariqin, the content and tone of the reports quoted above suggest just that.15 It is clear, for instance, that the builders of these citadels were rediscovering the defensive advantage of citadels as royal residences rather than reproducing established patterns or copying existing examples. Furthermore, the reports do not indicate that the move to the citadel came as a response to a specific assault, such as an invasion or a revolution. It was instead a cumulative process that began with deciding to build a citadel and involved trial and error, change of emphasis, rebuilding, expansion and even desertion of previously built citadels, all in the pursuit of the most appropriate setting for the new ruler and his army. The same process is repeated in many more examples during the twelfth and thirteenth centuries all across the area.16 Here is ‘Izz al-Din Ibn Shaddad reporting on the chronology and shift of construction in the Citadel of Damascus from its foundation in the eleventh century up to his day, that is during the reign of al-Zahir Baybars (1260–1276): No palace of government (dar imara) was left in Damascus until the accession of Taj al-Dawla Tutush (the Seljuqid) in 471 (1078), who built a small citadel, made it a dar imara, and moved to it. He built a house for his son Ridwan in the Citadel, which – 63–

still carries his name until our days [. . .] When Nur al-Din ibn Zengi took over, he built a nice house in the Citadel. It is attributed to him until today. He also built a magnificent palace known as The House of Joy (Dar al-Massarra) and constructed a hammam next to it. He also built the Grand Mosque at the Citadel with a minaret and a basin and added a fountain at the mosque’s entrance. He then established a small mosque, or an oratory, in the durkah (vestibule or most probably chicane). When al-Malik al-‘Adil (the Ayyubid) took Damascus, he demolished the Citadel and distributed the work for its rebuilding among his amirs. He planned it with twelve huge towers and ordered each amir to build a tower. He also dug a moat and filled it with water. Its reconstruction was a great success because of the large sums of money spent by al-‘Adil’s amirs. Then al-‘Adel’s son, al-Malik al-Mu‘azzam ‘Issa renovated its palaces and houses. When al-Malik alAshraf Musa took over in 626 (1228), he demolished Dar al-Massarra and rebuilt it anew. He also constructed a new palace called al-Bahra (the Bassin). His brother, al-Malik al-Kamil Muhammad added a new palace that became known as al-Dar al-Kamiliyya after him [. . .]. The Mongols destroyed the citadel’s curtain walls and crenellations and damaged its towers when they occupied Damascus in 658 (1260). So when our lord al-Sultan al-Zahir Baybars acquired Damascus he renovated the citadel and rebuilt what the Mongols had destroyed. He constructed a high and elegantly built lookout (mushtaraf) atop the Tower of the Corner (Burj al-Zawiya) which overlooks the Maydan (Hippodrome). He also added a new hall (qa‘a) near al-Bahra Palace for his son al-Malik al-Sa‘id. Finally, he erected a hammam on the side of the Citadel’s gate from the city. The construction work is still going on in the Citadel until the date of our present recording, which is the year 675 (1276) [the year of Baybars’ death].17

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

Most citadels built or refurbished during the same period display similar patterns of organisation. Palaces and halls for the ruling family and the courtiers occupied the centre of the citadel or its high point, whereas residences for the troops, stables and storehouses were dispersed on the periphery, at lower levels, or within the defensive towers. In some instances, notably the Citadel of Cairo, the two areas, royal and military, were separated by a curtain wall pierced by a ceremonial gate. Thus the topography and architectural setting of citadels visually and spatially reflected the sociopolitical hierarchy imposed by the new rulers. They emphasised the centrality of the ruler’s abode among his troops and in the city and conveyed obvious messages of power, seclusion and grandeur. These signs were aimed equally at the army residing within the citadels and the local population restricted to the city outside their ramparts, except on the days of audience when they were admitted to a proscribed area inside. This practice of contiguity to the city coupled with physical separation and visual or regulated contact was engendered by the cultural and social changes brought about by the shift in the political structure of the entire region.18 Other aspects of the militarisation of architecture went beyond the building of citadels to affect the forms and meanings of surface articulation applied to their walls, towers or the palaces inside them. Most prolific were the ubiquitous inscription bands with the elaborate honorific titles that appeared on the most conspicuous places in the citadels: on towers and machicolations, above gates and windows, along curtain walls and in stone slabs affixed in special areas (fig. 11). They emphasised the warlike qualities of the new rulers and their commitment to the defence of Islam and the jihad against its enemies.19 They also stressed the sociopolitical, civic and pietistic attributes of the patron to complete the paradigmatic image of the true Muslim ruler

as it had developed by the medieval period. These verbal descriptions were complemented and visually and artistically enhanced by the art form that is least expected to be found in a medieval Islamic context: sculptural and painted figures. Many walls and towers of the citadels and cities fortified by the new rulers were decorated with stone reliefs depicting human and animal figures, mythical creatures and architectural scenes. Some were reused classical spolia, but most were newly carved after ancient or Central Asian models, or, in some exceptional cases, ex nihilo.20 Examples of both kinds are still visible on the walls of the citadels in Konya, Aleppo, Diyarbakr, Divrigi, Damascus, Palmyra, Bosra and Cairo, among others. They are conspicuously positioned around areas of major inscriptions, that is, areas of high visibility and heavy traffic in and out of the citadels or the cities. They are obviously intentional FIG. 11. Three-level inscriptions, main gate, the Citadel of Aleppo

–64–

T H E M I L I T A R I S A T I O N O F T A S T E I N M E D I E VA L B I L A D A L - S H A M

and meaningful, but their exact significance is still not clear. It has to be deduced from their forms, references and positions since they are generally devoid of any inscription that identifies them and they are not mentioned by the chroniclers of the period.21 The same situation arises in the case of nonmilitary architecture, artefacts and even coinage of the period, especially in Anatolia and the Jazira, which display various anthropomorphic, zoomorphic and zodiacal reliefs as well. Their meanings and references too are still subjects of guesswork and imaginative theory.22 The Citadel of Aleppo has perhaps the most complete sequence of carved figures positioned along the path from the main gate through the winding vaulted passage leading to the main audience hall above the entrance tower. The series starts with the motif of two interlacing tubular snake bodies, with a dragon’s head at their four ends, carved above the arch of the main portal known as the Serpent Gate (Bab al-Hayyat) and attributed to al-Zahir Ghazi’s rebuilding (fig. 12). This is a well-known design. It is found in so many other contemporary examples all over Anatolia, Bilad al-Sham, and even Iraq and Egypt.23 The most famous of these dragons’ interlaces is the no longer extant carved motif above the Talisman Gate in Baghdad, dated to 1221–2. It is attributed to the caliph al-Nasir li-Din Allah, who, although Abbasid himself, was a product of the militarised culture and a keen player in the power struggle of his time. He is said to have intentionally represented himself seated cross-legged between the two dragons, which represent the two dangerous sultans whose plots on his authority he managed to halt: the Khwarazmshah and the Mongol, or the Isma‘ili lord of Alamut and the Khwarazmshah (fig. 13).24 The dragon itself is a widespread talismanic symbol, which the Turks may have brought with them from their shamanistic Central-Asian past or appropriated from the Chinese cosmology. Although we still don’t

FIG. 12. The Serpent Gate (Bab al-Hayyat), the Citadel of Aleppo

know the precise meaning of the various dragon motifs, as most of them are of the paired variant and are sited above entrances to palaces, towers and cities in medieval Anatolia and Bilad al-Sham, they were probably a mixture of a talisman and a political manifesto. The dragon magically protected the space within and warned any malefactor of the might and swift retaliation of the defender of that space, the amir or sultan who commissioned the motif in the first place. This must be how the builders of the Bab al-Hayyat understood and intended their dragons’ interlaces, otherwise why would they put them atop the main entrance to the Aleppo Citadel where everyone would have to pass under them. This interpretation of the dragons is substantiated by the next two carved motifs found along the ascending vaulted path in

– 65–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 13. The Talisman Gate in Baghdad

the Citadel of Aleppo. High above the first inner door is a carved scene of two lions facing each other across a diminutive palm tree, which is most probably meant to be the Tree of Life (fig. 14). Paired animals on the two sides of the Tree of Life is an ancient motif of the region, which goes back to preclassical times but appears profusely in Christian and early Islamic mosaic floors in Syria and Palestine, and again in carved stone examples in medieval architecture. Its precise meaning and references elude the modern interpreters, but it must have been connected to the notions of protection, vigilance and shielding. This broad attribution is predicated primarily on the meaning of the lion, the symbol of power and invincibility in almost all of the ancient cultures. The lion itself appears in countless examples and in various combinations and postures covering the entire region from Azerbaijan to

Egypt. Stone carved lions can be found in Seljuqid and post-Seljuqid architecture, and in the Mamluk realm, where the lion passant appears on all Sultan al-Zahir Baybars’ buildings, as it was his emblem (rank). But the next motifs encountered at the Citadel of Aleppo, though still using the lion figure, seem to be a rarity. From the two jambs of the final door along the vaulted path leading to the interior of the citadel protrude two stiffly carved lions, dubbed, probably in recent times, the smiling and crying lions because of their facial expressions (figs 15 and 16). Similar projecting lions exist in a small number of khans, madrasas and citadels in Anatolia. However, the Aleppo lions are unique in their expressions, which must have been intentional. Could they signify the welcoming and the threatening? They most probably do, as these two qualities belonged to what the medieval mentality

– 66 –

T H E M I L I T A R I S A T I O N O F T A S T E I N M E D I E VA L B I L A D A L - S H A M

FIG. 14. Carved scene of two lions, the Citadel of Aleppo

ascribed to the just, resolute and powerful ruler. The two qualities were likewise emphasised in inscriptions and highly praised in the medieval treatises on the prerequisites of good rule known as ‘Mirrors of Princes’. Another pertinent genre of representation was what has been termed ‘the Princely Cycle’. This included images of battles and victories, of sultans and their amirs in full military regalia, and of the cities and fortresses they had conquered. Palaces decorated with variations on the Princely Cycle are reported in the written sources for the citadels of Aleppo, Mosul and Cairo, among others. Although none of these examples survive, we can get a glimpse of what they may have looked like by reading the admiring reports in the chronicles. Two palaces in

the Citadel of Cairo, the first built by alZahir Baybars and the second by al-Ashraf Khalil stand out as the most illustrious examples. The Palace of al-Zahir Baybars, known as the Qubba al-Zahiriyya, had figures of the sultan and his amirs ‘painted’ on its interior walls.25 Ibn Shaddad says that the scenes represented Baybars and his retinue in the day of the procession (mawkib), and cites part of a poem by Abu al-Fityan Ibn Hayyus, a twelfth-century poet from Damascus, which he says almost depicts the representations on the qubba’s walls. It is a very meagre source of information to rely upon, but given the lack of information on figural representations, it does furnish a few clues to the composition and the topic represented:

– 67–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 15. The Crying Lion, the Citadel of Aleppo

FIG. 16. The Smiling Lion, the Citadel of Aleppo

And you behold the birds of the air upon its sides Some in flight, others perching And racehorses whose hooves never leave the ground As though they gallop beneath their riders Then those who hurl thunderous blows that do no harm And those who wear yalmaqs26 that cannot be removed A group of them have drawn their shining blades, and another Has drawn his bow, whose arrow has no target And cannot leave the bow And his snares are always fatal for the birds

The audience hall or iwan renovated by alAshraf Khalil ibn Qalawun in 1293 had, Ibn al-Dawadari reports, representations of his amirs, each with his own emblem (rank) above his head, probably as a means of identification. This composition suggests a convention of figural representation different from the Princely Cycle of Baybars’ qubba. Similar designs with earlier dates can be found in the Jazira. They, too, do not include composite scenes, but rather single standing figures, which differed from each other only

in the attribute of their office. The figure with a sword may signify the silahdar, the one with a bow and arrow the bunduqdar, and the one with a napkin and a beaker the saqi, and so on. In two examples – a stone bridge over the Tigris River built by the Artuqid ruler of Husn Kifa Qara Arslan (1148–1167) and a stone niche from the Gu’ Kummet at Sinjar, dated around 1240 – single figures are carved in relief, each standing by itself in its own frame (fig. 17). The figures have been interpreted as representations of the khassakiyya (the close Mamluks) of the ruler of these two principalities, and the whole composition as a symbol of sovereignty. The rulers are not shown, but their royal status is implied by the presence of their attendants and courtiers and, in the case of the Sinjar niche, a baldachin is carved at the pinnacle of the composition, probably symbolising the absent ruler who would normally have sat there.27 The ruling elite also acquired objects decorated with representations of military themes – princes and courtiers, campaigns, parades and hunt excursions – and they had their names and titles inscribed on them as a sign of their admiration of these objects. The most famous is the so-called Baptistère

–68–

T H E M I L I T A R I S A T I O N O F T A S T E I N M E D I E VA L B I L A D A L - S H A M

FIG. 18. Rank of Amir Qawsun, carved on a stone slab, garden of the Mosque of Sultan Hasan, Cairo

FIG. 17. Detail of the niche from the Gu’ Kummet at Sinjar, the Bunduqdar

de Saint Louis, whose date and patron have been debated for more than fifty years. The figures on the brass basin have been repeatedly interpreted to be the great amirs and khassakiyya of a Mamluk sultan – most probably Baybars according to Doris BehrensAbouseif – represented individually and in a ceremonial formation around a central figure that may have been the sultan himself.28 The central figure is seen engaged in various occupations: riding, hunting or sitting in an official majlis flanked by two pages. The amirs’ figures may have been divided along ethnic lines into the Turks and the Mongols who formed the two dominant groups in Baybars’ court. In a style that recalls the depiction of amirs in al-Ashraf Khalil’s iwan or the carvings of the Sinjar niche, the amirs’ figures on the Baptistère are represented with what appears to have been the attribute of their ceremonial office: sword,

bow and arrow, ceremonial axe and napkin. Other metalwork objects of the period had similar themes, though more restricted and less vibrant than the Baptistère’s representations. Still others used the same animal motifs we find in architecture, such as the beautiful pierced globe made around 1270 for the amir Badr al-Din Baysari al-Shamsi, one of Baybars’s great amirs, whose emblem might have been the double-headed eagle which appears four times on the sphere. The pieces of this transformation of architectural and artistic expression fit into a wider framework of change which reflects the mentality of the new ruling elite, and which can be explained as a search for appropriate and accessible ways to express and satisfy their needs, intentions and tastes. This interpretation may be applied to other artistic and literary innovations of the time, which may be seen as manifestations of the same cultural shift. Firstly, there was the system of emblems, called ranks, distant counterparts to coats of arms in Europe, which were adopted by sultans, amirs and perhaps high officials from as early as the eleventh century. Ranks seem to have codified many of the images that the new rulers had created to represent the attributes of

– 69 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

–70 –

T H E M I L I T A R I S A T I O N O F T A S T E I N M E D I E VA L B I L A D A L - S H A M

their offices, positions and ideals. They were carved on buildings, painted on glass, wood and pottery, engraved on metalwork, and embroidered or dyed on textiles (fig. 18).29 Secondly, there was the revival of Iranian, Turkish and antique mythical and heroic images in architecture, textiles, miniature paintings, and especially the coinage of several dynasties in the Jazira in the twelfth century with their varied and intriguing figural and mythical representations.30 Thirdly, there was the resurrection of royal themes from the pre-Islamic literary traditions that could serve as models for the panegyrics composed by court poets for the new rulers. The archetype was the Shahnameh of Firdawsi, which the poet composed for the last Samanid but dedicated to Mahmud of Ghazna, one of the first paragons of the new polity, around 1020. Fourthly, there was the renaissance of the classical masalik (geography) genre, exemplified by huge compendia such as Ibn Shaddad’s al-A‘laq al-Khatira bi Zikr Umara’ al-Sham wa-l Jazira and al-‘Umari’s Masalik al-Absar fi-Mamalik al-Amsar. Both works were sponsored by Mamluk sultans, Baybars and al-Nasir Muhammad respectively, and may have reflected their pride in their campaigns and their territorial possessions.31 The overall effect of these innovations, revivals and adaptations was to give the art and architecture of the medieval period an aura of a vibrant and fluid yet inconclusive search for an artistic idiom that glorified military attributes. Some experiments seem to have led nowhere and were dropped either immediately or after a few trials. Others were felt to be more satisfactory and were adopted for longer stretches of time or for a variety of instances. Still others became fixtures and were used over and over again, some of them even surviving the medieval period into later times. Art, naturally, was echoing what was happening  FIG. 19 (Above). The Madrasa of Sultan Hasan  FIG. 20 (Below). The Madrasa of Sultan al-Ghuri

in the political and social spheres while the Turkic dynasties had been constructing their realms and their image with little preconceived notions and established norms and against tremendous odds. But things began to change when the new polity became secure, and both external and internal threats had been neutralised. The once fiercely and strictly segregated Turkish-speaking elite began to lessen the emphasis on their roles as defenders of Islam and to soften their military image.32 They took to fraternising with the local upper class through marriage and business partnerships. They also began slowly to adopt the indigenous urban culture of their subjects in which few of their cherished martial attributes could be accepted as marks of status and distinction. Other symptoms of acculturation were soon to follow, and by the beginning of the fifteenth century the rulers’ transformation was effected not only through the new habits and manners they affected, but also through the more sedate and poised artistic and architectural forms they commissioned and sponsored. These are the forms that we have come to identify with medieval Islamic art and architecture. However, their contemporaries did not necessarily see them as such. They seem to have favoured the earlier, bolder and more expressive features. This is best illustrated by an anecdote attributed to a fine connoisseur of military qualities, the Ottoman sultan Selim I, who conquered Egypt and eliminated the Mamluk sultanate in 1517. During his stay in Cairo he toured the city and inspected its monuments. He stopped in front of the marvellous Madrasa of Sultan Hasan (1357–1362) and exclaimed, ‘This is a great hisar (citadel).’ But when he saw the Madrasa of Sultan al-Ghuri (1510), his opponent who had died a year earlier fighting him in the battle of Marj Dabiq, north of Aleppo, he said, ‘This is the qa‘a [hall] of a merchant’ (figs 19 and 20).33

– 71 –

7. AL-AZHAR MOSQUE: AN ARCHITECTUR AL CHRONICLE OF CAIRO’S HISTORY

I

n 1924, Martin Briggs, a British architect who was embarking on a study of Islamic architecture in Egypt and Palestine, had these words to say about al-Azhar Mosque in Cairo: To a European, al-Azhar offers an oriental spectacle, unparalleled save by the Mecca pilgrimage, where one may realize at the same time the backwardness of Islam and its tremendous power. Nor does this picturesque scene lose anything by its staging. The dazzling white arcades that surround the sahn with their quaint battlements silhouetted against the brilliant blue sky, the duster of minarets above them – some bizarre, one at least graceful – all enhance the glow of colour presented by the many hued robes of the students and their teachers.1

To Taha Hussein, the pioneering Egyptian thinker and educator who came to al-Azhar in 1908 as a young blind fellah, the mosque sparked a totally different feeling. He wrote about his impression in the third person, ‘It was enough for his bare feet to touch the stone paving, and for his face to be caressed by the fresh breezes in the sahn to have his heart filled with peace and hope.’2

For a single structure to induce such powerful feelings 1,000 years after it was built is surely a sign of vitality. Over the centuries, al-Azhar has played a significant role in the cultural, intellectual and political life of Egypt and the Islamic world generally. Its authority, sometimes rising, sometimes ebbing, whether triumphant or vanquished, fought for or fought against, has survived the vicissitudes of history in Islamic Egypt from the end of the tenth century until today. The mosque was first built in 970 by Jawhar al-Siqilli, the Fatimid general who had conquered Egypt for his master, alMu‘izz li-Din Allah, a year earlier. He intended this to be the Friday mosque for the new city he founded and named alMansuriyya, probably after the earlier Fatimid capital near Qayrawan in Ifriqiya (Tunisia) built by al-Mu‘izz’s father, alMansur (r. 946–953).3 Soon afterwards, in 972, al-Mu‘izz himself arrived in Egypt and the mosque underwent a facelift. This was followed by a succession of expansions, additions, alterations and annexations of new dependencies and semi-independent institutions that went on until the twentieth century. The sequence of changes in al-Azhar’s architecture reciprocates and reflects its rise

–72–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

to become the foremost religious learning institution in Egypt, and the concomitant political influence its denizens enjoyed among both the ruling classes and the general population. It also closely follows the fortune of the city of Cairo itself in its progress from the capital of the selfconsciously religious Fatimid dynasty to the centre of the aggressive and expansionist Mamluk military state, to provincial capital of the Ottoman empire, and finally to contemporary metropolis. By the end of the sixteenth century, if not earlier, al-Azhar had reached such a high degree of sanctity that it was considered Islam’s fifth most important mosque, after those in Mecca, Medina, Jerusalem and Damascus.4 In the public mind, the mosque was both a sanctuary and a space for spontaneous acts of civic resistance. Many riots against cruel or foreign rulers began there or converged in its courtyard, including the revolt of the people of the Hussayniyya Quarter against the rapacious Mamluk amirs in 1785 and the uprising against the French occupation in 1798. 5 But, although al-Azhar became the people’s assembly place par excellence, and although it functioned as an independent institution with its shaykhs and students forming a self-governing community, its upkeep, expansions and embellishments were initiated and paid for by Egypt’s rulers. In fact, there seems to have been, and to a large extent still is, a discernible correlation between the political order in Cairo and the care and attention bestowed on al-Azhar both as a structure and as an educational and religious institution. With a few exceptions, one can read the intentions of the rulers in the type of work they effected at al-Azhar, or in the neglect they showed towards its maintenance. Consequently, the architectural development of al-Azhar can be seen as a chronicle of the rise and fall of leaders and factions in the religious and political history of the city and the country.

THE ARCHITECTURE OF THE MOSQUE The mosque’s original core is now totally enclosed in a cluster of later dependencies and secondary structures ranging in date from early fourteenth century to recent times. The expansion of the mosque could be described as following a spiral, that is, new structures grew up around its circumference until they completed a full circle, then, in the eighteenth century, a new series started to the south and east around the older circle (fig. 21). These additions continuously changed the mosque’s perimeter and did not always respect the neighbouring buildings, especially in the late nineteenth century. They also replaced sections of the mosque’s original walls against which they were built. Consequently, only a tiny portion of the Fatimid western façade remains; it tells us that the mosque was originally built of brick and was plastered over on several occasions.6 On approaching the mosque from the Maydan al-Azhar, which was laid out in the late nineteenth century, the north-western façade displays an amalgam of pseudoMamluk patterns fashionable at the turn of the century. Above this façade, three minarets and one pointed dome frame the main entrance. They are, from north to south, the minaret and dome of the Madrasa alAqbughawiyya (1339; rebuilt several times), the minaret of Qaytbay (1495) and the double-finial minaret of Qansuh al-Ghuri (1509) (fig. 22). The present main entrance of al-Azhar is the Bab al-Muzayinin (Gate of the Barbers), a double-arched portal built of stone with recessed arches surrounding the two doors and four panels of stone-cut, floriated ornaments with roundels in between. The gate is attributed to ‘Abd al-Rahman Katkhuda (d. 1776), and was constructed in the mid eighteenth century (fig. 23). The Bab al-Muzayinin opens into a rectangular marble-paved court flanked on the north-east by the main façade of the Madrasa al-Aqbughawiyya, rebuilt in 1888

–73 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 21. Aerial view of al-Azhar from the south

FIG. 22. The minarets of al-Aqbughawiyya, of Qaytbay, and of Qansuh al-Ghuri

– 74 –

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

FIG. 23. Bab al-Muzayinin (Gate of the Barbers) of ‘Abd al-Rahman Katkhuda

by the Khedive Tawfiq (1879–1892), and on the south-west by the polychrome façade of the Madrasa al-Taybarsiyya. The fourth side is occupied by the Gate of Qaytbay, built in 1495, with the splendid minaret above it to the right of centre. This gate leads to the court (sahn) of the mosque proper. The rectangular sahn, paved with white stone today, is surrounded by arcades on its four sides. The two lateral arcades are three aisles deep; the north-western, or entry, arcade is one aisle deep. The south-eastern arcade leads into the mosque’s sanctuary. Its central arch is a little advanced and surmounted by a higher, rectangular panel, forming a Persian framing gate (pishtaq), which visually marks the entrance to the sanctuary (haram) (fig. 24). The arches of the arcades are all keelshaped (what used to be called a Persian arch). They are composed of two-centred arches at the beginning of the curves which are joined together by means of two straight lines forming the tangents of the two arches.

They are credited to the Fatimid caliph alHafiz li-Din Allah (r. 1131–1149) who sponsored a major refurbishing of the mosque. Two types of stucco ornaments, which stylistically belong to al-Hafiz’s time but were redone in 1891, alternate along the walls’ surface above the arches. The first consists of shallow niches surmounted by a fluted hood on small, engaged columns and surounded by a band of Qur’anic inscription in a floriated Kufic script. The inscriptions were added to the original scheme at a later date, but still in the Fatimid period, and repaired in the late nineteenth century. The second type is a sunken roundel with a circular band of vegetal motifs (inserted in 1893) and twenty-four lobes. The roundels appear above the apex of each arch in the arcades. The upper cresting of the façades is made of an elaborate geometrical starshaped band topped with the usual triangular, tiered crenellations (fig. 25). The ceiling of the arcades is lower than the level of the stucco facing which is used as a screen,

–75 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 24. Main or south-eastern façade of al-Azhar Mosque’s haram

FIG. 25. Roundel and niche stucco ornaments above the arches of al-Azhar’s courtyard

giving the mosque’s interior façades greater monumentality. But even with the addition of this fake height the proportion of the elevation to the sahn’s width is still one to five. Behind the central south-eastern arch is a dome, attributed to the reconstruction of alHafiz, which rests on columns without any transitional zone. It is supported on four keel arches and elaborately decorated with stucco ornaments. The old haram of the mosque is oriented at 40 degrees south-east, which is slightly off the qibla (fig. 26). It is a great hypostyle hall, 85m by 24m (260ft by 75ft), consisting of five aisles made up of four arcades on marble columns running parallel to the qibla wall. They are intersected by a central transverse aisle running from the outer portico dome to the mihrab.7 Its arches are supported on double columns. All the columns in the hall are reused from old Roman, Coptic and perhaps even Pharaonic monuments. They are of various heights, brought to the same level by raising them –76 –

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

‘

FIG. 26. Plan of al-Azhar Mosque in the early twentieth century

–77–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

on bases of different thicknesses. The transverse aisle is made higher than the rest of the hall, probably to bring in light. The last arch of the transverse aisle is cut across by the first arcade of the haram to provide the four side supports for the dome in front of the mihrab, where two columns are added to provide more strength. Creswell dated this dome to Qansuh al-Ghuri’s time (1501) on stylistic grounds.8 The inner spandrels of the two lateral arcades of the transversal aisle are decorated with foliated stucco ornaments; the edges of the curves have Qur’anic inscription bands in Kufic. The mihrab is a round niche with a semi-dome and two marble columns flanking its arched frame. The stucco frame and inscription found there are probably Fatimid, belonging to the renovation of alHafiz, for the mihrab’s conch and frame had been covered by wooden boards for seven centuries, from the reign of al-Zahir Baybars until 1933. The interior of the mihrab was modified at an unknown date in the early Mamluk period, and a polychrome marble facing was installed there over what may have been an earlier stucco-carved decoration. A photograph taken in 1985 shows that the Mamluk facing, which appears in earlier illustrations of the mihrab, was removed at some unknown but recent date and replaced by the plain marble facing with small golden pattern seen there today. Three steps above this sanctuary and behind the mihrab area, which is the only part of the original qibla wall still standing, is the hall added by ‘Abd al-Rahman Katkhuda in 1751, and rebuilt by Khedive Tawfiq in 1888. It measures 69m by 20m (210ft by 60ft) and consists of four aisles formed by three arcades. Its mihrab axis is almost 10m (30ft) off the old axis. It has a transverse aisle dividing the longitudinal aisles, but its arches, unlike those of the old haram, are supported on piers. It, too, has a dome over its mihrab. On the eastern end of this hall, in the centre of the fourth aisle from the qibla wall, a door opens to the

Madrasa al-Jawhariyya. The qibla wall has a raised passage in its left side, half a metre high, which leads to the Mausoleum of Sitt Nafisa al-Bakriyya, a mystic who died around 1588.9 On the right is the Bab alShourba (the Soup Gate), built by the same ‘Abd al-Rahman Katkhuda in the style of the period, through which food for the students (usually rice soup) was brought for distribution along the raised passage. At the other, or western, end of the new hall stands the second main gate of alAzhar, the Bab al-Sa‘ayida (Gate of the Sa‘idis, after the people of Upper Egypt). It is similar in form and decoration to the Bab al-Muzayinin; it was built at the same time and by the same patron, ‘Abd al-Rahman Katkhuda. The other structures around the mosque are mostly arwiqa (pl. riwaq, used in the Azharite context to designate the residence halls for students) where the various ethnic or regional groups that formed the student body at al-Azhar lived. They were built at vairous times and by various sponsors.10 These structures are of minor architectural importance, though they may be relevant to a study of patronage and social status in pre-modern Egypt.

R ECONSTRUCTING THE FATIMID AL-A ZHAR Creswell meticulously and methodically reconstructed the original plan of al-Azhar in his magisterial work on the Islamic architecture of Egypt. He showed that it was a simple rectangle composed of the five-aisled prayer hall bordered on the north-west by a rectangular court (sahn), which was framed by porticoes on at least the two lateral sides (fig. 27). Creswell also proved that the keelshaped arches of the inner porticoes and the pishtaq and dome in the centre of the northwest façade were the work of al-Hafiz li-Din Allah. Sometime after 1125 al-Hafiz decided to unify the appearance of the mosque by surrounding its court with a continuous

–78 –

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

40

FIG. 27. Proposed reconstruction of the original plan of al-Azhar Mosque

portico and by emphasising the main axis leading to the mihrab through the addition of the dome and the pishtaq. Creswell was correct in most of his assertions and interpretations, although he was too eager to locate the architecture of the mosque in a primarily Egyptian tradition, which caused him to underestimate the influence of earlier Fatimid Ifriqiyan examples.11 Jonathan Bloom, in his thesis on early Fatimid architecture, identifies two aspects of Creswell’s reconstruction where his emphasis on the local model led him to interpret or fill the gaps in his data wrongly.12 Firstly, he overlooked the possible articulation of the external façade and the existence of a projecting portal in the original mosque, and secondly, he postulated the clerestory of the haram to have been a later modification, though no source reports such an addition. Neither Creswell nor Bloom include a minaret in the original scheme, nor do they discuss the location and function of the belvedere of al-Azhar (manzarat al-Azhar), though it formed the connection between the mosque and the caliphal Eastern Palace (al-Qasr al-Sharqi), the physical and political centre of Fatimid al-Qahira.

The earliest reference I found to a minaret at al-Azhar is a passing remark in a chronicle compiled by Ibn ‘Abd al-Zahir around the year 1280, which says that ‘the minaret of al-Azhar was increased in height during the time of the late judge (qadi) Sadr al-Din’; an individual who is otherwise unidentified.13 Two judges with this title were directly involved in al-Azhar’s affairs before the date of the chronicle of Ibn ‘Abd al-Zahir. The first is Sadr al-Din ‘Abd alMalik ibn Dirbass, the first Shafi‘ite judge of al-Qahira and al-Fustat (the old Islamic capital of Egypt) appointed by Salah al-Din al-Ayyubi, who is said to have banned the Friday prayer at al-Azhar around 1171.14 This makes him an unlikely candidate for a patron who rebuilt the minaret of a congregational mosque he had just downgraded to a neighbourhood mosque.15 The second is the qadi al-qudat Sadr al-Din Suleiman alAdhra‘i al-Dimashqi al-Hanafi (d. 1278), a Hanafite supreme judge whose school allows more than one Friday mosque in a city. He was a close consultant of al-Zahir Baybars; he accompanied him in several campaigns and was his tutor when Baybars went on his hajj in 1269. Baybars was apparently partial to him, since he authorised him to hold his court in any city of the sultanate.16 Baybers may have solicited his legal opinion ( fatwa) when he sought the agreement of the ulama about the restoration of al-Azhar to its former status as a Friday mosque, and may also have entrusted him with the repair of the minaret as part of the general programme of renovation undertaken by a number of high officials in 1266.17 This is all the more plausible as the mosque would have needed a higher minaret after its reinstitution as a Friday mosque. This reference proves that al-Azhar had a minaret by the end of the Ayyubid period. None of the Ayyubid rulers of Egypt could have built it, for they had neglected the mosque and closed it for Friday prayer. The brick minaret must therefore have been erected in the Fatimid period and probably

–79 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

was even included in the original scheme of the mosque. Most of the mosques built before al-Azhar in both Ifriqiya and Egypt that could have influenced its architecture had minarets.18 In Ifriqiya, the Mosque of Qayrawan (rebuilt in 836) had an imposing minaret facing the mihrab axis across the court. In al-Fustat, the Mosque of ‘Amru (c. 684)19 and the Mosque of Ibn Tulun (876–879) had minarets as well. The Ibn Tulun minaret was a prominent and original feature of the mosque that enhanced its character and connected it to Abbasid prototypes in Samarra. Another Fatimid mosque built in alFustat, the Mosque of al-Qarafa (built in 976 and since destroyed) had a minaret and was said to have been built on the model of al-Azhar Mosque, which implies that the original example must have had a minaret as well. The short text reporting the correspondence between the two mosques, found in al-Maqrizi’s Khitat, is typical of this famous medieval topographer’s tangled prose. It is presented in the form of a quotation from the Fatimid chronicler al-Quda‘i (d. 1061) who says that ‘the Mosque of alQarafa was built in the fashion of al-Azhar in al-Qahira. It had a lovely garden to the west and a cistern. Its door, from which it is entered, is a large one with mastabas. It is the central door under the high minaret built above it, and is covered with iron sheets [. . .]’ (wa huwa‘ala nahuw bina’ al-Jami‘ al-Azhar. Wa kana bi hadha al-jami‘ bustanun latifun fi gharbihi wa sahrijun. Wa babuhu alladhi yudkhalu minhu dhu al-masatib al-kabir alawsat taht al-manar al-‘ali alladhi ‘alayhi musaffahun bi-l-hadid [. . .]).20 The wording of this text needs reshuffling to be understood, but otherwise it is clear in stating that the Mosque of alQarafa had a minaret, and in the details of its location and height. If one accepts that the Mosque of al-Qarafa copied al-Azhar in every detail, then one must also accept that the original al-Azhar had a minaret, and that it too stood above the central portal

facing the axial transversal aisle of the haram, an arrangement used in many early mosques. The second text about the minaret, written around 1430, comes from al-Maqrizi; he says that the short minaret at al-Azhar was rebuilt by Barquq in 1397, but he does not mention its location or its date of construction, though a few lines down he offers further details that may help elucidate that point.21 In 1414, Barquq’s minaret started to lean dangerously and had to be torn down. An amir named Taj al-Din al-Shawbaki, who was the wali and muhtasib of al-Qahira, sponsored its rebuilding in stone and reconstructed the bahri (north-western or main) entrance gate, which supported it, with stone as well. Another rebuilding was undertaken in 1432, which lasted until 1495 when al-Ashraf Qaytbay (r. 1468–1496) rebuilt the mosque’s main entrance and erected his marvellous minaret on a massive base which formed the southern frame of the portal.22 This place is probably where the earlier Fatimid minaret of al-Azhar must have been built. It was demolished and rebuilt not only to correct defects in its construction but also because it was a brick tower and probably not sufficiently high to suit later Mamluk tastes. Al-Maqrizi’s pivotal text about the Mosque of al-Qarafa also states that there were mastabas flanking the gate. A mastaba is a low stone platform against a wall, usually used as a bench. In later Cairene practice, mastabas became ubiquitous features on portals, where two of them facing each other across the entrance space could be found in religious and residential structures alike. But no such arrangement is attested to in Fatimid al-Qahira or in Mahdiyya, the Ifriqiyan Fatimid capital before Egypt was occupied. Alexandre Lézine, who excavated Mahdiyya, reconstructed its tenth-century mosque façade with a projecting portal like a Roman triumphal arch with niches on either side of the central door.23 The same Mahdiyya scheme appears again in the

–80–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

Mosque of al-Hakim and in al-Aqmar Mosque (1125) in al-Qahira, which may be taken to indicate a continuous tradition of adapting the triumphal arch form to Fatimid monumental vocabulary.24 In order to reconcile the textual reference with the architectural one and to posit the portal of the Mosque of al-Qarafa in the same Mahdiyya pattern, the lower sills of the portal niches would have to be seen as mastabas to sit on.25 However, if the Mosque of alQarafa followed the Mahdiyya example, the more proper term to use for its niches in the text would have been kuwwa or haniya, two known words meaning niche.26 Clearly, the question cannot be solved with so many lacunae in our data, though one fact is definitive: the mastabas in question, whether they were niches or benches, would have required that the gate project from the wall in order to accommodate either the thickness of the niches or the width of the benches. The gate of the Mosque of al-Qarafa and consequently that of al-Azhar Mosque, were thus more articulated than Creswell admitted the latter to have been; a portal projecting out to accommodate the two flanking mastabas and a minaret of brick built above it. Another hypothesis about the outer appearance of the mosque which was advanced by Creswell and accepted by Bloom needs to be revised. This is the question of whether there were two small lateral domes; their presence was postulated by Creswell on the basis of a text by al-Maqrizi and the later appearance of two lateral domes on the Mosque of al-Hakim.27 AlMaqrizi says that ‘the foundation text of al-Azhar Mosque was inscribed by Jawhar al-Siqilli in 960 on the inner drum of the dome inside the haram in the first aisle to the right of the mihrab and the minbar’ (al-qubba allati fi al-riwaqi al-awwal wa hia ‘ala yumnati al-mihrab wa-l minbar).28 Creswell interpreted this text as referring to the lateral dome at the right end of the aisle, and supposed that it must have been balanced

by another dome at the other end of the aisle. If they ever existed, these two domes have left no trace, not even of the columns which were supposed to have been added for extra support as Creswell suggested. Besides, there are no precedents for this arrangement in the proposed models, either Ifriqiyan or Egyptian, of al-Azhar. The only known example comes from the Mosque of al-Hakim in al-Qahira, built after al-Azhar, which introduced a number of innovations like the twin minarets and probably the lateral domes. Another way of refuting this hypothesis stems from the significance of the text inscribed on the dome drum as reported by al-Maqrizi. Clearly, this was the foundation text for the mosque; it would have been an integral inscription with no conceivable equivalent or corresponding text occupying a parallel position in the same structure. It is very difficult indeed to propose what the contents of an inscription on the suggested other, left-side dome might have been to complement or echo the one on the right. It would be more plausible to suppose that the inscription band was on the initial central dome (rebuilt by Qansuh al-Ghuri in 1501), which was indeed in the first aisle, and which may be viewed as standing to the right of the minbar and mihrab. Such a location for a foundation text that identifies and honours the patron of the mosque would have been more prominent than on a lateral dome, which would have been seen only by someone standing at the end of a row of worshippers. Later Mamluk examples in Cairo, notably the Mosque of al-Nasir Muhammad at the citadel (built 1318 and 1335), have their foundation texts inscribed around the drums of their central domes. ‘Ali Pasha Mubarak locates the belvedere of al-Azhar next to the door of the Madrasa al-Jawhariyya. He states that the bahri (north-western) gate through which the caliph entered the mosque is still extant, but blocked, and quotes an earlier chronicler, al-Sakhawi, who wrote that the

– 81 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

Madrasa al-Jawhariyya was built near the secret gate (bab al-sirr) to the mosque on the bahri side.29 This ‘secret gate’ could not have been where the window of the Madrasa al-Jawhariyya is today, for this window was cut in the mosque’s wall when the madrasa was built and Jawhar al-Qanqaba’i, its patron, needed a fatwa from a judge to do it.30 The only other possibility for the location of the caliph’s gate is the door of the Riwaq al-Sharqawiyya (Hall of the Sharqiyya Natives), attributed to Muhammad ‘Ali alKabir (1780), which is adjacent to the northeastern wall of al-Jawhariyya. The door, however, has an inscription from the period of Qaytbay, which dates it to 1470, and the entire structure may be older. Creswell was puzzled by the two salients north-east of Qaytbay’s door. He dates them, unconvincingly, to the reigns of the khedives Tawfiq and ‘Abbas, having first proved that they were not meant to be buttresses of the wall since it was leaning forward and they were not. These two salients may be the bases that supported the belvedere of al-Azhar where the caliph and his wife sat to watch the festivities in the mosque on the nights of illuminations and on Fridays. The door attributed to Qaytbay should then be considered a renovated rather than a new door, replacing the original Fatimid gate of the caliph. AL-A ZHAR IN THE

FATIMID PERIOD

Jawhar established al-Mansuriyya when he first arrived in Egypt as an encampment (manakh) for the estimated 30,000 Maghribi soldiers (this is the lowest estimate) in his army and their families and animals and as a fortress and stronghold (hisnun wa ma‘qalun) in front of the capital al-Fustat on the road to Syria from where his enemies the Qarmatians were rumoured to plan an attack.31 The original manakh for Jawhar’s army may have occupied the area of the maydan that Kafur, who ruled Egypt as

regent for the young Ikhshidid ‘Ali and later as a prince shortly before the Fatimid conquest (961–968), had laid out next to his gardens, which were later incorporated into al-Qahira.32 Jawhar, following the tradition of early Islamic conquests in planning amsar, started the building of the qasr – which should be understood here as the walled enclosure that would later become the royal palace – at once and allocated khitat (pl. khitta, lot or quarter) for the various groups in the army to build their homes inside its walls; he forbade them to live in al-Fustat.33 He also built a musalla al-‘id, or open space for the two holiday prayers, just outside the walls to the north of the planned Bab alNasr.34 This may have been a practical decision. The Egyptian capital apparently lacked an open space at that time, although the Arabs had founded a musalla in it after the conquest, and Jawhar needed it in time to celebrate the prayer of the ‘Id al-Fitr, which fell in the month following his arrival.35 Unlike contemporary cities with a Sunni majority, where musallas had become a thing of the past, they appear to have been required in Fatimid cities, possibly as a sign of adherence to the prophetic custom of praying in the open on holidays and as another confirmation of the Fatimid claim to be the true descendants of the Prophet Muhammad through his daughter Fatima and his cousin ‘Ali.36 Jawhar, however, in a clear departure from the traditional layout of the amsar, does not seem to have included a congregational mosque in his plan for al-Mansuriyya. The Maghribi soldiers were able to perform their ‘id prayers at the musalla, but they had two choices for fulfilling their Friday prayer obligations: they could go either to the Mosque of Ibn Tulun (one mile south) or to the Mosque of ‘Amru (two miles to the south).37 For at least three years (Sha‘ban 358/July 969–Ramadan 361/June 972), they did just that, despite the antagonism of the Sunni Egyptians and their initial reluctance to accept the Shi‘ite prayer rituals in both

– 82–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

mosques. Jawhar’s policy was slowly but firmly to compel the unwilling khatibs and muezzins to institute the Shi‘ite adhan formula with the Isma‘ili addition which exalts the Fatimid caliph. This he achieved nine months after the conquest (Jamada I 359/ April 970). At the same time, he ordered the construction of a new congregational mosque in al-Mansuriyya, which was inaugurated the next Ramadan (22 June 971), almost two years after the establishment of the new Fatimid capital.38 The completion of the mosque must have come as a welcome relief for the Maghribi inhabitants of the city, providing them as it did not only with an accessible and enclosed space for prayer but also with a symbol of their doctrine’s ascendance and dominance that could stylistically and architecturally compete with the older congregational mosques in alFustat. Al-Maqrizi, who culled his account from the reports of at least four Fatimid historians of Cairo, does not offer any reason for Jawhar’s decision to built the mosque at that particular time, nor does he provide its original name. It is possible, however, that by then either al-Mansuriyya had developed into a fairly settled city that required its own congregational mosque, or that alMu‘izz informed his general of his imminent arrival and ordered him to prepare the new city for the occasion by building royal palaces and a mosque. Both explanations are hypothetical and difficult to verify, because of the dearth of Fatimid sources.39 However, whether either proposition is correct or not, the late date of the mosque’s building suggests that the Fatimids first intended their new settlement to be temporary or an extension of the existing capital and then changed their minds and decided to develop it into a fully fledged city, with the requisite congregational mosque. This shift probably occurred after Jawhar secured Egypt for his master and al-Mu‘izz decided to move there to pursue the Fatimid dream of winning the entire Islamic world to the Isma‘ili cause

from a base more central geographically, and closer to Abbasid Baghdad than his caliphal seat in Ifriqiya. The mosque’s original name is not known but, given earlier Islamic practices, it may have been the Jami‘ al-Mansuriyya. When the caliph al-Mu‘izz came to Egypt in 972 he changed the name of al-Mansuriyya to al-Qahira,40 and the mosque consequently acquired the name Jami‘ al-Qahira (the congregational Mosque of al-Qahira), which is the name we first encounter in Arabic sources.41 The mosque acquired its current name, al-Jami‘ al-Azhar, at an even later date, but no later than the end of the reign of al-‘Aziz (r. 975–996), the second Fatimid caliph in Egypt, for it appears in the waqf of his son al-Hakim, dated 400 (1009).42 Neither the origin of the name nor its intended association is quite clear. Scholars have proposed several explanations. The most widely accepted among them is that alazhar, which means ‘the magnificent’, is the masculine form of al-zahra’, the honorific title of Fatima, the prophet’s daughter and the eponym for the Fatimid dynasty.43 But this possibility is weakened by the belated appearance of the name – the Fatimids would have asserted their affiliation with Fatima al-Zahra’ from the start. Another explanation for the name alAzhar is that it derived from the nearby Fatimid palaces known as al-Qusur alZahira (the Brilliant Palaces) to which the mosque was spatially and ceremonially connected.44 The caliphal palace begun by Jawhar al-Siqilli before the arrival of his master al-Mu‘izz in Egypt developed into an enclosure of eleven royal structures known together as the Eastern Palace. The Eastern Palace was ultimately linked to alAzhar by a series of posterns, named alKuwakh al-Sab‘a (the Seven Posterns) although their number may have been less than seven.45 The posterns opened onto a pavilion named the Manzarat al-Azhar (Belvedere of al-Azhar) to which the caliph and his wife came from their palace to

– 83–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

watch the festivities on the nights of illuminations.46 The palaces of the caliph were not completed until some time during the reign of al-‘Aziz who finished the Eastern Palace and built the Western Palace complex opposite it, and who also renovated the mosque.47 He called the entire palatial complex alQusur al-Zahira and may have renamed the mosque after the palaces to show that both spaces essentially framed the caliph, who was considered to be an emanation of God’s light; he himself bore the epithet al-azhar among several that conveyed the idea of brilliance and resplendence.48 A similar phenomenon can be seen in the later proliferation of equivalent names for Fatimid mosques such as al-Anwar (the most luminous) for the Mosque of al-Hakim; al-Aqmar (the moonlit) for the small mosque adjacent to the northern corner of the Eastern Palace built by Ma’mun al-Bata’ihi, the vizier of alAmir (r. 1101–1130); al-Afkhar (the most splendid) for the mosque built by al-Zafir (1149), known today as the Mosque of alFakahaniyyin; and, most telling, for three belvederes built by al-Amir known as alZahira, al-Fakhira (the luxuriant) and alNadira (the exhilarating), suggesting that the names referred to the attributes of the caliph rather than the function of the structure.49 The close symbolic connection between the palaces and al-Azhar Mosque was evident in the spatial relationship between them and in the kind of functions assigned to the mosque. Throughout the Fatimid period, al-Azhar alternated with the Mosque of al-Hakim and the two older mosques of al-Fustat as the place where the caliph led the Friday prayers, competed with the Mosque of al-Hakim as the dynastic mosque of the Fatimids, and played a major ceremonial role in the programme for every caliphal procession.50 An analysis of the layout of Fatimid al-Qahira would corroborate the significance of al-Azhar to its planning and function. Though conjectural, a basic plan of alQahira’s earliest stages can be reconstructed

from textual sources, despite the fact that most of the city has been totally rebuilt since the time of the Fatimids and the caliphal palaces disappeared shortly after the fall of the dynasty. Such a plan was attempted by Paul Ravaisse more than a hundred years ago.51 Ravaisse painstakingly collected the bits and pieces in al-Maqrizi’s Khitat that pertain to the topography of the Fatimid city to produce his plan, but it still needs many revisions, for despite being the most complete source on Fatimid al-Qahira, al-Maqrizi’s text is misleading. He combines quotations from early and late Fatimid sources indiscriminately and without regard for the chronology of construction in the city.52 Thus, in examining the urban relationship between the Eastern Palace and alAzhar Mosque the successive planning of the area and the dates of the few structures in it have to be taken into consideration. In conflating the plan of the royal centre of al-Qahira and the chronology of building, it becomes clear that the area was developed according to a ceremonialreligious theme attached to the person of the caliph (fig. 28). This must have happened early in the Fatimid period, probably during the caliphate of al-‘Aziz who enlarged and monumentalised the palatial complex, and, with the help of his genius vizier, Ya‘qub ibn Kilis, shaped the Fatimid ceremonial to enhance his semi-sacred image during the 970s and 980s.53 Ravaisse’s plan shows the palatial complex as it may have appeared at the end of the Fatimid period with the two major units, the Eastern Palace started by Jawhar for alMu‘izz and completed by al-‘Aziz and the Western Palace which the latter ordered built, and the huge parade ground between them which received the convenient toponym Bayn al-Qasryn (Between the Two Palaces). The complex had acquired its general contour by the end of al-‘Aziz’s reign. Later caliphs implemented new minor structures, such as the al-Aqmar Mosque built by alAmir in 1125, or modified existing ones

–84–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

FIG. 28. Plan of Fatimid al-Qahira

–85–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

without any noticeable alteration to the overall spatial and urban arrangement of the entire complex. The Eastern Palace had two gates on its southwestern side, the Gate of the Turbat al-Za‘faran, and, to the east of it, the Gate of al-Daylam. The space between the two gates was originally the palace cemetery, Turbat al-Za‘faran (Cemetery of Saffron), where the Fatimid caliphs and their families were buried, beginning in the reign of alMu‘izz. Al-Mu‘izz even brought with him the coffins of his three predecessors from Ifriqiya for burial there, probably as a tangible mark of continuity of the imamate in the line of the Fatimids. This cemetery played a major role in instituting the cult of the Fatimid family, for the caliphs visited it after every religious procession and distributed alms there.54 To the south of the cemetery were the palace storehouses (khaza’in alqasr), which were apparently built during the reigns of the first two caliphs, al-Mu‘izz and al-‘Aziz. The next gate of the palace complex to the east, the Gate of Qasr al-Shawk, was pierced in the eastern façade and opened onto the court named after it.55 The court was bordered on the south by Dar al-Fitra, built by al-‘Aziz, where the caliph distributed sweets and victuals to his soldiers and courtiers in religious ceremonies.56 To the west of Dar al-Fitra was the Istabl al-Tarima (Rotunda Stables), where the caliph’s horses were kept. (Ravaisse locates the Istabl alTarima to the east of Dar al-Fitra, but alMaqrizi clearly specifies that the stables were adjacent to the Maydan al-Azhar and the seven posterns, which means that they were to the west of Dar al-Fitra.) South of Dar al-Fitra and Istabl al-Tarima stood the Maydan al-Azhar where the caliphal procession from the palace to the mosque on Fridays and holidays culminated. This court must have been open when al-Azhar was first built, but it was encroached upon by the later additions of al-‘Aziz and al-Hakim.

The seven posterns or the seven private doors (depending on how the word khukha is interpreted) must have led from between the gates of Turbat al-Za‘faran and alDaylam on the south-eastern side of the Eastern Palace to this court in a passage paralleling the Istabl al-Tarima.57 Al-Maqrizi’s descriptions of Fatimid Cairo’s layout reveal the ceremonial function in the siting of al-Azhar, its relation to the Eastern Palace and surrounding lesser structures, its architecture and furnishings, and even the monumental articulation of the mosque gates (the mosque had at least three gates), which served as points of entry for the various processions going into the mosque. A glance at the waqf attributed to the Caliph al-Hakim (1009) for the benefit of al-Azhar, the Dar al-‘Ilm, which he established, and two other mosques shows that the first received more than half the shares of the deed (167 out of 300 shares).58 This reflects the importance of al-Azhar in the ceremonial programme of al-Hakim, who provided it with an income equivalent to that of the three other caliphal structures combined, even though he had completed his mosque shortly before as the new congregational mosque of the royal city.59 The waqf shows also that most of the allocations for al-Azhar went to support ritual and ceremonial items, such as silver lanterns, incense, camphor, musk, candles and so on. Around the same time, a new subject – religious education – was introduced at alAzhar. From the scattered references to its early years, the mosque seems to have had a modest debut as a learning centre. The first seminar in Isma‘ili jurisprudence ( fiqh) took place there in 976, four years after the mosque’s inauguration. The supreme judge (qadi al-qudat) ‘Ali b. al-Nu‘man led the seminar in the mosque in a way common to any congregational mosque at the time. He read al-Ikhtisar, the book compiled by his father al-Qadi al-Nu‘man b. Muhammad, the fore-

–86–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

most Fatimid theologian of the reign of alMu‘izz. The seminar differed from the usual only in the selection of the judge’s audience. They were chosen from the upper echelons of Isma‘ili Fatimid officialdom and from the royal family, to the exclusion of Egyptians; they included women who had had their names written down and had drawn lots for the right to attend.60 The next recorded instance of teaching at al-Azhar occurred in 988 when the crafty vizier, Abu al-Faraj Ya‘qub ibn Kilis, started reading a religious book he had compiled from the sayings of the two imams and caliphs, al-‘Aziz and his father alMu‘izz, to a select audience once every fortnight. Later in the same year, Ya‘qub ibn Kilis asked the caliph al-‘Aziz to sponsor regular teaching in the mosque every Friday after the prayer by a group of thirty-five jurists, of whom Ibn ‘Abd al-Zahir reports only the name of the leader, a certain Abu Ya‘qub, known as qadi al-khandaq (qadi of the ditch, an area south of al-Qahira).61 The caliph even built a house for these jurists next to the mosque, but they nonetheless moved their lessons to the Mosque of alHakim after its completion, suggesting that al-Azhar at the beginning was apparently not meant to be a centre of religious learning.62 It was used by famous scholars to deliver their lessons in its haram, but that was a common practice in congregational mosques everywhere. A few years later, the mosque attained a new role: it was made the centre for teaching and propagation (markaz al-da‘wa) of the official Isma‘ili creed, where new missionaries were trained before they were sent out to the various regions of the Islamic world. Soon afterwards, it vied for the position of supreme foundation of Fatimid propaganda with the more specialised institution, Dar al‘Ilm or Dar al-Hikma, founded by the third caliph, al-Hakim bi Amr Allah (r. 996–1021), in 1005.63 Dar al-Hikma ultimately became the official headquarters of Isma‘ili teaching

and jurisprudence with instruction in rhetoric, logic and philosophy, while al-Azhar retained its role as a centre of religious learning with few sectarian leanings. This was a status al-Azhar shared with all other congregational mosques in the joint Fatimid capital of al-Qahira and al-Fustat – the Mosque of ‘Amru, the Mosque of Ibn Tulun and the Mosque of al-Hakim – perhaps because most Egyptians remained Sunnis under the Fatimid regime. When al-Azhar became the single most prominent centre of Islamic teaching in al-Qahira and when its reputation spread to the rest of the world is not quite clear. It could not have happened during the Fatimid period, since the mosque was still patronised mainly by the Isma‘ili ruling class and could not have attained a universally acknowledged status in primarily Sunni Egypt and the Islamic world beyond. Nor could this efflorescence have occurred during the Ayyubid period, for the mosque was then relegated to minor status and was closed to Friday prayer through the late twelfth and early thirteenth centuries. It was not reinstated and refurbished until 1266, during the reign of al-Zahir Baybars. The mosque had acquired considerable popularity in the fourteenth century, and may first have attained its eminent position as the foremost centre of learning in Egypt around the year 1490. This hypothesis is based primarily on the architectural evidence, since the sources do not speak of how or when it became a famous institution, but only report a renewed and avid interest in its preservation and monumentalisation. This started as a series of restorations and additions in and around it after its partial destruction during the earthquake of 1302, and culminated in the major royal interventions undertaken in the last quarter of the Mamluk period, especially under the two late and great sultans, al-Ashraf Qaybay and Qansuh al-Ghuri.

– 87–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

THE PERIOD OF NEGLECT The teaching enterprise at al-Azhar and the seemingly impartial treatment of all Islamic schools of thought as pedagogical subjects may have saved it after the fall of the Fatimids and the abolition of the Isma‘ili da‘wa. Shi‘ite doctrines were banned in Egypt by the Ayyubids, and Sunni teaching in all its branches was officially implemented and promoted. Madrasas, the new institutional type which promulgated the official Ayyubid brand of Islamic Sunni teaching, were constructed all over Cairo and endowed with hefty waqfs. Al-Azhar does not seem to have received any Ayyubid financial support, probably because of its former association with the Fatimids or because of its intractable independence, but it was at least left alone to function as a place of learning. It ultimately grew in vitality and in prestige under the Mamluks and Ottomans to become the leading centre of Sunni studies. Sheltering what amounted to a religious university remained its main purpose until the present day, in addition to its more obvious function as a congregational mosque, a status it lost for 100 years under the Ayyubids. Salah al-Din al-Ayyubi eliminated the Fatimid caliphate and reinstated Sunni Islam as the state religion in Egypt in 1171. He founded two madrasas, one in al-Qahira for the Malikites and one in Fustat for the Shafi‘ites, and abolished the Isma‘ili teaching in al-Azhar Mosque. He dismissed the Isma‘ili supreme judge and appointed in his place Sadr al-Din ibn Dirbass, a fellow Kurd and Shafi‘ite qadi, who proceeded to ban Friday prayer at al-Azhar ostensibly because the Shafi‘i school of jurisprudence, to which Salah al-Din adhered, prohibited the holding of more than one Friday sermon in each community. Inside the walls of al-Qahira, the Friday prayer was held only in the Mosque of al-Hakim because it was considered larger than al-Azhar.64 However, this explanation is shaky, for we do not have a

similar report on the two mosques of Ibn Tulun and ‘Amru, which by that time belonged to the same urban unit, Misr alFustat, and which both seem to have remained functional during the Ayyubid period. Furthermore, even if we admit that the Shafi‘ite opinion dominated, the choice of al-Hakim as the congregational mosque is not fully justified. It is located on the edge of the city, while al-Azhar is central and easily accessible from all quarters.65 The real reason behind the demotion of al-Azhar from a royal mosque to a minor role in the communal life of the city must have stemmed from Salah al-Din’s attempts to erase the memories attached to the grandeur of his predecessors and, most of all, their religious heritage and claims. During the rest of the Ayyubid period, al-Azhar at least maintained one aspect of its function as a major mosque; it remained a teaching institution. The famous Baghdadi doctor, ‘Abd al-Latif al-Baghdadi, taught there between 1192 and 1198.66 Musa ibn Maymun (Maimonides), the physician of Salah al-Din, also gave some lectures there in medicine and astronomy. But this activity must have been modest in the context of Ayyubid Cairo, for the members of the ruling family were more interested in sponsoring madrasas. Introduced into Egypt shortly before the fall of the Fatimids, madrasas became the main vehicle during the Ayyubid period for the preparation of a new class of Sunni jurists to staff the administration. This preparation had two goals: to eradicate the Isma‘ili doctrine from the ruling class and to diminish the Christian presence in the government, which was very strong during the late Fatimid period.67 Contrary to the officially regulated madrasas, whose curriculum and budget were supervised by the Ayyubid patrons and sponsors, al-Azhar seems to have remained relatively independent of both interference in the type of teaching taking place in it and official financing. The lack of royal patronage, however, may have contributed to the decline in its reputa-

–88–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

tion and its physical maintenance, as evidenced by the emphasis on its sorry state in the sources of the early Mamluk period.

THE RISE TO PROMINENCE Friday prayer was restored to the mosque in 1266 during the reign of al-Zahir Baybars after one of his major amirs, ‘Izz al-Din Aydamar al-Hilli, interceded and convinced him to authorise its restitution. He was reportedly moved by piety, for he lived near the mosque and was outraged by the miserable state into which it had sunk after almost 100 years of neglect.68 However, the spirit of the times must have greatly facilitated his task. Al-Azhar’s association with the Fatimid regime had become only a dim memory; reopening it to Friday prayer was thus politically safe. It was also a popular move and may have been intended as another sign of the coming of age of the Mamluk sultanate under Baybars as an autonomous entity rather than just an inept imitator of the Ayyubid state.69 Moreover, al-Qahira’s population was growing so rapidly that mosque space for Friday prayer was greatly needed. Aydamar sponsored some repair work at al-Azhar that put it back in functioning condition. He and Baybars’s vicegerent, Baylabak al-Khazindar, each constructed a maqsura in the mosque intended for study gatherings and endowed it with waqfs for the sponsorship of Qur’anic and legal studies. In 1302, during the second reign of alNasir Muhammad (1298–1308), a powerful earthquake struck the Mamluk territories, destroying many large structures in Cairo and cracking many minarets, including those of al-Azhar, al-Hakim, ‘Amru and the dome of al-Mansur Qalawun in Bayn alQasryn. Al-Nasir Muhammad and his amirs divided up the damaged mosques, and took it upon themselves to rebuild them, with alNasir reserving for himself the restoration of his father’s complex and minaret.70

Al-Azhar’s reconstruction was undertaken by Amir Salar, the atabek al-‘asakir (the head of the army). This, however, was part of a general effort on the part of the ruling magnates to alleviate the damage caused by the earthquake to the major mosques and madrasas in Cairo. In fact, between the refurbishing effected under Baybars in 1266 and the rebuilding after the earthquake, alAzhar was left alone amid the frantic construction taking place all over Cairo in the early Mamluk period. Teaching and praying went on in it as usual, while new and stately madrasas and khanqahs, sponsored by sultans and amirs, were being erected around the city and in its adjacent cemeteries, alQarafa al-Kubra and al-Qarafa al-Sughra, to the east and south. Nonetheless, al-Azhar ended up indirectly benefiting from the steady rise of Cairo’s reputation as the foremost centre of Islamic learning, as the city’s geographical and political importance grew, especially after the fall of Abbasid Baghdad in 1258 and the consolidation of a powerful Mamluk empire in Egypt and Syria.71 The flurry of madrasa building in Cairo, shortly after 1302, led to the next step in al-Azhar’s architectural history. Two madrasas, the Madrasa al-Taybarsiyya (1309) and the Madrasa al-Aqbughawiyya (1339), were built against the exterior wall of the mosque on opposite sides right outside the main gate (see fig. 26).72 Taybars al-Waziri, the founder of the Taybarsiyya, stipulated that his structure was to function as a complimentary mosque (ziyada) to alAzhar and a Shafi‘ite madrasa. AlAqbughawiyya, which stood on the site of the house of Aydamar al-Hilli, al-Azhar’s first benefactor in the Mamluk period, was instituted by its founder, Aqbugha ‘Abd alWahid, as a mosque and a madrasa as well. It was designed by Ibn al-Siyufi, the master architect during al-Nasir Muhammad’s reign, and had a dome and an elegant stone minaret attached to it which still survives today. Another small structure, the Madrasa al-Jawhariyya, was built in 1440 along the

– 89 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

north-eastern wall of the mosque by the amir Jawhar al-Qanqaba’i, supervisor of the privy purse for Sultan Barsbay, for which he cut a door into the wall of the haram.73 This situation, unparalleled in Cairo but encountered in illustrious mosques elsewhere such as the Haram of Mecca and the Great Mosque in Isfahan,74 is probably a reflection of the importance of al-Azhar both as a central structure and as a venerated sanctuary, so that patrons would wish to build next to it while other large congregational mosques in the city were left alone.75 However, although the three structures were significant architectural additions to al-Azhar’s surroundings and contiguous with its outer walls, they were clearly and conspicuously meant to be physically and functionally separate. They stood outside the mosque proper and were entered through separate doors, though they provided more space for congregational prayer, the first explicitly as a ziyada, the second as an independent mosque with its own minaret and its own adhan, and the third as a small oratory (masjid). The three structures also served as independent madrasas right outside the door and in full view of this prestigious centre of learning. Mamluk historians treated them as independent institutions and listed them separately. Until today, each remains physically detached from the mosque and known after its founder, unlike the earlier maqsuras of the amirs of al-Zahir Baybars built inside the mosque, which seem to have disappeared soon after their institution. This change in the pattern of patronage at al-Azhar from the time of Baybars to that of al-Nasir Muhammad reflects the harshly competitive milieu Mamluk amirs lived in at that time, where building charitable institutions and attaching their names to them had become a prerequisite for amirial status. Up until that time, no Mamluk sultan had shown any interest in sponsoring any architectural work at alAzhar. Only amirs built in and around it,

and even those were of the second rank. Taybars was the chief of staff in Egypt (naqib al-juyush) for twenty years (1298–1319), a position subordinate to that of the atabek. Aqbugha was the master of the household (ustadar) of al-Nasir Muhammad, a powerful position, for it provided easy access to the sultan, but it was also second rank in the Mamluk hierarchy. Jawhar, who died in 1440, was head of the household eunuchs under Barsbay, obviously a position lower than that of the great amirs.76 It was not until the fifteenth century that direct Mamluk royal patronage was extended to al-Azhar. Three powerful sultans – Barquq, Qaytbay and Qansuh alGhuri – patronised al-Azhar during the Mamluk Circassian period. In 1397, al-Zahir Barquq, the first Circassian sultan, rebuilt the minaret and spent 15,000 dirhams of his own money in the undertaking. His minaret had to be reconstructed again in 1414 during the reign of al-Mu’ayyad Shaykh, this time in stone, as we have seen. But even this minaret did not last for long. It started leaning dangerously and had to be dismantled and rebuilt anew in 1423, during the reign of Barsbay, though we do not know if he himself was the patron.77 Al-Ashraf Qaytbay intervened at the mosque at least three times during his long reign (1468–1496), in 1469, 1476 and 1495. He razed a number of parasitical shacks, repaired the porticoes, rebuilt the main entry, the Bab alMuzayinin, and erected a new stone minaret on top of its arched opening.78 He also built at least two riwaqs along the periphery of the mosque to lodge the Turkish and Syrian students. Qaytbay’s minaret, with its shaft typically divided in three sections and topped with a bulb, is among the most exquisitely carved and masterfully proportioned late Mamluk examples. The effect of this extraordinary minaret apparently instigated Qansuh al-Ghuri to order a new minaret built for al-Azhar in 1509 to the south-west of Qaytbay’s minaret.79 However, the inability of his architect to surpass

–90–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

Qaytbay’s minaret may account for the unusual, huge and double-finial minaret he erected (see fig. 22). It is important to note that the Mamluk sultans chose minarets over any other addition; they either built them directly or through some amir, probably acting as their agent. Minarets were the ultimate symbol of power and dominion and the most effective proclamation of the buildings and their patrons in Cairo’s cityscape.80 Sultans and great amirs in the Mamluk period always wanted to emphasise the size and location of the minarets in their complexes as a means of setting it off from others. The competition is very obvious in the case of the last two minaret builders at al-Azhar – Qaytbay and al-Ghuri – for the latter did not demolish the minaret of his predecessor but simply tried to outdo it by building a new one next to it that was larger, taller and had two heads rather than one (in vain it has to be admitted, for his minaret lacks the grace and balance of Qaytbay’s). The succession of minarets built at al-Azhar, where liturgically one would have been enough, shows that these sultans sought a highly visible association with this specific mosque, which had by then become a most eminent university, attended by scholars from all over the Islamic world. The terminus ad quem proposed earlier to mark the mosque’s rise to prominence as a religious university is based on the royal architectural intervention under Qaytbay and al-Ghuri. These two sultans bestowed their sponsorship on the mosque after it had become the primary institution of learning in Cairo, in recognition of this status. This proposition diverges from the generally held opinion that al-Azhar, venerated as it was as an old and somewhat independent mosque, did not play a prominent role as a centre of teaching during the Mamluk period, primarily because of the profusion of generously endowed royal and amirial madrasas.81 An analysis of the biographical data available for the members of the Egyptian civilian

elite who died in Cairo between 1397 and 1495 shows that, indeed, al-Azhar did not occupy the first place in many of the classified categories: place of birth, of education, and of occupation. But it is significant to note that, although its competitors were extremely well-endowed royal madrasas, such as al-Zahriyya Madrasa (of al-Zahir Baybars) and al-Mu’ayyadiyya (of alMu’ayyad Shaykh), it did consistently come in second, and in a few significant categories – notably among people coming from the Delta and from North Africa – it was an emphatic first.82 There is no comparable analysis to prove that the situation rapidly changed in favour of al-Azhar at the end of the Mamluk period that could explain the sudden royal patronage, but another study – admittedly limited to one source – strongly supports such a conclusion. The analysis of ‘Abd al-Wahab al-Sha‘rani’s famous biographical compendium, Lawaqih al-Anwar fi Tabaqat al-Akhyar, showed that one third of the Egyptian ulama mentioned in it had either studied or taught at al-Azhar.83 AlSha‘rani (1491–1565), a famous Sufi and scholar whose life bridged the Mamluk and Ottoman period, compiled his book before 1554. This means that his data reflects the situation at the end of Mamluk times, when most of his ulama must have received their education, rather than the early Ottoman period.84 Royal patronage was abruptly interrupted after the fall of the Mamluk sultanate to the Ottoman sultan Selim I in 1517. Selim acknowledged al-Azhar’s status by attending Friday prayer there in his last week in Cairo, but he did not donate anything to the mosque.85 Ottoman walis (governors), who were appointed from Istanbul, all regularly attended Friday prayer there and many presented the mosque with cash gifts for its ceremonies and as stipends for its ulama. However, only a handful of them ever contributed to its upkeep, and none added any significant structure to it, probably because their terms of office were very

– 91 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

brief and generally too unstable to encourage extensive patronage. In fact, until the beginning of the eighteenth century, only two walis – Muhammad Pasha al-Sharif (1596–1598) and Hasan Pasha (1605–1607) – are known to have restored al-Azhar during their governorships. Hasan Pasha also constructed the riwaq of the Hanafis (the Hanafi mazhab was officially sanctioned by the Ottomans) in 1605.86 Nor did the local Mamluk beys (the title assumed by the local Mamluk grandees in place of the old title amir) contribute much to the maintenance of the mosque/ university throughout the sixteenth and seventeenth centuries. Some did designate a share of their waqfs for some charitable donation to the institution such as the sponsorship of students, but it was not until the eighteenth century that a number of great amirs took it upon themselves to restore alAzhar and to add new structures to its periphery. This two-century gap in patronage is primarily a consequence of the decline in power and wealth the Mamluk class suffered after the Ottoman conquest. After their devastating defeat in 1517, Mamluk survivors strove to restructure their households and recover their power base under a new regime in which they did not hold sovereign authority over the country. After an initial period of adjustment, Mamluk households began to emerge as the de facto power brokers in the city, each in control of its own urban area, while the overall administration was in the hand of the pashas who resided at the citadel with their Ottoman troops. For a long time Mamluk beys focused their attention on neighbourhood improvement projects, which were at the same time profitable; they established commercial foundations, restored religious complexes and, in many instances, added sabil/kuttab units to them.87 By the beginning of the eighteenth century, the hold of the central Ottoman authority was weakened enough to permit the Mamluk class to interfere more force-

fully in the governing of Egypt, with the pashas sent from Istanbul playing the role of arbitrators in power struggles among the dominant households. A number of Mamluk beys who had accumulated wealth and power began to act as true leaders and as benefactors of the city at large, restoring and renovating many of its most famous monuments.88 Al-Azhar was one such major Cairene landmark in dire need of repair. The mosque was resting on its accumulated glory and continuous fame, but its state of preservation and its capacity had become inadequate for its role as the prime educational centre in the Islamic world. In 1735, Amir ‘Uthman Katkhuda al-Qazdughli, the leader of a dominant Mamluk household, established there a hall for the blind, who constituted a large component of alAzhar’s student body and were usually destined to become qurra’ (Qur’an reciters) and muezzins. He also built – or more accurately rebuilt – the two riwaqs attributed to Qaytbay, the Turkish and the Syrian riwaq. ‘Abd al-Rahman al-Jabarti, the main source for Cairo’s history in the eighteenth and nineteenth centuries, explains that ‘Uthman Katkhuda craftily amassed huge sums of money from the unclaimed inheritances he collected after the plague of 1735, which permitted him not only to sponsor the riwaqs at al-Azhar, but also to construct a new mosque in alAzbakiyya, the new posh residential quarter where most beys resided.89 However, it was not until the middle of the eighteenth century that al-Azhar received its first major facelift since the fall of the Mamluk sultanate. This was done by Amir ‘Abd al-Rahman Katkhuda, the son of ‘Uthman Katkhuda’s master Hasan Katkhuda, who was called sahib al-‘ama’ir (the master of building) because of the large number of buildings he sponsored in Cairo.90 Between 1746 and 1763, he renovated most of the major shrines in the city and as a patron single-handedly altered al-Azhar more than anyone else since Jawhar al-Siqilli first built

–92–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

it. He began in the 1740s by building two gates – Bab al-Muzayinin and Bab alSa‘ayida – with sabil/kuttabs and minarets and then bought a number of properties south of the mosque to expand its prayer hall.91 After his appointment as katkhuda (head of the Janissaries) in 1749, ‘Abd alRahman embarked on a large-scale refurbishing and enlargement programme at al-Azhar. He added a new prayer hall south of the original one, so large that it doubled the surface area of the haram, and built or refurbished a number of dependencies and residence halls around the mosque that put it back in functioning order and accommodated the rising number of students. He also tried to give the mosque a unified urban aspect by delimiting its enclosure and regulating the circulation in and out of it through the three new gates he had constructed – the two early ones and the Bab al-Shourba (Soup Gate) – and by creating

a new main façade through remodelling the minaret of the Madrasa al-Aqbughawiyya to make it look like the two Ottoman minarets he had built (fig. 29).92 Finally, ‘Abd al-Rahman Katkhuda acquired great honour by building himself a mausoleum in this most prominent religious institution where he was buried in 1776, an act that has not been done for almost two centuries before him (Nafisa al-Bakriyya was interred in her mausoleum there around 1588), and never after him.93

MODERNISATION AND ITS AFTERMATH Al-Azhar’s role as the centre of resistance in Cairo intensified during the brief but influential French occupation (1798–1801). Despite a mutually rather beneficial first encounter between al-Azhar’s ulama and Napoleon Bonaparte, which resulted in the

FIG. 29. The four minarets of al-Azhar including the two Ottoman minarets of ‘Abd al-Rahman Katkhuda at Bab al-Muzayinin and al-Aqbughawiyya

–93–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

formation of a diwan made up of nine Azharite shaykhs to govern Cairo, the popular uprising on 21 October 1798 started from the mosque and ended with its bombardment, temporary closing and the disbanding of the diwan (which eventually was twice reinstituted, with Azharite shaykhs forming a sizable portion of its members). After the slaying of General Kleber by an Azharite, Sulayman al-Halabi, in March 1800, the mosque was closed for good and its entrances bolted until the return of the Ottomans with the help of the British in August 1801.94 The French occupation indirectly had two effects on al-Azhar that were initially promising but ultimately detrimental to maintaining its pedagogical and social dominance in Cairo. The involvement of the Azharite shaykhs in the diwans set up by the French and in the Cairene revolts made them more aware of the political potential their position conferred upon them.95 They tried to put their newly discovered capacity to effective use after the departure of the French. Under the energetic ‘Umar Makram, naqib al-ashraf (head of the descendants of the Prophet), they became uncommonly involved in the political struggle that finally brought Muhammad ‘Ali (1805–1848), the founder of modern Egypt and of the Khedival dynasty, to power. However, as the ulama were reluctant to go all the way in assuming political power and the canny Muhammad ‘Ali was unwilling to share with them his hard-won authority, their alliance was short-lived. In 1809, Muhammad ‘Ali dealt a heavy blow to their hesitant attempt at organising themselves by imposing a tax on and ultimately confiscating iltizam (taxfarmed) agricultural land, which cut deeply into their financial base, and exiled their leader and his erstwhile ally ‘Umar Makram to Damietta.96 This was the first among a series of actions undertaken by Muhammad ‘Ali and his descendants that, over the next century, curtailed al-Azhar’s independence and subordinated its ulama and students to the government’s structure.

