# Cities as Palimpsests?

Oxford & Philadelphia

# Cities as Palimpsests?

Responses to Antiquity in Eastern Mediterranean Urbanism

*Edited by* Elizabeth Key Fowden, Suna Çağaptay, Edward Zychowicz-Coghill and Louise Blanke

Published in the United Kingdom in 2022 by OXBOW BOOKS The Old Music Hall, 106–108 Cowley Road, Oxford, OX4 1JE

and in the United States by OXBOW BOOKS 1950 Lawrence Road, Havertown, PA 19083

© Oxbow Books and the individual contributors 2022

Hardback edition: ISBN 978-1-78925-768-7 Digital Edition: ISBN 978-1-78925-769-4

A CIP record for this book is available from the British Library

Library of Congress Control Number: 2021949463

An open-access on-line version of this book is available at: http://books.casematepublishing.com/ Cities\_as\_Palimpsests.pdf. The online work is licensed under the Creative Commons Attribution 3.0 Unported Licence. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/ or send a letter to Creative Commons, 444 Castro Street, Suite 900, Mountain View, California, 94041, USA. This licence allows for copying any part of the online work for personal and commercial use, providing author attribution is clearly stated.

Some rights reserved. No part of the print edition of the book may be reproduced or transmitted in any form or by any means, electronic or mechanical including photocopying, recording or by any information storage and retrieval system, without permission from the publisher in writing.

Materials provided by third parties remain the copyright of their owners.

Printed in the United Kingdom by Short Run Press

Typeset in India by Lapiz Digital Services, Chennai.

For a complete list of Oxbow titles, please contact:

UNITED KINGDOM UNITED STATES OF AMERICA Oxbow Books Oxbow Books

Telephone (01865) 241249 Telephone (610) 853-9131, Fax (610) 853-9146 Email: oxbow@oxbowbooks.com Email: queries@casemateacademic.com www.oxbowbooks.com www.casemateacademic.com/oxbow

Oxbow Books is part of the Casemate Group

*Front and back cover artwork by Sofia Greaves.*

## Contents



## Series preface

The present series of three volumes was made possible by a five-year Advanced Grant from the European Research Council under the European Union's Horizon 2020 research and innovation programme (grant agreement No 693418). By an ironic twist of fate, our 'Impact of the Ancient City' proposal was submitted to and accepted by the ERC at the very period when many in the UK seemed to have lost sight of the power of the ideals of free movement of goods, people and ideas. The University sector in general – and Cambridge in particular, the Classics Faculty of which hosted the project – felt that popular rhetoric had lost sight of solid advantages. The benefits reaped from the circulation of people and ideas go beyond the economic. The academic enterprise is and always has been a cosmopolitan one, and it is a relief that our government has agreed that this is one aspect of the old relationship with Europe worth preserving. Our gratitude to the European Union is therefore heartfelt.

In Europe and across the globe today there is tremendous interest in urbanism as a defining feature of our world, but often without sensitivity to the historical depth of cities. What we proposed was to think again about the relationship between cities with a Greco-Roman past and the long history of urbanism across the Mediterranean that has continued to the present. To do this, we felt it would not help to suggest a single story line. The story of 'Classical reception' increasingly concerns Classicists who, when challenged on the relevance of this past world to the present, point to a long and changing story of relevances. Strangely enough, there has been surprisingly little attention given to the 'reception' of ancient, Greco-Roman urbanism. To fill that gap, what we hoped to explore was how the city is not only a fundamental characteristic of Greco-Roman civilisation, but has acted as a vital mechanism by which that civilization was generated, transmitted and transmuted. Our project is about understanding changing responses to the urban past over the duration of two millennia, with a focus on the Mediterranean region.

The ERC Advanced Grant presented us with the exceptional opportunity to be ambitious in both scope and range while creating a small community of scholars with expertise from different periods and areas that reached beyond the capacities of any single scholar. From the outset the project was designed to range chronologically from late antiquity to the present, geographically across the Mediterranean, east and west, culturally across the Christian and Islamic worlds, and in disciplinary terms across the study of texts and physical remains. Despite the generous support, we soon discovered that it was impossible to do more than sample this vast area, selecting a group of scholars who both complemented and challenged each other: a late antique archaeologist specializing in Visigothic Spain (Javier Martínez Jiménez), an early medieval historian focusing on relations between the courts of Charlemagne and Umayyad Spain (Sam Ottewill-Soulsby), an Arabist and historian of the medieval Middle East (Edward Zychowicz-Coghill), an archaeologist working on late antique and early Islamic Jordan and Egypt (Louise Blanke), an architectural historian exploring the transition from Byzantine to Ottoman (Suna Çağaptay), a late antique historian who has turned her attention to Ottoman Greece (Elizabeth Key Fowden), a PhD student with a background in Classics studying urban planning in nineteenth- and early twentieth-century Italy (Sofia Greaves) and a principal investigator specializing in Roman social history and urban archaeology in Italy (Andrew Wallace-Hadrill).

Other Cambridge colleagues joined our discussions on a regular basis, notably Tom Langley, writing a PhD on ideas of the city in Greek Patristic writers, Professor Amira Bennison, a historian of the medieval Maghrib, especially its cities, Professor Rosamond McKitterick, a leading figure in the study of Carolingian France and papal Rome, and Professor Martin Millett, a Roman archaeologist with a longstanding interest in urbanism. We benefited from the support and advice of the members of our Advisory Committee, both in Cambridge (in addition to the above named, Cyprian Broodbank, Robin Cormack, Garth Fowden, Alessandro Launaro, Robin Osborne and John Patterson) and beyond – Luuk de Ligt (Leiden), Çiğdem Kafescioğlu (Istanbul), Ray Laurence (Sydney), Keith Lilley (Belfast) and from Oxford, Josephine Quinn, Bryan Ward-Perkins and Chris Wickham. We also enjoyed the invaluable support of two administrators, Nigel Thompson of the Classics Faculty and Beth Clark, whose calm efficiency facilitated conferences and seminars, enabled foreign travel and smoothed contact with the bureaucracies at both ends.

We invited many scholars, from Cambridge or further afield, to share their knowledge with us at our weekly seminars. We also organised one-day workshops, including one on the Roman and Islamic city in North Africa and one on Cities and Citizenship after antiquity (that led to an *Al-Masāq* special issue)1 , as well a panel for the 2018 Leeds International Medieval congress on 'Memory' and two three-day conferences, one in Istanbul and one in Rome. The last three underlie the three volumes in the present series. In each of those conferences, the members of our group contributed, but we knew that to cover the ground we needed to bring in international colleagues. The three volumes that constitute the present series are far from exhausting the output of the project, and each of us has papers and monographs in the pipeline or already out. Each of the three volumes has its own set of questions, but together they build up an overriding collective agenda of exploring how the cities of the Greek and Roman past, and such ideas of the city that were articulated around them, have impacted on the city and the idea of the city in later periods.

> *Andrew Wallace-Hadrill 29 July 2021*

<sup>1</sup> Javier Martínez Jiménez and Sam Ottewill-Soulsby, *Cities and Citizenship after Rome*, *Al-Masāq. Journal of the Medieval Mediterranean* vol. 32 no. 1 (2020).

## Acknowledgements

First and foremost the co-editors would like to thank Andrew Wallace-Hadrill, the Principal Investigator of the ERC Advanced Grant 'Impact of the Ancient City', as well as the other members of the project, Sofia Greaves, Javier Martínez-Jiménez, Sam Ottewill-Soulsby, our adopted member, Thomas Langley, our advisory board members Rosamond McKitterick and Amira Bennison, who have regularly participated in our weekly seminars, our project administrator, Beth Clark, particularly for her work preparing this volume, and finally the Faculty of Classics, for providing our project with a stimulating environment.

The origins of this volume can be traced back to a three-day conference in May 2018, generously hosted by Bahçeşehir University (BAU). Thanks to Enver Yücel, founder and chairman of BAU, the conference took place at the main campus of BAU overlooking the Bosphorus with views of the old city and the new. We are grateful to Necdet Kenar, secretary general of BAU and the students from the Faculty of Architecture and Design at BAU who worked in conference management; Berna Argun Habib for her planning and Hidayet Softaoğlu for help in designing the poster and booklet for the event.

Several scholars presented papers at the conference, but were not able to submit their papers for this collaborative volume, as they were either part of ongoing projects or promised for publication elsewhere: for their participation we thank Sotirios Dimitriadis, Asa Eger, Ahmet Ersoy, Çiğdem Kafescioğlu, Emily Neumeier, Alessandra Ricci and Alan Walmsley. Hugh Kennedy is to be thanked for drawing out themes and problems that emerged in the first day, Arietta Papaconstantinou for performing the same increasingly challenging task at the end of the second day and finally Cemal Kafadar for his concluding remarks that reflected on what we had learned over the three-day event, and what more we might want to explore. Our gratitude to Bob Ousterhout for a learned tour of the Chora Monastery and to Suna Çağaptay, not only for overseeing the conference organisation with Beth Clark, but also for devising and leading our project site trips in Istanbul and in western Anatolia which included Nicomedia/İzmit, Nicaea/İznik, Prusa/Bursa, Ephesus/Efes and Magnesia/Manisa. The fact that no one collapsed from mental or physical exhaustion is a tribute to Suna's humour, inventiveness and knowledge.

## Illustrations


xii *Illustrations*



xiv *Illustrations*



xvi *Illustrations*


## Contributors

#### **Co-editors**

Islam in the Greek world is an abiding theme in Elizabeth Key Fowden's research, which draws on architectural, visual and textual sources to analyse cultural exchange. Spanning from late antique Syria to Ottoman Greece (with an excursus into contemporary Arab art), her publications include *The Barbarian Plain: Saint Sergius between Rome and Iran* (1999), *Studies on Hellenism, Christianity and the Umayyads* (with Garth Fowden, 2004), 'The lamp and the wine flask: Early Muslim interest in Christian monasticism' (2007), 'Jerusalem and the work of discontinuity' (2019) and 'The Parthenon Mosque, King Solomon and the Greek Sages' (2019). She is Senior Researcher on the Impact of the Ancient City project at the University of Cambridge and College Post-Doctoral Associate at Jesus College, Cambridge.

Suna Çağaptay is a medievalist working on artistic and cultural interactions in the eastern Mediterranean and their reflections on the built environment. Currently, she is a research associate on the Impact of the Ancient City project at the University of Cambridge and a postdoctoral research associate at St Edmund's College, Cambridge. She is the author of *The First Ottoman Capital: The Religious, Architectural and Social History of Bursa* (2020), as well as several articles focusing on aspects of medieval frontiers, spolia and identity appearing in *Muqarnas*, *Byzantine and Modern Greek Studies Journal*, and *Dumbarton Oaks Papers*. She teaches in the Faculty of Architecture and Design at Bahçeşehir University (BAU), Istanbul.

Edward Zychowicz-Coghill is Lecturer in the History of Asia at King's College London. He was previously a research associate on the Impact of the Ancient City project. He is a cultural and intellectual historian of the early Islamic world whose work encompasses early Arabic historiography, visions of the pre-Islamic past, and economic imaginaries. Publications include *The First Arabic Annals: Fragments of Umayyad History* (2021) and *Writing the Conquest of Egypt: The Formation of Early Islamic Historiography* (forthcoming).

Louise Blanke is Lecturer in Late Antique Archaeology at the University of Edinburgh and previously a research associate on the Impact of the Ancient City project at the University of Cambridge. She has also held postdoctoral positions at the University of Oxford (sponsored by the Danish Carlsberg Foundation) and at the University of Aarhus. She is the author of *An Archaeology of Egyptian Monasticism: Settlement, Economy and Daily Life at the White Monastery Federation* (2019), *Monastic Economies* (edited with Jennifer Cromwell, forthcoming) as well as several articles on urbanism and monasticism in the Late Antique Near East. She has directed archaeological projects and participated in fieldwork at sites in Egypt, Denmark, Jordan and Qatar. She currently directs the ongoing *Late Antique Jerash Project.* 

#### **Contributors**

Göksun Akyürek is an architect and architectural historian. Her research interests and publications include nineteenth- and twentieth-century practices of history and architectural production of various scales, mostly focusing in Istanbul and Turkey. Currently, she teaches basic design, architectural design and architectural history at Bahçeşehir University in Istanbul. She holds a PhD in architectural history and theory from the Yıldız Technical University in Istanbul and an MA also in architectural history and theory from the Middle East Technical University (METU) and a BA again from METU.

Benjamin Anderson is Associate Professor of the History of Art and Classics at Cornell University. He studies late antique and Byzantine art and architecture, the urban history of Constantinople and the history of archaeology. Recent publications include an edition of Robert Wood's *The Ruins of Palmyra and Baalbek* (2021) and *The Byzantine Neighbourhood: Urban Space and Political Action* (2021), the latter co-edited with Fotini Kondyli.

Gideon Avni is Chief Archaeologist for the Israel Antiquities Authority and Professor of Archaeology at the Hebrew University in Jerusalem. His academic interests focus on various aspects of Classical, Late Antique, Early Islamic and medieval archaeology, cultural and religious transformations, the diffusion of technologies and movement of people in the Near East and beyond. His recent books are *The Byzantine – Islamic Transition in Palestine, an Archaeological Approach* (2014) and *A New Old City – Jerusalem in the Late Roman Period* (2017).

Nikolas Bakirtzis is Associate Professor at The Cyprus Institute in Nicosia. His research and publications focus on Byzantine monasticism, medieval cities and fortifications, and the island landscapes of the Byzantine, medieval and early modern Mediterranean. More recently his work explores issues of heritage and cultural identity in historic cities. As the Director of the Andreas Pittas Art Characterization Labs at the Cyprus Institute he is leading research on aspects of the history, materiality and provenance of medieval and early modern works of art enhanced through the use of advanced digital and analytical methods.

Amira K. Bennison is Professor in the History and Culture of the Maghrib at the University of Cambridge and a fellow of Magdalene College, Cambridge. Her research interests include the medieval Islamic West, Maghribi modes of legitimation and cultures of power and nineteenthcentury Muslim religio-political engagement with modernity. Her publications include *The Almoravid and Almohad Empires* (2016), *The Articulation of Power in Medieval Iberia and the Maghrib* (2014), *The Great Caliphs* (2009), *Cities in the Premodern Islamic World*, edited with Alison L. Gascoigne (2007), *Jihad and its Interpretations in Pre-Colonial Morocco* (2002), as well as numerous articles. Amira also contributes regularly to television programmes on Islamic history and is a frequent guest on BBC Radio 4's 'In Our Time'.

Ine Jacobs is the Stavros Niarchos Foundation Associate Professor of Byzantine Archaeology and Visual Culture at the University of Oxford. She received her doctorate from the University of Leuven. Her research interests include Roman and Byzantine architecture and urbanism, the experience and perception of the built environment and its decoration, long histories of display and reception of sites, statuary and artifacts and material religion. She has worked on excavations in Belgium, Italy, the Republic of North Macedonia and Turkey. She was a member of the Sagalassos team between 2003 and 2014. In 2016 she became field director of the Aphrodisias excavations.

Dimitri J. Kastritsis is Lecturer in Ottoman and Middle Eastern History at the University of St Andrews. He is the author of several books and articles on Ottoman history, historiography and literature in the fifteenth and sixteenth centuries, including *The Sons of Bayezid* (2007) and *An Early Ottoman History* (2017). He is currently working on a monograph about early Ottoman historical writing and ideas of the past, for which he has received support from the British Academy and the Institute for Advanced Study at Princeton.

Paul Magdalino is Emeritus Professor of Byzantine History at the University of St Andrews and a Fellow of the British Academy. He has worked on many aspects and most periods of Byzantine history, with a concentration on the ninth to twelfth centuries. His numerous publications have focused especially on imperial ideology and court culture, the textual evidence for lost monuments, astrology and prophecy, and the topography and development of the city of Constantinople.

Sam Ottewill-Soulsby is a research associate on the Impact of the Ancient City project at the University of Cambridge, where he studies urban ideals in the western European Mediterranean. He is also a fellow of Darwin College and Associate Member of the Faculty of History. He has published on late antique and medieval urbanism, early medieval diplomacy and foreign policy, and charismatic megafauna, including *The Emperor and the Elephant: Carolingian Diplomacy with the Islamic World* (forthcoming) and *Remembering and Forgetting the Ancient City* (edited with Javier Martínez Jiménez, forthcoming).

Robert G. Ousterhout is Professor Emeritus in the History of Art at the University of Pennsylvania. He is the author most recently of *Visualizing Community: Art, Material Culture, and Settlement in Byzantine Cappadocia* (2017) and *Eastern Medieval Architecture: The Building Traditions of Byzantium and Neighboring Lands* (2019) as well as co-editor of *Piroska and the Pantokrator*, with M. Sághy (2019) and *The Holy Apostles: A Lost Monument, a Forgotten Project, and the Presentness of the Past*, with M. Mullett (2020). His fieldwork has concentrated on Byzantine architecture, monumental art, and urbanism in Constantinople, Thrace, Cappadocia and Jerusalem.

Scott Redford is Nasser D. Khalili Professor of Islamic Art and Archaeology at the School of Oriental and African Studies, University of London. He studies the medieval art, architecture and archaeology of Turkey and the eastern Mediterranean. His interest in medieval encounters with remains of earlier civilizations in this region stems from his twin interests in archaeology and architecture.

Helen Saradi specialises in the civilization and social history of Byzantium and combines an interdisciplinary approach to a broad range of topics. Her research interests focus on Byzantine urbanism, the rhetoric of Byzantine cities and the acts of private transactions and related social aspects. She is the author of three monographs: *The Byzantine City in the Sixth Century: Literary Images and Historical Reality* (2006), *Notai e documenti greci dall'età di Giustiniano al XIX secolo* (1999) and *Le notariat byzantin du IXe au XVe siècles* (1991). She has published over seventy papers and four conference proceedings, and has lectured and participated in research projects at various academic institutions.

Assaad Seif is Associate Professor of Archaeology at the Lebanese University. He is a scientific adviser to the Lebanese National Council of Scientific Research and an editorial advisory board member of the journal *Conservation and Management of Archaeological Sites*. His research focuses on the history of archaeology and the modern uses of archaeological heritage, engaging theoretical frameworks and models in exploring the interactions between archaeology, heritage and urban space. Since 2005 he has focused on the theoretical debate regarding heritage construction processes and their ideological uses in relation to identity-building and post-conflict reconstruction.

Bethany J. Walker is Professor and Director of Islamic Archaeology at the University of Bonn. She is founding editor of the *Journal of Islamic Archaeology*, founding co-editor of Equinox's *Monographs in Islamic Archaeology* and co-editor of the recently published *Oxford Handbook of Islamic Archaeology*. She serves on numerous American and French editorial boards and is a long-time board member of the American Center of Research in Amman. Her monograph *Jordan in the Late Middle Ages: Transformation of the Mamluk Frontier*  was published in 2011. Director of Excavations at Tall Hisban, her field career in Jordan spans over two decades.

Introduction

## Chapter 1

## Historical distance, physical presence and the living past of cities

*Elizabeth Key Fowden, Suna Çağaptay, Edward Zychowicz-Coghill and Louise Blanke*

*How many a city We have destroyed in its evildoing, and now it is fallen down upon its turrets! How many a ruined well, a tall palace!*

*Qur'an* 22.45*<sup>1</sup>*

*It is not quite known: Is it the work of humans for jinn to live in or the work of jinn for humans?*

al-Buḥturī, *Īwān Kisrā <sup>2</sup>*

*Ruins that don't take you back to the past, but coexist on the same plane as buildings still living.* Nikos Gabriel Pentzikis, 'Thessaloniki and life'3

*Time is made to curl up end to end, so that distance draws near and the past becomes present; depth disappears in a flattening effect that brings up to the surface what once lay buried.*

Marina Warner, 'Freud's couch: A case history'4

### **A metaphor for time and space in flux**

Contradiction, paradox, disjunction: eastern Mediterranean cities seem to offer these qualities in profusion, attracting visitors in pursuit of the immersive experience of

<sup>1</sup> Tr. Arberry.

<sup>2</sup> Tr. Ali in 'Reinterpreting al-Buḥturī', line 43 (translation, 64; transliteration, 67).

<sup>3</sup> Extract from Pentzikis, *Mother Thessaloniki*, originally published in 1970, tr. Marshall, 178.

<sup>4</sup> Warner, 'Freud's couch', 158.

time compressed in place. We are fascinated to see buildings whose ancient and medieval shapes seem to collide with twenty-first century additions; to encounter the pagan, Jewish, Christian, Muslim and secular side by side. We want to feel the fusion of many disparate parts into a bricolage, a visual power clash of past and present in which we celebrate the separate parts for their distinctive historical contributions. This 'we' is not only tourists, eager to record experiences on Instagram, but also scholars concerned with understanding how the past accumulates in the present. Palimpsest is the word we reach for as shorthand for the historical complexity and cultural hybridity of the eastern Mediterranean city. But does this fashionable trope slyly force us to see contradiction where local inhabitants saw (and see) none, to impose distinctions that satisfy our own assumptions about historical periodisation and cultural practice, but bear little relation to the experience of ancient, medieval, early modern or modern persons? By visualising the city as distinctive strata do we blind ourselves to the porosity of urban tissue?

### *The history of a metaphor*

To imagine a city as a palimpsest is a mental experiment most commonly associated with Sigmund Freud. Concerned with preservation (*Erhaltung*) in the sphere of the mind he opened *Civilisation and its discontents*, first published in 1930, by playing with Rome as a 'psychical entity' in which 'all the earlier phases of development continue to exist alongside the latest one', as all experiences and memories co-exist preserved in the human mind.5 Freud soon abandoned the fanciful comparison as idle and absurd, arguing that 'if we want to represent historical sequence in spatial terms we can only do it by juxtaposition in space: the same space cannot have two different contents'. What is vital in this short passage for understanding eastern Mediterranean urban evolutions is that, without ever using the word palimpsest, Freud's fantastical game highlights the interpreting role of the informed viewer confronted with deciphering the thickly sedimented city. The human eye is needed to focus on and distinguish buildings, streets and clearings that were originally constructed in temporal sequences in the same space. Freud was not, of course, writing as an urban historian, but what is useful for those who do is the emphasis on human viewpoint in the city and the sensual experience of urban juxtapositions of time, space, material and memory.

A palimpsest is a support, be it parchment, wax or papyrus, on which a text is written and later expunged in order for the material to be inscribed again, possibly multiple times. The process of erasure is often imperfect or impermanent so that the

<sup>5</sup> Freud, *Civilisation*, 17: 'Nun machen wir die phantastische Annahme, Rom sei nicht eine menschliche Wohnstätte, sondern ein psychisches Wesen von ähnlich langer und reichhaltiger Vergangenheit, in dem also nichts, was einmal zustande gekommen war, untergegangen ist, in dem neben der letzten Entwicklungsphase auch alle früheren noch fortbestehen'. In the previous sentence Freud describes Rome as 'das Gewirre', which James Stratchey memorably translates as a 'jumble', a word several contributors in this volume use to evoke the entangled pasts in the city fabric: 'Es bedarf kaum noch einer besonderen Erwähnung, daß alle diese Überreste des alten Roms als Einsprengungen in das Gewirre einer Großstadt aus den letzten Jahrhunderten seit der Renaissance erscheinen.'

underwriting can also be seen, creating the visual effect of two or more simultaneously present texts. Transposing this onto the city, the urban tissue becomes the support on which successive waves of actors build, demolish and rebuild their environment, imbuing each change with different uses and meanings, some responding to past phases, others not. What is most attractive about the image of the city as palimpsest is that it makes present the simultaneity of multiple material expressions, both visible as well as invisible yet formative. The metaphor's shortcomings have to do with the inertness and inanimate nature of the actual palimpsest and the limited potential responsiveness between the multiple writings conceived as layers, problems we address in this introduction. Taking a city case study approach, this volume brings together art historians, archaeologists and historians to investigate from their distinctive disciplinary approaches the relationships between and responses to successive reworkings. More than this, we are concerned with a third exercise – and here Freud's interpreting eye is important – namely, to explore the interactions between the city's material accumulations and their interpreters.

The metaphor's power to compress complex, often non-linear, temporal evolutions into one spatial support has made the palimpsest useful in a range of humane disciplines concerned with the force of the past in shaping the present, and the role of the author/interpreter. Palimpsest has appeared widely as a 'vivid conceit' used freely, sometimes constructively, other times misleadingly.<sup>6</sup> In a parenthetical aside in 1897, the English historian F. W. Maitland described the Ordnance Survey map as a palimpsest.<sup>7</sup> This is as good an example as any of the term's loose application to mean an accumulation of past traces all represented at a single moment in time (in this case, that of the map's creation), rather than presenting all phases at once as Freud would attempt to imagine thirty years later. English landscape historians have toyed with the palimpsest in order to visualise relationships between landscape uses, rural and urban, over time. In the 1940s and 1950s W. G. Hoskins experimented with the metaphor in a way that drew out the powerful presence of the dimly visible underlayers, urging historians to study towns as landscapes 'to get behind the superficial appearances, to uncover the layers of the palimpsest and to see, for example, a piece of the tenth century in the way a street makes an abrupt turn or does something else unexpected'.<sup>8</sup>

From the second half of the twentieth century onwards, 'palimpsest' has been used to describe paratactic relationships in time and space by a wide range of scholars in distinct fields of enquiry coexisting without obviously direct connections. By the early 1980s, the palimpsest was serving parallel purposes among students of language and philosophy concerned with problems of polysemy and interconnected authorships. Literary critics such as Andreas Huyssen were attracted to think of the

<sup>6</sup> Gardiner and Rippon, 'Introduction', 1. For recent overviews of usages of 'palimpsest', see Osthues, *Literatur als Palimpsest*; Bartolini, 'Critical urban heritage' (discussed below); Dillon, *Palimpsest* and Dillon, 'Reinscribing De Quincey's palimpsest'.

<sup>7</sup> Maitland, *Domesday Book*, 15.

<sup>8</sup> Hoskins, *Making of the English Landscape*, 245. He concludes the book with a careful reading of the traces in the palimpsest outside his west Oxfordshire study window, 270–274.

city as a palimpsest, 'reading a city as a conglomeration of signs' where 'strong marks of present space merge in the imaginary with traces of the past, erasures, losses, and heterotopias'.9 The question of how to decipher traces of multiple urban pasts was posed in discussions about the memorialisation of historical trauma in the urban tissue. Urban theorists faced with the increasing uniformity of modern cities felt the urgent 'necessity of keeping the successive layers of urban form alive', warning 'lose depth and we sever our intimate connection to a living past'.<sup>10</sup> Memorians' and urban historians' concerns converged with those of heritage preservationists who awkwardly combine an insistence on lived experience with refusal to privilege any one 'layer' of the past. But the historian must ask whether the palimpsest comes close to becoming a modern totem when we claim that 'the accumulation is called culture'?11 These developments are directly related to the problem of the strangely static quality of the palimpsest, whose layers are to be observed and preserved as part of an ever-evolving heritage manifesto, a phenomenon to which we will return in our concluding remarks.

Less memory, more 'time depth' has been some archaeologists' emphasis in their use of 'palimpsest' since the 1980s as a tool for conceptualising time. Prehistoric archaeologist Geoff Bailey argues that the formation of the archaeological record is palimpsestic in nature and proposes that 'palimpsests are shown to be a universal phenomenon of the material world, and to form a series of overlapping categories, which vary according to their geographical scale, temporal resolution and completeness of preservation'.12 Bailey's advocacy of the palimpsest is compelling because of the concept's emphasis on human practice, defining it as 'a superimposition of successive activities, the material traces of which are partially destroyed or reworked because of the process of superimposition'.13 But superimposition is insufficient for the student of the city who would understand interactions and responsiveness between layers.

By the end of the twentieth century 'palimpsest' began to appear in studies by architectural historians and art historians as shorthand for buildings and architectural complexes with long histories of adaptation and reuse.14 It has proven more analytically productive to apply the metaphor to buildings (often viewed with the distancing label of 'monument') rather than the city. By 2008, the usage had become

<sup>9</sup> Huyssen, *Present Pasts*, 49, 7; Huyssen, *Other cities*, 3.

<sup>10</sup> Kroessler, 'City as palimpsest', n.p.

<sup>11</sup> Huxtable, 'Where did we go wrong?', D24; ours is a willfully de-contextualised quotation from Huxtable's stimulating diatribe against preservation, which must be understood in the context of late 1960s urban trends in the US. Cited recently by Kroessler, 'City as palimpsest', n.p.

<sup>12</sup> Bailey, 'Time perspectives', 198.

<sup>13</sup> Bailey, 'Time perspectives', 203; Bailey's near-contemporary articles 'Time perspectives' and 'Time perspectivism' present his conceptual work with the multi-valence of palimpsest for archaeologists, ongoing since the 1980s. He analyses different types of palimpsests with which the archaeologist contends, and not in isolation from each other, as true palimpsest, cumulative palimpsest, spatial

palimpsest, temporal palimpsest and palimpsests of meaning. 14 Especially targeted use of palimpsest for reading buildings and their contexts has been made by Flood, 'Pillars', and Necipoğlu, 'Dome of the Rock'.

widespread enough for the Islamic art historian Nebahat Avcıoğlu to protest that 'the word palimpsest has been used so often by writers and scholars in describing the city of Istanbul that it has almost become an orientalism of a postmodern kind'.15 Admitting the power of the palimpsest to evoke what critic Juan Goytisolo described, in his essay on 'The palimpsest city', as Istanbul's 'subtle interplay of synchrony and diachrony',16 Avcıoğlu nevertheless rightly called for closer attention to how the strata that embody this historical density 'are organised and how they interact with one another to form a coherent "whole", such as a city, which is also in a state of perpetual flux'.17 To demonstrate the practical deciphering involved in understanding a city as opposed to a text, she adapted and transferred Gérard Genette's hypertext and hypotext from a literary to an urban context, applying his relational thinking to the palace and the city in Ottoman Istanbul, and using the modulations between kiosk, tent and public fountain to explore deliberate intersectional nodes across the urban fabric.<sup>18</sup> A similar insistence on interrelations across the city in space and time, that refuses to see buildings in isolated abstraction, inspires the work of another Islamic art historian, Avinoam Shalem, who employs the palimpsest to consider ways of thinking about time expressed both horizontally and vertically in the city. We encounter the 'multi-synchronic' city both as we walk through the jumbled present, and as we penetrate the formations and transformations of the past, in a way that metro stations with archaeological exhibitions make increasingly possible.19

In the second part of the Introduction we consider in greater detail some of the ways in which we believe thinking with the concept of the palimpsest can help to expose problems in how we approach eastern Mediterranean cities. We highlight first one difficulty with much current usage that we hope this collection of studies will begin to rectify: palimpsests need decoders to discern the layers. Decoding itself is not always and only a detached exercise, but can be intended as a didactic practice in which past layers have lessons to teach, regardless of contemporary academic discomfort with such a view of past traces. Too often the city becomes an arid abstraction even when the writer's intent is to elucidate how the fabric instantiates vibrant transformations. A common thread linking many of our authors is the attempt to avoid lifeless abstraction and look closely at human agency, intention and choice, practical and accidental, that motivate repair and reuse over intangible spatial systems that undergird accumulation. To do this, historians and archaeologists must read their

<sup>15</sup> Avcıoğlu, 'Palimpsest city', 191.

<sup>16</sup> Juan Goytisolo, 'Palimpsest city', 72; quoted by Avcıoğlu; see also Yanıkkaya, '"Mirror mirror on the wall"', 137–139 on palimpsestic Istanbul in the light of Goytisolo, Lefebvre and Foucault.

<sup>17</sup> Avcıoğlu, 'Palimpsest city', 191.

<sup>18</sup> Genette, *Palimpsestes*, 1–9; *Architext*, 1–2, 83–84.

<sup>19</sup> Shalem, 'Against gravity', 58. For discussion of controversy raised by metro works and in situ archaeological presentation as represented by the case of Thessaloniki, see Aslanidis *et al*., 'From the late antique *decumanus maximus* to the middle Byzantine *mese*'; Konstantinidou and Miza, 'Μεταμορφώσεις του αστικού τοπίου της Θεσσαλονίκης'; Vasileiadou and Tzevreni, 'Μετασχηματισμοί του αστικού χώρου της Θεσσαλονίκης'; Plantzos, 'Το παρελθόν ως τραυματική εμπειρία στο παρόν'; and Plantzos, 'Hellas *Mon Amour*'.

evidence with attention to the cultural background that leads to the choices that shape what is kept and what is erased. Many of our studies focus on this problem of who decides, and how we might understand motivation from the evidence. Running through our enquiry is the centrality of the viewer in the encounter with the living city, whether that city is represented in writing or material. We have asked our contributors to play with the palimpsest metaphor as a way of cross-questioning its now habitual use. We do this not because we want to pin down a single meaning of a metaphor meant to provoke reflection, but because once a word becomes habitual it loses its provocative force. Scholars before us have quarrelled over the aptness of the word and it is certainly worthwhile to examine the metaphor's strengths and limits.<sup>20</sup> Our aim is not to stop there, but to instrumentalise our frustration at the word's casual overuse to ask what it is that the palimpsest reveals about the interrelationship between time and space in the once-ancient city that keeps us returning to it. We would like this volume to leave a lingering question mark in the mind of anyone who applies the word palimpsest to cities.

#### *The challenge of the eastern Mediterranean city as palimpsest*

The attraction of the palimpsest city to so many disciplines, and to the wider public, lies in its power to conjure up the invisible city, its concealed pasts, its absent presences.21 The palimpsest metaphor stimulates us to look through present appearances to formative or disruptive choices in the past. And yet the problem remains that the multiple texts of an actual palimpsest rarely relate to each other, even though they have been written on the same material. How can the same be said of the urban evolutions of a living city that only with some distortion can be understood as discrete layers? What the enthusiasm for the palimpsest does not always take into account is that the accumulative jumbling does not stop; even now, as we write, the processes of exclusion, erasure, selection and promotion continue. These processes might even be regarded as signs of a healthy urban tissue, a useful metaphor that accommodates growth, injury and regrowth. What happens when competing parties continue the practice of destroying and reworking the city and its built environment for their own purposes, claiming different antiquities to justify their own view of history's value in the present and future? What are the mechanisms behind the material and conceptual

<sup>20</sup> Gardiner and Rippon, 'Introduction', 1.

<sup>21</sup> Examples of the Mediterranean palimpsest's wide appeal include the blog (2011 to present) www. mediterraneanpalimpsest.wordpress.com that paratactically assembles the author's observations about 'The history and culture of Greece and the eastern Mediterranean', using rebetika music as the (loose) connecting tissue; an exhibition design project for the Fourth Thessaloniki Biennale of Contemporary Art (2013–2014), entitled *Μεσογειακά παλίμψηστα. Τρία αινίγματα φθοράς και αφθαρσίας. Μια συνομιλία του Δημήτρη Ξόνογλου με τον πάπυρο του Δερβενίου/Mediterranean palimpsests. Three enigmas of decay and incorruption: An open dialogue between Dimitris Xonoglou and the Derveni papyrus,* Archaeological Museum of Thessaloniki, www.loopostudio.com/mediterranean-palimpsests-hp4ps; and the international research project 'Mediterranean Palimpsests: Connecting the art and architectural histories of medieval and early modern cities' (Getty Foundation grant, 2017–2019), directed by Nikolas Bakirtzis and D. Fairchild Ruggles.

manipulations of the city that make these multiple pasts such powerful tools? How has the city, by its nature an organism in constant flux, served as such a useful vehicle for the expression of particular cultural, political and religious values? How do we take into account both pragmatic responses to the need for renewal, often small scale due to limited funds as in the reuse of parchment, and ostentatious demolition and recreation in the urban fabric, frequently a sign of economic prosperity?

The questions we raise in this volume emerge from the 'Impact of the Ancient City' project, a project conceived as two interlinking strands: one focused on the physical city, investigating the survival, adaptation and reformulation of Mediterranean cities that flourished in the Greco-Roman period, bearing constantly in mind what meaning, if any, material traces retained that was carried over from the past. The other strand focuses on the conceptual city, re-examining medieval, early modern and modern writings about cities and their projections of ancient urban ideals that often owe more to the imagination than evidence on the ground*.* Exploring multiple modes of interpretation available in Mediterranean cities, the project attempts to bring urban discourses into relation with urban fabrics.

Part of the challenge for those of us concerned with eastern Mediterranean cities has been to expose less noticed 'impacts', which may be better understood as a ricochet than direct hit. In response to this, a focus of many contributors to this volume is on the variety of Muslim responses to the cities they inherited and reshaped. We want to understand better how awareness of the Greco-Roman city was mediated by its Christian successor and nourished by accumulated monotheist re-readings of inherited urban spaces, buildings and the explanatory legends associated with them. The medieval and early modern concentration of this volume allows us to avoid the common attention to eighteenth- and nineteenth-century European 'reception' of an imagined ancient city, adapted to fit modern ideological needs. We have attempted, instead, to retrieve how knowledge of the ancient city was transferred in different circumstances while recognising how often the only identifications left to us today, thanks in large part to scholarly priorities and colonial agendas, are those that have erased the sometimes culturally alien medieval understandings and 'restored' a more familiar (if partly invented) original. To this material we pose the question of what we can learn from pre-modern ways of knowing about the past an relating past, present and future, which may not correspond to dominant Euro-American epistemologies.

The eastern Mediterranean region has inherited a particularly rich variety of pasts – local, national, imperial, colonial, polytheist, Jewish, Orthodox or Catholic or Protestant Christian, Sunni or Shiite Muslim, to mention only the large, unwieldy identities that are most prominent in discussions of the region. Different inhabitants and outsiders have drawn from these local, imported and invented pasts in order to shape and reshape cities. None of the pasts has achieved and maintained a monopoly, but Greco-Roman antiquity has sometimes been treated as a sort of secularism *avant la lettre* in a world of religion-fired, contentious reworkings of history. In opposition to this, one of the primary aims underlying this volume is to bring to light and examine the often unspoken assumption that Greco-Roman antiquity is not only the most important 'Antiquity', but that it carries with it an aura of neutrality.

## **What can, and can't, the palimpsest do for us?**

### *Privileged palimpsests: Seeking the original*

The palimpsest metaphor raises another related challenge that this volume aims to tackle, namely the privileging of the 'original'. In the case of the textual palimpsest, what is found underneath the visible surface is in most cases what most excites scholars. The parallel problem with cities is obvious since the pursuit of the 'classical' original underneath medieval, early modern and contemporary cities has long been the focus of antiquarian, archaeological and historical interest. Two major issues emerge from this.

Firstly, cities have more than two texts. Beneath the modern cities of the eastern Mediterranean are layers written upon layers. Choosing the 'original' stratum to be exposed, studied and exhibited inevitably involves choices about origins which are tied teleologically to modern self-conceptions. The frequent scholarly fixation on the Greco-Roman as original is inseparably fused with European conceptions of connectedness to an Enlightened Classical Antiquity in which the subsequent periods are characterised by decline, fanaticism and ignorance. Other attempts at communal self-fashioning, often associated with nation-building, have sifted Levantine soil and history for other genealogies: modern Israeli identification with Iron Age Israel; Lebanese attempts to claim the Phoenicians; Mesopotamian Christians' intimate association with Assyrian heritage; while Pharaonic Egypt has been a similar site for modern national contestation.22

These selective emphases on alternative palimpsestic pasts in the eastern Mediterranean have often been embedded in challenges to Islamic political and cultural hegemony in the region, forwarding alternative claims by presenting the Muslim and/ or Arab presence as a text overwritten onto an earlier, idealised ethnic, political, or religious script. Such strategies are not only modern, as several contributions to this volume show. Use of the distant past in the service of a contested present has a long history in the eastern Mediterranean. Earlier scholarly practices have often idealised or focused excessively on particular origins or periods that served to buttress the relationship between archaeology and history on the one hand, and modern political and cultural claims on the other.23 In response, this volume attempts to investigate

<sup>22</sup> Ohana, *The Origins of Israeli Mythology*. On Lebanon, see Quinn, *In Search of the Phoenicians*, 1–24, undergirded by Kaufman, *Reviving Phoenicia*. On the emergence of modern Assyrianism, with bibliography, see Butts, 'Assyrian Christians', 599–612. On the modern struggle for Egypt's past, see Reid, *Whose pharaohs?* and Reid, *Contesting Antiquity*, with El-Daly, *Egyptology: The Missing Millenium*, who shows that there was a long pre-modern tradition of making sense of Egyptian antiquities in Islamic Egypt.

<sup>23</sup> As in early Islamic adaptation and reuse of Sasanian ideas about kingship and the ancient Iranian past, see Zychowicz-Coghill, 'Remembering the ancient Iranian city'.

more closely the ideologies which informed pre-modern 'archaeological' thinking in the region, while at the same time drawing attention to some of the distorting results and missed opportunities of earlier practice.

Secondly, an archaeological focus on original phases of construction, as well as tourism-driven attempts to curate ancient sites as exhibitions of major classical era constructions, can conjure up an artificial ancient city of perfectly finished, monumental stone constructions (even if those constructions were in fact from different periods of a city's history). Any inevitable deviation from such a pristine material state in the course of a building or cityscape's lifetime then reads as decline. We have asked the contributors to consider what if, instead, we view such changes as healthy signs of the living city? How does the archaeologist's or historian's understanding of the city change if we make a concerted effort not to separate out layers or periods, but to focus on their intersections? In an effort to put this view into practice, our volume embraces studies of urban evolutions through the lenses of reuse, adaptation, duration and resilience in the face of changing circumstances. While less negative and prejudicial perspectives on change have become more prevalent in scholarship concerned with the city's material form,24 our volume extends this scholarly shift to evaluations of literary evidence as well so that medieval and early modern ideas about the city and its past can also benefit from being taken on their own terms within their own cultural framework, rather than dismissed as ignorance or error, as has often been standard practice in the past.

#### *Text and material in the palimpsest city*

The palimpsest metaphor can prompt us to think about how and why modern scholars or historical actors might come to define an 'original text' in a city palimpsest, and also to contemplate the implications of such quests for the reception of other phases of a city's history. But can it do more for us?

In *Palimpsests: Buildings, Sites, Time*, published in 2017, the co-editors Nadja Aksamija, Clark Maines and Phillip Wagoner suggest that thinking about how textual palimpsests are made can be useful for the study of the architectural development of buildings and sites.25 They propose that the initial construction corresponds to the writing of the original text, acts of demolition or dismantling to the erasure of that text and the rebuilding or expansion of the structure to the act of re-inscription. They add a fourth layer of 'recognition' by a 'perceptive beholder'.<sup>26</sup> There is much to be said for this approach, and especially for their insistence that architectural studies incorporate the ways in which the vestiges of time become embedded in both buildings and sites. Yet their analogy cannot be fully sustained: textual palimpsests are rarely,

<sup>24</sup> See Hugh Kennedy's article 'From *polis* to *madina*'.

<sup>25</sup> We encountered this beautifully produced volume in the course of writing our introduction and are grateful to Nadja Aksamija for providing us with their introduction and the chance to engage with it.

<sup>26</sup> Aksamija *et al*., *Palimpsests*, 9: 'the fourth moment, one of recognition, remains a historically contingent cognitive act experienced by perceptive beholders at various points along the palimpsest's temporal arc.' See also Bailey, 'Time perspectives', esp. 199–210 and 220.

if ever, half-erased. Even if the earlier text remains to some extent exposed, the palimpsest's new text (or texts) usually bears no relationship to the old text, which is reduced to redundancy. Rather, the notable feature of a palimpsest is the absence of correspondence between the original text and the re-inscription. A palimpsest is defined by material traces which alert the viewer that the writing support has been reappropriated as the vehicle for a new communication. The reused support is a material space for the immaterial constructions of language. In this, a palimpsest stands in contrast to the evolution of the urban space in which the new layers are conditioned by the old.

The crucial contrast between an evolving building and a palimpsest, then, is that the palimpsest is language on top of language, rather than material on top of material. While the over-writing on a palimpsest – and the ideas which it conveys – relies on the materiality of the support and is itself materially constituted, its relationship to immaterial language has the potential to utterly transform the meaning of its material support in ways which are discontinuous: turning the manuscript from, say, a mathematical proof to a prayer.27 Yet, the materiality of the support bears testimony to this transformation: it betrays its re-inscription in traces of characters reminding the reader that here there were other lines of thought that were followed, lines that one might be able to reconstruct. Thus, the palimpsest directs our thoughts to the ways in which new ideas can be projected onto the same material in transformative ways that seek to render previously projected ideas redundant. But it also asks us to consider whether all types of material are as suitable for such radically disjunctive projections as a manuscript page, the ways in which other types of material might condition the new messages which they can bear and the material traces which bear witness to such transformations.

The changes witnessed in the urban landscapes across the broad period covered by this volume give the lie to the specious Orientalist notion of the timeless East. New ideas, beliefs and regimes of knowledge and power arose that were applied to the cities of the post-classical eastern Mediterranean. But were these cities as fit to bear radically new messages as an erased parchment, or did some pre-existing features condition the new stories which could be told about them? Did the material remnants of antiquity remain as unread palimpsestic traces of the past or did they have a more dynamic role to play in the remaking of the city? In short, how did new understandings and uses of cities interact with the material traces of their antiquity?

Such reflections on the troubled discontinuity represented by the combination of materiality and language in a palimpsest compel our inclusion of archaeologists and textual scholars – and all points in between – in a volume considering the evolutions of eastern Mediterranean cities. Texts about cities can stress disjuncture or changelessness, exaggerate one aspect while concealing another. They evoke their own worlds and processes whose relationship to material realities lies somewhere between qualified and arbitrary. Yet despite texts' partiality they are precious testimonies to

<sup>27</sup> Netz *et al*., *The Archimedes Palimpsest*, vols. 1 and 2.

the thoughts, interpretations and beliefs that animated people's responses to evolving urban environments and their pasts. The physical remains of past urbanism, by contrast, are less inclined to hyperbolic reactions to the changing city. Rather, they more subtly bear their responsiveness to the ever-shifting limitations of space, time, resources, craft and technology that go into the remaking of a city. The material stuff reveals the concrete choices and changes which kept the city alive. Only through consideration of both of these durable sets of artefacts – the textual and material records of cities – can we trace a full picture of how the physical and imaginative reconfigurations of cities fed into one another.

#### *Accumulating metaphors*

Geographer Nadia Bartolini has recently challenged the dominance of the palimpsest metaphor in urban studies, arguing that a cityscape does not evolve through processes of neatly layered superimpositions, but more often through juxtapositions in material, space and time that co-exist across a city. She shares our frustration in the casual use of palimpsest as generalisable shorthand for 'the mingling of the past in the present'.<sup>28</sup> To complement the overly linear and rigid metaphor of the layer, so widespread in describing complex cities, Bartolini proposes the geological concept of brecciation, the agglutinative process by which pieces of different rock of diverse origins become consolidated. Brecciation, Bartolini suggests, can be used productively to understand how structures and objects from different times and with separate origins become enmeshed in the present to create an urban whole. Applied to the urban fabric, brecciation helps account for irregular and uneven development and allows greater precision in describing how 'variegated times and materials coexist without linearity'.29 What brecciation helps us to do is to represent the more complex displacements in city spaces and structures that cannot be made intelligible through reference to layers and strict periodisations.

Bartolini centres her discussion on Rome in order to illustrate the concept of brecciation, but Istanbul also offers an excellent example of brecciated urban development. Istanbul was the place where the contributors to this volume first met to discuss the usefulness of the city palimpsest trope, and five of our chapters approach the Queen of Cities from different disciplinary angles. While the city's history reaches back 8,000 years, it is conventionally approached diachronically through four isolated chronological blocks – Greco-Roman, Byzantine, Ottoman and Turkish – designated by the city's four primary names – Byzantion, Constantinople, Konstantiniyye and Istanbul. But the urban narrative is too convoluted and seeps through such sharply drawn lines. Peoples and cultures are entangled across time through the force of place. Across the city, the viewer encounters chance survivals from different times and contexts that disrupt easy chronologies: a Byzantine *hagiasma* (holy spring) is

<sup>28</sup> Bartolini, 'Critical urban heritage', 519: 'there are realities in the field that are overshadowed by the dominant use of the metaphor'. 29 Bartolini, 'Critical urban heritage', 530.

found in the basement of a modern-day carpet shop in Sultanahmet;30 in Galata, the Arap Cami's minaret clearly recalls a Latin bell tower, revealing its origins as a Genoese early Gothic church – with traces of early Byzantine construction – converted in the fifteenth century to Muslim use and acquiring a patina of legends associating the mosque with eighth-century Arab besiegers;31 during the excavations of the Marmaray commuter railway system connecting the two sides of the city a Theodosian harbour with dozens of tsunami-stricken ships dating to the Byzantine period was unearthed next to the remains of Ottoman-period workshops owned by Armenians.<sup>32</sup>

Such a focus on the interactive simultaneity of elements from different times in the same twenty-first century cityscape is at odds with a palimpsestic perspective that encourages us to conceive of the city in terms of discrete, superimposed layers. Understanding spatial and temporal relationships in terms of brecciation may come closer to replicating the clustered manner in which pre-modern and modern persons developed, perceived and interacted with their cities. And at the same time it avoids the abstracting distortions of 'palimpsest' that distance us from the people we are trying to understand from a historicising perspective. We do not intend to substitute 'brecciation' for 'palimpsest'. Both are offered as metaphors with limitations. Yet, changing metaphors may at least stimulate us to pose the historical question 'who sees the palimpsest'?

The examples of urban development Bartolini investigates belong to contemporary Rome and efforts by planners and architects to present tourists and locals with a sense of the city's deep material history.33 She argues persuasively for the more practical concept of brecciation, which like a city and its building projects is always in a process of consolidation, producing new conjunctures and spaces. We suggest that 'palimpsest', by contrast, tempts us to defuse the dynamic power of the city, the contention that is its lifeblood, by pushing us too far into the passive observer's stance. Such a position is dangerous for the historian and archaeologist whose work requires a degree of imaginative empathy. Laudable though aims of putting aside past sectarian conflicts may be, efforts to neutralise contested buildings and neighbourhoods as 'heritage monuments' impose yet another contested claim of value and meaning that carries its own price. Traces of the historical past in urban space can be instrumentalised as political and social statements about present and future communal space, as in examples as wide and varied as the destruction of the Ayodha

<sup>30</sup> Ousterhout, 'Water and healing', 71.

<sup>31</sup> Çetinkaya, 'Arap Camii', 169–172; Westphalen, 'Pittori greci', 51–62; Akyürek, 'Dominican painting', 328–330, with nn. 3 and 5; and most recently Marangon, 'Latin inscriptions', 107–110, with bibliography.

<sup>32</sup> On the harbour and the shipwrecks, see Pulak *et al*., 'Eight shipwrecks from the Theodosian harbor excavations', 39–73; Karamani Pekin, *Istanbul.*

<sup>33</sup> Bartolini discusses the Radisson Blues Hotel and Renzo Piano's 'Parco della Musica'. Two remarkable works of brecciated sites that deserve mention, both reworkings of churches and associated buildings partially destroyed in World War 2, are the Stiftskirche in Stuttgart, see esp. the chapter 'Von der Ruine zum Gesamtkunstwerk' in Auer *et al*., *Stiftkirche Stuttgart*, and architect Peter Zumthor's 'Kolumba', the Art Museum of the Archdiocese of Cologne. Available at: www.kolumba.de/?language=eng&cat\_select =1&category=14&artikle=61; with Zochiou, 'Palimpsest approaches'.

mosque by Hindu nationalists, the toppling of Edward Colston's statue in Bristol by Black Lives Matter activists or re-conversions of secular, World Heritage museums to active mosques in Istanbul by the Turkish President. In these living examples of the city as fulcrum of conflicting and evolving identities, the actors are not seeing a palimpsest from a distance as 'perceptive beholders'. But, used carefully, 'palimpsest' can remind us that the past matters in the present; it is not just decorative but also predicative, and today's urban conflicts warn us against creating too wide a gap between extremes of bloodless scholarship and violent engagements in cities with deep and conflicted histories across the world. Looking forward, 'brecciation' might even help city dwellers, historians, archaeologists and urban planners to rework contested urban spaces, taking into account that destruction and regrowth are integral parts of the urban phenomenon where 'the resurfacing of tangible evidence may provoke an accommodation of materials, temporalities and spaces,'34 or it may not.

### **Eastern Mediterranean urban evolutions**

The authors of this volume investigate already-ancient eastern Mediterranean cities in their evolutions from late antiquity to the contemporary period. The volume's geographical focus is on cities of the southern Balkans, Anatolia and the Levant, with several contributions addressing the urban complexities of Constantinople/Istanbul from different disciplinary angles (Figure 1.1).

Most chapters are city case studies, some address wider questions of the urban imagination and address more than one city. The contributions to this volume do not show pre-modern persons viewing the traces of the ancient in their cities in a distanced, disinterested manner; nor do they show people ignoring the vestiges of the past in favour of a contemporary urban superscript that ignores the old. Rather than engaging head-on with the usefulness of the palimpsest metaphor, each chapter evolves its own analysis of how cities have been the subject of sequential manipulations and interpretations, whose interrelations can be fittingly described in terms of palimpsest, brecciation or other explanatory metaphors. The volume is organised around four overarching and overlapping themes: 1) Accumulation and juxtaposition, 2) Erasure and selective memory, 3) The new and the old and 4) Whose past? We asked the contributors to each of the four sections to consider the following questions:

#### *Accumulation and juxtaposition*

Long-lived cities become spaces where structures in varying degrees of completion (or decay, depending on one's viewpoint) are tolerated, enhanced and ignored. Like living organisms, buildings and spatial organisation are not ossified as in models and plans, but always alive and changing. Cities evolve unevenly and collective memories that might relish certain juxtapositions as symbolic of local identity or future aspiration are

<sup>34</sup> Bartolini, 'Critical urban heritage', 531.

*map data available at www.naturalearthdata.com).*

lost through population shifts and abandonment.35 In light of such gaps and silences, how can one identify when ideological clashes infiltrate scholarly interpretations of archaeological and architectural evidence? How have people in the past read and interpreted the accumulated relationships between buildings in different stages of repair and reuse in the city? How do we, in turn, read their readings?

Gideon Avni's 'Between wars and peace: some archaeological and historiographical aspects of studying urban transformations in Jerusalem' revisits the narrative of war and conquest as monocausal drivers for change. Instead of presenting a storyline of sharp breaks in periodisation between Jewish, Christian, and Muslim domination, he suggests we understand this period as one of accumulating populations, monumental landscapes and urban complexity. Comparing the long-term aftermaths of the Roman destruction of Jerusalem in AD 70 and the Sasanian and Arab conquests in the seventh century, Avni takes an archaeological approach to complicate the 'replacement paradigm' that has been characteristic for much past research on Jerusalem's history. Through careful readings of the material remains, Avni nuances our understanding of prolonged processes of change which confute the radical, violent transformations depicted in literary or normative sources and which have underpinned past historiographical consensuses.

Robert Ousterhout, in 'Visualising Constantinople as a palimpsest', takes a similarly wide temporal scope in order to demonstrate how we may read the long metamorphoses of some of Constantinople's most famous structures – Hagia Euphemia, the Lips Monastery and the Chora Monastery – as microcosmic reflections of the changing, dynamic city. He argues that by the late Byzantine period we see patrons and architects renovating centuries old buildings in ways designed to expose and even draw viewers' attention to the contrasting styles of different building phases. Ousterhout suggests that such visible reminders of transition did not evoke the redundancy of the old, but the regenerative eternity of the city as a whole. He concludes with the fourteenth-century statesman, urban patron and writer, Theodore Metochites, who suggested the continual reuse of old building material in Constantinople as indicating not urban discontinuity and disrepair, but the falling and regrowth of a bird's plumage.

Suna Çağaptay discusses another case of urban patrons playing with the intentional juxtaposition of styles in a single building: in this case elements from supposedly different cultures architectural repertoires. 'Transcultural encounters in medieval Anatolia: The Sungur Ağa Mosque in Niğde' focuses on a Mongol-Ilkhanid mosque built in the Cappadocian heartland in the early fourteenth century. The chapter outlines the dynamic combination of Armenian, Seljuk and Latin Crusader construction techniques and decorative details attested in the building. Çağaptay argues that this heterogeneous scheme was chosen by its patron, Sungur Bey, the Ilkhanid ruler of the city, as a response to contemporary political rivalry, subtly and selectively employing material forms from past and present in order to assert his future legacy.

<sup>35</sup> See Pullan and Sternberg, 'The making of Jerusalem's "Holy Basin"'.

In 'The water of life, the vanity of mortal existence, and a penalty of 2,500 denarii: Thoughts on the reuse of classical and Byzantine remains in Seljuk cities', Scott Redford positions Seljuk display of classical stonework and inscriptions in Anatolian walls and gates in direct continuity with preceding and contemporary Byzantine practices. Taking as his example the spoliated pieces incorporated into the decoration of the minaret of the Sahip Ata mosque, Redford argues that past symbols such as Medusa's head were consciously incorporated into new cosmological frameworks, forming a dialogue between old and new beliefs, rather than simply proclaiming the victory of the present over the past.

Finally, Bethany J. Walker returns us to the Levant in an exploration of how accumulated urban fabric and infrastructure can contribute to the persistence of a site that has experienced severe discontinuity. 'Echoes of late antique Esbus in Mamluk Ḥisbān (Jordan)' examines the site of Tall Ḥisbān, which saw continuous occupation into the Abbasid period, but was then abandoned for centuries. In the Mamluk period, Ḥisbān was reoccupied in a sudden event that involved several family units refurbishing the ruined town and restoring the Byzantine cisterns and reservoirs. In the fourteenth century Ḥisbān became the capital of the Balqāʾ, resulting in overnight expansion. Through textual and archaeological sources, Walker examines the processes of keeping, discarding and culturally (re)appropriating elements of the built environment, and how the infrastructure of standing buildings and a functioning water supply became paramount for the settlement's success.

#### *Erasure and selective memory*

Can a city's past be erased and scraped away like the layers of a palimpsest? And if it can, how might traces of the past still be visible? The authors contributing to this section address questions of when, why, and how certain pasts have been forgotten as people have contemplated the histories of eastern Mediterranean cities. Some throw light on the long and diverse history of pre-modern partiality in remembering the ancient city. Others consider how modern scholarly tastes, priorities and assumptions have resulted in gaps in our record of the continuous life of cities, gaps which can be reified as genuine absences rather than the results of selection bias. It is now well known, for example, that early archaeologists' tendency to discard later archaeological layers onto the spoil heap while seeking an idealised Greco-Roman monumentality exacerbated historiographical perceptions of post-classical urban decline. But have historians of urban thought and memory equally erased or neglected alternative modes of post-classical seeing, reading, and remembering the ancient city and its remains? What can we learn from pre-modern knowledge of ancient cities which does not match our own perspectives or epistemologies?

In 'Constantinople's medieval antiquarians of the future', Ben Anderson traces a thread of medieval interpretation of the Column of Theodosios in Constantinople which saw in its reliefs the future defeat of the city, rather than Theodosios' triumph over his imperial rival Maximus. Anderson explores how the anachronistic presence of the old in the contemporary fabric allowed experimentation with temporalities: monuments of the past becoming heralds of the future. His 'antiquarians of the future' did not see this vestige of the past within their city as an inert remnant of a redundant urban layer, but as the bearer of a potent message concerning the future fate of their city. Moreover, despite the new interpretation's discontinuity with the original communication, Anderson demonstrates that this radical re-understanding resulted from close inspections of the monument's visual programme: the materiality of the city shaping the new messages which it could communicate. Provocatively, he suggests that his medieval viewers' interpretations of these antiquities were at least as 'scientific' as our own.

In 'William of Tyre and the cities of the Levant', Sam Ottewill-Soulsby introduces us to another medieval viewer of eastern Mediterranean cities who was fixated on their ancient elements. However, in contrast to the autoptic scrutiny discussed by Anderson, he shows us how exclusively the twelfth-century Crusader historian William of Tyre presented the cities of his Latin Levantine homeland through the lens of classical or biblical descriptions already a millennium or more years old. This classicising perspective enabled a vision of the city rooted in Greco-Roman antiquity, intentionally blinkered to the visible impacts on the cityscape of the intervening centuries of Islamic and Eastern Christian occupation. Ottewill-Soulsby argues that this conscious, erudite erasure was a means of arguing for the prestige of these cities to a western audience, making them relatable as part of a western world and positioning the Latins of Outremer as the natural inheritors of the Eastern Mediterranean.

In 'Portraits of Ottoman Athens from Martin Crusius to Strategos Makriyannis', Elizabeth Key Fowden considers practices of accumulation, correction and erasure not in the archaeological record or literary descriptions, but in visual representations made by outsiders. Through both well-known and unpublished early portraits, Fowden examines the uneven 'palimpsestic process … by which earlier visions of the city are rubbed away and partly redrawn to respond to evolving needs and visual tastes'. Some early portraits depict Athens as a living composite of ancient and modern, while in others a tendency grows to blank out contemporary details, a form of erasure or visual 'aphaeresis', selectively seeing and representing a city to fit a specific purpose or ideology that prefers the ancient city. Fowden's study ends with the first portraits of Athens made by local Greeks, arguing that these depictions resist the imposed European views and represent instead a complex, immersive experience of the city.

Nikolas Bakirtzis in 'Perceptions, histories and practical realities of Thessaloniki's layered past' examines pivotal moments in the city's tumultuous history in which 'religious practice was catalytic in the resilient continuity of spaces and sites within the ever-growing urban landscape'. Treating the city walls and Basilica of Saint Demetrios as resilient agents in diachronic urban change, Bakirtzis explores the impact of modern historiographies as well as socioeconomic and national realities on choices made in preserving and reconstituting the city after devastating fires and earthquakes in the late nineteenth and twentieth centuries. Waves of political upheaval, urban destruction and human dislocation have contrived to all but expunge earlier collective memories of the city's complex transformations. The 'ideological priorities for a hellenised Thessaloniki' encouraged the privileging of Byzantine buildings often stripped of their accumulated material surroundings and social history, now isolated as monuments and no longer interwoven into the urban fabric.

### *The new and the old*

The coexistence of elements from different times is a fundamental feature of a palimpsest and an urban fabric. Yet novelty and antiquity are subjective categories. This section asks when (and where) was it important for elements of a city to seem old and when was it important for them to seem new? At what point did the spatial configurations, monuments, or conceptions of the classical city start to feel oldfashioned, and what was achieved by invoking these old fashions? When was innovation adopted or celebrated, and why? Did renovation and upkeep – processes which might involve substantial change to both fabric and function – bring buildings up to date, so that they felt modern, or were they intentionally maintained in archaic form? What role did anachronism play in the planning or experience of the post-classical city?

In the first chapter of the section, Paul Magdalino in 'From Byzantion to Constantinople' revisits the story of the making of Constantine's capital city, on the site of Byzantion, the ancient city whose foundation legend attributed to mythic Byzas. Magdalino argues that the transformation of Byzantion into Constantinople happened in two phases: the old Greek city was upgraded into a Roman imperial residence and the new urban space that made Constantinople into the New Rome. The result presented a potent contrast to rival imperial capitals such as Split and Thessaloniki. The latter, for example, had been designed to recall the monuments – not replicate the topography – of Rome, creating the Balkan capital's urban structure along the lines of a long processional road. Magdalino suggests that the layering of Constantinople upon Byzantion did not, initially, form a palimpsest. While building his capital city, Constantine was keen to create an imperial megalopolis on top of an ordinary ancient city.

Ine Jacobs in 'Looking in two directions: Urban (re)building in sixth-century Asia Minor' demonstrates the persistence of classical ideals of holistic urban planning into the sixth century, alongside a contrasting example of their abandonment, reminding us of the contingent, local factors which influenced decisions about the evolving post-classical cityscape. By focussing on three cities levelled by earthquakes around the same time, the chapter pinpoints sixth-century choices about large-scale urban planning as revealed by the different programmes of rebuilding and deliberate abandonment. Two strategies emerge. At Aphrodisias and Sagalassos, a classicising approach reproduced the architecture of urban life as it had manifested in previous centuries, with regular street plans and colonnades restored and statues re-erected. At Assos, by contrast, a kind of palimpsest-settlement replaced the former HellenisticRoman city, marked by the erasure and irrelevance of older buildings, thoroughfares and public spaces. By rooting her discussion in specific case studies, Jacobs draws attention to the questions of resources, labour, topography, urban agency and civic governance which determined the evolution of the Eastern Mediterranean city.

Helen Saradi's 'Byzantine urban imagination: Idealisation and political thinking (eighth to fifteenth centuries)' explores the image of the city in Byzantine literary works from the eighth to the fifteenth century. Using orations, local legendary compilations and letters, Saradi traces the changing elements emphasised in Byzantine idealisations of the city over time and how these reflected shifting urban realities and literary tastes, particularly with regard to the reception of classical rhetorical strictures about writing the city. She shows how the reduced circumstances of provincial Byzantine cities caused by the seventh-century imperial contraction resulted in a genre of texts focused entirely on Constantinople, before outlining the reemergence of discussions a broader range of cities in the late Byzantine period. In this later era too we see how antiquities which had previously been discussed largely in relation to their imperial associations became a means for Byzantine writers to emphasise their special link to the ancient Greek heritage, particularly in contrast to the Latins whose presence and power now overshadowed the embattled Roman state.

In contrast to a literary tradition which projected an ideal of Byzantine imperial continuity and conservatism onto the city, Dimitri J. Kastritsis looks at how early Ottoman rulers adopted several cities of the Byzantine heartland as emblems of a new political order. 'Ottoman urbanism and capital cities before the conquest of 1453' traces the Ottoman attitude toward cities with a Greco-Roman past as the new rulers both repurposed some remains symbolically and also built anew. Kastritsis discusses both the Ottoman responses to pre-existing urban contexts, reusing particular buildings and forms, and the development of innovative architectural expressions, focussing on the Anatolian cities of Bursa, Edirne, Amasya and Manisa, and the Balkan cities of Dimetoka and Siroz. The spatial and urban shifts that emerged in these cities are contextualised as expressions of contemporary power struggles within the nascent Ottoman empire.

This section began with Constantine's refoundation of Byzantion into Constantinople. It closes with a chapter on the role his honorific column, still standing some 1500 years later, was to play in the development of a modern, historicised Istanbul in the nineteenth century. Göksun Akyürek in 'A new history for old Istanbul: Late Ottoman encounters with Constantinople in the urban landscape' sets out from an official competition for architects to restore Constantine's porphyry column to its original state, analysing the role it was intended to play in the constitution of the city's grand new boulevard, Divanyolu Avenue. She shows how explaining and showcasing the very ancient – notably the column and the iconic Hagia Sophia – became a feature of an Ottoman regime striving for modernity which fundamentally changed the relationship between such monuments and their urban contexts. The column's previous relation to the city, integrated as part of an uneven agglomeration of structures and spaces from different times, was radically shifted as it became part of a modern, sequential showcase of monuments literally cut through the former urban fabric.

#### *Whose past?*

Memory is always selective. Multi-period cities afford both their inhabitants, and outsiders, choices about which heritages to retain or rediscover. Social memories of a city's heritage might be kept alive through a variety of processes, such as ritual, restoration, exhibition, inscription, historical writing and more. Different constituencies within a city may seek to preserve different pasts, sometimes in conflict with others. The authors in this section consider what constraints, if any, limit the choice of which city to remember? And how are tensions exposed and/or defused by choices made about which past to celebrate, both in the historical periods which we study and today, through practices of heritage and restoration?

Edward Zychowicz-Coghill's chapter, 'Medieval Arabic archaeologies of the ancient cities of Syria' begins the section by exploring medieval Arabic ideas about the origins of the cities of Syria-Palestine, showing that not only modern archaeology has been bound up with political claims about belonging and territoriality. He demonstrates how the texts used by the greatest medieval Arab geographer, the thirteenth century Yāqūt al-Ḥamawī, locate the origins of Syrian urbanism in the repeopling of the world by Noah's children after the biblical Flood, thus integrating these cities into the Islamic explanation of the distribution of different ethnicities across the earth. Contemporary Islamic political and territorial boundaries are thereby explained as reflecting primevally established ethnic divisions or affinities, while pre-Islamic Syria is imaginatively de-Romanised by casting it as part of an original Arab homeland. The Greco-Roman remains of Syria become re-identified with Israelite sacred history and their ruined state invested with moral injunctions, rendering these cities' ancient pasts relatable and meaningful to their now largely Muslim inhabitants.

Louise Blanke in '(Re)constructing Jarash: History, historiography and the making of the ancient city' shows how ideological bias in early twentieth-century scholarship constructed an image of Jarash as a mainly Roman city. This image was, in turn, used in the making of a modern Jordanian national identity and in the forging of international relations, especially with the US. The second half of Blanke's paper explores the intricacies of the urban development of Jarash over the *longue durée*.

Amira K. Bennison's 'Constantinople in the sixteenth-century Maghribī imaginary: The travelogue of 'Alī al-Tamgrutī' analyses the description of Constantinople by ʿAlī b. Muḥammad al-Tamgrūtī (1534–1594), a Saʿdī official from Morocco who visited the Ottoman capital in the last decade of the sixteenth century. Bennison demonstrates how al-Tamgrūtī's selective description of the city's domestic and religious architecture, from Eyüp to Hagia Sophia, was fiercely competitive. Tailoring his description around the expectations of his Maghribī audience, the Saʿdī emissary highlights the Byzantine past rather than the Ottoman present, interpreting the city through the Arab Islamic heritage while ignoring (one might even say erasing) Ottoman reinterpretations of the city's past.

Assaad Seif's 'Beirut as a palimpsest: Conflicting present pasts, materiality and interpretation' unravels the multi-layered complexities involved in the forging of heritage in post-war Beirut. Through a series of case studies, Seif demonstrates how the material culture of the past has been appropriated by political, economic and religious groupings in order to construct narratives that serve their specific agendas. Seif's Beirut is a city of conflicting pasts and diverging memories in which heritage is a value-based construct that springs from the vested interests of its 'stakeholders'.

#### **Conclusion**

Taken together, the chapters that make up this volume affirm that 'what we call the past is actually part of our durational present'.36 Rather than fixating on or dismissing the 'original', the contributors pay attention to relationships between material changes in the city. They analyse their evidence with an eye not to separation but to seepage, sensitive to signs of impact, as in the title of the 'Impact of the Ancient City' project from which the questions guiding this collaborative volume grew. We may have retained the question mark in our title as a provocation, but the reader will still expect some answers to the many questions raised in this introduction. Each chapter offers its own answers, not all in agreement. Read together, what emerges from the studies in lieu of a single answer or an alternative metaphor is a general resistance to sealing off layers in a city's history. The analyses demonstrate that conceptual layering creates unnatural separations that can conceal change on the ground, expressed through conflict, accommodation or even negligence. Cities are places where differences – manifested both horizontally and vertically – meet. This meeting happens across the city unevenly, casually, destructively and creatively.

The student of the 'ancient world', so often used as a synonym for the 'Greco-Roman world' will, we hope, be delighted to find in this volume many other eastern Mediterranean antiquities as well, including Iranian, Arab and Anatolian. The reader will find Greeks, Syrians, Arabs, Jews, Iranians, Latins, Mongol-Ilkhanids, Turks, Ottomans, as well as Europeans, who reworked the successive urban forms and habits as the diverse heirs to the cities of the eastern Mediterranean.<sup>37</sup> All these inhabitants of the Greater eastern Mediterranean, however fleeting their stay or light their imprint, could and did choose what they required from these multiple urban pasts to promote their present and future ambitions. In these choices they silenced some pasts as they put words into the mouths of others. The process of silencing ancient remains is represented in Arabic tradition (and in our epigrams) by both the Qur'an and the *qasīda*, or ode, in which great achievements of the past are

<sup>36</sup> Bailey, 'Time perspectivism', 220.

<sup>37</sup> For Ottoman reworking and reinterpretation of the pagan and Christian past in Athens, see Fowden, 'The Parthenon Mosque', and Fowden, 'al-Bārthīnūn Barīklīs wa-l-malik Sulaymān'.

ossified into generic warnings or nostalgia for a lost world. In this way, threatening comparison with the pre-Islamic past is sealed off by reducing great antique cities and structures to ruins that warn of mortal error and the brevity of life. While the Arabic poetic manipulation of the ruined abode may arguably have reached the most exquisite height of the seemingly universal *Ubi sunt?* genre, living cities too taught and warned, and continue to do so today. However vague knowledge of their historical founders and perpetuators has become, cities provoke emotion, they become emblems, they are fought over, used and abused. Cities are still didactic, not inert, sanitised and preserved heritage spaces, but mirrors of our own fears about the past and aspirations for the future.

## **Bibliography**


## Accumulation and juxtaposition

## Chapter 2

## Between wars and peace: Some archaeological and historiographical aspects to studying urban transformations in Jerusalem

## *Gideon Avni*

*I've come back to this city where names are given to distances as if to human beings and the numbers are not of bus routes but: 70 After, 1917, 500 B.C., Forty-eight. These are the lines you really travel on.1*

The widely accepted paradigm of military events as distinctive markers of urban change has been reflected from the mid nineteenth century to the present in the historiography of research in Jerusalem.2 This approach has been supported by scholars from different fields, including history and archaeology.<sup>3</sup> However, the largescale archaeological work conducted in and around the city in recent decades calls for a revision of this paradigmatic approach in order to present a more complicated and nuanced picture of continuity and change.

The customary division of the history of Jerusalem into periods framed by military and political upheavals raises a methodological and historiographical question as to the relevance of such divisions to social and urban transformations. To what extent

<sup>1</sup> Amichai, *Poems of Jerusalem*, 93.

<sup>2</sup> See for example Silberman, *Digging for God and Country*; Silberman, 'If I forget thee, O Jerusalem' for the historiography of modern research in Jerusalem. See Avni and Seligman, 'Between the Temple Mount and the Holy Sepulchre' for a detailed discussion of the involvement of scholars in the research into these central religious monuments.

<sup>3</sup> It is frequently used in popular summaries of the history of Jerusalem. See for example Montefiore, *Jerusalem*, 1–10, where he divides the historical narrative first by the religions – Judaism, Paganism, Christianity and Islam – and then by the national – political 'participants' – Crusaders, Mamluks, Ottomans, Empires (mainly European) and Zionists. For a similar chronological concept see Galor and Bloedhorn, *Jerusalem*; Peters, *Jerusalem*; Asali, *Jerusalem*; but a different approach is presented in Goldhill, *Jerusalem*, which emphasises continuities rather than divisions.

has a single event influenced the *longue durée* processes of changes in Jerusalem, and how were these changes reflected in the writing of modern scholars of various origins and faiths? These questions have become more relevant with the development of archaeological research in Jerusalem and the refinement of chronological observations, which have provided some new insights into the processes of urban change.

While from ancient to modern history the events of wars, sieges, conquests, violent revolts and other disturbances have been used to mark changes in the history of the city,<sup>4</sup> only rarely has a methodological approach been taken in the evaluation of the impact of a single military event on the *longue durée* history of Jerusalem.5 Looking particularly at the first millennium AD, several events have been used as indicators for dramatic transformations in the history of the city and in the identities of its people: the conquest of Jerusalem by the Romans in AD 70 marked the end of Jewish sovereignty, which opened the door to the transformation from a Jewish to a pagan city; the pilgrimage of the Empress Helena in 324 is used as the watershed for the pagan to Christian transformation; and the Arab conquest of 638 marks the major political, religious and cultural change from Christianity to Islam. Beyond the first millennium, the 1099 Crusader conquest provides the clear-cut point of change from an Islamic to a Christian-European identity, whereas the conquest of Jerusalem by Saladin (Ṣalāḥ al-Dīn) in 1187 is considered to mark the regaining of its Muslim identity.<sup>6</sup>

In this chapter I concentrate on some of the major events that shaped the changes in settlement and society in Jerusalem during the first millennium AD, assessing the impact of single political or military events as reflected in the historiography of modern research, sometimes in contrast with more recent archaeological findings. In this framework, I address some of the existing paradigms for religious, cultural and political changes in Jerusalem and review them through the lens of archaeological research, which in some cases presents a different picture. The archaeological sequences revealed in large-scale excavations in Jerusalem challenge the traditional approaches to urban changes and transformations. The main question addresses whether the cause of such changes are best understood as political and military events or whether the *longue durée* processes of gradual urban development and transformation provide a more convincing explanation. The latter processes are connected to changes in the economy and society in periods of affluence and decline which opened the road to transformations in urban structures. This question is of relevance to other major urban centres with long historical sequences and narratives, such as Athens, Rome and Constantinople.

<sup>4</sup> See Cline, *Jerusalem Besieged*, 2, where he numbers 118 violent events that formed landmarks in the history of Jerusalem. 5 From this perspective see, for example, Kedar's interesting discussion of the impact of the 1967 war

as a landmark in the history of Jerusalem. Kedar, 'Wars as turning points in history'.

<sup>6</sup> Many scholarly works on Jerusalem adopt this rigid approach. See for example the monumental work of Vincent and Abel, *Jerusalem nouvelle*, 875–1006; recent works follow the same pattern: Kenyon, *Digging up Jerusalem*; Asali, *Jerusalem in History*; Bahat, *Carta's Great Atlas*; Ben Dov, *Historical Atlas of Jerusalem*; Kloner, *Survey of Jerusalem.*

My main argument is that the detailed archaeological record provides an excellent tool for the evaluation of the impact of a single major event as compared to the *longue durée* processes. I focus on two major events in their historiographical and archaeological context: the destruction of Jerusalem by the Romans in AD 70 and the Persian raid of 614, followed by the Arab conquest in 638. Apart from their significance for the historical timeline of Jerusalem, each event is connected with the modern religious and national identities of Jews, Muslims and Christians, adding to its impact on the historiography of scholarly attitudes.

### **The contribution of archaeology to the historiography of Jerusalem**

Comprehensive archaeological research in Jerusalem has been the main contributor to the reshaping of our knowledge of urban developments and transformations in the first millennium AD. Hundreds of excavations conducted from the mid-nineteenth century, augmented by detailed archaeological surveys, have provided a huge database on the city's layout, its public buildings and monuments, streets and residential areas. While early excavations focused on the study of major monuments, recent research yields information about household archaeology, the urban necropolis, the agricultural hinterland and the sophisticated water installations in and around the city. The exact number of archaeological excavations in Jerusalem between 1863 and 2020 is debatable. The Israel Antiquities Authority archives contain some 1,623 entries on excavations conducted in and around Jerusalem.<sup>7</sup> The total area explored throughout those years is exceptional, with *c*. 30 percent of the ancient city excavated. Large-scale excavations have been concentrated west and south of the Temple Mount/Haram al-Sharif in the Jewish Quarter, and along the present-day city walls.<sup>8</sup> These have been supplemented by hundreds of small-scale excavations and probes.9 Particularly

<sup>7</sup> The actual number, however, is probably far higher. Many 'unofficial' or clandestine explorations were conducted in and within proximity to the city, mainly between the second half of the nineteenth century and the 1970s. Those, unfortunately, have not left any detailed written records. In addition, numerous occasional finds, which were not revealed during proper archaeological investigations, have been documented. It appears that the actual number of excavations conducted in and around the Old City of Jerusalem over the last 150 years comes much closer to 2,250 single initiatives.

<sup>8</sup> Particularly notable were the excavations in several areas: the upper parts of the City of David by Crowfoot and FitzGerald in 1925–1927, Shiloh in 1978–1984, Ben Ami and Tchekhnovets in 2006–2012, Gadot and Shalev in 2016–2020, the western and southern walls of the Temple Mount by Kenyon in 1961–1967, Mazar and Ben Dov in 1968–1978, Reich in 1994–2000 and Barbe in 2012–2017, the Western Wall plaza by Weksler-Bdolah in 2005–2009, the Jewish quarter by Avigad in 1968–1982, the Citadel by Johns in 1934–1947 and the Damascus Gate and north wall by Hamilton in 1937–1938. A good summary of excavations in Jerusalem is provided in *NEAEHL* 5.1805–1839; Galor and Avni, *Unearthing Jerusalem*. See also Bieberstein and Bloedhorn, *Grundzüge*; Küchler, *Ein Handbuch*, and see also the abridged 2014 volume, widely considered one of the most authoritative works, though as it is in German it may not receive the attention it deserves.

<sup>9</sup> See Kloner, *Survey of Jerusalem* for a summary of the archaeological survey of Jerusalem which included small-scale excavations.

significant were the excavations south and west of the Temple Mount,10 in the Old City's Jewish Quarter,<sup>11</sup> the City of David,<sup>12</sup> the Citadel,13 along the city walls and in the vast urban necropolis.14 In addition to excavations, hundreds of ancient buildings, most of them from medieval and early modern times, have been surveyed and documented,<sup>15</sup> including the main religious monuments of Jerusalem – the Temple Mount/Haram al-Sharif16 and the Church of the Holy Sepulchre area.<sup>17</sup>

The accumulation of archaeological findings points towards an unprecedented urban expansion of the city in the early Roman period (first century BC to AD 70) and in Late Antiquity (between the fourth and the eighth centuries AD).18 New constructions expanded the urban area far beyond the city walls, especially to the north and east.19 This urban expansion persisted throughout most of the early Islamic period and it was only in the tenth and eleventh centuries that a considerable decline could be seen.<sup>20</sup>

Perhaps the best contribution to the study of *longue durée* processes in Jerusalem has been provided by the large-scale excavations conducted in the past 25 years around the Temple Mount/Haram al-Sharif. These unearthed some fine stratigraphic and chronological sequences which emphasise not only the impact of military events but also the intermediate periods and the process of transformation and change between them.21 A clear sequence of urban structures from the late Iron Age to the Ottoman period was revealed in this area, emphasising the continuity between periods rather than the destruction phases. For example, the excavations at the Western Wall plaza showed a gradual transformation from the Byzantine to the early Islamic periods (sixth to eighth centuries AD), with the contraction of the Eastern Cardo of Byzantine Jerusalem. The excavations in the upper City of David emphasised the transition from late Roman Aelia Capitolina to Byzantine Jerusalem, and then to the Islamic city. In both areas the devastating impact of the Roman conquest of AD 70 is highly visible, but also the evidence of the restorations and the foundation of the new city in the second century. This vast material furnishes the basis for the discussion which follows.

<sup>10</sup> Mazar, *The Mountain of the Lord*; Ben Dov, *In the Shadow of the Temple*; Reich and Billig, 'Excavations near the Temple Mount'; Baruch *et al*., 'A decade of archaeological exploration'.

<sup>11</sup> Avigad, *Discovering Jerusalem*; *NEAEHL* 2.729–736; Gutfeld, *Jewish Quarter Excavations*.

<sup>12</sup> Ben Ami and Tchehanovets, 'Lower city of Jerusalem'; Greenhut, 'Domestic quarter'.

<sup>13</sup> *NEAEHL* 5.1816. 14 Kloner and Zissu, *The Necropolis of Jerusalem*; Avni, 'Urban area of Jerusalem'.

<sup>15</sup> See Burgoyne, *Mamluk Jerusalem*; Hillenbrand and Auld, *Ayyubid Jerusalem*; Auld *et al*., *Ottoman Jerusalem*.

<sup>16</sup> Grabar and Kedar, *Where Heaven and Earth Meet*; Rosen-Ayalon*, Early Islamic Monuments*.

<sup>17</sup> Biddle, *Tomb of Christ*; Biddle *et al*., *Church of the Holy Sepulchre*; Vieweger and Gibson, *Church of the Redeemer and the Muristan*.

<sup>18</sup> Tsafrir, 'Topography and archaeology', 285–295, 330–342; Bahat, 'Physical infrastructures'.

<sup>19</sup> Kloner, *Survey of Jerusalem*.

<sup>20</sup> See Avni, *Byzantine Islamic Transition*, 109–159 for a summary.

<sup>21</sup> Particularly notable were the excavations at Western Wall plaza and in the upper section of the City of David. See for updated summaries: Weksler-Bdolah, *Aelia Capitolina*; Regev *et al*., 'Radiocarbon dating of Wilson Arch'; Ben Ami and Tchekhanovets, 'The lower city'; Ben Ami and Tchekhanovets, *Excavations in the Tyropoeon Valley*.

#### **The Roman conquest of Jerusalem in AD 70**

Pliny the Elder's much quoted phrase, addressing Jerusalem as 'the most famous of the cities of the East',<sup>22</sup> accurately described the city as it appeared in the mid-first century AD. That was the zenith of Jerusalem's urban expansion following King Herod's largescale constructions, and prior to its devastating destruction by the Roman legions in AD 70.23 Two sources provide an outstanding opportunity to reconstruct the urban layout of Jerusalem during this time: the commentaries of Flavius Josephus, who was an eyewitness to the events in Jerusalem, and the evidence of the material culture, as revealed in archaeological excavations in and around the city.24 The integration of both provides a reliable picture of Jerusalem's urban outline, its fortifications, major public monuments and private residences. It also sheds light on many aspects of its inhabitants' daily life. Jerusalem's population during this period is estimated at between 30,000 and 120,000.<sup>25</sup> It was a well-fortified city which enjoyed almost a century of development and expansion in a period of relative stability under the *Pax Romana*, resulting in a demographic boom and constant growth.<sup>26</sup> The city's layout consisted of two ridges divided by valleys – the eastern ridge forming the Temple Mount, and the Western Hill accommodating palaces and lavish residential quarters. The two sections of the city were separated by the Tyropoeon Valley, which became much steeper south of the Temple Mount and down to the Kidron Valley (Figure 2.1).

Wide streets crossed the Tyropoeon Valley in an east-west direction, including one constructed as a viaduct supported by arches, connecting the Western Hill and the Temple Mount.27 Jerusalem's main urban artery extended along the Tyropoeon from the Damascus Gate in the north to the City of David and the Siloam Pool in the south. It was constructed in the Roman fashion as a wide street flanked by shops, forming the main commercial hub of the city and serving the large number of pilgrims who packed the city during the Jewish holy days. Particularly impressive is a large section of the street adjacent to the western wall of the Temple Mount, where some 70 m of the street were exposed in excavations (Figure 2.2). It was paved with great stone slabs, some of them more than two or three metres long, and weighing *c*. 40–50 tons each. A row of small chambers, which served as shops, was constructed between the

<sup>22</sup> Pliny, *Natural History*, 5.70.

<sup>23</sup> See Levine, *Jerusalem*; Gafni *et al*., *Jerusalem in the Second Temple Period*; Avni, 'Temple, palaces and markets'.

<sup>24</sup> For the main summaries of these excavations, see Ben Dov, *In the Shadow of the Temple*; Avigad, *Discovering Jerusalem*; *NEAEHL* 2.743–771 sv. 'Jerusalem, the Second Temple Period'; *NEAEHL* 5.1801–1837; Reich and Billig, 'Excavations near the Temple Mount'. An update is provided in Galor and Avni, *Unearthing Jerusalem*.

<sup>25</sup> See Levine, *Jerusalem*, 340–343. This range rules out some extremely high and low estimations. See for example Geva, 'Estimating Jerusalem's population', yet it seems that a reasonable estimate is *c*. 60,000–70,000 *.*

<sup>26</sup> *NEAEHL* 5. 1811–1812 sv. 'Jerusalem'. 27 See Bahat, *Carta's Historical Atlas*, 40–41. For a different interpretation based on the recent excavations west of the Temple Mount see Onn *et al*., 'Jerusalem Wilson arch'; Weksler-Bdolah, *Aelia Capitolina*, 96–109.

*Fig. 2.1: Aerial view of the Old City from the south (G. Avni and the Israel Antiquities Authority).*

street and the western wall of the Temple Mount. The first stages of this street may have been formed already in the time of King Herod, yet it was not completed until the mid-first century, as coins and pottery found under it have been dated to the reign of Agrippa I (AD 40–44) and later. It seems that this section of the street and the shops facing it were in use only for a short time, perhaps less than one generation, and were destroyed in AD 70.

The war of AD 66–70 marked the end of the prosperous Jewish Jerusalem. As described by Josephus, during the last stages of the war, in the spring and summer of AD 70 , sections of the city were set on fire and violently destroyed.<sup>28</sup> The fighting at the Western Hill resulted in a large fire that destroyed the lavish mansions, and is highly visible in archaeological excavations in this area.<sup>29</sup> The most dramatic remains of Jerusalem's destruction were exposed near the south-western corner of the Temple Mount platform, where a huge pile of gigantic stones that sealed the adjacent street was unearthed. The stones were deliberately dismantled from the upper sections of the Temple Mount wall by the Roman legions following the conquest of the city.<sup>30</sup>

<sup>28</sup> Josephus, *Jewish War* 6, 353.

<sup>29</sup> Avigad, *Discovering Jerusalem*.

<sup>30</sup> Mazar, *Mountain of the Lord*; Mazar, *Excavations in the Old City*; Ben Dov, *In the Shadow of the Temple*; Reich, 'The destruction of Jerusalem'; Gurevich, 'Jerusalem during the First War'.

What happened after the conquest and destruction? Traditional narratives, based on the descriptions of Josephus, suggest devastating effects on local Jewish society, with mass killing and expulsion of the survivors from the city.31 The depiction in Rome on the triumphal Arch of Titus showing Jewish prisoners carrying the treasures of the Temple provided a vivid artistic representation of the Jewish exile and has framed this event in modern historical and ideological interpretations.32 It seems that the combination of the dramatic historical descriptions and the vivid evidence from excavations in Jerusalem furnished a clear-cut narrative that portrayed a sharp transformation from a Jewish to pagan city following the Roman conquest. The end of Jewish presence in Jerusalem opened the road to the establishment of the Roman Aelia Capitolina some sixty years later. This concept prevailed in the modern archaeological and historical scholarship

*Fig. 2.2: The street under Robinson Arch (G. Avni and the Israel Antiquities Authority).*

of Jerusalem and was further enhanced by the results of archaeological excavations between 1968 and 2000, particularly in the Jewish Quarter and near the Temple Mount.

Yet in contrast with this clear-cut historical narrative, new excavations in the last two decades have raised doubts as to these conclusions by presenting new data on the continuity of Jewish communities after the war. In addition, recent excavations within the city show that there was a rapid revival of some urban structures and that the construction of a new city, the harbinger of later Roman Aelia Capitolia, began as early as the late first century. It seems that parallel to the reconstruction of urban structures in the city a Jewish population, probably the survivors of the war against the Romans, settled in the immediate hinterland of Jerusalem, some five to ten kilometres from the ruined city. Rescue excavations conducted recently north of Jerusalem revealed segments of a large settlement, dated between AD 70 and 132, with distinctive Jewish characteristics such as ritual baths (*mikveh*) and the massive use of stone vessels, showing that Jews continued to inhabit the Jerusalem area between the two revolts against the Romans.33 Further evidence of continuity in the Jewish

<sup>31</sup> Josephus, *Jewish War* 6.414–419.

<sup>32</sup> Millar, 'Last year in Jerusalem'; Magness, 'Arch of Titus'; Goodman, *Rome and Jerusalem*.

<sup>33</sup> Bar Nathan and Sklar-Parnes, 'Jewish settlement'; Sklar-Parnes *et al*., 'Excavations in North West

population was found in the vast necropolis of Jerusalem, which contained thousands of family burial caves hewn in the local limestone rocks.34 While most of the tombs were abandoned around AD 70, a few continued into the late first and early second centuries and were perhaps still in use by the former Jewish population, given their continuity of typical Jewish burial practices.35 A small number of Hebrew and Aramaic inscriptions from the second to the fourth centuries were found in Jerusalem, but there is no agreement as to whether these represent the inhabitants of Jerusalem or the limited number of Jewish pilgrims who were allowed to visit the city.

The continuity of Jewish settlement in the Jerusalem area is also evidenced by some historical references. Two documents, dating from the second Jewish revolt in AD 132–135 and discovered in the caves facing the Dead Sea in the Judaean Desert, were written in Jerusalem, suggesting that a Jewish population still inhabited the city after AD 70. The existence of a Jewish presence in Jerusalem during the second Jewish revolt is disputed. While most scholars agree that Jerusalem was not conquered by the rebels, and that the establishment of the Roman colony Aelia Capitolina was the result of the revolt, there may have been an attempt to conquer the city and renew worship on the now abandoned Temple Mount.<sup>36</sup>

Thus, the story of the Roman conquest of Jerusalem and the swift change from a Jewish to a pagan city, although supported both by the archaeological evidence and eye-witness descriptions, seems to contain within it some elements of continuity rather than exclusive disruption. The Jewish population was not terminally eradicated from the Jerusalem area and the next 250 years of pagan domination were dotted with Jewish and early Christian presence in the city, contrary to previous evaluations. While Roman Jerusalem was developed as a pagan centre starting from the late first century AD,37 the new archaeological evidence for the continuity of a Jewish presence around the city shows that the transformation process was not swift and clear-cut, but rather longer and more complicated than previously considered.

Why has this change in scholarly attitudes on the transition of the city occurred only in the last two decades? It seems that the new approaches to the transformation of Jerusalem between AD 70 and 132 emerged with the updated archaeological data from rescue excavations around Jerusalem in which the remains of settlements postdating the Roman conquest were revealed. At the same time, the re-evaluation of historical descriptions and their relation to archaeological data have contributed to a change in historiographical attitude to this significant military and political event. The picture of total destruction and exile, which was much forwarded throughout 150 years of archaeological research on Jerusalem, is gradually changing into a more balanced view, in the light of recent archaeological findings and the refined

Jerusalem'. For the use of soft stone vessels as a typical Jewish characteristic see Magness, *Stone and Dung*, and references therein.

<sup>34</sup> Kloner and Zissu, *The Necropolis of Jerusalem*. 35 Kloner and Zissu, *The Necropolis of Jerusalem*, but see Avni and Greenhut, *Akeldama Tombs*; Avni, 'Christian secondary use' for a different interpretation.

<sup>36</sup> See Weksler-Bdolah, *Aelia Capitolina*, 19–60 for a summary of the main discussions on this topic.

<sup>37</sup> See Weksler-Bdolah, *Aelia Capitolina* 51–60 for an updated state of research.

chronological and stratigraphic observations resulting from new excavations. The accumulating data points to the renewal of settlement following AD 70, both within the city limits and in its rural hinterland. Spot finds in and around the Old City show that urban structures were reconstructed as early as the late first and second centuries AD, the urban necropolis was reused adding new sections of pagan graves and the hinterland of Jerusalem presented an intensification of settlement in the second and third centuries AD.<sup>38</sup> It seems that the construction of Aelia Capitolina as a pagan centre, from which Jews and Christians were restricted, was an ongoing process that lasted for more than 200 years. Jerusalem at that time did not regain its previous splendour, yet it developed a significant urban presence by the fourth century AD.

#### **The Persian and Arab conquests of Jerusalem, AD 614 and AD 638**

The last hundred years of Byzantine rule over Jerusalem are traditionally seen as the triumphal period of Christian rule which was ended by two major events – the sack of the city by the Sasanians in 614 and the Arab conquest in 638. As in the structural change between the Jewish and pagan city, these conquests were viewed as the turning-point from Christianity to Islam. The Persian Sasanian conquest of Jerusalem is attested in historical sources as a violent military raid that dramatically affected the political and administrative stability of Byzantine Palestine, involving large-scale damage to churches and a mass killing of the local Christian population.39 The common view has it that the conquest marked a turning-point in the history of the Near East and was one of the causes for the rapid Arab conquest twenty years later. Although Persian domination – lasting only fourteen years (AD 614–628) – was a very brief episode in the long historical sequence of Palestine, it was believed that the devastating effects of the conquest changed the urban and rural landscape of the country for many years to come. Archaeological studies generally adopted this clear-cut narrative and considered the presumed destruction of certain buildings in Jerusalem as consequences of the conquest.

As recent scholarship shows, the realities were very different. The Persian conquest, although involving a major massacre of the city's Christian population, had only limited effects on its urban structures.40 The capture of Jerusalem by the Arabs twentyfour years later did not involve violent conquest and the city capitulated through a surrender treaty.41 An Islamic presence was only gradually imposed, and Jerusalem preserved its Christian character throughout most of the early Islamic period.<sup>42</sup>

How did such a gap arise between the traditional historical descriptions and the realities of the archaeological findings? The roots of this traditional attitude may

<sup>38</sup> See Weksler-Bdolah, 'A plan of Jerusalem' for an updated list of excavations; Avni, 'Necropoleis' for the distribution of urban cemeteries; Kloner *et al*. 'Rural hinterland' for villages outside Jerusalem.

<sup>39</sup> See for example Avi Yonah, *The Holy Land*, 124; *The Jews*, 257–277; Schick, *Christian Communities*, 33–39.

<sup>40</sup> See Avni, 'Persian conquest' for a detailed evaluation.

<sup>41</sup> Levy-Rubin, *Non-Muslims*; Gil, 'Political history'; Gil, *History of Palestine*, 51–60.

<sup>42</sup> Avni, *Byzantine Islamic Transition*, 93–132.

have been embedded in the European colonial approaches of the nineteenth century, which emphasised the superiority of Christian cultures. According to these views, the inevitable struggle between Christianity and Islam led to the swift conquest of Palestine by the Arabs and to the end of a Christian presence in the Holy Land. The traditional view of the Arabs as nomads emerging from the desert to conquer the settled lands has been refuted by modern scholarship on many grounds. Archaeological findings since the 1980s have provided an alternative approach to evaluating the Byzantine-Islamic transition.<sup>43</sup>

The place of Jerusalem in the Byzantine Empire as a major religious centre and a focus of pilgrimage was much discussed in the historical scholarship and in the reconstructions of the city's layout at its zenith in the sixth century. Perhaps the best representation of Jerusalem as a Christian city is its depiction in the Madaba Map from the second half of the sixth century, which shows the city in an oval-shaped, east–west, bird's eye view (Figure 2.3).<sup>44</sup>

The map is dominated by the depiction of one large building which covers most of its central part – the Church of the Anastasis (the Holy Sepulchre), thus emphasising the central position of the Church within the urban landscape. Around it, the map shows the configuration of the city's layout: the western and eastern *cardines* are represented as wide colonnaded streets and the city wall with numerous towers and several gates represent the limits of the *intramural* area. Apart from the Church of the Anastasis, the internal area of Jerusalem is filled mainly with churches: the large 'New Church of the Holy Mother of God and Ever-Virgin Mary' (the Nea) along the western *cardo*, and the Church of Holy Zion on Mount Zion on the right (south) section of the map. Other churches are shown, some of them identified with those mentioned in historical sources. In contrast to the significant areas devoted by the mapmakers to the Anastasis and the Nea, the large open compound of the former Temple Mount platform can hardly be seen on the map. This contradiction perhaps represents the ideological message of Jerusalem as a homogeneous Christian city in which there is no place to commemorate other, extinct religions. Consequently, the triumph of Christianity and its absolute rule over Jerusalem is represented by magnificent churches and large colonnaded streets, while no evidence of the previous Jewish or pagan presence is to be seen.45

In light of the centrality of Jerusalem to Christianity, the presumed devastating effects of the Persian and Arab conquest have been much exaggerated both in the descriptions of Byzantine sources and in modern scholarship. While the traditional archaeological discussions unequivocally accepted the historical narrative of violence and destruction, a careful survey of archaeological finds reveals no clear evidence of

<sup>43</sup> Several comprehensive works published in the last two decades address various aspects of this transition. See particularly the recent works of Kennedy, Hoyland, Donner, Walmsley, Whitcomb and Magness, discussed in Avni, *Byzantine Islamic Transition*. 44 Avi Yonah, *Madaba Mosaic Map*; Donner, *Mosaic Map of Madaba*; Tsafrir, 'Topography and archaeology',

<sup>342–51.</sup> 

<sup>45</sup> Eliav, *God's Mountain*, 125–150.

*Fig. 2.3: The Madaba Map (The Israel Antiquities Authority).*

destruction layers that can be associated with the Persian conquest. On many sites, evidence for destruction is ambiguous, and sometimes burnt layers were associated with the Persian conquest without a firm ceramic or chronological basis.<sup>46</sup>

The Arab conquest of the city in 638, even more than the Persian conquest, left no immediate impact on Jerusalem. Its main contribution to the urban landscape – the re-creation of the former Temple Mount, now the Haram al-Sharif, as a centre of Islam, and the construction of the al-Aqsa Mosque (*al-masjid al-aqṣā*), the Dome of the Rock (*qubbat al-ṣakhra*) and the adjacent palatial complex – was brought about only several decades later, in the time of ʿAbd al-Malik and al-Walīd.47 Their monumental constructions created a new Muslim hub within Jerusalem, physically separated from other sections of the city. The Islamic section dominated the eastern part of the walled city, while the previous Christian centre remained in its western and northern sections and around the Church of the Anastasis. In contrast to the clear-cut narrative of urban change in the first half of the seventh century as an outcome of the Persian sack and

<sup>46</sup> See for example the different versions of the conquest and massacre in Antiochus Strategius. Avni, 'Persian conquest' for a summary and references therein; on the credibility of historical sources see Stoyanov, *Defenders and Enemies*.

<sup>47</sup> See for summaries: Grabar, *Shape of the Holy*; Elad, *Medieval Jerusalem*; Grabar and Kedar, *Where Heaven and Earth Meet*. See now Levy-Rubin, 'Why was the Dome of the Rock built' for a new interpretation. See also Mostafa, 'From the Dome of the Chain to *Miḥrāb Dāʿūd*', and Mostafa 'Early mosque revisited'. I thank Elizabeth Key Fowden and Suna Çağaptay for these references.

the Arab takeover of Jerusalem, the updated archaeological evidence points to a much slower process which began well before these political and military events, perhaps already in the mid-sixth century. As pointed out by Hugh Kennedy, the processes of urban change in the Byzantine East started as early as the mid-sixth century.<sup>48</sup> It seems that Jerusalem fits well into this model. The period of change in its urban structures is framed between two monumental enterprises: the construction of the Nea church, inaugurated in 543 as the largest church of Jerusalem, and, almost two hundred years later, the construction of the Dome of the Rock and al-Aqsa Mosque on the former Temple Mount platform as the main Muslim shrines. The Persian and Arab conquests occurred within this framework, but they were incorporated into the *longue durée* process of urban transformation. In a wider perspective, it seems that changes in settlement patterns and social stability in Jerusalem were not dictated by short-term military episodes, but were rather part of a long-term process of cultural and religious change which spanned several centuries and culminated only on the eve of the Crusader conquest of Palestine.

What were the significant points of change in this process? While the main political event in the history of Jerusalem in the second half of the first millennium AD is considered to be the Arab conquest, the preceding and following processes are of no less relevance to the urban transformation of the city. For example, the investment in the main churches of Jerusalem – the construction of the Nea under Justinian in the 540s49 and the renovations at the Church of the Anastasis under Maurice in the 570s,50 show that in spite of the Justinianic Plague and the deteriorating economic conditions in the second half of the sixth century,51 investment in religious institutions continued.52 A century later, the re-division of properties between Christians and Muslims and the development of the Jerusalem markets under the new Islamic regime in a different form were another indication of the gradual change in urban structures from Roman times.53

The evolution of markets in Jerusalem is one example which represents the interaction between the historical narratives and the evidence of the material culture. While in the times of Aelia Capitolina commercial activities were concentrated in a tight compound in the forum area, the Byzantine period had witnessed the development of linear markets along the main arteries of the city. Markets gradually developed along the western and eastern *cardines* and their adjacent streets. This process was further intensified in the eighth to tenth centuries with the contraction

<sup>48</sup> Kennedy, 'From *polis* to *madina*'; and see now Di Segni, 'Greek inscriptions'; Di Segni, 'Late antique inscriptions' for a detailed analysis which supports this dating on the basis of hundreds of dated inscriptions from Palestine and Jordan.

<sup>49</sup> Avigad, *Discovering Jerusalem*; Gutfeld, *Jewish Quarter Excavations*.

<sup>50</sup> Di Segni, 'Epigraphic finds'.

<sup>51</sup> See for example Morrison and Sodini, 'Sixth-century economy'.

<sup>52</sup> See Di Segni, 'Greek inscriptions'; Di Segni, 'Late antique inscriptions'; Di Segni and Tsafrir, 'Ethnic composition'.

<sup>53</sup> Elad, 'Early Arabic source'; Avni, 'Markets of Jerusalem'.

of the width of the main colonnaded streets through the introduction of commercial activities into them. The role of the markets in the local economy is evident both from historical sources and from archaeological findings.<sup>54</sup> It seems that the impact of local markets on the urban tissue was much stronger in Jerusalem compared to other cities in Palestine, as they were used both by the local population and by a large number of pilgrims. As in present times, the bazaars along the main urban arteries were a meeting-place for local inhabitants of different communities, as well as for foreign visitors. The main colonnaded streets of the city, the *cardo* and the *decumanus*, were narrowed and filled with shops and other commercial buildings during the early Islamic period, as was common in other cities of the Near East.<sup>55</sup>

## **Conclusion: Urban transformations in Jerusalem, between archaeology and historiography**

The long-standing interpretative commitment to the role of a single event as the primary trigger for the urban transformations of Jerusalem was bolstered by the supposed overhauls of the ethnic composition of its population following wars and massacres reported in literary sources. Such a view proposes that in most periods Jerusalem was composed of a monolithic ethno-religious population, with an urban structure dominated by religious institutions that reflected the homogeneity of its inhabitants. This type of narrative begins by arguing that the Hellenistic and early Roman city developed as the hub of Jewish population in Palestine, and Jerusalem of the first century AD was formed as a typical Jewish city under Roman domination, in contrast to other main cities which consisted of multicultural ethno-religious populations.56 Then, it is claimed, the Jewish War and the destruction of the Temple marked the Jewish to pagan transformation, leading to the construction of Aelia Capitolina by Hadrian. The development of Jerusalem in the second to the fourth centuries as a pagan *polis* is represented as a significant change in the city's physical shape and in the demographic composition of its population.57 Following the conversion of Constantine and the pilgrimage of the Empress Helena to Jerusalem in 324, the city is now perceived to have transformed into a Christian hub and a major pilgrim destination in the Holy Land. In a similar manner, the Arab takeover in 638 is presented as the watershed between the Christian and Muslim city. This was, and

<sup>54</sup> Elad, 'Early Arabic source'; Weksler-Bdolah *et al*., 'Jerusalem Western Wall'; contra Gil, *History of Palestine*, 407; Grabar, *Shape of the Holy*, 139.

<sup>55</sup> Weksler-Bdolah *et al*., 'Jerusalem Western Wall'. See also Kennedy, 'From *polis* to *madina*', 11–13, for a general evaluation. For examples from specific sites: Beth Shean: Tsafrir and Foerster, 'Urbanism at Scythopolis', 138–140; Gerasa: Kraeling, *Gerasa*, 116–117; Palmyra: Al-As'ad and Stepniowski, 'Umayyad suq'.

<sup>56</sup> Levine, *Jerusalem*; Goodman, *Rome and Jerusalem*; Avni, 'Temple, palaces and markets'. Jerusalem stands in sharp contrast with the seaport cities of Caesarea and Ashkelon, which contained mixed pagan, Jewish and Samaritan populations.

<sup>57</sup> Avni and Stiebel, *Roman Jerusalem* for a summary of updated research.

continues to be, the dominant perception of the transformations of Jerusalem during the first millennium. It relates particularly to the two monumental landmarks of the sacred city, the Temple Mount/Haram al-Sharif and the Church of the Anastasis. In this narrative each is emblematic of a religious sovereignty: Judaism up to the conquest of Jerusalem in AD 70, Christianity during the Byzantine period, between 324 and 638, and Islam, during the first Muslim rule of Jerusalem, which was interrupted by the Crusader conquest in 1099. All these events formed the 'replacement paradigm' which predominated in the archaeological and historical research of Jerusalem for more than a century: the Jewish temple was replaced by pagan temples, which were later replaced by numerous churches crowned by the Church of the Anastasis.58 The Christian to Muslim transformation is presented as a similar process of replacement, when the religious focus of the city was transferred back to the Temple Mount, now renamed Haram al-Sharif.59

Yet in contrast to this conceptual paradigm, the extensive collection and interpretation of data from archaeological excavations provides a refined and more balanced picture, emphasising *longue durée* processes. One good example of the contribution of a detailed analysis based on archaeological findings is the case of the Jewish War and the destruction of Jerusalem in AD 70, perhaps the most documented event in the city's history. The eyewitness account of Josephus Flavius, in conjunction with the evidence of violent destruction as revealed in the excavations, supply a clearcut picture of collapse and exile. At the same time, the updated archaeological data provides evidence of continuity of a Jewish presence in the Jerusalem area, with signs of reconstruction of the ruined city. This gradual urban change lasted a few decades, and when Aelia Capitolina emerged as a pagan city in AD 132, it bore imprints of the previous early Roman city.