For centuries, al-Azhar, as a centre of learning, had lagged behind the times. It had neglected the natural sciences and dealt only with the time-honoured religious sciences, such as Qur’anic exegesis, hadith and jurisprudence. Even in these fields, the teaching had generally deteriorated into pedantic exercises in verbal sophistries and scholasticism.97 This state of affairs did not change because of the French expedition, despite the repeated assertion that the work of the French was the catalyst of modernisation in the country.98 Some of the experiments pursued by the French savants engaged in collecting and studying all the available material on Egypt did in fact pique the curiosity of a few ulama but was totally incomprehensible to the majority among them. Those few who realised the limits of their scholarly enterprise, steeped as it was in outdated medieval tradition, urged reform at al-Azhar, but their proclamations went unheeded.99 It was not until two generations later that change was introduced to al-Azhar, promoted and facilitated by the same early advocates of change. Muhammad ‘Ali did not have the patience to wait until the Azharites came around to accepting change. Being a practical man who had just embarked on a programme of intense economic, social and military restructuring after Western models, he soon established a new school system, which was independent from al-Azhar, to train the new cadres for his army and administration. He also sent a number of students on academic missions to France to acquire a higher education. The graduates of this new system were clearly destined to occupy the choice positions in the new governmental structure, thus excluding the ulama, but this does not seem to have been an intentional replacement. In fact, although al-Azhar did not receive any direct support from the pasha’s educational reforms, Muhammad ‘Ali needed to hire Azharite teachers, who were the most educated in Egypt, to fill the positions in his new schools

–94–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

where Arabic was the language of instruction. He also sent some Azharites with his missions to Europe, among whom Rifa‘a alTahtawi (1801–1873), who was the religious tutor of the 1826 mission, was to become one of Egypt’s leading intellectuals and reformers.100 Starting with the Khedive Isma‘il, Muhammad ‘Ali’s grandson and Cairo’s moderniser, the mosque was again patronised by the ruling family. Isma‘il rebuilt ‘Abd al-Rahman Katkhuda’s Bab al-Sa‘ayida, renovated the additional structures attached to it, such as the sabil and maktab, and restored the Madrasa al-Aqbaghawiyya.101 Khedive Tawfiq renovated the prayer hall added by ‘Abd al-Rahman Katkhuda, aligned the south-east façade behind it with the street, remodelled the façades of the Madrasa al-Aqbughawiyya and built a new ablution area along the north-eastern façade of the mosque. Khedive ‘Abbas Hilmi II (r. 1892–1914) refurbished the main façade of the mosque and built his own Riwaq al‘Abbasi in a neo-Mamluk style along the eastern end of that façade (completed in 1901) (fig. 30). He also constructed a threestorey façade in front of the original southwestern façade of the mosque with several riwaqs tucked in between the two walls. During the same period, the Committee for the Conservation of Monuments of Arab Art (known by its French abbreviation, Comité), the mostly European agency responsible for the architectural conservation of Islamic monuments in Egypt, sponsored the restoration of the old Fatimid sanctuary, which resulted in some of the discrepancies noted by Creswell, such as the addition of a circular band of vegetal motifs in the sunken roundels above the arcades in 1893, and the alteration of some of the contents of the Qur’anic inscriptions.102 Three reasons can be advanced to explain the official architectural sponsorship during the reign of the three successive khedives, the last two of whom were neither distinguished nor inspired rulers. The first

FIG. 30. Portal of the Riwaq al-‘Abbasi

was the straightforward requirement to maintain the building to ensure the adequate functioning of this major pedagogical institution. The second was the desire of Isma‘il, and to a lesser extent his two successors, Tawfiq and ‘Abbas Hilmi II, to modernise, regularise and urbanise this central site in their capital. The task became more urgent with the opening of al-Azhar Street and Maydan al-Azhar in 1890–92; the entire northern façade needed to be aligned with the street, and the minaret of ‘Abd alRahman Katkhuda had to be demolished since it stood alone outside the Bab alMuzayinin. The third was the effort to win the acceptance and boost the popularity of the rulers among the Azharites and the general population by endowing this respected centre of learning with visible additions that would celebrate and perpetuate their names. The same religious policy lies behind the restoration of popular shrines and the launching of grand mosques

–95 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

projects – the Rifa‘i Mosque (1869–1880 and 1906–1911), the Sayyida Zaynab Mosque (1885) and the Sayyida Nafisa Mosque (1897) – by the same khedives to express their social and cultural commitment to the country.103 Khedival architectural patronage, however, was preceded and accompanied by an increasing interference into the institutional independence of al-Azhar and its methods of education. Sometimes shrewdly and sometimes crudely, al-Azhar has been used by successive regimes to achieve their own ends by sanctioning decisions and making them acceptable to a people that still trust the authority of this revered institution; a kind of political manipulation that has persisted until the present time. But the main impetus for intervention in al-Azhar’s internal affairs has, ever since then, been the drive for modernisation (tajdid)104 – and its inherent component Westernisation – the overarching goal pursued by various parties in the recent history of Egypt. The status of education at al-Azhar itself became the battleground for the two sides of the issue – the modernisers and the traditionalists.105 From the reign of Isma‘il to the revolution of 1952 and Gamal ‘Abd alNasser’s republic after that, rulers generally supported modernisation for political reasons. However, within al-Azhar itself, homegrown reformers sought to put the institution on a par with contemporary educational establishments so it could maintain its status as the first centre of learning in the country.106 The manoeuvering on both sides resulted in a succession of organising laws, the first of which was issued under Isma‘il in 1872 by the shaykh al-Azhar (rector) Muhammad Mahdi al-‘Abbasi, which sought to regulate the appointment of teachers at the institution and organise the examinations for its degrees, but did not attempt to reform its curriculum. During the reign of ‘Abbas Hilmi II, under the British occupation, laws were promulgated in 1895, 1896, 1908 and 1911. ‘Abbas Hilmi had, in Shaykh Muhammad ‘Abdu (1849–1905), an Azharite

revolutionary reformer and one of the most outspoken critics of conservative systems of thought, an ally to organise the administration and educational policy of al-Azhar.107 The laws of 1908 and 1911 were passed when Sa‘d Zaghlul, the leader of the Egyptian revolution against the British occupation in 1919, was minister of education.108 Under King Fu’ad (r. 1917–1936), two complimentary pieces of legislation were passed in 1930 and 1936 that attempted to arrange al-Azhar along the same lines as the newly instituted secular university. Another law was passed under the republican reform in 1954 and, finally, a sweeping reform was forced upon al-Azhar in the law of 1961.109 For the architectural history of the mosque, the last four acts are the most important. Starting with the law of 1930, alAzhar was rearranged as a modern university and split into three departments – of shari‘a, Arabic language and theology. In 1932, these three new departments were accommodated in three new buildings scattered around Cairo.110 In 1935–36, al-Azhar administration was moved into a new building located across from the mosque on Azhar street and designed by Ahmad Charmi using an eclectic Islamic style which resonates well with the concoction of styles represented at al-Azhar itself.111 A number of buildings for the primary and secondary institutes and a medical complex were constructed to the north of it. Between 1950 and 1955, several properties around alAzhar were bought and their structures demolished to make room for the creation of a modern campus. Two structures in a bland neo-Islamic style were added near the mosque to house the faculties of shari‘a and Arabic language, which were moved from their old buildings. A residential complex for international students was then constructed in an open space in ‘Abbassiya, to the north-east of old Cairo, thus extending the university of al-Azhar away from the immediate surroundings of the mosque.112 The pedagogical apparatus of al-Azhar

–96–

A L -A Z H A R MOSQU E : A N A RCH I T ECT U R A L CH RON ICL E

itself was reduced and eventually totally eclipsed when all of its teaching functions had been transferred to the new structures. Today, al-Azhar still stands in the heart of the old city of Cairo. Its sahn is no longer a place where turbaned, robed barely bearded students would sit or pace back and forth while memorising their antiquated treatises. Nor are its porticoes and aisles any longer the settings for halaqas (study circles). Some students of the new Azhar University went there every now and then to review their lessons, either to keep the old tradition alive or to find a quiet space. The content of the mosque’s only authentic and self-contained Fatimid inscriptions echoes its present situation. Creswell proved that most of the beautiful writings inside the mosque and around its courtyard were redone after 1891 as part of the renovation and restoration programme undertaken by the Comité under ‘Abbas Hilmi II. He has shown that many of these inscriptions were not always repaired according to the original plan, nor were they put back in their original locations.113 Consequently, no iconographic programme could be deduced from the extant bits and pieces since their present arrangement is only 100 years old.114 Two lines in the conch of the original mihrab, which were uncovered in 1933, form the only complete and sure Fatimid inscriptions in the mosque. They contain two

Qur’anic quotations. The first includes the first three verses of the Sura of the Believers (23). ‘Prosperous are the believers – who are humble in their prayer – and who turn away from idle talk.’ The second comprises the verses 162–3 of the Sura of the Cattle (6). ‘Say my prayer, my piety, my living and my dying all belong to God the Lord of all beings – no associate has He and thus I have been commanded, and I am the first of the Muslims.’ Both inscriptions establish a straightforward reference to the essential function of the space, namely prayer, and both were used in mosques around the Islamic world about the same time.115 Their exposition around the time of the 1930 law and its 1936 supplement, which reorganised the new Azhar University and practically dislodged it from the old mosque to the newly built schools to the east, was so timely as almost to be prophetic. The mosque functions today only as its name implies: as a house of prayer for the throngs of students in the new university, merchants and inhabitants of the crowded quarters around it, and as an attraction for the scores of visitors who tour it in search of its old glory or to marvel at the pell-mell architectural styles confined within its walls. This monument, however, is not only an important part of the rich heritage of Cairo but also a gauge for detecting major trends in its political, religious and social history during its first millennium.

–97–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

–98–

8. DOCUMENTING BUILDINGS IN THE WAQF SYSTEM

B

uildings are a peculiar brand of cultural artefact. They are both time- and energy-consuming, and require complicated planning, financing and a multitude of skills to build. Over time, they acquire multiple functions, meanings and associations that may or may not conform to the intentions of their original builders. But they have one consistent physical trait: they are usually anchored in place (save for temporary exhibition pavilions and the like, which can be disassembled). To look at buildings, the viewer must either see them in situ, or rely on illustration, description or representation. Thus, documenting buildings has been essential to conveying architectural knowledge; it involves a continuous process of refinement and calibration, on the one hand, and of deciding which of many distinct modes of documentation to use, on the other. Generally speaking, three modes of documentation exist: verbal, graphic and representational. The first includes any description of a building, whether oral or written. The second comprises all the graphic methods that aim at representing the various aspects of a building: its footprint or plan, its facades, cross-sections and three-dimensional representations, but not

its direct depiction. These usually involve some kind of codified depiction of a certain real or constituted spatial impression, one that requires specialised knowledge for applying the code on the part of the person who makes the drawings as well as the person who deciphers them. The third mode, which I refer to as ‘representational’, encompasses all the ways in which a building is represented as nearly as possible to its actual appearance. These include all pictorial, perspectival and photographic techniques from the simplest sketches to the most sophisticated photogrammetry and perspective-correcting photography, to computer rendering systems and, of course, modelling. These three modes of architectural documentation occupy a historical continuum but do not succeed one another in chronological order. We do not know exactly when and where each appeared, but we suspect that some form of representation preceded verbal description as a means of documenting buildings at the dawn of history. This is at least what some cave paintings, with their putative depictions of tent structures, seem to suggest. The three modes have at times coexisted in a certain place or tradition, as they do now all over

–101 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

the world. However, this has not always been the case. They sometimes existed in pairs or alone, and they sometimes disappeared from a certain place only to be replaced by another mode or to reappear at a later time. The choice of one or another mode depends primarily, but not absolutely, on technical skills and technological advances. Culture also plays an important role in the choice, especially where the technical skills exist that will allow all three modes to be used. In these cases, the dominance of a certain mode can be seen as a reflection of a cultural preference or a cultural attitude (and by culture, I mean both architectural culture and culture in the broad sense). Some cultures have favoured the visual over the verbal while others have opted for the verbal over the visual regardless of technological capacity. Those choices have always been a reflection of and operative in particular cultures. In the Western tradition, for example, one might consider the drawings of a Borromini or a Bernini as a kind of apogee of the graphic expression of architecture whose visual power supersedes the need for any textual description. By contrast, medieval Islamic cultures in general developed a highly complex system for the verbal description of buildings, all the while eschewing the need for graphic representation. Verbal description was adopted in a wide range of written genres: encyclopedic compendia, geographical treatises (masalik), topographical tracts (khitat), travel and pilgrimage literature (kutub al-ziyarat), chancery and taxation manuals (kutub al-kharaj) and cartography (surat al-ard). These flourished in the Islamic world from around the ninth to the early sixteenth century; by 1530 images of the urban fabric began to appear at the Ottoman court. The dependence on language is made most manifest in the legal domain, where waqf or endowment deeds used verbal description – without any graphic represen-

tation whatsoever – as the exclusive means of documenting buildings for property and appraisal purposes. The institution of the waqf is an old and venerable Islamic legalfiscal system for organising charity, social services, and the management and inheritance of real estate and agricultural land.1 Waqf is believed to have first appeared in the days of the Prophet Muhammad when he endowed a certain orchard for all Muslims, which implies that the term and the practice might be pre-Islamic; although this is impossible to verify given the sources at our disposal. By the medieval period, the waqf system had spread all over the Islamic world and beyond. It had also developed a language and a procedure for documenting buildings that satisfied contractual and legal requirements, and that reflected both an interest in the purely socioeconomic dimensions of architecture and a specific vision of the role of buildings in urban space.2 A waqf deed is a written document that lists a series of properties to be endowed in perpetuity for a specific charitable or personal purpose. This included the construction and maintenance of a religious or educational building, the support of a group of people devoted to a certain function or related to the waqf provider, and the provisions for filling the needs of a particular institution. Documenting a building in a waqf usually begins with recording its surroundings – the other buildings, streets and urban artefacts facing or abutting it in all directions. This set the boundaries of the building and framed it within its urban context. Then came the sequential description of individual spaces in the building as they are seen by a person walking though it. The description ordinarily begins with the entrance and then moves in a set direction enumerating the various aspects and features of each interior space. In most cases, the description covers an entire level before moving to the upper levels. The individual descriptions pay more attention

–10 2 –

D O C U M E N T I N G BU I L D I N G S I N T H E WAQ F S Y S T E M

to circulation – focusing on windows, doors, and the specific functions of spatial zones – than they do to appearance. With the exception of listing particularly expensive building materials or indicating that a certain ornamental surface was made by using a certain technique, physical or spatial qualities are not mentioned in the waqf description. Yet waqf authors were definitely engaged in documenting buildings. They brought to the verbal description of architecture a new level of sophistication in which even formulaic expressions carried specific connotations. These not only affected the value and desirability of the structure being described, but captured in words what the culture in general deemed important in buildings. Such codified expressions also revealed the architectural, urban and, ultimately, sociocultural preferences and biases of the authors. In their eyes, the status of the building as a visible and static whole was much less important than its experiential quality or its functional capacity, two attributes that can be verbally transmitted very well. A building, moreover, was never seen as a separate object: it could only make sense as a component in an urban context or in the landscape, probably a reflection of the prevalent forms of dense layout in the city. Otherwise, hardly any description of an urban façade can be found in waqf documents. Only the location of entrances and position of minarets were noted, emphasising the link between the public space of the street and the building proper. Even interior spaces were seen in the context of their connectivity and functionality and never as an abstract arrangement of spaces or volumes. Their architectural characteristics were, in sum, never noticed except to indicate how they were accessed and whether or not they

had built-in usable spaces, such as recesses, niches and alcoves. One of the pressing questions here is whether cultural proclivity dictated the technique of documentation (i.e. verbal versus graphic), or whether technical limitations imposed on a culture a mode of documentation. The answer is far from settled, to be sure, and cannot be definitive. Yet Islamic cultures used graphic documentation in many other fields such as mathematics, pharmacology, botany, zoology and even entertainment. That these cultures did not choose to use it in architectural documentation is therefore hardly the result of technical incapacity. My guess is that the choice had to do primarily with the division of labour in hierarchised early modern societies, where buildings and writing about buildings were the separate domains of two different social groups that did not easily and systematically communicate with each other. Builders belonged to modest social classes and building trades were relegated to the artisanal realm, with very little historical or intellectual interest shown in their pursuits.3 Describing buildings was left to the ‘men of the pen’, who belonged to the upper echelons of society and hardly acknowledged the builders or their language and means of representation. When they had to document buildings either for the waqf documents or in other documents, these ‘men of the pen’ did so in their usual medium of expression, which was rhetorical rather than graphical (at which they had no reason to be skilled and with which they were traditionally unfamiliar). I suspect that it is only in places and times where builders were also speakers for and interpreters of buildings that we see the dominance of graphic over verbal modes of documentation.

–10 3 –

9. THE IWANS OF THE MADR ASA OF SULTAN HASAN

I

n his book A History of Architecture, Spiro Kostof defined the word iwan as ‘an Islamic vaulted hall, open at one end’.1 However, Islamic medieval sources reveal a wider range of meanings for the same word. Three interrelated architectural denotations can be identified: a large alcove, a raised portion of a reception hall’s floor, or a whole palace.2 These three meanings appear to connote one underlying concept, monumentality, both in its formal and commemorative functions. Aesthetically, the image of iwan seems to be related to the concept of grandeur. Symbolically, the iwan itself seems to convey, in most cases, a memorial value as the place of honour in a structure.

THE ETYMOLOGY OF THE WORD IWAN The Arabic word iwan is borrowed from Persian. Arabic lexica, starting with Ibn Durayd’s al Jamhara fi-l-Lugha, give the word’s meaning as a ‘lofty suffa’ which is a canopy or a marquee, or as ‘a construction resembling an open faced azj’, which is a longitudinal structure. In every lexicon until the present day, the word is mentioned in connection with the legendary Iwan Kisra

(Arabic for Khosroe, the Sasanian king) (fig. 31). These persistent references suggest that this particular monument might have represented the archetype for iwans in the medieval collective memory, although very few people really saw the famed iwan. In other words, Iwan Kisra might have offered the source for the architectural denotations, as well as the memorial connotations of the word iwan. This may be demonstrated by examining some examples from the medieval Arabic poetry in which the word occurs. Poetic examples are countless in which Iwan Kisra is evoked as a symbol of grandeur. Fewer are those that deal with it in any specific formal meaning, and still less are those that combine the two modes. One of the few cases in point is a distich from a poem by the Baghdadi poet, Muhtar al Daylami (early 10th century): And my father Kisra upon his Iwan Who in the world has a father like mine?3

There is here an obvious link between the place of high honour, i.e. the seat of Kisra the great king, and the iwan. Nonetheless, the spatial configuration of the iwan as having a raised floor can be deduced from the use of the word upon ‘ala in Arabic.

–10 4 –

T H E I WA N S O F T H E M A D R A S A O F S U LTA N H A S A N

FIG. 31. Façade of the Iwan Kisra: elevation and in a photograph taken before the flood of 1883

Furthermore, Iwan Kisra becomes part of a monumental category, or what could be termed as a grand ceremonial tradition in medieval Arabic poetry. Fabled and real palaces, such as those of Palmyra and the High Pillars of Iram belong to the same group. In a poem by Abu Sa‘id al-Rustumi dedicated to al Mu‘tamid bin ‘Abbad, we read: By raising it, you devalued Kisra bin Hurmuz’s Iwan Which became worthless in the land of Ctesiphon If the High Pillars of Iram were to see its supports

Their heights would tumble down in shame And if the Heavens of Palmyra glimpsed its beauty They would learn how stone edifices ought to be built.’4

Eulogists used to invoke a variation of this class of buildings whenever they needed to extol the oeuvre of their patrons, which they included in the same esteemed list. The regular usage of Iwan Kisra in this particular context may have reinforced the monumental connotation of the word iwan itself.

–10 5 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

Another observation pertains more to the possible medieval understanding of the word. In some poetical instances, al Iwan was used as a proper name referring to the specific palace, such as in a verse from an eulogy by Ibn Hamdis the Sicilian, from the mid eleventh century: If the Palace’s beauty was to be compared to that of al Iwan The latter would have been found lacking.5

However, despite this later usage of the word, it is well documented that the architectural form, known as iwan, was adopted early on in Islamic palatial architecture. Al-Istakhri, the tenth-century geographer, described the Palace of Abu Muslim al Khurasani in Merv in Eastern Iran, built before 760, as having ‘four iwans each commanding a square court’, and when he reported on Iwan Kisra, he distinguished its form as being ‘arched’ (fig. 32).6 Thus, by comparing the textual and poetical uses of the word iwan it can be said that, by the eleventh century, the word has had two simultaneous meanings. The first FIG. 32. Reconstructed plan of the Palace of Abu Muslim al Khurasani in Merv

was the proper name of an architecturally little-understood referent, but a commemoratively highly charged one. The second was a reference to an architectural type, possibly copied from its model Iwan Kisra, yet not necessarily carrying any memorial value. We know now that the word eventually dropped the direct reference to Iwan Kisra and maintained the ‘monumentality’ connotation, in addition to the architectural significance. The question that arises then is were there any actual referents in which the two modes of meanings were merged? In other words, were there any Islamic monuments in which iwans were used to convey monumentality by their sheer forms without any direct reference to Iwan Kisra? The answer is yes, but we have to wait until the fourteenth century to start seeing new monuments that would take the place occupied by Iwan Kisra in the Muslims’ collective memory. One of the major structures that belong to this category is the Madrasa of Sultan Hasan in Cairo, built between 1356 and 1361 (fig. 33). However, this madrasa is not the only such monument. Other contemporary structures display the same understanding in the use of iwans in their spatial organisation, such as the Mosque of ‘Ali Shah in Tabriz. What should be investigated here is whether these buildings were seen in their own time as monumental by virtue of their iwans. FIG. 33. Aerial view of the Madrasa of Sultan Hasan

–10 6 –

T H E I WA N S O F T H E M A D R A S A O F S U LTA N H A S A N

Recently, it has been noticed that the Islamic artistic sensitivity was ready, by the fourteenth century, to define a style of the time and to identify contemporary monuments with it.7 For our purpose, this translates into the possibility of extracting some notion of aesthetic and architectural standards from our primary sources, for when dealing with documents written before the fourteenth century this exercise seems difficult. Thus, when reading a passage that relates the story of the inception of the Madrasa of Sultan Hasan, a new awareness of its relevance to our interpretation is necessary. Khalil al Zahiri wrote in the mid fifteenth century: ‘As for the Madrasa of Sultan Hasan, this edifice has no equivalent in the whole world. It was reported that Sultan Hasan, when he ordered its construction, summoned all the architects (muhandisin) from all the countries and asked them: which is the highest building in the world? He was told: Iwan Kisra anu Shirwan. So he ordered that the Iwan should be measured and revised ( yuharrar) and that his Madrasa should be 10 cubits higher than it, and it was constructed [. . .] Iwan Kisra has but one iwan, this Madrasa has four!’8 In this anecdote, Iwan Kisra is clearly the model of the proposed madrasa. But what is more important for our analysis is that iwan here is typologically understood; it is not the monument anymore, it rather becomes the monumental arched opening. This is probably the image of Iwan Kisra that was adopted for the Madrasa of Sultan Hasan.

rets are so imposing that Sultan Selim the Grim is said to have exclaimed upon seeing it in 1517, ‘this is a great hissar (castle)’.9 In addition, the madrasa manifests an unusual treatment in its exterior facades, which, unlike those of prior Mamluk madrasas, seem to have been planned for maximum monumental effect. This exterior monumentality is echoed internally in the sheer size of the four unequal iwans, arranged in a cross pattern around the central courtyard (fig. 34). All four iwans have raised floors above the court level. The southern iwan is the largest one, and was designated as the congregational mosque space. It has a raised platform (dikka), situated on its axis, a mihrab, and a minbar against its qibli wall (facing Mecca)

FIG. 34. Plan of the Madrasa of Sultan Hasan

THE M ADRASA OF SULTAN HASAN The Madrasa of Sultan Hasan is not an isolated phenomenon. It belongs to a sequence of Mamluk religious buildings, though it is architecturally the most impressive among them. From the outside, its walls, the huge portal, the dome and the two extant mina–107–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

(fig. 35). In the four corners of the cross plan are tucked four units with similar spatial organisation and varying sizes. They constitute the four Sunni schools’ madrasas, with the largest one for the Hanafis in the south-east corner, and the smallest, truncated one for the Hanbalis in the north-west. Each of these units is architecturally planned around a small courtyard with an iwan of its own facing north. It is obvious that the Madrasa of Sultan Hasan belongs to what came to be defined as a four-iwan plan type. It could be argued, however, that in its use of the iwan unit to emphasise monumentality, this madrasa differs from its preceding and contemporaneous four-iwan madrasas and khanqahs. To explain this characteristic, three possible causal factors could be proposed: historical, evolutionary or typological, and referential.

The historical one has to do with the aftermath of the 1348–9 bubonic plague, which ravaged the Mamluk sultanate, and especially the city of Cairo. It has been suggested that the concentration of inheritance wealth in the hands of the ruling Mamluk elite, the need for an urban reordering and refilling in the city, and especially the expression of religious piety on the part of the disaster’s survivors all prompted the magnificent growth in endowing and constructing religious institutions in Cairo.10 The Madrasa of Sultan Hasan may be considered the climactic representation of this tendency. The typological justification views this madrasa in the context of the development of four-iwan plan institutions in Cairo. Sultans endowed madrasas in Cairo ever since Salah al-Din al-Ayyubi sponsored few madrasas in the city. Those madrasas

FIG. 35. Interior view of the Madrasa of Sultan Hasan

–10 8 –

T H E I WA N S O F T H E M A D R A S A O F S U LTA N H A S A N

have disappeared, and the earliest royal madrasa still remaining is the Salihiyya, built by al-Salih Najm al-Din Ayyub in 1248. It was composed of two separate, selfcontained courtyard units, parallel in plan, and each having two large iwans (fig. 36). Al-Maqrizi calls them the two madrasas, and specifies that each iwan was allotted to one school’s teaching.11 The first Mamluk royal madrasa is the one built by al-Zahir Baybars, al-Madrasa al-Zahiriyya, in the year 1262/3. This madrasa is now destroyed and its plan was not recorded. However, the disposition of its iwans can be reconstructed from al-Maqrizi’s account. There were four iwans arranged around the courtyard: two housed the Shafi‘i and Maliki schools, whereas the other two accommodated classes in Hadith and in the Seven Readings of the Qur’an (fig. 37).12 In the next royal madrasa, al-Nasiriyya, of al-Nasir Muhammad

(1295–1303), the four iwans that served the four schools are arranged in a cross around the courtyard (fig. 38). The Madrasa of Sultan Hasan was designated as a four-school-madrasa.13 Moreover, its act of endowment (waqf ) postulates that the four iwans be used as a congregational mosque: a new usage for the four-iwan plan in Egypt. Also, the Madrasa of Sultan Hasan was planned to house half of its fuqaha’ students, around 300 men, unlike most previous madrasas, with the obvious exception of the Madrasa of Amir Sarghatmish (1356), which was considered in many ways its prototype (fig. 9.9). In the Madrasa of Sultan Hasan, the students’ cells are arranged in groups around the four peripheral courtyards on three storeys. By stretching vertically to accommodate all the students’ cells, the four iwans could take up virtually all the space around the central

FIG. 36. Plan of the Madrasa of al-Salih Najm al-Din Ayyub

–10 9 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 39. Plan of the Madrasa of Amir Sarghatmish

FIG. 37. Plan of the Madrasa of Sultan al-Zahir Baybars

courtyard, thus resulting in the tightly organised, visually as well as spatially, edifice.

M ADRASA AND IWAN FIG. 38. Plan of the Madrasa of al-Nasir Muhammad

The critical question here is, did the necessity to accommodate the congregational mosque and to lodge a large number of students, faculty and servants, in addition to the compositional limitations of the fouriwan plan create this upward thrust, or vice versa? In other words, was the massiveness of the madrasa the result of a ‘functionalist’ approach, or did the intended monumentality provide the designers with a possible vehicle, evidently well utilised, to contain the required functions? These questions lead to the third causal factor alluded to earlier to explain the singularity of the Madrasa of Sultan Hasan, namely the referential relationship between the madrasa and Iwan Kisra. This explanation clearly stresses the monumental intention over the functional one in interpreting the madrasa’s building. In a similar – 11 0 –

T H E I WA N S O F T H E M A D R A S A O F S U LTA N H A S A N

context, it was noticed that Mamluk architecture in general displays a striving for effect that could be identified as its ‘expressive intent’.14 Mamluk religious architecture, in particular, embodies a tension between its expressive intent – that is, its political and memorial messages carried out by its forms – and the religious requirements of its functions. In this analysis of the Madrasa of Sultan Hasan, it is suggested that the iwan as an architectural type loaded with formal connections to a popularly glorious monument of the far past, and more immediate connotations of an institution, the madrasa, which ennobles the religious inclination of the masses and honours its patrons, alleviated, or at least lessened, the ‘tension’ between the function and the intended message of the madrasa. Part of this role is accomplished on a subtle level by the inherent geometric quality of the iwan itself, that is, the obvious reciprocal relationship between the span of the iwan and its height. In Mamluk architecture, where all arches are either semicircular or fifth-point arches (Makhmus), the height could never be less than half the width of the iwan’s opening. Thus, to accommodate a congregational prayer in an iwan space, it was functionally necessary to widen the arch span as much as possible, and therefore elevate its roof. This was further emphasised in the Madrasa of Sultan Hasan, unlike some other four-iwan Mamluk madrasas, by raising the vertical sides of the iwans; which induces the conclusion that at least part of the interior immensity of the four iwans is due to a deliberate decision, and perhaps – although this could be considered as applying present-day artistic language – an aesthetic imperative.

THE FOURTEENTH-CENTURY MEANING OF THE WORD IWAN The iwans of the Madrasa of Sultan Hasan may be said to epitomise the semantic field of the word iwan as it was probably understood in the fourteenth century. This was certainly enhanced by the mere fact that the madrasa belongs to a historical period noted for its monumental architectural production all over the Islamic world, on the one hand, and the psychological, social, ceremonial and liturgical functions implied in the iwans themselves on the other hand. We have today lost at least two of the connotations evoked by the fourteenthcentury usage of the word: the liturgical and the specifically memorial ones. The former was lost because mosques are no longer built in iwan form. The latter was lost simply because Iwan Kisra is no longer the loftiest building in the world: an image that was very much alive in the fourteenth century. However, the iwans of the Madrasa of Sultan Hasan, along with other contemporary monuments, created new typological paradigms, which in turn preserved the monumental connotation of the word iwan and separated it from its association with Iwan Kisra. After all, the huge iwans of the fourteenth century were true products of their time for, as al-Maqrizi tells us, quoting an eleventh-century Andalusian poet:

– 111 –

When Kings wish to be remembered for their deeds They rely on the tongues of the monuments The grander an edifice is The more it indicates the greatness of its patron.15

10. QASR: AN AGENT OF MONUMENTALITY IN MAMLUK ARCHITECTURE

T

his chapter takes as its point of departure the proposition that medieval sources that directly or indirectly deal with architecture and decoration have not been thoroughly and appropriately used despite the many modern studies devoted to the explanation and analysis of Seljuq, Ayyubid, Ilkhanid, Mamluk and other important medieval architectural corpora. The inadequate use of primary sources goes beyond the normal and expected omission of some relevant accounts or the misreading of others to a basic attitude that survived until very recently in the study of Islamic architecture in general. This approach limited use of the written sources to searching for mention of building locations, dates of construction, patrons, reasons for building, and passing remarks about specific spaces or features. It failed to see in the seemingly haphazard, disconnected or repetitive uses of terms in medieval texts a reflection of the conceptions and views of the writers and, therefore, interpretable historical references. In the last thirty years, scholarship has begun to pay more attention to context through a better understanding of the terms used in medieval sources. Nowhere has this

approach proved to be more necessary, and at the same time more rewarding, than in studies dealing with material culture, especially architecture. By its nature, the field requires two basic elements to reconstruct the history of any architectural object: the physical remains of the object itself and contemporary documents that at least permit its identification and verify its attribution and function. This is the first and direct result of textual analysis and precise interpretation of terms. An appropriate terminological enquiry then allows us to penetrate behind the objects themselves, their dates and apparent use into the minds and manners of the people who named them, and to investigate the aesthetic, social and cultural factors that informed the processes of naming, and the later alterations of original names, or their initial meanings. In Egypt, scholars such as ‘Abd alLatif Ibrahim, Mona Zakarya, Laila ‘Ali Ibrahim, Muhammad Muhammad Amin, Alexandre Lézine, Jean-Claude Garcin, JeanCharles Depaule, Donald Little, Michael Rogers, Doris Behrens-Abouseif, Salih Lam‘i Mustafa, M. Hussam al-Din ‘Abd-al-Fattah and Hazem Sayed, have studied several kinds of Mamluk sources – chronicles, bio-

– 11 2 –

QASR: A N AGE N T OF MON U M E N TA L I T Y

graphical compendia, topographical tracts or khitat, and especially waqf documents – to propose new readings for some architectural terms and consequently to change many perceptions that were based on incorrect or anachronistic interpretations of these same terms.1 However, a number of construction and architectural terms that occur in Mamluk sources, especially in relation to palaces and residences, still elude strict definition. One such important term is qasr. Today it is used to mean palaces everywhere and at any time, but this does not seem to have been the case in the early, so-called Bahri, Mamluk period (1250–1382), when the word qasr had a specific and restricted meaning. That particular meaning appears to have been intentionally dropped in the Burji period (1382–1517) when both the social and political contexts changed, new architectural elements were introduced and novel ways of living were adopted. Not all Bahri Mamluk palaces were called qasr in contemporary sources. They were sometimes referred to as qasr and sometimes as something else, such as dar or bayt or, more often, qa‘a. The only exception to this seeming variation in usage is the Ablaq or Striped Palace (al-Qasr al-Ablaq, the palace acquired its name from its exterior walls built of successive courses of black and ochre stone called ablaq in Arabic), constructed at the Citadel of Cairo in 1313–14 by the sultan al-Nasir Muhammad ibn Qalawun, which is always called qasr in all Mamluk sources and whose units are constantly called qusur as well.2 The textual descriptions of the nowvanished Ablaq Palace provide the first clue for what the Bahri Mamluks meant by the word qasr. The earliest and probably most accurate account is that of Ibn Fadl-Allah al-‘Umari who, as a high-ranking administrator and a secretary of the sultan, was an eyewitness to what he describes, and whose report, recorded in 1325, was copied almost

verbatim by later chroniclers such as al-Maqrizi and al-Qalqashandi. Al-‘Umari writes:

– 11 3 –

On the side of the Great Iwan there is a passageway to the door of the Qasr alAblaq followed by a small court where the close amirs sit before they enter to the qasr for the service. From the door of the qasr one passes through corridors to a monumental qasr of splendid construction with two iwans, the larger being the northern [north-west], which overlooks the stables of the sultan, and from which one can see the horse market, Cairo and its suburbs as far as the Nile, and beyond to Giza and its villages.The second or qibla iwan [south-east] has a special door [al-Qalqashandi calls it the secret door] for the exit of the Sultan and his courtiers to the Great Iwan on the days of ceremonies. From this qasr one can enter three inner palaces (al-qusur al-juwwaniyya), of which one is on the same level of the first palace, whereas the other two are reached by a staircase. All these qusur have windows with iron grilles, whence the view is the same as the principal qasr. The qusur aljuwwaniyya communicates with the inner part (haram) of the harem, and the Abwab al-Sutur (Gates of the Veils). The facades of all these qusur are built of black and yellow stones, and within are dadoes of marble and gold and floriated mosaics, heightened with mother of pearl and colored paste and different colors. The ceilings are all gilded and painted with lapis lazuli. The light comes through windows filled with colored glass from Cyprus resembling necklaces of precious stones. All the floors are paved with marble transported from all the countries of the world, which has no equal. One can descend from the side of the iwan of the qasr to the stables of the sultan, then to a maydan covered with grass, which is so spacious that the eye travels in it. This maydan lies between the stables and horse market to its west

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

[north-west]. The sultan mounts his horse from a staircase next to his qasrahu al-juwwani, and he descends to his private stable, then to the maydan with the great amirs in his service.3

The main component of the Qasr al-Ablaq, the throne hall, had a qa‘a plan, with two unequal iwans and a durqa‘a in the middle topped with a dome. The large north-western iwan overlooked the sultan’s stables, the maydan, and Cairo beyond; the south-eastern one led to the private door through which the sultan and his retinue entered the Great Iwan on official occasions. The three other units, which had qa‘a plans with an iwan at either end as well, had a similar disposition to benefit from the same view over the city (fig. 40). The whole palace was connected on one end with the Great Iwan and on the other with the rest of the sultan’s palaces (al-addur al-sultaniyya), where his wives and concubines lived. Al-‘Umari says that the stairs that descended to the private stables

of the sultan were located next to al-Nasir Muhammad’s inner palace (sg. qasrahu aljuwwani), which may be taken to mean the last of the series of three inner palaces. The door to these stairs was in the passageway (dihliz) of the palace, which should be taken to mean the corridor connecting the Ablaq Palace to the Great Iwan. Clearly, the word qasr is used both to designate separate halls in a group and the whole structure that comprises all these units. The three other units were collectively called the inner qusur, but there is some later evidence to suggest that they had individual names.4 Is this dual use of the term a mistake caused by other sources copying the initial report of al-‘Umari? Or is it a special application of the term to mark the most important royal palace in the Bahri Mamluk period? Or does it stem from a Bahri Mamluk use of the word that restricts its meaning to a certain type of structure? If so, what is that type? A review of the four surviving princely palaces of the

FIG. 40. Schematic plan of the four palaces of al-Qasr al-Ablaq

– 114 –

QASR: A N AGE N T OF MON U M E N TA L I T Y

Bahri period in Cairo will allow us to answer these questions.

PRINCELY BAHRI PALACES The four Bahri Mamluk princely palaces still standing in Cairo today are those of Alin Aq (1293), Bashtak (1334–39), Qawsun (1337) and Taz (1352).5 The names refer to their original owners, all Bahri amirs, but the palaces were all occupied by a succession of amirs throughout the Mamluk and Ottoman periods. They were all also renovated and added to several times but their cores remained relatively intact and all display the same arrangement of a vaulted hall on the ground floor and a corresponding hall above. The upper-floor hall in the Palace of Qawsun is a four-iwan qa‘a (fig. 41); those of Alin Aq and Bashtak have two-iwan qa‘as (fig. 42). That of Taz no longer exists, but the plan of the ground-floor hall suggests a two-iwan qa‘a above.

The architecture of these four palaces has been thoroughly studied by a group of French and Egyptian scholars who also measured and produced their plans and sections. On the plans of all four palaces, published in the book Palais et maisons du Caire (1982), the vaulted halls on the ground level are labelled istabl and the upper-level halls qasr.6 Istabl means stable and the authors assign it with caution to the lower halls.7 As for qasr, which is clearly used in al-Maqrizi’s Khitat to designate the most important hall in his descriptions of three of these same palaces,8 the French authors note that in other Mamluk palaces similar spaces are usually called qa‘a. Garcin proposes an explanation for the difference between a qa‘a and a qasr. He defines the latter as the most important hall in the second level of the palace, reserved for the amir and his retinue.9 He sees the choice of this particular term as an intentional reference to the royal paradigm in the citadel, namely the Ablaq Palace of al-Nasir Muhammad. Garcin does

FIG. 41. General view of the Palace of Qawsun

– 11 5 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 42. Qa‘a of the Palace of Bashtak

not comment, however, on its disappearance from the terminology of princely palaces constructed in the Burji period (1382–1517).10 Of the four palaces, that of Alin Aq is the best documented (fig. 43). It is described in three waqf documents, which date from various times during the Mamluk and Ottoman periods, and offer the possibility of deducing the actual Mamluk names for the spaces of a Bahri palace and of tracing the changes in terminology over time. The three waqfs were drafted after the date of the palace’s building when the initial structure was incorporated on three different occasions in endowed architectural complexes. The first waqf is that of Sultan Barsbay and dates to the year 1438.11 In it, the whole structure is called a qasr. The ground-level hall is not named but we are told that it contains service spaces and storehouses (marafiq, ma‘azil and buyutat);12 the first floor hall is either a qasr or a man-

zara and is connected to a set of dependencies that include takhayen and ma‘azil.13 The second waqf, that of Amir Khayer Bak, is dated to the year 1521.14 It describes the state of the palace after it had been joined to the new madrasa and mausoleum built by Khayer Bak around that time. In this waqf, the whole structure is called the qasr, the lower-level space, a vaulted qa‘a (qa‘a musaqqafa ‘aqdan) and the upper-level space simply the qasr. The waqf further identifies the qasr as being of old construction, and describes it as having a durqa‘a and two iwans: a large one with two sleeping spaces (mabitat) and a small one with one sleeping space (mabit) and the dependencies (khizana, or closet, and the usual manafi‘ and marafiq). By the date of the last waqf, that of the amir Ibrahim Aga Mustahfizan drafted in 1652, the palace had become ascribed to Khayer Bak, and the lower-level hall is called a sitting qa‘a (qa‘at julus).15 The whole

– 11 6 –

QASR: A N AGE N T OF MON U M E N TA L I T Y

FIG. 43. General view of the Palace of Alin Aq from the street

structure is still called qasr, and the upperlevel hall is also called the qasr, with no further elaboration. Despite the differences in date between the waqfs and the palace construction, several pertinent remarks may still be drawn from them. Firstly, the word istabl is never used to identify the lower-level hall. The first two waqfs, those of Barsbay and Khayer Bak, tell us that the istabl (stable) of the palace formed a separate unit from the qasr, and that the space under the qasr contained a set of dependencies. Secondly, the word qasr is repeatedly used in the three documents to designate both the upper-level hall and the whole palace. But qasr was not the word generally used to indicate a residence in the Mamluk period; the more common term was dar. Laila Ibrahim notes that, of fifty-seven important residences in Cairo listed by al-Maqrizi, only four are called qasr and fifty-one are called dar.16 She does not

give an explanation for the restricted use of qasr but after a review of the word’s etymology in Islamic Egypt up to the Mamluk period, she suggests that it may have been a descriptive term signifying ‘a high rectangular construction’. Al-Maqrizi’s report provides further important data.17 Although he was writing in the early fifteenth century, the middle of the Burji period, his data are mainly relevant to the study of princely Bahri palaces and do not apply to residences in his own time. Of the fifty-seven residences listed, six are Fatimid, one is Ayyubid, forty-three are Bahri Mamluk and only seven are attributable to people who lived in the Burji period. Of the seven Burji residences, only four are explicitly said to have been built after 1382, the beginning of the Burji period. Only seven out of forty-three residences attributed to the Bahri period were owned by high administrators in the state; thirty-six were

– 117 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

owned by amirs and members of the royal Qalawunid family. The same is not true for the smaller sample of Burji residences: six were owned by high administrators and only one was owned by an amir from the early Burji period, Bahadur al-Muqaddam.18 A few applications of the word qasr can be gleaned from a cursory examination of a number of royal Mamluk waqfs, chief among them is that of Barsbay, dated 841/1438.19 It contains a list of princely palaces in Cairo that Barsbay had managed to appropriate and include in his endowment, most of which date from earlier periods. The frequent appearance of the word qasr in the document allows for a comparison between the various descriptions attached to it. In the Palace of Amir Salar, the qasr is a large riwaq (in the context, the word indicates a generic living unit in the upper level of a residence)20 composed of two opposing iwans and a durqa‘a in the middle with its dependencies.21 Qasr is also a riwaq in the Palace of Amir Baysari, but no description of its plan is given other than that it has ma‘azil and manafi‘, probably the requirements of a living space.22 In the residence of Amir Mughulatay, the qasr is also a riwaq with takhayen and ma‘azil and a tibaq in addition to a kitchen and dependencies.23 In the Palace of Amir Sudun Ba‘jad, who is unknown otherwise, the qasr is a raised qa‘a (qa‘a mu‘allaqa) that has four iwans arranged in a cross plan with four columns to support the durqa‘a’s roof, with a list of unspecified dependencies.24 It is evident from the different uses of the term in the waqf of Barsbay that qasr did not denote a specific structure. There are, however, a few common characteristics in all the descriptions of qasr, which make it possible to advance a more inclusive definition of the term than those provided by Garcin and Ibrahim. Firstly, all the descriptions use the term to designate a component in a structure, not the entire structure; those of Alin Aq and Salar use it for both. Secondly, qasr appears to have become an

obsolete term in the Burji Mamluk period: many descriptions in Barsbay’s waqf are preceded by the expression ‘the place that used to be known in the past as a qasr’ ( yu‘raf qadiman bi-l-qasr). In all these descriptions, qasr is equated with either a riwaq or a raised qa‘a; terms known to the readers to suggest what a qasr was like, thus implying that it had become an obscure word by then. From the locations of all the qusur and the constant use of the two terms riwaq and raised qa‘a in conjunction with qasr it is evident that the latter is always an upper-level space. Thirdly, and most importantly, all the halls named qasr exclusively belonged to palaces of Bahri amirs. Alin Aq was a khassaki (a member of the inner circle of retinue) of al-Ashraf Khalil. He was involved in the assassination of his master and is listed among those executed in 1293.25 Baysari was a great amir and a khushdash (companion in training and member of the same household) of al-Zahir Baybars. He died in prison during the reign of al-Mansur Lajin in 1298.26 Salar was the vicegerent during the second reign of al-Nasir Muhammad, and died in prison in 1310.27 Mughulatay was the vizier of al-Nasir Muhammad for a short while at the beginning of his third reign, and had a long career that lasted well after the death of al-Nasir in 1341.28 Sudun Ba‘jad is the only unknown amir mentioned in the waqf. Other palaces listed in the same waqf that belonged to high-ranking administrators from the Bahri period (such as the dar of al-‘Imadi ibn al-Mushrif) or those that definitely belong to the Burji period (such as the huge residence near Suwayqat Mun‘im) did not contain a space called qasr. In the Bahri period, then, a qasr was an upper-level unit in an amir’s palace. From the limited evidence of architectural remains, it can also be said that the qasr rested on a lower-level hall whose plan it reciprocated and which was vaulted in most cases. This is a specific and period-bound

– 11 8 –

QASR: A N AGE N T OF MON U M E N TA L I T Y

use of the term. Before and after the Bahri period, the word qasr was applied universally to mean the fortified palace of a king or a governor;29 a meaning that remained in use during the whole Mamluk period. The Ablaq Palace, al-Maqrizi’s four palaces and the waqf of Barsbay’s designation of the palaces of Salar and Alin Aq are all qusur. The new meaning was derived from the generic one and applied to a particular context. The development of the restricted and spatial meaning for the word qasr in Bahri Mamluk Egypt along with the old common one created confusion in the documents of the period. Using the term qasr in its general meaning to designate a whole structure seems to have been arbitrary, as evidenced in al-Maqrizi’s Khitat where he alternates between the terms dar and qasr when he speaks about several structures,30 and in other chronicles, where some of the palaces that al-Maqrizi calls qasr are called either a bayt (house) or a dar. 31 In the Burji period, it appears from the limited number of available documents that the narrow Bahri use of the term was dropped while the word retained its general palatial connotation. The upper-level hall in a Burji amir’s palace lost the semantic distinction that a princely Bahri equivalent had. Like non-princely residences throughout the Mamluk period and after, it was called either a raised qa’a or a riwaq.32 The word qasr was used in its Bahri meaning only to refer to princely Bahri halls that were still functioning in the Burji period.33 The lack of distinction between princely and non-princely palaces of the Burji period as opposed to those of the Bahri period was noticed by contemporaries, and even used as a snub by observers. In an anecdote attributed to the Ottoman sultan Selim, he was said to have reacted to the Mamluk structures he saw by remarking that the Mosque of Sultan Hasan was a magnificent fortress (hisar); while the Madrasa of Qansuh al-Ghuri (1509) was a merchant’s qa‘a (qa‘at tajir), lacking in royal monumentality.34

A qasr as an architectural unit in a palace could have any number of iwans around its durqa‘a (the examples include qusur with two iwans or four). Each iwan could have a number of sleeping spaces (mabitat, which are also called khaza’in nawmiyya) in alcoves on its sides. The lower-level space under the qasr is mostly referred to as a qa‘a (Alin Aq, Baysari, Mughulatay), sometimes a vaulted qa‘a (Alin Aq in Khayer Bak’s waqf), and sometimes undefined (Salar, Sudun Ba‘jad). However, the most important aspect of a qasr is that it is exclusively a princely hall from the Bahri period that rises above a lower level. This definition perfectly fits the Ablaq Palace of the citadel. The four halls of the Ablaq Palace reported by al-‘Umari are qusur par excellence: they are more than princely structures, they are royal ones, and they appear to have risen above, not only one lower level but two and perhaps even three. The Ablaq Palace stood on the site of a present-day platform adjacent to the outer courtyard (sahn) of the Mosque of Muhammad ‘Ali, which extends along the citadel’s walls to the south-west to the location of the men’s quarter (salamlik) of the Bijou Palace, built in 1812 (fig. 44). Two huge, superimposed and vaulted halls run underneath the platform.35 The upper hall, trapezoidal in shape and badly damaged, is composed of forty vaults arranged in six rows, the front two of which are groin vaults, the back ones are barrel vaults. The lower hall is made up of eleven square vaults: nine of them are groin vaults and two are barrel vaults (fig. 45). On its southern side, it opens onto a ruined area, which was hastily restored in the 1980s to complete the façade. Its remains, which extend to the upper hall above, cover the surface of six vaults. The two halls’ outer façades opens onto the maydan with eight arched windows for the upper hall and six corresponding square windows for the lower one. There is another set of similar windows, blocked with masonry, immediately below

– 11 9 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 44. Site of the Ablaq Palace at the Citadel

the windows of the lower hall. They may reveal a third level below the two vaulted halls, but this is impossible to verify at present. These two halls were most probably the two lower levels of the Ablaq Palace of al-Nasir Muhammad. The qa‘at, or qusur, of the palace may have been arranged on the present platform on top of the upper vaulted hall. Its surface area is large enough to have accommodated four halls aligned side by side in a row parallel to the wall of the citadel, in the way described by al-‘Umari and al-Maqrizi, so that the window view of each qasr in the palace encompasses the whole city of Cairo lying at the foot of the citadel. The lower halls were not part of the royal stables, for it is clearly stated in the

FIG. 45. Vaults of the underground hall below the Ablaq Palace

–120 –

QASR: A N AGE N T OF MON U M E N TA L I T Y

sources that the Ablaq Palace’s qusur overlooks (tattilu ‘ala) the royal stables, which must then have been separate from them.36 The lower halls were most probably reserved for the Mamluks in service and for some of the royal storehouses (al-buyut al-sultaniyya), just as the lower halls in princely palaces were, as described in waqfs.37

QASR AND MONUMENTALITY The question remains why the word qasr was used rather than a more precise term to designate what seems to have been only a variation on a prevalent architectural type, the riwaq (whose plan is that of a qa‘a) and its dependencies over one or more lower halls. The readily available explanation, and the one alluded to by Garcin and Revault, is that the Ablaq Palace provided the royal model for contemporary and later Bahri palaces, which copied the form and acquired the name qasr. Revault and Garcin did not differentiate, in their discussion of Bahri palaces, between the general meaning and the specific Mamluk uses of the word qasr, and thus they did not notice the contradiction inherent in explaining the adoption of a word from a general royal usage to an architecturally particular one. Furthermore, their explanation is based on the thesis of a downward-moving influence in architecture which assumes that the Ablaq Palace’s forms were new to Mamluk Cairo: that they were invented, or imported and transmitted to princely palaces, and then possibly spread into the local architectural repertoire. The history of residential architecture in Cairo, in which upper-level halls with a qa‘a plan had been common before the Mamluk period, refutes this assumption, as does the fact that Bahri palaces older than the Ablaq Palace, such as the palaces of Alin Aq, Baysari, Salar and Bektash Amir Silah, included qusur units and that these qusur may have furnished the model for the Ablaq Palace rather than the other way round.

The word qasr was used precisely for the meaning it would convey, namely monumentality, both in its memorial and formal connotations.38 Since qasr was normally used to designate the palace of a sultan or an amir, applying it to the place of honour in the palace, such as the qa‘a where the sultan or the amir sat, is an appropriate concentration of the meaning in a place that lends the entire structure its memorial value. This interpretation would clarify why a riwaq in a princely Bahri palace is called a qasr, while a riwaq in the palace of an administrator or a merchant is just a riwaq. However, this distinction cannot fully account for the disappearance of the term from the vocabulary of Burji palatial architecture, despite the fact that princely palaces were still being built with an upper riwaq and a lower hall. The second connotation of monumentality, grandeur in the formal sense, is what distinguished princely Bahri halls from Burji ones and made them deserving of the appellation qasr. Extant princely Bahri halls are in fact larger and higher than later Burji (of which an even smaller number exist) or Ottoman ones.39 Moreover, Bahri qusur were not only loftier (the word Humphreys likes to use when describing the physical monumentality of Mamluk architecture in general)40 than Burji ones, but they seem to have been positioned to appear even loftier than they actually were. The qusur of Alin Aq and Bashtak overlooked the street so that their mass and their height would impress the passersby (fig. 46).41 The palaces of Qawsun and Taz may originally have had the same disposition but this is difficult to verify today as their plans have been modified and the street configurations around them have been altered several times.42 In alMaqrizi’s report on the building of a qasr for Amir Arghun al-Kamili in 1346, a more explicit remark concerning the position of the qasr vis-à-vis the street is presented. Sultan al-Kamil Sha‘ban, the son

– 1 21 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

–122–

QASR: A N AGE N T OF MON U M E N TA L I T Y

FIG. 47. The Maq‘ad of the Amir Mamay al-Sayfi

of al-Nasir Muhammad, who had bestowed this palace that bordered on the Pond of the Elephant (Birket al-Fil) on his favourite amir Arghun, ordered that a qasr be constructed next to it and specified that it should overlook the street. Although the more expected choice would have been the pond side in order to exploit the view, achieving the monumentality effect on the street side was apparently more important to him than having a view of the water.43 The arrangement of architectural elements in princely palaces changed in the Burji period. Closed upper-level halls lost their prominence as the place of honour, to be replaced by the maq‘ad, or open loggia

 FIG. 46. Façade of the Palace of Bashtak

with several arches that overlooked the courtyard (fig. 47). The elevation to prominence of the maq‘ad in the Burji period necessitated the restitution of the courtyard to a central position in the palace, as shown in the plans of the few remaining palaces from the time of Qaytbay.44 The palace became an introverted composition where halls and rooms were arranged around, and opened onto, its courtyard. They could hardly have been seen from the street, and thus they could not have been intended to overwhelm. The Bahri qusur were conceived with exactly the opposite intention in mind. Constructed above lower halls, their iwans and durqa‘as were built to a great height and were planned to have an imposing effect on observers. The building of these qusur was to achieve utmost monumentality (in the formal sense), certainly to reflect and symbolise the memorial function of the structure

–123 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

as the palace of the sultan or the prominent amir. This effect was not realised by inventing or importing a new architectural type but by using a local one, adapting it to the

exigencies of the sultan’s or the amir’s image and giving it a name, qasr, that already carried the connotations of loftiness and commemoration.

–124 –

11. MAMLUK THRONE HALLS: QUBBA OR IWAN

T

he Mamluk sultans of Egypt and Syria were associated with the Citadel of Cairo from the foundation of their regime in 1250 until its ultimate fall in 1517. In the early part of this long period, four prominent sultans, al-Zahir Baybars (r. 1260–1277), al-Mansur Qalawun (r. 1280–1290), al-Ashraf Khalil (r. 1290–1294) and al-Nasir Muhammad (r. 1294–1340, with two interruptions) developed the citadel they had inherited from their masters the Ayyubids to become both their seat of government and their royal residence. They established a palatial complex in its southern part and separated it from the northern part, which was reserved for the administrative functions and for the lodging of the royal Mamluks. Each of these four sultans is known to have ordered the construction of a number of palaces and audience halls, which either replaced or complemented those of his predecessors and which extended the citadel’s southern enclosure to the west and south while maintaining its overall spatial division. Three of the four, Baybars, Qalawun and al-Nasir Muhammad, instituted ceremonies and processions that followed a strictly prescribed order aimed at engendering the rigid hierarchical structure of the Mamluk sultanate

in the form and plan of the palatial complex. In fact, the ceremonial programme they followed seems to have been so pervasive as to govern not only the relationships between the various components of the northern and southern enclosures but also those between the entire citadel and the city outside.1 One royal structure, the public audience or throne hall, was so pivotal in the representation of the Mamluk hierarchy with the sultan at its apex that it persistently played a central role in the conception of the citadel’s layout. Each of the four sultans destroyed the throne hall constructed by his predecessor and built a new one in its place soon after he ascended the throne. Early in his reign, Baybars built a throne hall as a replacement of or as an addition to the original one in the citadel. Qalawun ordered the demolition of Baybars’s hall in 1284 to build a new one in its stead. The building of the new hall can be ascribed to Qalawun’s desire to be the patron of this most visible and most public structure in the palatial complex, for otherwise he did not demolish any other structures Baybars had built. Both the sons and successors of Qalawun, al-Ashraf Khalil and al-Nasir Muhammad rebuilt the citadel’s audience

–125 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

hall within the next half century, presumably because they too wanted to have their own names attached to it. In 1293, al-Ashraf Khalil restored, or perhaps rebuilt (the sources are unclear about the extent of the work achieved during his short reign) the structure attributed to his father. This last structure was in turn destroyed by Khalil’s brother and successor al-Nasir Muhammad, to be replaced by his famous hall, the Iwan al-Nasiri, concurrently called Dar al-‘Adl (the Palace of Justice) and used as the official setting for the royal dispensation of justice (also called dar al-‘adl). The Iwan al-Nasiri remained in use throughout the Mamluk period, was neglected in the Ottoman period, and was finally razed to the ground during the reign of Muhammad ‘Ali (1805–1848) who replaced it with his monumental mosque. Its location and form, however, are known, for it was still standing at the beginning of the nineteenth century, when it was documented for the Description de l’Égypte. Its site is marked on the

Description map where it is labelled le Divan de Joseph. Its plan, elevations and sections were reproduced in a series of eighteenth- and nineteenth-century sources of which those of the Description are probably the most accurate (fig. 48).2 An interesting and revealing aspect of these four Mamluk halls is the vocabulary used to designate them in contemporary sources. Instead of standard Arabic terms, such as qasr, qa‘a, majlis or takht, they are consistently named either qubba or iwan in the sources. More interesting is the interchangeability of iwan and qubba in naming the halls of Baybars and Qalawun. The use of the term iwan only became fixed with the building of al-Ashraf Khalil ibn Qalawun, which was called the Iwan al-Ashrafi. Al-Nasir Muhammad’s hall was always referred to as the Great Iwan throughout the Mamluk period. However, its name became distorted as the Diwan of Sultan al-Ghuri under the Ottomans,3 and it changed again at a certain point to become Diwan Yusuf, which is

FIG. 48 Plan and sections of the Great Iwan

–126 –

M A M L U K T H RO N E H A L L S : Q U B B A O R I WA N

the name the French cartographers heard from the local residents and recorded on their Description map in 1800.4 The two words iwan and qubba present complex ranges of meaning in the medieval context. Both normally denoted architectural elements: the first a vaulted hall open at one end or a raised portion of the floor in a vaulted hall, the second any kind of dome. Yet both words were used to designate the whole structure in a number of instances throughout pre-Islamic and Islamic history. In commemorative architecture, the word qubba often signified the mausoleum of an amir or a pious man, which was usually but not always a cubical structure covered with a dome. 5 In the palatial context, qubba appears less frequently as the name of an entire structure but it is still encountered in a few famous examples, especially in the early or classical period. There are references to the Qubbat al-Khadra (the Green Dome) as the name of a number of Umayyad palaces in Damascus, Wasit and Rusafa, and the early Abbasid palace built by Abu Ja‘far al-Mansur in the centre of the round city of Baghdad.6 Later palaces with the same generic name, qubba, are also attested as far afield as Egypt and North Africa. They include the palace known as the Qubba (la Cuba) in Palermo, Sicily, built for the Norman king William II in 1182,7 and the famous Qubbat al-Hawa (Dome of the Winds) which was built between 809–11 by Hatim ibn Harthama, the Abbasid governor of Egypt, on the hill upon which the citadel was later constructed, and which was used as a pleasure pavilion by all subsequent rulers of the country until the end of the Tulunid period. 8 Similarly, the word iwan is used in several instances to designate an entire palace. The most important examples are the legendary Iwan Kisra (Arabic for Khosroe), the Sasanian palace in ancient Ctesiphon,9 and the Iwan al-Kabir (Great Iwan) which was the main ceremonial hall in the

Fatimid Eastern Palace in Cairo built by al-‘Aziz in 979.10 These palaces were presumably named after their most visually impressive element, be it the iwan or the dome. Both features appear to have represented one underlying concept: monumentality, in both its formal and spatial aspects. The iwan in a palace seems to have conveyed, in most cases, a ceremonial value as the place of honour in which royal audiences took place. The dome, too, seems to have symbolised authority and domination.11 However, as the examples at the Citadel of Cairo suggest, the process of naming structures was not always connotative. This chapter proposes a different explanation for the terminology used in Mamluk throne halls that raises a number of wider questions about the conscious use of the past in Mamluk architecture, and about the survival of ancient architectural paradigms in the medieval Mediterranean world in both its Islamic and Christian areas.

THE SEQUENCE OF BUILDING THRONE HALLS AT THE CITADEL At the Citadel of Cairo, the structure known as Iwan al-Qal‘a (Iwan of the Citadel) is first mentioned in the sources during the reign of al-Zahir Baybars.12 He, however, is not credited with its building, nor are any of his Mamluk predecessors. The iwan was most probably built by al-Kamil Muhammad (r. 1218–1238), for he is the sultan credited with most major structures at the citadel in the Ayyubid period. In fact, one early Mamluk chronicler, Baybars al-Mansuri, calls it the Iwan al-Kabir al-Kamili (the Great Iwan of al-Kamil), when he reports the ceremony of recognition of the second Abbasid caliph, al-Hakim, in 1261.13 To Baybars, however, is attributed another structure called the qubba, which he ordered built in the main court at the citadel. It was supported by twelve coloured, marble

–127–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

columns, was profusely ornamented, and figures of the sultan and his amirs were painted on its interior surfaces.14 The qubba was most probably added to, or part of, the Iwan of al-Kamil for we have no report that the iwan was demolished when the qubba was constructed. Ibn alDawadari, who generally speaks of Baybars holding audiences in the iwan, says in one instance that the sultan ‘sat in al-iwan wa-lqubba with the newly appointed caliph’, implying that the two words refer to a single structure.15 Al-Maqrizi offers us another clue in his description of the circumcision celebration of Baybars’ son in 1273: ‘he sat in the seat of his sultanate in the qubba alsa‘ida’ (the exalted qubba),16 which indicates that the same throne hall named in the earlier reference iwan and qubba, could also be referred to simply as qubba. The use of these two terms together or interchangeably in designating a single structure could be interpreted in one of three ways. It may signify that the original structure of al-Kamil did not have a dome, and that Baybars added a dome to it and caused observers to shift their emphasis from the iwan to the dome when they mentioned the new audience hall. It may also mean that both architectural elements belonged to Baybars’ rebuilding and both were prominent in the perception and description of the new hall so that the authors used them at will. The third possibility would discount the notion of looking for a correspondence between the name of the hall and its most impressive architectural element and seek an explanation for the shift in terminology in the wider context of early Mamluk architecture and its use of historical references. The nascent Mamluk architectural style appears to have developed a visual and symbolic system of references to venerated precedents, references that may also have been reflected in the terminology of buildings. The Qubba al-Zahiriyya did not last for long, nor was it described anywhere, so

there is no data on which to base an analysis of its possible referential or historicising form and layout. In 1284, al-Mansur Qalawun ordered its demolition and the building of a new qubba in its place.17 From the wording of the report on Qalawun’s order, it is very difficult to know whether the demolition of Baybars’ qubba was total or partial, and other Mamluk sources tend to pass over the event in silence.18 Another source tells us that the Qubba al-Mansuriyya had ninety-four small and large columns, not counting those on the porticos or aisles (depending on how we interpret the word riwaqat).19 This number, if true, must have included all the columns in the qubba, both those supporting the dome and those adorning the interior and exterior surfaces of the building. Decorative columns could be numerous, especially in the early Mamluk period when double-arched windows with three engaged columns each were the norm, as shown in other structures of Qalawun, such as his mausoleum (completed in 1285) and that of his wife Umm al-Salih or Fatima Khatun (built in 1283–84). By the time Khalil demolished his father’s hall in 1293, the sources were already calling it the Iwan al-Mansuri instead of the Qubba al-Mansuriyya. Afterwards, the structure’s name changed in the sources from the Iwan al-Mansuri to the Iwan al-Ashrafi, although it seems that no major structural or spatial changes were introduced to the building that essentially remained as Qalawun had built it.20 Khalil’s iwan also did not survive long after his death. Al-Nasir Muhammad’s first work in the citadel, after his return to the throne for the third and last time in 1310, was the demolition of his brother’s iwan and the building of a new one in 1311.21 Three reasons could be advanced to explain this rebuilding. The first is that al-Nasir disliked the gloom (ghils) of the old iwan and the additional, awkward supports (arkan) erected after the earthquake of 1303.22 The second is that he hated the iwan of his

–128 –

M A M L U K T H RO N E H A L L S : Q U B B A O R I WA N

brother because it carried the memory of previous sultans and the two earlier humiliating periods of his reign, when he held the title of sultan but had no real power.23 The third reason stems from the political and ceremonial changes introduced by al-Nasir Muhammad. Shortly after his return to the throne, he started to hold the regular sessions for the dispensation of justice, dar al‘adl, in the iwan where he sat twice weekly surrounded by all the important amirs of the realm who either sat or stood in a strict hierarchical order.24 The old iwan may have been inadequate for this novel ceremony, as al-Nasir had designed it, and thus may have prompted him to order its demolition and to build a new, more accommodating hall.25 Conditions seem to have changed again in the later part of al-Nasir’s reign, for he rebuilt the iwan a second time in 1333. The reports are not very clear about how much he actually had razed; some say the structure was levelled, others that only the dome in the middle of the structure was demolished.26 The new iwan remained the official throne hall at the citadel where coronations, iqta‘ distributions, reviews of troops, receptions of envoys and the biweekly dar al-

FIG. 49. The plan of the Great Iwan

‘adl sessions were held throughout the rest of the Qalawunid period until 1382. In the Burji period, it lost its function as the setting for dar al-‘adl,27 but was still used to receive foreign embassies, undoubtedly because of its size and spatial arrangement which made it the most impressive structure at the citadel and the most expressive of the sultan’s might. Otherwise, it was neglected throughout most of the Burji and Ottoman periods, and its site was finally cleared by Muhammad ‘Ali, along with those of all the other palaces and halls that the Mamluks and the Ottomans had built, to construct his new structures there.28

THE DESCRIPTION OF THE GREAT IWAN Fortunately, the Great Iwan was documented at the beginning of the nineteenth century in the Description de l’Égypte, although it was already in a ruined state and lacking its dome. These invaluable plan and views permit an architectural analysis of the structure that is not possible for the majority of the other Mamluk monuments that existed in the citadel (fig. 49). The French plan shows the Great Iwan to have been a rectangle (measuring 36m by 31m (120 ⫻ 95 ft) without the corridor behind it), open on three sides: the northeast, which constituted its main façade, the south-east and the north-west. The fourth side, which faced back towards the rest of the sultan’s palaces across from the passageway (called dihliz al-‘ubur, or the corridor of crossing, in Mamluk sources), was built up with a thick wall. This wall was pierced with doors in five places; the central door, according to the eighteenth-century traveller Pococke, ‘was adorned with that grotesque sort of work, which is common in the Eastern buildings.’29 He, of course, did not know the muqarnas conch, which was the element he observed above the central door (seen in Part II, fig. 48). The shape of this arched door was confused by Jomard, the author of the entry on

–12 9 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

the citadel in the Description de l’Égypte, with that of a mihrab, which led him to conclude that the iwan must have been used as a mosque.30 The door opened onto the sultan’s private domain behind the iwan through the passageway seen in the plan, which led to the Ablaq Palace and beyond it to the private quarters (al-dur al-sultaniyya).31 The iwan’s layout consisted of three parallel aisles formed by four rows of reused red granite columns. The central aisle was almost three times as wide as the lateral ones. The frontal one-third of its length was subdivided into three parts formed by two pairs of columns. The columns in the iwan numbered thirty-two in all,32 and were taken from pre-Islamic Egyptian temples. The back section of the central aisle, covering two-thirds of its length, was surmounted by a dome. That dome – which had already collapsed when the drawing was made – had once been the most striking feature of the iwan.33 It was constructed of wood, like most Bahri Mamluk domes, and covered on the outside with greenish tiles. It was supported by twelve columns which, together with the back wall, formed a square plan, almost 20m (63 ft) to the side. The transition from square to circle was achieved by four wooden muqarnas pendentives, whose units, to judge from the perspective of the Description de l’Égypte, were huge (fig. 50). A broad inscription band, whose characters were made of large carved and gilded wood units, ran around the full perimeter of the inner square under the dome and even followed the curve of the central aisle’s arch. Its text seems to have consisted of the full titulature of Sultan al-Nasir Muhammad and probably the construction date.34

THE INTERPRETATION OF THE GREAT IWAN’S ARCHITECTURE The iwan of al-Nasir Muhammad was certainly the most public and most ceremonial of the sultan’s palaces, and therefore made

FIG. 50. Interior of the Great Iwan

to impress and to inspire awe. Its royal and monumental character was visually and spatially articulated through its massiveness, height, the lavishness of its surface articulation and a number of other architectural features. Two of these features, the layout and the green dome, were particularly effective as historically recognisable signs of royalty: the first as a spatial and functional frame of action, the second as a visual referent. The plan of the iwan was different from the common hall type of Islamic Egypt, generally known as a qa‘a. This difference has been noted by many scholars, and a number of architectural precedents have been proposed. Doris Behrens-Abouseif suggests that the plan for the main hall in several Fatimid shrines (mashhads) in Cairo may have provided the model for the plan of the Great Iwan. She specifically mentions the shrines of al-Juyushi (1085) and Sayyida Ruqayya (1133),35 whose halls she compares to the Great Iwan in form and arrangement (fig. 51). However, the plans of these shrines are only variations on the majlis plan, the most widespread hall type in Egypt until the thirteenth century. A majlis has a T-shape plan with a large space in the

–13 0 –

M A M L U K T H RO N E H A L L S : Q U B B A O R I WA N

FIG. 51. Plan of the Mashhad al-Juyushi

FIG. 52. Plan of a Fustat house

centre and two smaller, and sometimes shallower, ones flanking it. It has a frontal gallery (riwaq) with a set of doors that separate it from the central space, whether it is an open court or a roofed durqa‘a (the name of the central square in a hall, usually covered with a lantern).36 The type first appeared in the houses of Abbasid Samarra and seems to have been imported to Egypt during the Tulunid period (868–905). The majority of houses excavated at Fustat, dating from the ninth through the twelfth century, had at least one majlis around their open courtyard, which probably functioned as the place of honour in the structure (fig. 52). The halls of both Fatimid shrines clearly derive from the majlis plan, but have in addition a dome above the central back space, probably to either honour the sacrosanct persons to whom they were dedicated or to emphasise the qibla side, an arrangement encountered in several large Fatimid mosques such as the Azhar Mosque (970–72) and al-Hakim Mosque (990–1013). Alexandre Lézine proposes as a prototype for the iwan and other royal Mamluk structures the hall of al-Salih Najm al-Din Ayyub at the Roda Citadel (fig. 53).37 This may have been the hall that Ibn Sai‘d alMaghribi, who visited the Roda Citadel shortly after its completion in the 1240s, identified as the iwan that al-Salih used for his audiences.38 The plan of this hall, which was still standing in the early nineteenth century, comprised two iwans facing one another across a huge durqa‘a. Four sets of columns, each composed of three columns arranged as a triangle, formed a smaller rectangle inside the durqa‘a. They framed the side iwans and the two alcoves (suffas) on the longitudinal sides and may have supported a dome, or perhaps a wooden lantern in the centre of a flat roof, similar to those found in later Cairene houses and called shukhshikha. Lézine considered this plan to have formed a transitional stage between the hall plan in Fustat houses, which we now know was another architectural type – 1 31 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 53. Plan of the hall of al-Salih Najm al-Din Ayyub

called majlis, and that of the later qa‘as of the Bahri Mamluk period. Though al-Salih Najm al-Din’s hall may have presented the model for other Mamluk halls, the Great Iwan’s plan is radically different from any of the stages of this simple linear development from majlis to qa‘a. It was a unique structure designed for a specific set of functions and with other architectural paradigms in mind. Reduced to its essence, al-Nasir Muhammad’s iwan was a roof supported on pointed arches carried by columns with open façades on three of its four sides, while both majlis and qa‘a plans are by definition enclosed spaces. The openness of the Great Iwan was one of its essential characteristics, since it functioned both as stage and reviewing stand for the sultan. He could be seen from all sides when

he sat to hear the grievances of his subjects on dar al-‘adl days or for embassy receptions,39 and he could view the parades taking place in the court in front of the iwan on audience (khidma) days.40 Only the back wall, which functioned as the screen that separated the iwan proper from the royal palaces, was solid. The Great Iwan was in fact considered by contemporaries to be a public structure standing alone ‘outside’ (zahir) the royal palaces.41 Besides, the form and function of the central door on the iwan’s back side, complete with its recessed arched opening, topped with the muqarnas conch and flanked by the two customary stone benches called mastabas or maksalas, give it the appearance of a typical Mamluk portal leading to the private royal palaces behind. Analysis of seventeenth- to nineteenthcentury drawings and descriptions suggests that the plan of the Great Iwan may be related to another established type, the basilica,42 though basilicas were uncommon in Cairo at the time. The basilical plan was used in only one other contemporary Cairene structure: the qibla iwan of the Madrasa of Sultan Qalawun in the complex he had established in 1284 on the site of the Western Fatimid Palace in Bayn al-Qasrayn (fig. 54).43 There, the three aisles, with four arches each, that form the tripartite arrangement appear to have been added to the qibla iwan as an afterthought to what was otherwise a madrasa with the usual ubiquitous two-iwan plan. It may have been that the madrasa was already started with its two iwans when an order was given to insert the three aisles in the qibla iwan. They may have been applied to provide the prayer space with a different treatment or an impressive façade, or to suggest a royal association.44 They may also have been meant as an imitation of other structures admired by the patron or suggested by his architects. The context was evidently different for the Great Iwan at the citadel. Its plan seems

–132 –

M A M L U K T H RO N E H A L L S : Q U B B A O R I WA N

FIG. 54. Plan of the Madrasa and Mausoleum of Qalawun

to have been a well-thought-out variation of a basilica without the anachronistic envelope of an iwan or a qa‘a. The basic tripartite division leading to a focal point in the centre of the back wall is kept in it, though the hall is modified by shortening its sides to become almost a square, thereby transforming the traditional longitudinal arrangement of basilicas. The apse is replaced with a monumental portal, and the sides are opened up to provide an unobstructed view to the outside, implying the accessibility of the sultan sitting within. These alterations, important as they may be, do not conceal the fundamental affinity of the Great Iwan’s plan with the domed basilica type, examples of which abound in Eastern Roman, Byzantine

and Umayyad urban and provincial architecture.45 But where did this basilical plan come from? It may have been inspired by its four direct predecessors at the Citadel in Cairo, but this is impossible to ascertain for we know little about their layouts beside each having had a dome and possibly iwans or surrounding porticoes. However, even if these halls did provide the models for the Great Iwan, this still does not explain the resurgence of a basilical plan in medieval Cairo which had no such palatial precedents from the Islamic era. This leaves us with two alternatives. The first is that the basilica type survived in Egypt through some unknown series of structures, to which the

–133 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

forerunners of the Great Iwan at the citadel may have belonged. The second is that the type was reintroduced to Mamluk Cairo after it had disappeared from the secular architectural vocabulary of the country, following the abandonment of the classical tradition with the fall of the Umayyads in 750. Both alternatives are plausible and both are difficult to verify on archaeological or architectural grounds. Ultimately, the Great Iwan plan has to be traced back to the Roman basilical model, which may have provided not only the original architectural paradigm, but also the ceremonial functions and the symbolism attached to it and developed for it.46 The historical circumstances of the Great Iwan’s building and the elaborate descriptions of the ceremonies that took place within it point in this direction, but the complete image is still too sketchy and the sources too inadequate to allow for any firm conclusion.47 However, the meaningful connection for the Great Iwan is not its ultimate origin in antiquity, which was probably obscure to Mamluk builders and patrons alike. Rather, its intentional formal and symbolic associations should be sought in the intermediate Islamic types that had already adapted the basilica form to new functions. This proposition would allow us to postulate a number of political and social reasons for the adaptation of such a plan for the Great Iwan. Basilicas must have existed in Hellenistic and Roman Egypt, though no hard evidence remains of them. The great majority of churches built in Egypt from the fourth century to the seventh, much as elsewhere in the Byzantine realm, followed a basilical plan. This certainly argues for a local tradition that might have survived in the country after the Islamic conquest. However, the absence of any trace of an Islamic palace based on a basilical plan before the thirteenth century makes this possibility unlikely. Furthermore, the ubiquitous majlis plan in all the known large

houses of Fustat, with its proven ties to Samarran models, weakens such a possibility. It is very difficult indeed to identify any intermediate model in Islamic Egypt that may have provided the inspiration for the Great Iwan at the citadel. The only preMamluk halls whose arrangement we know something about, besides that of al-Salih Najm al-Din Ayyub, are the two Fatimid halls, the Great Iwan of the Eastern Palace and the Qa‘at al-Dhahab (Hall of Gold). However, al-Maqrizi’s descriptions of them clearly indicate that both belonged to the majlis type.48 The situation is different in Syria, where the existence of a long tradition of basilical audience halls well into the early Islamic period has been demonstrated. The tradition was apparently localised before the coming of Islam, as evidenced by the basilical audience hall built between 569 and 581 for the Ghassanid chief, al-Mundhir ibn alHarith (fig. 55). This hall, which stood outside the north wall of the city of Rusafa (Sergiopolis), was for a long time considered to have been a church until Jean Sauvaget conclusively proved its palatial function.49 It had a dome or a lantern over the centre of the nave, and its apse, which is much smaller than apses in contemporary basilical churches, was absorbed in the thickness of the back wall with flanking rectangular rooms. Despite being the only known example of a basilica/audience hall of the period, al-Mundhir’s structure has been used to postulate a possible ‘Arabic’ precedent for a later Umayyad hall, Khirbat al-Mafjar, attributed to al-Walid ibn Yazid (743–744).50 Al-Mundhir’s hall, it has been claimed, provided not only the model for the architectural arrangement of the Umayyad hall, but also its functional and ceremonial references. Several other early Umayyad palaces, such as the Dar al-Imara at Kufa and the palace at ‘Anjar, took up basilical themes for their audience halls.51 In one celebrated case, Mshatta, the three naves led to an elab-

–13 4 –

M A M L U K T H RO N E H A L L S : Q U B B A O R I WA N

orate triple apse or a tricorium. The audience hall of Mshatta, similar to and yet distinguished from the examples of the local Syrian developments of triclinia, was considered to have been either a later imitation of a little-understood classical type,52 or a syncretic creation of a new Umayyad type that synthesised Sasanian models of audience halls with the local, classically inspired basilica.53 Our knowledge of palatial architecture in Islamic Syria after the Umayyads and until the twelfth century is very sparse. Few vestiges remain, and only the names, locations and extravagant decorations of palaces are mentioned in the medieval sources. This, obviously, renders the task of proving the persistence of Umayyad models in medieval Syrian architecture, and their subsequent transfer to fourteenth-century Cairo very difficult.54 There are, however, a few indices to suggest that indeed this was the case. These clues range from circumstantial

to interpretive ones and, put together, they make the association between the early Islamic halls and the Great Iwan plausible not only on the formal but also on the ideological level. The first level of argument is tangential to the specific question of the basilica, but it provides the impetus and the plausible conduit for the transfer of Syrian building traditions to Cairo. The motivation is suggested by the reports on the enthusiastic reaction of many of the early Mamluk sultans to a number of buildings in Damascus and their desire to replicate these structures in Cairo. Two examples are especially relevant for they refer to two sultans, Qalawun and al-Nasir Muhammad, who were responsible for two of the audience halls at the Citadel of Cairo. Qalawun visited Damascus several times, before and after he became sultan, and was once successfully treated for dysentery at the bimaristan of Nur al-Din (built 1154). He consequently renovated that

FIG. 55. View of al-Mundhir’s Basilica in Rusafa (Sergiopolis)

–135 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

bimaristan in 1281, and is said to have wanted to follow Nur al-Din’s example when he constructed his own bimaristan in Cairo three years later in Bayn al-Qasrayn as part of a complex which comprised also his madrasa with its basilica/iwan, and his funerary qubba (mausoleum).55 Al-Nasir Muhammad also visited Damascus many times in the early period of his sultanate. During many of these visits he stayed at the Ablaq Palace in Damascus, which was constructed by al-Zahir Baybars in 1264.56 In 1312–13, al-Nasir went to Damascus, after rumours of an imminent new Mongol invasion had reached him, and stayed at the Ablaq Palace for a month. When the invasion failed to materialise, he went to the Hijaz for the hajj and returned to Cairo via Damascus. Upon his return to Cairo, he at once ordered the building of an Ablaq Palace at the citadel, which was, we are told, modelled on that of Baybars in Damascus.57 Both Qalawun and al-Nasir Muhammad re-created an admired Damascene structure in Cairo, the first because he was inspired by the social and charitable role fulfilled by Nur al-Din’s bimaristan, the second because he was impressed by the striking appearance and opulence of Baybars’ palace. Both were also, in their sponsorship of these structures, trying to emulate the example of the two princes, Nur al-Din and Baybars, who had been well remembered after their death, the former as a just, pious and brave ruler and the latter as an energetic and valiant one. Qalawun and al-Nasir Muhammad also appear to have summoned workers from Damascus to help building their structures in Cairo. In Qalawun’s case, Michael Meinecke has argued a Damascene source for many decorative and structural techniques found for the first time in Cairo in his complex. He suggested a direct link between the decoration of a number of Zengid, Ayyubid and Mamluk structures in Damascus and Qalawun’s complex in Cairo,

and maintained that Syrian craftsmen were brought to Cairo to execute the decoration of the latter. This hypothesis was critical for Meinecke’s contention that the decoration of Qalawun’s complex functioned as the main vehicle for Damascene decorative influences in Mamluk Cairo.58 But here we can extend the proposition that Syrian influence accounted also for both the basilical plan of Qalawun’s madrasa and the octagonal plan of his mausoleum, which has been compared to the Dome of the Rock in Jerusalem, the first Umayyad monument built in 692 and rarely copied in Islamic architectural history.59 In the case of al-Nasir Muhammad’s Ablaq Palace, we possess a more direct reference. Mufaddal ibn-Abi al-Fada’il says that Christian marble cutters from Damascus were brought to build the Ablaq Palace at the citadel in 1311.60 He goes on to tell us that the citadel palace was built like (nazir) the Ablaq Palace of al-Zahir Baybars in Damascus and following the same arrangement (tartib), implying that the marble cutters brought with them not only their craftsmanship but their architectural traditions. The second level of argument concerning the provenance of the Great Iwan’s plan reveals its symbolic association with early Islamic precedents through its use of a green dome. Textual references to at least three Umayyad halls, in Damascus, Rusafa and Wasit, and an Abbasid one in Baghdad, call them by the generic name qubbat alkhadra, which is generally understood to have meant ‘green dome’. The medieval historians, who are our only source of information on the four palaces, had never seen the domes themselves. Thus it comes as no surprise that objections have been raised as to just how ‘green’ the domes of these palaces were.61 However, even if it were valid to pose the question in relation to the origin and meaning of the word khadra, the fact remains that for medieval authors the dome of any qubbat al-khadra was invariably assumed to have been green in colour.62

–13 6 –

M A M L U K T H RO N E H A L L S : Q U B B A O R I WA N

The first example, the Palace of Mu‘awiya in Damascus which he had built when he was still governor of Syria between 640–661, was called the Dar al-Imara (the Palace of Government), or the khadra (the green) or the qubbat al-khadra.63 The palace was surmounted by a great green dome, and was used as the official seat of the Umayyad caliphs in Damascus until the Abbasid revolution of 749–50. The concept and name of Mu‘awiya’s palace were taken up by al-Hajjaj, the mighty governor of Iraq and the east under ‘Abd al-Malik (r. 685–705), when he established the new capital Wasit in Iraq in 695 and built the governor’s palace in it with a monumental dome also called the qubbat al-khadra. The same was done a while later by Hisham ibn ‘Abd al-Malik (r. 724–43) when he moved the caliphal seat to Rusafa during his reign and constructed his qubbat al-khadra there. Abu Ja‘far al-Mansur, the second Abbasid caliph, built himself a qubbat al-khadra as well when he planned his round city of Baghdad. His structure was so monumental it could be seen from outside the city. It also had on top of it a figure of a mounted horseman with a long spear in his hand.64 Al-Mansur’s qubbat al-khadra was probably an appropriation of an already established architectural sign of dominion and authority and an assertion of the recently won transfer of that authority from the Umayyads to the Abbasids. The known qubbat al-khadra halls belonged to palace/mosque complexes that formed the administrative nucleus of the early Islamic centres and symbolised the visual and spatial expression of the rulers’ power. Modern scholars have traced the green dome’s symbolism to the wide range of meaning given to domes in two precursors of Islamic architecture – the Byzantine and Sasanian traditions.65 Arabic sources are silent on the iconography of the Umayyad halls, but the green dome of alMansur in Baghdad, which collapsed in 941 long before the sources at our disposal were

composed, is recognised as ‘the emblem of Baghdad and its crown, and the triumph of the Abbasids’.66 Similar praise for the Abbasid qubba appears in many medieval sources, which indicate that the memory of the green dome has survived through the literary and artistic tradition well into the Mamluk period although its actual examples were long gone.67 The green dome of the Great Iwan, whose prominent position is emphasised by all contemporary sources, appears to have been a deliberate reference to the image of the qubbat al-khadra, for otherwise green domes were unknown in medieval Egyptian architecture. The only mention we have of a pre-Mamluk audience hall with a dome is that of the Fatimid Great Iwan in the Eastern Palace of al-Qahira, which had a gilded, not a green, dome under which the caliph sat on audience days.68 The revival of the green dome as a recognisable royal sign may have been one way of forging the link with the early Islamic caliphate and dissociating the Mamluk ceremonial and official image from that of the Fatimids.

CONCLUSION If al-Nasir Muhammad’s hall was ultimately a descendant of early caliphal models, why then did a shift occur in terminology from qubba (the Umayyad and Abbasid term for the known audience halls) to iwan? The sources do not offer any explanation, but the prevalence of the latter term over the former in late thirteenth-century Cairo may have been related to a local, contemporary development.69 The meaning of the word qubba had evolved in such a way that it was abandoned in the secular architectural vocabulary of Egypt around that period and migrated to the domain of funerary architecture. The use of domes in both secular and commemorative architecture in the Islamic lands, as elsewhere, obviously served the same purpose as a symbol of preeminence

–137–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

and sovereignty. But in medieval Cairo, with the diffusion of domed mausolea for religious and public figures, the funerary connotations of the word qubba had become so dominant that it eventually obscured its secular association.70 The word iwan replaced it in naming the throne hall at the citadel primarily because it had never been applied to commemorative or funerary structures, and because its connotations, which were appropriate for a royal structure, remained fairly consistent throughout the medieval period. From early Islamic times, iwan, as a term for an entire structure, designated a commemoratively highly charged palace, the Iwan Kisra. The word’s meaning developed over time to encompass all types of audience or reception halls, but it maintained ceremonial and royal connotations. This is most probably why the structures of al-Ashraf Khalil and al-Nasir Muhammad were called iwans in the sources. They conveyed an

image of royal grandeur appropriate for the ceremonies for which they were built. If some of the palaces built in Syria between the eighth and thirteenth centuries can be shown to have perpetuated the basilical model of the Umayyad palaces, then the Great Iwan of al-Nasir Muhammad would only be the last of a line of development. However, if the palaces of post-Umayyad Syria did not continue the basilical tradition and were instead analogous to the development in palatial architecture in the neighbouring regions, Iraq, Jazira and Egypt, then al-Nasir Muhammad’s Great Iwan can be interpreted as a revival of the early caliphal tradition. Given the uniqueness of its layout among the known medieval Islamic palaces, we can conclude that it was a consciously historicising structure through which al-Nasir Muhammad wanted not only to reintroduce an early form but also its well-established associations with a caliphal golden age.

–13 8 –

12. WRITING THE HISTORY OF ISLAMIC ARCHITECTURE IN CAIRO

A

mong the cities usually associated with Islamic civilisation, Cairo is perhaps the most ‘Islamic’ in its outlook and image, and decidedly the richest in monuments. Its architectural legacy spans the entire gamut of styles, from the seventh through to the twentieth century, that we now call Islamic. Founded at the strategic head of the Nile delta as the new capital of the country after the Islamic conquest in 641, the city, originally called alFustat, grew by annexing its northern satellites. In 969, a rectangular camp was established for the invading Fatimids, also to the north, and soon became their new capital. It was called al-Qahira (the Victorious), whence came Cairo, its Italian distortion. Later on, a citadel, built by Salah al-Din (Saladin) in the 1170s on an outcrop to the east, became the hub of a newly walled area comprising the two older urban centres. The Mamluk period (1250–1517) produced a wealth of monuments that synthesised the achievements of previous ages and symbolised the image of the city for centuries to come. After this most spectacular epoch, Cairo was reduced to an Ottoman provincial capital until the end of the eighteenth century. Then the city witnessed a short and capricious renascence under its

independently minded ruler Muhammad ‘Ali (1805–1848), followed by a period of vacillation between conservatism and modernisation that is still with us, which was exacerbated by the late twentieth-century population explosion and its concomitant urban degradation.1 Yet Cairo still preserves many of its Islamic monuments (456 registered by the 1951 Survey of the Islamic Monuments of Cairo), although the number is dwindling at an exceedingly alarming pace. The destruction, brought about by a mixture of neglect, greed and chaotic expansion, went on almost unnoticed until the earthquake of October 1992, which focused the attention of international organisations on some of the perennial problems of architectural conservation in Cairo and prompted some studies that have yet to be implemented. All the while, Cairo’s architectural gems remain largely unknown to the world’s architectural community. This, however, is no fault of the city dwellers, its architects, planners and decision-makers, or even the international conservationists. It is the outcome of an historical process whose roots date back to the seventeenth and eighteenth centuries and the genesis of the field of architectural history when architectural historians of the

– 141 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

time were unwilling to integrate Islamic architecture within the general discourse on the history of architecture.2 The initial rejection, which stemmed from cultural arrogance, led to the polarisation of knowledge during the colonial period and the relegation of the study of Islamic architecture to area specialists who collected, analysed, codified and classified its specimens and presented them as examples of a seemingly endogenous and insular architectural tradition. Nowhere has this been truer than in the case of Cairo, for the effects of the biased conceptual frameworks of European historians and the modern drive to compartmentalise knowledge were compounded by the traditional cosmocentric attitudes of the native scholars. The particularist tendency is discernible from the early Islamic period but, due to a number of historical and geopolitical reasons, Egyptians’ sense of territorial and national identity intensified in medieval times. It led to the development of a novel historical genre in which these feelings found expression through careful and meticulous descriptions of everything that pertained to Egypt, and to a greater extent Cairo as the metropolis of the country. The genre later acquired the name khitat (the word means the planned quarters in an urban setting, hence its use to designate what we could call topographical/historical studies), probably after the influential book of Taqiyy al-Din al-Maqrizi, al Mawa‘iz wa-lI‘tibar bi-Dhikr al-Khitat wa-l-Athar, which, unfortunately, was never fully translated into any European language.3 Composed between 1415–42, this encyclopedic work records with loving care each and every street and important structure in Cairo, and to a lesser degree other Egyptian cities, and produces their description and the history of everything connected with them. Most modern histories of Cairo relied heavily on al-Maqrizi’s data and many, especially those written by Egyptians, even adopted his methods and reflected his idiosyncrasies

by considering the city’s architectural history an autonomous development. One nineteenth-century author, ‘Ali Pasha Mubarak, consciously modelled his massive compendium al-Khitat al-Tawfiqiyya al-Jadida, and even named it, after al-Maqrizi’s Khitat.4 In his book, Mubarak updated al-Maqrizi’s material and added sections on the urban developments in Cairo between the fifteenth and nineteenth centuries.5 The nineteenth century brought another intellectual tradition to bear on the study of the architecture of Cairo: the European empirical method of documentation and analysis of the arts and material culture. Its first application in Egypt, and perhaps its majestic paradigm, is the Description de l’Égypte, which was the joint effort of more than 100 scientists, scholars and engineers recruited by Napoléon when he embarked on his expedition to Egypt (1798–1801). Published in Paris between 1809–28 in nine volumes of text and twelve in-folio volumes of illustrations, the Description de l’Égypte is a catalogue raisonné of sorts that formed the basis of the modern understanding of Egypt, its Pharaonic and Islamic patrimony as well as its contemporary conditions. The two volumes on modern Egypt contain a selection of measured plans, perspective drawings and analytical details of important Islamic monuments that offer the first visually comprehensive and typologically codified analysis of Cairo’s architectural heritage.6 The immediate successor in the tradition of the Description’s recording method was by Pascal Coste, a French architect from Marseille who worked in Egypt for ten years between 1817 and 1827, and who published an impressive survey of Cairene architecture after his return to France, L’Architecture Arabe ou Monuments du Kaire mesurés et dessinés de 1818 à 1826.7 This book, which focuses exclusively on Islamic monuments, is conceived and presented visually and graphically in the grand tradition of the Beaux-Arts, where Coste was trained, and is

– 14 2 –

W R I T I NG T H E H I ST ORY OF I SL A M IC A RCH I T E C T U R E I N C A I RO

prefaced by a concise historical introduction and short, descriptive paragraphs to explain the drawings. After Coste’s study, several books on Cairo’s architecture were published in the nineteenth and early twentieth centuries, but none was as exhaustive architecturally or as speculative analytically. One study, Martin S. Briggs’s Muhammadan Architecture in Egypt and Palestine, stands out for its conscious attempt to explore the links between the architecture of Cairo and the Syro-Palestinian cities which were historically and culturally close to it, like Jerusalem, Damascus and Aleppo.8 Then came the most comprehensive and most architecturally correct study – the lifetime work of a diligent Englishman, K. A. C. Creswell, who single-handedly measured, photographed, researched and drew all known Cairene monuments up to 1311, and published his work on them in two heavy volumes which, with their two predecessors on early Muslim architecture, established the field of Islamic architecture on solid, definite grounds.9 Creswell’s systematic technique of investigation was perfect for the formal evaluation of buildings, but in his drive for methodical rigour and quantitative accuracy he overlooked or misread many subtleties about the architecture and its context. Thus, his deterministic proclamations have become superseded by subsequent scholarship, although his plans and architectural descriptions still furnish the basis for any serious study of Cairene architecture. Furthermore, Creswell’s passion for things Egyptian, his linear architectural chronology and the extensiveness of his documentation have indirectly furnished a number of scholarly pretexts for later arguments of an endogenic Cairene architectural tradition, despite his own fanatic preoccupation with ‘architectural origins’ and formal parallels between buildings regardless of their geographical or cultural connectedness.10 Hence it was only natural that the work of the first generation of the univer-

sity-trained Egyptian architectural historians would exude an aura of self-centredness and particularism, for, even if we set aside their nationalistic or religious tendencies, these authors were greatly influenced by three exclusivist scholarly currents: the long Egyptian literary tradition of khitat, the more recent European epistemological framework which resulted in a field of studies based on the binary oppositions of us/them, the West/ the Orient (generally named Orientalism), and the idiosyncratic but powerful Creswellian paradigm which was Egyptiancentric probably for reasons opposite to those of its Orientalist predecessors or contemporaries (with the exception perhaps of Pascal Coste). All Egyptian studies reflect this mixed intellectual ancestry in varying combinations but the legacy of the khitat genre dominates their discourse and permeates their language. In fact, some of these studies are so affected by the two old Egyptian khitats that they come across as essentially new renditions of them with a visually oriented outlook, such as the books of Hasan ‘Abdel-Wahab and that of Su‘ad Maher which deal with the mosques of Cairo and Egypt respectively.11 Other authors are more analytical, though they still depend on the khitat’s data and syntax. Some even extend the scope of their analysis in order to establish links with preIslamic architectural traditions around the Mediterranean and other Islamic developments outside Egypt. Their works diverge from the chronological or topographic order of khitat-inspired studies and adopt one of two more inclusive methods. The first is the typologically structured survey; its most successful representative is Ahmad Fikri’s three-volume book on Cairene religious architecture.12 The second is writing architectural history as a historical narrative; examples include the works of Creswell’s student Farid Shafe‘i, who at times engaged in a polemic against his former tutor, ‘Abd al-Rahman Zaki and Ahmad ‘Abd al-Razzaq Ahmad.13

– 14 3 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

The last addition to the corpus of studies on Cairene architecture is Doris Behrens-Abouseif’s recent book with a selfdescriptive title, Islamic Architecture in Cairo: An Introduction. The book is divided into two parts. Part one offers three concise chapters presenting background material on the city’s history and expansion, the stylistic evolution of its architecture and its domestic architecture (though this last one has many omissions). Part two is a survey of sixty-one, mostly religious complexes and monuments arranged in six chapters, roughly corresponding to the usual periodisation of Cairene history from the seventh to the nineteenth century, with an entry for each structure. The entries vary in length from half a page to five pages, with the shortest appearing for monuments built after alMaqrizi’s time and Creswell’s cut-off date of 1311. The book appears as an ideal textbook designed for a course on Islamic architecture in Cairo aimed at non-Arabic speakers, which is indeed a staple course at the department of Arabic Studies in the American University of Cairo, and which has been taught in the past by Dr BehrensAbouseif herself. However, a number of errors, exclusions of important structures and the absence of a general map in the text diminish its value as a teaching tool.14 The book undoubtedly fills a gap in the English architectural history library. No handy review of Cairo’s Islamic monuments had appeared since Martin Briggs’s study of 1924 and Mrs Devonshire’s Rambles in Cairo of 1931. Even the second edition of Creswell’s Muslim Architecture of Egypt, published in 1969, which was still incomplete chronologically and bulky for the quick perusal, went out of print very fast. But, despite the fact that Dr Behrens-Abouseif’s book is written in English, it still corresponds more to the model of recent Egyptian architectural history books than to their Western counterparts. This conformity is discernible in two interrelated aspects. First, Dr BehrensAbouseif, like her Egyptian peers, depends

on the Khitat’s of al-Maqrizi and Mubarak as primary sources of information, and sometimes misinformation. Al-Maqrizi is summoned for almost every discussion of a building dating before 1432 (a hefty twothirds of the book’s inventory), while Mubarak’s citations dominate the rest. Secondly, an impression of endogeny and self-sufficiency pervades Dr BehrensAbouseif’s analysis of the evolution of Islamic architecture in Cairo, which is similar to the one encountered in the works of ‘Abdel-Wahab, Maher, Shafe‘i and Zaki. The origin of this attitude is evidently traceable to the influence that the khitat’s paradigm exerted on recent Egyptian historiography not only regarding the information but also on the structural and ideological levels. But the persistence of a particularist inclination is more surprising in the case of Dr Behrens-Abouseif than in that of other Egyptian authors. Unlike most of them, she is familiar with the work of Western authors who have recently proved the existence of an active architectural interaction between Islamic Egypt and other regions, both in the Islamic world and beyond, such as Michael Meinecke and J. M. Rogers. She even lists their articles in her bibliography. The same applies to Creswell’s elaborate discussions of precedents and influences, for despite Behrens-Abouseif’s reliance upon his plans and architectural descriptions, little traces of his opinions appear in her book. Even in the sections on pivotal structures whose architectural provenance is undoubtedly foreign, such as the Fatimid city gates (pp. 67–72) and the complex of Sultan Qalawun (pp. 95–100), scholarly debates on the sources of their structural and decorative elements are summarily diffused through generalisations on the common Byzantine heritage in the Eastern Mediterranean. This tendency to project an independent Egyptian architectural evolution throughout Cairo’s history is troubling for it affirms and perpetuates the antiquated epistemological construct in which

– 14 4 –

W R I T I NG T H E H I ST ORY OF I SL A M IC A RCH I T E C T U R E I N C A I RO

self-conscious architectural identities (notably the Western one) can be interpreted to have historically developed with little indebtedness to external influences and interactions. Thus, despite its accessibility to a Western audience, its chronological completeness (as opposed to Creswell’s voluminous albums with their cut-off date in the year 1311), and its clarity of presentation and illustration, and perhaps because of all these positive characteristics, Islamic Architecture in Cairo: An Introduction is a missed opportunity. It is so because it does not try to introduce Cairene architecture into the disciplinary discourses of either architecture or architectural history, precisely

because it does not address their interest in formal connections and cultural linkages, respectively. Clearly, to expect a single, introductory book to right the wrong propagated by long-established and partial epistemological structures, or to respond to the varied needs of different audiences is an exacting and excessive requirement. But the history of Cairene architecture still awaits a critical treatment that presents architecture as a product of culture, history and environment, that pursues explanations of historical phenomena beyond set boundaries, and that challenges the discipline of architectural history to integrate what has been left out in the continuous project of rewriting the history of architecture.

– 14 5 –

13. THE IDEOLOGICAL SIGNIFICANCE OF THE DAR AL-‘ADL IN THE MEDIEVAL ISLAMIC ORIENT

M

edieval Islamic architecture presents the scholar with a fascinating set of historiographical problems. Some are methodological, others are related to the nature of the sources, and they are shared by various other branches of research in medieval history such as urban history, topography and the history of arts and crafts whose inquiries depend on the same sources. Still others are peculiar to the specific domain of architectural history. These last are the most challenging, for they require particular strategies that take into account the disparities in our knowledge of the two basic components needed to reconstruct the history of any architectural object: the physical remains and the contemporary documents related to them. There are three possible kinds of disparity: in the best cases, buildings that are still standing and in fairly good shape can be studied in light of relevant contemporary documents. In more difficult cases, the structures still exist, but supportive documents, written or otherwise, do not. Most difficult of all is when we have documents describing, or referring to, a structure or a group of structures for which we have no visible trace.

The medieval structure known as the Dar al-‘Adl belongs to this last category. This unique Islamic institution, which may be best translated in today’s context as ‘palace of justice’, was initially conceived for the qada’ al-mazalim service, that is, for the public hearings held once or twice each week and presided over by the ruler himself or his appointed deputies to review and redress grievances submitted by his subjects.1 The earliest known dar al-‘adl was built c. 1163 by Nur al-Din Mahmud ibn Zengi in his capital Damascus, and the last one was constructed by the Mamluk Sultan al-Nasir Muhammad ibn Qalawun at the Citadel of the Mountain (Qal‘at al-Jabal) in Cairo in 1315 (it was rebuilt in 1334). Four more dur al-‘adl are known to have been constructed between these two dates: one in Aleppo in 1189 by al-Zahir Ghazi, the son of Salah al-Din, and three in or next to the Citadel in Cairo beginning in c. 1206 and ending when the last one was erected by al-Nasir Muhammad. After this no more dur al-‘adl seem to have been built until modern times, when the palace of justice was introduced. The period when dur al-‘adl flourished corresponds to the age of Crusader and

– 14 6 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

Mongol attacks on the central Islamic lands. The geographic area in which they appeared is also well defined. The three cities in which they were built had been capitals of separate realms from the late eleventh to the mid twelfth century, but by 1171, they had all fallen under the control of Nur alDin, the first unifier of the Islamic front against the Crusaders. After Nur al-Din’s death, the three cities were taken over by Salah al-Din Yusuf ibn Ayyub (1174–1193), Nur al-Din’s general, his deputy in Egypt and his nemesis in his last years, and were integrated into his empire. They became loosely united and hotly contested Ayyubid capitals after Salah al-Din’s death, and, later, centres of Mamluk government, with Cairo leading as the sultan’s seat. From Nur al-Din’s time onwards, the three cities had been joined by a highly popular common cause: fighting the Crusaders. The Zengids, Ayyubids and early Mamluks were intensely engaged in the counter-Crusades, and the latter also repelled the Mongol forays into Bilad al-Sham. No trace of any of the five palaces of justice remains today. Their existence is known to us only from textual references, except for the last one, torn down in 1825, for which we have several plans, façades and a few views drawn by European visitors to Cairo in the eighteenth and nineteenth centuries. Nor can archaeology help in investigating the history of dur al-‘adl, for the areas the buildings stood in have been built over several times. The written sources, do, however, provide a substitute for this lack of material remains, because for Egypt and Syria, the medieval period is unusually rich in historical writing. Annals, biographical compendia, manuals for the chancery, geographical treatises (masalik) and topographical tracts (khitat) all exist in abundance for the period between the twelfth and the fifteenth centuries.2 The problem is that, numerous as they are, medieval Islamic sources rarely deal directly with architec-

ture or urban projects. Scattered in the texts, however, are little details about dates of buildings, locations, patrons, the reasons for building, the functions and ceremonies that took place in the structures once they were built and passing remarks about specific spaces or features inside them. By collating and analysing these clues, or indices, one can reconstruct a historical account of the dur al-‘adl and propose an explanation for both their emergence and their disappearance. The dar al-‘adl was not simply a development of the qada’ al-mazalim institution. It was an original product of an extraordinary time: the period of the counter-Crusade and the ideological revival that went with it, as ideal Islamic qualities were promulgated by both the ruling and religious classes and demanded by the people. The dar al-‘adl visually represented one of these qualities, justice, and provided the rulers with a forum to publicly claim their adherence to proper Islamic codes. That religious ardour, ignited by external threats and internal schisms, had withered away by the middle of the fourteenth century, after the Crusader and Mongol offensives had been thwarted and Egypt and Bilad al-Sham had been securely united under the rule of the Mamluks. The dar al-‘adl, along with other innovations of the period, such as the fada’il of the Holy Land and jihad literature, lost its raison d’être and vanished altogether when the circumstances that prompted its development had passed.

THE DEVELOPMENT OF QADA’ AL-MAZALIM

THE

The qada’ al-mazalim or al-nazar fi-l-mazalim, which literally means ‘to consider or to look into acts of injustice’, is an Islamic judicial institution with a complex history. It has been identified as the organisation that ‘brings the litigants to an agreement by fear and prevents the contestants from rebuffing

– 147 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

judgment by awe. It is a position that combines the authority of the ruler and the impartiality of the judge.’3 This definition implies that the supervisor of the mazalim must be a person who has greater executive power than a judge: he is able also to enforce his decisions. Most Islamic sources attribute the conception of the qada’ almazalim to some revered early caliphs, such as ‘Ali, Mu‘awiya, ‘Abd al-Malik and ‘Umar b. ‘Abd al-‘Aziz, or even to Sasanian, preIslamic precedents.4 But some modern scholars have questioned whether the formal division between simple general qada’ and the qada’ al-mazalim appeared so early. They argue that during both the Rashidi and Umayyad periods, there was no clear-cut separation between the political and judicial authorities. The caliphs and their provincial governors (wulat) were also judges. They could exercise their judiciary authority either directly in public or delegate it to an appointed qadi.5 It was not until the consolidation of the Islamic legal corpus that became known as the shari‘a in the early Abbasid period (the second half of the eighth century), that a practical need for religiously qualified judges had arisen. Gradually thereafter, the caliphs had to cede their judicial authority to the qadis. But because they considered themselves to be the leaders of all Muslims and the successors (khulafa’) of the Prophet Muhammad, the Abbasids upheld the claim to be the protectors of the shari‘a and the ultimate administrators of justice.6 This political motive helps to explain why the qada’ al-mazalim was detached from the nascent judicial system and retained under the caliph’s jurisdiction: it was seen as one of the symbols of the Abbasid right to rule.7 The third and fourth Abbasid caliphs, al-Mahdi (r. 775–785) and al-Hadi (r. 785–786), both supervised mazalim sessions in person. They also introduced the office of sahib al-mazalim, which was occupied either by a high-ranking official or a special qadi responsible for the regular sessions of maza-

lim.8 The next step was the institutionalisation of qada’ al-mazalim, proposed by Abu Yusuf (d. 798) in his book al-Khiraj, which he wrote for the fifth Abbasid caliph, Harun al-Rashid (r. 786–808).9 Abu Yusuf, who was the qadi al-qudat (chief judge) of al-Rashid, recommended that the caliph regularly preside over mazalim sessions, obviously not to undermine his own position, but because he saw the legitimising potential of supervising mazalim sessions and advised his caliph accordingly. That political function was not lost on the ambitious usurpers of Abbasid authority, whether they were in the provinces or in the capital, for they almost always took over the qada’ al-mazalim as part of their acquired power. Thus, Ibn Tulun (r. 868–884), the first governor of Egypt to break away from Baghdad successfully and to expand his domain into Syria, was also the first to hold mazalim sessions regularly.10 The Shi‘ite Buyids, who overpowered the caliphate and its Iranian provinces, including Baghdad between 945 and 1055, and reduced the Abbasid caliphs to figureheads, took control of the qada’ al-mazalim and passed it on to the Shi‘ite sharifs, whom they considered to be the imams.11 In the middle of the eleventh century, the qada’ al-mazalim became for the first time an integral part of a comprehensive political theory of Islamic rule. In his al-Ahkam al-Sultaniyya, Abu al-Hasan al-Mawardi (974–1058), a legist and a high officer in the Abbasid court, developed a full discussion of mazalim jurisdiction and its relationship to qada’, and decreed it to be one of the fundamental duties of Muslim rulers or their appointed deputies.12 It has been suggested that al-Mawardi’s theoretical formulations had direct political relevance.13 al-Mawardi’s career fell between two phases in the turbulent history of the Abbasids, the Buyid dominance and the Seljuq sultanate. He served two successive caliphs, al-Qadir (r. 911–1031) and al-Qa’im (r. 1031–75), who were trying in the ensuing hiatus to regain some of the caliphate’s

– 14 8 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

political authority, and was directly involved in that attempt. Seen in this context, alAhkam al-Sultaniyya represents a model structure for a true Islamic government and reflects the high hopes of the period that such a goal was finally at hand. But the Abbasid resurgence project did not fully succeed. The Seljuqs did wrest Iraq from the Buyids and restored the caliphate and Sunnism to religious supremacy, but they obliged the caliphs to delegate their political power (sultan) to them. Nizam al-Mulk al-Tusi (r. 1063–1092), the great vizier of the Seljuq sultans Alp Arslan (r. 1063–1072) and Malikshah (r. 1072–1092) who ingeniously planned their administration, realised the imperative political function of mazalim sessions and stressed the importance of regularly convening them in his treatise on politics, Siyasat-Namah.14 Some of the Seljuq sultans must have heeded Nizam al-Mulk’s advice, for they are reported to have held mazalim sessions, with their viziers and qadis present, while others delegated the responsibility to their viziers. In the late Abbasid period, when the caliphate managed for a short while to recoup some of its lost political authority with the weakening of the Seljuq overlords, individual caliphs, such as al-Muqtafi (r. 1136–1160) and al-Nasir (r. 1180–1225), resumed the duty of selecting mazalim supervisors, and one, al-Mustanjid (r. 1160–70), is even reported to have presided over mazalim sessions personally.15 The practice was maintained, at least informally, by many rulers of other Islamic states, especially of the Zengid, Ayyubid and Artuqid dynasties, whose dominions were carved out of the vast Seljuq empire and who inherited many of its political and bureaucratic structures. Sitting for two days a week to look into acts of injustice seems to have become part of the ruler’s routine. Sessions were held in different places depending on ruler and locale, but usually took place in a major hall in the palace of government.16

THE DAR AL-‘ADL OF DAMASCUS In the late twelfth century, Nur al-Din Mahmud ibn Zengi (r. 1146–1174) introduced an innovation that had no precedent in Islamic history. He built a special palace for mazalim sessions in Damascus, and named it the dar al-‘adl or dar kashf al-mazalim (house of justice or of mazalim’s inquest).17 We know very little about this structure for it is rarely mentioned in the sources and it entirely disappeared around the middle of the seventeenth century.18 This led a recent study to doubt its very existence on the basis of uncertainties regarding its location, history, and contradictory reports on exactly where Nur al-Din held his mazalim sessions.19 But all the chroniclers who report the building of the Dar al-‘Adl, including the great historian of Damascus Ibn ‘Asakir (1105–1176) who was Nur al-Din’s contemporary, are clearly speaking about a specific structure.20 They do not record the date of its construction, however. Some scholars have suggested that it must have been built shortly after Nur al-Din took the city in 1154, but the wording of the sources does not support this suggestion.21 Dar al-‘Adl was certainly built after 1160, possibly around 1163 when the prince finally settled on Damascus as his capital after a period of domestic political troubles in which he constantly moved between Damascus and Aleppo, his old seat.22 Nur al-Din sat in his Dar al-‘Adl to review mazalim at least twice a week. Salah al-Din also held mazalim sessions there when he was in Damascus, albeit intermittently.23 Later Ayyubids apparently maintained this custom, for there are at least two references to princes holding sessions in the Dar al‘Adl, one in 1195, the other in 1198.24 Early Mamluk sultans, who based their legitimacy on their claim of loyalty to the Ayyubid sultan of Egypt, al-Salih Najm al-Din Ayyub (r. 1240–1249), all headed mazalim sessions in the Dar al-‘Adl whenever they were in Damascus.25 Otherwise, the vicegerent

– 14 9 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

(na’ib) of the city was the one who regularly presided over mazalim sessions there as part of his normal duties.26 In the mid-Mamluk period, however, textual evidence becomes more complicated because the contemporary sources start using two names, Dar al-‘Adl and Dar alSa‘ada, interchangeably to designate what appears to be the same complex, without specifying how or when they merged.27 The Dar al-Sa‘ada (the Palace of Felicity) was the vicegeral palace in Damascus throughout the Mamluk period. It was initially called the Dar Farrukhshah, after a nephew of Salah al-Din’s who was vicegerent (na’ib) in Damascus until his death in 1182. It was ostensibly maintained as the private residence of his son al-Amjad Bahramshah until 1230.28 Its transformation from private property to the na’ib’s official residence in Damascus and its connection with the Dar al-‘Adl are not very clear.29 After Bahramshah’s death, the palace inexplicably passed to his cousin al-Ashraf Musa, the king of Damascus (r. 1229–1237), who deeded it to his only daughter Malaka Khatun upon his death in 1237.30 The sources are silent about the palace’s legal status thereafter, but it presumably passed to the Mamluk state treasury at some point, perhaps in 1287, when Malaka Khatun was striped of some of her possessions, though the palace had already been the official vicegeral residence for twenty years before that date.31 The Dar al-Sa‘ada appears to have preserved its official aura from the days when it was the residence of Salah al-Din’s na’ib Farrukhshah, even when it was still privately owned in the late Ayybid period. The Ayyubid kings of the city normally lived in the palaces of the citadel but twice at least they moved to Dar al-Sa‘ada as a temporary residence whenever the sultan, whose capital was usually in Cairo, was in town.32 This notion of hierarchy in domicile whereby the citadel’s palaces were reserved for the sultan and the Dar al-Sa‘ada for the malik (king)

was apparently institutionalised in the early Mamluk period, probably as early as the reign of Qutuz (r. 1259–1260).33 The sultan and his retinue were housed at the citadel whenever they came from Cairo, until al-Zahir Baybars constructed a royal palace in 1269, the Qasr al-Ablaq (the Striped Palace), to the west of the city along the river Barada, which was reserved for the sultan’s visits.34 The Dar al-Sa‘ada was designated as the vicegeral palace, where the na’ib resided and conducted business, while the citadel became the base of a different official, na’ib al-qal‘a, who reported directly to the sultan in Cairo. Though not recorded in any source, the Dar al-‘Adl must have been incorporated with the Dar al-Sa‘ada during that time, for its function complemented those under the jurisdiction of the city’s na’ib. After the merger, Dar al-‘Adl seems to have been used for a wide range of services, such as the ceremony of pledging allegiance (ba‘a) to the sultan, the reception of foreign envoys and the appointment of officials and qadis, although the sitting for mazalim sessions remained predominant among all these uses.35 The site of the Dar al-‘Adl is not mentioned in any of the sources, but the Dar alSa‘ada’s location is well established.36 The entire governmental complex, including the Dar al-‘Adl, occupied a block south of the citadel, across the street from its southern postern, and slightly to the east of the city’s Bab al-Jinan (Gate of Gardens), which later became known as the Bab al-Nasr (Gate of Victory), and still later, in the Mamluk period, as the Bab Dar al-Sa‘ada (fig. 56).37 The sources supply little information about the architecture of Dar al-‘Adl as an independent unit inside the na’ib’s complex.38 We know only that it was a grand qa‘a that had an iwan, in the centre of which sat the na’ib when the court sessions were convened; next to him was an empty seat covered with yellow silk (the official colour of the Mamluk sultanate) representing the sultan.39 These references allow us to suggest that the Dar

–15 0 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

‘ ‘ FIG. 56. The location of the Dar al-‘Adl in Damascus

al-‘Adl might have been similar to most private and princely reception halls of the period, for they had qa‘a plans with one, two, three or even four iwans flanking a central space.40 The form of the iwan and the appearance of the qa‘a are impossible to determine since the two words were used generically in medieval times to designate any type of arched opening and hall respectively. The Dar al-‘Adl was destroyed by fire and rebuilt a few times along with the rest of the Dar al-Sa‘ada, until the latter was moved to another place, extra-muros, at the end of the sixteenth century. By that time the term dar al-‘adl ceased to be used by chroniclers to refer to any part of the na’ib’s palace.

THE DAR AL-‘ADL OF ALEPPO Although Aleppo had been Nur al-Din’s capital before Damascus, it appears that neither he nor Salah al-Din after him established any palace for justice there.41 Aleppo’s Dar al-‘Adl, the second such palace after that of Damascus, was not started until 1189. It was built by the Ayyubid king alZahir Ghazi (r. 1186–1216), the son of Salah

al-Din, who ruled the city during his father’s reign.42 Located south of the citadel, which was being refurbished at the same time, the Dar al-‘Adl was a separate structure integrated within what may be considered the extension of the citadel’s royal complex outside the walls and towards the city. In what seems to have been a planned act, Ghazi enclosed the Dar al-‘Adl and the surrounding area, which contained the maydan (training field) built earlier by Nur al-Din for equestrian exercises, between the old walls of the city and new ones he had added especially for that purpose further to the east. The character of exclusivity of this new royal enclosure was conveyed through the control of its access and the establishment of a private passage leading to it from the royal palaces in the citadel (fig. 57). At the citadel end of this passage, Ghazi constructed a new gate, called the Bab al-Jabal (the Gate of the Mountain), while the gate at the other end was appropriately called the Bab Dar al-‘Adl. This passage was reserved only for the king when he rode out of the citadel to the Dar al-‘Adl or the city beyond. The petitioners coming from the city used another gate called the Bab alSaghir (Little Gate) to gain access to the

– 1 51 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

‘ ‘

FIG. 57. The location of the Dar al-‘Adl in Aleppo

enclosure of the palace. Construction of all the gates and fortifications around the Dar al-‘Adl ended in 1214.43 It is clear that the construction of the Dar al-‘Adl was initiated by Ghazi as a major step in a general plan of organisation devised to renovate the Citadel of Aleppo and to redefine its relationship to the city. The king, his officials and his army resided in the citadel or around it, while the population was restricted to the city extending to the west at the foot of the hill. The Dar al-‘Adl represented the transitional zone between the city and the citadel where the ruler and the ruled meet, and it visually and spatially underlined the pivotal role of the king in the state administration. This urban programme should be seen in the larger context of al-Zahir Ghazi’s reign: his involvement in the insidious world of Ayyubid politics, his constant struggle to

maintain his independence and to consolidate his hold on his territories, and his policy of administrative centralisation.44 From the beginning, Ghazi used his Dar al-‘Adl for many court ceremonials other than mazalim sessions; these included the reception of foreign envoys and learned debates among scholars presided over by the king. The day after his death, 200 slaves he had freed as an act of devotion were assembled in Dar al-‘Adl before they were to be released.45 This report indicates that the structure also played a role as a way station between the citadel and the city. The official mourning services after Ghazi’s death and the assembly that gathered to discuss the arrangement for the transition of power to his son took place in the Dar al-‘Adl as well.46 Ghazi’s two successors, al-‘Aziz Muhammad (r. 1216–1237) and al-Nasir Yusuf (r. 1237–1260), are reported to have

–152 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

maintained the custom of holding mazalim sessions in the Dar al-‘Adl twice a week, on Mondays and Thursdays.47 We know next to nothing about the appearance of the palace. Ibn Wasil (d. 1298) states in a passing reference that the caliph’s envoy to Ghazi was accommodated in the iwan of the Dar al-‘Adl.48 As in Damascus, the Dar al-‘Adl in Aleppo was built next to the citadel, which constituted the government centre and the royal residence in both cities. Like the Dar al-‘Adl of Damascus, the Dar al-‘Adl of Aleppo was enlarged in the Mamluk period to include the residence of the na’ib of Aleppo and ultimately acquired the name Dar al-Sa‘ada as well.49 It was used for various court ceremonials, among which the biweekly mazalim ceremony remained prominent.50 However, unlike Damascus, where the spatial link between the citadel and the palace of justice was ambivalent, the relationship of the Aleppine Dar al-‘Adl with the royal complex was made more pronounced by enclosing it between two parallel city walls and joining it to the citadel by a direct passage. This line of development reached its logical conclusion in Cairo. After two trials, which applied similar solutions to the one used in Aleppo, the last Mamluk Dar al-‘Adl was moved into the southern enclosure of the citadel, which was simultaneously rearranged and partitioned to accommodate both private and public functions of the sultan.

DAR AL-‘ADL AL-K AMILIYYA Of the three major capitals of the Ayyubid realm, Cairo, Damascus and Aleppo, Cairo was the last one to have its own Dar al-‘Adl. We have no straightforward reference to the building of a Dar al-‘Adl in Cairo prior to the coming of the Mamluks in 1250, but a few remarks indirectly indicate that one existed in Ayyubid times. When two Mamluk sources report the disobedience of

an amir in 1310 who barricaded himself in his residence, they refer to that structure as the Dar al-‘Adl al-Kamiliyya, and place it inside the citadel.51 This attribution to the Ayyubid al-Kamil is very plausible, for we know that he was the one responsible for the completion of the citadel and its endowment with palatial and administrative structures.52 The Dar al-‘Adl seems to have been located in the narrow end of the northern enclosure where today it meets the southern enclosure not very far from the Qulla Gate which separated the two enclosures (fig. 58). It probably belonged to the first stage in the construction of the citadel’s administrative section since we know that, in the Mamluk period, this area contained many other administrative buildings such as the Dar al-Niyaba (vicegeral palace) and the Qa‘at alSahib (hall of the vizier). It is impossible, however, to fix a date for the construction of any of al-Kamil’s structures since the sources speak of them only in passing without even providing their names. There is no mention of the Dar al-‘Adl al-Kamiliyya having been used as a palace of justice during al-Kamil’s time, although the name implies that it was, but during the reign of al-Salih Najm al-Din Ayyub, al-Kamil’s second son, the structure appears to have been used as it was intended. This is deduced from a vague account stating that in 1239, Sultan al-Salih delegated the authority to hold the sessions of mazalim in an unidentified dar al-‘adl to a triad of military men ( jund). Two of them are further identified: the first was al-Sharif Shams alDin, the judge of the army (qadi al-‘askar), and the second, named al-faqih ‘Abbas, was the preacher (khatib) of the citadel’s mosque.53 Both must have resided in the citadel in order to attend to their work, which implies that the dar al-‘adl they sat in may have been that of al-Kamil. The structure was apparently neglected after al-Salih’s death, for the sessions of mazalim were held in the madrasas he had built in Fatimid alQahira (1239–45) rather than at the citadel

–153 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

‘

‘

FIG. 58. The location of the three Dur al-‘Adl in Cairo

during the reign of the first Mamluk Sultan al-Mu‘izz Aybak (r. 1250–1257). They were presided over by the amir Aydakin alBundaqdari, Aybak’s vicegerent and the master of the future sultan al-Zahir Baybars, who was assisted by a number of qadis and administrators.54 The Dar al-‘Adl al-Kamiliyya was turned into a residence after the end of the Ayyubid period, for the sources we have speak of it as the living quarters of an amir who had official duties that required his permanent presence at the citadel.55

DAR AL-‘ADL AL-ZAHIRIYYA The second Cairene Dar al-‘Adl was the one built or renovated by the Mamluk sultan alZahir Baybars. His two biographers, Ibn Shaddad and Ibn ‘Abd al-Zahir, give inadequate reports about this structure, although they elaborate on his ardour to uphold the principle of justice and to attend mazalim sessions. Ibn Shaddad speaks only of a fenced, square mastaba (platform) installed in the middle of the court in front of the citadel gate, covered by a canopy to protect it

–15 4 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

from the sun and the rain and designated for the public sittings of the vicegerent and the vizier. In the same list of structures, he mentions a dar al-‘adl under the citadel, without attributing it to Baybars.56 Ibn ‘Abd al-Zahir reports in 1262 that Baybars ordered the renovation and remodelling of a building under the citadel and the establishment of the dar al-‘adl in it, suggesting that the structure existed prior to Baybars.57 Casanova, who wrote a history of the Citadel of Cairo, demonstrated that the structure was a mausoleum (turba) of a Fatimid family of princes, the Banu al-Muhtar, but mistakenly assigned its refurbishing and transformation into the Dar al-‘Adl to al-Kamil without any historical basis.58 Baybars is the patron who restored that turba, which was probably in disrepair like many other Fatimid remains, put it to a new use as the Dar al-‘Adl and added the canopied mastaba in front of it for less formal ceremonies.59 He is credited in the sources with sitting in the Dar al-‘Adl on Mondays and Thursdays, both to inspect mazalim petitions and to review the Mamluk army.60 After Baybars’s death, his structure became known as the dar al-‘adl al-qadima (the old), and was occasionally used for official events presided over by high-ranking administrators.61 It was eventually converted into the Tablakhana (Drummery, the place where the military band plays at specific hours as a sign of royalty) in 1322 during the reign of al-Nasir Muhammad.62 The site of Baybars’s Dar al-‘Adl is difficult to ascertain today, after the major changes in the topography of the area during the reign of Muhammad ‘Ali, when the new carriage route was completed in 1825. AlNasir’s Tablakhana, or Baybars’ Dar al-‘Adl, was reportedly located between the Bab alSilsila (Gate of the Chain) and the Mudarraj Gate.63 The Gate of the Chain (probably the present Katkhuda Gate) was the main royal entrance to the citadel from the maydan. The Mudarraj Gate, also named the Citadel Gate, was the public entrance to the citadel

and stood at the end of a stepped path that ascended from a spot near the Gate of the Chain to an elevation 40ft (13m) above the maydan. Baybars’s Dar al-‘Adl was somewhere along this path of nearly 600ft (200m), probably in a spot high enough to permit the sultan seated in the royal stand to review the parades that took place in the maydan. A brief reference in the waqf of the zawiya of Hasan al-Rumi, built in 1522, further establishes the location of al-Nasir’s Tablakhana, or Baybars’s Dar al-‘Adl. The waqf, dated to 1535, states that the Tablakhana was above the zawiya, which still stands today on the eastern side of the road leading to the original Mudarraj Gate.64 The waqf also specifies that the zawiya is situated between the Mudarraj Gate and the Chain Gate, exactly the site given in al-Maqrizi’s Khitat as that of al-Nasir’s Tablakhana. Baybars’ Dar al-‘Adl, then, might have stood where Muhammad ‘Ali’s Dar al-Mahfuzat (Archives Administration) stands today.

THE IWAN OF THE CITADEL By 1280, when Qalawun acceded to the throne, Baybars’s Dar al-‘Adl had ceased to be the setting for mazalim sessions. It may have been replaced by the iwan, which was the principal royal audience hall in the citadel, but the date of the transfer is not known. The iwan is first mentioned in the sources at the time of Baybars’s crowning ceremony in 1259.65 Chroniclers do not credit anyone with its building, but its most probable patron is al-Kamil, as he is the only sultan who undertook major works at the citadel. Baybars, however, built a domed hall as a replacement of, or addition to, the earlier iwan.66 In 1284, Qalawun ordered the demolition of Baybars’s hall to build a new one, named the Iwan al-Mansuri (of Qalawun).67 Al-Ashraf Khalil rebuilt, or perhaps only refurbished, the iwan of his father as the sources are very imprecise about the kind of work he ordered at the citadel.68 This last

–15 5 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

structure was in turn destroyed by Khalil’s brother and successor, al-Nasir Muhammad, to be replaced by his famous hall, the Iwan al-Kabir al-Nasiri (the Great Iwan of alNasir). The Great Iwan is better known to us. We have its plan, elevations and sections as documented by the savants of the French Expedition (1798–1801). We also have its location on the map of the Description de l’Égypte, where it is labelled as le divan de Joseph.69 From that map we can note that the Great Iwan stood where the court extending to the north-east of the Mosque of Muhammad ‘Ali towards the Police Museum is located today (fig. 59). The successive building and rebuilding of the main audience hall in the citadel in such a short period could be explained by the zeal of the sultans to be the patrons of the most visible, and most publicly accessible, structure in the citadel. We know that

the hall of Baybars was not intended as a Dar al-‘Adl, since there was another one outside the citadel. We know also that the Great Iwan of al-Nasir Muhammad was the stage of mazalim sessions twice a week during his reign. The sources are silent as to whether Qalawun or his son Khalil ever used their halls as settings for dar al-‘adl, although we encounter a few references to the practice as being sporadically observed during their reigns.70 In the case of Qalawun, we can assume that he never personally presided in this office, for the sources tell us that his Arabic was very poor, which would have prevented him from communicating with his subjects on dar al-‘adl days.71 He, on the other hand, sat in his iwan to review troops and distribute warrants of iqta‘ to amirs and soldiers, or to receive foreign dignitaries and ambassadors on official occasions.

FIG. 59. The Location of the Divan de Joseph in the Citadel of Cairo as reproduced in the Description de l’Égypte

–15 6 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

AL-IWAN AL-K ABIR AL-NASIRI

Al-Nasir Muhammad came to the throne as a boy of eight in 1294. His long reign was interrupted by two periods of usurpation (1295–1299, and 1309–1310), and it was not until 1310 that he finally became the supreme ruler of the Mamluk sultanate. In 1311, less than a year after he assumed the rule for the third time, al-Nasir demolished the iwan of his brother Khalil and built a new one.72 At the same time, he decided to regularly preside over mazalim sessions in the new iwan, which became known as the Dar al-‘Adl: a fundamental shift in policy from the one established under Qalawun, Khalil and even al-Nasir himself in his first two reigns when mazalim supervision was the duty of the vicegerent.73 He also elevated the event to a formal ceremony where all the important amirs of the realm and the members of the sultan’s inner circle had to be present and seated around him in a set order. These ceremonial innovations had their immediate roots in the political circumstances of the time.74 They tallied with the implementation of changes in the structure of the Mamluk hierarchy and the consolidation of the sultan’s role at its apex as if to engender them and give them their physical manifestations. When al-Nasir Muhammad returned to the throne in 1310, he orchestrated drastic shifts of power to ensure his throne. He pitted the strong amirs against each other, replaced most of them with his own Mamluks and systematically weakened the authority of many topranking officers by assuming some of their duties himself. Consequently, he not only controlled the internal affairs of the state, he effectively became the state. As such, al-Nasir surpassed all his predecessors in Mamluk Egypt, who were extreme autocratic rulers themselves. At the beginning of his reign, the sultan was also interested in strengthening the support he had enjoyed among the common people (al-harafish or

al-‘amma) of Cairo and Damascus in his struggle to maintain his throne.75 Appearing as a ruler concerned with the fair application of justice helped al-Nasir in maintaining this positive popular sentiment towards him. It is from within these two sets of considerations, dominating the Mamluk hierarchy and pleasing the populace, that rebuilding the iwan and holding mazalim sessions in it should be seen.76 The Great Iwan of al-Nasir Muhammad was a monumental stone structure made to impress and to inspire awe. The illustrations of the Description show it to have been open with arcades on three sides: the north-east, which formed its main façade, the southeast and the north-west. The southwestern side was built up with a thick wall pierced with five doors which led to the sultan’s private quarters (al-dur al-sultaniyya) through the dihliz al-‘ubur, or the passageway, behind the iwan.77 The plan of the iwan consisted of a wide, central aisle flanked by two lateral ones formed by rows of red granite columns which were taken from ancient Egyptian temples. The central aisle was surmounted by a huge wooden dome covered on the outside with green tiles. That dome – which had already collapsed when the French drawings were made – had been the most striking feature of the iwan.78 A broad inscription band, with characters made of large gilded-wood units, ran around the perimeter of the inner square under the dome. Its text seems to have consisted of the full titulature of al-Nasir Muhammad and probably the construction date.79 We have a detailed account of the dar al-‘adl ceremony during the reign of al-Nasir Muhammad written by Ibn Fadl-Allah al-‘Umari (1301–1349), who was a high administrator at the court.80 On dar al-‘adl days, usually Mondays and Thursdays except in Ramadan, al-Nasir would come out of his inner palaces through the vestibule behind the iwan and enter through the central door with the muqarnas conch.81 He would sit on a wooden chair covered with a

– 1 57 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

silk cloth (dast) next to his marble throne, in the centre of the iwan’s back wall (fig. 60). The marble throne, which resembled the minbar of a mosque, was only used on official occasions when foreign envoys were received.82 The sultan’s place lay at the apex of a concentric circle of dignitaries surrounding him in a strict hierarchical order. Nearest to him were those officials directly involved in the proceedings. To the right were the four supreme judges (qudat al-qudat) of the four Sunni schools of jurisprudence in the order of their importance: the Shafi‘ite judge closest to the sultan, followed by the Hanafite, the Malikite and the Hanbalite. Next to the Hanbalite judge came the treasury controller (wakil bayt al-mal), then the market inspector (muhtasib) of Cairo. To the left of the sultan sat his secretary (katib al-sirr), who between 1329 and 1332 was Ibn Fadl-Allah al-‘Umari himself, 83 followed by the army supervisor (nazir al-jaysh). The circle would be completed by a group known as the clerks of the chair (kuttab al-dast), after the royal dast, who sat facing the sultan. Their job was to record the minutes of the sessions. These functionaries were probably seated under the dome, with the sultan close to the centre, some distance from the back wall, as both court protocol and the sultan’s safety required sufficient space behind him for two rows of guards to his right and left, the silahdariyya, jamadariyya and the khassakiyya Mamluks (fig. 61). The great amirs of hundred, the highest rank in the Mamluk system, were seated in a row opposite each other on either side of the sultan and some fifteen cubits (approximately 15ft) from him. They were called amirs of the council (umara’ almashura), and they functioned as the sultan’s official advisers. There were twenty-four of them in al-Nasir Muhammad’s army, so they probably sat twelve on each side (fig. 62). The less important amirs of forty and other civil servants would be placed further away from these high-ranking amirs, completing the rows towards the entrance to the

iwan, but these amirs and administrators had to remain standing. Behind this first row stood several other rows of amirs of ten and of Mamluks. Attendants and clerks of the chancery formed the outermost circle around the three open sides of the iwan. The rows of important amirs probably reached as far as the first row of columns which supported the dome; the lesser amirs and Mamluks stood in the space between the inner and outer rows of columns. The attendants stood in the wide space in the front of the iwan and escorted the petitioners there to face the assembly (fig. 63). The order of seating in dar al-‘adl sessions matches almost exactly the plan of the Great Iwan (fig. 64). Whether this means that the dar al-‘adl’s ceremony was designed to follow the logic of the iwan’s spatial arrangement, or that al-Nasir Muhammad rebuilt the iwan to accommodate the ceremonial he had introduced we do not know. But the latter alternative is the more plausible one since the plan of the iwan was radically different from the common hall type in Islamic Egypt, generally known as a qa‘a.84 It may have been inspired by its four direct predecessors at the Citadel in Cairo, but this is impossible to ascertain for we have no idea what they looked like. Yet analysis of the iwan’s architecture suggests that, although it appears to be a synthesis of a variety of elements taken from existing and familiar structures, its plan bore a manifest resemblance to a specific type, the basilica, with its central nave and two arcaded side aisles. In the Great Iwan, however, the typical basilical plan is modified by opening the sides to provide an unobstructed view to the outside and to suggest the accessibility of the sultan sitting within, who could be seen from all sides when he sat for mazalim sessions on dar al-‘adl days or for embassy receptions.85 The formal affinity of the Great Iwan’s plan with the basilica type may have been a consequence of a functional one, for although basilicas are usually connected with early Christian churches,

–15 8 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

FIG. 60. Diagram of the sultan’s entry into the Great Iwan on the Dar al-‘Adl days

FIG. 61. Diagram of the circle forming around the sultan on the Dar al-‘Adl days

FIG. 62. Diagram of the circle around the sultan and the Amirs of Hundred

–159 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 63. Diagram of the complete seating layout on the Dar al-‘Adl days

the original Roman functions as a royal hall of justice and for public audiences symbolism of this widespread type was never lost or forgotten.86 The Great Iwan began to lose its prominence as the official Dar al-‘Adl during the undistinguished rules of al-Nasir Muhammad’s twelve powerless epigones who succeeded one another in a frantic turnover between 1341 and 1382. Their vice-

gerents became the effective power brokers in the sultanate, and their residence in the northern enclosure of the citadel, the Dar al-Niyaba, which was rebuilt in 1343, became the real centre of government.87 Officially, the Great Iwan had remained the throne hall where coronations and receptions of foreign envoys took place, but the day-to-day reviews of troops, the administration of iqta‘ and the biweekly dar al-‘adl

–16 0 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

FIG. 64. Plan of the Iwan with the seating layout of the Dar al-‘Adl superimposed on it

sessions were transferred to the Dar alNiyaba. With the advent of the Burji period, the Great Iwan regained some of its glory for a short while. After he acceded to the throne, Barquq (r. 1382–1389, 1390–1399), the first Burji sultan, started to sit in the iwan for dar al-‘adl sessions, probably as a sign of kingship and as an attempt to associate himself with established royal customs. The definitive shift in focus came upon in 1387, when Barquq completely broke with the Qalawunid tradition. He replaced the

Great Iwan as the setting for dar al-‘adl with an unspecified place in the royal stables (most probably the hall called al-Harraqa), and changed the days of the service to Tuesdays and Saturdays.88 This choice was probably dictated by Barquq’s first official position as amir akhur (stable master), and by the general mistrust that dominated this struggle-ridden transitional period, when controlling the stables meant blocking the movement in and out of the palatial area of the citadel. The ceremonial changes

– 1 61 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

instituted by Barquq signalled his intention to restructure the sultanate, and his introduction of a new site and different days for the ceremonies consecrated these changes. The convention of holding mazalim sessions in the royal stables was intermittently followed by Barquq’s immediate successors, but eventually the dar al-‘adl ceremony was downgraded, and at times totally suspended. The Great Iwan was still occasionally used to receive foreign embassies, undoubtedly because its size and spatial arrangement made it the most impressive structure at the citadel. Otherwise, it too had fallen into disuse by the middle of the Burji period, although a few sultans attempted for short periods to revive the biweekly review of Mamluks (khidma) in it, and at least two, Barsbay and Qaytbay, had it restored.89 The Great Iwan was still standing in ruins at the beginning of the nineteenth century, when it was documented for the Description de l’Égypte. It was razed by Muhammad ‘Ali in 1825 to clear the ground for his new mosque.90

INTERPRETING THE DAR AL-‘ADL PHENOMENON The Great Iwan of al-Nasir Muhammad was the last Dar al-‘Adl built in Islamic central lands. Although qada’ al-mazalim continued to be a duty of Muslim rulers, they appear to have reverted to the old practice of holding its sessions in a non-specific hall in their palaces. This conclusion leaves us with a few historical puzzles, such as why did Dar al-‘Adl come into existence in the first place? And why did it fade out of use 160 years after it was introduced? Why did it appear only in the three capital cities of Bilad al-Sham and Egypt? Elucidating these questions would help us better situate this unique and peculiar Islamic institutional structure in its wider cultural context. As already noted, building dur al-‘adl coincided with the height of the Crusader

and Mongol attacks on the Islamic world. The five structures were established in the Syrian and Egyptian capitals of the Islamic states that conducted the counter-offensives to these attacks: the Zengid, Ayyubid and Mamluk sultanates. The founders of these dur al-‘adl were all rulers of non-Arabic origin who belonged to a recently Islamised and staunchly Sunni military caste that dominated the political scene in the Eastern Mediterranean after the eleventh-century Seljuqid expansion.91 They led armies made up mainly of Turkish and Kurdish free and manumitted cavalry, expanded their principalities through war, conquest and intrigue, and distinguished themselves in jihad against a host of enemies: the Byzantines, splinter Shi‘ite groups, the Crusaders and, later, the Mongols. It is probably no historical coincidence that the first known Dar al-‘Adl was built by Nur al-Din, the first organiser of an Islamic front against the Crusaders. It is no coincidence either that the last palace was constructed by al-Nasir Muhammad, in whose early years of rule the Crusaders finally were driven out of the Orient, with the conquest of the island of Arwad off the Syrian coast in 1302, and the Mongols checked in 1303 on their last incursion into the country until Tamerlane’s invasion at the end of the fourteenth century. Evidently, however, the relationship between the upsurge of jihad and the building of dur al-‘adl was not simply causal or reciprocal. After all, many great warriors of the period, such as Salah al-Din and Qalawun, did not sponsor any such structures, and at least two dur al-‘adl builders, al-Zahir Ghazi and al-Nasir Muhammad, were better known for their diplomatic skills than their fighting abilities. The explanation for the connection lies in the wider context of the religious awakening in that time of intense ontological crisis. The fierce encounters between Christian Europe and the Islamic East during the Crusades generated a combative religious passion among

–16 2 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

all classes of society in the Islamic Orient. It also heightened the rulers’ awareness of their ideological obligations and, at the same time, provided them with a political platform around which popular approval and support would surely coalesce. A number of energetic rulers, notably Zengi, his son Nur al-Din, Salah al-Din and later Baybars, Qalawun and al-Ashraf Khalil, were able to channel the immense moral repercussions of the Crusades and Mongol attacks to mount their counter-offensives. They, and others, also consciously used their image as defenders and supporters of Islam to advance their political agenda both on the external and internal fronts.92 The emphasis on their achievements in furthering the cause of Islam can be observed in the diplomatic letters they sent to announce their conquests, to ask for military, financial and logistic support, to berate their opponents and competitors, and to request diplomas of investiture (taqlids) from the Abbasid caliphs in Baghdad, who represented the ultimate legitimising authority in the land of Islam. The letters of Nur al-Din, Salah al-Din and later Ayyubids addressed to the Abbasid caliphs in Baghdad are replete with references to their jihad, their veneration of the caliphate and readiness to defend it, and their emphasis on applying the shari‘a rules in their realms.93 Similarly elaborate discourses on jihad, shari‘a and the requirements of Islamic rule appear in the khutbas of investiture delivered and the taqlids written by the titular caliphs installed in Cairo by Baybars after the fall of the Abbasid caliphate in Baghdad. The khutba of investiture written in 1261 by the first Egyptian Abbasid caliph, al-Mustansir II, recognised Baybars not only as the sultan of Egypt and Syria but also as the universal sultan of Islam, the deputy of the universal caliph and the leader of jihad.94 From that point on, the Abbasid caliphs, kept in Cairo with no real political role to play, were employed as legitimising figureheads in ceremonies of investiture throughout the

Mamluk period, and even, in some instances, as tools in the hands of the sultans to bestow religious recognition on the rules of allies elsewhere in the Islamic world.95 The early Mamluk sultans also used the caliphal ratification of their rule, along with their jihad credentials, to boost their position in their correspondence with their bitter enemies, the Ilkhanid Mongols, during the late thirteenth and early fourteenth centuries.96 They deliberately contrasted their services to Islam and their instituted legitimisation by its utmost legal authority, the caliph, with the Ilkhanid history of destroying the caliphate and wrecking havoc in its eastern territories. This image must have been quite effective, not only in the Mamluk sultanate itself, where we have ample evidence of the pride felt by the Mamluk intelligentsia in the pivotal role played by their rulers in defending Islam but also, and most surprisingly, among some of the intellectuals who served in the Ilkhanid court. Sharaf al-Din Wassaf al-Hadra (1264–1330), the Persian historian who dedicated his treatise, Tajziyat al-Amsar wa Tazjiyat al-A‘sar, to the Ilkhan Öljeytü (r. 1304–1317), dared in the same text to praise the Mamluks for their steadfastness in jihad and their adherence to the tenets of Islam.97 On the home front, although most Ayyubid and Mamluk rulers of the period depended heavily on their repressive power, they still sought acceptance by, and perhaps popularity among, their subjects. To this end, they allied themselves with, employed and patronised the religious class, including both the learned fuqaha’ and popular shaykhs who sometimes served them as propagandists and apologists.98 They endowed civic structures to prove their piety and their support for religious activities: madrasas to educate a new class of fuqaha’, ribats and khanqahs to lodge the Sufis, and mausolea to commemorate themselves and to aggrandise their deeds. They actively publicised their enactment of religious regulations regarding

–16 3 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

social organisation and relied heavily on elaborate titulatures that stressed their religiously glorifying actions and qualifications which they inscribed on their buildings and objects.99 The introduction of new attributes in royal protocols is a strong case in point. Nur al-Din’s titulature changed drastically early in his reign, following the decisive battles he won against the Crusaders between 1146 and 1150.100 When the titles in his inscriptions are compared with those of his father and predecessor, what is notable is that, unlike his father’s protocol which mixed Turkish, Persian and Arabic titles, his were all Arabic and concentrated on Islamic virtuous traits as seen through the prism of traditional Sunnism. Out of the thirty-eight preserved inscriptions on structures built by Nur al-Din, the epithet al-mujahid (the jihad fighter) is present in sixteen. The title al-‘adil (the just), which was his regnal title, is to be found in all of them.101 There is no doubt that the adoption of these two titles by Nur al-Din was intended to accentuate his qualifications as a good Muslim ruler. His building of the first Dar al-‘Adl at the same time should be regarded as part of the same concern. Like later rulers, who all adopted titles with some reference to justice, such as al-‘adil or muhiyy al-‘adl fi al-‘alamin (the reviver of justice in the world), building a palace of justice was a magnificent propaganda tool. It was intended as another of their legitimising acts.102 ‘A just ruler is a legitimate ruler’ seems to have been the slogan embodied in the building of a palace of justice and the establishment of a ceremonial for its usage. The masses driven by religious fervour in that period of danger to Islam and to its lands was prone to appreciate such an endowment and laud its patron. This is evidenced not only in the special places reserved in Islamic historiography for the rulers who combined jihad and justice, such as that assigned to Nur al-Din as the exemplary just ruler comparable only to the

Rashidi caliphs,103 but also in the tales that developed from the popular lore concerning the same princes. Even today, we have songs, epics and stories that celebrate the heroism and justice of Nur al-Din, of Salah al-Din and of al-Zahir Baybars.104 By the time al-Nasir Muhammad acceded to the throne for the third time in 1310, the Crusaders were already routed and the Ilkhanid Mongol menace had been repeatedly thwarted; the Mamluk sultanate had finally achieved political maturity and regional supremacy. Al-Nasir’s third reign, stable and prosperous, proved to be a turning point in Mamluk history and in the character of the Mamluk state. Unlike his predecessors, al-Nasir was a better tactician and diplomat than a fighter and leader of armies. He preferred alliances and clientage bonds and, at times, relied on fidawiyya (hit men of the Isma‘ili sect) to eliminate his political opponents. He used limited military force only on rare occasions to reach a prominent position among the rulers of his time. He had his name pronounced in the Friday khutba and sometimes struck on coins in various regions in North Africa, Nubia and Anatolia without sending in his troops (both acts were considered signs of recognition of sovereignty).105 Because both external and internal threats had been removed and because alNasir relied mostly on negotiation and intrigue in his foreign policy, the military function of the state and the emphasis on the role of the ruling Mamluks as the warriors of Islam were slowly softening.106 The first sign of change surfaced in the 1330s when al-Nasir Muhammad formed a new circle of hand-picked great amirs who, contrary to established procedures, were not all accomplished fighters. The two most influential among them, Qawsun and Bashtak, came to Egypt as free men and sold themselves to the sultan who raised them to the highest ranks without their having to endure the prerequisite military training.107 Other signs of disintegration were soon to

–16 4 –

T H E I D E O L O G I C A L S I G N I F I C A N C E O F T H E D A R A L - ‘A D L

follow, but the effect of this shift on the elite’s character did not spread among the entire Mamluk class until the Burji period, for the structure of the army remained more or less intact during the reigns of al-Nasir’s twelve Qalawunid successors. Mamluks were still bought at a young age and lodged in special barracks at the citadel where they were put through rigorous training and a thorough religious education that inculcated upon them a military mentality and respect for Islamic tenets before they were manumitted and enlisted in the army.108 Barquq maintained the same strict programme that governed the Mamluks’ training during his first reign, which marked the transition between the Bahri and Burji periods, but relaxed it tremendously in his second reign. From then on, Mamluks were permitted to live in the city and to fraternise with the local population through marriage and business transactions. The system deteriorated even further after Barquq, when new Mamluks were brought at a fairly advanced age, after their character had already been formed, and were no longer required to undergo an extensive religious education before their manumission. Consequently, the once fiercely proud and strictly segregated Mamluks began slowly to adopt an urbane culture in which few of their glorified military and political attributes were still operative. By the beginning of the fifteenth century the Mamluks’ acculturation was discernible not only in their attitudes, tastes and preferences, but most of all, in the lack of interest they showed in emphasising their jihad role or their pro-religion stances in their political image making.109 Their contemporary observers no longer saw in them the deserv-

ing leaders they once were, skilfully and thoughtfully managing a great empire and fighting for the cause of Islam.110 It was during that same period that the dar al-‘adl ceremony lost its significance and the Great Iwan succumbed to neglect and was abandoned. This interpretation of the dar al-‘adl phenomenon is clearly influenced by the scope and nature of our written sources. Since they mostly wrote about the political, military and religious state of affairs, with the social and material conditions touched upon only as they became relevant to the narration of political events, our view of the same period is conditioned by these idiosyncrasies. It has to be stressed, however, that the general historical and ideological context of the period and the almost perfect concurrence of the upsurge in the jihad movement with the appearance of dur al-‘adl strongly support the politico-cultural interpretation. Finally, it is appropriate to note that the only short-lived attempt to revive the role of the Great Iwan as a Dar al-‘Adl during the Burji period was initiated by the one sultan, Barsbay (r. 1422–1437), who also tried to instigate a revival of the jihad mentality among his troops in order to conquer one of the last Crusader footholds in the Orient. Barsbay sent three successive flotillas against the Kingdom of Cyprus in 1424, 1425 and 1426, the last of which occupied the island, captured its king Janus, and brought him to Cairo where the sultan imposed on him tough terms of vassalage. Barsbay is also reported to have refurbished the Great Iwan in 1427 and reinstituted the dar al-‘adl ceremony in it for a short while in 1431.111 Was it just a coincidence?

–16 5 –

14. ‘AJIB AND GHARIB: ARTISTIC PERCEPTION IN MEDIEVAL AR ABIC SOURCES

T

he visual arts occupy a tiny corner in the prodigious medieval Arabic historical output of the period between the twelfth and fifteenth centuries. The written sources, which are on the whole expansive and rather chatty, hardly ever mention artists, their artworks or their reception by their contemporaries in a manner similar to the ways they speak about poets, udaba’, musicians and scholars and the meaning and impact of their work. In those very rare instances when the sources mention art objects, they do it in the context of royal patrons rather than artists, usually in lists of gifts sent by rulers to foreign sovereigns or objects looted or confiscated from fallen dignitaries and amirs.1 Even then, they focus primarily on the monetary value and functional aspects of the objects and almost always refrain from giving an opinion about their form, look or composition – in contrast to their typical expert judgement of literary work, be it poetry or prose. Moreover, where specific art objects are noted, the texts reveal a lack of familiarity with the most elementary visual vocabulary which is otherwise known from lexical works or from philosophical or optical treatises.2 They do not go beyond exclaiming the ‘ajib or

gharib, that is the marvellous, wonderous and extraordinary qualities of a painting or an object, or observing extra-artistic attributes such as the real jewels adorning representations of women in one instance or the ranks affixed above the images of Mamluk amirs in another.3 The venerable classical and medieval Islamic philosophical tradition that propounded some sophisticated reflections on art and beauty does not seem to have penetrated the chronicles and the biographical dictionaries that constitute the two main types of historical sources for the medieval period. Not even the towering Ibn Khaldun and his student al-Maqrizi, who otherwise show a keen interest in conceptual and theoretical questions, include in their work any discussion on artistic subjects such as representation or beauty as it was developed in the philosophical Islamic tradition. Likewise, the encyclopedist kuttab-historians – such as Ibn Shaddad, al-Nuwayri, Ibn Fadl Allah al‘Umari and al-Qalqashandi – who cover in their exhaustive and vast compendia all the theoretical sciences and practical skills a successful adib needs, such as literature, history, epistolography, cosmography, geography, botany, zoology, religion, law and politics – find no place for philosophy and its concomitant topics including aesthetics.