The Christian-Muslim transformation (or perhaps interaction) is more nuanced and complicated. The traditional narratives of the Persian and Muslim conquests as triggers of change have been replaced by a new approach which emphasises continuity rather than disruption. A gradual change developed during the 200 years between the sixth and eighth centuries and was framed by the construction of central religious monuments – the Nea church and the Dome of the Rock and al-Aqsa Mosque. Contrary to the 'replacement paradigm', the Christian-Muslim transformation did not alter the existing components of the city. The new Muslim compound at the Haram al-Sharif, constructed on the temenos of the former Jewish temple, only had a limited effect on other sections of the city. Although the two religious complexes,

<sup>58</sup> Corbo, *Santo Sepolcro*; Biddle, *Tomb of Christ*; Biddle *et al*., *Church of the Holy Sepulchre*.

<sup>59</sup> This concept is poetically described in Caseau, 'Sacred landscapes', 49: 'Like rivers that create currents when they meet the sea, the religions in competition in the Roman and Persian worlds altered the societies they encountered. In regions like Syria and Palestine the currents were the strongest. First the sounds and perfumes of Pagan processions had filled the streets of Syrian cities; then the music of Christian hymns and the odor of incense sanctified the urban space as Christians went in procession from one sanctuary to another; finally their voices were silenced by the chanting of the muezzin calling the Muslims to prayer from the minarets which dominated the ancient cities'.

the Christian in the west and the Islamic in the east, dominated the city's skyline, the urban development of Jerusalem in the early Islamic period left its impact mainly on the outer neighbourhoods of the city. The main arteries and colonnaded streets, the churches and monasteries which dotted the city and its hinterland and the residential quarters within and outside the city walls, all reflected a slow process in which Christian, Jewish and Muslim communities established their identities and shaped their well-defined urban quarters. Given these conclusions, it seems that a new approach is required for the evaluation of urban changes in Jerusalem in both transitional periods discussed here. While the process of change may have been influenced by a single event such as the Roman or the Arab conquest, it was in fact a prolonged, slow and nuanced process of urban transformation.

#### **Bibliography**

Amichai, Y., *Poems of Jerusalem: A Bilingual Edition* (Tel-Aviv, 1987).


Avni, G., 'Christian secondary use of Jewish burial caves in Jerusalem in the light of new excavations at the Aceldama Tombs' in F. Manns and E. Alliata (eds.), *Early Christianity in Context: Monuments and Documents*, Studium Biblicum Franciscanum Collectio Maior 38, (Jerusalem, 1993), pp. 265–275.


Ben Ami, D. and Y. Tchekhanovets, *Excavations in the Tyropoeon Valley (Giv`ati Parking Lot) vol. II: The Byzantine and Early Islamic Periods* (Jerusalem, 2020).


## Chapter 3

## Visualising Constantinople as a palimpsest

*Robert Ousterhout*

*For Paul Magdalino*

#### **Introduction**

Istanbul is a city of juxtapositions. We have grown accustomed to so many of them that we hardly look twice. The minarets of Hagia Sophia, for example, were added gradually during the fifteenth and sixteenth centuries, but it is difficult to imagine the building without them, as they seem to both frame and complete the monument (Figure 3.1).1

Other juxtapositions are perhaps more jarring. After the church of the Theotokos Pammakaristos was converted into the Fethiye Camii, for example, its bema was destroyed by a blocky addition to redirect the naos toward Mecca, while interior walls, piers and columns were cut down or eliminated to convert the labyrinth of Byzantine monastic and funerary spaces into a more uniform congregational mosque.2 The changes were so drastic that it is all but impossible to reconstruct its original form or even aspects of its relative history today (Figure 3.2). Juxtapositions such as these, whether functional or symbolic, whether jarring or not, are but one reflection of the transformation of the city through its troubled, millennial history. While it might be difficult to view the transformations at the Fethiye Camii as aesthetic 'improvements', the addition of the minarets at the Hagia Sophia both evoked a conquered Constantinople and established a template for Ottoman mosque designs to follow.

<sup>1</sup> Müller-Wiener, *Bildlexikon*, 84–96.

<sup>2</sup> Müller-Wiener, *Bildlexikon*, 132–135; Hallensleben, 'Untersuchungen zur Baugeschichte der ehemaligen Pammakaristoskirche'; Belting *et al*., *Mosaics and Frescoes of St. Mary Pammakaristos.*

*Fig. 3.1: Istanbul. Hagia Sophia, seen from the west (author).*

*Fig. 3.2: Istanbul. Fethiye Camii (Theotokos Pammakaristos), hypothetical plan, showing multiple phases (author).*

As the Fethiye Camii illustrates, Ottoman Istanbul often sits uncomfortably on its Byzantine past. For the predominantly Muslim Ottoman Empire, the Christian past represented something entirely different from its present reality. While other nations and capitals had appropriated their past to forge a modern identity, the pre-Ottoman past was not useful to the Ottomans; it stood in dramatic opposition to the present by both its Hellenism and its Christianity; it could never provide validation or symbolic underpinnings to an Ottoman regime – at least after the first Ottoman century, and clearly not in the late Ottoman period.3 Put another way, it could serve to empower the present regime only by remaining conquered – that is, if its churches continued to function as mosques and if its spaces of power remained unexplored. The past was best left buried. This attitude continued into the Republic, which (with the exception of the Hittites) emphasised modernity over antiquity.

*Fig. 3.3: Istanbul. Excavation for subway, near Aksaray, revealing multiple strata of architecture (author).*

But cities like Istanbul are forever in the process of becoming, despite the efforts of historians to fix them at specific moments in time. If we penetrate the Ottoman veneer in an attempt to recover the Byzantine city, even this becomes a complicated process, for Constantinople was *always* a city in transition. Witness the innumerable cultural layers uncovered in the subway construction or at the Yenikapı excavations (Figure 3.3).4 When I first visited Istanbul in the 1970s, I came away with the notion that one could dig anywhere and find something. I have since modified this view: one could dig anywhere and find *layers* of something. Indeed, the limited excavations of the twentieth century discovered just that. Archaeologists identified, for example, five Byzantine phases at the Chora Monastery (Kariye Camii), another five at the Kyriotissa Monastery (Kalenderhane Camii).5

While Constantinople is normally discussed as a late antique city, the vast majority of its monuments are later in date and represent different and evolving concepts

<sup>3</sup> As I have discussed elsewhere, see Ousterhout, 'The rediscovery of Constantinople'.

<sup>4</sup> Zeynep Kızıltan, *Gün Işιğιnda: İstanbul'un 8000 yιlι*.

<sup>5</sup> Oates, 'Summary report ', amplified by Ousterhout, *Architecture of the Kariye Camii*. For the Kalenderhane, see Striker and Kuban, *Kalenderhane*.

*Fig. 3.4: Istanbul. Çemberlitaş (Constantine's Column), general view (author).*

of urbanism.6 In the absence of urban archaeology, however, it is difficult to visualise the transformations on an urban level. For example, texts hint at the loss of function and meaning in the Forum of Constantine, but there is no archaeological evidence to illustrate change – other than the forlorn and much-altered column that once stood at its centre (Figure 3.4).<sup>7</sup> While this may represent our best evidence for the presence of Constantine in the city that bore his name, the column has been effectively muted by its later history – the dedicatory inscription for the restoration by Manuel Komnenos (r. 1143–1180) fails to mention Constantine. And it's not particularly attractive.

In this paper, I shall examine a few discrete examples for which there is archaeological evidence of transformation over the *longue durée* and discuss what they might tell us about the reconceptualisation

of the city around them. The image of Constantinople I would like to evoke is of a city in transition.8 Based on the evidence provided by its surviving monuments, it should be possible to envision the city not as something static and fixed in time, but as something dramatic and changing. Although we have little evidence for either the maintenance or the elaboration of the urban matrix, the transformation of specific monuments may be seen as reflecting the changing nature of the city as a whole. In this respect, the close analysis of surviving buildings may be informative.

## **Hagia Euphemia**

I begin with one of my favorite examples, the church of Hagia Euphemia at the Hippodrome, still partially standing but known from excavations (Figure 3.5).9 It began its life in the early fifth century as the centrepiece of an aristocratic palace built by a

<sup>6</sup> See Mango, *Le développement urbain*; Dagron, *Naissance d'une capitale*; Dagron, *Constantinople imaginaire*; Magdalino, *Constantinople Médiévale*; translated as 'Medieval Constantinople' in P. Magdalino, *Studies* 

*on the History and Topography of Byzantine Constantinople* (Aldershot, 2007), pp. 1–111. 7 See most recently, Ousterhout, 'The life and afterlife of Constantine's column'. See also the contributions by Göksun Akyürek and Paul Magdalino in this volume.

<sup>8</sup> For much of what follows, see Ousterhout, 'Constantinople and the construction of a medieval urban identity'.

<sup>9</sup> Naumann and Belting, *Die Euphemia-Kirche*; more recently, Akyürek, *Khalkedon'lu (Kadıköy) Azize Euphemia ve Sultanahmet'teki Kilisesi*.

high court official, an uppity eunuch of Persian origin named Antiochus, who served as *praepositus*, *cublicularius*, praetorian prefect, patrician and consul under Theodosius II (r. 402– 450). The palace was entered from a street perpendicular to the north side of the Hippodrome, accessible by means of an elegant semicircular portico and garden. Symmetrically disposed pavilions of intricate geometry opened onto the portico. On the central axis lay a large ceremonial hall or *triclinium*, which was hexagonal and niched in plan, covered by a dome. Small colonnaded porticoes opened outward to the surrounding gardens. We might imagine the palace complex to have been a smaller version of the now lost pavilions and gardens of the Great Palace, for when Antiochus

*Fig. 3.5: Istanbul. Palaces excavated by the Hippodrome, plan (author, after W. Müller-Wiener,*  Bildlexikon zur Topographie Istanbuls*, fig. 109).*

overstepped his position, he was accused of living 'like an emperor, not like an emperor's minister.'10 His palace was confiscated around 436, and he was forced to take holy orders. Sometime later – although it remains unclear exactly when – its hexagonal hall was converted into a church, dedicated to a local saint, Euphemia, the 'all-praised,' whose main shrine lay across the Bosporus in Chalkedon, on the site of her martyrdom in 303. In the troubled seventh century, when the Asian shore was threatened by Persian attack, the relics of Euphemia were transferred by the Emperor Heraclius for safekeeping, either in 615 or 626.11 Thomas Mathews had argued for a sixth-century conversion for the building – that is, before the transfer of relics.12 A later conversion seems more likely, perhaps even as late as the recorded 796 restoration by Eirene (r. 797–802) and Constantine VI (r. 780–797).13 Once Euphemia was established near the Hippodrome, however, tombs and mausolea were added around the building as her cult grew in importance.14 In the late thirteenth century, following the Latin Occupation, the church was restored once again and redecorated with a cycle of paintings chronicling the Life of Euphemia.15

<sup>10</sup> Lavin, 'The house of the Lord', 1–27; Zonaras, *Epitome historiarum*, 12.22.

<sup>11</sup> *Acta sanctorum*, Sept., 5: 275; Naumann and Belting, *Die Euphemia-Kirche*, 23–24.

<sup>12</sup> Mathews, *The Early Churches of Constantinople*, 61–67.

<sup>13</sup> Mango, 'The relics of St. Euphemia'. I thank Jordan Pickett for his observations of H. Euphemia.

<sup>14</sup> Naumann and Belting, *Die Euphemia-Kirche*, 49–53; with limited remains, the dates of the mausolea remain uncertain.

<sup>15</sup> Naumann and Belting, *Die Euphemia-Kirche*, 112–194.

Each step in the transformation of the site had important reflections in the transformation of the city around it. Antiochus's palace lay in a position indicative of his political importance. The emphasis on ceremonial spaces reflects the ritualisation of daily life among the aristocracy in Late Antiquity; Antiochus, as the charges against him suggest, was living like an emperor. The conversion of Antiochus's palace into a place of Christian worship reflects the rising power of the church, which gradually took over many official duties of the state administration and assumed a role in civic governance.16 Similarly, the incorporation of the saint's relics reflects both the reduction in scale of the city during the Transitional Period (seventh to mid-ninth centuries) and its increasingly sacred character. While Euphemia's original martyrium in Chalcedon seems to have fallen off the historical record, the relocation of her sanctuary at the very centre of the city accords with the functional and ideological redefinition of the city in the early middle ages. A restoration at the end of the eighth century would fit with the urban revival under the efforts of Constantine V (r. 741–775) and his successors.<sup>17</sup> The addition of burials around the church marks a fundamental transformation in the character of the city. While Roman law had prohibited burials inside the *pomerium* of the city, during the middle ages intramural burials became increasingly common, often set in relationship to sacred space or sacred objects.18 Finally, the redecoration of the church in the late thirteenth century represents the attempt to reassert the city's sanctity in the wake of the Latin Occupation, when many of its holy shrines were plundered. As a defender of Orthodoxy whose shrine lay at the heart of the city, Euphemia would have gained new resonance against the backdrop of Palaiologan attempts for a union with the Church of Rome.19

#### **Dynamic examples, static texts**

Similar analyses could effectively address urban transformations as reflected in the Kyriotissa church, the Myrelaion Palace or even the fortified reduction of the Great Palace into what's known as the Bukoleon Palace.<sup>20</sup> For each, the evidence provided by archaeology offers considerably more nuance than what we might find in the texts alone. This is an important point, for in terms of Byzantine scholarship Constantinople has been much more the realm of the philologist that the archaeologist, better known through its evocative descriptions and ceremonial texts than through its monuments or its urban infrastructure. Few Byzantine writers attempt to address the *longue durée*; most appear trapped in an eternal present. Indeed, the dynamic, changing urban fabric stands in sharp contrast to the static impression provided by the texts.

<sup>16</sup> Thomas, *Private Religious Foundations*.

<sup>17</sup> Magdalino, 'Constantine V and the middle age of Constantinople'.

<sup>18</sup> Constas, 'Death and dying in Byzantium'; Snively, 'Old Rome and new Constantinople'; Saradi-Mendelovici, 'The demise of the ancient city'.

<sup>19</sup> Talbot, 'The restoration of Constantinople under Michael VIII', 243–261, does not discuss Euphemia; for the period in general, see Geanakoplos, *Emperor Michael Palaeologus and the West*.

<sup>20</sup> Ousterhout, 'Constantinople and the construction of a medieval urban identity'.

*Fig. 3.6: Constantinople, plan of city showing Basil I's ceremonial route (author, after C. Mango,* Le développement urbain de Constantinople*, plan II).*

While noting the origins of the rituals it records, for example, the *Book of Ceremonies* reads as if all the ceremonies are being repeated in perpetuity, without alteration, in settings that do not change.

This is emphasised in the account of the triumphal entry celebrated by Basil I (r. 867–886) in 879 (Figure 3.6). Its stational ceremonies occurred at major late antique public spaces, with virtually no indication of the transformation of the city. The procession could have taken place centuries earlier with almost no alteration. Basil enjoyed two ceremonies outside the walls and stopped for ten receptions between the Golden Gate and Hagia Sophia.21 Following a military parade into the city, Basil was acclaimed by both factions and the demarchs in appropriate ceremonial dress. The cortege proceeded through the city on horseback, stopping for acclamations and receptions at the Sigma, the Exakionion, the Forum of Arcadius, the Forum Bovis, the Capitol, the Philadelphion, the Forum of Theodosius, the Artopolia and the Forum of Constantine. There they dismounted and were met by the patriarch outside the church

<sup>21</sup> Haldon, *Byzantium in the Seventh Century*, 140–147; McCormick, *Eternal Victory*, 154–157, 212–230.

*Fig. 3.7: Istanbul. Fenari İsa Camii (Theotokos tou Libos), view from the east (author).*

of the Theotokos, which Basil had founded. After changing into civilian garb, they continued on foot to the Milion, and to Hagia Sophia. Finally, they proceeded to the Great Palace, according to the usual ceremonial movements, terminating in a banquet.

What is perhaps most remarkable here is that with the exception of the church built by Basil, all monuments or public spaces used in the ceremony were pre-Iconoclastic; most were Roman in character. While the rhetoric surrounding Basil emphasised his renewal of the city, we are given no indication of the condition of the monuments and monumental spaces. The appearance of continuity is emphasised, but the ceremony may mark a transformation from actual order to symbolic order. The text of rituals like Basil's entry may demonstrate the gap between the desires and the realities of later Byzantine society.

### **The Monastery** *tou Libos*

Is there a difference between how the Byzantines saw their city and how they wrote about it? If we turn to the surviving monuments, there are a few clues, particularly with the restoration of older monuments. At the Monastery *tou Libos*, for example, the early tenth-century church was expanded in *c*. 1282–1303 by the dowager empress Theodora Palaiologina, widow of Michael VIII. The old church of the Theotokos was expanded by the addition of a funerary church to the south, dedicated to John the Baptist, enveloped by an outer ambulatory equipped for additional burials (Figures 3.7–3.8).<sup>22</sup> The connection of the two churches is not straightforward: the south annexed chapel of the older church was transformed into the prothesis of the new church, and the preservation of the older stair tower required the new narthex to be asymmetrical, with an off-axis dome. Most striking in the additions is the accommodation for burial throughout the complex. Clearly, privileged burial was of prime concern.23

Complexity becomes the watchword of its design, with disparate apses aligned along the east façade, and the roofline featuring multiple domes (all now missing or replaced). There is little attempt at integration: the parts read individually, without rigid conformity, and speak to the inherent significance (and multiple functions) of the whole. The east façade provides an impression of its original elegance: the careful ordering, stark simplicity and

*Fig. 3.8: Istanbul. Plan of Fenari İsa Camii (Theotokos*  tou Libos*) (author, after V. Marinis,* Architecture and Ritual in the Churches of Constantinople: Ninth to Fifteenth Centuries*, plan XXIII-4).*

symmetry of the tenth-century apses contrasting with the asymmetry, complexity and decorative profusion of the thirteenth-century apses – an object-lesson in the differences between Constantinopolitan architecture of the Middle and Late Byzantine periods – an impression which I would argue was intentional, meant to visually highlight the history of this important monastery. In effect, the building itself could be understood as a palimpsest.

#### **The Chora Monastery**

The Monastery of the Chora could be read in similar terms. Restored and lavishly decorated by the statesman and scholar Theodore Metochites, *c*. 1316–21, the twelfth-century naos was refurbished and stabilised, its dome replaced, and it was

<sup>22</sup> Macridy, 'The Monastery of Lips and the burials of the Palaeologi'; Marinis, *The Monastery tou Libos*. For the Typikon, see Talbot, '*Lips*: *Typikon* of Theodora Palaiologina'.

<sup>23</sup> Marinis, 'Tombs and burials in the Monastery *tou Libos* in Constantinople'.

*Fig. 3.9: Istanbul. Kariye Camii, view from east, showing multiple phases of construction (author).*

enveloped by additions.24 These included the rebuilding of the pastophoria, with the southern isolated from the bema to serve as a private chapel. A twostoried annex flanked the naos to the north, the upper level functioning as the founder's study, accessible by a vaulted staircase set into the northern wall. Two broad narthexes fronted the building to the west – the inner topped by two asymmetrical domes; the outer opened by a portico façade (now blocked). To the south a large, domed funeral chapel

or *parekklesion* was added to provide spaces for the privileged burials of Metochites, his family and compatriots. A belfry rose at the south-west corner (now replaced by a minaret), while the twelfth-century apse was stabilised by the addition of a flying buttress.

At first glance, the design may appear an incoherent jumble, but it is governed by a series of unrelated axes – both in terms of its major facades and its plan (Figures 3.9–3.10). From the west façade, one axis leads to the naos, the other to the *parekklesion*, although neither is framed symmetrically. From the south, one axis aligns a portal with the inner narthex and its dome, while the other aligns the *parekklesion* dome with the naos dome. The large size and the position of the south inner narthex bay are reflected in the detailing of the south and west facades. The *Deesis* mosaic fills its eastern wall.

In spite of the lack of clear relationships among the architectural elements and the odd juxtapositions of spaces, the fourteenth-century additions were nevertheless high in quality and the result of a single phase of construction – that is, the puzzling design was the result of intention, rather than happenstance. No attempt was made at symmetry, and the numerous functional units received individual expression. The formal organisation might be described as mannerist (or even postmodern), consciously breaking fixed patterns, creating surprising juxtapositions in the relationship of parts to the whole. At the same time, it is possible to read the architecture and decoration of the Chora as part of a discourse with the Byzantine past and its urban identity. In the building program, we find new architectural additions artfully set against older elements, which were left exposed. With his wealth, Metochites could have easily afforded to start from scratch, but he chose not to – instead, he chose to preserve and build around the older core of the complex. The juxtaposition of old and new was intentional. The new portions may be understood as a response to history, an attempt to establish a symbolic relationship with the past. The domes of the naos and the funeral chapel are aligned, for example, and the detailing of the older apse

<sup>24</sup> For what follows, see most recently Ousterhout, 'Reading difficult buildings'.

is reflected in that of the newer. Moreover, the fourteenth-century builders seem to have been inspired by the difficulties of adding to an older building, to design around, but to maintain the integrity of, the historical core of the monastery. The masons would appear to be addressing not just new functional considerations, but also the symbolic significance of the historical setting. A view of the east façade today is instructive, with exposed remains from the sixth, ninth, twelfth and fourteenth centuries. Although I was initially puzzled by the visibility of the early substructures, I would now

*Fig. 3.10: Istanbul. Plan of Kariye Camii, showing organisation by unrelated axes (author).*

argue that in Theodore Metochites' restoration of the complex, they were meant to be visible, palimpsest-like, to situate the monastery within the larger context of urban history. Metochites, I propose, wanted to highlight his own additions, clearly demarcating them from the older elements.

### **Conclusions**

In contrast to this willful avoidance of the present reality evident in earlier texts, Late Byzantine patriographic writings may offer a more transparent view.25 Most important among these works is the *Byzantios*, an oration in praise of Constantinople composed by Theodore Metochites, the patron of the Chora Monastery.26 The rhetorical attention to the city's past greatness coincides with the period of revival under Michael VIII (r. 1261–1282) and Andronikos II (r. 1282–1328).<sup>27</sup> But Metochites seems to recognise the diminished state of affairs, although he gives it a positive spin: Constantinople as a city is constantly regenerating herself. By way of comparison, he points out that as birds molt, new feathers appear amid the older plumage; in an evergreen plant, losses are not fatal but are replaced by new growth. In a like manner Constantinople renews herself, he argues, so that ancient ruins are woven into the city's fabric to assert their ancient nobility amid the new constructions. Similarly, he notes how the ruins of the city are recycled in new constructions both within the city itself and – as evidence of the city's generosity – in other cities as well. The intended message of Metochites'

<sup>25</sup> Fenster, *Laudes Constantinopolitanae*, 185 ff.

<sup>26</sup> Vienna, Nationalbibliothek, *Codex Vind. Phil. Gr.* 95, 274v-275r, as discussed by Magdalino, 'Theodore Metochites'.

<sup>27</sup> Magdalino, 'Theodore Metochites'.

encomium is of unchanging greatness, implying that the new creations replicate the pattern of their predecessors, while glossing over the tawdrier realities of ruin and spoliation. For Metochites, Constantinople could be simultaneously eternal *and* a city in transition. Or to put it another way, the palimpsest of the city only makes sense when all the layers are read at the same time.

## **Bibliography**

*Acta sanctorum*, 71 vols. (Paris, 1863–1940).


Naumann, R. and H. Belting, *Die Euphemia-Kirche am Hippodrom zu Istanbul und Ihre Fresken* (Berlin, 1966).

Oates, D., 'A summary report on the excavations of the Byzantine Institute in the Kariye Camii: 1957 and 1958', *Dumbarton Oaks Papers* 14 (1960): 223–231.

Ousterhout, R. G., *The Architecture of the Kariye Camii in Istanbul* (Washington, D.C., 1987).

Ousterhout, R. G., 'Constantinople and the construction of a medieval urban identity' in P. Stephenson (ed.), *The Byzantine World* (London, 2010), pp. 334–351.


Striker, C. L. and Y. D. Kuban, *Kalenderhane in Istanbul: The Buildings*, 2 vols. (Mainz, 1997, 2007).


Thomas, J. P., *Private Religious Foundation in the Byzantine Empire* (Washington, D.C., 1987).

Zonaras, Ioannes, *Epitome historiarum*, ed. I. Dindorf, 6 vols. (Leipzig, 1868–1875).

## Chapter 4

## Transcultural encounters in medieval Anatolia: The Sungur Ağa Mosque in Niğde

## *Suna Çağaptay*

On a gloomy day at the University Library in Cambridge, I was researching the cultural overlaps and built environment in medieval Anatolia when I found myself in the main stacks reading about the Mongol Empire. After a long day reading about

the Mongols and their offshoots, such as the Ilkhanids based in Iran and Anatolia, I decided to decompress by finding a book that had nothing to do with the Mongols. I was looking for something more artistic, and I landed upon an impressive work published in 1973 entitled *Armenian Manuscripts in the Walters Art Gallery*. In it, Sirarpie Der Nersessian, a renowned historian of Armenian and Byzantine art, examines an image labelled MS. W539, fol. 19 recto (Figure 4.1), which comes from an Armenian manuscript produced in 1262 in Cilicia that is now located in the collection of Baltimore's Walters Art Museum. The scene has an inscription that reads: 'The Magi returned to their country', and, as Der Nersessian notes, there is the appearance of a Mongol figure. Here I was again – back to familiar yet mystifying terrain: a transcultural encounter in Anatolia and all it implied and promised.

*Fig. 4.1: Return of the Magi, MS. W539, fol. 19 recto, (Walter's Art Museum, Baltimore).*

#### 62 *Suna Çağaptay*

In the 'Return of the Magi', a take on the idea of the Three Kings 'coming from the East' and offering gifts to the newborn Jesus, the figure garbed in a red turban refers to the East, where the Mongols originated. This figure in red headdress, or a Mongolised 'Magus', is perhaps meant to show a Mongol as perceived by the Armenians.<sup>1</sup> I saw again that I could not look at this region, this period, and omit mention of the Mongols and their arrival from the East. I could not, as the Armenians had once hoped, magically catapult them back to their homelands.

The arrival of the Mongols into Anatolia took place in the aftermath of the battle of Kösedağ in 1243, which resulted in the defeat of the last Rum Seljuk ruler.<sup>2</sup> The Ilkhanids – meaning the 'lesser *khanate*', subordinate to the Great Khans in China – controlled much of central and eastern Anatolia until their decline in the 1330s.<sup>3</sup> During this period, the interplay among cultural activity, religious practices and political alliances was striking.<sup>4</sup> The local architectural forms and details were already amalgamated from the architectural cultures of Anatolia, northern Syria, the Caucasus, Iran and Central Asia and these imported forms and details were then influenced by the Ilkhanids to create the cultural landscape of Anatolia from 1240 to the 1330s.5

Herein, I use the Mosque of Sungur Ağa (Figure 4.2), an Ilkhanid mosque built in 1335 in Niğde, to explore the translation of different forms, techniques and details that formed the cultural and political geography of Niğde, Anatolia and the eastern Mediterranean.<sup>6</sup> The mosque, as previous scholarship has discussed, demonstrates an amalgam of forms associated with the Gothic architecture of the Latin East, Armenian moulding and fenestration forms and several Rum Seljuk details. Previous scholars have dealt only narrowly with the Ilkhanids' hybrid vocabulary that emerged at the Sungur Ağa – the Crusader contribution has been discussed using a limited number of comparisons and there has been a failure in addressing the overlaps with Armenian architecture due to nationalist paradigms, instead seeing Ilkhanid architecture as paving the way to its Muslim successors the Eretnids and Karamanids.<sup>7</sup> Those studies

<sup>3</sup> Amitai Preiss, *Mongols and Mamluks.*

<sup>1</sup> Der Nersessian, *Armenian Manuscripts*, plate 45, fig. 59. Grigoryan, 'Manifestations of Mongol-Armenian relations', revisits this pictorial depiction within the context of visual manifestations (coinage and manuscript production), esp. 273–74.

<sup>2</sup> Cahen, *Formation of Turkey*, 71. I use the term 'Ilkhanids' when referring to the group in its own realm in Anatolia and Persia, while I use 'Mongols' in the larger context. For the dynastic history of the Ilkhanids, see Amitai, 'Il-Khanids, Dynastic History'.

<sup>4</sup> Komaroff, *Legacy of Genghis Khan*, 49.

<sup>5</sup> More recently, Blessing, *Rebuilding Anatolia*, for further bibliography.

<sup>6</sup> The Master's thesis written by Didem Esin, 'A study on possible foreign impacts', reviews the scholarship on the building and discusses the role of trade in the emerging architectural forms and details. Demircan, 'Sungur Ağa Camiisi ve Türbesi'. For a general history of Niğde, see Wittek, 'Nīgde'.

<sup>7</sup> For a discussion on the Uygur and Buddhist origins of the Eretnids and their conversion to Islam, see Cahen, 'Eretna'. For the Turkic origins of the Karamanids see Sümer, 'Ḳarāmān-Ogh̲ullari̊'; Aslanapa, *Turkish Art and Architecture*, 176; Sözen, *Türk Mimarisinin Gelişimi,* 43 and Akmaydalı, 'Niğde Sungurbey Camii', 147 accept it as an Eretnid building. Arseven, *Türk Sanatı Tarihi,* 47 attributes the building to the Karamanid period. The following scholars led by Diez, *Karaman Devri Sanatı*, 160; Gabriel, *Niğde Türk Anıtları*, 37, and more recently Özkarcı, *Niğde'de Türk Mimarisi*, 51, argue for an Ilkhanid patronage.

*Fig. 4.2: Sungur Ağa Mosque, general view from the east (author).*

primarily examined the buildings in Niğde independently or studied them from the perspective of architectural mouldings on the portals.<sup>8</sup>

In this paper, my aim is twofold: first, I will revisit the previous scholarship and discuss how the architectural and decorative forms were used, selected and translated beyond their points of origin into Ilkhanid territory.<sup>9</sup> Second, I will elaborate on how and why the Mongol-Ilkhanid hybrid architectural vocabulary at the Sungur Ağa differed strikingly from the broader Persian context from which it emerged. In doing this, I will contextualise the mosque by looking at examples of contemporary structures, specifically the Alaeddin Mosque, built in 1223, and the Mausoleum of Khudavand Khatun, built in 1312. Accordingly, I will argue that the dynamic process of transculturation at the Sungur Ağa emphasises a multidirectional exchange of forms as well as a simultaneous movement across and within, as seen in the moulding details, architectural forms and manuscript production. The term transculturation is introduced by the anthropologist Fernando Ortiz in 1947 and was subsequently

For an overview of the scholarship examining the cultural attributions see Esin, 'A study on possible foreign impacts', 20–22.

<sup>8</sup> Two such examples are Şaman-Doğan, 'Niğde'deki Türk Dönemi (13–15. Yüzyıl)'; Parla, 'Melike Hüdavend Hatun Türbesi'.

<sup>9</sup> For a comparative analysis on the use of the Islamic forms in Armenian settings, Guidetti, 'Islamicness', 155.

used in visual studies to explore the processes of when different cultures come into contact and the resulting hybrid vocabulary.10 Sungur Ağa, the mosque's patron and the governor of Niğde, used this process to leave his imprint on the landscape of the city, appropriating Crusader and local Islamic and Armenian architectural forms and showcasing his mosque as a product of the rivalry for control of his city.

### **Framing the Sungur Ağa Mosque**

Niğde, in the heartland of the Central Anatolian region of Cappadocia, was situated on a major trade route connecting Sinop on the Black Sea through Cappadocia to the Cilician Gates, and from there towards the eastward routes to the major trading centre of Tabriz.<sup>11</sup> In 1333, two years before the construction of the mosque, Ibn Battuta described Niğde as 'a large, thickly populated city, part of which is in ruins'.<sup>12</sup> Albert Gabriel,13 a French art historian, made the first detailed study of the Sungur Ağa Mosque in 1931, and he labelled it 'one of the most curious structures of medieval Anatolia'. He discussed how Sungur Ağa employed Muslim artists for the decorative aspects, such as the design of the mihrab and the minbar, and Christian artists for the stone masonry. He also referenced other influences such as the Lusignan Crusaders and Armenian Cilicia.<sup>14</sup> More recently, Baha Tanman corroborated Gabriel's findings and argued that two different teams of masons must have worked at the construction. He suggests the Gothic builders might have adapted their art of construction to make it appealing to the Anatolian taste.15

The architectural vocabulary of the Sungur Ağa Mosque differs strikingly from the mosques in the broader Persian context. The Ilkhanids converted to Sunni Islam under their ruler Ghazan (r. 1292–1304).<sup>16</sup> Prior to the Ilkhanids' arrival in Anatolia, their architectural idiom in Iran was characterised by monumentality touched with sophistication and grace, manifested in mosque and tomb complexes built in stone and baked brick and decorated with stucco, lustre and glazed tiles. Oljaytu's tomb located in Sultaniyya, Iran is an example of this. It was built as part of a socio-religious

<sup>10</sup> See for example, Latour, *We Have Never Been Modern*, 3, 11, 121; Flood, *Objects of Translation* studies transculturation in the Hindo-Islamic context.

<sup>11</sup> Turan, *Türkiye-İtalya İlişkileri,* 1.132–139. For the Seljuk history of Niğde, see Ertuğrul, *Niğdeli Kadı Ahmed'in*, (cf. reign of different sultans and fortifications repairs) and Cowe, 'Patterns of Armeno-Muslim interchange', 84–85; for the treaty signed between the Ilkhanids and the Venetians in 1331–1332, which secured the route extending from Caesarea to the Cilician port of Ayas and Tabriz, see Sinclair, *Eastern Trade*, 3–28 and Peacock, 'Black Sea trade and the Islamic world'.

<sup>12</sup> Gibb, *Travels of Ibn Battuta*, 433.

<sup>13</sup> Gabriel, *Niğde Türk Anıtları*, 39.

<sup>14</sup> Gabriel, *Niğde Türk Anıtları*, 39; Esin, 'A study on possible foreign impacts', 49–50. 15 Tanman, '14. ve 15. Yüzyılların Anadolu Türk Mimarlığında Gotik Etkiler', 213–225.

<sup>16</sup> Komaroff, *Legacy of Genghis Khan*, 120. The Ilkhanids chose Sunni Islam under Ghazan in 1295, while his successor Oljaytu (r. 1304–1316) opted for Shiism. Prior to Ghazan and his conversion to Islam, Hülagu (r. 1258–1265) wished to convert to Christianity. See Bardakjian, *Armenian Apocalyptic Tradition*, 111.

complex and bears an octagonal structure decorated with painted plaster. It is made of tile and brick and is marked by eight minarets.<sup>17</sup>

The cultural landscape the Ilkhanids entered in Anatolia was already a hybrid one. The period of the Islamisation of Anatolia from the late eleventh century onward featured a selective set of decorative borrowings, including muqarnas vaults, arabesque mouldings and ribbed vaults, that were circulated across different regions, such as Divriği and Trebizond, through itinerant masons and artists.18 Similarly, the Ilkhanid buildings constructed in Anatolia and Niğde in the aftermath of the 1243 battle of Kösedağ borrow from Rum Seljuk and Caucasian architectural practices. Examples include the Ilkhanid Çifte Minareli Medrese in Erzurum (1290–1300), built by the Ilkhanid vizier Shams-al Din Juwayni, and the Alaeddin Mosque in Niğde, built in 1223 by Zayn al-Din Bashara.19 This led to the creation of a hybrid and innovative architectural vocabulary. Hence, the forms we associate as Crusader, Armenian or Islamic cannot be interpreted at their face value, *i.e.*, they cannot be seen as Crusader, Armenian or Islamic but rather they were used by the elite or ruling figures as part of the chivalric wish to link themselves to the innovative artistic and architectural framework that was recognised by the nearby elites and rulers.20

#### **The Sungur Ağa: Architecture, epigraphy, decoration**

The Sungur Ağa Mosque was built in 1335, coinciding with the dissolution of the Ilkhanid power. It is named after its patron, Saif al-Dewlet wal-Din Sungur Ağa (shortened in modern scholarship to Sungur Ağa), the governor of Niğde and advisor to the Ilkhanid ruler Abu Sa'id (r. 1316–1335). The mosque is constructed entirely of ashlar blocks and follows a rectangular plan measuring 32.9 by 24.5 metres.<sup>21</sup> Portals are situated on the east and the north. Adjoined to the south side of the east wall is an octagonal tomb for the patron. The building has been rebuilt on several occasions, including after being destroyed by a fire in the eighteenth century. The current plan (Figure 4.3) is covered by a flat roof carried by 24 columns, which are arranged in six rows of four across the width of the space. This arrangement is largely due to the restoration efforts after the fire. Two of the mosque's minarets collapsed in the fire, and only one, on the north, has since been rebuilt.<sup>22</sup>

<sup>17</sup> Bloom and Blair, *Art and Architecture of Islam*, 5.

<sup>18</sup> For discussing such a transformation in Ani, Guidetti, 'Islamicness', 155–157; for Trebizond, Eastmond, *Art and Identity in Thirteenth-Century Byzantium*, and for Divriği, Pancaroğlu, 'The mosque-hospital complex in Divriği', 169–198.

<sup>19</sup> Given the lack of an inscription, previous scholarship has estimated the construction date of the Çifte Minareli to be between the 1220s and 1320s. Blessing, *Rebuilding Anatolia*, 142, dates the construction to sometime between 1290 and 1300. For the Alaeddin Mosque, Şaman-Doğan, 'Niğde'deki Türk Dönemi', 118.

<sup>20</sup> Redford, 'Byzantium and the Islamic world, 1261–1557', esp. 390; Redford, 'A grammar of Rum Seljuk ornament', 288 and 291; Guidetti, 'Islamicness', 177.

<sup>21</sup> Measurements come from Akmaydalı, 'Niğde Sungurbey Camii', 151.

<sup>22</sup> For an analysis on the damage, Akmaydalı, 'Niğde Sungurbey Camii', 151.

*Fig. 4.3: Sungur Ağa Mosque, restitution of the ground plan (author, redrawn based on A. Gabriel,* Niğde Türk Anıtları*, fig. 14).*

There are pointed blind arches on all four sides of the building: three on both the south and north, four on the west and two on the east. These blind arches are pierced with window openings on the upper level of the south and the west sides of the building.23 The west wall has windows on the lower level as well. The wall elevations on the outside correlate structurally to the inside, as evidenced by the placement of the bundles of colonnettes at regular intervals and their alignment with the blind arches on the façade.

The original construction of the roof was a matter of some debate. Gabriel argued that the nave was domed and the side aisles were covered with cross-vaults.<sup>24</sup> The nave is somewhat larger than the side

aisles and ends in a slightly protruding mihrab on the south wall. Alternatively, other scholars argued that the mosque's nave and side aisles might have been supported with ribbed cross-vaults.25 Based on the ground plan, previous scholarship has concluded that the nave and the side aisles were covered with ribbed cross-vaults and that the pre-fire plan and roofing was similar to the basilical plan of the Rum Seljuk examples.<sup>26</sup> This is indeed a common plan and elevation system used for creating sacred spaces in the eastern Mediterranean, and it also recalls the plans of churches built in Lusignan Cyprus, the Christian Levant and theArmenian highlands.<sup>27</sup> Most of the churches built under Lusignan rule have flat roofs, with a few exceptions carried on pitched roofs, while in the Armenian realm the nave was domed and the side aisles were vaulted.<sup>28</sup>

Despite the absence of an extant inscription on the foundation of the building, there are two other inscriptions in Arabic that provide information about the building, the artist employed and the patron. The wooden minbar of the Sungur Ağa Mosque, now relocated to the Ottoman-era Dışarı Mosque (literally 'built outside the city walls'),

<sup>23</sup> The pointed blind arches on the façade can also be noted at Bursa's Grand Mosque, built by Bayezid I in 1396.

<sup>24</sup> Gabriel, *Monuments Turcs d'Anatolie* and Gabriel, *Niğde Türk Anıtları*, 33. Gabriel's roofing reconstruction with a domed nave and vaulted side aisles has mostly been followed by later scholarship.

<sup>25</sup> Esin, 'A study on possible foreign impacts', 28–29 reviews the scholarship on various proposals for the original state of construction.

<sup>26</sup> For alternative roofing reconstructions, see Özkarcı, *Niğde'de Türk Mimarisi*, 52.

<sup>27</sup> For the plans in the Christian Levant from the eleventh to fourteenth centuries, Borowski, 'Placed in the midst of enemies', 89, fig. 2.17. For the Armenian highlands, Guidetti, 'Islamicness', 170, discusses how the presence of the Armenian church plan may be linked to the Islamic tradition.

<sup>28</sup> Borowski, 'Placed in the midst of enemies', 81; Guidetti, 'Islamicness', 169–171.

has inscriptions on both sides.29 The inscriptions on the left side include a reference to a certain Hoca Abu Bakr, the son of a teacher and scribe. The inscription on the right side identifies the patron of the edifice as Saif-al-Dewlet wal-Din Sungur Ağa, supporting the likely period of construction being 1316 to 1335.30

Interestingly, the Rum Seljuk borrowings found in the plan of the mosque cannot be extended to other structural and decorative details. The vaulting system, mouldings on the portals and the fenestration details strongly suggest borrowings from Crusader and Armenian architectural practices. Gabriel contended that the nature of the 'foreign elements' hinted at interactions with the Lusignans, the Crusader dynasty ruling on Cyprus31 from the twelfth to the fifteenth century, or Armenian Cilicia. Gabriel's ground-breaking analysis of the building attracted academic attention from a range of scholars, including Howard Crane. In a brief piece, Crane echoed Gabriel's argument about the likelihood of Lusignan input and contended that scholars must turn to the Ilkhanids' Christian ally in Anatolia, the Rubenid dynasty in Cilician Armenia, to understand the complexity of forms and details present on the edifice.32 Crane's basic chronology here, though, was off: the Armenian dynasty contemporaneous with the Ilkhanids was the Hetumids.

Scholars, including Gabriel and Crane, used an inscription that is located on the base of the minaret as seen on the left hand side of Figure 4.2, to argue for the presence of Christian builders from Cilicia or Cyprus working on the construction.<sup>33</sup> This inscription describes a decree on taxation from AD 1335–1336, passed by Governor Sungur Ağa, exempting Christians from paying taxes.<sup>34</sup> However, I am not sure if this inscription was intended for the mosque as it was added onto the base of the minaret at an unknown time. Instead it may be that the inscription was meant to inform the inhabitants about a decree on Christians' exemption from taxes under Ilkhanid rule during the reign of Abu Saʿid.

<sup>29</sup> The minbar was removed from the Sungur Ağa in the eighteenth century when the mosque was being repaired after a fire. For the inscription, Özkarcı, *Niğde'de Türk Mimarisi*, 70; Crane, 'The Ilkhanid Sungur Bey of Niğde', 67. (Upper left line): The work of the weak servant who needs God's compassion. (Lower left line): And forgiveness, the son of the teacher and scribe, Hoca Abu Bakr.

<sup>30</sup> Özkarcı, *Niğde'de Türk Mimarisi*, 70; Crane, 'The Ilkhanid Sungur Bey of Niğde', 67: (Upper right line): In the time of the great sultan Abu Sa'id, God grant that his domain be enduring. (Lower right line): At the command of the commander of commanders Saif al-Din Sungur Ağa. For a discussion on the construction date for the building, Gabriel, *Niğde Türk Anıtları*, 37.

<sup>31</sup> Gabriel, *Monuments Turcs d'Anatolie*, 123–135; Olympios, 'The shifting mantle of Jerusalem', 120, footnote 84 contends that the similarity is stylistic.

<sup>32</sup> Crane, 'The Ilkhanid Sungur Bey of Niğde'.

<sup>33</sup> Gabriel, *Niğde Türk Anıtları*, 37; Crane, 'The Ilkhanid Sungur Bey of Niğde', 64–70. Post-Gabriel scholarship, as discussed by Özkarcı, *Niğde'de Türk Mimarisi*, 68, discusses the likelihood of the workers coming from Cyprus. For an overview of scholarship, Esin, 'A study on possible foreign impacts', 22.

<sup>34</sup> Özkarcı, *Niğde'de Türk Mimarisi*, 68; Crane, 'The Ilkhanid Sungur Bey of Niğde', 68: (Line 1) In the name of God, praise to God and honor to Muhammad, His prophet, in the months of the year 736. (Line 2): In view of the future life, at the command of the great emir, Saif-al-Dewlet wal-Din, God protect his state. (Line 3): Let it be so: from this day forward at Niğde, well defended, the jizya [poll tax] and kharaj [land tax] will no longer be paid by the strangers of the Christian nation.

*Fig. 4.4: Sungur Ağa Mosque, arched iwan, east portal (author).*

Patricia Blessing has discussed three Ilkhanid inscriptions found on the southern gate of Ankara citadel at the Cacabey Medrese in Kırşehir as well as the now-lost inscription from the western wall of the Minuchir Mosque in Ani, dated to 1320.35 Written in Persian and dated to the reign of Abu Saʿid, the inscriptions refer to the abolition of taxes and warn against abuses.Only the inscription from Ani specifically refers to the Christian inhabitants of the city and other regions of Georgia being exempt from paying a series of unlawful taxes. Blessing has argued that these inscriptions were intended as a visual statement of Ilkhanid power in those cities at a time when the local figures holding power were imposing their own taxes.36 Other contemporaneous manuscript colophons refer to types of taxes as well as the exemptions existing for local Christians and clergyman.37 Thus the taxation decree at the Sungur Ağa Mosque may refer

to the Christians and their taxation rights in the city, not the Christians as builders of the mosque.

## **The Sungur Ağa: Portals, fenestration, moulded band**

The two portals of the Sungur Ağa Mosque, on the east and north, showcase the varied influences of the cultural and political landscape at the time. The east portal (Figure 4.4), or the arched *iwan*, at the Sungur Ağa displays bundled pilasters and high-relief bands decorated with stylised geometric animal and floral patterns. It has two side niches on either wall and has a pointed arch supported by a ribbed cross-vault. A photogrammetric analysis done by Şinasi Kılıç and work by art historian Ömür Bakırer have restituted the now-destroyed 'rose' window arrangement on the tympanum of

<sup>35</sup> Blessing, *Rebuilding Anatolia*, 179–182 for the original Persian text and transliterations of the three inscriptions as well as further bibliography.

<sup>36</sup> Blessing, *Rebuilding Anatolia*, 183.

<sup>37</sup> Dashdondog, *The Mongols and the Armenians*, 215. The colophons of 1321 describe the time of Abu Sa'id as a bitter period, Dashdondog, *The Mongols and the Armenians*, 153.

the east portal, which enables us to reconstruct most of the window details: a large circle at the centre of the tympanum had a rosette formed by an eight-pointed star.<sup>38</sup>

Based on the examination of buildings displaying similar decorative and structural features, scholars often place the architectural style of the Sungur Ağa's east portal within the Rum Seljuk realm, situating it in a chronology starting with the portal of the hospital in the Mengücekid Mosque-Hospital Complex at Divriği (1228) and ending with the main portal of the Menteshid Medrese of Ahmed Gazi in Beçin (1375).39 In these examples, as at the Sungur Ağa, the tympanum has a window opening, a feature often attributed to Gothic architecture.40 Following the approach of more recent scholarship on Divriği and Beçin, it is now possible to say that, despite offering stylistic similarities, each building must be considered within its own cultural context. Oya Pancaroğlu has convincingly argued that in the construction of Divriği the role of the artists from Ahlat (one with a Persian family background) and Tbilisi (a city then under the rule of Muslim-Arab control) must be considered.41 For Beçin, the evidence hints at a connection to the Aegean Islands and the Dalmatian coast via commercial networks.<sup>42</sup> In the case of the cross-ribbed vaulted entrance at the Sungur Ağa, comparisons come from the Gallery of the Knights Hall at the Krak des Chevaliers in Syria, dated to *c*. 1230, and from the Cathedral of St. Nicholas in Famagusta and the Cathedral of St. Sophia in Nicosia, both Lusignan Cypriot examples of the midfourteenth century.<sup>43</sup>

On the north portal, above a bicephalic gyrfalcon moulding on the lintel, there is a window displaying two concentric circles surrounded by a rectangle (Figure 4.5).<sup>44</sup> The inner circle of the window has a trefoil in the centre surrounded by a hexagonal star. The outer circle, enframing the smaller one, is partitioned into eight sections that each contain a trefoil, hexagon or quatrefoil, although a few of these have now been destroyed.

The round or 'rose' window detail seen on the north portal, as well as the previously discussed window on the east portal as seen in Figure 4.2, is quite common in the buildings of mid-fourteenth century Lusignan Cyprus, such as in the refectory at Bellapais Abbey.45 Further Lusignan comparisons are the rose window arrangement at the Cathedral of Saint Nicholas in Nicosia and the Church of Saint Mary the Augustinians in Famagusta in Cyprus.46 The visual analogy noted in the style of carving

<sup>38</sup> Bakırer, 'Niğde Sungurbey Camisi'nin Taçkapı', 79–81, fig. 368: 1–4.

<sup>39</sup> One such example is proposed by Çaycı, 'Korkueli (Istanoz) Tarihi ve Korkuteli Alaaddin Camii', 119. For Divriği: Pancaroğlu, 'The mosque-hospital complex in Divriği'. For Beçin: Akarca and Akarca, *Milâs-Coğrafyası, Tarihi ve Arkeolojisi*, 117–120; Demir *et al*., 'Sultan of the coasts'.

<sup>40</sup> Akmaydalı, 'Niğde Sungurbey Camii', 152; Sözen, *Anadolu Medreseleri*, I, 180.

<sup>41</sup> Pancaroğlu, 'The mosque-hospital complex in Divriği', 184–185.

<sup>42</sup> Kiel, 'Cross-cultural contacts in 14th century Anatolia'.

<sup>43</sup> For the most recent analysis and further bibliography on the medieval architecture in Cyprus and the Crusader mainland, Olympios, 'Reminiscing about the Crusader Levant'.

<sup>44</sup> Bakırer, 'Niğde Sungurbey Camisi'nin taçkapı', 79 and fig. 368: 1–4.

<sup>45</sup> Esin, 'A study on possible foreign impacts', 49–50.

<sup>46</sup> Esin, 'A study on possible foreign impacts', 53, fig. 65 (a) and (b); Enlart, *Gothic Art*, 149, fig. 86.

*Fig. 4.5: Sungur Ağa Mosque, window detail, north portal (author).*

*Fig. 4.6: Sungur Ağa Mosque, fenestration details (author).*

in the window tracery is even more apt when compared to the rose window detail of contemporaneous Anatolian buildings, such as the Alaeddin Mosque in Korkuteli in Isparta, located in Pisidia (five hundred kilometres south-west of Niğde). It was built in the post-Kösedağ realm, dating possibly to the 1370s.<sup>47</sup>

The window situated on the wall that separates the tomb from the mosque displays a round frame bearing a small hexagonal star inserted into the middle of a larger one. There are trefoils carved in the spaces between the rays of the large star and the circular frame as well as surrounding the edges of the circular frame.

The fenestration and decorative moulding on the west façade display fascinating aspects, particularly the pointed recessed arches and tympana over the fenestrations (Figures 4.6–4.7). The tympana over three of the windows display a variety of tracery details, notably

perforated with trefoils, quatrefoils and hexagonal stars. The closest comparisons are the quatrefoils found on the tracery of the Cathedral at Saint Nicholas in Famagusta, the flying buttresses pierced with quatrefoils at the Cathedral of Saint Sophia in Nicosia and the tracery noted in the fourteenth-century churches of the socalled Dominican and Carmelite churches in Famagusta.48 Breaking the pattern, the fourth window on the west facade has a square cap (not pictured); it used to be

decorated with a hexagonal star surrounded by pentagonal perforations.49 Inside the

<sup>47</sup> Çaycı, 'Korkuteli (Istanoz) Tarihi ve Korkuteli Alaaddin Camii', 118–121. For a different interpretation, Erken, *Türkiye'de Vakıf Abideler*, 601, who attributed both the Sungur Ağa and the Mosque of Alaeddin to Mongol influence and/or workmanship.

<sup>48</sup> Borowski, 'Placed in the midst of enemies', 74–78, figures 2.3–2.6. The so-called Dominican Church is also known as the Church of Saints Peter and Paul in the scholarship.

<sup>49</sup> Discussed and recorded by Gabriel, *Niğde Türk Anıtları*, plate 12; this detail is not visible anymore.

*Fig. 4.7: Sungur Ağa Mosque, fenestration details (author).*

*Fig. 4.8: The 'merged angled chevron' arch, St. Epiphanios (Thomas Kaffenberger).*

*Fig. 4.9: The 'merged angled chevron' arch, St. George Exorinos (Thomas Kaffenberger).*

building, a recessed arch with a chevron pattern and mouldings is visible around a window that was formerly a door leading into the tomb.

The profile of the recessed arches and the decorative details on the windows located on the west façade and on the inside are comparable to the south portal of the Famagustan Church of Saint Epiphanios, dated 1310–1330 (Figure 4.8). The 'merged angled chevron' noted at the Church of Saint Epiphanios is especially comparable to the profile drawings done by Gabriel.50 Similarly, the arch on the south façade of the Church of Saint George Exorinos, located in Famagusta and dated 1310–1330 (Figure 4.9), also displays the 'merged angled chevron' and is particularly comparable with the window on the inside that was formerly a door at Sungur Ağa. Both Cypriot examples display a cushion-like hood mould, when compared to those at the Sungur Ağa, where the hood mould directly extends into the masonry. Similarly, the voussoir from the refectory portal of the Bellapais Abbey near Kyrenia, dated 1340–1350 also has a chevron moulding.51

<sup>50</sup> Kaffenberger, 'Evoking a distant past', 160–189. I thank Thomas Kaffenberger for sharing a copy of his work as well as his photographs (Figs. 4.8 and 4.9 used in this piece) with me. For a discussion on the Sungur Ağa, Gabriel, *Niğde Türk Anıtları*, fig. 20.

<sup>51</sup> For a comparanda discussion, Kaffenberger, 'Evoking a distant past', 169–174, figs. 9.1, 9.2 and 9.3.

*Fig. 4.10: Dedication and ritual pages from MS 2027, (1270s) fols. 61v, 86v. Armenian Patriarchate, Library of the Monastery of St. James, Jerusalem.*

In the context of Anatolia, recessed arches with similar elegantly moulded zigzag designs are noted in Rum Seljuk buildings such as the Alaeddin Mosque in Konya, dated to the 1220s, and at the Ahi Yusuf Tomb in Antalya, dated to 1249 and built amid the battle of Kösedağ.52 In the Armenian context, the tenth-century Church of Aght'amar set the earliest example of similar recessed arches, which soon became quite widespread. The recessed arch detail is also found in contemporaneous manuscripts, such as the dedication pages from MS 2027, located in the Library of the Monastery of St. James in the Armenian Patriarchate, in Jerusalem (Figure 4.10).53 This emphasises that the chevron pattern, with its complex history of origins in both the East and the West, was introduced into the Cypriot architecture of the fourteenth century via the architects from Syria who were based in Cyprus after the dissolution of the Crusader states in the late thirteenth century.54

<sup>52</sup> Redford, 'Alaeddin', 54–74; Riefstahl, *Turkish Architecture in Southwestern Anatolia*, 49; Esin, 'A study on possible foreign impacts'; 57; Tanman, '14. ve 15. Yüzyılların Anadolu Türk Mimarlığında Gotik Etkiler', 213–225.

<sup>53</sup> Esin, 'A study on possible foreign impacts', 56–57; Grigoryan, 'Manifestations of Mongol-Armenian relations', figs. 4a and 4b.

<sup>54</sup> For the Eastern origins of the chevron pattern, one must remember the voussoirs with chevron mouldings from Qusayr Amra as well as the Zangid and Ayyubid and Rum Seljuk examples. In the West, it is often attributed to Anglo-Saxon and Carolingian architecture and later to Norman architecture. For a summary of origins, Kaffenberger, 'Evoking a distant past', 163–165, 180; Tanman, '14. ve 15. Yüzyılların Anadolu Türk Mimarlığında Gotik Etkiler', 213–225.

Besides the mouldings on the window arches and portals, another feature of the Sungur Ağa Mosque is the decorated moulded band bearing chevron weave and ellipse-shaped carving that runs through the midpoint of the façade (albeit being plain on the west wall). The moulded band is unique, with no forerunners in Crusader or Islamic architecture.

Finally, the profile of the mosque's bundles is not identical, instead following three different profiles, as discussed by Gabriel.55 The semi-standing bundles of colonnettes and moulded frieze inside the building echo any generic example of Crusader architecture, including the fourteenth-century Lusignan examples from Cyprus or Europe, as well as similar support systems in the Armenian context.56

The masons' marks on the stones of the Sungur Ağa as a tool to identify the ethnic background of the builders have been examined in three studies devoted to the medieval Islamic buildings of Anatolia. The marks noted in the previous century led Mehmet Çayırdağ, Zeki Sönmez and Demet Binan to discuss a genealogical continuity and the role of guild formations from the thirteenth into the fourteenth century.57 Studying those marks, Vryonis has argued the letters can be read as Greek letters and hence show the contribution of Greek builders.58 An exhaustive table created by Didem Esin combines data coming from medieval Anatolian buildings, European Gothic structures and the buildings in Lusignan Cyprus. She aptly demonstrates the appearance of common marks and notes that the masons' marks at Niğde's Alaeddin Mosque and the Mausoleum of Khudavand Khatun are strikingly similar to those of the Sungur Ağa.59

Overall, the variety of Gothicised forms can be linked to the architectural knowledge and input realised by the Crusaders, based on Cyprus, as they passed through Anatolia and engaged in cultural and political interactions with the Rum Seljuks, the Ayyubids and the Mamluks.60 The movement across particular, interrelated geographies that makes Anatolian architectural culture of this period so vital and invigoratingly creative brings to mind Thomas DaCosta Kaufmann's theory of the geography of art emphasises a focus on cross-regional networks in order to define a broader reading of cultural overlaps and move beyond narrowly defined geographical categories.<sup>61</sup> The cross-pollination between different architectural forms and styles blended influences we might classify as Anatolian and beyond-Anatolian, local and

<sup>55</sup> Gabriel, *Niğde Türk Anıtları*, fig. 21.

<sup>56</sup> Gabriel, *Niğde Türk Anıtları*, fig. 20. Bowie, *Sketchbook of Villard de Honnecourt*, plate 44. My Cypriot comparisons are based on the vaulting details from the New Mosque in Nicosia and the Cathedral Church of St. George of the Greeks in Famagusta. Enlart, *Gothic Art*, 151, fig. 91; 254, fig. 198; figs. 54(a), 54(b); Esin, 'A study on possible foreign impacts', 81; Donabedian, 'The architect Trdat', 52–53.

<sup>57</sup> Çayırdağ, 'Kayseri'de Selçuklu ve Beylikler Devri', 84–91; Sönmez, *Başlangıcından 16. Yüzyıla Kadar*, 15–18; Binan, 'Ortaçağ Anadolu Türk Mimarisinde', 128–129.

<sup>58</sup> Vryonis, *Decline of Medieval Hellenism*, 237. 59 Esin, 'A study on possible foreign impacts', 77–80, table 1. For Gabriel's discussion on the similarities, *Niğde Türk Anıtları*, 39 and 43 for the catalogue of the twenty-two masons' marks identified at the Sungur Ağa.

<sup>60</sup> Tanman, '14. ve 15. Yüzyılların Anadolu Türk Mimarlığında Gotik Etkiler', 213–225.

<sup>61</sup> DaCosta Kaufmann, *Towards a Geography of Art*.

*Fig. 4.11: Sungur Ağa Mosque, mouldings, side walls, east portal (author).*

global, as well as Islamic and Christian. The result was a visual language in which none of the details we have discussed can be clearly identified as Gothic or Islamic. This complex formation of the decorative details stands as a testimony to a dynamic transculturation process, as I will explain below.

## **Reading the Sungur Ağa Mosque: 'paintings in stone'**

Having explored the immense variety of their intersections, how can we interpret the similarities in the decorative and structural details discussed so far? In the absence of identified Armenian or Crusader buildings in Cappadocia,62 I would like to explore the idea that different media might offer knowledge about architectural and artistic overlaps.

The delicate carvings on the north portal, the side niches on the east portal and on the mihrab at the Sungur Ağa (Figure 4.11) find their closest comparison

in the decorated capitals from the *zhamatun* (narthex) thirteenth-century Church of Tigran Honents in Ani. Studying these capitals, Thierry and Thierry suggest a possible relationship between the forms found in the dedication pages and canon tables of Armenian manuscripts and carved mouldings decorating window and door details coming from the Islamic context, such as Dagestan.63 The mouldings at the Monastery of Surp Karapet in Baghesh located in present-day Muş and the Ilkhanid Gök Madrasa in Sivas are comparable as well.<sup>64</sup> The swirling floral patterns noted on the Canon Tables of MS 251 (Figure 4.12) are strikingly similar to the mouldings at Sungur Ağa.65 Similarly, the animal heads sticking out of floral patterns found on the carvings or the 'sculptural marginalia' can also be compared to the Church of Saint Nicholas in Famagusta.<sup>66</sup>

<sup>62</sup> Poorly preserved Cilician Armenian settlements such as Hromkla, Zeytun and Levonkla may also offer comparisons, and those sites need thorough archaeological surveys and studies.

<sup>63</sup> Guidetti, 'Islamicness', 160, ft. 23; Thierry and Thierry, *L'église Saint-Grégoire*, 19.

<sup>64</sup> Kuehn, *The Dragon in Medieval East Christian Art*, 60–61, plate 44, fig. 191; Blessing, *Rebuilding Anatolia*, 107–108.

<sup>65</sup> Der Nersessian, *Aght'amar Church of Holy Cross*, 11; Rapti, 'Art from another Byzantium'.

<sup>66</sup> Borowski, 'Placed in the midst of enemies', 76 and fig. 2.2.

Carvings around the doors and moulded bands are most closely related to the thirteenth-century Surb Astvatsatsin Church in Noravank, near Yerevan, and find further visual resonances in the dedication pages of the Cilician *Zeyt'un Gospels, c*. 1256.

How would the moulding details have been developed and incorporated into the Sungur Ağa? In an inspirational piece, Linda Komaroff, a scholar of Ilkhanid Persia, refers to courtly scenes appearing in manuscripts that evidently surfaced in an Ilkhanid candlestick. She discusses these as 'paintings in gold and silver'.<sup>67</sup> Just as an illuminator working on a manuscript transfers what is remembered or perceived from a pattern book onto a new surface, could a mason carve a moulding in the same way?68 I would like to argue the mouldings and fenestration details at the Sungur Ağa can be seen as 'paintings in stone'. For the overlapping patterns of animals and floral motifs seen between the ceramic tiles and textiles, Komaroff suggests verbal and visual (plans, sketches and drawings) ways of transmitting the knowledge.69 The now-lost muqarnas drawing once recorded on the south façade of the narthex in the Church of Astvatsankal offers insight into how decorative details were shared in the Armeno-Muslim realms.70 Although the art of construction is different from that of drawing and illuminating, what we see in the carved form closely follows the imagery in the manuscripts and book covers.71 In that regard, a possible avenue of transmission would be via the patterns noted on the Armenian manuscripts or the Ilkhanid book covers produced in Anatolia that employ similar floral and faunal assemblages.<sup>72</sup>

### **An elite chivalric ethos: Laying claim through transculturation**

Even Marco Polo made note of the vibrant flora and fauna and idiosyncratic geometry that characterise both the Sungur Ağa Mosque and Armenian and Rum Seljuk forms. The specific reference occurs in 1248–1250, when he reports that Smbat the Constable, a noble in the Cilician court, received a tablet of authority from the Ilkhanid ruler containing 'the image of a gyrfalcon or a lion or of different animals'.73 On folio 341 v. of MS 197, produced in Cilicia in 1289, we note Archbishop John, the brother of Het'um I, wearing a tunic embroidered with a Chinese dragon pattern, likely a gift

<sup>67</sup> Komaroff, 'Paintings in silver and gold'.

<sup>68</sup> Carruthers, *Craft of Thought*, 42; Marshall and Fryer 'Speak, memory!', 3–4.

<sup>69</sup> In Islamic architectural practice plans and actual drawings began to be used from the eighth century, reaching widespread use from the fifteenth century onwards. Holod, 'Text, plan and building', 1–12; Bloom, 'On the transmission of designs'; Komaroff, *Legacy of Genghis Khan*, 183–184, 186; Necipoğlu, *Topkapı Scroll*, 3–4. 70 Ghazarian and Ousterhout, 'A Muqarnas drawing from thirteenth-century Armenia', 141–154.

<sup>71</sup> The following are three different medieval contexts attempting similar suggestions. Mengücekids: Pancaroğlu, 'The mosque-hospital complex in Divriği', 169; Hindu-Muslims: Flood, *Objects of Translation*, 252; Armeno-Crusader: Kouymjian, 'The Holy Mother of God Armenian church in Famagusta', 146.

<sup>72</sup> Demircan Aksoy, 'İlhanlı-Memlûk Etkileşiminde', 267, especially figs. 14, 22, 24.

<sup>73</sup> Grigoryan, 'Manifestations of Mongol-Armenian relations', 287; Moule and Pelliot, *Marco Polo*, I: 226.

*Fig. 4.12: Floral and faunal moulding details, canon tables (1260) MS 251, fols. 2r and 3v, Armenian Patriarchate Library of the Monastery of St. James.*

given to him during his visits to the Ilkhanid court.<sup>74</sup> Examples such as these hint at how the exchange of gifts, the prevalence of courtly etiquette and acknowledgement by the various actors of the Cappadocian-Cilician frontier resulted in the creation of a shared aesthetic vocabulary.

To demystify the creation of this vocabulary, Scott Redford has suggested the prospect of an 'elite chivalric ethos'.75 The blend of Crusader, Armenian and Rum Seljuk details from traveling and local masons and artists at the Sungur Ağa illustrates the elements of this shared ethos largely emerging from the process of transculturation. The concept of transculturation, as introduced earlier is useful in our attempt to interpret the ways in which Armenian and Crusader architectural and artistic techniques, forms and details found expression at the Sungur Ağa Mosque. It is important to remember that Ortiz's notion of transculturation in the case of this enigmatic building. The workings of transculturation are particularly suited to the Anatolian material in that it accounts for subtle, implying selective losses and gains during the process of cultural transformation rather than the complete acquisition of a new culture (acculturation) alongside outright loss of the previous cultural heritage (deculturation).<sup>76</sup>

In the case of the Sungur Ağa, local forms blended with Gothic details. To better understand this, we can turn to a contemporaneous tomb structure commissioned by the Rum Seljuk princess Khudavand Khatun (daughter of Qilij Arslan IV, r. 1248–

<sup>74</sup> Kouymjian, 'Chinese motifs in thirteenth-century Armenian art', 303–324, plates 23–25. For the appearance of the imagery associated with the Ilkhanids, Grigoryan, 'Manifestations of Mongol-

Armenian relations', 284–285; Lane-Poole, *Coins of the Mongols in the British Museum*, xliv. 75 Redford, 'Byzantium and the Islamic world, 1261–1557', 390; Redford, 'A grammar of Rum Seljuk ornament', 288 and 291.

<sup>76</sup> Latour, *We Have Never Been Modern*, 3, 11, 121. Flood, *Objects of Translation* studies transculturation in the Hindo-Islamic context.

*Fig. 4.13: Tomb of Khudavand Khatun, Niğde (author).*

*Fig. 4.14: Sungur Ağa Mosque, North portal, door lintel, bicephalic gyrfalcon (author).*

1265), who was married to Arghun (r. 1284–1291), the fourth Mongol-Ilkhanid ruler. Her mausoleum (Figure 4.13) has an octagonal foundation topped by a sixteensided conical roof. Its plan is identical to that of the tomb Sungur Ağa appended to the east wall of his mosque. However, whereas Sungur Ağa's tomb as seen on the left hand side of Figure 4.2, only offers stalactite mouldings, Khudavand's tomb showcases a variety of rich floral and zoomorphic depictions on the tympana

and eave line, acting as figural representations of her Seljuk pedigree and her earned Ilkhanid patrimony via her marriage.<sup>77</sup>

While Sungur Ağa's tomb was left simple, the Sungur Ağa Mosque included a rich assemblage. In addition to the features previously discussed, a two-headed gyrfalcon carved in bas-relief and accompanied by palmette motifs perches on the lintel of the north portal (Figure 4.14). Bicephalic birds, commonly in the guise of eagles, were widely appropriated by Christian and Muslim monarchs in medieval Anatolia, and in this case the representation is likely a personal badge of Sungur Ağa, whose name means gyrfalcon. The placement in the midpoint of the door lintel – just as the Lusignan examples displayed it on the keystone of the arch leading into the main entrance on the west façade78 – is a further indication that this imagery stands for the patron himself.

But while the image of the bicephalic eagle may symbolise Sungur Ağa, I contend that it goes far beyond the personal connotation. Similar imagery can

<sup>77</sup> For an analysis of the figural representations, the possible impact of Khudavand spending time in the Mongol-Ilkhanid-controlled Iran with her husband and pre-Islamic Shaman resonances, Parla, 'Melike Hüdavend Hatun Türbesi', 153–162.

<sup>78</sup> Examples include the carved Jerusalem crosses from numerous churches in Cyprus, as discussed by Borowski, 'Placed in the midst of enemies', 86, fig. 2.13.

*Fig. 4.15: Tomb of Khudavand Khatun, Niğde, the crowned female head (author).*

be found on the Mausoleum of Khudavand Khatun. In one of the tympana in the north-western arch at Khudavand's mausoleum, there is a crowned female figure carved in half-profile (Figure 4.15). Canan Parla suggests this may be a representation of the Rum Seljuk princess.<sup>79</sup> Next to this image of Khudavand, there is a bicephalic eagle flanked by two symmetrically placed dragonheads with their tongues sticking out. If we follow Canan Parla's identification of the

other figural representations on the tympana as her father, the Rum Seljuk Sultan Qilij Arslan IV, and mother, Fatma Hatun, one may also argue that the 'amplified version' of a double-headed eagle-cum-dragon acted to emphasise her rulership and presence in the city.80

Both Sungur Ağa's and Khudavand's attempts to make themselves visible in Niğde can be tested against the literary evidence in this period. For example, in a Persian account known as the *Al-Walad al-Shafīq*, written in 1333 by a learned local figure called Ahmed of Niğde, there is surprisingly no reference to Sungur Ağa. Ibn Battuta also does not talk about Sungur Ağa at all, going against his usual practice of describing such rulers whilst visiting cities in Anatolia.<sup>81</sup> While Sungur Ağa gets no recognition, Ahmed of Niğde praises the late Khudavand as 'a great ruler, pompous queen, a notable, reverend female figure, a merciful, sinless benefactress, from the crown of the Rum Seljuks, the daughter of Qilij Arslan IV'.<sup>82</sup> Khudavand is also described in the inscription from a now-destroyed Beylerbeyi tomb (d. 1325), which refers to her as the 'Malika-i Muʿazzama', ruling in Niğde and wishing her state to be eternal.83 After her husband's death in 1291, she was granted Tokat and Niğde, in accordance with the Ilkhanid practice of dividing the land among the wives and sons.

Khudavand resided in Niğde and possibly ruled from there from 1292–1325, perhaps until her death.<sup>84</sup> Khudavand's claim to power coincided with the time

<sup>79</sup> Parla, 'Melike Hüdavend Hatun Türbesi', 160–161; Parla, 'İkonografik Yaklaşım', 1024–1027; Oral, *Niğde* 

*Tarihi Tetkiklerinden*, 29. 80 As discussed in Çağaptay, 'On the wings of the double-headed eagle', 322–323; Baer, *Sphinxes and Harpies in Medieval Islamic Art*; Kuehn, *The Dragon in Medieval East Christian Art*, 123.

<sup>81</sup> Peacock, 'Aḥmad of Niğde's', 106. In Bursa, for example, he talks about Orhan and in Birgi, he talks about Mehmed Bey.

<sup>82</sup> Ertuğrul, *Niğdeli Kadı Aḥmed'in*, 194.

<sup>83</sup> Oral, *Niğde Tarihi Tetkiklerinden*, 29; Parla, 'Melike Hüdavend Hatun Türbesi', 150–151.

<sup>84</sup> The inscription in the portal and her coffin, as well as Ahmed of Niğde's words, support the claim

when Niğde was under Ilkhanid control, from 1308 onwards, as well as with the construction of Sungur Ağa's mosque and the declaration of his independence upon the death of Abu Said in 1335. This shows how Niğde was the locus of an extraordinary political stratification. It has been argued that Ahmed of Niğde's grandfathers and father held important positions in the Rum Seljuk court, and in the *Al-Walad al-Shafīq* Ahmed is mostly concerned with glorifying the Rum Seljuk rulers and their deeds in Anatolia while ignoring Ilkhanid rule in the city.85 One can speculate that this selective referencing may be a sign of his taking sides amid the rivalling political powers. In this competitive environment, leaders would want to demonstrate and cement their power. When compared to the gyrfalcon image at the Sungur Ağa, the decorative assemblage at Khudavand's tomb evokes the Anatolian (read Armenian and Rum Seljuk) and Ilkhanid elite chivalric codes. Given the funerary functions of both buildings, I would like to argue that the depictions of the bicephalic gyrfalcon at the Sungur Ağa and the bicephalic eagle-cum-dragon at the Khudavand Khatun bear dual identities: politically these aquiline images represent both figures wielding full power over their city, but they also stand for the bird accompanying the soul of the deceased. Viewing death as an act of reunion with God, this grants the eagle a sacred role in representing the passage from this life to the next. Here, then, the images signal the visibility of Sungur Ağa and Khudavand Khatun in death.

In using the bicephalic bird of prey, a long-time symbol of power in Anatolia, the newly preeminent Ilkhanid governor Sungur Ağa sought to preserve his political hegemony by appropriating Crusader and local (Seljuk and Armenian) architectural forms and details for his mosque. When considered in relation to the matronage of Khudawand Khatun, 86 we begin to see how each laid claim to their rulership via architectural commissions. Both buildings are located outside the Rum Seljuk walls of Niğde and each demonstrates their cultural pedigree and tries to distinguish themselves from the other.<sup>87</sup> Both Sungur Ağa's and Khudavand Khatun's political presence is felt through their architectural commissions, which helped them present themselves to the people of Niğde.

### **Conclusion**

I have attempted in this chapter to read between and beyond the buildings and artistic forms in fourteenth-century Anatolia to offer deeper insight into the routes and networks of cultural borrowings and the process of transculturation. Using the

that she had resided and ruled there. The inscription from the Hilaf Gazi Zawiya in Tokat, 1291–1292, mentions her return to Tokat following her husband's death and the foundations she commissioned, Parla, 'Melike Hüdavend Hatun Türbesi', 150–151; Parla, 'İkonografik Yaklaşım', 1011–1012; Oral, *Niğde Tarihi Tetkiklerinden*, 29–31 uses waqf charters for this claim. 85 Peacock, 'Aḥmad of Niğde's', 106.

<sup>86</sup> The term matronage is used by Banks Findly, 'Women's wealth and styles of giving'.

<sup>87</sup> For a similar comparison, note the figures in English and French-inspired clothing, Borowski, 'Placed in the midst of enemies', 78–79, fig. 2.8.

example of the Sungur Ağa Mosque, I have focused on the ribbed cross-vault on the east portal; the fenestration, recessed arch and moulding details on the east, north and west façades; and the clustered piers inside the building in order to trace this dynamic process. Governor Sungur Ağa's mosque was a response to contemporary political rivalry that subtly and selectively employed material forms from past and present to assert his future legacy. The complex architectural vocabulary he used to adorn and structure the building proclaimed his dual commitments as patron of the building and ruler of the city. The Sungur Ağa Mosque, I believe marks a striking case in which Christian building techniques were translated and preserved in an Islamic context, showing how the elite chivalric ethos touched leaders of all affiliations and shaped a medieval region.

## **Acknowledgements**

Earlier versions of this paper were presented at the 2018 Leeds Congress of Medieval Studies and Wellesley College, MA in 2019. I am grateful to the audiences for their comments and questions, which helped me to crystallise my thoughts on the Sungur Ağa Mosque. Funding for the final research for this paper was provided by the European Research Council Grant/ ERC Advanced Grant Agreement no. 693418, 'The Impact of the Ancient City', in the Faculty of Classics at the University of Cambridge. My participation in 'The Art of the Crusades: A Re-Evaluation' research project, endowed by the Getty Foundation and SOAS and led by Scott Redford, kindled my interest in the Sungur Ağa Mosque, and the travels we did in the eastern Mediterranean helped me contextualise this enigmatic building. Unless, stated otherwise, all the images are by the author.

## **Bibliography**


Bloom, J. and S. Blair, *The Art and Architecture of Islam, 1250–1800* (New Haven, 1994).


Carruthers, M., *Craft of Thought: Meditation, Rhetoric and Making of Images, 400–1200* (Cambridge, 1998).


Erken, S., *Türkiye'de Vakıf Abideler ve Eski Eserler I* (Ankara, 1972).


Gabriel, A., *Niğde Türk Anıtları.*, tr. A. A. Tütenk (Ankara, 1962).

Ghazarian, A and R. Ousterhout, 'A Muqarnas drawing from thirteenth-century Armenia and the use of architectural drawings in the middle ages', *Muqarnas* 18 (2001): 141–154.


Olympios, M., 'Reminiscing about the Crusader Levant: Royal architecture and memory in Lusignan Cyprus' in I. Baumgärtner, M. Vagnoni, and M. Welton (eds.), *Representations of Power at the Mediterranean Borders of Europe (12th – 14th centuries)* (Florence, 2014), pp. 139–60.


Peacock, A. C. S., 'Black Sea trade and the Islamic world down to the Mongol period' in G. Erkut and S. Mitchell (eds.), *The Black Sea: Past, Present and Future. Proceedings of the International, Interdisciplinary Conference, Istanbul, 14–16 October 2004* (London, 2007), pp. 65–72.

Rapti, I., 'Art from another Byzantium: The sculptural decoration of the Church of Bghno Noravank' in I. Jevtić and K. Durak (eds.), *Identity and the Other in Byzantium* (Istanbul, 2019), pp. 245–268.

Redford, S., 'The Alaeddin Mosque in Konya reconsidered', *Artibus Asiae* 51.1–2 (1991): 57–74.

Redford, S., 'Byzantium and the Islamic world, 1261–1557' in H. C. Evans (ed.), *Byzantium: Faith and Power (1261–1557)* (New York, 2004), pp. 389–396.

Redford, S., 'A grammar of Rum Seljuk ornament', *Mesogeios* 25–26 (2005): 283–310.


Şaman-Doğan, N., 'Niğde'deki Türk Dönemi (13–15. Yüzyıl) Yapılarında Taç Kapı-Mihrap Tasarımı ve Bezeme İlişkisi', *Edebiyat Fakültesi Dergisi* 30.1 (2013): 115–139.

Tanman, M. B., '14. ve 15. Yüzyılların Anadolu Türk Mimarlığında Gotik Etkiler' in A. Ağır, D. Mazlum and G. Cephanecigil (eds.), *Afife Batur'a Armağan: Mimarlık ve Sanat Tarihi Yazıları* (Istanbul, 2005), pp. 214–225.

Thierry, N. and M. Thierry, *L'Église Saint-Grégoire de Tigran Honenc à Ani (1215)* (Paris, 1993).

Turan, Ş., *Türkiye-İtalya ilişkileri I: Selçuklular'dan Bizans'ın Sona Erişine* (Istanbul, 1990).

Vryonis, S., *The Decline of Medieval Hellenism in Asia Minor and the Process of Islamization from the Eleventh through the Fifteenth Century* (Berkeley and London, 1971).

Wittek, P., 'Nīgde' in E. van Donzel, B. Lewis and C. Pellat (eds.), *Encyclopaedia of Islam 2*, vol. 8 (Leiden, 1993), pp. 15–16.

## Chapter 5

## The water of life, the vanity of mortal existence and a penalty of 2,500 denarii: Thoughts on the reuse of classical and Byzantine remains in Seljuk cities

## *Scott Redford*

*These and similar issues cannot find accurate answers, as there is not enough information for the medieval Anatolian elite or townsman, his way of life, his habits or preferences. However, one is inclined to think that as these people were interested with all the happenings around them, be it nature, flora, fauna or buildings and building stones, so it is possible that they may have been interested with the stones that were laying around them merely with an artistic eye and placed them on their buildings to own and keep them in their possession.*<sup>1</sup>

What did medieval Muslim conquerors, rulers and inhabitants of formerly Byzantine cities in Anatolia find in these cities, and what did they do with it? This chapter takes as its subject medieval Anatolian cities during the Rum Seljuk sultanate, mainly in the period of its greatest prosperity, the first half of the thirteenth century. After a general overview, it will look at reuse of Roman and Byzantine marble reliefs in two Seljuk cities, principally the Seljuk capital of Konya (Iconium), but also the town of Akşehir (Philomelium), both in central Anatolia. Concepts like enframing and moralising will be proposed as acting within a more circular notion of time, one that contrasts with linear ones often found in academic studies of the past. The word recycle is a neologism, while the concept behind it is an old one. This chapter tries to contextualise by using an instrumentalist approach.

### **Introduction**

Any understanding of the encounter of the Seljuks and other Turco-Islamic dynasties ruling in medieval Anatolia with the classical urban past should be filtered through the centuries of Byzantine reuse and rebuilding using elements of Hellenistic and Roman stone buildings in the cities to which they fell heir. Byzantine refashioning and

<sup>1</sup> Bakırer, '"Continuity" by "Reuse"', 77.

reuse of remains of the Hellenistic and Roman past, to the extent that we can fathom reasons for them, explain many aspects of later transformations. More importantly, they permit us to propose, if not a continuous, then a *dis*continuous recycling and repurposing of buildings and architectural members, problematising a linear narrative of change, one often based solely on conquest in the name of religion, and revealing the topic to be one based on a series of choices made by individuals, be they patrons, builders or others, defined by the physical and cultural parameters of the time and place – the instrumentalism referred to above.

Anatolia shared with many regions of the Mediterranean traditions of monumental building in stone. Like many other parts of the Mediterranean it is rich in marble. Thus, despite another characteristic shared with other Mediterranean regions in the medieval era, namely the practice of using lime kilns to turn marble into the stuff of mortar and flooring, there was extensive spoliation and reuse of relict Hellenistic and Roman era marble reliefs and architectural elements, as well as those in other stones. Principally, but not exclusively, the stone blocks of classical cities constituted much of city and citadel walls in the late antique and so-called Dark Ages of the eighth and ninth centuries.<sup>2</sup> As the largest structures in medieval cities, fortifications loomed large over Byzantine cities and towns. Byzantine patrons and builders used the prominence of city and citadel walls and gates for adornment with arrays of reused architectural and funerary reliefs in marble. These arrays could and did signify many things, and often had apotropaic qualities to them, a quality that also inhered to the stone itself.<sup>3</sup>

There were many other ways of reusing elements of the past. Complete sarcophagi or sarcophagus panels were repurposed as feeding and watering troughs for animals, fountains for humans and chancel screens in churches, where putti became angels. Monolithic columns were used whole to bind the cut stone facades of fortification walls to their rubble cores, sectioned horizontally to provide decorative veneers for the interiors of churches and palaces, sectioned vertically to provide roof rollers allowing flat-roofed mud brick houses to be rolled smooth after rainfall, and hollowed to form large mortars for pounding grain. Columns and other architectural members were also reused as grave markers. Bath-gymnasium complexes were used as workshops for the production of such products as wine, ceramics, metal and glass, and also turned into churches, as were temples. Some theatres were dismantled, their seats used in the construction of fortification walls, while others were used as sites of industrial production, others as cemeteries, or as forts.<sup>4</sup>

<sup>2</sup> Ivison, 'Urban renewal', 1–46; Brubaker and Haldon, *Iconoclast Era*, 534–559 for an overview.

<sup>3</sup> Word limits constrain further discussion of this topic. Bevilacqua, 'Family inheritance', 203–209 provides a way in.

<sup>4</sup> The recycling of building material and the reuse of structures are also discussed in other contributions in this volume, see especially the chapters by Louise Blanke, Ine Jacobs and Bethany Walker.

By the time of the Muslim conquests of Anatolia in the late eleventh century, Byzantine cities were replete with reused Greco-Roman building elements employed in a variety of ways: practical, apotropaic, talismanic and otherwise. Anatolian cities like Antalya (Attaleia), İznik (Nicaea) and Sinop (Sinope) also preserved significant portions of earlier grid-plans, and in Antalya and Alanya (Kalonoros) spatial order imposed on settlement by Hellenistic era terracing.

Traditionally, histories of urbanism and architecture so focus on monumental buildings, and the monumental urban cores that housed them, that it is easy to forget that most of the buildings of classical and post-classical cities alike were not built of stone or (and, in Anatolia, to a much lesser extent) of baked brick at all; rather domestic and other architecture used the more perishable materials of wood and mud brick. These could and did reuse stones for their foundation courses, or were actually built atop ruined or half-ruined stone walls from earlier eras. In post-classical cities especially, this contrast between durable and ephemeral must have been striking, especially with depopulation leading not only to the use of more ephemeral building materials in former urban cores, but also in the introduction of vegetable gardens and orchards into the heart of intramural settlements. A logic both of impermanence and maintenance inheres to cities increasingly built of materials less enduring than stone, and must have informed the stories told of ancient stone buildings by medieval city and town dwellers living in proximity to the remains of large stone structures, as well as a (re)cyclical view of rise and fall.

Two photographs of the former Seljuk capital of Konya give an idea of what the contrast between building materials must have looked like in earlier eras. Well known British archaeologist and political agent Gertrude Bell took both in May of 1905. Figure 5.1 is a view of Konya citadel. It housed the main congregational mosque and dynastic tomb tower of the Seljuks, their palace and the middle Byzantine church of Ayios Amphilochios, and was surrounded by a fortification wall. At the time the photograph was taken, all of these structures stood in varying states of repair. In Bell's photograph, mud brick garden walls and houses flank the track leading to the citadel. An even more direct conjunction of monumental stone and baked brick architecture with mud brick can be remarked in Figure 5.2, Bell's photograph of the portal of the Seljuk mosque of Sahib Ata (1258). In this photograph, mud brick walls replace the original walls of the mosque, embracing the monumental stone and baked brick portal. The cycle did not stop spinning in later eras.

#### **Finding religious motivation in architectural production**

It is a mistake to think that the Seljuks rejected, in a spirit of fanaticism, the materials that formed pagan monuments, the walls of Konya are full of fragments of architecture and sculpture situated with care, even figures of men and lions are placed ornamentally on several towers of the city.<sup>5</sup>

<sup>5</sup> Texier, *Asie Mineure*, 413.

*Fig. 5.1: Konya Citadel (used with permission from Gertrude Bell Archive, Newcastle University).*

One of the clichés of the medieval era is as an era of close personal and dynastic identification with religiosity. How can this not be the case in the medieval eastern Mediterranean, you might ask, when we have the ultimate proof of this assertion in the two hundred year long movement we call the Crusades? Germane to the subject of this essay, it was the collapse of Byzantine rule in the eastern half of the Empire following the battle of Manzikert in 1071 against Seljuk armies that spurred Byzantine calls for help from the Pope, ultimately leading to the First Crusade at the end of the eleventh century. And yet, in addition to religious zeal, there were many other motivations for northern and western European Christians to travel to the eastern Mediterranean and, among other things, do battle with Muslims. That travel, and the battles that went with it, involved armies of the First, Second and Third Crusades with Anatolia, and of course, most notoriously, the Fourth Crusade with Constantinople. Crusader armies attacked Konya during the First and Third Crusades.<sup>6</sup>

Likewise, it is hard *not* to find religious motivation in the struggles that accompanied the establishment of Turco-Islamic states in the former eastern half of the Byzantine Empire. Certainly religious zeal was one motivation for competition between Byzantines, Franks, Seljuks and others for power and influence in medieval Anatolia. That said, here I would like to offer material cultural evidence that may be at variance

<sup>6</sup> Brisch, *Galatien und Lykaonien*, 177.

with chronicles of the time, opening up, if not irenic vistas of *convivencia*, then the sometimes surprising ways in which the remains of the near and distant past were put to use for poetic, political and other ends, ones sometimes congruent with religious sentiment, sometimes not.

As is well known, the Byzantines were inheritors of the Roman Empire, but did not call themselves Byzantines, but rather Romans. So did their neighbours: Muslims and others, who called them and their land Rum. The Seljuk sultanate ruled in Rum, the first Islamic state to do so. The combination of rivalry and awe felt by Muslims towards the 'Romans' was not limited to the medieval period, but rather had deep roots reaching into the earliest years of Islamic civilisation.

The Byzantine state was, of course, closely identified with Christianity. With the collapse of the eastern half *Fig. 5.2: The Sahib Ata Mosque, Konya, portal (used with permission from Gertrude Bell Archive, Newcastle University).*

of their empire in the late eleventh century, Byzantine chroniclers often mention the conversion of churches into mosques by Muslim forces.7 Undoubtedly such conversions took place, to be replaced in later decades and centuries by purpose-built mosques. But did they take place to the extent and in the way chronicles describe them, and do we have to view conflict in medieval Anatolia principally in this light?

The material record contains hints about different ways in which Muslim rulers recast the heritage of Rum. Here I would like to mention two examples. Unique to the Islamic states of medieval Anatolia, a new kind of mosque, the three-aisled basilica plan mosque appears. While completely different from middle Byzantine ecclesiastical architecture, the basilica mosque reminds of the continued building of churches with this plan, especially in the countryside. The phenomenon of the basilica mosque can be viewed as providing religious architecture intermediate to different Islamic and Christian traditions of the time.<sup>8</sup>

<sup>7</sup> Vryonis, *Decline of Medieval Hellenism*, 198, 'When the Turks took the towns, they usually took over many of the churches for their own cult or for other purposes.' The author gives many examples of this throughout the text.

<sup>8</sup> Necipoğlu, 'Anatolia and the Ottoman legacy', 143–144.

The second example of the Muslim use of religious architecture is more concrete, and limited, and engages more directly with a Byzantine architectural heritage in lands under Seljuk control. At the time, two Byzantine churches stood in citadels in Anatolia. Next to these citadel churches, Seljuk sultans constructed palaces without converting the churches to mosques or, at least in one of them, covering its frescos. One, mentioned above, was in the citadel of the Seljuk capital of Konya. The second still stands at the entrance to Seljuk citadel palace in the port town of Alanya. In order to explain this seeming challenge to preconceptions of the primacy of religious motivation of an Islamic dynasty towards Christian buildings, scholars have not been able to avoid formulating answers not rooted in religion. The conjunction of Byzantine church and Seljuk palace is presented as a sign of religious tolerance, or, alternatively, given a functional explanation: these churches were so preserved for the use of the Christian women in the sultan's harem. In opposition to these explanations, I have argued that to the Seljuks, these churches represented the Byzantine state and its authority as much as its religion, thereby challenging the paradigm of medieval religiosity as a prime motivator. In the land of Rum, the placing of visual markers of this displaced, venerable power, quite literally in an Islamic architectural framework, accrued prestige to the enframers.<sup>9</sup> Moreover, it is perhaps not a coincidence that these palaces were either built or in use at a time when the Seljuks promoted themselves as contenders for the ultimate prize, the capital of Constantinople, recently conquered by Frankish and Venetian forces of the Fourth Crusade (1204), and therefore as rightful rulers of Rum.

The enframing of Byzantine churches within Seljuk palace complexes parallels the Seljuk rebuilding of the walls of Konya in the early 1220s, with the north gate to the city especially, on the road leading to Constantinople, furnished with plentiful spolia, perhaps inspired by the prominent place of spoliated marbles on the Golden Gate in Constantinople. The western neighbours, allies and rivals of the Seljuks, the Laskarids of Nicaea, also employed spolia prominently when they rebuilt the walls of the city in the thirteenth century, with its most prominent employment, as might be expected, on the northern gate of the city, the one that led towards Constantinople, but also the eastern Lefke Gate, the one that led to Konya. Someone traveling between the Laskarid capital and the Seljuk capital, would have been met with similar displays of antique sculpture at both ends of the journey.<sup>10</sup>

In this way, then, the Seljuks followed Byzantine practice, a practice that also may be seen as adhering to the traditions of the land of Rum, while also building Islamic buildings and making Islamic inscriptions for them, and for the gates and fortifications of their cities. In the nineteenth century, many western Europeans travelled to the lands of the Ottoman Empire in search of remnants of the classical past, to document and study them, and sometimes to carry off objects and even buildings to display in imperial museums. Those Europeans traveling to central Anatolia also discovered the

<sup>9</sup> Redford, '*Mamālik* and *Mamālīk*', 329.

<sup>10</sup> Bevilacqua, 'Displaying the past', 145–150.

Seljuks, whose mosques, caravanserais and madrasas they admired. Another source of admiration for European travellers to Konya, one that overlapped with their search for the classical past, was the walls of the former Seljuk capital.

Visitors to Konya seeking marbles and inscriptions found an exploded classical city. Its situation then could be compared to an extinct volcano with an island in its crater: Konya's citadel mound. The high rim of the cratered city was the fortification wall, peppered with pieces of inscriptions, statuary and architectural and funerary reliefs. Because these walls also bore Islamic inscriptions, nineteenth-century western European and American visitors universally attributed them to the Seljuks. In the quote that heads this section, one of these, Charles Texier, situates a presentation of the walls of Konya in the context not only of religious strife, but also the iconoclasm with which Islam was and still is associated. Texier hints that the Seljuks, may not have been barbarous Muslims after all: witness their appreciation of figural art.

These nineteenth and early twentieth century travellers to Anatolia would not stop at a town or village if there were no classical ruins and inscriptions to record. Indeed, not finding ruined temples or other monumental buildings of classical date, many such travellers did not spend long in Konya. But when they did, the city and citadel walls, with their rich displays of inscriptions and architectural and funerary sculpture and reliefs, caught their attention. French government agent Guillaume Olivier, traveling in the Ottoman and Persian lands early in the nineteenth century, was one of these.<sup>11</sup> In a work published in 1838, another French traveller, Leon de Laborde, also attributed the construction of the walls of Konya citadel and city to the Seljuks and compared the display of pieces of the classical past there to a museum.

These vigorous founders of an empire which a thousand revolutions have not destroyed, and which the cupidity of their neighbours has barely diminished, found a small city. They replaced it with a redoubtable fortress raised in the middle of the city, itself surrounded by a formidable line of walls from which project many towers and a bastion that controls them. It is on these walls and towers, some of them round, others square, which flank them, that they inserted as if in the walls of a museum

<sup>11</sup> Olivier, *Voyage*, 390: 'Parmi les inscriptions, les unes sont en beaux caractères, les autres sont peu lisibles, et ressemblent a celles qu'on voit dans les monuments du Bas-Empire.' Ramsay, 'Iconium', 271: 'The walls, about 2 m[iles] in circumference, consisted of a core of rubble and concrete, coated with ancient stones, inscriptions, sculptures and architectural marbles, forming a striking sight, which no traveller ever examined in detail.' Texier, *Asie Mineure*, 662, counted 108 square towers in the city walls; von Vincke *et al*., *Plan Atlas*, 'Plan der umgegend von Koniah' represent 60 on their map. According to Sterrett, *Epigraphical Journey*, 225, after Ibrahim Pasha's sack of the city in 1833, '… these walls were used as quarries for the modern city of Konia.' Lightfoot, 'Amorium', 335, provides the best comparison for the walls of Konya. He states that the central-western Anatolian city of Amorium had a city wall built of large spoliated blocks and square towers, likely in the seventh century, giving us a sense of what the original walls must have been like. An unusual aspect of the walls of Konya, and one that must have contributed to their upkeep over the centuries, was the role they played in deflecting the annual flooding that came from streams in the hills to the west of town and, skirting the northern circuit of the walls, formed a seasonal lake to the east of the city. Von Vincke *et al*., *Plan Atlas*, 'Plan der umgegend von Koniah' shows the situation well.

all of the fragments of antique sculpture and all of the Greek inscriptions that they could gather.<sup>12</sup>

In the winter of 1910–11, British scholar William Ramsay was granted permission to excavate remains of the ruined Seljuk palace on the citadel hill in the middle of Konya, and found there many Roman inscriptions, mostly funerary. Although Ramsay's collaborator, William Calder, who published these inscriptions, does not state it directly, it seems that the Seljuk palace had been, at least in part, built on the foundations of the late antique or early Byzantine citadel fortification wall, one that furnished these inscriptions.<sup>13</sup>

All these writers, and many others, either explicitly or implicitly attributed the construction of the walls of Konya to the Seljuk period, presumably due to the presence of Seljuk inscriptions in Arabic or Persian and Seljuk relief sculptures there. Later scholars take literally the account of the Seljuk chronicler Ibn Bibi relating the building of Konya's walls by Seljuk Sultan Alaeddin Keykubad I (r. 1219–1237) early in his reign.14 Be that as it may, the walls seem very likely to have been of late antique or early Byzantine construction. Figure 5.3 reproduces another photograph taken by Gertrude Bell in May 1905. It shows one section of the city fortification walls in the process of being dismantled, with marble blocks prominent in the foreground. While several different wall fabrics are distinguishable in this image, the bottom right of the tower displays the header-stretcher construction typical of Byzantine workmanship and found in the citadel and city walls of Sinop and Antalya, among other Anatolian cities.<sup>15</sup>

The walls of the Seljuk capital employed the visual language of Byzantine spolia, placing it in a frame that was at once religious, mythical and political. The rebuilding of these walls in the early 1220s, along with new relief carvings and new spolia, also

<sup>12</sup> de Laborde, *Voyage en Orient*, 117.

<sup>13</sup> Calder, 'Les inscriptions', 48–49, found that the western wall of the Seljuk palace, when excavated, was chock-a-block with inscriptions and other marbles: '… il se trouva que le mur avait été construit en grande partie avec des pierres inscrites.' McLean, *Greek and Latin Inscriptions*, ix, mentions without further reference the 'late Roman city walls' of Konya. He also quotes Calder on the citadel excavations, which, he writes, included 'many new tombstones.' There were two citadel walls to Konya, see Sarre, *Der Kiosk*, 10, fig. 5, for a plan of the citadel mound made in 1896 that shows the former location of a lower set of walls with square towers very much like those of the city walls, and a later, higher wall on top of one of the towers of which Seljuk Sultan Kılıç Arslan II erected the tower pavilion known as the Alaeddin Köşkü. See also von Vincke *et al*., *Plan Atlas*, 'Plan der umgegend von Koniah' for the lower citadel wall. One tower of this lower wall was uncovered during the construction of the building that currently houses the Konya Açı Özel Öğretim Kursu, diagonally across from the Karatay Medrese. It can be seen preserved in the basement of this building, many metres below the current street level. To the best of my knowledge this tower remains unpublished.

<sup>14</sup> Ibn Bibi, *El-Evamirü'l-'Ala'aiyye*, 254–255, for the account of the Seljuk building of the walls of Konya. This, in reality, likely consisted of rebuilding the four city gates and repairing other towers. The citadel walls must similarly have been rebuilt, and a second citadel, by the walls, the Zindankale, added.

<sup>15</sup> Brisch, *Galatien und Lykaonien*, 177, notes earlier sieges of the city (and citadel) of Konya, including during the First and Third Crusades, implying that these sieges needed walls. Brubaker and Haldon, *Iconoclast Era*, 76 and 546, note that Konya was considered a major Byzantine fortress and administrative centre in earlier centuries, also implying the presence of fortifications.

included foundation inscriptions, proverbs or aphorisms (*ahkam*), Qur'anic quotations, traditions of the Prophet Muhammad (*hadith*), and excerpts from the Persian national epic, the *Shahname*, thereby inserting the pieces of the past found on the walls in a double framework, one religious and the other epic. The oftrebuilt Byzantine walls were being transmogrified and re-enframed, while being preserved and rebuilt. But the period of this construction was the time of competition for the reconquest of Constantinople, and as we have seen, this programme can also be seen in this light.<sup>16</sup>

*Fig. 5.3: Fortification walls of Konya, detail (used with permission from Gertrude Bell Archive, Newcastle University).* 

#### **Moralising spolia**

Even though there must have been Byzantine Greek inscriptions galore at the time, and travellers mention marble slabs bearing Byzantine crosses on the walls of Konya, Seljuk builders in Konya and elsewhere never reused slabs bearing the distinctive lettering of medieval Byzantine Greek, preferring earlier lapidary styles of Greek, largely from the Roman era. In addition to following established Byzantine practice of spoliation, might there have been other reasons for this preference?

A lost inscription may provide insight into this and other Seljuk practices concerning the reuse of ancient marbles. Two nineteenth century authors report on this inscription, which was most probably Seljuk and roughly contemporaneous with the rebuilding of the walls of Konya. The first was English Captain Francis Beaufort, who surveyed the Mediterranean coast of Anatolia in 1811–12. At the riverside castle just south of the town of Manavgat,<sup>17</sup> constructed using blocks, likely taken by boat from the nearby Hellenistic and Roman port city of Side, Beaufort remarked:

On a tablet in the north wall we found some lines in Arabic, which were difficult to decipher. The following, however, was given by our interpreter as the general meaning:

<sup>16</sup> Redford, 'The Seljuks of Rum', 148–156. For a new and different approach to this topic, see Yalman, 'Repairing the antique'.

<sup>17</sup> Hellenkemper and Hild, *Lykien und Pamphylien*, vol. 2, 711 note the spoliated blocks. They plump for a medieval date, possibly Seljuk, for this fortress, which I think is of twelfth century Byzantine construction, as it fits the pattern of Byzantine (re)building of fortifications along the south coast of Anatolia in order to aid in the projection of Byzantine naval power towards Antioch, then in Crusader hands.

Be not vain of thy splendid apparel: I have experienced those delusions: the world is open to people of all ranks.<sup>18</sup>

The Ottoman Armenian Catholic priest and scholar Ghewond Alishan later published the same inscription. He described it as an Arabic maxim:

Ne te vantes pas de ta belle apparence; moi aussi j'ai passé par cette vaine illusion. Le monde est ouvert et libre devant toute classe humaine.<sup>19</sup>

As far as it is comprehensible from these two translations, the sentiment expressed is one of regret over the transient and vain nature of human existence. Without a full architectural context and original text, it is hard to decide who is being addressed and by whom. These are not, as Beaufort noted, sentiments that one might expect to find expressed in an inscription on a fortification.<sup>20</sup> Unless, that is, one considers the spoliated nature of its construction material. None of the aphoristic inscriptions from the walls of Konya or elsewhere has survived, but, intermixed with ancient stones, some may have expressed similar sentiments.

And if this practice was found on two fortifications, why not elsewhere in Seljuk realms as well? Persian was the language of administration of the Seljuks, and of the palace: the names of Seljuk monarchs duplicate those of some of the heroic monarchs of the *Shahname*. A partially preserved inscription on the citadel walls of Sinop is the best testimonial to the use of walls as loci of both mythologising contemporaneous events and placing them in a Persian cultural context. Written in Persian in the metre of the *Shahname*, this inscription celebrates the conquest of the city in 1214.<sup>21</sup> It does not express regret over the transience of the world, but does demonstrate a different kind of poetic contextualisation on fortification walls which were being newly reframed following the Seljuk conquest of the city the previous year.

Akşehir, ancient Philomelium, is a town some 50 kilometres north-west of Konya. Traveling in Anatolia in search of classical inscriptions, American epigrapher Sterrett and his team did not spend any time there. Like Konya, at the time Akşehir was without ruined classical buildings. However, like other travellers to Anatolia, Sterrett noted with approval the Seljuk buildings of Akşehir:

But few remains of Greek antiquity are to be found in Philomelium; but, on the other hand, the traveller is surprised by some Seldjuk ruins of exquisite beauty. The accurate workmanship displayed, even in the execution of details, will compare favourably with Greek buildings of a good period.<sup>22</sup>

<sup>18</sup> Beaufort, *Karamania,* 165. This section draws on Redford, 'Words, books, and buildings'.

<sup>19</sup> Roughly: 'Do not feel proud of your beautiful appearance. I, too, have experienced that vain illusion. The world is open and free for all manner of humans.' Alishan, *Sissouan*, 367.

<sup>20</sup> Beaufort, *Karamania*, 165: 'This does not seem to be a very apposite inscription for a fortress; but in Turkey, verses of the Koran, or some trite aphorisms, are promiscuously placed upon fountains, tombs, and fortresses.'

<sup>21</sup> Peacock, 'Persian verse', 247–253.

<sup>22</sup> Sterrett, *Epigraphical Journey*, 165.

Sterrett was in search of inscriptions, and it is true that the Seljuk buildings of Akşehir he admired in passing do not incorporate many spoliated blocks containing inscriptions. Still, Seljuk buildings here do employ many reused blocks of Hellenistic and Roman period, to an extent that is perhaps unparalleled in a period of great building activity that often included spoliated blocks.<sup>23</sup>

Akşehir constitutes a good place to consider the coincidence of physical remains of the past with the Islamic and Persian inflections of the Seljuk elite, the patrons of these buildings. Persian literature of the time contains moralistic scenes of ruins and uses them as loci of reflection on the passage of time and power. In Akşehir, one surviving coincidence of Persian verse and spoliated remains points to the moralistic use of spolia discussed above.

The Ferruhşah Mescidi in Akşehir is a small single-domed mosque dated by inscription to 1224, one of a series of small neighbourhood mosques built by Seljuk patrons at this time. It is reported that the mosque had a crypt below it, which is why I assign it a memorial/mortuary as well as a religious function.24 The most striking thing about this otherwise unexceptional building is the number and variety of spoliated marble blocks and pieces of architectural sculpture and funerary reliefs placed prominently on its entrance facade. As is the case with many Seljuk inscriptions, the foundation inscription of this building (itself carved on a reused marble block) is surrounded by an asymmetrical array of pieces of Byzantine architectural sculpture: templon blocks and chancel screens. All of these pieces share one feature, knotted decoration, protecting the inscription.

To the right and left of the entrance façade, two Roman era grave steles complete the ensemble of relief carved spolia. I think it is possible to link the funerary steles to a Persian language aphorism prominently situated on the beautiful carved wooden door that led into the building and therefore also formed part of the ensemble of the façade:

Kas nakhwāhad [a]gar nīkat va gardūn/Kam baste badīn kārst gardūn No one and fate do not wish your good/Because fate has little to do with this business.<sup>25</sup>

Anyone entering this small mosque would have passed through this door, having apprehended a façade centred on an inscription surrounded with fragmented knot patterned reliefs, and flanked by two grave stele; these combining to give an oddly personal message of the transitory nature of existence similar to that on the Manavgat castle inscription. The word I translate as fate literally means wheel, the wheel of fortune.

<sup>23</sup> Bakırer, '"Continuity" by "Reuse"', 67, ft. 6: 'In Akşehir, almost 75% of the ordinary building stones are reused building stones, in addition sculptured and carved pieces are also used.'

<sup>24</sup> Demiralp, *Akşehir*, 19–22.

<sup>25</sup> Konyalı, *Nasreddin Hoca*, 312.

### **Moralia and religion**

### *The Minaret of the Ulu Cami, Akşehir*

Of course the evanescence of this world is the stock in trade of religion. Even though it is a mosque, the Ferruhşah Mescidi's inscriptional and decorative programme does not make explicit reference to Islam beyond the standard issue Arabic language foundation inscription; indeed the use of Persian for the more prominent door inscription links the patron to the Seljuk elite of which he was a part more strongly than it does to any religious sentiment. However, in another instance, also in Akşehir, it may be possible to see a religious message in the reuse of spoliated marble reliefs placed around the foundation inscription near the base of the minaret of the Ulu Cami, the town's congregational mosque, which is dated by inscription to 1213, just over a decade before the inscription of the Ferruhşah Mescidi.<sup>26</sup>

Today, we think of minarets as standard parts of mosques, but the minaret came late to medieval Anatolia, and the minaret of the Ulu Cami of Akşehir is among the first of these dated securely by inscription. It may well have been added to a preexisting mosque, as happened elsewhere in Seljuk domains. The inscription itself seems to have been carved onto a rectangular Byzantine templon screen, with the cross originally at the centre of the relief effaced, to be replaced by the Arabic language foundation inscription.27 In addition to being made out of a Byzantine ecclesiastical spolium, the inscription is surrounded by other such marble reliefs below it and to the left. To the right is a Roman grave stele, which depicts confronted figures under an aedicule, which may have had a Medusa head in its centre.

The placement of these reliefs surrounding the foundation inscription and their location at the base of the minaret of the congregational mosque seem to impart a message of replacement, of Christianity with Islam. This is effected by placing these reliefs around an inscription in Arabic made over the effaced cross of a chancel screen; reliefs of relatively recent date that would have been recognisably pieces of Byzantine ecclesiastical reliefs. And of course, above the reliefs and inscription towers the minaret, a novelty at the time and a potent representation of Islam in the cityscape.

#### *The Sahip Ata mosque in Konya*

If the minaret was new to Seljuk lands in the early thirteenth century, half a century later, the double minaret (or façade flanked by minarets) was just as new. The fashion for a façade flanked by minarets originated in Iran.<sup>28</sup> Its first manifestation in Anatolia was in Konya, at the mosque known today by the name of its patron, the grand vizier of the Seljuk sultanate, now a vassal state of the Mongols, and is dated by inscription to the year 1258.

The Sahip Ata mosque was the second collaboration between the patron and an architect named Kaluk bin 'Abdallah. His name tells us he was of Armenian origin,

<sup>26</sup> Demiralp, *Akşehir*, 15–16; Bakırer, '"Continuity" by "Reuse"', 67–68.

<sup>27</sup> Konyalı, *Nasreddin Hoca*, 350–351.

<sup>28</sup> Korn, 'Twin minarets'.

and his patronymic that he was a convert to Islam.29 This partnership produced two of the most inventive facades of Rum Seljuk architecture, the İnce Minareli Medrese, which is devoid of spolia, or anything made from spoliated marble, and the Sahip Ata Mosque, which, with two notable exceptions, is also spolia-free. It is these two exceptions that concern us here because they, two Roman sarcophagi, are positioned prominently at the very base of both of the minarets.30 And rather than being sampled they were placed whole (sans lids) in this prominent location. In fact, the dimensions of the minaret itself seem to have been taken from the sarcophagi themselves.

The Sahip Ata mosque formed part of a larger complex that eventually included a Sufi convent, a tomb chamber for Sahip Ata and his family and a double bathhouse. It was located just outside the major southern gate to the city, in a prominent, heavily trafficked location. As a result, the inclusion, for the first time in Rum Seljuk architectural history, of a fountain on the façade of a mosque can seem natural, as it could provide both for thirsty or dusty passers-by and those performing ritual ablution before entering the mosque for prayer. What is more, as noted above, sarcophagi were often used for fountains, with the trough formed by the empty sarcophagus serving as a pool for water pouring from a pipe or spigot.

However, as Figure 5.4, another Gertrude Bell photograph, shows, the water in these sarcophagi was accessed through a hole drilled in the front of the sarcophagus, as well as, awkwardly, through a small opening, a miniature entrance, in a miniature copy of the building's façade that was placed above both sarcophagi, creating one of the most jarring contrasts in style I know in Rum Seljuk architecture.

The sarcophagi are mismatched. Both are made of the greyish-blue marble of Ladik (Laodiceia Combusta), north-west of Konya, on the road to Akşehir, and as such are local products. Although they are made of the same stone, and have more or less the same dimensions, their decoration is different. While at one time both had Greek inscriptions on their central panels, the left hand sarcophagus has geometric aniconic decoration, while the right hand sarcophagus (Figure 5.5) features two Medusa heads.

A similar sarcophagus is in the Konya Archaeology Museum. Its inscription reads: 'Aelia Paulina made this sarcophagus at her own expense both for herself and her husband Publius Aelius Cyrillus. Whoever else should break in will be liable to the Treasury for 2,500 [denarii]',31 so we can assume that at least the right hand sarcophagus had a similar inscription; indeed the inscriptions of other sarcophagi at the museum also threaten large fines to those who rob them.

Although the composition of both minarets, as well as of the façade, is symmetrical, there are clues in the design that it was the right hand minaret that formed the basis for the overall modular decoration of the square bases of these minarets, with the minaret shaft only changing to prismatic and brick once it was free of the top of the entrance portal.

<sup>29</sup> Redford, '*Mamālik* and *Mamālīk*', 329, ft. 31, relating information provided by Rachel Goshgarian.

<sup>30</sup> Redford, 'Sarcophagus', 200–203.

<sup>31</sup> McLean, *Greek and Latin Inscriptions*, 59, no. 180 and figs. 209–210.

*Fig. 5.4: The Sahib Ata Mosque, Konya, portal detail showing sarcophagus (used with permission from Gertrude Bell Archive, Newcastle University).*

What are these clues to the importance of the right hand minaret, and therefore the importance I place on the Medusaheaded sarcophagus? They lie principally in the writing on the miniature portal that sits above this sarcophagus. This miniportal was chosen by Kaluk as the place to inscribe his name. More importantly, this mini-portal has a richer epigraphic programme than the left-hand miniportal, with a series of Qur'anic quotations running around its edge. All of these quotations concern water.32 They literally begin at the beginning, with creation, and quotes concerning the importance of water to God's shaping of the earth and humans, and end with the fountains of Islamic paradise. With the minaret mounting skyward, the names of Islamic intercessors – the Prophet Muhammad and the four rightly guided Caliphs – appear further up.<sup>33</sup>

The minaret therefore appears to have been conceptualised as a link between

heaven and earth, this world and the next, for the Muslim faithful. At the base, the water of the fountain/sarcophagus literally underlies a miniature entrance bearing a synopsis of the importance of water for Muslims at the beginning and end of time. Above it, any Muslim lifting his or her gaze, perhaps to watch the muezzin on the balcony making the call to prayer, would also see the names of intercessory figures to whom they could pray for entrance to paradise on judgment day.

According to Olivier, the city gate directly opposite the façade of the Sahip Ata mosque, the Larende Gate, in addition to purpose-made reliefs of winged genii or angels and lions (reused from the lids of Roman era sarcophagi), was also decorated with reliefs of winged dragons.34 The dragon, often represented in a fashion similar to that of a snake, was, at that time, viewed as both a creature associated with water, as well as a celestial one: representative of the pseudo-planet Jawzahar, which was thought to cause eclipses.<sup>35</sup>

<sup>32</sup> Redford, 'Minaret meets portal'.

<sup>33</sup> To the right is the name of the caliph Abu Bakr, and to the right the caliph 'Ali. There are enough instances of this practice that have survived complete to permit me to state with some certainty that the other faces of this part of the minaret base would once have borne the names of the other two rightly guided caliphs, the prophet Muhammad and perhaps Allah as well.

<sup>34</sup> Olivier, *Voyage*, 390.

<sup>35</sup> Kuehn, *Dragon*, 51–56 for the dragon's association with water, 138–144 for Jawzahar. It is possible

The cosmic conceits of the 1220s, when the south gate of Konya was constructed, will never be understood fully, as there are few descriptions like Olivier's that can allow us to reconstruct its iconographic programme. But I think that it is possible to posit that, several generations on, it constituted one of the stimuli to the formulation of a new, more overtly Islamic cosmic conceit by the patron, Sahip Ata, the architect, Kaluk, and others in their orbit. That this was a more explicitly Islamic programme fit both with the type of building, and the circumstances, the

*Fig. 5.5: The Sahib Ata Mosque, Konya, portal detail showing right hand sarcophagus (Benni Claasz Coockson, 2003).*

eclipse of Seljuk power by a non-Islamic power, the Mongols, who had defeated the Seljuks in the previous decade and made them their vassal. The choice of a sarcophagus with prominent representation of snake-locked Medusas may have been seen as a way to bridge the gap between the two programmes. Whether this is the case or not, I think that, despite the aesthetic disjunction, the placement of these sarcophagi at the base of twin minarets whose decorative and epigraphic programmes expressed a complete and coherent message related the past and the present, the elements (water and sky) and an Islamic cosmology and soteriology in a new and sophisticated way. This message can be contrasted with the way in which spolia were used on the base of another minaret from half a century before this construction, that of the Ulu Cami in Akşehir. I do not think that the sarcophagi at the bases of the Sahip Ata mosque's minarets represent the replacement of one religion by another, as the spolia in a similar position at Akşehir likely did. Rather, if we take the Medusa sarcophagus as the real subject, they may constitute a reaching across what we view now as cultural, chronological and civilisational boundaries to incorporate a classical figure within an Islamic cosmology. In this case the pagans were not the Romans, but the Mongols.

#### **Conclusion: Dragomans and Orientalists**

We spent two days in Konia copying inscriptions and taking photographs of the Seldjuk city. The people of this eastern country seem to have had little interest in the affairs of this world, and spent their surplus energy in preparing tombs and epitaphs for themselves …<sup>36</sup>

that the reliefs of the side niches at the entrance to the Susuz Han caravanserai, which dates to the late 1230s, display something similar to the original arrangement of the Larende Gate. Another comparandum is a marble slab that somehow ended up in Cairo; a spolium itself, it was transferred from the fifteenth century mosque of Sultan al-Muayyad Shaykh (1415–1420) to the Islamic Art Museum in Cairo, see O'Kane, *Illustrated Guide*, 92. It depicts two addorsed winged and legged dragons with heads twisted into a confronted pose.

<sup>36</sup> Sterrett, *Epigraphical Journey*, 225.

The architect of two of the most innovative and sophisticated buildings of Rum Seljuk architecture was an Armenian convert to Islam. The Seljuk inscriptional record furnishes us with other examples of non-Muslims, or converts to Islam, who served as architects. Indeed, the architect whose style so resembles that of Kaluk that he is sometimes posited as his former assistant had a Greek name. This, person, called Kaluyan al-Qunawi (Kaloyannis from Konya) built another building for Sahib Ata in Sivas in the following decade.

At the end of this chapter, I raise this matter due to the academic tendency to mention conquest as a final act; as if nothing remains of the previous cultures in the lands conquered, and, as adumbrated, the tendency to ignore or diminish continuity when faced with dynastic and other political and dynastic change. We read of intrepid western European and American explorers and savants many centuries on without thinking of their companions, or 'fixers' if you will, who constituted part of a long continuum of interaction between Muslims, Christians and the remains of pagan cultures in this region. Translators, guides, or to use an old term, dragomans, were essential for the success of any of the missions mentioned here, and many more. Most dragomans were Ottoman Christians: Armenians, or as is the case here, Greeks.

Two such men are relevant to this essay. The first is Prodromos Petridis; according to Calder, he was Ramsay's companion on many trips, and performed the same function for others.37 The second was not a guide/translator in the usual sense of a companion. Sabbas Diamantidis was a medical inspector in Ottoman service in Konya. But, as a participant in the Philhellenic revival amongst Greek communities in the nineteenth century Ottoman Empire, he became interested in classical era Greek inscriptions, made his own copies and published them, and was one of the chief interlocutors of Ramsay, Sterrett and others who came to Konya in the late nineteenth century. While descrying their inferior quality and unreliability, Ramsay and Calder used his transcriptions and doubtless his connections to further their own work.<sup>38</sup>

Thirteenth-century Konya, and medieval Anatolia in general, must have had many figures like Petridis or Diamantidis: educated Christians or converts who knew Greek and other non-Islamic languages, and could 'translate' the past; a mixture of informed (Petridis) and speculative (Diamantidis), based on written sources, traditions, hearsay and beliefs. Perhaps one of them told a story about the Medusa heads that related them to Turco-Islamic beliefs in snakes and dragons, permitting the cosmological conceit of the minarets of the Sahip Ata mosque façade to encompass them. And yet another warned about the curse on the sarcophagi, causing the inscriptions to be defaced.

<sup>37</sup> Calder, 'Les inscriptions', 48 '[il] avait voyagé plusieurs années avec Sir W. M. Ramsay et d'autres explorateurs, et acquis une grande habilité dans l'art de copier les inscriptions, se tint constamment sur les lieux et sauva plusieurs textes de l'oubli.' This was not the view of Ramsay, *Intermix of Races*, who presented himself there as always traveling without a translator and in direct (if basic) contact with Turkish speaking villagers and townspeople.

<sup>38</sup> McLean, *Greek and Latin Inscriptions*, ix: 'Epigraphy was not his field of expertise, and many of his transcriptions were inaccurate'.

The medieval Mediterranean, especially when viewed by classical archaeologists, has been, and in some quarters continues to be, seen through the lens of decline. In Eurocentric views of the Greco-Roman past, again, still prevalent in some quarters, western Europe and North America are seen as the inheritors and guardians of the classical past. A corollary of these views is that 'non-western' cultures have no meaningful relationship with Greco-Roman civilisations, and therefore are not concerned with it, and neglect it. This essay has attempted to present not a model of linear decline, but one of shifting and differing encounter, reinterpretation and reuse of pieces of the past in medieval Anatolia, noting overlaps between Byzantine and Seljuk praxis, and also classical archaeologists' and epigraphers' grudging admiration for Seljuk architecture partially due to its prominent display of classical antiquity. This encounter continues in different ways in subsequent centuries in this region. However, I think that, varied and shifting as it is, the constancy and intensity of the Seljuk encounter with antiquity bespeaks both the wealth of the dynasty at this time (and its ambitious building programmes) as well as an attempt of patrons, architects and others to locate themselves in relationship to the past of the land over which the Seljuks ruled it found through the display and reframing of pieces of that past.

### **Acknowledgements**

I would like to acknowledge an intellectual debt to Professor Ömür Bakırer, author of the article quoted at the beginning of this article, although she is likely to find this homage too speculative for her taste. I have written on inscriptions and monuments examined here before: I hope this article serves to raise new issues and deepen old ones. I would like to thank the editors for their comments and suggestions.

### **Bibliography**

Alishan, G., *Sissouan, ou l'Arméno-Cilicie. Description Géographique et Historique* (Venice, 1899).


Brisch, K., *Tabula Imperii Byzantini 4. Galatien und Lykaonien* (Vienna, 1984).

Brubaker, L. and J. Haldon, *Byzantium in the Iconoclast Era c. 650–850: A History* (Cambridge, 2011).

Calder, W. M., 'Les inscriptions d'Iconium', *Revue de Philologie, de Littérature, et d'Histoire Ancienne* 36 (1912): 48–77.

Demiralp, Y., *Akşehir ve Köylerindeki Türk Anıtları* (Ankara, 1996).


## Chapter 6

## Echoes of late antique Esbus in Mamluk Ḥisbān (Jordan)

*Bethany J. Walker*

One of the most striking settlement patterns of the medieval era in southern Bilād al-Shām is the widespread reoccupation in the thirteenth century of the ruins of towns and villages built in the Byzantine and early Islamic periods.<sup>1</sup> At these longabandoned sitesthe walls of housesstillstood, and their cisterns were stillstructurally sound, though not maintained for centuries. Reoccupation of these ruins was not gradual: it happened suddenly and involved several family units. Roofs were repaired, cisterns cleaned out and re-plastered, fields once again cultivated and agricultural terraces built to expand cultivation and control water runoff. The majority of these sites that have been excavated in the region were reoccupied in this manner. But intriguing questions remain unanswered: why were these particular places chosen for resettlement after as long as three hundred years of abandonment? How were they revitalised? What was special about their urban form that made them good candidates for rebuilding? And, finally, how might these factors have contributed to the construction of memories of place in the medieval era? The revival of old place names, the return of family units and even small communities to ruined towns and the revival of the 'longing for homeland' (*ḥanīn ilā-l-awṭān*) chronicles in the thirteenth century suggest that particular places carried cultural and social meaning for generations: they were neverforgotten.<sup>2</sup> One case in point isthe site of Tall Ḥisbān, generally recognised by Biblical scholars as the Heshbon of the Old Testament, and known in the Roman and Byzantine eras as the town of Esbus. It is the best preserved and most intensively studied of these settlements in southern Bilād al-Shām, and has passed in and out of the textual record as long as arguably any site in the Levant.

<sup>1</sup> For a survey of such sites in Transjordan and historical Palestine, see Walker, 'Ceramics to social theory'; Walker and Dolinka, 'Khirbet Beit Mazmīl'; and Walker, 'Southern Syria'.

<sup>2</sup> Antrim, *Routes and Realms*; Walker, 'Reconstructing homeland'.

As such, it serves as the ideal case study to explore this cluster of questions arising from reoccupation.

By nature of their formation processes, tells are palimpsests: the traces of earlier phases of occupation are preserved and visible in reused, though altered, form. Tall Ḥisbān, located twenty kilometres south of Amman and in the hinterland of modern Madaba, is actually a quasi-tell: a natural outcrop of limestone whose current form was achieved through millennia of occupation, periodic agricultural terracing of its archaeological ruins and five decades of archaeological interventions. With its long history of settlement, abandonment, resettlement and physical transformation, the site is uniquely suited to the study of long-term urban development. American excavations began there over fifty years ago and continue today under the direction of the author, as part of a joint German-American initiative, making Tall Ḥisbān one of the longest-lived foreign-led archaeological projects in the Middle East. While the site has revealed periods of occupation spanning the Paleolithic to modern periods, spikes in construction have been documented for the Roman, Byzantine, Abbasid (early Islamic), Mamluk (middle Islamic) and late Ottoman (late Islamic) periods. The historical record provides some context for these peaks of growth.

The site is well documented in historical sources, from the Iron Age to today. The 35 references in the Old Testament to a place called Heshbon was the main reason that excavations were originally launched there over five decades ago. The Roman town of Esbus, which was the Greek name for the Hebrew Heshbon, is richly attested in written sources. Between 63 BC and AD 330, it was a border town and semi-autonomous polis, located on the Via Nova Traiana.3 The Byzantine town of the same name was a bishopric under the authority of Antioch. Islamic Ḥisbān first becomestextually visible to us with a reference by the thirteenth-century geographer Yāqūt al-Ḥamawī to an uprising of a descendant of the Umayyad family against the Abbasid caliph al-Ma'mūn early in his reign (AD 813–833); his revolt culminated with a stand-off with Abbasid forces at Ḥisbān, where there was a 'strong fortress' (*wa-bihā ḥiṣn ḥasīn*).<sup>4</sup> With the exception of later, random references by chroniclers and geographers to a place by this name, the town (*madīna*) of Ḥisbān reappears with a vengeance in the fourteenth century. In the Mamluk period, chroniclers (both Egyptian and Syrian) describe in detail visits by Mamluk sultans to the Citadel there, the political loyalty of its local tribes, and the agricultural wealth of its lands.5 It reached its peak in the first half of the fourteenth century, when the Mamluk state transferred the capital of the Balqa district there and installed a small garrison on the summit of the tell.

<sup>3</sup> On the Roman roads that serviced the site, see Mitchel, *Hellenistic and Roman Strata*, 66–69.

<sup>4</sup> Yāqūt, *Muʿjam al*-*Buldān*, vol. 4, 240–241; Grabar, 'A small episode of early Abbasid times'; Whitcomb, 'Reassessing the archaeology of Jordan in the Abbasid period', 506–508.

<sup>5</sup> A full discussion of the history of Mamluk Ḥisbān is well beyond the scope of this article. It has been a focus of the author's monograph *Jordan in the Late Middle Ages* and nearly a decades' worth of publications by the author that have followed since then. For a bibliography, see the project website: www.islamic-archaeology.uni-bonn.de/field-projects/tall-hisban.

The archaeological site consists of an eight hectare space defined by a fence built by the Jordanian Department of Antiquities in the 1970s and surrounded by the Wadi al-Majarr and a village access road leading to theNāʿūr Exitfrom Madaba. The settlement in the Classical, late antique and medieval eras, however, was much larger than this space, and likely extended under the modern village. The area defined by the fence consists of the architectural remains of the summit of the tell and the clusters of houses that surround its base. Visible on the ground surface at the base of the tell and beyond is the undulating pattern of collapsed barrel vaults (Figure 6.1). The site is a veritable rock pile: centuries of earthquakes have sent the heavy limestone blocks of the fortification wall downslope, further obscuring the architecturalremains of the domestic structures along the slopes and at the base of the tell.

The primary focus of the original American excavations of the 1960s and 1970s was the summit of the tell.<sup>6</sup>

*Fig. 6.1: Aerial view of Tall Ḥisbān (courtesy of APAAME\_20090930\_MJN-50 © Michael John Neville, Aerial Photographic Archive for Archaeology in the Middle East).*

This part of the site is dominated by the remains of a fourteenth-century citadel, built in the ruins of pre-existing structures. Most of the standing architecture here and surrounding the base of the tell is Mamluk in date, making Tall Ḥisbān the perfect case study of a fourteenth-century settlement and frontier garrison. For that reason, it has become a type site for the Mamluk period and has raised the status of Mamluk archaeology in Jordan. The buildings on the tell and around its base, however, are largely reoccupations and revitalisations of earlier structures. Where new structures were built in the Mamluk period they co-existed with the abandoned ruins of earlier times, creating a bewildering combination of 'living' areas – fresh constructions and rebuilt or repurposed ones – and areas of rubble. The Mamluk-era site preserves, but obscures, extensive remains of the late antique and early Islamic settlements.

<sup>6</sup> The preliminary reports of the original Heshbon Expedition, which ran from 1968–1976, can be found in issues of those years of *Andrews University Seminary Studies*.

While all of the reasons for the reoccupation of the site after some two hundred years are beyond the scope of this paper, the hidden faces of late Byzantine Esbus and early Islamic Ḥisbān and the Mamluk-era remodeling of their ruins are at the centre of this chapter. In the process of looking for the traces of earlier periods in a site built largely through recycling, this paper aims to spark debate on the social nature of reoccupation and how memories about places are constructed. As a Mamlukist, I am particularly interested in the conditions under which repurposing was chosen over demolition at late antique and early Islamic sites, when demolition was a common strategy in Mamluk Transjordan and Palestine. In addressing this question, I hope also to bridge the historiographical gap between the scholarship of the early and middle Islamic periods.

#### **The late antique and early Islamic town**

Much of what we know about Tall Ḥisbān in the late antique and early Islamic periods is the result of the early phases of fieldwork in the 1970s. In an effort to find the Old Testament site of Heshbon, the seminary of Andrews University in Michigan planned a large-scale excavation in the village of Ḥisbān in 1967. Fragments of capitals and columns on the summit of the tell were the only visible archaeological evidence at the time. The Six Day War prevented the season from proceeding as planned and fieldwork was postponed for a year.

The seven seasons which followed centred on the summit of the tell in Field A, where the remains of the town's fifth-century basilica were uncovered just below subtopsoil. In later seasons long trenches were opened on the slopes of the tell, uncovering mostly Byzantine construction with Roman spolia and evidence of early Islamic occupation. As nothing structural from the Iron Age wasidentified to associate the site with the Biblical Heshbon – on the contrary, the standing remains largely appeared at the time to be medieval Islamic in date – excavations in this project phase came to an end in 1976.

Let me focus on features that best illustrate the questions of reoccupation. The occupational history of Field A reveals much about the nature of the late antique town and its complex history of construction, continued occupation and reoccupation. Little remains of the buildings that pre-date the Mamluk Sultanate.7 What we find are mainly wallstubs,spolia, installations and architecturalruins apparently ignored and left to be by the masons making the repairs. The oldest construction on the tell is what isleft of the foundation levels of a monumental Roman-era building, which once occupied the entire length of the summit. The building was subsequently dismantled and its building blocks, columns and capitals reused in Byzantine construction of the

<sup>7</sup> The Mamluk Sultanate was a Cairo-based Sunni Muslim dynasty of slave soldiers that controlled much of the eastern Mediterranean from 1250 (in Egypt) to 1517 (in Syria). The Mamluks were responsible for putting an end to the Crusaderstates(with the conquest of Acre in 1291) and checking the Mongol-Ilkhanid expansion west to the Mediterranean. For a discussion on this expansion and its architectural impact see the contribution by Suna Çağaptay in this volume.

fifth century. The Roman monument wasidentified at the time by Andrews University staff as a temple, on the basis of a third-century coin, minted at Esbus; the obverse portrays the entrance to a large building topped by a pediment and a cultic statue of a figure with upraised arm visible through the doorway. In the Abbasid period, a bathhouse (*ḥammām*) of hypostyle form was constructed directly on top of the extant foundation courses, even making use of preexisting water installations. For this reason, recent assessment of the late Roman building suggests that rather than a temple, it was built as a nymphaeum.8 Beyond the monumental building on the summit of the tell, archaeological evidence for what must have been a sizable town includes the monumental staircase at the south entrance to the summit, the paved plaza at its base, a necropolis to the south-west of the modern village and extensive sherd scatters in the fields beyond the Wadi al-Majarr.

As many as four Byzantine churches have been identified at Tall Ḥisbān: two by the Heshbon Expedition and two others by the Department of Antiquities (visible only through partially exposed mosaics).9 The first was the basilica of Byzantine Esbus on the summit (the Acropolis Church), incorporating the stylobate walls and columns of the Roman ruins, which were constructed firmly on bedrock. The spolia were reused creatively: the apse, for example, was formed with sections of cornices laid upside down and aligned in a circle. This basilica was built in the second half of the fifth century and, following an earthquake, rebuilt in the early sixth. Iconoclastic damage to the mosaic suggests that the church continued in its original use as late as the eighth century. Elements of the east end of the basilica that were reused in early Islamic constructions remained visible in the thirteenth century, when the summit was fortified by the Mamluks.

As for the later use of this church, as is the pattern throughout Transjordan, the church continued to be used as a church well after the Islamic conquests.<sup>10</sup> Then, after a brief period of abandonment – demonstrated by a few centimetres of aeolian deposits – the ruins of the partially collapsed structure were later repurposed for domestic use. This may have occurred as late as the late Umayyad or early Abbasid period, as a two metre-wide *ṭābūn*, or clay oven, cut through the mosaic pavement of the church nave, illustrates.<sup>11</sup> Smaller walls subdivided the larger space into smaller storerooms at this time: here large storage jars, of complete form, were found during the inaugural excavation season 1968 turned upside down on a layer of yellow soil.<sup>12</sup> This yellow clay surface covers much of the north and east ends of the summit of the tell and provides a convenient stratigraphic anchor, separating the strata associated with the continued use of the basilica as a church from its repurposing for domestic use. It covered the flagstone pavement of the nave and sealed against the apse, and

<sup>8</sup> Pers. comm. Ms. Elena Ronza, Jordan Coordinator, Hisban Cultural Heritage Project.

<sup>9</sup> The following summarises Walker, 'Islamization of central Jordan'.

<sup>10</sup> Schick, *Christian Communities of Palestine*.

<sup>11</sup> The date assigned to the associated pottery by the original Heshbon Expedition is being downdated now to the late Umayyad and early Abbasid period based on parallels from Pella and Jerash.

<sup>12</sup> Van Elderen, 'Heshbon 1968: Area A'.

the column bases, all of which were left in place.These newly created storerooms were more or less intact during the reoccupation of the summit in the thirteenth century. 13

Many years later, in 2001, two rooms of similar construction were identified inside the north postern gate of the tell. This 'Area N house' represents repairs to an older structure, the vaulted ceilings and upper wall courses of which had collapsed.<sup>14</sup> During the reconstruction process, some sections of the vaults were left in place on the floor. *Ṭābūn*s, cooking pots and storage vessels were recovered from these rooms. What appears to be the same layer of yellow soil separates this phase of repair from the original phase of construction. In typical fashion, elementsfrom the Byzantine basilica on the summit (apparently still accessible) were reincorporated into this domestic structure, including fragments of a chancel screen reused as flooring and a baptismal font as a possible trough or household water basin. The structure continued to be occupied for some time; it was finally abandoned in the ninth or tenth centuries, and then reoccupied in the Mamluk period, when the space was subdivided into three rooms and may have been converted into the garrison kitchen.15

The second church is located a few hundred metres to the north of the tell, and wassubject to excavationsfor only one briefseason in 1976, by Baptist Bible College.<sup>16</sup> Its history of construction and use is a bit later than that of the Acropolis Church and may reflect growth of the Christian community at Esbus. Built in *c*. AD 550, the North Church was renovated and provided with new mosaic floors in the last quarter of the sixth century. The North Church seems to have continued to be used as a church until well into the eighth century. The building had been abandoned for quite some time when it was reused as a cemetery in the Mamluk period. The excavations by John Lawlor in 1976 uncovered some eighty cist graves, found throughout the building. Two burials are of special note, as they were complete skeletons laid directly on the mosaics of the apse (which was cleared of debrisforthis purpose), and the individuals were buried with sgraffito bowls, some of apparent Cypriot provenance.17 It is a rite normally associated with Orthodox Christians in Crusader Cyprus, and this is the first

<sup>13</sup> There was occupation at Ḥisbān and construction on the summit prior to the transfer of the garrison there in the early fourteenth century. Ayyubid-era strata were isolated by the Heshbon Expedition on the summit; this was based mostly on early versions of Handmade Geometrically Painted Ware, as dated by Jim Sauer. In recent field seasons, we have recovered Ayyubid pottery from the Middle Islamic farmhouse cluster below the tell. These are mostly glazed wares of the thirteenth century and the so-called 'Raqqa Wares', molded lamps, monochrome purple glazed wares and sgraffitos. There are very few textual references to Ḥisbān in the Ayyubid period: Salāḥ al-Dīn withdrew here after a failed attack on the Crusader forces at Kerak (Ibn Shaddād, *al-Nawādir al-Sulṭāniyya wa'l-Maḥāsin al-Yusūfiyya*, as cited in Russell, 'Hesban during the Arab period', 28), and in the fifteenth century the historian al-ʿAsqalānī attributed the construction of a madrasa there to the Ayyubid sultan al-Muʿazzam ʿIṣā (r. 1218–1227 in Damascus), al-ʿAsqalānī, Ibn Ḥajar, *Inbāʾ al-Ghumr bi-Anbāʾ al-ʿUmr*, 206.

<sup>14</sup> For the full preliminary report, see Walker and LaBianca, 'The Islamic *Quṣūr* of Tall Hisban'.

<sup>15</sup> Iraqi 'splashed ware' was recovered from this stratum; no pottery later than this or earlier than the thirteenth century was recovered.

<sup>16</sup> Lawlor, 'Excavation of the north church'.

<sup>17</sup> Walker, 'Islamization of central Jordan', 15–17.

*Fig. 6.2: View of entrance to Ḥisbān Citadel from the south. The layers of plastered roads are visible in section to the left of the monumental staircase used in the Roman and Byzantine periods (author).*

instance of such practice in Jordan known to this author. Very little of this cemetery was published, so much is it to be learned about the nature of the community buried there and the circumstances of the sudden revival of the North Church for burial. Some cleaning and preliminary probes were conducted in 2018 to achieve a full plan of the building and better understand the Byzantine stratigraphy.

These churchesformed part of what was a large town in the Byzantine period, with well-built stone farmhouses, multiple monumental urban cisterns and an extensive, cave-centred necropolis. As a bishopric, it fell under the ecclesiastical authority of the patriarch of Antioch in the province of Arabia.

The Acropolis basilica was the largest and most centrally located of the churches. The entrance to the acropolis and accessto the summit were developed in this period. Several stairs were added above the original Roman staircase, which led from the paved plaza to the summit; a steeply inclined road, carefully paved in limestone and repaved multiple times, gave an alternative and wheeled access to the summit (Figure 6.2). A massive Byzantine-era lime kiln was excavated at the base of this road in the 1970s, providing evidence of the importance of maintaining this road over a long period of time.18

<sup>18</sup> Bird, 'Heshbon 1968: Area D'; Walker *et al*., 'Tall Ḥisbān 2016 excavation season'.

As for the domestic areas of Byzantine Esbus, two houses of this period have been excavated at the base of the tell: on the westslope (Field C), which waslater developed into a housing cluster in the Mamluk period, and on the south slope (Field B), in a domestic space reoccupied in the fourteenth century. In both cases, these grew into multi-room, barrel-vaulted structures. One further house was excavated at the top of the south slope of the tell (Field M) and will be discussed below. While the plans and extent of these buildings are heavily obscured by Mamluk-era rebuilding, a few observations can be made. Their construction was on bedrock and of good enough quality to have remained viable as living spaces hundreds of years later.

The Field B house, excavated each season since 2011, is of special interest. Two pits – filled with garbage – were sunk into the living space of one room of the original, Byzantine structures and subsequently plastered over, and the space then returned to domestic use once again, all during different phases of use within the late Byzantine period. The walls of the ruined house were preserved in some places four courses high for hundreds of years, and the structure was rebuilt and reoccupied in the Mamluk period. The central living space was subsequently subdivided by a wall partially constructed with spolia (in this case Byzantine incense stands) likely taken from the Acropolis Church. Roman and Byzantine spolia of many forms (capitals, column drums, stands) were heavily incorporated into the walls of the Mamluk-era revitalisation of this structure. Fragments of painted plaster, which fell off the walls during these interventions, give hints to what the interior of houses such as this one may have looked like in the fifth to seventh centuries.

In the late fourteenth century, the site suffered damage as the result of further tremors (as documented throughout the site) and many of the houses were temporarily abandoned. The room with the pits in the Field B house appears at this phase of occupation to have been used as a pantry, where three jars were stored. At the time containing fermenting goat milk and olive oil, the residents encircled them with a line of stones and covered them with soil, leaving them behind for some reason.19 They later returned, attempted structural repairs to the walls of the house and eventually left it for good. It is a fascinating story of revival, repurposing and phased abandonment that is repeated across the site.

While there islittle evidence of new construction at the site overall in the Umayyad period, we do find it in the Abbasid era. The Abbasid horizon is coming into clearer focus at Tall Ḥisbān, in part the result of a general trend in revising the chronology of early Islamic wares, but also the consequence of where excavation trenches have been placed in recent seasons. Abbasid-period levels have been identified in the Mamluk 'farmhouses' at the base of the tell, though hidden by the thirteenth- and fourteenth-century constructions, as well as on the summit of the tell. One important development in this regard was the re-dating of the three-room bathhouse, built on top of the narthex of the Acropolis Church and into the foundation levels of the Roman

<sup>19</sup> For a full analysis of the contents of the jars (subjected to lipid analysis at the Freie Universität Berlin), their stratigraphic contexts and ceramic parallels, see Walker *et al*., 'Residue analysis'.

monumental building, mentioned above. Originally attributed to the Mamluk period in the 1970s, fieldwork in 2007 identified stratigraphic connections with securely dated Abbasid surfaces to the west of the hammam entrance.<sup>20</sup> In terms of plan and structure, it fits more the typology of early Islamic private baths than anything of the Mamluk period.<sup>21</sup> While certainly cleaned and put back into use in the Mamluk period, when the adjacent rooms of the 'Governor's House' were built, it now appears to have been built in the Abbasid period, when there is further evidence of construction on the summit.<sup>22</sup> This general fortification of the summit may be associated with the al-Fudaynī revolt described by Yāqūt, as referenced above.

The early Islamic period, in short, generally becomes visible to us in the form of spolia and walls reused in Mamluk and Ottoman structures. A fortified farmstead on private land, located outside the site fence to the south and locally known as the 'Nabulsi *qaṣr*', provides a tantalising glimpse of what this place used to be before its dramatic facelift in the Mamluk period. While the complex in its current form evolved over the course of the nineteenth and early twentieth centuries,23 it makes considerable use of building material (which must have been available close by) and possibly even structuralruinsfrom much earlier periods. The entrance to an Ottoman building of some official function (villagers today call it 'the jail', *sijn*) is flanked by panels with Arabic inscriptions, used as spolia. The longest one, though preserved only in mid-section, is of special interest. It is a well-preserved inscribed block, with text in early Islamic Kufic (Figure 6.3). A preliminary reading is:


It could have been part of a tombstone, a boundary marker, or foundation text from the façade of an earlier building subsequently dismantled. Spolia are found in the walls of the entire Nabulsi complex in the form of decorative panels, columns, entablatures and capitals. In the 1940s a stable was added to the complex, making use of several earlier walls and installations, including a series of animal troughs. Careful study of architectural phasing and photogrammetry have documented several building

<sup>20</sup> De Vries, 'The Islamic bath at Tell Hesban'; Walker and LaBianca, 'Tall Hisban (2007)'.

<sup>21</sup> Dow, *Islamic Baths of Palestine*. Moreover, there are no known examples of Mamluk citadels with their own ḥammams. Even in the Cairo Citadel, the centre of power of the Mamluk sultans, the soldiers resident in the Citadel bathed in the baths of city below.

<sup>22</sup> In terms of general plan, the bath was entered from the east, and was in its foundational use likely part of domestic structures that once stood on the east side of the summit in the Abbasid period, rather than the Mamluk-era *qāʿa* to the west.

<sup>23</sup> Ahmad *et al*., 'Mabānī Qaryat Ḥisbān'; Carroll, 'Sowing the seeds of modernity'; Carroll *et al*., 'The Ottoman Qasr at Hisban'.

<sup>24</sup> Walker and LaBianca, 'The Islamic *Quṣūr* of Tall Hisban', 258. For an alternative reading, see Bisheh, 'Arabic inscriptions from Hesban', 10–11.

*Fig. 6.3: Inscribed panel as spolia in doorway of Ottoman building, 'Nabulsi* qaṣr*' (courtesy of Dr. Daniel Redlinger).*

phases that, on the basis of architectural parallels, span the late antique and early Islamic periods.<sup>25</sup>

## **Reoccupation – the Mamluk (middle Islamic) site**

Excavations since 1998 have, by design, been Islamic in focus. While Tall Ḥisbān is a gift to Mamlukists, it presents particular challenges to the study of the Byzantine and early Islamic towns. Methodologically it requires stripping away the Mamluk layers to see what lies below, or inside. Typical Mamluk construction in the

Transjordan central highlands is of stone construction – heavily reusing Byzantine and early Islamic building blocks – with barrel vaulting; everything is built directly on bedrock. Earlier buildings that were deemed structurally unsound, or for some reason in the way, were demolished. On the citadel, much of the Roman, Byzantine and early Islamic construction was simply dismantled or reused.

In converting the summit of the tell into a Citadel in the Mamluk period, construction centred on the western half of the summit, where a barrel-vaulted complex was built. Identified as the residence of the Governor of the Balqā, the complex centred on an open-air, flagstone-paved courtyard and based on the abbreviated form of the *qāʿa* plan common to Mamluk urban architecture. On the west side of the courtyard was the large vaulted room with plastered floor and walls that we have identified as the *īwān* where the governor may have held audience.26 On the east side the complex incorporated and rehabilitated the small hammam built hundreds of years earlier, as discussed earlier. The function of the smaller vaulted room on the north side may have served multiple functions; the two long parallel vaulted rooms on the south, located behind a staircase and smaller vaulted rooms, comprised the Citadel storeroom.27 Another row of barrel-vaulted rooms framed the southern side of the summit, producing the monumental façade at the entrance to the Citadel, but maintaining the original Byzantine staircase. There appears to have

<sup>25</sup> An initial study was done by architecture students and faculty at the University of Jordan in 2001 (producing the report by Ahmad *et al*., 'Mabānī Qayrat Ḥisbān'). In 2018 documentation of architectural phasing resumed, with the aid of photogrammetry, by Dr. Nicolò Pini.

<sup>26</sup> On the many visits of Sultans al-Nāṣir Muḥammad and Barqūq to Ḥisbān, see Walker, *Jordan in the Late Middle Ages*.

<sup>27</sup> This storeroom is unique for excavated citadels of the Mamluk period. The collapse of the vault in the mid-fourteenth century preserved much of its contents, producing a rare sealed context for the most complete assemblage of display, table, cooking and storage wares, lamps and military accoutrements (though fragmentary) in Bilād al-Shām outside Damascus, see Walker and LaBianca, 'The Islamic *Quṣūr* of Tall Hisban'.

been, however, no new construction in the eastern half of the summit. The storerooms that were inserted into the basilica's nave in the early Islamic period were simply filled in, and whatever other buildings and installations that still stood there were otherwise dismantled to create a level surface for an unpaved courtyard. The series of vaulted rooms that lined the northern, southern and western sides of the summit of the tell, then, together produced a U-shaped complex, which faced this second open-air courtyard to the east.

The tell as a whole is riddled with a maze of subterranean passageways that form channels between large cisterns, and at least two similar multi-room complexes can be found at the bases of the south and north slopes. These urban cisterns appear to have been constructed in the Byzantine period and remained fully functional until the mid-Abbasid abandonment of much of the site.28 In the early Mamluk period they were carefully cleaned, re-plastered and put back into use. A large Byzantineera reservoir located half a kilometre from the tell, was also cleaned, re-plastered and restored as a water-holding installation at this time. Over the course of the late fourteenth and fifteenth centuries, most of these facilities went out of use for water storage and were repurposed as urban middens.29

The largest of the rehabilitated cistern complexes is located at the bottom of the south slope of the tell, in Field B. As hand-mapping of the complex proved precarious, laser mapping and 3-D reconstruction were adopted in 2016,30 in addition to small excavation probes in rooms with secure superstructure during the 2010 season.31 This complex had two stories and was covered by a barrel vault; everything today is below ground, the roof obscured by extensive alluvium and colluvium. Excavation and architectural analysis suggest an original construction date in the Byzantine period. As with most of the other large cisterns on the site, portions of it were re-plastered in the early Mamluk era, and it went largely out of use in the late fourteenth century, when it was repurposed as an urban midden.

Below the tell to the west and south, the Mamluk village took the form of clusters of single-room, vaulted houses, each built against the next in a line facing a shared courtyard (Figure 6.4). This is a regional koine of vernacular architecture found in the central highlands of Jordan and, while of long history, is rather typical of the Middle Islamic period.32 In the centre of these courtyards were small cisterns, which have been identified as family cisterns. Bedrock has been reached in two of these housing clusters. One is on the north slope in Field M, where a row of such structures faces downslope and makes use of sumps, drains and cisterns designed in

<sup>28</sup> Only three cisterns (two on the summit of the tell and one on the west slope) can be attributed to the Mamluk period, see Walker, 'The struggle over water'.

<sup>29</sup> Walker, 'The struggle over water'.

<sup>30</sup> Walker *et al*., 'Tall Ḥisbān 2016 excavation season'.

<sup>31</sup> Walker and LaBianca, 'Tall Hisban (2012)'.

<sup>32</sup> For parallels at Khirbat Fāris and Tall Dhibān,see McQuitty, 'Khirbat Faris' and Porter, 'Locating middle Islamic Dhiban'. At both sites, the Mamluk-era structures are largely rebuildings of Byzantine and early Islamic ones.

*Fig. 6.4: Site plan of Tall Ḥisbān (courtesy of Dr. Nicolò Pini).*

the Roman, Byzantine and early Islamic periods. The doorway and vaulted chamber of one of the vaulted structures (dated to the Mamluk period) is built directly on top of the standing walls of an early Islamic and late Byzantine building, which, in turn, used sections and stubs of Roman walls (Figure 6.5). The Byzantine levels contained covered drains and cooking and storage areas. Throughout the site, three phases of Mamluk occupation could be identified, which in this field correspond to the use of the vaulted buildings as domestic space, possible *ṭābūn* houses and storage facilities, in turn, reflecting their developing functional relationship with the Citadel.33 Individual buildings went out of use at different times, however, as the settlement was slowly abandoned from the late fourteenth century. 34

The best preserved and largest of the housing clusters, in Field O, has produced the richest ceramic and small finds assemblages, as well as macro- and micro-botanical remains. Four independent vaulted structures of 5–7 metres in dimension were built in a row against one another (see Figure 6.4). On the basis of coin- and ceramic-dated foundation trenches and analysis of architectural phasing, the original construction of the North House has been attributed to the Abbasid period. It is likely that some of the other buildings in this cluster were as well. The South House was reoccupied in the thirteenth century, and the rest of the building soon followed suit. Renovations

<sup>33</sup> Walker *et al*., 'Tall Hisban 2013 and 2014 excavation seasons'.

<sup>34</sup> Walker, 'Planned villages and rural resilience'.

in the Mamluk period consisted of adding a few wall courses and replacing the vault, with interior dividing walls added in later phases. The stone working was local: the debris left from finishing the limestone blocks litter the outside courtyard. Floors and walls were heavily plastered and replastered. The abandonment of the North House sometime in the fifteenth century may be specifically related to the presence of a cannonball embedded in the plaster of the doorway and discovered in 2018, as yet unpublished.

The association of agriculturaltools and animal accoutrements with this cluster,

*Fig. 6.5: Mamluk-era vaulted structure with doorway in Field M, view upslope. Note Early Islamic, Byzantine and Roman walls below (project files).*

and others on site, are expected for small finds in a rural community reliant on agriculture and animal husbandry. Less expected were the gaming pieces, a hoard of sixty silver coins, fragments ofrather high quality enamelled and lustered table wares and lamps and market weights.35 Every housing cluster has produced such material, with the highest concentrationsin the North House of Field O. Such assemblagesspeak to a more urban environment with a diverse population and should be related to the important economic and social changesthatresulted from the transfer of the garrison here during Sultan al-Nasir Muhammad's third reign (AD 1310–1341). Contemporary narrative sources call Ḥisbān a city/town (*madīna*), rather than a village (*qarya*), which was its nomenclature in later centuries.36 Further evidence of local flourishing has been found in an annex to the North House, added in the Mamluk period and excavated in 2018, which would seem to have had a specialised function, as it was equipped with plumbing and drains, a plastered basin and areas of what appear to have been worked clay. This, in addition to the recovery in previous seasons of firing tripods and wasters, as well as the results of recent petrographic and XRD analyses of glazed wares, suggests local ceramic production – perhaps in such a space.37

Understanding the historical context of these changes requires recourse to texts. Unlike the Umayyad and Abbasid eras, the Mamluk period istext-rich, offering a wide range of written sources of different genresthat are of direct utility for archaeologists. While the texts, largely in manuscript form, are not easy to use (or access, or even

<sup>35</sup> While chemical analysis of the enameled glass is planned for this coming year, it is likely that they were imports from Damascus, which specialised in such products and with which the town of Ḥisbān at the time had very close professional, family and economic ties. For the 2016 season preliminary report, see Walker *et al*., 'Tall Ḥisbān 2016 excavation season'.

<sup>36</sup> For the historical documentation on Ḥisbān in the Mamluk period, see Walker, *Jordan in the Late Middle Ages*.

<sup>37</sup> Kassem, 'Reconstructing rural social networks through archaeometry'; Walker, 'Made on order for the Amir'.

find), they often reveal the reasons for patterns we see in the archaeological record. An endowment document (*waqfiyya*) of the late fourteenth century in Cairo describes single-room, barrel-vaulted structures of more or less identical shape and form in the village of Adar near Shobak.38 Outside of the physical description of houses contemporary with those at Ḥisbān (and very similarin dimensions and construction), we learn about their function and condition. First, most (but not all) were houses, and all were apparently rental properties, the revenues from which were earmarked for the financial support of Sultan Shaʿbān's madrasa in Cairo. Second, many of them were derelict: no one lived in them and they are described as being 'in ruins' (*kharāb*). They are, nonetheless, listed as part of the village holdings and endowment, likely with an eye to renovating them and making them profitable once again.

In addition to the eighty-three houses, the Adar waqfiyyah describes all of the 157 cisterns (127 of which were still in use) as *ṣahārīj rūmāniyya* – 'ancient', that is rehabilitated, cisterns.39 One of the cisterns belongs to the village mosque, and the remainder to households. The urban-scale cistern complex that is described in this document in detail is referred to as a *ṣaḥrīj rūmānī*. Clearly the practice in the early Mamluk period was to restore these facilities rather than construct new ones, if possible.

### **Construction, settlement and the Mamluks' investment mentality**

Building practice throughout the Mamluks' territories was, above all, cost- and timeeffective: one built quickly atthe least cost. Ruins of older buildings were repurposed, if they were stillstructurally sound; otherwise they were demolished and new structures were built directly on bedrock, largely reusing old building material. This kind of construction is typical of the recycling of ruins (*khirab*) that defined the building industry of Mamluk Cairo;<sup>40</sup> it is increasingly being identified through excavations of contemporary sites throughout Transjordan and Palestine, providing archaeological evidence for the kind of building practices described in Mamluk-era chronicles and *waqfiyyāt*. <sup>41</sup> Most buildings were an amalgam of old and new elements. As in Cairo

<sup>38</sup> Waqfiyyah 8/49 (Wizārat al-Awqāf); Ghawanmeh, *Tārīkh Sharqī al-Urdunn*, 243–244; Ghawanmeh, 'al-Qarya fī junūb al-Shām'; al-Qaḥṭānī, *Awqāf al-Sulṭān al-Ashraf Shaʿbān*, 66; Walker, *Jordan in the Late Middle Ages*, 154–164. The modern village of Adar has never been subjected to archaeological study and there are no standing remains of any structures from the medieval period.

<sup>39</sup> In medieval Arabic sources, *rūmānī* was a term typically used to describe anything old, the origins of which go beyond communal memory, whether an olive tree, terrace, orchard or water facility.

<sup>40</sup> Louiseau, *Reconstruire la Maison du Sultan*. The forthcoming PhD dissertation by Ms. Hend El Sayed (University of Bonn), entitled 'Modelling economic change in late medieval Cairo through religious architecture', includes an economic and spatial assessment of urban ruins. For more on the changing meaning of 'khirbah' in the Mamluk and Ottoman periods, see Walker, 'Khirbeh'.

<sup>41</sup> For archaeological evidence of construction on bedrock at Tall Ḥisbān, see Walker, 'Planned villages and rural resilience' and at Khirbet Beit Mazmīl in Jerusalem, see Walker and Dolinka, 'Khirbet Beit Mazmil'; for a review of published reports throughout in Israel, see Walker, 'From ceramics to social theory'. While construction on bedrock was the general practice, there were a range of localised building preferences beyond Cairo, such as building on fill, as at the north-west Quarter of Jerash and

today, the residents of Mamluk Cairo became accustomed to living side-by-side with ruins; the sight of rubble and collapse was commonplace and one unfortunate byproduct of the legal complexities of the medieval *waqf* system.<sup>42</sup> It appears that it was no different in other cities – large and small – in Egypt and Syria. Mamluk-era construction at Tall Ḥisbān follows the same pattern. Some collapsed walls and vaults were pushed off the summit. Most walls, however, were simply dismantled and the building material reused in new construction. Quarrying of new stone was kept to a minimum, even though soft limestone was readily available in the natural caves below the tell (which were quarried in antiquity) and in easily accessible limestone beds in the hills across the Wadi Majarr. Construction and remodelling of houses in the town at the base of the tell was done in the same way.

In making the decision on where to move garrisons, courts, markets and people, the Mamluk state displayed the same attitude towards investment that they did with construction: try to achieve immediate results with least effort and funds. The administrative structure of the Mamluk territories was quite fluid, with regular transfers of district (*ʿamal*) and even provincial (*wilāya*) capitals from one town to another. The borders between districts often shifted, it appears largely for political and military reasons: to reign in recalcitrant tribes and reward the faithful among them. Because of this frequent restructuring, it was in the interest of the state to not invest too much time or resources in one particular town. District capitals seem to have been selected on the basis of their economic feasibility as centres of government; they needed to have adequate water supplies and a water infrastructure already in place, stable markets, a large (and professionally qualified) population and be located on major transport routes.43 Ḥisbān was certainly qualified, as it had a very well developed water and transport infrastructure (connecting it to Cairo and Damascus and the ports of Palestine), vast cereal fields and a diverse orchard-based economy, the remnants of a fortification on the summit of the tell, its own marketplace that was said to have served overthree hundred local villages, a court, educational institutions and a self-sufficient population of peasants and artisans.<sup>44</sup>

In terms ofsize, public services and diversity of population, Ḥisbān lay somewhere between a village and a town in the Middle Islamic period. Textual sources are ambivalent about its status, as they often are in regards to site typology, but perhaps with a good reason: until the fourteenth century, Ḥisbān was the size of a village and it was largely dependent on agriculture and animal husbandry, but it had, at the same time, several services and institutions typical of towns (a madrasa, a marketplace, a

at Tall Dhibān, see Kalaitzoglou, 'A middle Islamic hamlet in Jerash'.

<sup>42</sup> Endowed properties were, in theory, to remain unchanged, and budgets were set for building maintenance. Many legal loopholes developed – hotly disputed among the ulema of the day – to deal with the inevitable problems of building collapse. For a review of the literature and debatesin Mamluk Studies, see Walker, *Jordan in the Late Middle Ages*, 279–281.

<sup>43</sup> Walker, *Jordan in the Late Middle Ages*, 108–112. For the special requirements of local administrative centres in the Ottoman period, see Akarlı, *Some Ottoman Documents on Jordan*.

<sup>44</sup> The main study on the town of Ḥisbān in the Mamluk period, as well as Transjordan as a whole, is Walker, *Jordan in the Late Middle Ages*.

court). When the decision was made to move the capital of the Balqāʾ district to Ḥisbān during the sultan al-Nāṣir Muḥammad's third reign, the place (now called a *madīna* by contemporary chroniclers) grew rapidly overnight. A garrison was established on the summit of the tell, old buildings were repurposed for new use (as stables, storage facilities and possibly workshops), administrators and merchants were resettled here and business in the marketplace boomed.45 There was manic building activity in the first half of the fourteenth century. <sup>46</sup> None of it, however, would have been possible in the short amount of time it took to transform this otherwise rural settlement into a garrison-town, if water and building infrastructures were not already in place.

## **Concluding thoughts**

The phenomenon of the later medieval reoccupation of Byzantine and early Islamic ruins should probably be understood in the context of large-scale movements of peoples and financial prosperity in the Mamluk period. Ḥisbān's population grew rapidly in the early fourteenth century, partially as the result of population transfer by state fiat; the town lost population under the same conditions at the end of the century, as the soldiers, merchants, clerics, administrators and many others were 'transferred' to Amman. The summit of the tell wasthen abandoned, as were buildings outside the gates of the Citadel used as stables and storerooms. The Damascene historian Ibn Hijji al-Ḥisbāni (whose family hailed from Ḥisbān) explains the transfer as a financial move: at one point Ḥisbān was a good investment for a Mamluk officer, and then it was not.47 Along the lines of the information in the Adar *waqfiyyah*, could we envision that the ruins of Byzantine Esbus and Abbasid Ḥisbān were quickly renovated to house the newly arrived state officials, and that at least some of them were rental properties for such a purpose?

Quick construction at the least cost would be an expedient way to house large numbers of newly arrived people. I suggest that we are witnessing a pattern of demographic of growth in the thirteenth and fourteenth centuries that is partially to be credited for the remarkable economic growth of the Mamluks' territories in this period. Construction and (re)occupation in this manner might also help us understand

<sup>45</sup> Population transfer, while normally associated with the Ottomans as a political strategy, was known, as well, in the Mamluk period, where it was occasionally practiced for financial reasons. The Governor of Syria, Amir Tankiz, for example, moved people to Ajlun, after the flood of 738/1328 depopulated the town and destroyed much of its marketplace, see references in Walker, *Jordan in the Late Middle Ages*, 230 ff.

<sup>46</sup> There is no question about the chronology for this. Mamluk glazed wares of the fourteenth century are quite distinctive, and we have numerous textual sources, a large corpus of legible coins, a growing series of C14 dates, and excellent stratigraphic contexts (series of plastered and flagstone floors, structural collapse sealing deposits,*etc*.), and objects with inscriptions and blazons that provide rough dates. The storeroom of the Mamluk Citadel is unique for Greater Syria, as the contents were intact and sealed by the collapse of the vaults in mid-century.

<sup>47</sup> Transmitted by his student Ibn Qāḍī Shuhba in the latter's *Tārīkh*, 3.174. See the discussion of the historical narrative in Walker, *Jordan in the Late Middle Ages*, 77–78.

the apparent ease of the residents with living among ruins, as we see on the Ḥisbān citadel, in the settlement below and in villages such as Adar.

Let us recall that in the Cairo document many of the structures mentioned were derelict: no one lived in them and they are described as being 'in ruins' (*kharāb*). The reoccupation of places of 'ruin' teaches us an important lesson about urban resilience. The water infrastructure, settlement organisation and domestic architecture of late antique and early Islamic towns and villages were stable and flexible enough to be re-appropriated centuries later by a markedly different regime. In the case of Ḥisbān, there is some evidence for an enduring memory of place in the form of the toponym, which was retained over a very long time, even when no one lived year-round on the tell anymore and there was only limited evidence that the fields nearby were in use. The Ḥisbānīs ofthe thirteenth and fourteenth centuriesrelated in complex waysto the ruins they appropriated. Most were reused and repurposed; others, like many of the structures at the base of the tell, were simply ignored and became a tolerated visual experience of the new urban landscape, devoid of cultural meaning and communal connections. Unfortunately, the material evidence is not much helped by the textual record, rich as it is for the Mamluk period, to reconstruct what these ruins meant to the inhabitants of the village.

Recent interviews with the villagers reveal at least one way in which the tell has become part of communal consciousness in recent decades.48 For today's residents the summit of the tell has always been a *muṣallā*, an open air place of prayer where the community would gather especially for the great feasts, but also for rain prayers, commonly located just outside a settlement.49 In recent years, the architectural ruins uncovered by excavations have become the backdrop for village celebrations and family excursions. The archaeological site has, thus, become a stage for activities that are entirely unconnected from their original function.

What the archaeological record suggests is a return of people, over and over again, to a place that had well preserved buildings and a well-developed water infrastructure. The place name was retained over centuries. Even when most of the architecturalruins were not visible anymore, the cisterns and reservoirs were, and the extensive of subterranean channels that connected many of them were either known or rediscovered. That subterranean world of caves, channels and cisterns way have been one of the most important factors in bringing people back to this site time and again. There is a strong pattern of water use on the summit of the tell that includes the Roman nymphaeum, the Byzantine basilica (and its many baptismal fonts), the Abbasid *ḥammām* and the water associations connected to communal prayer on the summit in more recent times. The urban water systems below the tell were used time and again, revived either for their original purpose, or repurposed for domestic use,

<sup>48</sup> Pers. comm. Prof. Frode Jacobsen, social anthropologist at Bergen University College, Norway.

<sup>49</sup> In village parlance, a *muṣallā* is simply a large, outdoor space where people can gather for prayer. It does not need to be associated with a built structure. On other definitions of the term, see Wensinck and Hillenbrand, 'Musallā'.

stabling, storage, or garbage disposal.50 Large built spaces have a way of bringing people back to a place.

Isuggest that we understand the process of keeping, discarding and culturally (re-) appropriating elements of the built environment as a mode in which people in the past have read the connections between ancient buildings, landscapes and the people who lived there before them. By focusing on the changes that new peoples make to old places, we can begin to trace this process, if not fully understand its meaning.

## **Bibliography**

#### *Primary*

al-ʿAsqalānī, Ibn Ḥajar, *Inbāʾ al-Ghumr bi-Anbāʾ al-ʿUmr*, ed. M. A. Duhmān (Damascus, 1979). Ibn Qāḍī Shuhba, Ṭāqī al-Dīn, *Tārīkh Ibn Qāḍī Shuhba*, ed. ʿA. Darwish (Damascus, 1997). Waqfiyyah 8/49, *Ḥajaj 'umara' wa salāṭīn, Dar al-Wathā'iq* (Cairo). Yāqūt al-Ḥamawī, *Shihāb al-Dīn ibn ʿAbdullah, Muʿjam al-buldān* (Beirut, 1957).

### *Secondary*

Ahmad, A. S., D. al-Huseini, D. ʿAnbatawi, and ʿA. Musa, 'Mabānī Qaryat Ḥisbān', unpublished project architectural report, University of Jordan (2001).

Akarlı, E., *Some Ottoman Documents on Jordan: Ottoman Criteria for the Choice of an Administrative Center in the Light of Documents on Hauran, 1909–1910* (Amman, 1989).


Bisheh, G., 'The Arabic inscriptionsfrom Hesban' in P. Ray (ed.), *Small Finds: Studies of Bone, Iron, Glass, Figures, and Stone Objects from Tell Hesban and Vicinity*, Hesban 12 (Berrien Springs, 2009), pp. 9–16.


<sup>50</sup> A brief report on the excavation of the largest of the Byzantine cisterns can be found in Walker and LaBianca, 'Tall Hisban (2010)'. For a vivid example of reuse of water systems at a contemporary hilltop site in Jerusalem, see Walker and Dolinka, 'Khirbet Beit Mazmīl'.


## Erasure and selective memory

## Chapter 7

## Constantinople's medieval antiquarians of the future

*Benjamin Anderson*

*Denn gründen alle sich nicht auf Geschichte? Geschrieben oder überliefert! – Und Geschichte muß doch wohl allein auf Treu Und Glauben angenommen werden? – Nicht? –*

Lessing, *Nathan der Weise* 

The antiquarian is a stock figure in the drama of historical research: the one whose inspection of objects in the present produces insight into affairs of the past.<sup>1</sup> This expertise is by definition sensory-empirical, direct and unmediated. The one who receives and transmits stories about ancestral deeds is a different character: the storyteller, the bard. The antiquarian claims to circumvent all the links in the chain of transmission and directly access the past by means of its material remains. The procedure enables leaps beyond text and tradition. 'These antiquities are so exceeding old', said John Aubrey on the megalithic monuments of England, 'that no books do reach them: so that there is no way to retrieve them but by comparative antiquity, which I have writ upon the spot, from the monuments themselves.'<sup>2</sup>

Monuments represent not only their own times, but also their futures. So it was that Dürer's 1498 *Apocalypse* could portend, if not foresee, the Reformation; and David's 1785 *Oath of the Horatii* could anticipate the Tennis Court Oath of 1789. This drama has a different protagonist: the preternaturally sensitive artist, whose work registers 'a very early change in barometric pressure', a shift in the world spirit, and thus unwittingly generates prophecy.<sup>3</sup>

<sup>1</sup> For this definition of the antiquarian, see Anderson, 'Forgetting Athens', 186–189. For the archaeophile, a close relative, see Rojas, *Pasts of Roman Anatolia*, 18–19.

<sup>2</sup> Quoted in Miller, 'Not for lumpers only', 211.

<sup>3</sup> Haskell, *History and its Images*, 389–430, the quote at 415.

This chapter is about a third character who, while distinct from the first two, is kin to both: the antiquarian of the future, whose inspection of ancient monuments yields knowledge, not of ancient times, but of those to come. Such were the medieval interpreters of the Forum of Theodosius in Constantinople, in which a spiral relief column and equestrian statue were built to celebrate that emperor's defeat of a usurper, and in which some medieval Constantinopolitans saw a depiction of their city's future. As we shall see below, an anonymous source of the tenth century (the *Patria*) recorded that the Forum depicted a future defeat, a prophecy whose truth witnesses of the Fourth Crusade (1204) confirmed.

What kind of knowledge does the antiquarian of the future produce? Is it fundamentally similar or different to the antiquarian's knowledge of the past? We will consider this question by comparing the *Patria* to a second tenth-century description of the Forum of Theodosius, one written by Constantine of Rhodes and firmly focused on the past. But it may help first to sketch some possible responses on a more general level.

Let us start with an argument for the fundamental difference of the two pursuits. Monuments made in the past are direct indices of events in the past (above all, but not only, those of their own making), and may also convey knowledge (conscious and unconscious) from their makers and patrons about the time and place in which they were made. However, they cannot convey knowledge about a future far removed from the time of their facture and unknown to their makers. To think that they could requires an irrational belief in magic or (less judgmentally) an ontology different from that of the scholar, one in which time does not work in the same way as it does for us.

An argument for similarity may begin from the same principle – namely, that objects transmit knowledge about the past through two kinds of sign. The index is the direct physical trace of a past action, and thus by definition cannot record the future. Symbols, by contrast, transmit two distinct kinds of knowledge: intentional messages, which the object's makers and patrons meant to convey, and evidence legible only to the historian, which the object's makers and patrons unconsciously revealed.<sup>4</sup> Now, historians regularly assume that knowledge of future events was present in potential (if rarely active) in the experience of historical subjects. For example: 'if any man is to be held responsible for the disaster of 1204, it is Manuel Comnenus', the Byzantine emperor who died in 1180.<sup>5</sup> By the same logic, the symbols borne by past monuments may reveal the causes of future events.

Much has been written on the prophetic monuments of medieval Constantinople.6 However, the most explicit attempt to explain the nature of this knowledge, the relationship between ancient sculptures and future events, remains that of Richard

<sup>4</sup> 'Just as the degree of politeness in lifting a hat is a matter of the will and consciousness of the person doing the greeting, but it is not in his power to control what message about his innermost nature others may take from his gesture, so likewise even the artist knows only "what he parades" but not "what he betrays".' Panofsky, 'Describing and interpreting', 479–480.

<sup>5</sup> An anoymous 'verdict' cited at two removes: Magdalino, *Empire of Manuel Komnenos*, 4.

<sup>6</sup> Note, *e.g.*, Kraft, 'Prophecies'; Jouette, 'Divination'; Anderson, 'Classified knowledge'.

Dawkins, published in 1924. Dawkins adopts the argument for fundamental difference outlined above. Medieval Constantinopolitans had an 'incorrect' knowledge of 'ancient statues and buildings,' just as 'incorrect as their notions of [the monuments'] origin and nature'.<sup>7</sup>

Ancient structures … were rather regarded … as the works of powerful magicians of the days of heathendom; days of which the legend had remained that men were then more learned and had more mastery over nature than Christians had, or perhaps ought to have … The men of old were regarded as wonder-workers, magicians, doctors, and sages, and the works they left had been endowed by their arts with all sorts of magical powers and gifts, and could give lessons in wisdom to such as were able to read their meaning.<sup>8</sup> […]

The power with which the statues or other works were gifted appeared in various ways. It will be convenient to begin with cases of the simplest kind, where we find that a statue or inscription is believed to embody a prophecy which could be read by the understanding spectator.<sup>9</sup>

Dawkins then considers, as his first example, the anonymous tenth-century account of the Forum of Theodosius. He thus places the knowledge that it claims explicitly in the realm of the storyteller, tracing its source to a 'legend'.

We too shall begin with the same account, but will entertain a different hypothesis: namely, that it rests upon empirical inspection of the monuments, not upon tradition. In particular, it responds to the Forum's representations of the losing side in a civil war. In this respect, it is closer to the antiquarian's knowledge than to the bard's. However, a past-oriented antiquarian of the tenth century viewed the same images very differently. Where the anonymous author of the *Patria* saw defeated Romans of the future, the poem of Constantine of Rhodes saw instead ancient barbarians.

Many other Constantinopolitan monuments were subject to competing interpretations. The particular interest of the Forum of Theodosius is that the dispute was settled already in the middle ages. Crusader historians report, namely, that the residents of Constantinople saw the sack of their city by the Franks in 1204 as the fulfillment of the prophecies recorded on the column.

There is, then, a fundamental difference between Constantinople's medieval antiquarians of the past and those of the future, but it does not oppose empirical inspection to received wisdom, reason to tradition. Rather, it is a question of verifiability. Hypotheses about the past cannot be verified by experience, but hypotheses about the future can be, as the pictures in the Forum of Theodosius were.

#### **The** *Patria*

The first of the two tenth-century descriptions, which saw in the Forum of Theodosius a record of future events, appears in the *Patria Konstantinopoleos*, 'antiquities of

<sup>7</sup> Dawkins, 'Ancient statues', 213–214.

<sup>8</sup> Dawkins, 'Ancient statues', 217.

<sup>9</sup> Dawkins, 'Ancient statues', 218.

Constantinople'. This compendium of various (some much earlier) notices on the monuments of the city was assembled by 'a single, anonymous author' in 989/90.10 In the following we are concerned with a single notice 'On the Tauros', the familiar name for a monumental complex of the fourth century.

The emperor Theodosius (r. 379–395) defeated his erstwhile co-ruler, Maximus, in a civil war. The second imperial forum of Constantinople, built to the west of the Forum of Constantine, commemorated his victory.<sup>11</sup> Both fora straddled the Mese, the capital's primary east–west artery. The points at which the street entered and exited the Forum of Theodosius were articulated by monumental arches. Also like the Forum of Constantine, the Forum of Theodosius boasted a lofty column atop which stood a statue of the imperial honoree. But the Forum of Theodosius simultaneously hearkened back to the second-century Forum of Trajan in Rome. Its column, like Trajan's, was encircled by a continuous spiral relief depicting military campaigns. Also like the Forum of Trajan, the Forum of Theodosius featured an equestrian statue of the emperor atop a massive plinth.

This is how the Forum of Theodosius is seen by a twenty-first century archaeologist. Let us now consider a description of the same monuments written in the tenth century, over five hundred years after their construction. I quote the account of the *Patria* in full, in the translation of Albrecht Berger:

A statue of Theodosius the Great, which was formerly silver, stands in the Tauros where he used to receive those who came from the foreigners. Formerly, palaces and a hostel of the Romans were there, that is, at the place called 'the threshing floor'. Theodosius is installed atop the big column. His sons are above the lofty great quadruple columns: Honorius stands on the stone arch to the west, Arcadius on the stone arch to the east. In the middle of the courtyard is a huge equestrian statue, which some people call Joshua son of Nun, others Bellerophon. It was brought from Antioch the Great. The four-sided stone-cut plinth of the rider has relief narratives of the final days of the city, of the Rus who will conquer this city. And that impediment, namely the very short man-shaped bronze object tied kneeling under the left foot of the huge horse, signifies the same as that which is depicted there. Similarly, both the huge, hollow column there and the Xerolophos [that is, the Forum of Arcadius] have the story of the final days of the city and its defeats depicted as reliefs.<sup>12</sup>

This passage consists both of direct observation and of hearsay: things seen in the monuments, and things heard from other people. For example, the claim that the Forum occupied the former site of state facilities (palaces and hostel) likely rests on tradition, not observation of the material remains. Mention of a threshing floor, by contrast, evokes a smooth expanse, be it of packed earth or of paving stones. It helps to sketch the forum in plan, just as the following remarks construct its elevation. The statues, namely, appear in order from highest to lowest: first the statue of Theodosius atop its lofty column, then the statues of Arcadius and Honorius atop the two gates, and finally the equestrian statue atop its plinth.

<sup>10</sup> Berger, *Accounts of Medieval Constantinople*, xi-xii.

<sup>11</sup> For collection and analysis of the written and archaeological sources, see Mayer, *Rom ist dort*, 130–143; for further analysis, see Anderson, 'Forum of Theodosius'.

<sup>12</sup> Berger, *Accounts of Medieval Constantinople*, 82–83 (translation slightly modified).

*Fig. 7.1: Remains of one gate of the Forum of Theodosius, Beyazıt, Istanbul, as exposed in 1969. Deutsches Archäologisches Insitut Negative No. D-DAI-IST-69-3 (W. Schiele).*

The account of the *Patria* fills these broad outlines with directly observed realia, such as the 'quadruple columns'. Each of the forum's gates was anchored by four quartets of ten-metre tall columns carved in the form of cypress clubs, remains of which are still visible *in situ* near the tram stop at Beyazıt (Figure 7.1).<sup>13</sup> By contrast, the identities of the individual emperors may be based on inscriptions found within the forum, or on traditions regarding the place, written or oral, but not direct observation of the statues. The portrait types of late Roman emperors are so similar as to make differentiation difficult even when viewed by an expert up close: for example, the head of a member of the Theodosian dynasty found in Beyazıt and now in the Istanbul Archaeological Museum (Figure 7.2).<sup>14</sup>

*Fig. 7.2: Portrait of a Theodosian emperor, discovered in Beyazıt, now in the Istanbul Archaeological Museum (Inv. 5028 T) (B. Anderson).*

<sup>13</sup> Naumann, 'Neue Beobachtungen'.

<sup>14</sup> Stichel, *Römische Kaiserstatue*, 39 (difficulties of differentiating) and 51–52 (Beyazıt head).

*Fig. 7.3: Joshua (at far left) commands sun and moon to stand still; the Israelites rout the Amorites (at centre); the kings of the Amorites flee, as Joshua hears from two soldiers (at right). Biblioteca Apostolica Vaticana, Pal. gr. 431, fol. 13r (© Biblioteca Apostolica Vaticana).*

The author's primary focus is the equestrian monument, which receives four full sentences. The first registers the statue's exceptional scale, then turns to its subject. The *Patria* never considers that the rider might be Theodosius, for reasons to be discussed below. The two possibilities are cited as hearsay: some people call the rider Joshua, with others call him Bellerophon. At first this reads as a structural opposition between two heroes: one pagan, the other biblical. But both options are specific and respond to visible aspects of the statue.<sup>15</sup>

The choice between Bellerophon and Joshua was still debated centuries later, even after the statue was destroyed by the Crusaders in 1204. Writing in Nicaean exile, the historian Niketas Choniates supplied the visual basis of the biblical identification (here as translated by Harry Magoulias):

others contended that it was Joshua, the son of Nave, mounted on the horse, and that this was evident from the composition of the figure which extended his right hand in the direction of the chariot-driving sun and the moon's course to stay them in their procession …16

The reference is to Joshua 10:12–13, the triumph of the Israelites over the Amorites at Gibeon. Joshua's command to sun and moon appears in the tenth-century 'Joshua Roll', complete with plumed helmet, spear, and outstretched arm (Figure 7.3).<sup>17</sup>

Unlike Joshua, Bellerophon was conventionally depicted on horseback. For Choniates, the identification of the steed came first, and its rider followed as a consequence:

The horse was no less a wonder because of the perfection of its art; it was depicted without trappings, snorting and leaping with its ears pricked up as though at the sound of the war

<sup>15</sup> Stichel, 'Bellerophon oder Josua', 725.

<sup>16</sup> Magoulias, *City of Byzantium*, 353, cf. 358; Van Dieten, *Historia*, 643 and 649.

<sup>17</sup> Weitzmann, *Joshua Roll*, 27.

trumpet. Some claimed that the horse was Pegasus and its rider Bellerophon …<sup>18</sup>

Thus while interpretation as Joshua focused on the rider, interpretation as Bellerophon focused as much on the steed. It may also have invoked the vanquished foe under the horse's hoof (on which further below), just as the Chimera conventionally appeared in depictions of Bellerophon's triumph.<sup>19</sup>

The second sentence in the *Patria*'s account of the equestrian monument addresses the statue's provenance: 'It was brought from Antioch the Great.' This can only be hearsay, and serves at least two functions.20 First, it speaks to a broader interest in the cities of antiquity that

*Fig. 7.4: Members of the imperial guard of Theodosius on the northwest side of the obelisk base in the Hippodrome, Istanbul (Deutsches Archäologisches Institut Negative No. D-DAI-Z-NEG-9399. Photographer unknown).*

supplied Constantinople with statues. Second, it introduces a note of contemporary relevance. Antioch had been reconquered by the Byzantine army in 969, two decades before the composition of the *Patria*.

The third sentence shifts from the statue to its plinth, on which appear 'relief narratives of the final days of this city, of the Rus who will conquer this city'. Of the few preserved bases for Roman equestrian statues, none are decorated with reliefs.<sup>21</sup> But we know of reliefs on column bases, and, very close to home, the reliefs on the base of the Egyptian obelisk erected by Theodosius in the Hippodrome of Constantinople. The latter suggests an empirical source for the *Patria*'s identification of Rus on the base of the equestrian statue: the emperor's fine-haired, be-torqued guardsmen (*e.g.*, Figure 7.4). They are understood by archaeologists today as 'Germanic',22 in the tenth century perhaps simply as northerners.<sup>23</sup>

The fourth and final sentence about the equestrian monument returns from plinth to statue, more precisely to the bound captive ('the very short man-shaped bronze object tied kneeling') under the horse's hoof. This figure, the author asserts, depicts the same thing as the reliefs of the base. The captive's diminutive size and submissive posture recall the subdued foe beneath the emperor's horse on the Great Trajanic Frieze (Figure 7.5). Such a captive once stooped beneath the equestrian statue of Marcus Aurelius in Rome.<sup>24</sup>

<sup>18</sup> Magoulias, *City of Byzantium*, 353; Van Dieten, *Historia*, 643.

<sup>19</sup> Stichel, 'Bellerophon oder Josua', 726.

<sup>20</sup> The tradition repeats at *Patria* I.62, Berger, *Accounts of Medieval Constantinople*, 36–37.

<sup>21</sup> Højte, *Roman Imperial Statue Bases*.

<sup>22</sup> Mayer, *Rom ist dort*, 124.

<sup>23</sup> Books VIII and IX of the *History* of Leo the Deacon supply an account, precisely contemporary with the *Patria*, of a Byzantine campaign against Rus (Scythians, Tauroscythians) and other northerners.

<sup>24</sup> Kinney, 'Horse', 377.

*Fig. 7.5: Segment of the 'Great Trajanic Frieze' reinstalled in the passage of the Arch of Constantine, Rome (Deutsches Archäologisches Institut Negative No. D-DAI-ROM-37.328).*

*Fig. 7.6: Surrender of the Guard of Maximus, relief fragment from the Column of Theodosius now built into the foundations of Beyazıt Hamam, Istanbul (Deutsches Archäologisches Institut Negative No. D-DAI-IST-R6408. U. Peschlow).*

This is an account of direct observation, although the captive would soon disappear from view and become a topic of hearsay instead. Choniates wrote of the same figure in the thirteenth century:

There was an ancient tradition which came down to us and which was in the mouths of all, that under this horse's front left hoof there was buried the image of a man which, as it had been handed down to some, was of a certain Venetian; others claimed that it was of a member of some other Western nation not allied with the Romans, or that it was a Bulgarian. As the attempt was often made to secure the hoof, the statue was completely covered over and hidden from sight. When [in 1204] the horse was broken into pieces and committed to the flames, together with the rider, the statue was found buried beneath the horse's hoof; it was dressed in the kind of cloak [*chlaina*] that is woven from sheep's wool.<sup>25</sup>

This is a fascinating passage, not least because the captive's rediscovery fails to resolve his identity. The account of the *Patria*, composed before the captive disappeared behind wedges and struts, is equally generic: 'the very short manshaped bronze object'. However, it does specify that this object 'signifies the same as that which is depicted there', that is, on the reliefs of the base. In other words, the *Patria* sees both base and statue as representations of Roman defeat.

Without the statue, we cannot fully understand this claim. But the final sentence helps, by asserting that the spiral relief columns of Theodosius and of his son, Arcadius, likewise showed the defeats of the city. Fragments of the

Column of Theodosius are preserved in the Istanbul Archaeological Museum and in the foundation of Beyazıt Hamam. Here we see Roman soldiers, partisans of the defeated usurper Maximus, in attitudes of surrender (*e.g.*, Figure 7.6).26 Perhaps the author of the *Patria* saw something similar in the bound captive beneath the horse's hoof.

<sup>25</sup> Magoulias, *City of Byzantium*, 358; Van Dieten, *Historia*, 649.

<sup>26</sup> Speidel, 'Garde des Maximus'.

Everywhere in the Forum of Theodosius – the base of the equestrian statue, the statue itself and the reliefs of the column – the *Patria* sees Roman defeat. This perception explains the author's refusal to make the obvious identifications: Theodosius on the horse, his victories on the column. We need only supply the underlying principle that no state commemorates its own subjection. Thus results the conclusion that the reliefs do not commemorate past actions of the Romans, but foreshadow instead their future defeats.

The *Patria*'s account of the Forum of Theodosius contains both hearsay and direct observation. However, its identification of the statue and column as representations of the future rests not on hearsay – a belief in the wisdom of the ancients – but on direct observation of the Forum's representations of the vanquished.

#### **Constantine of Rhodes**

An interpretation of the Forum of Theodosius as a testimony to past events was also available in the tenth century. Several decades before the *Patria*, in the first half of the tenth century, Constantine of Rhodes, civil servant and imperial favorite, composed in verse an 'account of the Church of the Holy Apostles and a partial description of the statues of [Constantinople] and its tall and very great columns'.27 The poem devotes some forty lines to the Forum of Theodosius. Constantine's account is very different from that of the *Patria*. He identifies the rider as Theodosius, and the vanquished as barbarians of the past.

Constantine does not evoke the totality of the forum. He omits its paved expanse and twin gates, moving directly to the column and equestrian statue.28 He is unambiguous in his identification of the rider (here in the translation of Liz James):

… this best and great horseman Theodosius, standing there, the marvellous man, there on the topmost step of the great street …29

As for the column, he writes that Theodosius's son,

that famed Arcadius set [it] up long ago glorifying all his father's prowess and his trophies and incomparable battles … For in pictures especially well-arranged and well-carved to beautiful effect, it presents on all sides all manner of slaughter of barbarians and Scythians and their cities destroyed forever.30

<sup>27</sup> James, *Constantine of Rhodes*, 18–19 (title) and 131–144 (author and date).

<sup>28</sup> 'The *Patria* focuses on places; Constantine of Rhodes is concerned with columns and sculptures.' James, *Constantine of Rhodes*, 169.

<sup>29</sup> James, *Constantine of Rhodes*, 34–35.

<sup>30</sup> James, *Constantine of Rhodes*, 32–33.

Constantine sees here, as scholars see now in the fragments of the column, a historical representation. But his view also diverges from that of contemporary archaeologists, who see there Romans defeating Romans. For Constantine, by contrast, the column shows Romans defeating Scythians. And this places him in diametric opposition to the author of the *Patria*, who sees Rus defeating Romans.

Like the *Patria*, Constantine's poem draws upon both hearsay and observation. Constantine repeatedly attributes erection of column and equestrian statue to Arcadius, not his father. For this he must rely on some tradition no longer available to us. Constantine's most obvious direct observations relate to the technical aspects of the monuments. He delights in describing his own ascent of the spiral stair inside the column, and devotes particular attention to the skillful rendering of the horse:

Anyone seeing the horse and its violent snorting, frozen in bronze by the force of the sculpted art, bristling its mane and tossing its hair and champing at the bit in its eagerness, thinks that it holds out its neck like a great tower in its most haughty and marvelous snorting and that he expects its hoof to move, and that the horse is perhaps neighing and is alive, bearing its victorious lord …31

At the same time, and despite this focus upon the horse, Constantine entirely omits the bound captive beneath its hoof.

Thus the two accounts are very different, but not because one relies on tradition and the other on observation. Rather, both interpret a key, visible aspect – the depictions of vanquished soldiers on the column – in opposite fashion.

Both Constantine and the *Patria* accept that no state commemorates its own subjection. Constantine has read the historians, and he knows that the Forum depicts 'the rebellion of Maximus'.<sup>32</sup> But when he turns to the dead, he sees only 'slaughter of barbarians and Scythians';<sup>33</sup> and again, 'the slaughter of Scythians and the butchering of barbarians'.34 The authors of the *Patria*, by contrast, turn the vanquished into Romans of the future.

At stake in this distinction is the visual differentiation of the Roman army from its adversaries. It is a question primarily of visual conventions and the expectations that they create in viewers, secondarily therefore of ideology.

Antonine triumphal monuments, such as the Columns of Trajan and Marcus Aurelius, maintained clear distinctions of dress and grooming between Roman soldiers and their barbarian foes. Also the narrative frieze of the Arch of Constantine, which in fact depicted a civil war between Romans, clad Constantine's soldiers in

<sup>31</sup> James, *Constantine of Rhodes*, 34–35.

<sup>32</sup> James, *Constantine of Rhodes*, 34–35.

<sup>33</sup> James, *Constantine of Rhodes*, 32–33.

<sup>34</sup> James, *Constantine of Rhodes*, 36–37.

armor formally distinct from that of Maxentius's troops.<sup>35</sup> Such distinctions were abandoned in fourth- and fifth-century Constantinople, whose monuments made no visible distinction between victor and vanquished, with the result that Romans were depicted in attitudes of subjection.36

Both visual conventions – distinction and resemblance of friend and foe – were maintained into the tenth century. Resemblance was maintained especially in the depiction of sacred history. Consider, for example, manuscript paintings of battles between the Israelites and their various adversaries.

In the Joshua Roll, the helmets and armor of the victorious Israelites are identical in form to those of the routed Amorites, who are labeled generically as *allophyloi*, 'foreigners' (Figure 7.3). The two sides are distinguished only through color. The duotone shields of the Amorites contrast to the monotone shields of the Israelites; and while soldiers on both sides wear broad bands across their chests, only those of the Amorites are colored blue.<sup>37</sup>

The Paris Psalter, another product of the tenth century, also confronts Israelite troops to foreign (*allophyloi*, again), in its depiction of the duel between David and Goliath.<sup>38</sup> The distinction between monotone and duotone shields is maintained, while that between chest bands is abandoned.

These two tenth-century manuscripts help to understand Constantine's interpretation of the Forum of Theodosius. The painters distinguish foreign armor from native (the Israelites stand for the Romans in middle Byzantine political thought) on the basis of color, not form. So too might Constantine justify his distinction between two identically clad armies, especially as the painted details of the relief will have long since faded.

A very different visual convention, one of absolute distinction between soldier and foe, persisted in late Roman and Byzantine depictions of mounted warriors, be they Solomon or saints.<sup>39</sup> These were especially common on amulets, and staged an opposition between male human saint (first Sissinios of Antioch, then George and Theodore) and the prey whom his steed tramples, and whom he often spears for good measure. The vanquished are animal, demonic, female or indeed (like the Chimera) all three. Even when a male human foe is depicted, the artists distinguish the form of his dress from that of the riders. Consider, for example, the headdress of the vanquished enemy on the medieval relief from Amaseia today in the Benaki Museum.40

As we have seen, the account of the *Patria* devotes close attention to the trampled foe beneath the equestrian statue in the Forum of Theodosius, a figure whom Constantine passes over in silence. The *Patria* considers it a representation of Roman defeat, 'the same as that which is depicted' on the base. Whatever visual distinction it

<sup>35</sup> Zanker, 'Konstantinsbogen', 88–90.

<sup>36</sup> Mayer, *Rom ist dort*, 150–160.

<sup>37</sup> On the bands, see Grotowski, *Arms and Armour*, 280, fn. 589.

<sup>38</sup> Paris, BNF, Par. Gr. 139, f. 4v.

<sup>39</sup> Grotowski, *Arms and Armour*, 74–85.

<sup>40</sup> For the Amaseia relief, see Grotowski, *Arms and Armour*, 307–308.

might have registered between victor and foe was too slight to satisfy the expectations of the anonymous author.

It is helpful now to revisit Choniates's account of the fate of this figure. Visible in the tenth century, it was gradually obscured by supports meant to secure the horse. The memory of the captive persisted, however, as did a dispute regarding its identity: 'the image of a man which, as it had been handed down to some, was of a certain Venetian; others claimed that it was of a member of some other Western nation not allied with the Romans, or that it was a Bulgarian'.41 This passage describes a spectrum from near-identity (until 1204, the Venetians were Roman allies, even subjects) to substantial otherness (Bulgarian or Frank).

Clearly the captive (Roman or not) was as much a topic of dispute as the rider (Bellerophon or Joshua). Its ambiguity was especially disturbing in the context of a motif, the mounted warrior trampling his foe, that conventionally maintained an absolute visual distinction between opponents. The captive thus assumes a central role in the account of the *Patria*, and its interpretation of the complex as a whole as a representation of future events.

Constantine's poem and the *Patria* both rely on a mixture of hearsay and direct inspection in their accounts of the Forum of Theodosius. However, when it comes to direct inspection, they employ distinct approaches to the interpretation of visual evidence. Constantine, writing a poem in praise of emperors past and present, passes over disagreements about identification and sees the Forum as an unambiguous boast of triumph over foreign nations.42 The *Patria*, by contrast, foregrounds the disagreements and refuses to ignore the ambivalence of the bound captive. Constantine is thus the more willful interpreter, the *Patria* more faithful to the evidence. This fidelity underlies the view of the Forum as a representation of future defeats.

## **Robert of Clari and Gunther of Pairis**

In fact, the experience of Constantinopolitans verified the *Patria*'s interpretation of the Column of Theodosius as a depiction of the future. Their city did fall to northerners in 1204, just as the reliefs had shown. Two such foreigners, Robert de Clari and Gunther of Pairis, report that, as the sack occurred, the residents of the city returned to the Column of Theodosius (thus Gunther and Robert) and to that of Arcadius (thus Robert) and confirmed the truth of their images.<sup>43</sup>

Gunther never visited Constantinople, but wrote his history (an artful combination of Latin prose and verse) in the abbey of Pairis, on the basis of his abbot's eyewitness reports. Robert, by contrast, participated in the sack, and wrote about it in his

<sup>41</sup> Magoulias, *City of Byzantium*, 358; Van Dieten, *Historia*, 649.

<sup>42</sup> For the imperial loyalty of Constantine's poem, see James, *Constantine of Rhodes*, 177.

<sup>43</sup> Macrides, 'Constantinople', 202–205. A third account, by Geoffrey of Villehardouin, also invokes prophecy, but more vaguely. On the Frankish historians of the Fourth Crusade, see now Demacopoulos, *Colonizing Christianity*. On their accounts of the spiral relief columns, compare Kraft, 'Prophecies'.

native French. Here is his account of the columns of Theodosius and Arcadius, in the translation of Edgar Holmes McNeal:

There are two columns … On the outside of these columns there were pictured and written by prophecy all the events and all the conquests which have happened in Constantinople or which were going to happen. But no one could understand the event until it had happened, and when it had happened the people would go there and ponder over it, and then for the first time they would see and understand the event. And even this conquest of the French was written and pictured there and the ships in which they made the assault when the city was taken, and the Greeks were not able to understand it before it had happened, but when it had happened they went to look at these columns and ponder over it, and they found that the letters which were written on the pictured ships said that a people, short haired and with iron swords, would come from the west to conquer Constantinople.<sup>44</sup>

Robert claims to transmit, not his own interpretation of the columns, but one generated by Constantinopolitans after the conquest. The claim is rendered plausible by Gunther, whose report of the Constantinopolitan reaction shares much in common with Robert's, including the emphasis on naval assault. For Gunther too (here in the translation of Alfred Andrea), the relevant images on the Column of Theodosius were the figures 'of ships, with ladders of a sort projecting from them, on which armed men were climbing. They seemed to be storming and capturing a city which was also sculpted there'.<sup>45</sup>

In fact, the Column of Arcadius, as we know it from a set of sixteenth-century drawings, depicted on its central windings a naval assault on a city (Figure 7.7). A preserved fragment of the Column of Theodosius likewise portrays soldiers at sea (Figure 7.8). There is however no preserved trace of Robert's 'letters … written on the pictured ships'.

Neither Gunther nor Robert knew Greek, much less the *Patria*. Indeed, both claim that the correct interpretation of the figures was obscure until the assault on the city began. Robert simply writes that 'the Greeks were not able to understand it before it happened.' Gunther's account is more complex. First, he acknowledges that the Column of Theodosius was already known to depict future events:

various representations of events since events since antiquity were sculpted on it, which are said to depict in sundry scenes the prophecies of a Sibyl, largely concerning their kingdom.46

However, Gunther continues, the residents of Constantinople chose to ignore what was shown there: 'Until that time the Greeks had disregarded the sculpture, thinking that nothing was less possible than that such a thing could ever befall a city such as their own.'<sup>47</sup> Gunther is writing about the column, but his claim recalls the omission of the captive by Constantine and its literal disappearance in later centuries. The

<sup>44</sup> McNeal, *Conquest of Constantinople*, 110–111; Lauer, *Robert*, 89.

<sup>45</sup> Andrea, *Capture of Constantinople*, 117; Orth, *Gunther*, 166.

<sup>46</sup> Andrea, *Capture of Constantinople*, 117; Orth, *Gunther*, 166.

<sup>47</sup> Andrea, *Capture of Constantinople*, 117; Orth, *Gunther*, 166.

*Fig. 7.7: Scenes of naval combat on the east side of the Column of Arcadius. 'The Freshfield Album', Cambridge, Trinity College, MS O.17.2, fol. 11 (detail) (Courtesy of the Master and Fellows of Trinity College, Cambridge).*

*Fig. 7.8: Two shipboard soldiers, relief fragment from the Column of Theodosius now built into the foundations of Beyazıt Hammam, Istanbul (Deutsches Archäologisches Institut Negative No. D-DAI-IST-2565. Photographer unknown).*

account of the *Patria*, by contrast, shows that not all of 'the Greeks' participated in this suppression.<sup>48</sup>

The accounts of the *Patria* (in the tenth century) and of Choniates, Gunther and Robert (all writing in the thirteenth) demonstrate remarkable continuity in the interpretation of the fourth-century Forum of Theodosius. One such continuity is the identification of the rider as Bellerophon or Joshua, which appears first in the *Patria c*. 990, then again post-1204 in the history of Choniates. A second is the perception of future defeats on the reliefs of the columns, claimed by the *Patria* then enacted in the histories of Gunther and Robert.

It is one thing to note such continuities in local interpretation, another to claim that the events of 1204 confirmed the interpretation of the *Patria*. The *Patria* is primarily concerned with the equestrian monument, especially the reliefs on the plinth and the bound captive beneath the horse's hoof, and mentions only in passing that the Columns of Theodosius and Arcadius also depict the future. Gunther and Robert, by contrast, mention only the depictions of the naval battles on the columns. While the Crusader historians share with the *Patria* a general concern with conquest and columns, perhaps the objects of their specific interest are merely proximate, not meaningfully similar.

Alternatively, perhaps both the tenth- and the thirteenth-century accounts respond to the same underlying phenomenon (the depiction of defeated Romans on the Theodosian monuments of Constantinople), from which they draw the same conclusion (that the city was destined to fall). The shift in attention to the naval battles would then be a refinement of the hypothesis to accommodate new data (the arrival of the Crusader ships in the Golden Horn). The ultimate conclusion would

<sup>48</sup> Compare Kraft, 'Prophecies', *e.g*. at 100–101, distinguishing between 'Byzantine' and 'Latin' interpretations of the Forum.

be the same: the depictions of defeated Romans on the works of the city's founders foretold its eventual defeat.

Finally, the thirteenth-century accounts retrospectively illuminate the difference between Constantine, an antiquarian firmly focused on the past, and the *Patria*'s antiquarian of the future. It is not a difference between the popular and the intellectual, superstition and reason.49 The key distinction, rather, is one neatly expressed by R. G. Collingwood:

the truths which science discovers are known to be true by being found exemplified in what we actually perceive, whereas the past has vanished and our ideas about it can never be verified as we verify scientific hypotheses.50

The claims that historians and antiquarians make about the past do not admit empirical verification; the claims that the *Patria* made about the future of Constantinople, by contrast, did.

For Dawkins, the *Patria* and its kin exhibit little 'scientific' interest, 'or rather their appreciation of ancient statues and buildings as objects of knowledge was as incorrect as their notions of their origin and nature'.51 This chapter, by contrast, has sought to understand the *Patria*'s account of the Forum of Theodosius not only as scientific (verifiable), but also as correct (verified).

### **Acknowledgements**

Thank you to the editors and organisers for the invitation to contribute and for helpful advice; thank you to Ine Jacobs for references and discussion.

## **Bibliography**


<sup>49</sup> *Contra* Mango, 'Antique statuary', 59: 'Byzantine attitudes towards ancient statuary should be considered on two levels: the popular and the intellectual.'

<sup>50</sup> Collingwood, *Idea of History*, 5.

<sup>51</sup> Dawkins, 'Ancient statues', 213–214.


## Chapter 8

## William of Tyre and the cities of the Levant

## *Sam Ottewill-Soulsby*

When the armies of the First Crusade set off on their expedition to the Holy Land in 1096, they were animated by the thought of one ancient city, Jerusalem, holiest of cities.1 But as they left familiar lands and passed into those of the Eastern Mediterranean, they found not one city, but many. Their extraordinary journey was punctuated by encounters with cities, most notably the traumatic sieges of Nicaea and Antioch. Even after the capture of Jerusalem in 1099, the Crusaders were not finished with cities. The survival of the fledgling Frankish Kingdom of Jerusalem depended on the rapid acquisition of centres like Jaffa, Acre, Beirut, Sidon and Tyre, cities that were only dubiously part of any previous conception of the Holy Land, but were essential for maintaining communication and trade with the wider Latin world.2 The castles of the Latin East may be famous today, but it was the cities that were the true bulwark of the kingdom. Rural military elites in the West found themselves converted into urbanites in the East, as the majority of the new Frankish migrants who came to Outremer3 resided in the great cities of the kingdom.<sup>4</sup> As the capital of the young realm, a huge building program took place in Jerusalem.<sup>5</sup>

Given all this, it is hardly surprising that cities featured prominently in the explosion of historical writing that appeared after the First Crusade. Participants such as Fulcher of Chartres and the anonymous author of the *Gesta Francorum* included descriptions of cities amid their accounts of the things they had seen.6 But of all

<sup>1</sup> Although see the doubts of Cheveddin, 'Canon 2 of the Council of Clermont'.

<sup>2</sup> Murray, 'Sacred space and strategic geography'.

<sup>3</sup> On the use of the term Outremer, Morreale and Paul, *The French of Outremer*, 1. 4 Murray, 'The demographics of urban space'; Pringle, 'Churches and settlement in Crusader Palestine'. Although see the useful corrective in Ellenblum, *Frankish Rural Settlement.* 

<sup>5</sup> Hamilton, 'The impact of Crusader Jerusalem on western Christendom', 699–700.

<sup>6</sup> *Gesta Francorum* 10.32, 76; Fulcher of Chartres, *Chronicle of the First Crusade*, 15.1–4, 41–42; Albu, 'Antioch and the Normans'.

the written work that emerged in the Kingdom of Jerusalem, none is quite so filled with cities as the enormous *Chronicle* written by William of Tyre (*c*. 1130–*c*. 1186), the greatest of the historians of the Latin East.<sup>7</sup> As Wolfgang Giese has argued, William's *Chronicle* is obsessed with cities and his account routinely pauses to allow for a description and history of individual cities.8

In addition to these shorter insertions, William also included three great setpieces on the cities of Antioch, Jerusalem and Tyre, stopping his narrative in order to give them more extended treatments. Of these, the description of Tyre is the most spectacular. William begins by discussing Tyre's appearances in classical sources, addressing its origins and the characteristics of its inhabitants, their commercial activities and links with the Carthaginians. William then turns to Tyre's role in the bible and as the death place of the Christian Father Origen. Finally he describes the landscape around Tyre and the great buildings of the city in loving detail.9

William's cities punctuate both his history and his landscape. They are grand, proud and almost entirely ancient. A reader of his description of Tyre might reasonably suppose that nothing had changed there since the second century. Throughout his history William jumped cheerfully from the ancient world to his present, ignoring the way in which Eastern Christians and Muslims had shaped the cities he wrote about. Indeed, there is barely any acknowledgement that these cities were ever controlled by Muslim rulers. In writing his history, William was seeking to place himself and other Franks in the East within a longer context than just the aftermath of the First Crusade. Antiquarianism alone did not drive William to write about the ancient city. Rather his attention to the ancient layers of these cities, at the expense of the very visible and more recent periods, served as part of William's efforts to create a deep history for Latins living in Outremer.

#### **A Hundred Acres**

An example of William's focus on the ancient city is provided by his description of Acre:

A city on the seacoast in the province of Phoenicia, one of the suffragan cities which is under the metropolis of Tyre. Its double harbour, lying both inside and outside the walls, offers a safe and tranquil anchorage to ships. It is well located between the mountains and the sea and possesses wide domains with rich and fertile fields. The river Belus flows past the city. According to the story generally received, it was founded by two brothers, Ptolemais and Acco. They fortified it with walls of solid masonry and divided it into two sections, which were called from their names. Hence even today it is known indifferently as Ptolemais or Acre, as is the case with most of the cities of Syria, each of which has, as a rule, two or even three names.10

<sup>7</sup> William of Tyre, *Chronicon*.

<sup>8</sup> Giese, 'Stadt-und herrscherbeschreibungen bei Wilhelm von Tyrus'.

<sup>9</sup> William of Tyre, *Chronicon*, 13.1–5, 584–592.

<sup>10</sup> William of Tyre, *Chronicon*, 10.26, 485: 'Civitas maritima provincie Phenicis, una de urbibus suffrageneis que ad Tyrensem metropolim habent respectum, portum habens infra menia et exterius etiam, ubi

Perhaps the first thing to note about William's description of Acre is how dependent it is on written sources. Apart from a few generic notes on the walls, harbour, river and surrounding fields, there is little in his account of Acre that suggests that he drew much information from the physical structures of the city. This is striking given that in 1165 he was granted a prebend in the cathedral of Acre by the city's bishop, so would presumably would have spent some time in Acre.11 Acre is not unusual in this regard, as his picture of many of the other cities of the kingdom in his *Chronicle* is dependent on written material. William employed an extraordinary range of classical and biblical texts in reconstructing the past of the city. In his great set piece description of Tyre alone he drew upon Ulpian, Virgil, Ezekiel, Isaiah, Josephus, Jerome and Isidore, among others.12

William did not entirely disregard the physical landscape of cities he was familiar with. When discussing his native city, he professed himself to be surprised that both Solinus' *Polyhistor* and 2 Chronicles described Judea as being filled with springs, because in his Jerusalem the inhabitants depended upon rainwater and cisterns for their water.13 Likewise, when talking about Tyre, William notes 'this same city still guards the body of Origen, as may be proved by the testimony of one's own eyes', before supporting this by quoting Jerome.14 His description of the countryside by Tyre, including the way one can ride to the top of a tower of a nearby reservoir used to irrigate sugar cane ('a most precious product, very necessary for the use and health of mankind'), suggests personal experience or oral report.<sup>15</sup> His emphasis on written sources probably partly reflects unfamiliarity with many of the places discussed. Jerusalem and Tyre also receive two of the longest descriptions in the *Chronicle*, giving him more space to draw in details.

Acre serves as a useful example of William's city-writing because it featured in a number of travel accounts written by Muslims in the twelfth century. Their descriptions of the city are very different from William's. Among the details not mentioned by William is that while the original fortifications of Acre might have gone back to antiquity, the walls encircling it in his day had been raised by the Muslim ruler Ibn

tranquillam navibus possit prebere stationem. Est autem et satis commode sita inter montes et mare, pingue et opimum habens latifundium, Belo flumine preterfluente civitatem. Hanc frequens fama est geminos fratres fundasse Ptolomeum et Acconem et muris vallasse solidioribus eamque quasi per medium divisam suis nominibus appellasse. Unde et hodie binomia est, ut Ptolomaida dicatur et Accon, sicut pene omnes Syrie civitates duo vel tria habent nomina'; tr. Atwater Babcock and Krey, *A History of Deeds Done Beyond the Sea*, 453.

<sup>11</sup> Edbury and Rowe, *William of Tyre*, 15.

<sup>12</sup> William of Tyre, *Chronicon*, 13.1–5, 584–592.

<sup>13</sup> William of Tyre, *Chronicon*, 8.4, 389–390. On water supply in twelfth-century Jerusalem, see Boas, *Jerusalem in the Time of the Crusades*, 171–177.

<sup>14</sup> William of Tyre, *Chronicon*, 13.1, 586: 'haec eadem et eximii doctoris, Origenis videlicet, corpus occultat, sicut oculata fide etiam hodie licet inspicere', He quotes Jerome, *Ep.* 84.7. On medieval traditions of the tomb of Origen in Tyre, see Crouzel, *Origen*, 35.

<sup>15</sup> William of Tyre, *Chronicon*, 13.3, 589: 'preciosissima, usibus et saluti mortalium'. On sugar production in the kingdom of Jerusalem, see Ouerfelli, *Le Sucre*, 37–46.

Ṭūlūn in the ninth century.16 The Arab geographer and poet Ibn Jubayr, who visited Acre in 1185, noted that the city was dominated by the principal mosque that stood in its centre and was taller than any other building, although only a part of it was permitted to be used as a Muslim place of worship.<sup>17</sup> By the mosque lay the tomb of the pre-Islamic Prophet Ṣāliḥ, which was a major pilgrimage point even in Ibn Jubayr's day.

As a Christian archbishop, it is perhaps unsurprising that William does not mention Muslim religious sites in his account of Acre. It is quite clear throughout his writing that he was hostile to Islam, although not necessarily to individual Muslims.18 More interesting is William's silence concerning Acre's other important centre of pilgrimage, the Ox Spring, where legend had it that the oxen Adam put to the plough came from. William was normally very keen to list holy sites that could be connected to biblical figures.<sup>19</sup> The Persian traveller ʿAlī of Herat wrote about the Ox Spring in 1173, calling it sacred to Jews, Christians and Muslims alike, although his description is hostile to Christians. 20 Ibn Jubayr concurred, saying 'in the hands of the Christians its venerableness is maintained'.21 The spring only seems to have acquired these sacred connotations after the Arab Conquest.22 It might have been a site of great significance to Christians in the twelfth century, but to William it was still a product of the period of Muslim control, making it an innovation he disdained to comment on.

## **William's history**

William was born in Jerusalem in about 1130 to a family of burghers. There is reason to believe that both of his parents were also born in Jerusalem. Where his ancestors were from before then is unknown and seems to have been unimportant to William.23 He referred to himself throughout his life as a Latin, a proud son of the Kingdom of Jerusalem.<sup>24</sup> He was extremely well-educated, spending sixteen years studying the liberal arts and theology in the finest schools of Paris and Orleans, followed by another three years in Bologna learning canon law.25 Upon his return to Jerusalem in 1165 he was fast-tracked through the administration, becoming Chancellor in 1174 and Archbishop of Tyre in 1175. He was the only member of the higher clergy to have been born in the kingdom.26 Over his career he was sent as a diplomat to Constantinople and the West. In 1170 King Amaury I appointed

<sup>16</sup> Al-Muqaddasī, *Best Divisions for Knowledge of the Regions*, 138–139.

<sup>17</sup> Ibn Jubayr, *Travels of Ibn Jubayr*, 318.

<sup>18</sup> Marten, 'William of Tyre's attitude towards Islam', 13–23.

<sup>19</sup> See for example, William of Tyre, *Chronicon*, 8.1, 381–382; 19.14, 883–884; 21.8, 972.

<sup>20</sup> ʿAlī of Herat, *Guide des Lieux de Pèlerinage*, 57.

<sup>21</sup> Ibn Jubayr, *Travels of Ibn Jubayr*, 319.

<sup>22</sup> See the discussion in al-Ṭabarī, *History Vol 1*, 294, 299.

<sup>23</sup> Huygens, 'Editing William of Tyre', 461–473, 461.

<sup>24</sup> Murray, 'Ethnic identity in the Crusader states', 59–73, 61.

<sup>25</sup> Huygens, 'Guillaume de Tyr étudiant', 811–829.

<sup>26</sup> Murray, 'Prosopography and onomastics of the Franks', 283–294.