–16 6 –

A R T I S T I C P E R C E P T I O N I N M E D I E VA L A R A B I C S O U R C E S

Ignoring the representational in art is to be expected in a tradition that shunned any attempt to express its ideals or to embody its values and virtues in figural representation. Early Muslim historians seldom notice the figural representation in their surroundings – in the form of ancient statuary and murals say, or contemporary modest attempts at depicting the human form in drawing or sculpture such as the few examples known from Umayyad Syria, Abbasid Samarra or Fatimid Egypt. However, their medieval successors go one step further by even failing to report the vigorous emergence of representational tendencies in Seljuq, post-Seljuq and early Mamluk art, which covered a wide array of art forms including reliefs, murals, sculptural ranks, coins, metalwork, woodwork, glass and miniature painting, and popular figurines, and which lasted well into the fourteenth century. This state of affairs is perplexing to modern students of medieval culture, for it is clear that the flourishing of representational art underlies an appreciation of its visual and symbolic meanings, at least among those who sponsored, acquired, displayed, and viewed and admired it. Yet the laconic information on art in the sources and the ways it was recorded imply that both authors and readers were little moved by art objects and the effect they had on their environment. One famous and often-quoted example is the comparatively long (all of five lines!) and detailed citation from al-Maqrizi on three specimens of realistic representational painting.4 The wording of the reference reinforces the general impression just mentioned of how rudimentary the understanding of the visual arts was among the literati/historians – those (such as alMaqrizi) who were very open-minded and were otherwise highly sensitive to other creative forms, i.e. architecture. While describing the Fatimid Mosque of al-Qarafa in his Khitat, al-Maqrizi pauses to marvel at a visual ‘ajiba in the painting of the intrados of

its arcade that, through apparently colouristic means, makes its surface appear like a three-dimensional muqarnas if seen from the centre and flat if seen from the side. Not uncharacteristically, he then shifts to a discussion of various painters who lived in the eleventh century and their chefs d’oeuvre, as reported in the unfortunately lost Tabaqat alMuzawwiqin, or Daw’ al-Nibras wa Anas al-Jullas fi Akhbar al-Muzawwiqin min al-Nas, the sole prosopography of painters mentioned in medieval sources.5 Al-Maqrizi reports a competition between two otherwise totally unknown artists, the Egyptian al-Qasir and the Iraqi Ibn ‘Aziz, which was sponsored by the Fatimid vizier al-Yazuri (r. 1049–1058) who, we are told, ‘loved painting and never tired of looking at illustrated books’. Ibn ‘Aziz boasted that he could ‘paint ( yusawwiru) a dancer that looks as if she was coming out of the surface of the wall’. Al-Qasir claimed that his dancer ‘will look as if she was going into the wall’, to which the listeners responded, ‘This is more wonderous (a‘jab).’ To demonstrate their claims, they used simple painterly techniques of contrasting colour in the foreground and background (black on white and yellow on red respectively), to give the image of a dancer framed within a niche the illusion of depth and motion either into or out of the painted surface.6 Al-Maqrizi’s comparison is certainly at the heart of the question of art as illusion of life and movement. Traditional views on Islamic art would have us believe that such an endeavour would have aroused religious misgivings on the part of this fiqh-trained author who would be expected on principle and by formation to oppose the representation of life in art.7 But al-Maqrizi’s text betrays nothing of the sort. No condemnation, no injunction, not even a trace of legal or religious polemic can be discerned. Instead, the text, especially when it describes the paintings themselves,8 is a simple, awkward and clearly inexperienced analysis. The choice of terms suggests that

–167–

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

al-Maqrizi was perplexed by this apparently unusual representational possibility and was uncertain how to deal with it or cast it into words. Representational art in this instance seems to have been an amazing, even an enthralling, visual feast, wonderful perhaps, but clearly not intellectually and aesthetically understandable or easily explainable. This superficial and somewhat amateurish approach to the visual arts is not limited to al-Maqrizi. It is shared by other chroniclers and biographers, such as Ibn ‘Abd al-Zahir, Ibn Shaddad, Ibn alDawadari, al-Safadi and al-‘Ayni, who record, in short and trite sentences, the instances when a sultan decorated his palace with representations of himself and his amirs and Mamluks or other figures, but fail to describe the images, reflect on their meaning, or register any objection to the practice.9 In none of the references is the blanket ban on figural representation even mentioned. On the contrary, incidents of effacing images or demolishing statues by some zealous individuals are reported with arguments for and against the legality of the action.10 In one extraordinary instance, al-Maqrizi straightforwardly condemns an act of destruction as hypocritial and insensitive to the majesty and beauty of the objects destroyed. The incident happened in 1379, when a certain Sufi shaykh by the name of Muhammad Sayem al-Dahr (the Eternally Fasting) tried to deface the feline ranks of Baybars inscribed on the Qanatir al-Siba‘ (Bridge of the Lions) in Cairo and the face of the Sphinx and claimed it to be his moral obligation. Al-Maqrizi ends the story with a moral that roughly translates, ‘Those who have reached the high rank they covet have but one purpose: to trick the people with all sorts of chicaneries (hiyal).’11 The prosaic attitude towards representational art encountered in medieval sources, therefore, seems not to have been the reflection of a legal or religious imperative. Some form of religious censure (tahrim) or aversion (karahiya) may have induced the

disposition to ignore figural art in early Islamic writing, or at least sustained it and gave it shape.12 However, this is not sufficient to explain the scope of the phenomenon in the medieval period when figural art itself underwent a real revival in a variety of media. Many other religiously prohibited practices, drinking and homosexuality for example, seem to have thrived unabated from pre-Islamic times despite the opprobrium attached to them. They flourished as literary topoi-khamriyyat (Bacchic poetry) and ghulamiyyat (love of youth) – if not as practice. The two types were interconnected and belonged to an extraordinary libertine literary tradition whose ups and downs depended on the religious mood of the time, but which never totally disappeared.13 In the medieval period, in fact, numerous scholars, including those definitely known to have had a religious inclination, such as Ibn ‘Abd al-Zahir and al-Safadi, excelled in detailed, realistic and highly informed descriptions of drinking sessions, homosexual trysts or wine itself, but there is no evidence that they either condoned, engaged in or even had a fleeting experience with any of them, although they were popular among certain classes.14 Not all religiously banned practices – and there is no definite evidence that figural representation ever fell into that category – were omitted from the repertoire of the literati interests and their intellectual, literary or recreational pursuits. Unlike the medieval khamriyyat and ghulamiyyat, which handle these literary types with subtlety and mastery despite their moral dubiousness, contemporary references to the visual arts are inept. Many authors resorted to poetry, a more familiar terrain, to describe paintings in terms of established literary tropes. Even more revealing, they appropriated verses by others to do their descriptions for them.15 This was not just a personal preference. It was probably a symptom of an intellectual rustiness or perhaps visual illiteracy, from which literati had suffered for a long time. Ever

–16 8 –

A R T I S T I C P E R C E P T I O N I N M E D I E VA L A R A B I C S O U R C E S

since a set of subjects and opinions about what was proper and what was not for a member of the literati to deal with began to take form in the mid ninth century and was fully formulated and universally embraced by the eleventh, the visual arts had been excluded.16 A medieval Islamic litterateur was expected to be well-versed in ‘ilm (religious sciences) and adab (general literary education), and, in the early period, in some form of philosophy (but that was almost totally banished by the eleventh century). ‘Ilm included Qur’anic and Hadith studies and its concomitant sciences of transmission and textual criticism. Abab involved proficiency in Arabic language and poetry, in Arabic and Islamic historical traditions, and in some secular sciences such as geography and cosmology. However, the medieval litterateur had no exposure to the visual arts and no training in art appreciation or concern for art history, aside from calligraphy, either through formal curricula or the informal agenda that governed educational and literary settings (madrasas, mosques, diwan, or even majalis al-adab and their long manuallike compendia). The uncertain handling of images and figures by al-Maqrizi and other historians thus illustrates a historical condition in which many generations of literati have found themselves with a visual-less education, initially prompted and perhaps later maintained by the religious abhorrence and rarity of images. Even the ostensibly technical terms frequently used in reporting any painting, ‘ajib and its analogue gharib,17 seem not to have been based on visual perception or artistic vocabulary. They were both probably borrowed or appropriated from literary categories whose elaborate discussions formed the basis of at least two genres – one in ‘ilm and one in adab – which were particularly common in the medieval period.18 The first is the branch of Qur’anic and Hadith studies that could be collected under the rubrics Gharib al-Qur’an and Gharib al-Hadith. The second could be termed al-‘aja’ib wa-l ghara’ib,

an adab genre that seems to encompass several interrelated subgenres from among the ones that dealt with natural and supernatural wonders: astronomy, astrology, zoology, mineralogy, geography, cosmology, paradoxology, mirabilia and miracula.19 ‘Ajib in this context is usually translated as wonderous, gharib as strange or singular. Together, they seem to span the scope of cognitive reactions to the extraordinary and unusual, with ‘ajib as the more encompassing term. Gharib was little used and may have been borrowed from its original lexicographic niche to function more or less as a rhyming complement to ‘ajib rather than as a denotative or connotative extension of it.20 ‘Aja’ib and ghara’ib motifs appeared early on in Arabic literature, at least since the early ninth-century treatise of al-Jahiz, al-Hayawan (the Animal), but the first systematic compilation on the subject was the famous book by Zakariyya al-Qazwini (1203–1283), ‘Aja’ib al-Makhluqat wa Ghara’ib al-Mawjudat, the prototype that influenced many medieval authors. Historians were all versed in the discourses of Gharib al-Qur’an and Gharib al Hadith literature, which constituted part of their philological, grammatical and Qur’anic training. Some of them, such as Dia’ al-Din ibn al-Athir and Ibn Hajar al‘Asqalani, even composed their own Gharib al-Qur’an or Gharib al-Hadith treatises.21 On the other hand, the popularity of alQazwini’s book, which was copied and illustrated many times in the medieval period, meant that the literati were also probably familiar with his definitions of the wonderous and extraordinary.22 Many Mamluk encyclopedists, including al-Nuwayri, Ibn Fadl Allah al-‘Umari and al-Qalqashandi, appear to have incorporated most of alQazwini’s material into those sections of their compendia dealing with cosmography and geography, and even devoted sections to the ‘aja’ib in nature.23 The literati were thus accustomed to think of ‘ajib and gharib in a literary sense as more or less technical terms designating rare and unfamiliar

–16 9 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

language in the Qur’an and Hadith. To paraphrase al-Qazwini’s definition, they were the expression of puzzlement vis-à-vis a thing or an event, not because it was unobservable but because it occurred so rarely or because its cause and/or effect were not readily graspable or because the way to react to it was unknown.24 Al-Qazwini’s ‘aja’ib and ghara’ib are natural and supernatural phenomena that are either observed directly, or reported by trustworthy authorities, or accepted and believed to exist because the scriptures said they did. They range from astral and celestial bodies, to the angels and demons and other heavenly spirits, to atmospheric and terrestrial phenomena, and finally to human beings, animals, plants and minerals, and even man-made objects and monuments. They are so intricate or extraordinary or rare as to escape immediate comprehension, hence their status as wonderous. In transposing the terms ‘ajib and gharib to painting, medieval authors seem to have combined the two ranges of connotation, the rare and the puzzling, perhaps because for them they were exactly what these images signified. Paintings were uncommon and, when encountered, difficult to comprehend since they had lain outside the cognitive range of the literati’s interests for such a long time. The literati’s reactions to them had to be verbalised using concepts and terminology that already existed for other well-charted intellectual and scholarly categories, which share some perspectival qualities with representational painting. Extraordinary natural phenomena, supranatural occurrences, myths of the ancients and stories of the prophets, all fell into the wonderous category. So did intricate literary and poetic inventions and paintings. But it was not the wonderous that induced further investigation or fired up the imagination. It was rather the explanation for the unexplainable, or, perhaps more accurately, the exclamation in front of the unexplainable.

That the literati did not expand their vocabulary to include notions and terms more appropriate to discussing painting – i.e. the professional terminology of the artists themselves or the more theoretical aesthetic terminology of philosophy, geometry, music and the like – suggests that they were either cut off from or not seriously trained in these fields. The absence of artistic or artisanal idiom reinforces the impression that literati and painters, and probably other artisans as well, belonged to different social spheres that did not easily and systematically communicate with each other.25 The dearth of philosophical, theoretical terms, on the other hand, points to an intellectual rather than a social impediment. It could be taken as evidence that the conservative elements among the ulama in late medieval times were successful in stamping out most of the suspect fields of al-‘ulum al‘aqliyya (philosophical sciences) from scholarly enquiry. Their specific propositions, concerns and even terminology ceased to be part of the literati discourse. Geometry and music, and to a lesser extent philosophy, were still taught and written about, though on a much smaller scale and sometimes in hiding to avoid the risk of being denounced by the establishment. They seem to have become so marginal that they no longer even furnished the proper vocabulary, as they once apparently had done, where it was truly needed, such as in aesthetic or visual appreciation.26 This is not to say that a religious aversion to figures and figural art did not exist. Quite the opposite: exegetical and legal treatises of the medieval period routinely reiterate an unbending position against them, citing the famous hadith against the musawwirun (a word that in this context can mean either painters or makers of figures of living things), whose final abode will be in hell, and elaborating on its implications.27 Figures may even have been banned and the ban customarily upheld in whatever milieu religious scholars controlled (mosques,

– 17 0 –

A R T I S T I C P E R C E P T I O N I N M E D I E VA L A R A B I C S O U R C E S

madrasas and the like), since no images existed in them. But this attitude should not be seen as either objective or collective. That is, the sources that report the ban or neglect art could not have been impartially recording the situation as it was (a claim that they never make, yet we always expect it of them), nor could their stance be representative of general attitudes towards art in the medieval society at large. Despite their apparent editing by the literati, several reports in the sources suggest that the two other main groups in the medieval society, the ruling elite and the common people, had different opinions on the question of figural art and the function of art in society in general. The military and administrative elite patronised and enjoyed figural art not only in their private residences and their illustrated books, but also in public spaces, such as royal palaces, hammams and citadels, and even in books that they endowed as waqfs. They sponsored figural paintings, reliefs and murals, sculptural ranks, metalwork with figures and miniature painting, in addition to textiles adorned with images. They even used images on temporary structures and models built for celebratory processions and festivals or, in a few instances, as warnings for people to desist from some prohibited public behaviour. An example of the first case was for the triumphant entry into Cairo in 1303 of al-Nasir Muhammad and his army after defeating the Mongols. Seventy model qal‘as (citadels) were built by the amirs along the road from the Bab alNasr to the Citadel of the Mountain.28 An example of the second case was Amir Manjak al-Yusufi’s putting images (suwar) of executed women (?) in 1351 on the walls of the city to dissuade Cairene women from wearing men’s cloaks.29 The common people, especially the urban riffraff, seem to have enjoyed and responded favourably to the public display of images, figures and unusual artefacts sponsored by sultans and amirs. They also seem

to have used similar representational techniques, though more crudely executed in cheaper versions, coupled with chants, slogans and zajal (popular strophic poetry) to communicate their hopes, fears and discontent, and convey sarcasm and mockery, or even mark some unusual events and holidays.30 A remarkable example, not unlike an instance of contemporary advertising, took place in the 1370s in Cairo when the commander of the royal harariq (warships), a certain Ibn ‘Abid, managed to construct a hoist to transport two particularly cumbersome marble columns from the citadel mount to the city. The event was considered so extraordinary that people composed and sang sonnets and made models of the hoist and even embroidered the machine on handkerchiefs and silk clothing, dubbed jarr al- ‘amud (the pulling of the column), just as T-shirts are routinely used today to promote events and companies.31 Another example was when a kawwaz (jug maker) parodied the entry of the famous Ayyubid prince, Abu al-Hayja’ al-Samin (the Fat), into Baghdad in 1197, by casting his corpulent form on his horse in a clay figurine. The figurine was soon duplicated all over Baghdad. The amir responded with an admirable good humour.32 However, not all popular figurines were made as joyous farces. On more than one occasion, the people expressed their rancour against fallen officials by making grotesque figures of them, along with demonstrations, looting their properties, and even sometimes desecrating their corpses. Both the formidable Amir Qawsun and alNashw, al-Nasir Muhammad’s hated supervisor of the privy purse (1333–1339), were portrayed in sugar figures (named ‘alaliq)33 at their executions: Qawsun was nailed on a camel, and al-Nashw was shown on the gallows.34 These examples suggest that there was some sort of a public dialogue between the rulers and the ruled in part carried on by figures and figural representations, despite the fact the literati appear to have been

– 171 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

oblivious to it.35 They record some of the instances in their chronicles, though they recast them in their language. They sometimes evince real and heartfelt appreciation of an artistic element that happened to catch their fancy, as al-Maqrizi did in his report. However, as historians, they do not seem to notice the eloquent potential of images and figures in conveying meaning. Again, there is no evidence that this was the result of socioreligious stands – such as the condemnation of figures – consciously adopted by a proper religiously minded literati. It appears rather as just another

aspect of the lack of visual acuity caused by the long neglect of philosophical reflection and artistic appreciation among the literati. These intellectual shortcomings and not the usually postulated collective religious attitude is what characterised how the literati experienced, appreciated and dealt with representational art. They formed part of their mental and social structures. They are the most conspicuous simply because they conditioned how art is reported in the historical sources. Modern studies of medieval Islamic art have to take this into account whenever they use the written record of the period.36

– 17 2 –

15. THE FORMATION OF THE NEO-MAMLUK STYLE IN MODERN EGYPT

T

he pretext that the ‘lure of the Orient’ lay behind the interest Europeans took in the lands and cultures that lie east and south of Europe was critically challenged by Edward Saïd in his seminal book Orientalism.1 He, and subsequent authors who took up his arguments, assert that innocent curiosity and pure, detached scholarly exploration do not suffice to explain all of the vast enterprise of collecting, processing and interpreting data on the non-European world that has come to be referred to as ‘Orientalism’. Cultural, political and economic dominance, or the justification of it in a new world order, largely loom in the background as other motives for the surge in Orientalist activities in late eighteenth- and nineteenthcentury Europe.2 The nineteenth century began with the first European military interventions in the East and ended with most of it under colonial rule. The infiltration began in India by several European powers and culminated in the British elimination of the Mughal empire in 1857 and the establishment of colonial rule in the subcontinent. The expedition of Napoleon Bonaparte to Egypt in 1789–1801, though brief, ushered in the age of Westernisation and modernisation and

opened the eyes of Europeans to the riches of the ‘Near’ East as well as to its relative political weakness.3 The French occupation of Algeria in 1831 introduced the practice of appropriating the land and denying the country its national identity. It was followed by the gradual occupation of the whole of North Africa by France, Italy and Spain. Egypt’s turn came in 1882, when it was seized by Britain after decades of manoeuvering and machination. Finally, the last surviving Oriental power and Europe’s ‘sick man’ for a century, the Ottoman empire, collapsed at the end of the First World War, and in the 1920s most of its Arab provinces were divided up among the victorious European powers and placed under mandates. Architecture, like all institutionalised and socially informed disciplines, was affected by these new political realities. Not only were European powers invading the East and colonising it, but European building styles were cropping up in its cities and shaping their appearance. European and European-trained architects created new images for a new East and informed the tastes of its dominant classes. Some of them also began to act as the interpreters of the architectural heritage of the countries in

– 17 3 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

which they worked. They documented, analysed and classified the structures they encountered, which permitted them to introduce these formerly unexplored architectural traditions to Europe, America and the East itself. Consequently, hybrid styles of building and decoration were produced in both East and West that borrowed freely and imaginatively from the varied repertories of non-Western architecture and blended them with European structural, constructional, functional and decorative conventions.4 But to see the East–West discourse in architecture, as well as in any other domain, as only a reflection or an expression of a power relationship – that is, as a unilateral exchange in which the East passively received the imprint of the West and, at the same time, unveiled its structures to its probing gaze – is a one-dimensional and incomplete interpretation. Although such an assessment is probably justified to the extent that imbalance governed the entire intercultural modern project, it nonetheless overlooks the specifics both of the history of the various disciplines and of each instance of interchange. The result is that almost any inquiry is critical, but never really fully demonstrated historically or constructive methodologically, although even a cursory examination of the various shifts and turns through which European ways were imposed on the various cultures of the East, and the ensuing reactions of these cultures, would reveal the complexities and intricacies informing the nature and range of interrelations between the two sides.5 An example taken from nineteenthcentury Egypt can be used to illustrate the working of multiple internal and external programmes in an East-West architectural exchange and to affirm its grounding in an operative historiography of architecture. This example is the neo-Mamluk style, also known as ‘Islamic revival’, ‘neo-Islamic’ or, in one appropriately deferential article penned by a statesman, ‘the Fu’ad I style’

after King Fu’ad of Egypt (r. 1917–1936),6 which in the early twentieth century was regarded as a national Egyptian style.7 The neo-Mamluk style owed its creation to the confluence of various intellectual, architectural and political currents which developed over the course of the nineteenth century. The first stage of its formation was the arrival of Napoleon and his French Expedition in Egypt and the investigations into everything Egyptian, including architecture, carried out by the Commission des sciences et arts. This effort was continued with the documentation and design projects of the French architect Pascal Coste (1787–1879) who worked under the patronage of Muhammad ‘Ali Pasha, the independent-minded ruler of the country, between 1817 and 1827. These early and inconclusive steps finally led to the blossoming of the neo-Mamluk style sixty years after Coste, when architects began to search for representations of an Egyptian identity for the twentieh century. One of the earliest and most monumental examples to adopt the neo-Mamluk style was al-Rifa‘i Mosque, constructed in two stages between 1869–80 and 1906–11, that is, preceding and following the British occupation of the country in 1882. The first stage was designed and supervised by the Egyptian Hussein Fahmi, the second by the Austro-Hungarian Max Herz, with others who were hired as consultants and interior designers for shorter periods. Cairo is incredibly rich in architecture. Its architectural legacy encompasses the entire gamut of styles referred to collectively as Islamic. The most spectacular date from the Mamluk period (1250–1517), when the city first assumed an imperial status and produced a wealth of religious, palatial and commemorative monuments that synthesised the achievements of earlier times and symbolised the image of the city for centuries to come. Cairene Mamluk structures are distinguished by their ingeniously negotiated volumes and façades that simultaneously

– 174 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

accommodate and dominate their urban surroundings; by their emphasis on verticality, which they achieved through attenuated portals, tall minarets and high and slender domes; and by their use of polychromatic surface articulation, especially the ubiquitous ablaq, consisting of alternate courses of black and white stone. In 1517, Egypt lost its independence and was incorporated into the Ottoman empire until the end of the eighteenth century. The three intervening centuries of subordinate status were not conducive to grand architectural achievements. Institutional and religious structures were still being built, albeit less frequently and on a smaller scale and with an uncertain architectural character made up of an uneasy blend of local Mamluk forms with imperial Ottoman ones. Consequently, no new predominant image emerged from the Ottoman presence in Cairo, and it was essentially a Mamluk city that the French army under Napoleon Bonaparte entered on 25 July 1798.8 With his expedition to Egypt, Napoleon was embarking on a grand imperial enterprise carefully and imaginatively planned and executed. Along with the army, he had recruited more than 150 scientists, scholars, artists, engineers and students, many of whom were to form the Commission des sciences et arts and later the Institut d’Égypte.9 Housed in the sumptuous Palace of Hasan Kashef in Cairo appropriated for the purpose, the Institut d’Égypte was quickly organised and assigned its principal tasks. According to Napoleon’s decree, the institute had two main objectives: first, to propagate the ideals of progress and enlightenment in Egypt and, second, to study and publish the natural, industrial and historical facts pertaining to Egypt’s past and present.10 The first aim translated into the defensive, administrative and public work installed by the new authorities primarily to secure French control over the country. To fulfil the second objective, the savants attached to the institute systematically col-

lected, studied, classified and represented all the available material on the history, geography and culture of the country. They ultimately produced the monumental Description de l’Égypte, which was published after many difficulties in Paris between 1809 and 1828 in nine folio volumes of text and twelve grand folio volumes of illustrations.11 This was a catalogue raisonné of sorts that formed the basis of the modern understanding of Egypt, its geography and topography, its flora and fauna, its Pharaonic and Islamic patrimony as well as its contemporary condition. The two volumes of plates on modern Egypt (published in 1809 and 1817) contain a selection of measured plans, perspective drawings and analytical details of important Islamic monuments that offered the first visually comprehensive and typologically codified analysis of this architectural tradition to Europe. Yet, the Description de l’Égypte cannot be seen as the result solely of a vast project of appropriation that represented a country and presented it to its prospective colonisers. The work has to be considered in its European intellectual context, and evaluated from within the paradigm of scholarly endeavours of its time. Following the model of Diderot’s and D’Alembert’s Encyclopédie, ou Dictionnaire raisonné des sciences, des arts et des métiers (1751–65), the Description de l’Égypte represented a magnificent chapter in the out-and-out cultural project of late eighteenth-century Europe that attempted to recover all available knowledge and to classify and codify it. The Description belongs to the Enlightenment not only in its methods and epistemological principles of exhaustive coverage and scientific accuracy but also in its aims, assumptions and ramifications. It was the precursor of many later projects of documentation and analysis of the arts and material culture in a number of countries and regions in Europe and elsewhere, and it launched the new field of Egyptology as a soundly defined scholarly pursuit. It also brought back an empirically

– 17 5 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

constructed and scientifically ordered body of knowledge about Egypt that helped establish the fundamental place of its ancient civilisation as the fountainhead of the Western historical evolution and verify the extent of its influence. Thus, for instance, Pharaonic architecture was presumed to be at the beginning of a continuous line of development that led ultimately to contemporary European architecture.12 True, the authors of the sections on Islamic architecture did not attempt to integrate their subject in the same line of development because of the prejudiced ideological and political imperatives that operated on the European nineteenth-century scene, but they nonetheless provided a basis for analysing and abstracting it, and ultimately for methodically using it in the design of new buildings. These three stages were accomplished single-handedly and almost inadvertently by Pascal-Xavier Coste, a little-known French architect from Marseilles. Coste lived and worked in Egypt as the designer and supervisor of large engineering projects all over the country for almost ten years (1817–1827), with only one interruption between 1822 and 1823, when he returned to France for a visit.13 Coste was the immediate successor in his use of the Description’s exhaustive method of recording architecture in all of its details. In fact, it was Coste’s acquaintance with Edme François Jomard, a founding member of the Institut d’Égypte and the editor of the Description de l’Égypte, that led to his opportunity to work for Muhammad ‘Ali Pasha, the viceroy of Egypt. Muhammad ‘Ali had commissioned his friend Jomard to hire French technocrats to direct and supervise the organising and modernising projects he had launched in Egypt. Coste was one of the earliest recruits. He planned and constructed irrigation canals in the Delta, factories and mills in and around Cairo, and a telegraphic line between Alexandria and Cairo. He was later commissioned to build palaces and pavilions for a number of European and Egyptian digni-

taries and for Muhammad ‘Ali Pasha himself, of which little has survived. However, Coste’s lasting impact as an architect in Egypt did not come from his built projects, important as they may have been, but rather from L’Architecture Arabe ou Monuments du Kaire mesurés et dessinés de 1818 à 1826, the book he published in 1837 after his return to France, and from two projects for mosques commissioned by Muhammad ‘Ali in 1821 and 1824 but never built.14 These projects anticipated and probably influenced a whole string of religious and public structures constructed in the second half of the nineteenth century and the early decades of the twentieth century in a style that was historicising in its inspirations and nationalistic in its aspirations. The mosque projects and the genesis, scope and focus of the book on Cairene architecture are in fact historically and conceptually intertwined. After almost five years of grand engineering projects, Coste’s architectural break came in 1821–2 when Muhammad ‘Ali asked him to design a funerary mosque at the Citadel of Cairo. Using the excuse that he needed to have direct knowledge of Islamic religious architecture in order to design the mosque, Coste obtained permission to visit and study a number of Cairene mosques. He spent two years (1822–4) documenting and analysing many Fatimid, Ayyubid and Mamluk structures using several drafting techniques. He later published many of them in his Monuments du Kaire which, in approach, technique and aim, represents a step beyond the Description.15 Coste’s project dealt exclusively with Islamic architecture, whereas in the Description selected Islamic monuments form only one facet of an exhaustive ethnographic study of the country. The Description’s drawings are presented as illustrations of historical types, and are supplemented by studies on the social, cultural and technical aspects of life in modern Egypt. In Coste’s book, it is the drawings that, so to speak, ‘do the talking’.16 His work is conceived and presented in

– 17 6 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

the grand tradition of the Beaux-Arts, where Coste had been trained in the ateliers of Vaudoyer père and Labadye. It is prefaced with a concise historical introduction, then each drawing is presented, preceded by a short explanatory paragraph. These drawings constituted the repertoire upon which Coste based his Mamlukinspired style for the design of the mosques commissioned by Muhammad ‘Ali at the citadel and in Alexandria. The detailed analytical drawings of Mamluk monuments, many of which are still unpublished (they are kept at the Bibliothèque municipale de Marseille), seem to have been necessary for the architect, who was educated in an academic, neo-classical tradition, to make the leap from admiring historical examples to adopting their distinctive elements for design purposes. The two designs for the mosques in fact reveal how Coste, with a concern for symmetry and clarity characteristic of his architectural background, manages to synthesise and order a basic vocabulary for a neo-Mamluk style. Both designs display the main characteristics of all monumental Mamluk mosques: hypostyle plans, tightly composed façades with layered surfaces, elevated portals with trilobed conches and geometric decorative patterns framing the entrance proper, carved stone domes with high drums and tapered profiles, tiered minarets with bulbous finials and ablaq surface articulation. The Citadel Mosque has four hypostyle halls arranged around a central courtyard with four domes marking the four corners. Four minarets à la turque – two at the ends of each of the two external aisles that Coste added along the longitudinal sides of the mosque – frame the squarish composition. A huge funerary dome protrudes from the qibla wall in a direct reference to the arrangement of the funerary Mosque of Sultan Hasan (1356–1361), the most monumental of the Mamluk structures. Two arcaded arms extend from the two ends of the entrance façade and enclose

an open space that may have been intended as a forecourt. They each end in a sabil/kuttab (a public fountain surmounted by a classroom for the reading of the Qur’an, which was usually open on all sides) (fig. 65). The façades and sections in the sketchbook show that the emphasis is on the five domes, especially the one behind the qibla wall, designated as the pasha’s funerary dome. They all have high Mamluk profiles, are all ribbed or carved in a chevron pattern and their drums are stepped and rest on four triangular corners. These are integral features encountered in all Mamluk Burji (or Circassian) domes of the fifteenth century. The Mosque of Alexandria is smaller than the Citadel Mosque; it has a hypostyle plan that recalls those of the late Fatimid mosques, especially al-Aqmar Mosque (1125) and al-Salih Tala’i‘ Mosque (1160). Coste added two square chambers on the two sides of the qibla wall, with huge domes carved in chevron and diamond patterns. The façade – clearly influenced by that of the complex of Sultan Faraj ibn Barquq in the northern cemetery (1400–11) – has two Mamluk minarets flanking a typical projecting portal surmounted by another dome carved in a chevroned pattern, with the two domes in the back framing the composition. Two aisles run parallel to the long side of the mosque from the domes towards the entrance façade and end with two diminutive sabil/kuttab units. These structures, which appear to have been detached from the mass of the mosque proper, constitute the only nod towards contemporary tastes with their semi-circular plans and their slanted roofs. The Mosque of Alexandria never saw the light of day, purportedly because of a sudden lack of funds caused by the expedition to the Morea that Muhammad ‘Ali had to dispatch to help the Ottomans against the Greek nationalist insurgents, but the Citadel Mosque was apparently started during the last months of Coste’s stay in Cairo.17 In a note attached to his sketches,

– 17 7 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 65. Façade of the mosque designed by Coste for the Citadel

Coste says that the foundations for his mosque were excavated in June 1827.18 He left Egypt for good shortly after, but the project was still alive in 1831, when the Russian consul in Cairo reported that the pasha was collecting marble for the funerary mosque at the citadel designed by Coste.19 Coste’s name, however, ceased to appear in all later reports on the mosque’s building. By 1833, Coste’s design had been abandoned, things had changed and the mosque’s appearance now seemed ‘barbarous’ and ‘in the Armenian [read Ottoman] style’ by the British traveller Robert Curzon.20 Many visitors to the citadel in the 1830s and 1840s scorned the mosque Muhammad ‘Ali was building as being of foreign design and unsuitable for the site, but neither Curzon nor any later visitor mentions either the date in which Coste’s design had been dropped or the name of the architect responsible for its replacement.21 From circumstantial evidence, however, it

appears that Coste was replaced around 1832 by an architect, or a group of architects, trained in the Ottoman tradition but well aware of contemporary European practices.22 The new architect totally revamped Coste’s design, but apparently had to use the foundation already laid for the mosque, since the Mosque of Muhammad ‘Ali today and the one designed by Coste have the same outer dimensions and are both arranged in 4m square grids.23 To ignore how Coste’s design for the mosque of the Citadel was replaced by a totally different structure after his departure form Egypt would be to miss an important aspect of the interplay between architecture and politics and overlook the ideological shift behind the resurrection of the style, pioneered by Coste, a half century after him. The Mosque Muhammad ‘Ali ultimately built at the Citadel has a distinctly Ottoman, pure central-domed plan and two slender and high pencil minarets, a

– 17 8 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

FIG. 66. Façade of the actual Mosque of Muhammad ‘Ali

– 17 9 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

far cry from Coste’s neo-Mamluk design (fig. 66). The built mosque, however, does not follow the example of contemporary Ottoman mosques erected in Istanbul in heavy baroque or rococo styles, such as the Nusretiye Mosque (1822–6). It looks more like older mosques built by early great Ottoman sultans. Contemporary guides and scholars assert that it was a copy of the Nuruosmaniye Mosque, completed by Sultan Osman III in 1755, but its plan and general structure more closely resemble a much earlier mosque, that of Sultan Ahmet, known as the Blue Mosque, built between 1609 and 1617.24 The Mosque of Muhammad ‘Ali was meant as a rhetorical composition designed to emulate early models. This archaic Ottomanism must have been intentional, especially if we take into consideration the psychological makeup of the pasha and the political conditions of the period during which his mosque was designed and constructed (1830–48). Muhammad ‘Ali, an exceedingly cunning Ottoman officer from Kavala in Macedonia, came to Egypt with the Ottoman army that replaced the French in 1801. By 1805, he had managed to rise from the position of second-in-charge of the Albanian contingency to secure his appointment by the Ottoman sultan as the governor of Egypt. But it took him many years of political manoeuvering – which culminated in the notorious massacre of the Mamluk amirs, his sworn enemies, in 1811 – to become the sole master of the country. He soon embarked on a programme of intense economic, social and military restructuring that was aimed primarily at modernising the country after Western models and creating a power base for his plans of territorial expansion. Coste was only one among many Europeans, mostly French, who were engaged by the pasha to carry out his reforms, especially in building a powerful modern army and navy. Hiring foreign experts, however, was only meant to speed up the modernisation process. The pasha

realised that he needed to train a local technocratic class if he wanted his reforms to take root. During the 1820s, he established a new school system to educate the cadres for his army and administration, founded a governmental press at Bulaq, north of Cairo, to publish the translated textbooks needed for the new schools and for the administration, and sent a number of students on academic missions to France to acquire the higher education deemed indispensable for the proper management of the new army and state structures.25 The reforms were not limited to the upper classes; their effects spread to the peasant class as well. In staffing the New Army (Nizam Jadid), the pasha faced the problem of recruiting soldiers. He was wary of the unruliness that characterised the Albanian troops who initially formed his main support in Egypt and of the arrogance and rapaciousness displayed by the Mamluks who made up the local military before his rule. He was also reluctant to recruit local Egyptians, who had been excluded from the army in Egypt for more than seven centuries. This lack of a military tradition was compounded by the strong and widespread belief in the superiority of the TurcoCircassian ‘race’ as soldiers and rulers, a belief shared by the pasha himself. However, after his attempt to recruit Sudanese slaves to form the new infantry failed, Muhammad ‘Ali reluctantly conscripted Egyptian fellahs, a move that forcibly introduced them to new and unfamiliar ideas and structures. The effectiveness of the Nizam Jadid was successfully put to the test in the various campaigns the Egyptian army, under the command of Muhammad ‘Ali’s son Ibrahim Pasha, participated in alone or with the Ottoman army. In these, Egyptian soldiers, only recently introduced to the art and discipline of war, outperformed their Ottoman counterparts.26 Muhammad ‘Ali himself remained the Ottoman gentleman he had made himself into when he first came to Egypt. He only

–18 0 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

spoke Turkish and dressed in the Turkish Ottoman fashion. He displayed the most exquisite manners of the Ottoman court. His immediate entourage was made up of an elite of Ottoman-trained men, among whom he fostered an Ottoman cultural identity, even when he was engaged in a war against the Ottomans.27 The pasha also cultivated a taste for classical Ottoman literature and for the lore of the great Ottoman sultans of yore, such as Selim I (r. 1512–1520), the conqueror of Egypt, and Süleyman the Magnificent (r. 1520–1566), whose biographies he ordered printed at the Bulaq press. He also had the divans of celebrated Persian poets – such as Hafiz, Sa‘di, Jalal al-Din Rumi and Farid al-Din ‘Attar – which had been de rigueur reading for a true Ottoman aristocrat, printed there as well.28 By 1830, Muhammad ‘Ali, who had long been aware of the intrinsic weakness of the Ottoman empire, decided to risk an allout attack despite his awareness that Europe would resist such a daring move. Between 1830 and 1833, his army marched north and threatened Istanbul before it was halted by European pressure. For eight years afterwards Muhammad ‘Ali defiantly reigned over an empire that extended into the Ottoman heartland of Anatolia, with Ibrahim Pasha as the effective governor of greater Syria. He seems not to have totally relinquished his dreams of rising to the pinnacle of the empire until 1841, when his army was driven out of Syria by a combined European-Ottoman assault. Soon afterwards, in the Treaty of London forced upon him by the European Powers, he agreed to recognise the sovereignty of the Ottoman sultan in return for a guaranteed hereditary entitlement to rule in Egypt for his family. This act established a semi-autonomous Egyptian kingdom which was ruled by Muhammad ‘Ali’s descendants until 1952. The Mosque of Muhammad ‘Ali was constructed during that period of struggle for control in the Ottoman empire, and the pasha probably intended it to articulate

these aspirations. The references to the glorious Ottoman past and the independence from the decadent Ottoman present are equally well expressed in the mosque’s classical style, monumentality and location inside the Citadel of Cairo, which had been the seat of government in the country for at least seven centuries. This may have been Muhammad ‘Ali’s idea of visually asserting his aims to take the place of the Ottoman sultan and to revitalise the empire, and perhaps to move its capital from Istanbul to Cairo.29 Evoking the memory of earlier Ottoman mosques in the outer appearance of his own also reflects the mentality of the pasha, who was after all the product of the Ottoman system and the Ottoman cultural milieu. For him, classical Ottoman architecture must have represented the most appropriate and most revered image of power itself. He did not want to destroy that symbol; he wanted instead to revive and appropriate it. Coste’s neo-Mamluk design, had it been constructed, would have given the wrong signals. It would have connected Muhammad ‘Ali to the Mamluk legacy that he had eradicated when he massacred its last representatives, the Mamluk amirs in 1811.30 But this same neo-Mamluk style was to prove itself convenient when Muhammad ‘Ali’s successors needed to proclaim their independence from the hegemony of the Ottoman sultanate and to assert a modern Egyptian image as the symbol of their reigns. Sa‘id Pasha (r. 1854–1863), Muhammad ‘Ali’s son and third successor, and especially Isma‘il Pasha (r. 1863–1879), his grandson, sought to free themselves from subordination to the Ottomans imposed upon Egypt in 1841 by the European Powers. Isma‘il finally succeeded in 1866 through bribery and diplomacy to extract from the sultan a few improvements and the title khedive (an old Persian title equivalent to king) which underscored his special status.31 In Egypt, Isma‘il tried to lessen his reliance on the Turco-Circassian class in the army

– 1 81 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

and administration and to replace them with Egyptians. He also initiated a few minor measures of Arabisation and Egyptianisation such as substituting Turkish with Arabic as the official language of correspondence.32 He himself, however, usually spoke Turkish or French and was totally fascinated with European, especially French, culture, to the point that he adopted French manners in his personal life and encouraged his entourage to follow suit.33 He also was an impatient moderniser who wanted to turn Egypt into a piece of Europe despite all adverse circumstances. But, lacking his grandfather’s farsightedness, this led to his opening up the country to foreign investors and adventurers and borrowing heavily from exploitative European banks to finance his projects, which ultimately resulted in his deposition, the country’s bankruptcy in 1879 and its consequent occupation by the British in 1882.34 Isma‘il’s passion for Europeanisation was exemplified by his grand urban projects in Cairo. When he visited the Exposition Universelle in Paris in 1867, he was charmed by the new plan of the Second Empire capital envisaged by Louis Napoleon and designed and executed by Baron Haussmann. Back in Cairo, he wanted to turn his capital into another Paris, complete with wide, straight avenues planted with trees, palaces, planned gardens and pavilions, and all the amenities of modern city life such as theatres, cafés and even an opera house. He commissioned his minister of public works, ‘Ali Pasha Mubarak, a member of the academic mission of 1844–49 to France and one of the most influential figures in modern Egyptian history, to draw the city’s new master plan. But the new city was only a façade concealing the traditional city behind it. With the exception of two boulevards – Shari‘ Muhammad ‘Ali and al-Sikka al-Jadida (New Avenue) or Shari‘ al-Muski – that cut across the old city’s dense fabric and razed many medieval structures, the new city

extended westwards towards the Nile along a north-south axis with streets radiating from central squares to form star patterns à la Haussmann. This Parisian-style Cairo was built in haste to impress Isma‘il’s guests, the European monarchs, who had been invited to Egypt for the inauguration of the Suez Canal in 1869, among them Empress Eugénie of France, wife of Napoleon III, who was the guest of honour.35 The Rifa‘i Mosque was commissioned in the same year by Princess Khoshiar, the powerful mother of Khedive Isma‘il, as a congregational mosque surrounding an already existing shrine devoted to a popular saint named al-Rifa‘i, and including mausolea for the royal family members.36 Choosing the tomb of a saint as the nucleus for a royal funerary mosque suggests that the khedival family was beginning to identify with the culture of the people it ruled. But, more than the sanctity attached to the shrine, it was its prominent location that made it attractive for a royal monument. Situated at the end of the new Shari‘ Muhammad ‘Ali opposite the famous Sultan Hasan Mosque, the most prodigious of Mamluk mosques built in the fourteenth century, it was in plain view of the citadel, where the royal family lived and where the Mosque of Muhammad ‘Ali dominated the skyline (fig. 67). The proximity of these two gigantic mosques presented both a dilemma and a challenge to any designer. Their sheer sizes required any competing structure to be truly monumental but their conflicting styles did not leave much room for a stylistic compromise in its appearance. The style for the new mosque, chosen apparently by the princess, was the one later to be called ‘neo-Mamluk’. It was conceived by Hussein Pasha Kojak Fahmi al-mi‘mar (the architect), who had earned the title for the excellence of his designs and drafting, which he had studied in France as a member of the fifth Egyptian educational mission, the same mission that included ‘Ali

–18 2 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

FIG. 67. Aerial view of the Mosques of Sultan Hasan and the Rifa‘i

Pasha Mubarak and Khedive Isma‘il himself. Fahmi later became the supervisor of religious endowments (awqaf), a highranking administrative position that required him to inspect the Islamic monuments of Cairo.37 He produced a design with many of the accoutrements of Mamluk religious architecture, as is apparent from a sketch of the southern façade later published by Max Herz (fig. 68). The strictly symmetrical façade with its two minarets, two arched entrances and five bays suggests that Fahmi was well aware of the academic imperative of symmetry and most probably familiar with the modern synthesis of Mamluk architectural elements pioneered by Coste.38 In fact, given the great similarity between the Rifa‘i’s façade and the façades of the mosques designed by Coste, one can assume that Fahmi was probably acquainted with Coste’s analysis of Mamluk architecture. This may have even hap-

pened when he was a student at the Muhandiskhaneh (School of Engineering), before he was sent to France, for we know that Muhammad ‘Ali had purchased ten copies of the Monuments du Kaire, some of which were donated to the library of the Muhandiskhaneh, where Fahmi could have consulted them.39 The Rifa‘i Mosque as designed by Fahmi was never completed. Construction was repeatedly halted because of spiralling costs or technical difficulties until it was stopped altogether in 1880. Mubarak, who describes the mosque as it was in 1892, records a telling story about one of the reasons why the mosque’s construction was suspended only two years after it began. He says that an unidentified European architect who was highly recommended to Khedive Isma‘il as knowledgeable in ‘Arabic’ architecture was given all the drawings of the mosque and that he drastically revised them

–18 3 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 68. Southern façade of the Rifa‘i Mosque as designed by Hussein Pasha Fahmi

while recommending the demolition of what had already been built. Princess Khoshiar rejected the new design and insisted on the original one. This led to a disagreement between mother and son, which interrupted the work on the mosque. When the work was resumed, it was continued according to Fahmi’s design.40 Mubarak himself, who was the minister of public works in several governments under Isma‘il in the 1870s, tried to solve the structural problems by proposing a large steel dome to cover the prayer hall. He even went ahead and hired an unidentified European consulting firm, which produced a design and a feasibility study that showed the dome to be less expensive than the original column-supported roof. Again, Isma‘il approved Mubarak’s proposal but Princess

Khoshiar insisted on the original design by Hussein Fahmi.41 The mosque’s construction was not resumed again until 1906, long after both patron and architect were dead. A new architect, the Austro-Hungarian Max Herz, was appointed to complete the project. He had been the technical director of the Committee for the Conservation of Monuments of Arab Art and involved in the restoration of a large number of Mamluk monuments, including the Mosque of Sultan Hasan, about which he published a detailed and beautifully illustrated architectural study.42 Herz held Mamluk architecture in high esteem as the most ingenious response to its urban, aesthetic and historical setting.43 In design, he was something of a purist, who,

–18 4 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

FIG. 69. Southern façade of the Rifa‘i Mosque as built by Herz

while praising Fahmi’s scheme because of its Mamluk inspiration, criticised it because of its strict symmetry which he regarded as inadequate, both because it was not responsive to the contextual and topographic requirements of the site and because it was

alien to the more fluid arrangements of the Mamluk precedents (fig. 69).44 His hands were tied, however, because most of the exterior was already in place, including the symmetrically positioned bases of the two minarets on the southern façade.

–18 5 –

M A M LU K H I ST ORY T H ROUGH A RCH I T E C T U R E

FIG. 70. Western façade of the Rifa‘i Mosque

FIG. 71. View of the Egyptian National Library (Dar al-Kutub)

–18 6 –

T H E FOR M AT ION OF T H E N EO -M A M LU K ST Y L E I N MODE R N EGY P T

Consequently, Herz kept all of Fahmi’s Mamluk references in his new design, if anything he clarified and stressed their direct lineage from ideal Mamluk models, such as the complex of Sultan Qalawun (1284) and the Mosque of Sultan Hasan, a preference seen in Coste’s designs as well (fig. 70). The Rifa‘i Mosque was the first of a series of mosques and other public and cultural structures that adopted the neoMamluk style. Others that followed include the Sayyida Zaynab Mosque (1885), the Sayyida Nafisa Mosque (1897), the Ministry of Endowments (Awqaf) (built in three stages in 1898, 1911 and 1929 by Mahmud Fahmi, the chief architect of the waqfs ministry) and the Egyptian National Library (Dar al-Kutub) (built in 1904 by the Italian architect Alfonso Manescalo) (fig. 71).45 It was probably the desire to give a modern

Egyptian identity to these culturally specific public buildings – especially the mosques which were endowed as proof of the khedival family’s social and cultural commitment to the country – that the neo-Mamluk style was used. Later on, this style was assimilated by architects working on large public projects during the first half of the twentieth century and came to be regarded as an Egyptian national style, relating contemporary Egyptian architecture to a glorious phase of its history.46 The metamorphoses it had undergone from the documentary and classificatory stage of the savants of the French Expedition, to the pioneering work of Pascal Coste in synthesising its elements into a new style, to its adoption in khedival public buildings, and finally to its appropriation by a national and international elite in modern Egypt were forgotten.

–18 7–

GLOSSARY Ablaq Abwab al-Sutur Adab Al-Addur (or Dur) al-Sultaniyya Adhan Adib Agha Ahdath Ahl al-Balad ‘Ajib ‘Alaliq Al-‘Amma Arbab al-hiraf wa-al sana’i‘ Arkan ‘Asabiyya Atabek al-‘Asakir ‘Azim Badhahanj Bahri Bay‘a qahriyya Bayt Al-Bilad Bimaristan Da‘wa Dahhan Dar al-‘adl Dar Dast Devshirme Dihliz Dikka Divan Diwan al-‘ama’ir Diwan Durkah

striped, alternate black and white. gates of the veils, leading to the royal private palaces. belles-lettres. royal private palaces. call to prayer. belletrist. Mamluk trainer. youth, but also urban militia. locals. wondrous. sugar figures, akin to lollipops. common people. artisans. buttresses, or supports (sing. rukn). solidarity, especially tribal solidarity as defined by Ibn Khaldun. the commander of the armies. large. wind-catcher. facing the Bahr, the Nile, which is the north-west orientation in Cairo. oath under duress. house. home country. hospital. Isma‘ili missionary network painter. the palace of justice. house. a wooden chair covered with a silk cloth for the sultan. Ottoman children levy system. passageway or a vestibule. raised platform for sitting. poetry collection. building department. chancery or council. vestibule. –18 8 –

GL O S S A RY

Durqa‘a Fada’il Fakhir Fatwa Fellah Fidawiyya Futuh Gharib Ghulamiyyat Ghutumi Halaqa Al-Halaqa Haniya Hara Al-Harafish Haram Harariq Hisar Hiyal handasiyya Huquq Hurmiyya Husn ‘Ilm ‘Imran Iqta‘ Istabl ‘Itaqa Iwan Jamadariyya Jamil Jashinkir Jihad jund/jundi Kabir Karahiya Katib al-Sirr Kawwaz Khamriyyat Khanqah Khassakiyya Khatib Khawarij Khidma Khitat Khizana Khushdash

the central part of the qa‘a, which literally means the entry to the qa‘a. merits. monumental, impressive. legal opinion or legal decree. peasant. Isma‘ili self-sacrificing assassins. chronicles of the conquest. strange. love of youth poetry. inarticulate. circle of learning. A Mamluk military unit made out mostly of local recruits. niche. quarter. common people. sanctuary. Mamluk warships. citadel. engineering devices. dependencies. either a part of a qa‘a or a certain type of qa‘a. castle. religious sciences. Khaldunian urbanization. alloted land revenues. stables. manumittance. a vaulted or roofed structure totally open on one end. of jamadar, the wardrobe master. beautiful. taster. fighting for the cause of religion. Mamluk soldiers/soldier. big, large. religious aversion. private secretary. jug maker. Bacchic poetry. sufi lodge. Mamluks of the privy. preacher. outsiders to Islam (derogatory). service at the palace. topographical tracts. nook. comrade in training. –18 9 –

GL O S S A RY

Khutba Kuttab al-Dast Kuttab Kuwwa Lala Ma‘azil Mabitat Madhhab Madrasa Manzara Maq‘ad Maqsura Marafiq Masjid Mastaba Maydan Mi‘mar Misr Mu‘alim al-Mu‘alimin Mu‘alim Al-Mujahid Mu’talif Mudhahhib Muhandis Muhandiskhana Muhtasib Muntazam Muqarnas Musawwir Mutanasib Muzammik Muzayyin Na’ib al-Saltana Naqib al-Ashraf Naqib al-juyush Naqqash Nazir al-Jaysh Pishtaq Qa‘a musaqqafa ‘aqdan Qa‘a Qadi al-Jaysh or Qadi al-‘Askar Qadi al-Qudat Qadi Qarafa Qibla

sermon. clerks of the throne. clerks. niche. governess. unidentified part of the common area of the residence. sleeping spaces. school of jurisprudence. institution of high religious learning. belvedere. Mamluk loggia with an arcaded opening. enclosed area inside a mosque. household dependencies oratory a square stand, or a bench. hippodrome. architect. country, province in an empire. chief of masters. master. the wager of Jihad, the fighter for Islam. proportionate. gilder. engineer, surveyor. school of engineering. market-inspector. harmonious. decorative elements mostly in the shape of niches arranged linearily as corbels or arcuately as conches and domes. painter. proportionate. illuminator. painter. vicegerent. head of the descendants of the prophet. head of the army. carver. army supervisor. Persian framing gate. vaulted qa‘a. dominant type of hall. army judge. chief judge. judge. cemetery. orientation toward the Ka‘ba. –19 0 –

GL O S S A RY

Qubba Rank Rassam Riwaq Sabil/Kuttab Sahn Salamlik Sani‘ Saqi Shadd Shari‘a Al-Sharq Shukhshikha Silahdariyya Suffah Surat al-Ard Tablakhana Tahrim Tajdid Takhayen Takht Tarima Tibaq Tricorium Umara‘ al-Mashura Ustadar Ustadh Wafidiyya Wakil bayt al-mal Wali Waqf Yalmaq Zajal Zardakash Ziyada

dome, but also a mausoleum. emblem. painter. living unit in the upper level of a residence. It may also mean a gallery or a portico. a particularly Cairene institution with a public fountain on the street level and a children’s school above. court. men’s section in a house, reception quarter in the Ottoman period. apprentice cup bearer. building supervisor. religious law. Orient. wooden lantern. of the silahdar, the arms-bearer. awning, but a suffah in a qa‘a is a shallow and small alcove. Cartography. drummery. religious censure. reform. probably a sleeping nook. throne. baldachin. apartment. Roman triple apse. amirs of the council, the sultan’s advisory group. the master of the household. the owner and the master of a Mamluk, and the head of a military household. Mongol immigrants to the Mamluk realm. the treasury controller. governor in the Ottoman period. endowment deed. Mongol furred coat. rhymed song. surveyor of the armory. overflow space in a mosque.

– 1 91 –

NOTES CHAPTER 1 1. On the Mamluk institution in Islamic history, see David Ayalon, ‘Preliminary Remarks on the Mamluk Military Institution in Islam’, in War, Technology and Society in the Middle East, ed. V. Parry and M. E. Yapp (Oxford, 1975), 44–58; idem, ‘Mamluks of the Seljuks: Islam’s Military Might at the Crossroads’, Journal of the Royal Aisatic Society Series 3, 6/3 (1996): 305–33. 2. Marius Canard, ‘Quelques observations sur l’introduction géographique de Bughyat al-talab de Kamal al-Din Ibn al-‘Adim’, Annales de l’Institut d’Études Orientales de la Faculté des Lettres d’Alger 15 (1957): 41–53, where he reports the wala’ (loyalty) of ghilman to the caliph, their ribats and arms, their organisation, their strict discipline and the fear of temptation by their beauty that required that a pious sheikh accompanied whoever among them needs to perform his ablutions during general assemblies. 3. A review of the institution of mounted archers in Islamic armies can be found in Shihab al-Sarraf’s, ‘L’impact des techniques militaires sur l’évolution politique et sociale dans le moyenorient médiéval: le cas de l’archerie’, Études Orientales-Dirasat Sharqiya 7/8 (1990): 6–28; on Turkish mounted archers’ effectiveness, see Walter E. Kaegi, Jr., ‘The Contribution of Archery to the Turkish Conquest of Anatolia’, Speculum 39 (1964): 96–108. 4. Claude Cahen reminds us that their fame had reached the Frankish chansons de geste, where regiments of mounted ghulams were known as agolants; see Cahen, Orient et Occident au temps des Croisades (Paris, 1983), 64, n. 14. 5. On the controversy surrounding the saqaliba and on the rarity of black Mamluks, see David Ayalon, ‘Aspects of the Mamluk Phenomenon’, Der Islam 53, 2 (1976): 196–225, esp. 199–202; idem, ‘On the Eunuchs in Islam’, Jerusalem Studies in Arabic and Islam 1 (1979): 67–124, esp. 92 ff. 6. Shihab al-Din al-Nuwayri, Nihayat al-Arab fi Funun al-Adab, 33 vols, various eds (Cairo, 1923–93), vol. 29 (Cairo, 1992), 415–17, dates the arrival of Qipchaqs in large numbers on the slave market after 1229. For a brief history of the Qipchaqs’ gradual penetration into the Mamluk system, see Robert Irwin, The Middle East in the Middle Ages: The Early Mamluk Sultanate 1250–1382 (Carbondale, Ill., 1986), 12–18; also, Reuven Amitai-Preiss, Mongols and Mamluks: the MamlukIlkhanid War, 1260–1281 (Cambridge, 1995), 18–19. 7. Recent accounts of the rise of the Mamluks under al-Salih include R. Stephen Humphreys, From Saladin to the Mongols: The Ayyubids of Damascus 1193–1260 (Albany, NY, 1977), 301–5; Amalia Levanoni, ‘The Mamluks’ Ascent to Power in Egypt’, Studia Islamica 72 (1990): 121–44. 8. Irwin, Mamluk Sultanate, 26–9; Amalia Levanoni, ‘The Consolidation of Aybak’s Rule: An Example of Factionalism in the Mamluk State’, Der Islam 71 (1994): 241–54. 9. This was not the first time that Mamluks assumed the rule in Islamic history. There are many short-lived precedents practically everywhere, including Egypt (Ahmad ibn Tulun, Tughuj al-Ikhshid and Kafur). 10. A parallel case, the sultanate of Delhi, was ruled between 1206 and 1290 by a succession of Turkish Mamluks and their sons, see P. Hardy, ‘Ghulam- India’, Encyclopaedia of Islam, 2nd edn, 2: 1084–5. The Indian Mamluks, however, do not seem to have developed a real Mamluk system: many of them – notably the founder Qutb al-Din Aybak (notice the parallel in name –19 2 –

N O T E S T O PAG E S 5 –7

11. 12. 13. 14.

15.

16. 17.

18.

19.

20. 21. 22. 23.

24. 25.

with the first Egyptian Mamluk sultan) – and Iltumish were not manimutted until after they had become amirs. Moreover, their armies, like almost all other Islamic armies, remained under the control of a mixture of free and Mamluk officers. See the perceptive assessment in Irwin, Mamluk Sultanate, postscript, 152–60. This applies to such great names in Mamluk historiography such as Ibn Wasil, Abu Shama, al-Nuwayri and later on al-Maqrizi, al-‘Ayni and Ibn al-Furat. Badr al-Din ibn Jama‘a, Tahrir al-Ahkam fi Tadbir Ahl al-Islam, Hans Kofler, ed., Islamica, vols 6–7 (1933–35), 6: 353–414, 7: 1–64, especially chapters 2 and 5; Radwan al-Sayyid, al-Umma wa-l-Jama‘a wa-l-Sulta: Dirasat fi-l-Fikr al-Siyasi al-‘Arabi al-Islami (Beirut, 1984), 139–43. Abu Shama, al-Dhayl ‘ala-al-Rawdatin, ed., M. Zahed al-Kawthari (Cairo, 1947), 208, composed two verses after ‘Ayn Jalut to this effect; a translation is provided by David Ayalon, ‘The European-Asiatic Steppe: A Major Reservoir of Power for the Islamic World’, Proceedings of the 25th Congress of Orientalists, 2 vols (Moscow, 1963), 2: 47–52, esp. 49. For the khutba of al-Mustansir, see Ibn ‘Abd-al-Zahir, al-Rawd al-Zahir fi-Tarikh al-Malik al-Zahir, ed. ‘Abd al-‘Aziz al-Khuwaiter (Riyadh, 1976), 102–10; Ibn al-Dawadari, Kanz al-Durar, vol. 8, al-Durra al-Zakiyya fi Akhbar al Dawla al-Turkiyya, ed. U. Haarmann (Cairo, 1971), 73–9; al-Maqrizi, al-Suluk li-Ma‘rifat Duwal al-Muluk, 4 vols, ed. M. Zyada et al. (Cairo, 1934–72), 1: 453–7; al-Qalqashandi, Subh al-A‘sha fi Sina‘at al-Insha, 14 vols (Cairo, 1913–18), 10: 111–16. For the khutba of al-Hakim, see Ibn ‘Abd al-Zahir, Rawd, 141–7. For the second delivery of the same khutba, see al-Maqrizi, Suluk, 1: 771–2; Ibn al-Furat, Tarikh al-Duwal wa-l-Muluk, ed. K. Zurayk and N. Izzedine, 3 vols (Beirut, 1939–42), 8: 128–9, 135. See also P. M. Holt, ‘Some Observations on the Abbasid Caliphate of Cairo’, Bulletin of the School of Oriental and African Studies 47 (1984): 501–7. Baybars al-Mansuri, al-Tuhfa al-Mulukiyya fi al-Dawla al-Turkiyya, ed. A. H. S. Hamdan (Cairo, 1987), 25–6. For a discussion of Ibn Khaldun’s passage on the Mamluks, see David Ayalon, ‘Mamlukiyyat: (B) Ibn Khaldun’s View of the Mamluk Phenomenon’, Jerusalem Studies in Arabic and Islam 2 (1980): 340–49; also Tarif Khalidi, Arabic Historical Thought in the Classical Period (New York, 1994), 228–31. The different steps initiated by Baybars and Qalawun are discussed in Amalia Levanoni, A Turning Point in Mamluk History: The Third Reign of al-Nasir Muhammad ibn Qalawun (1310–1341) (Leiden, 1995), 15–27. See also Nasser Rabbat, The Citadel of Cairo: A New Interpretation of Royal Mamluk Architecture (Leiden, 1995), 96–100, 132–7. Al-Maqrizi, al-Mawa‘iz wa-l-I‘tibar bi-Dhikr al-Khitat wa-l-Athar, 2 vols (Bulaq, 1854), 2: 213–14; also Ibn Shahin al-Zahiri, Zubdat Kashf al-Mamalik wa-Bayan al-Turuq wa-l-Masalik, ed. Paul Ravaisse (Paris, 1894), 115–16, gives an even shorter description while discussing different divisions of the army. Ulrich Haarmann, ‘Regicide and “The Law of the Turks”’, in Intellectual Studies on Islam: Essays Written in Honor of Martin B. Dickson, ed. Michel M. Mazzaoui and Vera B. Moreen (Salt Lake City, 1990), 127–35. David Ayalon, ‘The Great Yasa of Chingiz Khan: A Reexamination’, Studia Islamica 36 (1972): 113–58 (part C1) ‘The Position of the Yasa in the Mamluk Sultanate’, and 38 (1973): 107–56 (part C2) ‘al-Maqrizi’s Passage on the Yasa under the Mamluks’. Being legally free was one of the fundamental conditions in the Islamic legal code for a ruler; cf. Irwin, Mamluk Sultanate, 8–9. David Ayalon’s several studies on the Mamluk institution and the Mamluks of Egypt do not unequivocally clear the question of manumission, cf. his ‘L’esclavage du Mamelouk’, Oriental Notes and Studies 1 (1951): 1–66, esp. 20–2. Levanoni, Reign of al-Nasir, 19, only briefly mentions the manumission ceremony in her otherwise detailed review of the Mamluk system. Al-Maqrizi, Suluk, 1: 386–9. Taj al-Din al-Subki, Tabaqat al-Shafi‘iyya al-Kubra, ed. M. al-Tanahi and ‘A.-F. al-Huluw (Cairo, 1966), 8: 216–17, includes this story in a hagiographical biography of Ibn ‘Abd al-Salam. The report appears in an abridged form in al-Suyuti, Husn al-Muhadara fi Tarikh Misr wa-l-Qahira –19 3 –

N O T E S T O PAG E S 7 – 8

26. 27.

28. 29.

30.

31.

32. 33.

34.

35.

(Cairo, 1968), 2: 161–3. I could not find it in any other biography of this famous Shafi‘ite qadi including al-Safadi, al-Wafi bi-l-Wafi yyat, 29 vols so far, various eds (Leipzig and Wiesbaden, 1931–), 18: 520–22; al-Nuwayri, Nihayat al-Arab, 29: 294–9, and 30: 66–77; Ibn Wasil, Mufarrij al-Kurub fi Akhbar Bani Ayyub, ed. Hasanayn M. Rabi‘ (Cairo, 1977), 5: 302–304. Al-Subki’s report is not dated, which obviously further weakens its authenticity, but it could not have occurred before Aybak because he was the one who instituted the post of vicegerent. For a French translation of the passage and a discussion of its relevance to understanding the relationship between the ulama and the Mamluks, see Jean-Claude Garcin, ‘Le Sultan et Pharaon (Le politique et le religieux dans l’Égypte mamluke)’, in Hommages à François Daumas (Montpellier, 1986), 261–72. Al-Maqrizi, Khitat, 2: 214. On the outstanding career of Isma‘il ibn Muhammad, al-Majd al-Sallami both as a slave trader and diplomat, see al-Yusufi, Nuzhat al-Nazir fi Sirat al-Malik al-Nasir, ed. Ahmad Hutait (Beirut, 1986), 201–3, 282–3; al-Safadi, Wafi, 9: 220–21; Shams al-Din al-Shuja‘i, Tarikh al-Malik al-Nasir Muhammad ibn Qalawun al-Salihi wa-Awladahu, ed. Barbara Schäfer (Wiesbaden, 1978), 251; Ibn Hajar al-‘Asqalani, al-Durar al-Kamina fi A‘yan al-Mi’a al-Thamina, 4 vols (Hyderabad, 1929–32), 1: 381–2; al-Maqrizi, Khitat, 2: 43; Ayalon, ‘Esclavage’, 3. Altinbugha al-Maridani was sent to al-Nasir Muhammad as a gift from the prince of Mardin; al-Yusufi, Nuzhat, 265; al-Shuja‘i, Tarikh, 266. Kitbugha was captured after the first battle of Homs in 1259; al-Safadi, Wafi, 24: 178; Ibn alFurat, Tarikh, 8: 192; Amitai-Preiss, Mongols and Mamluks, 51. Qarasunqur may have been a Christian of Qara captured in 1264 although al-Safadi, Wafi, 9: 331–3, rejects this report; Almalik al-Jukandar was captured in the battle of Abulustayn (Elbistan) in 1277; al-Maqrizi, Khitat, 2: 310. Levanoni, Reign of al-Nasir, 17. Numerous sources – with the revealing exception of Ibn ‘Abd alZahir, the sultan’s official chronicler – mention that Baybars attacked Qara in 1264 because its inhabitants were abducting travelling Muslims and selling them to the Franks. He killed the men and captured the women and boys who were then sold as slaves. Al-Nuwayri, Nihayat al-Arab, 30: 294–5, adds that the boys of Qara were raised among the Mamluks, spoke Turkish, and many among them attained high positions. At least this is the impression given by later historians who otherwise wax poetic on the beauty and bravery of the sons and daughters of the wafidiyya, during Kitbugha’s reign, who belonged to the tribe of Oirats known to the Mamluk sources as the Uwayratiyya, see al-Maqrizi, Khitat, 2: 22–3; Ibn al-Furat, Tarikh, 8: 204–5. Ayalon, ‘The Wafidiyya in the Mamluk Kingdom’, Islamic Culture 25 (1951): 81–104, provides numerous references to the wafidiyya. Baybars once told the emissaries of the king of France (Rey-de-France in the Arabic sources): ‘Thanks to God, no more war is to take place between us [Mamluks] and the Mongols, who are, after all, of the same race ( jins), and they should not let each other down’, al-Qalqashandi, Subh, 8: 37. For a collection of other references in the Arabic sources about the issue of racial affinity see David Ayalon, ‘Great Yasa, (part C1)’, 117–26. Ayalon, twice (122, 127) attributes the aforementioned statement to al-Nasir Muhammad, who ruled after the Crusaders had been expelled from Palestine and had no reason to negotiate with any French king about the status of Jerusalem or of any Palestinian port. This mistaken attribution may have originated with Ayalon’s source, Ibn Fadl-Allah al-‘Umari, al-Ta‘rif bi-l-Mustalah al-Sharif (Cairo, 1894), 64, or with the copyist of the manuscript, which in fact deals mostly with al-Nasir’s period. Ibn Shaddad, Tarikh al-Malik al-Zahir, ed. Ahmad Hutait (Wiesbaden, 1983), 338–9, lists forty out of 3,000 Mongol wafidiyya during Baybars’ reign who were given amirships, the most famous of whom is Nuqay or Nugay al-Tatari, who was the father of Ardukin, the wife of al-Ashraf Khalil and of al-Nasir Muhammad after his brother’s death. The halaqa seems to have lost the crucial position it held under Salah al-Din and the early Ayyubids after al-Salih Najm al-Din Ayyub and later Baybars shifted the dependence of the –19 4 –

N O T E S T O PAG E S 8 –12

36.

37. 38. 39. 40.

41.

42.

43.

44. 45. 46.

47.

army to the royal Mamluk corps, cf. David Ayalon, ‘Studies on the Structure of the Mamluk Army’, Bulletin of the School of Oriental and African Studies 15/3 (1953): 448–59. On the Mamluk training system, see Ayalon, ‘Esclavage’, 9–26; Hassanein Rabie, ‘The Training of the Mamluk Faris’, in War, Technology and Society in the Middle East, 153–63; R. Stephen Humphreys, ‘The Emergence of the Mamluk Army’, Studia Islamica 45 (1977): 67–99; 46 (1977): 147–82. See Ayalon’s critique of Humphreys’s analysis of the difference between the Ayyubid and Mamluk use of Mamluks in his ‘From Ayyubids to Mamluks’, Revue des Études Islamiques 49 (1981): 43–57 (reproduced in Islam and the Abode of War (London, 1994)); see also Levanoni’s rebuttal of Ayalon’s notion of continuity from Ayyubid to Mamluk times in Reign of al-Nasir, 6–7 and notes. On tibaqs and qa‘as, as well as the division of Mamluks according to function or ethnicity, see Rabbat, Citadel of Cairo, 110–19, 138–42, 283–91. The most detailed account can be found in al-Shuja‘i, Tarikh, 149–55. Ahmad ‘Abd al-Raziq, ‘al-Mamalik wa mafhum al-usra ladayhim’, Majallat Kulliyat al-Athar 2 (1987): 188–207; Ayalon, ‘Esclavage’, 21–31. The obvious anomaly is that a Mamluk tibaq family was totally masculine even, curiously, in their sexual conduct. Sodomy, that is sex with beardless boys (murdan), and perhaps even a more equitable form of homosexuality, was widespread in the Mamluk class as is clear in countless reports but it was rarely openly acknowledged and discussed, see al-Maqrizi, Khitat, 2: 104; also Ahmad ‘Abd al-Raziq, La femme au temps des Mamlouks en Égypte (Cairo, 1973), 102. A number of examples are listed, without noting their contradictory nature, in ‘Abd al-Raziq, ‘al-Mamalik wa mafhum al-usra ladayhim’, passim. See also the discussion in Irwin, Mamluk Sultanate, 88–90. For the Burjiyya, see al-Maqrizi, Suluk, 1: 756; Ibn Duqmaq, al-Jawhar al-Thamin fi Siyar alKhulafa wa-l-Muluk wa-l-Salatin, ed. S. A. F. ‘Ashur (Mecca, 1982), 308; Ibn Iyas, Bada’i‘ al-Zuhur fi Waqa’i‘ al-Duhur, ed. M. Mustafa, 5 vols in 6 books (Wiesbaden, 1960–75), 1, 1: 362; David Ayalon, art. ‘Burdjiyya’ in Encyclopaedia of Islam, 2nd edn, 1: 1324–5. The most elaborate explanation of the dependence of Kitbugha on his ethnic ‘asabiyya is presented in the reports by al-Yusufi preserved in al-‘Ayni, ‘Iqd al-Juman fi Tarikh Ahl al-Zaman, vol. 5, ed. M. M. Amin (Cairo, 1989), 5: 304–15; also Ayalon, ‘Wafidiyya’, 92–3, 99–100. For Baybars alJashankir, see especially Ibn Taghri-Birdi, al-Nujum al-Zahira fi Muluk Misr wa-l-Qahira, 16 vols, ed. Muhammad Ramzi (Cairo, 1930–56), 8: 227, 232–7. See also Irwin, Mamluk Sultanate, 90–2. Al-‘Ayni, ‘Iqd al-Juman, 5: 234–8; Ibn Taghri-Birdi, Nujum, 8: 41–6. See the indignant reactions to that state in al-‘Ayni, ‘Iqd al-Juman, 5: 305–7; al-Maqrizi, Suluk, 1: 812–13; Ibn al-Furat, Tarikh, 8: 204–5. Nasser Rabbat, ‘The Ideological Significance of the Dar al-‘Adl in the Medieval Islamic Orient’, The International Journal of Middle Eastern Studies 27, 1 (Feb 1995): 3–28. [Reproduced in this volume]. Dorothea Krawulsky, introduction to al-‘Umari, Masalik al-Absar fi Mamalik al-Amsar: Dawlat alMamalik al-Uwla, ed. D. Krawulsky (Beirut, 1986), 36–7, gives a similar interpretation in her review of the changes in the literary production after the reign of al-Nasir Muhammad. Levanoni in her Reign of al-Nasir develops an argument first suggested by Ayalon in ‘The Muslim City and the Mamluk Military Aristocracy’, Proceedings of the Israel Academy of Sciences and Humanities (1968): 311–29, that the change happened during al-Nasir’s rule and because of his autocratic policies.

CHAPTER 2 1. On Mamluk historiography, see Li Guo, ‘Mamluk Historiographic Studies: The State of the Art’, Mamluk Studies Review 1 (1997): 15–43; Ulrich Haarmann, Quellenstudien zur frühen Mamlukenzeit –19 5 –

N O T E S T O PAG E S 12 –13

2.

3. 4.

5.

6.

7.

8.

(Freiburg, 1970); Donald Little, An Introduction to Mamluk Historiography: An Analysis of Arabic Annalistic and Biographical Sources for the Reign of al-Malik an-Nasir Muhammad ibn Qalawun (Wiesbaden, 1970); idem, ‘al-Safadi as Biographer of his Contemporaries’, in Essays on Islamic Civilization Presented to Niyazi Berkes, ed. Donald Little (Leiden, 1976), 190–210; Shakir Mustafa, al-Tarikh al-‘Arabi wa-l-Mu’rikhun: Dirasa fi Tatawwur ‘Ilm al-Tarikh wa-Rijaluhu fi-l-Islam, 4 vols (Beirut, 1978–93), vol. 2, 139–304, vol. 3, vol. 4, 7–227; Franz Rosenthal, A History of Muslim Historiography 2nd edn (Leiden, 1968), passim; Tarif Khalidi, Arabic Historical Thought, 182–231. Michael Chamberlain, Knowledge and Social Practice in Medieval Damascus, 1190–1350 (New York, 1994), 2, n. 1, reminds us that despite its comparatively large and varied corpus in comparison to other Islamic historiographies, Mamluk historiography still suffers from many lacunae and discontinuities in time, geographic area and subject matter; as for original documents, it is hopelessly dwarfed by the prodigious archives of contemporary Europe. For a pioneering evaluation of the patterns of dependence and relative originality of Mamluk prosopobiographies and chronicles, unfortunately limited to the reign of al-Nasir Muhammad, see Little, Mamluk Historiography, 94–9, 132–6. Dorothea Krawulsky in her introduction to al-‘Umari, Masalik, 15–37, presents a perceptive interpretation of the political and cultural impetus behind Mamluk historical writing and distinguishes between the early Bahri period and the period after al-Nasir Muhammad’s reign. Remke Kruk, ‘History and Apocalypse: Ibn al-Nafis’ Justification of Mamluk Rule’, Der Islam 72 (1995): 324–37, offers a politico-cultural analysis of Ibn al-Nafis’s unusual treatise; D. W. Morray, An Ayyubid Notable and his World: Ibn al-Adim and Aleppo as Portrayed in his Biographical Dictionary of People Associated with the City (Leiden, 1994), is a comprehensive interpretive study of a historian, his milieu and his writing, albeit in the Ayyubid period. For a general critique of Mamluk history study and its limited use of theorisation, see W. W. Clifford, ‘Ubi Sumus? Mamluk History and Social Theory’, Mamluk Studies Review 1 (1997): 45–62. I join here two groups that are usually thought of as distinct. Typical stratifications of the Mamluk society distinguish between two middle groups: the ulama (sometimes combined with the kuttab) and the merchants. See Ira Lapidus, Muslim Cities in the Later Middle Ages (Cambridge, 1967), 79–85; Boaz Shoshan, Popular Culture in Medieval Cairo (Cambridge, 1993), 3. However, my interest is in how these strata are articulated in the historical texts not in how they appear in urban and economic settings. For the most recent treatment of awlad al-nas, see Ulrich Haarmann, ‘Joseph’s Law: The Careers and Activities of Mamluk Descendants before the Ottoman Conquest of Egypt’, in The Mamluks in Egyptian Politics and Society, ed. T. Philipp and U. Haarmann, (Cambridge, 1998), 77–83. A somewhat overlooked field of research, despite the abundance of information, is income levels in Mamluk society. These offer important clues to understanding the dynamics of social movements; cf., Éli Strauss (Eliyahu Ashtor), ‘Prix et salaire à l’époque mamlouk’, Revue d’Études Islamiques 17 (1949): 49–94; David Ayalon, ‘The System of Payment in Mamluk Military Society’, Journal of the Economic and Social History of the Orient 1 (1958): 37–65, 257–96; Hassanein Rabie, ‘Size and Value of the Iqta‘ in Egypt AH 564–741, 1169–1341 AD’, in Studies in the Economic History of the Middle East, ed. M. A. Cook (Oxford, 1970), 129–38, and his larger, The Financial System of Egypt AH 564–741 AD 1169–1341 (Oxford, 1972). In the late Mamluk period, the case of Shahin al-Khalwati, a Mamluk turned master Sufi under Qaytbay, is a famous one. A khanqah built in 1538 carries his name. In another instance cited by al-Khatib al-Jawhari al-Sayrafi, Inba’ al-Hasr bi-Abna’ al-‘Asr, ed. Hasan Habashi (Cairo, 1970), 275, Qaytbay is said to have been particularly irked by a Mamluk whom he called zuqaqi (riffraff) and ordered doubly beaten as a punishment, not only for his offensive behaviour (he left his tabaqa and got married without permission), but also because, while bastoned, he was using the commoners’ invocation of ‘ya sitti nafisa’ instead of asking for mercy by imploring à la mamluk, ‘tuba khujam’. For a lesser-known case from the early Mamluk period of a Mamluk who became a Shafi‘ite faqih and changed his name from Sanjar to Talha, see al-Safadi, Wafi, 16: 460. On the manipulation of ulama posts in Damascus, see Chamberlain, Knowledge, 57–113. –19 6 –

N O T E S T O PAG E S 13 –15

9. For Baybars al-Mansuri, see al-Safadi, Wafi, 10: 352; the introductions to the three published books of his work: al-Tuhfa al-Mulukiyya fi-l-Dawla al-Turkiyya, 5–13; idem, Mukhtar al-Akhbar, ba’-ya’; and S. M. Elham to Zubdat al-Fikra fi Tarikh al-Hijra, vol. 10; Little, Introduction, 4–5; for Zetterstéen’s author see ibid., 18–19; for al-Shuja‘i, see P. M. Holt, ‘Shams al-Shuja‘i: A Chronicler identified?’ Bulletin of the School of Oriental and African Studies 58(1995): 532–4. 10. An insightful and singular chronicler, Musa ibn Yahya al-Yusufi wrote a history that apparently covered the reigns of fourteen Bahri sultans from Qalawun to Hasan in fifteen volumes. This history is now lost, except for the copious quotations taken from it by al-‘Ayni and others and the single volume discovered by Donald Little in the Ayasofya Library which covers only five years of al-Nasir’s reign, see Donald Little, ‘The Recovery of a Lost Source for Bahri Mamluk History: al-Yusufi’s Nuzhat al-Nazir fi Sirat al-Malik al-Nasir’, Journal of the American Oriental Society 94 (1974): 42–54. Al-Yusufi’s single volume is now published, not by Little, who nonetheless had stated in 1974 that he was preparing an edition of the manuscript, but by Ahmad Hutait. It is strange, however, that the editor fails to credit Little with the discovery of the manuscript and identification of the author. Hutait instead ascribes the discovery to his advisor, Ihsan ‘Abbas, and the identification of the author as al-Yusufi to himself. 11. See the discussion in Ulrich Haarmann ‘Arabic in Speech, Turkish in Lineage: Mamluks and their Sons in the Intellectual life of Fourteenth-Century Egypt and Syria’, Journal of Semitic Studies 33 (1988): 81–114, esp. 82–5. In Baybars al-Mansuri’s case in particular, another reason may be that he was accused of having used chancery kuttab to compose his chronicles for him, which, if true, and that is impossible to prove, would explain the absence of clear Mamluk insights in these works. 12. Al-Jawhari, Inba’ al-Hasr, 175–82. It should be remembered, however, that the two were academic rivals in more than one way. 13. Al-Dhayl ‘ala al-Rawdatayn, a combination chronicle and biographical dictionary in which this late thirteenth-century author mixes the account of public events with the affairs of his private life, including his feelings and meditations upon the circumstances of some randomly encountered individuals, most of them common people. See Abu Shama, Dhayl; Louis Pouzet, ‘Abu ˇSama (599–665/1203–1268) et la société damascaine de son temps’, Bulletin d’Études Orientales 37–8 (1985–6): 115–26; Joseph E. Lowry, ‘Time, Form and Self: The Autobiography of Abu Shama’, Edebiyat 7: 2 (Autumn 1996): 313–25. 14. Ulrich Haarmann addressed several facets of the complex relationship between the Turks and the Arab literati, see his ‘Arabic in Speech’, also ‘Rather the Injustice of the Turks than the Righteousness of the Arabs: Changing ‘Ulama’ Attitudes Towards Mamluk Rule in the Late Fifteenth-Century’, Studia Islamica 68 (1989): 61–79. 15. Emmanuel Sivan, L’Islam et la Croisade, Idéologie et Propagande dans les Réactions Musulmanes aux Croisades (Paris, 1968), 67–70, 102–8; Hadia Dajani-Shakeel, al-Qadi al-Fadil ‘Abd al-Rahim al-Baysani al-‘Asqalani: Dawruhu al-Takhtiti fi Dawlat Salah al-Din wa Futuhatuhu (Beirut, 1993), 183–4, 217–22, 231–9. 16. Nizam al-Mulk, Siyasat-Namah, trans. al-Sayyed al-‘Izzawi (Cairo, 1976); Ayalon, ‘The Mamluks of the Seljuks’, 307–8. 17. Immediately before the Mamluk rule in the Ayybid period, ulama or their sons could attain high executive positions in the government and the army as illustrated by cases such as the four sons of Sadr al-Din ‘Abd al-Rahman ibn Hamaweyh Shaykh al-Shuyukh (al-Safadi, Wafi, 18:121–2), Fakhr al-Din, ‘Imad al-Din, Mu‘in al-Din and Kamal al-Din, who served the late Ayyubid princes (al-Maqrizi, Suluk, 1: 302, idem, Khitat, 2: 33–4). One of them, Fakhr al-Din, stopped wearing the turban and wore the sharbush and qubba’ (dress items of soldiers), or the faqih ‘Issa alHakkari who was a great amir under both Nur al-Din and Salah al-Din as well as a religious scholar. On the high position of the civilian elite under Salah al-Din and the Ayyubids in general, see Humphreys, From Saladin to the Mongols, 21–9, 377–80. 18. For Mamluk dress codes see L. A. Mayer, Mamluk Costume: A Survey (Geneva, 1952), 15–35 for the Mamluks, and 49–55 for the ulama. For rank, see idem, Saracenic Heraldry (Oxford, 1938); Michael –197–

N O T E S T O P A G E S 1 5 – 17

19.

20. 21.

22.

23.

24. 25.

26.

27. 28.

29. 30. 31. 32.

Meinecke, ‘Zur mamlukischen Heraldik’, Mitteilungen des Deutschen Archäologischen Instituts Abteilung Kairo 28 (1972): 213–87; Estelle Whelan, ‘Representations of the Khassakiyah and the Origins of Mamluk Emblems’, in Content and Context of Visual Arts in the Islamic World, ed. P. Soucek (Philadelphia, 1988), 219–43. Promoting the Muslim-born sons of Mamluk amirs to high amirial positions seems to have begun under al-Nasir Muhammad, but accelerated considerably after his death and during the turbulent reigns of his sons, especially under al-Nasir Hasan (1347–1351, 1354–1361), who used them expressly to counter the influence of the Mamluk amirs. See Amalia Levanoni, Reign of al-Nasir, 42–52. Chamberlain, Knowledge, 37–54, discusses these developments in Damascus. The three encomia are al-Rawd al-Zahir fi-Sirat al-Malik al-Zahir; Tashrif al-Ayyam wa-l-‘Usur fi-Sirat al-Malik al-Mansur, ed. Murad Kamil (Cairo, 1961); al-Altaf al-Khafi yya fi-l-Sira al-Sharifa al-Sultaniyya al-Malakiyya al-Ashrafi yya, ed. Axel Moberg (Lund, 1902). The last two are incomplete. For a short biography of Ibn ‘Abd al-Zahir see P. M. Holt, ‘The Virtuous Ruler in Thirteenth-Century Mamluk Royal Biographies’ Nottingham Medieval Studies 24(1980): 27–35; idem, ‘Three Biographies of al-Zahir Baybars’, in Medieval Historical Writing in the Christian and Islamic Worlds, ed. D. O. Morgan (London, 1982), 19–29. Sivan, L’Islam et la Croisade, 165–80, discusses a number of themes found in Mamluk sources concerning the primacy of jihad but overstates, in my opinion, the propagandist intent of these writings. A keener, though unfortunately much shorter, analysis is by Krawulsky, introduction to al-‘Umari, Masalik. These are the words of al-Maqrizi, a student of Ibn Khaldun and one of the best critics of his age, who went on to satirise the Mamluks of his time as ‘more lustful than monkeys, more ravenous than rats, and more harmful than wolves’, Khitat, 2: 214; see also Ibn Taghri-Birdi, Nujum, 7: 328–9, where he deplores the changes in the Mamluk army structure from the time of Qalawun and describes the Mamluks of his time as ‘holding their buttocks in the water and their nose in the sky’ (meaning that they were both impotent and arrogant). There is a tendency among the Burji historians to idealise the Bahri period, but as David Ayalon (‘Harb’, Encyclopaedia of Islam, 2nd edn, 3: 189) remarks ‘this tendency is by no means without foundation’. Examples provided in Haarmann ‘Arabic in Speech’, 95–100. Al-Yusufi, Nuzhat, 329–30, notes that Qibjaq al-Mansuri and Aitamish al-Muhammadi, both of Mongol extraction, spoke and wrote ‘Mongol’. Ibn Taghri-Birdi, Nujum, 11: 224, mentions that when Barquq’s father, Anas, arrived in Cairo, they spoke in ‘Circassian’, presumably not Oghuz Turkish, since Ibn Taghri-Birdi was a native Turkish speaker and he clearly did not understand Barquq’s speech. Haarmann, ‘Arabic in Speech’, 90–2; Barbara Flemming, ‘Literary Activities in Mamluk Halls and Barracks’, in Studies in Memory of Gaston Wiet, ed. Myriam Rosen-Ayalon (Jerusalem, 1977), 251. Maya Shatzmiller, ‘The Crusades and Islamic Warfare: A Re-Evaluation’, Der Islam 69 (1992): 247–88, esp. 272–88, argues that the training manuals, all Mamluk and all written in Arabic, can be seen as a sign of the reintegration of the soldier in the cultural life of the thirteenth and fourteenth centuries. Flemming, ‘Literary Activities’, 251–3. Al-Safadi, Wafi, 15: 341–2; he also cites an unusual poem composed by Ibn Katib Sunqur in which he mixes Arabic and Turkish in describing a Turkish lover living in Cairo. Al-Safadi mentions only one famous precedent to this most unusual form of poetic composition. Haarmann, ‘Arabic in Speech’, 92; al-Shuja‘i, Tarikh, 275; Ibn Hajar, al-Durar al-Kamina, 2: 75–6. For al-‘Ayni, see W. Marçais, art ‘al-‘Ayni’, Encyclopaedia of Islam, 2nd edn, 1: 790–1; for ‘Ayntab see Charles Pellat, art. ‘‘Ayntab’, Encyclopaedia of Islam, 2nd edn, 1: 791–2. Flemming, ‘Literary Activities’, 255, n. 39, 40. Ibn Taghri-Birid, a historian and member of the awlad al-nas, was belittled by other historians precisely for his alleged weak command of the Arabic language. He returned the favour by show–19 8 –

N O T E S T O P A G E S 17 – 1 8

33.

34.

35.

36. 37.

38.

39. 40. 41. 42. 43.

44.

ing how senior scholars, such as Ibn Hajar al-‘Asqalani, made mistakes because of their ‘ignorance of Turkish and unfamiliarity with Turkish customs’, two drawbacks that he obviously did not suffer from, see Nujum, 6: 555. In the case of Qalawun, see al-Maqrizi, Suluk, 1: 673 and 756. Amir Bashtak also refused to speak Arabic, though the sources say that he knew it, and always used an interpreter in public, see al-Safadi, Wafi, 10: 142; al-Maqrizi, Khitat, 2: 34; Haarmann, Quellenstudien, 77, n. 2. Even Baybars, who apparently was fluent in Arabic, spoke in Turkish on some occasions, such as during a session in the Dar al-‘Adl where he needed a translator for the Arabic for the judges present, see Ibn ‘Abd al-Zahir, Rawd, 85 and n. 4. See al-Yusufi, Nuzhat, 250, where a particularly vulgar and savage amir, Sayf al-Din al-Akuz, who was mushshid al-dawawin and the torture officer for al-Nasir Muhammad admits being a ghutumi to emphasise his rudeness. Ulmas al-Hajib, a disreputable and altogether despicable amir, is also called ghutumi, see al-Safadi, Wafi, 9: 371. Ibn Taghri-Birdi, Nujum, 11: 224, raises the same question in refuting the reports by al-Maqrizi and by Ibn Khatib al-Nasiriyya (an Aleppine historian of the late fourteenth century) concerning the name of Barquq. He notes that their chain of transmission starts with the slave-trader who brought Barquq, ‘Uthman ibn Musafir, who did not speak a word of Arabic. Yet he allegedly communicated with their second transmitter, Burhan al-Din al-Mahalli, who did not speak a word of Turkish. Haarmann, ‘Arabic in Speech’, 91, n. 38. The exchange between Bahadur (Halawa) al-Ushaqi and Amir Tankiz (al-Safadi, Wafi, 10: 302–3) was not in Arabic as stated by Haarmann. It was in fact in Turkish as specified by al-Safadi himself. Examples of discussions that were most likely fabricated abound. See for instance the numerous dramatic exchanges narrating how Qarasunqur al-Jukandar, Aqush al-Afram and other amirs escaped from al-Nasir and took refuge in the Ilkhanid court of Abu Said in 1311 in alSafadi, Wafi, 9: 331–3 for al-Afram; and ibid., 24: 217–20 for Qarasunqur. Compare them with the discussion between Qabjaq al-Mansuri and the father of Ibn Fadl Allah, where similar topoi are used in a similar context before Qabjaq’s escape to the Ilkhanid Ghazan in 1298, ibid., 24: 181–2. See also the self-reflective remark attributed to Qawsun after Baktimur al-Saqi’s death on the way back from the Hijaz, ibid., 10: 196; or the long-distance exchange between the amirs Taz and Baybugha Arus in the 1350s, ibid., 16: 383–4. Another example is the story explaining the rift between al-Nasir Muhammad and his older son Ahmad, al-Shuja‘i, Tarikh, 47–9. See Ulrich Haarmann, ‘Auflösung und Bewahrung der klassischen Formen arabischer Geschichtsschreibung in der Zeit der Mamluken’, Zeitschrift der Deutschen Morgenländischen Gesellschaft 121 (1971): 46–60, esp. 49. A succinct restating of Haarmann’s historiographical observations appears in his review of Bernd Radtke, Weltgeschichte und Weltbeschreibung im Mittelalterlichen Islam, in Journal of the American Oriental Society 115/1 (1995): 134. Al-Safadi, Wafi, 10: 143–4; al-Maqrizi, Khitat, 2: 35; and Ibn Taghri-Birdi, Nujum, 10: 19–20, repeat al-Safadi’s story verbatim (al-Maqrizi reports his source). See Rabbat, ‘Dar al-‘Adl’, 3–28 [reproduced in this volume]. Al-Safadi, Wafi, 24: 277. Al-Shuja‘i, Tarikh, 159–60; al-Maqrizi, Khitat, 2: 307–8. Qawsun was a young man around eighteen years of age when he arrived in Cairo in the company of Tulbiyya, the Mongol princess sent to al-Nasir from the Golden Horde court of Uzbek in 1320, see al-Maqrizi, Khitat, 2: 307. Bashtak was also from the Golden Horde. We don’t know at what age he reached Cairo as a Mamluk, but he probably was a young man as well. He was tutored by none other than Qawsun, before he became his main competitor for the sultan’s favours, see al-Shuja‘i, Tarikh, 219. Haarmann, Quellenstudien, 179–80, notes the tension between reporting events as they were spoken or recasting them in grammatically correct Arabic in Mamluk chronicles. He distinguishes between al-Maqrizi who strove to maintain a high level of Arabic throughout his text and Ibn –19 9 –

N O T E S T O PAG E S 18 – 2 2

45. 46.

47. 48.

al-Dawadari who included lots of colloquialisms either because he wanted to give it an authentic appearance or because he himself did not know any better. See a contextualisation of this contempt in Haarmann ‘Arabic in Speech’, 81–5. A telling example is offered by al-Yusufi, Nuzhat, 178, where a famously abusive officer uses the lowest dialectical curses in addressing a group of administrators. The sultan and his amirs did not understand the words and asked another administrator about their meanings. He feigned his ignorance of these words as well and said that he heard them spoken only by the riffraff. Al-Yusufi, Nuzhat, 199–207. Clifford, ‘Ubi Sumus?’ exempts only two Mamluk historians, Ira Lapidus and Michael Chamberlain, from his general conclusion that Mamluk historical studies have failed to develop adequate theoretical and interpretive frameworks.

CHAPTER 3 1. For a recent review of the publications on Mamluk art and architecture, see Jonathan Bloom, ‘Mamluk Art and Architectural History: A Review Article’, Mamluk Studies Review 3 (1999): 31–58. 2. See R. Stephen Humphreys, ‘The Expressive Intent of the Mamluk Architecture in Cairo: A Preliminary Essay’, Studia Islamica 35 (1972): 69–119; Oleg Grabar, ‘Reflections on Mamluk Art’, Muqarnas 2 (1984): 1–12; Michael Meinecke, ‘Mamluk Architecture, Regional Architectural Tradition: Evolutions and Interrelations’, Damaszener Mitteilungen 2 (1985): 163–75; idem, Patterns of Stylistic Change in Islamic Architecture: Local Traditions versus Migrating Artists (New York, 1995); Rabbat, Citadel of Cairo; Bernard O’Kane, ‘Monumentality in Mamluk and Mongol Art and Architecture’, Art History 19, 4 (December 1996): 499–522. Some of the still unpublished recent Ph.D. theses proclaim the new, more interpretive directions that the field in general is following; see, for example, Lobna Abdel Azim Sherif, ‘Layers of Meaning: An Interpretive Analysis of Three Early Mamluk Buildings’, Ph.D. Thesis, University of Michigan, 1988; Howyda N. al-Harithy, ‘Urban Form and Meaning in Bahri Mamluk Architecture’, Ph.D. Thesis, Harvard University, 1992; Jane Jakeman, ‘Abstract Art and Communication in “Mamluk” Architecture’, Ph.D. Thesis, Oxford University, 1993. 3. Michael Meinecke, Die Mamlukische Architektur in Ägypten und Syrien (648/1250 bis 923/1517), Teil II: Chronologische Liste der Mamlukischen Baumassnahmen (Glückstadt, 1992), vii–ix, provides thorough estimates of the number of Mamluk monuments in all the major Syrian and Egyptian cities. 4. This is apparent in al-Sakhawi’s, Kitab al-Tibr al-Masbuk fi-Zayl al-Suluk, which is a chronicle continuing al-Maqrizi’s Suluk, and in his little article, al-Tuhfah al-latifah fi tarikh al-Madinah al-sharifah. It comes across more distinctly in al-Suyuti, Husn al-Muhadara, which is his modest attempt at producing a khitat book. 5. Ibn Khaldun, al-Muqaddima, ed., ‘Ali A-W Wafi, 4 vols (Cairo, 1960) 3: 829–36; for an abridged text in English, The Muqaddimah, trans. Franz Rosenthal, ed. N. J. Dawood (Princeton, 1967), 263–7. 6. See the condemnation of al-Maqrizi, one of the best critics of his age, in his Khitat, 2: 214; see also Ibn Taghri-Birdi, Nujum, 7: 328–9. 7. See Nasser Rabbat, ‘Representing the Mamluks in Mamluk Historical Writing’, in The Historiography of Islamic Egypt, c. 950–1800, ed. Hugh Kennedy (Leiden, 2000), 60–71. [Reproduced in this volume.] 8. Ibn Khaldun, al-Muqaddima, 3: 832–6; abridged English text, The Muqaddimah, 265–7. 9. Except for some odd and not immediately explainable cases, such as the Madrasa al-Mu‘izziyya, of al-Mu‘izz Aybak (1250–57), in Misr al-Fustat. Ibn Kathir, al-Bidaya wa-l Nihaya fi-l-Tarikh, 14 vols (Cairo, 1932–9), 13: 196, comments that ‘although the madrasa’s span from the outside is of the best construction, its interior space is not so impressive’; al-Yunini, Dhayl Mir’at al-Zaman, 4 vols (Heyderabad, 1954–61), 1: 60, and Ibn Taghri-Birdi, Nujum, 7: 14, say that ‘its dihliz is very wide and very long, while the structure itself is proportionally small’. See also al-‘Ayni, ‘Iqd al–200 –

N O T E S T O PAG E S 2 2 – 24

10.

11.

12.

13.

14.

15. 16. 17.

18. 19.

20.

Juman, 1: 44; Ibn Duqmaq, al-Intisar li Wasitat ‘Iqd al-Amsar, ed. K. Vollers (Cairo, 1893), 4: 35, 53–4, 92–3. A. I. Sabra in his edition of Ibn al-Haytham, The Optics of Ibn al-Haytham: Books I-III, On Direct Vision, 2 vols (London, 1989), 2: 99, discusses the example of the famous essayist al-Jahiz (767–869) in his Risalat al-Qiyan (The Essay on Singer-Slaves), ed. ‘Abd al-Salam Muhammad Harun in his Rasa’il al-Jahiz, 2 vols (Cairo, 1965) 2: 162–3. Al-Jahiz explains physical beauty in terms of two aesthetic principles: tamam (fullness) and i‘tidal (moderation), both are dependent on wazin (measure, balance, rhythm), which varies according to every case under consideration. Al-Jahiz goes on to say that wazin also governs the beauty of vessels, furnishings, embroidered textiles and water channels, all of which have to achieve balance in form and composition (al-istiwa’ fi al-khart wa-l-tarkib). (Sabra considered tamam, i‘tidal and wazin to be three separate principles, although it seems that Jahiz suggests that tamam and i‘tidal both derive from wazin.) This area of research is not well covered. One notable pioneer is George Makdisi; his Rise of Humanism in Classical Islam and the Christian West (Edinburgh, 1990), passim, presents one of the most thorough discussions of the types of knowledge and kinds of settings available to medieval Islamic ‘humanists’ (to use Makdisi’s term). Makdisi (Appendix A, 355–61) provides a summary of Dia’ al-Din ibn al-Athir’s eight scholarly requisites for poets and kuttab from his al-Mathal alSai’r fi Adab al-Katib wa-al-Sha‘ir (Riyadh, 1983–4), which shows clearly that no visual concerns belonged in those lists. A single exception, to my knowledge, can be found in the memoirs of the Iraqi physician, ‘Abd al-Latif al-Baghdadi, a very sharp and perceptive resident of Cairo in the later part of the Ayyubid period (he wrote his text in 1204), al-Ifada wa-l-I‘tibar fi al-Umur al-Mushahada wa-lHawadith al-Mu‘ayana bi-Ard Misr (Cairo, 1869), where he uses aesthetic notions to analyse the naturalness and proportionality achieved in the ancient Egyptian statues. This exceptional short treatise deserves a study on its own. For a discussion of the status of the building professions in the Mamluk society, see Nasser Rabbat ‘Architects and Artists in Mamluk Society: The Perspective of the Sources’, Journal of Architectural Education 52/1 (September 1998): 30–7 [reproduced in this volume]; Doris BehrensAbouseif, ‘Muhandis, Shad, Mu‘allim: Note on the Building Craft in the Mamluk Period’, Der Islam 72 (1995): 293–309; and the pioneering Leo Mayer, Islamic Architects and Their Works (Geneva: 1956), 20–7. An interesting example of a philosophical framework is al-Qalqashandi’s chapter on nafs al-khatt (‘the writing itself’, used here in the sense of the nature of penmanship) in his voluminous Subh, 3: 1–149, esp. 41–3, a well structured and competent, if platitudinous, discussion that relies heavily on older texts and poetic quotations. Badr al-Din al-‘Ayni, al-Maqrizi’s rival, who served seven times as a muhtasib is a case in point. For their dates of service as muhtasib, see Ahmad ‘Abd al-Raziq, ‘La hisba et le muhtasib en Égypte au temps des Mamluks’, Annales Islamologiques 13 (1977): 115–78, 148–53. O’Kane, ‘Monumentality in Mamluk Art’, passim. On various aspects of this heated propaganda war, see Rabbat, ‘Dar al-‘Adl’, 24–8; Adel Allouche, ‘Teguder’s Ultimatum to Qalawun’, International Journal of Middle East Studies 22/4 (November 1990): 437–46; Donald P. Little, ‘Notes on Aitamis, A Mongol Mamluk’, Beiruter Texte und Studien 22 (1979): 387–401. O’Kane, ‘Monumentality in Mamluk Art’, 500, no. 4; see also Nasser Rabbat, ‘al-Iwan: Ma‘nahu al-Faraghi wa Madlulahu al-Tadhkari’, Bulletin d’Études Orientales 49 (1997): 249–67. O’Kane, ‘Monumentality in Mamuk Art’, 510 and nn., discusses comparisons made in two Mamluk and Ilkhanid sources between two major monuments – the Mosque of ‘Ali Shah in Tabriz and the Madrasa of Sultan Hasan in Cairo on the one hand, and the Iwan Kisra, on the other. See the analysis of Dorothea Krawulsky concerning the change in historical production in the Mamluk period in ‘al-Intaj al-Thaqafi wa-Shar‘iyat al-Sulta’, her introduction to al-‘Umari, Masalik, 15–37, reprinted in a volume of her collected articles, al-‘Arab wa Iran, Dirasat fi al-Tarikh wa al-Adab min al-Manzur al-Aydiologi (Beirut, 1993), 94–116. – 2 01 –

N O T E S T O PAG E S 24 – 2 6

21. See Nasser Rabbat, ‘Mosaics of the Qubba al-Zahiriyya in Damascus: A Classical Syrian Medium Acquires a Mamluk Signature’, Aram 9–10 (1997–8): 1–13; also Rabbat ‘Mamluk Throne Halls: Qubba or Iwan’, Ars Orientalis 23 (1993): 201–18, for discussions of Umayyad echoes in early Mamluk architecture [both reproduced in this volume]. See also Jonathan Bloom, ‘The Mosque of Baybars al-Bunduqdari in Cairo’, Annales Islamologiques 18 (1982): 50–5; Hana Taragan, ‘Politics and Aesthetics: Sultan Baybars and the Abu Hurayra/Rabbi Gamliel Building in Yavne’, in Milestones in the Art and Culture of Egypt, ed. Asher Ovafiah (Tel Aviv, 2000), 117–43, esp. 124–30, for discussions of conscious Fatimid references in the architecture of Baybars’s mosque. 22. Methods for the use of waqfs as historical documents in analysing architecture have been developed by many authors in the recent past. The pioneering scholar was ‘Abd al-Latif Ibrahim, ‘Wathiqat al-Amir Akhur Kabir Qaraquja al-Hasani’, Majallat Kulliyat al-Adab 18, 2 (December 1956): 183–251; idem, ‘al-Watha’iq fi Khidmat al-Athar’, in al-Mu’tamar al-Thani li-l-Athar fi alBilad al-‘Arabiyya (Cairo, 1958), 205–88. See also Michael Rogers, ‘Waqfiyyas and WaqfRegisters: New Primary Sources for Islamic Architecture’, Kunst des Orients 11 (1976–7): 182–96; Muhammad M. Amin, al-Awqaf wa-l-Hayyat al-Ijtima‘iyya fi-Misr. AH 684-923/1250-1517 AD (Cairo, 1980); Mona Zakarya, Deux palais du Caire médiéval, Waqfs et architecture (Marseilles, 1983); Donald P. Little, ‘The Haram Documents as Sources for the Arts and Architecture of the Mamluk Period’, Muqarnas 2 (1984): 61–72; Leonor Fernandes, ‘Notes on a New Source for the Study of Religious Architecture during the Mamluk Period: The Waqfiya’, al-Abhath 33 (1985): 3–12. One of the best studies of the architectural particularities of waqf formulae and terminology is Hazem Sayed, ‘The Rab‘ in Cairo: A Window on Mamluk Architecture and Urbanism’, Ph.D. Thesis, MIT, 1987; it is unfortunately still unpublished. For a summary of his research, see his ‘Development of the Cairene Qa‘a: Some Considerations’, Annales Islamologiques 23 (1987): 31–53. 23. See Rabbat, Citadel of Cairo, 9–14, 59–60, for an analysis of the texts describing the fortifications of the Cairo Citadel. 24. See, for instance, Ibn ‘Abd al-Zahir, Rawd, 278–9, who copied the section on al-Azhar from an original waqf document redacted for the Fatimid caliph al-Hakim. Al-Maqrizi, Khitat, 2: 273–4, reproduces what appears to be a more complete text from the same waqf. 25. A recent discussion of the purposes and techniques of historical writing in the Mamluk period is Ulrich Haarmann, ‘al-Maqrizi, the Master, and Abu Hamid al-Qudsi, the Disciple: Whose Historical Writing Can Claim More Topicality and Modernity?’, in The Historiography of Islamic Egypt, c. 950-1800, ed. Hugh Kennedy (Leiden, 2000), 59–75. 26. See my discussion of a conversation between the two powerful amirs of al-Nasir Muhammad, Bashtak and Qawsun, in Rabbat, ‘Representing the Mamluks’, 72–5. 27. Haarmann, ‘al-Maqrizi, the Master, and Abu Hamid al-Qudsi, the Disciple’, 149–51. 28. See Haarmann, Quellenstudien, 129–37; idem, ‘Auflösung und Bewahrung’, 49. See also Guo, ‘Mamluk Historiographic Studies’, 33–7. 29. See Haarmann ‘Arabic in Speech’, 82–5; idem, ‘Joseph’s Law’, 55–84. 30. Little, Mamluk Historiography, 73–99, offers a comparative examination of specific years in the annals of six historians, which shows their complicated patterns of interdependence. 31. David Ayalon went so far as to assert that al-Nasir Muhammad’s predilection for grand building projects drained the Mamluk economy so much that it never recovered; see his ‘Muslim City and the Mamluk Military Aristocracy’, 311–29; idem, ‘The Expansion and Decline of Cairo under the Mamluks and Its Background’, in Itineraires d’Orient: Hommages à Claude Cahen, ed. Raoul Curiel and Rika Gyselen (Paris, 1994), 14–16. The decline of Cairo, which was congruent with the downfall of the Egyptian economy in the second half of the Mamluk period, is complicated and cannot be blamed solely on internal political factors; it still needs a thorough study. Janet Abu-Lughod, Before European Hegemony: The World System A.D. 1250–1350 (New York, 1989), 224–47, presents a well-balanced synthesis of Egypt’s economic plight in the fourteenth and fifteenth centuries. 32. Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 293–5, lists a number of instances in which Mamluk patrons, most notably al-Nasir Muhammad, played a direct role in the design of the –202–

N O T E S T O P A G E S 2 6 – 31

33.

34. 35. 36.

37. 38. 39. 40. 41.

42. 43. 44.

45. 46. 47. 48. 49.

50.

51. 52. 53.

buildings they sponsored; see also Rabbat, Citadel of Cairo, 186–90, 277–80, for a discussion of al-Nasir Muhammad’s involvement in the remodelling of the citadel and its surroundings. On Ibn Shaddad, see Yoel Koch, ‘‘Izz al-Din ibn Shaddad and His Biography of Baybars’, Annali: Istituto Universitario Orientale, Sezione Slava 43 (1983): 249–87; Holt, ‘Three Biographies of al-Zahir Baybars’, 19–29. Ibn Shaddad, Tarikh al-Malik al-Zahir, 339–61. The buildings are analysed in Rabbat, Citadel of Cairo, 100–31. For references to Abu Hamid al-Qudsi’s predilection for architecture and urbanism which characterises all of his work, see Haarmann, ‘al-Maqrizi, the Master, and Abu Hamid al-Qudsi, the Disciple’, 154–61. For an analysis of the book, see Muhammad Sa‘id Rida, ‘Ibn Shaddad fi Kitabihi al-A‘laq al-Khatirah, ‘Qism al-Jazirah’, Majallat al-Mu’arrikh al-‘Arabi 14 (1980): 124–204. Ibn Taghri-Birdi, Nujum, 7: 191–7; Ibn Shakir al-Kutubi, Fawat al-Wafi yyat wa-l-Dhayl ‘Alayha, 4 vols, ed. Ihsan ‘Abbas (Beirut, 1973), 1: 242. Grabar, ‘Reflections on Mamluk Art’; Humphreys, ‘Expressive Intent’. Oleg Grabar, ‘Reflections on the Study of Islamic Art’, Muqarnas 1 (1983): 1–14, esp. 10–11. This new awareness seems to have affected even the traditional form of adab collections. A fascinating example is al-Baha’i al-Ghuzuli, Matali‘ al-Budur fi Manazil al-Surur (Cairo, 1882), which integrates in an unprecedented way a number of architectural elements, such as fountains, tanks and wind catchers, in the list of topics that an adib needs to be able to discuss and to summon literary quotations about in his function as a literary companion. André Miquel, La géographie humaine du monde musulman jusqu’au milieu du 11e siècle, 3 vols (Paris, 1967–80), 1: 267–330; Ulrich Haarman, ‘Auflösung und Bewahrung’, 46–60. Dorothea Krawulsky, Introduction to ‘Umari, Masalik, 15–37. The idea that Egypt had a specific character and was a clearly defined entity is the theme of many Egyptian historical and analytical studies, see especially Jamal Hamdan, Shakhsiyyat Misr, Dirasa fi ‘Abqariyyat al-Makan, 4 vols (Cairo, 1980–84), passim. More recent studies include Milad Hanna, The Seven Pillars of the Egyptian Identity (Cairo, 1994); ‘Izzat ‘Ali ‘Izzat, al-Shakhsiyya alMisriyya fi al-amthal al-sha‘biyya (Cairo, 1997); Muhammad Nu‘man Galal and Magdi Mutawalli, Hawiyyat Misr (Cairo, 1997); Tal‘at Radwan and Fathi Radwan, Ab‘ad al-Shakhsiyya al-Misriyya: Bayna al-Madi wa-al-Hadir (Cairo, 1999). Claude Cahen ‘Khitat’, Encyclopaedia of Islam, 2nd edn, 5: 22; Jack A. Crabbs, Jr., The Writing of History in Nineteenth-Century Egypt (Detroit, 1984), 115–19. Muhammad ‘Abdallah ‘Inan, Misr al-Islamiyya wa Tarikh al-Khitat al-Misriyya (Cairo: 1969), 52–4. Al-Maqrizi, Khitat, 1: 2–3. Ibid., 1: 2. Al-Maqrizi, Durar al-‘Uqud al-Farida fi Tarajim al A‘yan al-Mufida, ed. A. Darwish and M. al-Masri, 2 vols (Damascus, 1995), 2: 63, 193; idem, Khitat, 1: 50, 2: 76, for the passages directly copied from Ibn Khaldun’s dictation and bearing dates spanning more than ten years. The influence of Ibn Khaldun’s interpretive framework is evident in a number of short thematic books by al-Maqrizi, such as his treatise on the calamity of the early fifteenth century, Ighathat al-Umma bi-Kashf al-Ghamma, and his analysis of the rivalry between the Umayyads and the Abbasids, al-Niza‘ wal-Takhasum fima bayn Bani Ummaya wa-Bani Hashim. See M. Mustafa Ziyada, ‘Tarikh Hayat al-Maqrizi’, in Dirasat ‘an al-Maqrizi, Majmu‘at Abhath, M. Ziyada et al. (Cairo, 1971), 13–22; see also Adel Allouche, Mamluk Economics: A Study and Translation of al-Maqrizi’s Ighathat al-Ummah bi-Kashf al-Ghummah (Salt Lake City, 1994). The most clearly structured cycles are those of Tulunid al-Qata’i‘ and Fatimid Cairo; see al-Maqrizi, Khitat, 1: 313–26 and 360–5 respectively. Victor Hugo, ‘ Ceci Tuera Cela’, Livre Cinquième, pt. 2., Notre Dâme de Paris (Paris, 1830–2). As convincingly argued by Haarmann, ‘al-Maqrizi, the Master, and Abu Hamid al-Qudsi, the Disciple’, 154–5. Abu Hamid al-Qudsi’s predilection for architecture and urbanism characterises –203–

N O T E S T O P A G E S 31 – 3 6

54.

55.

56. 57.

58.

most of his work. See, for instance, his short treatise, al-Fawa’id al-Nafissa al-Bahira fi Bayan Hukm Shawari‘ al-Qahira fi Madhahib al-A’imma al-Arba‘a al-Zahira, ed. Amal al-‘Umari (Cairo, 1988). Some, of course, reject the whole historical method and emphasise the particularity of Islam as a religion in endowing all of its art and architecture with somewhat suprahistorical, spiritual and esoteric qualities. For a discussion of this démarche with an emphasis on the Mamluk period see Aly Gabr, ‘The Traditional Process of the Production of Medieval Muslim Art and Architecture: With Special Reference to the Mamluk Period’, Edinburgh Architectural Research 20 (1993): 133–59. Bloom, ‘Mamluk Art and Architectural History’, 40, dropped the whole issue by exclaiming, ‘It remains to be proven that Mamluk builders gave a hoot about symbolic meaning.’ I am not sure whether he meant ‘builders’ specifically or was referring to the entire Mamluk society. See my interpretation in Rabbat, ‘Architects and Artists’, 30–7. Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 293–5, reminds us that many early Mamluk patrons, most notably al-Nasir Muhammad, played a direct role in the design of the buildings they sponsored. I am here obviously pushing Ulrich Haarmann’s salient observation about the presumed objectivity of the sources to locate their subjectivity in their collective mindset and their complicated relationships to the Mamluk elite. See Haarman, ‘al-Maqrizi, the Master, and Abu Hamid alQudsi, the Disciple’, 150. See my full argument on the problem of representation in Mamluk sources in general in ‘Representing the Mamluks’.

CHAPTER 4 1. Mayer, Islamic Architects, 20–7. 2. Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 293–309. 3. The Geniza documents cover the correspondence and legal contracts of the Jewish community of Fustat with all others, both Jewish and non-Jewish; see S. D. Goitein, A Mediterranean Society: The Jewish Communities of the Arab World as Portrayed in the Documents of the Cairo Geniza 6 vols (Berkeley, 1967–93), vol. 1, Economic Foundations, 113; vol. 2, The Community, 467, no. 106. 4. For the text of the inscription, see Max van Berchem, Matériaux pour un Corpus Inscriptionum Arabicarum, Jerusalem, 2 vols (Cairo, 1920–22), 2: 381–92. 5. Ahmad Taymur, al-Muhandisun fi al-‘Asr al-Islami (Cairo, 2nd edn, 1979), 68–74. 6. David James, Qur’ans of the Mamluks (New York, 1988), 65–7. 7. J. W. Allan, ‘Muhammad ibn al-Zain: Craftsman in Cups, Thrones and Window Grilles?’ Levant 28 (1996): 199–207. 8. Mayer, Islamic Architects, 26, and Taymur, Muhandisun, 121–2, opt for the term muhandis, but a muhandis is the designer only in the sense of surveying and laying out the plot. 9. Ibn Khaldun, al-Muqaddima, 3: 935–7, explains how muhandis developed from primarily a surveyor and builder to a real-estate expert and arbitrator. See also, idem, The Muqaddimah, 2: 359. For complementary definitions of muhandis, see Goitein, Mediterranean Society, 1: 113, 4: 38–9; Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 293–5. 10. For a somewhat legalistic definition of the shadd’s duties, see al-Subki, Mu‘id al-Ni‘am wa Mubid al-Niqam, ed. M. A. al-Najjar, A. Z. Shalabi, and M. A. al-‘Uyun (Cairo, 1993), 129; al-Qalqashandi, Subh, 4: 22. 11. Abu Shama, Kitab al-Rawdatayn fi-Akhbar al-Dawlatayn al-Nuriyya wa-l-Salahiyya, ed. M. H. M. Ahmad, 2 vols (Cairo, 1956–62), 1, 2: 488, and al-Maqrizi, Itti‘az al-Hunafa bi-Akhbar al-A’imma al-Fatimiyyin al-Khulafa, ed. M. H. M. Ahmad et al., 3 vols (Cairo, 1967–73), 3: 321; K. A. C. Creswell, Early Muslim Architecture, 2 vols (Oxford, 1932–40), 2: 2. Building fortifications seems to have become Qaraqush’s speciality from then on. Ibn Wasil, Mufarrij al-Kurub, 2: 253, and al–204 –

N O T E S T O PAG E S 3 6 – 4 0

12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

22. 23.

24.

25.

26.

Maqrizi, Suluk, 1: 126, mention that he was put in charge of the repairs of the fortification of Acre after it was briefly conquered by Salah al-Din in 1189. Al-Shuja‘i, Tarikh, 113–14; Ibn Taghri-Birdi, Nujum, 9: 178–98; al-Maqrizi, Khitat, 2: 306; idem, Suluk, 2: 130. Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 295. Ibn ‘Abd al-Zahir, Sirat al-Malik al-Mansur, 55; al-Maqrizi, Khitat, 2: 406–8; Amir al-Shuja‘i’s biography in al-Safadi, Wafi, 15: 475–8. See also Rabbat, Citadel of Cairo, 148–9. Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 296–304. In addition to the references in Behrens-Abouseif’s article, see the larger biography of Shihab al-Din Ahmad ibn al-Tuluni in al-Maqrizi, Durar al-‘Uqud, 1: 235–6. Al-Maqrizi, Khitat, 2: 212; Behrens-Abouseif, ‘Muhandis, Shad, Mu‘allim’, 295. Taymur, Muhandisun, 46. For other faqihs with craftsmen nisbas, such as Ibn al-Naqqash, see alSafadi, Wafi, 4: 209; Ibn Taghri-Birdi, Nujum, 9: 252. For a faqih known as Ibn al-Rassam (Son of the Draftsman), see al-Safadi, Wafi, 22: 179. For the palace, see al-Maqrizi, Suluk, 2: 129; al-Qalqashandi, Subh, 4: 94; Ibn Taghri-Birdi, Nujum, 9: 36–7. For the mausoleum, see Meinecke, Mamlukische Architektur, 1: 37. Taymur, Muhandisun, 45–6. Meinecke, ‘Mamluk Architecture: Regional Architectural Traditions’, 171–2, used this single piece of information to support his theory of regional architectural interaction between Syria and Egypt in the early Mamluk period. S. D. Goitein, ‘Review of Carl Petry, The Civilian Elite of Cairo in the Late Middle Ages (Princeton, 1981)’, Speculum 59, 1 (January 1984): 195–6. Al-Maqrizi, Khitat, 2: 318. Thomas Arnold, Painting in Islam: A Study of the Place of Pictorial Art in Muslim Culture (Oxford, 1928), 22, erroneously attributes the book to al-Maqrizi, when in fact al-Maqrizi does not make such a claim. On the importance of al-Safadi’s lexicon, see Little, ‘al-Safadi as Biographer of his Contemporaries’, 190–210. Three is not the final number of biographies of artists or architects in al-Wafi. Other biographies might still be there to be discovered, for I have systematically checked only seven of the twenty-two published volumes out of a total of twenty-nine. However, three out of approximately 2500 biographies checked gives a fair idea of the relative unimportance of architects among the people worthy of being recorded in the early Mamluk period. Al-Safadi, Wafi, 24: 304; Ibn Wasil, Mufarrij al-Kurub, 5: 343–4; Ibn Nazif, al-Tarikh al-Mansuri, ed. Abul‘id Dudu (Damascus, 1981), 177; Ibn Khallikan, Wafi yyat al-A‘yan, ed. Ihsan ‘Abbas, 6 vols (Beirut, 1969–72), 5: 315–16; al-Udfawi, al-Tali‘ al-Sa‘id, ed. S. M. Hasan (Cairo, 1966), 469–71; Leo Mayer, Islamic Astrolabists and Their Works (Geneva, 1956), 80–1. M. Amari is the only commentator I found who tried to explain the strange nickname Ta‘asif as derived from rakib alta‘asif, which he explains as ‘cavaliere che corre furioso qua e la senza scopo, e pià breve: cavalcante all’impazzata’, see his ‘Estrattai del Tarih Mansuri’, Archivio Storico Siciliano 9 (1884): 119, no. 2. Amari’s interpretation seems to be in accord with the Arabic derivation of the nickname from ta‘sif, namely blindly and stubbornly pursuing a certain way (cf. Ibn Manzur, Tahdhib Lisan al‘Arab (Beirut, 1993), 173). It is very curious that such a man should be described as the plural of ‘impetuous’ or ‘headlong’, which is what the term seems to indicate, although unfortunately no source says why. Charles H. Haskins, Studies in the History of Mediaeval Science 2nd edn (Cambridge, MA, 1927), 253, reports, ‘While in the East Frederick II [of Hohenstaufen] asked for an interview with someone learned in astronomy, and in response Sultan al-Kamil sent him a most learned astronomer and mathematician surnamed al-Hanifi [sic].’ See also E. Wiedemann, ‘Fragen aus dem Gebiet der Naturwissenschaften, gestellt von Friedrich II’, in Archiv für Kulturgeschichte 11 (1914): 483–5; Michele Amari, ‘Questions philosophiques adressées aux savants musulmans par l’Empereur Frédéric II’, Journal Asiatique 5eme série, 1 (1853): 240–74. Zakariyya al-Qazwini, however, in his Kitab Athar al-Bilad wa-Akhbar al-‘Ibad, ed. Ferdinand Wüstenfeld (Göttingen, 1848), 309–10, –205 –

N O T E S T O PAG E S 4 0 – 4 9

27. 28. 29. 30. 31. 32.

attributes the solution of the geometric riddles to al-Shaykh Kamal al-Din ibn Yunus of Mosul, who was in fact one of Ta‘asif’s shaykhs. See the biography of ‘Ala’ al-Din al-A‘ma al-Rukni al-Zahid (Aydghadi or Taybars?) in al-Safadi, Wafi, 9: 485; also al-Safadi, Nakt al-Himyan fi-Nukat al-‘Umyan, ed. A. Zaki (Cairo, 1911), 123; al-Jazari, Jawahir al-Suluk fi al-Khulafa’ wa-l-Muluk, mss.7575 H, Dar al-Kutub, Cairo, 262–3. Fedwa Malti-Douglas, ‘Mentalités and Marginality: Blindness and Mamluk Civilization’, in Essays in Honor of Bernard Lewis: The Islamic World from Classical to Modern Times, ed. C. E. Bosworth et al. (Princeton, 1989), 211–37. Leo Mayer, Islamic Metalworkers and Their Works (Geneva, 1959), 16. Al-Safadi, Wafi, 11: 213–14, copied almost verbatim in Ibn Taghri-Birdi, al-Manhal al-Safi wa-al-Mustawfi ba‘d al-Wafi, ed. Muhammad M. Amin et al., 5 vols (Cairo, 1984–90), 5: 32–3. Taymur, Muhandisun, 41–2. A fuller biography of al-Harithy is in Ibn Abi Usaybi‘a, ‘Uyun al-Anba fi Tabaqat al-Atibba, ed. Nizar Rida, 2 vols (Beirut, 1956–7), 2: 190–1. Beside the geometrical basis of most of what carpenters do, there might have been a symbolic connection as well in the carpenters’ interest in Euclid. He appears in the medieval Arabic sources with the epithet al-Najjar (the Carpenter).

CHAPTER 5 1. Partial lists of the Mamluk mosaics of Cairo appear in L. Hautecœur and G. Wiet, Les mosquées du Caire, 2 vols (Paris, 1932), 1: 116; K. A. C. Creswell, Muslim Architecture of Egypt, 2 vols (Oxford, 1959) 2: 138, 226, and plate 81 for Lajin’s Mihrab; Michael Meinecke, ‘Das Mausoleum des Qala’un in Kairo: Untersuchungen zur Genese der Mamlukischen Architekturdekoration’, Mitteilungen des Deutsches Archäologisches Institut, Abteilung Kairo, vol. 27 (1971): 62, n. 91. The Damascus ones are less well-known. Meinecke’s article discusses those of the Zahiriyya, and a few articles on the mosaics of the Umayyad Mosque mention them in passing. For the three examples in Jerusalem see Miriam Rosen-Ayalon, ‘A Neglected Group of Mihrabs in Palestine’ in M. Sharon, Studies in Islamic History and Civilization in Honor of Professor David Ayalon (Jerusalem, 1986), 553–63. For Tripoli, see Hayat Salam-Leibish, The Architecture of the Mamluk City of Tripoli (Cambridge, MA, 1983), 49–50. 2. The preliminary report on the excavation was published by M. al-Hadidi and F. ‘Abd al-‘Alim, ‘’A‘mal Tarmim al-Qasr al-Ablaq bi-Qal‘at Salah al-Din’, ‘Alam al-Athar 26 (April 1986): 4–16. 3. Mufaddal ibn Abi al-Fada’il, al-Nahj al-Sadid wa-l-Durr al-Farid fi-ma Ba‘d Tarikh ibn al-‘Amid, ed. E. Blochet, in Patrologia Orientalis (Paris, 1919–20), 14: 443–4. See also Muhammad Ahmad Duhman, ‘al-Madrasa al-Zahiriyya’, in his Fi Rihab Dimashq (Damascus, 1982), 112–29. 4. Meinecke, ‘Mausoleum des Qala’un’, 64. 5. A complete description of the Qubba al-Zahiriyya’s non-architectural representations is in ibid., 65–6. 6. This group appears as a residue of the depictions of exedrae and porticoes found in the Umayyad Mosque’s mosaics such as the one discussed by Barbara Finster, ‘Die Mosaiken der Umayyadenmoschee von Damaskus’, Kunst des Orients 7, 2 (1972): 83–136, esp. 93–7. If this interpretation of the Mamluk scene is correct, it will explain the existence of the mysterious green dots between the rear columns that may be the remainder of the windows found in the back wall of the exedra’s representation, such as the one published in Eustache De Lorey, ‘Les mosaïques de la mosquée des omayyades à Damas’, Syria 12 (1931), pt. 10. 7. See Finster, ‘Mosaiken’, figs 33 and 35. The closer example attributed to Baybars is reproduced in De Lorey, ‘mosaïques’, fig. 12. 8. These two bowls greatly resemble another Damascene example closer to them in date: the stylised vase found under the Nur al-Din inscription panel on the eastern portico of the Umayyad Mosque, uncovered in 1953, which may be dated to 1159. See Nikita Élisséeff, ‘Les monuments de Nur al-Din’, Bulletin d’Études Orientales 13 (1949–51): 5–43; the panel is reproduced in ‘Abd –206 –

N O T E S T O PAG E S 4 9 – 5 5

9.

10.

11.

12.

13.

14. 15.

16.

al-Qadir al-Rihawi, ‘Fusayfusa’ al-Jami‘ al-Amwi’, Les annales archéologiques arabes syriennes 10 (1960), fig. 6. Marguerite van Berchem, ‘The Mosaics of the Dome of the Rock at Jerusalem and of the Great Mosque in Damascus’, in Creswell, Early Muslim Architecture, 1: 150–252, 247. She identifies a whole group of architectural elements in the mosaics of the transept of the Umayyad Mosque that may have been derived from the medieval Islamic repertoire. This feature was similarly used by Sauvaget in his attempt to prove that an Italian painting from the late Renaissance effectively represented Mamluk Damascus, see Jean Sauvaget, ‘Une ancienne représentation de Damas au Musée du Louvre’, Bulletin d’Études Orientales 11 (1945–6): 5–12. Meinecke, ‘Mausoleum des Qala’un’, 63–9. He had to adjust the chronology of mosaic appearance to preserve the precedence of the Qubba al-Zahiriyya’s mosaics in Damascus, for two of the Cairene structures decorated with mosaic were constructed before it: the mausolea of al-Salih (1250) and Shajar al-Durr (1256). He achieved this on stylistic grounds by contending that the mosaic representation in the mihrab of Shajar al-Durr’s mausoleum – that of al-Salih’s had disappeared some years ago – is subsequent to Qalawun’s madrasa’s mosaics, which he argued was decorated by Syrian craftsmen who were brought to Cairo after they finished the Qubba alZahiriyya. Baybars in fact effectuated mosaic repairs both at the Dome of the Rock and the Umayyad Mosque, see Mujir al-Din al-‘Ulaymi, al-Uns al-Jalil bi Tarikh al-Quds wa-l-Khalil, ed. M. Bahr al-‘Ulum (Najaf, Iraq, 1968), 2: 87–8; Ibn-Shaddad, Tarikh al-Malik al-Zahir, 355. Ibn Fadl-Allah al-‘Umari, Masalik al-Absar fi-Mamalik al-Amsar, (reproduced from MS. 2797/1, Ahmet III Collection, Topkapi Sarayi Library, Istanbul) ed. Fuat Sezgin (Frankfurt, 1988) 1: 144. Tankiz must have used the same atelier in the decoration of his two madrasas in Palestine, at Jerusalem and Hebron (both built in the late 1320s), for it is very difficult to posit the presence of qualified mosaicists in the area then, see Michael Burgoyne, Mamluk Jerusalem: An Architectural Study (London, 1987), 236, and n. 70. There is another possible and curious parallel between the Mamluk mosaics and the branch of manuscript illustrations in which there was a high degree of realism in architectural representations, namely the miniatures of Maqamat al-Hariri. That is that they were synchronous: they flourished together and went out of fashion almost at the same time. See, Oleg Grabar, ‘The Illustrated Maqamat of the Thirteenth Century: The Bourgeoisie and the Arts’, in The Islamic City: A Colloquium, ed. A. H. Hourani and S. M. Stern (Oxford, 1970) 208–10. The implications of this similarity are beyond the scope of this study, but they surely would elucidate wider aspects of the social and cultural history of the period. Several of the manuscripts of the Maqamat alHariri contain such depictions, such as the Leningrad MS. S 23, Forty-Second Maqama, dated to 1230; also Rasa’il Ikhwan al-Safa, MS. Essad Effendi 3638, Topkapi Saray, dated to 1287. Both manuscripts are attributed to the Baghdad School. Kashf al-Asrar, MS. Lala Ismail 565, Süleimaniye Library, Istanbul, f16v, see Duncan Haldane, Mamluk Painting (Warminster, 1978), 52–3. Both Baybars and Qalawun spent long stretches of time in Damascus and both effectuated repair works and additions in the Umayyad Mosque and the other charitable institutions around it, such as the bimaristan of Nur al-Din, which was renovated by Qalawun in 1281. The representation of conquered cities as a means of asserting the power of rulers in either painting or low relief on the walls of palaces is known from the earliest periods of Egyptian and Mesopotamian kingdoms. Maps were not a familiar means of visual communication in the fourteenth century, nor were models and mock-ups. The only other way that was conceivable, and which was used, was writing, and it was exhibited in the addition of new attributes in the titulatures of rulers. See the analysis of Nur al-Din’s titulature development in Yasser Tabbaa, ‘Monuments with a Message: Propagation of Jihad under Nur A-Din (1146–1174)’, The Meeting of Two Worlds: Cultural Exchange between East and West during the Period of the Crusades, ed. V. P. Goss and C. V. Bornstein (Kalamazoo, MI, 1986), 223–40, esp. 226. –207–

N O T E S T O PAG E S 5 6 – 5 9

17. The inferior quality of imitation in the repaired scenes at the Umayyad Mosque and in new Mamluk structures were noticed even by the contemporary historian al-‘Umari, Masalik, vol. 1, 193, suggests that the craftsmen did not master the medium well enough to break with the old conventions and venture into new compositions. 18. Ibn-‘Abd al-Zahir, Sirat al-Malik al-Mansur, 139. Ibn al-Furat, Tarikh, 8: 38, and al-Maqrizi, Khitat, 2: 212, give similar reports but do not mention the images. 19. Obviously a modern positivist sensibility could never accept such an interpretation and, in fact, Marguerite van Berchem rejected the assertions of al-Muqqadasi and that of a later Damascene, Ibn Shakir, that they could see in the mosaics of the Umayyad Mosque the depictions of various locales; see van Berchem, ‘Mosaics’, 163. The notion of exactness between original and copy in Western architecture in the medieval period is discussed in Richard Krautheimer’s seminal work, ‘Introduction to an “Iconography of Medieval Architecture”’, Journal of the Warburg and Courtauld Institutes 5 (1942): 1–33; reprinted in Studies in Early Christian, Medieval and Renaissance Art (London and New York, 1969), 115–50, esp. 126–8. 20. Meinecke, ‘Mausoleum des Qala’un’, 64. 21. Al-Muqqadasi, Ahsan al-Taqasim fi-Ma‘rifat al-Aqalim, ed. M. J. de Goeje (Bibliotheca Geographorum Arabicorum, vol. 3) (Leiden, 1906) 157; English text in van Berchem, ‘Mosaics’, 163. 22. Ibn-‘Asakir, Tarikh Madinat Dimashq wa Zikr Fadlaha wa-Tasmiyat man Hallaha min-al-Amathil aw Ijtaza bi-Nawahiha min Waridiha wa-Ahlaha, 19 vols (reproduced from MS in al-Zahiriyya in Damascus), ed. M. al-Tarhuni (Damascus, 1980?), 1: 311–13; French translation by Nikita Élisséeff, La déscription de Damas d’Ibn ‘Asakir (Damascus, 1959), 57–9. 23. Yaqut al-Hamawi, Mu‘jam al-Buldan, ed. Ferdinand Wüstenfeld (Leipzig, 1866–70), 8: 465. 24. Al-Qazwini, Kitab Athar al-Bilad, 137. 25. Sheikh al-Rabwa al-Dimashqi, Nukhbat al-Dahr fi ‘Aja’ib al-Barr wa-l-Bahr, ed. M. A. F. Mehren (Leipzig, 1923), 193. 26. Ibn Shakir al-Kutubi, ‘Uyun al-Tawarikh, trans. Henri Sauvaire, ‘La Description de Damas’, Journal Asiatique 3, 7 (1896): 369–72; English translation in van Berchem, ‘Mosaics’, 1: 161, no. 9. ‘Abd al-Basit al-‘Ulmawi, Mukhtasar Tanbih al-Talib; Shams al-Din al-Busrawi, Tuhfat al-Anam fi Fada‘il al-Sham, both are published in Henri Sauvaire, ‘La Description de Damas’, 422; English translations in van Berchem, ‘Mosaics’, 163. 27. Creswell, Early Muslim Architecture, 1: 119, n. 5. 28. Abu al-Baqa’ al-Badari, Nuzhat al-Anam fi-Mahasin al-Sham (Beirut, 1980), 25. Ahmad Taymur, al-Taswir ‘Ind al-‘Arab, ed. Z. M. Hassan (Cairo, 1942) 4, reported the story of al-Badari and traced its probable origin to al-Muqqadasi, although he did not note the introduction of the identification of the Ka‘ba.

CHAPTER 6 1. On the history of the Turks in Central Asia and the process of their Islamisation, see V. V. (Vasilii Vladimirovich) Bartol’d, Turkestan Down to the Mongol Invasion, ed. and trans. T. Minorsky, with addenda and corrigenda by C. E. Bosworth (London, 1968); Yuri Bregel, The Role of Central Asia in the History of the Muslim East (New York, 1980); Rene Grousset, The Empire of the Steppes: A History of Central Asia (New Brunswick, NJ, 1970). A similar process occurred in the direction of India with the disintegration of the Ghurid empire in the twelfth century, but this article will focus on the westward movement of the Turks. 2. On the Turkish migration to the Middle East, see Claude Cahen, ‘La première pénétration turque en Asie-Mineure’, Byzantion 18 (1946–8): 5–67; Shakir Mustafa, ‘Dukhul al-Turk al-Ghuzz ila Bilad al-Sham’, in The International Congress on the History of Bilad al-Sham, ed. A. K. Gharaybeh, A. A. al-Duri and O. al-Madani (Beirut, 1974), 303–98. 3. The counter-Crusade is the most studied movement of medieval Jihad. See Hadia DajaniShakeel, ‘Jihad in Twelfth-Century Arabic Poetry: A Moral and Religious Force to Counter the –208 –

N O T E S T O PAG E S 5 9 – 6 3

4.

5. 6. 7. 8. 9.

10.

11. 12. 13. 14.

15.

Crusades’, The Muslim World 66, 2 (April 1976): 96–113; ‘Imad al-Din Khalil, al-Imara al-Artuqiyya fi al-Jazira wa al-Sham (AH 456–812/1072–1409 AD): Adwa’ Jadida ‘ala al-Muqawama al-Islamiyya lil-Salibiyyin wal-Tatar (Beirut, 1980); Sivan, L’Islam et la Croisade; Margaret E. Bertsch, ‘Countercrusade: A Study of Twelfth Century Jihad in Syria and Palestine’, Ph.D. Thesis, University of Michigan, 1993; Carole Hillenbrand, ‘The Career of Najm al-Din Il-Ghazi’, Der Islam 58 (1981): 250–92; Malcolm C. Lyons, ‘The Crusading Stratum in the Arabic Hero Cycles’, in Crusaders and Muslims in Twelfth-Century Syria, ed. M. Shatzmiller (Leiden, 1993); Shatzmiller, ‘The Crusades and Islamic Warfare’, 247–88. For a discussion on the linguistic barriers in the Mamluk society, see Rabbat, ‘Representing the Mamluks’, 59–75. For an analysis of the Mamluk phenomenon in medieval Islam, the work of David Ayalon offers an exhaustive source, albeit strictly textual in its interpretation. See his ‘Preliminary Remarks on the Mamluk Military Institution in Islam’, 44–58; idem, ‘Aspects of the Mamluk Phenomenon’; idem, ‘Mamlukiyyat: (A) A First Attempt to Evaluate the Mamluk Military System (B) Ibn Khaldun’s View of the Mamluk Phenomenon’, 321–33; idem, ‘Mamluks of the Seljuqs: Islam’s Military Might at the Crossroads’, Journal of the Royal Aisatic Society Series 3, 6, 3 (1996): 305–33. Haarmann, ‘Arabic in speech’, 81–114; idem, ‘Injustice of the Turks’, 61–79; idem, ‘Ideology and History, Idendity and Alterity: The Arab Image of the Turk from the Abbasids to Modern Egypt’, International Journal of Middle Eastern Studies 20, 2 (May 1988): 175–96. Chamberlain, Knowledge, 37–54, discusses these developments in Damascus. See a more contextualised argument in Rabbat, ‘Dar al-‘Adl’, 3–28. A broader argument that emphasises the internal struggle in the Islamic world is Yasser Tabbaa, The Transformation of Islamic Art during the Sunni Revival (Seattle, 2001). The witty observer Ibn Jubair, al-Rihla (Beirut, 1980), 216, commenting on the profusion of elaborate titles, remarks that many of these princes do not deserve the honours they attribute to themselves, but excludes Salah al-Din from his general judgement. Few scholars noticed this phenomenon. Most of those who have paid attention were motivated by a nationalistic Turkocentrism, see for instance Emel Esin, ‘The Turk al-Agam of Samarra and the Paintings Attributable to them in the Gawsaq al-Haqani’, Kunst des Orients 7, 2 (1972): 47–88; idem, ‘L’arme zomorphe du guerrier turc (Étude iconographique)’, in Sprache, Geschichte und Kultur der Altaischen Völker: Protokollband der Permanent International Altaistic Conference 1969 in Berlin, ed. G. Hazai and P. Zieme (Berlin, 1974), 193–217. One exception is J. M. Rogers, ‘The 11th century: A Turning Point in the Architecture of the Mashriq’, in Islamic Civilization 950–1150, ed. D. S. Richards (Oxford, 1973), 211–49. For a fuller discussion on the fortification of cities under the Seljuqs due to the changes in the military balance of power prior to the Crusades, see Stefan Heidemann, Die Renaissance der Städte in Nordsyrien und Nordmesopotamien: Städtische Entwicklung und wirtschaftliche Bedingungen in ar-Raqqa und Harran von der Zeit der beduinischen Vorherrschaft bis zu den Seldschuken (Leiden, 2002), 263–6. ‘Izz al-Din Ibn Shaddad, al-A‘laq al Khatira fi Dhikr Umara’ al-Sham wal-Jazira: Tarikh Halab, ed. Dominique Sourdel, vol. 1, pt 1 (Damascus, 1953), 23–7. See the analysis of Yasser Tabbaa, Constructions of Power and Piety in Medieval Aleppo (University Park, PA, 1997). For a fuller analysis of the uneven relationship between the Arab Bedouins and the Turkic states in the region, see Stefan Heidemann, ‘Arab Nomads and Seljuq Military’, in Militär und Staatlichkeit, ed. Irene Schneide (Halle, 2003), 201–20. See, for instance, Ulrich Haarmann, ‘Turkish legends in the popular historiography of medieval Egypt’, in Proceedings of the VIth Congress of Arabic and Islamic Studies, ed. F. Rundgren (Stockholm; Leiden, 1972), 97–107; Shihab al-Sarraf, ‘L’impact des techniques militaires sur l’évolution politique et sociale dans le moyen-orient médiéval: le cas de l’archerie’, Études Orientales-Dirasat Sharqiya 7/8 (1990): 6–28. Other citadels in the same area, Diyar Bakr in the Jazira, date to the same general period and show that Bedouin tribal dynasties were the first builders of citadels but that the new Turkic –209 –

N O T E S T O PAG E S 6 3 –71

16. 17.

18. 19.

20.

21.

22.

23. 24. 25. 26. 27. 28. 29. 30.

rulers either rebuilt the extant citadels or built new ones. See the examples of Nusaybin, Harran and Sanjar in Ibn Shaddad, al-A‘laq al Khatira, vol. 3, pt 1, ed. Yahya ‘Abbara (Damascus, 1978), 124, 140 and 155, respectively. I have studied the process in detail for the Citadel of Cairo, see Rabbat, Citadel of Cairo, passim. Ibn Shaddad, al-A‘laq al Khatira: Tarikh Madinat Dimashq, ed. Sami al-Dahhan (Damascus, 1956), 37. The Citadel of Damascus is slowly receiving the scholarly interest it deserves, see ‘Abd alQader Al-Rihawi, Qal‘at Dimashq (Damascus, 1980); Paul E. Chevedden, ‘The Citadel of Damascus’, Ph.D. Thesis, UCLA, 1986; Hanspeter Hanisch, Die ayybidischen Toranlagen der Zitadelle von Damaskus: ein Beitrag zur Kenntnis des mittelalterlichen Festungsbauwesens in Syrien (Wiesbaden, 1996). See Jere L. Bacharach, ‘The Court-Citadel: An Islamic Urban Symbol of Power’, in The Proceedings of International Conference on Urbanism in Islam (ICUIT), ed. T. Yukawa (Tokyo, 1989) 206–45. See Tabbaa, ‘Monuments with a Message’, 223–40; idem, ‘The Transformation of Arabic Writing: Part 2: The Public Text’, Ars Orientalis 24 (1994): 119–46; Erica C. Dodd, ‘The Image of the Word (Notes on the Religious Iconography of Islam)’, Berytus 18 (1969): 35–62; Irene Bierman, Writing Signs: The Fatimid Public Text (Berkeley, 1998): 1–59. Joachim Gierlichs has documented most examples in his two books, Mittelalterliche Tierreliefs in Anatolien und Nordmesopotamien: Untersuchungen zur figurlichen Baudekoration der Seldschuken, Artuqiden und ihrer Nachfolger bis ins 15. Jahrhundert (Tubingen, 1996); idem, Drache, Phonix, Doppeladler: Fabelwesen in der islamischen Kunst (Berlin, 1993). One of the most telling examples is the double-headed eagle at the Citadel of Cairo. Although it is most probably Ayyubid, it was never noted by any of the Mamluk historians who otherwise filled pages of description of the citadel. The double-headed eagle was first reported by Evliya Çelebi (c. 1670) in Seyahatnamesi, ed. Mümin Çevik (Istanbul, 1984) 9–10: 385–6. For a brief discussion of the relief, see Rabbat, Citadel of Cairo, 24. Examples include: Gönül Öney, ‘Mounted Hunting Scenes of Anatolian Seljuqs in Comparison with Iranian Seljuqs’, Anatolia 11 (1967): 121–89; idem, ‘Elements from Ancient Civilizations in Anatolian Seljuk Art’, Anatolica 12 (1970): 17–38; Katharina Otto-Dorn, ‘Figural Stone Reliefs on Seljuq Sacred Architecture in Anatolia’, Kunst des Orients 12, 1/2 (1978/79): 103–46; Eva Baer, ‘The Ruler in Cosmic Setting: A Note on Medieval Islamic Iconography’, in Essays in Islamic Art and Architecture in Honor of Katharina Otto-Dorn, ed. Abbas Daneshvari (Malibu, CA, 1981), 13–19; Whelan, ‘Representations of the Khassakiyah’, 219–43. See, for instance, Laila ‘Ali Ibrahim, ‘ Dragons on a Cairene Madrasa’, Art and Archaeology Research Papers 10 (1976): 11–19, for a Cairene example; for a survey of the Anatolian examples see Gönül Öney, ‘Dragon Figures in Anatolian Seljuk Art’, Belleten 32/130 (1971): 171–216. Max van Berchem, Amida: Matériaux pour l’épigraphie et l’histoire musulmanes du Diyar-Bekr (Paris, 1910), 82–3. Ibn Shaddad, Tarikh al-Malik al-Zahir, 340; Ibn Kathir, Bidaya, 13: 275; Ibn Taghri-Birdi, Nujum, 7: 190. For another discussion of this qubba, see Meinecke, Mamlukische Architektur, 1: 29–31. A yalmaq, a word of Persian origin, means ‘a furred coat of the Tartars (Mongols)’, see J. G. Hava, Al-Fara’id al-Duriyya, ‘Arabi/Inglizi, (Beirut, 1982), 904. Whelan, ‘Representations of the Khassakiyah’, 221–5. Doris Behrens-Abouseif, ‘The Baptistère de Saint Louis: A Reinterpretation’, Islamic Art 3 (1988–9): 3–13. Meinecke, ‘Zur mamlukischen Heraldik’, 213–87; J. W. Allan, ‘Mamluk Sultanic Heraldry and the Numismatic Evidence: A Reinterpretation’, Journal of the Royal Asiatic Society 2 (1970): 19–112; Nasser Rabbat, article ‘Rank’, Encyclopaedia of Islam, 2nd edn, 7: 431–3. Helen Mitchell Brown, ‘Some Reflections on the Figured Coinage of the Artuqids and Zengids’, in Near Eastern Numismatics, Iconography, Epigraphy and History: Studies in Honor of George C. Miles, ed. D. Kouymjian (Beirut, 1974), 353–8; Nicholas Lowick, ‘The Religious, the Royal and the Popular

– 21 0 –

N O T E S T O PAG E S 71 –7 9

in the Figural Coinage of the Jazira’, in The Art of Syria and the Jazira 1100–1250, ed. Julian Raby (Oxford, 1985), 159–74. 31. See the analysis of Dorothea Krawulsky concerning the change in historical production in the Mamluk period in ‘al-Intaj al-Thaqafi wa-Shar‘iyat al-Sulta’, her introduction to al-‘Umari’s Masalik, 15–37. She presents a perceptive interpretation of the political and cultural impetus behind Mamluk historical writing and distinguishes between the early Bahri period and the period after al-Nasir Muhammad’s reign. Remke Kruk, ‘History and Apocalypse’, 324–37, offers a politico-cultural analysis of Ibn al-Nafis’s unusual treatise. 32. Al-Maqrizi, Khitat, 2: 214, analyses the degradation of the Mamluk recruiting system in the Burji period and connects it to the general decline of the sultanate. See also Ibn Taghri-Birdi, Nujum, 7: 328–9. For a more complete analysis of al-Maqrizi’s attitudes towards the Mamluks see Rabbat, ‘al-Madina wa al-Tarikh wa al-Sultah: al-Maqrizi wa Kitabuhu al-Ra’id al-Mawa‘iz wa al-I‘tibar bi Dhikr al-Khitat wa al-Athar’, Annales Islamologiques 35 (2001): 77–100, see also Sami Massoud, ‘al-Maqrizi as a Historian of the Reign of Barquq’, Mamluk Studies Review 7/2 (2003): 119–36. 33 Muhammd ‘Abd al-Mu‘ti al-Ishaqi al-Munufi, Akhbar al-Uwal fi man Tasarrafa fi Misr min Arbab alDuwal (Cairo, 1890), 140; quoted in Hasan ‘Abd al-Wahab, Tarikh al-Masajid al-Athariyya allati salla fiha faridhat al-jum‘a hadret sahib al-jalala al-malik al-salih Farouq al-Awwal (Cairo, 1946), 166.

CHAPTER 7 1. Martin S. Briggs, Muhammadan Architecture in Egypt and Palestine (New York, 1924; reprint 1974), 67. 2. Taha Hussein, al-Ayyam, 2 vols (Cairo, 1962), 2: 16. 3. Al-Maqrizi, Itti‘az, 1: 111, quoting Ibn Zulaq who saw al-Mansuriyya while it was under construction. Jamal al-Din al-Shayyal, editor of vol. 1 of Itti‘az, suggests in his note 1 that the name of the Ifriqiyan city may have been reused in Egypt because the new Fatimid cities bore a similar topographical relation to the older capitals, Qayrawan and Fustat. 4. Mustafa ‘Ali, Mustafa ‘Ali’s Description of Cairo of 1599, trans. and ed. A. Tietze (Vienna, 1975) 32. 5. For a list of popular movements that began at al-Azhar, see André Raymond, Le Caire (Paris, 1993), 239 and 259; also Mustafa M. Ramadan, Dawr al-Azhar fi al-Hayat al-Misriyya (Cairo, 1986), 111–52. 6. The mosque’s architecture and its development over time are described in great detail in Creswell, Muslim Architecture of Egypt, 1: 36–64. 7. Ahmad Fikri, Masajid al-Qahira wa Madarisaha: al-‘Asr al-Fatimi (Cairo, 1965), 127–40, argues that it is wrong to use the word transept in designating this transverse aisle, and proves that it never functioned in a mosque as a transept in a church. 8. Creswell, Muslim Architecture of Egypt, 1: 45. 9. Ibid., 1: 46, mistakenly identified this Nafisa as the daughter of ‘Ali ibn Abi Taleb. For the correct identification of the Nafisa buried at al-Azhar, see ‘Ali Pasha Mubarak, al-Khitat al-Tawfiqiyya al-Jadida, 20 vols (reprint Cairo, 1969), 2: 92. 10. Mubarak, Khitat, 4: 49–58, and K. Vollers, ‘Azhar’, Encyclopaedia of Islam, 1st edn, 1: 533 offer detailed lists of these arwiqa. 11. On Creswell’s methods, see J. M. Rogers, ‘Architectural History as Literature: Creswell’s Reading and Methods’, Muqarnas 8 (1991): 45–54. 12. Jonathan Bloom, ‘Meaning in Early Fatimid Architecture: Islamic Art in North Africa and Egypt in the 4th c. H/10th c. A.D.’, Ph.D. Thesis, Harvard University, 1980, 99–105. 13. Ibn ‘Abd al-Zahir, Rawd, 278; Doris Behrens-Abouseif, The Minarets of Cairo (Cairo, 1985), 62, reports the same instance without any reference. 14. Al-Maqrizi, Khitat, 2: 276; idem, Itti‘az, 3, 2: 320, says only that the sessions for missionaries (majalis al-da‘wa) at al-Azhar were eliminated.