William tutor to his son and heir, Baldwin, whose leprosy William was allegedly the first to diagnose.27

William was working on his history throughout this time. The *Chronicle* itself starts with the First Crusade, prefaced by an account of the fall of Jerusalem to ʿUmar. It then continues until 1184, where it was discontinued, probably because of William's ill-health.28 His historical narrative becomes much more detailed from 1167, when he appears to have started taking notes. Prior to this he drew upon written sources and oral report, particularly the former which are much richer for the period prior to 1127, where the chronicle of Fulcher of Chartres ends.<sup>29</sup> He was at the heart of the running of the Kingdom of Jerusalem as he wrote. William attributed the idea for the history to King Amaury, who had relevant material sent to him.<sup>30</sup>

William seems to have begun his work with a local readership in mind. After William's death, the *Chronicle* circulated in Outremer, with a large number of manuscripts being produced in Acre in the thirteenth century.31 In the prologue to his *Chronicle*, he wrote that he was driven by 'an insistent love of my country'.32 Despite having had an education that would have afforded him employment in any court in the West, William chose to return to Jerusalem, and spent the rest of his life in its service. Alan Murray has convincingly portrayed William of Tyre's *Chronicle* as an *origo gentis*, in which the historical narrative serves to cement the Latins of the East as a national group. The First Crusade became the migration story of a people seeking a new home.33 William was seeking to provide a history for his people, giving them a sense of identity and justifying their presence in the Holy Land.

William's exclusion of the Eastern Christian and Muslim city was not due to ignorance on his part. Although he did not speak Arabic, he was formidably wellinformed about the Islamic world.34 He was commissioned by King Amaury to write a history of the Muslims, which does not survive, although it was later consulted by Jacques de Vitry. The king aided William in this endeavour by delivering a large amount of written sources to him, among them the tenth-century chronicle composed by Patriarch Eutychius of Alexandria, also known as Saʿīd b. Baṭrīq.35 Despite the amount of material contained in the chronicle on the late antique past of cities such as Tyre, none of it appears in William's history.36 That William knew quite a lot about cities in the Islamic world is demonstrated by his description of Cairo, which he admired for

<sup>27</sup> William of Tyre, *Chronicon*, 21.1, 961.

<sup>28</sup> On the date of William's death, see Hiestand, 'Zum leben und zur laufbahn Wilhelms von Tyrus', 345–380.

<sup>29</sup> Edbury and Rowe, *William of Tyre*, 26, 44.

<sup>30</sup> William of Tyre, *Chronicon*, Prologue, 99; 19.12, 882; 20.31, 957.

<sup>31</sup> Rubin, *Learning in a Crusader City*, 24–27.

<sup>32</sup> William of Tyre, *Chronicon*, Prologue, 99: 'Sed urgentissimus instat amor patrie'.

<sup>33</sup> Murray, 'Ethnic identity in the Crusader states', 68.

<sup>34</sup> Möhring, 'Zu der geschichte der orientalischen'; Murray, 'William of Tyre and the origins of the Turks'.

<sup>35</sup> William of Tyre, *Chronicon*, 100.

<sup>36</sup> Saʿīd b. Baṭrīq, *Naẓm al-Jawhar*, tr. Bredy, 101–103, 119–120.

its 'magnificent buildings'.37 His account of the history of the city is very accurate, identifying both the Fatimid general who founded the city and the Caliph in whose name he did so.38 If the way he chose to present the cities of his homeland in his History of Jerusalem was one that largely misses out their time as places under Muslim control, it was not from lack of knowledge about that period of history.

William was a master of the genre of urban description. His treatments show the influence of late antique manuals in rhetoric which emphasised discussing the origins of a city, its fortifications and the fertility of the landscape.39 This would have reached William via the teaching he received.40 His education in the most prestigious schools in the Latin world also placed him in the midst of the 'twelfth-century renaissance', an explosion of intellectual activity partly powered by a revival in interest in ancient texts. As the cities of western Europe rapidly expanded in size and complexity, educated people responded with a new range of writing about the city.41 In addition to standalone poetry, or speeches in favour of particular cities, urban descriptions were integrated into a wide range of other genres.42 Descriptions of cities also punctuate histories written in the twelfth century, such as William of Malmesbury's *Gesta Pontificum Anglorum* (1125) or Otto of Freising's *Gesta Friderici Imperatoris* (1157–1158).<sup>43</sup> William of Tyre was writing his history at a time when inserting set-piece descriptions of relevant cities was by no means unexpected in the Latin West.

Twelfth-century Latin descriptions often emphasised the ancient histories of cities.44 The common inheritance of both the classical and the biblical past to much of Western culture gave them a universal importance, which was useful for stressing the significance of the city to a wider audience. But engagement with the more modern past was not uncommon.45 In the early 1140s, Falco of Benevento placed his description of his native city within a chronicle concerned with recent events in Southern Italy.<sup>46</sup> At the end of the century, Boncompagno da Signa centred his account of Ancona on the city's resistance to being besieged by Frederick Barbarossa in 1173.47 Writers in England such as William of Malmesbury celebrated the urban developments carried out by Anglo-Saxon kings.48 In light of the willingness of other authors to discuss the more recent past of their cities, William of Tyre's apparent lack of interest in the post-Roman urban histories of his cities appears more striking.

<sup>37</sup> William of Tyre, *Chronicon* 19.15, 884: 'magnificis decoratum edificiis'.

<sup>38</sup> William of Tyre, *Chronicon* 19.15, 884–885.

<sup>39</sup> Kempshall, *Rhetoric and the Writing of History*, 2–12; Ruth, *Urban Honor in Spain*, 14–33.

<sup>40</sup> Morse, *Truth and Convention in the Middle Ages*, 8, 71, 86–109.

<sup>41</sup> Oldfield, *Urban Panegyric*.

<sup>42</sup> Hyde, 'Medieval descriptions of cities'.

<sup>43</sup> William of Malmesbury, *Gesta Pontificum Anglorum*; Otto of Freising, *The Deeds of Frederick Barbarossa*.

<sup>44</sup> Beňes, *Urban Legends*, 13–36.

<sup>45</sup> Oldfield, *Urban Panegyric*, 168–169.

<sup>46</sup> Falco of Benevento, *Chronicon Beneventanum.*

<sup>47</sup> Boncompagno da Signa, *Liber de Obsidione Ancone*.

<sup>48</sup> See for example, William of Malmesbury on Exeter, *Gesta Regum Anglorum*, 2.134, 216–217; and on Shaftesbury, *Gesta Pontificum Anglorum*, 2.86.1, 293. See also, Brett and Woodman, *Long Twelfth-Century View.*

William was not the only person writing descriptions of cities that had recently been controlled by Muslims. An example is the description of Palermo contained in Pseudo-Hugo Falcandus' *Letter to Peter the Treasurer*, composed in 1190.49 While the description is not particularly expansive on the Muslim past, it is nonetheless filled with references to places such as 'the Arab palace', 'the Forum of the Saracens' and 'the residence of Ṣiddīq the Saracen'.50 Critics of Muslim rule could employ urban space as a polemical tool. The anonymous account of the *Conquest of Lisbon* in 1147 cleverly uses a depiction of Muslim Lisbon as overcrowded, insalubrious and poorly planned to help justify the seizure of the city by the Portuguese.<sup>51</sup>

#### **Dangerous pasts**

All of which is to say that William was making clear decisions when he chose to ignore the Eastern Christian and Islamic past. This erasure made sense in the wider context of what he was trying to do with the History of Jerusalem as a whole. As mentioned earlier, William was creating a past for his people, the Latins of the East. William's history acted as both a justification of the Latins as a people and a claim to their right to inhabit the landscape. In order to truly claim a land, one had to master its history. The German pilgrim Theoderic, who visited the Holy Land in about 1169, began his description of Palestine with the observation that:

It is clear to all who read the pages of the Old and New Testaments, that the land of Canaan was … of old enriched by many cities, towns, and castles. The names and locations of all these cities were in antiquity well known to everyone; but the moderns, being strangers in the land, and not its native inhabitants, know only the names of a few places which we shall discuss in their place.52

Lack of knowledge about the urban history of a land was a sign of a temporary presence. Writing at the same time, William's demonstration of his understanding of the ancient city strengthened his people's claim to the territory.

Almost nowhere on Earth has as complicated and intensely emotional historical landscape as the Levant. Any effort to possess that space has to be able to cope with a whole range of inconvenient pasts preceding it. William was engaged in just such an exercise, deliberately erasing dangerous layers of the past in his descriptions of cities. The biblical past was safe for him, part of the wider heritage claimed by the

<sup>49</sup> Loud and Wiedemann, *History of the Tyrants of Sicily*, 28–38, 50.

<sup>50</sup> 'Epistola ad Petrum Panormitane Ecclesie Thesaurarium de Calamitate Sicilie', 167–186, 181: 'palatium Arabum … forum Sarracenorum'; 182: 'domum Sedicti sarraceni'; Eng. tr. 'A letter concerning the Sicilian tragedy to Peter, Treasurer of the Church of Palermo', 260.

<sup>51</sup> *De expugnatione Lyxbonensi*, 94–95.

<sup>52</sup> Theoderic, *Peregrinationes tres*, 143–197, 143–144: 'Sicut omnibus novi ac veteris testamenti paginas legentibus liquet, terra Chanaan … multis civitatibus et villis atque castellis antiquitus erat locupletata. Et harum quidem urbium omnia loca vel nomina antiquitus cunctis patebant, a modernis vero utpote adventiciis, non nativa habitatione ibidem manentibus, preter paucorum nomina locorum, que postea suo loco narrabimus, ignorantur'.

Latins which could usefully be employed as part of their deep history.53 Classical antiquity was also eminently serviceable, being both prestigious in the world of the twelfth-century renaissance and far enough away to be appropriated without any risk. Indeed, to the extent that the Crusaders understood themselves to be restoring lands to their rightful owners through their conquest, the mantle of that ancient past was extremely valuable.<sup>54</sup>

The Eastern Christian and Muslim layers of history were more perilous for William. He was living in a space where Catholic Christians were dramatically outnumbered by both of those groups, and very aware of the dangers posed by the growing power of the Zengid dynasty and Saladin beyond the kingdom. The Eastern Christian and Muslim presence in both the history and the physical geography of the cities of Outremer threatened to subvert the legitimacy of the Crusader state. In his writing of the history of his cities, William sought to cement the Latins as the true heirs to the Levant, by associating the Kingdom of Jerusalem with the ancient urban landscape of the Holy Land.

He was far from alone in being engaged in such an enterprise. Many of the Latins of the East were well aware of the way the physical reminders of the recent past challenged the legitimacy of their kingdom. The Kings of Jerusalem responded by emphasising their connection to the biblical and early Christian structures of their capital city. The claim made by the kings of Jerusalem to be the heirs of biblical kings was articulated in the epitaph to the first king, Baldwin I, by his grave in the Holy Sepulchre, which proclaimed him to be 'a second Judas Maccabeus'.<sup>55</sup> The Hasmonean era citadel, believed since the fourth century to have been constructed by King David, also acquired major importance for the Crusader monarchs.<sup>56</sup> The significance of both structures is indicated by their appearance on coinage. Baldwin III placed the Tower of David on the reverse of his coins, while the Holy Sepulchre featured on those of his successor, Amaury I.<sup>57</sup>

Perhaps the most obvious example of the erasure of the Muslim urban past is offered by the Frankish treatment of the Dome of the Rock and the al-Aqṣā Mosque, which competed with ancient Christian buildings such as the Holy Sepulchre to dominate the skyline.<sup>58</sup> Frankish writers responded by presenting the two great buildings on the Temple Mount as part of the biblical or early Christian heritage of the city. From the beginning of the kingdom, the al-Aqṣā Mosque was connected to the biblical past by being interpreted as the Palace of Solomon.<sup>59</sup> This took place even as the Templars continued to allow access to Muslims seeking to pray in the al-Aqṣā Mosque.<sup>60</sup>

<sup>53</sup> John, 'The "Feast of the liberation of Jerusalem"'.

<sup>54</sup> Flori, 'Guerre sainte et rétributions spirituelles', 628.

<sup>55</sup> 'Rex Baldwinus, Iudas Alter Machabeus', *Corpus inscriptionum crucesignatorum*, 57–58.

<sup>56</sup> Mostafa, 'From the Dome of the Chain to *Miḥrāb Dāʿūd*', 12.

<sup>57</sup> Metcalf, *Coinage of the Crusades*, 14–16.

<sup>58</sup> Grabar, *Shape of the Holy*, 160–173; Avni, *Byzantine-Islamic Transition in Palestine*, 128–133; Murray, 'Constructing Jerusalem as a Christian Capital', 5.

<sup>59</sup> John of Würzburg, *Peregrinationes Tres*, 93; Theoderic, *Peregrinationes tres*, 164.

<sup>60</sup> ʿUsāma b. Munqidh, An Arab-Syrian gentleman and warrior, 163–164.

The Dome of the Rock was renamed the *Templum Domini* in memory of the temple built by Solomon on the site.61 Despite the Arabic inscriptions on its walls, the Dome was viewed as an originally Christian structure. In the history written between 1118 and 1131 by Achard of Arrouaise, prior of the Augustinian community placed inside the Dome of the Rock,<sup>62</sup> Achard admitted to some uncertainty, stating that:

Some think that Justinian, prince of the Romans, built it [the Temple]; some ascribe it to Helena, King Constantine's Mother, who when looking for the Cross and the place of Calvary, rebuilt the Temple as it looks today. Some say that it was built by Heraclius, killer of Khosrow, when he restored the cross. Whoever built the Holy Temple of the Lord, as it is so it will be until the end of days.63

All of these were important figures from the Roman Christian past. Albert of Aachen (*c*. 1100) wrote that the building was erected 'by modern people and Christian worshippers.'64 Theoderic the pilgrim was more confident in attributing the Temple to Helena and Constantine.<sup>65</sup>

These accounts placed the current management of the Dome of the Rock within a longer tradition of Christian rule. This was probably a deliberate rebranding, with many of those who attributed the Dome to Solomon being well aware of its actual origin.66 In doing so they removed a dangerous presence from their midst while placing themselves as the heirs to Israel. The importance of the Dome of the Rock to the imagery of the kingdom is indicated by its appearance on the royal seal, together with the Tower of David and the Holy Sepulchre. Frankish maps of Jerusalem prominently displayed the *Templum Domini* of Solomon, making it serve as an epitome of the Crusader city.<sup>67</sup>

Interestingly, the Temple Mount structures offer a rare instance of William actively acknowledging the Muslim urban past. He was well aware that the Dome of the Rock was built by the Caliphs, because of the 'ancient inscriptions upon the walls of the building both within and without', although he attributed it to ʿUmar rather than to ʿAbd al-Malik or al-Maʿmūn.<sup>68</sup> His unusual willingness to comment on the Muslim

<sup>61</sup> Busse, 'Vom Felsendom zum Templum Domini', 19–32; Schein, 'Between Mount Moriah', 175–195.

<sup>62</sup> Hiestand, 'Gaufridus abbas Templi Domini', 48–59.

<sup>63</sup> Lehmann, 'Die mittellateinischen Dichtungen der Prioren des Tempels' 329 *'*Quidam a Iustiniano, Romanorum principe, factum putant atque quidam hoc ascribunt Elene, matri regis Constantini, quod eodem tempore, quando crucem requisivit et locum Calvarie, templum reedificavit, quod apparet hodie. Ab Eraclio constructum sunt qui velint dicere, quando crucem reportavit, interfector Cosdroe. A quocumque sit constructum templum sanctum Domini, fuit, est erit usque ad extrema seculi*'; Schein, 'Between Mount Moriah', 181*.

<sup>64</sup> Albert of Aachen, *Historia Ierosolimitana*, 432–435.

<sup>65</sup> Theoderic, *Peregrinationes tres*, 163.

<sup>66</sup> Boas, *Jerusalem in the Time of the Crusades*, 89–91; Pringle, *The Churches of the Crusader Kingdom of Jerusalem*, 400–401; Ousterhout, '"Sweetly refreshed in imagination"', esp. 157; Di Cesare, 'The *Qubbat al-Ṣaḫrah'*, 233–255.

<sup>67</sup> Spaer, 'A seal of Baldwin I'; Levy-Rubin and Rubin, 'The image of the Holy City', esp. 354; Mahoney, 'The Frankish icon', 22.

<sup>68</sup> William of Tyre, *Chronicon*, 8.3, 386: 'esse antiqua litterarum monimenta in eodem edificio intus scripta et deforis'.

origins of the structure probably owes much to the deep antipathy he felt towards the Knights Templar who occupied the Dome of the Rock from 1119.<sup>69</sup>

This wider activity suggests that William's omission of the medieval past would be well received by a domestic audience. It was also potentially useful for an overseas market. Upon his return to Tyre after taking part in the Third Lateran Council in 1179, William appears to have done some serious editing of his manuscript. Among the new material he added in the 1180s are the great setpiece descriptions of Antioch, Jerusalem and Tyre.<sup>70</sup> These changes were designed to make the *Chronicle* more exportable to the West, in order to act as an apologia for his homeland.

Jerusalem in the 1180s was looking increasingly rickety. Rival factions jockeyed for power within the kingdom as the Muslims of Egypt and Syria were united under the leadership of Saladin. William's revised *Chronicle* was intended at least in part to encourage readers in the West to support the embattled Latin East. But it was also intended to assert Outremer's place among the realms of Western Christendom.<sup>71</sup> Perceptions of the inhabitants of the Latin East by their counterparts in the West was often not very positive, with the former being assumed to have gone native, corrupted into oriental habits.72 Jerusalem was also an intellectual backwater.73 In his *Chronicle* William sought to counterbalance that, by emphasising the links between the Kingdom of Jerusalem and the West.

The set-pieces concerned with cities fit into this program well. They celebrate the cities of William's native land as beautiful and important and therefore worthy of respect in a manner that was immediately recognisable to an overseas audience. The biblical and Classical past was highly prestigious in Western Europe.74 Positive descriptions of cities throughout the Latin world such as Trier, Bergamo and London emphasised their antique heritage.75 Positioning a city as in some way related to early Christianity also served to elevate their status, as demonstrated by the famous *Mirabilia Urbis Romae* of 1143.<sup>76</sup> This happened even if the writers in question had to fabricate said ancient history, as in the case of Florence and Venice.<sup>77</sup>

William's set-pieces also demonstrated the vitality of Latin culture in the East. Through his portraits of Antioch, Jerusalem and Tyre, William was showing off his mastery of classical and biblical literature in a bravura performance of his scholarship, and by implication, the wider learning to be found in the Latin East.<sup>78</sup> These displays

<sup>69</sup> Nicholson, 'Before William of Tyre'.

<sup>70</sup> Edbury and Rowe, *William of Tyre*, 27–29.

<sup>71</sup> Kedar, 'Some new light', esp. 9.

<sup>72</sup> Schein, 'The image of the Crusader Kingdom of Jerusalem'.

<sup>73</sup> Kedar, 'Intellectual activities in a Holy City'.

<sup>74</sup> See Benson and Constable, *Renaissance and Renewal in the Twelfth Century*.

<sup>75</sup> Oldfield, *Urban Panegyric*, 160–185; *Gesta Treverorum*; Mosè del Brolo, *Liber Pergaminus*; Geoffrey of Monmouth, *History of the Kings of Britain*, 3.53, 66–67.

<sup>76</sup> 'Mirabilia Urbis Romae', *Codice topografico della città di Roma*.

<sup>77</sup> Baldassarri, 'Like fathers like sons'; Brown, 'History as myth'.

<sup>78</sup> Edbury and Rowe, *William of Tyre*, 35–37.

of erudition were in part a simple reflection of his education, they also added to the effect he was trying to create of portraying his homeland as part of a wider Latin world, inhabited by people who could communicate in a way that western readers would recognise. William does not seem to have been immediately successful in this endeavor, as there are only a handful of Latin manuscripts of his *Chronicle* in the West.<sup>79</sup> The Old French translation on the other hand did very well indeed, proving to be extremely popular in France and England.<sup>80</sup>

### **Conclusion**

The lacunae in William's descriptions of the cities of the Holy Land were a deliberate part of an exercise in claiming the urban landscape of the Kingdom of Jerusalem for the Latins of the East. Ancient history was prestigious and safe, that of the Eastern Christians and Muslims was not. In choosing the layers of the urban past that he did, William created a history that was useful both for his fellow countrymen and a means of communicating a shared identity with the Franks of the West.

William of Tyre's last years were unhappy.81 But in another way William was a very fortunate man indeed. Although he saw his beloved Outremer in peril, he died before the Horns of Ḥaṭṭīn and Saladin's conquest of Jerusalem in 1187. Fundamental to William had been his deep and abiding love for his homeland, proved not just by words, but through a lifetime in its service. It was this desire to support the nation that he felt he was a part of that prompted him to sideline the many difficult pasts that the cities of the Levant play host to. But within five years of William's breathing his last, perhaps even within one year, the world that he had been born into, that he had lived through and loved, was another inconvenient layer in the history of these ancient cities.

#### **Acknowledgements**

This chapter was researched and written as part of the 'Impact of the Ancient City' project, University of Cambridge, funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 693418). I would like to express my gratitude to the other members of the project, the attendees of the 2019 Istanbul Palimpsests conference and to James Kane for their advice and comments.

### **Bibliography**

#### *Primary*

Albert of Aachen, *Historia Ierosolimitana: History of the Journey to Jerusalem*, ed. and tr. S. B. Edgington (Oxford, 2007).

ʿAlī of Herat, *Guide des Lieux de Pèlerinage*, tr. J. Sourdel-Thomine (Damascus, 1957).

<sup>79</sup> Edbury and Rowe, *William of Tyre*, 4.

<sup>80</sup> Edbury, 'The French translation of William of Tyre's *Historia*'.

<sup>81</sup> Edbury and Rowe, 'William of Tyre and the patriarchal election of 1180'.

Boncompagno da Signa, *Liber de Obsidione Ancone*, ed. and tr. P. Garbini (Rome, 1999).


*De expugnatione Lyxbonensi: The Conquest of Lisbon*, ed. and tr. C. W. David (New York, 1936).

Falco of Benevento, *Chronicon Beneventanum*, ed. E. D'Angelo (Tavarnuzze, 1998).

Fulcher of Chartres, *Chronicle of the First Crusade*, tr. M. E. McGinty (Philadelphia, 2016).


John of Würzburg, *Peregrinationes Tres: Saewulf, John of Würzburg, Theodoricus*, *Corpus Christianorum Continuatio Mediaevalis* 39, ed. R. B. C. Huygens (Turnholt, 1994), pp. 79–141.


#### *Secondary*


Benson, R. L. and G. Constable (eds.), *Renaissance and Renewal in the Twelfth Century* (Oxford, 1985).


## Chapter 9

## Portraits of Ottoman Athens from Martin Crusius to Strategos Makriyannis

*Elizabeth Key Fowden*

*For my mother, Jane Crusius Key*

On two occasions, in 1436 and 1444, diplomat and antiquarian Cyriac of Ancona visited Athens and sketched the city's antiquities. Most famous is his drawing of the Parthenon's west facade, which he depicted not as it was – the entrance to a Catholic church with a high tower of reused ancient marble – but as he wanted it to be – an ancient temple in pristine condition.1 In 1458, the Ottoman sultan Mehmed II also ascended the citadel of Athens and 'from the ruins and the remains, he reconstructed mentally the ancient buildings, being a wise man and as a great king, and he conjectured how they must have been originally'.2 That, in any case, was the report of the contemporary chronicler Kritovoulos of Imbros, who had not been with the sultan when he visited the recently conquered city. Mehmed's interest in drawing and in maps is well-attested and the Constantinopolitan historian Doukas even depicts the Conqueror sketching the walls of the Queen of Cities as he prepared his final assault.<sup>3</sup> We might quite plausibly imagine the Ottoman prince, like Cyriac some twenty years earlier, sketching the Parthenon, or perhaps a more distant view of the acropolis with its impressive Frankish defences.

Neither Cyriac's nor Mehmed's admiring visits to Athens did much to stimulate antiquarian pilgrimage to Athens in the fifteenth century. Adapting Henri Pirenne's comment on America, 'discovered by the Norsemen and then forgotten, because, in the eleventh century, the world had as yet no need of it,' it seems that the world

<sup>1</sup> Copies of Cyriac's sketch survive; see the excellent studies by Howard, 'Responses' and Tanoulas, 'Through the broken looking glass'. Also, Tanoulas, *Τὰ Προπύλαια*, 40–41, 310–314. 2 Kritovoulos, *Histories*, 3.9.5–6, tr. 136. Inserted after 'a wise man', the marginalium 'and a Philhellene'

is found at 3.9.6, lines 27–28 of the unique manuscript dedicated to the sultan and kept in his private library.

<sup>3</sup> Doukas, *History*, 35.6.

in the fifteenth century did not yet need the 'discovery' of Athens.<sup>4</sup> And neither did it need an eye-witness portrait of the city. For that we would wait another two centuries, when Athens would suddenly receive a cluster of visitors who commented extensively on what they saw and made depictions that would set the standard for later portraits of Athens. My purpose in this chapter is to ask: when the need did finally arise in the late seventeenth century, who needed Athens and its portrait, and what purposes did those first portraits serve? That new interest in Athens was driven by religion, war and antiquities, so my discussion will follow the contours of these three intertwined concerns.

Like the discovery of America, what we miss when considering the early portraits of Athens is the inside view, the population waiting to be 'discovered'. What the portraits allow us to glimpse is an intense period of roughly two decades in the 1670s and 1680s when visitors to Athens underwent a process of recognition not entirely different from the exoticism and wonder of New World encounters.<sup>5</sup> In the case of Athens, the discoverers had books from the past to guide them, even though ancient Greek and Latin authors led them to expect a city and inhabitants different from what they saw. These discoverers of Athens also had the benefit of a handful of missionaries and consuls who could act as translators between local legends about the past *in situ* and the ancient texts. The portraits reveal some of this process of translation, through their manipulation of the city, labelling of its buildings and juxtapositions of European representational conventions used to communicate what the city, both past and present, meant to different foreign viewers. In these early portraits it is possible to capture modes of engagement with an accumulated local understanding of the city's past, before it was corrected, rewritten and either forgotten or transformed into the antique or the picturesque.

In parallel with the trend from accumulated to selective representation in the early portraits, this chapter examines another epistemic stance made visible through them. This is the difference between the static city portrayed from a distance and the depiction that re-presents immersive experience of it. This difference is more commonly analysed in writing, as a method of conquest and ordering, than in visual representations of cities although the two were brought together in the work of the Jesuit scholar Michel de Certeau.6 Travel literature offers itself for the examination of the outsider's retrieval of knowledge and rendering of the local informant's voice. Some of the portraits studied here accompanied writings or were framed by written legends; others stood on their own. By beginning and ending with views in which the local voice and experience was part of the 'dictation' of the city's visual representation I bring to the foreground the interactive processes involving very different actors

<sup>4</sup> Pirenne, *Histoire*, 143, Eng. tr. 196; Yakovaki, *Ευρώπη*, 168; cf. Calis, 'Reconstructing', 184.

<sup>5</sup> One cannot overestimate the importance of Yakovaki, *Ευρώπη* as a detailed contextualisation of these pivotal decades, based on the written sources.

<sup>6</sup> I am thinking particularly of *L'écriture de L'histoire* and *L'invention du quotidien*; see Hähnle, 'Knowledge', esp. 44–46, 62–63. Sixteenth-century immersive, multi-focal representation of Istanbul has been fruitfully studied by Kafescioğlu in, for example, 'Picturing the square'.

that led to the creation of these early portraits. In this I hope to honour Certeau's exemplary reticence with regard to the scholar's role as resolver, given that as scholars we are always still part of the human practice of writing and ordering, never offering a resolution of the problematic relationship between local voice and outsider writer, but hopefully drawing attention to the previously absent others in our discourses.

#### **What cities can say**

There are many reasons to portray a city, especially a city with a long history. The artist may wish to project patriotic sentiment, economic prosperity, religious authority, territorial claims, cultural distinction or historical priority, or simply flatter a patron's vanity. All of these reasons call for enhancing or omitting particular features of the urban tissue. In an elegant study of humanist representations of Athens, George Tolias has argued that between the late fifteenth and mid-seventeenth centuries humanist geographers and cartographers from across Europe deliberately cultivated through their writings and the fantastical images that accompanied them the notion of Athens as a ruined city onto which they could project their own vision of ancient Greece and its lost 'capital'.7 Instead of selecting a 'consistent corpus of evidence with internal coherence' as Tolias has done in order to outline one characterisation of Athens, this study is motivated by the competing purposes and ideologies that stand out when the whole range of early depictions of Athens is viewed together.

My use of 'portrait' is deliberately broad and includes the sketch, engraved plan, illustration to accompany a text, and broadsheet print, as well as the more formal painted city view. By bringing together different modes of representation I highlight the depictors' varying expectations of the city and the different technical means by which those expectations were projected. As for the term 'city', I do not aim to establish a technical definition. What matters here is that a defined settlement is the object of depiction, so that the conventional terms 'city view' and 'city portrait' will serve even for smaller-scale settlements such as early modern Athens whose population seems to have risen as high as 18,700 during the sixteenth-century flourishing of the Pax Ottomanica.<sup>8</sup> What the city views under consideration all share is that they are the earliest known attempts to visualise the city of Athens as a whole based on eye-witness experience.

As a contribution to a volume whose authors explore ways in which inhabitants of ancient cities in ongoing use selected which elements from the past were useful to them as both material and metaphor, and which were not, I consider in this chapter the process of selection implicit in depicting such a city visually. I am not concerned here with tracing the linear development of a single visual genre, nor

<sup>7</sup> Tolias, 'Ruined city'. For recent overviews of European depictions of Athens that focus on the idea of the city as isolated and ruined, see Koutsogiannis, 'Η εικόνα' and Kouria, *Η Ελλάδα*, both with extensive bibliographies. I thank Aphrodite Kouria for a copy of her book, which draws especially on lesser known material in the Efstathios Finopoulos Collection (Benaki Museum) and other private collections. 8 Kiel, 'Central Greece', 399–405; Kiel, 'Athens'.

with producing an exhaustive catalogue of all early portraits, but with examining a representative assemblage of views of a single city that allow us to analyse their range of motivations and approaches. I give special attention to the various reuses of the so-called Capuchin plan, which I will refer to more neutrally as the first known city plan of Athens. The reason for my focus on Athens is its surprising absence from the explosion of interest in city portraits by European artists that erupted in the sixteenth century. Rome, Constantinople/Istanbul and Jerusalem were already frequently pictured from the fifteenth century (and earlier) as artists experimented with modes of representing cities of great political, cultural and religious importance. Not until the late seventeenth century did Athens catch up in a quickening of efforts under Louis XIV and Venetian Captain-General Francesco Morosini to exert control over the eastern Mediterranean. Visualising cities helped to articulate their ambitions.

## **Marginal Athens**

I begin with the necessary reminder that Athens was not always a great political or cultural capital. The point is clearly made on an early fifteenth-century Greek map, thought to have been owned by Mehmed II, in which Athens is not identified by a walled city vignette (as are Salonica and Mistras), but is indistinguishable from the many other small settlements marked with only a conventional symbol.9 Subsequent Ottoman rulers commissioned maps and city views in which the Italianate bird's eye view was merged with Persianate traditions of architectural representation drawing on manuscript painting in order to accompany narratives of territorial expansion and idealisations of Ottoman urban order.10 Matrakçı Nasuh was the sixteenthcentury master of city portraits and his *History of Sultan Bayezit* includes portraits of four ports in the Morea retaken from the Venetians in 1499: Lepanto, Modon, Koron and Navarino.11 But no known Ottoman city views picture Athens. In fact, before the Greek uprising against Ottoman rule that began in 1821, no local residents or Ottoman visitors are known to have produced visual representations of the 'City of Sages', as Athens was known in Arabic and Ottoman writings.12

<sup>9</sup> The map survives in Topkapı Saray Library, see Pinto, 'Maps', 156–158, fig. 2 (G.I. 27 fols 83b-84a); Necipoğlu, 'Visual cosmopolitanism', 7, 11.

<sup>10</sup> Antrim, *Mapping*, 63–109 with bibliography; Kafescioğlu, 'Viewing'; Beyazit, 'Defining Ottoman realism' on sixteenth- and seventeenth-century depictions of Mecca and their artistic context.

<sup>11</sup> For a recent discussion of Matrakçı Nasuh with bibliography, see Beyazit, 'Defining Ottoman realism' 217–218; also Ebel, 'Representations', 11 with pl. 3. For a possible indication of how Evliya Çelebi might have depicted Athens, had he produced a map to accompany this Balkan travels, see the publication of drawings believed to illustrate his Nile explorations, see Dankoff *et al*., *Ottoman Explorations.*

<sup>12</sup> One makes such statements in order to be refuted. Cities sometimes appear in Ottoman-period icons, a possible future source of portraits of Athens. Machiel Kiel has pioneered the use of what some scholars call 'post-Byzantine' painting for visual evidence of Ottoman architecture, see Kiel, 'Icon and the mosque'; Kiel, 'Byzantine architecture'. It has also been argued that Michael Choniates, a twelfth-century bishop of Athens from Constantinople, had a portrait of ancient Athens made, see Speck, 'Byzantine depiction'.

While eye-witness portraits of Athens by Europeans appear only in the 1670s, one should resist imagining that no Europeans passed through the flourishing sixteenthcentury Ottoman city, the third largest in the Balkans after Edirne and Salonica.13 By 1570 Athens was adorned with Islamic buildings, two mosques and a school, woven into the existing urban tissue to serve the growing number of Muslim households in the lower town, north of the citadel, in addition to the citadel mosque used for prayer by the garrison and their families who were resident on the acropolis.14 Athens was not yet part of the European story of competing territorial, religious and cultural claims that led to its depictions, but neither was it completely forgotten. In October 1588, an apothecary from Königsberg named Reinhold Lubenau visited Athens on his return to Venice after spending a year and a half in Constantinople with the embassy of the Holy Roman Emperor Rudolph II. Lubenau claimed that his notebook had been confiscated while he was sketching on the citadel.<sup>15</sup> Besides confirming the obvious fact that travellers with different purposes sketched what they saw, and quite commonly lost their sketches along the way, Lubenau's story also reminds us that in the sixteenth and seventeenth centuries that form our main interest, Constantinople was the urban pole en route from and to which visitors passed through Athens.

Despite frequent travel to the Ottoman capital, most sixteenth-century humanists learned from books, correspondence and visitors who came back with eye-witness news. For the majority who wrote about Athens on the basis of ancient texts, the ancient city was of primary but not exclusive interest. Contemporary religious conditions also attracted some attention. In 1576 Martin Crusius, Professor of Greek at Tübingen and author of *Turcograecia*, wrote to a Greek correspondent in Constantinople asking for a sketch of Athens to help orient his students in their reading of Thucydides.16 Besides helping his students to understand the disposition of the Long Walls in relation to Athens, Piraeus and the surrounding landscape, Crusius had an equally burning desire: he wanted to dispel the rumour that the city had fallen into utter ruin and was now nothing but a collection of fishing huts.17 One of Crusius's correspondents was Theodosios Zygomalas, *protonotarios* at the Greek Patriarchate and a native of Nauplion who had visited Athens several times before he moved to Constantinople

<sup>13</sup> Kiel, 'Athens'.

<sup>14</sup> Kiel, 'Quatrefoil plan', 115–117.

<sup>15</sup> Paton, *Chapters*, 45–50, esp. 49. Lubenau's diaries shed light on a range of tensions in the late sixteenth century Mediterranean that are the background to Athens' marginality, see Freller, '"On the enemies' side"'; Koder, 'Early modern times', 141–148; Koder, 'Reinhold Lubenau'.

<sup>16</sup> Crusius, *Turcograecia*, 446: 'ἀπεικόνισμα τῶν Ἀθηνῶν οἱονδήποτε (μόνον τῇ σῇ χειρὶ) βλέπειν, Θουκυδίδου εἵνεκα, ὃν τοῖς φοιτηταῖς διασαφῶ', in Martin Crusius's letter 14 to Theodosios Zygomalas dated 15 April 1576, 444–450. For the chronology of their correspondence, see Rhoby, '"Friendship"', 253–255; Calis, 'Reconstructing', 176. The Corfiote cartographer Nikolaos Sophianos claimed to have visited Athens, but the vignette of Athens and Piraeus on his map of 1540 shows no indication of eyewitness acquaintance with the city, reflecting ancient descriptions instead. On Sophianos, see Tolias, 'Nikolaos Sophianos's *Totius Graeciae Descriptio*'; Della Dora, 'Mapping', 465–466, fig. 9.

<sup>17</sup> Crusius, *Turcograecia*, 446, letter 14. Tolias, 'Ruined city', 25–26, on Melanchthon.

*Fig. 9.1: Martin Crusius, sketch map of Athens, unpublished diary (Universitätsbibliothek Tübingen, UBT Mh 466, vol. 3, fol. 302).*

at age ten.18 Crusius rejoices in having found a 'trustworthy eye-witness' ('ἀξιόπιστος μάρτυς') who had seen Athens himself ('αὐτόπτης τοῦ ἄστεως').<sup>19</sup> *Turcograecia* is a monument to the omnivorous curiosity of a Lutheran humanist and Hellenist, and among the many documents Crusius published therein were letters from Theodosios and Ioannis Zygomalas and Symeon Kabasilas, who described for him some of the surviving ancient monuments of Athens. But there is no indication that Crusius ever received his requested sketch of the city from Zygomalas's own hand.

Crusius's unpublished diaries do, though, contain several sketches he made himself, patiently piecing together information he could extract from the succession of Greeks who made their way to his house in Tübingen. There is even one annotated sketch of Athens viewed from north to south, made nearly a decade after his request to Zygomalas (Figure 9.1).20 The sketch and annotations (a mix of Latin and Greek) are in Crusius's hand, as he was taking dictation from Daniel Palaiologos, an Athonite monk who had grown up in Athens where his father was a merchant.21 The unique sketch should be seen as a collaborative effort between a local informant and a Tübingen scholar. It does not exactly provide a local view of the city, first and foremost because

<sup>18</sup> Crusius, *Turcograecia*, 428–435, letter 10.

<sup>19</sup> Crusius, *Turcograecia*, 446, letter 14.

<sup>20</sup> Calis, 'Reconstructing', 174. For correspondence about Martin Crusius, I warmly thank Richard Calis, to whom I owe my knowledge of the sketch map of Athens. For discussion of the sketch, I also thank John Davis, Garth Fowden and Andreas Rhoby.

<sup>21</sup> Daniel Palaiologos, self-identified as *oikonomos* of Iviron Monastery, visited Crusius in September 1585. For Crusius's notes, see Universitätsbibliothek Tübingen (UBT), Mb 37, fol. 161; sketch of Athens with labels: UBT Mh 466, vol. 3, fol. 302; see also fol. 303 of Mount Athos, with labels.

the idea of representing the city at all came from an outsider. But the recollection of what is plotted is local, to which Crusius has added some of his own notes connecting up Palaiologos's information with other sources.22

All the buildings noted are identified by their contemporary, overwhelmingly Christian names and functions, not their classical names. The acropolis is identified as the 'castellum', a not-so-high rocky height where the Turks lived, according to his annotation.23 Along the top of the acropolis hatching seems to indicate the fortification wall.24 It is a rough sketch and the east–west orientation has been reversed. At the east end a tower-like doodle likely represents the medieval tower that stood at the west end of the inhabited acropolis. At the centre stands a single large building with a peaked east end identified by the word 'meskita' (that is *mezquita*, mosque), written to the building's right.25 This is, of course, the Parthenon, converted to a church at whose south-west corner was erected a tower that had been reworked into a minaret when the church became a mosque. The substantial building in the centre of the lower city, south of the 'Forum' is labelled as the 'great church of S. Anna' (again impressionistic or just poorly remembered?). A few other buildings are identified only by words: north of Crusius's 'Forum' is a 'church of S. Chrysostom' and a 'church of S. George' is placed inside the walls on the west side (wrongly, as this must be the intact temple-turned-church now called the Hephaisteion or, locally, the Theseion). One domestic habitation is also noted: the house of the prominent Chalcocondyles family. We can hear Crusius interrogating Daniel about the relationship of the landmarks he mentions to the acropolis and to the city walls. The fact that he drew a circuit around the city is intriguing – did Palaiologos indicate there was a wall, or did Crusius assume there had to be one, even though in one of his notebooks he records the information that neither Corinth nor Athens had walls?<sup>26</sup>

Picturing Athens was always related to how the person who held the pen or brush understood the relationship between the classical and Christian histories of Greece, and the continuing material legacy of these histories under contemporary Ottoman rule. What is intriguing and exceptional about Protestant Crusius is that for him contemporary Orthodox Greeks were a vital part of his multi-temporal encounter with Athens. Although eventually disillusioned, Crusius had courted ties with the Greek Orthodox Patriarchate in Constantinople on account of his conviction that Lutherans and Greek Orthodox shared theological views potentially useful against Catholic claims. For Crusius, Athens was 'rediscovered' as a living city not exclusively or even primarily for its antiquities, but as part of European intra-Christian polemic and intertwined political aspirations in Ottoman-dominated Christian lands. And he

<sup>22</sup> For example, Crusius's marginalium along the left side notes the tripartite division of the city, which his correspondent Symeon Kabasilas had mentioned in a letter, *Turcograecia* 462, letter 19.

<sup>23</sup> UBT Mh 466, vol. 3, fol. 302: 'Tὸ καστέλιο, ἐν ὄρει οὐχ ὑψηλῷ, ἐν πέτρᾳ. Ibi, Turcae.'

<sup>24</sup> Or possibly an impression of rooftops visible on the inhabited castle as seen from below, as suggested by Guy Meyer (pers. comm. 3 May 2021).

<sup>25</sup> The citadel mosque was also named a 'mezquita' in the Spanish description by Diego Galán, a native of Toledo who briefly visited Athens in the 1590s while serving as an Ottoman galley slave, see Fowden, 'Parthenon mosque', 69, 76–77. On Galán, see Boll, 'Tale' 7–9; Hershenzon, *Captive Sea*, 31, 234 n. 29. 26 UBT Mb37, fol. 102. On the walls of Athens, see below 171–173 with n.57 and 181–183.

rediscovered Athens through individuals with eye-witness experience from whom he could gather authentic, local information. In this, Crusius foreshadowed the collaborative dependence of Athens' depictors a century later.

## **The rise of the city portrait**

There was, then, a little information circulating among scholars, merchants and monks about the ancient city as a still-living organism, but knowledge of Athens under early Ottoman rule was chronologically mixed. Nonetheless, given the lack of contact with, or concern for, local circumstances, for the most part Athens presented an unusual opportunity for the artist, mapper and writer. The city's obscurity made it easier to invent elements free from any local knowledge of contemporary Athens, or to depict an imagined classical Athens inspired by ancient texts, fantasy and by analogy to portraits of other ancient cities.27

Meanwhile, the idea that a city portrait could project a city's political, cultural and commercial status was galvanised by the first printed atlas of the world's cities, the *Civitates orbis terrarum*, published in Cologne between 1572 and 1617. It was a multi-artist production edited by Georg Braun with engravings by Frans Hogenberg, including city images assembled from many other local artists, but most importantly the Antwerp merchant/artist Joris Hoefnagel. The transformation from medieval depictions of ideal or stereotyped cities to detailed portraits relied on the vital tool of eye-witness experience, often explicitly highlighted in the texts which accompanied the images.28 Ancient buildings were not singled out, but woven into the complex urban fabric, in various stages of decay or reuse as they appeared in real life. Both practical and symbolic, walls evoked civic pride and protection, as well as offering the artist a crisp line mediating between the more impressionistically represented, open countryside and the densely built-up interior. Multiple modes of representation – bird's eye view or 'true perspective', bird's flight view combining vertical plan and perspective views and a more formal street plan – were used within the same representation to create a multifocal portrait of the city.

Once visualised, a particular angle for viewing a city could be strikingly tenacious. The example of Jerusalem is instructive (Figure 9.2): the city seen from the Mount of Olives, for example, appears early in pilgrim guides as the best vantage point for seeing the Christian (but also Muslim) holy places and came to be the expected view, as here in the *Civitates*, and also in later portraits, as we shall see. A common feature of the cities in the Braun and Hogenberg collection, and in particular those drawn by Hoefnagel, is the inclusion of figures in the foreground. Sometimes the artist was shown at work in the corner of the picture, taking in the view, which intimated accuracy and

<sup>27</sup> Tolias, 'Ruined city'; Kouria, *Η Ελλάδα*, 21; Tsigakou, *Rediscovery*, 14.

<sup>28</sup> The European development is much discussed, see Nuti, 'Mapped views'; Ballon and Friedman, 'Portraying'; Nuti, 'Perspective plan'; Maier, 'A "true likeness"', esp. 716–717; Lichtert *et al*., *Portraits of the City*, esp. 1–8, all with extensive bibliography. See also Shalem, 'The city objectified'.

*Fig. 9.2: Franz Hogenberg, Jerusalem, in Georg Braun and Franz Hogenberg,* Civitates orbis terrarum*, vol. 2, pl. 54 (Cologne 1575) (Princeton University Library).*

autopsy.29 Very commonly the middle foreground figures wore local dress, another way to heighten the city portrait's specificity by showing local crafts and traditions.

Such city portraits were conceived not only to satisfy curiosity but also as part of a very particular political conjuncture, as Braun explains in the introduction to the 1572 edition:

Nobody has to be afraid that our work may harm the Christians in any way because with its help their major cities could be conquered by the enemy. That danger, which is rather real, is avoided in this way. We asked that the different kinds of clothing of all nations and several people be drawn at each city, both of the upper and lower classes. The reason is because the bloodthirsty Turks, who are not allowed to look upon representations of the human form, will never allow this book, however great their use for it may be!30

With the invention of such an apotropaic screen to protect potentially revealing details from Muslim eyes, Braun argued in his project's defence that the detailed depictions of cities that had been conquered by the Ottomans could actually help the Christians in retaking them. It was a live issue. Braun's argument appeared in the

<sup>29</sup> Nuti, 'Mapped views' on Hoefnagel's artistic response to mapping conventions, esp. 553 on the artist depicted, 569 on autopsy.

<sup>30</sup> English tr. (with alterations): van Egmond, 'Town atlas'.

first volume, published in 1572, the year after the Battle of Lepanto in which the Holy League inflicted what it proclaimed to be a glorious defeat of the Ottoman fleet and interpreted as a sign that the tide was turning and Christian reconquest was in sight.

In this period of rapid visual and conceptual development for the city portrait, the only cities to appear in the *Civitates orbis terrarum* from what we call Greece today belonged to the Venetian Aegean: Chios, Crete and Rhodes.31 Athens was absent. Yet immediately prior to our first surviving depictions, Athens was becoming a testing ground for the complex relations between religious identities, not only between Christian and Muslim, but also between Protestant, Catholic and Orthodox. The first views to which I will now turn belong to a discourse about the aspiration to Christian political and cultural domination of Ottoman-held Athens in which the city was depicted holistically, without privileging its antique features. These first portraits of Athens employ a variety of visual conventions, all appeared in the 1670s and most made a lasting impression on how artists and other travellers depicted the city, even those who see the city with their own eyes.

### **Louis XIV and the cities of the Levant**

The need for portraits of Athens finally became politically pressing during the reign of Louis XIV and accompanied his pursuit of religion, war and antiquities to consolidate his power across the Levant. The first three portraits are associated with the Sun King: the first published drawing was provided by the king's Antiquarian, the first detailed veristic painting was commissioned by the king's Ambassador to the Sublime Porte and the first city plan highlighting modern and ancient buildings of particular note has been associated with the French Capuchin missionaries settled in Athens, and if not actually produced by them was created in the overlapping circles of French information-gatherers at this time. By the 1660s an intricate and far-reaching network of Catholic priests and missionaries was in position which provided local information, mostly orally transmitted and tapped by consuls and merchants, to enhance France's strategic standing in the Levant. The manner in which they had made themselves part of the local scenery, thereby attracting less attention and facilitating their work, 'reflects something of the uniqueness of French interests in the Ottoman world in this period'.32 Sometimes orally communicated information was committed to writing. For example, most of our earliest descriptions of contemporary Athens are from Jesuits or Capuchins who had either passed through Athens or lived there. Nearly three years

<sup>31</sup> Kouria, *Η Ελλάδα*, ch. 1 illustrates two cities from *Civitates* (figs. 4–5), as well as other Greek city views, some rare. Greene, 'Beyond the northern invasion' argues for the continuing importance of religious conflict in the seventeenth-century Mediterranean, vital background to the early portraits of Athens. On the intersection of Greek and Venetian interest in antiquities in sixteenth- and seventeenthcentury Crete, see Stouraiti, 'Collecting', 36–37. For an important study of the complex inter-religious developments that cut across Ottoman, Greek, Orthodox and French interests across the seventeenthcentury eastern Mediterranean, exposing the difficulties of parsing religious, linguistic, cultural and political identities, see Koutzakiotis, *Αναμένοντας το τέλος του κόσμου*.

<sup>32</sup> Ghobrial, *Whispers*, 46–48, 80.

before he had the opportunity to see the city's antiquities with his own eyes, a rising orientalist in the king's circle, Antoine Galland, recorded in his journal a description of Athens he heard in Constantinople from the Capuchin Père Alexis in 1672.33

Louis XIV's court patronised skills that would promote the expansion of his influence, notably oriental languages, including contemporary spoken Greek, as well as numismatics, epigraphy and manuscripts, specialisations that were enriched by the acquisition of new, previously unknown material. Royal missions were regularly accompanied by instructions for the collection of manuscripts and antiquities from Greek priests and monks that would be sent back to Paris so that France would be 'adorned with spoils from the East'.34 As a seat of missionary activity, Athens acted as a minor node where French interests in religion, conquest and antiquities converged. The earliest portraits respond to all three. Ottoman Athens sat ambiguously between ancient and modern, and the missionaries might pose as agents who moved between the two sometimes competing spheres.

Already in 1641 a Jesuit mission had been founded in Athens. In 1657 the Jesuits moved to Negroponte and were replaced in Athens by the Capuchin mission. In 1669, Fr Simon purchased the 'Lantern of Demosthenes', as the choragic monument of Lysicrates was known locally, and incorporated it into the Capuchin monastery.35 Later they opened a school, as the Jesuits had done. Just as diplomatic and religious missions were inseparably mixed, so the convent served as base for persons with entangled tasks such as Fr Robert de Dreux, the French ambassador's chaplain in Constantinople (1665–1669) who stopped to see the antiquities of Athens while conveying diplomatic letters from Larissa to Melos.36 The Capuchins seem to have acted as hosts and informants for many of the growing number of European visitors, as the Franciscans did for Jerusalem pilgrims. They did not operate alone, but in tandem with local authorities.37 It was, for example, through the intercession of a member of the leading Palaiologos family and guided by a 'mémoire' on antiquities provided by the Lyonnais Jean Giraud, formerly the French consul, that Fr Robert was able to visit the 'château'. He records his visit on the citadel to what he and many other early visitors understood to be the Temple to the Unknown God with an unproblematic

<sup>33</sup> Schefer, *Journal d'Antoine Galland* 1.38–39, entry for 27 January.

<sup>34</sup> For Colbert's November 1674 letter of instruction for acquiring manuscripts, see de Laborde, *Documents*, 26–27; cp. Omont, *Missions* for fuller documentation of this trend. On clerics and antiquities, see Augustinos, *French Odysseys*, 85–92.

<sup>35</sup> La Guilletière [Guillet], *Athènes*, 222–230 and Spon, *Voyage*, 2.243–444, on Jesuit departure from Athens and Capuchin purchase of the 'Lantern'; also de Laborde, *Athènes* 1.71–81; Haussouiller, 'Athènes', 130; Frazee, *Catholics*, 124–125 (1658 given as the monastery's foundation).

<sup>36</sup> Robert de Dreux, *Voyage*, 135–156; Paton, *Chapters*, 13–17.

<sup>37</sup> Spon, *Réponse*, 231–232, quotes a letter from Antoine Galland, who had visited Athens in 1674, in which he criticises Guillet's reliance on the Capuchins as sources of knowledge about antiquities. Colbert, on the other hand, in the aforementioned letter of 1672 explicitly urges French consuls in the Levant to collect manuscripts and find 'either a Capuchin or someone else who is knowledgeable and therefore able to choose well': Omont, *Missions*, 1.223. On the awkward co-existence of mythologising and historicising approaches to the Holy Land, see Beaver, 'Scholarly pilgrims', esp. 273–274.

reverence for both pagan and Christian antiquity.38 The Capuchins of Athens had also insinuated themselves into local social and religious life and formed close relations with the aforementioned French, later English, consul Jean Giraud, who spoke Turkish, modern Greek and Italian, and had lived many years in Athens.39 In 1662 they had even celebrated Giraud's marriage to the daughter of the Palaiologos who facilitated Fr Robert's acropolis tour. In addition, the monks allegedly acted as confessors to Orthodox laymen and even clergy. Martin Crusius would have been appalled.

## **The marquis de Nointel's city portraits**

The first naturalistic portrait of Athens burst onto the scene in full sophistication, executed by an unknown artist (or artists) in the employ of Charles-François Olier, marquis de Nointel, Louis XIV's ambassador at the Sublime Porte from 1670 to 1679.40 Nointel was not only a successful diplomat, but also a flamboyant antiquarian and orientalist, with family links to the Port-Royal abbey. Nointel's foremost task in Constantinople was the re-negotiation of the Capitulations, involving both trading privileges for French merchants in the Ottoman Empire and French protection of the empire's Christians. Following his successful accomplishment of this charge and as an affirmation of French interest and protection, in September 1673 Nointel set off – with a suite of scholars, artists and staff – on an extended tour of the Aegean islands, Syria and Palestine, where he collected Greek and Arabic manuscripts as well as antiquities, inscriptions and coins. Along the way he also gathered intelligence about Ottoman-held cities and oriental Christian communities, but also information such as the state of city defenses that might be useful for the eventual, hoped-for Christian reconquest of the East. It seemed a not impossible dream in the 1670s and 1680s: surveyors and spies had been circulating in the region and only a decade after Nointel's Aegean tour, Louis XIV commissioned in 1685 a comprehensive, undercover re-mapping of the Aegean and eastern Mediterranean designed to advance France's geo-political and scientific advantage *vis-à-vis* the Ottomans. Led by the naval officer Étienne Gravier, marquis d'Ortières, this culminating secret mission produced scores of harbour and city views with detailed attention to walls and topography, made to a scale of 1 to 5000. These included, for example, Alexandria, Famagusta, Sidon and the seaside ruins of ancient Tyre, as well as Jerusalem and Athens. As we will see below, the plan of Athens was not original, but copied from a pre-existing city view.41 That

<sup>38</sup> Robert de Dreux, *Voyage*, 148–156, on the acropolis and Parthenon. On Giraud's 'mémoire', see the description of Athens kept among Nointel's papers, reproduced in Collignon, 'Documents', 60–71, to be read together with Collignon, 'Le Consul Jean Giraud', esp. 373–375, in which he reattributes the 'Relattion

des antiquittés d'Attènes dans l'estat qui se treuve à présent' from the Capuchins to Giraud himself. 39 Collignon, 'Le Consul Jean Giraud', 379. On Giraud, see La Guilletière [Guillet], *Athènes*, 84, 119, and Spon, *Voyage* 2.125–131.

<sup>40</sup> Meyer, 'L'Entrée', for a detailed discussion of the painting, its attribution and its historical context.

<sup>41</sup> Documentation from the Gravier d'Ortières mission is scattered between archives, including the Bibliothèque nationale de France, Archives nationales, Service historique de la défense in Vincennes,

*Fig. 9.3: Anonymous, The Marquis de Nointel at Jerusalem in 1674 (J.-L. Losi for B. Janson. © Oscar de laRenta).*

Ottoman domains could be reconquered was demonstrated (with great cost to the population and urban fabric) in 1687 when the Venetians famously bombarded the citadel of Athens, briefly taking it from the Ottomans fourteen years after Nointel's visit to the city. Fortunately, Nointel had taken the initiative to commission detailed drawings of the Parthenon sculptures and it is thanks to the Nointel Album that we may today admire on paper the intact, more than two-millennia-old work that spellbound the early travellers.

Near the end of his extended journey, in December 1674, Nointel wrote a letter back to Paris from Athens that encapsulates the three main areas of concern through which he saw the Levantine cities he visited: the state of the Christian communities, the conditions of commerce and the quality of the antiquities.42 Theatrical Nointel hired artists to record his ambassadorial performances in the Orient. We know he had four large canvases painted: one of his audience with the Grand Vizier on the occasion of the Capitulations renewal, another of the Holy Fire Ceremony at the Holy Sepulchre and portraits of himself which happen also to be superbly detailed city portraits of Jerusalem and Athens. Unfortunately the names of the artist(s) who painted the two city portraits are not known.43 The Jerusalem portrait had been

and even the Munich Staatsbibliothek. Two sets of maps of the Cyclades survive, in Thessaloniki and Vincennes, beautifully reproduced in the exhibition catalogue, Livieratos, *Αρχιπέλαγος.* See also de Laborde, *Athènes* 2.55–57. Bilici, *XIV. Louis*, publishes the French text of Gravier d'Ortières's report to

Louis XIV about how to take Constantinople. I thank Guy Meyer for correspondence about the mission. 42 Nointel's entire letter to the marquis de Pomponne from Athens is published by Vandal, *L'Odyssée*, 330–353, esp. 350–353 on Athens; Omont, *Missions* 1.193–194.

<sup>43</sup> Dispute over the identity of the artist(s) has continued since the discovery of the Nointel Album and the Athens portrait. For previous scholarship and discussion, see Meyer, 'L'Entrée' and Meyer, 'À la

nearly forgotten but came to light again in January 2019 during restoration work for a new Oscar de la Renta boutique in Paris – a coincidence wardrobe-conscious Nointel would have relished. In its current condition the enormous painting measures 2.87m × 5.35m, but both its height and width on both sides may originally have been greater (Figure 9.3).44

Jerusalem had rarely, if ever, been depicted in such finely-wrought detail. Careful study of this beautiful image of the walled city, bathed in morning light, has only just begun. Posing on the Mount of Olives, the magnificent representative of the Sun King and Protector of Christians employs the exotic, ancient Holy City as the stage set for his nearly-royal progress, lavishly recorded in order to be displayed in his Pera palace and eventually in Paris.45 From Nointel's letters we know that his visit in March 1674 coincided with Palm Sunday, a feast the Ottomans allowed the Latin church to celebrate publicly as a procession following Jesus's progress into Jerusalem from the Mount of Olives. The French embassy was timed to arrive only a few days before the procession, and the palm tree arching over Nointel's head registers this conjuncture. The west-facing view of the city takes in the Dome of the Rock and the Haram al-Sharif with the Church of the Resurrection, clearly flanked here by the later minarets of the ʿUmar and al-Khānqāh al-Ṣalāḥiyya mosques. Nointel actually entered through the Bethlehem gate, but artistic liberty allowed the adoption of this more conventional viewpoint, taken by the majority of city portraits both before and after Nointel's. Through this choice Nointel, not known for his modesty, is immortalised entering the city both armed and mounted along the same route as Christ. Guy Meyer has suggested that the portrait with its view of the walled-up Bāb al-Raḥma (Gate of Mercy), known to Christians as the Golden Gate, playfully evokes the legend that a Frankish prince would enter that gate to liberate the Christians and take back the city,46 an interpretation that recalls the intertwining of religious and political intent articulated in the *Civitates*-style city portraits. By the early seventeenth century city portraits were frequently employed to commemorate armed victory, two-part compositions with besieged cities in the background and violent clashes or victorious celebrations in the foreground. So conventional had these scenes become that it is

recherche' for thorough examination of the evidence for the identities and backgrounds of the artists known to have worked for Nointel, citing earlier discussions. Both city portraits would have been produced in the same atelier, if not by the same artist(s). Meyer convincingly argues that previous attributions to the French Jacques Carrey and Flemish Arnould de Vuez, both students of Le Brun, are unsustainable. Instead he attributes the Athens painting to another, anonymous Flemish painter in Nointel's employ, possibly in collaboration with Carrey.

<sup>44</sup> I thank Benoit Janson and his colleagues at Nouvelle Tendance for providing me with this photograph and their conservation report, with preliminary historical contributions by Raphaëlle Merle and Guy Meyer, see Janson, *Restauration*.

<sup>45</sup> Nointel writes to Colbert that it will be hung 'sur un endroit de la salle du palais de France', see Meyer, 'L'Entrée', 267.

<sup>46</sup> Meyer, 'L'Entrée', 268–269. The legend arises from Heraclius's triumphal return of the True Cross to Jerusalem, riding down to the city from the Mount of Olives on Palm Sunday in 630. For discussion with medieval sources, see Drijvers, 'Heraclius', esp. 178–179 with n. 12 and 186 with n. 49.

hard to understand Nointel's mounted portrait without an implied frisson that the Holy City might one day be restored to Christendom.

Even without these possible evocations, the scene corresponds to written accounts by members of Nointel's suite who record that the party was welcomed outside the city by the minister, or *musallim*, of the pasha of Jerusalem who was absent in Mecca at the time.47 The act of honouring Nointel, the King's ambassador, with the gift of a fresh and lavishly caparisoned mount frames the eternal city: on the left, staring directly at the viewer, is the exhausted silver steed still adorned with the precious textiles bearing the ambassador's family crest48, and on the right a placid Nointel offers the viewer his profile from his heroically rearing stallion. Nointel's hair and his horse's mane overlap with the north-east corner of the city wall, as if to interlock the scene in the foreground with the Holy City, almost dwarfed in the background. The *musallim* in his red turban and riding a light grey horse is pictured in rear view, already returning toward the city at the centre of the painting's lower register. Pictured is the moment when, honoured by Ottoman recognition, Nointel and his suite descended from the Mount of Olives to enter the city in what would appear to be a triumphal procession. That they were allowed by special privilege to bear arms would have been noticed by all spectators, Muslim and Christian. Grumblings by the Orthodox hierarchy did not go unnoticed by Nointel and his suite, and Christian clerics of any variety are absent from the scene as depicted. Louis XIV's delegation would naturally have seemed the embodiment of French aspiration to wrest control of the holy sites from the Eastern Churches (even if not resolving the fierce rivalries among Catholic orders in Ottoman territory).49

Slightly smaller than the Jerusalem painting, Nointel's second city portrait measures what is nonetheless an extravagant 2.60 m × 5.20 m (Figure 9.4). Like the Jerusalem view, the Athens painting is a composite, a mix of autopsy and convention, fusing contemporary fashions in mapping, mural painting, costume album figures and architectural drawing. As if in a nod to *Civitates orbis terrarum* conventions, the foreground is occupied by figures of various social groups: oriental stereotypes are joined by Nointel, dressed in a fur-trimmed Ottoman kaftan, French wig and wide-brimmed hat, members of his retinue and Capuchin monks present not only as cultural guides but as strategic state agents whose missionary efforts across the Aegean supported Nointel's presence as the representative of his king.50 The absence

<sup>47</sup> Meyer, 'L'Entrée', esp. 262–266, assembles accounts of the entry by Nointel's suite.

<sup>48</sup> Meyer, 'L'Entrée', 272.

<sup>49</sup> Meyer, 'L'Entrée', 267–274, offers identifications of the French delegation on the basis of extensive examination of the written evidence. Nointel reciprocated the Greek hierarchy's distaste. He reported back to Paris about 'l'instabilité, l'ambition et la vengeance qui règne absolument entre eux', see Vandal, *L'Odyssée*, 301–302. For an eye-witness account of confessional rivalry in the Holy Sepulchre the same Easter by the Jesuit Michel Nau and his diminishing comments on the Greeks in Jerusalem, see Armstrong, 'Journeying', 183–186. For an overview of Orthodox-Franciscan tensions in Jerusalem, see Frazee, *Catholics*, 59–60, 62–63, 145–148, esp. 147; Kitromilidis, 'Orthodoxy', 187.

<sup>50</sup> In 1672 Nointel wrote to Louis XIV that both Catholics and oriental Christians were in need of his protection as an 'asile assuré', see Omont, *Missions*, 179–180. On the intertwined religious and political

*Fig. 9.4: Anonymous, The Marquis de Nointel at Athens in 1674 (Chartres, Musée des Beaux-Arts, on permanent loan to the Museum of the City of Athens-Vouros-Eutaxias Foundation).*

of figures in authentic local dress brings us back to the question of the missing local view. Where we might have expected local inhabitants in the distinctive dress of Athens and environs, or even a local Orthodox priest (Nointel had, after all, attempted to extract a manuscript from the bishop), we find instead types recognisable from Ottoman costume albums already from the late sixteenth century.51 In the lefthand corner of the Athens painting we see two artists, or proto-archaeologists, whose presence served to authenticate the on-the-spot rendering of Athens with its rarely seen antiquities.52

The walled city occupies the centre, crowned by the inhabited citadel, the looming Parthenon with its delicate white marble minaret, the Frankish tower and the Propylaea. In the open countryside surrounding the lower town we see a few ancient and medieval buildings rendered disproportionately large, outside the walled settlement: the Temple of Olympian Zeus, the Church of Soteira Lykodemou and a

nature of the Capuchin mission in particular, see Papaïliaki, 'Des schismatiques', esp. 306–309.

<sup>51</sup> The Lambert de Vos Ottoman costume album in the Gennadius Library, Athens, dating to the 1570s, includes groups of figures that became widely stereotyped and appear readapted among the 'local' figures in the Nointel painting, for example women walking to the bath (fol. 72) and cavalrymen (fol. 4). On the de Vos images, as well as the production and use of Ottoman costume albums by European artists in the late sixteenth century, see Kynan-Williams, 'Play' and Kynan-Williams, 'Ottoman imagery'. Nointel collected clothing and accessories on his travels and asked his artists to draw portraits of people in local dress; see his letter addressed to Pomponne on 10 December 1673, from Naxos in Omont, *Missions*, 1.191; also Vandal, *L'Odyssée*, 476.

<sup>52</sup> For a detailed discussion of the setting and an interpretation of the scene at the far left as a verification of autopsy, see Meyer, 'À la recherche', 261–282. On the sketches of the city and Temple of Olympian Zeus in the Nointel Album, see Fowden, 'Parthenon mosque', esp. 85; Fowden, 'Rituals of memory'.

corner of the Ionic portico of Hadrian's reservoir on the lower slopes of Mt Lykabettos. This is not a portrait of an antique city, a city of ruins, but a living city that has accumulated temples, churches and mosques, as well as a compact and attractive assemblage of houses, some quite substantial and architecturally differentiated.<sup>53</sup> There is no tension between past and present, but a contrast is drawn between the animated foreground and the static city in the background, made to stand for the East, its antiquity and even its numinosity, over which Louis XIV's emissary stood as guardian and connoisseur. Reporting to Paris, Nointel would describe the monuments of Athens using the venerable local names handed down to him, such as the tomb of Socrates for the Tower of the Winds.54

Our appreciation of the Athens portrait is now increased by comparison with the Jerusalem view, most notably the air of a magnificent French presence in the East projected by the vast canvases, and commented on by contemporaries. Both portraits present not idealised but contemporary Ottoman cities in which Nointel represents Christian authority and protection. The absence of local Orthodox figures in both paintings is intriguing and accounts of his Athens visit too are peppered with disparaging comments about local clergy, their petty internal rivalries and their attitudes toward their Catholic counterparts.55 In style Martin Crusius and the marquis de Nointel shared little, but they were both deeply disillusioned by contemporary Greeks and their beliefs. In neither portrait are we shown any experience of the city or its inhabitants, represented only by the oriental figures at the fringes of Nointel's suite. The city is viewed from a distance, realistically detailed but also apart, an object to observe, or even a longed-for trophy. Both portraits bear the same message of an authoritative French presence in an oriental city: indeed, in a letter from Athens to Paris Nointel describes the 'temple de Minerve' as his 'arc de triomphe'.56

I have not yet mentioned the most striking feature shared by both Athens and Jerusalem, namely their walls, that most prominent feature of the sixteenth- and seventeenth-century city portrait. Two contemporary English visitors to Athens, George Wheler and the merchant Bernard Randolph, noted that Athens did not have a fortification wall: Athenians protected their city by building their houses and garden walls close together, and gates at the end of the streets were closed at night to protect the inhabitants from pirate raids.57 Seen from a distance this melange of

<sup>53</sup> Much closer analysis of particular buildings in comparison with other depictions is required. For a start see Bodnar *et al*., 'Church of St. Dionysios', 189–190; Kizis, 'Restoration', 137; Ameen, 'Küçük Cami'; Karidis, *Athens*, 1.75. 54 Vandal, *L'Odyssée*, 173; de Laborde, *Athènes* 1.122–123. See n. 68, below. 55 Meyer, 'L'Entrée', 264–266. See Nointel's letter to Pomponne on the Greeks, Vandal, *L'Odyssée,* 301–302.

<sup>56</sup> Letter in a private collection, Vandal, *L'Odyssée*, 173. 57 Randolph, *Present State*, 22 and Wheler, *Journey*, 346. Expectations from ancient sources and observation conflicted. Johannes Meursius was a crucial source, especially for Guillet and Spon. In his *Atticarum lectionum libri VI* (1617), 4, Meursius identified from the written sources the names of ten gates and in *Athenae Atticae* (1624), 175–184, he devoted two chapters to the walls and gates. His work retained the status of a reference work into the mid-nineteenth century. However, visitors even before Spon and Wheler had noted the decay of the ancient walls and the over-estimation of what survived

walls and gates could appear more linked up than upon closer inspection. We might conclude that by representing the city wall with idealised, sharp lines, Nointel's artist was motivated by the expectation that a city worth picturing (or conquering) would have walls, but possibly also by knowledge that the ancient city had walls, although this is unlikely given the painting's realism. Both portraits were drawn on site and executed in Constantinople. Despite degrees of idealisation present in both, they offer an unusually accurate picture of the two cities' urban fabric in 1674. But because neither was widely seen after its execution, the portraits were deprived of the influence they might have had.

## **The first published city view of Athens**

The first published portrait of Athens based on autopsy appeared alongside a short description of the city published in Lyon in 1674, the same year in which Nointel visited Athens (Figure 9.5). Fr Jacques-Paul Babin, a Jesuit who had been based in Negroponte and had visited Athens on five occasions, was asked by his fellow Jesuit, abbé Pecoil of Lyon, to write up a description of what he had seen. That letter was supplemented by antiquarian commentary by the Lyonnais medical doctor Jacob Spon and published as a small book. The multiple contributors to the *Relation de l'état présent de la ville d'Athènes* have inspired numerous ways of referring to the volume's authorship, or rather, editorship, best considered as Spon's. The drawing is commonly but inaccurately attributed to Babin. On the penultimate page of the Preface Spon states that he included a city view as well as some drawings of coins by Jean Vaillant, 'Antiquaire du Roi', who is known to have visited Athens, a vital piece of information that Guy Meyer has argued should be taken seriously as identifying the drawing's author.58 Consequently, I will refer to this Athens portrait as both the *Relation* portrait and the Vaillant portrait.

The walls are less tidied up than in the Nointel painting and correspond more closely to the eye-witness descriptions of adjoining garden walls encircling the settlement. Sufficient accuracy in the main features and orientation within the landscape gives the impression of a portrait made or dictated by someone who had seen the city. Intimations of scientific veracity mark the picture, not just the drawings of ancient coins, but also the banner's explanation, 'The city of Athens partly hidden behind the hill'. Only someone who had been there could explain that in order to be true to

according to accounts they had heard. Note, for example, Nicolas du Loir who in 1641 commented that the remains of the enclosure wall could not be made out, see Paton, *Chapters*, 63. Homolle, 'Vue d'Athènes', 519–520 points out the correspondence of the joined-up walls in the Nointel painting with the Verneda plan, executed in 1687. For varying descriptions of the walls of Athens from ruinous to strong in fourteenth- to seventeenth-century travelers' accounts, see Theocharaki, *Τα αρχαία τείχη*, 75–100, esp. 98. See also the discussion of the walls in the early depictions in Karidis, *Athens*, 58–63.

<sup>58</sup> Meyer, 'Non pas Athènes', esp. 289–290; Meyer, 'Un voyage', 15. I thank Guy Meyer for his correspondence about the attribution.

*Fig. 9.5: 'Vue d'Athènes dont une partie est cachée derniere la colline', drawing provided by Jean Vaillant and published in Jacob Spon,* Relation de l'état présent de la ville d'Athènes, ancienne capitale de la Grèce, bâtie depuis 3400 ans avec un abrégé de son histoire et de ses antiquités *(Lyon 1674) (American School of Classical Studies, Gennadius Library).*

how the city really looks, the artist has rendered only a partial view.59 In actual fact, the artist has rotated the acropolis so that we get the more interesting view from the west, with the citadel gate, the west entrance of the great mosque and the remains of the Odeon of Herodes Atticus (identified as the Areopagus). Although the artist has standardised the Propylaea/Frankish tower construction, the overall presentation is accurate. But there are obvious omissions, over-embellishments and inventions, such as the missing Soteira Lykodemou church outside the densely inhabited area, the oversized and crenelated gate, or the immense mosque dome looming over the outer row of houses. Despite its imaginative round arches, composite capitals, extra

<sup>59</sup> The effort to record accurate geographical orientation as a sign of authenticity is exemplified in two unpublished line drawings of Athens made by Francis Vernon on 26 August 1675. They can be found online among the digitised pages from his travel notebook in the Library of the Royal Society, London. Fol. 6 shows the city from the south-east; fol. 31 is a bird's eye view outline of the acropolis with the Saronic Gulf, its islands and the Peloponnese in the distance. On Vernon, see Walker, 'Frances Vernon'; Fowden, 'Parthenon mosque', 87–90.

windows and swirling patterns on the west facade, the citadel temple with walled-in columns and imposed minaret effectively sums up and fuses the building's ancient and modern phases.

The chosen vantage point (which was to be favoured by many future visitors) allows the artist to create a spatial distinction between the impressive ancient ruins in the foreground and the modern city in the background.60 The stock figures in eastern dress bind together the two planes of the ruinous ancient and oriental modern city. This distinction is also evoked in the title of Spon's compilation, whose full title brings together both the modern city and its ancient history: *Relation de l'état présent de la ville d'Athènes, ancienne capitale de la Grèce, bâtie depuis 3400 ans.* Labels on the city portrait give only contemporary, local names for ancient buildings, most notably the 'Palace of Hadrian', the same identification for the Temple of Olympian Zeus that is recorded by most contemporary visitors, including Antoine Galland who annotated the Nointel album drawings.61 The titles of both the book and the city portrait denote the beginning of a process of city viewing that involved visual excision and mental separating out of the ancient from the modern that was imposed by outsiders and eventually adopted by locals too.

### **The first known city plan of Athens**

Local names, rather than names derived from ancient sources, featured prominently on what is conventionally called the first plan of Athens, widely believed (though with little solid evidence) to have been drawn by the city's Capuchin monks. It was Léon de Laborde's conjecture, published in 1854, that the Capuchins drew up the plan themselves in response to the growing numbers of foreign visitors who asked them for guidance to the ancient city.62 Given the technical knowledge required to produce such a plan, the question of authorship (and transmission) needs to be reopened.63 But what concerns me here are the visual conventions of the three versions of the early plan that are known today. One was published in 1675 by Georges Guillet (also known as Guillet de Saint-George) under the pseudonym Sieur de La Guilletière in *Athènes ancienne et nouvelle* (Figure 9.6). Another known copy is a pencil drawing found among the papers of the aforementioned Gravier mission sent by Louis XIV to survey the geophysical and strategic features of Levantine ports and cities in 1685 (Figure 9.7).64 This second example, which I will designate the Gravier plan, diverges

<sup>60</sup> Another contemporary, eye-witness city portrait, known as the 'Bassano drawing' (*c*. 1670), depicts the citadel from Philopappos Hill, the best vantage point from which to capture the impressive medieval fortifications at the west end, see Tanoulas, 'Reconsidering', 60–65 with bibliography.

<sup>61</sup> Some descriptions call it simply a palace (Giraud's 'Relattion' in Collignon, 'Documents', 65 [*pace* his n. 3], see n. 35 above), others identify it with the Queen of Sheba, see Fowden, 'Rituals of memory'.

<sup>62</sup> de Laborde, *Athènes*, 1.76–81. 63 Guy Meyer is preparing a detailed study of the plan, its alleged Capuchin and other possible authorship.

<sup>64</sup> BnF MS français 7176, fol. 162, no. 34. Omont, *Athènes*, pls. XXXIX and XL with discussion 13–16; reprinted in de Laborde, *Athènes*, 1, opposite page 78, and see 77 n.1. The Gravier mission plan is sometimes wrongly associated with the engineer Plantier.

only slightly to the west of the acropolis from the plan published by Guillet in 1675.65 Guillet explains in the unpaginated Preface that Capuchin monks – he names pères Simon de Compiegne, Louis de Paris and Pierre François de Paris – had provided him with information that he supplemented with what he could learn from books in Paris since there were 'beaucoup d'Antiquitez que les gens du pays ne reconnoisent plus'. Guillet does not state that the Capuchins provided him with the plan. Regarding the relationship of the two versions of the early plan, de Laborde suggested that Guillet's plan represented an unfinished version of one provided to him by the Capuchins, while Louis XIV's engineer made his copy after later additions had been made to the Capuchins' copy in Athens.66 In 1901 J. R. Wheeler offered a slightly revised hypothesis, suggesting that Guillet might have been sent by the Capuchins 'some sort of imperfect sketch which showed the general lie of the land'. Wheeler suggests that Guillet would have used that sketch as the basis for the plan that he would have filled with features gleaned from ancient sources and published as his own drawing – the plan bears the attribution 'Guillet delin.' – in *Athènes ancienne et nouvelle*. 67 But were Guillet or the Capuchins, about whom so little is known, capable of creating such a scientifically executed (though imperfect) plan? The probability is overwhelming that the same plan, either the same original or copy (possibly copies) of it, lie behind both Guillet and Gravier plans, but for now the author of that original plan is best considered anonymous.

Our understanding of the puzzling process of creation and transmission is augmented, not satisfied, by a third version with a legend in French that was acquired in 1975 by the Benaki Museum in Athens (Figure 9.8) and not known to de Laborde or Wheeler, who were aware only of the Guillet and Gravier plans.68 The anonymous coloured, engraved plan is identical to the Gravier plan, with a significant exception: in the Benaki plan the city blocks are filled in with houses, sketched in less detail than the major buildings and represented in bird's eye view perspective. We are reminded of the *Civitates*-style conventions for urban portraiture in which both prominent and unremarkable buildings are represented, even if the later are often generic. At the heart of the *Civitates* conventions from which the first Athens plan emerged is the detailed view of a contemporary city, albeit without inhabitants in the streets – as we

<sup>65</sup> See Wheeler, 'Notes', for detailed comparison of the two plans. 66 de Laborde, *Athènes*, 1.231–232, n. 2. There is no evidence that an original plan was kept in Athens, this too is de Laborde's conjecture.

<sup>67</sup> Wheeler, 'Notes', esp. 222 and 230.

<sup>68</sup> Anonymous, 'Plan d'Athenae', Benaki inv. no. 22955, acquired in 1975 (for the acquisition date I thank Polina Kosmadaki, Benaki Museum. No further information is recorded). The engraving is illustrated without discussion in Tsigakou, *Η Αθήνα*, 107, fig. 88, where it is described as 'from the publication *Athènes Ancienne et nouvelle et l'état present de l'Empire des Turcs par le Sieur de la Guilletière*, Paris, 1675', which is incorrect since, as described above, the perspective rendering of the city in the Benaki coloured engraving is not identical to the plan published by Guillet. The Benaki plan is also illustrated in Koutsogiannis, 'Η εικόνα', 85. Lest confusion arise, it should be noted that fig. 115 in the original 1960 edition of Travlos, *Πολεοδομική εξέλιξις των Αθηνών* was the Gravier plan. The Benaki plan was substituted in the 1993 reprint without any alteration in the attribution*.*

have seen, more often local figures in the foreground, a feature the Nointel artist(s) retained. But both Guillet and the Gravier copyist have flattened out the city blocks which are presented as greyed-in spaces from which rise only a selection of notable buildings, depicted three-dimensionally as in the Benaki plan. It is striking that the Gravier plan picks out exactly the same buildings as Guillet's, a shared feature that may well point to a common source copied by both.

The relationship of the Benaki engraving to a conjectured original plan is not known and the acquisition record reveals nothing of its origin. At present we cannot prove or exclude the possibility that the Gravier draughtsman made his pencil drawing from the Benaki engraving, or that both the engraving and Gravier plan were copied from a 'master' plan which was more complete than whatever was available to Guillet, or which Guillet modified to fit his own interests and erudition. Regardless of the uncertain relationship between the three plans and their origins, we should not lose sight of what is especially important for our purposes, namely that in the hands of Guillet and the royal engineer much of the local detail (even if ultimately generic) filled in by the Benaki mapper has been erased in the process of reproduction. The Benaki portrait would then allow a glimpse into a palimpsestic process of mapmaking by which earlier visions of the city are rubbed away and partly redrawn to respond to evolving needs and visual tastes.

*Fig. 9.6: Engraving of the first known plan of Athens, published by Georges Guillet,* Athènes ancienne et nouvelle, et l'estat present de l'empire des Turcs, contenant la vie du Sultan Mahomet IV., le ministère de Coprogli Achmet Pacha, Grand Vizir, & son campement devant Candie. Avec le plan de la ville d'Athènes, par le Sr. de la Guilletière *(Paris 1675) (American School of Classical Studies, Gennadius Library).*

*Fig. 9.7: Gravier d'Ortières mission, 'Plan d'Athenae', pencil drawing,* c. *1685. 162v, in* Estat des places que les princes mahométans possèdent sur les côtes de la Mer Méditerrannée et dont les plans ont esté levez par ordre du Roy à la faveur de la visitte des Eschelles de Levant, que Sa Majesté a fait faire les années 1685, 1686 et 1687, avec les projets pour y faire descente et s'en rendre maistres *(Bibliothèque nationale de France).*

*Fig. 9.8: Anonymous, 'Plan d'Athenae', coloured engraving,* c*. 1670 (Benaki Museum, Athens).*

What we see in all three versions of the plan is a complex city portrait which depicts the contemporary shape of the city as a street plan seen from above set within a topographical surrounding seen from an oblique, bird's eye view. The countryside surrounding the city plan is drawn with great attention to detail, creating the impression of a careful survey while remaining evocative of a pastoral world (without shepherd or sheep). The topography suggests that the viewer's gaze is west-northwestward with the city held in place by roads whose destinations are the only features identified by words. The roads suggest connectivity with other settlements, implying a wider social and commercial context.

The maker adopted a different strategy from Vaillant's to the problem of picturing the whole of the settled area below the citadel. Here the citadel and city are rotated away from each other, a choice that misrepresents the more intimate relationship of the two, but also allows the artist to separate the city plan out pictorially. The inhabited citadel, itself a miniature settlement dominated by its mosque (labelled by Guillet as 'Le Temple de Minerve, Sainte Sophie, Mosquée'), is part of the obliquely viewed landscape, while a completely different representational mode is adopted for the lower city, depicted as a city plan seen from overhead. Even with the flattening of the urban profile in the Guillet and Gravier plans, the lingering impact is discernible of the *Civitates* standard for what a city should look like: the densely built up city centre with its streets, the roads leading from it into the countryside, which is depicted in bird's eye view perspective imagined as if hovering over the foothills of Mt Hymettos, with the Ilissos river and hills of Ardettos in the foreground.69

The Gravier copy is the simplest and does not identify any of the buildings it features. The Benaki plan notes thirty-two buildings by numbers identified in a short legend on the right side of the plan. In contrast to the Vaillant portrait where only contemporary names were given, on the Benaki plan the Parthenon and Hephaisteion (as we know them today) are not identified by their contemporary use as the citadel mosque and church of St. George, respectively. Instead what is recorded is the contemporary understanding of what they had been in the pagan past: temples of Minerva and Theseus. The process of sifting through local and current identifications to find the 'authentic' is captured in the choice of names. Despite this, there is no obvious attempt to select out only ancient monuments in the Benaki plan. Instead we find a mix of buildings, such as 'Église Grecque', 'Monastère', 'Ancien monastère des capucins', 'Mosquée', 'Serail du Serdar', as well as the 'Didascalion, Palati tou Adrianou', local names for the Temple of Olympian Zeus.70

All identifications on the Benaki plan are also found on Guillet's amplified list on which he often provides supplementary information. Guillet's foldout plan is scattered with numbers and letters keyed to an extensive legend with 185 identifications from multiple periods.71 We can trace Guillet's use of the plan to build up his own cumulative approach to the city's toponymy using the example of the

<sup>69</sup> Cp. Curtius and Kaupert, *Karten von Attika*, 1, pl. 4 'Athen-Hymettos'.

<sup>70</sup> La Guilletière [Guillet], *Athènes*, 351, no. 115 reproduces the same name. Original spelling retained.

<sup>71</sup> La Guilletière [Guillet], *Athènes*, 345–353; reprinted in de Laborde, *Athènes*, 228–231. de Laborde discovered in what was then the Bibliothèque Impèrial, in a different file from that containing the Gravier plan, a list entitled 'explication de la nouvelle Athènes. Les chiffres sont derrière la carte', written in a seventeenth-century hand. Close correspondences with Guillet's identifications led de Laborde to suggest that the 'explication' had been provided to Guillet by the Capuchins, see de Laborde, *Athènes*, 1.77–80, esp. 80 n. 5.

only building within the urban tissue to be identified in writing on the Benaki plan, as opposed to its marginal legend. On the street leading leftwards (southwards) from the open square towards a polygonal building we read the local name 'Anemi', the Tower of the Winds, which Nointel mentioned by another, probably older, appellation, the tomb of Socrates.72 Retaining this name and building on it, Guillet's no. 34 reads 'Tour d'Andronicus Cyrrhestes. Anemoi [with corrected orthography] Maison des Vents'. Guillet's consistent practice is, where possible, to add information about buildings and places that is derived from ancient texts and to combine this with local traditions belonging to the world of local legend, a mix of half-remembered or reimagined appellations, such as 'Tour de l'Arcenal de Lycurgue' for the Frankish tower in the Propylaea,73 or the 'Arc de Trajan'74 for the Philopappos monument. Domestic architecture also featured, such as the 'Palais du Cadi'75, the Houses of 'Dimitrios Beninzellos'76 and 'Calchondile',77 which Crusius's informant Daniel Palaiologos had also noted on his sketch map of Athens. Guillet's work, in both his text and his boldly expanded annotation of the early plan, was a composite of old and new that made space for multiple views of Athens – the modern Greek, Catholic, Muslim city and the classicising antiquarian one.

Who, then, was Georges Guillet and why did he publish the map with these accumulative labels? Guillet was a somewhat shadowy figure who emerged from the Parisian theatre world into the Royal Academy of Sculpture and Painting, which he would serve as its first *historiographe*. He was gifted at cultivating connections, not only with leading politicians, intellectuals and artists such as Jean-Baptiste Colbert and Charles Le Brun, but also Capuchin missionaries. They provided from Athens detailed descriptions of the city that Guillet used as the eye-witness bedrock of his *Athènes ancienne et nouvelle*, which would be fabulously successful and go quickly through multiple French editions from its original publication in 1675, and a prompt English translation in 1676. Guillet shared with Martin Crusius a capacious approach to source material, drawing from comments on liturgical and ritual practices, food and clothing and language, as well as, in Guillet's case, the modern urban layout of Athens and its antiquities in continued use. But while Crusius was a meticulous

<sup>72</sup> Socrates was the most common association with the building, attested in writing already in the mid-fifteenth century text by the so-called Vienna Anonymous (de Laborde, *Athènes* 1.17) and two centuries later in the description associated with Giraud (Collignon, 'Documents', 61), though both describe the monument as a school. In the 1660s Robert de Dreux, *Voyage*, 145, recorded the building's association with Socrates, while the Ottoman traveler Evliya Çelebi associated it with Plato, and the early eighteenth-century local historian, Mahmud Efendi, with Socrates, see Fowden, 'Parthenon mosque', 88, n. 87 and 90, n. 101. Cp. La Guilletière [Guillet], *Athènes*, 216. We await comparison with the anonymous, unpublished mid-seventeenth-century Ottoman travel account (BNF Supplement turc 1027), an edition of which is under preparation by Machiel Kiel, see Kiel, 'Quatrefoil plan', 121, n. 37. 73 La Guilletière [Guillet], *Athènes*, 346, no. \*. 74 La Guilletière [Guillet], *Athènes*, 350, no. 99.

<sup>75</sup> La Guilletière [Guillet], *Athènes*, 347, no. 27.

<sup>76</sup> La Guilletière [Guillet], *Athènes*, 347, no. 24 and 347, no. 23 'Janis Beninzellos'.

<sup>77</sup> La Guilletière [Guillet], *Athènes*, 347, no. \* and 347, no. 44 of 'Stamatis Paleologue'.

scholar compiling evidence for the current state of Greek Orthodoxy in the Ottoman world (of which Athens was just one small part), Guillet was an erudite and inventive literary artist who went a step beyond compiling evidence to create what we might almost call an experiential guide to the city. By publishing the richly annotated plan of uncertain origin, Guillet transformed the plan's intended use as a practical guide into the object of an armchair reader's gaze.

Guillet was somewhat unusual among the early describers of Athens for his arguably less judgmental attention to the Orthodox and Islamic character of the modern city.78 Guillet's book is fascinating and relatively neglected for many reasons, but one notable feature of the work is that he aspires to give voice to local Greek inhabitants of Athens (even if his stories are not first-hand). He relates, for example, a long but perfectly-paced story in which an Athenian monk upbraids a party of overweening English, Italian and German travellers for coming to 'teach' the local Greeks about their great ancestors from whose height they have fallen so far.79 In some ways the dense urban tissue of Nointel's city portrait with its churches, mosques and ancient buildings better illustrates the holistic, multifaceted quality of Guillet's written account than the early city plan, in which we witness the beginning of the process by which impressive, mainly ancient buildings are picked out, and unremarkable ones omitted, turned into the plan's grey spaces. The absence of human figures is also incongruous, since Guillet's literary city is animated by human encounters. But access to a plan of contemporary Athens was a major publishing coup and Guillet's lively and detailed descriptions made up for some of the blanking out of the urban fabric.

Nevertheless, an incipient schizophrenia between ancient and modern seeps into the plan through the partial erasure of contemporary houses in the Guillet and Gravier versions. In this practice the portrait resembles maps of European cities which already used this greying technique to represent the city by its main streets, distinctive buildings and formally demarcated spaces, for example in Gomboust's celebrated 1652 plan of Paris.80 For future depictions of Athens this precedent aided the mental process of scraping down the present to find the ancient city.

#### **Responses to the early portraits**

Vaillant's *Relation* portrait and Guillet's plan formed the visual starting point for subsequent depictions of Athens. Jacob Spon, who had coordinated the publication of

<sup>78</sup> Fowden, 'Parthenon mosque', 88–89. See Augustinos, *French Odysseys*, 62–64 on the more catholic interests in both contemporary and antique among travellers before Jacob Spon. The controversy between Spon and Guillet has overshadowed appreciation of Guillet on his own terms. He deserves independent study. 79 La Guilletière [Guillet], *Athènes*, 233–249; Spon, *Voyage*, 2.152, mentions the story and corrects Guillet.

Guillet defends his inclusion of the story in the Preface to the subsequent editions, without naming Spon.

<sup>80</sup> Available at: https://upload.wikimedia.org/wikipedia/commons/0/05/1652\_Gomboust\_Map\_of\_ Paris%2C\_France\_%28c.\_1900\_Taride\_reissue%29\_-\_Geographicus\_-\_ParisSm-gomboust-1900.jpg.

Babin's description of contemporary Athens, visited the city himself in 1676 together with his English traveling companion George Wheler. Like most Europeans who visited Athens, Spon and Wheler arrived via Constantinople, and there they had visited Nointel and seen his artists' drawings.81 Spon's extended account of Athens included a few architectural drawings and a city view (Figure 9.9).82 When creating his own depiction of Athens, Spon built on the Vaillant view which he had commissioned for the Babin volume, as well as the plan which he knew from Guillet's book.

Vaillant's drawing seems to have influenced Spon's attempt to recreate the way in which part of the city is hidden behind the citadel when viewed from the south-east. Spon eschewed the distorting swerve by which the mapper behind Guillet made the city plan visible by separating it from the citadel, instead firmly juxtaposing plan and citadel. At the same time, he follows the Guillet plan by maintaining the bird's eye view and its stark contrast with the lower settlement represented as an urban plan. Spon has reduced the number of features that appear in the surrounding countryside and added some, such as the 'Lyon de marbre' and 'Temple de Thesée' that are not on Guillet's version, but do appear on the Benaki and Gravier plans.83 For Spon the early plan rendering offered an authentic grounding which he adapted as he added his own observations made *in situ*.

Retaining the the early plan's orientation also made Spon's corrections of Guillet's misidentifications – road destinations, for example – more apparent to anyone who compared their two versions. Spon and his traveling companions began to suspect Guillet had not seen the city himself, so the act of correcting was an act of disproof.84 Spon's plan appears as an accompaniment to his detailed description of Athenian monuments in which he does not spare his disparaging comments about local interventions.85 For example, Spon mentions the 'miserable' Greek frescoes painted inside the temple of Artemis Agrotera by the Ilissos, labelled 35 on his plan. Spon labels far fewer features, cutting Guillet's 180 down to 50, leaving only a few contemporary churches, mosques and houses and highlighting ancient monuments. Spon reduces and pays much less attention to the countryside except for roads and major features such as olive groves, again represented in close correspondence to what we find in Guillet's copy. Spon uses his 'explication' to express his hesitancy over contemporary identifications that he is not able to correct to their original

<sup>81</sup> Spon, *Voyage*, 1.262–264. On Athens in 1672–1682, see Yakovaki, *Ευρώπη*, 245–305.

<sup>82</sup> Wheler, *Journey*, 338 and 340, published his own line drawings of Athens, viewed from the South and the North, neither of which responds to the plan published in Guillet, unlike Spon whose plan aimed to prove his greater accuracy over Guillet.

<sup>83</sup> Spon's inclusion of these argues in favour of de Laborde's theory that Guillet, who never visited Athens, was working with an unfinished plan and had to fill in the area north-west of the acropolis from literary references, while the Benaki, Gravier and Spon versions all represent what could be seen on the ground in the 1670s. 84 On this controversy, see Omont, 'Athènes', 282–289; Constantine, 'Question', esp. 1–9; Augustinos,

*French Odysseys*, 109–115.

<sup>85</sup> Spon, *Voyage*, 2.162; cf. Wheler, *Journey*, 379.

*Fig. 9.9: Jacob Spon, plan of Athens, in* Voyage d'Italie, de Dalmatie, de Grèce et du Levant, fait aux années 1675 et 1676 *(Lyon 1678) (American School of Classical Studies, Gennadius Library).*

ancient names, adding distancing phrases like 'they say'.86 He omits noting that the Lysicrates monument had been incorporated into the Capuchin monastery, an identification that appeared in Guillet (and the Benaki plan). The blanking out of modern urban detail in Spon's plan is an example of what I would call 'visual aphaeresis', a form of erasure, selectively seeing and representing a city to fit a particular purpose or ideology. Aphaeresis is an extreme representational mode practiced both by people who had seen the city with their own eyes and those who had not.

But Spon not only subtracts, he also adds. Spon's major departure from the earlier portraits was to depict the walls of Athens as a circuit with what the viewer is left to imagine might be towers. There are gates, but no walls were shown on any of the three versions of the early plan, instead Athens seems to be ringed by a road. The Nointel painter produced sharp, almost Venetian-style walls, while in the Vaillant drawing the walls appear more like garden walls. Spon's walls and above all his suggested towers seem to run counter to his own scientific insistence on autopsy. Instead they respond to the expectation that an ancient city would have had walls, and to his inclination to render the ancient city at the expense of the modern.

<sup>86</sup> For instance, 4. 'Tour qu'on pretend être de l'Arsenal de Lycurgue; 5. Eglise Panagia Spiliotissa, dans une antre du rocher qu'on pretend être celuy de Creuse,' followed by '6. Veritable endroit où est la grotte de Creuse à la muraille *Pelasgicon*.' Where he relates the unqualified local identification '7*.* Fanari tou Dimosthenis' he keeps it in (faulty) demotic to manifest his distance from the attribution.

#### **The city as theatre of war**

Walls were more than the battleground of the antiquarian controversy between Spon and Guillet. Real armed conflict over the Morea, and the hoped-for reversal of Muslim control to Christendom's advantage, imparted practical purpose to these early depictions. Athens' putative city walls were embellished beyond Spon's wildest dreams by two of the best known artists to project Venetian power against the Turk, casting Christian good against Muslim evil. Cartographer Vincenzo Coronelli and engraver/ printer Giovanni Giacomo de Rossi both reworked known portraits of Athens, guided by oral accounts and drawings by engineers and other participants in the Venetian assault.87 They recycled in order to insert, enhance and abstract features of the city that would project their political message, inadvertently continuing the palimpsestic practice that had been the fate of the city plan from an early stage.88

The dual forces of missionising religion and political geography enlivened the career of Vincenzo Coronelli, who despite his ambitious tracking of the heavens and the New World did not travel widely, but spent his life in Venice, Paris and Rome. Coronelli rose to be Vicar of the Franciscan order in Rome and Cosmographer of the Venetian Republic, and (to bring us back to the Sun King's territorial ambitions) presented two elaborate globes to Louis XIV in 1683.89 A few years later Coronelli's city portraits propagated support for Morosini's Catholic campaign to recapture the Morea from Ottoman control.Having never seen the Morea himself, Coronelli adapted to his purposes three city views of Athens I have discussed: Vaillant, Guillet and Spon.90 Since Coronelli's work served to spread Venetian propaganda across Europe, his reworking of these three city portraits served in turn as templates for subsequent Athens views that would be retraced and elaborated, often by artists who had never seen the city themselves. But as Coronelli's work has been more frequently discussed, I will focus on de Rossi's view (Figure 9.10).

<sup>87</sup> On Coronelli, see Navari, 'Vincenzo Coronelli'; Wheeler, 'Notes' and 'Coronelli's maps'; Della Dora, 'Mapping'. Some dozen portraits of the city under bombardment in 1687 survive, made by draughtsmen, mainly engineers, of various nationalities under Morosini's command. Giraud, 'War', includes high quality reproductions and discussion. Verneda's work is especially highly valued for its accuracy, see also n. 57 above. Older reproductions and discussions of city views from the time of the campaign include, Omont, *Athènes*, ills. 32–37 and Bowie and Thimme, *Carrey drawings*, 30–36, as well as Tsigakou, *Rediscovery*, 111–112 and Hadjiaslani, *Μοροζίνι*. Other depictions include a rare broadside printed in Venice by Girolama Albrizzi, Efstathios Finopoulos Collection (Benaki Museum); a fold-out drawing of the city contained in a rare Italian description of the city of Athens in 1687 (Gennadius Library MSS 247); a unique watercolour of Athens at the time of the Venetian campaign that normally hangs, together with the keys to the citadel given by Francesco Morosini to Count Tomaso Pompei, governor of Athens, in the Villa Sagramoso-Pompei, near Verona. The latter was exhibited in 'Francesco Morosini and his legacy in Venice, Athens and the Morea', curated by Maria Georgopoulou and Alexis Malliaris, Gennadius Library, Athens (19 November 2019 – 29 February 2020).

<sup>88</sup> Cosgrove, 'Global illumination', on Coronelli's work as universal, yet parochial; entrepreneurial, yet recycling.

<sup>89</sup> Cosgrove, 'Global illumination', 35–36.

<sup>90</sup> Coronelli selection: https://eng.travelogues.gr/travelogue.php?view=173&creator=961916&tag=8852.

*Fig. 9.10: Giovanni Giacomo de Rossi,* Teatro della Guerra contro il Turco *(Rome 1687) (Efstathios Finopoulos Collection, Benaki Museum).*

De Rossi's volume of twenty engravings entitled *Teatro della Guerra contro il Turco* is a luxurious testament to Morosini's glorious reconquest of Greece for Christendom. De Rossi based his work on pre-existing views to produce his portraits of cities as theatres of war, depicting Venetian forces wresting control from the Ottomans. But he also innovates, intruding human figures into Guillet's plan, either single figures on foot in the city streets and on the country roads leading into the city, or in conventional squadrons of pike- and banner-carrying soldiers. His portrait of the campaign shows city gates and walls in use as access to the besieged citadel. Despite its high technical quality, we might ungenerously describe de Rossi's composition as a rendering of a battle scene put together from a borrowed, lifeless plan and animated by three-dimensionalised monuments and stick figures.<sup>91</sup> He remains remarkably true to the early plan's juxtaposition of a perspective rendering of the citadel and surrounding countryside with the urban plan, although perpetuating Guillet's greyed-out city blocks.

<sup>91</sup> De Rossi's introduction of figures was conventional for the time; for example, see two early eighteenthcentury portraits of Mecca by European artists in Beyazit, 'Defining Ottoman realism', 214–217 with nn. 31–33, figs. 6 and 7; one portrait is briefly discussed and illustrated in Tütüncü, 'Uppsala Mecca painting', 60–61, fig. 15.

De Rossi exercised considerable freedom in his inscriptions, expanding the banner and adapting Guillet's legend to fit into a continuous legend at the bottom of the page. He includes 130 identifications, a number close to Guillet's, far more than Spon's and, like Guillet's, a mix of ancient and modern drawn from local legends, ancient texts and contemporary use.92 To the right of the legend siege positions are identified, reminding us that this is not Athens imagined as a ruin-field of antiquities, but as contested territory between rival empires. By superimposing a depiction of the siege on what was designed as a tourist map, de Rossi engineers a particularly poignant visual collision. The soldiers are as stereotyped as the oriental figures that appear in the Vaillant and Nointel portraits. Yet their presence reinforces how empty of life all the early city portraits of Athens have been – with the major exception of Nointel and his companions, outsiders enjoying a day's excursion.

## **The Ottoman city alive and the antique city restored**

De Rossi's illustration of the glorious reconquest of Athens for Christendom unsentimentally depicted the citadel's bombardment by Morosini and his mercenaries, for whom wresting the city from the Ottoman rival was worth the cost of destroying the well-preserved ancient temple in its most accreted form.93 In 1687 the garrisoned citadel was not an archaeological preserve, but a living part of a city whose Orthodox Christian inhabitants found themselves caught between Venetian and Ottoman ambitions. The city's past had not yet been separated out from its present to be preserved as relics and survivals. The men who represented Venice and Paris in the struggle against the Turk had not yet cultivated the boldly aphaeretic eye of the antiquarian or scholar who saw through the city's accumulated layers and uses to the 'original', the practice better attested in the drawings of Cyriac of Ancona and Jacob Spon.94 After the citadel mosque's explosion and the subsequent Ottoman surrender, Venetian control of the city lasted less than a year. Eventually Ottoman order was restored and as the numbers of European visitors increased, so did the city portraits. Now representation developed into separate genres, in keeping with European fashion: picturesque city vignettes that focused on ruins, architectural drawings of single monuments, formal city plans.95 The depictions of Athens by European artists and travellers from the late eighteenth century onwards have been much discussed. The city views are part of the European process of self-discovery and

<sup>92</sup> The anonymous Bonn drawing, a partial city view from the years before 1687, also includes local identifications, such as the schools of Plato and the Peripatetics, see Omont, *Athènes*, pl. 29; Necipoğlu, 'Visual cosmopolitanism', 9–10, fig. 4.

<sup>93</sup> de Laborde dedicates the second volume of *Athènes* to materials relating to the 1687 siege and aftermath. 94 On the quest for the 'original' in city histories, see the chapter by Edward Zychowicz-Coghill in this volume.

<sup>95</sup> For recent overviews of eighteenth- and nineteenth-century trends, see Koutsogiannis, 'Η εικόνα', 103–133 with bibliography; Korres, *Οι πρώτοι χάρτες*; Karidis, *Athens*, especially chapter two, illustrations (no page numbers) and 103–114; Kaika, *City*, 92–105.

self-invention galvanised by the cluster of travellers who visited Athens in the 1670s. These eighteenth- and early nineteenth-century city views add little to the present discussion other than to drive home the hegemony of the outsider's view. While the numbers of European visitors grew, Athens would remain in Ottoman hands until the 1821 uprising when warfare again stimulated depictions of Athens. Not until 1836–1839 when Strategos (General) Ioannis Makriyannis and his artist Dimitrios Zographos portrayed Athens as the war theatre between Greeks and Turks did Athenians finally 'need' a locally conceived and executed portrait, teeming with life and death.96 It is with the problem of the missing local view that has recurred throughout this study of the early city views that I will conclude.

## **The first Greek portraits of Athens**

In 1829, Ioannis Makriyannis – revolutionary against the Turks and patriotic supporter of the Hellenic constitution – began writing his *Memoirs* in a leather-bound notebook which he kept hidden in a tin box. Soon after he began writing, Makriyannis came up with the idea to 'dictate' ('υπαγορεύει') to a painter the battles and allegories he would describe, thereby creating what he called a history in images.97 The collaborative method calls to mind Martin Crusius sketching from Daniel Palaiologos's dictation. These annotated and detailed pictures would, he hoped, provide a direct and undistorted account for the many illiterate Greeks whom he wanted to inspire and inform about the sacrifices made for 'homeland and religion', as if they were seeing events with their own eyes.98 Makriyannis states clearly that the pictures were designed to disprove the 'lies and deceptions'99 in circulation (one is again reminded that Crusius wanted a portrait to disprove the rumours that the city of Athens no longer existed). Makriyannis first employed a 'Frangos' to do the work, but they could not communicate and Makriyannis did not like the three paintings he produced so he paid him and sent him away.100 Makriyannis writes that he then found a war veteran and icon-painter named Panayiotis Zographos who was able to

<sup>96</sup> The uprising also inspired the first known Ottoman plan of Athens, annotated in Ottoman and dated to 1827, according to the most probable reading. A full publication is awaited; for a preliminary discussion, see Stathi, 'Putting Athens on the Ottoman map'; Stathi, 'Carta incognita'; Ambraseys, 'Ottoman archives', 26–31. Unlike the Venetian maps produced in response to the 1687 conflict, this one does not combine antiquities with warfare. A few ruins are labelled but neither they nor any other topographical features such as gates are identified by local names, a striking indication that the map was not designed to record the past, or local knowledge, but to aid the citadel's capture.

<sup>97</sup> An inscription on the bottom border of picture no. 24 of the Gennadius watercolor series (Georgopoulou and Papadaki, *Ιωάννης Μακρυγιάννης*, 165) records: 'εσχεδιάσθη κατά στοχασμόν, υπαγόρευσιν και ιδίων εξόδων του συνταγματάρχου Μακρυγιάννη᾽ ('drawn according to the thoughts, dictation and at the personal expense of colonel Makryiannis'); transcribed in Papadaki and Tsokani, 'Οι πίνακες', 164.

<sup>98</sup> 'Πατρίδα και θρησκεία᾽ is the memoir's mantra, see especially his Prologue and Introduction, Makriyannis, *Απομνημονεύματα* vol. 1, 15–17, 19–23; Holton, 'Ethnic identity', 149–160.

<sup>99</sup> 'Ψευτιές και χαμέρπιες᾽, Makriyannis, *Απομνημονεύματα*, bk 3, ch. 4, vol 2, 115.

<sup>100</sup> Makriyannis, *Απομνημονεύματα*, bk 3, ch. 5; vol. 2, 130–131.

portray figures, settings, narrative and symbolism in an idiom that complemented Makriyannis's own vision.101 Between the two of them Dimitris Zographos and his son Panayiotis executed 125 paintings between 1836 and 1839, a first series of twenty-five tempera on wood panels (like icons) and then four sets of copies in body colour on paper which Makriyannis had made as gifts for the sovereigns of England, France, Russia and Greece.102 Three of the numbered paintings on paper, which Makriyannis called 'εικονογραφίες' ('iconographíes', literally, writing in images),103 picture Athens and its citadel: 10 'Μάχη πρώτη των Αθηνών' ('The first battle of Athens'), 17 'Μάχη εις Ανάλατον' ('The battle at Analaton') and 19 'Η πολιορκία Αθηνών' ('The siege of Athens'). The lower margin of each painting has a legend that identifies names of people, places and actions depicted keyed to numbers on the picture.

'The first battle of Athens', no. 10, (25 April 1821 – 9 June 1822, O. S.) depicts the second siege of the Athenian citadel, which the local Turkish population finally surrendered (Figure 9.11).104 The number 1 inscribed on the Parthenon's west pediment corresponds in the legend below to 'The Acropolis of Athens held by the local Turks᾽ ('οι εντόπιοι Τούρκοι'). Zographos has shifted the Parthenon 45 degrees so that we see the west end but not the domed mosque rebuilt inside the bombed-out temple after the Venetian attack. Number 7 identifies 'part of the houses of Athens', 'Hadrian's Gate' and 'Columns of Olympian Zeus' on the right side of the battle scene, and two more stone-built ancient buildings are identified in the legend by their contemporary names: the Séntzos for the Philopappos monument and the Serpetzés for the fortification incorporating the Odeion of Herodes Atticus. In the foreground women in local costume (not stereotyped from costume albums) tend the wounded and visibly mourn the dead.

The battle at Analatos was a critical battle joined on 24 April 1827 (O. S.) on the level plain known as Analatos, today built over by the neighbourhoods of Neos Kosmos and Nea Smyrni. In 'The battle at Analatos', no. 17 (Figure 9.12), the vantage point from which the scene is viewed is Mt Hymettos, not far, in fact, from that adopted in the

<sup>101</sup> The relationship between Dimitrios and Panayotis Zographos (father and son respectively) and their pictures was long debated by Greek scholars. It is now agreed that Dimitrios painted in collaboration with Makriyannis between 1836 and 1839, quite possibly assisted by his son Panayotis, who from 1837 was a student at the newly founded Higher Technical School (Ανώτατη Τεχνική Σχολή), established by King Otto in 1836, later known as the Athens School of Fine Arts. A critical discussion is Fenerli, 'Ζωγράφοι', with supporting archival material; for bibliography and recent discussion, see the 2018 exhibition catalogue edited by Georgopoulou and Papadaki, *Ιωάννης Μακρυγιάννης*.

<sup>102</sup> Eight of the original, unsigned, pictures on wood survive, housed in the National History Museum in Athens. Two complete sets of the twelve pictures also survive, signed 'hand of D. Zographos' and 'thought (στοχασμός) of Makriyiannis', in the Royal Collection in Windsor Castle and the Gennadius Library, see Asdrachas, 'Μακρυγιάννης᾽, 23. On the Windsor Castle set, see now Mavrikiou and Petropoulos, 'Η Ελληνική επανάσταση'. The Gennadius set was made for King Otto and purchased by Ioannis Gennadius in 1909, two years after the publication of the Vlachoyannis edition of the *Απομνημονεύματα*, see Mavrikiou and Petropoulos, 'Ιωάννης Γεννάδιος᾽.

<sup>103</sup> Makriyannis, *Απομνημονεύματα*, *e.g*. bk 3, ch.4, vol. 2, 115–117; bk 3, ch. 5, vol. 2, 130.

<sup>104</sup> The legends are transcribed in Papadaki and Tsokani, *Οι πίνακες*, 108.

*Fig. 9.11: Ioannes Makriyannis and Dimitrios Zographos, 10 'Μάχη πρώτη των Αθηνών' (The first battle of Athens) (American School of Classical Studies, Gennadius Library).*

early city plan and its subsequent adaptations in Figures 9.6–9.10 above. Whereas the European plans foreground Ardettos Hill with its temple converted to a church and foreshorten the distance between the Ilissos and the peak of Hymettos, Makriyannis/ Zographos, by contrast, stretch out the distance between the foothills and the columns of the Olympieion at the city's periphery in order to focus on the details of the battle. Squadrons appear in blocks not unlike those de Rossi drew (following a well-established convention in depicting European battle-scenes) inserted into the early plan to depict the Venetian assault on the citadel one hundred and forty years earlier. Athens, over whose possession the battle is fought, is depicted as a miniature version of picture no. 10, with only 'Acropolis' noted in the legend (16).105 The Greek forces, which included Makriyannis and his cavalry, suffered a devastating defeat that led to the citadel's surrender a month later. If 'The battle at Analatos' is a zoomed-out version of 'The first battle of Athens', picture no. 19, 'The siege of Athens' zooms in on the citadel (Figure 9.13). The visual narration of the siege of Athens defended by the Greeks from 3 August 1826 (O. S.) until its fall to Reşid Mehmed Paşa's forces on 25 May 1827 keeps a tight focus on the acropolis with its mosque and contemporary fortifications. Makriyiannis mentions only a few landmarks in the legend to indicate where the Turkish cannon were ranged, from the church of Agia Marina to the columns of Olympian Zeus. The

<sup>105</sup> The legends are transcribed in Papadaki and Tsokani, *Οι πίνακες*, 136–138.

*Fig. 9.12: Ioannes Makriyannis and Dimitrios Zographos, 17 'Μάχη εις Ανάλατον' (The battle at Analaton) (American School of Classical Studies, Gennadius Library).*

*Fig. 9.13: Ioannes Makriyannis and Dimitrios Zographos, 19 'Η πολιορκία Αθηνών' (The siege of Athens, 3 August 1826–25 May 1827 (O. S.)) (American School of Classical Studies, Gennadius Library).*

last number is 15, '*Séntzos*', the contemporary name for the Philopappos monument, a sliver of which is visible on the treeless, rocky hill at the far left.106

The Makriyannis/Zographos portraits condense the city into its citadel and a few major landmarks, and inject that stage with 'vital expressions' of a life and death struggle.107 Zographos expressed the scenes described to him by Makriyannis through a visual language that evokes multiple artistic practices, an accumulation that also mirrors the complexity of nineteenth-century Greek identity. Dimitrios was an icon painter and the foundation of these 'εικονογραφίες' ('iconographies') is the abstractive, multitemporal and non-perspectival idiom of the Orthodox icon in its nineteenth-century expression, by which time older Byzantine visual traditions were infused with elements of both perspectival and sentimental style. Scholars have also seen in Zographos's work features that appear in European maps and patterns from arts and crafts such as embroidery.108 What is important for our study of the early portraits of Athens is that as in the story about the rejected European artist, the visual language devised here represents a choice of resistance and autonomy.109 The Makriyannis/Zographos city views are the polar opposite of the detached plan view with its soldier caricatures inserted into the Venetian portraits of Athens as a war theatre, based on early city plans. Both the Venetian and the Greek views depict features of the contemporary and historical city simultaneously. But in the Makriyannis/Zographos rendering, past and present, vision and history are intertwined using a single visual idiom. The city view mediates human experience and the symbolic meaning of the city for which the human actors fight. The immersive abstraction of the Makriyannis/Zographos renderings is designed to draw the viewer in to participate in the city and its struggle, and this intention is far removed from the Venetian city views that allow the viewer to observe the battle over the city from a safe, aerial distance.

In a process beginning among contemporaries of Martin Crusius, the valorisation of the classical past, rather than Christian culture, came to dominate European identity – and the monumentalised, often also aphaeretic views of Athens were

<sup>106</sup> The legends are transcribed in Papadaki and Tsokani, *Οι πίνακες*, 144–146.

<sup>107</sup> The phrase is George Seferis's 'ζωντανὴ ἔκφραση᾽ in 'Ἕνας Ἕλληνας', 229.

<sup>108</sup> Seferis, 'Ἕνας Ἕλληνας', 239–240 is reminded of traditional embroideries. On Dimitrios Zographos's extant icons, see Fenerli, 'Ζωγράφοι', 57–58; for stylistic discussion, including resemblances to European maps, see the brilliant Tsarouchis, 'Ἀθώα καὶ σοφὴ ζωγραφικὴ᾽, 31–33; also Asdrachas, 'Μακρυγιάννης', 324–349; Petris, *Μακρυγιάννης*, 217–218; Drandakis, 'Ό Μακρυγιάννης᾽, 139–169.

<sup>109</sup> The combined impact of the payment for and rejection of the European painting both resists and at the same time mocks the European monetarisation of 'antiquities', a trend accommodated by local populations (see the famous story in the memoirs about the statues of a 'woman and prince' that Makriyannis caught Greek soldiers carting off to sell to European buyers; Makriyannis later gave them as a gift to the king for the benefit of the country, see Makriyannis, *Απομνημονεύματα*, bk 3, ch. 1, vol. 2, 75α). See Holton, 'Ethnic identity', 146–148, for acute comments on the background to Makriyannis's appreciation for ancient Greece; Petris, *Μακρυγιάννης*, 54–56, on Makriyannis's understanding of antiquities. Drandakis, 'Ό Μακρυγιάννης᾽, 145 on the room in Makriyannis's house set aside as a shrine, full of icons. On the work of Makriyannis/Zographos as an emphatic expression of Greek traditions against European, see Souyoultzoglou-Kallmyer, 'Στρατηγός Μακρυγιάννης᾽, 64–65, with bibliography, who raises the question again.

soon adopted by Greek artists from the mid-nineteenth century. But our first Greek portraits of Athens by Makriyannis and Zographos reveal an inhabited city of flesh and blood, another world in which the multiple strands of identity are not separated but remain intricately intertwined in a city that was both Orthodox Christian and ancient. Strategos Makriyannis was heir to a complex political, geographical, cultural and religious identity, but he was no romantic. After Dimitrios Zographos finished the paintings, in around 1840, Makriyannis had him make a black and white pebble mosaic in his garden and described the mosaic in his *Memoirs*: 'Further on there are the columns of the Olympian Zeus and the Gate [of Hadrian] and the sign of the owl. Further on a dance is going on: a man in Frankish dress is dancing with a Greek. The man in Frankish dress wants his own kind of dancing, the Greek his own and they will soon be at odds for no man can learn another's steps'.110

The European view of Athens in the fifteenth to seventeenth centuries is poised between an accumulation of information and an idealised, visualised concept. But what the Zographos/Makriyannis pictures show is that a city portrait can be both abstractive and full-blooded, representational and experienced, ancient and contemporary. The abstracting power of the icon that directs the viewer to vital elements combines with the lived experience of a city as war theatre to create a unique combination, the product not of an outsider's gaze, but of local participation. Makriyannis and Zographos transform the distant view, which like the outsider's city plan is a means to express power and knowledge, and to control place, by infusing it with an animated immersion in space to produce – for once – a portrait dictated by a local voice. The city is the stage for the actors' struggle, but it is not an inert backdrop. One could say, adopting Makriyannis's own mix of the visceral and the metaphoric, that the city is sanctified through its participation in the struggle, obliterating the objective distance of earlier city portraits.

Makriyannis speaks for a complex local inheritance and his collaboration with Zographos communicates the ancient city's complex present, combining objective and subjective experience of locality while at the same time joining visual conventions of the 'city portrait' with immersive action inside the pictured city.111 And yet their work is not usually discussed among Athens 'city portraits', which are more commonly separated out according to genre and other invented categories. What one witnesses in the Athens of Makriyannis and Zographos – the first *Greek* portraits of Athens – is not a fossilisation or a revival of an antique city, nor a passive reception of a foreign visual idiom and imaginative impulse, but an act of creative resistance that conveys the accumulated history of experience in a city whose ancient traces form part of the material to be constantly reshaped.

<sup>110</sup> Makriyannis, *Απομνημονεύματα*, bk 3, ch. 5, vol.3, 135. Petris, *Μακρυγιάννης*, has a chapter on the mosaic, 148–162.

<sup>111</sup> Makriyannis even planned to make the pictures widely available through lithographic reproductions, as is indicated by his adoption of a format that includes a wide legend at the bottom of the image, keyed to numbers on the picture, see Souyoultzoglou-Kallmyer, 'Στρατηγός Μακρυγιάννης᾽, 42 and 65.

### **Acknowledgements**

This chapter has been researched and written as part of the 'Impact of the Ancient City' project, University of Cambridge, which has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 693418). For comments on earlier versions I am grateful to Suna Çağaptay, Garth Fowden, Guy Meyer and Deniz Türker, as well as the many other colleagues who are thanked in the notes. I take full responsibility for what is presented here.

## **Bibliography**


Hähnle, M., 'Knowledge about the "Orient" between voice and scripture: Michel de Certeau and the Royal Danish Expedition to Arabia Felix (1761–1767)', *InterDisciplines* 2 (2017) 42–68.

Haussouiller, B., 'Athènes au XVIIe siècle. Capucins et consuls', *Journal des savants* 14 (1916): 128–136.


Rhoby, A., 'The "Friendship" between Martin Crusius and Theodosios Zygomalas: A study of their correspondence', *Medioevo Greco: Rivista di Storia e Filologia Bizantina* 5 (2005): 249–267.

Robert de Dreux, R. P., *Voyage en Turquie et en Grèce du R.P. Robert de Dreux, aumônier de l'ambassadeur de France (1665–1669)*, ed. with comm. H. Pernot (Paris, 1925).


Travlos, I.*, Πολεοδομική εξέλιξις των Αθηνών (Athens, 1993).*


Wheeler, J. R., 'Coronelli's maps of Athens', *Harvard Studies in Classical Philology* 7 (1896): 177–189.


## Chapter 10

## Perceptions, histories and urban realities of Thessaloniki's layered past

## *Nikolas Bakirtzis*

Commanding the Thermaic Gulf and the northern Aegean beyond it, Thessaloniki is a dense, sprawling modern city whose layered past continues to reverberate throughout its long and continuous history.1 The physical remains of Thessaloniki's past constitute an integral part of its contemporary urban fabric as they contain aspects of the socio-economic networks, the cultural identity and the lived experience of its former inhabitants. The ancient and medieval formation of Mediterranean cities, including Thessaloniki, shaped their subsequent layered growth and spatially framed the experience of urban life through time. Consequently, as the stratigraphy of cities keeps growing, the surviving parts of past layers continue to have agency through time, making the preservation of what we might call an urban palimpsest an essential aspect of the cities' heritage. Furthermore, modern perceptions of the past have significantly shaped attitudes towards its remains, now seen as monuments. In this context, the preservation, selective appropriation or obliteration of the palimpsest of cities reflects the sociopolitical priorities, ideologies and national narratives of their societies.

This chapter explores aspects of Thessaloniki's layered heritage and the making of its city narrative as they were shaped by the resilience of key elements of its architectural landscape and the selective appropriation of its past layers in the early decades of the twentieth century. The long use of buildings and sites through time reflects their role and significance in the life of the city. For example, the continuity of Thessaloniki's fortifications, which defined its urban experience for almost 1,500 years, reflects the most basic need for protection. Additionally, the strength of religious traditions, first and foremost the cult of Saint Demetrios, sustained the use of church buildings such as the basilica of the city's patron saint to this day. Efforts to

<sup>1</sup> Vacalopoulos, *Ιστορία της Θεσσαλονίκης*; Hassiotis, *Τοις αγαθοίς βασιλεύουσα*; Mazower, *Salonica*.

modernise the historic city in the nineteenth and twentieth centuries combined with historical developments, leading to selective perceptions of the city's past layers. As a result, memory of the past was appropriately curated to accommodate political and ideological priorities, while the preservation of heritage adjusted to socioeconomic realities. Milestone events such as the outset of Greek control in 1912, the great fire of 1917, as well as the 1978 earthquake and the ensuing attempts to protect and restore the damaged monuments, accelerated developments and permitted the selective erasure and appropriation of Thessaloniki's palimpsest, thus shaping the city's experience of heritage and cultural identity.

## **Resilient layers**

A useful way to understand the persistent impact of past layers on the historical growth of cities like Thessaloniki is to turn to surviving monuments and examine their evolution, role and continuing agency. Possibly the most instructive example of the resilience and continuity of the city's monumental past in relation to its historical growth is its fortified enclosure, which defined its spatial experience through to the end of the nineteenth century.2 Providing protection to its inhabitants, the fortifications of Thessaloniki were an integral part of the city's growth with building additions, restorations, patrons' inscriptions and the re-use of spolia forming a 'patina' on the facades of walls, towers and gates accrued over time.3

Founded in 315 BC by Cassander (358–297 BC), the king of Macedonia, Thessaloniki acquired a fortified enclosure which was significantly strengthened in the third century AD by the extensive re-use of materials from various buildings of the ancient city, thus reflecting the hasty efforts to protect the city from the incursions of the Goths.4 Towers like that of 'Klaudianos', named after the official mentioned in a Roman inscription displayed on the tower's façade, were built almost entirely with spolia.<sup>5</sup> By the end of the third century, a triumphal arch originally erected outside the city to honour Octavian after his victory at the battle of Philippi (42 BC) was integrated into the city's western walls, becoming known as the 'Golden Gate'.6 The major building phase of the city's walls dates from the end of the fourth to the middle of the fifth century and is largely attributed to the Emperor Theodosius I (AD 379–395).7 Once more, construction utilised the remains of the ancient city. Materials from redundant institutions of the city's pagan past, such as temples and gymnasia, were used in this momentous defensive endeavour.<sup>8</sup> The facades of the western land walls bear the

<sup>2</sup> Tafrali, *Topographie*, 30–114; Spieser, *Thessalonique*, 25–80; Spieser, 'Les remparts', 557–574; Vitti, *Πολεοδομική εξέλιξη*, 119–130, 159–172; Velenis, *Τείχη*.

<sup>3</sup> Bakirtzis, 'Visual language', 15–34; Bakirtzis, 'H Θεσσαλονίκη και τα τείχη της', 79–103.

<sup>4</sup> Velenis, *Τείχη,* 17–88; Spieser, 'Les remparts', 558–569.

<sup>5</sup> Spieser, *Thessalonique*, 64; Velenis, *Τείχη*, 53–54, fig. 52.

<sup>6</sup> Vitti, *Πολεοδομική εξέλιξη*, 170–171; Spieser, *Thessalonique*, 55–56.

<sup>7</sup> Spieser, *Thessalonique*, 59–77; Velenis, *Τείχη*, 107–124.

<sup>8</sup> Bakirtzis 'Visual language', 17–20.

visual evidence of prominently displayed spolia (Figure 10.1).<sup>9</sup>

Although the city's population fluctuated and declined in the Middle and Late Byzantine periods, the physical size of Thessaloniki did not change and continued to be bounded by its late antique enclosure. The continuing relation of the walls with the city is reflected in the array of dedicatory inscriptions that span the centuries, marking the work of church and state officials. The inscriptions of Hormisdas, dating from the end of the fourth century, of Andronikos Lapardas

*Fig. 10.1: View from the western walls of Thessaloniki (N. Bakirtzis).*

and Michael Prosuch, members of the Komnenian court of Thessaloniki in the 1160s, or of George Apokafkos on behalf of Manuel Palaiologos, governor of the city between 1369 and 1373, are indicative examples of the long tradition of the practice and the continuous use of the walls.10 The continuing agency of fortifications relied on their symbiotic relationship with Thessaloniki, which was manifested in the city's cultural life.11 Mosaics in the Basilica of Saint Demetrios provide instructive reflections of an embattled society. For example, a seventh-century mosaic panel shows the city's patron saint embracing the bishop and the eparch as they stand in front of the city's walls.12 In time, walls came to represent the city and acquired a sacred dimension associated with its patron saint.13 The 'Miracles of Saint Demetrios' echo this notion, describing the city's patron fighting alongside Thessalonians on the walls as they repel attacking enemies.14 Ioannes Kameniates' description of Thessaloniki in 904 identified Saint Demetrios as a warrior, responsible for its survival and victory against all enemies.<sup>15</sup> These descriptions were visualised in artistic representations such as the twelfth-century image of the saint defending the walls of the city depicted on a silver reliquary from the Vatopedi Monastery and a late sixteenth-century icon from Corfu depicting the saint as a mounted warrior with his fortified city in the background.<sup>16</sup>

Religious practice catalysed the resilient continuity of spaces and sites within the ever-growing urban landscape of Thessaloniki. Undoubtedly, the best example is the cult of the city's patron saint which in effect sustained the use of its basilica

<sup>9</sup> Spieser, *Thessalonique*, 569–572; Crow, 'Fortifications and urbanism', 93–98.

<sup>10</sup> Spieser, 'Inventaires', 151–152, no. 3, 165–166, no. 16 and no. 29, 176–177, no. 29.

<sup>11</sup> Bakirtzis, 'Byzantine fortification', 358–360.

<sup>12</sup> Soteriou and Soteriou, *Βασιλική Αγίου Δημητρίου*, 193–194, pl. 63–64.

<sup>13</sup> Bakirtzis, 'H Θεσσαλονίκη και τα τείχη της', 84–90.

<sup>14</sup> Lemerle, *Miracles de Saint Démétrius*.

<sup>15</sup> Böhlig, *De Expugnatione Thessalonicae*, 8.

<sup>16</sup> Grabar, 'Quelques reliquaires', 3–6, figs. 1–5; Vocotopoulos, *Εικόνες της Κέρκυρας*, 99–100.

through time.<sup>17</sup> The extensive architectural complex was inserted into the fabric of the ancient city, along with other Christian church buildings like the Virgin Acheiropoietοs and Hagia Sophia, thus creating what was described by Slobodan Ćurčić as an urban iconography facilitating Thessaloniki's transition into an early Christian metropolis.<sup>18</sup> These complexes occupied central positions situated along major thoroughfares, public spaces, markets and bath houses, appropriating spaces and transforming public life. Anchored around the tomb and relic of the saint, the long and complicated building history of the basilica exemplifies the resilience of the cult.19 Occupying the location of a Roman bath complex, where according to tradition the saint was martyred and buried, the five-aisle basilica with a three-aisle transept at its east end is the result of multiple building phases.20 These are related to the patronage of church and state officials as well as to restoration and rebuilding efforts due to damage from fires and earthquakes. As the city's foremost church complex, the basilica's art and architecture contained within it, as if in concentrated form, the wider city's rich history.

Thessaloniki's transition to Ottoman rule provides useful insights regarding the role of past layers, specifically in relation to the city walls and the Basilica of Saint Demetrios.21 Continuing the centuries-long practice, the new rulers of the city added their presence on the facades of the city's fortifications. A marble inscription over the main gate of the citadel crowning the northern edge of the city, known as Eptapyrgion or Yedikule ('seven towers'), celebrates the city's 1430 capitulation to Çavuş Bey.<sup>22</sup> Furthermore, an eclectic compilation of re-used sculptural and architectural pieces from Roman, early Christian and Byzantine buildings enriched the façade of the gatetower.23 Over the centuries of Ottoman rule, the walls gradually lost their defensive role, yet continued to outline the city and to punctuate the pace of daily life.

Within the city, the conversion of major churches into mosques, along with the construction of new religious and public benefit institutions such as covered markets, baths, hospitals, schools and public fountains transformed Thessaloniki's urban landscape.24 The Saint Demetrios basilica was converted into the Kasımiye Cami in 1493, largely maintaining the architectural integrity of the complex.<sup>25</sup> The cult of the local saint was sidelined yet not removed as it continued in the underground compartments of the Roman bath that were integrated into the foundations of the

<sup>17</sup> Bauer, *Stadt und Patron*; Skedros, *Saint Demetrios*.

<sup>18</sup> Ćurčić, 'Christianization of Thessaloniki', 213–244.

<sup>19</sup> Bakirtzis, 'Pilgrimage to Thessaloniki', 175–192.

<sup>20</sup> Xyngopoulos, *Η βασιλική του Αγίου Δημητρίου*; Soteriou and Soteriou, *Βασιλική Αγίου Δημητρίου.*

<sup>21</sup> Vacalopoulos, *Ιστορία της Θεσσαλονίκης*, 185–231; Ηassiotis, 'Η τουρκοκρατούμενη Θεσσαλονίκη', 103–116; Mazower, *Salonica*, 15–45.

<sup>22</sup> Lowry, *Shaping of the Ottoman Balkans*, 120; Dimitriadis, *Τοπογραφία*, 2011.

<sup>23</sup> Bakirtzis, 'Visual language', 24.

<sup>24</sup> Taddei, 'Conversion of Byzantine buildings', 201–214; Ćurčić, *Architecture in the Balkans*, 748–752; Zygomalas, *Η προστασία των αρχιτεκτονικών μνημείων*, 78–102.

<sup>25</sup> Dimitriadis, *Τοπογραφία,* 292–294; Bauer, *Stadt und Patron*, 439–450.

basilica. The tomb of the saint was transferred there and continued to be worshiped by the Christian as well as the Muslim population of the city. The religious importance of the site helped to preserve the physical remnants the past.

Balancing between the emergence of new layers and the resilience of past ones, the social, ethnic and cultural transformation of the city during the Ottoman period was greatly enriched by the arrival of thousands of Sephardic Jews following their expulsion from Spain in 1492.26 From this point until the middle of the nineteenth century, Thessaloniki's labyrinthine urban fabric consisted of dense neighbourhoods of Jews, Greeks, Turks and other minorities, with synagogues, churches and mosques serving as cultural and social points of reference.27 The Jewish population was concentrated around the port and in the central part of the lower city, where its commercial centre was situated. The Greeks occupied the central and eastern part of the city with the Turks primarily residing in the sloping and better ventilated upper city.<sup>28</sup>

#### **Appreciation of the past**

Interest in the city's history and antiquities had been growing steadily from the eighteenth and especially nineteenth century onwards as European travellers began to include Thessaloniki in their journeys to the Levant, recording their findings and impressions.29 As in other areas of the Ottoman Empire, remnants of Thessaloniki's antique heritage became the focus of collectors, antiquarians and archaeologists, who noted the importance of its antiquities.30 Among them, a set of eight sculptures of deities associated with the cult of Dionysus known locally as 'the idols' or 'Las Incantadas', meaning the 'enchanted ones', stood out. They were part of a secondcentury portico whose remnants became incorporated into the courtyard of a local residence in the densely populated Jewish neighbourhood of Rogos.<sup>31</sup> Drawn by James Stuart and Nicholas Revett in 1753, the monument became a highlight for travellers to the city eager to discover and collect its ancient past.32 Despite the protests of local inhabitants, but with the permission of Ottoman officials, the Incantadas were eventually dismantled, packed and sent to the Louvre by the French paleographer Emmanuel Miller in 1864.33 One of the most important surviving monuments of Thessaloniki's Roman heritage layer was singled out, extracted and shipped to Paris

<sup>26</sup> Mazower, *Salonica*, 46–65; Nehama, *Histoire, ii*; Nehama, *Histoire, iii*.

<sup>27</sup> Lowry, 'Portrait of a city', 254–293.

<sup>28</sup> Perdrizet, 'L'Incantada de Salonique'; Dimitriadis, *Τοπογραφία,* 16–19.

<sup>29</sup> Mazower, 'Travellers', 59–111; Tambaki, S., *Η Θεσσαλονίκη στις περιγραφές των περιηγητών*; Zygomalas, 'Ερευνητική δραστηριότητα', 120–122.

<sup>30</sup> Mazower, 'Travellers', 79–109; Shaw, *Possessors and Possessed*; Anderson, '"Alternative discourse"', 450–460.

<sup>31</sup> Dimitriadis, *Τοπογραφία*, 160–162.

<sup>32</sup> Stuart and Revett, *Antiquities of Athens*, 53–55.

<sup>33</sup> Mazower, 'Travellers', 80–90; Anderson, '"Alternative discourse"', 454–456.

*Fig. 10.2: Postcard showing the promenade of Thessaloniki in the early years of the twentieth century with the White Tower in the background (N. Bakirtzis).*

as a collector's item. Ironically, it was thanks to Miller that the important sculptures survived the disastrous fire of 1917, which burnt through the city's centre. Absolutely nothing survives on-site from the important monument except its memory.

Beyond the fascination with Greco-Roman Thessaloniki, the early Christian apostolic legacy of Paul's passage through the city was another focus of attention and interest for travellers to the city.34 Furthermore, the emerging appreciation of Byzantium gradually helped to highlight the remarkable monuments of the Byzantine city, especially from the second half of the nineteenth century onwards.<sup>35</sup> The foundation of the modern Greek state and related efforts towards the creation of a national narrative – driven primarily by the ideals of Classicism – gradually embraced Byzantine culture and its monuments by highlighting its connecting role between ancient and modern Greece.36 In this context, Michael Hadji-Ioannou's *Astygrafia* of Thessaloniki, published in 1880, provided a historical topography of the city focusing on its early Christian and Byzantine monuments.37

The progressive environment of the Hatt-ı Humayun of 1856 and the subsequent Tanzimat reforms in the Ottoman Empire promoted efforts to modernise Thessaloniki

<sup>34</sup> Mazower, 'Travellers', 93–95; Davies, *St. Paul in Greece*, 120–125.

<sup>35</sup> Chlepa, *Τα Βυζαντινά μνημεία*, 13–29; Mazower, 'Travellers', 90–92.

<sup>36</sup> Chlepa, *Τα Βυζαντινά μνημεία*, 31–114; Zygomalas, 'Ερευνητική δραστηριότητα', 123–135.

<sup>37</sup> Hadji-Ioannou, *Αστυγραφία Θεσσαλονίκης*.

in the light of urban planning developments in Europe.<sup>38</sup> Attention was given to the improvement of public spaces, hygiene and services, as well as the overall beautification of the city. At the same time, Ottoman attitudes towards antiquities and monuments of the past were shifting.39 In 1870, the demolition of the sea walls by governor Sabri Paşa was a milestone development with practical and public health implications.40 Symbolically, it was an act that liberated the city from its medieval confines as it transitioned into modernity. Thessaloniki had now acquired a cosmopolitan waterfront promenade embellished with hotels, restaurants and cafés aspiring to the contemporary lifestyles of European cities (Figure 10.2). A little later, parts of the lower western and eastern walls were also demolished along with the Golden Gate and the Kalamaria Gate to permit the growth of suburbs. Other parts of the fortifications were kept, such as the circular tower at the south-eastern corner of the city's enclosure, known as Kanlı Kule (Tower of Blood), which ceased to function as a prison; it was whitewashed and by 1884 was referred to as the White Tower, and today is an icon of the city's long history.41

Beyond the seafront and the city's cosmopolitan façade, the daily lives of an array of ethnic and religious populations persisted unchanged in the dense urban fabric of neighbourhoods organised around synagogues, mosques and churches.42 In the labyrinthine streets of Thessaloniki, aspects of the exotic oriental city described by Western travellers showed resilience.43 Change became possible as a result of a fire which on 4 September 1890 burnt a significant part of the city's centre, destroying the Orthodox Cathedral of Saint Gregory Palamas and the British and Greek Consulates along with numerous other buildings.44 A renewal plan was ready by 1892, providing the first major planning intervention in the historic city. A new grid of regular streets aimed at providing better organisation and access to the port while newly developed residential blocks and their related commercial activities gentrified this part of the city.

A highlight of the reconfiguration of public space in response to Thessaloniki's palimpsestic urban tissue was the design and creation of a new open square in front of the eighth-century Church of the Hagia Sophia, which had been used as a mosque since its conversion in 1524. This project was complemented by restoration works, initiated in 1906/7, addressing the damage caused by the 1890 fire.45 As part of this

<sup>38</sup> Gounaris, 'Salonica', 499–518; Gounaris, 'Θεσσαλονίκη 1830–1912', 156–175; Anastassiadou, *Salonique*, 89–201; Dimitriadis, 'Ottoman port-city', 37–180; Dimitriadis, 'Transforming a late-Ottoman port-city', 207–221; Yerolympos and Kolonas, 'Un urbanisme cosmopolite', 158–176.

<sup>39</sup> Eldem, 'A new look at an ancient city', 105–125; Eldem, 'Ottoman perceptions of antiquities, 1799–1869', 281–329; Zygomalas, *Η προστασία των αρχιτεκτονικών μνημείων*, 120–124.

<sup>40</sup> Yerolympos, *Urban Transformations*, 61–70; Vacalopoulos, *Ιστορία της Θεσσαλονίκης*, 334–340.

<sup>41</sup> Gounaris, '1884', 100–105.

<sup>42</sup> Anastassiadou, *Salonique*, 203–381; Molho, *Οι Εβραίοι της Θεσσαλονίκης*.

<sup>43</sup> Mazower, *Salonica*, 222–252.

<sup>44</sup> Yerolympos, *Μεταξύ Ανατολής και Δύσης*, 165–182; Dimitriadis, 'Transforming a late-Ottoman port-city', 216–218.

<sup>45</sup> Dimitriadis, 'Ottoman port-city', 109–113; Zygomalas, 'Ερευνητική δραστηριότητα', 133.

effort the French government dispatched funds along with the archaeologist Marcel Le Tourneau, who collaborated with art historian Charles Diehl to study Hagia Sophia and other Byzantine monuments of the city.46 These studies complemented earlier and ongoing work by international and local scholars, thus providing the basis for the growing appreciation of the artistic and cultural value of Thessaloniki's Byzantine past.47 Other churches had also been converted into mosques but were now suffering from neglect. As a result of the changing attitude towards the monuments of the past, the historic buildings of Saint Demetrios, the Rotunda, the Virgin Acheiropoietos, Saint Panteleimon and Panagia Chalkeon all benefited from repair works.48 Writing for the *National Geographic* in 1916, Harry Griswold Dwight offered his impressions of Thessaloniki, unequivocally declaring that the 'finest remains of the ancient city are its churches' and that 'they make up between them a museum of the lost Byzantine art of mosaic'.49 Reflecting on an earlier visit to the then Ottoman-controlled city, Dwight also reminisced about the tolerant attitudes of the Ottomans towards the figural religious art preserved in churches converted to mosques, as well as towards the cult of Saint Demetrios.

With the city being the centre of the dramatic political developments of the 'Young Turks' revolution, the visit of sultan Abdulhamid in June 1909 and the related efforts to beautify and modernise the city were the last Ottoman urban planning interventions in the city.50 Restorations at the Hagia Sophia Mosque were hurried to prepare for the sultan's prayers, streets were remodeled and the walls surrounding the White Tower were demolished.

### **Adjusting to a new narrative**

The early decades of the twentieth century transformed Thessaloniki. In the course of the First Balkan War, the advancing Greek army captured the growing city in November 1912.51 Appointed as the first representative of the city's Greek administration by the Prime Minister Eleftherios Venizelos, Konstantinos Raktivan led Thessaloniki's cautious transition to its new, hellenised reality.<sup>52</sup> Among his important initiatives was the establishment of a Beautification Committee (Epitrope Exoraismou) aimed at the preparation of a new plan to improve Thessaloniki's infrastructure and embellish its public life, priorities well aligned with eighteenth-century urban planning in

<sup>46</sup> Diehl *et al*., *Monuments Chrétiens*; Zygomalas, 'Ερευνητική δραστηριότητα', 59.

<sup>47</sup> Important earlier work was conducted by Theophilus Tafel, Charles Texier, Auguste Choisy, Robert Weir Schultz, Sidney Barnsley, Walter S. George, Oreste Tafrali, Théodore Uspenskij and Nikodim Kondakov, as well as local experts like Petros Papageorghiou. See Zygomalas, 'Ερευνητική δραστηριότητα', 123–134; Trakosopoulou-Tzimou, 'The care of monuments in modern Thessaloniki', 356–360.

<sup>48</sup> Chlepa, *Τα Βυζαντινά μνημεία*, 208; Zygomalas, *Η προστασία των αρχιτεκτονικών μνημείων*, 180–203.

<sup>49</sup> Dwight, 'Saloniki', 203–232.

<sup>50</sup> Dimitriadis, 'Urban transformation and revolution', 126–138.

<sup>51</sup> Vacalopoulos, *Ιστορία της Θεσσαλονίκης*, 375–379; Mazower, *Salonica*, 301–304; Petridis, 'Θεσσαλονίκη 1912–1940', 137–149; Dimitriadis, 'Ο πληθυσμός της Θεσσαλονίκης', 88–116.

<sup>52</sup> Vogli, 'Aftermath of the Balkan wars', 169–182.

Europe.<sup>53</sup> Combining technical and artistic expertise, the committee consisted of architect Aristotle Zachos, artist Petros Roumpos, chief engineer of the Municipality of Thessaloniki Konstantinos Maleas, also a major modern Greek painter, the previous municipal engineer Georgios Menexes and the prefect Pericles Argyropoulos, who served as its chair. Although little documentation has survived from their work, through press coverage we have a general sense of the plan prepared by Zachos.<sup>54</sup> In an interview he stated that Thessaloniki's urban development 'must relate to its particular local conditions; its geography, climate, traditions and more importantly its monuments'. Zachos emphasised the need to position monuments at 'prominent locations in the urban plan of the city and to open around them squares and wide boulevards'.

For the Beautification Committee, the city's architectural heritage was an urban planning priority. In the new plan, landmark sites like the White Tower, which had barely escaped demolition in the early days of the arrival of the Greek army, had pivotal roles in anchoring the broad public spaces around them. The city's important late antique and Byzantine monuments such as the Arch of Galerius, the Rotunda, Saint Demetrios, the Holy Apostles, Saint Catherine, Saint Panteleimon and the city walls, as well as markets such as those around the Ottoman Alaca Imaret, were highlighted through the opening of squares and thoroughfares, which were named after glorious figures of the Greek past. Moreover, according to Zachos's plan, which was in alignment with the Greek administration, the grand Basilica of the Acheiropoietos was to function as a museum. Furthermore, the sparsely populated area of the acropolis and the Eptapyrgion citadel was designed as a multi-purpose park for leisure and culture, accommodating gardens, fountains, playgrounds, a small zoo, sports facilities, concerts, museums and exhibitions. Part of the acropolis walls would be used to support terraces providing views of the lower city. Yet despite all the planning, financial difficulties and the onset of World War I caused the Beautification Committee's activities to cease.

Regardless of the final outcome, the plans of Zachos provided a selective perspective on the city's past and reflected the priorities of the new Greek administration. Byzantine churches and monuments were highlighted in the public spatial experience of the city, while mosques and synagogues did not receive the same attention, though retaining their role as focal points for the religious life of Turkish and Jewish neighbourhoods.55 Furthermore, it is worth mentioning that in 1913 eight Byzantine churches, the Rotunda and the fortification walls of the city were recognised as 'national monuments'.56 The early years of Thessaloniki's transition into Greek administration are particularly instructive for tracing the formulation of a new city history aligned with the historical

<sup>53</sup> Yerolympos, *Ανάδυση*, 103–115.

<sup>54</sup> Yerolympos, *Ανάδυση*, 107–113.

<sup>55</sup> Yerolympos, *Ανάδυση*, 113.

<sup>56</sup> Mastora, 'The French armée d'Orient at the Rotunda', 216; Zygomalas, *Η προστασία των αρχιτεκτονικών μνημείων*, 384–386.

narrative of modern Greece and the continuity of Hellenic cultural identity in the broader region of Macedonia.<sup>57</sup> With the capital of Athens having assumed its role as a key reference for Greece's Classical past, the Byzantine heritage of Thessaloniki and the 'return of Saint Demetrios', as beautifully put by historian Mark Mazower, at the central stage of the city's cultural life, helped to locate the city in the Greek national narrative.58 It is worth mentioning that the Acheiropoietos basilica was intended to host the Central Byzantine Museum of Greece. According to Adamantios Adamantiou, Ephor of the Christian and Medieval Monuments of Greece, Thessaloniki was 'a city-symbol for the national ideal of the re-foundation of the great Greek empire' and 'as the second capital of the Byzantine Empire' should host the central Byzantine Museum and be the 'base for the study of Christian and Byzantine art'.59 The return of the major Byzantine churches to their ecclesiastical use carried serious implications for the social fabric and the cultural life of the city. Moreover, the foundation of an Ephorate of Antiquities under Georgios P. Oikonomos, days after Thessaloniki was surrendered to the Greek army, underscores the immediate attention given to the management and protection of the city's archaeological heritage.60 Indicative of these developments in the context of Balkan states' rivalry for the control of Thessaloniki was the reaction of the Serbian Consul, who accused the Greek administration of promoting the Hellenisation of the city to serve its political priorities.61

The outbreak of World War I found Thessaloniki at the forefront of dramatic developments. Culminating in his showdown with King Constantine I (1913–17 and 1920–22) over which side Greece would support – known at the National Schism – Venizelos declared the city the seat of the Provisional Government of National Defence, which aligned Greece with the Entente.62 The important port had critical significance for the war effort and its Balkan front, and from 1915 onwards thousands of allied troops arrived in Thessaloniki. French, British, Italian, Russian and Serbian Entente forces joined Greek troops, transforming the city into a massive army base with vast camping grounds on the western outskirts of the city.63 The city's history and archaeology received the attention of the European armies, particularly the French and the British who, in their colonial tradition, studied, excavated and collected the remains of the broader region's glorious past.64 In the context of these activities

<sup>57</sup> Stefani, 'National ideology', 19–39.

<sup>58</sup> Bastéa, *The Creation of Modern Athens*; Mazower, *Salonica*, 293–295.

<sup>59</sup> Kourkoutidou-Nikolaidou *et al*., *Βυζαντινοί Θησαυροί της Θεσσαλονίκης*, 8–11.

<sup>60</sup> Akrivopoulou, 'Η μέριμνα του ελληνικού κράτους', 57–63; Vocotopoulou, 'Τα πρώτα 50 χρόνια', 1–5; Adam-Veleni, 'Η τύχη των αρχαιολογικών ιστών', 91–92.

<sup>61</sup> Hassiotis, *Ελληνοσερβικές σχέσεις 1913–1918*, 167, 178.

<sup>62</sup> Petridis, 'Προσωρινή κυβέρνηση Θεσσαλονίκης', 131–139; Mazower, *Salonica*, 305–310; Vacalopoulos, *Ιστορία της Θεσσαλονίκης*, 379–387.

<sup>63</sup> Kolonas, *Σκιά του μεγάλου πολέμου*; Yerolympos and Hekimoglou, 'Η Θεσσαλονίκη του πρώτου παγκοσμίου πολέμου', 167–182; Hekimoglou, 'Επιδράσεις της συμμαχικής παρουσίας', 51–63; Katsaridou and Motsianos, *The Allied Army of the Orient in Thessaloniki.*

<sup>64</sup> Andreou and Eukleidou, *Αρχαιολογία στη γραμμή του πυρός*; Satrazanis, *Η γαλλική στρατιά στη Θεσσαλονίκη*; Farnoux, 'Archaeology and the armée d'Orient', 83–89.

*Fig. 10.3: Postcard showing Thessaloniki from the east in the beginning of the twentieth century (N. Bakirtzis).*

and with the support of the Venizelos government, the Rotunda was declared a 'Macedonian Museum' to host collections of antiquities and archaeological finds from the broader area of Macedonia.<sup>65</sup> Yet, besides the interest of Europeans and the active presence of the Greek Archaeological Service in the city, the prevalent uncertainty and volatility of the times did not leave much room for meaningful considerations of Thessaloniki's past. According to Harry Dwight, writing in 1916, Thessaloniki 'is too much interested in what she is and what she is going to be, to think very much about her past' (Figure 10.3).<sup>66</sup>

## **The opportunity of a fire disaster**

When a fire broke out at a small house in the central Mevlane neighbourhood on the afternoon of 5 August 1917, nobody expected the size of the devastation it would cause.<sup>67</sup> As a densely populated city, Thessaloniki had experienced its share of fires throughout history but nothing compared to the 1917 disaster, which burnt approximately thirty-two percent of the city. Various factors contributed to fire's

<sup>65</sup> Mastora, 'The French armée d'Orient at the Rotunda', 217–218; Zygomalas, *Η προστασία των αρχιτεκτονικών μνημείων*, 362–364.

<sup>66</sup> Dwight, 'Saloniki', 221.

<sup>67</sup> Yerolympos, *La chronique*; Yerolympos, *Οχυρωμένο στρατόπεδο*, 42–51.

cause and the extent of the devastation: the hot and dry summer, the strong northwest winds which fuelled the flames, the shortage of water, the absence of organised fire response measures and the density of the city's urban fabric.68

Within days after the fire was extinguished and the scale of the destruction realised, Venizelos' government announced the reconstruction of the city with Minister of Transport Alexandros Papanastasiou assuming the responsibility for the major effort, a chapter in the history of Thessaloniki expertly discussed in the pioneering work of Aleka Yerolympos.69 The fire had burnt the heart of Thessaloniki's centre including its busy markets and areas of economic activity around the port, affecting primarily areas occupied by the city's Jewish population. Also burnt were numerous key monuments and religious sites such as the Basilica of Saint Demetrios, the Monastery of Saint Theodora, the Saatli Cami and numerous synagogues including the central Talmud Torah Hagadol.70 To tackle the challenge, Papanastasiou established the International Committee for the New Plan of Thessaloniki, appointing the French architect and archaeologist Ernest Hébrard as its chair.<sup>71</sup> The seven-member committee also included the British landscape architect Thomas Mawson, the French engineer and architect Joseph Pleyber, the architects Aristotle Zachos and Konstantinos Kitsikis, the port planner Angelos Ginis and the city's mayor Konstantinos Angelakis. The fire disaster was an opportunity for the city's rebirth as a modern Greek metropolis, leaving behind its Ottoman past.

The new plan for the city engaged its past and especially its Byzantine monuments.72 It provided a curated experience of key buildings that had survived the fire or at least could be restored, such as the Basilica of Saint Demetrios. In this context, its Byzantine past was acknowledged and highlighted as its foremost heritage layer, thus finding its appropriate place in the Greek national narrative.<sup>73</sup> Hébrard conducted archaeological investigations in Thessaloniki on behalf of the French army and had first-hand knowledge of the city's stratigraphy.<sup>74</sup> Although his plan provided a new city centre for administration and trade with its design well rooted in his Beaux Arts background, selected monuments were kept, thus providing familiar topographical references in a rapidly changing urban context.75 Late antique monuments and remains, basilicas, churches, fortification walls, Ottoman mosques and baths were highlighted as heritage sites through the opening of public spaces and squares around them, while broad

<sup>68</sup> Adam-Veleni and Karliambas, *Monuments in Flames,* 28–38.

<sup>69</sup> Yerolympos, *Ανοικοδόμηση*.

<sup>70</sup> Adam-Veleni and Karliambas, *Monuments in Flames*, 39–79; Pallini and Scaccabarozzi, 'Salonika's lost synagogues', 16–25; Ηekimoglou, 'Οι καμένες συναγωγές της Θεσσαλονίκης', 159–190.

<sup>71</sup> Yerolympos, *Ανοικοδόμηση*, 90–93; Yiakoumis *et al*., *Ernest Hébrard*.

<sup>72</sup> Yerolympos, *Ανάδυση*, 121–122; Yerolympos, *Οχυρωμένο στρατόπεδο*, 59–73.

<sup>73</sup> Mazower, *Salonica*, 467–468. 74 Hébrard, 'Les travaux du service archéologique', 5–40; Mastora, 'The French armée d'Orient at the Rotunda', 214–229.

<sup>75</sup> Hastaoglou-Martinidis, 'Urban aesthetics', 161–165; Yerolympos, *Urban Transformations*, 109–114; Trakosopoulou-Tzimou, 'The care of monuments in modern Thessaloniki', 362–365.

avenues were drawn between these sites offering beautiful vistas of them across the city.76 It is worth noting that the upper town with its Ottoman vernacular architecture was preserved as a historic neighbourhood. Centrepieces of the plan were grand boulevards such as the eclectic grand axis designed to be the centre of administrative and social life connecting the upper city and Saint Demetrios with the sea (Figure 10.4).77 According to the design, a new open public space, named as the 'Square of Nations', occupied the area of the ancient agora to the south of the patron saint's basilica and featured the Municipality and the Court House. The Middle Byzantine Church of the Panagia Chalkeon and the fifteenth-century Bey Hamam occupied the southern part of the square, adding historical context to the grand public space. A boulevard framed by multi-storey buildings with neo-Byzantine style arcades on their ground floor led south to the sea front where a second open

*Fig. 10.4: Views of the churches of Saint Panteleimon and Hagia Sophia along an axial thoroughfare following the design of the New Plan for Thessaloniki (N. Bakirtzis).*

space, known as Aristotelous Square, framed views of the port and Mount Olympus in the distance.

Of great significance in efforts to strengthen the city's Byzantine heritage and its religious traditions was the restoration of the Basilica of Saint Demetrios, which had been badly damaged by the fire. For Georgios Soteriou, the leading Byzantine archaeologist involved with the project, the basilica was the 'Heroon' (ancient shrine dedicated to a hero) of Thessaloniki. Soteriou likened the church to a stage on which 'the fortunes of Hellenism' were performed.<sup>78</sup> The restoration of the complex was an arduous process and lasted until the 1950s, as it faced a range of financial challenges during periods of war and political volatility.79 In addition, the project was at the centre of heated debates about the course and methods of its rehabilitation. Aristotle Zachos directed the architectural restoration of the complex while Soteriou led its archaeological study. Upon Zachos's death in 1939, Anastasios Orlandos assumed

<sup>76</sup> Yerolympos, *Ανάδυση*, 119–124; Yerolympos, *Ανοικοδόμηση*, 150–157, 165–167; Zygomalas, *Η προστασία των αρχιτεκτονικών μνημείων*, 388–392. 77 Yerolympos, *Ανοικοδόμηση*, 148–167.

<sup>78</sup> Soteriou, 'Έκθεσις', 3.

<sup>79</sup> Chlepa, *Τα Βυζαντινά μνημεία*, 159–197.

responsibility for its restoration. Archaeologists Andreas Xyngopoulos, Stylianos Pelakanidis and Eustratios Pelekidis along with architects Heraklis Leontidis, Leonidas Thanopoulos, Eustathios Stikas and civil engineer Dimitris Saklambanis, also contributed to the grand project.

A closer look at the conflicting visions and approaches to the basilica's revival reveals the significance of the building in efforts to redefine Thessaloniki's historical narrative after the 1917 fire. Zachos championed extensive rebuilding with the aim of returning the basilica to its early Christian state, while archaeologists like Soteriou, Xyngopoulos and Pelekanidis, who led the final stages of the project, supported the preservation of as much original material *in situ* as possible.80 Important to mention here is that in the early clearing of debris and the stabilisation efforts to secure what had survived the fire, additions dating from the Ottoman period and its use as a mosque were largely removed.<sup>81</sup> Additionally, archaeological work strengthened the building's legacy and shed light upon its layered history.<sup>82</sup> The 'rediscovery' of the crypt associated with the early cult of the saint and its integration in the monument's rehabilitation enriched the basilica's religious significance.

Although the ambitious plan of Hébrard did not fully materialise, it nevertheless outlined Thessaloniki's urban experience and helped shape the city's Hellenised historical narrative and the perception of its layered past.<sup>83</sup> Monuments were cleared of their social and cultural contexts and treated as precious relics in the ambitious gentrification of the city. The overall planning effort was driven by contemporary political priorities that aspired to redefine the socio-economic organisation of the city by looking towards the models of colonial Western Europe, also manifested in other urban planning projects like the work of Henri Prost in Smyrna.84 In Thessaloniki, the plan's implementation was sustained by a legal framework that addressed issues of property and land ownership.85 Led by Minister Papanastasiou, new policies and procedures reshaped the city's socio-economic fabric.86 Thessaloniki's Jewish population, already adjusting to the city's Greek administration, and whose properties, residences and heritage had been greatly affected by the fire, reacted with suspicion and frustration to these efforts, which effectively appropriated the urban centre for new use.87 As a result, the ancient Jewish quarters of the city were not treated as heritage worth preserving and were thus effectively removed from the city's spatial experience. Moreover, in the midst of political volatility, economic downturn and successive wars, the city

<sup>80</sup> Chlepa, *Τα Βυζαντινά μνημεία*, 174–197; Adam-Veleni and Karliambas, *Monuments in Flames*, 107–113.

<sup>81</sup> Chlepa, *Τα Βυζαντινά μνημεία*, 176–183*.*

<sup>82</sup> Soteriou, 'Έκθεσις', 25–46.

<sup>83</sup> Yerolympos, *Ανοικοδόμηση*, 212–218; Mazower, *Salonica*, 351–353.

<sup>84</sup> Amygdalou, 'Building the nation', 1–19.

<sup>85</sup> Yerolympos, *Urban Transformations*, 115–124. 86 Yerolympos, *Ανάδυση*, 188–197.

<sup>87</sup> Molho, 'The Jewish community of Salonika', 391–403; Yerolympos, *Ανάδυση*, 142–151; Mazower, *Salonica*, 326–329.

and its population were in flux. The arrival of tens of thousands of refugees from Asia Minor after 1922, the related exodus of the city's Turkish population and the extermination of Thessalonian Jews in World War II altered irreversibly the city's population and helped the erasure of the spatial presence of these communities in the collective memory of the rapidly growing city.<sup>88</sup>

Such traumatic events affected attitudes towards the preservation of past layers and accommodated Greek nationalistic efforts to redefine and claim urban space.89 An indicative development was the demolition of the city's minarets in 1924–25 (the minaret of the Rotunda survived).<sup>90</sup> Moreover, the eradication of the vast Jewish cemeteries outside the eastern walls of the city and the re-use of tombstones in projects like the restoration of Saint Demetrios remains a sad chapter in the history of modern Thessaloniki.91 The 1920s through to the 1940s were particularly challenging for the preservation of heritage, notwithstanding the efforts of the archaeological service and the gradual declaration of sites as protected archaeological monuments under the strengthened Archaeological Law of 1932.92 The public's changing attitudes towards the city's past were evident in the strong opposition to the construction, following Hébrard's plans, of the city's Court House in the so-called 'Plateia Dikasterion' (Court House Square).<sup>93</sup> In 1966, excavations for the building's foundations revealed the architectural remains of the Roman Agora, sparking the interest of the public who, under the leadership of the Ephor of Antiquities Fotis Petsas and with the support of local newspapers, protested against the Court House plans. In 1969 the site was declared an archaeological monument and the name of the square became 'Plateia Archaias Agoras' – today one of the most important archaeological sites in the city. The systematic excavations of the Galerian palatial complex in the 1950s and the 1960s under the leadership of the Ephor of Antiquities Charalambos Makaronas protected and promoted another major archaeological site for Thessaloniki.94

An attempt to connect with the Byzantine past while reaping cultural and economic benefits for the city was the 1966 establishment of the annual festival of Demetria by the Greek Tourism Organisation, which continues to this day. Reminiscent of the primarily commercial and market activities taking place in Byzantine Thessaloniki related to the feast day of the saint on 26 October, the modern version of Demetria involves cultural events held across the city in the months of September and October,

<sup>88</sup> Mazower, *Salonica*, 332–370; Hastaoglou-Martinidis, 'Προσφυγική εγκατάσταση', 43–88; Katerinopoulos, 'Refugee resettlement', 183–192; Molho, 'Nationalism and antisemitism', 217–228; Kavala, *Η καταστροφή των Εβραίων*, 46–107.

<sup>89</sup> Koumaridis, 'Urban transformation', 213–242; Trakosopoulou-Tzimou, 'The care of monuments in modern Thessaloniki', 364–367.

<sup>90</sup> Adam-Veleni, 'Η τύχη των αρχαιολογικών ιστών', 95; Mazower, *Salonica*, 351–352.

<sup>91</sup> Saltiel, 'Thessaloniki's Jewish cemetery', 11–46; Adam-Veleni and Karliambas, *Monuments in Flames*, 110–111.

<sup>92</sup> Adam-Veleni, 'Η τύχη των αρχαιολογικών ιστών', 94–97.

<sup>93</sup> Adam-Veleni, 'Ο Φώτης Πέτσας και η Αρχαία Αγορά Θεσσαλονίκης', 31–63.

<sup>94</sup> Hastaoglou-Martinidis, 'Νεωτερικότητα και μνήμη', 169–171.

culminating in parades and processions around the saint's basilica.<sup>95</sup> This was a properly reinvented tradition, to borrow from Eric Hobsbawm, which fused a set of practices, rituals and behavioural norms into an annual festival that implied continuity with the past, preferably 'a suitable historic past'.96

## **From earthquake to World Heritage and recent realities**

A different disaster, this time a major earthquake in 1978, became an opportunity to address monument preservation in the city more methodically. The 6.5 Richter quake damaged most monuments, triggering an extensive programme of restorations and archaeological work. With a formal focus on the city's 'Byzantine and post-Byzantine monuments', churches, mosques, baths and markets from the Byzantine and Ottoman periods were systematically studied and restored.97 In her presentation at an international conference on the restoration of Thessaloniki's monuments, the Director of Byzantine Antiquities Chrysanthi Mavropoulou-Tsioumi acknowledged that 'the 1978 earthquakes offered the opportunity to address the monuments of the city as a whole, being works of large scale and global significance'.98 In 1986, the approval of the Thessaloniki Master Plan provided the important framework of urban planning regulations for the protection of monuments and sites.99

The results of the systematic preservation programme and the related archaeological studies significantly contributed to the 1988 inclusion of the 'Palaeochristian and Byzantine Monuments of Thessaloniki' in the UNESCO World Heritage list for their 'outstanding design and major artistic value' as 'among the most significant of the Byzantine period'.<sup>100</sup> It is notable that the UNESCO description not only asserts that the monuments are 'intact', but also mentions the urban setting and the presence of 'boundaries adequate to protect the fabric of the monuments and there are buffer zones for some of them'. There is even acknowledgement that the 'continual use of the buildings (from their foundation until today) proved beneficial to their state of conservation, despite certain inevitable interventions'. In effect, UNESCO's World Heritage designation incorporated the urban planning priorities of the post-1917 city, and along with the restoration works that followed the 1978 quake helped establish the monuments of the early Christian and Byzantine period as the predominant heritage layer for Thessaloniki. Moreover, the opening of the Museum of Byzantine Culture in 1994 satisfied the long-standing desire for the foundation of a major Byzantine museum in Thessaloniki.<sup>101</sup> Notably, its magnificent exhibitions

<sup>95</sup> Vryonis, 'Panegyris', 202–204.

<sup>96</sup> Hobsbawm, 'Introduction', 1–14.

<sup>97</sup> Theocharidou-Tsaprali and Mavropoulou-Tsioumi, *Aναστύλωση*.

<sup>98</sup> Mavropoulou-Tsioumi, 'Σεισμοί', 14.

<sup>99</sup> Bastéa and Hastaoglou-Martinidis, 'Urban change and the persistence of memory', 272–274.

<sup>100</sup> UNESCO, 'Monuments of Thessalonica'.

<sup>101</sup> Kourkoutidou-Nikolaidou *et al*., *Βυζαντινοί Θησαυροί της Θεσσαλονίκης*, 8–18; Kourkoutidou-Nikolaidou, 'Ένα μουσείο γεννιέται', 4–9.

included important 'treasures' from the Byzantine monuments of the city that had been kept in Athens since 1915 and were now returned in an atmosphere of civic elation and pride. Almost at the dusk of the century, Thessaloniki's designation as Cultural Capital of Europe in 1997 was celebrated with a rich program of events and projects highlighting the its history and heritage.102 The city's renewed historical narrative was now properly integrated in the broader European experiment of the EU.

More recently, two major infrastructure projects have provided a unique opportunity to explore the city's archaeology systematically. Between 2000 and 2002, a systematic archaeological excavation preceded the installation of a new water pipe along Saint Demetrios Avenue, which runs over the course of the thoroughfare second most important in the Roman and Byzantine city (Figure 10.5).103 The project allowed the exploration of the city's stratigraphy across its historic centre, locating a series of residential complexes,

*Fig. 10.5: Excavations during the installation of Thessaloniki's new water pipe (N. Bakirtzis).*

public buildings and churches, as well as evidence of the historical development of the city's road and water management infrastructure. Outside the city's walls, excavations uncovered late Roman and early Christian tombs belonging to Thessaloniki's *necropoleis*. Besides the collection of archaeological finds and the systematic recording of architectural remains, the excavated layers of the past were not exposed but were carefully covered as the water pipe work progressed.

Works for Thessaloniki's Metro (an ongoing project) necessitated systematic archaeological investigations. These commenced in 2006 in the areas of the underground stations located both within and on the outskirts of the historic city.104 In the historic centre, the Metro line runs approximately along the main Egnatia Avenue, which coincides with the course of the Roman city's *decumanus maximus*, the 'Leophoros' or 'Mese' of Byzantine Thessaloniki. Excavations at the Venizelou and Hagia Sophia metro stations have uncovered a relatively

<sup>102</sup> Bastéa and Hastaoglou-Martinidis, 'Urban change and the persistence of memory', 274–275.

<sup>103</sup> Bakirtzis, *Σωστική ανασκαφή αγωγού ύδρευσης.*

<sup>104</sup> Adam-Veleni and Mylopoulos, *The Metro-nome of Thessaloniki's History*.

undisturbed stratigraphic panorama of the city's long history. Of critical importance is the discovery of central parts of the 'Mese' and its continuous use through time to accommodate communication and commercial life. Archaeological finds and architectural remains help trace the city's evolution and its related socio-economic transformations. The importance of the archaeological discoveries has sparked heated debates about their protection, especially in the case of the Venizelou Street station. The proposed removal of the remains of the 'Mese' to permit construction and their re-assemblage as an integrated part of the station structure have met with fierce opposition from Thessalonians as well as the local and international archaeological community. Work is progressing as planned with tensions continuing to be high. Once more in the city's modern history, the preservation of the city's layered past clashes with the needs, realities and priorities of the growing city.<sup>105</sup>

Political, economic and cultural developments at the beginning of the twentyfirst century continue to shape perceptions of Thessaloniki's historical narrative and the related attitudes towards the remains of its layered past.106 Nonetheless, we can conclude that it is impossible even to begin considering the urban palimpsest of Thessaloniki without understanding the multifaceted impact of nineteenth and twentieth century events and the role of modern perspectives on heritage. Wars, political transitions, socio-economic volatility and both man-made and natural disasters provided opportunities and moments which demanded confrontation with the remains of the past. The integration of Thessaloniki into modern Greece prioritised the Hellenisation of its historical narrative, thus promoting the selective erasure and appropriation of surviving remains which highlighted its Byzantine heritage. In the end, the experience of the city's layered past relies on the living presence of its inhabitants and as such can only be known in memory. Nikos Gavriel Pentzikes, whose poems, texts and art obsessively searched for his mother city's true essence, explained: 'The place named by memory is ours … Shaped by the breath of memory we experience the reality of Thessaloniki'.<sup>107</sup>

### **Acknowledgements**

Aspects of this essay draw upon research pursued in the context of project 'Mediterranean Palimpsests: Connecting the Art and Architectural Histories of Medieval and Early Modern Cities' (Co-PIs: N. Bakirtzis, The Cyprus Institute and D. Fairchild Ruggles, University of Illinois at Urbana-Champaign) funded by the Getty Foundation's Connecting Art Histories Initiative. Many thanks to the organisers of the 'Cities as Palimpsests?' conference in Istanbul and to the Impact of the Ancient

<sup>105</sup> On the clash between different pasts and contemporary priorities in Beirut, see Assaad Seif in this volume.

<sup>106</sup> Hastaoglou-Martinidis and Christodoulou, 'Preservation of urban heritage', 121–147.

<sup>107</sup> Pentzikis, *Μητέρα Θεσσαλονίκη*, 133.

City ERC project team for their invitation to present and contribute some thoughts on Thessaloniki's layered past.

## **Bibliography**


Mazower, M., *Salonica, City of Ghosts: Christians, Muslims and Jews 1430–1950* (London, 2004).

Molho, R., 'The Jewish community of Salonika and its incorporation into the Greek State, 1912–19', *Middle Eastern Studies* 24 (1988): 391–403.

Molho, R., *Οι Εβραίοι της Θεσσαλονίκης 1865–1919. Μια ιδιαίτερη κοινότητα* (Athens, 2001).


The new and the old

## Chapter 11

## From Byzantion to Constantinople

## *Paul Magdalino*

The foundation of Constantinople was undoubtedly the most spectacular and the best-recorded act of urban transformation in the ancient world.1 It cannot, however, be regarded as a typical case study of urban evolution in late antiquity. While the object of transformation, the city of Byzantion, was in terms of size and status a fairly average example of a middle-ranking Greek *polis* under Roman imperial rule, the result of the transformation was exceptional by any standards and surpassed only by Rome itself. It is therefore without any pretence of providing or describing a paradigm of urban change in the late Roman world that this paper sets out to analyse the scale and the scope of the process by which Byzantion became Constantinople – how the city of the mythical Byzas turned into the city of the historical Constantine.

The scale of the operation is clear. We know more or less the line of the new land wall that Constantine built, enclosing a space that more than tripled the surface area of the pre-existing settlement;2 we know too that he instituted grain imports from Egypt on an industrial scale that emulated the food supply of Rome.3 But the scope of the foundation of Constantinople is more difficult to evaluate. We have practically no archaeological evidence, and sparse written evidence, for the articulation and typology of the built environment with which Constantine covered, or planned to

<sup>1</sup> Dagron, *Naissance d'une capitale* remains fundamental. The most important subsequent publications on the foundation are: Mango, *Le développement urbain*; Bassett, *The Urban Image of Late Antique Constantinople*; Bardill, *Constantine: Divine Emperor of the Christian Golden Age*; Van Dam, '"Constantine's beautiful city"'.

<sup>2</sup> The source for the line of the wall is the *Patria*, 1.53, *Scriptores rerum Constantinopolitanarum*, 141–142; the most recent discussion of the topography, is by Asutay-Effenberger and Effenberger, 'Eski-İmaret Camii, Bonoszisterne und Konstantinsmauer'; Asutay-Effenberger and Effenberger, 'Zum verlauf der Konstantinsmauer zwischen Marmarameer und Bonoszisterne und zu den Toren und Strassen'.

<sup>3</sup> On the grain supply of Constantinople, see now Van Dam, *Rome and Constantinople*.

cover, the intramural space. Moreover, the shape of Constantine's foundation, not to mention the pre-Constantinian city, was rapidly subsumed and superseded, in the following two centuries, by the prodigious expansion of Constantinople as an imperial and Christian capital. The narrative of the city's existence developed in the light of this expansion. The first and only systematic description of the city, the *Notitia Urbis Constantinopolitanae,* was written one century after the foundation, and what it documents is not so much Constantine's transformation of Byzantion as the upgrading of Constantine's megalopolis into the even greater megalopolis of Theodosius I and his dynastic successors.4

It is possible, nevertheless, to bring some clarity to the narrative of the foundation by comparing the city that Constantine found on the shores of the Bosporus in 324 with the city that took shape in his mind, as well as on the ground, over the next thirteen years of his life. The evidence for both is more informative than we might think, if we examine it without preconceptions. We can get a fairly precise idea of the layout of Roman Byzantion at the beginning of the fourth century, and we can get a fairly clear sense of how and why Constantine wanted the city that bore his name to differ from Byzantion and the kind of city that Byzantion represented.

### **Byzantion before Constantine**

Ever since the Greek colonisation of the Propontis and the Black Sea, Byzantion had played a prominent role in the history of the region, owing to the strategic and commercial importance of its site.5 Apart from its exceptional maritime position, however, there was little to distinguish Byzantion before Constantine from many another Greek *polis* that had flourished under Roman imperial rule. Like many cities throughout the East, it had benefitted from the patronage of Hadrian, who provided an aqueduct,6 and of at least one of his successors, whom later Byzantine tradition identified as Septimius Severus, although modern scholarship has been sceptical.<sup>7</sup> Like every Greek *polis*, the city was composed of acropolis and agora, and as in some cities in Asia Minor, the latter was further duplicated into an upper and lower agora.8 Otherwise, Byzantion was a classic example of the topographical and functional division between acropolis and agora. The agora area comprised, in addition to the market place, administrative buildings, schools, gymnasia and baths. The acropolis housed the temples and, on its lower slopes, the theatres and the stadium; the other

<sup>4</sup> Seeck, *Notitia Dignitatum*, 227–243. German tr. and commentary by Berger, 'Regionen und Strassen im frühen Konstantinopel'. English tr. and commentary by Matthews, 'The *Notitia Urbis Constantinopolitanae*', 81–115. Cf. Magdalino, 'Renaissances d'une capitale'.

<sup>5</sup> Russell, *Byzantium and the Bosporus*.

<sup>6</sup> Crow *et al*., *Water Supply of Byzantine Constantinople*, 10–14; for the line of the aqueduct within the city, see now Crow, 'The imagined water supply of Byzantine Constantinople'.

<sup>7</sup> Mango, 'Septime Sévère et Byzance'; Pont, 'Septime Sévère à Byzance: l'invention d'un fondateur', 191–198. The sceptical trend has now been reversed, however, with serious arguments for a Severan re-foundation, by Chatzilazarou, *Η Βασίλειος Στοά.*

<sup>8</sup> Notable examples were Pergamon and Ephesos: Saradi, *Byzantine City in the Sixth Century*, 211.

main entertainment venue, the Hippodrome, was a late addition and its location outside both the acropolis and the agora area was determined by the physical geography of the site.

The approximate layout of Roman Byzantion has long been known to scholarship, but a closer, comprehensive, contextual reading of the textual sources in relation to the physical geography of the peninsula and the sporadic archaeological evidence can help us to identify, locate and interrelate some sites more exactly, and thus to visualise more clearly the place of the ancient city in Constantine's vision of the future city (Figure 11.1). This has been demonstrated most recently by the work of Dimitris Chatzilazarou on the monumental centre of Constantinople.9 A key point in his analysis is the identification of the Basilica or 'Royal Portico', the large colonnaded courtyard whose site is marked by the Yerebatan cistern, as the upper agora of Roman Byzantion. As we shall see, this has important implications for the significance of the ensemble into which Constantine integrated the Basilica complex. It also indicates that the upper agora stood in a meaningful visual and axial relationship to the lower agora, the monumental square known as the Strategion. Although the site of the Strategion has not been pinpointed by archaeological finds, an emerging consensus locates it somewhere to the east of Sirkeci Station.10 This brings it very close to the axis of the Basilica, which follows the prevailing street alignment of Byzantion. In any case, the Strategion would have been clearly visible from the top of the grand staircase (72 steps) of the Basilica, which as Chatzilazarou, developing an observation of Cyril Mango, has argued, must have been on the north-eastern side.11

Apart from the Basilica, only two pre-Constantinian structures can be located on the basis of their physical remains: the *thermae* of Zeuxippos,12 in the monumental centre, and the monument to Tyche/Fortuna on the Acropolis, whose site is marked by the column of the Goths.13 However, a number of other structures can be mapped more exactly by closer scrutiny and coordination of the textual references. Byzantion had four main temples on the Acropolis, that of Poseidon at the foot of the hill,14 and those of Apollo, Aphrodite and Artemis on the hilltop.15 The second-century *Anaplus Bospori* of Dionysios of Byzantion indicates that the temple of Poseidon lay right beside the sea close to the tip of the peninsula. The subsequent fate of the building allows

<sup>9</sup> Chatzilazarou, *Η Βασίλειος Στοά*.

<sup>10</sup> Mango, 'The triumphal way of Constantinople and the Golden Gate', 187–188; Westbrook, 'Notes towards the reconstruction of the Forum of the Strategion'.

<sup>11</sup> Chatzilazarou, *Η Βασίλειος Στοά*; cf. Mango, *Brazen House*, 44. The staircase is mentioned by Zosimos, *New History*, 104; for the number of 72 steps, see *Parastaseis*, 40 (§37).

<sup>12</sup> See Müller-Wiener, *Bildlexikon zur Topographie Istanbuls*, 52 ; Mango, 'Septime Sévère et Byzance', 599–602.

<sup>13</sup> Müller-Wiener, *Bildlexikon zur Topographie Istanbuls*, 52; Peschlow, 'Betrachtungen zur Gotensäule in Istanbul'; Mango, *Studies on Constantinople*, 2; Mango, 'The triumphal way of Constantinople and the Golden Gate', 177; Croke, 'Poetry and propaganda', 462–463; Magdalino, 'The "Columns" and the Acropolis Gate', 147, 153.

<sup>14</sup> First mentioned by Dionysios of Byzantium, *Anaplus Bospori*, 5.

<sup>15</sup> John Malalas, *Chronicle*, XII 20, 221; XIII 13, 248; XIII 38, 267; *Chronicon Paschale*, 495. Cf. Mango, *Le développement urbain*, 18, 33.

*Fig. 11.1: Byzantion excluding the Hippodrome (adapted from W. Müller-Wiener,* Bildlexikon von Byzantion-Konstantinopel-Istanbul*, 497). Blue: pre-Constantinian structures, locations approximate except for the Basilica, the Baths of Zeuxippos and the Temple of Fortuna Redux (column of the Goths). Red: constructed or planned by Constantine. Yellow: schematic lines of streets corresponding to the boundaries of the urban* regiones *in the fourth-century* Notitia*. Green: hypothetical line of other streets.*

us to locate it more exactly. According to the *Patria* of Constantinople, the temple of Poseidon was replaced by the church of St Menas.16 Two later sources state that St Menas lay close to the complex of the Mangana,17 whose exact location beside the Bosporus was determined by the French excavations of 1921–1923.18 It is therefore on the eastern shore of the peninsula, between the Mangana and the Acropolis point, though closer to the latter, that the temple of Poseidon must be sought.

The other major pagan sanctuary that left a detectable trace in the topography of the Acropolis area was the temple of Aphrodite. According to Malalas, Theodosius I had it converted into a carriage house for the Praetorian Prefecture of the East.19 As Cyril Mango has noted, it is clear from the *Chronicle Paschale*'s description of the great fire of the Nika Riot in 532 that the Prefecture building was close to the church

<sup>16</sup> *Patria*, ed. Preger, *Scriptores*, I 15, 7; I 51, 140 (where the temple is said to have been dedicated to Zeus).

<sup>17</sup> Abbess Sergia, 'Narratio de translatione Sanctae Olympiadis', 45 (7th c.); Anonymus Mercati, 'Une description de Constantinople traduite par un pèlerin anglais', 250 (12th c.).

<sup>18</sup> Demangel and Mamboury, *Le quartier des Manganes et la première région de Constantinople*.

<sup>19</sup> John Malalas, *Chronicle*, 267.

of Hagia Eirene (Aya Irini).20 The location of the temple of Aphrodite in this area (the first court of the Topkapı palace), and that of the temple of Artemis further north on the ridge (the second court) can be confirmed by the evidence for the two theatres of Roman Byzantion. Malalas and the *Chronicon Paschale* record that Septimius Severus rebuilt both temples, 'building a very big amphitheatre [κυνῆγιν] opposite the sanctuary of Artemis, and a theatre opposite the sanctuary of Aphrodite'.21 'Opposite' (κατέναντι) here must mean 'to the east of' (on the assumption that their entrances faced the sunrise), or 'directly downhill from'. In the *Notitia* of *c*. 425 both theatres are clearly on the side of the hill overlooking the Bosporus, just inside Region II but on the boundary with Region I.22 The *amphitheatrum* or *theatrum maius* lies just beyond the north end of the first region, and the *theatrum* or *theatrum minus* at the point where 'the Second Region … rises in a gentle, almost imperceptible ascent': a description which, in this area, can only apply to the slope downhill from Aya Sofya and Aya Irini.23 We may note that it was near the temple of Aphrodite that the ancient city wall turned eastward to descend to the Bosporus coast.24 If the theatre was in this area, this would place the amphitheatre under the steep, precipitous slope to the north,25 beside the sea, in the area later occupied by the palace and monastery complex of the Mangana. Indeed, in the eleventh century, the imperial residence associated with the foundation of the monastery was originally known as 'the house of the Kynegion'.26

<sup>20</sup> Mango, *Studies on Constantinople*, 1–3.

<sup>21</sup> John Malalas, *Chronicle*, 221; *Chronicon Paschale*, 495.

<sup>22</sup> Seeck, *Notitia Dignitatum*, 230–231: Matthews, 'The *Notitia Urbis Constantinopolitanae*', 86–87. 23 Matthews, 'The *Notitia Urbis Constantinopolitanae*', 87, 101–102. Matthews' interpretation of the topographical details of the theatres in the *Notitia* is generally more convincing than that of Berger, 'Regionen und Strassen', 357–360.

<sup>24</sup> See Zosimos, *New History*, 2.30.3, to be read in conjunction with *Patria*, I 52, *Scriptores*, 104, which says that the wall descended to the Mangana via the Topoi. This location figures in the *Parastaseis* (38, § 32) and elsewhere in the *Patria* (2.27, 3.26, 164, 222) as a structure composed of 'seats' (*βάθρα*) where the emperor Zeno, after overcoming the usurpation of Basiliskos (475), sat in judgement on the usurper together with senate and clergy. This clearly refers to some sort of tiered seating, and hence may be identified with the *theatrum minus* of the *Notitia*. In any case, the fact that *theatrum minus* and *theatrum maius* were both connected and contained within Region II by its boundary with Region I suggests not only that the boundary was marked by a road between the two theatres, but also that the road and the regional boundary followed the line of the pre-Constantinian city wall.

<sup>25</sup> A hollow in the side of the hill beneath the kitchens of the Topkapı Saray has been identified as the possible site of a theatre: Tezcan, *Topkapı Sarayı ve Çevresinin Bizans Devri*, 120–123. Van Millingen placed the Kynegion closer to the sea, behind the Deïrmen Kapısı, in 'a hollow now occupied by market gardens', *Byzantine Constantinople*, 251. Proximity to the sea is confirmed by a law in the Theodosian code (*CTh*, 14.6.5) as noted by Matthews, 'The *Notitia Urbis Constantinopolitanae*', 103. The proposal by Demangel and Mamboury, *Le quartier des Manganes et la première région de Constantinople*, 15–17 and map, to locate the arena at the end of the peninsula, near the Column of the Goths, seems to me completely untenable.

<sup>26</sup> John Zonaras, *Epitome historiarum*, 647; cf. 619–620. There is no satisfactory study of the Mangana complex in all its documented phases from the sixth to the eleventh century; see provisionally Oikonomides, 'St. George of Mangana, Maria Skleraina, and the "Malyj Sion" of Novgorod', 239–246; Magdalino, *Medieval Constantinople*, 48–49, 87–88; Spingou, 'Snapshots from the eleventh century', 61–65.

In addition to the theatre and amphitheatre, Byzantion possessed a stadium, which probably dated from the city's pre-Roman past. Dionysios of Byzantion indicates that it was on the level ground inside the city wall along the Golden Horn, and Procopius, writing in the sixth century when Justinian converted it into a hostel, states that it was 'very close to the sea.'27

Having located the major monuments that have disappeared, we are now in a position to crack the enigma of the one pre-Constantinian monument that is still standing on the Acropolis of Byzantion: the column of the Goths. The Latin inscription on its base advertising its dedication to *Fortunae reduci ob Gothos devictos* indicates that it must have been put up at a time when it was normal to thank Fortune, rather than the Christian God, for victory over one's enemies; thus it most likely commemorated the defeat of the Gothic invasions of the 260s. The Greek translation of and commentary on the inscription by John the Lydian show that by the sixth century the column was associated with the memory of an adjoining temple of Fortune that had been turned into a tavern.28 Such a conversion is the subject of four epigrams by the poet Palladas (*AP* IX 180–183), who, it has recently been demonstrated, was active in Constantinople at the time of Constantine.29 Here it is important to point out that the monument occupied a key, nodal position in the Acropolis area. It stood at the crossroads of two major routes, one linking the city's two main entertainment arenas, the stadium and the amphitheatre, and the other the 'sacred way' linking Byzantion's two major cult locations, the temple of Poseidon by the sea and the temples of Artemis, Aphrodite and Apollo up on the hill.30

When we consider the continuation of these routes to the south, an even more interesting pattern emerges: the column of the Goths and the adjacent temple of Fortune was one intersection between the sacred axis of the Acropolis and the peripheral circuit of the Acropolis, the route linking all the main civic monuments that flanked the Acropolis hill, from the two *agorai* (Strategion and Basilica) on the western side to the theatres (amphitheatre and *theatrum minus*) on the eastern side. In fact, the constructions attributed to Septimius Severus – the three rebuilt temples along the Acropolis ridge, the two theatres on the eastern slope, the Baths

<sup>27</sup> Procopius, *De aedificiis*, 1.24–1.27. The Stadium was evidently not far from the temple of Poseidon; Dionysios of Byzantium implies that it was the first building that one saw after the temple when sailing around the headland into the Golden Horn from the Bosporus (see above n. 5), and it is listed along with St Menas, which replaced the temple, in Region IV of the fifth-century *Notitia*: Seeck, *Notitia Dignitatum*, 233; Matthews, 'The *Notitia Urbis Constantinopolitanae*', 89.

<sup>28</sup> John Lydus, *De mensibus*, 161.

<sup>29</sup> Wilkinson, 'Palladas and the age of Constantine'; Wilkinson, 'Palladas and the foundation of Constantinople', esp. 181–185, where the author attempts to locate the converted temple in Constantinople with reference to the passage in Lydus (see previous note), but without recognising the site of the Latin inscription that Lydus translates.

<sup>30</sup> It may also have stood on the shortest route from the temple of Poseidon to the Stadium and the city centre; the 'front' of the column, *i.e*. the side with the inscription, overlooks the sea coast at the point where I would be inclined to place the temple, in view of the proximity of the latter's replacement, St Menas.

of Zeuxippos and the Hippodrome to the south – gave Byzantion the plan of a secular ellipse bisected by a sacred axis. Yet the plan was not quite complete. There was a gap in both the sacred axis and the secular circuit at the point where their second, southern intersection should have been. The gap was to be filled by the imperial Great Palace and the Great Church of Constantinople, consisting of the cathedral churches of Holy Peace and Holy Wisdom. Thus the nerve centre of Constantinople was firmly embedded in the urban fabric of Byzantion.

## **Constantine's vision of Constantinople**

Discussion of Constantine's vision for Constantinople has tended to revolve around two questions: firstly, did he conceive of it as a New Rome, and secondly, did he intend it to be a Christian capital? Both questions have been widely aired in the extensive literature on Constantine, and both have been answered with a mixture of yes and no, more yes in the first case, more no in the second case. Revisiting the questions will not significantly change the answers, and will not get us any further unless we push them further down the line of enquiry, so that they follow on from, and do not precede, an attempt to reconstruct the design that Constantine had in mind as the foundation took shape on the ground from 324 to 330. This design may be approached in terms of the following considerations:


## **1. Why Constantinople?**

Constantine's motives in founding Constantinople are ultimately a matter for speculation, but the sources provide three credible explanations: he wanted to build a city that bore his name, he wanted this city to imitate and emulate Rome and he wanted to mark his victory over his co-emperor Licinius in the civil war that had broken out between them.31 The last of these explanations is the key to all three.32 It is confirmed by the time and place of the operation: the site of the new city was formally inaugurated on 8 November 324, two months after Licinius had surrendered to Constantine at Chrysopolis. The victory not only put an end to two years of civil

<sup>31</sup> Dagron, *Naissance d'une capitale*, 19–31, 43–47.

<sup>32</sup> It is made explicit by the Anonymus Valesianus, MGH AA, 9.10.

war, and eight years of strained relations between the rulers of the eastern and western halves of the Roman Empire, it also ended almost forty years of power-sharing among a college of Augusti and Caesars. Just as the battle of Actium had marked the bitter end of the Second Triumvirate, the battle of Chrysopolis was the last gasp of the Tetrarchy, and the 'victory city' that commemorated the event symbolised the reunification of the Roman world.33 But the city of Constantine was an even more powerful statement of unity than the Nikopolis of Augustus.34 It not only reaffirmed the domination of Rome, but also re-imposed the physical presence of Rome in the Greek East. It said goodbye to the decentralisation of the Tetrarchy, to the detachment of imperial power from Rome and to the establishment of alternative power bases with a predominantly military function. This point is important to grasp. Constantinople was not just one more Tetrarchic capital, but it marked a conscious break with the Tetrarchic model of urban development.

At the same time, Constantinople was the logical outcome of the Tetrarchic system, in the sense that it did not put back the clock to the pre-Tetrarchic world, when all roads led only to Rome; rather, it finalised the trend to establish a Rome away from Rome, albeit a Rome with a difference. It answered to the strategic need, evident from the mid third century, for a major imperial power base in the Balkan-northern Aegean area. It also provided a focus for the ambitions and loyalties of the civic aristocracies of the Greek East; in terms of recent politics, it facilitated the task of taking over and consolidating the regional patronage networks operated by Diocletian, Galerius and Licinius from their court at Nicomedia. Constantinople provided the eastern cities with a power centre and power structure that was closer to home, culturally as well as geographically, than Rome. In turn, the Greek East enveloped Constantine in a style of ruler cult that still went against the grain of the Roman Senate. To go back to the comparison with Augustus, Constantine's *nikopoli*s was not founded in Rome's shadow or on Rome's behalf. In its origins, it had most in common with the city that would have become the capital of the Greek East if the battle of Actium had gone the other way: Alexandria. Constantine signalled his adherence to Hellenistic tradition by having his city personified as a goddess with a turreted crown, and by crowning himself with a Hellenistic style diadem on the day of the city's inauguration; as the chronicler John Malalas remarked, 'None of the emperors who had ruled before him ever wore such a thing'.35

It had important implications for the design of Constantinople that although it duplicated Rome as an *urbs regia/βασιλεύουσα πόλις*, it did not replicate, but inverted, the arrangement by which that unique status was achieved: in Rome, the senate and people preceded the emperor, and formally invested the emperor with his power; in Constantinople, the emperor created the city and its institutions, and he created

<sup>33</sup> Dagron, *Naissance d'une capitale*, 25, for Augustus' foundation of Nikopolis. 34 For the Augustan precedent, see Jones, 'Cities of victory', 99–108.

<sup>35</sup> John Malalas, *Chronicle*, 8.7–8, ed. Thurn, 245–247; cf. Lenski, 'Constantine and the Tyche of Constantinople'; Chatzilazarou, 'Le centre monumental de Constantinople', 44–45.

them as an inheritance for his family. Constantinople was thus, as Dagron and others have emphasised, in origin an exclusively dynastic foundation, and it never really lost this dynastic character.36 Moreover, although undoubtedly a Christian city, it was created less obviously as an act of piety, than as a function of the divinity in which Constantine assumed that his piety entitled him to share.37 The religious dimension of the foundation was subordinate to the political dimension, and only became meaningful in retrospect.

As for the siting of the new 'victory city', once Constantine had decided to found it in the general area of the Propontis, the choice of Byzantion, where he had besieged Licinius prior to defeating him across the Bosporus, was inevitable. The advantages of the site outweighed the disadvantages, of which only the inadequate water supply was serious though not insuperable. To the well-known advantages we might add the point that Byzantion was already equipped with the accoutrements of a Roman city, including an amphitheatre and perhaps the beginnings, at least, of a hippodrome. Later tradition, as we have seen, ascribed these improvements to Septimius Severus, but the intriguing suggestion has been made that the Hippodrome was begun by Licinius,38 which of course Constantine could not acknowledge, any more than he had acknowledged the buildings of Maxentius that he had appropriated and reused in Rome.39

### **2. The other cities in Constantine's life**

What cities other than Rome had made an impression on Constantine that might have influenced his project for Constantinople? He had travelled extensively throughout the empire, both before and after becoming emperor in 305. He knew the highly urbanised provinces of the northern Mediterranean as well as the militarised zones behind the British, Rhine and Danube frontiers. Yet the only cities apart from Rome where he had resided for any length of time were the new provincial capitals of the Tetrarchy: Nicomedia, Trier, Milan, Sirmium, Thessalonica and, in a lower register, places like Arles and Sardica. He had no direct exposure to the classical *poleis* of the Aegean area, although he was apparently a great admirer of Athens.40 He had visited the great Hellenistic cities of Antioch and Alexandria in his youth, and he revisited

<sup>36</sup> Dagron, *Naissance d'une capitale*, 26–28; Van Dam, '"Constantine's beautiful city"', 91–93. Cf. Magdalino, 'Renaissances d'une capitale'.

<sup>37</sup> The most complete and balanced study of Constantine's religious profile is Bardill, *Constantine: Divine Emperor of the Christian Golden Age*.

<sup>38</sup> A passing comment by Albrecht Berger has been seized on by other scholars and expanded into a hypothesis that Licinius was developing Byzantion as his capital: see Berger, 'Streets and public spaces', 165; Van Dam, '"Constantine's beautiful city"', 86; Stephenson, *Constantine*, 92–94, 139; Barnes, *Constantine*, 112–113. 39 Bardill, *Constantine: Divine Emperor of the Christian Golden Age*, 83–84, 92–93, 99.

<sup>40</sup> Julian, *Oration* I 8 C-D, 20–23. The sophist Praxagoras of Athens wrote an encomiastic biography of Constantine: Photius, *Bibliotheca*, cod. 62; Barnes, *Constantine*, 195–197; Krallis, 'Greek glory, Constantinian legend'.

*Fig. 11.2:* Folles *of Constantine I, with the image of the camp (or city?) gate on the reverse.*

Antioch while Constantinople was under construction. But his experience of the Greek city was mainly limited to the two Tetrarchic capitals of Nicomedia and Thessalonica, important Greek regional centres that Diocletian and Galerius, respectively, had expensively upgraded to serve as imperial residences.41 Constantine spent two years (322–324) in Thessalonica while preparing his final showdown with

Licinius.<sup>42</sup> He had been based at Nicomedia with the court of Diocletian from 302 to 305, 43 and he returned there in 324, using the city as his main residence while Constantinople was being built.44 Both cities were therefore fresh in his mind at the time of the foundation. Both were possible alternatives to Byzantion – Thessalonica is mentioned as such, and Nicomedia was too obvious to need mentioning. In rejecting them both, Constantine was not only voting in favour of Byzantion's superior site, but also escaping from the shadow of the Tetrarchic past that hung over them both, and would have cramped the style of his programme to create a city in his own name and according to his own specifications.

Thessalonica and Nicomedia clearly served as precedents for Constantine's intention of re-imposing a Roman imperial presence on a Greek civic environment. Of all the cities in which he had resided, they were probably the most influential. But the point not to neglect here is that Constantine, like all his Tetrarchic predecessors, was often on the move between imperial residences. Movement to and from cites was just as important as sedentary residence in his experience of urban space. The type of urban building most commonly depicted on the reverse of his coins, like those of Diocletian and Galerius before him, was the city gate opened to receive the emperor arriving in triumphal *adventus*; significantly, most of Constantine's *folles* bearing this image were issued in 324–329, exactly when Constantinople was under construction (Figure 11.2).45

### **3. Sources for Constantine's urban design**

Few texts give details of Constantine's building programme in Constantinople; fewer still attempt to explain its rationale, and even so, their accounts do not add up to a consistent picture of urban planning and transformation. While they make it very clear that the decisive factor was the hugely enlarged perimeter, they give a confused,

<sup>41</sup> Foss, 'Nicomedia and Constantinople', 181–183; Tourta, 'Thessalonike', 75–79.

<sup>42</sup> Zosimos, *New History*, II 22, 1; Pseudo-Symeon, ed. Halkin, 'Le règne de Constantin', 17. Cf. Makri, 'Έργα και ημέρες του Μεγάλου Κωνσταντίνου στη Θεσσαλονίκη', 77–87.

<sup>43</sup> Barnes, *Constantine*, 51–5.

<sup>44</sup> *Chronicon Paschale*, 527; Foss, 'Nicomedia and Constantinople', 181–183.

<sup>45</sup> I follow the reinterpretation of the 'camp gate' iconography by Woods, 'The late Roman "Camp gate" reverse type and the Sidus Salutare', 159–174.

vague and incomplete impression of the intended layout of the intramural space. The text that is closest in time to the foundation, Eusebius' *Life of Constantine*, is also the most partial, effectively claiming that Constantine programmed Constantinople to be a Christian city, with 'very many places of worship, very large martyr shrines', with Christian imagery and symbolism on public display, and with a ban on pagan worship and images,46 apart from those Hellenic cult statues that Constantine had brought from elsewhere and exposed to public ridicule by setting them up in the streets and squares.47 Mythological statuary did indeed have a very prominent place in the cityscape of Constantinople from the beginning, but its purpose was not quite so simple as Eusebius alleges.48 Eusebius gets closer to Constantine's purpose when he refers to Constantinople as 'his own beautiful city' ('τῆς αὐτοῦ καλλιπόλεως').<sup>49</sup> *Kallipolis –* 'Beautiful City' or 'Fair City' – was almost the standard designation for Constantinople in the fourth century, until it fully adopted the title of 'Royal City' or 'Sovereign City' (urbs regia/βασιλεύουσα πόλις) in its capacity as a New Rome.<sup>50</sup>

As for Eusebius' statement that Constantine provided his city with 'very many places of worship, very large martyr shrines',51 this is impossible to prove or to disprove, precisely because these churches were not a structural part of the city's armature, in the way that the temples on the Acropolis had formed the backbone of ancient Byzantion. The only exception was the Christian cathedral, which was added to the pre-Christian sacred axis of the Acropolis.52 The same point may be made with respect to the houses that Constantine built for the senatorial aristocracy; they were unquestionably a vital component of his vision for a grand capital city, but they apparently followed and did not shape the urban layout.53

The sources that emphasise Constantine's imitation of Rome mention only selected buildings – the Palace, the Hippodrome and the Senate – by way of example.<sup>54</sup>

<sup>46</sup> *Vita Constantini* III, 47–49

<sup>47</sup> *Vita Constantini* III, 54).

<sup>48</sup> Bassett, *The Urban Image of Late Antique Constantinople*, 37–78; Magdalino, 'The myth in the street'.

<sup>49</sup> Sozomenos, *HE*, 2.5.3, states that 'the wonderfully worked bronze statues were brought from everywhere to the emperor's eponymous city as ornament' ('τὰ δὲ ἐν χαλκῷ θαυμασίως εἰργασμένα πάντοθεν εἰς τὴν ἐπώνυμον πόλιν τοῦ αὐτοκράτορος μετεκομίσθη πρὸς κόσμον'); *Vita Constantini* III, 55.1.

<sup>50</sup> Dagron, *Naissance d'une capitale*, 52, citing in n. 5 the numerous references to Constantinople as *kallipolis* in the orations of Themistios. It is notable that Themistios first uses the expression in an oration (3.41a, 42a, 44b–c) which was written to be delivered at Rome in 357 and was largely concerned with the relationship of Constantinople and Rome, broaching the subject of Constantinople as a second capital: Schenkl and Downey, *Themistii orationes quae supersunt*, 59, 60, 63; Heather and Moncur, *Politics, Philosophy and Empire in the Fourth Century. Select Orations of Themistius*, 125, 126, 130. As the translators pertinently note (n. 242), *kallipolis* was the epithet used of the ideal city by Plato in the *Republic*.

<sup>51</sup> An assertion amplified by Sozomenos, *HE*, 2.3.7–8.

<sup>52</sup> *i.e*. the church of Hagia Eirene: Dagron, *Naissance d'une capitale*, 392–393.

<sup>53</sup> Zosimos, *New History*, 2.31.3, ed. Paschoud, 104; Sozomenos, *HE*, 2.3.4, says that Constantine built these 'very great houses at intervals along the streets' ('μεγίστας οἰκίας ἀνὰ τὰς ἀγυιὰς σποράδην οἰκοδομήσας').

<sup>54</sup> Zosimos, *New History*, 2.31.1, ed. Paschoud, 104; John Malalas, *Chronicle*, 13.7, ed. Thurn, 245; *Chronicon Paschale*, 527–528; Philostorgios, *Historia ecclesiastica*, 2.8–9. ed. Bleckmann and Stein, *Philostorgios, Kirchengeschichte*, 190.

However, two of these sources convey a more holistic impression in the structure of their foundation narratives. Philostorgios and Zosimos both emphasise the pivotal importance of the site, at the former land gate of Byzantion, where Constantine laid out his eponymous Forum and erected the porphyry column topped by his colossal statue.55 Philostorgios describes only the development of the city in the direction of the new land wall, and this may be why the only Roman institution that he mentions is the Senate, which had one of its two buildings at the Forum.56 Zosimos' narrative moves from the Forum in the other direction, to describe the imperial Palace and the improvements that Constantine made to previously existing structures, the Hippodrome and the Basilica.57 But in a later section of his *New History*, Zosimos returns to the subject of Constantinople to describe and deplore the phenomenal expansion of the city, which he clearly presents as a secondary phase of development, associated with the promotion of (the last of) Constantine's sons to the rank of Caesar.58 It is hard to avoid the impression that both historians are implying a distinction between the old town of an upgraded Byzantion and a new town that was Constantine's city properly speaking.

There is only one narrative of the foundation that explicitly describes the plan of the intramural space. This is in the tenth-century *Patria of Constantinople*, the collection of stories and short notices about the origins, statues and buildings of the city. In the last section of Book I, which the tenth-century author added to the legendary account of Byzantion's early development from Byzas to Constantine, it is said that Constantine:

built four porticoes with brick and mortar vaults from the Palace as far as the land walls. One led from the [Palace] polo-ground via the Mangana and the Acropolis and *ta Eugeniou* and went to St Anthony's. Another went from the Daphne via the [harbour of] Sophiae as far as Rabdos. The other two went from the Chalke [Gate] via the Milion and the Forum to [the Forum of] the Bull, [the Forum of] the Ox and the Exokionion. On top of the porticoes were walkways paved with stone slabs and innumerable bronze statues stood there for the embellishment of the city.59

Almost no-one has taken this information seriously, and it is easy to dismiss as the fantasy of a notoriously inventive author, who at this point in his composition was using the hagiography that had grown up around the person of Constantine and the legendary figure of the eunuch Euphratas who had helped in the construction of the city.60 It is impossible to find parallels and precedents in any other ancient city for the arrangement described here, according to which the intramural space of

<sup>55</sup> Ousterhout, 'The life and afterlife of Constantine's Column', 304–326; Kaldellis, 'The Forum of Constantine in Constantinople'.

<sup>56</sup> Philostorgios, *Historia ecclesiastica*, 2.3–9, 188–190; Yoncacı Arslan, 'Towards a new honorific column'.

<sup>57</sup> Zosimos, *New History*, 2.30.4–31.3, 103–104.

<sup>58</sup> Zosimos, *New History*, 2.35, 108.

<sup>59</sup> *Patria*, 1.68, ed. Preger, 148–149; Berger, *Accounts of Medieval Constantinople: The Patria*, 41–43.

<sup>60</sup> See now Gerhold, 'The legend of Euphratas'.

Constantinople was framed by a web of porticoes, two central and two peripheral, emanating from the Palace. The arrangement also does not correspond to the street layout that can be surmised from the alignment of standing and excavated Byzantine structures.61 Nevertheless, the author of the *Patria* never demonstrably invents constructions that were not consistent with, or suggested by, structures visible to himself and his readers. There are traces of sections of porticoes running parallel to the coast.62 There is also one written testimony that has been entirely overlooked. Themistius, in an oration to Constantius II, praises the emperor for having 'girded the city with a roofed corridor like a precious girdle'.63 It is difficult to escape the conclusion that Constantius had completed his father's project to build a covered portico around the coast of the peninsula, just as the *Patria* describes. In any case, the existence and importance of the two central porticoes that ran from the Palace to the land wall cannot be in doubt, in view of all the other evidence for the central avenue (Mese) of Constantinople and the porticoes that lined it on either side. The *Patria* thus confirms that in this, at least, the planning of Constantinople conformed to the dominant trend of late antique urbanism, whose defining feature was the colonnade lining the processional way.64

#### **4. Constantine's building programme**

All these partial indications are valuable, but they cannot be used to reconstruct the initial design of Constantine's city unless they are integrated into the pattern that emerges from the ensemble of evidence for the topography and typology of the constructions that can reliably be attributed to Constantine. The analysis must start with the line of the new land wall, which, all the sources agree with hindsight, was Constantine's chief priority. It is remarkable that he did not build this wall over the isthmus between the Golden Horn and the Sea of Marmara, but some 300 metres to the west, where it had to be four times longer.65 Being that much longer, it was correspondingly all the more impressive, and to this extent was an end in itself. At the same time, Constantine evidently wanted the wall to enclose the hill on which he later built his mausoleum, and to protect the bays on either side of the isthmus, which would later become important port and commercial areas. He also, presumably,

<sup>61</sup> Berger, 'Regionen und Strassen'; Berger, 'Streets and public spaces'.

<sup>62</sup> Magdalino, 'The maritime neighborhoods of Constantinople', 224.

<sup>63</sup> Schenkl and Downey, *Themistii orations quae supersunt*, 1.75: 'δρόμῳ τε αὐτὴν ὑποστέγῳ οἵα πολυτελεῖ μίτρᾳ διαζωννύς'.

<sup>64</sup> Dey, *The Afterlife of the Roman City*.

<sup>65</sup> The existence of an isthmus about 900 m wide between the modern location of Unkapanı and the mouth of the river Lykos at Yenikapı was postulated by Cyril Mango twenty years ago: Mango, 'The shoreline of Constantinople in the fourth century'; Mango, *Le développement urbain*, 16–17. Mango's hypothesis was confirmed, at least for the Marmara shore, by the subsequent excavations at Yenikapı, which still await publication. It has been suggested that the isthmus was already fortified prior to Constantine by the 'forewall' (*proteichisma*) mentioned in later, semi-legendary narratives of the city's foundation: Effenberger, 'Zur wiederverwendung der venezianischen Tetrarchengruppen in Konstantinopel'.

*Fig. 11.3: Constantinople in the fourth century (adapted from C. Mango, 'The shoreline of Constantinople in the fourth century', p. 21), showing the main Constantinian monuments: Golden Gate (*Porta Aurea*), Constantine's Mausoleum and the Church of the Holy Apostles (HA), the Capitol (C), Tetrapylon (T, possibly not Constantinian), Forum (F). The monumental centre, comprising: Basilica (B), Hagia Eirene (E), Milion (M), Hagia Sophia (S), Hippodrome (H), Baths of Zeuxippos (Z), Augusteum (A), Senate House (S), Great Palace (P).*

wanted to make room for a large future expansion of the settlement. But in that case, it is interesting that his intramural building programme does not reflect a symmetrical, centric development of the intramural space. His Forum was slightly off-centre, and it was not at the centre of his main monumental constructions (Figure 11.3). The monumental centre of Constantinople was further east, in the old city.66 It was a huge ensemble, consisting of monuments that Constantine took over and renovated or completed (the Basilica, the Baths of Zeuxippos, the church of Hagia Eirene, probably the Hippodrome), along with others that he added (the Milion, the Palace, the Augousteion) and one that he almost certainly projected but left for his son Constantius II to attend to: the church of Hagia Sophia.67 It is important to recognise that this ensemble, which housed all the functions of the new capital apart from defence and commerce, was a seamless extension of Roman Byzantion, following the street plan and contained within the precinct of the ancient city, except for the Palace,

<sup>66</sup> See in general Bassett, *The Urban Image of Late Antique Constantinople*, chap. 1; Chatzilazarou, *Η Βασίλειος Στοά*; Chatzilazarou, 'Le centre monumental de Constantinople'.

<sup>67</sup> This is probably the best way to reconcile the sources that attribute the building to both emperors: Dagron, *Naissance d'une capitale*, 397–399; Chatzilazarou, 'Le centre monumental de Constantinople', 53–54.

which was built just outside the line of the wall indicated by Zosimos and the *Patria*. 68 There was no radical remake or dislocation of the existing settlement. Constantine may have turned the Temple of Fortuna Redux into a tavern, and converted the temple of Poseidon into the church of St Menas, but he did not convert or demolish the temples of Artemis, Aphrodite and Apollo. Rather, the churches of Hagia Eirene (Holy Peace) and Hagia Sophia (Holy Wisdom) were added to the sacred axis formed by these temples. Neither did Constantine interfere with the theatre, amphitheatre and stadium of Byzantion: all three are landmarks in the *Notitia* of *c*. 425. He adorned the ancient agora, the Strategion, with an equestrian statue of himself, and it was here, in the heart of ancient Greek Byzantion, that he is said to have posted the monumental inscription designating Constantinople as the New Rome, with all the privileges of the mother city.69

What we have been describing so far is the insertion of a Roman monumental and imperial presence into the fabric of an already Romanised Greek city, in a manner that was by now becoming conventional, through the precedents of Nicomedia and Thessalonica. Indeed, Constantine seems to have have conserved the pre-existing fabric of Byzantion more faithfully than Diocletian and Galerius had done in their respective capitals. In both Nicomedia and Thessalonica, the imperial palace and Hippodrome complex was created in an already built-up, intramural area, 70 while in Byzantion, the Palace did not intrude upon the urban space, and its location may well have been determined by the prior existence of a hippodrome. The only thing that seriously differentiated Constantinople from these cities was the thing that allowed Constantine to be more conservationist: the addition of the enormous extra intramural space defined by the new land wall. It is hard not to see the urbanisation of this space as a secondary development. It certainly presents an enigma in terms of urban planning, because it shows no coordinating principle apart from the line of the former access road that now became its central avenue, with a branch to the north-west. Two monuments in this area can definitely be attributed to Constantine: the Forum just outside the former Thracian Gate of Byzantion, and the founder's mausoleum, which doubled as a sanctuary of the Apostles, on the top of the highest intramural hill.71 Less certain, though probable, is the Constantinian origin of the Capitol at the point where the central avenue divided.72 The spatial and geographical

<sup>68</sup> See above, n. 24.

<sup>69</sup> Socrates, *HE*, 1.16,1; Hesychios, *Patria*, 39, ed. Preger, *Scriptores*, 17.

<sup>70</sup> The intervention seems, however, to have been less severe in Thessalonica than in Nicomedia, see Foss, 'Nicomedia and Constantinople', 182; Spieser, 'Réflexions sur le Palais de Galère à Thessalonique', 19–20.

<sup>71</sup> For the origins and development of this monument see Mullett and Ousterhout, *The Holy Apostles*, especially chapters 6 (Johnson) and 8 (Magdalino).

<sup>72</sup> It is attributed to Constantine in only one manuscript of the *Patria* (Vat. Ang. Gr. 22, ed. Preger, *Scriptores*, 18), and is otherwise first mentioned in connection with a storm and earthquake in 407 that brought down 'the *signochriston* [a cross or chi-rho symbol ?] of the Capitol': *Chronicon Paschale*, 570. For further discussion and bibliography, see Effenberger, 'Zur wiederverwendung der venezianischen Tetrarchengruppen in Konstantinopel', 221–227, 245–251; Magdalino, 'Renaissances d'une capitale', 62–64.

*Fig. 11.4: The regions of fifth-century Constantinople, illustrating the street layout of the Constantinian 'new town' between the Capitol and the Forum, as reflected in the boundaries between regions VI–X.*

logic of their situation seems clear, and so does their symbolic function, because all three monuments echoed distinctive features of Rome. The sanctuary of the Apostles recalled the mausolea of Augustus and Hadrian, as well as the martyrium of St Peter, on the western edge of Rome. The Capitol was obviously meant to recall the sacred Roman citadel, and its significance is underlined by the fact that it was the only one of its kind in the Greek East, apart from the complex that Hadrian had established at Aelia Capitolina in Palestine, the former and future Jerusalem.73 The Forum echoed both the Forum Romanum and the imperial fora nearby, especially through the presence of a Senate House that was unique outside Rome. These monuments clearly stamped the new city as the Roman extension to Byzantion, the space that gave the new foundation its distinctively Roman identity (which the Palace and Hippodrome alone did not provide). But as such, it was a remarkably free reinterpretation of its model, not only in its architectural language and its personality cult of its creator, but also in the spatial deployment of its principal elements, above all the separation of the Capitol both from the Forum, and, even more, from the monumental centre of imperial power. It is as if the point was to enhance the impact of the ensemble by spacing out its constituent monuments and extending the connections, in the form of avenues lined with porticoes, which linked them.74

<sup>73</sup> See Quinn and Wilson, 'Capitolia'.

<sup>74</sup> Dey, *The Afterlife of the Roman City*, 77–85.