– 211 –

N O T E S T O PAG E S 7 9 – 8 2

15. Idem, Suluk, 1: 150, reports that Ibn Dirbass built several shops and stables in the ziyada of al-Azhar next to his own house, showing how intent he was on downgrading the mosque by appropriating its ziyada for his personal use. 16. Ibn al-Furat, Tarikh, 7: 119–20; see also the biography of Suleiman al-Hanafi in al-Safadi, Wafi, 15: 404–5. 17. Ibn ‘Abd al-Zahir, Rawd, 277; al-Maqrizi, Khitat, 2: 275. 18. For a different interpretation of the ‘towers’ in pre-Azhar Cairene and Ifriqiyan mosques, with which I disagree for reasons that will become apparent, see Jonathan Bloom, Minaret: Symbol of Islam (Oxford, 1989), 125–44. 19. This is the date given by al-Maqrizi, Khitat, 2: 268, for the building of the first minaret (manar) at the ‘Amru Mosque. 20. Al-Maqrizi, Khitat, 2: 318. Jonathan Bloom, ‘The Mosque of the Qarafa in Cairo’, Muqarnas 4 (1987): 7–8, misinterpreted al-Maqrizi’s text and reconstructed the plan of both mosques as having a clerestory (which is how he understood manar) in the middle of the haram and no minaret. The same assertion is repeated in idem, Minaret, 130. 21. Al-Maqrizi, Khitat, 2: 276. 22. The minaret is dated by Qaytbay’s inscriptions, van Berchem, Corpus Inscriptionum Arabicarum, Egypt, 1, no. 21. 23. Alexandre Lézine, Mahdiya: recherches d’archéologie islamique (Paris, 1965), 93. 24. Bloom, ‘Early Fatimid Architecture’, 187–94; also, idem, ‘The Mosque of al-Hakim in Cairo’, Muqarnas 1 (1983): 24–6, offers a more elaborate explanation of the monumental Fatimid portals than the one based on a readapted classical model, first suggested by Alexandre Lézine. 25. Bloom, ‘Early Fatimid Architecture’, 103, reached this point, but opted for the correct linguistic interpretation. 26. For definitions of the three terms haniya, kuwwa and mastaba with references to medieval Arabic lexica, see Laila Ibrahim and M. M. Amin, al-Mustalahat al-Mi‘mariyya fi-l-Watha’iq al-Mamlukiyya (Cairo, 1990), 38, 97, 106. 27. Bloom, ‘Early Fatimid Architecture’, 105. 28. Al-Maqrizi, Khitat, 2: 273. 29. Mubarak, Khitat, 4: 48. 30. Ibn Taghri-Birdi, al-Manhal al-Safi, 5: 41. 31. Al-Maqrizi, Khitat, 1: 348, 359, 377. 32. Ibid., 1: 360, states in passing that Jawhar incorporated the garden and maydan in his new city. The existence of this planned open ground, which would have provided the best accommodations for an army camp, may have been what attracted Jawhar to the site in the first place, for otherwise we read echoes of al-Mu‘izz’s disatisfaction with the site after he came to Egypt and pointed to two better sites nearby, al-Maqs for its proximity to the river and al-Rasd for its defensibility; see al-Maqrizi, Itti‘az, 1: 112–13. 33. Idem, Khitat, 2: 138; Ibn al-Dawadari, Kanz al-Durar wa-Jami‘ al-Ghurar, vol. 6, al-Durra al-Mudiyya fi Akhbar al-Dawla al-Fatimiyya, ed. Salah al-Din al-Munajjid (Cairo, 1961), 137–42, elaborates more on the khitat of Cairo as established by Jawhar’s army around the qasr, and specifies that he is copying from a manuscript of Ibn ‘Abd al-Zahir’s al-Rawda al-Bahiyya fi Khitat al-Qahira al-Mu‘izziyya, dated by the author to AH 647 (1249), which was the source for al-Maqrizi as well. 34. The existence and location of this musalla is first reported in 985 by al-Muqqadasi, Ahsan alTaqasim, 200; al-Maqrizi, Khitat, 1: 451–7, describes the varying ceremony of ‘Id prayer in the musalla at three different instances during the Fatimid period. The musalla lost its ceremonial role after Fatimid times and became a space of prayer for the dead, probably the one known throughout the Mamluk period as the Musalla of Bab al-Nasr; al-Maqrizi, Khitat, 2: 5, 138. 35. Ibid., 1: 125, 2: 454–5. 36. Bloom, ‘Early Fatimid Architecture’, 90–3; see also Paula Sanders, Ritual, Politics, and the City in Fatimid Cairo (Albany, NY, 1994), 45, 50. – 21 2 –

N O T E S T O PAG E S 8 2 – 8 6

37. On his first Friday in Egypt, Jawhar and his soldiers prayed at the Mosque of ‘Amru; nine months later, he prayed at Ibn Tulun’s mosque on the occasion of the introduction of the Shi‘ite phrase ‘hayy ‘ala khayr al-‘amal’, to the adhan, al-Maqrizi, Itti‘az, 1: 114, 120 38. Bloom, ‘Early Fatimid Architecture’, 94, uses al-Maqrizi, Khitat, 2: 270, 273, to establish the chain of events that led to the institution of the Shi‘ite prayer in Egypt after the arrival of the Fatimids. 39. Bloom, ‘Early Fatimid Architecture’, 79, quotes al-Maqrizi, Itti‘az, 1: 133, to argue that Jawhar started the building of the caliphal palace in Rabi‘ II, 362 (972) when he heard that his master was coming to Egypt, almost a year after the reported date of the mosque’s consecration. AlMaqrizi’s text (wa akhadha fi-‘amarat al-qasr wa-l-ziyada fihi) clearly indicates that Jawhar expedited the building of the qasr and added to it, which, contrary to Bloom’s explanation, means that the construction had begun earlier but may have slackened until the news of the imminent arrival of al-Mu‘izz. This date would have given Jawhar only four months to complete the palace, hardly enough time to build a royal structure. 40. Ibn al-Dawadari, al-Durra al-Mudiyya, 143, cites a poem recited to mark al-Mu‘izz’s arrival to his new capital in which the qusur are qualified as qahira, the victorious, an otherwise unusual attribute for palaces in Arabic poetry. 41. Bloom, ‘Early Fatimid Architecture’, 95–6. Al-Maqrizi, Khitat, 1: 362, says that the name al-Azhar is recent and that the mosque’s name appears as Jami‘ al-Qahira in ‘history books’. 42. Ibn ‘Abd al-Zahir, Rawd, 278; al-Maqrizi, Khitat, 2: 273. 43. Creswell, Muslim Architecture of Egypt, 1: 36. 44. Ibid., 1: 62; Muhammad ‘Abdallah ‘Inan, Tarikh al-Jami‘ al-Azhar fi al-‘Asr al-Fatimi (Cairo, 1942), 20. 45. The significance of the number seven in Egyptian Islamic architecture has yet to be deciphered. We encounter the seven qa‘as and the seven hadarat at the Citadel of Cairo (al-Maqrizi, Khitat, 2: 59), the seven domes (a series of Fatimid domes in the desert near Cairo that apparently only numbered six, al-Maqrizi, Khitat, 2: 459) and the tower of the seven qa‘as in Damietta (Mufaddal ibn Abi al-Fada’il, al-Nahj al-Sadid, 20: 180), among other names incorporating the number seven. Omar al-Daqqaq, ‘Manzilat al-‘Adad Saba‘a fi al-Fikr al-‘Arabi’, ‘Adiyat Halab 1 (1975): 53–108, suggests that the number seven may have stood for completeness and totality, and was not meant to be a literal count of the components. 46. See al-Maqrizi, Khitat, 1: 362 and 465–7, for a detailed description of the four nights of illuminations (layali al-wuqud) and the procession of the caliph and his family to al-Azhar. 47. Ibid., 2: 273. 48. Sanders, Ritual in Fatimid Cairo, 41. 49. Bloom, ‘Mosque of al-Hakim’, 17, shows that the the Mosque of al-Hakim did not acquire the epithet al-anwar until 1024–5; Caroline Williams, ‘The Cult of the ‘Alid Saints in the Fatimid Monuments of Cairo, Part 1: The Mosque of al-Aqmar’, Muqarnas 1 (1983): 37–52, suggests that the name is a reference to Husayn, the model of martyrs for the Shi‘ites, who was likened to the moon. For al-Afkhar Mosque, and the three belvederes see al-Maqrizi, Khitat, 2: 193 and 1: 404 respectively. 50. See Sanders, Ritual in Fatimid Cairo, 46–73, for a succinct description of the rituals developed in early Fatimid al-Qahira which either centred on or involved the mosque and the palace. 51. Paul Ravaisse, ‘Essai sur l’histoire et la topographie du Caire d’après al-Maqrizi’, Mémoires publiés par les Membres de la Mission Archéologique Française au Caire (Paris, 1886 and 1889), 3:1, 3, plate 3. 52. Bloom, ‘Early Fatimid Architecture’, 79–81. 53. Ibid., 84–5; al-Maqrizi, Itti‘az, 1: 294–5, for a list of al-‘Aziz’s buildings. 54. Al-Maqrizi, Khitat, 1: 407; Sanders, Ritual in Fatimid Cairo, 42. 55. The name Qasr al-Shawk (made famous by the novelist Naguib Mahfouz) predates the building of al-Qahira and was reportedly the name of a structure that belonged to the clan of ‘Azra, an Arabic tribe that participated in the Islamic conquest of Egypt. Jawhar incorporated the structure in the new palace, Ibn al-Dawadari, al-Durra al-Mudiyya, 138. – 21 3 –

N O T E S T O PAG E S 8 6 – 9 0

56. Al-Maqrizi, Khitat, 1: 425. 57. Ibid., 1: 362; Ravaisse, ‘Essai’, 1:430–32. 58. Ibn ‘Abd al-Zahir, Rawd, 278–9, copied the section on al-Azhar from an original document; alMaqrizi, Khitat, 2: 273–4 reproduces what appears to be a more complete text from the same source. 59. Ibn al-Dawadari, al-Durra al-Mudiyya, 286, reports that al-Hakim banned the khutba from alAzhar after the inauguration of his new mosque, but this decision is not corroborated by any other source. 60. Al-Maqrizi, Khitat, 2: 341, and Itti‘az, 1: 227. 61. Ibn ‘Abd al-Zahir, Rawd, 278, credits the story to Ibn Zulaq who reported it on the authority of the Fatimid historian al-Musabihi, but disputes the use of jurists ( fuqaha’), because for him Isma‘ili fiqh is not considered valid fiqh. Al-Maqrizi, Khitat, 2: 273, 341 and al-Qalqashandi, Subh, 3: 367, copy either Ibn ‘Abd al-Zahir or Ibn Zulaq. 62. Al-Maqrizi, Khitat, 2: 277, on the authority of the same al-Musabihi, probably from his still missing Akhbar Misr (only vol. 40 has been discovered and published [Cairo, 1978]). 63. On Dar al-‘Ilm or Dar al-Hikma, al-Maqrizi, Khitat, 1: 458–60; al-Qalqashandi, Subh, 3: 366. 64. Al-Maqrizi, Khitat, 2: 275–6. 65. Ibid., 2: 274, a few lines after he reported the choice of the Mosque of al-Hakim for its size, reiterated the centrality of al-Azhar when he noted that the people of Cairo were pleased when the Friday prayer was reinstituted in it because it is more accessible for them. 66. See the biography of ‘Abd al-Latif al-Baghdadi in Ibn Abi Usaybi‘a’, Uyun al-Anba’, 689. 67. Ira Lapidus, ‘Ayyubid Religious Policy and the Development of the Schools of Law in Cairo’, in André Raymond et al., Colloque international sur l’histoire du Caire (27 mars–5 avril 1969) (Cairo, 1969), 279–86; Gary Leiser, ‘The Madrasa and the Islamization of the Middle East: The Case of Egypt’, Journal of the American Research Center in Egypt 22 (1985): 29–47. 68. Ibn ‘Abd al-Zahir, Rawd, 277; al-Nuwayri, Nihayat al-Arab, vol. 30, 135–6; al-Maqrizi, Khitat, 2: 275; idem, Suluk, 1: 556–7. 69. This move was preceded by other decisions in the same vein, such as the one in 1265 to appoint four chief judges rather than the one Shafi‘ite judge, which had been the practice during the Ayyubid period; see J. S. Escovitz, ‘The Establishment of Four Chief Judgeships in the Mamluk Empire’, Journal of the American Oriental Society 102 (1982): 529–31. 70. Al-Maqrizi, Suluk, 1: 944. Ibn Iyas, Bada’i‘, 1, 1: 416–17, says that al-Nasir ordered each of the amirs to restore the mosque of which he was the supervisor (nazir). This is not corroborated by earlier reporters, and it is very difficult to imagine how amirs could have been the supervisors of major congregational mosques such as al-Azhar. 71. Jacques Jomier, ‘al-Azhar’, Encyclopaedia of Islam, 2nd edn, 1: 813. 72. Al-Maqrizi, Khitat, 2: 383–4. 73. Mubarak, Khitat, 4:48. This madrasa caused heated debate among the major legal authorities of the time for having a door into the haram; see al-Khatib al-Jawhari al-Sayrafi, Nuzhat al-Nufus wa-l-Abdan fi Tawarikh al-Zaman, ed. Hasan Habashi, 4 vols (Cairo, 1970–94), 4: 225–6. 74. On the analogous development in the Great Mosque in Isfahan, see Oleg Grabar, The Great Mosque of Isfahan (New York, 1990), 36–41, 49–59. 75. With the qualified exception of the two madrasas endowed by Salah al-Din around the Mosque of ‘Amru when he was still the vizier of the Fatimid caliph and eager to promote Sunni Islam in Egypt, see al-Maqrizi, Khitat, 2: 263–4. Another exception is the Madrasa of Sarghatmish (1356) built next to the north-western wall of the Mosque of Ibn Tulun. None of these madrasas, however, was built along the entrance of the mosque; they all merely used the existing walls as support and benefited from proximity to a centre of worship. 76. For biographical notes on the three men, see Mubarak, Khitat, 4: 44–9. 77. Al-Maqrizi, Khitat, 2: 276. 78. Ibn Iyas, Bada’i‘, 3: 124 for the work of 1476; 3: 306 for the renovation of 1494–5, which was sponsored by a certain Mustafa ibn Mahmud al-Rumi (the Anatolian). – 214 –

N O T E S T O PAG E S 9 0 – 9 4

79. Ibid., 5: 94; van Berchem, Corpus Inscriptionum Arabicarum, Egypt, 1, nos. 26–7. 80. Humphreys, ‘Expressive Intent’, 111–12. 81. Doris Behrens-Abouseif, Egypt’s Adjustment to Ottoman Rule: Institutions, Waqf and Architecture in Cairo (16th and 17th Centuries) (Leiden, 1994), 89. 82. Carl Petry, The Civilian Elite of Cairo in the Later Middle Ages (Princeton, 1981), 145–64, esp. n. 35–7, and the conclusion on al-Azhar’s position, 334. 83. Jean-Claude Garcin, ‘L’insertion sociale de Sha‘rani dans le milieu Cairote (d’après l’analyse des Tabaqat de cet auteur)’, in André Raymond et al., Colloque international, 159–68. 84. Behrens-Abouseif, Egypt’s Adjustment, 91, uses the same study to conclude that al-Azhar attained its prominent position in the first half of the sixteenth century. 85. Ibn Iyas, Bada’i‘, 5: 205; Behrens-Abouseif, Egypt’s Adjustment, 90, reads Ibn Iyas’s text as saying that the donations were given to al-Azhar. The reference, however, clearly indicates that the sultan gave out alms to the people (tasadaqa bi-mal lahu sura). 86. Ahmad Chalabi ibn ‘Abd al-Ghani, Awdah al-Isharat fi-man Tawala Misr al-Qahira min-l-Wuzara’ wal-Bashat, ed. A. R. Abdel-Rahim (Cairo, 1978), 158, 162; Behrens-Abouseif, Egypt’s Adjustment, 180. 87. Ibid., 165–72 and references. 88. Other monuments that were restored in the same period include the Mosque of ‘Amru, the Mosque of al-Fakhani, and the shrines of al-Shafi‘i and of ‘Umar ibn al-Farid, see ibid., 168; ‘Abd al-Rahman al-Jabarti, ‘Aja’ib al-Athar fi-l-Tarajim wa-l-Akhbar, 3 vols (Beirut, n.d.), 1: 250. 89. Al-Jabarti, ‘Aja’ib al-Athar, 1: 251; Mubarak, Khitat, 4: 49, 52, 53. This explanation is reminiscent of that given by some chroniclers for the Madrasas of Sultan Hasan in 1356, built with money collected from unclaimed inheritances after the Black Death of 1348. 90. Al-Jabarti, ‘Aja’ib al-Athar, 1: 490. 91. Ahmad al-Damardashi, Kitab al-Durra al-Musana, ed. ‘Abd al-Rahim A. ‘Abd al-Rahim (Cairo, 1989), 250. 92. Today, the minaret has regained its Mamluk top that reproduces the familiar silhouette typical of a minaret built in 1339, but this rebuilding cannot be dated from the available sources. It must have happened after 1932, because the photograph in Louis Hautecoeur and Gaston Wiet, Les mosquées du Caire, 2 vols (Paris, 1932) shows the minaret of al-Aqbughawiyya to have had an Ottoman third shaft and a finial. 93. Al-Jabarti, ‘Aja’ib al-Athar, 1: 490–5, gives a biography of ‘Abd al-Rahman Katkhuda and a list of his architectural works. The list is reproduced in all later Arabic sources such as Mubarak, Khitat, 4: 34–5. See also André Raymond, ‘Les constructions de l’émir ‘Abd al-Rahman Katkhuda au Caire’, Annales Islamologiques 11 (1972): 239; Doris Behrens-Abouseif, ‘The ‘Abd al-Rahman Katkhuda Style in 18th-Century Cairo’, Annales Islamologiques 26 (1992): 117–26. 94. ‘Abdulaziz M. al-Shinawwi, Suwar min Dawr al-Azhar fi Muqawamat al-Ihtilal al-Faransi li Misr (Cairo, 1971), 100–76; ‘Inan, Tarikh al-Azhar, 151–86. 95. Anwar Louca, ‘La renaissance égyptienne et les limites de l’œuvre de Bonaparte’, Cahiers d’histoire égyptienne 7, 1 (February 1955): 1–20; Afaf Lutfi al-Sayyid Marsot, ‘The Role of the Ulama in Egypt during the Early Nineteenth Century’, in Political and Social Change in Modern Egypt, ed. P. M. Holt (London, 1968), 264–80. 96. Al-Jabarti, ‘Aja’ib al-Athar, 3: 365; Vollers, ‘Azhar’, 1: 535. For a review of events between 1801 and 1809, see Afaf Lutfi al-Sayyid Marsot, Egypt in the Reign of Muhammad ‘Ali (Cambridge, 1984), 36–70; Jamal Mohammed Ahmed, The Intellectual Origins of Egyptian Nationalism (Oxford, 1960), 1–9. 97. For a brief analysis of al-Azhar’s teaching method, see Jacques Berque, Egypt: Imperialsim and Revolution, trans. J. Stewart (London, 1972), 76–83. 98. Louca, ‘Renaissance égyptienne’, 15–18. 99. Mubarak, Khitat, 4: 82–4; see also Ahmed Abdul Wahab Azmy, ‘University Tradition and Continuity in Architecture as a Stimulation for the Future Development of al-Azhar University and Old Cairo’, Ph.D. Thesis, Princeton University, 1966, 196–7, for a translation of a very – 21 5 –

N O T E S T O PAG E S 9 4 –10 2

100.

101. 102. 103.

104.

105. 106.

107. 108. 109. 110. 111. 112. 113. 114. 115.

revealing passage from one of the first reformers, Shaykh Hasan al-‘Attar, Hashiyat al-‘Attar ala Jami‘ al-Jawami‘ (Beirut, 1983), 1: 225–6. Crabbs, Jr., Writing of History, 88–9; P. J. Vatikiotis, The History of Modern Egypt from Muhammad Ali to Mubarak (Baltimore, 1991), 93–101; for al-Tahtawi, see Anouar Louca, ‘Rifa‘a al-Tahtawi (1801–73) et la science occidentale’, in D’un Orient l’autre: les metamorphoses successives des perceptions et connaissances, vol. 2, ed. Marie-Claude Burgat, Identifications (Paris, 1991), 201–18. Mubarak, Khitat, 4: 38. Creswell, Muslim Architecture of Egypt, 1: 52; Azmy, ‘University Tradition’, 212–16. For the Sayyida Zaynab and Sayyida Nafisa mosques, see the Ministry of Waqf, The Mosques of Egypt, 2 vols (Cairo, 1949), 2: 122–3, figs 180–81, for the first, and 124, figs 184–5, for the second; for the Rifa‘i Mosque, the most recent study is Muhammad al-Asad, ‘The Mosque of al-Rifa‘i in Cairo’, Muqarnas 10 (1993): 108–24. Tajdid, a term used in the nineteenth century to signify modernisation, was equally applied to designate the renovation of any old building. This semantic correlation perhaps reveals something about the notion of modernisation in the mind of nineteenth-century reformers: they were renovating an existing system, not remaking it; see Berque, Egypt, 82. A recent critical analysis of the contrast between the traditional and new methods of education is Timothy Mitchell, Colonizing Egypt (Berkeley, 1988), 74–92. They were also prompted by the loss of al-Azhar’s monopoly on the legal and educational professions with the rise of a new class of graduates from the European-inspired secular schools; see Donald Malcolm Reid, Cairo University and the Making of Modern Egypt (London, 1990), 11–14; Vatikiotis, Modern Egypt, 302. Ibid., 194–200; Ahmed, Egyptian Nationalism, 33–43. Vatikiotis, Modern Egypt, 301–4. Ibid., 472–7; Berque, Egypt, 508–10. A melancholy review of al-Azhar’s recent history is ‘Inan, Tarikh al-Azhar, 218–66. Azmy, ‘University Tradition’, 256–72. Tarek Sakr, Early Twentieth-Century Islamic Architecture in Cairo (Cairo, 1992), 60–2. Jomier, ‘al-Azhar’, Encyclopaedia of Islam, 2nd edn, 1: 815. This displacement of the original inscriptions may have occurred earlier than the nineteenth century since the mosque’s wall surfaces were repaired and whitewashed several times; see Creswell, Muslim Architecture of Egypt, 1: 37–8; Bloom, ‘ Early Fatimid Architecture’, 113–17. Ibid., 117–22; Bloom nonetheless goes to great lengths to try to read two levels of meaning in the original inscriptions, while recognising that they are fragmentary and incomplete. Some of the examples are listed in Erica Dodd and Shereen Khairallah, The Image of the Word: A Study of Quranic Verses in Islamic Architecture, 2 vols (Beirut, 1981), 2: 40, 83. Another (comprising the first three verses of Sura (23)) is in the southern dome of the Mosque of Isfahan built between 1072–75 by the vizier Nizam al-Mulk, the pioneer of the Sunni revival that ultimately swept away the Fatimid presence in Egypt and reduced the Isma‘ilis to an underground movement.

CHAPTER 8 1. For an introduction to the waqf system, see Miriam Hoexter, ‘Waqf Studies in the Twentieth Century: The State of the Art’, Journal of the Economic and Social History of the Orient 41/4 (November 1998): 474–95; Murat Cizakca, A History of Philanthropic Foundations: The Islamic World from the Seventh Century to the Present (Istanbul, 2000). 2. Many authors have developed methods for the use of waqfs as historical documents in analysing architecture. The pioneering scholar was ‘Abd al-Latif Ibrahim, ‘al-Watha’iq fi Khidmat alAthar’, 205–88. See also Rogers, ‘Waqfiyyas and Waqf-Registers’, 182–96; Leonor Fernandes, ‘The Waqfiya’, 3–12; Randi Deguilhem and André Raymond, Le Waqf dans l’espace islamique: Outil de pouvoir socio-politique (Damascus, 1995); Richard van Leeuwen, Waqfs and Urban Structures: – 21 6 –

N O T E S T O P A G E S 1 0 2 – 114

The Case of Ottoman Damascus (Leiden, 1999). One of the best studies of the architectural particularities of waqf formulae and terminology is Hazem Sayed, ‘The Rab‘ in Cairo’. 3. For a discussion of the status of artists and builders in the Mamluk society in particular, see Rabbat, ‘Architects and Artists’, 30–7.

CHAPTER 9 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Spiro Kostof, A History of Architecture (New York, 1985), 764. Oleg Grabar, ‘Iwan’, in Encyclopaedia of Islam, 2nd edn, 4: 287–9 A. A. al-Fallal, Dirasa Tahlilyia li-Shi‘r Muhtar al-Daylami (Cairo, n.d.), 46–7. Al-Nuwayri, Nihayat al-Arab, 1: 406. Ibid, 408. Al-Istakhri, al-Masalik wa-l-Mamalik, ed. M. J. al-Hini (Cairo, 1961), 147. Grabar, ‘Reflections on the Study of Islamic Art’, 10–11. Ibn Shahin al-Zahiri, Zubdat, 31. ‘Abd al-Wahab, Tarikh al-Masajid al-Athariyya, 166. I am indebted to my colleague Jean Inamorati for permission to use her unpublished research paper, ‘The Black Death and Institutional Building in Cairo, The Funerary Complexes of Sultans Qalaoun and Hasan’ (MIT, 1983). Al-Maqirzi, Khitat, 2: 374–5. Ibid., 2: 378. M. M. Amin, Watha’iq Waqf al-Sultan al-Malik al-Nasir Hasan (Cairo, 1986), 47. Humphreys, ‘Expressive Intent’, 69–119. Al-Maqrizi, Khitat, 2: 183.

CHAPTER 10 1. See the following studies: ‘Abd al-Latif Ibrahim, ‘al-Watha’iq Fi-Khidmat al-Athar’, 205–88; Alexandre Lézine, ‘Les Salles Nobles Des Palais Mamelouks’, Annales Islamologiques 10 (1972): 149–205; Rogers, ‘Waqfiyyas and Waqf-Registers’, 182–96; Laila A. Ibrahim, ‘Middle-Class Living Units in Mamluk Cairo: Architecture and Terminology’, Art and Archaeology Research Papers 14 (1978); idem, ‘Residential Architecture in Mamluk Cairo’, Muqarnas 2 (1984): 47–59; Salih Lam‘i Mustafa, al-Watha’iq wa-l-‘Imara, Dirasa fi-l-‘Imara al-Islamiyya fi-l-‘Asr al-Mamluki al-Jarkasi (Beirut, 1980); Jacques Revault et al., Palais et Maisons Du Caire, I- Époque Mamelouke (Paris, 1982); Zakarya, Deux Palais Du Caire; Doris Behrens-Abouseif, ‘The Qubba, an Aristocratic Type of Zawiya’, Annales Islamologiques 19 (1983): 1–7; Hazem Sayed, ‘The Development of the Cairene Qa‘a’, 31–53; Hussam al-Din ‘Abd-al-Fattah, ‘Arba‘ Buyut Mamlukiyya min al-Watha’iq al‘Uthmaniyya’, Annales Islamologiques 24 (1988): 49–102; Donald P. Little, ‘The Nature of Khanqahs, Ribats, and Zawiyas under the Mamluks’, in Islamic Studies Presented to Charles J. Adams, ed. Wael B. Hallaq and Donald P. Little (Leiden, 1991), 91–105. 2. Al-Nuwayri, Nihayat al-Arab, vol. 30, fol. 63; Ibn al-Dawadari, Kanz al-Durar, vol. 9, al-Durr alFakhir fi-Sirat al-Malik al-Nasir, ed. H. R. Roemer (Cairo, 1960), 266; K. V. Zetterstéen, Beitrage zur Geschichte der Mamlukesultane in den Jahren 690–741 (Leiden, 1919), 161; al-Maqrizi, Khitat, 2: 209; idem, Suluk, 2: 129; Ibn Taghri-Birdi, Nujum, 9: 36–7. 3. Al-‘Umari, Masalik, 142–4; al-Maqrizi, Khitat, 2: 209–10, 229–30; al-Qalqashandi Subh, 3: 369–72; part of the translation is taken from Creswell, Muslim Architecture of Egypt, 2: 260. Two later chroniclers, Ibn Iyas and Ibn Shahin al-Zahiri, complicate matters concerning the morphology of the Ablaq Palace. Ibn Shahin al-Zahiri, Zubdat, 27, says in passing that the Ablaq Palace comprised three palaces. Ibn Iyas, Bada’i‘, 1, 1: 445, asserts that it is composed of three – 217 –

N O T E S T O P A G E S 114 – 11 8

4.

5.

6. 7.

8. 9.

10. 11. 12.

13. 14. 15. 16.

17. 18. 19.

interconnected palaces, with five qa‘as and three marqad, which could mean a sleeping space as an alcove or a suffa. Both descriptions should be rejected: they are both anachronistic, and they appear to have included later additions to the original Ablaq Palace in their depictions. Paul Casanova, Histoire et description de la citadelle du Caire (Cairo, 1894), published in Arabic as Tarikh wa-Wasf Qal‘at al-Qahira, trans. Ahmad Darraj (Cairo, 1974), 134, reports the names of two qa‘as, but he does not ascribe them to the Inner Palaces, the Qa‘a of Silver and the Qa‘a of Copper. The name Qa‘a of Silver (Qa‘at al-Fidda) is corroborated from Mamluk sources, one even implies that it was in the Ablaq Palace, Ibn al-Furat, Tarikh, 9, 1: 105, 129. Laila A. Ibrahim, ‘Residential Architecture in Mamluk Cairo’, 55 and n. 40. A fifth Bahri palace, that of Manjak al-Yusufi, was standing until the beginning of the twentieth century as evidenced by a photograph published in Gaston Migeon, Les arts musulmans (Paris-Brussels, 1926), plate xiii. The palace vestibule with a beautiful flat dome above it with the titles of Manjak inscribed around it still stands today. Jacques Revault, ‘L’architecture domestique du Caire à l’époque mamelouke’, 49–74, and JeanClaude Garcin, ‘Habitat médiéval et histoire urbaine à Fustat et au Caire’, 180–7, in Revault, Palais et maisons du Caire. All plans are labelled istabl with a question mark. Garcin, ‘Habitat médiéval’, 185, advances the reading of the lower hall in the Palace of Bashtak as either stables or a hall for guards. He also notes (182) that the notion that istabl sometimes designates both the stables in an amir’s palace and the total structure, as is the case of the Palace of Qawsun, is incorrect. It is specific to that palace because the stables existed before the palace and gave their name to the complex. Al-Maqrizi, Khitat, 2: 70 (the qasr of Bashtak), 72 (Istabl Qawsun), 73 (the dar of Taz). Garcin, ‘Habitat médiéval’, 183, basing himself on al-Maqrizi, Khitat, 2: 72; the term qasr is used to describe the amir’s private hall in Istabl Qawsun, which was apparently the highest part of the complex. Zakarya, Deux palais du Caire, 143, gives a similar definition of qasr based on two waqfs of the fifteenth century. Garcin, ‘Habitat médiéval’, 205, records, however, the use of the expression ‘the place known in the past as qasr’ in the waqf of Barsbay. Awqaf 880, waqf of Sultan Barsbay, dated 24 Rajab, 841/1438, 106–8. Buyutat is a generic term for household dependencies, which include the tishtakhana (the washbasin room), the hawa’ijkhana (the pantry) and the firashkhana (the tent room), al-Qalqashandi, Subh, 4: 10–11; Marafiq has the same general meaning. Hazem Sayed, ‘The Rab‘ in Cairo’, does not define the word ma‘azil in his study. Zakarya, Deux palais du Caire, 128, says that ma‘azil is an unidentified part of the common area of the residence. Manzara is an upper floor qa‘a with a view to the outside. Sayed, ‘The Rab‘ in Cairo’, proposes a conjectural definition of takhayen as sleeping rooms. Zakarya, Deux palais du Caire, does not identify it. Awqaf 292/44, waqf of Amir Khayer Bak, dated 8 Muharram, 927/1521, 8–10; text published in ‘Abd al-Fattah, ‘Arba‘ Buyut Mamlukiyya’, 82–3. Awqaf 952, waqf of Amir Ibrahim Aga Mustahfizan, dated 10 Safar, 1062/1652, 117–18. The reference is taken from ‘Abd al-Fattah, ‘Arba‘ Buyut Mamlukiyya’, 86–7. Laila A. Ibrahim, ‘Mamluk Monuments of Cairo’, Quaderni dell’ Instituto Italiano di Cultura per la R.A.E. (Cairo, 1976), 9–29. The number of qusur mentioned by al-Maqrizi is five not four; they all belong to the period of al-Nasir Muhammad. They are the palaces of Baktimur, Bashtak, Yalbugha, Altunbugha and Tatar al-Hijaziyya, a daughter of al-Nasir Muhammad (al-Maqrizi, Khitat, 2: 68–72). Al-Maqrizi, Khitat, 2: 51–79. Ibid., 2: 74. The abbreviated text of the waqf has been published and annotated; Ahmad Darrag, L’acte de waqf de Barsbay (Cairo, 1963), but the one used here is the complete text of Awqaf 880. The waqf of Barsbay was chosen because of the frequent appearance of the word qasr in it, which gives a sampling of the range of meaning applied to this term in a legal Mamluk document. Five – 21 8 –

N O T E S T O P A G E S 11 8 – 1 21

20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32.

33.

34. 35. 36. 37.

halls in five different residential structures are called qasr in this waqf, in addition to the qasr of Alin Aq. Riwaq, like qasr, acquired a specific meaning in the Mamluk period. For a definition of the word, see Abd al-Latif Ibrahim ‘Ali, ‘Wathiqat Qaraquja al-Hasani’, n. 41; Zakarya, Deux palais du Caire, 146; Laila Ibrahim, ‘Middle-Class Living Units’, 24; Ibrahim and Amin, Mustalahat, 57–8. Waqf of Sultan Barsbay, 62: 9–10. Tibaq here appears to mean a small separate room, and khaza’in is the equivalent of mabitat, or sleeping spaces, see Zakarya, Deux palais du Caire, 125. Waqf of Sultan Barsbay, 77, l. 8–9. Waqf of Sultan Barsbay, 113, l. 7–8. Takhayen is not identified by either Sayed or Zakarya. Waqf of Sultan Barsbay, 122, l. 7–8. A raised qa‘a (qa‘a mu‘allaqa) is a qa‘a in the second floor, raised above a first floor. A raised qa‘a and a riwaq denote basically the same space in term of location: both are first floor halls built above some ground floor space. Garcin did not identify Alin Aq; al-Maqrizi, Khitat, 2: 112–13, lists him among the khassakiyya of Khalil. The name is written A‘naq al-Husami, which is obviously a corruption of the name Alin Aq. For his execution, see al-Nuwayri, Nihayat al-Arab, vol. 29, fol. 77; al-Maqrizi, Suluk, 1: 795. Ibn Shaddad, Tarikh al-Malik al-Zahir, 290, tells us that Baysari was Baybars’ companion from the beginning of their careers as Mamluks. For the date of his death, see al-Maqrizi, Khitat, 2: 69–70. Idem, Suluk, 2: 88–9. Idem, Khitat, 2: 392–3. Laila A. Ibrahim, ‘Mamluk Monuments of Cairo’, 18–19; ‘Abd al-Rahim Ghaleb, Mawsu‘at al-‘Imara al-Islamiyya (Beirut, 1988), 315–16. Namely istabl Qawsun, the dar of Taz, and the dar of Sarghatmish, see al-Maqrizi, Khitat, 2: 72, 73, 74. A fourth example, the dar of Baysari, is called a qasr only in the context of its popular name, 69. Al-Maqrizi’s usage of the word qasr in these last examples is always coupled with the usage of istabl, as if he implies that a dar of an amir has two separate units: a qasr and an istabl. This denotation of qasr is related to the second use noted in the description of Alin Aq’s palace in the waqfs of Barsbay and Khayer Bak. Al-Shuja‘i, Tarikh, 25, calls the residence of Yalbugha bayt and istabl; on page 68, he calls the residence of Bashtak, dar; Ibn Taghri-Birdi, Nujum, 9: 188, calls the dar of Tashtimur qasr. See, for example, the waqf of Sultan Qaytbay, Awqaf 886, dated 15 Dhi-l-Hujja, 895/1490, where the upper-level halls are called riwaq in two residences attributed to him; one is in fact said to have been built by him and called the Grand Dar (al-Dar al-Kubra), 47–8 for the first house and 260–61 for the second. Other examples, all belonging to Burji princely palaces, are collected in Abd al-Latif Ibrahim ‘Ali, ‘Wathiqat Qaraquja al-Hasani’, 231–2, n. 41. This is difficult to prove for the Palace of Amir Janim al-Sharifi described in a waqf of Sultan Qaytbay dated 1481. In it, a space in the upper level is called qasr. It has a durqa‘a of the type called Iraqi, two mabitat, one large and one small. Zakarya, Deux palais du Caire, 49–81. She notes, however, that the palace appears to have been renovated and enlarged many times, 80–81, and that it was ascribed to a certain Baybars al-Tawil, who resided in it in the past (qadiman). She does not identify Baybars, and I could not identify him from the sources available to me. The initial palace could have been built in the Bahri period since the area in which it is located, Suwayqat al-‘Izzi, was first zoned by ‘Izz al-Din Aybak al-‘Izzi who was one of the amirs of al-Ashraf Khalil, al-Maqrizi, Khitat, 2: 107. Al-Munufi, Akhbar al-’Uwal, 140. They were first noted on the plan of the citadel drawn by the British Lt. Colonel Green in 1896 where they were marked ‘ruined vaults’. The plan was published in the Comité de Conservation des Monuments de l’Art Arabe, Exercice 1896, Fasc. 15, 145. Al-‘Umari, Masalik, 142; al-Maqrizi, Khitat, 2: 210; al-Qalqashandi, Subh, 3: 371. Ibn al-Furat, Tarikh, 9, 1: 169, reports an event that shows that the buttery overlooked the gate of the maydan (which means that it was located near or in the lower halls of the Ablaq Palace) and that the Qa‘a al-Ashrafiyya and the Ablaq Palace both opened into the stables underneath. – 21 9 –

N O T E S T O P A G E S 1 21 – 1 2 7

38. 39.

40. 41. 42.

43. 44.

Al-Maqrizi, Suluk, 3: 682, produces a short version of the same report. Ibn Shahin al-Zahiri, Zubdat, 124, says that the saddlery is attached to the royal stables. The concept of monumentality as embodying a memorial function and an aesthetic of immensity is discussed in Françoise Choay, ‘Alberti, the Invention of Monumentality and Memory’, Harvard Architectural Review 4 (Spring, 1984): 99–105. Laila A. Ibrahim, ‘Residential Architecture in Mamluk Cairo’, 55; Lézine, ‘Salles nobles’, 128–30, gives the plan dimensions of the extant Bahri qa‘at. Many of his remarks have been disproved but his measurements show that there is a decrease in the average size of a princely qa‘a from the Bahri to the Burji period. Humphreys, ‘Expressive Intent’, 98. Al-Shuja‘i, Tarikh, 68; al-Maqrizi, Khitat, 2: 70, and Suluk, 2: 501, note that the view from the Palace of Bashtak encompasses the whole city of Cairo. They use almost the same language to describe the view from the Ablaq Palace. Garcin, ‘Habitat médiéval’, 187, notes the fundamental difference between the three palaces and that of Taz in terms of their organisation in relation to the open court. This courtyard appears to have been the central space onto which the other components of the palace opened, just like the later Burji palaces. The Palace of Taz, which is later than the three other palaces, might have been a transitional model between the period of al-Nasir Muhammad and the Burji period. Al-Maqrizi, Suluk, 2: 687. Garcin, ‘Habitat médiéval’, 211–16.

CHAPTER 11 1. The development of the citadel under these sultans and its impact on its urban environment are studied in detail in my Ph.D. Thesis, ‘The Citadel of Cairo, 1176–1341: Reconstructing Architecture from Texts’, MIT, 1991. For the relationship between the Mamluk ceremonials and the architecture of the palatial complex, see Doris Behrens-Abouseif, ‘The Citadel of Cairo: Stage for Mamluk Ceremonial’, Annales Islamologiques 24 (1988): 25–79. 2. Other plans include that of Richard Pococke, A Description of the East and some Other Countries, vol. I, Observations on Egypt, (London, 1743–5), 33, plate 14; and that of L. Cassas, drawn in 1799 and published in Voyage pittoresque de la Syrie et Basse Égypte, reproduced in Behrens-Abouseif, ‘Citadel of Cairo’, 62. 3. Evliya Çelebi, Seyahatnamesi, 9–10: 389; Mustafa Ibrahim (writing in the mid eighteenth century), Waqa’i‘ Misr al-Qahira Bayn 1100 wa 1150, MS H.O.38, Vienna National Library (copy at Creswell Library, AUC, Cairo), 19. 4. It is not clear whether the name Yusuf refers to the patriarch Joseph, whose hagiography is connected to Egypt in many ways, or to Salah al-Din al-Ayyubi, whose given name was Yusuf. Both attributions are of course erroneous, and both reflect the popularisation of the citadel’s toponyms in the late Ottoman period with the complete deterioration of its architecture and its role as the centre of power in the country. 5. For the different types of qubbas as mausolea, see Ernest Diez, ‘Kubba’, Encyclopaedia of Islam, 2nd edn, 5: 289–96. 6. References are collected in Oleg Grabar, ‘al-Mushatta, Baghdad and Wasit’, in J. Kritzeck and R. B. Winder, The World of Islam: Studies in Honour of Philip K. Hitti (London, 1959), 105–6; Jacob Lassner, The Topography of Baghdad in the Early Middle Ages: Texts and Studies (Detroit, 1970) n. 22, 239–40. There might have been another early Abbasid palace named also al-Khadra in alMansur’s early capital of al-Hashimiyya. Muhammad ibn Jarir al-Tabari, Tarikh al-Rusul wa alMuluk, ed. M. Abu al-Fadhl Ibrahim (Cairo, 1963), 8: 83, says that the Rawandiyya (a rebellious religious group) ascended the caliph’s Khadra during their revolt in 758 and jumped off it as if to fly. Lassner, Topography, 136, doubts the existence of such a palace in this unofficial Abbasid –220 –

N O T E S T O PAG E S 12 7 –12 9

7. 8. 9.

10. 11. 12. 13. 14. 15. 16. 17. 18.

19. 20.

21. 22. 23.

24.

capital and raises the possibility that Tabari may have confused the palaces of Baghdad and Hashimiyya in describing the event. This is very plausible especially since no other reference exists to this early Abbasid Khadra. A recent book on la Cuba is Giuseppe Caronia and Vittorio Noto, La Cuba di Palermo (Arabi e Normanni nel XII Secolo) (Palermo, 1988); Lézine, ‘Salles nobles’, 70, and references listed in n. 3. Al-Maqrizi, Khitat, 2: 202; J. M. Rogers, ‘Kubbat al-Hawa’, Encyclopaedia of Islam, 2nd edn, 5: 297–8. In every Arabic lexicon until the present day, the word iwan is mentioned in connection with Iwan Kisra, see A. A. al-Bustani, al-Bustan (Beirut, 1927), 1: 85; E. W. Lane, Arabic-English Lexicon (Edinburgh, 1863), book 1, pt 1: 129. These persistent references suggest that this particular monument might have represented the archetype for iwans in the medieval collective memory; see Nasser Rabbat, ‘The Iwans of the Madrasa of Sultan Hasan’, The American Research Center in Egypt Newsletter 143/144 (Autumn/Winter 1988/89): 5–9. [Reproduced in this volume]. Al-Maqrizi, Khitat, 1: 388. For the different denotations of the word iwan, see Oleg Grabar, ‘Iwan’, Encyclopaedia of Islam, 2nd edn, 4: 287–9; for the dome, see E. Baldwin Smith, The Dome: A Study in the History of Ideas (Princeton, NJ, 1951, reprint 1978). Ibn al-Dawadari, al-Durra al-Zakiyya fi-Akhbar al-Dawla al-Turkiyya, ed. U. Haarmann (Cairo, 1971), 63; al-Maqrizi, Suluk, 1: 438; Ibn Taghri-Birdi, Nujum, 7: 102. Al-‘Ayni, ‘Iqd al-Juman, 1: 348, on the authority of Baybars al-Mansuri from his still unpublished Zubdat al-Fikra fi-Tarikh al-Hijra. Ibn Shaddad, Tarikh al-Malik al-Zahir, 340; al-Kutubi, Fawat al-Wafi yyat, 1: 242; Ibn Kathir, Bidaya, 13: 275; Ibn Taghri-Birdi, Nujum, 7:190. Ibn al-Dawadari, al-Durra al-Zakiyya, 73, for the reference to the iwan and qubba, 63, 94, 303 where he mentions only the iwan. Al-Maqrizi, Khitat, 2: 112. Ibn al-Furat, Tarikh, 8: 38. Perhaps except for al-Maqrizi, Khitat, 2: 212, who states that Qalawun demolished the qulla (keep) of Baybars and built his qubba in its place in 1285, which is the same date given by Ibn al-Furat for the demolition of the Qubba of Baybars. The concurrence of dates in these reports permits the correction of the spelling of al-qulla in al-Maqrizi’s sentence (most probably a typographical error in this generally inaccurate printed text), which should read al-qubba instead when reporting Qalawun’s replacement of Baybars’ structure. This should be differentiated from the qulla, also built by Baybars inside the wall separating the two enclosures of the citadel, which was not rebuilt until the time of al-Nasir Muhammad. Ibn-‘Abd al-Zahir, Sirat al-Malik al-Mansur, 139. Ibn al-Dawadari is the one who offers the most direct reference, for he reports in the events of 1293 that ‘al-Iwan al-Ashrafi was completed by Amir ‘Alam al-Din al-Shuja‘i’, Ibn al-Dawadari, al-Durra al-Zakiyya, 345; al-Maqrizi, Khitat, 2: 206; Ibn Iyas, Bada’i‘, 1, 1: 378, states specifically that Khalil is the builder of al-Iwan al-Ashrafi and al-Qa‘a al-Ashrafiyya, thus preventing any confusion between the two structures. Al-Nuwayri, Nihayat al-Arab, 30, fol. 66; Zetterstéen, Geschichte der Mamlukesultane, 156; Ibn al-Dawadari, al-Durr al-Fakhir, 238; al-Maqrizi, Suluk, 2: 107; Ibn Taghri-Birdi, Nujum, 9: 51. Baybars al-Mansuri, Tuhfa, 232–3. Many chroniclers have noted the zeal with which al-Nasir was building his autocratic rule after the two first reigns in which he had only nominal authority. See P. M. Holt, The Age of the Crusades: The Near East From the Eleventh Century to 1517 (London, 1986), 107–20; Irwin, Mamluk Sultanate, 85–124. Baybars al-Mansuri, Tuhfa, 231, 233–4, gives an elaborate description of the ceremony, listing the names of the amirs who were required to sit around the sultan; al-Nuwayri, Nihayat, vol. 30, fol. 66. The anonymous author edited in Zetterstéen, Geschichte der Mamlukesultane, 158, reports another activity, the review of the troops, which was instituted in the iwan. – 2 21 –

N O T E S T O PAG E S 12 9 –13 2

25. Ibn Taghri-Birdi, Nujum, 9: 180, noted the complex rituals introduced by al-Nasir in the iwan, and detected the important function of these rituals as inspiring awe among envoys of foreign rulers, who were more common in al-Nasir’s court than any Mamluk sultan before him. 26. Ibn al-Dawadari, al-Durr al-Fakhir, 372, only says that the Iwan of al-Ashraf was demolished on the third of Sha‘ban 733 (1333), along with other structures; the anonymous author of Zetterstéen, Geschichte der Mamlukesultane, 186, gives the same date of completion but a different one for the beginning of construction, and specifies that the dome was the only part of the old iwan that al-Nasir destroyed; al-Shuja‘i, Tarikh, 113, only says that al-Nasir demolished the Iwan of al-Ashraf twice, presumably in the same dates given by Ibn al-Dawadari, but unfortunately the part of al-Shuja‘i’s chronicle covering the years between 1310 and 1338 is missing. The same vague report is repeated by al-Maqrizi, Suluk, 2: 538, and Ibn Taghri-Birdi, Nujum, 9: 180. 27. Al-Maqrizi, Khitat, 2: 207; Ibn Iyas, Bada’i‘, 1, 2: 388. 28. Al-Jabarti, ‘Aja’ib al-Athar, 3: 372–3. 29. Pococke, Observations on Egypt, 33. 30. Edme François Jomard, ‘Déscription de la ville et de la citadelle du Kaire’, Déscription de l’Égypte, état moderne (Paris, 1821–9), 18: 354–5. 31. Al-‘Umari, Masalik, 141–2, offers the most complete description of the iwan’s layout. His text was copied in al-Maqrizi, Khitat, 2: 210. 32. Evliya Çelebi, Seyahatnamesi, 9–10: 389, says that the diwan of Sultan al-Ghuri had thirty-five columns. Casanova, Tarikh, 124, reports that Maillet counted thirty-four columns, probably including two of the square pillars as columns. Pococke’s plan has forty-four columns, but it is doubtlessly wrong as he extends the middle two rows all the way to the end wall, thus adding six columns, even though he himself says in his text that the middle rows were designed to support a dome. 33. Al-Maqrizi, Khitat, 2: 206; Ibn Shahin al-Zahiri, Zubdat, 26; Ibn Iyas, Bada’i‘, 5: 441. 34. Casanova, Tarikh, 127, was the first to correctly read the remainder of the inscription, which confirms that the drawing represented al-Nasir’s iwan. A similar inscription band runs around the drum of the Nasiri Mosque’s dome. Both inscriptions were possibly done at the same time. 35. Behrens-Abouseif, ‘Citadel of Cairo’, 77–8. 36. Hazem Sayed proved that the majlis was a specific type and traced its development in his, ‘The Development of the Cairene Qa‘a’, 32–9. 37. Lézine, ‘Salles nobles’, 65 and 71, the plan of this qa‘a is known from its reproduction in the Déscription de l’Égypte, état moderne, planches, vol. 1, plate 23; redrawn by Creswell, Muslim Architecture of Egypt, 2: 86. 38. Al-Maqrizi, Khitat, 2: 183–4, quoting from Ibn Sa‘id al-Maghribi’s al-Mugharrib. 39. The Florentine traveller Brancacci, who reported on the audience he attended there during the reign of Barsbay, says that the sultan was seated on a raised platform inside the iwan and was perfectly visible from all sides; see Behrens-Abouseif, ‘Citadel of Cairo’, 42–3. 40. Al-Maqrizi, Khitat, 2: 206. 41. Ibn Shahin al-Zahiri, Zubdat, 26, says exactly that about the iwan. He was writing in the mid fifteenth century, however, when the iwan was only used for important ceremonies. 42. Though basilicas are mostly connected with early Christian churches, the original Greco-Roman functions and symbolism of this widespread type were never lost or forgotten; see Irving Lavin, ‘The House of the Lord’, Art Bulletin 44 (1962): 16–17; William L. MacDonald, The Architecture of the Roman Empire: An Introductory Study (New Haven, 1982), 53 and n. 21. It is in its royal and congregational connotations that the word basilica is used in this chapter. 43. Behrens-Abouseif, ‘Citadel of Cairo’, 76, notes this precedent. 44. In fact, the entire complex of Qalawun may have alluded to palatial symbolism in the original functions of its site, its architectural references and the lavishness of its decoration. The conversion of an exclusive, old Fatimid palace to a public socioreligious complex may have been an intentional gesture aimed at advertising the generosity and charity of the ruler. See the description of the buildings, the waqf and the social functions of the bimaristan and the mausoleum –222–

N O T E S T O PAG E S 13 2 –13 6

45.

46.

47.

48. 49. 50. 51.

52. 53.

54.

55. 56. 57. 58. 59.

60.

in al-Nuwayri, Nihayat, vol. 29, fol. 28 ff., reproduced in al-Maqrizi, Suluk, 1: 997–1001, and translated in Creswell, Muslim Architecture of Egypt, 2: 191. A variety of domed basilicas, modified in different ways to accommodate the dome, span the entire historical and functional ranges of late Roman and Byzantine architecture: see Cyril Mango, Byzantine Architecture (New York, 1985), 52–88, where the domed basilicas of the age of Justinian are discussed, including the most famous of them all, the Hagia Sophia of Constantinople. Jean Sauvaget was the first to persuasively argue the persistence of royal connotations in the adaptation of the basilica to early Islamic architecture; see La Mosquée oméyyade de Médina: Étude sur les origines architecturales de la mosquée et de la basilique (Paris, 1947), 158–84; also Elias J. Bickerman’s review of the book in Classical Philology 44/1 (January 1949): 142. Many of Sauvaget’s ideas on the adaptations and transformations of the basilical plan are still unchallenged and warrant further research despite recent scholarship that modified, and sometimes disproved, his conclusions concerning the Prophet’s Mosque at Medina and the chronological development of the mosque in general. On the continuity of different classical palatial traditions in the Islamic part of the Mediterranean world, see Oleg Grabar, The Alhambra (Cambridge, MA, 1978), 103–8, 207–8; idem, The Formation of Islamic Art (New Haven, 1973), 148–50; idem, ‘From Dome of Heaven to Pleasure Dome’, Journal of the Society of Architectural Historians 49 (March 1990): 15–21. Al-Maqrizi, Khitat, 1: 386, 388. Jean Sauvaget, ‘Les Ghassanides et Sergiopolis’, Byzantion 14 (1939): 115–30 (reprinted in Mémorial Jean Sauvaget (Damascus, 1954), 147–64). Richard Ettinghausen, From Byzantium to Sasanian Iran and the Islamic World (Leiden, 1972), 50. The basilical hall with a domed room behind it at the Dar al-Imara at Kufa is dated to the early Umayyad period and believed to have been the work of Ziyad ibn Abihi during the caliphate of Mu‘awiya, see K. A. C. Creswell, A Short Account of Early Muslim Architecture, ed. James W. Allan (Cairo, 1989), 10–15, and fig. 2. For the basilical hall in the palace at ‘Anjar, dated to the end of the reign of al-Walid I (705–715), see Maurice Chéhab, ‘The Umayyad Palace at ‘Anjar’, Ars Orientalis 5 (1963): 17–25, and fig. 2. Lavin, ‘House of the Lord’, 3–5, 9–11. Henri Stern, ‘Notes sur l’architecture des châteaux omeyyades’, Ars Islamica 11–12 (1946): 89–92. Ettinghausen was further to elaborate the interpretation and to make the influence tripartite – local, classical and Sasanian – in his discussion of Khirbat al-Mafjar in Ettinghausen, From Byzantium, 62–4. The same remarks apply to the organisation of the hall in Dar al-Imara at Kufa. There exist, however, some intriguing palatial examples, scattered in Syria and datable to the medieval period, which may prove the survival of Umayyad models; see the plan of Qasr alBanat in Raqqa, c. 1168, reproduced in Yasser Tabbaa, ‘Circles of Power: Palace, Citadel and City in Ayyubid Aleppo’, Ars Orientalis 23 (1993): 181–200. It is a basic majlis plan whose riwaq was extended to form a basilical layout. Al-Maqrizi, Khitat, 2: 406, 408. Textual references to the Ablaq Palace in Damascus are gathered in ‘Abd al-Qader al-Rihawi, ‘Qusur al-Hukkam fi-Dimashq’, Les annales archéologiques arabes syriennes, vols 22–3 (1972–3), 22: 42–8. Al-Maqrizi, Suluk, 2: 129; al-Qalqashandi, Subh, 4: 94; Ibn Taghri-Birdi, Nujum, 9: 36–7. Meinecke, ‘Mausoleum des Qala’un’, 64. Creswell, Muslim Architecture of Egypt, 2: 203, accepts the opinion of Saladin and Herzfeld before him concerning the affinity between the qubba of Qalawun and the Dome of the Rock, although he noted the differences in the arrangement of supports under the dome in the later structure. He also established a possible context for the transfer of the Dome’s plan to Cairo by remarking that Qalawun visited Jerusalem many times during his reign and had ordered the building of a ribat there in 1282 that still stands today. Mufaddal ibn Abi al-Fada’il, al-Nahj al-Sadid, 20: 236–7. –223 –

N O T E S T O P A G E S 1 3 6 – 14 2

61. For a discussion of the possibilities of colours and meanings attached to the qubbat al-khadra see Charles Wendell, ‘Baghdad: Imago Mundi and Other Foundation Lore’, International Journal for Middle Eastern Studies 2 (1971): 117–20. Wendell seems to have overlooked the fact that his sources, all post-eleventh century, refer to green-coloured domes and differently coloured domes. One contemporary example is the dome called the qubbat al-zarqa’ (Blue Dome) because it was covered with blue tiles, which was finished in 1292 at the Citadel of Damascus, see Ibn Habib, Tazkirat al-Nabih fi-Ayyam al-Mansur wa-Banih, 3 vols, ed. M. M. Amin (Cairo, 1976), 1: 140; also, Ibn Kathir, Bidaya, 13: 323, 327. 62. Although Ibn-‘Asakir, Tarikh Madinat Dimashq, ed. S. al-Munajjid (Damascus, 1951–4), 2: 134, speaks only of al-Khadra of Damascus, Ibn Kathir, Bidaya, 9: 143, explains that the palace was called the Qubbat al-Khadra after the green dome built in it. For the other palaces, the references always mention green domes. 63. References to the Qubbat al-Khadra of Mu‘awiya are in al-Rihawi, ‘Qusur’, 34–6. 64. Lassner, Topography, 52–3, 134–5; Creswell, Short Account, 239. 65. Wendell, ‘Baghdad’, 118–19; Grabar, ‘ Dome of Heaven’, 15–17. 66. This is a verbatim translation of the attributes of this dome as reported by al-Khatib alBaghdadi, Tarikh Baghdad, 14 vols (Beirut, n.d.), 1: 73, repeated by al-Qazwini, Kitab Athar al-Bilad, 310. See Lassner, Topography, 53, for his translation of this passage. 67. The memory of the Baghdad dome with its horseman on top may have been preserved in the design of an arbiter for a drinking session illustrated in Badi‘ al-Zaman al-Jazari, Kitab fi Ma‘rifat al-Hiyal al-Handasiyya, trans. Donald Hill (The Book of Knowledge of Ingenious Mechanical Devices) (Dordrecht, Holland, 1974), 99, fig. 82, 219, plate 13. Al-Jazari composed his work and illustrated it for Nasir al-Din Mahmud, the Artuqid ruler of Amid (Diyarbakir) (1200–1222), around 1204. Hill’s figure 82 belongs to a manuscript dated to 1486. His plate 13 belongs to a manuscript dated to 1354. 68. Al-Maqrizi, Khitat, 1: 388. 69. With the completion of Qalawun’s funerary qubba in Bayn al-Qasrayn, also known as the Qubba al-Mansuriyya, in 1284 the use of the same name for his throne hall at the citadel would have caused some confusion. 70. Two additional designations of the word qubba in Mamluk Cairo, though in the Burji period, which would strengthen the argument that the term has definitely migrated to the funerary domain, were noted by Behrens-Abouseif, ‘Aristocratic Type of Zawiya’, 1–7.

CHAPTER 12 1. The best source on the urban history of Cairo remains Janet Abu-Lughod, Cairo, 1001 Years of the City Victorious (Princeton, 1971). 2. An illuminating discussion of the processes by which the Enlightenment age constructed an exclusively European cultural evolution is Martin Bernal, Black Athena: The Afroasiatic Roots of Classical Civilization. Volume 1: The Fabrication of Ancient Greece 1785–1985 (New Brunswick, NJ, 1987). 3. Claude Cahen, ‘Khitat’, Encyclopaedia of Islam, 2nd edn, 5: 22. 4. Mubarak, al-Khitat al-Tawfiqiyya al-Jadida. 5. See the discussion on the khitat genre in modern Egyptian historiography in Crabbs, Jr., Writing of History, 115–19. 6. See the introduction of Charles Coulston Gillispie and Michael Dewachter (eds), Monuments of Egypt: The Napoleonic Edition: The complete archaeological plates from la Description de l’Égypte (Princeton, 1987), 1–29. 7. Pascal-Xavier Coste, Architecture Arabe ou Monuments du Kaire mésurés et dessinés de 1818 à 1826 (Paris, 1839). For a recent appraisal of Coste’s work see the articles collected in Daniel Armogathe and Sylviane Leprun, Pascal Coste ou l’architecture cosmopolite (Paris, 1990). – 2 24 –

N O T E S T O P A G E S 14 3 – 14 9

8. Briggs, Muhammadan Architecture. 9. Creswell, Early Muslim Architecture and Muslim Architecture of Egypt. A recent volume of Muqarnas, vol. 8 (1991) was devoted to Creswell and his legacy. 10. See the analysis of J. M. Rogers, ‘Architectural History as Literature: Creswell’s Reading and Methods’, Muqarnas 8 (1991): 45–54. 11. ‘Abdel-Wahab, Tarikh al-Masajid al-Athariyya; Su‘ad Maher, Masajid Misr wa Awliya’uha al-Salihun, 3 vols (Cairo, 1971). An earlier French survey of Cairo’s mosques may have supplied another frame of reference to these books: Louis Hautecœur and Gaston Wiet, Les mosquées du Caire (1932). 12. Fikri, Masajid al-Qahira wa Madarisaha. 13. Farid Shafe‘i, al-‘Imara al-‘Arabiyya fi Misr al-Islamiyya, ‘Asr al-Wulat (Cairo, 1970), and al-‘Imara al‘Arabiyya al-Islamiyya, Madiha wa-Hadiraha wa-Mustaqbalaha (Riyadh, 1982); ‘Abd al-Rahman Zaki, al-Qahira, Tarikhaha wa-Atharaha (Cairo, 1966), and Mawsu‘at Madinat al-Qahira fi Alf ‘Am (Cairo, 1969); Ahmad ‘Abd al-Razzaq Ahmad, Tarikh wa Athar Misr al-Islamiyya (Cairo, 1977, reprint 1993). 14. For a list of some of these mistakes see the book review by Caroline Williams, Journal of the American Research Center in Egypt 29 (1992): 226–9.

CHAPER 13 1. For the definition of mazalim, see Jørgen Nielsen, art. ‘Mazalim’, Encyclopaedia of Islam, 2nd edn, 6: 933–5; also idem, ‘Mazalim and Dar al-‘Adl under the Early Mamluks’, The Muslim World 66, 2 (April 1976): 114–21. 2. For the analysis of historical writing in medieval Egypt and Syria, see Little, Mamluk Historiography, 94–9, 112–36; also Shakir Mustafa, al-Tarikh al-‘Arabi, 2: 139–304, all of vol. 3, 4: 7–227. 3. This is the classical definition of Abu al-Hasan al-Mawardi, al-Ahkam al-Sultaniyya (Cairo, 1909), 64, copied almost verbatim by Abu Ya‘la al-Farra’, al-Ahkam al-Sultaniyya, ed. M. H. al-Fiqqi (Cairo, 1966), 73–4, and translated somewhat differently in Jørgen Nielsen, Secular Justice in an Islamic State: Mazalim under the Bahri Mamluks, 662/1264–789/1387 (Leiden, 1985), 20. 4. A representative text is al-Maqrizi, Khitat, 2: 207–8; for a review, see Émile Tyan, Histoire de l’organisation judiciaire en pays d’Islam, 2 vols (Leiden, 1960), 1: 512–20. 5. Radwan al-Sayyid, ‘Qada’ al-Mazalim: Nazra fi Wajh min Wujuh ‘Alaqat al-Din bi-l-Dawla fi-l-Tarikh al-Islami’, Dirasart 14, 10 (1987): 157–64. 6. The development is outlined in Tyan, Histoire, 1: 87–98; Nielsen, Secular Justice, 2–3. 7. The political interpretation is adapted from Sayyid, ‘Qada’’, 169–74. Nielsen, Secular Justice, 4–7, traces the evolution of mazalim as it alternated between the executive and judicial authority, but does not explain the reason behind keeping the office in the hands of the caliph or why powerful rulers tended to supervise mazalim sessions personally. 8. Al-Mawardi, Ahkam, 65; Abu Ya‘la, Ahkam, 59; Sayyid, ‘Qada’’, 172–3; Nielsen, Secular Justice, 4–5. 9. Abu Yusuf, al-Khiraj (Beirut, 1979), 111–12. Qadi al-qudat was a new position created by al-Rashid for Abu Yusuf. 10. Al-Maqrizi, Khitat, 2: 207. 11. Nielsen, Secular Justice, 6 and references. 12. Al-Mawardi, Ahkam, 64–82. 13. Nielsen, Secular Justice, 18–19; Henri Laoust, ‘La pensée et l’action politiques d’al-Mawardi’, Revue d’Études Islamiques 36 (1968): 12–92. 14. Nizam al-Mulk, Siyasat-Namah, 39 and 70. 15. Sayyid, ‘Qada’’, 172–4; Nielsen, Secular Justice, 7 and references. 16. The Fatimid case is interesting because the place of mazalim sessions depended on the background of the vizier. If he was a ‘man of the sword’, the ceremony took place in the Hall of Gold. –225 –

N O T E S T O P A G E S 14 9 – 1 5 0

17.

18. 19.

20. 21. 22.

23. 24. 25. 26. 27.

28.

29.

If he was a ‘man of the pen’, then another official, the Master of the Gate, would preside over the session outside the Gate of Gold in the great palace, al-Maqrizi, Khitat, 1: 402; al-Qalqashandi, Subh, 3: 491. Ibn ‘Asakir, Tarikh, 16: 294, says in Nur al-Din’s biography that the sultan built in most [cities] of his realm palaces of justices (bana fi akthar mamlakatihi adur al-‘adl) without specifying which cities. The earliest to state that Nur al-Din built the first Dar al-‘Adl in Damascus is the pro-Zengid Ibn al-Athir in his work al-Tarikh al-Bahir fi-l-Dawla al-Atabikiyya fi-l-Musil, ed. A. Q. Tulaymat (Cairo, 1963), 168; also, idem, al-Kamil fi-l-Tarikh, 13 vols (Beirut, 1965–7), 11: 404. He was followed by Abu Shama, Kitab al-Rawdatayn, 1: 8; Sibt ibn-al-Jawzi, Mir’at al-Zaman fi-Tarikh al-A‘yan (Heydarabad, 1952), 8, 1: 309; Ibn Kathir, Bidaya, 12: 280. For extensive research on, and evaluation of, references to Dar al-‘Adl and Dar al-Sa‘ada, see al-Rihawi, ‘Qusur’, 48–70. W. M. Brinner, ‘Dar al-Sa‘ada and Dar al-‘Adl in Mamluk Damascus’, in Studies in Memory of Gaston Wiet, ed. Myriam Rosen-Ayalon (Jerusalem, 1977), 243–4. For a conclusive rebuttal of Brinner’s arguments, see Yasser Tabbaa, ‘The Architectural Patronage of Nur al-Din (1146–1174)’, Ph.D. Thesis, New York University, 1982, 222–5. It is true that the accounts of the three historians, Abu Shama, Sibt ibn al-Jawzi and Ibn Kathir are apparently copied from Ibn al-Athir, but they all lived and worked in Damascus, and they would have noted the fabrication, had the report of Ibn al-Athir been dubious. Tabbaa, ‘Architectural Patronage’, 222, and Nielsen, Secular Justice, 49. Ibn al-Athir, al-Dawla al-Atabikiyya, 168, however, says that Nur al-Din built his Dar al-‘Adl ‘after he had been living for a long time in Damascus’, (tala maqamuhu fi Dimashq). On the problems of the period (1154–63) and Nur al-Din’s constant moving, see Nikita Élisséeff, Nur ad-Din, un grand prince musulman de Syrie au temps des Croisades (511–569/1118–1174), 3 vols (Damascus, 1967), 2: 485–519. Ibn al-Qalanisi, Dhayl Tarikh Dimashq, ed. S. Zakkar (Damascus, 1983), who ends his chronicle in 1160 does not mention the building of Dar al-‘Adl, but in 548, he records in his last entry the selection of Kamal al-Din al-Shahrazuri as the qadi of Damascus. Al-Shahrazuri appears in Ibn al-Athir’s story relating the building of Dar al-‘Adl as the qadi of Damascus, which means that it was constructed after the date of his appointment. Ibn Kathir, Bidaya, 12: 329; al-Bandari, Sana al-Barq al-Shami, (abridged from al-Katib al-Isfahani, al-Barq al-Shami), ed. Fathia al-Nabrawi (Cairo, 1979), 278. Ibn Kathir, Bidaya, 13: 12, for the former, and Abu-Shama, Dhayl, 16, for the latter. Al-Rihawi, ‘Qusur’, 51–2. Baybars apparently repaired the palace, al-Maqrizi, Suluk, 1: 446. Abu-Shama, Dhayl, 220, for the year 1261; Ibn Kathir, Bidaya, 13: 345, for 1295. Early Mamluk sources alternate between the two terms, al-Rihawi, ‘Qusur’, 48–52. The first one to state that Dar al-‘Adl and Dar al-Sa‘ada were one and the same is al-Badari, Nuzhat al-Anam, 18; also Ibn Tulun al-Salihi, al-Sham‘a al-Mudiyya fi Akhbar al-Qal‘a al-Dimashqiyya (Damascus, 1929), 8, equates Dar al-Sa‘ada with Dar al-‘Adl when reporting the founding of the latter. For Farrukhshah, see Abu Shama, Kitab al-Rawdatayn, 2: 19, 23, 33; Sibt ibn al-Jawzi, Mir’at alZaman, 366, 372. Bahramshah was assassinated in Dar Farrukhshah, ibid., 667; Ibn Wasil, Mufarrij al-Kurub, 4: 284–5, explains that Dar Farrukhshah became known in his time as Dar al-Sa‘ada, and that it became the official residence of the na’ib. Nielsen, Secular Justice, 49–50, 52–3, gives an incorrect interpretation of the transfer of the name Dar al-Sa‘ada to Dar al-‘Adl and of al-Qasr al-Ablaq as the new vicegeral palace. Nikita Élisséeff, ‘Les monuments de Nur al-Din: inventaire, notes archéologiques et bibliographiques’, Bulletin d’ Études Orientales 13 (1951): 20–1, constructs a history of Dar al-‘Adl and attributes to Salah al-Din its merging with Dar al-Sa‘ada after 1187. This event does not appear in any of the sources I have consulted, including those listed by Élisséeff (al-Badari, Nuzhat al-Anam; Abu alFida’, al-Mukhtasar fi Akhbar al-Bashar; and Sauvaire’s ‘Description de Damas’, in Journal asiatique (May–June 1894), 425, 477 n. 200, 486, n. 259, (September–October 1895), 266, 270, 307, (May–June 1896), 426).

–226 –

N O T E S T O PAG E S 15 0 –15 3

30. Sibt ibn al-Jawzi, Mir’at al-Zaman, 716, calls the palace Dar al-Sa‘ada; Ibn Kathir, Bidaya, 13: 147. 31. Ibn Kathir, Bidaya, 13: 310; al-Maqrizi, Suluk, 1: 735–6. The manoeuvre, arranged by Sanjar alShuja‘i, who was the vizier during Qalawun’s sultanate (1280–1290), used the excuse that Malaka Khatun had become an imbecile and thus legally ineligible to own property. 32. Ibn Kathir, Bidaya, 13: 228, when Sultan al-Kamil was in Damascus, al-Ashraf moved to Dar alSa‘ada and left the citadel for him; Sibt ibn al-Jawzi, Mir’at al-Zaman, 719, speaks of al-Jawad Yunis moving to Dar al-Sa‘ada in 1237, as a sign of submission to al-Salih Najm al-Din Ayyub who stayed at the citadel. 33. It is very difficult to decide whether this plan took place under Qutuz or his successor Baybars, better known for his restructuring of the Mamluk state. Al-Maqrizi, Suluk, 1: 432, and al-Makin Jurjus ibn al-‘Amid, ‘Akhbar al-Ayyubiyyin’, ed. Claude Cahen, Bulletin d’ Études Orientales 15 (1955–7): 175–6, say that Qutuz stayed at the citadel and his na’ib at Dar al-Sa‘ada when he was in Damascus after the battle of ‘Ayn Jalut. When Baybars became sultan a year later, Qutuz’s na’ib in Damascus, Sanjar al-Halabi, revolted against Baybars and declared himself an independent sultan. As a sign of his revolt, he moved his quarters to the citadel, which indicates that the residential protocol was already in place: see Ibn Kathir, Bidaya, 13: 223. 34. Reference collected in al-Rihawi, ‘Qusur’, 45–8. 35. Ibid., 54–7. 36. Jean Sauvaget, Les monuments ayyoubides de Damas, 2 vols (Damascus, 1940), 2: 57–63, established Dar al-Sa‘ada’s location by identifying the ruins of an Ayyubid madrasa, al-‘Adhrawiyya, known from the sources to have been adjacent to it. 37. Al-Rihawi, ‘Qusur’, 58–61; al-Maqrizi, Suluk, 1: 671, offers a description of the location. 38. Dar al-‘Adl appears to have remained a distinct unit within Dar al-Sa‘ada until late in the Mamluk period. During an uprising in 1415, the dissident amir occupied Dar al-‘Adl and bombarded the citadel from there, which caused the burning of the gable roof ( jamalun) of Dar alSa‘ada. This indicates that the two were separate; see al-Maqrizi, Suluk, 4: 321; Ibn Taghri-Birdi, Nujum, 14: 32–4. 39. Ibn-Tulun al-Salihi, I‘lam al-Wara bi-man Wulliya Na’iban min-al-Atrak bi-Dimashq al-Sham al-Kubra, ed. M. A. Duham (Damascus, 1984), 92; Ibn Kathir, Bidaya, 14: 317, for the qa‘a; al-Qalqashandi, Subh, 4: 194–7, offers the most elaborate description of the mazalim ceremony in Damascus, probably as it was in the middle of the fourteenth century. 40. Ibrahim and Amin, Mustalahat, 87–8. 41. A passing sentence in a fifteenth-century chronicle (Sibt ibn-al-‘Ajami who died in 1479, Kunuz al-Zahab fi-Tarikh Halab) wrongly attributes Dar al-‘Adl in Aleppo to Nur al-Din. Jean Sauvaget, Alep, Essai sur le developpement d’une grande ville syrienne, des origines au milieu du XIXe siècle (Paris, 1941), 126, n. 424, favoured that reference to earlier sources which credited Ghazi with the structure; see also Tabbaa, ‘Architectural Patronage’, 221. 42. Ibn Shaddad, al-A‘laq al-Khatira, 1, 1: 17; Ibn al-Shuhna, al-Durr al-Muntakhab fi-Tarikh Mamlakat Halab, ed. Joseph Elyan Sarkis (Beirut, 1909), 33–4, repeats the text of Ibn Shaddad and brings the information on the city’s topography up to date. 43. Sauvaget, Alep, 146, for a concise description of Dar al-‘Adl. Further details are collected from Ibn Shaddad, al-A‘laq al-Khatira, 1, 1: 17, 21, 25; Ibn al-Shuhna, al-Durr al-Muntakhab, 34, 41–2, 51; Kamel al-Ghazzi, Nahr al-Zahab fi-Tarikh Halab (Aleppo, 1920), vol. 2, 4, 7, 8, 111. 44. On Ghazi’s reign and policy, see Ibn al-‘Adim, Zubdat al-Halab min Tarikh Halab, 3 vols, ed. Sami al-Dahhan (Damascus, 1951–68), 3: 136–41; Gaston Wiet, ‘Une inscription de Malik Zahir Ghazi à Lattakieh’, Bulletin de l’Institut français d’archéologie orientale du Caire 30 (1930–31): 273–92, esp. 282–6. 45. Ibn Wasil, Mufarrij al-Kurub, 3: 239. 46. Ibid., 248–9. 47. Ibn al-‘Adim, Zubdat, 3: 205, 267. 48. Ibn Wasil, Mufarrij al-Kurub, 3: 180.

–227–

N O T E S T O PAG E S 15 3 –15 6

49. Sauvaget, Alep, 169. 50. Al-Qalqashandi, Subh, 4: 222–5, offers a description of the mazalim ceremony in Aleppo. 51. Al-Nuwayri, Nihayat al-Arab, vol. 30, fol. 63; and Baybars al-Mansuri, Tuhfa, 224, say that Dar al-‘Adl al-Kamiliyya overlooked the durqah of the citadel (inside the Qulla Gate). 52. Al-Maqrizi, Khitat, 2: 204; idem, Suluk, 1: 202. 53. Sawirus ibn al-Muqaffa‘, Siar al-bay‘a al-Muqaddasa, ed. A. Khater and O. H. E. Burmester, (Cairo, 1974), 4, 2: 107; Ibn Wasil, Mufarrij al-Kurub, 5: 241–2, events of 637/1239; a truncated report in al-Maqrizi, Suluk, 1: 306–7. 54. Idem, Khitat, 2: 208, 374; idem, Suluk, 1: 373. 55. Al-Nuwayri, Nihayat al-Arab, vol. 31, fol. 99, the amir lived there in the year 1330. 56. Ibn Shaddad, Tarikh al-Malik al-Zahir, 341–2. 57. Ibn ‘Abd al-Zahir, Rawd, 182. 58. Casanova, Tarikh, 103–4. The sources we have today refute the ascription of this building to alKamil for his Dar al-‘Adl was inside the Citadel, whereas that of Baybars was outside. AlMaqrizi, Suluk, 2: 236; and Ibn Taghri-Birdi, Nujum, 9: 74, corroborate the report of Shafi‘ ibn ‘Ali, cited by Casanova, about the Dar al-‘Adl having been a Fatimid mausoleum by recording the discovery of buried bodies under the structure when it was rebuilt as a tablakhana in 1322. 59. Al-Maqrizi, Suluk, 1: 501; Ibn ‘Abd al-Zahir, Rawd, 210, say that Baybars sat under the awning (suffah) in the mastaba next to Dar al-‘Adl to review the parades in the maydan. 60. Al-Maqrizi, Khitat, 2: 205; Ibn-Shaddad, Tarikh al-Malik al-Zahir, 277–82. For a list of cases reviewed by Baybars in Dur al-‘Adl in Cairo and Damascus, see Nielsen, Secular Justice, 144–7. 61. Ibn al-Furat, Tarikh, 7: 259; and al-Maqrizi, Suluk, 1: 712, report that the assessment of the polltax ( jawali) from non-Muslims in 1283 took place in the Dar al-‘Adl under the citadel, and was supervised by the vizier. Al-Maqrizi, Khitat, 2: 206, says that Dar al-‘Adl al-Zahiriyya was used for mazalim sessions presided over by the na’ib until the reign of Qalawun. 62. Al-Maqrizi, Khitat, 2: 205, 213, and, Suluk, 2: 236; Ibn Taghri-Birdi, Nujum, 9: 74. 63. Casanova, Tarikh, 105–6. 64. ‘The zawiya stands below (sifl) the tablakhana of the Citadel’, awqaf 1079, waqf of Hasan b. Ilyas al-Rumi, dated 8 Shawwal, 941(1535), line 21. Reference from M. Hussam al-Din ‘Abd al-Fattah, ‘Mantiqat al-Darb al-Ahmar (the district of al-Darb al-Ahmar)’, Master’s Thesis, Asyut University, 1986, 311, lines 34–5. 65. Ibn ‘Abd al-Zahir, Rawd, 141; Ibn al-Dawadari, al-Durra al-Zakiyya, 63; al-Maqrizi, Suluk, 1: 438. 66. Ibn-Shaddad, Tarikh al-Malik al-Zahir, 339–40; al-Kutubi, Fawat al-Wafi yyat, 1: 242; Ibn TaghriBirdi, Nujum, 7: 190. 67. Ibn al-Furat, Tarikh, 8: 38; Ibn-‘Abd al-Zahir, Sirat al-Malik al-Mansur, 139. Al-Maqrizi, Khitat, 2: 212, says that Qalawun demolished the qulla to build his qubba. I think this is a typographical error, which has qulla where it should have qubba. 68. Chroniclers are not unanimous in ascribing the construction of an iwan to Khalil. Ibn alDawadari, al-Durra al-Zakiyya, 345, states that ‘al-Iwan al-Ashrafi was completed by Amir ‘Alam al-Din al-Shuja‘i. In Zetterstéen, Geschichte der Mamlukesultane, 156, the unknown chronicler who lived during that period refers to the iwan as al-Ashrafi when al-Nasir ordered its destruction in 1311; al-Nuwayri, Nihayat al-Arab, vol. 30, fol. 66, however, calls the same structure the Iwan al-Mansuri when he reports that event. 69. Description de l’Egypte, état moderne, planches, vol. 1 (Paris, 1809), plate 26 for the map of Cairo with the citadel, plates 70–2 for views, sections and elevations of the Great Iwan. 70. Al-Maqrizi, Suluk, 1: 734; and Ibn al-Furat, Tarikh, 8: 49, 58, report an incident in 1287, which indicates that the four qudat al-qudat sat in the Dar al-‘Adl in some order, but they do not mention the location of this structure nor whether Sultan Qalawun attended the service. Al-Maqrizi, Suluk, 1: 772, says that Khalil sat in the Dar al-‘Adl but does not say where that structure was. Ibid., 1: 830, also reports that Sultan Lajin (1297–1299) decided to sit in the Dar al-‘Adl twice a week but, again, no mention of where it was. 71. Al-Maqrizi, Suluk, 1: 756. –228 –

N O T E S T O PAG E S 157 –16 3

72. Ibn al-Dawadari, al-Durr al-Fakhir, 238; al-Maqrizi, Suluk, 2: 107; Ibn Taghri-Birdi, Nujum, 9: 51. 73. Al-‘Umari, Masalik, 116–17; al-Maqrizi, Khitat, 2: 215; al-Qalqashandi, Subh, 4: 16–17. Both copied al-‘Umari with a few additions. 74. For a concise summary of al-Nasir’s reign, see Irwin, Mamluk Sultanate, 85–124; also, Holt, Crusades, 107–20. 75. The chroniclers offer many examples of the ‘amma, who were generally believed to be indifferent to government changes, helping al-Nasir. For an instance in 1307 see al-Maqrizi, Suluk, 2: 34–5; for another one in 1310, ibid., 2: 68; and Baybars al-Mansuri, Tuhfa, 204. For reactions to his opponent Baybars al-Jashankir, see Ibn Taghri-Birdi, Nujum, 8: 244; al-Maqrizi, Suluk, 2: 71, Baybars al-Mansuri, Tuhfa, 199. 76. Hayat al-Hajji, The Internal Affairs in Egypt during the Third Reign of Sultan al-Nasir Muhammad B. Qalawun (Kuwait, 1978), 78–9, noted the political importance of al-Nasir’s decision to attend mazalim sessions, but went on to advance an improbable and historically unverified reaction of the amirs to this decision. 77. Al-‘Umari, Masalik, 141–2 offers the most complete description of the iwan’s layout. His text was copied in al-Maqrizi, Khitat, 2: 210. 78. Ibid., 2: 206; Ibn Shahin al-Zahiri, Zubdat, 26; Ibn Iyas, Bada’i‘, 5: 441. 79. Casanova, Tarikh, 127, was the first to read the remainder of the inscription correctly. 80. Al-‘Umari, Masalik, 100–2, who was an eyewitness; al-Maqrizi, Khitat, 2: 208–9; and alQalqashandi, Subh, 4: 44–5, 62, copy al-‘Umari and add new information pertaining to the custom in their period, but they confuse it with the original report. Baybars al-Mansuri, Tuhfa, 231, 233–4, gives an elaborate description of the event and lists the names of the amirs who were required to sit around the sultan. An English translation of al-Maqrizi’s report can be found in Holt, Crusades, 144–5; Karl Stowasser, ‘Manners and Customs at the Mamluk Court’, Muqarnas 2 (1984): 17; Nielsen, Secular Justice, 54–9. 81. Jomard, ‘Déscription de la ville et de la citadelle du Kaire’, 354–5, thought it was a mihrab and concluded that the iwan might have been a mosque. 82. By asserting that the sultan’s throne was on a level higher than his officials, Nielsen, Secular Justice, 56, missed the point about the symbolic humility implied in the sultan’s consciously abandoning the higher throne on dar al-‘adl days in favour of the lower dast. 83. See the references to al-‘Umari’s tenure as katib al-sirr in Nielsen, Secular Justice, 160. 84. A number of architectural precedents to the Great Iwan have been proposed; see Doris BehrensAbouseif, ‘Citadel of Cairo’, 77–8; Lézine, ‘Salles nobles’, 65, 71. 85. The Florentine traveller Brancacci, who reported on the audience he attended there during the reign of Barsbay (1422–1437), says that the sultan was seated on a raised platform and was perfectly visible from all sides; see Behrens-Abouseif, ‘Citadel of Cairo’, 42–3. 86. Irving Lavin, ‘The House of the Lord’, Art Bulletin 44 (1962): 16–17, begins to chart the survival of basilicas in royal Islamic architecture. For a fuller discussion of the architecture of the Great Iwan, see Nasser Rabbat, ‘Mamluk Throne Halls Qubba or Iwan’, Ars Orientalis 23 (1993): 201–9. [Reproduced in this volume.] 87. Al-Shuja‘i, Tarikh, 235; al-Maqrizi, Khitat, 2: 214–15; Ibn Taghri-Birdi, Nujum, 10: 32. 88. Al-Maqrizi, Khitat, 2: 207, 241, and Suluk, 3: 566; Ibn al-Furat, Tarikh, 9: 17; Ibn Iyas, Bada’i‘, 1, 2: 388. 89. For Barsbay’s work, see Ibn Taghri-Birdi, Nujum, 14: 318; al-Sayrafi, Nuzhat al-Nufus, 3: 238–9. For Qaytbay’s work, see idem, Inba’ al-Hasr, 294, 327, 339; Ibn Iyas, Bada’i‘, 3: 60–1, 329. 90. For a collection of eyewitness reports on the destruction of the iwan by Muhammad-‘Ali, see Gaston Wiet, Mohammed Ali et les beaux-arts (Cairo, 1949), 265–88. 91. See the general remarks on the ‘Turkish’ penetration of Anatolia, Syria and Jezira in Claude Cahen, ‘La première pénétration turque en Asie-Mineure’, 61–6; Musatafa, ‘Dukhul al-Turk alGhuzz ila al-Sham’, 303–4. 92. Sivan, L’Islam et la Croisade, is the most complete study to date on this important yet neglected topic. –229 –

N O T E S T O PAG E S 16 3 –16 4

93. Examples abound, especially from the periods of Nur al-Din and Salah al-Din; as samples, see the letter Nur al-Din sent to al-Mustadi’ in 1173, Abu Shama, Kitab al-Rawdatayn, 1: 215; Ibn alAthir, Kamil, 11: 395, for a description of the contents. For a series of letters composed by alQadi al-Fadil for Salah al-Din and sent to the caliphs in Baghdad, see al-Qalqashandi, Subh, 6: 496–516, 7: 126–30. For the Arabic text and an English translation of the letter Salah al-Din sent after the conquest of Tiberias in 1187, see C. P. Melville and M. C. Lyons, ‘Saladin’s Hattin Letter’, in The Horns of Hattin, ed. B. Z. Kedar (Jerusalem, 1992), 208–12. 94. For the khutba’s text, see Ibn ‘Abd al-Zahir, Rawd, 102–10; Ibn al-Dawadari, al-Durra al-Zakiyya, 73–9; al-Maqrizi, Suluk, 1: 453–7; al-Qalqashandi, Subh, 10: 111–16. For the khutba of the second caliph, al-Hakim, installed by Baybars in 1262 after the death of al-Mustansir in his illfated campaign against the Mongols, see Ibn ‘Abd al-Zahir, Rawd, 141–7. The same khutba was delivered by the same caliph, more than thirty years later when al-Ashraf Khalil released him from house arrest in 1291 and ordered him to praise his conquests publicly, see al-Maqrizi, Suluk, 1: 771–2; Ibn al-Furat, Tarikh, 8: 128–9, 135. See also P. M. Holt, ‘Some Observations on the Abbasid Caliphate of Cairo’, Bulletin of the School of Oriental and African Studies 47 (1984): 501–7. 95. Ibn Shahin al-Zahiri, Zubdat, 89–90, lists the Muslim rulers who sought investitures from the caliph in Cairo, and adds that the title sultan should be given only to the ruler of Egypt because of his direct investiture by the legitimate caliph. Al-Qalqashandi, Subh, 10: 129–34, reproduces a letter of investiture written in 1410 in the name of the Caliph al-Musta‘in Billah (1406–1414) to Shams al-Din Muzaffar Shah, king of Delhi (probably Zafar Khan Muzaffar Shah (1391–1411), the last Tughluqid governor and founder of the sultanate of Gujarat, but asserts that this was the only diploma for a sultan other than the Mamluks. 96. Adel Allouche, ‘Tegüder’s Ultimatum to Qalawun’, 437–46, noticed a pertinent discrepancy between the Mamluk and Ilkhanid versions of Qalawun’s reply to Tegüder, whereby the two Ilkhanid sources include Qalawun’s praise for Caliph al-Hakim, whom he in fact put under house arrest, at the beginning of the text. The Mamluk chroniclers, with one exception, omit that reference, probably to mitigate any possible angry reaction of the sultan. Allouche accurately explains this political stratagem on the part of the Mamluk sultans to underscore their Islamic legitimacy and seniority vis-à-vis the Ilkhanids. 97. See the illuminating analysis of Dorothea Krawulsky concerning the roles of the intellectuals, the resurrected caliphate and the leadership of jihad in providing the legitimisation for the Mamluk state in her introductory study to al-‘Umari, Masalik, 15–37, text of Wassaf, 33–4. 98. Examples in Sivan, L’Islam et la Croisade, 67–70, 102–8, 133–4, 178–80. 99. Ibn Jubair, al-Rihla, 216, commenting on the profusion of elaborate titles, remarked that many of these princes do not deserve the honours they attribute to themselves, but excluded Salah al-Din from his general judgment. 100. Nikita Élisséeff, ‘La titulature de Nur al-Din d’après ses inscriptions’, Bulletin d’ Études Orientales 14 (1952–4): 155–96. 101. Tabbaa, ‘Monuments with a Message’, 223–40, analyses the development of Nur al-Din’s titulature. 102. At the end of the thirteenth century, one of the leading legists placed the dispensing of justice second only to the jihad; Ibn Jama‘a, Tahrir al-Ahkam, esp. chapters 2 and 5. 103. Regardless of his known favouritism towards the Zengids, Ibn al-Athir, Kamil, 11: 403, was clearly expressing a true feeling when he stated that he ‘had read the biographies of the sovereigns of old, and other than among the first [Rashidi] caliphs he had not found a man as virtuous and just as Nur al-Din’. 104. The most glaring example is the Sirat Baybars (Roman de Baîbars) translated by Georges Bohas and Jean-Patrick Guillaume from an Aleppine manuscript written in the early nineteenth century, of which eight volumes have been published since 1985. It is too early to conduct any fullscale study on the relationship between the real-life Baybars and the epic hero, but for a preliminary discussion see J. P. Guillaume, ‘Présentation’, in Les enfances de Baîbars (Paris, 1985), –230 –

N O T E S T O PAG E S 16 4 –16 7

105. 106. 107.

108. 109. 110. 111.

20–35. Unfortunately, the Islamic epic literature of the Crusades and of the post-Crusades period is a subject that has been barely touched on in a scholarly fashion, see Sivan, L’Islam et la Croisade, 195–206. Al-Maqrizi, Suluk, 2: 536–7; Ibn Taghri-Birdi, Nujum, 9: 176. This change of mentality is not well charted, but a passage in al-Maqrizi, Khitat, 2: 213–14, bewails the Mamluk character during the Qalawunid period and outlines the changes that took place with the advent of the Burji age. This was a clever tactic by which al-Nasir effectively prevented the usurpation of the throne from his designated heir by appointing Qawsun and Bashtak as regents, as they had no right to become sultans themselves because neither of them was a real Mamluk nor a son of a sultan. As apparent from a dialogue reported in Ibn Taghri-Birdi, Nujum, 10: 19–20, they realised that they were limited in their scheming to manipulate the sultan. On the Mamluk training system, see David Ayalon, ‘L’esclavage du Mamelouk’, 9–26; also, idem, ‘Mamlukiyyat: (B) Ibn Khaldun’s View of the Mamluk Phenomenon’, 345–6, for a translation of Ibn Khaldun’s passage on the education of Mamluks. Dorothea Krawulsky, in her introduction to ‘Umari, Masalik, 36–7, gives a similar interpretation in her review of the changes in the literary production after the reign of al-Nasir Muhammad. These are the words of al-Maqrizi, the bitter critic of his age. On the campaigns against Cyprus and the jihad propaganda initiated by the sultan, see alMaqrizi, Suluk, 4: 721–6, 738–43; Ibn Taghri-Birdi, Nujum, 14: 287–90, 292–304; al-Sayrafi, Nuzhat al-Nufus, 3: 76–95; Ibn Iyas, Bada’i‘, 2: 97–100, 106–10; also Ahmad Darrag, L’Égypte sous le règne de Barsbay 825–841/1422–1438 (Damascus, 1961), 239–61. For Barsbay sitting in the Great Iwan for mazalim sessions on Tuesdays and Saturdays, see al-Maqrizi, Suluk, 4: 871; Ibn TaghriBirdi, Nujum, 14: 361–2; Ibn Iyas, Bada’i‘, 2: 173–4.

CHAPTER 14 1. See, for instance, the list of artefacts taken from the Fatimid palace in Cairo, al-Maqrizi, Khitat, 1: 414–16; or the list of presents sent to Baraka of the Golden Horde by Baybars in Ibn ‘Abd al-Zahir, Rawd, 172–3; or the list of objects sent to al-Mu’ayyad by the Doge of Venice in al-Maqrizi, Suluk, 4: 325; or the list of objects sent out as dowry for the daughter of Baktimur al-Saqi when she married Anuk the son of al-Nasir Muhammad in 1332, in al-Safadi, Wafi, 10: 194–5; al-Maqrizi, Khitat, 2: 68; or the objects confiscated from the estate of Tankiz after his fall in 1340, in al-Safadi, Wafi, 10: 428. The terms used need a special philological and contextual study before they can yield useful information on these objects. 2. For a succint discussion of notions of aesthetics in classical and medieval Islamic writing, see Gulru Necipoglu, The Topkapi Scroll: Geometry and Ornament in Islamic Architecture: Topkapi Palace Museum Library MS H. 1956 (Santa Monica, CA, 1995), 185–215. 3. For the first instance, the hall was called the Bayt al-Dhahab (House of Gold) in Khumarawayh’s palace (884–896). It was covered with larger-than-life painted wooden reliefs depicting him and his favourite concubines and singers, see al-Maqrizi, Khitat, 1: 316–17; for the second, the diwan of al-Ashraf Khalil had representations of his amirs, each with his own rank above his head, see Ibn al-Dawadari, al-Durra al-Zakiyya, 345; al-‘Ayni, ‘Iqd al-Juman, 3: 79–80; al-Maqrizi, Khitat, 2: 213. 4. Al-Maqrizi, Khitat, 2:318. For a translation and an analysis of al-Maqrizi’s text, see Arnold, Painting in Islam, 21–2; Richard Ettinghausen, ‘Early Realism in Islamic Art’, in Studi Orientalistici in onore di Giorgio Levi Della Vida (Rome, 1956), 1: 250–73, esp. 267–71. See also idem, ‘Painting in the Fatimid Period: A Reconstruction’, Ars Islamica 9 (1942): 112–13. Both articles are reprinted in idem, Islamic Art and Archaeology: Collected Papers, ed. M. Rosen-Ayalon – 2 31 –

N O T E S T O PAG E S 16 7 –16 8

5. 6.

7.

8. 9. 10.

11. 12.

13. 14.

15.

(Berlin, 1984). Al-Maqrizi compares the two painters, al-Qasir and Ibn ‘Aziz to Ibn Muqla and Ibn al-Bawwab respectively, further indicating that for a member of the literati, calligraphy was the more fathomable artistic referent. Oleg Grabar, ‘Patronage in Islamic Art’, in Islamic Art & Patronage: Treasures from Kuwait, ed. E. Atil (New York, 1990), 35, thinks the report dubious because of its use of a kind of rhetoric narrative known from antiquity. Arnold, Painting in Islam, 22, erroneously attributes the book to al-Maqrizi, when in fact al-Maqrizi does not make such a claim. Ettinghausen, Arab Painting, 55, suggests that the illusion of movement may have been achieved by foreshortening in addition to the colour contrast. Al-Maqrizi’s text cannot be interpreted to support this proposition, although some surviving images from the Fatimid period, notably at the Capella Palatina in Palermo, hint at foreshortening. The debate on whether a ban on representational art was effectively imposed still rages, despite the fact that examples from every period and place exist to just make the proposition itself moot. It is true that many famous legal authorities, such as al-Ghazali and al-Subki (Mu ‘id al-Ni ‘am, 135), unequivocally forbid representational art, but the mere fact that they admonish their readers to efface or destroy statues and images that they encounter especially in hammams betrays the idealistic nature of their exhortations, cf. Ahmad Taymur, Taswir, 10–11. Also see Hasan al-Basha, al-Taswir al-Islami fi-l-‘Usur al-Wusta (Cairo, 1959); Richard Ettinghausen, Arab Painting (Geneva, 1977); David James, Arab Painting (Edinburgh and London, 1978); Terry Allen, ‘Aniconism and Figural Representation in Islamic Art’, in his Five Essays in Islamic Architecture (Sebastopol, CA, 1988), 17–37. This is especially true of the third painting mentioned there by al-Maqrizi, Khitat, 2: 318, that of the Patriarch Joseph in the well painted on a black background in a house in al-Qarafa. Try as I might, I could not make sense of what al-Maqrizi is actually saying about this painting. Ibn-‘Abd al-Zahir, Rawd, 246, for Baybars’ Dar al-Dhahab; Ibn al-Dawadari, al-Durra al-Zakiyya, 345; al-‘Ayni, ‘Iqd al-Juman, 3: 79–80; al-Maqrizi, Khitat, 2: 213, for Khalil’s iwan or rafraf. Taymur, Taswir, 65–6, cites a report from the still-unpublished chronicle of ‘Abd al-Basit alHanafi, al-Rawd al-Basim fi Hawadith al-‘Umr wa-l-Tarajim, about the events of 1461 when the father of the author, Khalil al-Zahiri, a Mamluk amir who was a famous chronicler himself and who opposed statues for religious reasons, refuted the argument of ‘some alleged ‘ulama’ in the presence of Sultan Khushqadam and convinced the sultan to remove a gilded statue of a falcon from his reception hall. Another instance cited by Taymur shows that al-Qadi al-Fadil, Salah al-Din’s vizir, was inclined to condemn the shadow-play for its use of figures, but after he watched it for a while, declared that he finds it to be of ‘great effect as conveyer of moral stories’, Ahmad Taymur, Khayal al-Zill (Cairo, 1957), 12. Al-Maqrizi, Khitat, 2: 147, 177. This state of affairs has frustrated many authors who dealt with early and classical Islamic art, whose examples of figural representations appear to have been unnoticed by the contemporary historians who otherwise commented on many other aspects of cultural life. Citations are unnecessary for this lament has become a regular routine in studies on Umayyad and Abbasid art and architecture. An overview is J. E. Bencheikh, art. ‘Khamriyya’, Encyclopaedia of Islam, 2nd edn, 4: 998–1009. For Ibn ‘Abd al-Zahir’s poetry see his biography in al-Safadi, Wafi, 17: 257–90; or his Diwan, ed. Gharib M. Ahmad (Cairo, 1990). An outstanding example is ‘Abd al-‘Aziz ibn Saraya, Safiyy alDin al-Hilli, the most famous poet of the early fourteenth century in Egypt whose work ranges from extremely pietistic poems to outrageously sensual and homosexual ones. His contemporary al-Safadi, Wafi, 18: 481–512, lists all of them in his biography with palpable admiration and without any hint of shame or trepidation. It is odd that the modern anonymous editors of his poetry are the ones who seem to have censured his libertine poetry by eliminating it from his published diwan (Beirut, n.d.). Ibn Shaddad, Tarikh al-Malik al-Zahir, 340, only reports the existence of images in Baybars’ qubba but does say anything about them. Instead, he quotes a few verses from a poem of the long-dead –232–

N O T E S T O P A G E S 1 6 8 – 17 1

16.

17. 18. 19.

20.

21.

22. 23.

24. 25. 26.

27.

28.

Ibn Hayyus praising a different setting to describe their subject matter and stops short of quoting the full poem because, as he says ‘the rest of its verses mention images that do not exist in the qubba’. Another example is a poem recited by al-Rashid al-Nabulsi on the occasion of the inauguration of Dar al-Shukhus (House of Figures) built by al-Zahir Ghazi at the Citadel of Aleppo in 1193, see Ibn Shaddad, al-A‘laq al-Khatira, 1, 1: 25–6; Ibn al- Shihna, al-Durr al-Muntakhab, 52–3; al-Ghazzi, Nahr al-Dhahab, 2: 26. Makdisi, Rise of Humanism, presents one of the most thorough discussions of the types of knowledge and kinds of settings available to medieval Islamic ‘humanists’ (to use Makdisi’s term). Makdisi, (Append. A, 355–61) provides a summary of Dia’ al-Din ibn al-Athir’s eight scholarly requisites for poets and katibs from his al-Mathal al-Sai’r fi Adab al-Katib wa-al-Sha‘ir (Riyadh, 1983–4), which shows clearly that no visual concerns penetrated those lists. See S. A. Bonebakker, art. ‘Gharib’, Encyclopaedia of Islam, 2nd edn, 2: 1011. See the discussion of the concept of ‘aja’ib in Qur’anic studies in Mohammed Arkoun, ‘Peut-on parler de merveilleux dans le Coran?’ in his Lectures du Coran (Paris, 1982), 87–144. For a recent survey of the field, see Caroline W. Bynum, ‘Wonder’, American Historical Review (February 1997): 1–26, for the various theoretical approaches to wonder in Europe in the medieval period. For the relationship between geography and ‘aja’ib in medieval Islamic texts, see Miquel, Géographie humaine, 2: 484 ff. I differ here from earlier attempts to conceptually distinguish between the two terms, cf. Tawfiq Fahd, ‘Le merveilleux dans la faun, la flore et les minéraux’, and the ensuing discussion, in M. Arkoun et al., L’étrange et le merveilleux dans l’islam médiéval (Paris, 1978), 117–65. My position is similar to that of Miquel, Géographie humaine, 138 –9. Two of the three famous Awlad al-Athir brothers compiled a Gharib al-Hadith, Dia’ al-Din ibn alAthir, al-Nihaya fi Gharib al-Hadith wa-al-Athar (Tehran, 1852, reprint 1904), and Majd al-Din Ibn al-Athir, Manal al-Talib fi Sharh al-Gharaib (Mecca, 1983); Ibn Hajar al-Asqalani, Tafsir Gharib al-Hadith: Murattaban ‘ala al-Huruf (Cairo, n.d.). See art. ‘Qazwini’ by T. Lewicki, Encyclopaedia of Islam, 2nd edn, 4: 865–7. Al-‘Umari, Masalik al-Absar, vol. 2, sec. 3, art. 3, entitled zikr nibza min ‘aja’ib al-barr wal bahr (listing of some ‘aja’ib of the sea and earth) of which many items appear to have been glossed from al-Qazwini, ‘Aja’ib al-Makhluqat wa Ghara’ib al-Mawjudat, ed. F. Sa‘d (Beirut, 1973). Al-Nuwayri, Nihayat al-Arab, vol. 1 deals with the celestial and the terrestial phenomena as they appear in alQazwini’s ‘Aja’ib section. Al-Qalqashandi, Subh, quotes al-Qazwini several times and devotes half of volume 2 to the celestial and the terrestial phenomena as well. It seems that at least the ‘scientific’ repertoire of ‘aja’ib had, by the fourteenth century, moved out of its specialised niche and became incorporated in the usual curriculum of kuttabs and literati in general. Al-Qazwini, ‘Aja’ib, 29, 31, 38. For a discussion of the status of artists and builders in the Mamluk society, see Nasser Rabbat, ‘Architects and Artists’. Sabra, Optics of Ibn al-Haytham, 2: 99, discusses the example of the famous essayist al-Jahiz, Risalat al-Qiyan, ed. ‘Abd al-Salam Muhammad Harun in his Rasa’il al-Jahiz, 2 vols (Cairo, 1965), 2: 162–3. Another example, closer in time and place to the Mamluk authors is al-Baghdadi, al-Ifada, 30–31, where he employs the notion of tanasub (proportionality) to analyse the beauty achieved in the ancient Egyptian statues. See Arnold, Painting in Islam, 1–40; Creswell, Early Muslim Architecture, 1: 269–71, for an interpretation tinged with what we would call today racist expressions; expanded in idem, ‘The Lawfulness of Painting in Early Islam’, Ars Islamica 11–12 (1946): 159–67. The usual list of Islamic references is amended by Bishr Farés, Sirr al-Zakhrafa al-Islamiyya (Cairo, 1952), 31–4, to include legal opinions that see no problem with painting. Al-Maqrizi, Suluk, 1: 938–40; Ibn Taghri-Birdi, Nujum, 8: 165–8; for a discussion in the context of popular culture in Cairo, see Shoshan, Popular Culture, 67–76. On page 133, n. 80, Shoshan noted that displaying model citadels seems to have been used on more than one occasion, perhaps it was a tradition, but still the sources give us no idea about its exact meaning. –233 –

N O T E S T O P A G E S 17 1 – 17 4

29. Al-Maqrizi, Khitat, 2: 322. 30. Popular culture in medieval Egypt is a subject that has often been evoked but rarely studied despite the wealth of information that can be gleaned from particular sources. Some recent studies began to investigate some aspects of Cairene popular culture such as Shoshan, Popular Culture; and Huda Lutfi, ‘Manners and Customs of fourteenth-century Cairene Women: Female Anarchy Versus Male Shar‘i Order in Muslim Prescriptive Treatises’, in Women in Middle Eastern History: Shifting Boundaries in Sex and Gender, ed. Nikki Keddie and Beth Baron (New Haven, CT, 1991), 99–121. 31. Al-Maqrizi, Khitat, 1: 405. 32. Abu Shama, Dhayl, 11, Sibt Ibn al-Jawzi, Mir’at al-Zaman, 8, 1: 290. 33. Al-Maqrizi, Khitat, 2: 99–100, gives a definition of ‘alaliq, and a description of their more common and benign forms: fruits, vessels, animals. 34. For Qawsun, see al-Maqrizi, Suluk, 2: 574–7, 579, 586–95; Ibn Taghri-Birdi, Nujum, 10: 24–30, 38–48, 51–2, 60–61; for al-Nashw, see Mufaddal ibn Abi al-Fada’il, al-Nahj al-Sadid, 20: 78; Shoshan, Popular Culture, 53–8, discusses both instances as aspects of the ‘political expression’ of the Cairene crowd. 35. Can this imply that the literati, not unlike their European counterparts who produced and consumed written texts, considered the visual arts to be the texts of the illiterates and therefore unworthy of their attention? For a discussion in connection to medieval Europe, see Lawrence Duggan, ‘Was Art Really the Book of the Illiterate?’ Word and Image (1989): 227–51. 36. Gulru Necipoglu in her review of Risale-i Mi‘mariyye: An Early Seventeenth Century Ottoman Treatise on Architecture, ed. Howard Crane (Leiden, 1987), in Journal of the Society of Historians of Architecture 49, 2 (June 1990): 210–14, notes how the religious training of Ca‘fer Efendi, the author of the treatise, tinged his approach, understanding and appreciation of architecture and its terminology. Her concern about the use of this Ottoman source in interpreting the status of architecture at the time resonates with the caution presented here about the medieval case.

CHAPTER 15 1. Edward Said, Orientalism (New York, 1979). 2. For an analysis of orientalising architecture, see Zeynep Çelik, Displaying the Orient: Architecture of Islam at Nineteenth-Century World’s Fairs (Berkeley, 1992); for a nuanced view, see Sibel Bozdogan, ‘Journey to the East: Ways of Looking at the Orient and the Question of Representation’, Journal of Architectural Education 41, 4 (Summer 1988): 38–45; for a critical analysis of Orientalist painting, Linda Nochlin, ‘The Imaginary Orient’, Art in America 71 (May 1983): 119–29, 187–91. 3. For a French view of the effects of the expedition on Egypt, see Edouard Driault, ‘La Renaissance de l’Égypte’, Napoléon, revue des études napoléoniennes 14, 1 (January-February 1925): 5–22; for a collection of Arabic views, see Ra’if Khuri, Modern Arab Thought: Channels of the French Revolution to the Arab East, trans. I. ‘Abbas (Princeton, 1983); for a critical view, see Louca, ‘Renaissance égyptienne’, 1–20. 4. For a survey of the incorporation of Islamic motifs in nineteenth-century British architecture, see Michael Darby, Islamic Perspective: An Aspect of British Architecture and Design in the 19th Century (London, 1983); a more complete and more critical study is John Sweetman’s The Oriental Obsession: Islamic Inspiration in British and American Art and Architecture 1500–1920 (Cambridge, 1987). For the peculiar case of Carl von Diebitsch, a Berlin architect who wanted to standardise and industrialise Islamic elements for mass use in Europe but who ended up exporting the style back to the Egypt of Khedive Isma‘il, see Elke Pflugradt-Abdel Aziz, ‘The Mausoleum for Soliman Pasha ‘el-Faransawi’ in Cairo’, Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 44 (1988): 205–14. 5. Long before the rise of postcolonial studies, Jacques Berque used a historicising approach in his study of the colonisation of Egypt in the nineteenth century, which was first published in 1967, –234 –

N O T E S T O P A G E S 17 4 – 17 6

6. 7.

8.

9. 10. 11.

12.

13.

14.

15. 16.

see Berque, Egypt, 1012. A more recent study is Mitchell, Colonizing Egypt. A cursory inquiry of an East-West interaction in a cultural field is Donald Malcolm Reid, ‘Cultural Imperialism and Nationalism: The Struggle to Define and Control the Heritage of Arab Art in Egypt’, International Journal of Middle Eastern Studies 24/1 (February 1992): 57–76. Berque, Egypt, 468. This is a seriously neglected period of architecture in Egypt. A preliminary discussion of its major examples is Robert Ilbert and Mercedes Volait, ‘Neo-Arabic Renaissance in Egypt, 1870–1930’, Mimar 13 (1983): 26–34; for an attempt to contextualise it see Mona Zakarya, ‘L’inscription du discours occidental dans l’architecture et l’urbanisme orientaux’, in D’un Orient l’autre: les metamorphoses successives des perceptions et connaissances, vol. 1. Configurations, 559–74; a new study that lists and classifies the monuments but does little else is Sakr, Islamic Architecture in Cairo. The best source on the urban history of Cairo in English remains Abu-Lughod, Cairo, 1001 Years of the City Victorious; see also André Raymond, Le Caire, which incorporates some of the recent archaeological and architectural findings. For the Cairene architectural legacy, see Creswell, Muslim Architecture of Egypt, which stops at 1311; a recent and more exhaustive survey of Mamluk architecture is Meinecke, Mamlukische Architektur. For the history of this institution, see Jean-Édouard Goby, ‘La composition du premier Institut d’Égypte’, Bulletin de l’Institut d’Égypte 29 (1946–7): 343–67 and 30 (1947–8), 81–100; also Gabriel Guémard, ‘Éssai d’histoire de l’Institut d’Égypte’, Bulletin de l’Institut d’Égypte 6 (1924): 43–84. Mémoires sur l’Égypte publiés pendant les campagnes du Général Bonaparte dans les années VI et VII (Paris, 1801), 1; Louca, ‘Renaissance égyptienne’, 1–3. There exist two editions of the Description de l’Égypte: the first is the édition Imprimerie Nationale (1809–28), the second is the édition dite Panckoucke, 26 octavo vols of text and 12 folio vols of illustrations (Paris, 1821–9). For a review of the history of this major work, see the introduction in Monuments of Egypt: The Napoleonic Edition, ed. Gillispie and Dewachter, 1–29. Jean-Marie Carré, Voyageurs et écrivains français en Égypte, 2 vols (Cairo, 1956), 1: 152–62, relates the story of the first French encounters with the Pharaonic marvels of Upper Egypt during the summer and autumn of 1799 and the realisation of the French savants that these monuments must have been the inspiration behind the later Greek monuments. There has been a resurgence in the interest in Coste after a long period of neglect. A partial list of recent publications is Armogathe and Leprun, Pascal Coste ou l’architecture cosmopolite, which comprises twelve essays on Coste’s life and work; Paule Guiral, ‘Un architecte français en Afrique du Nord dans la première moitié du XIXe siècle’, Revue de l’Occident Musulman et de la Mediterranée 6 (1969): 103–12; idem, ‘L’architecte marseillais Pascal Coste et son oeuvre en Égypte sous Mehmet Ali’, Provence Historique 19 (1976): 133–42; Denise Jasmin, ‘Le travail d’un architecte marseillais en Orient’, Perspectives Mediterranéennes 14 (June 1983): 3–9; Sylviane Leprun, ‘Rives, dérives ethno-architecturales: Pascal-Xavier Coste, architecte anthropologue’, Les cahiers de la recherche architecturale 27/28 (1992): 209–18. The two mosque projects are in Coste’s sketchbook at the Bibliothèque municipale de Marseille; their existence was known, but not much attention has been given to them. They are introduced and discussed in Kara Marietta Hill, ‘Pascal-Xavier Coste (1787–1879): A French Architect in Egypt’ (1992), vol. 1, 144–8. Hill also includes a set of photos from Coste’s archives of the citadel mosque (2: 319–20, figs. 27–9) and the Alexandria mosque (2: 320–22, figs. 30–33), which I depended upon for this book. See also André Raymond, ‘Muhammad ‘Ali et Pascal Coste’, in Armogathe and Leprun, Pascal Coste, 29–31; a section of the citadel mosque is published on page 33. The façade of the Alexandria mosque has been published in Denise Jasmin, ‘Pascal Coste et l’Égypte’, Monuments Historiques 75 (1983): 29; also Les Orientalistes Provençeaux, a catalogue of an exposition held at Marseilles: Musée des Beaux-Arts, Novembre 1982–Février 1983, fig. 66, 47. Carré, Voyageurs français, 1: 287, calls it the ‘most rigorous and most complete documentation of Islamic architecture in Egypt’. See the analysis of Coste’s drawing methods in Claude Jasmin, ‘Le dessin de Coste’, in Armogathe and Leprun, Pascal Coste, 133–41. –235 –

N O T E S T O P A G E S 17 7 – 1 8 3

17. Coste, Mémoires d’un artiste: Notes et souvenirs de voyages 1817–1877 (Marseilles, 1878), 43. 18. Raymond, ‘Muhammad ‘Ali et Pascal Coste’, 30. 19. A sequence of quotations on the development of the construction of the Muhammad ‘Ali Mosque from Coste’s design to the final stages is collected in Wiet, Mohammed Ali et les beaux-arts, 265–88. 20. Robert Curzon, Visits to the Monasteries in the Levant (London, 1851), 41. 21. Hautecoeur and Wiet, Les mosquées du Caire, 1: 344, report that his name was Yusuf Bushnaq. ‘Abd al-Wahab, Tarikh al-Masajid al-Athariyya, 378, lists the names of many Egyptian draftsmen and aides to the architect and an Anatolian designer who did the window grilles, but says that in his thorough search in the archives he was unable to find the name of the architect. 22. Muhammad al-Asad, ‘The Mosque of Muhammad ‘Ali in Cairo’, Muqarnas 9 (1992): 48–9, n. 26, questions the attribution of the mosque to Yusuf Bushnaq, who is otherwise unknown, and points out Wiet’s mistake in translating Robert Curzon’s account where ‘Armenian architecture’, in the original became ‘Armenian architect’ in the French translation. 23. Hill, ‘Pascal-Xavier Coste’, 1: 151. 24. Al-Asad, ‘Mosque of Muhammad ‘Ali’, 39–55, discusses the mosque’s architecture. 25. A succinct review of Muhammad ‘Ali’s educational endeavours in English is in Sophie Rentz, ‘The Effect of the Introduction of the Printing Press on Learning in Egypt in the Early Nineteenth Century’, Master’s Thesis, San Jose State College, 1972, 121–211. 26. For a concise review of Muhammad ‘Ali’s career, see Afaf Lutfi al-Sayyid Marsot, A Short History of Modern Egypt (Cambridge, 1986), 51–65; for Muhammad ‘Ali and the Ottomans, see idem, Reign of Muhammad ‘Ali, 196–248. 27. Ibid., 263, calls Muhammad ‘Ali’s regime a Turco-Circassian-Albanian raj in Egypt similar to the British raj in India. 28. Of the 243 books printed at Bulaq between 1822 and 1842, none is an Arabic history book. Instead, Ottoman history, the biographies of great Ottoman sultans and Persian and Turkish classical poetry make up a considerable proportion of the non-pedagogical output. See the list in ThomasXavier Bianchi, ‘Catalogue général des livres arabes, persans, et turcs imprimés à Boulac en Égypte depuis l’introduction de l’imprimerie dans ce pays’, Journal Asiatique series, 2 (July–August 1843): 24–61. 29. This last goal is exemplified well by the rehablitation of the Citadel of Cairo as his seat of government, which was undertaken during the 1820s; see Nasser Rabbat, The Citadel of Cairo (Geneva, 1989), 25–31. 30. Hill, ‘Pascal-Xavier Coste’, 1: 48–53, weighs all the possible reasons for the change of plans for the Muhammad ‘Ali Mosque. 31. Vatikiotis, History of Modern Egypt, 67–77. 32. Gabriel Baer, ‘Social Change in Egypt: 1800–1914’, in Political and Social Change in Modern Egypt, ed. Holt, 135–61. 33. A sound review of Isma‘il’s personality and his fascination with Europe can be found in Elias alAyyubi’s Tarikh Misr fi ‘Ahd al-Khidiwi Isma‘il Pasha, 2 vols, 2nd edn (Cairo, 1994), 1: 258–99; a somewhat dramatised and caricatured one can be found in Trevor Mostyn, Egypt’s Belle Epoque: Cairo 1869–1952 (London, 1989), 42–124. 34. For an assessment of Isma‘il’s reign, see Vatikiotis, History of Modern Egypt, 77–89. 35. For Isma‘il’s urban projects, see Mubarak, Khitat, 1: 210–17; Abu-Lughod, Cairo, 98–117; Raymond, Le Caire, 306–15. 36. An account of the mosque’s early construction history before Herz can be found in Mubarak, Khitat, 4: 237–45; for an architectural evaluation of the different stages of the building by its last architect see Max Herz, La Mosquée el-Rifaï au Caire (Milan, 1912); the most recent study of the mosque is al-Asad, ‘Mosque of al-Rifa‘i’, 108–24. 37. For a short biography of Hussein Pasha Fahmi, see Abdul Rahman al-Rafi‘i, ‘Asr Muhammad ‘Ali, 4th edn (Cairo, 1982), 419, 484.

–236 –

N O T E S T O PAG E S 18 3 –18 7

38. Herz, Mosquée el-Rifaï, 22, reports that Fahmi boasted of his borrowing from medieval precedents, but he immediately faults him for not having studied them more carefully, probably an indirect reference to his own knowledge of Mamluk architecture. 39. For the ten copies of the Monuments du Kaire, see Hill, ‘Pascal-Xavier Coste’, 1: 226. Coste (known in Egyptian documents as khawaga Kosti), in fact, was not unknown at the Muhandiskhaneh; he was appointed there in September 1820 to teach calculus and drafting, see Rentz, ‘Printing Press’, 124–5; al-Rafi‘i, ‘Asr Muhammad ‘Ali, 399. 40. Mubarak, Khitat, 4: 239; Herz, Mosquée el-Rifaï, 21, says that the European architect in question may have been an Italian by the name of Gaï, whom he had met at the Rifa‘i Mosque thirty years before the date of his writing – that is, around 1880. 41. Mubarak, Khitat, 4: 240. 42. Max Herz, La Mosquée du Sultan Hassan au Caire (Cairo, 1899). 43. Not only did Hertz work on the conservation of Mamluk monuments in Cairo, but he also designed a huge number of structures in the ‘Arab’ style – another variation on the style’s name – including his own house in Qasr al-Doubara, which no longer exists, see Tawfiq Iskarus, ‘Maks Hirts Basha’, al-Hilal 27, 10 (1 July 1919): 921–8. 44. Herz, La Mosquée el-Rifaï, 13–18. 45. For the Sayyida Zaynab and Sayyida Nafisa mosques, see the Ministry of Waqfs, The Mosques of Egypt, 2: 122–3, figs. 180–81 for the first and 124, figs. 184–5 for the second; the architects of both are unknown. For the Ministry of Awqaf, see Sakr, Islamic Architecture in Cairo, 19–21; for the Dar al-Kutub, ibid., 22–4. 46. Buildings such as the Mosque of Abu al-‘Abbas al-Mursi in Alexandria (built by the Italian Mario Rossi) were intentionally made in the neo-Mamluk style; see James Dickie, ‘Modern Islamic Architecture in Alexandria’, Islamic Quarterly 12 (1970): 183–91. Al-Asad, ‘The Mosque of al-Rifa‘i’, 124, n. 37, mentions the rise of another national style in Egypt, the Pharaonic Revival style, that competed for predominance with the neo-Mamluk style in the 1920s and 1930s. But its popularity was politically induced and did not survive the demise of the political ideology that promoted it; for further discussion, see Mercedes Volait, L’architecture moderne en Égypte et la revue al-‘imara (1939–1959) (Cairo, 1988), 43–50.

–237–

BIBLIOGR APHY ‘Abd al-Fattah, M. Hussam al-Din, ‘Mantiqat al-Darb al-Ahmar’, Master’s Thesis (Asyut University, 1986). ——, ’Arba‘ Buyut Mamlukiyya min al-Watha’iq al-‘Uthmaniyya’, Annales Islamologiques 24 (1988): 49–102. ‘Abd al-Raziq, Ahmad, La femme au temps des Mamlouks en Égypte (Cairo, 1973). ——, ‘La hisba et le muhtasib en Égypte au temps des Mamluks’, Annales Islamologiques 13 (1977): 115–78, 148–53. ——, ‘al-Mamalik wa mafhum al-usra ladayhim’, Majallat Kulliyat al-Athar 2 (1987): 188–207. ‘Abd, al-Razzaq, Ahmad, Ahmad, Tarikh wa Athar Misr al-Islamiyya (Cairo, 1977; reprint 1993). ‘Abd al-Wahab, Hasan, Tarikh al-Masajid al-Athariyya allati salla fiha faridhat al-jum‘a hadret sahib al-jalala al-malik al-salih Farouq al-Awwal (Cairo, 1946). ‘Ali, Mustafa, Mustafa ‘Ali’s Description of Cairo of 1599, trans. and ed. A. Tietze (Vienna, 1975). ‘Inan, Muhammad ‘Abdallah, Tarikh al-Jami‘ al-Azhar fi al-‘Asr al-Fatimi (Cairo, 1942). ——, Misr al-Islamiyya wa Tarikh al-Khitat al-Misriyya (Cairo: 1969). Abu Shama, al-Dhayl ‘ala-al-Rawdatin, ed. M. Zahed al-Kawthari (Cairo, 1947). ——, Kitab al-Rawdatayn fi-Akhbar al-Dawlatayn al-Nuriyya wa-l-Salahiyya, ed. M. H. M. Ahmad, 2 vols (Cairo, 1956–1962). Abu-Lughod, Janet, Cairo, 1001 Years of the City Victorious (Princeton, 1971). ——, Before European Hegemony: The World System A.D. 1250–1350 (New York, 1989). Afaf Lutfi al-Sayyid Marsot, ‘The Role of the Ulama in Egypt during the Early Nineteenth Century’, in P. M. Holt, ed., Political and Social Change in Modern Egypt (London, 1968), 264–80. Ahmed, Jamal Mohammed, The Intellectual Origins of Egyptian Nationalism (Oxford, 1960). Al-‘Attar, Hasan, Hashiyat al-‘Attar ala Jami‘ al-Jawami‘ (Beirut, 1983). Al-’Ayni, ‘Iqd al-Juman fi Tarikh Ahl al-Zaman, vol. 5, ed. M. M. Amin (Cairo, 1989). Al-‘Ulaymi, Mujir al-Din, al-Uns al-Jalil bi Tarikh al-Quds wa-l-Khalil, ed. M. Bahr al-‘Ulum (Najaf, Iraq, 1968). Al-Asad, Muhammad, ‘The Mosque of Muhammad ‘Ali in Cairo’, Muqarnas 9 (1992): 39–55. ——, ‘The Mosque of al-Rifa‘i in Cairo’, Muqarnas 10 (1993): 108–24. Al-Ayyubi, Elias, Tarikh Misr fi ‘Ahd al-Khidiwi Isma‘il Pasha, 2nd edn, 2 vols (Cairo, 1994). Al-Badari, Abu al-Baqa’, Nuzhat al-Anam fi-Mahasin al-Sham (Beirut, 1980). Al-Baghdadi, ‘Abd al-Latif, al-Ifada wa-l-I‘tibar fi al-Umur al-Mushahada wa-l-Hawadith al-Mu‘ayana bi-Ard Misr (Cairo, 1869). Al-Baha’i al-Ghuzuli, Matali‘ al-Budur fi Manazil al-Surur (Cairo, 1882). Al-Basha, Hasan, al-Taswir al-Islami fi-l-‘Usur al-Wusta (Cairo, 1959). Al-Bindari, Sana al-Barq al-Shami, ed. Fathia al-Nabrawi (Cairo, 1979). Al-Damardashi, Ahmad, Kitab al-Durra al-Musana, ed. ‘Abd al-Rahim A. ‘Abd al-Rahim (Cairo, 1989). Al-Daqqaq, Omar, ‘Manzilat al-‘adad saba‘a fi al-fikr al-‘arabi’, ‘Adiyat Halab 1 (1975): 53-108. Al-Dimashqi, Sheikh al-Rabwa, Nukhbat al-Dahr fi ‘Aja’ib al-Barr wa-l-Bahr, ed. M. A. F. Mehren (Leipzig, 1923). Al-Fallal, A.A., Dirasa Tahlilyia li-Shi‘r Muhtar al-Daylami (Cairo, n.d.). Al-Farra’, Abu Ya‘la, al-Ahkam al-Sultaniyya, ed. M. H. al-Fiqqi (Cairo, 1966). –238 –

BIBLIOGR A PH Y

Al-Ghazzi, Kamel, Nahr al-Zahab fi-Tarikh Halab (Aleppo, 1920). Al-Hadidi M., and F. ‘Abd al-‘Alim, ‘‘A‘mal Tarmim al-Qasr al-Ablaq bi-Qal‘at Salah al-Din’, ‘Alam al-Athar 26 (April 1986): 4–16. Al-Hajji, Hayat, The Internal Affairs in Egypt during the Third Reign of Sultan al-Nasir Muhammad B. Qalawun (Kuwait, 1978). Al-Hamawi, Yaqut, Mu‘jam al-Buldan, ed. Ferdinand Wüstenfeld (Leipzig, 1866-70). Al-Harithy, Howyda N., ‘Urban Form and Meaning in Bahri Mamluk Architecture’, Ph.D. diss., Harvard University, 1992. Al-Istakhri, al-Masalik wa-l-Mamalik, ed. M.J. al-Hini (Cairo, 1961). Al-Jabarti, ‘Abd al-Rahman, ‘Aja‘ib al-Athar fi-l-Tarajim wa-l-Akhbar, 3 vols (Beirut, n.d.). Al-Jahiz, Rasa’il al-Jahiz, ed. ‘Abd al-Salam Muhammad Harun, 2 vols (Cairo, 1965). Al-Jawhari al-Sayrafi, Inba’ al-Hasr bi-Abna’ al-‘Asr, ed. Hasan Habashi (Cairo, 1970). ——, Nuzhat al-Nufus wa-l-Abdan fi Tawarikh al-Zaman, ed. Hasan Habashi, 4 vols (Cairo, 1970–94). Al-Jazari, Badi‘ al-Zaman, Kitab fi Ma‘rifat al-Hiyal al-Handasiyya, trans. Donald Hill (The Book of Knowledge of Ingenious Mechanical Devices) (Dordrecht, Holland, 1974). Al-Jazari, Jawahir al-Suluk fi al-Khulafa’ wa-l-Muluk, MSS 7575 H, Dar al-Kutub, Cairo. Al-Khatib al-Baghdadi, Tarikh Baghdad, 14 vols (Beirut, n.d.). Allan, J. W., ‘Mamluk Sultanic Heraldry and the Numismatic Evidence: A Reinterpretation’, Journal of the Royal Asiatic Society 2 (1970): 19–112. ——, ‘Muhammad ibn al-Zain: Craftsman in Cups, Thrones and Window Grilles?’ Levant 28 (1996): 199–207. Allen, Terry, Five Essays in Islamic Architecture (Sebastopol, CA, 1988). Allouche, Adel, ‘Teguder’s Ultimatum to Qalawun’, International Journal of Middle East Studies 22/4 (November 1990): 437–46. ——, Mamluk Economics: A Study and Translation of al-Maqrizi’s Ighathat al-Ummah bi-Kashf al-Ghummah (Salt Lake City, 1994). Al-Makin Jurjis ibn al-‘Amid, ‘Akhbar al-Ayyubiyyin’, ed. Claude Cahen, Bulletin d’ Études Orientales 15 (1955–57): 109–84. Al-Maqrizi, al-Mawa‘iz wa-l-I‘tibar bi-Dhikr al-Khitat wa-l-Athar, 2 vols (Bulaq, 1854). ——, al-Suluk li-Ma‘rifat Duwal al-Muluk, 4 vols, ed. M. Zyada et al. (Cairo, 1934–72). ——, Itti‘az al-Hunafa bi-Akhbar al-A’imma al-Fatimiyyin al-Khulafa, ed. M. H. M. Ahmad et al., 3 vols (Cairo, 1967–73). ——, Durar al-‘Uqud al-Farida fi Tarajim al A‘yan al-Mufida, ed. A. Darwish and M. al-Masri, 2 vols (Damascus, 1995). Al-Mawardi, Abu al-Hasan, al-Ahkam al-Sultaniyya (Cairo, 1909). Al-Munufi, Muhammd ‘Abd al-Mu‘ti al-Ishaqi, Akhbar al-Uwal fi man Tasarrafa fi Misr min Arbab al-Duwal (Cairo, 1890). Al-Muqqadasi, Ahsan al-Taqasim fi-Ma‘rifat al-Aqalim, ed. M. J. de Goeje (Leiden, 1906). Al-Nuwayri, Shihab al-Din, Nihayat al-Arab fi Funun al-Adab, 33 vols, various eds (Cairo, 1923–93). Al-Qalqashandi, Subh al-A‘sha fi Sina‘at al-Insha, 14 vols (Cairo, 1913–18). Al-Qazwini, Zakariyya, Kitab Athar al-Bilad wa-Akhbar al-‘Ibad, ed. Ferdinand Wüstenfeld (Göttingen, 1848). ——, ‘Aja’ib al-Makhluqat wa Ghara’ib al-Mawjudat, ed. F. Sa‘d (Beirut, 1973). Al-Qudsi, Abu Hamid, al-Fawa’id al-Nafissa al-Bahira fi Bayan Hukm Shawari‘ al-Qahira fi Madhahib al-A’imma al-Arba‘a al-Zahira, ed. Amal al-’Umari (Cairo, 1988). Al-Rafi‘i, Abdul Rahman, ‘Asr Muhammad ‘Ali, 4th edn (Cairo, 1982). Al-Rihawi, ‘Abd al-Qadir, ‘Fusayfusa’ al-Jami‘ al-Amwi’, Les annales archéologiques arabes syriennes 10 (1960): 37–47. ——, ‘Qusur al-Hukkam fi-Dimashq’, Les annales archéologiques arabes syriennes, 22–23 (1972–73), 22: 42–48. ——, Qal‘at Dimashq (Damascus, 1980). Al-Safadi, Nakt al-Himyan fi-Nukat al-‘Umyan, ed. A. Zaki (Cairo, 1911). –239 –

BIBLIOGR A PH Y

——, al-Wafi bi-l-Wafi yyat, 29 vols, various eds (Wiesbaden, 1931–99). Al-Sarraf, Shihab, ‘L’impact des techniques militaires sur l’évolution politique et sociale dans le moyenorient médiéval: le cas de l’archerie’, Études Orientales-Dirasat Sharqiya 7/8 (1990): 6–28. Al-Sayyid, Radwan, ‘Qada’ al-Mazalim: Nazra fi Wajh min Wujuh ‘Alaqat al-Din bi-l-Dawla fi-l-Tarikh al-Islami’, Dirasart 14/10 (1987): 157–64. ——, al-Umma wa-l-Jama‘a wa-l-Sulta: Dirasat fi-l-Fikr al-Siyasi al-‘Arabi al-Islami (Beirut, 1984). Al-Shinawwi, ‘Abdulaziz M., Suwar min Dawr al-Azhar fi Muqawamat al-Ihtilal al-Faransi li Misr (Cairo, 1971). Al-Shuja‘i, Shams al-Din, Tarikh al-Malik al-Nasir Muhammad ibn Qalawun al-Salihi wa-Awladahu, ed. Barbara Schäfer (Wiesbaden, 1978). Al-Subki, Taj al-Din, Tabaqat al-Shafi‘iyya al-Kubra, ed. M. al-Tanahi and ‘A.-F. al-Huluw, 10 vols (Cairo, 1966–76). ——, Mu‘id al-Ni‘am wa Mubid al-Niqam, ed. M. A. al-Najjar, A. Z. Shalabi, and M. A. al-‘Uyun (Cairo, 1993). Al-Suyuti, Husn al-Muhadara fi Tarikh Misr wa-l-Qahira (Cairo, 1968). Al-Udfawi, al-Tali‘ al-Sa‘id, ed. S. M. Hasan (Cairo, 1966). Al-Yunini, Dhayl Mir’at al-Zaman, 4 vols (Heyderabad, 1954–61). Al-Yusufi, Nuzhat al-Nazir fi Sirat al-Malik al-Nasir, ed. Ahmad Hutait (Beirut, 1986). Amari, Micehele, ‘Questions philosophiques adressées aux savants musulmans par l’Empereur Frédéric II’, Journal Asiatique 5e serie, 1 (1853): 240–74. Amin, M. M., al-Awqaf wa-l-Hayyat al-Ijtima‘iyya fi-Misr. 684–923 A.H./1250–1517 A.D. (Cairo, 1980). ——, Watha’iq Waqf al-Sultan al-Malik al-Nasir Hasan (Cairo, 1986). Amitai-Preiss, Reuven, Mongols and Mamluks: the Mamluk-Ilkhanid War, 1260–1281 (Cambridge, 1995). Arkoun, Mohammed, Lectures du Coran (Paris, 1982). —— et al., L’étrange et le merveilleux dans l’islam médiéval (Paris, 1978). Armogathe, Daniel and Sylviane Leprun, Pascal Coste ou l’architecture cosmopolite (Paris, 1990). Arnold, Thomas, Painting in Islam: A Study of the Place of Pictorial Art in Muslim Culture (Oxford, 1928). Ayalon, David, ‘L’esclavage du Mamelouk’, Oriental Notes and Studies 1 (1951): 1–66. ——, ‘The Wafidiyya in the Mamluk Kingdom’, Islamic Culture 25 (1951): 81–104. ——, ‘Studies on the Structure of the Mamluk Army’, Bulletin of the School of Oriental and African Studies 15/3 (1953): 448–59. ——, ‘The System of Payment in Mamluk Military Society’, Journal of the Economic and Social History of the Orient 1(1958): 37–65, 257–96. ——, ‘The European-Asiatic Steppe: A Major Reservoir of Power for the Islamic World’, Proceedings of the 25th Congress of Orientalists, vol. 2 (Moscow, 1963). ——, ‘The Muslim City and the Mamluk Military Aristocracy’, Proceedings of the Israel Academy of Sciences and Humanities (1968): 311–29. ——, ‘The Great Yasa of Chingiz Khan. A Reexamination’, Studia Islamica 36 (1972): 113–58 (part C1) ‘The Position of the Yasa in the Mamluk Sultanate’, and 38 (1973): 107–56 (part C2) ‘Al-Maqrizi’s Passage on the Yasa under the Mamluks.’ ——, ‘Aspects of the Mamluk Phenomenon’, Der Islam 53/2 (1976): 196–225. ——, ‘On the Eunuchs in Islam’, Jerusalem Studies in Arabic and Islam 1(1979): 67–124. ——, ‘Mamlukiyyat: (B) Ibn Khaldun’s View of the Mamluk Phenomenon’, Jerusalem Studies in Arabic and Islam 2 (1980): 340–49. ——, ‘From Ayyubids to Mamluks’, Revue des Études Islamiques 49 (1981): 43–57. ——, ‘The Expansion and Decline of Cairo under the Mamluks and Its Background’, in Itineraires d’Orient: hommages à Claude Cahen, ed. Raoul Curiel and Rika Gyselen (Paris, 1994), 14–16. ——, ‘Mamluks of the Seljuks: Islam’s Military Might at the Crossroads’, Journal of the Royal Asiatic Society Series 3, 6/3 (1996): 305–33.

– 24 0 –

BIBLIOGR A PH Y

Azmy, Ahmed Abdul Wahab, ‘University Tradition and Continuity in Architecture as a Stimulation for the Future Development of al-Azhar University and Old Cairo’, Ph.D., Princeton University, 1966. Bacharach, Jere L., ‘The Court-Citadel: An Islamic Urban Symbol of Power’, in The Proceedings of International Conference on Urbanism in Islam (ICUIT), ed. T. Yukawa (Tokyo, 1989), 206–45. Baer, Eva, ‘The Ruler in Cosmic Setting: A Note on Medieval Islamic Iconography’, in Essays in Islamic Art and Architecture in Honor of Katharina Otto-Dorn, ed. Abbas Daneshvari (Malibu, CA, 1981), 13–19. Bartol’d, Vasilii Vladimirovich, Turkestan Down to the Mongol Invasion, ed. and trans. T. Minorsky, with addenda and corrigenda by C. E. Bosworth (London, 1968). Baybars al-Mansuri, al-Tuhfa al-Mulukiyya fi al-Dawla al-Turkiyya, ed. A. H. S. Hamdan (Cairo, 1987). Behrens-Abouseif, Doris, ‘The Qubba, an Aristocratic Type of Zawiya.’ Annales Islamologiques 19 (1983): 1–7. ——, The Minarets of Cairo (Cairo, 1985). ——, ‘The Baptistère de Saint Louis: A Reinterpretation’, Islamic Art 3 (1988–89): 3–13. ——, ‘The Citadel of Cairo: Stage for Mamluk Ceremonial’, Annales Islamologiques 24 (1988): 25–79. ——, ‘The ‘Abd al-Rahman Katkhuda Style in 18th-Century Cairo’, Annales Islamologiques 26 (1992): 117–26. ——, Egypt’s Adjustment to Ottoman Rule: Institutions, Waqf and Architecture in Cairo (16th and 17th Centuries) (Leiden, 1994). ——, ‘Muhandis, Shad, Mu‘allim – Note on the Building Craft in the Mamluk Period’, Der Islam 72 (1995): 293–309. Bernal, Martin, Black Athena: The Afroasiatic Roots of Classical Civilization. Vol. 1: The Fabrication of Ancient Greece 1785–1985 (New Brunswick, NJ, 1987). Berque, Jacques, Egypt: Imperialsim and Revolution, trans. J. Stewart (London, 1972). Bertsch, Margaret E., ‘Counter-crusade: a study of twelfth century Jihad in Syria and Palestine’, Ph.D. Thesis, University of Michigan, 1993. Bianchi, Thomas-Xavier, ‘Catalogue général des livres arabes, persans, et turcs imprimés à Boulac en Égypte depuis l’introduction de l’imprimerie dans ce pays’, Journal Asiatique 4 series, 2 (July–Aug 1843): 24–61. Bierman, Irene, Writing Signs: The Fatimid Public Text (Berkeley, 1998). Bloom, Jonathan, ‘Meaning in Early Fatimid Architecture: Islamic Art in North Africa and Egypt in the 4th c. H/10th c. A.D’, PhD Thesis, Harvard University, 1980. ——, ‘The Mosque of Baybars al-Bunduqdari in Cairo’ Annales Islamologiques 18 (1982): 50–55. ——, ‘The Mosque of al-Hakim in Cairo’, Muqarnas 1 (1983): 24–26. ——, Minaret: Symbol of Islam (Oxford, 1989). ——, ‘Mamluk Art and Architectural History: A Review Article’, Mamluk Studies Review 3 (1999): 31–58. Bohas, Georges and Jean-Patrick Guillaume, trans. Sirat Baybars (Roman de Baîbars) Les enfances de Baîbars (Paris, 1985). Bozdogan, Sibel, ‘Journey to the East: Ways of Looking at the Orient and the Question of Representation’, Journal of Architectural Education 41/4 (Summer 1988): 38–45. Bregel, Yuri, The Role of Central Asia in the History of the Muslim East (New York, 1980). Briggs, Martin S., Muhammadan Architecture in Egypt and Palestine (New York, 1924; reprint 1974). Brinner, W. M., ‘Dar al-Saa‘da and Dar al-‘Adl in Mamluk Damascus’, in Studies in Memory of Gaston Wiet, ed. Myriam Rosen-Ayalon (Jerusalem, 1977), 235–47. Burgat, Marie-Claude, ed., D’un Orient l’autre: les metamorphoses successives des perceptions et connaissances 2 vols (Paris, 1991). Burgoyne, Michael, Mamluk Jerusalem: An Architectural Study (London, 1987). Bynum, Caroline W., ‘Wonder’, American Historical Review (February 1997): 1–26. Cahen, Claude, ‘La première pénétration turque en Asie-Mineure’, Byzantion 18 (1946–48): 5–67.

– 2 41 –

BIBLIOGR A PH Y

——, Orient et Occident au temps des Croisades (Paris, 1983). Canard, Marius, ‘Quelques observations sur l’intorduction géographique de Bughyat al-talab de Kamal al-Din Ibn al-‘Adim’, Annales de l’Institut d’Études Orientales de la Faculté des Lettres d’Alger 15 (1957): 41–53. Caronia, Giuseppe and Vittorio Noto, La Cuba di Palermo (Arabi e Normanni nel XII Secolo) (Palermo, 1988). Carré, Jean-Marie, Voyageurs et écrivains français en Égypte, 2 vols (Cairo, 1956). Casanova, Paul, Histoire et description de la citadelle du Caire (Cairo, 1894), published in Arabic as Tarikh wa-Wasf Qal‘at al-Qahira, trans. Ahmad Darraj (Cairo, 1974). Çelik, Zeynep, Displaying the Orient: Architecture of Islam at Nineteenth-Century World’s Fairs (Berkeley, 1992). Chalabi, Ahmad, Awdah al-Isharat fi-man Tawala Misr al-Qahira min-l-Wuzara’ wa-l-Bashat, ed. A. R. ‘AbdelRahim (Cairo, 1978). Chamberlain, Michael, Knowledge and Social Practice in Medieval Damascus, 1190–1350 (New York, 1994). Chéhab, Maurice, ‘The Umayyad Palace at ‘Anjar’, Ars Orientalis 5 (1963): 17–25. Chevedden, Paul E., ‘The Citadel of Damascus’, Ph.D. Thesis, UCLA, 1986. Choay, Françoise, ‘Alberti, the Invention of Monumentality and Memory’, Harvard Architectural Review 4 (Spring, 1984): 99–105. Cizakca, Murat, A History of Philanthropic Foundations: The Islamic World from the Seventh Century to the Present (Istanbul, 2000). Clifford, W. W., ‘Ubi Sumus? Mamluk History and Social Theory’, Mamluk Studies Review 1 (1997): 45–62. Coste, Pascal-Xavier, Architecture Arabe ou Monuments du Kaire mésurés et dessinés de 1818 à 1826 (Paris, 1839). ——, Mémoires d’un artiste: Notes et souvenirs de voyages 1817–1877 (Marseilles, 1878). Crabbs, Jr., Jack A., The Writing of History in Nineteenth-Century Egypt (Detroit, 1984). Crane, Howard, ed. and trans., Risale-i Mi‘mariyye: An Early Seventeenth Century Ottoman Treatise on Architecture (Leiden, 1987). Creswell, K. A. C., Early Muslim Architecture, 2 vols (Oxford, 1932–40). ——, ‘ The Lawfulness of Painting in Early Islam’, Ars Islamica 11–12 (1946): 159–67. ——, Muslim Architecture of Egypt, 2 vols (Oxford, 1959). ——, A Short Account of Early Muslim Architecture, ed. James W. Allan (Cairo, 1989). Curzon, Robert, Visits to the Monasteries in the Levant (London, 1851). Dajani-Shakeel, Hadia, ‘Jihad in Twelfth-Century Arabic Poetry: A Moral and Religious Force to Counter the Crusades’, The Muslim World 66/2 (April 1976): 96–113. ——, al-Qadi al-Fadil ‘Abd al-Rahim al-Baysani al-‘Asqalani: Dawruhu al-Takhtiti fi Dawlat Salah al-Din wa Futuhatuhu (Beirut, 1993). Darby, Michael, Islamic Perspective: An Aspect of British Architecture and Design in the 19th Century (London, 1983). Darrag, Ahmad, L’Égypte sous le règne de Barsbay 825–841/1422–1438 (Damascus, 1961). ——, L’acte de waqf de Barsbay (Cairo, 1963). De Lorey, Eustache, ‘Les mosaïques de la mosquée des omayyades à Damas’, Syria 12 (1931): 326–49. Deguilhem, Randi, and André Raymond, Le Waqf dans l’espace islamique: Outil de pouvoir socio-politique (Damascus, 1995). Description de l’Égypte, édition Panckoucke (Paris, 1821–29). Dickie, James, ‘Modern Islamic Architecture in Alexandria’, Islamic Quarterly 12 (1970): 183–91. Dodd, Erica C., ‘The Image of the Word (Notes on the Religious Iconography of Islam)’, Berytus 18 (1969): 35–62. ——, and Shereen Khairallah, The Image of the Word: A Study of Quranic Verses in Islamic Architecture, 2 vols (Beirut, 1981). Driault, Edouard, ‘La Renaissance de l’Égypte’, Napoléon, revue des études napoléoniennes 14/1 (Jan–Feb. 1925): 5–22. – 242 –

BIBLIOGR A PH Y

Duggan, Lawrence, ‘Was Art Really the Book of the Illiterate?’ Word and Image (1989): 227–51. Duhman, Muhammad Ahmad, Fi Rihab Dimashq (Damascus, 1982). Élisséeff, Nikita, ‘Les monuments de Nur al-Din: inventaire, notes archéologiques et bibliographiques’, Bulletin d’ Études Orientales 13 (1949–51): 5–43. ——, ‘La titulature de Nur al-Din d’après ses inscriptions’, Bulletin d’ Études Orientales 14 (Damascus, 1952–54): 155–96. ——, La déscription de Damas d’Ibn ‘Asakir (Damascus, 1959). ——, Nur ad-Din, un grand prince musulman de Syrie au temps des Croisades (511–569/1118–1174), 3 vols (Damascus, 1967). Escovitz, J. S., ‘The Establishment of Four Chief Judgeships in the Mamluk Empire’, Journal of the American Oriental Society 102 (1982): 529–31. Esin, Emel, ‘The Turk al-Agam of Samarra and the Paintings Attributable to them in the Gawsaq al-Haqani’, Kunst des Orients 7/2 (1972): 47–88. ——, ‘L’arme zomorphe du guerrier turc (Étude iconographique)’, in Sprache, Geschichte und Kultur der Altaischen Völker. Protokollband der Permanent International Altaistic Conference 1969 in Berlin, ed. G. Hazai and P. Zieme (Berlin, 1974), 193–217. Ettinghausen, Richard, ‘Painting in the Fatimid Period: A Reconstruction’, Ars Islamica 9 (1942): 112–13. ——, ‘Early Realism in Islamic Art’, in Studi Orientalistici in onore di Giorgio Levi Della Vida (Rome, 1956), vol. 1: 250–73. ——, From Byzantium to Sasanian Iran and the Islamic World (Leiden, 1972). ——, Arab Painting (Geneva, 1977). ——, Islamic Art and Archaeology: Collected Papers, ed. M. Rosen-Ayalon (Berlin, 1984). Evliya Çelebi, Seyahatnamesi, ed. Mümin Çevik, 10 vols (Istanbul, 1984). Farés, Bishr, Sirr al-Zakhrafa al-Islamiyya (Cairo, 1952). Fernandes, Leonor, ‘Notes on a New Source for the Study of Religious Architecture during the Mamluk Period: the Waqfiya’, al-Abhath 33 (1985): 3–12. Fikri, Ahmad, Masajid al-Qahira wa Madarisaha: al-‘Asr al-Fatimi (Cairo, 1965). Finster, Barbara, ‘Die Mosaiken der Umayyadenmoschee von Damaskus’, Kunst des Orients 7/2 (1972): 83–136. Flemming, Barbara, ‘Literary Activities in Mamluk Halls and Barracks’, in Myriam Rosen-Ayalon, ed., Studies in Memory of Gaston Wiet (Jerusalem, 1977), 249–60. Gabr, Aly, ‘The Traditional Process of the Production of Medieval Muslim Art and Architecture: With Special Reference to the Mamluk Period’, Edinburgh Architectural Research 20 (1993): 133–59. Galal, Muhammad Nu‘man and Magdi Mutawalli, Hawiyyat Misr (Cairo, 1997). Garcin, Jean-Claude, ‘Le Sultan et Pharaon (Le politique et le religieux dans l’Égypte mamluke)’, in Hommages à François Daumas (Montpellier, 1986), 261–72. Gierlichs, Joachim, Drache, Phonix, Doppeladler: Fabelwesen in der islamischen Kunst (Berlin, 1993). ——, Mittelalterliche Tierreliefs in Anatolien und Nordmesopotamien: Untersuchungen zur figurlichen Baudekoration der Seldschuken, Artuqiden und ihrer Nachfolger bis ins 15. Jahrhundert (Tubingen, 1996). Gillispie, Charles Coulston and Michael Dewachter, eds, Monuments of Egypt: the Napoleonic Edition: the complete archaeological plates from la Description de l’Égypte (Princeton, 1987). Goby, Jean-Édouard, ‘La composition du premier Institut d’Égypte’, Bulletin de l’Institut d’Égypte 29 (1946–47): 343–67 and 30 (1947–48): 81–100. Goitein, S. D., A Mediterranean Society: The Jewish Communities of the Arab World as Portrayed in the Documents of the Cairo Geniza, 6 vols (Berkeley, 1967–1993). Grabar, Oleg, ‘al-Mushatta, Baghdad and Wasit’, in J. Kritzeck and R. B. Winder, The World of Islam: Studies in Honour of Philip K. Hitti (London, 1959), 99–108. ——, ‘The Illustrated Maqamat of the Thirteenth Century: The Bourgeoisie and the Arts’, in A.H. Hourani and S.M. Stern, eds, The Islamic City: A Colloquium (Oxford, 1970), 207–22. ——, The Formation of Islamic Art (New Haven, 1973). ——, The Alhambra (Cambridge, MA, 1978). – 243 –

BIBLIOGR A PH Y

——, ‘Reflections on the Study of Islamic Art’, Muqarnas 1 (1983): 1–14. ——, ‘Reflections on Mamluk Art’, Muqarnas 2 (1984): 1–12. ——, ‘From Dome of Heaven to Pleasure Dome’, Journal of the Society of Architectural Historians, 49, (March, 1990): 15–21. ——, The Great Mosque of Isfahan (New York, 1990). ——, ‘Patronage in Islamic Art’, in Islamic Art &Patronage: Treasures from Kuwait, ed. E. Atil (New York, 1990). Grousset, Rene, The Empire of the Steppes; a History of Central Asia (New Brunswick, NJ, 1970). Guémard, Gabriel, ‘Éssai d’histoire de l’Institut d’Égypte’, Bulletin de l’Institut d’Égypte 6 (1924): 43–84. Guiral, Paule, ‘Un architecte français en Afrique du Nord dans la première moitié du XIXe siècle’, Revue de l’Occident Musulman et de la Mediterranée 6 (1969): 103–12. ——, ‘L’architecte marseillais Pascal Coste et son oeuvre en Égypte sous Mehmet Ali’, Provence Historique 19 (1976): 133–42. Guo, Li, ‘Mamluk Historiographic Studies: The State of the Art’, Mamluk Studies Review 1 (1997): 15–43. Haarmann, Ulrich, Quellenstudien zur frühen Mamlukenzeit (Freiburg, 1970). ——, ‘Auflösung und Bewahrung der klassischen Formen arabischer Geschichtsschreibung in der Zeit der Mamluken’, Zeitschrift der Deutschen Morgenländischen Gesellschaft 121 (1971): 46–60. ——, ‘Turkish legends in the popular historiography of medieval Egypt’, in F. Rundgren, ed. Proceedings of the VIth Congress of Arabic and Islamic Studies (Stockholm, 1972), 97–107. ——, ‘Arabic in Speech, Turkish in Lineage: Mamluks and their Sons in the Intellectual life of Fourteenth-Century Egypt and Syria’, Journal of Semitic Studies 33 (1988): 81–114. ——, ‘Ideology and History, Idendity and Alterity: The Arab Image of the Turk from the ‘Abbasids to Modern Egypt’, International Journal of Middle Eastern Studies 20/2 (May 1988): 175–96. ——, ‘Rather the Injustice of the Turks than the Righteousness of the Arabs – Changing ‘Ulama’ Attitudes Towards Mamluk Rule in the Late Fifteenth-Century’, Studia Islamica 68 (1989): 61–79. ——, ‘Regicide and “The Law of the Turks” ’, in Intellectual Studies on Islam: Essays Written in Honor of Martin B. Dickson, ed. Michel M. Mazzaoui and Vera B. Moreen (Salt Lake City, 1990), 127–35. ——, ‘Joseph’s Law – The Careers and Activities of Mamluk Descendants before the Ottoman Conquest of Egypt’, in T. Philipp and U. Haarmann, eds, The Mamluks in Egyptian politics and society (Cambridge, 1998), 77–83. ——, ‘al-Maqrizi, the Master, and Abu Hamid al-Qudsi, the Disciple – Whose Historical Writing Can Claim More Topicality and Modernity?’ in The Historiography of Islamic Egypt, c.950–1800, ed. Hugh Kennedy (Leiden, 2000), 59–75. Haldane, Duncan, Mamluk Painting (Warminster, England, 1978). Hamdan, Jamal, Shakhsiyyat Misr, Dirasa fi ‘Abqariyyat al-Makan, 4 vols (Cairo, 1980–4). Hanisch, Hanspeter, Die ayybidischen Toranlagen der Zitadelle von Damaskus: ein Beitrag zur Kenntnis des mittelalterlichen Festungsbauwesens in Syrien (Wiesbaden, 1996). Hanna, Milad, The Seven Pillars of the Egyptian Identity (Cairo, 1994). Haskins, Charles H., Studies in the History of Mediaeval Science, 2nd edn (Cambridge, MA, 1927). Hautecoeur, Louis, and Gaston Wiet, Les mosquées du Caire, 2 vols (Paris, 1932). Heidemann, Stefan, Die Renaissance der Städte in Nordsyrien und Nordmesopotamien. Städtische Entwicklung und wirtschaftliche Bedingungen in ar-Raqqa und Harran von der Zeit der beduinischen Vorherrschaft bis zu den Seldschuken (Leiden, 2002). ——, ‘Arab Nomads and Seljuq Military’, in Militär und Staatlichkeit, ed. Irene Schneide (Halle, 2003), 201–20. Herz, Max, La Mosquée du Sultan Hassan au Caire (Cairo, 1899). ——, La Mosquée el-Rifaï au Caire (Milan, 1912). Hill, Kara Marietta, ‘Pascal-Xavier Coste (1787–1879): A French Architect in Egypt’, Ph.D. Thesis, Massachusetts Insititute of Technology, 1992. Hillenbrand, Carole, ‘The Career of Najm al-Din Il-Ghazi’, Der Islam 58 (1981): 250–92. – 24 4 –