We shall return to consider the visual implications of movement through and towards the city, but what should be observed here is that the line of the avenue mattered much more, in the planning of the new city, than the space through which it passed. This space shows no monumental construction away from the central axis, with the slight exception of Constantine's mausoleum, and no coherent geometric design. From the mention in later sources of a tetrapylon on the avenue west of the Forum of Constantine, at a junction where a well attested cross-street led down to the Golden Horn, Cyril Mango inferred the existence of a major intersection of *cardo*  and *decumanus*, which would have implied a Hippodamian street grid, in the section of the city between the Forum and the Capitol.75 This intersection has found its way into almost all subsequent plans of Constantinople. But the regional divisions of the *Notitia*, which are our only indirect evidence for the street layout in this area, do not support such an interpretation (Figure 11.4).76 While the boundaries between regions V, VI, VII and X, to the north of the Mese, suggest that the main streets here ran down to the Golden Horn at right angles to the avenue, the boundary between regions VIII and IX indicates a mainly east-west orientation along the terraces of the steep slope. It possibly followed the line of the Hadrianic aqueduct, which, according to the latest thinking, passed to the south of the Mese.77 Moreover, the breakdown by region of the numbers of houses shows that the northerly regions (V–VII, X) were more densely populated than those to the south, suggesting that the Golden Horn side of the 'new town' was built up earlier and more densely than the Marmara side of the ridge.78 In this scenario, the Mese would initially have functioned as the southern limit, rather than the central axis, of the city's expansion to the west.

The possibility that the foundation of Constantinople occurred in stages, in which Constantine's vision of his city evolved as his building programme progressed, is supported by chronological evidence that has not been fully taken into account. While there is no reason to doubt that Constantine conceived the idea of founding a city in his name immediately after his victory over Licinius, and that he formally inaugurated the project on 8 November 324, at the same time that he raised his son Constantius to the rank of Caesar,79 it cannot be assumed that the dimensions and status of the city were fully worked out at this point. Indeed, three sources clearly indicate that work did not begin in earnest until three to four years later. The *Chronicon Paschale* dates the construction to after Constantine's return from Rome in 328, when he was based at Nicomedia and made regular visits to Byzantion. Zosimus says that it was the emperor's bad experience in Rome which prompted him to create an alternative imperial capital. 80 According to the *Patria*, the foundations of the new city wall were

<sup>75</sup> Mango, *Le développement urbain*, 30–31.

<sup>76</sup> Magdalino, 'Le premier urbanisme byzantin', 85–86.

<sup>77</sup> Crow, 'The imagined water supply of Byzantine Constantinople' (above n. 6), maps on 224 and 226.

<sup>78</sup> For an analysis of the figures with a different argument, see Anderson, 'Social clustering in 5th-c. Constantinople'.

<sup>79</sup> Schenkl and Downey, *Themistii orations quae supersunt*, 1.83; Dagron, *Naissance d'une capitale*, 26.

<sup>80</sup> Zosimos, *New History*, 2.29–30.1

laid on 26 November 328.81 All this suggests that what Constantine envisaged in 324 may have been altogether more modest than what took shape in the accelerated development of 328–330. There may be an echo of his original intention in the rather puzzling information of the *Chronicon Paschale* that he 'renovated the first wall of Byzas' city, making not a few additions which he joined to the ancient city wall'82. This hardly works as a description of Constantine's land wall in its final form, but it could plausibly describe an earlier project to shift the southern part of the wall of Byzantion further to the west, so as to follow a course from the Thracian land gate south to the Sea of Marmara and thus enclose the Hippodrome and the Palace fully within the intramural area.83 Be that as it may, it seems clear that the decision to transform Byzantion into a *megalopolis* was taken a good three years after the decision to make it the site of Constantine's *nikopolis*, and barely two years before the formal inauguration of Constantinople on 11 May 330.

## **5. Approaching the city**

In considering Constantinople as *megalopolis*, we should reflect that city founders and city dwellers in antiquity did not have a bird's eye view of their cityscapes. Their normal perspective was the 'street view' of horizontal vision and horizontal movement. This is relevant for understanding Constantine's vision of Constantinople and the function of the huge extra space that he added to the area of Byzantion. When we view this new town from the viewpoint of the itinerant soldier emperor who had spent his whole career on the move between military camps, fortresses and Tetrarchic residences, we can get a better idea of what he wanted from it. Constantinople was designed as a city not only to be lived in but also to be moved to, and moved through, from west to east for the best effect. It was a city built to maximise the effect of a triumphal imperial *adventus*.

It is unlikely that the western approaches to the city were highly developed before the mid fourth century, although Rhegion (Küçükçekmece) may already have been an important outpost. But the new city wall – extended, as we have seen, for maximum effect, and entered through an imposing monumental gateway84 – was a spectacular introduction to what lay inside. The visual impact of what lay inside was then enhanced by the suspense of not seeing it all at once, but building up to the main monumental centre by moving through a drawn out series of progressively imposing monumental landmarks: the Capitol, possibly the Tetrapylon at the central T-junction, and the

<sup>81</sup> *Patria*, 1.55, ed. Preger, *Scriptores*, 142–3.

<sup>82</sup> *Chronicon Paschale*, 628: 'ἀνενέωσε τὸ πρωτον τεῖχος τῆς Βύζου πόλεως, ποιήσας καὶ προσθήκας τῷ αὐτῷ τείχει ούκ ὀλίγας, καὶ συνῆψεν τῷ παλαιῷ τείχει τῆς πόλεως'.

<sup>83</sup> This is in fact the line that both Mango (*Le développement urbain*, fig. 1, and 'The shoreline of Constantinople in the fourth century', 21) and Berger ('Regionen und Strassen', 361) assumed for the pre-Constantinian wall, without citing any evidence but overlooking that of Zosimos, *New History*, 2.30, 3 (see above, n. 24).

<sup>84</sup> On the Constantinian 'Golden Gate', see Kazan, 'What's in a name?'.

Forum of Constantine. The effect of the monumental clusters was enhanced by the long colonnades connecting them; they were meant to be serial rather than simultaneous and symmetrical. The relief of the peninsula was also used to effect. Arriving at the lowest point on the route, where it crossed the mouth of the Lykos river, the traveller in 337 would have seen Constantine's mausoleum on top of the hill to the left, and as he followed the road up the long slope to the east, he would have seen the Capitol above and ahead of him. The situation of this enigmatic monument, or monumental complex, begins to make sense when we think in terms of approaching it from the outside and from below, as one would approach the Capitol in Rome. The distance of the Capitol from the other monumental centres, marked by the column of Constantine and the Milion, also becomes more explicable when we consider the question of its symbolic function. If it was dedicated, as it should have been, to the three tutelary deities of Rome – Jupiter/Zeus, Mars/Ares, Minerva/Athena – its paganism was effectively neutralised by its liminal, transitional position in the city's monumental and ceremonial space. Unlike the Capitol in Rome, it was not the end of the road; the sanctuary of the Roman gods was but the first step in a hierarchical progression that led, via the Forum dedicated to the solar divinity of the emperor, to the Christian temples of Wisdom and Peace beside the final destination, which was the imperial Palace.

#### **Conclusion**

There are many more imaginative games that one can play with the symbolic topography, and indeed the symbolic chronology, of the foundation of Constantinople. Given the paucity of evidence, the field is wide open to speculation. There is also much more that can be said to amplify and to nuance the reading of the evidence presented here; in particular, patterns of symbolic duplication and symmetry between the 'old town' and the 'new town' – most obviously the existence of two senate houses, or the possible polarity of two Capitols, one Roman and one Christian – deserve to be investigated. But enough has been said to support the following conclusions.

First, the transformation of Byzantion into Constantinople occurred in two distinct phases, corresponding, on the one hand, to the upgrading of the old Greek city into a Roman imperial residence, and, on the other hand, to the addition of a new urban space that stamped Constantinople as a New Rome. Second, the new space was designed to recall the monuments, not to reproduce the topography of Rome, and it was designed, in the first instance, not to be covered and filled from coast to coast, but to be traversed by a long processional way punctuated by nodal points. Third, while Constantinople was ultimately intended to be a ruling city, it was immediately conceived as a beautiful city, a *kallipolis*, and its *kallos* consisted in an abundance of statues and a proliferation of porticoes. Finally, the imposition of Constantinople upon Byzantion did not, initially, create a palimpsest: that came later. What Constantine oversaw was the grafting of an extraordinary imperial *megalopolis* on to the body of an average ancient city.

## **Bibliography**

#### *Primary*


#### *Secondary*


Berger, A., 'Regionen und Strassen im frühen Konstantinopel', *Istanbuler Mitteilungen* 47 (1997): 349–414. Berger, A., 'Streets and public spaces in Constantinople', *Dumbarton Oaks Papers* 54 (2000): 161–172. Chatzilazarou, D., 'Le centre monumental de Constantinople, espace de synthèse des traditions

urbaines gréco-romaines', *Travaux et mémoires* 22/1 (2018): 35–54.


## Chapter 12

## Looking in two directions: Urban (re)building in sixth-century Asia Minor

*Ine Jacobs*

Since the city in late antique Asia Minor was very much a continuation and adaptation of the city of classical antiquity, the metaphor of a palimpsest is rarely applicable on a site-wide scale. Individual buildings may have been dismantled and replaced, but for the most part the built environment that was present in the third century AD remained standing until the end of antiquity and beyond. If we were to compare the late antique city with the biography of a manuscript, it would be more apt to say that marginalia continued to be added in between the lines as well as in the margins. These later additions are not an easy or uniform object of study. They have also long been neglected in favour of the original manuscript text. But that is where the comparison ends. Contrary to manuscripts, where the original content, comprising letters and illustrations, remained meaningful until it was erased, it is much harder to assess the relevance of individual buildings and urban spaces for contemporary society decades or even centuries after they had been created. The continued presence of a building, street or square may have been the result of continuous maintenance as well as thoughtless neglect. The archaeological record, or rather our current understanding of it, very rarely makes it possible to distinguish between these two scenarios. In fact, it is astounding to realise how little we know about the continuous use of large monuments or urban infrastructure after their construction and especially after the start of late antiquity. Stratigraphic evidence occurs only very rarely: in a few cases a stylistic dating of architectural decoration has been possible, at times there may be related epigraphic evidence, and, in some cases, the continued dedications of statuary can indicate the continued importance of certain locations.1

Getting a clear idea of what local inhabitants regarded as important buildings and relevant institutions, and, on a larger scale, what they thought their urban

<sup>1</sup> For a longer review of the evidence, see Jacobs, 'Clinging to tradition'.

surroundings should look like, is possible only on rare occasions. One of the best ways to do so is through examining choices made when rebuilding after a catastrophic event, when large parts of the urban fabric had collapsed or were severely damaged and suddenly cried out for attention. During such organised campaigns of rebuilding urban palimpsests could be created or, alternatively, the local population could prefer to retrace the original text.

In this brief article I will review three such rebuilding campaigns that took place at the end of the fifth and in the first half of the sixth century in the cities of Aphrodisias, Sagalassos and Assos. More precisely, I will focus on rebuilding made necessary after major earthquakes and the degree to which the renovation campaigns looked back to earlier planning principles in the settlement. Although one should always be careful when using earthquakes to date the destruction or renovation of a single monument,2 their occurrence and, especially, what happens in their aftermath is very useful to gain insight into the priorities of a community. Earthquakes affected entire cities. Hence, the ensuing necessity-driven responses make it possible to trace communitywide decisions concerning the urban fabric.

The case-studies of Aphrodisias, Sagalassos and Assos represent cities of diverse standing: Aphrodisias was a medium-sized city but had already in the mid-third century become the provincial capital of Caria and in late antiquity was the seat of the metropolitan bishop of the province.3 Assos and Sagalassos were smaller provincial cities, both with a bishop.4 Whereas Sagalassos was located inland, high up in the Taurus Mountains, Assos had, at least up until the Roman period, been a busy and well-equipped port.5 From the fourth century onwards, and after the ascension of Alexandria Troas as the new capital of the area, its importance may have declined but its port remained functional. In all three case-studies a strong argument can be made for earthquakes occurring in the later fifth or the early sixth century, even if we cannot determine the exact date or epicentre. The event that shook Assos may have been the event that brought havoc to the region of the Hellespont in 484. The main source, a late eighth-century author of chronicles known as the Great Chronographer, claims that the buildings on the island of Tenedos, located only 60 km from Assos, were destroyed as a result of this earthquake.6 Although no reference is made to Assos, the sheer amount of debris and the extent of the changes to the urban fabric in the later fifth and early sixth century makes us believe that the impact here had

<sup>2</sup> Kristensen, '"And Christ-loving Antioch"', for a more detailed critique.

<sup>3</sup> Aphrodisias became the capital of the province of Caria in the middle of the third century, see Roueché, 'Rome, Asia and Aphrodisias'. The city's bishop is attested for the first time in 325, see Honigmann, 'The original list'.

<sup>4</sup> The first time a bishop of Assos turns up in literary sources is at the council of Ephesus in 431, see *Notitiae Episcopatuum* 1.85, Darrouzès, *Notitiae*, 206; Laurent, *Corpus* 1, 193–196. A bishop of Sagalassos attended the Council of Constantinople in 381, see Belke and Mersich, *Phrygien und Pisidien*, 368–369.

<sup>5</sup> Arslan *et al*., 'Der Hafen von Assos'. 6 Ambraseys, *Earthquakes*, 176. On the Great Chronographer, see Nicholson, *The Oxford Dictionary of Late Antiquity*, 678.

also been great. The seismic event that shook Aphrodisias could have been that of 494 mentioned by Marcellinus Comes as being catastrophic for the cities of Hierapolis, Laodikeia and Tripolis, located at a distance of respectively *c*. 43 km, 37 km and 41 km from Aphrodisias.7 A combination of epigraphic and the stratigraphic evidence from the site itself makes it possible to situate a very heavy earthquake between *c*. 485 and 500.8 Around the same time, Sagalassos suffered a similar fate. Again, there are no literary attestations for the events, but widespread repairs and renovations dated independently to the same period as well as the nature of the damage make a strong case for a large-scale natural event being the cause.9

In these three cities, the population was thus forced to consciously engage with past urban settings and to make decisions and choices about what to leave in ruin, what to restore, re-adapt or even recreate, as well as in which sequence interventions would take place. One can imagine a series of formal meetings of the local government, informal discussions among city-dwellers, and vibrant if not heated debates on how to spend money and how to invest time and energy. As we will see, the outcomes of these deliberations could be extremely diverse. Yet, each of these case-studies makes it possible to further nuance and even counter the still prevalent idea that sixth-century city dwellers lived their life indifferently within the slowly deteriorating framework that they inherited from classical antiquity.10

For the sake of brevity and coherence, I will use the urban squares and streets of Aphrodisias, Sagalassos and Assos as a proxy for their entire built environment. The reasons for doing so are threefold. First, they have been well researched in all three cities, making it possible to compare their later history. Second, they are in many ways examples of excessive urban infrastructure that was not necessary for the everyday functioning of the settlement, and yet was more independent from political and religious changes than other traditional forms of architecture such as gymnasia or theatres.11 And third, a major part of classical and late antique city life took place in its streets and squares. People came here to socialise, play games, sell and buy all kinds of goods, watch and participate in all kinds of processions,

<sup>7</sup> Marcellinus Comes, *Annales* anno 494 (ii Asterii et Praesidii). 8 Wilson, 'Water, nymphs', 134; Wilson *et al*., 'Excavations in an urban park', 89–90; Wilson, 'Aphrodisias in the long sixth century', 203; Wilson, 'Earthquakes at Aphrodisias', 479.

<sup>9</sup> Sintubin *et al*., 'Seismic catastrophes', 6–15. The long-supposed date of this earthquake was confirmed in 2012 by soundings underneath the mosaic floor of frigidarium I in the Imperial Baths, see Waelkens, *Sagalassos - Jaarboek 2011–2012*, 104–105. A seismic fault has been identified underneath the city of Sagalassos, see Similox-Tohon *et al.* 'Identification of an active fault', 81, 91.

<sup>10</sup> Depending largely on the background and site experience of the researcher involved, the sixth century has been considered either an extension of classical antiquity, or the prelude to the less prosperous centuries to follow. See for instance the gloomy picture painted by Niewöhner, 'Urbanism', 43–46. For an overview of the state of research, see Jacobs and Elton, *The Long Sixth Century*.

<sup>11</sup> On the fate of traditional entertainment buildings in the sixth century, see amongst others Leyerle, *Theatrical Shows*, 13–74; Saradi, *Byzantine City*, 310–319; Puk, *Römische Spielewesen* for spectacles in general.

hear official pronouncements, see justice done and so on.12 By extrapolating from attitudes adopted towards these elements of urban infrastructure I will argue that these case-studies are representative of two opposite strategies of urbanism present in sixth-century Asia Minor, both of which reflected local organisation, concerns and also constraints. Aphrodisias and Sagalassos, on the one hand, consciously and actively continued the cityscape and its classical antique functionality. Assos, on the other hand, is representative of another type of urban-like settlement that had been appearing both in Asia Minor and in the rest of the Roman empire already from the fourth century onward, and which would become more prevalent as time progressed.

#### **Sixth-century Aphrodisias, Sagalassos and Assos**

In all three cities, the extent and character of the renovations make it possible to assume the presence of a thoroughgoing strategy. At Aphrodisias, comprehensive interventions datable to the later fifth or the early sixth century are attested in public buildings all over the city centre, including in the bouleuterion, the Hadrianic Baths, the North Agora, the Place of Palms, the Tetrapylon Street and the Temple-Church.13 Recent excavations have demonstrated that both in the Place of Palms and in the Tetrapylon Street, the main north–south street of Aphrodisias, the scale of the renovation work is astounding.

The Place of Palms, formerly known as the South Agora of Aphrodisias, was a lush urban park just to the south of the North Agora with a length of 212 m, centred on a pool almost 170 m long and provided with plenty of shade by surrounding stoas and palm trees. It was created at some point in the first century and survived into the seventh century. In the later fifth and/or the early sixth century it underwent a complete makeover.14 Due to the city's rich epigraphic record, the main initiators and additional benefactors of the building works are known. Thus the western stoa of the area was rebuilt by the *clarissimus* Albinus, whose benefaction was commemorated in a series of acclamations eternalised on the stoa's columns and rewarded with a statue.15 A certain Philippos paid for the creation, or roofing, of part of the south

<sup>12</sup> For activities in streets, see Saradi, *Byzantine City*, 266–267 and the very extensive discussion in Lavan, *Public Space*, 150–262. For a summary on activities on agorai, see Lavan, 'Fora and agorai', 206–234;

Lavan, *Public Space*, 308–332, 358–363. 13 Post-earthquake interventions, renovations to and rebuilding of these monuments are all discussed in Wilson, 'Earthquakes at Aphrodisias' and 'Aphrodisias in the long sixth century'. In addition, for the bouleuterion see Hallett and Quatember, 'Three bouleuteria', 355–356; for the Hadrianic Baths, McDavid, 'Renovation of the Hadrianic baths', with suggested changes to chronology in Wilson, 'The Olympian (Hadrianic) baths'; for the North Agora see Ratté and Smith, 'Archaeological research', 720; Wilson *et al*., 'Excavations in an urban park', 89–90; Place of Palms see Wilson and Russell, *Place of Palms*, chap. 4. For a broader overview of building activity throughout the centuries at Aphrodisias, see Dalgıç and Sokolicek, 'Aphrodisias', 269–279, 270 table 23.1. Many of the dates cited in this last

article have been slightly adjusted in the meantime. 14 Wilson *et al*., 'Excavations in an urban park'; Wilson and Russell, *Place of Palms*, chap. 4.

<sup>15</sup> *Ala2004*, 82–83; *IAph2007*, 4.20–4.21.

stoa, which encompassed a whole series of newly carved Ionic capitals.16 In addition, parts of the masonry wall supporting the theatre hill were rebuilt, and columns and cornice blocks of the north stoa were repaired. The overarching responsibility for the renovation project appears to have been shared by two officials, the provincial governor, Doulkitios, and Flavius Ampelios, *pater tes poleos*. Doulkitios rebuilt the east gate to the area.17 Flavius Ampelios is said to have restored 'wonder and beauty to this place of palms',18 which its excavators interpret as confirmation that he paid for a whole set of measures ensuring the further functioning of the pool in the centre and of the larger area.19 They included the repair and re-setting of the stones that constituted the ring drain surrounding the pool, the filling of the ring drain probably to prevent leakage and, most impressively, the raising of the ground level of the entire park with a fill of soil and rubble about 40 cm thick – a measure that also dealt with large quantities of earthquake debris in a convenient manner. The refurbishment project was concluded with the replanting of palm trees, albeit probably in smaller numbers than in the previous phase.

Renovation activities in the Tetrapylon Street were, as far as we can tell, not commemorated with new building inscriptions, but were equally substantial. The Tetrapylon Street was the main street of Aphrodisias, running in a north–south direction and linking some of the city's main monuments and public spaces.20 As in the Place of Palms, the level of the street's pavement was raised considerably in the stretch in between the tetrapylon and the theatre. Whereas the level of the pavement directly in front of the tetrapylon apparently remained constant, some 75 m further south the new pavement was relaid about 1.10 m above the previous level. In front of the propylon to the Sebasteion, the level change was already as high as 1.60 m.21 Considering that all water was supposed to flow through the Place of Palms before continuing its course to the river Morsynus 2 km to the south-west of the site, a rise in the ground level here was bound to necessitate interventions in adjoining areas. In addition, the colonnades alongside the street were constructed only in this period. The west side of the road was apparently still flanked by a row of columns, whereas the lower storey of the east colonnade consisted of a mixture of piers and columns composed of various building elements, topped by brick arcades and carrying a second storey that belonged to a series of luxurious houses at the back. Only in the very heart of the city centre, where the street passed the east gate of the Place of Palms, was the colonnade uniform, composed of Ionic columns of which the capitals and bases were certainly newly carved in late antiquity (Figure 12.1). In this section, the street's

<sup>16</sup> Inscription testifying to the rebuilding of the south stoa: *Ala2004*, 66; *IAph2007*, 4.19. The newly carved building decoration is discussed in Kidd, '"The Ionic capitals"'. 17 *Ala2004*, 39–40; *IAph2007*, 4.202.ii-iii.

<sup>18</sup> *Ala2004*, 38; *IAph2007*, 4.202.i. 19 Wilson and Russell, *Place of Palms*, chap. 4. 20 Sokolicek, 'Excavations', 60. The results of the street excavation will be published as Jacobs, *Tetrapylon Street*. 21 As shown by the 2018 and 2019 excavations.

*Fig. 12.1: Late antique ionic capital from the Tetrapylon Street (© Aphrodisias Excavations).*

continued importance as a ceremonial setting after the renovations was made clear in a series of acclamations applied (and reapplied) to plaster on columns, one of which praises the Christian God, another the emperor, 'lord of the inhabited world'.22

Excavation of the main drain underneath the new street pavement revealed that the wall sections added on top of the early imperial street drain included a sizeable number of statuary fragments (Figure 12.2). They indicate that the statuary population was being thinned out, a phenomenon to

which the later fifth-century earthquake no doubt contributed. However, certainly at Aphrodisias the rich statuary finds retrieved in the site's excavations demonstrate that enough statues remained in place and new statues were still being added, so that the sixth-century situation could compete with previous centuries. Especially at the Place of Palms, marble statuary was found (in a variety of contexts) in impressive quantities decorating the façade of the park's east gate and west stoa, as well as the pool edge, into the seventh century.23

Both the street and urban park were therefore reinstated with an impressive investment of time and energy. The sixth-century spaces were not identical to their predecessors, but they integrated as much of their predecessors as possible, both physically and in essence. The desire to reinstate urban life as it was before the city was devastated is undeniable.

A similar sentiment can be clearly recognised in the rebuilding operations at Sagalassos. Renovation and rebuilding was initiated all over the city centre.24 In addition to several churches, the city's bath complex, monumental fountains and street network were restored in a form that was reminiscent of that of previous centuries. Deserted stacks of building materials ready for re-use suggest that renovation of the Odeion was started, but never finished (Figure 12.3).25 The east and west stoas of both of the city's agoras were re-erected upon the old stylobates using a variety of columns, bases and capitals.26 The colonnades of the main east–west colonnaded street were repaired and a new pavement was installed.27 The most extensive building project

<sup>22</sup> Tr. Angelos Chaniotis. 23 For an overview of the statuary still present in the Place of Palms, see Thomas, 'Sculptural life'. Smith, 'Three statues', 293–300 for a preliminary and partial report on statuary found in the Tetrapylon Street. An exact composition of what statues were present where is difficult because of more numerous and

intense post-antique interventions in the area of the street. 24 These interventions are discussed in more detail in Jacobs, 'Clinging to tradition'.

<sup>25</sup> Jacobs, *Aesthetic Maintenance*, 608. 26 Jacobs, *Aesthetic Maintenance*, 178–179.

<sup>27</sup> Martens, 'Late antique urban streets', 348.

*Fig. 12.2: Sixth-century street drain constructed with mortared rubble and statuary fragments (© Aphrodisias Excavations).*

taking place in the second quarter of the sixth century was the renovation of the 10 m wide north–south colonnaded street, the main traffic axis.28 Its colonnades were reconstructed with a mixture of re-used Corinthian columns placed on a variety of bases as well as brick-and-tuff pillars, all carrying arches.

As at Aphrodisias, statuary remained an integral part of all major renovation projects. However, since Sagalassos did not have a local statuary workshop and its supply had dried out already at some point in the fourth century, older statuary and reliefs were relocated from other sites in the city to be integrated into the more eye-catching sixth-century renovation works. The street section underneath the Lower Agora became the location of an entirely new statuary display comprising at least eight small-scale statues, mounted on top of statue brackets integrated into the brick-and-tuff piers alongside the street. They included an Apollo, a Hygieia or Tyche, a Hygieia with Hypnos, a third Hygieia, an Aphrodite, the central figure from a group of The Three Graces and two smaller statuettes that have disappeared in subsequent centuries.29 Likewise, the town's monumental fountains on the north sides

<sup>28</sup> The sixth-century renovation phase of the street is discussed in detail in Jacobs and Waelkens, 'Five centuries'.

<sup>29</sup> This statuary collection, its origins, reconstruction and meaning is discussed in detail in Jacobs and Stirling, 'Re-using the gods'.

*Fig. 12.3: Abandoned stack of column drums discovered in the Odeion at Sagalassos (©Sagalassos Archaeological Research Project).*

of both the Upper and the Lower Agora were redecorated with new assemblages of statues: a combination of statues surviving from the original display and newcomers from elsewhere, most of which were retrieved in the excavations.30

By contrast, the building works taking place in Assos were of an entirely different nature. Assos is mainly known among archaeologists as a model Archaic to Hellenistic city. Its Archaic Doric temple of Athena on top of the acropolis is a mainstay of books of Greek architecture. Its walls feature in overviews of classical-Hellenistic fortification works;31 the agora

is a textbook example of Pergamene planning whereby the north and south stoas direct attention to buildings on the short side of the square, in this case a temple and bouleuterion. Apparently, not much changed in the city centre during the first centuries of Roman rule and the Hellenistic framework was largely maintained.32 Yet on top of the Hellenistic-Roman city, at least on top of its south-western and southern quarters, lies a late fifth-/early sixth-century settlement, which was surveyed and examined only very recently.33

This settlement, which developed after the earthquake of the late fifth century, was entirely different in form and organisation from what had been there before. None of the old public buildings retained their function in the new early Byzantine configuration, even though remnants of their architecture remained present. Changes are most conspicuous when we review briefly what happened to the old agora area. In contrast to the situations encountered at Aphrodisias and Sagalassos, the agora of Assos practically ceased to exist after the rebuilding. Even though the American excavations of the late nineteenth century cleared the centre of the square more than a hundred years ago, their plans still show some of the structures that took over the area. They are only indicated as 'late' and not interpreted.34 The more recent survey of the town has however clarified that the north stoa and the area to the north of it were redeveloped, with new streets leading through and up from the area of the agora.35 Considering that the temple to the west of the agora was also redeveloped

<sup>30</sup> For an overview of the sixth-century building operations comprising re-used statuary, see Jacobs, 'Pagan-mythological statuary'.

<sup>31</sup> For the fortifications, see Türk, 'Befestigungsanlagen von Assos'.

<sup>32</sup> Böhlendorf-Arslan, 'The glorious sixth century', 224.

<sup>33</sup> Mohr and Rheidt, 'Der Assossurvey 2010–2012', 137–151. 34 Clarke *et al*., *Investigations at Assos*, 21.

<sup>35</sup> Böhlendorf-Arslan, 'Assos', 220; 'The glorious sixth century', 228.

as a residence, complete with reception or dining hall,36 it is logical to assume that the agora was converted into a domestic city quarter. The bouleuterion on the short east end of the agora apparently was restored, but only to be reused as storage space, as indicated by finds of animal bones and water jugs.<sup>37</sup>

The road network of the early Byzantine city was mapped by connecting the locations of still-standing doorways of houses. The streets that arose on top of the rubble frequently departed from the old street network, with houses being built with quite diverse orientations on top of the old pavement slabs and passages, and became narrower than their precursors (Figure 12.4).38 A main street cannot be recognised and street colonnades could have no reason to exist in such an organisation. The Assos archaeological team has suggested that the debris on top of the old street network was so substantial that the streets were cut off at several points, having been too damaged to be used further without substantial interventions.39 Yet, in theory, clearance of the streets and evacuation of the debris should have been possible. Or, if we compare the situation encountered at Aphrodisias, a continuation of the previous organisation and planning at a higher level should have been an option as well. Although a more extensive inquiry is certainly needed, it does seem that the contemporary local government of Assos decided against a return to the old situation and instead opted to take a new direction that was better adapted to the local topography.40

Finally, in contrast to both Aphrodisias and Sagalassos, there are no indications that statuary was in any way used in the early Byzantine phase of Assos. A small number of statue bases were found during excavation, including one for a statue of Constantius at the entrance to the agora and bases for Germanicus and his family in the bouleuterion.41 Whereas the first may at least in theory still have carried an imperial statue in the sixth-century phase (if it survived the seismic event), the other two are more likely to have been used as building material in the sixth-century rebuilding of the structure.42 It is therefore quite clear that statuary did not play the same active part in sixth-century Assos as it did at Sagalassos and certainly not as at Aphrodisias.

#### **Discussion**

The late fifth- and sixth-century officials and populations of Aphrodisias and Sagalassos obviously went through a lot of trouble in recreating prestigious urban spaces, using

<sup>36</sup> Böhlendorf-Arslan, 'The glorious sixth century', 227–228.

<sup>37</sup> Clarke *et al*., *Investigations at Assos*, 21; Böhlendorf-Arslan, 'Nothing to remember?', 23–24 for the rebuilding; Böhlendorf-Arslan, 'The glorious sixth century', 228.

<sup>38</sup> Böhlendorf-Arslan, 'The glorious sixth century', 231–234. This course of action was also followed in later centuries in, for instance, Hierapolis, where such houses were found to the west of the Frontinus Gate. This makes sense if the buildings behind the street are too damaged to be repaired.

<sup>39</sup> Böhlendorf-Arslan, 'The glorious sixth century', 231.

<sup>40</sup> Böhlendorf-Arslan, 'The glorious sixth century', 230.

<sup>41</sup> Böhlendorf-Arslan, 'The glorious sixth century', 224 with further references. 42 Özhan, 'Two inscribed pedestals'.

*Fig. 12.4: Roman street built over in the late fifth, early sixth century at Assos (© Assos excavations archive).*

inventive solutions when necessary and including statuary to finish the total picture. At both sites, the inhabitants of the late antique cities were treading in the footsteps of their Roman predecessors by again investing in symbols of city status that were inherited from preceding centuries. Moreover, both Sagalassos and Aphrodisias rebuilt not just one but both of their urban open areas, even though other cities may have kept only one functional.43 The continuity of the Place of Palms at Aphrodisias is especially remarkable in this respect. This urban park was no doubt very pleasant and served, according to the inscription dedicated to Doulkitios, as the location where the Maiouma festival was celebrated in Aphrodisias.44 Yet it could be argued that a grand park centred on a vast pool and lined with marble columns was not essential for the daily running of the city. The Place of Palms was reinstated as an enjoyable assembly place for citizens of the city to sit and relax on the many benches along the porticoes or at the edge of the water as their parents and grandparents had done before them.

<sup>43</sup> Lavan, 'Fora and agorai', 235–236. In his recent monograph, Lavan no longer thinks this to be the case for Asia Minor, see Lavan, *Public Space*, 357. For further examples of agorae created, renovated or repaired in the sixth century, see Lavan, *Public Space*, 341–346. Note that Lavan, 'Fora and agorai' still assumes a decline in importance for the sixth century.

<sup>44</sup> *Ala2004*, 40; *IAph 2007*, 4.202.iii. For the celebration of the Maiouma, which involved nocturnal festivities and aquatic shows, see Wilson, 'Water, nymphs', 132–135; Wilson and Russell, *Place of Palms*, chap. 4.

Like the renovation and revitalisation of the Place of Palms, the maintenance of the Tetrapylon Street was, in the strict sense, unnecessary. When the city of Aphrodisias was surrounded by a city wall around 360, the circuit blocked the continuation of the Tetrapylon Street, thereby turning it from a major connection to the hinterland into an inward facing boulevard.45 The lack of wheel ruts confirms that the late antique street was never used by wheeled traffic.46 Although with a road surface of just under 8 m wide it was much more modest than some magnificent colonnaded streets in the Roman East, it was still significantly larger than the standard 3 to 4 m wide street, and very much in line with other colonnaded streets in western Asia Minor.47 Likewise, at Sagalassos, although the location of the main access route to the centre was solidly fixed by the local topography, there was no need to rebuild it on such a monumental scale, including colonnades. The colonnaded street below the Lower Agora was again never used by wheeled traffic. There were no traces of wheel ruts and, moreover, the road incorporated staircases at multiple points.48 Considering that both the Tetrapylon Street at Aphrodisias and the main colonnaded street at Sagalassos were only used by pedestrians and pack animals, the borders of their pavements could have easily been used for the building of shops and workshops profiting from the many passers-by, or even houses, as happened in the Roman streets of Assos in early Byzantine times.

The contrast between the reconstruction operations in the late fifth and early sixth century makes one wonder what the influencing factors in the decision-making process were. Local topography may have been highly influential. Assos is built on the slopes of a hill, whereas Aphrodisias is located in a valley with a relatively level floor, which makes large open spaces and routes easier to establish and maintain, albeit on a higher level. Alternatively, it may very well be that the extent of the earthquake damage and hence the quantities of debris were much more substantial at Assos. Whereas the debris could be either cleared as happened at Sagalassos or reused to re-establish old patterns at a higher level as was the case at Aphrodisias, the amounts of debris may simply have been too much at Assos. We may also wonder how decisive the exact composition of the local government was for the form of the rebuilding. In a provincial city like Aphrodisias, where there was a larger concentration of wealth, officials and magistrates, and thus a higher number of processions and other official and highly ceremonial events, a greater attachment to traditional authoritative architecture may indeed be easier to explain. At Sagalassos, where at least the spirit, if not quite the extent, of the renovations and interventions was

<sup>45</sup> Sokolicek, 'Excavations', 60; Dalgıç and Sokolicek, 'Aphrodisias', 271.

<sup>46</sup> Wheel ruts have not been identified anywhere in Aphrodisias, which suggest that the city government had strict rules regulating wheeled traffic within the city centre.

<sup>47</sup> For standard street widths in the Roman period, see Adam, *Roman Building*, 280. Jacobs, *Aesthetic Maintenance*, 127–129 and table 2.5 for comparisons with other main streets in cities of western Asia Minor; Lavan, *Public Space*, 41–43 for comparisons with late antique streets across the Empire.

<sup>48</sup> Overall, no continuous wheel ruts have been discovered in the pavements at Sagalassos so far. Although topographical factors obviously hindered wheeled traffic, it seems highly unlikely that the entire city was off-limits for carts. For a wider discussion, see Martens, 'Late antique urban streets', 340–341.

comparable to Aphrodisias, the enduring presence of a traditional government can be hypothesised, even if it is no longer explicitly mentioned in inscriptions or literary sources. The traditional bouleuterion had already gone out of use before the end of the fourth century, but civic meetings could have continued inside the local odeion, which was being used until the end of the fifth century and for which renovations had been planned. Alternatively, a mosaic inscription in the sixth-century floor of a former frigidarium in the Imperial Baths of Sagalassos mentions that it was a *demosion* or public meeting place. Before the earthquake, a wooden, amphitheatreshaped auditorium had even been present in the eastern arm of the same large hall.49 Large halls inside bath complexes were also used as venues for civic meetings in the sixth century elsewhere, as is for instance suggested by the inscriptions referring to the *gerousia* and *boule* found on the opus sectile floor of the so-called Marble Court of the baths of Sardis.50 By contrast, the only authority that can be pinpointed with certainty in the sixth century at Assos is the local bishop who took up residence in a new complex established on a terrace to the south-west of the old agora.51 Finally, in addition to the presence or absence of traditional elements of government in a settlement, the direction of renovations may have been impacted by very specific and personal factors, whereby certain individuals wielding power at the moment decisions were being made drove renovation in one direction or the other.

On its own terms, nothing about the new organisation of Assos is particularly exceptional for late antiquity. Extensions of classical cities such as Sardis, Skythopolis or Jerusalem no longer incorporated regularly planned quarters with monumental streets and squares.52 Moreover, the focus of modern research on a few, exceptional, newly constructed sites such as Justiniana Prima, Dara-Anastasiopolis, Resafa-Sergiopolis and Zenobia, which indeed had colonnaded streets and open squares, hides the fact that the vast majority of newly founded settlements rarely had regular street plans, main streets with colonnades, monumental urban squares or other large open spaces.53 As early as the Tetrarchic period, the settlement of Döşeme Boğazı (Pamphylia) had an irregular layout, no recognisable main street and no open square. The settlement has been tentatively identified with the town of Maximianopolis, built in honour of the emperor Maximian, but imperial interest was limited to the site's role in the provisioning of the army.54 Hence, the only public architecture is an enormous horreum. By the sixth century, such settlements were appearing everywhere in the eastern Mediterranean. The sixth-century site of Mokisos, modern-day Viranşehir in Cappadocia, with more than a thousand houses spread out over about 50 hectares,

<sup>49</sup> Waelkens, *Sagalassos - Jaarboek 2013*, 25. 50 Yegül, *The Bath-Gymnasium Complex at Sardis*, 49, 51. Yegül, *Baths*, 313, 329 for similar functions in baths elsewhere. 51 Böhlendorf-Arslan, 'The glorious sixth century', 228–230.

<sup>52</sup> Jacobs, *Aesthetic Maintenance*, 140.

<sup>53</sup> Ćurčić, *Architecture in the Balkans*, 182–183, 209–216 with examples from the Balkans; Rizos, 'New cities', for a broad overview and further references. 54 Mitchell, 'Archaeology in Asia Minor'; Rizos, 'New cities', 36.

is impressively large.55 Yet, the only non-residential buildings recognisable today are a small fort and several churches, including a centrally located cathedral church and associated episcopal residence that housed the archbishop of the church province *Cappadocia III*. Even Procopius' description of this 'metropolis' only features 'churches, hospices, and public baths that are the mark of a prosperous city', but no characteristically classical elements.56 The form of the settlement of Assos in the sixth century is therefore remarkable foremost because it diverged so dramatically from what had been there before. In this particular case, the page was scraped clean and it was decided to write down a new text, (quasi-)unconnected to the one that had been there before.

In summary, impressive quantities of labour and equipment could still be rallied for civic projects into the early sixth century. Yet, instead of one unified idea of what should constitute an urban community, two diverging strategies can be recognised. One is aimed at establishing a state of affairs that is a direct descendent of the classical past. The other takes a new direction, abandoning older buildings and spaces altogether. The second strategy, represented in this article by the site of Assos, comes closest to creating an urban palimpsest, with the underlying layer of buildings and streets largely erased and no longer relevant, and its new layer only occasionally and only out of practical considerations engaging with what had been there before. However, it should be noted that this erasure was forced upon the population of Assos by a natural disaster, without which such drastic changes would probably not have come about. The first strategy does not result in a palimpsest. Instead, both main streets and public squares, the most visible parts of the cityscape, were recreated. This may have been partially the result of less extensive damage; scraping the page clean would have been much more labour-intensive than reviving faded but still readable text passages. However, in both Sagalassos and Aphrodisias the nature and the extent of the operations suggests they were fuelled by a genuine desire to create a urban stage that could be put to use in much the same way as it had been in preceding centuries.

The rebuilding of both these cities in the late fifth and early sixth century thus closely adhered to planning principles and architectural forms of the classical age, even though the materials and execution had changed somewhat and the late antique initiators and builders may not have had the funds to carry out reconstruction everywhere. There were certainly modifications compared to previous centuries: building materials were more varied and often recut from older elements, the execution of individual building elements was cruder than before, new elements were integrated next to older ones taken from elsewhere, and the composition of columnar rows was nowhere near as uniform as it had been under the High Empire. But how important is all of this in the face of the continuity of the *idea* of the classical city? Do we focus on piers being added in between marble columns, or on the continuity of shaded walkways? Does it matter that the details of a sixth-century capital were

<sup>55</sup> For a detailed description of the site, see Berger, 'Viranşehir' and 'Mokisos'.

<sup>56</sup> Procopius, *De Aedificiis* 5.4.15–18.

cruder than those of a second-century one or that its ornament was not identical to that of its neighbour? Or is it more important that we can still immediately identify them as Ionic?

The different building materials and skills of execution of later architectural creations have been noticed by modern researchers – often trained as classical archaeologists – and labelled as bad quality copies of older prototypes.57 However, it is highly doubtful that contemporaries judged their surroundings in such a comparative way. Through a re-establishment of the overall form and function of what was there before, they were enabled to resume all their activities, from spontaneous gatherings to organised political events, in the same location and in the same way as before. If we return to the comparison with a manuscript one last time: the script may have changed, the ink may have been different and individual words may even have been replaced, but the message was carefully preserved.

## **Bibliography**

#### *Primary sources*


#### *Secondary*

Adam, J.-P., *Roman Building. Materials and Techniques* (London and New York, 1994).


Belke, K. and N. Mersich, *Phrygien und Pisidien*, *Tabula Imperii Byzantini* 7 (Vienna, 1990).


<sup>57</sup> The usage of terms like 'copy', 'quotation', 'imitation' and 'decline' is increasingly coming under attack as attempts to imposing modern value judgements on historic material and ancient aesthetic perceptions. See Böhme *et al*., *Transformation*.


Yegül, F. K., *Baths and Bathing in Classical Antiquity* (Cambridge, 1992).

## Chapter 13

## Byzantine urban imagination: Idealisation and political thinking (eighth to fifteenth centuries)

## *Helen Saradi*

The Byzantine city was the continuation of the Greco-Roman city. Byzantium inherited from Rome an empire made up of cities, the number and size of which were unprecedented. Cities were adorned with splendid monuments displaying the ecumenical civilisation of the Roman empire and the power of its emperors, manifesting the people's religious beliefs, serving the urban community and defining its civic identity.

Profound changes in urban life and the urban landscape were evident everywhere by the end of the early Byzantine period.<sup>1</sup> The Greco-Roman form of the city with magnificent public buildings adorned with spectacular sculpture, pagan temples, large baths and gymnasia, agoras, colonnaded avenues and aristocratic luxurious mansions of the *domus* type, was gradually transformed by reason of cultural and religious shifts, administrative changes and invasions. In the sixth century in the Balkans and the seventh century in Anatolia, invasions ushered in the final stage of the urban development that had begun in the fourth century. The new type of city which emerged was the *kastron*. Its form is described in a sixth-century anonymous military treatise (*Strategikon*). The author rejects the Greco-Roman type of urban landscape and the aesthetic considerations formerly adopted in the construction of new cities, stating that people used to trust that the prosperity of their time would continue and therefore when they planned to build large cities, cared for these cities' pleasing appearance rather than for their security ('οὐ μᾶλλον τῆς ἀσφαλείας ἢ τῆς εὐπρεπείας ἐφρόντιζον'). Thus, he continues, they built cities on level ground and beautified them with gardens, parks and lawns. By contrast, given the uncertainty of his time, the author considers security more important than urban beauty ('τὴν

<sup>1</sup> The bibliography on this is vast. For a general picture see Liebeschuetz, *Roman City*; Saradi, *Byzantine City*.

ἀσφάλειαν μᾶλλον τῆς εὐπρεπείας προκρίνοντες').<sup>2</sup> In the chapters which follow the author describes the military character of the Byzantine city in detail. The new city, smaller in size, stands on naturally defensible sites – a hilltop, a promontory or an island – and is protected by one or more circuit walls. Thus, in the context of the military priorities of the time the Greco-Roman urban form was consciously left behind by the state administration. The old city did not serve the new needs of the population created by enemy invasions and changes in the empire's system of defence. Several cities were relocated to fortified sites, the size and population of other previously large cities shrank, while demonumentalisation and ruralisation also mark middle Byzantine cities.<sup>3</sup> As in the *Strategikon*, this new form of the city is reflected in the terminology used in later texts. For example, when Theophanes follows earlier sources in his *Chronographia*, which stretches from Creation until 843, he uses the term *polis*, but when he refers to events after the reign of Heraclius (610–641), he uses the term *kastron*. 4 He only continues to describe Constantinople and Thessalonike with the term 'city'.

In this chapter I will explore what difference this physical urban change made to how the city was imagined and described in Byzantine rhetorical laudatory accounts of the city, local legendary compilations and letters. These literary genres provide the most characteristic trends in the description of the Byzantine city. I will also outline the complex theme of urban antiquity, whether located in particular urban monuments or in discussions of the cities' glorious histories, as antiquity played a prominent role in creating and maintaining a city's identity. Aesthetics, legendary or historical associations, and realism or idealisation are recurring themes in descriptions of Byzantine cities. I will investigate how these ideas were dictated by literary models, historical development and the authors' political commitments.

## **The legendary imperial city in the middle Byzantine period: emperors and monuments**

Throughout the early Byzantine period cities were praised in rhetorical works, produced on the occasion of public events, which supplied the urban residents elaborate images of cities' architectural beauty and splendour and praised their cultural achievements.<sup>5</sup> From the seventh century, despite the changes to the city, authors continued to apply the vocabulary of aesthetics, the rhetorical *topos* of *kallos*. 6 Thus cities/*kastra* were said to possess beautiful walls and beautiful churches.<sup>7</sup> Both of

<sup>2</sup> *Strategikon*, chap. 11, 25–32 (32).

<sup>3</sup> On the structure of the middle Byzantine city see Bouras, 'Byzantine city' and other articles in Laiou, *Economic History*; Haldon, *Byzantium*, 92–124; Niewöhner, *Byzantine Anatolia*.

<sup>4</sup> Kazhdan, '*Polis* and *kastron*', 345–350.

<sup>5</sup> See Menander, 'On how to praise a city'*,* 32–74. For the city praise in ancient rhetoric see Pernot, *La rhétorique de l'éloge*, vol. 1, esp. 79–82, 178–216; Bouffartigue, 'L'éloge'.

<sup>6</sup> For the *kallos* in the ancient sources see Maupai, *Die Macht der Schönheit*, 203–239.

<sup>7</sup> See Saradi, '*Kallos*'; Maguire, 'The beauty of castles'; Bakirtzis, 'Τα τείχη'.

these were of the greatest importance to the Byzantine city and defined its military and religious character. Mention is also frequently made of the antiquity and renown of the cities,<sup>8</sup> and of the safety of their location. From the seventh century, however, laudatory descriptions (*enkomia* and *ekphraseis*) of cities in the form of orations ceased to be produced.9 Cities had lost their civic independence and the capital Constantinople was promoted by authors of the imperial entourage as the only city deserving of praise. Because of this diminution of literary interest in provincial cities, their topography in this period can be gleaned only from scattered references. The literary image of the city is thus clearly conditioned by the historical reality of the empire.

The absence of urban descriptions until the thirteenth century10 is partially filled by patriographic works on the monuments of Constantinople. These are the *Parastaseis syntomoi chronikai* (*Brief Historical Notes*), and the *Patria of Constantinople*, both deriving from the late antique genre of *patria*. The *Parastaseis*, dated to between the late eighth and the early ninth century, relate stories about the monuments of Constantinople, and of ancient and Byzantine statues in particular. Most of these stories relate to Constantinople's history, emperors and monuments,<sup>11</sup> but are confused in their identifications of statues and builders and, in general, appear to be absurd and fictitious.

The beginning of the *Parastaseis* exemplifies its general character. It starts with a chapter on the Church of Saint Mokios, whose patronal festival, 11 May, coincided with the day on which Constantinople was dedicated.12 We learn that the church was built by Constantine I (324–337) on the site of a temple of Zeus and reusing its stones, that it collapsed in the reign of Constantius (337–361), was given by Theodosius I (379–395) to the Arians and, after a destructive earthquake, was rebuilt by Justinian. The second chapter relates that the Church of St Agathonikos was built by Anastasius (491–518) and rebuilt by Justinian, noting that seven Patriarchs held office there and Emperors were crowned there, too. The third chapter deals with the city walls,13 repaired by Tiberius Apsimar (698–705), with the western walls further restored by Leo III (717–

<sup>8</sup> Menander, 50.2: 'τὸ πρεσβύτατον τιμιώτατον εἶναι'; Heller, 'Ἀρχαιότης et εὐγένεια'.

<sup>9</sup> On the *ekphrasis* see in general Maguire, *Art and Eloquence*, 22–52; Nelson, 'To say and to see'; Webb, 'Ekphrasis'; Webb, *Ekphrasis*; Metse and Agapitos, *Εικών και Λόγος*; Vavřínek *et al*., *Ekphrasis*. On the *ekphrasis* of cities see Fenster, *Laudes*; Classen, *Stadt*; Hunger, 'Laudes'; Konstantakopoulou, *Bυζαντινή Θεσσαλονίκη*; Saradi, '*Kallos*'; Saradi, 'Η *Έκφρασις* της Τραπεζούντας'; Saradi, 'The antiquities'; Saradi, 'The monuments'; Saradi, 'Οι ταυτότητες της Κωνσταντινούπολης'; Rhoby, 'Stadtlob'; Rhoby, *Reminiszenzen*; Rhoby, 'City *Encomia*'.

<sup>10</sup> The description of Thessaloniki by Ioannes Kameniates lamenting the city's destruction by the Arabs on 31 July 904 is believed to have been written or rewritten in the fifteenth century on the basis of a tenth-century story, because the rhetoric and urban images resemble those of the late period: Kazhdan, 'Kaminiates'; Christides, 'Caminiates'; Konstantakopoulou, *Bυζαντινή Θεσσαλονίκη*, 43–94, especially pp. 84ff.

<sup>11</sup> Chatterjee, 'Viewing the unknown'.

<sup>12</sup> *Parastaseis*, 56–59 and notes 167–169.

<sup>13</sup> *Parastaseis*, 58–59 and notes 169–171.

741). We are told that on this occasion a procession was held whose members chanted forty times 'Kyrie eleison' ('Lord, have mercy') and the Emperor was applauded by the faction of the Greens as having surpassed Constantine the Great. In these first three chapters, Constantinople is introduced as the Christian city that has replaced the pagan Byzantium. It is the seat of emperors and patriarchs. Its history is woven around churches. It is protected by walls restored by emperors. Public ceremonies marked by Christian and imperial features are part of urban life.

In addition to churches, the Constantinople of the *Parastaseis* was still full of pagan statues, the demonic power of which only Christian intervention could neutralise. This is a central theme of the text, introduced in chapter four with the story of the statue of Fidalia, wife of Byzas, the legendary founder of ancient Byzantium. When the statue was removed, the place trembled violently for a considerable time. The emperor sent a procession to the spot and the shaking ceased with the prayers of St Sabbas (439–532).<sup>14</sup> Crosses were carved on ancient statues to Christianise them, as was the case with the four Gorgons which were removed from the Temple of Artemis at Ephesus and placed over the vestibule of the Chalke, the main entrance to the palace.<sup>15</sup> Most of the statues were associated with emperors: they set them up or destroyed them and their destinies were bound up with secret messages conveyed by the statues, or entire areas where particular statues were set up were linked in some way with imperial intervention.

The foundation of Constantinople and related monuments are mentioned in several chapters (5, 38, 56, 68a), which combine various traditions. In chapter 56, the object of interest is the statue standing on a pillar in the Forum, the famous porphyry column of Constantine.<sup>16</sup> Emphasis is placed on the ceremonies that inaugurated Constantinople's foundation. The dignitaries processed, dressed in white, holding candles and carrying the statue on a carriage, all chanting the 'Kyrie eleison' a hundred times. Talismans were placed on top of the pillar where the statue stood. The description ended with the naming of the city and a prayer for eternal life.17 The festivities lasted forty days. On the next day, Constantinople's birthday was celebrated in the Hippodrome. We are told that Constantine instituted 'these celebrations as an eternal memorial'.<sup>18</sup> In fact, the *Parastaseis* wrongly combines two separate events held in separate years: the celebration of the foundation of Constantinople in 324, and the inauguration of the city on 11 May 330.19

It has been suggested that the purpose of the *Parastaseis* was to glorify Constantinople but denigrate its emperors,<sup>20</sup> or that it was a parody of the Iconoclast

<sup>14</sup> *Parastaseis*, 58–61 and note 171.

<sup>15</sup> *Parastaseis*, 158–159 and note 271.

<sup>16</sup> *Parastaseis*, 130–133 and notes 242–245. On the porphyry column see Ousterhout, 'The life and afterlife of Constantine's column'.

<sup>17</sup> *Parastaseis*, 132.3–4, 133.

<sup>18</sup> *Parastaseis*, 132.9–10, 133.

<sup>19</sup> Dagron, *Naissance d'une capitale*, 32–42.

<sup>20</sup> Dagron, *Constantinople imaginaire*, 315ff.

Controversy, which centred on the power of religious images.<sup>21</sup> Certainly the text conveyed a strong political message.

In the *Parastaseis* the antiquity of some monuments is appreciated, as antiquity was usually revered in Byzantium. That there is no concern about the statues' aesthetic value can be explained by the literary norms of the *Patria*, on which the *Parastaseis* was based. This genre connected monuments with the history and destiny of cities, leaving their artistic value aside. Similarly, the author of the *Parastaseis* is interested only in the secret power of statues and in their symbolism. The accounts of Constantinople's monuments reveal various layers of the city's past which form the present and predict the future of the Byzantine capital and of the Empire. Past, present and future are interconnected through prophetic stories of monuments related with emperors. In this text the monuments render urban space powerfully symbolic. Thus, through monuments of the past, the imaginary Constantinople of the *Parastaseis* is shot through with historical and political symbolism.

Other writings attest to the interest in the monuments of Constantinople as wonders. In the early tenth century, Constantine of Rhodes (870 or 880–931) praises the seven wonders of Constantinople in a poetic narrative followed by a description of the Church of the Holy Apostles. The monuments were all located on the triumphal route from the Golden Gate through three imperial Fora to Hagia Sophia, thus calling attention to the imperial rule in Constantinople. Five of these monuments were columns, four of which were associated with emperors, symbolising the imperial power defending the city: the column of Justinian supporting his equestrian statue, the porphyry column of Constantine with its venerated relics, the column of the Forum Tauri supporting the statue of Theodosius, the column of Arcadius and the column with the cross adding divine to imperial protection. Two buildings are mentioned. One is the Senate House in the Forum of Constantine, which was believed to have been built by Constantine I and re-founded by Justinian. It was adorned with statues, including the beautiful statue of Athena from Lindos, and the doors taken from the Temple of Artemis at Ephesus, depicting the awe-inspiring combat between giants and gods, though Constantine condemns these sculptures as pagan. The second monument is the Anemodoulion, a bronze weathervane decorating a tetrapylon, erected by Theodosius I and adorned with sculptures of animals and plants, 'sweetly laughing' naked Erotes and personifications of the winds.22 It has been argued that this monument may have been remodelled with spolia on the orders of the emperor Leo III, to commemorate the ending of the 717–718 Arab siege of Constantinople.<sup>23</sup> By describing these monuments, Constantine of Rhodes produced an imperial image of Constantinople as majestic and powerful,<sup>24</sup> connecting this vision with the famous Church of the Holy Apostles, an annex of which served as an imperial mausoleum. In

<sup>21</sup> Kazhdan, *Literature*, 311–313; Saradi, 'The antiquities', 102.

<sup>22</sup> Constantine of Rhodes, *On Constantinople and the Church of the Holy Apostles*, 18–84.

<sup>23</sup> Anderson, 'Anemodoulion'.

<sup>24</sup> Anderson, 'Anemodoulion', 171–180.

this text again, the description of magnificent old monuments, both Byzantine and ancient Greek, conveys a political message.

Around 900, Leo Choirosphaktes wrote a poem on the bath built by the emperor Leo VI (886–912), decorated with mosaics, statues, relief sculptures and water spouts in the form of animals.<sup>25</sup> He praises the splendour of the structure, which probably utilised much antique spolia. Again the poem's dominant idea is the connection of the monument with the emperor.

In the late tenth century, when Byzantine cities started to recover economically and Constantinople reached its apogee, a more systematic work on the *Patria of Constantinople* was written. It joined the sixth-century *Patria of Constantinople* by Hesychius of Miletus, the *Parastaseis*, a narrative (*Diegesis*) on the construction of the Church of Hagia Sophia and various other stories about Constantinople's monuments, thus producing an expanded medieval version of the late antique *Patria*. Despite its historical inaccuracies and the fictional character of many of the stories, the *Patria of Constantinople* is important in that it reveals how medieval Byzantines perceived their city.<sup>26</sup> Constantinople unfolds in layers of its past, present and future through its monuments. Monuments and statues are mentioned in connection with real or fantastic stories regarding their foundation, alteration or destruction. Ancient and Byzantine monuments form Constantinople's urban landscape, which is often romanticised and fictional. Churches are also mentioned. The two principal churches of Constantinople, Hagia Sophia and the Holy Apostles, form the final part of the *Patria*. Yet churches are not the principal theme of the work. The *Patria* expresses the amazement and mystification that the Byzantines would have felt before the city's statues and monuments. It also stresses Constantinople's central position in the empire. Constantinople was the city *par excellence*, the centre of imperial power, the only city that deserved a *Patria*.

## **Archaising and idealising the city (twelfth to thirteenth century)**

After the tenth century, interest in antiquity increased and references to cities in historiographical accounts occasionally include ancient monuments.27 During this period, Byzantine cities developed socially and economically and a middle class of merchants emerged. International trade and contacts with Italy opened up new horizons for the Byzantines and the inhabitants of towns and cities prospered. A striking new trend is to be observed: Byzantine authors – the vast majority of whom were part of the imperial milieu in Constantinople – drew on antique vocabulary and urban ideology to define the Byzantine city and give its inhabitants an urban identity while still following views that served state or ecclesiastical policy. The

<sup>25</sup> Leo Choirosphaktes, in *Cinque Poeti Bizantini,* 94–107. See Magdalino, 'The bath'; Mango, 'The palace of Marina'.

<sup>26</sup> See Berger, *Patria*, 17–18. See now Benjamin Anderson's chapter in this volume.

<sup>27</sup> Magdalino and Macrides, 'The fourth kingdom', 141–145; Saradi, 'The antiquities', 101ff.

historian Niketas Choniates (1155/1157–1217), an imperial bureaucrat, provincial officer and *logothetes ton sekreton* in Constantinople, wrote his *Chronike Diegesis* in three versions (the last one in exile in Nicaea) employing court rhetorical vocabulary28 and classicising language. Choniates uses the word *polis* primarily for large settlements, instead of *kastron*. For smaller settlements, he uses the derivatives of *polis* (*polisma, polismation, polichne,* etc.) and *chora*, *kome*, *phrourion, eryma, ochyrōma* and *kastron*. <sup>29</sup> A little later Theodore Skoutariotes (b. *c*. 1230), a high-ranking ecclesiastical official and Metropolitan of Cyzicus, in his *addenda* in the *Chronike Syngraphe* of George Akropolites (1217–1282), explains that small and insignificant settlements were called *phrouria* (forts), rather than cities.<sup>30</sup> He also emphasises the multiple lines of walls constructed to defend settlements: tower after tower, circular walls one after another. In Demotic texts, however, the term *kastron* continues to be used for urban settlements.<sup>31</sup> Niketas Choniates and Theodore Skoutariotes project an official urban image of the empire, namely the view held by educated court circles. The change in terminology applied to cities is expressed by the antique language used by these authors. However, the use of the term *polis* for large urban settlements also reflects the contemporary growth and prosperity of cities, which created a strong urban consciousness among the literati. Now not only Constantinople and Thessaloniki were identified as cities, but other provincial urban centres were perceived as such by classicising authors.

In the context of growing urban development and increased interest in antiquity, orators began to idealise cities and foster the illusion of a city's spiritual identity by creating images of ideal cities, populated by citizens with great intellectual and moral qualities. At the turn of the twelfth century, Michael Choniates (*c*. 1138–*c*. 1222), the intellectual Metropolitan of Athens (1182–1204), recounted the characteristics of a prosperous city according to the rules of ancient rhetoric: its good climate, location, proximity to the sea, products from earth and sea, population and social classes.<sup>32</sup> In several of his letters and addresses to provincial civil administrators (*praitors*) Choniates repeatedly expressed a negative view of Athens on the grounds that contemporary Athenians did not have the qualities of their ancient ancestors. In a poem, he represents himself falling in love with Athens, like the mythical Ixion who fell in love with Hera, but he realised that he was in the arms of an idol.33 In the same poem he refers to a painting of ancient Athens which he commissioned out of love for the glorious city. Choniates did not describe the painting, but states that his love (*ἔρως*) for ancient Athens overshadowed the nebulous remains of the ancient city.<sup>34</sup> Such feelings of disappointment were often expressed by ecclesiastics

<sup>28</sup> Magdalino, *Empire of Manuel Komnenos*, 443, 457–458.

<sup>29</sup> Kazhdan, 'Byzantine town'.

<sup>30</sup> Akropolites, *Opera*, 1.285.9–11.

<sup>31</sup> Karapidakis, 'Οι πόλεις', 212ff.

<sup>32</sup> Michael Choniates, 1, 180.1–2; Menander, 32–74.

<sup>33</sup> See the texts cited in Efthymiadis, 'Choniates', 63, n. 1 and 2. On Athens in Byzantine literature see Hunger, 'Athen in Byzanz'; Rhoby, *Reminiszenzen*, 29–92.

<sup>34</sup> Speck, 'A Byzantine depiction', 29–32.

obliged to leave Constantinople and serve in provincial cities. They show how great the gap in erudition, culture and social contacts was between Constantinople and provincial cities. Choniates, however, devoted himself to serving the interests of his metropolis and his flock.<sup>35</sup> In his inaugural address as metropolitan (1182),<sup>36</sup> delivered in the Church of the Virgin in the Parthenon, Choniates praised Athens for its glorious past and the moral and intellectual values of the ancient Athenians. He wondered whether Athens had been maintained the same as in the past ('μηδὲ τὰς Ἀθήνας ἔτι μεμάθηκα καθαρὰς καθαρῶς σώζεσθαι') or if it had become only a legend of its past glory, noting that the ancient monuments which did persist proved the city's former glory. He referred briefly to famous ancient sites: the Peripatos, the Stoa, the Acropolis, Piraeus and the Lamp of Demosthenes (the monument of Lysicrates).<sup>37</sup> However, the first oration of the Metropolitan addressing his flock was not an appropriate occasion to elaborate extensively on the topography of the pagan antiquities of Athens. Furthermore, his purpose as an ecclesiastical orator was to focus on the spiritual and moral qualities of the ancient Athenians and not their technical achievements. As a bishop, Choniates created an image of Christian Athens through the praise of the metropolitan church in the Parthenon, where he delivered his oration. He described the Christian Parthenon as a place of ascent to the heavens, identifying Athens as a New Zion and the City of God, an association usually reserved only for Constantinople.38 Choniates idealised Byzantine Athens and regarded its inhabitants as spiritually superior to their ancestors not on account of their education and intellectual achievements, but thanks to the Christian religion. In Choniates's works we observe the different concepts of the Byzantine city according to the rules of each literary genre: in his letters he complaints about the simplicity and lack of education in the provincial city, in his poem he expresses strong and unreserved love for ancient Athens and in his oration he maintains models of antique rhetoric and highly classicising language, while as a Christian orator attributing a strong religious quality to the city, consciously ignoring its antiquities.

A different urban image is drawn by Choniates's brother, Niketas, in his lament for the destroyed statues of Constantinople after its sack by the Crusaders in 1204.39 Niketas ended his historical narrative of the looting of Constantinople in highly emotional tones, comparing it to the fall of Jerusalem. He recounted that as the Latins rampaged through the defeated city they destroyed ancient statues that had been displayed in public places for centuries. Bronze statues were

<sup>35</sup> Herrin, 'Realities', 258–266.

<sup>36</sup> Michael Choniates, 93–106. See Rhoby, 'Studien'.

<sup>37</sup> Michael Choniates, 97–98; see Efthymiadis, 'Choniates', 69.

<sup>38</sup> Michael Choniates, 317. On the mystical image of Athens in this text see Efthymiadis, 'Choniates', 72–79.

<sup>39</sup> Niketas Choniates, *Historia*, 647–655. On this work see Cutler, '*De Signis*'; Saradi, 'Perceptions', 66–73; Papamastorakis, 'Interpreting the *De Signis*'; Kuttner-Homs, 'Interpréter le *De signis*'; Chatterjee, 'Sculptured eloquence'.

melted down for their metal and marble ones were smashed. Choniates repeatedly accused the Latins of greed, ignorance and barbaric behaviour. He described the statues in a highly classicising language of aesthetics and expresses an array of powerful emotions: grief and anger for the statues' destruction, admiration for their artistic value, pride in their national identity, passionate attachment to the culture they expressed and anguished hope for salvation from the enemy. He admired the naturalism, beauty and psychological insight given to the statues by their creators.

Choniates gave a symbolic meaning to the statues' destruction. He begins with the statue of Hera in the Forum of Constantine, a statue of Paris and one of Aphrodite offering Paris the gold apple. This was probably a complex of statues recalling the Trojan War and Rome's Trojan connection.<sup>40</sup> The statue of Heracles, a work of Lysippus which stood in the Hippodrome, is described in detail with admiration for its realism.<sup>41</sup> He reminds the reader that the statues of the donkey and its keeper from Nicopolis, commemorating the victory of Augustus at Actium and the statues of the hyena and the she-wolf which nourished Romulus and Remus, were revered monuments of the nation ('τὰ παλαιὰ σεμνώματα τοῦ γένους').<sup>42</sup> The most elevated lyrical description is reserved for the statue of Helen<sup>43</sup> of extravagant beauty. He declares that the statue was destroyed by the Latins, not as a punishment for Troy's destruction by the Greeks, as could have been suggested, but because of the Latins' greed and ignorance of the Homeric epic, a chant about Helen. Niketas Choniates concludes with the theme of justice being re-established. A statue group of two animals fighting suggests, through metaphorical Biblical allusions, that right and justice will be restored and the Latins will ultimately be defeated.<sup>44</sup> Niketas painted a picture of the topography of Constantinople through reference to Greek and Roman culture and historical tradition: the Trojan origins of the Romans, the Roman origin of Byzantium, the unsurpassed Greek artistic and literary tradition. By contrast, he makes only general reference to Christian icons and relics being destroyed or looted.<sup>45</sup> In Niketas' view, statues symbolised the connection of Constantinople to Greek and Roman antiquity, a connection which the Latins did not possess.

### **Praising independent cities: the utopian city and political identity (thirteenth to fifteenth centuries)**

From the thirteenth century onwards, when the Byzantine Empire had dissolved into city-states after the Latin occupation of Constantinople in 1204, intellectuals

<sup>40</sup> Bassett, *Urban Image of Late Antique Constantinople*, 205.

<sup>41</sup> Niketas Choniates, 649–650.

<sup>42</sup> Niketas Choniates, 650.

<sup>43</sup> Niketas Choniates, 652–653.

<sup>44</sup> Saradi, 'Perceptions', 70–73.

<sup>45</sup> Chatterjee, 'Charisma', 244–245, 261.

responded to the regional independence and clout of cities. Intellectuals at the courts of local magnates expressed local attachments and feelings and a commitment to their cities, and sought a prestigious social identity for them. One way to achieve this was idealisation, a mode provided by the rhetorical tradition. Thus, as in the early Byzantine period, praises of cities (*enkomia*) were once more written in the form of orations publicly delivered in accordance with the rules of ancient rhetoric. Although these texts included brief descriptions of the urban topography, they idealised their subjects unreservedly in contrast to the reality of the cities' physical appearance. Cities are said to possess unsurpassed beauty, be surrounded by idyllic scenery and have a glorious history and ancestors. Their citizens appreciated high culture and enjoyed a supreme level of education. They were endowed with a religious spirit and displayed lofty moral values. The city and its residents mingle in one image of beauty, spirituality and morality. Each of these texts emphasises a different aspect of the city depending on the author's preferences and the political message he wanted to convey. Features of ancient urban topography are rarely mentioned. Instead, the emphasis is placed on inherited ancient wisdom and education, and on the piety and Christian spirituality of the urban inhabitants.

In the early 1250s, Theodore II Doukas Laskaris (1254–1258), son of the emperor John Vatatzes (1222–1254), wrote an oration concerning Nicaea, the capital of the empire of Nicaea, formed after the fall of Constantinople in 1204. Laskaris delivered the oration before the emperor. Laskaris, an intellectual, employs philosophical ideas and complex metaphors to draw an idealised picture of Nicaea which focuses on the beauty of nature and on the intellectual supremacy and spirituality of the inhabitants. He first praised the city walls, because they secured the city's freedom and preserved the empire after the fall of Constantinople in 1204.46 They were strong, with high battlements and towers,<sup>47</sup> skillfully constructed from magnificent materials ('τῆς περιφανεστάτης ὕλης'), probably a reference to the antique gates and sections of the walls still standing today.<sup>48</sup> The city's true adornment was, however, not its appearance, but the incomparable *paideia* of its citizens.49 As a centre of advanced education, it resembled the 'divine mind' ('νοῦν θεῖον'), and the citizens resembled 'sublime thoughts' ('διανοήματα εὐγενῆ').50 Walls and education are intertwined in a variety of figures of speech. The absence of detailed descriptions of buildings opens the way for a series of idealised urban images: towers and walls seemed alive ('καὶ ὥσπερ ἔμψυχα φέρουσα τὰ πυργώματα καὶ ἐν

<sup>46</sup> Rhoby, 'Die Rezeption', 119–121. On the topography of Nicaea see Foss, *Nicaea*, 89–122; Peschlow, 'Nicaea'.

<sup>47</sup> On other references to the strong walls of Nicaea and its towers see Rhoby, 'City *Encomia*', 90–91. Nicaea had been praised for its strong walls in the *Dionysiaca* of Nonnos, see Lauritzen, '*Ekphraseis*', 187, 213.

<sup>48</sup> Laskaris, *Theodorus II,* chap. 1 (l. 10). On this oration see the comments of Foss, *Nicaea*, 128–129, 155–163.

<sup>49</sup> Laskaris, *Theodorus II*, chap. 1 (l. 43–47), 69–70 and *passim*.

<sup>50</sup> Laskaris, *Theodorus II*, chap. 2 (l. 68–69), 70.

ζωῇ τὰ περιτειχίσματα').51 Secular wisdom on its own had no value unless it was combined with the knowledge of God.<sup>52</sup>

Laskaris outlined the topography of Nicaea among images conveying the beauty of nature around and within the city, which formed a city of the graces.<sup>53</sup> Laskaris also stressed Nicaea's abundance of water: in addition to streams around the city, a raised aqueduct brought in water, making it available to every house.<sup>54</sup> The most beautiful churches and splendid monasteries made Nicaea a city of God.55 Thus people were guided to reflect on the divine and to create the idea that Nicaea was a city of God to which angels daily descended to give orders to the emperors and ascend again.56 The image of descending and ascending angels had an apocalyptic meaning, alluding to the *Revelation* (22.6–7) of Saint John, in which angels descend from and ascend to the sky, the last showing the Celestial Jerusalem and the Last Judgement. I have suggested elsewhere that this image conveys a historical message as it implies a hope for the recovery of Constantinople and the punishment of the Latins.<sup>57</sup> Laskaris also compared Nicaea, with its strong walls like pillars of adamant, to the House of Wisdom celebrated by Solomon (*Proverbs*, 9.1–6).<sup>58</sup> Thus Nicaea was connected with both Jerusalem and Constantinople, the New Jerusalem,59 as it was invested with apocalyptic characteristics symbolising the rebirth of the empire.<sup>60</sup> Orators of the empire of Nicaea related emperors to Old Testament figures, especially David, implying that they had saved the empire, and Laskaris ended his oration too with praise of the emperor.<sup>61</sup>

Laskaris eloquently affirmed the Byzantine image of a city. He highlighted Nicaea's strong walls and churches, while stressing the supreme education and piety of its inhabitants and the power of the emperor. By contrast, Laskaris referred only fleetingly to the ancient city when he mentions the antiquity of the walls and the aqueduct, both vital features for the city's survival. The brief reference to other monuments, apart from the walls, is explained by the rhetorical model Laskaris follows in his oration, as defined by Menander: the praise of the city by reason of the achievements and qualities of its inhabitants. In this model, topographical descriptions are limited.

As we noticed in the case of Michael Choniates, different literary genres demanded different perspectives on cities. In contrast to the rhetorical models of the city praise, epistolography allowed the expression of personal feelings and personal observations. While in Laskaris's praise of Nicaea only discreet reference was made to the remains of

<sup>51</sup> Laskaris, chap. 2 (l. 61–62), 70. See Saradi, 'The monuments', 181–182.

<sup>52</sup> Laskaris, chap. 3.71–74.

<sup>53</sup> Laskaris, chap. 5 (l. 200–201), 76.

<sup>54</sup> Laskaris, chap. 6 (l. 234–238), 78; Foss, *Nicaea*, 160.

<sup>55</sup> Laskaris, chap. 5 (l. 204), 76 and chap. 7 (l. 248), 78.

<sup>56</sup> Laskaris, chap. 5 (l. 204–207), 76.

<sup>57</sup> Saradi, 'The monuments', 183.

<sup>58</sup> Laskaris, chap. 5 (l. 220–221), 77. 59 On Constantinople as New Jerusalem see Rhoby, 'Stadtlob', 286–287.

<sup>60</sup> See Saradi, 'The monuments', 183–186.

<sup>61</sup> Angelov, *Imperial Ideology*, 127–131 .

the ancient city, in a letter addressed to George Akropolites Laskaris spoke of the ruins of ancient Pergamon and expressed his profound admiration. In this case, the remains revealed past splendour and the authority and wisdom of the ancient builders. This was ancestral glory and, compared with the buildings of his day, the buildings of Pergamon were marvellous.<sup>62</sup> While Laskaris was conscious of the continuity from the ancient city to the Byzantine, he also recognised the rupture in Pergamon's urban topography.

Another oration on Nicaea, written by Theodore Metochites (1270–1332) in 1290 on the occasion of the arrival of Emperor Andronikos II Palaiologos (1282–1328), focuses on other aspects of the city praise. Metochites refers pointedly to the ancient history of Nicaea,<sup>63</sup> from which the city derived great fame, representing the Emperor Trajan as the city's benefactor who raised it to the eminence it enjoyed in Metochites's time ('ὡς ἔχει νῦν'), a reminder of and monument to Trajan's generosity. One can thus regard the city both as old and new ('ὥστε τὴν αὐτὴν ἐξεῖναι παλαιάν τε ὁμοῦ καὶ νέαν ὁρᾶν'). Metochites reveres antiquity and recognises that it redounds to the honour of the city ('καὶ εἰ δή τῳ σεμνὸν ἡ ἀρχαιότης, καὶ τοῦτο μετὰ τῆς πόλεως'). In reality, Hadrian, the adopted son of Trajan, rather than Trajan himself, had restored Nicaea. The confusion might have arisen from two inscriptions, visible on the eastern gate of the walls, commemorating the restoration by Hadrian, named there as 'Trajan Hadrian'.<sup>64</sup> The topographical connection between antiquity and Byzantine cities would be increasingly emphasised as Byzantine intellectuals, in order to define their cultural identity and to distinguish themselves from the Latins, turned to their ancient Greek roots.<sup>65</sup>

Metochites employs idyllic imagery to depict Nicaea's location, its lake, and the surrounding countryside. He describes the walls of the city in elevated tones and mentions its lofty buildings built close to one another, its baths, its philanthropic institutions and its monasteries, which he praised as places of spirituality and meditation.<sup>66</sup> There is an extended description of an unnamed church, perhaps the Church of Hyacinth (dedicated to the Dormition of the Virgin), which may be unidentified because it was where the oration was delivered.<sup>67</sup> Under the Laskarids, the monastery attached to the church housed the patriarch, while members of the imperial family were buried there. Metochites praises the church's gleaming golden decoration, its paintings, its mosaics and its columns.<sup>68</sup> He describes in elevated tones the natural landscape surrounding the church and made eloquent reference to the raised aqueduct.69 The holiness of the church is demonstrated by the annual miracle

<sup>62</sup> Laskaris, *Epistles*, ep. 32, 107–108.

<sup>63</sup> Metochites, *Nikaeus*, chap. 2 (315).

<sup>64</sup> Foss, *Nicaea*, 197–198.

<sup>65</sup> Koder, 'Griechische Identitäten'; Kaldellis, *Hellenism*, 317ff.; Malamut, 'De l'empire des Romains'.

<sup>66</sup> Metochites, *Nikaeus*, chap. 6 (318).

<sup>67</sup> Foss, *Nicaea*, 97–101, 115–117; Rhoby, 'City *Encomia*', 93.

<sup>68</sup> Metochites, *Nikaeus*, chap. 7 (320).

<sup>69</sup> Metochites, *Nikaeus*, chap. 7 (320). 'Πάρεστι δὲ καὶ ὑδάτων θαυμάζειν ἐπιρροίας, ὅσας τε δαψιλῶς ἡ γῆ χορηγεῖ καὶ ὅσας ἐπὶ τῇ φύσει μηχανᾶται τέχνη, τὰς ἀναδόσεις ἐκ μετεώρου σοφιζομένη'.

of its patron, Saint Tryphon, and by the two ecumenical councils that took place there in 325 and 787. Metochites emphasises that Nicaea played a vital role in preserving the empire after 1204. The oration ends with praises of the emperor.

For Metochites, Nicaea's antique past and the Nicaea of the present formed a unity, although the antiquity of Nicaea occupies only a comparatively small part of the oration. The theme of a city's past is, however, central to *Byzantios*, a praise of Constantinople written by Metochites between 1305 and 1311/1312.70 This is a hymn to Constantinople and expresses a sense of optimism and belief in the empire after the re-occupation of the city in 1261. Metochites explains how Constantinople enjoys a splendid natural environment. Nature is praised repeatedly and at great length in elaborate, picturesque images, infused with great emotion and aesthetic sensibility.<sup>71</sup> He states that Orthodox Christianity was embedded in Constantinople's history at its foundation by Constantine.<sup>72</sup> The eternal presence of Christianity secured the primacy of Constantinople over the rest of the Christian world at the time when the Pope had moved to Avignon.<sup>73</sup> Constantinople had a historical role as the cradle of Orthodox Christianity and the capital of an ecumenical empire. It was the birthplace of the nation ('τῆς ὅλης τοῦ γένους ἀρχῆς').<sup>74</sup>

There is general praise of the city's buildings, without any description of specific monuments. The praise of the walls, which comes first, as in the praises of Nicaea, is intertwined with the account of the foundation of Constantinople, when the nations of the Romans and the Greeks merged with each other to create the new empire. He states that the walls had been beautifully constructed using superb technique, combining strength with beauty. The walls lead Metochites into a meditation on the interaction of time and the city. While time is by nature transient, constantly leading everything into decay and change, in the case of Constantinople's walls time acts in a different way. The passage of time, rather than destroying them, had constantly preserved the greatness of the walls, nurturing and adding new adornments to them.<sup>75</sup> This is followed by the praise of the marvels and beauties ('θαύματα καὶ κάλλη') within the walls, which resemble a treasury, of the richness of the markets, and of the beauty of all kinds of buildings, old and new.76 Metochites dwells eloquently on the constant renewal of the old buildings that demonstrate the old glory of Constantinople. The old buildings were still standing from long ago and new ones were constantly added. Those that were damaged were replaced with new ones, thus adding new grace and beauty to the city.<sup>77</sup> Even the ruined old buildings

<sup>70</sup> Rhoby, 'City *Encomia*', 84.

<sup>71</sup> Metochites, *Byzantios*, chaps. 4.69–9 (140–198).

<sup>72</sup> Metochites, *Byzantios*, chap. 1.51–58 (114), chap. 17.22–29 (252–254); Saradi, 'Οι ταυτότητες της Κωνσταντινούπολης', 24–25.

<sup>73</sup> Magdalino, 'The beauty of antiquity', 110–111.

<sup>74</sup> Metochites, *Byzantios*, chap. 5.4 (140).

<sup>75</sup> Metochites, *Byzantios*, chap. 19 (262–270).

<sup>76</sup> Metochites, *Byzantios*, chap. 20.3–4 (270).

<sup>77</sup> See Robert Ousterhout's chapter in this volume for a discussion of this theme in Metochites.

gave prestige ('τῆς παρὰ τῶν ἐρειπίων αὐτῶν καὶ τῶν παλαιῶν λειψάνων αἴγλης'), fabulous radiance and beauty to the city ('καὶ λαμπρότητος ἀμυθήτου καὶ καλλονῆς'). The remains of the splendid old buildings indicated the past glory of the city. In a powerful yet gloomy image Metochites compares the old buildings to tombstones and the epitaphs on them which testify to the glory of the dead. Metochites draws a dark picture of Constantinople, severely damaged by the Latins, in which the ruined buildings recalled the city's past glory. However, there was no need to refer to the past in order to show Constantinople's distinguished origin, the marks of which in his time were fading. The buildings that still existed and were visible were enough to demonstrate the city's greatness. Furthermore, in Constantinople both old and new buildings adorned the city.<sup>78</sup> There is no clear indication that Metochites is referring to any ancient Greek or Roman remains in Constantinople. He is instead probably referring to buildings damaged by time and ravaged by the Latins during the Fourth Crusade. Yet he claims that many buildings remained intact despite the passage of time, entirely undamaged as if they had been built yesterday or the day before. Metochites reminds his readers of the greatness of the buildings in Constantinople, evident from what is left of those which were damaged, using proverbial phrases such as 'to recognise the lion by its claws' or 'the beauty of the clothing is shown by its hem'.79 In the lengthy section on the old and renewed buildings, Metochites calls attention to the efforts of the emperor to renew the city after the re-taking of Constantinople in 1261.<sup>80</sup> The theme of the eternally renewed Constantinople, so central to *Byzantios*, had been promoted by the Komneni.<sup>81</sup>

Of the buildings inside the walls, Metochites praises first the churches and monasteries without describing any of them.82 Other urban structures are mentioned in general terms: the porticoes to shelter beggars and to offer protection from heat and rain, areas for exercise, broad streets, theatres, cisterns, water fountains, aqueducts, baths, hospitals, columns and statues.<sup>83</sup> Statues, an old literary theme with regard to Constantinople, are in particular praised eloquently for the symmetry and complexity of their art, since Constantinople claimed to have a greater number of more beautiful statues than anywhere else.<sup>84</sup> Constantinople's landmark was the Church of Hagia Sophia, which is praised at great length for its majestic architecture and sublimity, but without describing its adornments.<sup>85</sup> There is only passing reference to the inhabitants' *paideia*. 86

<sup>78</sup> Metochites, *Byzantios*, chap. 19.65–80 (268–270), chaps. 20–21 (270–284). On this theme see Magdalino, 'The beauty of antiquity', 109–110, 114; Saradi, 'Οι ταυτότητες της Κωνσταντινούπολης', 33–34.

<sup>79</sup> Metochites, *Byzantios*, chap. 21 (280–282).

<sup>80</sup> Macrides, 'The new Constantine'; Talbot, 'The restoration', 251, 260.

<sup>81</sup> See, for example, Manassis, *Breviarum Chronicum*, vv. 2321–2322, 2507–2508.

<sup>82</sup> Metochites, *Byzantios*, chap. 22.31–61 (286–290).

<sup>83</sup> Metochites, *Byzantios*, chap. 23 (290–296).

<sup>84</sup> Metochites, *Byzantios*, chap. 23.33–83 (292–296).

<sup>85</sup> Metochites, *Byzantios*, chap. 32 (342–356).

<sup>86</sup> Saradi, 'Οι ταυτότητες της Κωνσταντινούπολης', 31–32.

Constantinople, rather, is elevated to celestial heights. She is eternal, splendidly beautiful, the leader of the world which overcomes time, endowed with supreme Orthodox Christian values. The city itself is a divine temple, purified and dedicated to God.<sup>87</sup> The repeated emphasis on the inhabitants' faith in Orthodox Christianity is inspired by the anti-unionist policy of the emperor Andronikos II, whose imperial policy Metochites would have endorsed and promoted, being a member of the imperial bureaucracy.<sup>88</sup> An urban topography, sketched in very general terms, leads to an idealisation of Constantinople. Numerous words and expressions derive from the orations *Panathenaikos* and *To Rome* of Aelius Aristides, thus endowing Constantinople with the qualities of Athens and the might of the ecumenical Roman empire.89 The reality behind these images was, however, very different. The population had shrunk and, dispirited, was scarcely renewing itself. Buildings had been damaged and patched up. Constantinople was poor, militarily weak and threatened on all sides. Metochites presents Constantinople through its past, using images of nature and politically inspired symbolism to conceal the city's real condition.

About one hundred years later, Manuel Chrysoloras (*c*. 1350–1415) compared Constantinople with Rome in a letter addressed to the Emperor John VIII Palaiologos (1425–1448).90 Chrysoloras was a distinguished scholar who promoted Greek letters in Italy. He was also a diplomat who in his missions tried to obtain support from the western powers against the advancing Turks. The promotion of Greek letters in Italy was part of a cultural policy, as it formed an argument addressed to the leaders of the west in favour of the salvation of Byzantium for the sake of its Greek learning.91 As a teacher of the Greek language in renaissance Italy, he instilled enthusiasm for the Greek letters in prominent Italian humanists. His ties with Italy were very strong and in the end he converted to Catholicism. Just as Michael Choniates and Theodore Laskaris showed interest in ancient monuments in their letters, Chrysoloras too describes ancient monuments of Rome and Constantinople in his letter to the emperor, most of them ruined. Now, however, influenced by the spirit of the Italian renaissance, Chrysoloras makes urban antiquities the focus of his letter. For the first time, a Byzantine intellectual represented himself as walking through Rome's ancient ruins, admiring them and reconstructing their ancient history. In comparison to Rome's antiquities, those of Constantinople were very few: there were column bases that once supported statues and two of the twelve reliefs of the Golden Gate, which depicted Heracles' labours and the torture of Prometheus, an obvious allusion to the enormous difficulties faced by the Byzantine emperor.92 Other statues were hidden, a reference to their power as talismans of the city. In contrast to these antiquities,

<sup>87</sup> Metochites, *Byzantios*, chap. 30.2–5 (328).

<sup>88</sup> Metochites, *Byzantios*, chaps. 30–31 (328–342), chaps. 38–41 (392–424); Saradi, 'Οι ταυτότητες της Κωνσταντινούπολης', 28–31.

<sup>89</sup> Saradi, 'Οι ταυτότητες της Κωνσταντινούπολης', 23, 25–27, 34–40.

<sup>90</sup> Dagron, 'Manuel Chrysoloras'; Kioussopoulou, 'La ville'; Saradi, 'The antiquities', 106–108.

<sup>91</sup> Thomson, 'Manuel Chrysoloras'.

<sup>92</sup> Chrysoloras, *Confronto tra l'Antica e la Nuova Roma,* chaps. 47–49 (20–21).

the Constantinopolitan church of Hagia Sophia, eloquently described, was the city's real landmark.93 While Rome is represented through its past glory, Constantinople – as also in Metochites' *Byzantios* – was superior, in that it was the capital of the Christian Empire and the seat of an emperor. While Metochites referred to Byzantine buildings, of which the ancient ones would be included but not explicitly mentioned, Chrysoloras, influenced by the Italian renaissance, refers to ancient and Byzantine buildings in Constantinople. In both texts, however, as Constantinople is presented in layers of its past and present prestigious monuments, commitment to the empire was the central idea.

The oration on Trebizond by Bessarion (1403–1472), written around 1436, represents the final step in the evolution of encomiastic descriptions of the city.94 The oration follows the *Panathenaikos* of Aelius Aristides, although it borrows from other texts, too, including the *Antiochikos* of Libanios. Its central idea is that divine destiny preserved the freedom of Trebizond, capital of an independent empire in the last Byzantine centuries, at a time when other Byzantine cities were falling to the Turks.95 Bessarion praises Trebizond for its ancient origins which connected it with Athens, the mother of learning for the Greeks, by way of her colonies of Miletos and Sinope, for the military spirit of its inhabitants, for their love for freedom, and for their philanthropy and magnanimity.96 Throughout the oration, Trebizond is shown as possessing the qualities of Athens, and Bessarion specifically describes the city as an image and imitation of Athens ('… καὶ πρὸς τὴν Ἀττικὴν καὶ τὴν πόλιν τῶν Ἀθηνῶν ἐξεικασμένη τοῖς πλείοσιν ἐκείνης εἰκὼν καὶ ἀκριβὲς ὡς ἐνῆν ἀπείργαται μίμημα').97 He gives a lengthy account of the city's history from ancient times.98 Bessarion emphasises the supremacy of the Greeks over the Romans and the influence of Greek civilisation on Roman.99 After the Roman occupation was over, Trebizond remained free ('ἀνάλωτος').<sup>100</sup> The consciousness of the past is so strong that it forges a Greek national identity for the people of Trebizond which turns Trebizond from a Byzantine into a Greek city. Greek national consciousness emerged gradually in the works of the late Byzantine literati, who were faced with the threat of the advancing Turks. They turned to ancient Greece and ancient Greek culture, while the masses looked to their Church for salvation. Bessarion also promoted approaches to the Latin Church. This may explain his silence on the piety of the Trebizondians and the Orthodox Church, a theme prominent in earlier texts. For him, the ancient Greek qualities of Trebizond were stronger than religious faith.

<sup>93</sup> Chrysoloras, chaps. 52–57 (22–24).

<sup>94</sup> Giarenis, 'H *Έκφρασις* του Βησσαρίωνα', 180–187.

<sup>95</sup> Saradi, 'Η *έκφρασις* της Τραπεζούντας', 39–40; Giarenis, 'H *Έκφρασις* του Βησσαρίωνα', 194–195, 205.

<sup>96</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 5 (24.7–10).

<sup>97</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 11 (29–30); Saradi, 'Η *έκφρασις* της Τραπεζούντας', 41–42, 44–47; Giarenis, 'Η *Έκφρασις* του Βησσαρίωνα', 192–194.

<sup>98</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chaps. 18–21 (42–54).

<sup>99</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 19 (50).

<sup>100</sup>Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 20 (51.29–31).

Trebizond is praised for its central position in Asia, an ancient rhetorical topos, usually applied to Constantinople.101 This idea was appropriate in the historical circumstances of the time, as Trebizond was promoted as a rival to Constantinople. Trebizond's two harbours made her an international city, because they promoted international trade and contact with other peoples.<sup>102</sup> The openness offered by the sea is contrasted with the medieval characteristics of a naturally defensive site. Mountains were a barrier to the barbarians who dwelt in the area around the city. Deep gorges with rivers and strong walls secured the city.<sup>103</sup>

The best proof of the city's survival lay in its layout and topography, of which Bessarion gives a lengthy description, of about four pages, at the end of the oration.<sup>104</sup> Its condition shows that rule by the Grand Komneni had been successful and that the city was free, albeit surrounded by enemies. Three lines of walls protected the lower city, the upper city, the acropolis and the harbour. Wooden bridges set over a deep gorge linked the walls of the city to its territory.<sup>105</sup> Suburbs extended outside the walls amidst gardens and meadows. There were workshops and shops hosting local and international trading activity. The houses were built with timber from the forests in the mountains around.106 Bessarion describes the palace on the acropolis in detail.<sup>107</sup> He mentions vestibules, halls, balconies, verandas, servants' quarters and the central hall. Its ceiling was decorated with paintings depicting the sky and stars and its painted walls displayed portraits of the Grand Komneni, their ancestors and historical events. The throne stood on a platform covered by a pyramidal cupola set on four columns and surrounded by a screen of white marble. In a second hall there was another lofty throne.

Bessarion praises Trebizond for its glorious history, its survival over the centuries and its divine destiny, these all being ideas previously used in praises of Constantinople. Constantinople is referred to as Byzantion,<sup>108</sup> and, while its Roman imperial legacy remained unquestionable, its prestige was tarnished because its surroundings had succumbed to Ottoman rule.109 The praise of the Grand Komneni for their descent from the Komnenian dynasty subtly alludes to the connection of Trebizond with Constantinople.<sup>110</sup> However, in Bessarion's praise Trebizond does not replace Constantinople. Trebizond resembles Athens in her defence of the Greeks and

<sup>101</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chaps. 12–14 (30–34); Metochites, *Byzantios*, chap. 6.171–172 (164). On this *topos* see Koder, 'Anmerkungen'.

<sup>102</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chaps. 15–16 (32–37).

<sup>103</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chaps. 17–18 (40–41).

<sup>104</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 18 (41–42).

<sup>105</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 22 (60.20–61.35).

<sup>106</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 22 (62–63.11).

<sup>107</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 22 (63–65); Dorattiotto, 'La Trebisonda', 8–14; Saradi, 'The monuments', 190–191.

<sup>108</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chaps. 20 (51.17, 32; 52.23), 21 (53.20).

<sup>109</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 20 (51.31–33).

<sup>110</sup> Bessarion, '"Εἰς Τραπεζοῦντα"', chap. 21 (57–58). See also Karpov, 'Trapezunt'; Giarenis, 'Η *Έκφρασις* του Βησσαρίωνα', 203–205.

in her philanthropic rule over them. Athens is no longer referred to as a pagan city. While Bessarion only in passing mentions Athens as a centre of *paideia*, he represents it instead as a model of military might and benevolence.

## **Conclusion**

The idea of the Byzantine city was shaped by the rules of literary genres: idealism and commitment to the imperial rule dominate epideictic rhetoric; emperors and local history traced through ancient and Byzantine monuments are the focus of *Patria*; observations of ancient remains are a theme in epistles. In addition, the authors' personal views and political agendas mark all genres.111 Urban topography is part of Byzantine epideictic rhetoric only for the imperial capitals, and even then its use is limited. Topographical description increased in the last century of Byzantium, perhaps reflecting the influence of the Italian renaissance. Statuary is praised only in connection with Constantinople, usually in symbolic terms fitting for the *Patria*  genre – a genre reserved for the capital, since from the seventh century provincial cities lost their civic identity.

During the final centuries of Byzantium, as the empire fell prey to foreign invaders, the image of the city was increasingly idealised. Walls which had been the most essential element of urban survival and identity now became ideal symbols and their praise is combined with ancient education and imperial symbolism. Byzantine authors emphasised churches less and less, as humanism directed their attention to classical literature. The terrifying vicissitudes of the age compelled the authors to create urban images of abstract beauty and to revert to memories of power and grandeur. Ultimately, the perception of urban space ceased to be dominated by buildings, but rather by the beauty of its natural environment, the excellence of its intellectuals, the spirituality of the religious piety of its inhabitants, the values of the past and pride in history.

## **Bibliography**

### *Primary*


Laskaris, *Epistles*: *Theodori Ducae Lascaris Epistulae CCVII*, ed. N. Festa (Florence, 1898).

Laskaris, *Theodorus II Ducas Lascaris. Opuscula Rhetorica*, ed. A. Tartaglia (Leipzig, 2000).

Leo Choirosphaktes, in *Cinque Poeti Bizantini. Anacreontee del Barberino Greco 310*, ed. F. *Ciccolella* (Alessandria, 2000), 94–107.

<sup>111</sup>For a clear reference to a political message through a praise of Constantinople, see Akropolites, 188–189.


#### *Secondary*

Anderson, B., 'Leo III and the Anemodoulion', *Byzantinische Zeitschrift* 104.1 (2011): 41–54.


Liebeschuetz, J. H. W. G., *Decline and Fall of the Roman City* (Oxford, 2001).

Macrides, R., 'The new Constantine and the new Constantinople – 1261?', *Byzantine and Modern Greek Studies* 6 (1980): 13–41.

Magdalino, P., *The Empire of Manuel I Komnenos, 1143–1180* (Cambridge, 1993).

Magdalino, P., 'The bath of Leo the Wise and the "Macedonian Renaissance" revisited: topography, iconography, ceremonial, ideology', *Dumbarton Oaks Papers* 42 (1988): 117–156.

Magdalino, P., 'The beauty of antiquity in late Byzantine praises of Constantinople' in P. Odorico and Ch. Messis (eds.), *Villes de toute beauté, l'ekphrasis des cités dans les littératures byzantines et byzantino-slaves. Actes du colloque international, Prague, 25–26 novembre 2011*, Dossiers Byzantins 12 (Paris, 2012), pp. 101–121.

Magdalino, P. and R. Macrides, 'The fourth kingdom and the rhetoric of Hellenism' in P. Magdalino (ed.), *The Perception of the Past in Twelfth-Century Europe* (London, 1992), pp. 117–156.


## Chapter 14

## Ottoman urbanism and capital cities before the conquest of Constantinople (1453)

*Dimitri J. Kastritsis*

By the time the Ottomans captured Byzantine Constantinople and made it the capital of their own empire, they already had a long history of urbanism spanning over a century. Operating from the beginning of their existence in a region that had always been at the very centre of the eastern Roman world, their early conquests included many towns with a continuous history stretching back to antiquity. In 1331 they captured Nicaea, site of two ecumenical councils and former capital of the most successful Byzantine successor state following the Latin sack of Constantinople in 1204. The Byzantine court of Nicaea had moved back to Constantinople in 1261, only forty years before the Ottomans first made their appearance in the region. So when Nicaea became Ottoman, its role as a Byzantine capital-in-exile was still within living memory. If Nicaea was ever considered as an Ottoman capital, this did not last; five years earlier, they had taken another important Byzantine town, Bursa (Proussa), which was to fulfill this role into the fifteenth century. Shortly after they crossed the straits to Europe around 1350, as allies of the Byzantine claimant John VI Kantakouzenos, they went on to capture other historic towns, including Gallipoli (Kallipolis, Gelibolu), Dimetoka (Didymoteichon) and Edirne (Adrianople), all with a Roman and Byzantine past.1 In the first half of the fifteenth century, Edirne would gradually replace Bursa as Ottoman capital, while Gallipoli would continue to function as the main Ottoman port, and Dimetoka as a fortified stronghold.<sup>2</sup>

<sup>1</sup> Gallipoli was conquered in 1354 and Dimetoka by 1361. The date of the Ottoman conquest of Edirne is still disputed, see Beldiceanu-Steinherr, 'La conquête d'Adrianople par les Turcs'; İnalcık, 'The conquest of Edirne'; Zachariadou, 'The conquest of Adrianople by the Turks'.

<sup>2</sup> It is still not uncommon to find the view that immediately after its conquest by the Ottomans (whenever this actually occurred) Edirne replaced Bursa as Ottoman capital: *e.g*. Blessing, 'Seljuk past and Timurid present', 227. In the fourteenth century, Edirne was certainly the main military base for Ottoman expansion into the Balkans, and capital of Ottoman Rumelia; but as I will show below, Bursa

These are but a few examples of cities with a long Roman past taken by the Ottomans in the first century of their history. They all have in common their central location around Constantinople, the undisputed centre of the eastern Roman world. By this time, Constantinople was an expanse of partly occupied or abandoned buildings, ancient monuments and open fields, all contained within the massive Theodosian walls that still define the old town of Istanbul today. When it was finally captured by the Ottoman Sultan Mehmed II 'the Conqueror' (*fatih*) in 1453, the city was in desperate need of resettlement and reconstruction. Mehmed began this work shortly after the conquest, and by his death in 1481 had already transformed Constantinople into a suitable capital for his empire. From the perspective of early modern and later Ottoman history, fields practically synonymous with the academic study of the Ottoman Empire today, it is tempting to view the empire as only really beginning in 1453. According to this view, what is past is prologue to the great Ottoman Empire of Mehmed the Conqueror, Süleyman the Magnificent (r. 1520–1566) and their successors. It is certainly true that by the second half of the sixteenth century, people living both within and outside the empire's borders had come to think of it as it had become during the reign of Süleyman. By then, its borders stretched from the Caucasus to Algiers and from Hungary to Yemen. At least in name, all of this was ruled from Istanbul, which the Ottomans still formally called Constantinople (*Ḳosṭanṭıniyye*).

Under Mehmed and his successors, the old Byzantine capital was repopulated and rebuilt with new palaces and foundations. These included Mehmed's own 'new' mosque complex (Fātih), which would serve as a model for such foundations in the future, and the Topkapı Palace, which would continue to develop and function as the centre of Ottoman government for centuries to come.3 By the mid-sixteenth century, the city was flourishing once more with a population comparable to that of the Justinianic era,4 and the great architect Sinan had developed the classical Ottoman style whose culmination was the mosque complex of Süleyman I.5 This style would radiate outwards from Istanbul to the provinces as a projection of Ottoman culture and authority, where it is still rightly associated with the Ottoman legacy. However great the significance of Mehmed's actions following 1453, it is true in more ways than one that he and his successors were not beginning from nothing when they rebuilt Constantinople as the new Ottoman capital. First of all, despite the poor state of its infrastructure, in 1453 the city still boasted many famous buildings dating from both

was still thought of as the main Ottoman capital ('abode of sovereignty', *dār al-mülk*), a situation that lasted well into the fifteenth century.

<sup>3</sup> On the Topkapı Palace, see Necipoğlu, *Architecture, Ceremonial, and Power*; Kafescioǧlu, *Costantinopolis/ Istanbul*, 56–66. On Mehmed's 'new' mosque complex (Fatih) and its relationship to the Holy Apostles, see Kafescioǧlu, *Costantinopolis/Istanbul*, 54, 66–92; Dark and Özgümüş, *Constantinople*, 83–96. The last work is based on archaeological research and points to continuities between the two structures, including reuse of a Byzantine limestone wall.

<sup>4</sup> For Constantinople under Justinian, a common population estimate is ca 350,000. Another common estimate for the Ottoman city, 700,000 in the sixteenth century, is probably exaggerated at least for the city *intra muros*. See İnalcık, 'Istanbul', 244.

<sup>5</sup> See Necipoğlu, *The Age of Sinan*, 189–256.

ancient and medieval times. These included the cathedral of Hagia Sophia, which was converted to a mosque and had a great influence on Ottoman architecture in the fifteenth century and beyond, and the crumbling church of the Holy Apostles, which Mehmed would raze to build his own 'new' mosque complex in the years 1463–1470. At the same time, the Ottoman sultan and his court brought to Constantinople their existing architectural and urban culture, as it had been developing elsewhere for over a century. This in turn had been created in dialogue with the existing material heritage they found in these towns; the architectural traditions of medieval Anatolia, Byzantium and the wider Islamic world; and last but not least, an awareness of the past and their own history as it was evolving.6

After Mehmed the Conqueror's decision to rebuild Constantinople and make it the capital of his empire, the former Byzantine metropolis would come to play a central role in Ottoman history surpassing that of the previous capitals. In many respects, this was similar to the role the city had played for Byzantium, which had been deemed in its own turn a direct continuation of Rome. While this much is clear, what is significantly less so is the extent to which the city continued trends already present in previous Ottoman capitals. Thanks to palaces, state-funded religious complexes, and other important social buildings, Bursa, Edirne and other key cities (*e.g*. Amasya, Manisa, Gallipoli) played a major role in the political struggles of the first half of the fifteenth century. With or without the consent of their inhabitants, these towns served as seats of Ottoman courts and administrative structures, and not infrequently switched hands between different contenders for the throne, as well as becoming targets for hostile neighbouring states. Given this turbulent history, it is fair to say that architecture dating from this time serves the modern historian both as a source of 'hard' data, such as inscriptions providing information on who ruled a particular city at a given time, as well as more complex political and cultural messages. Buildings were imbued with multiple layers of meaning, which could derive equally from their religious and social functions, architectural style and association with one or more founders. But such messages and their reception can be difficult to interpret in the absence of a wider source base. For example, it is still unclear to what extent the strikingly Byzantine brickwork of Murad I's complex in Bursa, or the Timurid tiles of Mehmed I's 'green' (Yeşil) complex in the same city, represented or were perceived as deliberate attempts to communicate political or cultural messages, or should simply be attributed to the availability of artisans skilled in particular architectural styles.7

<sup>6</sup> For a wide-ranging discussion of the engagement of Ottoman architecture with the past, see Ousterhout, 'The east, the west, and the appropriation of the past'. On the development of Ottoman architecture before 1453, see Necipoğlu, *The Age of Sinan*, 71–103; Yürekli, 'Architectural patronage and the rise of the Ottomans'; Necipoğlu, 'Anatolia and the Ottoman legacy'.

<sup>7</sup> On this complex and its tiles, see Ayverdi, *Osmanlı Mi'marisinde Çelebi ve II. Sultan Murad Devri*, 46–118; Necipoğlu, 'From international Timurid to Ottoman'; Blessing, 'Seljuk past and Timurid present'. Veronika Poier (Harvard University) is currently completing a PhD thesis on the Yeşil complex. I would like to thank her for sharing some of her preliminary findings with me. For a broader perspective, see also Blessing, 'The blue-and-white tiles of the Muradiye in Edirne'.

To understand the history of Ottoman urbanism and architectural patronage before Constantinople, it is especially important to consider the role of the previous capitals of Bursa and Edirne, which continued to be important sites of imperial patronage even after 1453. The development of these towns must be considered not only in the context of the region's architectural history, but also more broadly of the early Ottoman state and the society it represented. The present contribution aims to shed light on the role of these former capitals and some of their main structures in early Ottoman history, by placing the research of architectural historians in a broader historical context. More specifically, I will focus on the first two Ottoman capitals, Bursa and Edirne, their place in Ottoman history, and their relationship to one another in the tumultuous half century prior to 1453. This will provide insight into how these former capitals were presented in some of the narrative accounts of the time, as well as the political and cultural significance of some of their most important royal foundations.

#### **Bursa and its multiple pasts**

A city is more than its buildings, and a larger question to be answered as part of any investigation of early Ottoman urbanism, its antecedents and its aspirations is what did it mean for a city to be the main Ottoman capital during the period in question. As early as the 1330s, the Moroccan traveller Ibn Battuta who travelled through Anatolia referred to the second Ottoman ruler Orhan as 'the Sultan of Bursa', while at the same time suggesting that no small part of Orhan's success came from the fact that he maintained a network of castles which he visited regularly.8 From this early remark, it is already clear that there was a main Ottoman capital associated with the ruler, but that it was also normal for the him to be frequently absent. This largely itinerant nature of the Ottoman ruler and his court, as well as the fact that there was nonetheless a main administrative centre which was initially Bursa, is confirmed by the testimony of Gregory Palamas, a Byzantine archbishop and intellectual who was captured in 1354 after the Ottomans took Gallipoli. Palamas spent most of his captivity in Nicaea, following an audience with Orhan and members of his court in a location somewhere in the mountains around Bursa.<sup>9</sup> Returning to Ibn Battuta, it is clear from his description that a large part of Bursa's importance came from the fact

<sup>8</sup> Ibn Battuta, *The Travels of Ibn Baṭṭūṭa*, 449–452.

<sup>9</sup> For Palamas' own written account of the meeting, see Philippidis-Braat, 'La captivité', 146–151. The most famous aspect of the encounter is a debate arranged by Orhan and recorded by his Byzantine physician, in which Palamas discussed matters of theology with a group of wise men called the *Chionai*  (text in Philippidis-Braat, 'La captivité', 168–185). The identity of these people has been the subject of scholarly controversy for some time; most recently, Ruth Miller has proposed a new theory that they may have come from Ilkhanid Iran (Miller, 'Religious v. ethnic identity', 40–41). While this has gained some acceptance in Ottoman circles, it is based on a weak understanding of the texts and Byzantine literary context. A detailed discussion of the evidence is beyond the scope of the present contribution, but will form the subject of a future study.

that it had large and well-provisioned markets, mentioned by the traveller alongside its famous hot springs.10 By this time, the town had probably already emerged as a western terminus of the overland trade routes from Iran and the east (the so-called Silk Road).11 Of course, it is also important to bear in mind that Ibn Battuta's account was only written down later in the fourteenth century, by which time the fame of the Ottomans and their capital city had increased, possibly colouring the traveller's earlier impressions. In the meantime, the dynasty had continued to invest in Bursa's markets and other buildings.

Particularly worthy of mention in Bursa are the Ottoman royal complexes(*ʿimāret/ külliye*),12 which in addition to their founder's purpose-built tomb (*türbe*) included a central multifunctional building and other structures serving social, religious and educational purposes (madrasas, hospitals, baths etc.). These complexes would take much grander form in Constantinople, beginning with the already mentioned mosque complex of Mehmed II the Conqueror (*fātiḥ*). Unlike Mehmed's foundation, however, the earlier complexes did not yet include Friday mosques (*cāmiʿ*), which were still treated at the time as a different, parallel type of construction. Before 1453, the central buildings of Ottoman royal complexes (*ʿimāret*) were clearly multifunctional, and similar in purpose to dervish convents (*zāviye, ḫānḳāh*) used for prayer, lodging, Sufi ceremonies and other social functions.13 In other words, the Ottomans were following precedents established by previous Muslim rulers of Anatolia, notably the Seljuks and Mongol-Ilkhanids. Over the course of the thirteenth century AD, despite the political turmoil of the period, members of the ruling classes of Muslim Anatolia had constructed many inns (*ḫan, kervānserāy*), madrasas, hospitals, hospices and dervish convents, which can still be seen today in Konya, Sivas, Erzurum, Amasya, Kayseri and other towns. These construction practices were continued by the emirates (*beyliks*) of western Anatolia into the fourteenth century, including the Ottomans, who expanded them to the region around Constantinople and into the Balkans.

The early Ottoman response to the pre-existing urban fabric of Bursa reveals examples of straightforward re-use and others of re-deployment of Byzantine architecture for new purposes. One of the first such re-deployments was the use of a Byzantine monastic complex in the city's citadel for the burial of, first, the founder of the dynasty, Osman, and later his son Orhan, who probably conquered the city in 1326 right after his father's death.14 The first Ottoman palace was also located in the

<sup>10</sup> Ibn Battuta, *The Travels of Ibn Baṭṭūṭa*, 450.

<sup>11</sup> For Ottoman Bursa's importance as a trading centre from the middle of the fourteenth century, see İnalcık, *An Economic and Social History*, 218–224.

<sup>12</sup> The term *külliye* is a neologism dating to the nineteenth century; in early Ottoman sources, the entire complex is generally called *ʿimāret* ('foundation'), a term often translated today as 'hospice' or 'soup kitchen'. See the discussion in Dark and Özgümüş, *Constantinople*, 87. In the passage already cited, Ibn Battuta mentions a hospice in Bursa, presumably that of Orhan (see below).

<sup>13</sup> Necipoğlu, *The Age of Sinan*, 49–50.

<sup>14</sup> Çağaptay, 'Prousa/Bursa', 52–62; Çağaptay, *First Capital*, 34–42, 60–61. It is also likely that Orhan's Friday mosque, famous for its 1337 inscription which has been the subject of much scholarly controversy,

citadel, although it no longer survives. It was probably located near the royal tombs, standing above the developing town, much as Mehmed II's first palace in Istanbul would later dominate the densely populated district of Tahtakale.15 As in the case of Mehmed II's later Fatih complex on the site of the Holy Apostles in Constantinople, this was sacred space already, albeit for a different religion. Under Orhan, this space would be extended down the slopes to the citadel's north-east, where Orhan constructed a socio-religious complex. Since Byzantine Bursa at the time of the Ottoman conquest was mostly confined to the area within the city walls, Orhan's complex would have benefitted from proximity to the royal tombs, while at the same time playing an important role in promoting the growth of the town outside the citadel walls. Its landscaping, perched on a hilltop incorporating the sloping terrain, has been seen as setting a precedent for later Ottoman royal complexes.16

Not unlike classical and late antique cities, the pious foundations of Orhan's son Murad I (r. 1363–1389) and grandson Bayezid I (r. 1389–1402) situated on hilltops offering panoramic views of the city and surrounding plains, served as the nucleus for royally sponsored urban development. Just as Orhan's complex just outside the acropolis and the royal tombs located within it came to function as the heart of the lower city, these complexes, which included purpose-built mausolea (*türbe*), came to function as the nuclei of new urban districts. The complex of Murad I developed on another hilltop, in the Kaplıca district, with its hot springs mentioned by Ibn Battuta and famous already from Byzantine times. However, his Friday mosque (known as Şehadet, 'martyrdom') is in the citadel in the older part of town. This suggests that he wished to make his mark on the central part of the city, while also being associated with the development of a new, more peripheral neighbourhood. The same is true for the architectural patronage of Bayezid I, who provided the city with its greatest Friday mosque (the famous Ulu Cami) in the old town and commercial centre, but also commissioned a *külliye* including his own tomb on a hilltop in a completely different, as yet undeveloped part of town.17 Apart from the buildings still visible today, namely the tomb, main multifunctional building on a 'T-type' plan, madrasa and bathouse, the complex originally also included a hospital, garden palace and fountain supplied by an aqueduct. Like Orhan's complex, Bayezid's thus seems originally to have functioned as a royally sponsored nucleus for religious, social and political life in a new part of the city.

In order to gain insight into the central role of Bursa in the early fifteenth century, it is worth taking a look at the Ottoman civil war or interregnum of 1402–1413, a

was a converted building located in the vicinity of these tombs. I would like to thank Suna Çağaptay for this observation, which is based on the testimony of Ibn Battuta, Evliya Çelebi, John Covel and other primary authors.

<sup>15</sup> On Mehmed II's first palace in Istanbul and the neighbourhood of Tahtakale (*taḥt al-ḳalʿa*, 'under the castle') see Kafescioglu, *Constantinopolis/Istanbul*, 22–24, 28–35.

<sup>16</sup> On Orhan's complex and palace in the context of the history of Ottoman architecture and Bursa's development, see Pancaroğlu, 'Architecture, landscape, and patronage', 42–43.

<sup>17</sup> On the complex of Bayezid I see Necipoğlu, *The Age of Sinan*, 49–51.

period of instability following Timur's defeat of Bayezid at the Battle of Ankara in 1402, which put an end to the first Ottoman attempt at empire. In the ensuing dynastic wars and relations with foreign powers, Bursa played a pivotal role, leaving no doubt as to the fact that it was still perceived as the Ottoman capital.18 One valuable source on the importance of the city to the Ottoman princes competing for the throne is the tomb of the defeated sultan Bayezid I, their father over whose legacy they competed. An examination of its history serves to demonstrate the extent to which royal authority at the time revolved around Bursa as a whole, and in particular Bayezid's socio-religious complex, where the late ruler had intended to be buried. In addition to the mausoleum and its inscription, evidence for this competition is provided by a contemporary epic account of the dynastic wars from the perspective of the winner, Mehmed I (r. 1413–1421).19 According to this account, in the immediate aftermath of Timur's departure from Anatolia in early 1403, when he left behind the corpse of Bayezid who had died in captivity, all four of the deceased Ottoman ruler's sons still capable of making a bid for the succession became involved in a struggle to take control of Bursa. In so doing, they clearly intended to show that they were each worthy to succeed him as 'the Sultan of Bursa' (to use Ibn Battuta's phrase). Following his victory in 1402, in an attempt to keep the Ottomans weak so he could turn his attention to other matters, Timur had burnt and pillaged the Ottoman capital Bursa, the seat of Ottoman sovereignty and central administration, then allowed two different Ottoman princes to claim authority over the city in his name. As was his intention, this led to outright civil war. The main outlines of this civil war or interregnum (*fetret devri*) will be sketched out briefly below, with a view to underestanding the place of Bursa as well as Edirne in the Ottoman struggles for the throne.

When Timur departed in spring of 1403, Bursa was in the hands of a prince named Isa, an older brother of the eventual winner Mehmed I (r. 1413–1421). However, the prince who looked most likely to succeed Bayezid I was another older brother, Emir Süleyman (d. 1411), who had taken refuge after the battle with key members of Bayezid's court and army on the European side of the straits, away from the threat of Timur and his nomadic armies. At the time, Mehmed was only fifteen, and was in a weaker position than his brothers, since he lacked authority over the more central parts of the Ottoman domains. However, he and his army were able to assert themselves in the area around the northern Anatolian town of Amasya, which Mehmed had previously governed with the help of his tutor and princely court. Finally, a fourth prince, Musa, was still a minor and had been captured in the battle of Ankara with his father. When Bayezid died in captivity, Timur left Musa with his corpse and a diploma of appointment over Bursa, in the custody of a rival ruler, the Turkish emir of Germiyan. As is clear from the epic account of Mehmed I's

<sup>18</sup> The following discussion of these events is based on Kastritsis, *Sons of Bayezid*, 79–110, which contains a detailed description of the sources and scholarly literature.

<sup>19</sup> See Kastritsis, 'The historical epic *Ahvāl-i Sultān Mehemmed*'. A full translation is available in Kastritsis, *An Early Ottoman History*, 97–151.

exploits written in his court, as well as a diplomatic document, Mehmed was able to make an alliance with the ruler of Germiyan, through which he gained possession of his father's corpse and the young prince Musa. In this way, he was able to challenge Isa's control of Bursa. Following a battle with his brother outside the city, Mehmed was able to drive out his brother and enter Bursa, where he held elaborate funeral ceremonies for his father, followed by enthronement ceremonies for himself. However, Mehmed's success prompted the intervention of Süleyman, who took control of Bursa and used it to cement his own claims. Although Mehmed had already buried Bayezid, his tomb was still unfinished, so Süleyman was able to complete it and take credit on the inscription.

From the above, it is clear how important Bursa still was for Ottoman legitimacy following the defeat by Timur in 1402. Eager to assert their claims to the succession, three sons of the defeated Bayezid I fought each other for what was left of his political legacy: an incomplete attempt at empire centred around Bursa, still perceived as the Ottoman capital. However, despite the continued importance of Bursa throughout the interregnum, during this time Edirne (Adrianople) began to be treated for the first time as an alternative capital by princes whose power did not extend to the Anatolian side of the straits. In 1402, Edirne, which like Bursa had an important Roman-Byzantine past and had long functioned as the main Ottoman centre on the European side of the straits, became the seat of the Ottoman prince Süleyman. It is clear that his intention was always to expand his power to Anatolia and rule from Bursa, and it was not long before he was able to achieve this aim. The timing of Süleyman's capture of Bursa is unclear, but should probably be dated to late 1403, or possibly a year later.20 Süleyman's inscription on Bayezid's mausoleum is dated early 809 in the Islamic calendar (late 1406), but this may well refer to the final stages of the building's construction.21 In any case, various sources suggest that after occupying Bursa, Ankara and other Anatolian cities, Süleyman remained mostly in Anatolia until 1410, when he was forced to return to Europe to face his brother Musa, who had earlier invaded his territory from the north. Musa's bid for power was the result of an alliance between Mehmed, in whose custody Musa had remained until that time, and a number of neighbouring powers threatened by Süleyman. It had the desired result, since Mehmed was able to take advantage of Süleyman's departure and recapture Bursa. As for Süleyman, having lost control of Ottoman Anatolia, he was once again forced to make Edirne his capital. After a series of confrontations with Musa, which ended when Musa surrounded Edirne and most of Süleyman's court and army deserted to him, Süleyman was forced to flee the city and was killed in early 1411.22 Thus Musa became the ruler of Ottoman Rumelia and its capital until 1413, when Mehmed was finally able to overthrow and kill him, becoming sole Ottoman ruler until his death in 1421.

<sup>20</sup> Kastritsis, *Sons of Bayezid*, 112–113.

<sup>21</sup> Kastritsis, *Sons of Bayezid*, 99–100.

<sup>22</sup> Kastritsis, *Sons of Bayezid*, 153–158.

It is important to note that despite Edirne's prominent role in the careers of Süleyman and Musa, after Mehmed was able to eliminate the threat posed by their rival courts, he probably continued to treat Bursa as his main capital. Written sources suggest that Bursa was burned no fewer than three times during the period 1402–1413: first by Timur in 1402; then by Isa in retaliation against Mehmed after he lost Bursa, probably in 1403; and finally, by the neighbouring principality of Karaman, a major enemy of the Ottomans which took advantage of the conflict in Rumelia between Mehmed and Musa to attack it in 1413. According to the Byzantine chronicle of Doukas, the Karamanids even went so far as to exhume and burn the corpse of Bayezid I, which as we have seen played such an important role in the dynastic claims of both Mehmed I and his older brother Süleyman.23 All this leaves no doubt that in 1413 Bursa was still perceived as the Ottoman ancestral capital, where members of the dynasty were to be buried. Further evidence is provided by the fact that when Musa was killed by Mehmed in Rumelia, ending the interregnum, his corpse was sent for burial across the straits to Bursa, ending Karaman's attack on the town. Like Mehmed I, whose 'green' (Yeşil) funerary socio-religious complex is among the most famous monuments in Bursa today, Murad II, under whose rule Edirne definitively became the main Ottoman capital, nonetheless chose to be buried with his ancestors in Bursa in a foundation he also constructed there, a practice that continued for lesser members of the Ottoman dynasty after Mehmed II and his successors chose instead to be buried in Constantinople. His complex, the Muradiye, is still visited today for the large number of Ottoman tombs it contains.

## **When did Edirne become the main Ottoman capital?**

We have seen that although Bursa was still thought of as the main Ottoman capital during the interregnum of 1402–1413 and subsequent reign of Mehmed I (r. 1413–1421), this was also the time when Edirne had first emerged as an alternative capital, at least for princes without access to Bursa. Sometime during the long reign of Murad II (1421–1444, 1446–1451), Edirne finally surpassed Bursa, so that on the eve of the 1453 conquest of Constantinople it was considered the Ottoman capital.24 However, is is difficult to determine at exactly what point the transition took place. For the first two years of Mehmed II's second reign (1451–1481), as well as his previous brief reign as an adolescent (1444–1446), there is no doubt that the capital was Edirne. Morover, even after 1453, the transition to Istanbul was not immediate and Edirne would continue to be used as a second capital for centuries to come.25 In addition to the two palaces he built in Istanbul, Mehmed also completed the Edirne palace begun by his father Murad II around 1450, which like the Topkapı Palace in Istanbul would be expanded in the sixteenth century and added to until the end of the seventeenth.26 All this leaves

<sup>23</sup> Doukas, *Decline and Fall*, 115 (chapter 21.1).

<sup>24</sup> Singer, 'Enter, riding on an elephant', is largely an attempt to answer the question of when Edirne became the Ottoman capital. On this question, its conclusions are broadly in agreement with my own.

<sup>25</sup> Kafescioğlu, *Constantinopolis/Istanbul*, 16–18.

<sup>26</sup> Kontolaimos, 'A landscape for the Sultan', 21; Özer, *Ottoman Imperial Palace*, 7–12, 48–51.

no doubt that on the eve of 1453, Edirne had become the main Ottoman capital. But the nature of a capital city for the Ottoman state of the time is by no means clear, any more than the precise moment at which Bursa ceded its place to Edirne.

We have seen that a key moment in this process was the Battle of Ankara (1402) and the ensuing period of civil wars, which had special significance for both Bursa and Edirne. It was at this time that it first became apparent that Bursa was unsuitable to be the Ottoman capital, due to its vulnerability to attack by other Muslim rulers. In this sense, Edirne came for a time to resemble a safe haven, since Timur and other Muslim enemies never crossed the straits to Europe. However, this situation would not last, because the Ottomans also had Christian enemies in Europe. That is why the first action of Emir Süleyman and his officials after crossing the straits in 1402 had been to negotiate a peace treaty with Byzantium, Venice and other local Christian powers. The vulnerability of Edirne to attack from the north would become fully apparent in the 1440s, when Hungary would lead military campaigns into the Balkans and even threaten Edirne itself (see below). However, earlier in the century the threat must have appeared more remote, especially compared with the ever present danger posed by Karaman and other Muslim enemies in Anatolia.

Be that as it may, notwithstanding the continuing importance of Bursa in the early fifteenth century, at that time Edirne began to function as an alternative Ottoman capital in its own right. But there was an essential tension to be overcome before Edirne, known as 'the abode of the gazi raiders' (*dār al-ġuzāt*), could replace Bursa as 'the abode of sovereignty' (*dār al-mülk*, *i.e*. the main administrative capital).27 This tension was inherent in the very nature of the Ottoman state and its expansion until that time, in which the newly conquered territories in Rumelia had played a key role. It was only by making peace with the Ottomans' Christian enemies there in 1403 that Süleyman had been able to then cross the straits to Anatolia and take control of Bursa. But as we have seen, his power was ultimately undermined by his brother Musa, who was able to invade Rumelia from the north. This was accomplished with the support of several Muslim and Christian powers threatened by Süleyman's success, but crucially also of the gazi raiders of Rumelia and other military elements there, for whose livelihood the resumption of warfare against Christians was essential. In order for Edirne to become the main capital of a centralised Ottoman state and its court, it was first necessary to overcome the opposition of such local elements to the growth of courtly authority in their home region. This growth entailed all the trappings of the central state and empire-building: a palace with a treasury and central taxing bureaucracy; control over marcher lords and other local power brokers; and an elite standing army loyal to the ruler, among other things. As demonstrated by well-known

<sup>27</sup> The term *dār al-mülk* is used for Bursa in the endowment deeds of both Bayezid I and Mehmed I for their complexes there. I would like to thank Veronika Poier for this observation. As for Edirne's place as 'abode of the gazi raiders' (*dār al-ġuzāt*), the epic-hagiography *Saltuḳnāme* compiled in the late fifteenth century uses this term, as well as providing a sense of how Edirne was perceived by the raiders of Rumelia at the time. For an English translation of a relevant passage, see Karamustafa, 'Sarı Saltık', 140–142. See also Kafadar, *Between Two Worlds*, 148–149.

passages from a body of critical texts known as 'the Anonymous Chronicles'(*Chronicles of the House of Osman*), this was far from a smooth process, and one that took over a century to complete.28

Such resistance notwithstanding, during the long and eventful reign of Murad II (r. 1421–1444, 1446–1451), slowly but surely Edirne came to replace Bursa as the main Ottoman capital. This is apparent from contemporary historical narratives, which discuss Murad's investment in the city and its surroundings. The following is a passage that forms part of the comprehensive Ottoman history known as Oxford Anonymous, not to be confused with the previously mentioned Anonymous Chronicles:

The pious foundations of the Sultan of Islam and of the Muslims, Sultan Murad, know no bounds. One of the many is the bridge over the Ergene river, in what had previously been a forest. In the winter, the place would become muddy and the Muslims would be unable to pass. It would become a lair for brigands, and much evil would take place. So Sultan Murad took great pains to clear the forest and build a large bridge there, founding towns on either side as well as a hospice [ʿ*imāret*]. He went there in person to cook the hospice's stew and distribute it to the poor, granting many other favours as well. Moreover, in Edirne he built a large Friday mosque and two madrasas, one for the study of the hadith. There he also built a hospice [*ʿimāret*] and a lodge for dervishes of the Mevlevi order. He assigned *waqf* property [charitable endowments] to these, so that every day his bounty reaches many poor people. Furthermore, in Bursa he built a hospice [*ʿimāret*], and next to that a madrasa and a mausoleum intended as his own resting place. His pious foundations are without end.<sup>29</sup>

From this description, it is clear that Murad II invested in both cities, but that the nature of his investment was different in each one. In the case of Edirne, his aim was to build essential infrastructure for further growth, much as Orhan, Murad I and Bayezid I had done for Bursa. He also aimed to improve communications in a previously underdeveloped region, as one might expect since Edirne was in the process of becoming the main Ottoman capital. The Uzunköprü bridge was constructed in the years 1426–1443, exactly the period when the transformation of Edirne seems to have taken place. The fact that Murad built two madrasas in Edirne and only one in Bursa also points in this direction. However, that he chose Bursa to build the complex that included his own tomb would suggest that the city still retained its significance as ancestral capital and resting place of the Ottoman dynasty.

<sup>28</sup> For these passages see Giese, *Die altosmanischen anonymen Chroniken*, 21–33. Translations of some of them into English may be found in Lewis, *Islam*, 135–141, 226–227, though the translations are not entirely accurate.

<sup>29</sup> Oxford University, Bodleian Library, MS Marsh 313, ff 120r-120v. Tr. Kastritsis, *An Early Ottoman History*, 170: 'Sulṭānü 'l-İslām ve 'l-Müslimīn Sulṭān Murād'uñ ḫayrātı bī-ḥaddür. Cümlesinden biri Ergine köprüsi'dür, ki evvelde bir orman olup, ḳışın çamur olup Müslümānlar geçmeyüp, ḥarāmī ṭuraġı olup ziyāde fesād olurdı. Sulṭān Murād ormanı zaḥmet ile giderüp, bir büyük köpri yapdı, ve iki ṭarafına şehirler yapup bir ʿimāret yapdı. Gendü üzerine varup, aşın bişirüp fuḳarāya ülüşdürdi, ve nice inʿāmlar itdi. Andan Edrene'de bir ulu cāmiʿ ve iki medrese—biri dārü'l-ḥadīs̱—ve bir ʿimāret ve bir Mevlev<ī>ḫāne yapup, evḳāf taʿyīn itdi, her gün nice faḳīre niʿmeti yetişür. Brusa'da daḫı bir ʿimāret ve anuñ yanında bir medrese yapdurup, ve gendü āsūde olmaġ içün bir türbe yapdurdı. Anuñ ḫayrātı bī-nihāyedür'.

Needless to say, Friday mosques were an essential part of urban growth in both Bursa and Edirne. In Bursa, this need was fulfilled by the Friday mosques mentioned earlier, especially Bayezid I's large congregational mosque (Ulu Cami). In Edirne, no fewer than three of Bayezid's sons played a role in constructing the town's first large congregational mosque, which would later come to be known as the Old Mosque (Eski Cami). The building's construction closely follows the development of the city during this time, as noted twice in the anonymous *Chronicles of the House of Osman*:

Emir Süleyman ruled for seven years. Then Musa Çelebi became ruler. He stayed in Edirne, where he laid the foundations of the Old Mosque [Eski Cami]. He had the building constructed to the point where it came above the ground.

[…]

After [suppressing the Şeyh Bedreddin revolt in 1416], Sultan Mehmed [I] went and stayed in Bursa. And in Edirne, Emir Süleyman had begun [work on] the Old Mosque [Eski Cami]. Then Musa Çelebi had raised it above the ground. So in the end, it fell to Sultan Mehmed's lot to complete it. He also built a hospice [*ʿimāret*] in Bursa. He was the first to begin the palace in Edirne. He built it, and after that the padishahs resided there. The prominent lords [*beg*] resided in Bursa. At that time, there were no houses in Edirne outside the fortifications. After that, everything outside them also became part of the city. In Edirne, Sultan Mehmed attained the command of God [*i.e*. he died]. May God's mercy be upon him.30

From the above, it is clear that upon Mehmed I's death in 1421, in addition to a large Friday mosque, Edirne already had a functioning palace, which was used and possibly expanded by Mehmed I.31 Although the above text creates the impression that the entire palace was first constructed by Mehmed, in fact it must already have existed by the late fourteenth century. Already under Murad I (r. 1262–1289), Edirne had become the European capital of the Ottoman dynasty, alongside Dimetoka which had long functioned as a royal residence and stronghold for the treasury.32 This 'old' Edirne palace had probably developed gradually from military installations, since by

<sup>30</sup> Giese, *Die altosmanischen anonymen Chroniken*, 49, 55, my translation: 'Emīr Süleymān yedi yıl beglik itdi. Andan ṣoñra Mūsā Çelebi beg oldı. Edrene'de ḳarār itdi. Eski Cāmiʿi bünyād bıraḳdı, tā yir yüzine çıḳınca bināsı yapdurdı … Andan ṣoñra Sulṭān Meḥemmed varup Burṣa'da ḳarār itdi. Ve Edrene'de Eski Cāmiʿi Emīr Süleymān başladı, Mūsā Çelebi yirden yuḳaru ḳaldurdı, āḫırı Sulṭān Meḥemmed'e naṣīb oldı, ol tamām itdürdi. Andan Burṣa'da daḫı bir ʿimāret yapdurdı. Evvel Edrene'de ol sarāy başladı, düzdürdi, andan ṣoñra pādişāhlar Edrene'de durur oldılar. İlerü gelen begler Bursa'da ṭururlardı. Ol vaḳit Edrene'nüñ ḥiṣārından ṭaşrasında evler yoġdı. Andan ṣoñra ṭaşrası daḫı hep şehir oldı. Andan ṣoñra Sulṭān Meḥemmed Edrene'de Allāh emrine vāṣıl oldı, Raḥmetullāhi ʿaleyh'.

<sup>31</sup> Amy Singer has suggested that there may have been no palace in Edirne under Mehmed I, based on the fact that it does not appear in the chronicle of Aşıkpaşazade; she points out that Murphey has dismissed the mention of a palace under Murad I as a 'phantom appearance' in the sources (Singer, 'Enter, riding on an elephant', 96; Murphey, *Exploring Ottoman Sovereignty*, 50). However, despite its undeniable importance, Aşıkpaşazade's work is but one available account. Also, Murphey does not so much dismiss the existence of the earlier palace as point out that not enough is known about it, and that nothing existed on the scale of the later Edirne palace begun around 1450.

<sup>32</sup> On the 'old' Edirne palace, its location in the town, and the role of Dimetoka as royal residence, see Kontolaimos, 'A landscape for the Sultan', 20; Zachariadou, 'The sultanic residence and the capital'; Bessi, 'The topographic reconstruction of Ottoman Dimetoka', 47.

this time the town was also the main centre for training the Janissaries. Its location was apparently outside the city walls to the north-east, where Sinan's famous Selimiye mosque may still be seen today. It still existed in the second half of the seventeenth century, when it was described by Evliya Çelebi, but since nothing remains today its precise character and extent are unclear. In at least some respects, however, it must have served as a model for the more extensive 'new' Edirne palace begun by Murad II, about which much more is known.33 Murad II also built another Friday mosque in the centre of the city, so that the old one came to be known as the Old Mosque (Eski Cami) and the new one as the New Mosque (Yeni Cami). Murad II's mosque is known today as Üç Şerefeli ('three balconied') because of the inclusion of a minaret with three balconies. In addition to the minaret, which is one of four probably reflecting Timurid inspiration, the plan shows the growing influence of Byzantine architecture, in that it includes a large dome and half dome. This has been seen as a stage in the formation of classical Ottoman architecture, and shows the influence of Byzantine architecture even before the conquest of Constantinople and its famous cathedral of Hagia Sophia.34

For our purposes, Murad II's choice to construct in Edirne a new palace as well as a second Friday mosque with two attached madrasas, one for studying the hadith as would later be the case later with the Süleymaniye in Istanbul, all point to a conscious decision to make Edirne the primary Ottoman capital. Of course, Murad's patronage would also produce the Muradiye complex in Bursa, renowned today as the resting place not only of Murad himself but of also of other members of the Ottoman dynasty. The main building's construction on a 'T-type' plan suggests that with this complex, Murad intended to follow in the footsteps of his ancestors who had also built such structures in Bursa. It is the last royal complex built on such a plan, following Murad's earlier construction in Edirne of a convent mosque on the same 'T-type' plan, originally intended as a Mevlevi lodge but later converted to a Friday mosque.35 Such foundations remained popular in Rumelia even after 1453, albeit with non-royal patrons, implying that they had become associated with an earlier Ottoman culture still revered by Rumelian Muslims. In any case, it is clear that by the end of Murad's reign we see the transition of Bursa from Ottoman capital to ancestral city of the dynasty, where the tombs of its earlier members were located and it was appropriate for later princes also to be buried. Further evidence that by this time Edirne had become the main Ottoman capital is provided by the complex events surrounding Murad's decision to abdicate in favour of his son Mehmed II. It was 1444, and by this time the importance of Europe for the survival and advancement

<sup>33</sup> On the 'new' Edirne palace and its structures, especially the famous Cihannüma Kasrı ('world-surveying tower') which shows Byzantine and Latin influence, see Özer, *Ottoman Imperial Palace*; Kontolaimos, 'A landscape for the Sultan'; Arel, 'Cihannüma Kasrı'.

<sup>34</sup> On the Üç Şerefeli and Murad II's other monuments in the context of Ottoman architectural history, see Necipoğlu, *The Age of Sinan*, 79. See also Tanman, 'Ekrem Hakkı Ayverdi'nin'.

<sup>35</sup> This is known today as the Muradiye of Edirne. See Blessing, 'The blue-and-white tiles of the Muradiye in Edirne'.

of the Ottoman state had grown, not just because the territories newly conquered for Islam were beyond the claims and military reach of other Muslim dynasties, but also because the threat posed by Christendom had increased and had shifted to the forefront of Ottoman concerns. The reasons for Murad II's abdication in favour of Mehmed are still not entirely clear; however, he was probably motivated at least in part by concerns over the succession. At the time, there was still an Ottoman pretender living in Byzantine Constantinople, and succession had been a problem of the Ottoman state for more than forty years.36

Whatever his motivation, when Murad chose to retire in 1444, he placed his son Mehmed on the throne in Edirne and retired to the Anatolian city of Manisa, where Mehmed had previously been governor. According to the history of Aşıkpaşazade:

Then Sultan Murad Khan Gazi came to Edirne. He said to [Çandarlı] Halil Paşa, 'I will place my son on the throne. For I have conducted many campaigns of holy raiding, and would now like to see during my own lifetime what kind of padishah my son will become.' So straight away, he ordered that they bring over from Manisa his son, namely Sultan Mehmed Gazi.37

The above passage comes between two others in the same history concerning Murad's ordering of the affairs of Rumelia. All of this points to the fact that by this time, Rumelia had become the main focus of Ottoman politics. To be sure, Ottoman interests in Anatolia were always threatened by Karaman and other enemies, a situation Murad presumably hoped to control from Manisa with the help of subordinates in other towns. In this connection, it is worth noting that Aşıkpaşazade also presents the ruler of Karaman as corresponding with the Hungarian king, suggesting common action against the Ottomans in Edirne.38 Although, like all histories of the period, Aşıkpaşazade's account had its own agendas, there is little doubt that by the middle of Murad II's reign Ottoman strategic use of evolving cities to promote their claims to sovereignty had taken a new turn and Edirne had become the main theatre of action as far as the Ottoman state was concerned. But we would be misguided to propose a straightforward progression with the ascent of one city and the regress of another. The geographical context of foreign policy and internal power play expressed through cities was complex and shifting, rather than linear. Prior to his abdication, Murad had reached an accommodation with Serbia and Hungary, which he apparently considered adequate for securing his son Mehmed's position in Edirne. Murad was proven wrong almost immediately, when the Hungarians crossed the Danube and attacked deep into Ottoman territory, so that the threat was felt even as far south as Edirne. As a result, Murad was forced to leave his retirement in Manisa and lead the Ottoman armies at

<sup>36</sup> On these complex events, the best discussion is still İnalcık, *Fatih Devri*. 37 Aşıkpaşazade, *Âşıkpaşazâde Tarihi*, 174, my translation: 'Andan ṣoñra Sulṭān Murād Ḫān Ġāzī daḫı Edrene'ye geldi. Ḫalīl Paşa'ya eydür "Ben oġlımı taḫta geçürürin" dir. "Ben ḫayli ġazā seferlerin itdüm. İmdi benüm oġlum daḫı benüm ḥayātumda göreyüm ne ṣūretilen pādişāh olur" didi. Hemān oġlını Maʿnisa'dan getürtdi kim ol Sulṭān Meḥemmed Ġāzī'dür'.

<sup>38</sup> Aşıkpaşazade, *Âşıkpaşazâde Tarihi*, 174.

Varna. The Council of Ferrara-Florence (1438–1439) had resulted in unified Christian action. An anonymous Ottoman account of the events surrounding Varna makes it clear that knowledge of this unified Christian front was circulating in Ottoman society, which makes Murad's choice to abdicate his throne and retire to the quiet western Anatolian town of Manisa all the more surprising.<sup>39</sup> In fact, he returned from retirement not once but twice: first in 1444 to face the Hungarian threat, then again in 1446, when events and court politics suggested that the young Mehmed was not yet suitable for the throne.

From our perspective, what is again significant about these events is the shift of Ottoman concerns to Edirne, where trenches were dug in 1444 to resist a possible Hungarian attack.40 As the anonymous account makes clear, the threat was such that the viziers and other magnates in Edirne had moved their property to the better fortified Dimetoka (Didymoteichon). According to this narrative, they justified their action to Edirne's residents by saying that they were men of affairs, so their situation was different from that of the town's residents; they may be required to carry out their duties outside Edirne, in which case they would require a treasury. However, the anonymous author is sceptical of their motives and adds that in fact they just wished to protect their property.41 The account of the Burgundian Bertradon de la Broquière, who travelled through the region in 1433, also suggests that around this time the stronghold of Dimetoka was the location of the Ottoman treasury.42 In fact, this had long been the case, as shown by the research of Elizabeth Zachariadou and Ourania Bessi.43 In any case, it is clear from both the Ottoman account and that of Bertrandon that by the 1430s and 1440s, Edirne had come to occupy the place previously held by Bursa as the location of the Ottoman central administration, and an obvious target for enemy invaders. A further indication of Edirne's prominent status at the end of Murad II's reign is Mehmed II's marriage to a princess from the eastern Anatolian Turcoman principality of Dulkadır. This diplomatic union, which took place in 1449 after Murad's final return to the throne, followed a precedent established by Mehmed I (d. 1421), who had also married a princess from Dulkadır.<sup>44</sup>

We have seen that toward the end of his reign, Murad II had begun constructing the new Edirne palace, which survives in ruined form even today. It was still the

<sup>39</sup> Anonymous, *The Holy Raids of Sultan Murad Son of Sultan Mehmed Khan*, see Imber, *Crusade of Varna*, 42–44, 47–51.

<sup>40</sup> For a detailed account of these events, including a translation of the source in question, see Imber, *Crusade of Varna*, 5–30 (esp. 26–27).

<sup>41</sup> Imber, *Crusade of Varna*, 79–80.

<sup>42</sup> Broquière, *Le voyage d'Outremer*, 172–173, 180, 186–188.

<sup>43</sup> Zachariadou, 'The sultanic residence and the capital'; Bessi, 'The topographic reconstruction of Ottoman Dimetoka', 47.

<sup>44</sup> Thanks to a set of Byzantine paintings commemorating the event, we are able to picture Mehmed II's nomadic bride riding into Edirne on an elephant. For a detailed discussion of the wedding and Byzantine image, see Singer, 'Enter, riding on an elephant', 100–106. For Mehmed I's earlier marriage alliance with Dulkadir, see Kastritsis, *Sons of Bayezid*, 107, 188; Kastritsis, *An Early Ottoman History*, 125–126, 146.

main Ottoman palace until Mehmed II constructed others in his new capital, and continued to be used later as an alternative to the Topkapi palace in Istanbul, for which it had served as a model.45 Thanks to photographs from the nineteenth century, when it was still standing, as well as more recent excavations, it is possible to know a fair amount about its layout and the function of its main buildings. However, we have seen that there was already an older palace in Edirne. This was the palace still in use during the first half of the fifteenth century, and would come to be known after the construction of the new palace (Sarāy-i Cedīd) as 'the old palace' (Sarāy-ı ʿAtīḳ). It is this palace that was visited by Bertrandon de la Broquière in 1433, and that also appears in the various narratives describing the struggles for Edirne during the civil wars of 1402–1413. In the years preceding the final confrontation between Emir Süleyman and Musa, which ended with Süleyman's death in early 1411, Musa had established his control in Ottoman Rumelia and fortified the palace, but was later ousted by Süleyman who took up residence there once again. Eventually, Musa was able to drive his brother out of Edirne by winning over his military commanders and officials. But a few years later, Musa himself was forced to flee when his older brother Mehmed took control of the surrounding region (1413). We saw that the anonymous *Chronicles of the House of Osman* present Mehmed I as the first Ottoman ruler 'to stay in Edirne', and claim that it was he who began construction of the Edirne palace. However, this source as it has come down to us was compiled later, at a time when the reigns of Mehmed's brothers Süleyman and Musa were not recognised as legitimate by Ottoman historiography. Most probably, the old palace of Edirne existed already under Murad I in the 1380s. As Colin Imber has suggested, after the sack of Bursa by Timur in 1402 it may well have surpassed any royal residence that still existed in Bursa.46 Nevertheless, at least in some cases the function of the palace in this period of Ottoman history was probably as much military as courtly; given the importance of Edirne for Ottoman expansion into Rumelia, even before 1402 the Edirne palace may well have surpassed the one in Bursa in size, if not in significance. As discussed earlier with regard to Bursa's role in the dynastic wars of the Ottoman interregnum, even in the absence of a substantial palace a town could still be thought of as the main Ottoman capital on historical and cultural grounds.

To conclude, we have seen that in the period before the conquest of Constantinople Ottoman notions of a capital city were still fluid and very different from the situation post-1453. Already under Orhan in the mid-fourteenth century, as suggested by Ibn Battuta's visit and other evidence, Bursa was being treated as the main capital of a largely itinerant court. Bertrandon de la Broquière's account from a century later suggests that Edirne was not so different under Murad II, who like his predecessors was largely an itinerant ruler. Nonetheless, as the Ottoman state developed into an

<sup>45</sup> Özer, *Ottoman Imperial Palace.* Although Özer's book is a useful resource on the later, more famous Edirne palace, including some old photos and archeological data from recent work on the site, the author does not provide citations for his claims about the earlier palace.

<sup>46</sup> Imber, *Ottoman Empire 1300–1650*, 144.

empire over the second half of the fourteenth century, it stands to reason that Bursa's status as capital of the enriched and empowered Ottoman sultanate would also have increased, giving the city the status of capital or 'abode of sovereignty' (*dār al-mülk*, the term used in Bayezid I and Mehmed I's endowment deeds). Evidence of Bursa's enduring importance can be found in the many royal construction projects there, which continued under Murad II even as Edirne came to replace Bursa as the main Ottoman capital. Here it is also worth mentioning the funerary complex in Bursa associated with the important Sufi Emir Sultan of Bukhara, who married into the Ottoman family and is still viewed today as the patron saint of the city. Although the mosque itself probably began life as a Sufi lodge (*tekke*) in the fourteenth century, and no longer survives in its original form, it was further developed during this time to include a mosque and the holy man's tomb, who died under Murad II.47

When discussing Ottoman urbanism prior to the conquest of Constantinople, it is also important to bear in mind other cities that functioned at one or another time as Ottoman provincial capitals, or otherwise important centres. These included Amasya, Mehmed I's provincial capital and main power base during the Ottoman civil war of 1402–1413; nearby Tokat, whose fortifications functioned as a military base and prison for Ottoman grandees and diplomatic hostages; and other strongholds and provincial capitals in Anatolia and the Balkans, such as Ankara, Kütahya, Manisa, Sofya and Serres (Siroz), to name only a few. The types of buildings discussed above – Friday mosques and socio-religious complexes – were also constructed in some of these places during the same time, and played a key role in Ottoman urbanism during the period in question. There would also have been palaces in the various provincial capitals, such as Birgi, Manisa and Ayasoluk, which the Ottomans either took over from previous dynasties or constructed themselves. A comprehensive discussion of these is beyond the scope of this paper, which has primary concerned itself with the role of Bursa and Edirne as Ottoman capitals in the century before 1453. Although Bursa's importance continued into the fifteenth century, by the middle of Murad II's reign (*c*. 1430) Edirne was well on its way to replacing Bursa as the main Ottoman capital. The repeated sacks of Bursa during the troubled period 1402–1413, as well as the use of Edirne as capital by several Ottoman princes during the same time, were probably both factors in these developments, as was the shifting focus of Ottoman politics to Europe during Murad II's reign. As a result, on the eve of the Ottoman conquest of Constantinople in 1453, Edirne had emerged as the main capital of an empire now bent on championing the cause of Islam against the threat of unified Christian action. Thanks to Murad II's 'New Palace' and Üç Şerefeli mosque in Edirne, as well as the other infrastructure built there in the first half of the fifteenth century, Murad's successor Mehmed the Conqueror would have ready inspiration for rebuilding Constantinople, once he had finally conquered it and decided to make it the capital of his own empire.

<sup>47</sup> Tanman, 'Emîr Sultan Külliyesi', 149; Algül and Azamat, 'Emîr Sultan', 147.

## **Bibliography**


## Chapter 15

## New history for old Istanbul: Late Ottoman encounters with Constantinople in the urban landscape

*Göksun Akyürek*

## **Making of a modern monument: The Column of Constantine re-appropriated**

The honorific column of the emperor Constantine, preserved on its original site and renamed Çemberlitaş (meaning literally 'the banded stone'), stands today beside the tram stop on Divanyolu Street, a humble observer of the busy daily routine in the Çemberlitaş neighbourhood that bears its name (Figure 15.1). Because the city has changed dramatically in the millennium and a half that followed its erection, the banded column seems to be the only remaining trace of the city's founder-emperor. Despite changes in its role and identity, it has stood for seventeen centuries, its timeworn appearance the result of numerous alterations and repairs in the Byzantine, Ottoman and Republican eras.<sup>1</sup> The column has been kept intact and visible, serving as a link to the past in multiple ways, supported by myths which have helped preserve it as a cryptic presence in the urban imagery. The mid-nineteenth-century rediscovery of the column as part of the city's ancient legacy started the process by which the 'banded stone' became a modern historic monument, with new meanings for an expanding audience. This process has had its own complexities, entailing a dynamic relationship between the column and its evolving urban context. This chapter aims to discuss this process as a parallel to, or even test case for, the relationship between the modern Ottoman city and its ancient past, newly emerging and contested in the aftermath of the Tanzimat reforms. I also point to a shift in the column's relationship to its urban surroundings triggered by the incorporation of the Column of Constantine into the new setting of Divanyolu Avenue. Through this recontextualisation, its former juxtaposition with structures and spaces created in different times, which

<sup>1</sup> Ousterhout, 'The life and afterlife of Constantine's column', 304–326; Sönmez, 'Osmanlı'da Bürokratik Tartışmalar', 1–26.

*Fig. 15.1: Present day Çemberlitaş (author).*

could usefully be called palimpsestic, was exchanged for a sequential showcase of monuments of various ages.

First, a brief history of the column. In AD 324–330 the column was originally set up in the centre of Constantine's (unusually) round-shaped forum, at the point where the new city of Constantine met the older city of Byzantion.2 The Column of Constantine was originally 36 m tall with a gilded bronze statue of Constantine (around 8 to 10 m high) standing on its top. It had a tall base supporting seven solid drums of porphyry, each 2.90 m in diameter, the joints banded with laurel wreaths. Sited on the crest of a hill, it was tall and highly visible, in keeping with the prominence which served to identify the early city with its founderemperor.3 It was also a significant site for civic rituals and religious ceremonies for

many centuries. Because of its pagan connotations the statue of Constantine had an uneasy relationship with its audience in subsequent centuries. The statue fell during a windstorm in 1105 or 1106 and a cross was erected instead, but this did not last either. Robert Ousterhout has argued that in its late Roman life the monument's significance was elusive: it was gradually Christianised, with rumours of Christian relics being incorporated at its base. By the early Byzantine period, the column had become mythologised with talismanic interpretations which claimed that statue protected the city from vermin.4 Throughout the centuries of Ottoman rule the column was preserved in its place while the urban context evolved. Its talismanic power seems to have captivated its new audience as well. In his *Book of Travels* the renowned Turkish traveller Evliya Çelebi (1611–1685) ascribes the column to Constantine, describing it with characteristic exaggeration as made of a thousand pieces, reaching 100 *arşın* in height (around 75 m), and encrusted with legends of a mythical past.<sup>5</sup> The myths attributed to the column might have helped its survival as part of the

<sup>2</sup> For the column's history see Müller-Wiener, *Bildlexikon zur Topographie Istanbuls*, 255–256; Mango, 'Constantine's porphyry column', 103–110; Ousterhout, 'The life and afterlife of Constantine's column', 304–326; Yoncacı Aslan, 'Towards a new honorific column', 121–145. For the monument's previous history see the contribution by Paul Magdalino in the present volume.

<sup>3</sup> Ousterhout, 'The life and afterlife of Constantine's column', 307.

<sup>4</sup> Ousterhout, 'Building medieval Constantinople', 42, ft. 38; Ousterhout, 'The life and afterlife of Constantine's column', 305.

<sup>5</sup> Çelebi, *Seyahatnâmesi*, 32–33.

urban fabric and even its preservation through repairs in various eras. During its last renovation by the Ottomans, the bands were renewed, and the original base and the lowest porphyry drum were encased in a new masonry envelope measuring 11 m in height. There is no official record explaining the process or justification of this repair, and different dates are proposed by different historians, varying from the mid-seventeenth century to the end of the eighteenth century.6 Even though the column was preserved and repaired over the centuries it had become hemmed in by a densely inhabited neighbourhood and Western visitors to the city named it 'la Colonne Brûlée' because of its darkened surface.7

In 1866, an unusual announcement of an architectural competition regarding the column was published in the local newspapers.8 It was requested that the Column of Constantine (Çemberlitaş) be 'restored to its original, ancient form' ('hey'et-i asliyye ve kadimesi üzere ta'miri'). This was the first architectural competition ever held by the Ottoman authorities in Turkey.9 The newspapers announced that all engineers and architects willing to display their knowledge of the 'science of architecture' ('fenn-i mi'mari') were invited to enter the competition. The prize for the winning project was announced not in the Ottoman currency of piastres (*kuruş*), but as 1,000 francs. The proposals were to be submitted to the Ministry of Commerce (*Ticaret Nezareti*) two months after the invitation, in the form of two drawings: an architectural survey of the column in its recent condition and the proposed design. An additional report on the original features and the history of the column was also sought, which meant that historical research into the column was mandatory for the competitors.

The advertisement insisted that the proposals be loyal to the original form of the column. This concern, publicly broadcast in the Ottoman newspapers by this competition, marked a qualitative change to earlier practices of repair. According to this novel perspective the column would no more stand as an old, organic feature of the urban setting, but would rather be stripped of its time-worn layers in order to be transformed into a modern monument. Its modernity as a monument thus demanded

<sup>6</sup> Nineteenth-century Greek historian Kostandiyos wrote that the column was shabbily repaired with a cheap wall 160 years ago which would be around the late seventeenth century; *Tercüman-ı Ahval*, 13 (2 Receb 1277 [14 January 1861]). Cornelius Gurlitt dated this addition to 1701; Gurlitt, *İstanbul'un Mimari Sanatı*, 15. Wolfgang Müller-Wiener also dates the repair to late seventeenth century after being severely damaged by the earthquake in 1648, see *Bildlexikon zur Topographie Istanbuls*, 256. For Semavi Eyice, 'Çemberlitaş', 483, the Baroque features of the masonry base implied that the renovation was made in the second half of eighteenth century. Ousterhout, 'The life and afterlife of Constantine's column', 307 argued that the column was last renovated in 1779. Armenian historians Cosimo Comidas de Carbognano and Gugas Inciciyan suggest that it was significantly destroyed during the great fire in 1779 and was repaired afterwards as discussed by Sönmez, 'Osmanlı'da Bürokratik Tartışmalar', 8. 7 One of the earliest records on the use of this name after the column is French Joseph Pitton de

Tournefort's (1656–1708) travelogue; de Tournefort *et al*., *Relation d'un voyage du Levant*, 230.

<sup>8</sup> The competition was announced two months earlier and the deadline was extended two more months. See; *Ruzname-i Ceride-i Havadis*, 440 (24 Safer 1283 [8 July 1866]); *Tasvir-i Efkar*, 403 (26 Safer 1283 [10

July 1866]). 9 Another architectural competition prior to this was organised in 1856 by the British Government for a new church design in Istanbul in memory of the Crimean War; Crinson, *Empire Building*, 138.

the priority of its initial ancient form over its whole history, and the revelation of the column's barely known pre-Ottoman past.

Unfortunately, no record of the competition or the architectural drawings submitted to it have come to light through research in the Ottoman archives. However, as was first noted by Cyril Mango, the Swiss-Italian architect Guiseppe Fossati prepared a proposal together with a report for the competition, which survive in the Fossati Collection of the Bellinzona Archives in Switzerland. Guiseppe Fossati was the younger brother of Gaspare Fossati and had worked as his assistant on numerous prestigious commissions which they had received from the Ottoman government and from the bureaucratic and commercial elite of Istanbul in the years 1840–1860. Fossati was living in Switzerland at the time but notes in his archival records that he heard about this competition from the *Journal de Constantinople.*10 Despite his discovery, Mango did not say anything about the drawing's relationship to the competition, nor did he find the project worth publishing because he thought that it was 'extremely ugly'. Another historian of the Fossati brothers, Tito Lacchia, published the report, again without the drawings of the project, considering it a personal survey of the column by Guiseppe Fossati.<sup>11</sup> In this published report by Lacchia, Fossati describes the column's poor condition at the time from memory; he also judged that the last repair had been entrusted to 'ignorant hands'. His proposals for the competition involved excavation of the ground to reach its original level, strengthening the base, completion of the column's missing parts with Marmara marble, polishing the column and renewal of the bands if necessary. On top of the column he proposed placing a marble sphere representing the Earth (probably the sphere the emperor originally held), and a bronze or copper statue of the Emperor Constantine. Alternatively, to add an Islamic-Ottoman flavour, he suggested placing an *âlem* (crescent) on the very top, as an option. He additionally proposed building balusters around the column; these were never installed, unlike the iron balusters which were installed around the columns of the Hippodrome in 1856.

Important background to the competition is that in August 1865, almost a year prior to this competition, the disastrous Hocapaşa fire destroyed the dense urban fabric surrounding the column. It was one of the biggest fires in Ottoman Istanbul, making thousands of people homeless. It started in the Hocapaşa neighbourhood of Eminönü and very soon spread into Kadırga and Kumkapı in the south, and Sirkeci, Divanyolu and Beyazıt to the north. Soon afterwards a comprehensive urban renovation, including the planning and reconstruction of these areas, was initiated by the government, fundamentally changing the overall layout of this vast urban landscape.12 This was a remarkable process, deploying new approaches to urban planning and architecture which unexpectedly facilitated a new relationship with the

<sup>10</sup> Mango found them in the Bellinzona Archives, Albume no 35; see Mango, *Mosaics of St Sophia*, 9.

<sup>11</sup> Lacchia, *I Fossati architetti del Sultano*, 96–98.

<sup>12</sup> Altınyıldız, 'The architectural heritage', 285; Çelik, *Remaking of Istanbul*, 53–59; Tanyeli, 'Düşlenmiş Rasyonalite', 503–537.

city's pre-Ottoman legacy. The Commission for the Improvement of Roads (*Islahat-ı Turuk Komisyonu*), established after the Hocapaşa fire, initiated the first clearing operation along Divanyolu, the main thoroughfare of ancient Constantinople as well as contemporary Istanbul, as far as Unkapanı. Historically, all the commercial activities had been sited on the Mese, the urban spine of Constantinople, together with the major public squares and several Fora of the Emperors, such as the Augusteion at the southern edge of Hagia Sophia and the Fora of Constantine, Theodosius I and Arcadius. The Ottoman Divanyolu followed this course and supported its spinal character with the addition of public monuments such as mosques and their annexes, including baths, madrasas and mausoleums.

This added another phase to the significant changes in the urban morphology of Istanbul which had occurred in the Byzantine and Ottoman periods, and which had also affected the form and the role of this main thoroughfare. Maurice Cerasi called the Ottoman instantiation of this thoroughfare the Divan 'axis' because it was constituted by a set of streets rather than being a single, unitary corridor. Nevertheless, this complex of streets running in a general direction formed an axis, flanked by important public monuments, which was used for ceremonial processions.13 Cerasi used the term Divanyolu only to refer to the part running as a single street from Ayasofya to the Beyazıt Square (the former Forum of Theodosius), because this eastern end was the most recognisable heir of the Mese. However, not even this section had maintained its former course and linear alignment, having moved some 10 m to the south, while the street level had been raised 2.35 m by fourteen centuries of infill.14 In 1865, however, the street was radically widened from an average of 5 m to 16 m, literally slicing the facades off the contiguous historical buildings and uniformly refashioning them with ribbed pointed arches, referencing a novel 'neo-Ottoman' repertoire.<sup>15</sup> This operation, executed by Italian Giovanni Battista Barborini, reclaimed Divanyolu's axiality, like that of former Mese, and gave it the prospect of an avenue with flanking monuments.16 The old street was remade into a new urban spectacle. Bayezid Square, a prominent urban centre of intramural Istanbul, was connected to the newly created southern plaza of Hagia Sophia by a linear vehicular strip with a new tram line flanked by sidewalks shaded by uniformly planted trees.

The ruins of the burnt-down buildings around the Column of Constantine/ Çemberlitaş were also cleared out in this comprehensive process. Information about the reorganisation of the area, together with a short history of the ancient column, was published in *Ruzname-i Ceride-i Havadis*, on 8 March 1866, five months after the fire.17 A

<sup>13</sup> Cerasi *et al*., *Istanbul Divanyolu*, 26, 33.

<sup>14</sup> Cerasi *et al*., *Istanbul Divanyolu*, 48. See also the chapter by Paul Magdalino in this volume.

<sup>15</sup> Ahmet Ersoy interpreted these new facades added to the sliced old monuments as an idiosyncratic Venetian-Gothic translation of Ottoman tradition by Italian Giovanni Battista Barborini. This belonged to the late Tanzimat historicism, including stylistic references to the fifteenth-century Tiled Pavilion (*Çinili Köşk*) in the public monuments of the era; Ersoy, *Architecture and the Late Ottoman*, 238.

<sup>16</sup> Çelik, *Remaking of Istanbul*, 55–63. 17 *Ruzname-i Ceride-i Havadis*, 360 (20 Şevval 1282 [8 March 1866]).

*Fig. 15.2: 'The Tomb of Sultan Mahmoud; the Burnt Column in the Distance', in J. Robertson,*  Photographic Views of Constantinople *(London, 1853) (Digital image courtesy of Getty's Open Content Program).*

notable statement in the article described the new aura surrounding the column as 'extremely ancient and artistic and unique'. Three months later, in May 1866, in tune with the re-organisation of Divanyolu and the creation of a new triangular plaza around Çemberlitaş, the competition for the column was announced.18 The expectation was that it would be repaired with a 'new method', intended to return it to its original form, in contrast to earlier repairs which had merely intended to keep it intact. However, for reasons now unknown, the competition was never held. Thus, the column was not renovated at this period, though its urban appearance was altered into a more visible and accessible one, facing the new Divanyolu Street right

at its edge (Figures 15.2 and 15.3). Fossati probably never sent his drawings and the report; these are still in his personal archive. Who the organisers were and why the competition process failed remain an obscure episode in Istanbul's urban history.

Late nineteenth century images, photos and etchings of Divanyolu commonly tend to show the street from a linear perspective. The Ottoman public monuments belonging to different centuries line up on both sides, including the dark silhouette of Çemberlitaş. The column stands immediately behind the mausoleum of Sultan Mahmud II (1785– 1839), the father of the Tanzimat reforms, built by his son Abdulmecid in 1840. With its up-to-date architectural features, showing neo-classical and French Empire stylistic influences, the mausoleum introduced a novel urban character to Divanyolu, and was a prominent site in the Ottoman political context, listed in the official itinerary for the city tour by foreign visitors.19 Thus, Çemberlitaş, still in its dirt-encrusted condition, was integrated into the modern frame of an imperial capital bearing the imprint of many historical accumulations, as well as the cultural constructs of a glorious imperial past. The avenue defined a new configuration whereby objects from different pasts

<sup>18</sup> The houses surrounding the column were demolished during these operations. The new triangular plaza created behind the column is the intersection of several paths leading to the Nuruosmaniye and Atik Ali Pasha mosques, the Vezir Han and the Grand Bazaar. For a critical evaluation and a new architectural projection over this narrow and special urban void, see Bugatti, 'The Istanbul Divanyolu', 47–52.

<sup>19</sup> During the stay of the Vatican's ambassador, he was made to visit the Topkapı Palace, Mausoleum of Sultan Mahmud II, Hagia Sophia, Sultan Ahmed Mosque and the Museum; *Journal de Constantinople*, 77. French Empress Eugénie, during her official stay in 1869, also visited Hagia Sophia, the Royal Mint, the Mausoleum of Sultan Mahmud II, imperial mosques, Grand Bazaar and the Museum; Giz, 'Batılı Prensler', 27.

became linked in a spatial relationship, in which different periods of the city's history were encountered in succession rather than overlapping (Figure 15.4). Eventually, the newly ordered layers of the old urban fabric were deliberately made to stage the historic city by forming a modern 'palimpsestic' scene of monuments defined by their origin in different periods. The contribution of the Column of Constantine to this new urban imagery seems crucial. Given the clearly stated official will to reinstate the column's 'original' appearance through the competition, it was intended to take its part as the oldest element of this palimpsest, in accordance with the city's rediscovery of its pre-Ottoman history.

*Fig. 15.3: 'The tomb of Sultan Mahmoud constructed in 1840', G. Berggren, No. 16, 1875 (Digital image courtesy of Getty's Open Content Program).*

### **The 'other' history of Istanbul: The re-contextualisation of the city**

The competition of 1866, which remains intriguing despite its failure, did not arise out of the blue. Rather, it was part of a history of ambivalence about the column's place in the city's history and, by association, the evolving Ottoman identity. The question of why the Ottomans consistently repaired this column, an ancient object with a history to which they were indifferent and which served no practical function, is not easy to answer. Originally, the colossal Column of Constantine belonged to the ancient Mediterranean tradition of freestanding urban column monuments which might have votive, funerary, honorary and triumphal connotations.20 These were paralleled with the old Roman imperial practice of installing ancient objects in cities as new components of imperial iconography, notably including obelisks brought from Egypt.21 These objects have always provided an evident and dynamic connection between past and present, while the symbolic meaning and significance they embodied constantly changed over time. The Ottomans borrowed and transformed former Byzantine narratives about these objects to fit their new Islamic/Ottoman context.22 Ancient objects might be preserved and re-used as building materials, often placed visibly in walls as spolia, or simply kept standing on their original sites. Wendy Shaw has suggested that such displays contributed to keeping the idea of conquest alive in the public memory.23 Thus, by the mid-nineteenth century, the traditional context of these

<sup>20</sup> Yoncacı Arslan, 'Towards a new honorific column', 128.

<sup>21</sup> James, '"Pray not to fall"', 13.

<sup>22</sup> Yerasimos, *Kostantiniye ve Ayasofya Efsaneleri.*

<sup>23</sup> Shaw, *Possessors and Possessed*, 43–44.

*Fig. 15.4: 'Entrance of Sultan Abdul-Aziz to Mausoleum of Mahmoud II',* Le Journal illustré*, vol. III, 28 (9 Juillet 1876) (Courtesy of SALT Research).*

ancient objects would have been transformed, demythologised and re-mythologised, including through new forms of historical narrative appearing in the emerging popular press. New visibilities were produced through evolving archaeological and architectural encounters with the city.

The emergence of new ways of seeing the pre-Ottoman legacy of Istanbul parallels the transformation in Ottoman Turkish historiography through the nineteenth century. Existing literature on Ottoman historiography widely acknowledges that Ottoman interest in the histories of other countries and civilisations, past or contemporary, was surprisingly limited.24 The dominant historiographical attitude of mystification and/or neglect of pre-Ottoman times was also prevalent in the Ottoman historiography of Istanbul's pre-Islamic past and its buildings. In his eighteenthcentury catalogue of the mosques of Istanbul, for example, Ayvansaraylı Hüseyin Efendi focusses almost exclusively on the Islamic period of the history of Hagia Sophia, with just a short note on its former function as a church*.* 25 Despite their general indifference to the 'other' history of Istanbul and its architecture, Ottoman authors writing about the city's physical scenery did discuss ancient buildings and objects

<sup>24</sup> Yinanç, 'Tanzimattan Meşrutiyete', 573–595; Arıkan,'Tanzimat'tan Cumhuriyet'e', 1584–1594; Ursinus, 'Byzantine history', 211–222; Neumann, *Araç Tarih Amaç Tanzimat*, 33–35; Strauss, 'The Millets', 189–249.

<sup>25</sup> Hüseyin Efendi *et al*., *Hadikatü'l Cevami'*, 42.

in various contexts.26 Visual sources also bear witness to Ottoman attention to the ancient objects of Istanbul. The ancient columns of the Hippodrome, for example, were frequently depicted as standard features of the urban setting of imperial ceremonies in the albums of *Surname-i Hümayun* miniatures.

Remarkably, an article published in the first semi-private newspaper, *Ceride-i Havadis*, in April 1856, announced the latest excavations at the bases of the stillstanding obelisks of the Hippodrome and provided a brief history and past uses of these monuments and the Hippodrome itself.27 Ten days later another article informed the readership of the recent constructions around the base of the Serpent Column, again with a brief history of its origin.28 Whereas excavations at the bases of these ancient columns had been carried out earlier by various people without being advertised, these articles now announced the latest works on these columns, re-contextualising their existence in the urban space as new historic objects of display.29 Accordingly their bases, several metres below ground level, were fully excavated, the pits lined with retaining walls and iron balusters, leaving them fully exposed but inaccessible to their audience. In 1861, several years after this highly visible operation to showcase the ancient legacy of the city, the first non-official Turkish newspaper, *Tercüman-ı Ahval*, started to publish a new series of articles on the architectural history of this 'other' Istanbul. The series was published twice a week for two months and was translated from the book of Kostandiyos (Konstantios) (1770–1859), who had briefly served as patriarch of Constantinople.30 The book, originally written in Greek and published in Venice in 1824, mainly focused on the history, topography and monuments of Byzantine Constantinople and the later Ottoman additions. Later, in 1861, the series would be published in an abridged Turkish version entitled *Heyet-i Sabıka-i Kostantiniyye* (Order of the Old Constantinople), in which the original text of approximately two hundred pages was summarised in thirty-nine pages, with the tables and etchings of major monuments omitted. In a short preface to the series, the editor explained the motive behind this publication as the 'apparent necessity of getting to know those ancient objects and buildings which people see and walk by every day without ever knowing what they actually are'.

<sup>26</sup> Two examples discussing the role of Ottoman testimony on Byzantine heritage are Ménage, 'The serpent column', 169–173; Raby, 'Mehmed the Conqueror', 141–152.

<sup>27</sup> *Ceride-i Havadis*, 786 (19 Şaban 1272 [25 April 1856]). *Ceride-i Havadis,* later named *Ruzname-i Ceride-i Havadis*, is the second newspaper published in Istanbul after 1840 with a semi-official character*.* Its British owner, William Churchill, received funding from the Sultan and it occasionally included international news and content translated into Turkish. See Topuz, *II.Mahmud'dan Holdinglere*, 17. 28 *Ceride-i Havadis*, 794 (29 Şaban 1272 [5 May 1856]). 29 The excavation at the columns of the Hippodrome was a two year process. In 1855, Charles Thomas

Newton had cleared the earth around the Serpent Column with a team of Croatian workers, but left the excavation incomplete. It was Otto Frick and Philip Anton Déthier who cleared the base and inscriptions of the column. The other two column bases were exposed in 1856 in another British enterprise. See Ousterhout, 'The rediscovery of Constantinople', 191.

<sup>30</sup> For his biography see Eyice, 'İstanbul'un Fetihten Önceki', 85–90; Sümertaş, 'Patrik Konstantios'un İstanbul'u', 48–53.

#### 316 *Göksun Akyürek*

A new journal, *Mecmua-i Fünun* (Journal of Sciences), of *Cemiyyet-i İlmiyye-i Osmaniyye* (Ottoman Society of Sciences) was founded by Münif Paşa in 1861, the same year when Alexander Paspates founded *Syllogos*, the Greek Literary Society of Istanbul. *Mecmua-i Fünun* played a crucial role in popularising scientific and scholarly information, including texts on the history of pre-Islamic Istanbul and its architectural history. Another facet of this history is the crucial part played by Ottoman Greek intellectuals in the production of archaeological knowledge. Their linguistic and historical affinities with Greco-Roman and Byzantine history along with active networks in both European circles and the official mechanisms of the Ottoman Empire facilitated this role.31 Their mediating role as authors and translators seems to have been central in the appearance of this new historical concern in the Turkish popular press, in which the scope of these texts seems to have been confined to Constantinople and its most visible monuments. Another series on Hagia Sophia by Kostantinidi Paşa (Alexander Constantinidis) published in *Mecmua-i Fünun* under the title *Tarih-i Ayasofya*, was also reprinted in book form in 1868.32 Earlier, in June 1866, parallel with the abortive competition concerning the Çemberlitaş, *Mecmua-i Fünun* also published an in-depth article on the history of the column by another Ottoman Greek, Fardis (Alexander Themistoklis Phardys).33 Such simultaneity demonstrates the widening context of interest in Istanbul's ancient legacy as it was rediscovered by varying actors through connected practices.

### **Mid-nineteenth century encounters with Istanbul's ancient legacy**

In 1839, inside the publicly inaccessible imperial gardens of the Topkapı Palace, the new Tanzimat Edict was announced in the presence of Sultan Abdülmecid. This is seen as a significant step in political reformation in nineteenth-century Ottoman history. It represented an internationally sanctioned statement of the construction of a new order for the state, favouring a modern, centralised bureaucratic government and a gradual transition from subjecthood to citizenship.34 The Tanzimat reforms had significant impact on the urban space of Istanbul, offering new relations with its historical legacy. In 1840, twenty-six years before the competition for the restoration of the Column of Constantine, Gaspare Fossati was commissioned to design a new monument to commemorate the Tanzimat Edict's first anniversary. *Ceride-i Havadis* announced that two identical monuments would be erected on two sites on 3

<sup>31</sup> A key figure in this process was antiquarian Alexander G. Paspates and his institution, *Syllogos*, established in 1861. It was a scholarly institution for literary and scientific studies within the Greek millet and would soon become the hub of a widening intellectual network in and out of the Ottoman Empire. See Sümertaş, 'From Paspates to Sillogos'; Tağmat, 'İstanbul'da Sosyo-Kültürel'.

<sup>32</sup> The serial was published in two year period, between December 1864 and June 1867, in changing intervals. 33 *Mecmua-i Fünun*, 35 (Safer 1283 [June 1866]), 46–49.

<sup>34</sup> Findley, *Bureaucratic Reform*, 190–201; Quataert, *The Ottoman Empire, 1700–1922*, 54–72.

November 1840, the date of the Edict's promulgation.35 One would be in the imperial gardens of Gülhane and the other would be installed in the more central Bayezid Square. The monument, called *nişan-ı adalet taşı* ('stone/column as marker of justice'), was described having 'the shape of the Obelisk standing at At Meydanı, only seven to eight *arşın* [*c*. 5 m]taller'. The full text of the Edict would be inscribed on its body. Four lion-shaped bronze fountains would also be sited at its four corners. Yet the publicly announced plans for this unusual and unprecedented monument were not ultimately realised. According to the records of the late nineteenth-century chronicler Ahmed Lütfi Efendi, the monument had provoked objections on the basis of its affinity with 'European traditions'.36 His account suggests that there was opposition to the general idea of commemorating a political event by erecting a self-standing monument in public space, irrespective of the specific choice to build a replica of the Obelisk. What is remarkable about this incident is the implicit appropriation of the Obelisk at the Hippodrome as a legitimate, Ottoman model for a new type of modern monument which used recognisable formal resemblance to communicate its new message. The appropriation of the Obelisk's form for a new monument evidently differs from reviving the Column of Constantine as a modern monument as an instance of relating to Istanbul's ancient legacy. Nevertheless, the common fact that they both faced resistance after the plans were made public indicates some invisible boundaries which restricted these new architectural practices in the urban context.

In 1846, Gaspare Fossati received another prestigious commission: the design and construction of the Darülfünun building, intended to be the first Ottoman university. Announced immediately after the Tanzimat education reforms, it was planned as a multi-ethnic and multi-religious school to serve as the highest level of public education. The selected location for the building was the former site of the Old Armoury, an imperial property next to Hagia Sophia on its eastern edge, along the path leading to the Topkapı Palace to its north. The giant mass of Fossati's final project was vast, with a longitudinal façade around 100 m long. Not only this, but it was orientated on a north–south axis, altering the silhouette of the area and partially blocking the visibility of Hagia Sophia from the Marmara Sea. The proximity of these two monuments with completely different backgrounds was never mentioned in the official documents regarding the process of site selection (Figure 15.5).37 A conservative objection to their juxtaposition claimed that having a multi-religious school next to such a sacred sanctuary was inappropriate.38 According to the contrasting western accounts, the positioning of the Darülfünun next to Hagia Sophia proved that a new generation's future was secured by religion.39 The two contradictory discourses surrounding the same project reveal the manifold identity and symbolism of Hagia

<sup>35</sup> *Ceride-i Havadis*, 11 (11 Ramazan 1256 [06 December 1840]).

<sup>36</sup> Lütfi Efendi, *Vakanüvis Lütfi Efendi Tarihi*, 1068.

<sup>37</sup> For an in-depth survey and discussion of this project, see Akyürek, *Bilgiyi Yeniden İnşa Etmek*, 65–103. <sup>38</sup> Cevdet Paşa, *Tezakir*, 13–14.

<sup>39</sup> *Journal de Constantinople*, 106 (06 Septembre 1846); Ubicini, *1855'te Türkiye*, 193.

*Fig. 15.5: 'Panoramic view of the Seraglio Point from one of the minarets of Hagia Sophia' in G. Fossati, lithographed by L. Haghe,* Aya Sofia Constantinople, as recently restored by order of H.M. the Sultan Abdul Medjid *(London, 1852) plate 20 (Courtesy of the Library of Congress Prints and Photographs Division, Washington, D.C.).*

Sophia at the dawn of its re-invention as an internationally acknowledged 'modern monument'.

In 1847, most probably because of the visibility of the new Darülfünun project, the dilapidated condition of Hagia Sophia became disturbingly apparent and was officially acknowledged. Gaspare Fossati, already managing the Darülfünun's construction, eventually received another commission for its 'repair'. However, no extra payment was offered to him for this operation because of his contract for the Darülfünun project. Moreover, there is also an official record saying that he was the second choice for this task after Garabet Balyan refused on account of his involvement with the Dolmabahçe Palace construction.40 Contemporary testimonies about the condition of Hagia Sophia relate that there were human-sized cracks in the main dome. Fossati himself believed that the building had been decaying as a result of negligence ever

<sup>40</sup> Ottoman Archives (OA), İ.MSM., No. 672, 18 Rebiülevvel 1263 [6 March 1847].

since the 1766 earthquake.41 During the repair process Fossati tried to repair the structural problems he observed, built new structures in the 'Greco-Roman style' (his own words) and added new decorations to revive its imperial prestige and grandeur as a ceremonial mosque. The unforeseen discoveries of substantial fragments of mosaics which Fossati and his brother made were certainly not part of the repair programme.42 Remarkably, the Sultan and his bureaucrats inspected and admired these newly found mosaics, an occasion fully documented by the *Journal de Constantinople*. 43 As a consequence of official concern about conservative opposition, Fossati had to cover the figurative mosaics 'for their preservation until a future liberal age'. Only the seraphim on the pendentives remained, their faces now covered by star medallions after Fossati had recorded them in detailed drawings.44

The official ceremony for the re-opening of Hagia Sophia, held in the presence of Sultan Abdülmecid on 13 July 1849, became an international spectacle for the government.45 More importantly, the building acquired a new visibility in Europe, with two successive publications following the renovation which had brought about the discovery of its long-forgotten mosaics.46 The first one, by Gaspare Fossati, was published in London in 1852 under the sponsorship of Sultan Abdülmecid, with his imperial signature on the book cover. In his drawings Fossati represented the building in its recent form, decorated and used as a mosque, in its urban context with adjacent houses and the unfinished Darülfünun building. These were impressively coloured drawings, conveying a true sense of the building's scale and complexity. A second book was published in Berlin in 1854 by Wilhelm Salzenberg, who had been commissioned by the German Emperor and allowed to enter the building during the repairs with the help of Gaspare Fossati. This book showed Hagia Sophia fully as a Byzantine church, with detailed architectural drawings of its structural layout and decorative programme. By means of these two publications, the building was gradually rediscovered in Europe as a significant Byzantine monument, tapping into the growing European interest in the middle ages, and Byzantine art and culture as a part of them. The building gained enormous

<sup>41</sup> Mango, *Mosaics of St. Sophia*, 11.

<sup>42</sup> Fossati briefly explained his works in Hagia Sophia in an undated letter addressed to an unknown authority found in the Bellinzona Archives. While Mango believed that the addressee was a high level Ottoman bureaucrat, Schlüter wrote that it was written to the president of the St. Petersburg Academy of Fine Arts in order to ask for financial support for the book he planned. See Mango, *Mosaics of St. Sophia*, 112–114; Schlüter, 'Gaspare Fossati'nin', 63. In his public lecture about his ongoing work in Hagia Sophia in a theatre at Pera, Fossati also explained that his major source on the building's history was the book of Konstantios. See Mango, *Mosaics of St. Sophia*, 107. 43 *Journal de Constantinople*, 161 (9 Avril 1849); 155 (9 Mai 1849).

<sup>44</sup> Mango, *Mosaics of St. Sophia*, 14.

<sup>45</sup> The renovation process was also followed by the European public and the opening ceremony was presented in *L'Illustration* with Adalbert de Beaumont's engraving '*Cérémonie d'inauguration de la Mosquée de Sainte-Sophie de Constantinople, restaurée par Messieurs Fossati*', *L'Illustration, Journal Universel*, 13 (Juillet 1849).

<sup>46</sup> Fossati, *Aya Sofia*; Salzenberg, *Alt-Christliche Baudenkmale*.

popularity through visits by European tourists and its visual reproduction on postcards by virtue of the progress and popularisation of photography.47 These photographic images of Hagia Sophia and other Byzantine monuments in their contemporary urban context provided a new visual experience of the classical model and were also influential in the emergence of Byzantine archaeology as a scholarly practice in the late nineteenth century.48

Hagia Sophia, as a key monument with a dynamic relationship between past and present, is central to the discussion of how the late Roman and Byzantine past were appropriated in the long centuries of Ottoman rule. According to Gülru Necipoğlu, it was an uneasy and long process for the Ottomans to Islamise this unique Byzantine church after Mehmed II 'Ottomanised' it as the main imperial mosque of his new capital, without any change to its name.49 Architecturally, it was already an outstanding structure, embedded in multiple layers of legends and mythology by Byzantine and later Ottoman narratives.50 The Ottoman perception of the building involved an awareness of its former significance, as well as a shared language of architectural forms which had been translated into a new repertoire of imperial mosque architecture. Over the course of two centuries rich figurative mosaics were gradually plastered over in response to a changing socio-political context which increasingly demanded a more Islamic identity for the monument. Until the mideighteenth century, the Islamic inscription programme was deliberately modest in size and selectively preserved figurative mosaics remained visible to remind the Muslim congregation of the long life of Hagia Sophia.51 This twofold identity of the monument was exposed by the differing representations of the building in the two mid-nineteenth-century publications which resulted from the Fossati repairs. Subsequently, in 1866, in the aftermath of the Hocapaşa fire, the houses surrounding the building were demolished, its full view in perspective was made available through the newly created southern plaza and 'the great sanctuary so uniquely ancient' was finally fully exposed to its (inter)national audience.52 The newly laid out Divanyolu Avenue was now connected to the Hippodrome and the south-west corner of Hagia Sophia, so that the chief monuments of Constantinople were made into a new itinerary of tourist spectacle. Beyond Hagia Sophia, the thoroughfare continued further east to the modern monument of the Darülfünun building, finally

<sup>47</sup> For further discussion on these two publications, their authors' interaction and the process of how Hagia Sophia was rediscovered in Europe as a Byzantine monument see Nelson, *Hagia Sophia*, 73–104.

<sup>48</sup> Ousterhout, 'Historic photography', 63–64.

<sup>49</sup> Necipoğlu, 'The life of an imperial monument', 202.

<sup>50</sup> The ninth century text *Diegesis*, is a history of Hagia Sophia incorporated into a collection of descriptions of the city known as *Patria*, which seems to have found a broad audience and it was translated into languages such as Latin, Turkish and Persian. A Greek copy survives in the Topkapı Palace. The text forms the basis of the later Ottoman-Islamic mythology of Hagia Sophia and Constantinople. See Necipoğlu, 'The life of an imperial monument', 198–199; Ousterhout, 'From history to myth', 51–56.

<sup>51</sup> Necipoğlu, 'The life of an imperial monument', 220.

<sup>52</sup> Cited from the commission's report in Ergin, *Mecelle*, 953. For the historical investigation of the building's changing urban context see Nelson, *Hagia Sophia*, 73–104.

reaching at its northern end the emergent Museum and the Topkapı Palace, already in the process of its own musealisation.53

#### **Conclusion**

The rise of a nationalist ideology in the late nineteenth century came along with the search for a dynastic/national identity and its definition through the preservation and representation of an Ottoman artistic and architectural heritage.54 By this point, however, the discourse on national culture excluded any claim on the pre-Islamic ancient legacy. The introduction of archaeology as a modern institution, initiated with an imperial museum, was a conscious state enterprise, holding manifold relations with the claims of European countries (primarily Britain, France and Germany) on the Greco-Roman heritage from within Ottoman territories. State policy and management of antiquities, as well as the Imperial Museum in Istanbul, were not the sole means of changing cultural perceptions of the distant past, as shown by Ahmet Ersoy's work highlighting the contribution of the print industry to the creation of hybrid and dynamic visions of history and archaeology during the reign of Sultan Abdulhamid II.<sup>55</sup> Thus, the history of how the pre-Islamic legacy of the Ottoman territories was appropriated through modern practices of archaeology and historiography, as well as their inclusion in urban settings, is becoming an expanding field of research with multiple important perspectives.56

This study has been concerned with acknowledging the initial, mid-nineteenthcentury instances of the transformation of regimes of knowing and seeing the city's pre-Ottoman historical legacy. The conversion of the Column of Constantine into a modern monument, its imposing scale re-emphasised through its positioning on a major new avenue, is one of the cases through which we may observe this change. This interactive process of urban planning and re-contextualisation was accompanied by the diffusion of new scholarly accounts of the history of pre-Ottoman Istanbul. With the reorganisation of Divanyolu as a modern avenue, its complex spatial and architectural configuration was made to enfold its multi-layers in a linear perspective, privileging the visual experience of the historical city over the physical. The facilitating role of the Hocapaşa fire in this large-scale reorganisation of the dense urban fabric was crucial in allowing contemporary ideas of urban planning to be tested in the most historic part of the city. In keeping with the newly laid out residential fabric with lower density and a new, regular road network, houses adjacent to the historic monuments were 'cleared' to satisfy the principle of isolating monuments, a current trend in Europe.57

<sup>53</sup> Özlü, 'Single p(a)lace, multiple narratives'; Aykaç, 'Musealization as an urban process'. 54 Ersoy, 'Architecture and the search'; Ersoy, *Architecture and the Late Ottoman.*

<sup>55</sup> Ersoy, 'Ottomans and the Kodak Galaxy', 330.

<sup>56</sup> Bahrani *et al*., *Scramble for the Past*. 57 Preservation of monuments in the urban space is part of the long process of the rise of heritage in Europe. For an analytical history on heritage-formation in France, Germany and Britain between 1789 and 1914 see Swenson, *Rise of Heritage*.

Zeynep Çelik has argued that Haussmann's Paris operations were influential in the replanning activities aimed at this partial 'embellishment' of the city, which went beyond the healing of the city's wounds from this great fire.58 Cerasi, on the other hand, has provided an overview of the interventions on the Divanyolu which criticises the losses caused by the volumetric alterations to the architectonic complexity of the Ottoman monuments (mausoleums, madrasas, graveyards, walls, old trees), which had previously been juxtaposed along the non-linear artery for centuries, creating a unique sense of urban space.59 Thus, two versions of the palimpsest city collided in this case of the construction of Divanyolu Avenue through the partial erasure of its old layers.

Invoking multi-layered manuscripts overwritten and re-used, the metaphor of the palimpsest when applied to cities draws attention to the temporal configuration of urban space, and the layers, sediments or fragments of the various pasts inherent in it.60 The two palimpsest images of the Divanyolu, however, offer alternative engagements with the interwoven and multi-layered urban space, as regards the changing context of how the city was historicised through modern practices of urban planning. The spontaneity of the acts of erasing and re-writing in the palimpsest was interrupted by a staged, deliberate multi-layering of the modern urban scenery. The unrealised attempt to restore the Column of Constantine to its 'original' state is a prime example of how this new process sought to revive particular obscured layers. Yet we still do not know why the competition for the Column failed to materialise. The answer to the question of why the Ottomans occasionally repaired the Column in the previous centuries remains equally obscure. According to Evliya Çelebi (d. 1682), Ottoman audiences up to his time barely knew that the Column was that of the Emperor Constantine, and most probably appreciated the Column's prestige simply for the monumental scale of its imposing physical presence and the precious porphyry stone of Egyptian origin from which it was made. Besides this, it had been appropriated as a talisman of the city through popular narratives about its apotropaic powers. The dialogue with its changing audiences kept the column alive as part of the dense urban fabric surrounding it. This type of relationship may be related to Yannis Hamilakis's discussion of local or indigenous archaeologies.61 As opposed to the linear, developmental narrative of archaeology in the late eighteenth and nineteenth centuries, he suggests including local discourses and practices to do with the material past within the concept of an 'indigenous archaeology' which predates the modern one. These included different perceptions and attitudes towards ancient objects, buildings and landscapes, expressed through folktales and narratives leading to active engagements in a multisensory manner, as well as their spoliation, which had been a common practice in many cultures.62

<sup>58</sup> Çelik, *Remaking of Istanbul*, 169, ft. 61, 67.

<sup>59</sup> Cerasi *et al*., *Istanbul Divanyolu*, 156–157.

<sup>60</sup> Huyssen, *Present Pasts*, 7.

<sup>61</sup> Hamilakis, 'Indigenous archaeologies', 61.

<sup>62</sup> For a recent compilation of studies on spoliation practices in Anatolia in varying contexts, see Jevtić and Yalman, *Spolia Reincarnated*. For an exploration of spoliation in the medieval Anatolian context see the chapter by Scott Redford in this volume.

Likewise, David Harvey has described heritage as a human condition which has always been produced by people according to their concerns about and experiences of any 'present' in history.63 He suggests viewing heritage and memory not only as particular modernist products of the late nineteenth century, but as processes in evolution over the longer term. The various instances from nineteenth-century Istanbul described here display the changes to the Ottoman cultural imagery regarding heritage and memory, through the emergence of new historical narratives and architectural practices involving its historic core and foremost monuments. These changes were made to modernise the city by historicising it, even if this was a partial attempt which neglected many other historic monuments of the Ottoman or pre-Ottoman era. In the 1950s, the historic city would experience even greater demolition and re-planning, this time de-historicising a vast area in the name of its modernisation, destroying many of its historic monuments.64 The Column of Constantine, kept visible and alive as an object of long-term heritage practices by its unique and fortunate urban context, was not 'saved' by those modern interventions, but re-made. In the long run, either as a historic monument of universal meaning and value now staged in a novel urban setting, or as a strange ancient object in relation to its current context, it sustains its power to stimulate for its beholders the multiple pasts, or invisible histories of the city.

### **Bibliography**

#### *Newspapers*

*Ceride-i Havadis Journal de Constantinople Le Journal illustré L'Illustration, Journal Universel Mecmua-i Fünun Ruzname-i Ceride-i Havadis Tasvir-i Efkar Tercüman-ı Ahval*

#### *Primary*

Ottoman Archives of the Prime Ministry, Istanbul, (BOA), İ.MSM., No. 672.

#### *Secondary*

Akpınar, I., 'The making of a modern Pay-ı Taht in Istanbul: Menderes' executions after Prost's plan' in C. Bilsel and P. Pinon (eds.), *From the Imperial Capital to the Republican Modern City: Henri Prost's Planning of Istanbul (1936–1951)*, (Istanbul, 2010), pp. 167–199.

<sup>63</sup> Harvey, 'Heritage pasts', 320.

<sup>64</sup> In 1956, then Prime Minister Adnan Menderes aimed to revive the Ottoman Istanbul with a claim 'to conquer Istanbul once again', and started an extensive urban development program for which two boulevards were built in the historical peninsula. This operation meant the relocation, mutilation and destruction of many historic monuments besides thousands of houses. See Akpınar, 'The making of a modern Pay-ı Taht'.

Akyürek, G., *Bilgiyi Yeniden İnşa Etmek: Tanzimat Döneminde Mimarlık, Bilgi ve İktidar* (Istanbul, 2011).


Huyssen, A., *Present Pasts: Urban Palimpsests and the Politics of Memory* (Stanford, 2003).

James, L., '"Pray not to fall into temptation and be on your guard": Pagan statues in Christian Constantinople', *Gesta* 35.1 (1996): 12–20.


Tağmat, Ç. D., 'İstanbul'da Sosyo-Kültürel Helenizm'in Temsilcisi Bir Cemiyet: 'Syllogos' (1861–1923)', *Tarihin Peşinde Uluslararası Tarih ve Sosyal Araştırmalar Dergisi* 13 (2015): 101–128.

Tanyeli, U., 'Düşlenmiş Rasyonalite Olarak Kent: Türkiye'de Planlama ve Çifte Bilinçlilik' in S. İlkin, O. Silier and M. Güvenç (eds.), *İlhan Tekeli İçin Armağan Yazılar* (Istanbul, 2004), pp. 503–537.

Topuz, H., *II.Mahmud'dan Holdinglere Türk Basın Tarihi* (Istanbul, 2003).

Ubicini, A., *1855'te Türkiye*, tr. A. Düz (Istanbul, 1971).

Ursinus, M., 'Byzantine history in late Ottoman Turkish historiography', *Byzantine and Modern Greek Studies* 10 (1986): 211–222.

Yerasimos, S., *Türk Metinlerinde Konstantiniye ve Ayasofya Efsaneleri*, tr. Ş. Tekeli (Istanbul, 1993).

Yinanç, M. H., '*Tanzimattan Meşrutiyete Kadar Bizde Tarihçilik*', Tanzimat I (Istanbul, 1940) pp. 573–595.

Yoncacı Arslan, P., 'Towards a new honorific column: The column of Constantine in early Byzantine urban landscape', *METU Journal of the Faculty of Architecture* 33.1 (2016): 121–145.

Whose past?

## Chapter 16

## Medieval Arabic archaeologies of the ancient cities of Syria

*Edward Zychowicz-Coghill* 

#### **Introduction**

In 1985, Hugh Kennedy's '*Polis* to *Madina*' article signalled a shift in the terms of debate about the changing form of public urban space in the early Islamic period. Kennedy argued that the infilling of wide, colonnaded streets and agoras and the end of monumental public construction, aside from religious buildings and smaller bathhouses, should not be understood as signs of general decline, embodying the inauguration of the radically new, debased urban form of the Islamic city. Rather, these changes were a continuation of pre-Islamic, late antique urban trends and represented different, but not necessarily less vital, life-styles, customs and uses of the city, a view which has been supported by subsequent archaeology.<sup>1</sup> Kennedy's correction was necessitated by a western academic tradition of idealising classical Greco-Roman urban forms and ideals, any perceived Islamic deviation from which could only be deterioration.2

This volume emerges from a project which seeks to study the afterlife of the Greco-Roman city: the way people across the post-Roman world built on the classical pasts of their cities, literally and figuratively, physically and imaginatively. One outcome has been, unsurprisingly, to see that, when talking about either the city in the abstract or particular ancient cities, Muslims writing in Arabic – despite their

<sup>1</sup> Walmsley, *Early Islamic Syria*, 77–90.

<sup>2</sup> The assumed otherness of Islamic urbanism is well illustrated by Tsafrir and Foerster, 'Urbanism at Scythopolis-Bet Shean', who consistently essentialise planned urbanism as 'western' and unplanned urbanism as 'eastern', the latter's 'victory' in the Umayyad period, they suggest, best being explained by 'psychology' (of the 'eastern' mind!?), 141. The predetermination of their approach can be seen by their relabelling what had once been considered a 'Byzantine street of shops' as an 'oriental bazaar' due not to any distinction in morphology but the discovery of an Arabic building inscription, *ibid*., 138–139, a point made by Walmsley, *Early Islamic Syria*, 87.

wholesale incorporation of certain genres of Greek knowledge, such as medicine and philosophy – drew far less on classical writings than their counterparts in the Latin west or Greek east.<sup>3</sup> Rather than lament this as a lack, or a failure, this article explores some of the discourses about the ancient city which did emerge in Arabic writing. Just as we now think of the dense, winding streets of the *madīna* as an alternative to the *polis* model which was nevertheless a productive approach to urban space, Islamic ideas about the obvious antiquity of the cities which Muslims came to inhabit in the eastern Mediterranean had their own logics and utilities, irrespective of whether they represented a transmission of classical urban ideas or ideals.

This chapter focuses on one of the questions raised by the volume and the metaphor of the palimpsest, the issue of privileging the original: valorising the underwriting of the city-palimpsest to the neglect of the later inscriptions and uses. Claim-making about the peculiar significance and precedence of such 'original' layers is often done by actors seeking to connect themselves to that past, while erasing or denigrating the city's subsequent pasts. This was a salient feature of colonial archaeologies in the Middle East which focused disproportionally on exposing classical monuments from cities founded or flourishing in the Hellenistic or classical Roman periods, while discarding Islamic or Byzantine/Medieval Roman<sup>4</sup> era evidence for the continuing life of these cities in later periods.5 As we shall see, medieval Arabic scholars were also interested in making claims about the origins of the ancient cities of Syria-Palestine, yet for them these original pasts generally bypassed the classical layers in which the colonial archaeologists were so interested, penetrating back to the time of the Israelites or even the first generations of mankind after the Flood. They accessed these pasts through different epistemologies to those of modernist archaeologists, epistemologies rooted in the Qurʾān and the Arabic-Islamic reception of traditions derived from the Torah's account of ancient history and genealogy. These conflicting premises mean that pre-modern Islamic views have not been much accounted as part of the archaeological historiography of the classical city and its remains.6

This raises the question of which ancient city different people at different times have found it useful to remember, and how seriously we are willing to take such

<sup>3</sup> With the partial exception of Arabic philosophers who engaged directly with Greek ideas about the ideal state as expressed in the form of a city, though the relationship between this and thinking practically about urbanism (rather than, say, metaphysics), is questionable, see Vallat, *Farabi et l'école d'Alexandrie*.

<sup>4</sup> In this chapter I use Medieval (New) Rome to refer to the polity typically known in scholarship as Byzantium, on the basis that both Medieval Roman and Arabic writers conceived it and its population in continuity with classical Rome and referred to them with terms cognate with Roman, in the Arabic case, al-Rūm.

<sup>5</sup> Diaz-Andreu, *World History of Nineteenth-Century Archaeology*, 264–275.

<sup>6</sup> This may be attributed to a widespread sense in archaeology that there was an 'Islamic disdain for everything before the advent of Islam', Liverani, *Imagining Babylon*, 350, quoted in Rojas, 'Review of *Imagining Babylon*', where Rojas argues contrarily that 'the notion that there was no Muslim interest in the material traces of the past is, at best, misleading. Local antiquarians of various sorts have existed … throughout the Dar al-Islam'.

memories if they do not accord with our own. The Greco-Roman city is a category which is intelligible and important to us through our participation in contingent academic disciplines and modern discourses: of history, classics, architecture and urbanism. This article questions whether the Greco-Roman city was a category which was recognised and useful to pre-modern Arabic writers discussing the antiquity of cities. A modern western cultural formation might reify a particular combination of classical forms of and ideas about the city as an essential legacy of the Greco-Roman past, but it would be both anachronistic and Eurocentric to judge medieval Arab-Islamic civilisation for developing other intellectual agendas and preoccupations which resulted in quite different visions of the ancient city. In order to engage receptively with these pre-modern Islamic conceptualisations we must challenge the modern separation of history, myth, archaeology, theology and geography. The usefulness is to draw attention to a very different manner of approaching the material traces of the past and fostering senses of belonging and meaning between living populations of cities and their pasts, a manner which was both constructive and didactic.

To explore these issues this article will examine the presentation of the ancient origins of the cities of Syria-Palestine in Yāqūt al-Ḥamawī's *Muʿjam al-buldān* (*Dictionary of the Lands*), the largest pre-modern Arabic geographical work, written in the early thirteenth century. Yāqūt is useful because he preserves now lost earlier Arabic sources discussing the origins of cities, while drawing together fleeting references to Greco-Roman remains from a diverse range of Arabic sources. As such, he allows us to draw broader conclusions about how the ancient history and remains of the cities of the eastern Mediterranean under discussion came to be understood in Arabic thought. By focusing on Syria-Palestine (by which I mean the classical Arabic usage *bilād al-Shām*, henceforth referred to as Syria for convenience), the study contributes to a growing body of scholarship on the construction of territory in the pre-modern Islamic world and the place of Syria within that.<sup>7</sup>

#### **Yāqūt al-Ḥamawī's city biographies**

Yāqūt al-Ḥamawī (d. 1229) was born around 1179 in Medieval Roman territory, but as a child was captured in a raid and from the age of five or six raised as a slave in Baghdad. Well educated by his master, he was sent on business trips as far afield as Aleppo in Syria and Kish, an island in the Persian Gulf. Around the age of twenty he was freed, after which he worked as a copyist and bookseller while continuing his travels throughout the central and eastern Islamic world, including Egypt, Syria, Iraq, Iran and central Asia. By the time he settled in Aleppo, under the patronage

<sup>7</sup> The *bilād al-Shām* incorporates roughly the territory of the modern nation states of Syria, Lebanon, Palestine, Israel and Jordan. Major recent works on the construction of place are Zadeh, *Mapping Frontiers* and Antrim, *Routes and Realms*, while for Syria specifically see Antrim, 'Ibn ʿAsakir's representations of Syria' and '*Waṭan* before *Waṭaniyya*'; Borrut, 'La Syrie de Salomon'; Cobb, 'Virtual sacrality'; Webb, 'Pre-Islamic *al-Shām*'.

of the Ayyūbid vizier Ibn al-Qifṭī, he had studied with and visited many of the most prominent teachers and libraries in the Islamic world.<sup>8</sup>

One of the fruits of his education and his remarkable travels was his geographical encyclopaedia, the *Muʿjam al-buldān*. In this work Yāqūt attempted to compile information about the entire world, verify it and organise it accessibly. To do so, he not only drew upon his own experiences but also extensively quoted earlier works, thereby granting us access to a sophisticated, encyclopaedic presentation of the accumulated tradition of medieval Arabic geography up to the thirteenth century.9 Yāqūt asserted in his introduction that his work fulfilled a need shared by all educated persons, describing in turn how all different classes of learning stood in need of geographical knowledge.10 His work was, therefore, a programmatic attempt to present everything an elite, cultivated person needed to know about places. Therefore, his presentation of the pre-Islamic origins of cities tells us what perhaps the most influential medieval Arabic geographer thought you needed to know about the ancient city.

In a passage in his introduction, Yāqūt spelt out exactly what kind of information he would consistently try to give about the cities he lists in his work.11 After noting his adoption of an alphabetical arrangement of entries – explicitly inspired by the dictionaries of the lexicographers (*ahl al-lugha*) – he tells the reader that he will give precise information about how each place's name should be vowelled and its etymology. Then, he tells us, he will physically locate each city by the region in which it falls and its geographical relationship to other settlements. Next, he will also locate it cosmologically, by giving its Ptolemaic co-ordinates and horoscope. Unique features and marvels of the city will be discussed, as well as poetry which has been written about it. Finally, he will list the famous Muslims who were associated with each town, often the most extensive part of his entries in practice.

In terms of the history of these cities, Yāqūt wrote that he was consistently interested in three aspects: who founded them, their conquest by the Muslim community and who held dominion there in his own time. This can be considered the basic historical narrative of a place which Yāqūt thought it necessary to know. Towns, in this understanding, had to be founded by someone in order to exist. The next historical event of significance, however, was the coming of Islamic rule. The

<sup>8</sup> Gilliot, 'Yāḳūt al-Rūmī'; Shboul, 'Yāqūt Ibn ʿabdallāh'.

<sup>9</sup> Antrim, *Routes and Realms*, 145–146, describes Yāqūt as 'a veritable archive of the discourse of place as it had developed up to the thirteenth century'. He discusses his own place in the Islamic geographical tradition in his introduction, see Jwaideh, *Introductory Chapters*, 10–13. On 'encyclopaedism' as a useful category for thinking about medieval Islamic texts, with references, see Muhanna, *World in a Book*, 5–28.

<sup>10</sup> Yāqūt argues in his introduction that geographical knowledge is necessary for anyone with any claim to education: 'the need for this particular knowledge is shared by everyone who has had a measure of learning … all men alike stand in need of this knowledge', Yāqūt, *Buldān*, vol. 1, 8, tr. Jwaideh, *Introductory Chapters*, 3–4. He goes on to explain why it is necessary for pilgrims, administrators, jurists, biographers, historians, *ḥadīth*-scholars, doctors, astrologers, linguists, grammarians, poets and other litterateurs, *ibid*., 4–8.

<sup>11</sup> *Ibid*., 13.

gap between the Muslim conquest of the city and Yāqūt's own day he bridged by biographical information about famous Muslims who were associated with the town: jurists, scholars, poets and the like. These biographies function in his text as emblems for a continuous, living city whose inhabitants contributed to the great stock of Islamic achievements.12 In contrast, while Yāqūt always tried to say who first built a city, he set out no programmatic intention to indicate a continuous history which bridged the time-span between its foundation moment and the Islamic conquests. Yāqūt was interested in the origins of towns, but not their continuity, at least not until the Islamic period.13

In this sense, the historical perspective of Yāqūt's entries does present us with a kind of palimpsest. The pre-Islamic history of the city is scraped away, leaving only the primordial foundation of the city as a blank parchment on which its Islamic history will be written. In doing so Yāqūt consigned to oblivion the millennium of direct Greco-Roman influence on the cities of the Levant from the conquests of Alexander onwards. This is not because it would have been impossible for Yāqūt to find and reproduce information about that pre-Islamic period. Syriac and Arabic texts from the medieval period show some Christian writers remembering rulers and events from the late antique, classical Roman and even Hellenistic periods, and of course celebrating the coming of Christianity to their cities and the early saints who continued to act in death as the spiritual patrons of their communities.14

Moreover, some Muslim writers did incorporate these resources into geographical or historical works. In the tenthand eleventh centuries, for example, the Iranian and Andalusī writers Ḥamza al-Iṣfahānī and al-Bakrī included brief but fairly accurate chronicles of Roman and Byzantine history in their historical and geographical texts, respectively. Ibn al-Azraq, a Muslim writing in Mesopotamia just before Yāqūt's birth in the twelfth century, used a Christian saint's life as the basis for his local history's account of the foundation of the city of Mayyāfāriqīn. And Ibn al-ʿAdīm, a close associate of Yāqūt, related what he could find of Aleppo's Hellenistic, classical Roman and late antique history in his local histories of Aleppo.<sup>15</sup> Thus, Yāqūt could

<sup>12</sup> On collective biography as an alternative form of Islamic historical writing see al-Qadi, 'Biographical dictionaries'.

<sup>13</sup> For a comparable imaginative leap and lack of interest in historical continuity, see Sam Ottewill-Soulsby's discussion of William of Tyre in this volume. 14 A clearly Christian Arabic text, perhaps dating to *c*. 960–1080, includes an extensive, fabulous account of

the foundation of Antioch after Alexander's death, before continuing with the tale of its Christianisation at the hands of several apostles, see Stinespring, *Description of Antioch* and Margoliouth, 'Review of Guidi'. In Egypt, the originally Coptic *History of the Patriarchs*, translated into Arabic in the eleventh century, had copious information about Alexandria's past from the first century CE onward, see den Heijer, *Mawhūb Ibn Manṣūr*. Christian communities in Iraq and Iran likewise retained pre-Islamic local history, see Wood, *Chronicle of Seert*. The *Life of Mar Shabbay*, for example, preserved in Syriac, Soghdian and Arabic texts, celebrated a fourth century saint and recounted Alexander the Great's foundation of the city of Marw, see Wood, *Chronicle of Seert*, 171.

<sup>15</sup> Ḥamza, *Taʾrīkh*, 52–*61*; al-Bakrī, *Masālik*, vol. 1, 306–320; Munt, 'Ibn al-Azraq'; Ibn al-ʿAdīm, *Zubda*, 7–13, with similar information found scattered through his *Bughya*, *e.g*. vol. 1, 51.

have found and included information about the Greco-Roman pasts of the cities of the Levant, but chose not to. What was achieved by bypassing immediate pre-Islamic history and instead claiming knowledge of the original foundation acts of the ancient cities of Syria? And what evidence did Yāqūt, or his sources, actually have about this very distant past?

## **Who founded the ancient cities of Syria?**

Effectively, the evidence used to discover the original founder appears to have been the name of the city in question. In almost all cases when Yāqūt gave a founder, he named an eponymous figure who supposedly gave their name to the city. Thus we read that 'Latakia is an ancient city named for the man who built it', 'Jarash is the name of a man, namely Jarash, son of ʿAbd Allāh, son of …', and that '[Homs] was built by a man called Homs, son of al-Muhr, son of …'.16 The idea of a postulated original founder figure whose name survived as the name of a contemporary city, territory or people was of course a long-standing one in near-eastern literatures, from the Torah to ancient Greek legend.<sup>17</sup> Following a tradition of Hellenistic monarchs actually naming cities for themselves, the late antique near east saw a multitude of cities founded or re-founded and named for emperors in both the Roman empire and its Sasanian counterpart further east.18 As such, there was both literary precedent and real historical grounds for Yāqūt's idea that Syrian place names reflected ancient founder figures.

In some cases, Yāqūt's ascription of city-founders was based on received knowledge of the tradition of royal, eponymous foundations from the Hellenistic period onwards, for example in his discussion of the cities founded by Alexander the Great, or of the many of the cities eponymously named by Sasanian emperors in Iraq and Iran.19 In the case of the cities of Syria, however, Yāqūt's ascriptions of these types are limited to an assertion that Tiberias was founded by and named for a Roman emperor, only the latter being true. Though Yāqūt did have access to a snippet of genuine information from a Christian historian, Yaḥyā b. Jarīr al-Takrītī (d. 1103–1104), about Seleucus Nicator's foundations of Antioch, Latakia, Beroea, Edessa and Apamea, he did not tap consistently into residual, Christianised memories of the pagan stories of divine

<sup>16</sup> Yāqūt, *Buldān*, vol. 2, 127 (Latakia), 302 (Homs), vol. 5, 6 (Jerash).

<sup>17</sup> On Ancient Greek city foundation stories, Dougherty, *Poetics of Colonization*; focussing on ancient Greek eponymous founders, Malkin, 'What's in a name?'; comparing Greek and Hebrew foundation stories, Darshan, 'Origins of the foundation stories genre'.

<sup>18</sup> For Imperial Roman and Byzantine practices of city-founding, see Angelova, *Sacred Founders*. The *Šahrestānīhā ī Ērānšahr*, an ʿAbbāsid-era middle Persian gazetteer of cities based on Sasanian tradition, illustrates both the prevalence of Sasanian cities named for emperors and the textual memorialisation of this practice.

<sup>19</sup> For discussion of Alexander's cities, see Yāqūt, *Buldān*, vol. 1, 182–189. For eponymous Sasanian cities, see, for example, *ibid*., vol. 1, 377 (Bi-Ardashīr in Kirmān), 516 (Bahman Ardashīr in Iraq), vol. 2, 170 (Jundīsābūr in Iraq).

and semi-divine city founders in the Greek and Syriac chronicle traditions.20 As such, Yāqūt's accounts of city origins – in the main – do not appear to have rested on pre-Islamic tradition, but rather were Islamic-era innovations.

The form in which these founders are presented suggests not a mythic but rather a genealogical interest in moments of city foundation. Yāqūt's founders do not come with epic tales recounting their deeds, the origins of the settlers or their relationships to other cities, as occurs in the Greek tradition. Instead, the names of his founders are usually only accompanied by the further information of their genealogies, which link them into the descent of man from Noah after the Flood. Tracing the descent of a people from a particular, often eponymous, descendant of Noah, as grafted onto the Table of Nations of Genesis 10, was one of the fundamental late antique strategies of explaining the differentiation between peoples of different regions. Through it, ethnic difference was established on the basis of differing lines of descent which had been separated by the migration of peoples to different places.<sup>21</sup> This strategy was taken up by the earliest Arabic writers to whom we have access. Thus, when the ninth-century author al-Yaʿqūbī discusses the origins of the Chinese, he tells us:

[T]he first to rule China was Ṣāyin son of Bāʿūr son of Yaraj son of ʿĀmūr son of Japheth son of Noah son of Lamech. He had built a ship in imitation of Noah's ship, and accompanied by many of his children and kinsmen he set sail and crossed the sea. When he came to a certain place of which he approved, he settled there and called it al-Ṣīn [the Arabic name for China], naming it after himself. He had many offspring and his descendants multiplied.<sup>22</sup>

In this vision, ethnic groups are constituted by the offspring of a descendant of Noah whose name usually becomes an eponymous ethnic label. Indeed, one of the props of the construction of an overarching 'Arab' identity in the early Islamic period was a harmonised genealogy uniting the tribal lineages of Arabia into a common descent

<sup>20</sup> *Ibid*., vol. 4, 18 (Tiberias). The information on Seleucus' foundations is given through quotations of Yaḥyā b. Jarīr al-Takrītī, see *ibid*., vol. 1, 128, 266, vol. 2, 282. Ibn al-ʿAdīm, *Zubda*, *9*, or *Bughya*, vol. 1, 83–84, also quotes al-Takrītī, and says that Yaḥyā has a book (now lost) 'containing the times of the constructions of cities' ('kitabin lahu ḍimnahu awqātu -bināʾi -l-mudun'). Ibn Shaddād, *al-Aʿlāq al-khaṭīra*, quotes Ibn al-ʿAdīm naming Yaḥyā's book as *The Comprehensive Book on Chronology Including the Beginning of States, Formation of Kingdoms, Births of Prophets, and Times of Constructions of Cities, as well as Mention of Famous Events* (*al-Kitāb al-jāmiʿ li-l-taʾrīkh al-mutaḍammin dhikr mabdaʾ al-duwal wa-manshaʾ al-mamālik wa-mawālīd al-anbiyāʾ wa-awqāt bināʾ al-mudun wa-dhikr al-ḥawādtih al-mashhūra*), saying that it ran from the time of Adam to the Marwānid period, though I do not find this full reference in Ibn al-ʿAdīm's work. On such Christian interest in city foundations, see Robinson, 'Ibn al-Azraq', 23–25. On Yaḥyā, see Teule and Swanson, 'Yaḥyā b. Jarīr'. On the rationalisation of Greek mythic figures into human dynasts in the chronicle tradition see, Jeffreys, 'Malalas' World View', 62.

<sup>21</sup> Pohl, 'Strategies of identification', 33–34. Such explanations were often bound up with justifications of racial hierarchies, see Goldberg, *Curse of Ham*.