BIBLIOGR A PH Y

Hoexter, Miriam, ‘Waqf Studies in the Twentieth Century: The State of the Art’, Journal of the Economic and Social History of the Orient 41/4 (Nov 1998): 474–95. Holt, P. M., ‘The Virtuous Ruler in Thirteenth-Century Mamluk Royal Biographies’, Nottingham Medieval Studies 24 (1980): 27–35. ——, ‘Three Biographies of al-Zahir Baybars’, in Medieval Historical Writing in the Christian and Islamic Worlds, ed. D. O. Morgan (London, 1982), 19–29. ——, ‘Some Observations on the Abbasid Caliphate of Cairo, ‘Bulletin of the School of Oriental and African Studies 47 (1984): 501–7. ——, The Age of the Crusades: The Near East From the Eleventh Century to 1517 (London, 1986). ——, ‘Shams al-Shuja‘i: A Chronicler identified?’ Bulletin of the School of Oriental and African Studies 58 (1995): 532–34. Hugo, Victor, Notre Dâme de Paris (Paris, 1830–32). Humphreys, R. Stephen, ‘The Expressive Intent of the Mamluk Architecture in Cairo: A Preliminary Essay’, Studia Islamica 35 (1972): 69–119. ——, From Saladin to the Mongols: The Ayyubids of Damascus 1193–1260 (Albany, NY, 1977). ——, ‘The Emergence of the Mamluk Army’, Studia Islamica 45 (1977): 67–99; 46 (1977): 147–82. Hussein, Taha, al-Ayyam, 2 vols (Cairo, 1962). Ibn ‘Abd al-Zahir, al-Altaf al-Khafi yya fi-l-Sira al-Sharifa al-Sultaniyya al-Malakiyya al-Ashrafi yya, ed. Axel Moberg (Lund, 1902). ——, Tashrif al-Ayyam wa-l-‘Usur fi-Sirat al-Malik al-Mansur, ed. Murad Kamil (Cairo, 1961). ——, al-Rawd al-Zahir fi-Tarikh al-Malik al-Zahir, ed. ‘Abd al-‘Aziz al-Khuwaiter (Riyadh, 1976). ——, Diwan, ed. Gharib M. Ahmad (Cairo, 1990). Ibn Abi al-Fada’il, Mufaddal, al-Nahj al-Sadid wa-l-Durr al-Farid fi-ma Ba‘d Tarikh ibn al-‘Amid, ed. E. Blochet, in Patrologia Orientalis (Paris, 1919–20). Ibn Abi Usaybi‘a, Uyun al-Anba fi Tabaqat al-Atibba, ed. Nizar Rida, 2 vols (Beirut, 1956–57). Ibn al-‘Adim, Zubdat al-Halab min Tarikh Halab, ed. Sami al-Dahhan, 3 vols (Damascus, 1951–68). Ibn al-Athir, al-Tarikh al-Bahir fi-l-Dawla al-Atabikiyya fi-l-Musil, ed. A. Q. Tulaymat (Cairo, 1963). ——, al-Kamil fi-l-Tarikh, 13 vols (Beirut, 1965–67). ——, Majd al-Din, Manal al-Talib fi Sharh al-Gharaib (Mecca, 1983). Ibn al-Dawadari, Kanz al-Durar wa-Jami‘ al-Ghurar, vol. 9, al-Durr al-Fakhir fi-Sirat al-Malik al-Nasir, ed. H. R. Roemer (Cairo, 1960). ——, Kanz al-Durar wa-Jami‘ al-Ghurar, vol. 6, al-Durra al-Mudiyya fi Akhbar al-Dawla al-Fatimiyya, ed. Salah al-Din al-Munajjid (Cairo, 1961). ——, Kanz al-Durar wa-Jami‘ al-Ghurar, vol. 8, al-Durra al-Zakiyya fi Akhbar al Dawla al-Turkiyya, ed. U. Haarmann (Cairo, 1971). Ibn al-Furat, Tarikh al-Duwal wa-l-Muluk, ed. K. Zurayk and N. Izzedine, 3 vols (Beirut, 1939–42). Ibn al-Haytham, The Optics of Ibn al-Haytham: Books I-III, On Direct Vision, ed. A. I. Sabra, 2 vols (London, 1989). Ibn al-Muqaffa‘, Sawirus, Siar al-bay‘a al-Muqaddasa, ed. A. Khater and O. H. E. Burmester (Cairo, 1974). Ibn al-Qalanisi, Dhayl Tarikh Dimashq, ed. S. Zakkar (Damascus, 1983). Ibn al-Shuhna, al-Durr al-Muntakhab fi-Tarikh Mamlakat Halab, ed. Joseph Elyan Sarkis (Beirut, 1909). Ibn Duqmaq, al-Intisar li Wasitat ‘Iqd al-Amsar, ed. K. Vollers (Cairo, 1893). ——, al-Jawhar al-Thamin fi Siyar al-Khulafa wa-l-Muluk wa-l-Salatin, ed, S. A. ‘Ashur (Mecca, 1982). Ibn Fadl-Allah al-‘Umari, al-Ta‘rif bi-l-Mustalah al-Sharif (Cairo, 1894). ——, Masalik al-Absar fi Mamalik al-Amsar: Dawlat al-Mamalik al-Uwla, ed. D. Krawulsky (Beirut, 1986). ——, Masalik al-Absar fi-Mamalik al-Amsar (reproduced from MS. 2797/1, Ahmet III Collection, Topkapi Sarayi Library, Istanbul) ed. Fuat Sezgin (Frankfurt, 1988). Ibn Habib, Tazkirat al-Nabih fi-Ayyam al-Mansur wa-Banih, ed. M. M. Amin, 3 vols (Cairo, 1976). Ibn Hajar al-‘Asqalani, al-Durar al-Kamina fi A‘yan al-Mi’a al-Thamina, 4 vols (Hyderabad, 1929–32). ——, Tafsir Gharib al-Hadith: Murattaban ‘ala al-Huruf (Cairo, n.d.). Ibn Iyas, Bada’i‘ al-Zuhur fi Waqa’i‘ al-Duhur, ed. M. Mustafa, 5 vols in 6 books (Wiesbaden, 1960–75). – 245 –

BIBLIOGR A PH Y

Ibn Jama‘a, Tahrir al-Ahkam fi Tadbir Ahl al-Islam, ed. Hans Kofler, Islamica, vols 6–7 (1933–35), 6: 353–414, 7: 1–64. Ibn Jubair, al-Rihla (Beirut, 1980). Ibn Kathir, al-Bidaya wa-l Nihaya fi-l-Tarikh, 14 vols (Cairo, 1932–39). Ibn Khaldun, al-Muqaddima, ed. ‘Ali A-W Wafi, 4 vols (Cairo, 1960). ——, The Muqaddimah, trans. Franz Rosenthal, ed. N. J. Dawood (Princeton, 1967). Ibn Khallikan, Wafi yyat al-A‘yan, ed. Ihsan ‘Abbas, 6 vols (Beirut, 1969–72). Ibn Nazif, al-Tarikh al-Mansuri, ed. Abul‘id Dudu (Damascus, 1981). Ibn Shaddad, al-A‘laq al Khatira fi Dhikr Umara’ al-Sham wal-Jazira: Tarikh Halab, ed. Dominique Sourdel, vol. 1, pt. 1 (Damascus, 1953). ——, al-A‘laq al Khatira: Tarikh Madinat Dimashq, ed. Sami al-Dahhan (Damascus, 1956). ——, al-A‘laq al Khatira, vol. 3, pt. 1, ed. Yahya ‘Abbara (Damascus, 1978). ——, Tarikh al-Malik al-Zahir, ed. Ahmad Hutait (Wiesbaden, 1983). Ibn Shahin al-Zahiri, Zubdat Kashf al-Mamalik wa-Bayan al-Turuq wa-l-Masalik, ed. Paul Ravaisse (Paris, 1894). Ibn Shakir al-Kutubi, Fawat al-Wafi yyat wa-l-Dhayl ‘Alayha, ed. Ihsan ‘Abbas, 4 vols (Beirut, 1973). Ibn Taghri-Birdi, al-Nujum al-Zahira fi Muluk Misr wa-l-Qahira, ed. Muhammad Ramzi, 16 vols (Cairo, 1930–56). ——, al-Manhal al-Safi wa-al-Mustawfa ba‘d al-Wafi, ed. A. Y. Najati et al., 6 vols (Cairo, 1956–2000). Ibn Tulun al-Salihi, al-Sham‘a al-Mudiyya fi Akhbar al-Qal‘a al-Dimashqiyya (Damascus, 1929). ——, I‘lam al-Wara bi-man Wulliya Na’iban min-al-Atrak bi-Dimashq al-Sham al-Kubra, ed. M. A. Duhman (Damascus, 1984). Ibn Wasil, Mufarrij al-Kurub fi Akhbar Bani Ayyub, ed. Jamal al-Din al-Shayyal et al., 5 vols (Cairo, 1953–77). Ibn-‘Asakir, Tarikh Madinat Dimashq, ed. S. al-Munajjid (Damascus, 1951–54). ——, Tarikh Madinat Dimashq wa Zikr Fadlaha wa-Tasmiyat man Hallaha min-al-Amathil aw Ijtaza bi-Nawahiha min Waridiha wa-Ahlaha, 19 vols (reproduced from MS in al-Zahiriyya in Damascus), ed. M. al-Tarhuni (Damascus, 1980). Ibrahim, ‘Abd al-Latif, ‘Wathiqat al-Amir Akhur Kabir Qaraquja al-Hasani’, Majallat Kulliyat al-Adab 18/2 (Dec. 1956): 183–251. ——, ‘al-Watha’iq fi Khidmat al-Athar’, in al-Mu’tamar al-Thani li-l-Athar fi al-Bilad al-‘Arabiyya (Cairo, 1958), 205–88. Ibrahim, Laila A., ‘Mamluk Monuments of Cairo’, Quaderni dell’Instituto Italiano di Cultura per la R.A.E. (Cairo, 1976). ——, Dragons on a Cairene Madrasa’, Art and Archaeology Research Papers 10 (1976): 11–19. ——, ‘Middle-Class Living Units in Mamluk Cairo: Architecture and Terminology.’ Art and Archaeology Research Papers 14 (1978). ——, ‘Residential Architecture in Mamluk Cairo.’ Muqarnas 2 (1984): 47–59. —— and M. M. Amin, al-Mustalahat al-Mi‘mariyya fi-l-Watha’iq al-Mamlukiyya (Cairo, 1990). Ilbert, Robert, and Mercedes Volait, ‘Neo-Arabic Renaissance in Egypt, 1870-1930’, Mimar 13 (1983): 26–34. Irwin, Robert, The Middle East in the Middle Ages, The Early Mamluk Sultanate 1250–1382 (Carbondale, IL, 1986). Izzat, Izzat ‘Ali, al-Shakhsiyya al-Misriyya fi al-Amthal al-Sha‘biyya (Cairo, 1997). Jakeman, Jane, ‘Abstract Art and Communication in ‘Mamluk’ Architecture’, Ph.D. Thesis, Oxford University, 1993. James, David, Arab Painting (Edinburgh and London, 1978). ——, Qur’ans of the Mamluks (New York, 1988). Jasmin, Denise, ‘Le travail d’un architecte marseillais en Orient’, Perspectives Mediterranéennes 14 (June 1983): 3–9. Kaegi, Walter E., Jr., ‘The Contribution of Archery to the Turkish Conquest of Anatolia’, Speculum 39 (1964): 96–108. – 24 6 –

BIBLIOGR A PH Y

Kedar, B. Z., ed., The Horns of Hattin (Jerusalem, 1992). Kennedy, Hugh, ed., The Historiography of Islamic Egypt, c.950–1800 (Leiden, 2000). Khalidi, Tarif, Arabic Historical Thought in the Classical Period (New York, 1994). Khalil, ‘Imad al-Din, al-‘Imara al-Artuqiyya fi al-Jazira wa al-Sham (456–812 H/1072–1409 AD):Adwa’ Jadida ‘ala al-Muqawama al-Islamiyya lil-Salibiyyin wal-Tatar (Beirut, 1980). Khuri, Ra’if, Modern Arab Thought: Channels of the French Revolution to the Arab East, trans. I. ‘Abbas (Princeton, 1983). Koch, Yoel, ‘‘Izz al-Din ibn Shaddad and His Biography of Baybars’, Annali: Istituto Universitario Orientale, Sezione Slava 43 (1983): 249–87. Kostof, Spiro, A History of Architecture (New York, 1985). Krautheimer, Richard, ‘Introduction to an “Iconography of Medieval Architecture”’, Journal of the Warburg and Courtauld Institutes 5 (1942): 1–33; reprinted in Studies in Early Christian, Medieval and Renaissance Art (London and New York, 1969), 115–50. Krawulsky, Dorothea, al-‘Arab wa Iran, Dirasat fi al-Tarikh wa al-Adab min al-Manzur al-Aydiologi (Beirut, 1993). Kruk, Remke, ‘History and Apocalypse: Ibn al-Nafis’ Justification of Mamluk Rule’, Der Islam 72 (1995): 324–37. Laoust, Henri, ‘La pensée et l’action politiques d’al-Mawardi’, Revue d’Études Islamiques 36 (1968): 12–92. Lapidus, Ira, Muslim Cities in the Later Middle Ages (Cambridge, 1967). ——, ‘Ayyubid Religious Policy and the Development of the Schools of Law in Cairo’, in André Raymond et al., Colloque international sur l’histoire du Caire (27 mars–5 avril 1969) (Cairo, 1969), 279–86. Lassner, Jacob, The Topography of Baghdad in the Early Middle Ages: Texts and Studies (Detroit, 1970). Lavin, Irving, ‘The House of the Lord: Aspects of the Role of Palace Triclinia in the Architecture of Late Antiquity and the Early Middle Ages’, Art Bulletin, 44 (1962): 1–27. Leiser, Gary, ‘The Madrasa and the Islamization of the Middle East: The Case of Egypt’, Journal of the American Research Center in Egypt 22 (1985): 29–47. Leprun, Sylviane, ‘Rives, dérives ethno-architecturales: Pascal-Xavier Coste, architecte anthropologue’, Les cahiers de la recherche architecturale 27/28 (1992): 209–18. Levanoni, Amalia, ‘The Mamluks’ Ascent to Power in Egypt’, Studia Islamica 72 (1990): 121–44. ——, ‘The Consolidation of Aybak’s Rule: An Example of Factionalism in the Mamluk State’, Der Islam 71 (1994): 241–54. ——, A Turning Point in Mamluk History: the Third Reign of al-Nasir Muhammad ibn Qalawun (1310–1341) (Leiden, 1995). Lézine, Alexandre, Mahdiya: recherches d’archéologie islamique (Paris, 1965). ——, ‘Les Salles Nobles Des Palais Mamelouks’, Annales Islamologiques 10 (1972): 149–205. Little, Donald P., An Introduction to Mamluk Historiography: An Analysis of Arabic Annalistic and Biographical Sources for the Reign of al-Malik an-Nasir Muhammad ibn Qalawun (Wiesbaden, 1970). ——, ‘The Recovery of a Lost Source for Bahri Mamluk History: al-Yusufi’s Nuzhat al-Nazir fi Sirat alMalik al-Nasir’, Journal of the American Oriental Society 94 (1974): 42–54. ——, ‘al-Safadi as Biographer of his Contemporaries’, in Essays on Islamic Civilization Presented to Niyazi Berkes, ed. Donald Little (Leiden, 1976). ——, ‘Notes on Aitamis, A Mongol Mamluk’, Beiruter Texte und Studien 22 (1979): 387–401. ——, ‘The Haram Documents as Sources for the Arts and Architecture of the Mamluk Period’, Muqarnas 2 (1984): 61–72. ——, ‘The Nature of Khanqahs, Ribats, and Zawiyas under the Mamluks’, in Islamic Studies Presented to Charles J. Adams, ed. Wael B. Hallaq and Donald P. Little (Leiden, 1991), 91-–105. Louca, Anwar, ‘La renaissance égyptienne et les limites de l’œuvre de Bonaparte’, Cahiers d’histoire égyptienne 7/1 (Feb. 1955): 1–20. Lowick, Nicholas, ‘The Religious, the Royal and the Popular in the Figural Coinage of the Jazira’, in The Art of Syria and the Jazira 1100–1250, ed. Julian Raby (Oxford, 1985), 159–74. – 2 47 –

BIBLIOGR A PH Y

Lowry, Joseph E., ‘Time, Form and Self: The Autobiography of Abu Shama, Edebiyat 7/2 (Autumn 1996): 313–25. Lutfi, Huda, ‘Manners and Customs of fourteenth-century Cairene Women: Female Anarchy Versus Male Shar‘i Order in Muslim Prescriptive Treatises’, in Women in Middle Eastern History: Shifting Boundaries in Sex and Gender, ed. Nikki Keddie and Beth Baron (New Haven, CT, 1991), 99–121. MacDonald, William L., The Architecture of the Roman Empire: An Introductory Study (New Haven, CT, 1982). Makdisi, George, Rise of Humanism in Classical Islam and the Christian West (Edinburgh, 1990). Malti-Douglas, Fedwa, ‘Mentalités and Marginality: Blindness and Mamluk Civilization’, in Essays in Honor of Bernard Lewis: The Islamic World from Classical to Modern Times, ed. C. Bosworth et al. (Princeton, 1989), 211–37. Mango, Cyril, Byzantine Architecture (New York, 1985). Marsot, Afaf Lutfi al-Sayyid, Egypt in the Reign of Muhammad ‘Ali (Cambridge, 1984). ——, A Short History of Modern Egypt (Cambridge, 1986). Massoud, Sami, ‘al-Maqrizi as a Historian of the Reign of Barquq’, Mamluk Studies Review 7/2 (2003): 119–36. Mayer, Leo, Saracenic Heraldry (Oxford, 1938). ——, Mamluk Costume: a Survey (Geneva, 1952). ——, Islamic Architects and Their Works (Geneva, 1956). ——, Islamic Astrolabists and Their Works (Geneva, 1956). ——, Islamic Metalworkers and Their Works (Geneva, 1959). Meinecke, Michael, ‘Das Mausoleum des Qala’un in Kairo: Untersuchungen zur Genese der Mamlukischen Architekturdekoration’, Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 27 (1971): 47–80. ——, ‘Zur mamlukischen Heraldik’, Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 28 (1972): 213–87. ——, ‘Mamluk Architecture, Regional Architectural Tradition: Evolutions and Interrelations’, Damaszener Mitteilungen 2 (1985): 163–75. ——, Die Mamlukische Architektur in Ägypten und Syrien (648/1250 bis 923/1517) 2 vols (Glückstadt, 1992). ——, Patterns of Stylistic Change in Islamic Architecture: Local Traditions versus Migrating Artists (New York, 1995). Mémoires sur l’Égypte publiés pendant les campagnes du Général Bonaparte dans les années VI et VII (Paris: Baudouin, 1801). Migeon, Gaston, Les arts musulmans (Paris and Brussels, 1926). Miquel, André, La géographie humaine du monde musulman jusqu’au milieu du 11e siècle, 3 vols (Paris, 1967–80). Mitchell Brown, Helen, ‘Some Reflections on the Figured Coinage of the Artuqids and Zengids’, in Near Eastern Numismatics, Iconography, Epigraphy and History: Studies in Honor of George C. Miles, ed. D. Kouymjian (Beirut, 1974), 353–58. Mitchell, Timothy, Colonizing Egypt (Berkeley, 1988). Morray, D. W., An Ayyubid Notable and his World: Ibn al-‘Adim and Aleppo as Portrayed in his Biographical Dictionary of People Associated with the City (Leiden, 1994). Mostyn, Trevor, Egypt’s Belle Epoque: Cairo 1869–1952 (London, 1989). Mubarak, ‘Ali Pasha, al-Khitat al-Tawfiqiyya al-Jadida 20 vols (Reprt. Cairo, 1969). Mustafa, Salih Lam‘i, al-Watha’iq wa-l-‘Imara, Dirasa fi-l-‘Imara al-Islamiyya fi-l-‘Asr al-Mamluki al-Jarkasi (Beirut, 1980). Mustafa, Shakir, ‘Dukhul al-Turk al-Ghuzz ila Bilad al-Sham’, in The International Congress on the History of Bilad al-Sham, ed. A.K. Gharaybeh, A. A. al-Duri, and O. al-Madani (Beirut, 1974), 303–98. ——, al-Tarikh al-‘Arabi wa-l-Mu’rikhun: Dirasa fi Tatawwur ‘Ilm al-Tarikh wa-Rijaluhu fi-l-Islam, 4 vols (Beirut, 1978–93).

– 24 8 –

BIBLIOGR A PH Y

Necipoglu, Gulru, The Topkapi Scroll: Geometry and Ornament in Islamic Architecture: Topkapi Palace Museum Library MS H. 1956 (Santa Monica, CA, 1995). Nielsen, Jørgen, ‘Mazalim and Dar al-‘Adl under the Early Mamluks’, The Muslim World 66/2 (April 1976): 114–21. ——, Secular Justice in an Islamic State: Mazalim under the Bahri Mamluks, 662/1264–789/1387 (Leiden, 1985). Nizam al-Mulk, Siyasat-Namah, trans. al-Sayyed al-‘Izzawi, (Cairo, 1976). Nochlin, Linda, ‘The Imaginary Orient’, Art in America 71 (May 1983): 119–29, 187–91. O’Kane, Bernard, ‘Monumentality in Mamluk and Mongol Art and Architecture’, Art History 19/4 (December 1996): 499–522. Öney, Gönül, ‘Mounted Hunting Scenes of Anatolian Seljuqs in Comparison with Iranian Seljuqs’, Anatolia 11 (1967): 121–89. ——, ‘Elements from Ancient Civilizations in Anatolian Seljuk Art’, Anatolica 12 (1970): 17–38. ——, ‘Dragon Figures in Anatolian Seljuk Art’, Belleten 32/130 (1971): 171–216. Otto-Dorn, Katharina, ‘Figural Stone Reliefs on Seljuq Sacred Architecture in Anatolia’, Kunst des Orients 12, 1/2 (1978/79): 103–46. Parry, V. and M.E. Yapp, eds, War, Technology and Society in the Middle East (Oxford, 1975). Petry, Carl, The Civilian Elite of Cairo in the Later Middle Ages (Princeton, 1981). Pflugradt-Abdel Aziz, Elke, ‘The Mausoleum for Soliman Pasha ‘el-Faransawi’ in Cairo’, Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 44 (1988): 205–14. Pococke, Richard, A Description of the East and some Other Countries, vol. I, Observations on Egypt (London, 1743–45). Pouzet, Louis, ‘Abu ˇSama (599–665/1203–1268) et la société damascaine de son temps’, Bulletin d’Études Orientales 37–38 (1985–86): 115–26. Rabbat, Nasser, The Citadel of Cairo (Geneva, 1989). ——, The Citadel of Cairo: A New Interpretation of Royal Mamluk Architecture (Leiden, 1995). ——, ‘al-Iwan: Ma‘nahu al-Faraghi wa Madlulahu al-Tadhkari’, Bulletin d’Études Orientales 49 (1997): 249–67. ——, ‘al-Madina wa al-Tarikh wa al-Sultah: al-Maqrizi wa Kitabuhu al-Ra’id al-Mawa‘iz wa al-I‘tibar bi Dhikr al-Khitat wa al-Athar’, Annales Islamologiques 35 (2001): 77–100. Rabie, Hassanein, ‘Size and Value of the Iqta‘ in Egypt 564-741 AH, 1169–1341 A.D.’, in Studies in the Economic History of the Middle East, ed. M. A. Cook (Oxford, 1970), 129–38. ——, The Financial System of Egypt A.H. 564–741 AD 1169–1341 (Oxford, 1972). Radwan, Tal‘at, and Fathi Radwan, Ab‘ad al-Shakhsiyya al-Misriyya: Bayna al-Madi wa-al-Hadir (Cairo, 1999). Ramadan, Mustafa M., Dawr al-Azhar fi al-Hayat al-Misriyya (Cairo, 1986). Ravaisse, Paul, ‘Essai sur l’histoire et la topographie du Caire d’après al-Maqrizi’, Mémoires publiés par les Membres de la Mission Archéologique Française au Caire (Paris, 1886 and 1889), 1:409–79, 3:1–114. Raymond, André, ‘Les constructions de l’émir ‘Abd al-Rahman Katkhuda au Caire’, Annales Islamologiques 11 (1972): 235–51. ——, Le Caire (Paris, 1993). Reid, Donald Malcolm, Cairo University and the Making of Modern Egypt (London, 1990). ——, ‘Cultural Imperialism and Nationalism: The Struggle to Define and Control the Heritage of Arab Art in Egypt’, International Journal of Middle Eastern Studies 24/1 (Feb. 1992): 57–76. Rentz, Sophie, ‘The Effect of the Introduction of the Printing Press on Learning in Egypt in the Early Nineteenth Century’, Master’s Thesis, San Jose State College, 1972. Revault, Jacques, et al. Palais et Maisons Du Caire, I- Époque Mamelouke (Paris, 1982). Rida, Muhammad Sa‘id, ‘Ibn Shaddad fi Kitabihi al-A‘laq al-Khatirah, ‘Qism al-Jazirah’,’ Majallat al-Mu’arrikh al-‘Arabi 14 (1980): 124–204. Rogers, J. M., ‘The 11th century – A Turning Point in the Architecture of the Mashriq’, in Islamic Civilization 950–1150, ed. D. S. Richards (Oxford, 1973), 211–49.

– 24 9 –

BIBLIOGR A PH Y

——, ‘Architectural History as Literature: Creswell’s Reading and Methods’, Muqarnas 8 (1991): 45–54. ——, ‘Waqfiyyas and Waqf-Registers: New Primary Sources for Islamic Architecture’, Kunst des Orients 11(1976–77): 182–96. Rosen-Ayalon, Miriam, ‘A Neglected Group of Mihrabs in Palestine’, in Studies in Islamic History and Civilization in Honor of Professor David Ayalon, ed. M. Sharon (Jerusalem, 1986), 553–63. Rosenthal, Franz, A History of Muslim Historiography, 2nd edn (Leiden, 1968). Said, Edward, Orientalism (New York, 1979). Sakr, Tarek, Early Twentieth-Century Islamic Architecture in Cairo (Cairo, 1992). Salam-Leibish, Hayat, The Architecture of the Mamluk City of Tripoli (Cambridge, MA, 1983). Sanders, Paula, Ritual, Politics, and the City in Fatimid Cairo (Albany, NY, 1994). Sauvaget, Jean, ‘Les Ghassanides et Sergiopolis’, Byzantion 14 (1939): 115–130. ——, Les monuments ayyoubides de Damas, 2 vols (Damascus, 1940). ——, Alep, Essai sur le developpement d’une grande ville syrienne, des origines au milieu du XIXe siècle (Paris, 1941). ——, ‘Une ancienne représentation de Damas au Musée du Louvre’, Bulletin d’Études Orientales 11 (1945–46): 5–12. ——, La Mosquée oméyyade de Médina: Étude sur les origines architecturales de la mosquée et de la basilique (Paris, 1947). Sayed, Hazem, ‘Development of the Cairene Qa‘a: Some Considerations’, Annales Islamologiques 23 (1987): 31–53. ——, ‘The Rab‘ in Cairo: A Window on Mamluk Architecture and Urbanism’, Ph.D. Thesis, MIT, 1987. Shafe‘i, Farid, al-‘Imara al-‘Arabiyya fi Misr al-Islamiyya, ‘Asr al-Wulat (Cairo, 1970). ——, al-‘Imara al-‘Arabiyya al-Islamiyya, Madiha wa-Hadiraha wa-Mustaqbalaha (Riyadh, 1982). Shatzmiller, Maya, ‘The Crusades and Islamic Warfare: A Re-evaluation’, Der Islam 69 (1992): 247–88. ——, ed., Crusaders and Muslims in Twelfth-Century Syria (Leiden, 1993). Sherif, Lobna Abdel Azim, ‘Layers of Meaning: An Interpretive Analysis of Three Early Mamluk Buildings’, Ph.D. Thesis, University of Michigan, 1988. Shoshan, Boaz, Popular Culture in Medieval Cairo (Cambridge, 1993). Sibt ibn-al-Jawzi, Mir’at al-Zaman fi-Tarikh al-A‘yan (Hyderabad, 1952). Sivan, Emmanuel, L’Islam et la Croisade, idéologie et propagande dans les réactions musulmanes aux Croisades (Paris, 1968). Smith, E. Baldwin, The Dome: A Study in the History of Ideas (Princeton, NJ, 1951, repr. 1978). Stern, Henri, ‘Notes sur l’architecture des châteaux omeyyades’, Ars Islamica 11–12 (1946): 89–92. Stowasser, Karl, ‘Manners and Customs at the Mamluk Court’, Muqarnas 2 (1984): 13–21. Strauss, Éli (Eliyahu Ashtor), ‘Prix et salaire à l’époque mamlouk’, Revue d’Études Islamiques 17 (1949): 49–94. Sweetman, John, The Oriental Obsession: Islamic Inspiration in British and American Art and Architecture 1500–1920 (Cambridge, 1987). Tabbaa, Yasser, ‘The Architectural Patronage of Nur al-Din (1146–1174)’, Ph.D. Thesis, New York University, 1982. ——, ‘Monuments with a Message: Propagation of Jihad under Nur A-Din (1146–1174)’, The Meeting of Two Worlds: Cultural Exchange between East and West during the Period of the Crusades, ed. V. P. Goss and C. V. Bornstein (Kalamazoo, MI, 1986), 223–40. ——, ‘Circles of Power: Palace, Citadel and City in Ayyubid Aleppo’, Ars Orientalis 23 (1993): 181–200. ——, ‘The Transformation of Arabic Writing: Part 2: The Public Text’, Ars Orientalis 24 (1994): 119–46. ——, Constructions of Power and Piety in Medieval Aleppo (University Park, PA, 1997).

–250 –

BIBLIOGR A PH Y

——, The Transformation of Islamic Art during the Sunni Revival (Seattle, 2001). Taragan, Hana, ‘Politics and Aesthetics: Sultan Baybars and the Abu Hurayra/Rabbi Gamliel Building in Yavne’, in Milestones in the Art and Culture of Egypt, ed. Asher Ovafiah (Tel Aviv, 2000), 117–43. Taymur Pasha, Ahmad, al-Taswir ‘Ind al-‘Arab, ed. Z. M. Hassan (Cairo, 1942). ——, al-Muhandisun fi al-‘Asr al-Islami, 2nd edn (Cairo, 1979). Tyan, Émile, Histoire de l’organisation judiciaire en pays d’Islam, 2 vols (Leiden, 1960). Van Berchem, Max, Amida: Matériaux pour l’épigraphie et l’histoire musulmanes du Diyar-Bekr (Paris, 1910). ——, Matériaux pour un Corpus Inscriptionum Arabicarum, Jerusalem 2 vols (Cairo, 1920–22). Van Leeuwen, Richard, Waqfs and Urban Structures: The Case of Ottoman Damascus (Leiden, 1999). Vatikiotis, P. J., The History of Modern Egypt from Muhammad Ali to Mubarak (Baltimore, MD, 1991). Volait, Mercedes, L’architecture moderne en Égypte et la revue al-‘imara (1939–1959) (Cairo, 1988). Wendell, Charles, ‘Baghdad: Imago Mundi and Other Foundation Lore’, International Journal for Middle Eastern Studies, 2 (1971): 99–128. Whelan, Estelle, ‘Representations of the Khassakiyah and the Origins of Mamluk Emblems’, in Content and Context of Visual Arts in the Islamic World, ed. P. Soucek (Philadelphia, 1988), 219–43. Wiedemann, E., ‘Fragen aus dem Gebiet der Naturwissenschaften, gestellt von Friedrich II’, in Archiv für Kulturgeschichte 11 (1914): 483–85. Wiet, Gaston, ‘Une inscription de Malik Zahir Ghazi à Lattakieh’, Bulletin de l’institut français d’archéologie orientale du Caire 30 (1930–31): 273–92. ——, Mohammed Ali et les beaux-arts (Cairo, 1949). Williams, Caroline, ‘The Cult of the ‘Alid Saints in the Fatimid Monuments of Cairo, Part 1: The Mosque of al-Aqmar’, Muqarnas 1 (1983): 37–52. Zakarya, Mona, Deux palais du Caire médiéval, waqfs et architecture (Marseilles, 1983). Zaki, ‘Abd al-Rahman, al-Qahira, Tarikhaha wa-Atharaha (Cairo, 1966). ——, Mawsu‘at Madinat al-Qahira fi Alf ‘Am (Cairo, 1969). Zetterstéen, K. V., Beitrage zur Geschichte der Mamlukesultane in den Jahren 690–741 (Leiden, 1919). Ziyada, M. Mustafa, et al., Dirasat ‘an al-Maqrizi, Majmu‘at Abhath (Cairo, 1971).

– 2 51 –

INDEX ‘Abbas (jurist), 153 ‘Abbas Hilmi II (Khedive), 82, 95, 96, 97 Al-‘Abbasi, Muhammad Mahdi (Rector), 96 Abbasids, 3, 5, 6, 7, 14, 24, 29, 42, 59, 62, 65, 80, 83, 127, 136, 137, 148, 149, 163, 167 ‘Abbassiya, 96 ‘Abd, 3 ‘Abd al-Fattah, M. Hussam al-Din (historian), 112 ‘Abd al-Latif al-Baghdadi (physician), 32, 88 ‘Abd al-Malik (Caliph), 136, 148 ‘Abd al-Nasser, Gamal (President), 96 ‘Abd al-Rahman Katkhuda (Amir), 73, 75, 78, 92, 93, 95 ‘Abdel Wahab, Hasan (art historian), 143, 144 ‘Abdu, Muhammad (Reformer), 96 Abjij (architect), 37 Abkhaz, 9 Ablaq, 113, 175, 177 Abu al-Hayja’ al-Samin (Amir), 171 Abu al-Qasim (Vizier), 62 Abu Bakr, al-Malik al-‘Adel (King), 63 Abu Ja‘far al-Mansur (Caliph), 126, 136, 137 Abu Yusuf (judge), 148 Abwab al-Sutur, 113 Adab, 16, 22, 23, 25, 29, 60, 169 Adib, 166 Al-‘Adil, 164 Al-Addur (or Dur) al-Sultaniyya, 114, 129, 157 Adhan, 83, 90 Al-Adhra‘i, Sadr al-Din Suleiman (judge), 79 Al-Afkhar Mosque (al-Fakahaniyyin), 84 Agha, 9 Ahdath, 13 Al-Ahkam al-Sultaniyya, 148, 149 Ahl al-Balad, 13 Ahmad, Ahmad ‘Abd al-Razzaq (Art historian), 143 ‘Aja’ib al-Makhluqat wa Ghara’ib al-Mawjudat, 169 ‘Ajib, 23, 166, 167, 169, 170 ‘Alaliq, 171 Alamut, 65 Alanya, 61

Aleppo, 7, 20, 27, 28, 33, 61, 62, 66, 71, 143, 146, 149, 151–3 Al-A’laq al-Khatira fi-Zikr Umara’ al-Sham wa-l-Jazira, 27, 28, 71 Alexandria, 176, 177 Algeria, 173 ‘Ali (Caliph, Imam), 82, 148 ‘Ali ibn al-Nu‘man (judge), 86 ‘Ali, al-Mansur (Sultan), 5 Alin Aq (Amir), 115, 118 Allan, J. W. (art historian), 35 Alp Arslan (Sultan), 14, 149 America, 174 American University of Cairo, 144 Amin, Muhammad Muhammad (historian), 112 Al-Amir (Caliph), 84, 85 Amir akhur, 162 Amir of Forty, 60, 158 Amir of Hundred, 60, 158, 159 Amir of Ten, 59, 158 Amir of Twenty, 60 Al-‘Amma, 13, 157 Anatolia, 7, 9, 10, 16, 23, 59, 61, 65, 66, 164, 181 Ani, 9 Al-Anwar Mosque (al-Hakim Mosques), 84 Aqbugha (Amir), 37, 89, 90 Al-Aqbughawiyya (Madrasa), 73, 74, 89, 93, 95 Al-Aqmar Mosque, 81, 84, 86, 177 Al-Aqsa, Mosque, 35 Aqsunqur al-Muzaffari (Amir), 13 Arbab al-hiraf wa-al sana’i‘, 34 L’architecture Arabe ou Monuments du Kaire mesurés et dessinés de 1818 a 1826, 142, 176 Arghun al-Kamili (Amir), 121 Arkan, 128 Armenian Style, 178 Armenians, 8, 9, 15, 59, 60 Arwad, 162 ‘Asabiyya, 6, 9, 10 Atabek al-‘Asakir, 6, 89, 90 Atabek, 4, 90 ‘Attar, Farid al-Din (poet), 181

–252–

IN DEX

Awlad al-Nas, 10, 12–16, 26 Aybak al-Musta‘rib, 6 Aybak al-Turkmani, ‘Izz al-Din (Sultan), 4, 5, 7, 154 Aydaghdi al-Rukni (Amir), 40, 42 Aydakin al-Bundaqdari (Amir), 154 Aydamar al-Hilli (Amir), 89 Al-‘Ayni, Badr al-Din (historian), 17, 21, 168 ‘Ayntab (Gaziantep), 17 Ayyub, al-Salih Najm al-Din (Sultan), 4, 109, 130, 131, 134, 149, 153 Ayyubids, 4, 5, 7, 14, 15, 27, 30, 39, 63, 79, 88, 117, 125, 147–154, 162, 163, 171, 176 Azbakiyya, 92 Azerbaijan, 16, 66 Al-Azhar Mosque, 72–75, 77–97, 130 Al-Azhar University, 96, 97 Al-‘Aziz (Caliph), 72, 83, 84, 86, 127 ‘Azim, 23 Azj, 104 Bab al-Hayyat, 65 Bab al-Jabal, 151 Bab al-Jinan, 150 Bab al-Muzayinin, 73, 75, 90, 93, 95 Bab al-Nasr, 82, 150, 171 Bab al-Sa‘ayida, 78, 93, 95 Bab al-Saghir, 151 Bab al-Shourba, 78, 93 Bab al-Silsila, 155 Bab Dar al-‘Adl, 151 Bab Dar al-Sa‘ada, 150 Al-Badari, Abu al-Baqa’ (historian), 57 Badhahanj, 50, 51 Baghdad, 5, 6, 7, 30, 65, 83, 89, 126, 136, 137, 148, 163, 171 Bahadur al-Muqaddam (Amir), 118 Bahij, 22 Bahramshah, al-Amjad (King), 150 Al-Bahriyya, Bahri Mamluks, 4, 5, 9, 113 Al-Bilad, 10 Banu al-Muhtar, 155 Baptistère de St Louis, 35, 68, 69 Barada River, 150 Barada Scene, 48 Barquq, al-Zahir (Sultan), 37, 80, 90, 161, 162, 165, 177 Barsbay, al-Ashraf (Sultan), 17, 90, 116–119, 162, 165 Bashtak (Amir), 17, 18, 115, 116, 121, 123, 165 Basilica of al-Mundhir ibn al-Harith, 134, 135 Basilica, 132–137, 158, 160 Battle of ‘Ayn Jalut, 5 Battle of Dandiqan, 59 Battle of Malazkurd (Manzikurt), 59 Battle of Marj Dabiq, 71

Bay‘a qahriyya, 5 Baybars al-Jashankir (Sultan), 10 Baybars al-Mansuri (historian), 6, 13, 127 Baybars, al-Zahir (Sultan), 5–10, 15, 24, 26–8, 38, 47, 48, 54–56, 63, 66–9, 71, 78, 79, 87, 89–91, 109, 118, 125–8, 135, 136, 150, 154–6, 163, 164, 168 Baylabak al-Khazindar (Amir), 89 Bayn al-Qasryn, 84, 89, 132, 135 Baysari, Badr al-Din (Amir), 69, 118, 119 Bayt, 113, 119 Beaux-Arts, 142, 177 Behrens-Abouseif, Doris (art historian), 34, 37, 69, 112, 130, 144 Bektash Amir Silah (Amir), 121 Bernini, 102 Bibliothèque municipale de Marseille, 177 Bijou Palace, 119 Bilad al-Sham, 59–62, 65, 147, 162 Bimaristan al-Mansuri (of Qalawun), 37, 56, 57, 135 Bimaristan al-Nuri (of Nur al-Din), 40, 135 Birket al-Fil, 123 Bloom, Jonathan (Art historian), 79, 81 Borromini, 102 Bosra, 61, 64 Briggs, Martin S. (art historian), 72, 143, 144 Bubonic Plague (Black Death), 108 Bulaq, 180, 181 Bunduqdar, 68, 69 Burj al-Zawiya, 63 Al-Burjiyya, Burji Mamluks, 9, 10 Al-Busrawi (historian), 57 Al-Buyut al-Sultaniyya (Royal Storehouses), 121 Buyutat, 116 Byzantines, 3, 47, 50, 59, 61, 132, 134, 137, 144, 162 Cairo, 7, 17, 18, 20, 21, 24, 27, 28, 30, 33, 36–9, 47, 48, 56, 69, 71–3, 81, 83, 86, 88–97, 106, 108, 113, 114, 115, 117, 118, 120, 121, 125, 130, 132, 135–7, 141–5, 147, 150, 153, 154, 158, 163, 165, 168, 171, 174–8, 180–3 Casanova, Paul (Arabist), 155 Caucasia, the Caucasus, 3, 7, 10 Central Asia, 3, 4, 59, 64, 65 Charmi, Ahmad (architect), 96 Chengiz Khan, 4 Citadel of Aleppo, 61, 64–8, 151–3 Citadel of Cairo (Citadel of the Mountain), 8, 27, 36, 37, 47, 48, 51, 54–6, 61, 64, 67, 113, 120, 125, 127, 132, 135, 146, 153, 155, 157, 158, 176, 181 Citadel of Damascus, 63, 64, 151 Comité (Committee for the Conservation of Monuments of Arab Art), 95, 97, 184 Commission des sciences et arts, 174, 175 Copts, 14, 15

–253 –

IN DEX

Cordoba Mosque, 47 Coste, Pascal (architect), 142, 143, 174, 176–8, 180, 181, 183, 187 Creswell, K. A. C. (architectural historian), 76, 78, 79, 81, 95, 97, 106, 109, 110, 131, 133, 143, 144, 145 Crimea, 3, 7, 18 Crusaders, 8, 10, 14, 15, 24, 29, 55, 59, 61, 146, 147, 162, 164, 165 La Cuba, 126 Curzon, Robert (traveller), 178 Cyprus, 113, 165 D’Alembert, Jean le Rond (Encyclopedist), 175 Da‘wa, 88 Dahhan, 35 Damascus, 7, 8, 20, 24, 27, 28, 33, 37, 38, 40, 47–9, 51, 53–7, 61, 63, 67, 73, 126, 135, 136, 143, 146, 149–51, 153, 157 Damietta, 4, 94 Al-Daylam Gate, 86 al-Daylami, Muhtar (poet), 104 Dar, 113, 117, 118, 119 Dar al-‘Adl (Palace of Justice), 60, 62, 126, 128, 146–65 Dar al-‘adl (session of justice), 126, 132, 147, 157, 158, 161, 162, 165 Dar al-‘Adl al-Kamiliyya, 153, 154 Dar al-‘Adl al-Qadima, 155 Dar Farrukhshah, 150 Dar al-Fitra, 86 Dar al-Hikma, 87 Dar al-‘Ilm, 86, 87 Dar al-‘Imadi ibn al-Mushrif, 118 Dar al-Imara, 63, 135, 136 Dar al-Islam, 3 Dar al-‘Izz, 62 Al-Dar al-Kamiliyya (Damascus), 63 Dar Kashf al-Mazalim, 149 Dar al-Kutub (Egyptian National Library), 187 Dar al-Mahfuzat, 155 Dar al-Massarra, 63 Dar al-Niyaba, 153, 160, 161 Dar al-Sa‘ada (Aleppo), 153 Dar al-Sa‘ada (Damascus), 150, 151 Dast, 158 Daw’ al-Nibras wa Anas al-Jullas fi Akhbar al-Muzawwiqin min al-Nas, 38, 167 Depaule, Jean-Charles (historian), 112 Description de l’Égypte, 126, 129, 130, 142, 156, 157, 162, 175, 176 Devonshire (Mrs) (historian), 144 Devshirme, 8 Diderot, Denis (Encyclopedist), 175

Dihliz al-ubur, 129, 157 Dihliz, 114 Dikka, 107 Divan (poetry collection), 181 Divrigi, 61, 64 Diwan (chancery), 39, 169 Diwan (council), 94 Diwan al-‘ama’ir, 36 Diwan of Sultan al-Ghuri, 126 Diwan Yusuf (le divan de Joseph), 126, 156 Diyarbakr (Amid), 61, 64 Dome of the Rock, 47, 49, 54, 136 Durkah, 63 Durqa‘a, 114, 116, 118, 119, 123, 130, 131 Eastern Palace (al-Qasr al-Sharqi), 79, 83, 84, 86, 127, 134, 137 Egypt, 3–6, 10, 11, 14, 15, 29–33, 35, 36, 55, 59, 60, 65, 66, 71–4, 78–83, 87–92, 94–6, 109, 112, 117, 119, 125–7, 130, 132–4, 137, 142–144, 147–9, 156–8, 162, 163, 165, 167, 173–83, 185–7 Elements of Geometry, 40 Encyclopédie, ou Dictionnaire raisonné des sciences, des arts et des métiers, 175 Enlightenment, 175 Euclid, 40 Eugénie (Empress), 182 Europe, 3, 4, 59, 69, 72, 95, 142, 143, 163, 173–6, 178, 182-184 Fada’il, 31, 147 Al-Fada’il al-Bahira fi Mahasin Misr wa-l-Qahira, 31 Fahmi, Hussein (architect), 174, 182–5, 187 Fahmi, Mahmud (architect), 187 Fakhir, 23 Al-Fakhira (belvedere), 84 Faqih, fuqaha’ (jurist), 5, 39, 109, 153, 163 Faraj ibn Barquq (Sultan), 177 Fatima al-Zahra’ (Daughter of the Prophet), 82, 83 Fatima Khatun, Umm al-Salih, 37, 128 Fatimids, 5, 30, 36, 62, 72, 73, 75, 78–89, 95, 117, 127, 130, 134, 137, 141, 144, 155, 167 Fatwa, 79, 82 Fawat al-Wafi yyat, 28 Fellah, 72, 180 Fidawiyya, 164 Fikri, Ahmad (art historian), 143 France, 94, 142, 173, 176, 180, 182, 183 Frederick II (Emperor), 40 French Expedition, 73, 93, 94, 142, 156, 173, 174, 175, 177, 187 Fu’ad I (King), 96, 174 Fu’ad I Style, 174 Al-Fustat, 30, 79, 80–4, 87, 88, 130, 131, 134, 141

–254 –

IN DEX

Futuh, 29 Futuh al-Sham, 17 Garcin, Jean-Claude (historian), 112, 115, 118, 121 Geniza Documents, 35 Georgian, 3 Gharib al-Hadith, 169 Gharib al-Qur‘an, 169 Gharib, 23, 166, 169, 170 Ghassanids, 134 Ghazi, al-Zahir (King), 62, 65, 146, 151–3, 162 Ghaznavids, 7, 59 Ghulam (young slave), 3 Ghulamiyyat, 168 Ghutumi, 17 Ghuzz, 59 Goitein, S.D. (Arabist), 35, 38 Golden Horde (Mongols), 7, 8 Great Iwan, al-Iwan al-Kabir (Fatimid), 127 Great Iwan, al-Iwan al-Nasiri, 113, 114, 126, 127–37, 156, 157–62, 165 Greek, 9, 177 Haarmann, Ulrich (historian), 22, 25 Hadith, 94, 109, 169, 170 Hafiz (poet), 181 Al-Hafiz (Caliph), 75, 76, 78 Hajj, 79, 135 Al-Hajjaj (Governor), 136 Hajji, al-Muzaffar (Sultan), 13 Al-Hakim (Abbasid Caliph in Egypt), 6, 127 Al-Hakim (Fatimid Caliph), 72, 81, 83, 86, 88 Al-Hakim Mosque, 81, 84, 87, 88, 89, 130 Al-Halaqa, 8, 14, 40, 97 Halaqa, 97 Hama, 7, 37, 40, 42, 61 Hamadanids, 61, 62 Hanafis, 79, 92, 108, 158 Hanbalis, 108, 158 Haniya, 81 Hara, 30 Al-Harafish, 157 Haram, 75, 76, 78–81, 87, 90, 93, 113 Harariq, 171 Harat al-Burjuwan, 30 Al-Harithy, Muhammad ibn ‘Abd al-Karim (geometrician), 40, 42 Al-Harraqa, 161 Harun al-Rashid (Caliph), 42, 148 Hasan Katkhuda (Amir), 92 Hasan Pasha (Governor), 92 Hasan, 22 Hatim ibn Harthama (Governor), 126 Al-Hayawan, 169

Hebron (al-Khalil), 40, 47 Herz, Max (architect), 107, 174, 183–5 High Pillars of Iram, 105 Hijaz, 135 hisar, 71, 107, 119 Hisham ibn ‘Abd al-Malik (Caliph), 136 Hisn Kifa, 61, 68 Hiyal handasiyya, 40 Hiyal, 168 Hollywood, 18 Homs, 61 Hugo, Victor (novelist), 31 Humphreys, R. Stephen (historian), 28, 121 Husn al-Muhadara fi Tarikh Misr wa-l-Qahira, 31 Hussayniyya Quarter, 73 Hussein, Taha (writer), 72 Ibn ‘Abd al-Salam, ‘Izz al-Din (judge), 7 Ibn ‘Abd al-Zahir (historian), 15, 27, 31, 56, 79, 87, 154, 155, 168 Ibn Abi Usaybu‘a (physician), 39 Ibn ‘Abid (master craftsman), 171 Ibn Aja, Shams al-Din (historian), 17 Ibn Amir al-Gharb, ‘Izz al-Din (polymath), 40 Ibn ‘Arab Shah (historian), 21 Ibn ‘Asakir (historian), 57, 149 Ibn Aybak al-Safadi, Khalil, Salah al-Din (historian), 16, 17, 18, 39, 168 Ibn ‘Aziz (painter), 167 Ibn al-Dawadari (historian), 13, 68, 127, 168 Ibn Dirbass, Sadr al-Din (judge), 79, 88 Ibn Duqmaq (historian), 21, 31 Ibn Durayd (lexicographer), 104 Ibn al-Furat (historian), 21 Ibn Ghana’im, Ibrahim (architect), 38 Ibn Hajar al-‘Asqalani (jurist), 21, 169 Ibn Hamdis (poet), 106 Ibn Hayyus, Abu al-Fityan (poet), 67 Ibn Kathir (historian), 24 Ibn Katib Sunqur, Sulayman ibn Ibrahim (judge), 16 Ibn Khaldun (philosopher of history), 6, 21–2, 29–31, 166 Ibn al-Muhandis (jurist), 37 Ibn Sa‘id al-Maghribi (traveller), 130 Ibn Shaddad, ‘Izz al-Din (historian), 20, 23, 27, 28, 32, 61, 63, 67, 154, 166, 168 Ibn al-Siyufi (architect), 89 Ibn Taghri-Birdi (historian), 13, 21, 28 Ibn Tulun (Governor), 148 Ibn al-Tuluni, Ahmad (master builder), 37 Ibn al-Tuluni, Muhammad (master builder), 37 Ibn Wasil (historian), 153 Ibn Zahira (historian), 31 Ibn al-Zain, Muhammad (master metalworker), 35

–255 –

IN DEX

Ibrahim Jamal al-Kufat (judge), 16 Ibrahim Pasha (General), 180, 181 Ibrahim, ‘Abd al-Latif (historian), 112 Ibrahim, Laila ‘Ali (architectural historian), 112, 117, 118 Ibrahim, Prophet, 62 Ifriqiya, 72, 80, 83, 86 Ikhshidids, 30, 82 Al-Ikhtisar, 87 Ilkhanids, 7, 23 ‘Ilm, 16, 60, 169 ‘Imran, 21 India, 173 Institut d’Égypte, 175, 176 ‘Iqd al-Juman fi Tarikh Ahl al-Zaman, 17 Iqta‘, 7, 40, 128, 157, 161 Iran, 59, 106, 148 Irbil, 61 Islamic architects and their Works, 38 Islamic architecture in Cairo: An Introduction, 144, 145 Islamic Revival Style, 174 Isma‘il (Khedive), 95, 96, 181–184 Isma‘il, al-Salih (Sultan), 16, 37 Isma‘ilis, 65, 83, 86, 87, 88 ‘Issa ibn Muhanna (Bedouin chief), 19 ‘Issa, al-Malik al-Mu‘azzam (King), 63 Al-Istakhri (geographer), 106 ‘Itaqa, 8 Istabl al-Tarima, 86 Istabl, 115, 117 Istanbul, 37, 91, 92, 180, 181 Italy, 42, 173 Iwan, 68, 69, 104–111–16, 118, 119, 123, 125, 127, 130, 131, 132, 135, 137, 150, 151, 153, 155, 156 Al-Iwan al-Ashrafi, 128, 157 Al-Iwan al-Kamili, 127, 155 Iwan Kisra, Iwan-i Kisra, 23, 104-7, 110, 11, 127, 137 Al-Iwan al-Mansuri, 128, 155 Iwan al-Qal‘a, 127 Ja‘bar, 61 Al-Jahiz (belletrist), 57, 169 Jamadariyya, 8, 158 Al-Jamhara fi-l-Lugha, 104 Jamil, 22 Janus (king), 165 Jarr al-‘amud, 171 Jashinkir, 4 Jawhar al-Qanqaba’i (Amir), 82, 90 Jawhar al-Siqilli (General), 72, 81, 82–4, 92 Jazira, 27, 61, 65, 68, 71, 137 Jerusalem, 20, 33, 40, 47, 54, 73, 136, 143 Jihad, 14, 15, 60, 64, 147, 162-5 Jomard, Edme François (editor), 129, 176

jund/jundi, 8, 13, 153 Ka‘ba, 57, 58 Kabir, 23 Kafur (Governor), 82 Karahiya, 168 Kashf al-Asrar, 55 Katib al-Sirr, 29, 158 Katkhuda Gate, 155 Kavala, 180 Kawwaz, 171 Kayseri, 61 Khalil, al-Ashraf (Sultan), 6, 15, 37, 55, 67, 68, 69, 118, 125, 126, 128, 137, 155–7, 163 Khamriyyat, 168 Khanqah, 25, 27, 60, 89, 108, 164 Khassakiyya, 37, 68, 69, 158 Khatib, 83, 153 Khawarij, 7 Khayer Bak (Amir), 116 Khaza’in al-qasr, 86 Khidma, 132, 162 Khirbat al-Mafjar, 135 Khitat, 12, 20, 21, 29, 30, 33, 82, 102, 112, 142, 143, 147 Khitat, al-Mawa‘iz wa-l I‘tibar bi-Dhikr al-Khitat wa-l-Athar, (Book), 6, 21, 27, 29, 30, 31, 31, 37, 80, 84, 115, 119, 142, 143, 144, 155, 167 Al-Khitat al-Tawfiqiyya al-Jadida, 142 Khizana, 116, 119 Khoshiar (Princess), 182, 184 Khosroe (Shah), 104, 127 Khushdash, 9, 10, 118 Khutba, 6, 163, 164 Al-Khuwakh al-Sab‘a, 83, 86 Khwarazmshah (Sultan), 65 Kitab al-kharaj, 148 Kitbugha, al-‘Adil (Sultan), 8, 10 Kleber (General), 94 Konya, 61, 64 Kostof, Spiro (architectural historian), 104 Kufa, 135 Kufic script, 75, 78 Kurds, 59, 60, 62, 88, 162 Kuttab al-Dast, 158 Kuttab, 12, 13, 15, 26, 28, 34, 166 Kutub al-Kharaj, 29, 102 al-Kutubi, Ibn Shakir (historian), 28, 57, 58 Kuwwa, 81 Labadye, Jean Baptiste-Auguste (architect), 177 Lajin, al-Mansur (Sultan), 118 Lala, 9 Lam‘i Mustafa, Salih (architect) Law of the Turks, 6

–256 –

IN DEX

Lawaqih al-Anwar fi-Tabaqat al-Akhyar, 91 Lebanon, 27, 40 Lézine, Alexandre (architectural historian), 80, 112, 130, 131 Literati, 12, 26, 28, 29, 32, 34, 38, 167–72 Little, Donald (historian), 112 Ma‘azil, 116, 118 Mabitat, 116, 119 Macedonia, 180 Mada’in (Ctesiphon), 23, 24, 105, 127 Madhhab, 26 Madrasa, 15, 27, 37, 56, 60, 66, 71, 88, 89, 90, 91, 107, 108, 109, 110, 164, 169, 171 Al-Madrasa al-Aqbughawiyya, 73, 89, 93, 95 Madrasa (and Mosque) of Sultan Hasan, 69, 71, 104, 106–11, 119, 177, 182, 183, 184, 187 Al-Madrasa al-Jawhariyya, 78, 81, 82, 89 Madrasa of Amir Khayer Bak, 116 Al-Madrasa al-Mu’ayyadiyya, 91 Al-Madrasa al-Nasiriyya (Madrasa of al-Nasir Muhammad), 109, 110 Madrasa of Amir Sarghitmish, 109, 110 Madrasa of Qalawun, 132, 133, 135, 136 Madrasa of Qansuh al-Ghuri, 71, 119 Al-Madrasa al-Salihiyya (Madrasa of al-Salih Najm al-Din), 109, 153 Al-Madrasa al-Taybarsiyya, 75, 89 Al-Madrasa al-Zahiriyya, 91, 109, 110 Maghribis, 82, 83 Mahdiyya, 80, 81 Maher, Su‘ad (art historian), 143, 144 Mahmud, al-Muzaffar (King), 40, 42 Mahumd of Ghazna (Sultan), 71 Majalis al-Adab, 169 Majlis, 69, 126, 130–2, 134 Makhmus, 111 Makram, ‘Umar (religious leader), 94 Malaka Khatun (Princess), 150 Malikites, 88, 109, 158 Malikshah (Sultan), 14, 149 Al-Mamalik al-Sultaniyya, 7 Mamluk Studies Review, 20 Ma’mun al-Bata’ihi (vizier), 84 Manafi‘, 116, 118 Manescalo, Alfonso (architect), 187 Manjak al-Yusufi (Amir), 171 Al-Mansur (Fatimid Caliph), 72 Al-Mansuriyya, 72, 82, 83 Manzara, 53, 62, 79, 116 Manzarat al-Azhar, 79, 83 Maq‘ad, 123 Maq‘ad of Mamay al-Sayfi, 123 Maqam Ibrahim, 62

Al-Maqrizi, Taqiyy al-Din (historian), 6, 7, 20, 21, 23, 27, 29, 30, 31, 32, 37, 38, 80, 81, 83, 84, 86, 109, 111, 113, 115, 117, 119, 120, 121, 127, 134, 142, 144, 155, 166–9, 172 Maqsura, 89, 90 Marafiq, 116 Marinids, 23 Markaz al-da‘wa, 87 Marseille, 142, 176, 177 Marwanids, 62 Masalik al-Absar fi-Mamalik al-Amsar, 29, 71 Masalik, al-Masalik wa-l Mamalik (geography treatises), 12, 20, 21, 29, 31, 33, 71, 102, 147 Mashhad al-Juyushi, 130, 131 Mashhad Sayyida Ruqayya, 130 Masjid, 90 Mastaba, 80, 81, 132, 154, 155 Al-Mawardi, Abu al-Hasan (jurist), 148 Mawkib, 67 Maydan al-Azhar, 73, 86, 95 Maydan, 61, 62, 63, 82, 113, 114, 119, 151, 155 Mayer, Leo (art historian), 34, 38, 40 Mayyafariqin, 61, 63 Mecca, 72, 73, 90, 107 Medina, 40, 73 Mediterranean, 47, 59, 63, 127, 143, 144, 162 Meinecke, Michael (architectural historian), 54, 56, 110, 136, 144 Men of the Pen, 42, 103 Merdasids, 61 Merv, 106 Mihrab, 56, 57, 76, 78–81, 97, 107, 129 Military Museum of Damascus, 40 Mi‘mar, 35, 36, 182 Minaret of ‘Abd al-Rahman Katkhuda, 93, 95 Minaret of al-Aqbughawiyya, 73, 74, 89, 93 Minaret of Qansuh al-Ghuri, 73, 74, 90, 91 Minaret of Qaytbay, 73, 74, 75, 90, 91 Minbar, 81, 107, 158 Ministry of Awqaf, 187 Misr (Egypt or Cairo), 30, 88 Misr, 84 Mongols, 4–10, 15, 16, 23, 24, 27, 29, 55, 59, 60, 63, 65, 69, 135, 147, 162–4, 171 Morea, 177 Mosque of ‘Ali Shah, 106 Mosque of ‘Amru, 80, 82, 87, 88 Mosque of Ibn Tulun, 80, 82, 87, 88 Mosque of Isfahan, 90 Mosque of Muhammad ‘Ali, 119, 156, 178–82 Mosque of al-Qarafa, 80, 81, 167 Mosque of Sultan Ahmet (Blue), 180 Mosque of Tankiz, 54 Mosul, 61, 67

– 2 57 –

IN DEX

Mshatta, 135 Mu‘alim, 35, 37 Mu‘alim al-Mu‘alimin, 37, 39 Al-Mu’ayyad (King), 37 Mu‘awiya (Caliph), 136, 148 Mubarak, ‘Ali Pasha (Minister, historian), 81, 142, 144, 182, 183, 184 Mudarraj Gate, 155 Mudhahhib, 35 Mufaddal ibn-Abi al-Fada’il (historian), 136 Mughal Empire, 173 Mughulatay (Amir), 118, 119 Muhammad ‘Ali (Governor), 94, 95, 119, 126, 129, 141, 155, 156, 162, 174, 176–183 Muhammad ‘Ali al-Kabir (Amir), 82 Muhammad, al-Kamil (Sultan), 40, 42, 63, 127, 153, 155 Muhammad ibn Marwan, Nasr al-Dawla (King), 62 Muhammad, al-Nasir (Sultan), 7, 8, 10, 11, 13–18, 26, 29, 36, 40, 71, 81, 89, 90, 109, 110, 113–15, 118, 120, 123, 125, 126, 128, 130, 132, 135–8, 146, 155–8, 160, 162, 164, 171 Muhammad Pasha al-Sharif (Governor), 92 Muhammad, The Prophet, 82, 102, 148 Muhammadan architecture in Egypt and Palestine, 143 Muhandis, 35–40, 42, 107 Muhandiskhana, 183 Muhiyy al-‘adl fi al-‘alamin, 164 Muhtasib, 23, 80, 158 Al-Mu‘izz (Caliph), 72, 83, 84, 86, 87 Al-Mu‘izziya, 5 Al-Mujahid, 164 Muntazam, 22 Al-Muqaddima, 29 Muqarnas, 38, 129, 132, 158, 167 Al-Muqqadasi (geographer), 57 Al-Muqtafi (Caliph), 149 Musa ibn Maymun (Maimonides) (physician), 88 Musa, al-Ashraf (King), 4, 63, 150 Musalla al-‘id, 82 Musawwir, 35, 170 Mushtaraf, 63 Al-Mustanjid (Caliph), 149 Al-Mustansir II (Egyptian Abbasid Caliph), 6, 163 Mu’talif, 22 Mutanasib, 22 Al-Mu‘tasim (Abbasid Caliph), 3 Muzammik, 35 Muzayyin, 35 Al-Nadira (Belvedere), 84 Na’ib, 150, 151, 153 Na’ib al-Qal‘a, 150 Na’ib al-Saltana, 7

Nafisa al-Bakriyya, 78, 93 Napoleon Bonaparte, 93, 142, 173–5, 182, Napoleon III, 182 Naqib al-ashraf, 94 Naqib al-juyush, 90 Naqqash, 35, 42 Al-Nas, 13 Al-Nashw, 171 Al-Nasir li-Din Allah (Caliph), 65, 149 Nazir al-Jaysh (army supervisor), 158 Neo-Islamic Style, 96, 174 Neo-Mamluk Style, 95, 173, 174, 175, 177, 180, 181, 182, 187 Nights of Illuminations, 82, 84 Nizam al-Mulk al-Tusi (vizier), 14, 149 North Africa, 23, 91, 126, 164, 173 Notre Dâme de Paris, 31 Nubia, 164 Nubian (language), 17 Al-Nu‘man ibn Muhammad (judge), 87 Al-Nujum al-Zahira fi Muluk Misr wa-l-Qahira, 28 Nuruosmaniye Mosque, 180 Nusretiye Mosque, 180 Al-Nuwayri (Encyclopedist), 24, 166, 169 Oghuz Turkish (language), 16 Orientalism, 173 Orientalism, 173 Orontes, 39 Ottoman Empire, 73, 173, 175, 181 Ottoman Turkish (language), 16 Ottomans, 8, 31, 34, 88, 91–94, 102, 115, 116, 119, 121, 126, 129, 177, 180 Oxus River (Amu Darya), 59 Palace of Abu Muslim al-Khurasani, 106 Palace of Alin Aq, 115–119, 121 Palace of Bashtak, 115, 116, 121, 123 Palace of Hasan Kashef, 175 Palace of Qawsun, 115, 121 Palace of Taz, 115, 121 Palace of Theodoric, 48 Palais et maisons du Caire, 115 Palermo, 126 Palestine, 27, 47, 66, 72 Palmyra, 61, 64, 105 Paris, 142, 175, 182 Pishtaq, 75, 78, 79 Pococke, Richard (traveller), 129 Police Museum, 156 Princely Cycle, 67, 68 Qa‘a, 27, 28, 47, 62, 63, 71, 113, 114, 116, 118–21, 130–2, 150, 151, 158

–258 –

IN DEX

Qa‘a (Citadel of Cairo), 48, 51–53, 55, 56 Qa‘a musaqqafa ‘aqdan, 116 Qa‘at al-Dhahab, 134 Qa‘at julus, 116 Qa‘at al-Sahib, 153 Qada’ al-Mazalim (al-Nazar fi-l Mazalim), 146, 147, 150, 152–157, 160, 162 Qadi, 7, 16, 79, 88, 148–50, 154 Al-Qadi al-Fadil (Vizier), 14 Qadi al-Jaysh or Qadi al-‘Askar, 17, 153 Qadi al-Qudat, 79, 86, 148 Al-Qadir (Caliph), 148 Al-Qahira (Fatimid Cairo), 30, 79–84, 87–9, 137, 141, 153 Al-Qa’im (Caliph), 148 Qal‘a, 56 Qalawun, al-Mansur (Sultan), 5–7, 9, 10, 15, 17, 24, 37, 48, 55, 56, 68, 89, 113, 125, 126, 128, 132, 135, 136, 144, 146, 155–7, 161–3, 187 Al-Qalqashandi (Encyclopedist), 21, 113, 166, 169 Qanatir al-Siba‘, 168 Qansuh al-Ghuri, al-Ashraf (Sultan), 37, 71, 73, 74, 78, 81, 87, 90, 91, 119, 126 Qara, 8 Qara Arslan (King), 68 Al-Qarafa al-Kubra, 89 Al-Qarafa al-Sughra, 89 Qaraqush (Amir), 36 Qarmatians, 82 Al-Qasir (painter), 167 Qasr, 82, 112–124, 126 Qasr al-Ablaq (Cairo), 113–15, 119–21, 129, 135, 136 Qasr al-Ablaq (Damascus), 38, 135, 136 Qasr al-Shawk Gate, 86 Al-Qazwini, Zakariyya (Cosmographer), 57, 169, 170 Qawsun (Amir), 9, 17, 18, 38, 39, 69, 115, 121, 165, 171 Qayrawan, 72, 80 Qaytbay, al-Ashraf (Sultan), 26, 73, 74, 75, 80, 82, 90–2, 123, 162 Qibla, 48, 76, 78, 113, 130, 132, 177 Qipchaq, 4, 7–10, 16 Qubba, 48, 50, 56, 81, 125, 126, 127, 128, 135, 137 Al-Qubba al-Mansuriyya, 56, 128 Qubba al-Sa‘ida, 127 Al-Qubba al-Zahiriyya (Cairo), 67, 128 Al-Qubba al-Zahiriyya (Damascus), 47–53, 55, 56 Qubbat al-Hawa, 126 Qubbat al-Khadra, 126, 136, 137 Al-Quda‘i (Hisorian), 80 Al-Qudsi, abu Hamid (historian), 20, 28, 31, 32, 61, 63, 67, 154, 166, 168 Qulla Gate, 153 Qur’an, 8, 40, 92, 109, 170, 177

Qur’anic inscriptions, 75, 78, 89, 95, 97 Qurra’, 92 Al-qusur al-juwwaniyya, 113, 114 Al-Qusur al-Zahira, 83, 84 Qutuz (Sultan), 5, 6, 15, 150 Rambles in Cairo, 144 Rank, 15, 60, 66, 68, 69, 166–8, 171 Rassam, 35 Al-Rawd al-Zahir fi Tarikh al-Malik al-Zahir, 27 Renaissance Italy, 42 Ribat, 40, 60, 164 Ridwan (King), 63 Rifa‘a al-Tahtawi (reformer), 95 Al-Rifa‘i Mosque, 96, 174, 182, 183, 187 Riwaq al-‘Abbasi, 95 Riwaq al-‘Abbasi, 95 Riwaq al-Shraqawiyya, 82 Riwaq, 56, 78, 81, 90, 92, 95, 118, 119, 121, 128, 130 Roda Citadel, 130 Rogers, Michael (art historian), 112, 144 Roman, 76, 80, 132, 133, 160 Rumi, Jalal al-Din (poet), 181 Rusafa (Sergiopolis), 126, 134, 136 Russia, 3 Al-Rustumi, Abu Sa‘id (poet), 105 Sabil/Kuttab, 92, 93, 177 Sa‘di (poet), 181 Sahib al-‘Ama’ir, 92 Sahib al-mazalim, 148 Sahn, 72, 75, 76, 78, 97, 119 Said, Edward (critic), 173 Al-Sakhawi (historian), 21, 82 Salah al-Din al-Ayyubi, Yusuf, al-Malik al-Nasir, 4, 14, 15, 27, 36, 39, 61, 62, 79, 88, 108, 141, 146, 147, 149, 150, 151, 162, 163, 164 Salamlik, 119 Salar (Amir), 89, 118, 119, 121 Al-Salih Tala’i‘ Mosque, 177 Al-Salihiyya (Mamluks and Amirs of al-Salih Najm al-Din), 4, 9 Al-Sallami, al-Majd (slave trader), 7 Samanids, 71 Samarra, 80, 130, 134, 167 Sani‘, 35 Sanjar al-Shuja‘i (Amir), 10 Saqaliba, 4 Saqi, 68 Sasanians, 104, 127, 135, 137, 148 Sauvaget, Jean (historian), 134 Sayed, Hazem (architectural historian), 112 Sayem al-Dahr, Muhammad, 168 Sayf al-Dawla al-Hamadani (King), 61

–259 –

IN DEX

Sayyida Nafisa Mosque, 96, 187 Sayyida Zaynab Mosque, 96, 187 Selim I (Sultan), 37, 71, 91, 107, 119, 181 Seljuq, 4, 5, 7, 14, 59, 61–63, 66, 112, 148, 149, 162, 167 Sha‘ban, al-Kamil (Sultan), 123 Shadd, 23, 35–37 Shafe‘i, Farid (architectural historian), 143, 144 Shafi‘ites, 79, 88, 89, 109, 158 Shahnameh of Firdawsi, 71 Shajar al-Durr (Sultana), 4 Shams al-Din, al-Sharif (judge), 153 Shari‘ al-Muski, 182 Shari‘ Muhammad ‘Ali, 182 Shari‘a, 6, 96, 148, 163 Al-Sha‘rani, ‘Abd al-Wahab (Sufi), 91 Al-Sharq, 10 Al-Shawbaki, Taj al-Din (Amir), 80 Shaykh, al-Mu’ayyad (Sultan), 90, 91 Sheikh al-Rabwa al-Dimashqi (geographer), 57 Shi‘ite, 59, 83, 88, 148, 162 Shukhshikha, 131 Al-Shuja‘i, ‘Alam al-Din (Amir), 37 Al-Shuja‘i, Shams al-Din (historian), 13 Al-Sikka al-Jadida, 182 Silahdar, 68 Silahdariyya, 8, 158 Sinjar Niche, Gu’ Kummet, 68, 69 Sinjar, 61, 68, 69 Sivas, 61 Siyasat Namah, 14, 149 Spain, 47, 173 Sphinx, 168 St-Appolinaire-Nuevo, Ravenna, 48 Al-Subki, Taj al-Din (jurist), 7 Sudan, 4, 180 Sudun Ba‘jad (Amir), 118, 119 Suffa, 104, 131 Sufi, 13, 60, 91, 164, 168 Sulayman al-Halabi, 94 Suleyman the Magnificent (Sultan), 181 Sunni, 59, 82, 87, 88, 108, 149, 158, 162, 164 Sura of the Believers, 97 Sura of the Cattle, 97 Surat al-Ard, 29, 102 Susun (Amir), 18 Suwayqat Mun‘im, 118 Al-Suyuti (historian), 21, 31 Syria, 3–5, 7, 9–11, 14, 27, 33, 47, 55, 61, 66, 82, 89, 125, 134, 135–7, 147, 148, 163, 167, 181 Ta‘asif, ‘Alam al-Din Qaysar (polymath), 39, 40, 42 Tablakhana, 155 Tabriz, 7, 106

Tahrim, 168 Tajdid, 96 Tajziyat al-Amsar wa Tazjiyat al-A‘sar, 163 Takhayen, 116, 118 Takht, 126 Takrit, 61 Takruri (language), 17 Talisman Gate, 65, 66 Tankiz (Amir), 40, 42, 54 Tarikh Madinat Dimashq, 57 Tawfiq (Khedive), 75, 78, 82, 95 Taybars al-Waziri (Amir), 89, 90 Taymur, Ahmad Pasha (historian), 35 Taz (Amir), 115 Tha‘lab, Husn al-Din (Amir), 7 Tibaq, 8-10, 118 Tigris River, 68 Transoxania, 3 Treaty of London, 181 Tree of Life, 66 Triclinia, 135 Tricorium, 135 Tripoli (Lebanon), 20, 47 Tuhfat al-Anam fi Fada’il al-Sham, 57 Tulunids, 30, 127, 130 Tunisia, 72 Turan Shah (Sultan), 4 Turbat al-Za‘faran, 86 Turkoman Turkish (language), 16 Tutush, Taj al-Dawla (King), 63 Ulama, 10, 12–15, 26, 28, 29, 32, 34, 37, 38, 60, 79, 91, 93, 94, 170 Al-‘Ulmawi (historian), 57 Al-‘ulum al-‘aqliyya, 170 ‘Umar b. ‘Abd al-‘Aziz (Caliph), 148 Al-‘Umari, Ibn Fadl Allah (historian), 24, 29, 32, 54, 71, 113, 114, 119, 120, 157, 158, 166, 169 Umara‘ al-Mashura, 158 Umayyad Mosque of Damascus, 47, 48, 53, 54, 56, 58 Umayyads, 47, 133, 135, 136, 148, 167 Upper Egypt (Sa‘id), 78 Ustadar, 90 Ustadh, 9, 10 ‘Uthman Katkhuda al-Qazdughli (Amir), 92 ‘Uyun al-Anba fi Tabaqat al-Atibba, 39 Uzbek (Sultan), 8 Vaudoyer père, Antoine-Laurent-Thomas (architect), 177 Al-Wafi bi-l-Wafi yyat, 18, 39 Wafidiyya (Mongols), 8, 9, 10

–260 –

IN DEX

Wakil bayt al-mal, 158 Wali, 80, 91, 92 Al-Walid ibn Yazid (Caliph), 135 Waqf, 15, 24, 25, 83, 86, 88, 89, 92, 101–3, 109, 112, 116–19, 121, 155, 171, 187 Waqf of Amir Khayer Bak, 116, 117, 119 Waqf of Sultan Barsbay, 116–19 Waqf of Ibrahim Aga Mustahfizan, 116 Al-Waqidi (historian), 17 Wasit, 126, 136 Wassaf al-Hadra, Sharaf al-Din (historian), 163 Western Palace (al-Qasr al-Gharbi), 84 William II (King), 126 Yalmaq, 68 Ya‘qub ibn Kilis (vizier), 84, 87 Yaqut al-Hamawi (geographer), 57 Yashbak min Mahdi (Amir), 17 Al-Yazuri (Vizier), 167

Yusuf II, al-Nasir Salah al-Din (King), 27, 152 Al-Yusufi, Musa ibn Yahya (historian), 14, 19 Al-Zafir (Caliph), 84 Zaghlish al-Shami, Muhammad (architect) 38, 39 Zaghlul, Sa‘d (Prime Minister), 96 Al-Zahabi (historian), 24, Al-Zahira (belvedere), 84 Al-Zahiri, Khalil (historian), 31, 107 Zajal, 171 Zakarya, Mona (architectural historian), 112 Zaki, ‘Abd al-Rahman (art historian), 143, 144 Zardakash, 40 Zawiya of Hasan al-Rumi, 155 Zengids, 5, 14, 61, 136, 146, 147, 149, 162 Zetterstéen, K.V. (historian), 13 Ziyada, 89, 90 Zubdat Kashf al-Mamlik wa-Bayan al-Turuq wa-l-Masalik, 31

– 2 61 –

IN DEX

–262–