<sup>22</sup> On this strategy in general, Bashear, *Arabs and Others*, 71. This example of China is taken from al-Yaʿqūbī, *Works*, vol. 2, 481–482, in which see further examples at 271, 275, 277, 428, 432, 451, 479–480, 488, 493, giving the Noahic line of descent for mythical ancient Arabian peoples such as ʿĀd and Thamūd, biblical peoples such as the Amalekites, and more historical peoples such as the Greeks, Romans, Indians, Parthians, Daylamites, Khazars, Copts, Berbers, Ethiopians and Sudanese.

from Ishmael, whose place in the Noahic descent was clear.23 Given the role of lineages from Noah in establishing ethnic difference in Arabic-Islamic discourse, therefore, Yāqūt's choice to provide Noahic genealogies for city founders is most plausibly read as a comment on the ethnic origins of their populations. These genealogies situated the populations of these ancient cities as offshoots from broader peoples, their foundations a more granular instantiation of the processes of the dispersal of humanity after the Flood or the confusion of tongues at Babylon.

In some cases, this scheme clearly follows biblical precedent. For example, Sidon, the namesake of the city in modern Lebanon, had long beforehand appeared in the biblical Table of Nations as the son of Canaan, son of Ham, son of Noah. In Yāqūt's entry on the city he gives an eponymous founder with the same genealogy except that an extra figure has been inserted between Canaan and Sidon.24 Thus, the practice of establishing a Canaanite identity for the inhabitants of Sidon through a genealogy back to Noah was as old as Genesis. In other cases, however, the Arabic geographical tradition appears to have been doing more innovative things with these genealogies. Jarash, for example, was given a founder with an Arab genealogy leading back to the tribe of Kalb, perhaps providing a primordial aetiology for a particular connection which had developed between the city and this Arab tribe which had dominated the desert approaches to its region in the early Islamic period.25

In other instances, these genealogies appear to have been an attempt to lend ancient precedent to contemporary Islamic provincial conceptions and the geographical demarcations which underpinned them. In Yāqūt's entry on Aleppo/ Ḥalab, for example, one theory of the origin of the city is that it was founded by 'Amalekites' (seemingly standing in for the biblical Canaanites) who had been driven north from Jericho and Amman by the invading Israelites under Joshua. The claimed Amalekite founder of Aleppo, Aleppo Ibn al-Muhr, is said to have taken possession of the territory of Qinnasrīn – presented as virgin land – and founded Aleppo as a fortress for his people and their wealth. No doubt the contemporary appearance of the citadel of Aleppo as an archetypal stronghold contributed to this imagined origin. However, this narrative is particularly concerned with emphasising the association of the foundation of Aleppo with the territories of Qinnasrīn and al-ʿAwāṣim, the twin Islamic-era provinces of which Aleppo was by Yāqūt's time the metropolis.26 The

<sup>23</sup> Webb, *Imagining the Arabs*, 211–213.

<sup>24</sup> Compare Genesis 10:6, 15 with Yāqūt, *Buldān*, vol. 3, 437.

<sup>25</sup> Yāqūt, *Buldān*, vol. 2, 127. Webb, 'Kalb b. Wabara', on the Kalb's traditional dominance of the Wādī Sirḥān, the main access route from central Arabia to Syria.

<sup>26</sup> Qinnasrīn had been detached from the province of Homs under the Umayyads, while Hārūn al-Rashīd further separated al-ʿAwāṣim from Qinnasrīn. Qinnasrīn had been originally ruled from its eponymous capital – ancient Chalcis – but by the tenth century, as stated by the tenth century author al-Muqaddasī, *Best Divisions*, 132–133, Aleppo was Qinnasrīn province's 'seat of government and the location of the government offices'. Al-ʿAwāṣim province was frequently ruled alongside Qinnasrīn, see Elisséeff, 'Ḳinnasrīn', and geographers often present the couplet 'Qinnasrīn and al-ʿAwāṣim' as a unit, for example al-Yaʿqūbī, *Works*, vol. 1, 193, 220–221, 224.

eponymous founder figure Aleppo is introduced as the 'overlord of the virgin plot of land of Qinnasrīn' (*al-mustawlī ʿalā khiṭṭat Qinnasrīn*). Then, immediately after Aleppo founds his city we are told that the Amalekites 'then, after that, founded [the cities of] al-ʿAwāṣim', suggesting the subordination of the latter to the former.27 Historically, these provinces were geographical constructs of the caliphal administration of northern Syria in the seventh and eighth centuries, and the suzerainty of Aleppo over them a development of the tenth century. However, by presenting an Israelite-era story in which the construction of Aleppo was part of the foundation of those areas' very existence as owned and organised territory, Yāqūt's narrative coded Aleppo's contingent position as the metropolis of Qinnasrīn and al-ʿAwāṣim as a primordial reality. The Greco-Roman foundations or phases of the cities which made up these provinces – such as Aleppo (Beroea), Qinnasrīn (Chalcis-on-Belus), Antioch, Manbij (Hieropolis), Bālis (Barbalissos), Alexandretta, etc. – were not how Yāqūt and his informants sought to establish the origins of grand cities, geographical territory or provincial hierarchies. Instead, they turned to the Israelite past28, a past which perhaps provided more flexible resources with which to present the pre-eminence of the city of Yāqūt's patrons in the Ayyūbid administration based in Aleppo.

We find a similar strategy at play arguing for a Israelite-era origin for the main provincial divisions (*ajnād*) of Syria, along with the uniquely prestigious site of Jerusalem. In Yāqūt's entry on Aelia (Arabic: *Īliyāʾ*), a name used for Jerusalem in some medieval Arabic texts, rendering its Roman name of Aelia Capitolina, Yāqūt wrote that:

<sup>27</sup> Yāqūt, *Buldān*, vol. 2, 283: 'The overlord of the virgin plot of land [*khiṭṭa*] of Qinnasrīn was Aleppo b. al-Muhr, one of the sons of al-Jānn b. Mukannif [vocalisation speculative] of the Amalekites. He founded a city which was named after him … The Amalekites had been displaced from those regions [Jericho and Amman] to the land of Zobah, which is [the province of] Qinnasrīn. They built Aleppo and made it a fortress for themselves and their wealth. Then, after that, they founded [the cities of] al-*ʿAwāṣim*: كان ّ ّ ف من العماليق فاختط مدينة سميت به ... ارتفع ا العماليق عن تلك ّ لمستولي على خطة قنسرين حلب بن المهر أحد بني الجان بن مكن ّسرين وبنوا حلب وجعلوها حصنا ألنفسهم وأموالهم ثم اختطوا بعد ذلك العواصم a of nature The *.* الديار إلى أرض صوبا وهي قن *khiṭṭa* as land being claimed without prior possession, is emphasised in the classical definitions, Lane, *Lexicon*, vol. 1, 760. It seems likely that he is talking about the province of Qinnasrīn, rather than the individual settlement as it is described as the equivalent of the 'land of Zobah' (an Aramean kingdom mentioned in Samuel and Kings in the bible). I translate '[the cities of] al-ʿAwāṣim', because al-ʿAwāṣim, the name of the province, is the plural form of *ʿāṣim*, literally meaning 'the defenders', referring to the cities of north-western Syria which functioned as military bases near the caliphate's border with the Byzantines. Thus, the narrative is saying that the Amalekites founded the cities which made up the province, whose defensive role in biblical times (against the expanding Israelites) mirrors that of the Islamic era (against the Byzantines). with the Romans. Thus, the narrative is saying that the Amalekites founded the cities up whose role (against expanding Israelites) mirrors that of the Islamic era (against the Romans).

<sup>28</sup> By the Israelite past, and the Israelite era, I refer to the ancient period of sacred history recounted extensively in the Torah/Old Testament, many of the major stories or figures of which also feature in the Qurʾān and more extensively in the Islamic genre of 'stories of prophets' ('*qiṣaṣ al-anbiyāʾ*'). I include strictly speaking pre-Israelite history, *i.e*. the time of Noah or Abraham (which might be referred to as Patriarchal), under this rubric for convenience. I use this term because it reflects the Arabic labelling of the monotheists of these times as the Banū Isrāʿīl and the stories about them which were derived from Jewish and Christian literature as Isrāʿīliyyāt.

It has been said that Aelia was named for its builder, Aelia, son of Iram, son of Shem, son of Noah. Aelia was the brother of Damascus, Homs, Jordan, and Palestine.<sup>29</sup>

The Roman origins of this name for Jerusalem – derived from the *nomen gentile* of the emperor Hadrian, who refounded the city – are totally forgotten here. Nor, curiously, are its origins discussed in terms of its relationship to either cosmology or to the patriarchs David and Solomon, typical Islamic means of talking about the origins of Jerusalem.30 Instead, Aelia is claimed to be the name of one of Noah's great-grandsons who founded the city. Therefore, the foundation of the city and one of its names are projected back into the depths of Israelite-era history.

The claim not only establishes Aelia as an ancient city populated by descendants of Noah, but also helps to construct the territories of Syria as a united entity with a common origin. Aelia is said to have been one of five brothers, with each of the other four giving his name to one of the *jund*s, the four provinces which made up the original administrative districts of Syria. These administrative districts were, according to the Islamic historical tradition, established by the first rightly guided caliph Abū Bakr, and continued to be used or referenced into the Mamluk period.<sup>31</sup> By claiming that these provinces – alongside the symbolically significant city of Aelia/Jerusalem – were originally founded and populated in ancient times by a band of brothers, this origin story argues that the borders and administrative divisions of the Islamic world were not just arbitrary, accidental, political configurations. Instead, it claims that the territorial borders of the Islamic world, as established by the caliphs, reflected an inherent, essential reality which dated back to time immemorial. Therefore, even as the Roman pasts of the cities of Syria were being left aside, new ideas about the ancient origins of the cities of the region were emerging which were a fundamental part of the Islamic construction of the space and boundaries of the Levant and beyond.<sup>32</sup>

This is not to say that none of the traditions which Yāqūt reproduced had a concept of the ancient Roman city. He did, in fact, claim that some cities of the eastern Mediterranean had a Roman origin. However, rather than corresponding to our own vision of Greco-Roman antiquity, this was likewise articulated through a post-Noahic

إنما سميت إيلياء باسم بانيها وهو إيلياء بن إرم بن سام بن نوح عليه السالم وهو أخو دمشق 293: 1, .vol ,*Buldān* ,Yāqūt <sup>29</sup> وفلسطين وأردن وحمص .

<sup>30</sup> For which, Antrim, *Routes and Realms*, 48–55; Borrut, 'La Syrie de Salomon', 109–110. These issues receive more attention in Yāqūt's work under the entry for al-Maqdis, another name for Jerusalem, see Yāqūt, *Buldān*, vol. 5, 166–172.

<sup>31</sup> Sourdel, 'Djund'. Yāqūt repeats this passage, or something similar, in his entries on Damascus, Jordan, Palestine and Homs, see *Buldān*, vol. 1, 147, vol. 2, 302, 464, vol. 4, 274.

<sup>32</sup> This strategy of using table of nations type material to explain the origins of provinces as well as peoples appears to date back to some of the earliest surviving Arabic histories. According to the sources of Ibn ʿAbd al-Ḥakam (d. 871), the territory of Egypt was established by being divided between four great-great-grandsons of Noah, who each lent their name to major cities, see Ibn ʿAbd al-Ḥakam, *Futūḥ*, 9. At the beginning of his history Al-Dīnawarī (d. 896) recounts various grandsons of Shem dispersing from the confusion of tongues at Babylon, some whose name related to ethnonyms such as al-Rūm (Romans) or Hayṭal (Huns), but others bearing the name of familiar Arabic regional divisions such as Khurāsān, Fārs, Armīn and Kirmān, see Al-Dīnawarī, *Ṭiwāl*, 3.

vision of the origins of human settlement and responded to Islamic-era territorial boundaries. In Yāqūt's discussion of Antioch (Arabic: *Anṭākiya*), for example, he relays a tradition which says that the eponymous founder of the city was a woman called Anṭākiya, whose father was a great-grandson of Noah called al-Rūm.<sup>33</sup> Al-Rūm is of course the Arabic term for 'the Romans', used for both the classical Romans and the Medieval Romans of their own time. Here, therefore, Yāqūt claimed an ancient Roman foundation of Antioch by attaching its eponymous founder to an established Arabic genealogical idea about the Romans' place in the descent from Noah.<sup>34</sup>

This was not an exceptional attribution for the case of Antioch, however, but part of a systematic explanation of the origin of cities in and beyond the northern and north-western borders of Syria. In the *Muʿjam al-buldān* the foundations of at least eleven eponymously named cities are attributed to different children of al-Rūm, almost all clustering around the traditional Arab-Roman frontier zone between Syria-Mesopotamia and Anatolia (Figure 16.1).<sup>35</sup> It is these cities which Yāqūt presented as having an original Roman pedigree, though they map suspiciously well onto the cities which had been reclaimed by the Medieval Romans during their conquests of the tenth and eleventh centuries, perhaps providing a clue as to the time when this idea originated.36 Again, therefore, Yāqūt's assessment of the ancient origins of cities appears to have little to do with the observation of remains of what we consider archetypically Greco-Roman monuments, which were just as prevalent in the Syrian heartlands as this region. Rather, it appears to have stemmed from Arabic traditions which reified the territorial boundaries of their day by invoking supposed knowledge of the deep past in order to understand these boundaries as reflecting ancient ethnic distributions of peoples.

One implication of this understanding is that it argues for an autochthonous population of Syria which was inherently non-Roman (in the sense of the Arabic 'al-Rūm', used as a label for a people). As we have seen, Yāqūt reproduced claims

<sup>33</sup> Yāqūt, *Buldān*, vol. 1, 266: 'The first who built [Antioch] and lived there was Antioch, daughter of al-Rūm, son of al-Yafar, son of Shem, son of Noah – peace be upon him – [who was] the sister of Antalya

they that was One .authors Arab by proposed Romans the for genealogies different several are There <sup>34</sup> . أول من بناها وسكنها أنطاكية بنت الروم بن اليقن *)*اليفز*(* بن سام بن نوح عليه السالم أخت أنطالية :'[Attaleia .:Gr[ descended from the biblical Esau, a continuation of a late antique Rabbinic theme, see Hacohen, *Jacob & Esau*, 55–90. Yāqūt's genealogy seems to be derived from that of al-Dīnawarī, *Ṭiwāl*, 3, or a common source, where it is said that the Romans descended from a certain al-Rūm b. al-Yafar b. Shem b. Noah.

<sup>35</sup> Apart from Antioch, see Yāqūt, *Buldān*, vol. 1, 270 for Anṭāliya (Attaleia); vol. 1, 328 for Bālis (Barbalissos); vol. 2, 359 for Kharshana (Charsianon); vol. 3, 106 for al-Ruhā (Edessa); vol. 3, 131 for Zibaṭra (Sozopetra); vol. 4, 28 for Ṭarsūs (Tarsos); vol. 4, 41 for Ṭimmīn (? said to be in the *bilād al-Rūm*, *i.e*. Roman territory); vol. 4, 158 for ʿAmmūriya (Amorion); vol. 5, 144 for Maṣṣīṣa (Mopsuestia); vol. 5, 398 for Ḥiraqla (Heraclea Cybistra?).

<sup>36</sup> In the face of a disintegrating Abbasid caliphate and before the reassertion of the more powerful Fatimid and Buyid states projecting power into the border region from Cairo and Iraq, the Medieval Romans came to dominate this area, through direct control, imposition of protectorates or treaties, or exaction of tribute, see Whittow, *Making of Byzantium*, 310–334. Things changed radically with the incursions of Turkic groups into Anatolia following the battle of Manzikert in 1071, vanquishing Roman rule from this area and providing a possible cut-off point for the emergence of these traditions.

*Fig. 16.1: Cities in Anatolia and Syria which Yāqūt claims were founded in ancient times by the Romans. Map after M. Whittow,* Making of Orthodox Byzantium*. (Design: Beth Clark, made with Natural Earth. Free vector and raster map data @naturalearthdata.com).*

that the territories of the provinces of Syria were settled by the sons of Iram, grandson of Noah. Al-Dīnawarī (d. *c*. 894–903), an early witness to Arabic traditions about the descendants of Iram, wrote that Iram was the eldest son of Shem, that Iram's children were known as 'the first Arabs' ('al-ʿArab al-ūlā'), that they were endowed with the Arabic language at the confusion of tongues, and that they went on to settle the Arabian peninsula.37 This idea clearly remained current, as through the *Muʿjam* Yāqūt repeated various authorities who explained Arabian place-names through original, eponymous founders descended from Iram.38 Thus,

<sup>37</sup> The Qurʾān mentions the word Iram ambiguously, in relation to the ancient Arabian people of ʿĀd, and while it ultimately became associated with an ancient city, early Islamic exegetes tended to conceive of Iram as an ancient people, Webb, 'Iram'.

<sup>38</sup> For Iramite genealogies giving founders of places or regions in Arabia see Yāqūt, *Buldān*, vol. 2, 151 (Jildhān, near al-Ṭāʾif); vol. 2, 410 (Khaybar); vol. 3, 139 (Zarūd, al-Shuqra, al-Rabadha); vol. 3, 331 (Sharāf); vol. 5, 354 (Wāqiṣa); vol. 5, 356 (Wabār); vol. 5, 430 (Yathrib). These Yāqūt sometimes attributes to the Abbasid-era polymath Ibn al-Kalbī (d. *c*. 819–821) (who in turn invoked the authority of an al-Sharqī), sometimes to the grammarian Abū -l-Qāsim al-Zajjājī (d. *c*. 948–950). Al-Zajjājī was a grammarian whose extant works do not include the passages attributed to him by Yāqūt. Ibn al-Kalbī, on the other hand, was a famed genealogist and polymath who also wrote several lost geographical works, including one

Yāqūt's tradition that identified Aelia, Damascus, Jordan, Palestine and Homs as sons of Iram made Syria, like Arabia, a part of the division of the world originally apportioned to and inhabited by the 'first Arabs'. In this model the Romans (al-Rūm) who ruled Syria at the time of the Muslim conquests were a separate ethnicity to the indigenous population, an alien group who were driven back to their original territories by the caliphs' armies, a similar idea to that expressed in early Muslim historical traditions about the distinct fate of the 'Romans' and indigenous 'Copts' in Egypt.39 Traditions about the Roman foundation of cities in Anatolia and northern Syria explained where those Roman territories started. Thus, the genealogical ideas expressed in these city foundation narratives were a part of the process of imaginatively de-Romanising the bulk of Syria and contrarily reconceiving of it as part of an Arab homeland.

### **The cities and the prophets**

As we have seen, Yāqūt's focus on the foundational moment of ancient cities in Syria firmly locates their origins in the time of the descendants of Noah and the world of sacred history first recounted by the Torah and frequently referenced by the Qurʾān. This focus on a deep, Israelite-era antiquity allowed him to identify many of the imposing Greco-Roman monuments which still survived in the cities of the region as part of this Israelite heritage.

In the case of some cities this tradition of linking foundation moments to sacred history used elements of the bible itself. Writing about the foundation of Amman, for example, Yāqūt explicitly informed his readers that he was following a Jewish informant when he retold an adapted version of Genesis 19:37–38. This biblical story originally explained that the names of the Ammonites and Moabites – antagonists of the Israelites in the holy land – were derived from Hebrew terms referencing those people's incestuous descent from Lot and his two daughters. Yāqūt's Arabic version introduces an uncle to the proceedings and so Lot's daughters name their respective sons Maʾāb (supposedly a shortening of *min ab*, meaning 'from a father') and Amman (playing on the Arabic *ʿamm* for uncle). His story continues that later in life each of these brothers built a city in Syria and named it after himself.<sup>40</sup> Thus the originally biblical aetiology for the name of the Ammonite people was repurposed by Yāqūt as an aetiology for the name of the city of Amman.

referred to by Yāqūt in his introduction called *The Etymology of Places* (*Ishtiqāq al-buldān*), making him a plausible source of such genealogical explanations for place names (see also Ibn al-Nadīm, *Fihrist*, 126, who mentions that Ibn al-Kalbī wrote two differently sized books named *The Countries* (*Kitāb al-Buldān*) as well as one called *The Division of the Lands* (*Kitāb qismat al-arḍīn*)). If Yāqūt's ascription of these quotations to Ibn al-Kalbī, probably indirectly via an intermediary source, are correct then the tradition of an Iramite settlement of Arabia would go back to the eighth century.

<sup>39</sup> See Coghill, 'Minority representation' and Zychowicz-Coghill, *Conquests of Egypt*, 65–72, 281–282.

<sup>40</sup> Yāqūt, *Buldān*, vol. 4, 151. The same pun for Moab also worked in the original Hebrew, though the pun for Ammon is related to the Hebrew word for nations.

Once Yāqūt had located Amman's origins in Israelite time, it became easy for him to explain the remains of the Greco-Roman city as vestiges of that past. Thus, he wrote:

The palace of Goliath is on the mountain overlooking the city. In the city is the tomb of Uriah the prophet, over which is a mosque, and the theatre of Solomon son of David.<sup>41</sup>

By Goliath's palace it seems likely that he meant the remains of the Roman Temple of Hercules, on the tell overlooking Amman.<sup>42</sup> The identification of the tomb of Uriah the prophet was probably derived from the biblical story of the death of Uriah the Hittite at Rabbah-Ammon, *i.e*. Amman, upgrading Uriah the Hittite to Uriah the Prophet in order to enable a local cult.43 The theatre (*malʿab*) which he mentioned must have been the Roman theatre which remains such an imposing presence in the centre of Amman. Here, a definitively Roman ruin was resignified as a construction of King Solomon. It is well known that early Muslims represented Syria as a holy land through emphasising the presence of sites linked to Qurʾānic people or places.44 As we see from the case of Amman, this tendency could build on existing Jewish and Christian traditions and associations, while adapting them to the new Arab-Islamic discourse in which it was being presented. An important aspect of this which we see here is the identification of Greco-Roman remains as remnants of Israelite history.

This attribution of ancient Greco-Roman buildings to Solomon and his *jinn* is a repeated theme running through Yāqūt's discussion of classical remains, building on a long history of such attributions in Islamic texts.45 The impressive nature of such constructions became attributed not to Greco-Roman engineering but to the *jinn* whom the Qurʾān states that God ordered to obey Solomon.<sup>46</sup> Yāqūt credited the magnificent colonnades and ancient buildings at both Damascus and Palmyra to Solomon, as well as a particularly spectacular and curative bath complex near Tiberias, surely a monumental Roman bathhouse.<sup>47</sup> Baalbek, perhaps the most remarkable of

<sup>41</sup> السالم عليه داود بن سليمان وملعب مسجد وعليه السالم عليه النبي أورياء قبر وبها عليها يطل جبل على جالوت وقصر *.* 

<sup>42</sup> It is possible that the ruined remains of the Umayyad palace were also considered part of this.

<sup>43</sup> 2 Samuel 11 for Uriah the Hittite, and Jeremiah 26:20–3 for the prophet.

<sup>44</sup> Cobb, 'Virtual sacrality', 44–50.

<sup>45</sup> Borrut, 'La Syrie de Salomon', 110–113, notes Roman or other ancient remains attributed to Solomon at Jerusalem, Palmyra, Lydda, Baalbek, Persepolis, Hebron, Tiberias, Amman and Ascalon on the basis of historical links noted in the bible or appreciation of the aesthetic or monumental features of the buildings there.

<sup>46</sup> On the Qurʾān's remarkable image of Solomon, see Mottahedeh, 'Eastern travels of Solomon', 247–248.

<sup>47</sup> Yāqūt, *Buldān*, vol. 2, 199, claims that Solomon built a lengthy colonnade at Damascus which is known as the Jayrūn, the name of the *jinn* who built it for him, though he does then alternatively report traditions attributing it to ancient descendants of Iram. Damascus itself is associated with almost the entire cast of the Arabic imagination of antiquity, from Iramites, other descendants of Shem, the grand antagonist of ancient Iranian epic al-Ḍaḥḥāk, the Arabian prophet Ḥūd, Canaanites and other biblical figures such as Noah, Abraham and Nimrod, *ibid*., vol. 2, 463–470. For Palmyra, see *ibid*., vol. 2, 17, where the attribution to Solomon is tempered by the locals' statement that people will say any marvel with an unknown builder gets attributed to Solomon, but their city is twice as old as anything built by Solomon. For the baths near Tiberias see *ibid*., vol. 4, 17–18.

the Greco-Roman remains in the eastern Mediterranean, he identified as a palace of Solomon given to Bilqīs, the Queen of Sheba, as her bride price. Yāqūt further remarked that its ancient temple complex – which had been transformed into an ecclesiastical complex in late antiquity and then a fort after the Muslim conquest – housed a Station of Abraham as well as graves of the Israelite descendants of Isaac. He also noted the presence in Baalbek of the grave of the prophet Elijah, an association which had presumably arisen from the Qurʾānic recounting of Elijah's struggle against Baal-worship and the name of the city.<sup>48</sup> Therefore we see that – in Yāqūt's encyclopaedia at any rate – the ancient monuments of Syria had undergone a thorough reinterpretation which redirected the educated viewer's contemplation not to the Greco-Roman past, but to the sacred history of the Israelites found in the Qurʾān and biblically-derived Arabic 'stories of prophets' ('qiṣaṣ al-anbiyāʾ').

Western academic literature has traditionally had little time for these readings of Greco-Roman monuments, viewing them through the lenses of error, ignorance or superstition. While acknowledging that professional archaeology has made great progress in understanding the original development and functions of these sites, it is worth also engaging with their other interpretative histories, which saw these structures incorporated into other systems of meaning.49 We may consider these resignifications of classical ruins as evidence of alternative premodern archaeologies, through which cultivated observers such as Yāqūt, as well as local populations, sought to read and curate the ancient heritage of their cities in productive ways that made sense according to the ruling epistemologies of the day.50 One feature of these reassociations of Greco-Roman sites with prophets or Israelite rulers was that it allowed the incorporation of spectacular elements of the ancient built environment into communal practices of worship or other forms of devotion, as indicated by the numerous references to mosques (*masjids*), stations (*maqām*s) or graves (*qabr*s) associated with Israelite figures at these places. The Temple of Bel at Palmyra, infamously destroyed by ISIS militants in 2015, had been a mosque from early Islamic times until French archaeologists reconverted it into a classical Roman temple/ heritage site.51 Yāqūt's identifications of Roman monuments as Israelite-era remains perhaps tell us something about how the earlier, living form of heritage practiced by those worshipping at sites such as the Temple of Bel/Palmyra mosque envisaged these reused elements of their urban environments. Rather than being remnants of a Roman civilisation that should be considered alien, in Yāqūt's vision, at least, they should be read as remnants of meaningful prophetic stories which were relatable to the inhabitants of these cities. This, perhaps, allowed these inhabitants to identify with and take pride in the ancient heritage of their cities within an Islamic scheme.

<sup>48</sup> Yāqūt, *Buldān*, vol. 1, 454. For Elijah and Baʿal, see Q 37:123–127, and for early associations with the site of Baalbek see al-Ṭabarī, *Tafsīr*, vol. 21, 95–99,

<sup>49</sup> For this problem, see the chapters in this volume by Benjamin Anderson and Elizabeth Key Fowden.

<sup>50</sup> For the distinction between professional or modernist and premodern archaeologies I follow Hamilakis, 'Indigenous archaeologies', 49–53, 63–64.

<sup>51</sup> Mulder, 'Imagining localities of antiquity', 231–233.

### **Ancient ruins, moral lessons**

Pride, however, was not the only, or indeed the explicit response that Yāqūt expected from the Muslim viewer of such sites. While Yāqūt's geography remains invaluable as a reference work for modern scholars, we must not mistake our frequent use of it as a neutral repository of facts for its intended function. Yāqūt was quite open, at the beginning of his book, about the moral framework in which he expected his reader to think about the settlements he would describe. The opening sentence of his book reads:

God has guided His servants to take homes for themselves and to make solid buildings and dwelling places. Thus, they have raised buildings and built towns, carved chambers from the mountains, and dug wells and cisterns. He has instilled their desire to raise high what they have built, to make firm the constructions they have erected and supported upon columns, as an admonition to the neglectful and a lesson to those who remain after them.<sup>52</sup>

Here there is no Aristotelian sense that city-dwelling is inherent to sociable man, nor a Khaldūnian association of cities with the exertion of sovereign power.<sup>53</sup> Rather, Yāqūt introduces his work by arguing that man's impulse to build towering, lasting structures is a divinely inspired means of prompting reflection in later generations. Yāqūt immediately clarifies, and justifies, this interpretation of human settlements with a quotation from the Qurʾān:

Do they not travel through the earth and see what was the End of those before them? They were more numerous than these and superior in strength and in the traces [*āthār*] [they have left] in the land: Yet all that they accomplished was of no profit to them.<sup>54</sup> [Q 40:82]

The grand scale of the *āthār* – the Arabic term used for ruined monuments – left by mighty ancient peoples is a permanent, visible reminder of the transience and vanity of worldly power.

This Qurʾānic injunction is one of a repeated set of commands made in the Qurʾān that believers should 'journey in the land' ('al-sayr fī-l-arḍ') in order to contemplate and take heed of the consequences of the impiety of past generations: the divine destruction of populations and their dwellings.<sup>55</sup> Yāqūt drew on this Qurʾānic discourse, the origins of what has recently been dubbed an 'Islamic theology of ruins',

وهدى عباده إلى اتخاذ المساكن وإحكام األبنية والمواطن فشيدوا البنيان وعمروا البلدان ونحتوا من 7: ,i ,*Buldān* ,Yāqūt <sup>52</sup>

 <sup>.</sup>d (Ṭūsī see ,century thirteenth Islamic the in writers to available was sense Aristotelean an Such <sup>53</sup> . الجبال بيوتا واستنبطوا آبارا وقلوتا وجعل حرصهم على تشييد ما شيدوا وإحكام ما بنوا وعمدوا عبرة للغافلين وتبصرة للغابرين 1274), *Nasirean Ethics*, 190, who caps his discussion of man's necessary inter-reliance leading to urban life with the statement, 'This is what the Philosophers mean when they say that Man is naturally a city-dweller'. In contrast, see the opening statements of chapter four of Ibn Khaldūn, *Muqaddimah*, 263, that 'Towns and cities are secondary products of royal authority'.

<sup>54</sup> Yāqūt, *Buldān*, vol. 1, 7. I use Yusuf Ali's Qurʾān translation.

<sup>55</sup> On the city in the Qurʾān see Khalidi, 'Some classical Islamic views', 266, and specifically on contemplating destroyed traces, Rubin, 'Remnant', a topic which is also the starting point for Feener, 'Muslim cultures and pre-Islamic pasts', 23–24.

quoting two further instances (Q 22:46, 6:11) and glossing them as God's commands to 'see how their mansions have become razed and sand-swept, their monuments obliterated, and their lights extinguished in order to punish them for discarding His commands and contravening His restrictions'.56 Thus Yāqūt introduced the cause of the destruction which proves the vanity of worldly power: God's wrath at those falling away from his instructions.

Therefore a providential historical scheme of capricious mankind falling away from God's covenant, derived from the Qurʾān, was forwarded by Yāqūt as a universal meaning for the ruins encountered by the Muslim.57 By invoking this discourse, Yāqūt argued for the benefit of contemplating such ancient remains through the lens of moral reflection on the consequences of impiety. He thus provided his readers with a clear interpretative framework for reading the grand vestiges which they might encounter, either through their own travels or when reading his work. Arabic stories of the Israelites repeatedly straying from the true path in Qurʾān, *ḥadīth* and 'stories of prophets' literature – derived of course from the same biblical theme – made up the richest body of illustrations of the Qurʾānic cycle of covenant and its subsequent betrayal and punishment. These stories, of course, geographically fitted the Syrian context perfectly.<sup>58</sup> Thus the fabulous remains of the Greco-Roman past in Syria were resignified as remnants of the rule of Solomon or other events of Israelite history, becoming physically incarnate invitations to meditate upon the contemporary implications of the ancient working out of God's providence in the tales of the Israelites in the holy land.

#### **Conclusions**

Yāqūt al-Ḥamawī justified writing his great encyclopaedia by claiming – amongst other things – that he was enabling Muslim readers of his work to vicariously fulfil God's command to travel the lands and contemplate the providential fate of past peoples. While this is not, in fact, one of the most prevalent themes of his book, the origins of cities and their ancient remains did have a programmatic role to play in his work in general, and his presentation of the cities of Syria in particular. By providing Noahic genealogies for eponymous city-founders, Yāqūt claimed knowledge of the origins of the cities of Syria. By focusing on those ancient founding moments, to the

<sup>56</sup> For the theology of ruins, see Devecka, *Broken Cities*, 64. Yāqūt, *Buldān,* vol. 1, 7:درست كيف ديارهم إلى انظروا qurʾānic this uses Yāqūt . وإلى آثارهم وأنوارهم كيف انطمست عقوبة لهم على اطراح أوامره وارتكاب زواجره 'command clearly rendering an obligation' ('ظاهرا الوجوب يقتضي أمر ('to justify his composition of his work, which allows those for whom 'the means to [journey in the land and] see may prove difficult' َسباب النظر') َّر أ تتعذ وقد ('to cultivate their awareness of God's providence as manifested in the wider world, see *ibid*., vol. 1, 7–8.

<sup>57</sup> Termed 'the myth of covenant, betrayal, and redemption', by Humphreys, 'Qurʾānic myth and narrative structure', 277–278.

<sup>58</sup> On the role of the Israelites in early Islamic self-fashioning, see Rubin, *Between Bible and Qurʾān*, though with scepticism towards his clear chronological development of attitudes to the Israelites, which could just as easily represent contesting perspectives within the Muslim community.

exclusion of the urban pasts that had immediately preceded Islam, he proposed a vision of the pre-Islamic history of these cities which had little place for the notion that the cities conquered by the Muslims in their seventh century campaigns were fundamentally Roman.

Rather, Yāqūt's origin stories integrated these cities into long-established patterns of understanding the peopling of the earth after the Flood by the descendants of Noah. Some of these traditions did conceive of some eastern Mediterranean cities as having primeval Roman origins, claiming that they were founded by the children of a common ancestor of the Romans, al-Rūm, a great-grandson of Noah. However, these Roman-founded cities were uniquely clustered in the *thughūr*, the border region in the north of Syria which had been retaken by the Medieval Romans in the tenth century. As such, this conception of the Roman-ness of these cities seems a reification of contingent medieval middle eastern political boundaries, claiming that they reflected a primordial division of the earth between peoples.

By contrast, Yāqūt repeated traditions which claimed that the Syrian provinces proper (the *jund*s) had been first settled by a group of sons of Iram, son of Shem, making Syria, like Arabia, a part of the land settled by the 'first Arabs' ('al-ʿArab alūlā'). Thereby such traditions naturalised as a return to its origins Syria's transition into part of the Arab world through the long subsequent processes of Arabisation following the seventh century conquests. Other schemas which linked together groups of ancient founders of cities or territories also did so to provide ancient precedent for contemporary Islamic territorial conceptions, such as the subordination of the provinces of Qinnasrīn and al-ʿAwāṣim to the metropolis of Aleppo, Yāqūt's adopted home. Therefore we see that Yāqūt's pre-modern 'archaeology' – in the strict sense of 'knowledge of beginnings' – of Syrian cities was thoroughly bound up with medieval Islamic discourses of territory and ethnicity which creatively adapted the resources provided by the Arabic reception of Israelite history.

Beyond theoretical knowledge of city-origins, Israelite history also informed Yāqūt's readings of the concrete remains of antiquity in the cities of Syria. Time and again Yāqūt presents what we think of as Greco-Roman monuments as remnants of the Israelite-era past of the region. As such, they could be celebrated as the signs of the prophets who had practiced the true religion in the Syria before the Muslims brought it back again with their conquests. These remains, though, also carried the warning of the destruction that would come to those who opposed God's community – such as Goliath's ruined palace in Amman – or to that community when it strayed from the wrong path, a continuous theme in Qurʾānic and other Islamic stories of the Israelites. Read thus, they fitted the moral framework which Yāqūt set out at the beginning of his book. By ascribing the ruined remnants of these cities' antiquity to Israelite times and figures, Yāqūt inscribed into the Greco-Roman remains of Syria a quite different meaning to the ones we take from them: the command to contemplate God's power and the consequences of impiety.

### **Acknowledgements**

This chapter has been researched and written as part of the project 'Impact of the Ancient City', University of Cambridge, which has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 693418). I would like to thank the much-missed Andrea Zerbini, who invited me to give an early version of this paper as a public lecture at the Council for British Research in the Levant, and Carol Palmer, who hosted me graciously at the CBRL. Thanks also are due to Amira Bennison, Sam Ottewill-Soulsby and Philip Wood for their comments while workshopping this paper, and to Elizabeth Key Fowden for sharing with me some of her ideas, questions and ever-perceptive readings.

## **Bibliography**


Darshan, G., 'The origins of the foundation stories genre in the Hebrew Bible and ancient eastern Mediterranean', *Journal of Biblical Literature* 133 (2014): 689–709.


Gilliot, C., 'Yāḳūt al-Rūmī', in P. Bearman, T. Bianquis, C. Bosworth, E. van Donzel and W. Heinrichs (eds.), *Encyclopaedia of Islam*, Second Edition (Brill Online) [available at: dx.doi.org/10.1163/1573- 3912\_islam\_COM\_1356] (accessed 29 September 2021).

Goldberg, D., *The Curse of Ham: Race and Slavery in Early Judaism, Christianity, and Islam* (Princeton, 2003). Hacohen, M., *Jacob & Esau: Jewish European History between Nation and Empire* (Cambridge, 2019).

Hamilakis, Y., 'Indigenous archaeologies in Ottoman Greece' in Z. Bahrani, Z. Çelik and E. Eldem (eds.), *Scramble for the Past: A Story of Archaeology in the Ottoman Empire, 1753–1914* (Istanbul, 2011), pp. 49–69.

Ḥamza al-Iṣfahānī, Ibn al-Ḥasan, *Taʾrīkh sinī mulūk al-arḍ wa-l-anbiyāʾ* (Beirut, 1961).


Ibn al-ʿAdīm, Kamāl al-Dīn ʿUmar b. Aḥmad al-ʿUqaylī, *Bughyat al-ṭalab fī taʾrīkh Ḥalab*, ed. S. Zakkār, 12 vols. (Damascus, no date).

Ibn al-ʿAdīm, Kamāl al-Dīn ʿUmar b. Aḥmad al-ʿUqaylī, *Zubdat al-ḥalab fī taʾrīkh Ḥalab*, ed. K. al-Manṣūr (Beirut, 1996).

Ibn Khaldūn, Walī al-Dīn Abū Zayd ʿAbd al-Raḥmān b. Muḥammad al-Ḥaḍramī al-Ishbīlī, *The Muqaddimah*, tr. F. Rosenthal (Princeton, 2015).


Rubin, U., *Between Bible and Qurʾān: The Children of Israel and the Islamic Self-Image* (Princeton, 1999).


## Chapter 17

## (Re)constructing Jarash: History, historiography and the making of the ancient city

*Louise Blanke*

'The past is not a natural growth but a cultural creation.'1

#### **Introduction**

Elena D. Corbett opens her volume on *Competitive Archaeology in Jordan* by describing the 1999 banknotes that were issued in celebration of King Abdullah II of Jordan's accession to the Hashemite throne.2 The banknotes bear images, among others, of the Hijaz Railway, the first Parliament and the first Hashemite Palace in Ma'an. These images replaced depictions of the archaeological and touristic sites of Jarash and Petra, which had appeared on Jordanian banknotes issued since the creation of the Hashemite Kingdom of Transjordan in 1946. The reasoning behind the omission of Jarash and Petra in favour of the new images was articulated by Ammar Khammash, the artist and architect who designed the notes, in an interview with Corbett. In his words, 'Jerash was Roman, or Arab under the Romans', whereas 'the Hijaz Railway is more us.'3

Each of the new notes bears a portrait of one of the five Hashemite rulers (from 1908 to the present) alongside a depiction of a landscape or building relevant to their time. King Abdullah I (r. 1921–1951), for example, is depicted on the obverse of the five-dinar note with the Ma'an palace on the reverse, while his successor King Talal (r. 1951–1952) is depicted on the ten-dinar note with Jordan's first parliament after independence on the reverse.4

The erasure of Jarash and Petra from the banknotes, a medium whose message would travel widely across the entire socio-political spectrum, reflects the decision

<sup>1</sup> Assmann, *Cultural Memory and Early Civilization*, 33. 2 Corbett, *Competitive Archaeology*.

<sup>3</sup> Corbett, *Competitive Archaeology*, 1. 4 Corbett, *Competitive Archaeology*, 1–3. For images of the banknotes, see www.atsnotes.com/catalog/ banknotes/jordan.html.

to downplay the record of western archaeologists who were concerned with the retrieval of a Biblical or Greco-Roman past.<sup>5</sup> In the second half of the twentieth century, Jarash and Petra – two of the most impressive archaeological sites in Jordan – had become the historical anchor and the public face of the new Jordanian nation.6 The sites appealed to foreign (European and American) scholars and travellers, their monuments were pictured on stamps and banknotes, and artefacts and architectural elemensts were even sent as gifts to American museums and universities to foster economic and diplomatic ties and to promote a Jordanian identity that would be palatable to 'the west'.7 This way, the material remains of the Biblical and Greco-Roman periods were associated not with the local Arab and largely Muslim culture but with a European cultural heritage that drew on multiple tendentious claims to the region's past. Ancient cities with long histories, such as Jarash, were examined for their Hellenistic, Roman and Byzantine monumental remains, and well into the second half of the twentieth century excavations were commonly carried out at speed to unearth the material culture associated with these periods, devoting little attention to recording post-Byzantine evidence. This meant, in particular, that strata dating to the Islamic period were either entirely overlooked or largely omitted from final publications.<sup>8</sup> Archaeological practices of this kind 'contributed to fragmenting the pasts of the region, segregating them from their land, times and people', and consequently misled many Jordanians into believing that sites like Jarash and Petra did not belong to them, but formed part of an imposed western heritage.9

Faced with such a deeply embedded colonial legacy, Khammash set out to depict new imagery for an alternative identity – one that was tied to the recent history of the Hashemite dynasty, its lineage and the formation of the nation state of Jordan. Rather than drawing on images of 'internationally renowned archaeological sites', he emphasised modernity and developed images that underpinned the historical legitimisation of the ruling dynasty.10

What were the processes of archaeological and historical interpretation that lay behind such disjunctures of perspective? In this chapter, I use Jarash as a case study to examine how such a long-lived city came to be perceived as mainly Roman, how the site became associated with western cultural heritage and how its image was utilised in the Jordanian nation-building project. I argue, as others have done before me, that the association of Jarash with a European historical past, with its accompanying cultural mythology and political ideology, resulted in the gross neglect

<sup>5</sup> For the development of archaeology as a discipline in Jordan, see Abu-Khafajah and Miqdadi, 'Prejudice, military intelligence, and neoliberalism'; Corbett, *Competitive Archaeology*; Maffi, 'The emergence of cultural heritage in Jordan'. 6 Corbett, *Competitive Archaeology*, 134–139. 7 Katz, *Jordanian Jerusalem*; Macaulay-Lewis and Simard, 'From Jerash to New York'.

<sup>8</sup> See, for example, Alan G. Walmsley's comments on Kraeling, *Gerasa, City of the Decapolis*. Walmsley,

<sup>&#</sup>x27;Urbanism at Islamic Jerash', 247, note 36. 9 Abu-Khafajah and Miqdadi, 'Prejudice, military intelligence, and neoliberalism', 3.

<sup>10</sup> Corbett, *Competitive Archaeology*, 1.

of the city's early Islamic history and a profound disassociation between the ancient site and the people of Jordan. The first part of my chapter addresses how Jarash came to be interpreted as a Roman site and how this image served the formation of a Jordanian national identify after independence in 1946: I outline the archaeological exploration of Jarash and examine the use of its monuments on stamps, banknotes and as diplomatic gifts. The second part of my chapter examines the archaeology of Jarash itself in order to highlight the stages of the city's long history. I conclude by reflecting on the metaphor of the palimpsest, the conceptual theme of this volume and consider whether it is a helpful way of describing the processes that have formed our understanding of the city of Jarash.

## **Part I: The making of a Roman city**

Jarash, in northern Jordan, is among the most intensively excavated urban centres in the Eastern Mediterranean. Since 1907, European and American historians and archaeologists have sought to transform its ruined landscape through targeted excavations of monumental remains that belong mainly to the Roman and Byzantine periods.11 These efforts have produced an impressive cityscape of colonnaded thoroughfares dotted with temples, theatres, churches and bathhouses, which have made Jarash among the best-known archaeological sites in the Near East (Figure 17.1).

Eva Mortensen has summarised early western encounters with Jarash by outlining how European travellers made their way to the city from the early nineteenth century onwards, returning with sketches and descriptions which depicted a vast ruined landscape of limestone columns with Corinthian capitals.12 The first excavations took place under Ottoman rule in 1907 when German archaeologists uncovered a mosaic floor featuring muses and poets, part of which is now on display at the Staatliche Museen zu Berlin.<sup>13</sup>

Some twenty years later, in 1928, when Transjordan was under the British Mandate, a joint mission was initiated, led by Yale University and the British School of Archaeology in Jerusalem.14 Over the course of six years, monuments of Roman and Byzantine date were unearthed throughout the city, including the hippodrome, the temple of Artemis, the city gates and public squares, as well as ten churches. These excavations were undertaken at a time when stratigraphic recording was not usually a part of the archaeological repertoire, which meant that deposits and architectural

<sup>11</sup> Lichtenberger and Raja, *Archaeology and History of Jerash*; Mortensen, 'The early research history of Jerash'. Archaeological work since the 1980s has also included Australian, Japanese and Jordanian Missions to the site.

<sup>12</sup> Mortensen, 'The early research history of Jerash'.

<sup>13</sup> Mortensen, 'The early research history of Jerash', 179–180. Other panels went to Damascus where they were purchased by C. Nelda and H. J. Lutcher Stark and brought to Texas. Five of these panels were sold by Sotheby's in 2003. www.sothebys.com/en/auctions/ecatalogue/2003/antiquities-n07949/

lot.74.html. 14 Kraeling, *Gerasa, City of the Decapolis*.

*Fig. 17.1: Map of Jarash showing the city's major monuments (© Thomas Lepaon).*

modifications of a later date were stripped away with minimal recording in order to create a pristine image of Jarash as a Roman and, to some extent, a Byzantine city.

In a recent article, Alan Walmsley described how Clarence Fisher, one of the leading excavators of the Joint Mission in its early years, noted in his field diaries that there was evidence of the continuation of settlement not only beyond the earthquake of AD 749 – which until recently was seen by scholars as marking the end of urban life in Jarash – but well into the twelfth or thirteenth centuries.15 However, the final publication, which was edited by Carl H. Kraeling, omitted Fisher's observations and instead constructed a narrative of decline, which began in the Byzantine period and accelerated under Arab rule. Kraeling dismissed the post-earthquake architectural remains as 'wretched hovels' invading the city's streets and squares and described its inhabitants as squatters.16 In this way, he rendered the post-earthquake settlement insignificant and took a leading role in shaping the myth of a glorious Roman city, which declined in the post-classical era.

Kraeling was not exceptional. On the contrary, Irene Maffi reminds us that 'excavations conducted in Transjordan during the Mandate period were an entirely western affair, and as a consequence the choice of sites and the interpretation of the material were determined by the cultural agenda and the historical vision of non-indigenous archaeologists.'17 Two types of archaeology were at play here: one explored sites associated with a Biblical past while the other concentrated on sites of Hellenistic, Roman and Byzantine origin that were 'regarded as places belonging not to the history of the modern local inhabitants but to the western past.'18 As a consequence, entire epochs of the region's history were overlooked as they were considered irrelevant to the archaeologists' concerns, especially the Prehistoric and Islamic periods.19

While the work of European and American archaeologists has often been emphasised as the main reason why Jarash was perceived as a 'classical' city, less attention has been paid to other stakeholders in this project, such as the Jordanian state itself. The dissolution of the British Mandate and the formation of the Hashemite Kingdom of Transjordan in 1946 was a defining period in which Jordan sought to build a national identity while also forging diplomatic ties with western countries and especially with the United States.20 Historic sites came to play an important role in this process.

It was particularly Jerusalem – which was under Jordanian rule from 1948 to 1967 – and sites of Biblical significance, such as Mount Nebo, where Moses first gazed on the Holy land, that were promoted. Petra and Jarash were also considered to be important as their classical heritage allowed Jordanians to present themselves as the heirs of

<sup>15</sup> Walmsley, 'Urbanism at Islamic Jerash', 247, note 36. 16 Kraeling, 'The history of Gerasa', 67–69.

<sup>17</sup> Maffi, 'The emergence of cultural heritage in Jordan', 15.

<sup>18</sup> Maffi, 'The emergence of cultural heritage in Jordan', 16–17.

<sup>19</sup> Maffi, 'The emergence of cultural heritage in Jordan'.

<sup>20</sup> Macaulay-Lewis and Simard, 'From Jerash to New York'.

a Greco-Roman antiquity.21 Kimberly Katz has examined how images of these sites were actively used by Jordanians themselves in the forging of a Jordanian national identity.22 She has demonstrated how the archaeological sites were used to construct a narrative of Jordan as the site of an ancient civilisation, which in turn boosted tourism as a major contributor to the country's economy.23 This can, according to Katz, be observed in the designs placed on bank notes and postage stamps issued from the 1950s onwards, as well as in the use of antiquities as diplomatic gifts.24

Monuments from Jarash had appeared sporadically on postage stamps from the 1930s, but in the 1950s and 1960s the Jordan Tourism Authority took an active role in promoting the site as a major tourist attraction. In 1965, Jarash was featured in a series of eight stamps with images of iconic monuments such as the South Theatre, the Temple of Artemis and the colonnaded streets.25 In an impressive redoubling of the message for foreign consumption, every stamp sent abroad was cancelled with a counterstamp that read in English, 'Visit Gerasa [Jerash] of the Decapolis,' or 'Visit Transjordan of the Arabs, Country of ancient sites and scenes,' or 'Petra! Immortal! Wonderful! Rose-red Rock-hewn City.'<sup>26</sup>

The message was also directed towards an internal audience. For example, in 1959 when banknotes were issued for the first time during King Hussein's reign, the images that appeared on the notes were, as Katz argues, 'significant for their role in the development of a collective Jordanian identity.'27 Jordan was at the time (and still is today) primarily a cash-based economy where most people depend on coins and banknotes for day-to-day expenditure. The small notes were in particularly high demand and the images that received the widest circulation were Jarash (on the half-dinar note), the Dome of the Rock (on the one-dinar note) and Petra (on the five-dinar note).

Elizabeth Macaulay-Lewis and Jarad Simard have studied the use of antiquities from Jordan as diplomatic gifts.28 In an article from 2015, they investigated how in the 1960s and 1970s King Hussein used archaeological heritage to 'forge goodwill with the United States and to strengthen academic relationships with American Universities.'<sup>29</sup> Macaulay-Lewis and Simard record how column capitals from Jarash were sent to the university museums at Harvard and the University of Pennsylvania, while Princeton

<sup>21</sup> Corbett, *Competitive Archaeology*; Macaulay-Lewis and Simard, 'From Jerash to New York'.

<sup>22</sup> Katz, *Holy Places and National Spaces*; Katz, 'Building Jordanian legitimacy'; Katz, *Jordanian Jerusalem*. 23 Katz, *Holy Places and National Spaces*, especially 49–59. 24 Corbett, *Competitive Archaeology*, 134–139; Katz, *Holy Places and National Spaces*, 49–59; Macaulay-Lewis and Simard, 'From Jerash to New York'.

<sup>25</sup> For images of the stamps, see www.freestampcatalogue.com/sjrp0524-roman-architecture-8v.

<sup>26</sup> Katz, *Holy Places and National Spaces*, 57. The mid-sixties promotion of tourism to Jordan coincided with and was orchestrated around the pilgrimage of Pope Paul VI to the Holy Land in 1964 – the first visit by a Catholic Pope since late antiquity – and the 1964–1965 World Fair in New York (see below): Katz,

*Holy Places and National Spaces*, 203–210. 27 Katz, *Holy Places and National Spaces*, 98.

<sup>28</sup> Macaulay-Lewis and Simard, 'From Jerash to New York'.

<sup>29</sup> Macaulay-Lewis and Simard, 'From Jerash to New York', 353.

received two Umayyad capitals from Qasr al-Muwaqqar, and the city of Philadelphia received a column from Jordan's capital Amman (ancient Philadelphia).30 These architectural elements were gifted in order to reward existing research efforts and encourage further scholarship. In a press release marking the gifting of the capitals to University of Pennsylvania, Jarash was described as 'the Pompeii of the Middle East'.<sup>31</sup> The reference to Pompeii served as a rhetorical device aiming to draw attention to Jarash as a magnificently preserved site of Roman date and to capitalise on Jordan's claim to a Greco-Roman antiquity.<sup>32</sup>

Finally, Jordanian contribution to the World Fair of 1964–1965, which took place in Flushing Meadows in Queens, New York, included not only a miniature model of the Dome of the Rock, but also the permanent gift by

*Fig. 17.2: Column from Jarash in Flushing Meadows, Queens, New York (Kyle Brunner).*

King Hussein and the government of Jordan to the city of New York of a 4.94 m tall column with a base and capital from Jarash (Figure 17.2).33 The gift of the column was representative of, and immediately identifiable with, the remains of a classical site. As to the archaeological deficit incurred by its transfer to a public park in New York City, it was perhaps thought that one less column would be no great loss to a site in which columns were found in abundance. The accompanying plaque states that it was retrieved from the temenos of the temple of Artemis, but Macaulay-Lewis and Simard have argued persuasively on the basis of comparative evidence that it most likely came from the colonnade of one of the city's thoroughfares (Figure 17.3).34 The column was thus one of hundreds that once lined the public spaces of Jarash, rather than a rare specimen from one of the city's most celebrated monuments.35

<sup>30</sup> Macaulay-Lewis and Simard, 'From Jerash to New York', 348.

<sup>31</sup> Macaulay-Lewis and Simard, 'From Jerash to New York', 351.

<sup>32</sup> More recent scholarly literature has seen a revival of the tradition of comparing Jerash to Pompeii. See, for example, Lichtenberger and Raja, 'Mosaicists at work'.

<sup>33</sup> Macaulay-Lewis and Simard, 'From Jerash to New York'.

<sup>34</sup> Macaulay-Lewis and Simard, 'From Jerash to New York', 343–347.

<sup>35</sup> The column and the misnaming of its origins is discussed by Macaulay-Lewis and Simard, 'From Jerash to New York', especially 343–348, 354.

*Fig. 17.3: View towards the north of the re-erected colonnade on Jarash main north-south thoroughfare (author).*

The deployment of Jarash's architectural remains for the purposes of encouraging tourism and facilitating diplomatic initiatives was also enacted within the site itself. In the 1960s the Jordanian military cleared the city's main thoroughfares and re-erected their columns to create an impressive vista for visiting tourists.36 The columns were placed along the north–south and east–west running streets, but in many cases they did not reflect the original configuration of these public thoroughfares. Sections of column

drums that matched neither in shape nor colour were fixed together with cement and the reconstituted columns were re-erected in urban spaces from which they had been absent long before Jarash was abandoned. The re-erection of columns was undertaken in an attempt to recreate a visual impression, based on modern assumptions, of the city at the time of its Roman origins. It is important to note that the motivation of the Jordanian state was economic and should be seen as part of a campaign to highlight aspects of the urban cultural heritage that would appeal to tourists.

The 1980s saw renewed archaeological interest in the city when the Department of Antiquities instigated the Jarash Archaeological Project. According to Alan Walmsley, the main purpose was 'to further tourism and heritage and to bring more structure to the exploration and conservation of Jerash.'37 Teams of American, Australian, British, French, Italian, Jordanian, Spanish and Polish archaeologists were invited to excavate hitherto unexplored sections of Jarash.38 For the first time, the early Islamic period was not automatically assumed by western scholars to have been a period of decline.39 In fact, the new archaeological explorations revealed a city that flourished under early Islamic rule. Some excavators, like Michel Gawlikowski, who directed the investigation of the so-called 'Umayyad house', were also able to document building activity that took place after the great earthquake in AD 749.40 In this way, the Islamic period in Jarash was, for the first time, extended beyond the reign of the Umayyads.

It was not until 2002, however, when Alan Walmsley instigated the Islamic Jarash Project (IJP), that efforts were made to actively explore the city's history during

<sup>36</sup> Pers. comm., Alan Walmsley, December 2020.

<sup>37</sup> Walmsley, 'Urbanism at Islamic Jerash', 242. 38 Zayadine, *Jerash Archaeological Project*. 39 Asem Barghouti proposed an Umayyad presence prior to the initiation of the Jarash Archaeological Project. Barghouti, 'Urbanization of Palestine and Jordan in Hellenistic and Roman times'.

<sup>40</sup> Gawlikowski, 'A residential area by the South Decumanus'.

the first centuries of Arab-Muslim rule.41 The focus of the project was an Umayyad/ Abbasid-period congregational mosque in the city centre and the nearby commercial, administrative and residential structures. The results of the extensive archaeological work carried out in this area showed that settlement in Jarash continued on a large scale after the earthquake of AD 749.42 Through careful excavation and examination of ceramic and numismatic materials, IJP documented how the mosque, shops and residential units were not only rebuilt, but also expanded and developed in the second half of the eighth century. IJP also devoted substantial resources to the conservation of the city's early Islamic heritage and its presentation to the public. The mosque was conserved (Figure 17.4) and an explanatory panel in three languages (Arabic, English and French) was set up on the city's main thoroughfare to bring the mosque to the attention of visitors.

In 2010, a Danish-German project was initiated in Jarash's north-west quarter exploring a hitherto largely unexplored part of the city. The project has exposed remains dating mainly to the late antique and Umayyad periods as well as a smaller settlement that utilised the area in the Ayyubid/Mamluk period. Among the project's main discoveries is an Umayyad residential building complete with household goods, which was abandoned after the earthquake of AD 749.43

In 2015, this author initiated a new archaeological project to explore Jarash's southwest district by means of geophysical examination, recording of surface remains and excavation. Building on the innovative work of Alan Walmsley, this new initiative examines a residential neighbourhood over the *longue durée* with the aim of furthering our understanding of two understudied aspects of the history of the site: its residential neighbourhoods and the early Islamic settlement. In the three campaigns undertaken so far, we have studied material evidence ranging in date from the first century BC to the tenth century AD.44

### **Part II: Creating the past anew**

Although recent initiatives have sought to construct a more inclusive view of Jarash's development from its foundation to the present, the city is understandably still best known for its Roman-period monumental remains, which dominate the urban

<sup>41</sup> See, for example, Walmsley, 'The Friday mosque of early Islamic Jarash in Jordan'; Walmsley, 'Urbanism at Islamic Jerash'. See also Rattenborg and Blanke, 'Jaraš in the Islamic ages'.

<sup>42</sup> Rattenborg and Blanke, 'Jaraš in the Islamic ages'.

<sup>43</sup> Lichtenberger *et al*., 'A newly excavated private house in Jerash'; Lichtenberger and Raja, 'The Danish-German Jarash North-West quarter project: results from the 2014–2015 Seasons'. In addition, Lichtenberger and Raja, *Middle Islamic Jerash*, summarise the results of their recent excavations of an Ayyubid-Mamluk period hamlet in north-west Jerash. See also Lichtenberger and Raja, 'Ğaraš in the middle Islamic period'.

<sup>44</sup> See, for example, Blanke, 'Abbasid Jerash reconsidered'; Blanke *et al*., 'The 2011 season of the Late Antique Jarash Project'; Blanke *et al*.,'Excavation and magnetic prospection in Jerash's southwest district'; Blanke *et al*.,'A millenium of unbroken habitation in Jarash's southwest district'; Pappalardo, 'The late antique Jerash project'.

*Fig. 17.4: Umayyad/Abbasid period mosque in the intersection of Jarash's two main streets (© Alan Walmsley/Islamic Jarash Project).*

landscape.45 The second part of this chapter aims to create a more balanced view of Jarash's history by drawing on the archaeological evidence to illustrate how it developed from the Roman to the early Islamic period. I will explore three noncontiguous periods in the city's history: the third century, the sixth century and the ninth century. Although these snapshots cannot do justice to the complex changes involved in the city's long-term transformation, they are intended to highlight three critical themes that governed the development of the urban landscape at different times: the shifting patterns of investment within the city's built environment and the transformative effects of new economic priorities as well as changing political and religious ideals upon the urban fabric.

### *The third century: Building Roman Jarash*

Third-century Jarash can be described as the culmination of a two-century-long building boom.46 By this point, most if not all of the city's famous monuments had been completed. While the origins of Jarash can be traced to the Hellenistic period,

<sup>45</sup> Even Macaulay-Lewis and Simard describe Jerash as 'one of the Hashemite Kingdom of Jordan's preeminent Greco-Roman cities that flourished in the second century CE' and ascribe the secondcentury transformation of its urban fabric to the Roman emperors. Macaulay-Lewis and Simard, 'From Jerash to New York', 343, 347.

<sup>46</sup> Kraeling, 'The history of Gerasa', see also articles in Lichtenberger and Raja, *The archaeology and history of Jerash*.

only a few remains from this early period have been recovered.47 The archaeological remains suggest that the Hellenistic settlement began in the second century BC with the founding of a town by the name of Antioch on the (river) Chrysorhoas.48 The name Gerasa is a Greek version of a Semitic name (Jarash) for the city that was in use continuously throughout its history.49 In 64 BC, Jarash was incorporated into the Roman empire as part of the newly founded province of Syria. It was probably around this time that Jarash

*Fig. 17.5: The Oval piazza, one of Jarash's most iconic monuments, which also appeared on stamps and banknotes issued in 1965 (author).*

became a member of the loose confederation of cities in Syria-Palestine known as the Decapolis.50 The Decapolis confederation lasted until the annexation of Nabataea by Trajan in AD 106, which entailed a reorganisation of the region and the formation of the province of Arabia, in which Jarash was included.

During the first and second centuries AD, the city's elite invested enormous sums to re-structure Jarash as a Roman provincial city by building monumental architectural complexes organised around colonnaded thoroughfares that were lined with shops and public fountains and intersected in large open squares (Figure 17.5). Some of the major building activities that took place at this time included the inauguration of the rebuilt temple of Zeus in AD 163, a series of fountains on the main street of which the nymphaeum was the most elaborate, as well as the construction of the south and north theatres, the macellum and the East and the West Baths.51 The emperor Hadrian visited Jarash in the winter of AD 129–130.52 This event was commemorated by the construction of a triumphal arch at the city's southern entrance and possibly also by the construction of the hippodrome, which, with a seating capacity of just 16,000, is the smallest known hippodrome in the Roman world.53

<sup>47</sup> Kraeling, 'The history of Gerasa', 28–34; Uscatescu and Martín-Bueno, 'The *macellum* of Gerasa', 67; Zayadine, 'The Jarash Project for excavation & restoration'; Blanke *et al*., 'A millenium of unbroken habitation in Jarash's southwest district'. 48 Kennedy, 'The identity of Roman Gerasa', 55.

<sup>49</sup> For the etymology of Jerash's two ancient names, see Kraeling, 'The history of Gerasa', 27, note 3. 50 Kennedy, *Gerasa and the Decapolis*. 51 Temple of Zeus: Seigne, 'Le sanctuaire de Zeus', 287. Fountains and Nymphaeum: Pickett, *Water after Antiquity*; Seigne, 'Remarques préliminaires à une étude sur l'eau dans la Gerasa antique'; Seigne, 'Fountaines et adductions d'eau à Gerasa'. For the *macellum*, see Uscatescu and Martín-Bueno, 'The *macellum* of Gerasa'; Uscatescu and Martín-Bueno, 'Evergetes and restorers' and for West Baths, see Lepaon, 'Les édifices balnéaires de Gerasa de la Décapole', 52–57.

<sup>52</sup> Seigne, 'Porquoi Hadrian a-t-il passé l'hiver de 129/30 à Gerasa?'.

<sup>53</sup> Kennedy, 'The identity of Roman Gerasa', 62.

The apparent decline in building activity during the third century caused Kraeling to propose that this was a time of economic recession for Jarash.54 This notion was based on general assumptions about the 'third-century crisis'. However, research over the past 30 years or so suggests that the challenges of the third century did not affect the eastern Mediterranean as much as previously believed.<sup>55</sup>

The vast array of monuments that can be dated to the first, the second and, to a lesser extent, the third century suggest that through the sponsorship of the local council and with the financial help of private benefactors, Jarash was reinvented as a provincial Roman city in this period. It seems that much of this public building activity formed part of a substantial, long-term plan for the visual development of the city.56 Arthur Segal has argued that the extensive building programme that unfolded contemporaneously in neighbouring cities within the former Decapolis should be seen as part of a process of self-definition by the city dwellers themselves, as well as an attempt to compete with neighbouring urban centres.57 This aim was achieved through the construction of Roman-influenced architecture on a massive scale in order to create a Roman imperial 'feel' to the cities.

More recently, David Kennedy has argued that although the architecture appears Roman in style, its use continued local practices to some extent.58 The hippodrome, for example, was based on a western Roman design, according to Kennedy, but appears to have had a non-Roman function.<sup>59</sup> A Roman-style racing track required twelve gates to accommodate the four circus factions, but the Jarash hippodrome had only ten gates, which was more suitable for Greek-style racing 'in which teams were entered by professional organisations rather than factions.'60 The neighbouring cities of Gadara and Skythopolis also constructed hippodromes in the second century. However, neither were employed for their original purpose. The racing track at Gadara was never completed and the one in Skythopolis was converted to an amphitheatre soon after its completion.61

#### *The sixth century: Life amidst the crumbling ruins of a distant past?*

While Kraeling described Jarash in the Roman period as the apogee of urban life, he characterised the following centuries as 'a life lived in the midst of the crumbling

<sup>54</sup> Kraeling, 'The history of Gerasa', 57.

<sup>55</sup> Cameron, *The Later Roman Empire*, 1–12.

<sup>56</sup> For an overview of the inscriptions uncovered by the Yale/ British School of Jerusalem Joint Mission, see Welles, 'The inscriptions'.

<sup>57</sup> Segal, 'Roman cities in the province of Arabia'.

<sup>58</sup> Kennedy, 'The identity of Roman Gerasa'. 59 Kennedy, 'The identity of Roman Gerasa', 62. 60 Kennedy, 'The identity of Roman Gerasa', 62. However two inscriptions hint at some local support for the factions within the city: see inscription 349 in Welles, 'The inscriptions', 490. See also, Feissel, *Chroniques d'épigraphie byzantine 1987–2004*, 273. However, following Jeffreys, 'Entertainments, theatre, and hippodrome', it seems reasonable to assume that the factional support was not necessarily associated with activities that took place in the hippodrome. 61 Kennedy, 'The identity of Roman Gerasa', 63–64.

ruins of a distant past.'62 In the sixth century, most monuments of the Roman period had indeed seen a discontinuation of their original use, but rather than the urban decay that Kraeling perceived, today we are more inclined to interpret the evidence as indicative of a gradual reorganisation of the city to accommodate religious change as well as shifting economic priorities. A close reading of the archaeological remains reveals that although the Roman-period monuments were no longer used for their original purposes, every inch of the city was utilised to its maximal potential.63

The institutionalisation of Christianity and its inauguration as the state religion in the fourth century introduced extensive changes to Jarash, which were manifested in the physical transformation of the cityscape. The most direct consequence of the rise of Christianity was the gradual abandonment of the city's two great temples and the construction of more than twenty churches and shrines in the fourth, fifth and sixth centuries. From the fourth century onwards, Jarash was the seat of bishops who represented the city in the councils of Seleucia in 359 and Chalcedon in 451.<sup>64</sup>

The presence of Christianity was felt not only in the layout of the city, but also in Jarash's ritual calendar. Epiphanius of Salamis described how the miracle of Cana, the marriage ceremony in which Jesus transformed water into wine, was commemorated annually in the cathedral in Jarash.65 The cathedral (built, according to the excavators, in the second half of the fourth century), was located adjacent to the temple of Artemis and was, according to Kraeling, built on the foundations of a former temple dedicated to Dionysus, in which a yearly wine festival had been celebrated.66 The accuracy of Kraeling's report cannot be reliably confirmed, but the replacement of a pagan wine festival with a Christian one could be perceived as part of a negotiation of religious identities in which a Christian commemoration superseded a similar pagan practice.67

The elevated locations of the temples of Artemis and Zeus mean that the buildings of the old cults would have remained as powerful visual reminders of the past. However, by the sixth century the significance of the structures had been radically transformed. The temple of Zeus, for example, served as a stone quarry including a stone cutter's workshop, in which building blocks were recut for use elsewhere in town. Other parts of the temple served as a metal workshop and as a dumping ground for waste products from a nearby ceramic kiln.68

<sup>62</sup> Kraeling, 'The history of Gerasa', 67. 63 See, for example, Kehrberg-Ostrasz, 'Jerash seen from below'.

<sup>64</sup> Kraeling, 'The history of Gerasa', 64. 65 Epiphanius *Panarion* Haer. 51: 30, 1–2 cited in Kraeling, 'The history of Gerasa', 63. 66 Crowfoot, 'The Christian churches'; Kennedy, *Gerasa and the Decapolis*, 101; Kraeling, 'The history of Gerasa', 37, 63; Walmsley, 'Byzantine Palestine and Arabia', 135.

<sup>67</sup> Though the existence of a temple of Dionysus in the area that was later occupied by the cathedral has never been fully established, excavations below the fountain court by the Yale Joint Mission as well as a Swiss team led by B. Brenk have documented the presence of a structure that has been interpreted as the remains of a pagan temple. Brenk *et al*., 'Gerasa, fountain court of the cathedral'; Brenk *et al*., 'The fountain court at Jerash Cathedral reconsidered'; Brenk *et al*.,'The buildings under the "Cathedral"

of Gerasa'. 68 Kehrberg-Ostrasz, 'Jerash seen from below', 120–122.

Other monuments were also diverted from their original purposes. The hippodrome, for example, saw a multitude of uses after its discontinuation as a racecourse in the fourth or fifth century. Most importantly, the cavea that lined the structure on its north, east and west sides were converted into tanneries and workshops for the manufacture of ceramics.69 Tanneries were also found within the former macellum, while an industrial-sized stone-saw was located near the premises of the temple of Artemis.70

#### *The ninth century: urban life after the earthquake of AD 749*

The increasing popularity of Islam and the great earthquake of AD 749 were probably the two most transformative factors for the city's urban landscape in the seventh and eighth centuries. A congregational mosque was constructed in the intersection of the city's two main thoroughfares in the early eighth century, but after only a few decades, the mosque and most of the city's buildings were devastated by one, or perhaps multiple, massive seismic events.71 The consequences of the earthquake(s) have been documented by three archaeological projects:72 Michel Gawlikowski's excavation of the so-called Umayyad house on the north side of the south *decumanus*  under the auspices of the Jarash Archaeological Project;<sup>73</sup> the Islamic Jarash Project's excavation of a mosque and its adjacent buildings in the town centre under direction of Alan Walmsley;74 and the ongoing project that I initiated in 2015, which examines a residential section of the city's south-west quarter.75

In the city centre, existing foundations were reused to reconstruct the buildings that had been lost in the earthquake. The mosque, for example, was rebuilt to its former dimensions. Shops were constructed alongside the mosque and residential and administrative buildings to the east and west of the mosque were rebuilt to continue the functions of the structures that they replaced (Figure 17.6). At a later date, a minaret was inserted into the mosque's northeast corner.

A similar development has been documented at the so-called 'Umayyad House', which is located across the street from the mosque, on the north side of the south *decumanus.* The original structure was erected around the middle of the seventh century and it was used for commercial and artisanal purposes.76 After the earthquake of AD 749, the foundation and surviving walls were refurbished and partitioned to

<sup>69</sup> See Bessard, 'Between localism and a desire for greater openness', 397–406; Kehrberg-Ostrasz, 'Jerash

Byzantine potters and their wares'. 70 Seigne, 'A sixth century water-powered sawmill at Jerash'.

<sup>71</sup> See summary in Ambraseys, *Earthquakes in the Mediterranean and Middle East*, 230–238. 72 See summary in Rattenborg and Blanke, 'Jaraš in the Islamic ages'.

<sup>73</sup> Gawlikowski, 'A residential area by the South Decumanus'.

<sup>74</sup> For a recent overview with a comprehensive bibliography, see Walmsley, 'Urbanism at Islamic Jerash'. 75 See, for example, Blanke, 'The late antique Jarash Project'; Blanke, 'Abbasid Jerash reconsidered'; Blanke *et al*., 'The 2011 season of the Late Antique Jarash Project'; Blanke *et al*., 'Excavation and magnetic prospection in Jerash's southwest district'; Blanke *et al*., 'A millennium of unbroken habitation in Jarash's southwest district'; Pappalardo, 'The late antique Jerash Project'.

<sup>76</sup> Gawlikowski, 'A residential area by the South Decumanus', 111.

*Fig. 17.6: Plan of early Islamic central Jarash* c*. 750–850 (© Rune Rattenborg).*

create smaller structural units.77 The final phase of its occupation in the ninth century comprised the construction of several ceramic kilns within the walls of the building.78

The evidence from the residential area on the south-west hilltop confirms that 'substantial efforts, at a private, communal, and public level, must have gone into rebuilding and refurbishing houses, neighbourhoods, and the congregational mosque soon after the earthquake of 749.'79 So far, excavations in nine trenches have documented buildings of a residential nature with cisterns (perhaps reused from previous structures) for the collection of water. The ceramic evidence supported by C14 dating suggests that the area was abandoned in the ninth century, probably as a consequence of a series of disasters. One house was abandoned after a fire broke out, which caused the roof to collapse. Another house was abandoned after an earthquake. The recovery of large quantities of domestic items confirms that both houses were in use at the time of their collapse.

At this point, the archaeological evidence from the mosque and its adjacent residential structures suggests that the city gradually contracted towards its centre at some time during the ninth or tenth century. Part of the mosque remained in use, while Jarash steadily transitioned from an urban settlement into a village community.80

<sup>77</sup> Gawlikowski, 'A residential area by the South Decumanus', 114–115.

<sup>78</sup> Gawlikowski, 'A residential area by the South Decumanus', 117.

<sup>79</sup> Rattenborg and Blanke, 'Jaraš in the Islamic ages', 13.

<sup>80</sup> Blanke *et al*., 'The 2011 season of the Late Antique Jarash Project'; Lichtenberger and Raja, 'Ğaraš in

#### 366 *Louise Blanke*

#### **Concluding remarks**

I opened this chapter by quoting the historian and Egyptologist Jan Assmann's observation that 'the past is not a natural growth but a cultural creation.'81 This is certainly true of the (re)construction of the history of Jarash over the past century, which has involved a careful process of selection and conservation of monuments dating mainly from the first to the third century AD, and to a lesser extent to the following three centuries. Although fieldwork undertaken since the 1980s has sought to redress the imbalance with regard to post-sixth century evidence, past archaeological and conservation practices have ensured that Jarash is irrevocably identified with its pre-Islamic remains. It is in part this European, colonial, exclusivist interest in the Roman and early Christian cultural heritage that prompted Ammar Khammash to dissociate Jordan's modern history and its national identity from the remains of its ancient sites.

To come finally to the metaphor of the 'palimpsest', this chapter would suggest that it is not a particularly helpful device for describing the development of Jarash as an urban space. A palimpsest implies disconnected layering, which does not accommodate the physical reality of archaeological sites or the creative processes of selection, reuse, refurbishment, destruction and abandonment. The metaphor is ill-suited to the analysis of the complex entanglement of heritage, memory, politics, economy, diplomacy and national as well as international identity that played a part in the consolidation of Jarash (and other Levantine cities) as Roman sites. Consequently, the idea (or ideal) that Jarash and its physical manifestation as an archaeological park sits as a palimpsestic layer on top of an actual historical past is misleading.

What the palimpsest metaphor *can* do for us is perhaps to remind us of the dangers of privileging some periods over others. The traditional emphasis on Jarash's Roman and Byzantine remains has caused a sense of disruption and discontinuity in the local perception of the past, and in the scholarly literature it has engendered narratives of decline and rupture rather than continuity, transformation and eventual organic abandonment. The physical (re)construction of Roman Jarash as a result of conservation strategies that were driven mainly by political motivations has created a distorted image of the city's development, because it identifies the terminal point of its evolution as the end of the 'classical' era.

By contrast, the brief survey of post-Roman Jarash as presented above reveals a resilient city whose urban fabric continued to develop and grow through the early Islamic period.82 Urban life in Jarash from the fourth to the sixth century was not a pale reflection of a glorious Roman past, but a period of selective repurposing which facilitated the process of adaptation to new religious ideals and urban priorities. Similarly, the seventh to ninth century saw the manifestation of new religious

the middle Islamic period'; Lichtenberger and Raja, *Middle Islamic Jerash*. 81 Assmann, *Cultural Memory and Early Civilization*, 33. 82 Rattenborg and Blanke, 'Jaraš in the Islamic ages' and for Greater Syria more generally, see Walmsley, *Early Islamic Syria*.

practices as well as constructive responses to natural disaster. The rigid periodisation which traditional approaches to urban history have taken for granted do not apply to Jarash because the archaeological record at Jarash, when viewed in the round, argues against them. We would do well to remember that archaeology was – and still is – a politically charged undertaking.83 Archaeological and historical interpretation plays an important role in shaping current historical narratives which in their turn set the course for the narratives of future pasts.

## **Acknowledgements**

This chapter was largely written and researched while I was a Postdoctoral Research Associate on the 'Impact of the Ancient City' project at the University of Cambridge, which has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 693418). I am grateful to Alan Walmsley, Luke Treadwell, Elizabeth Key Fowden and Edward Zychowicz-Coghill for their comments. Also many thanks to Pierre-Louis Gatier for providing the references and translations of the inscriptions concerning the Blues in Jarash. My fieldwork in Jarash was carried out with the financial support of the following funding bodies: The Danish Institute in Damascus, Lorne Thyssen Research Fund for Ancient World Topics, The Gerald Averay Wainwright Fund, The Barakat Trust, H.P. Hjerl Hansen Mindefondet for Dansk Palæstina Forskning, The Royal Danish Academy of Sciences and Letters, C.L. Davids Fond and Elizabeth Munksgaards Fonden. Most importantly, I am grateful to my friends and colleagues at the Jordanian Department of Antiquities without whom my research at Jarash would not have been possible.

## **Bibliography**


Barghouti, A. 'Urbanization of Palestine and Jordan in Hellenistic and Roman times', *Studies in the History and Archaeology of Jordan* 1 (1982): 209–29.


<sup>83</sup> Corbett, *Competitive Archaeology*, 11–12.


Zayadine, F. (ed.), *Jerash Archaeological Project, 1981–1983 vol I-II* (Amman, 1986–1989).

## Chapter 18

## Constantinople in the sixteenth-century Maghribī imaginary: The travelogue of ʿAlī al-Tamgrūtī

*Amira K. Bennison*

#### **Introduction**

Every city has a variety of 'pasts' generated consecutively or in tandem by its inhabitants and successive generations of rulers. Great political and cultural capitals, such as Constantinople, also exist in the imagination of much wider audiences, creating not only diachronic but also synchronic variations in the history(s) of a place. This chapter will explore one such synchronic variation, the sixteenth-century Maghribī view of Constantinople, which reflected an entirely different idea of the city's past to that held by the Ottomans themselves. The context for this view was the expansion of Ottoman power across the Mediterranean in the sixteenth century and their conflict with the Spanish Habsburgs, which sparked a series of interactions between the Ottomans in North Africa and Constantinople and the Saʿdī rulers of what is now Morocco during the sixteenth century.1

From the Ottoman perspective, the Saʿdī sultanate was a small but useful buffer between their North African territories and Spain but by the late sixteenth century the Saʿdīs had a rather different view. They could not deny Ottoman power but they perceived themselves to be a more 'noble' Arab lineage than the Turkish Ottomans and thus more worthy to be caliphs. Descriptions of the Ottoman capital written by Maghribī visitors in the early modern period, mostly Saʿdī ambassadors, must therefore be read in terms of the competitive trans-Mediterranean politics of the era which influenced how such authors adapted the literary tradition of travel writing and geographical description which had flourished in the Muslim Mediterranean for centuries. The result was a picture of Constantinople's past in which various aspects were elided or brought into focus to serve Saʿdī religio-political interests.

<sup>1</sup> See Hess, *The Forgotten Frontier*.

The interpretation of Constantinople's past and present penned by ʿAlī b. Muḥammad al-Tamgrūtī, whom the Saʿdī sultan Aḥmad al-Manṣūr sent to the Ottoman capital in 1589–1590, illustrates this point. By framing his account according to the literary norms of the travel account (*riḥla*) which privileged the religious landscape and identified 'wonders' of various kinds, the author can focus on religious buildings such as Hagia Sophia while minimising his description of the Topkapi Saray, thereby avoiding the implication that the Ottoman sultan's palace and court outshone that of his own master in Marrakesh. He also uses Islamic scripture, especially the *ḥadīth*, to place Constantinople in an eschatological narrative which minimised the Ottoman contribution to Constantinople in favour of flagging its Byzantine and early Islamic Arab phases, alongside conventional urban description. This was part of a larger Maghribī vaunting of the Islamic west, especially Umayyad Córdoba, in comparison to the Islamic east, observable in al-Maqqarī's compendium of Andalusī history and literature, written for an Ottoman Arab audience some four decades after al-Tamgrūtī visited Constantinople.

## **The political and literary background to al-Tamgrūtī's travelogue**

Saʿdī-Ottoman interactions commenced many decades before al-Tamgrūtī visited Constantinople and they had negative and positive aspects. The Ottomans were implicated in the assassination of the Saʿdī sultan Mawlay Muḥammad by his Turkish bodyguard in 1557 but several of his sons resided in Ottoman lands in the succeeding decades. One such prince, ʿAbd al-Malik, spent time at the court of Selim II (r. 1566–1574), fought in the Ottoman army at Lepanto and Tunis, and received military assistance from Murād III (r. 1574–1595) to launch a bid for the Saʿdī sultanate when his brother, the Saʿdī sultan ʿAbd Allāh, died. Similarly, the Saʿdī sultan who dispatched al-Tamgrūtī's embassy, Aḥmad al-Manṣūr (r. 1578–1603), had spent time in Ottoman Tlemcen as a youth and probably visited Algiers and Istanbul.<sup>2</sup> Aḥmad al-Manṣūr was much less enamoured of the Ottoman paradigm than his brother ʿAbd al-Malik had been and his sense of an Ottoman threat was compounded by the marriage of ʿAbd al-Malik's widow to Ḥasan Veneziano, the Pasha of Algiers, who promoted the claim to the Saʿdī sultanate of ʿAbd al-Malik's son, Ismāʿīl, in Constantinople.<sup>3</sup> To protect himself, Aḥmad al-Manṣūr made judicious approaches to Spain and other European countries, even though fraternisation with the 'infidel' came with its own domestic political risks.<sup>4</sup>

On the ideological front, he took a more assertive stance, claiming that as his lineage were *shurafāʾ*, that is descendants of the prophet, they were the true heirs to the caliphate.5 This type of claim had deep historical roots in the Maghrib but with

<sup>2</sup> Mouline, 'Sens et puissance', 119–120; García Arenal, *Ahmad al-Mansur*, 22–24, 29.

<sup>3</sup> Mouline, *Le califat imaginaire*, 305; Cory, *Reviving the Caliphate*, note 19, 64.

<sup>4</sup> For further detail on the extensive diplomatic manoevering of Aḥmad al-Manṣūr between the Ottomans,

Iberians, English and West Africans, see Mouline, *Le califat imaginaire*, part three. 5 Yahya, *Morocco in the Sixteenth Century*; Mouline, *Le califat imaginaire*; Cory, *Reviving the Caliphate*, especially chapter 3.

Aḥmad al-Manṣūr, it moved from regional assertion to a rejection of the pretensions of the Turkish Ottomans, whose undoubted power and large empire could not mask their non-Arab origins and late conversion to Islam. He was particularly keen to circulate this propaganda in his own domains, in West Africa and in the Arab lands of the Ottoman Empire in the late 1580s when a number of laudatory chronicles appeared, praising the noble prophetic ancestry of the Saʿdī lineage, and Aḥmad al-Manṣūr as its best incarnation. This cluster of works included Ibn al-Qāḍī's *al-Muntaqā al-maqṣūr* and the surviving portion of ʿAbd al-ʿAzīz b. Muḥammad b. Ibrāhīm al-Fishtālī's *Manāhil al-ṣafā*. 6 Their composition may well have been triggered by Aḥmad al-Manṣūr's irritation with the depiction of the Saʿdīs in an Ottoman chronicle by Muṣtafā al-Janābī, penned in Arabic and presumably therefore directed at educated readers in the Arab lands,<sup>7</sup> since the Saʿdī sultan's commissions appear to have been designed for the same audience, notably the urban, educated population of Cairo and Damascus.8

By 1589, however, Aḥmad al-Manṣūr had developed an ambitious plan to invade West Africa and he wanted to secure the neutrality of the Ottomans as well as the Spanish. He therefore dispatched his 1589–1590 embassy to Constantinople.<sup>9</sup> The precise reasons for Aḥmad al-Manṣūr's choice of ʿAlī al-Tamgrūtī for the mission and his status *vis-à-vis* his 'associates', Muḥammad b. ʿAlī al-Fishtālī, cousin of ʿAbd al-Azīz al-Fishtālī, author of the *Manāhil al-ṣafā*, and Muḥammad b. ʿAlī b. Abī -l-Qāsim, remain opaque. He describes both himself and al-Fishtālī as 'ambassadors' but the latter was a well-known private secretary (*kātib sirr*) to Aḥmad al-Manṣūr and probably the leader of the deputation.10 Our knowledge of al-Tamgrūtī himself comes primarily from his travelogue. He came from Tamgrūt, a way-station for trans-Saharan caravans and the site of one or more Sufi lodges located in the Darʿa valley, the natal region of the Saʿdī lineage.<sup>11</sup> His older brother, Muḥammad, was a religious scholar and Sufi who had served as tutor to various Saʿdī princes and emissary to Constantinople in the reign of the earlier Saʿdī sultan ʿAbd Allāh.12 The transmission of an Egyptian mystic's prognostication that Aḥmad al-Manṣūr would be a great caliph is also attributed to Muḥammad al-Tamgrūtī, confirming the family's religious prestige and affiliation to the Saʿdīs.<sup>13</sup>

Whatever his role, ʿAlī al-Tamgrūtī undertook to pen an account of the trip which he entitled, *al-Nafḥa al-miskiyya fī -l-sifāra al-turkiyya*, (*The Aromatic Scent of a Turkish Embassy*) which Cory sees as inspired by the same religio-political assertiveness as the contemporaneous writings of Ibn al-Qāḍī and al-Fishtālī, whom al-Tamgrūtī most

<sup>6</sup> Aḥmad b. al-Qāḍī, *al-Muntaqā al-maqṣūr ʿalā maʾāthir al-Manṣūr*; ʿAbd al-ʿAzīz b. Muḥammad b. Ibrāhīm

al-Fishtālī, *Manāhil al-ṣafā*. 7 Cory, *Reviving the Caliphate*, 59–62. 8 Cory, *Reviving the Caliphate*, 72–84, 135–136. 9 Mouline, *Le califat imaginaire*, 355–361.

<sup>10</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 26; Gannūn, *Rasāʾil saʿdiyya*, 25.

<sup>11</sup> Gutelius, 'The path is easy and the benefits large', 29.

<sup>12</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 5, 87.

<sup>13</sup> Mouline, *Le califat imaginaire*, 77. This prophecy is discussed by Cory but he does not identify the transmitter as Muḥammad al-Tamgrūtī. Cory, *Reviving the Caliphate*, 108.

likely knew.<sup>14</sup> Certainly, al-Tamgrūtī's travelogue reflects contemporary Saʿdī ideology, encapsulated in a well-known passage demeaning the Ottoman sultans and vaunting his Saʿdī masters:

The Ottomans are among those slaves and clients whom God uses to defend the Muslims, making them a walled fortress for Islam, even if the words of [the Prophet] – peace be upon him – that God supports this religion with debauched men, hold true for most of them and most of their followers! If they are charged with leadership and appointed to rule, in truth it is as deputies and custodians on behalf of those more deserving and their people, that is our lords and masters, the descendants of the Prophet, the kings of the Maghrib, who bestow nobility upon the imamate and the caliphate.15

However, the work can also be situated in the literary tradition of travel writing and geographical description which had flourished in the Muslim Mediterranean for centuries and was part of the intellectual formation of Arabic-educated individuals of the time. Al-Tamgrūtī's references show his desire to position himself as an erudite man fully conversant with the Arabo-Islamic heritage of the western Mediterranean and its many literary genres.16 He paraphrases or quotes several Andalusī and Maghribī literary and scholarly figures such as the Umayyad poet Ibn ʿAbd Rabbihi al-Qurṭubī (d. 940), the famous scholars Ibn ʿAbd al-Barr (d. 1071) and Qāḍī ʿIyāḍ of Ceuta (d. 1149), and the geographers and travellers al-Bakrī (d. 1094), Ibn ʿAbd Rabbihi al-Marrākushī (d. 1205),<sup>17</sup> Khālid b. ʿĪsa al-Balawī (fl. mid-fourteenth century)<sup>18</sup> and Ibn Baṭṭūṭa (d. 1368 or 1377).

Al-Tamgrūtī himself describes his work as a *riḥla*, thus situating it within a longestablished genre of travel writing, the most famous western Islamic exponents of which were Ibn Jubayr and Ibn Baṭṭūṭa. As Netton notes, the medieval Maghribī *riḥla* tends to be presented either as a journey to perform the pilgrimage to Mecca, or travel in search of religious learning, the so-called *ṭalab al-ʿilm*, or a combination of the two.19 In this type of travelogue, the culmination of the journey is Islam's religious pole, Mecca, and the stations on the way are part of a sacred landscape: towns are described with primary reference to their tangible and intangible religious heritage, their mosques, shrines, Sufi lodges, *madrasa*s and eminent religious personages, alive and dead.20

<sup>14</sup> Cory, *Reviving the Caliphate*, 85–6.

<sup>15</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 135. 16 De Castries, however, considers al-Tamgrūtī's work to be largely plagiarised, irrelevant and unnecessarily verbose. De Castries, *En-Nafhat el-Miskiya*, i–iv.

<sup>17</sup> In the introduction to his translation of the text, De Castries identifies Ibn ʿAbd Rabbihi as the famous tenth-century literary figure. The editor of the Arabic text, al-Shādhilī also notes only one Ibn ʿAbd Rabbihi in his index but the geographical citations attributed to Ibn ʿAbd Rabbihi appear to be paraphrases of the *Kitāb al-Istibṣār*, written by an Almohad official of the same name. Al-Tamgrūtī himself seems confused on the matter, attributing geographical descriptions to the *ʿIqd al-farīd* of the earlier writer! De Castries, *En-Nafhat el-Miskiya*, xi–xiii; Al-Tamgrūtī, *al-Nafḥa al-miskiyya*, 174. 18 For al-Balawi's biography, see De Castries, *En-Nafhat el-Miskiya*, xiii–xvi.

<sup>19</sup> Netton, 'The *Riḥla* of Ibn Jubayr', 57–74, 57.

<sup>20</sup> For more detail on Maghribī travel to the east, see El Moudden, 'Ambivalence'; Frenkel, 'Muslim travellers to Bilad al-Sham', 109–120.

The rhythm of al-Tamgrūtī's narrative follows the conventions of the medieval *riḥla*  by creating a structure in which time, in the form of dates, intersects with space, in the form of the places visited *en route*. As al-Tamgrūtī sails along the North African coast, stopping off at towns such as Tlemcen (Tilimsān), Oran (Wahrān), Bougie (Bijāya) and Tunis, he carefully notes his efforts to 'seek knowledge' from religious scholars and to visit the shrines of the great scholars and saints of the Maghribī past, eliding the contemporary Ottoman political presence by presenting his journey as a typical religiously-inspired *riḥla* rather than a political embasssy (*sifāra*).21 However, subtle choices in the description of urban locations were not devoid of political significance even in such a religious frame. Ibn Jubayr composed his *riḥla* in the late twelfth century when the Almohad caliph Yaʿqūb al-Manṣūr claimed universal caliphal status as heir to the Umayyads of Córdoba and their Damascene ancestors, and he expresses this in his work by privileging Umayyad Damascus while exaggerating the ruin and decline of ʿAbbāsid Baghdad.<sup>22</sup> As we shall see, al-Tamgrūtī also uses superficially religious observations to political effect.

The majority of al-Tamgrūtī's narrative describes the sea journey he took to and from the Ottoman capital, including its many terrors, his coastal stops, and the grand celebrations of the Prophet's birthday, the *mawlid*, in the splendid new palace of Aḥmad al-Manṣūr in Marrakesh that coincided with his return.23 The mission's departure from Tripoli to sail north to Constantinople marks a break in the narrative, a literal turn away from the normal *riḥla* route to Egypt and the Hijaz, and a departure from the familiar world of the Maghrib. Al-Tamgrūtī's description of Constantinople occupies fifteen out of a total of 146 pages in the published edition and focuses on the city rather than his mission, approaching it from a distinctively Arab perspective that taps into the Andalusī-Maghribī literary tradition in its composition, style and approach to the description of a city, downplaying the Ottoman contribution at several points.

From this latter perspective, it is instructive to place the *Nafḥa al-miskiyya* alongside the *Nafḥ al-ṭīb* of Aḥmad al-Maqqarī (*c*. 1577–1632), a native of the western Maghrib who lived for a time in Saʿdī Marrakesh and Fes and was acquainted with ʿAbd al-ʿAzīz al-Fishtālī and other members of the Saʿdī elite. Around 1617 al-Maqqarī departed for the Ottoman Arab lands and, just over a decade later, he wrote his famed compendium of Andalusī history and literature in Ottoman Cairo after being urged to do so by Ottoman Arab notables in Damascus.24 This was the very group targetted in the late sixteenth-century by Aḥmad al-Manṣūr's propaganda and the *Nafḥ al- ṭīb* both encapsulates Maghribī pride in the western Islamic heritage and sheds light on the material that a man like al-Tamgrūtī would have used to shape his account. Al-Maqqarī's monumental work includes lengthy descriptions of Umayyad Córdoba, its great mosque, its suburbs and the nearby palatine cities of Madīnat al-Zahrāʾ

<sup>21</sup> For other Maghribī embassies to Constantinople, see El Moudden, 'Sharifs and Padishahs'.

<sup>22</sup> See Dejugnat, 'Voyage au centre du monde'.

<sup>23</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 142–143. 24 al-Maqqarī, *al-Nafḥ al-ṭīb*, vol. I, 69–70.

and Madīnat al-Zāhira, drawn from western Islamic writers, which combined with experience of the contemporary Saʿdī sultanate to provide the cultural and literary background to al-Tamgrūtī's particular 'take' on Constantinople's past.

## **Al-Tamgrūtī's description of Constantinople**

As a city outside the purview of the medieval Maghribī tradition of travel writing, al-Tamgrūtī's account of Constantinople uses an experiential mode, incorporating his own observations and information gleaned from the Ottoman environment around him to inform his Moroccan audience about a destination that would have seemed exotic to many.25 Although his description has an unedited quality, with topics jumbled together and asides that break the narrative flow, he frames it according to the established conventions of Arab-Islamic writing on cities in geographical works as well as *riḥla* narratives, flagging sacred, topographical and ethnographic features. He gives his narrative a particular twist, however, that minimises the role of the Ottomans in the history of the city and privileges Arab and early Islamic associations, thereby placing it in an alternative historical sweep that culminates with the Arab Saʿdī 'caliphs' rather than their Ottoman competitors.

Al-Tamgrūtī commences his description by positioning Constantinople within prophetic and eschatological time. Rather than identifying it as the Ottoman capital, he alludes to its imperial and Christian past as the 'City of Caesar' founded by Constantine and, while he acknowledges its transformation into one of the 'most dominant and mighty cities in the world' under Muslim rule, he says nothing of the Ottoman conquest at this point but talks instead of the Prophet's foretelling of its Islamic conquest and early Arab Muslim attacks on the city.

Reports of the Prophet – peace be upon him – in a number of *ḥadīth* testify to [Constantinople's] fame. Among them, his comment – peace be upon him – 'The first fighters from my community who attack the city of Caesar will be pardoned [for their sins].' This city is the one meant by 'city of Caesar' and Yazīd b. Muʿāwiya attacked it during the caliphate of his father. During this campaign Abū Ayyūb al-Anṣārī al-Najārī died – may God have mercy on him – and they left the city without conquering it.<sup>26</sup>

The location of a city within Islamic sacred history was a fairly common technique but by taking this stance al-Tamgrūtī also creates an association with the Saʿdī sultans, given their much vaunted descent from the Prophet and dedication to *jihād* against Christian enemies. He then quotes a number of eschatological *ḥadīth* in which the Prophet linked the conquest of Constantinople to the coming of the final Hour, in other words, the Apocalypse, taken from the *Tadhkira* of the Andalusī scholar, al-Qurṭubī. These *ḥadīth* are reported on the authority of Muslim, Ibn Māja and al-Tirmidhī and

<sup>25</sup> On the 'citational' and 'experiential' modes of writing about place, see Antrim, *Routes and Realms*, 67–68, 70–74.

<sup>26</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 88.

give a variety of scenarios for the fall of Constantinople to the Muslims, the lengthiest of which states:

Muslim reported from Abū Hurayra – may God have mercy on him – from the Prophet – peace be upon him -–who said, 'Have you heard of a city with one side facing the land and the other facing the sea?' 'Yes, Messenger of God,' they replied and he said, 'The Hour will not come until 7,000 men from the Banī Isḥāq have marched against it, but when they reach it and make camp they will not fight with weapons or fire arrows.27 They will say, "There is no god but God and God is great", and one side of the city will fall.' The raconteur added, 'Only those at sea will realise this and they will call again, "There is no god but God and God is great" and the way will be opened for them and they will enter the city and plunder it. While they are dividing the booty a shriek will reach them, indicating that the Anti-Christ has departed.'<sup>28</sup>

Al-Tamgrūtī's stress on this aspect picks up on the widespread millenarian expectations generated by the approach of the first Islamic millenium in 1591–1592 CE, the long dramatic encounter between the Christian Habsburgs and the Muslim Ottomans and the messianic claims of the Saʿdīs themselves.<sup>29</sup> He makes no attempt to reconcile this with the fact that the city had been in Muslim hands since 1453 and he only alludes to Mehmed's conquest in a brief aside within a discussion of Ottoman slavery.<sup>30</sup> It is an open question whether this implies that the city is yet to fall to the true caliphs of the Islamic world, or whether it is a device to elide the Ottomans by focusing on Constantinople's symbolic past rather than current political status.

After this rousing start, al-Tamgrūtī moves to a conventional summary of the city's facilities, saying, 'This city is very large and walled with numerous gates, buildings, local mosques, great mosques, markets, baths and caravanserais,' a generic list common to descriptions of Baghdad, Córdoba and many other major cities in the Islamic world. He shows great interest in the harbour and waterfront, and the array of ships he sees there, many of which he identifies using terms derived from French and Spanish, perhaps useful intelligence for the Saʿdī sultan as well as a reflection of the city's impressive maritime facilities. He captures the bustle of the harbour, which must have been a phenomenal sight for a man accustomed to living on the Saharan fringe, by comparing the skiffs and lighters rushing around the harbour to ants, and their constant use for errands to Moroccan use of mules.31

After a brief comment on the general topography of the city above the harbour, with 'Istanbul' on the right and 'Galata' on the left, and its extensive markets, al-Tamgrūtī sweeps through Istanbul from the shoreline to the landward limits, describing its key religious buildings and palaces commencing with Hagia Sophia, followed by its

<sup>27</sup> The Banī Isḥāq, the 'Sons of Isḥāq (Isaac)', means the Arabs whose Biblical progenitor was Isḥāq son of Ibrāhīm (Abraham).

<sup>28</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 88–89.

<sup>29</sup> Mouline comments on the messianic tone of the era. Mouline, 'Sens et puissance', 152–154.

<sup>30</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 103.

<sup>31</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 90.

Ottoman competitor, the Süleymaniye Mosque, discussed below. He diverts briefly to stress the size and complexity of the city through reference to the vast numbers of buildings damaged by a recent fire that affected only a small portion of the city. He then moves through the urban space from the Golden Horn promontary towards the outer land walls, noting the two Ottoman palaces, the Topkapı Saray and the Eski Saray, before moving beyond the walls to give a description of the shrine of Abū Ayyūb al-Anṣārī (Turkish: *Eyüp*), the companion of the Prophet mentioned at the start of al-Tamgrūtī's account, who was said to have died during a seige of Constantinople.

As for the landward third of the city, that is where the tomb of Abū Ayyūb al-Anṣārī is situated outside the town itself: it has a large shrine complex, a great mosque and numerous pious endowments. The tomb itself is covered by a silk cloth surrounded by copies of the Qurʾān on stands from which the pilgrims may read. A number of candles gathering in gilded candelabra around the enclosure illuminate it. There are gate-keepers and custodians at the entrance and people flock to visit it all the time. The great men of state compete to be buried near [the tomb] and pay large sums for funerary plots, as do those who are able from amongst the pious and the common folk. We visited it and sought his blessing and prayed in his presence – may God have mercy upon him.32

Although the shrine complex was a fifteenth-century Ottoman construction and the rites associated with it entirely Ottoman, al-Tamgrūtī makes no mention of the Ottoman 'discovery' of the tomb and he glosses over its role in commemorating the Ottoman fulfilment of the prophecy that Rome (Rūm) would be conquered. Instead he recounts further *ḥadīth* related to Abū Ayyūb and early Islamic history from the Andalusī corpus.

The imam and *ḥāfiẓ*, Abū ʿUmar Yūsuf b. ʿAbd Allāh b. Muḥammad b. ʿAbd Allāh [b. ʿAbd] al-Barr al-Namarī al-Andalusī,33 said in his *Kitāb al-istiyʿāb* in a description of Abī Ayyūb, based on information from Khālid b. Zayd b. Kulayb b. Thaʿlaba, that Abū Ayyūb al-Anṣārī al-Najārī was from the Banū Ghānim b. Mālik b. al-Najār. His mother Hind, daughter of Saʿīd b. ʿAmr b. Imru'l-Qays b. Malik b. Thaʿlaba b. Kalb b. al-Ḥāritha b. al-Khazraj the elder, imposed his *kunya* on him. He was present at al-ʿAqaba, Badr and other early battles, and the Messenger of God – peace be upon him – stayed with him when he left the Banū ʿAmr b. ʿAwf and migrated from Mecca to Medina. He remained with him until he built the mosque and his own dwellings during that [first] year and moved to his own home.<sup>34</sup>

This section emphasises the Arabian ancestry of Abū Ayyūb, his involvement in key early Islamic battles and the intimate, almost familial, relations between him and the Prophet, which subsequent *ḥadīth* reiterate. While such *ḥadīth* were widely known, al-Tamgrūtī's choices carefully distance Abū Ayyūb from the Ottoman context and

<sup>32</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 95.

<sup>33</sup> This is the famous Andalusī scholar and man of letters, Ibn ʿAbd al-Barr (d. 1071). The full name of the work was *al-Istīʿāb fī asmāʾ al-aṣḥāb*, a biographical dictionary of the companions of the Prophet. Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, note 1, 95; Lucas, 'Ibn ʿAbd al-Barr', EI3, consulted online on 4 April

<sup>2019. 34</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 95–96.

highlight his status as part of the intimate circle around the Prophet, the illustrious ancestor of the Saʿdī sultans. His description of Constantinople thus metaphorically girds it with *ḥadīth* that stress its Arab and early Islamic associations in keeping with contemporary Saʿdī religio-political theory rather than Ottoman reality.

Al-Tamgrūtī moves on to further general description of the urban fabric, somewhat randomly organised, with observations about the use of wood in domestic construction, the ancient obelisks and serpent column at the hippodrome with their friezes of people, animals and objects, the ever present dangers of fire, mingled with an illustrative account of his visit to an unnamed minister's house where rich textiles abounded in dangerously close proximity to the fire. He then returns to urban topography with a mention of Üsküdar on the Asian side and its gardens and parks, before making more general observations about the city's theological colleges and Sufi retreats which offer food to travellers, and the shipyards in the Galata area. Although it would be difficult to find one's way around based on al-Tamgrūtī's description, his sense of the constituent parts of the city is fairly accurate.

After sketching the city, his brief report on the Maghribī embassy's audience with Murad III acts as a preface to a lengthy description of Ottoman court protocol and governance that blends into a discussion of the Friday prayer and sermon in Constantinople, and religious life more generally. Throughout al-Tamgrūtī's work, he studiously avoids giving too much praise or recognition to the Ottoman sultans themselves. Murad III and his dynasty are never called more than *khāqān* and *khān*, Turkish titles, or *sulṭān*, a term in Arabic that simply denoted a holder of power, with none of the honorific adjectives normally added to embellish such a title, despite the multiplicity of titles they used for themselves. He alludes surprisingly briefly to the actual audiences with Murad III, further downplaying the status of the Ottoman sultan in comparison to Aḥmad al-Manṣūr. In a similar vein, his detailed description of the sophistication of Ottoman government, and the consequent ability of the ministers and judges to act without constant reference to the sultan, provides him with an opportunity to caricature Murad III as irresponsible and given over to pleasures in contrast to Aḥmad al-Manṣūr, who appears in Maghribī sources as an active warrior, statesman and religious scholar.35

Al-Tamgrūtī then makes a series of comments on the character of Ottoman society, noting the Ottomans' commercial acumen, people's love of the good things in life, and preference for slaves over free-born Turks as sexual partners, servants, soldiers and companions in a manner that El Moudden describes as alternating between 'fascination and rejection'.<sup>36</sup> Al-Tamgrūtī concludes his section on Constantinople by returning to religious learning and the great number of books available, with a cursory mention of the Maghribī embassy's leave-taking of the sultan.

<sup>35</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 99. On Maghribī depictions of Aḥmad al-Manṣūr see, Cory, *Reviving the Caliphate*, 73–80, 109–113. On mixed Ottoman attitudes to Murad III, see Woodhead, 'Murad III and the historians'.

<sup>36</sup> El Moudden, 'Ambivalence', 80.

### **Constantinople's great mosques and the memory of Umayyad Córdoba**

Maghribī travellers combined their sense of awe at the monuments of the Islamic east with pride in the western Islamic heritage which was often focused upon Córdoba, the Andalusī Umayyad capital, and, to a lesser degree, Qayrawān, capital of early Islamic Ifrīqiya (eastern Algeria, Tunisia and western Libya). In a number of medieval geographical works and travelogues, descriptions of these two cities are dominated by lengthy and detailed accounts of the wondrous architecture and decoration of their great mosques which become incarnations of the city as a whole, otherwise described with the broadest of brushstrokes. In a similar fashion, al-Tamgrūtī pays considerable attention to Hagia Sophia and the sixteenth-century Süleymaniye Mosque built to rival it in the course of his description of Constantinople.

His account of Hagia Sophia is made up of two parts, a mytho-historical section and a lengthy and fulsome description of its architectural form and decoration based on his own visit(s) to the building. As with his introduction to the city, al-Tamgrūtī focuses on the pre-Islamic Christian past when recounting the history of Hagia Sophia.<sup>37</sup> In keeping with the expectation that founders of cities found their chief religious buildings, he reports:

Prior to Islam, [Hagia Sophia] was a mighty church – God alone knows if it was built on the same scale of magnificence as in Jerusalem – which some histories mention was built by a king called Constantine who became a Christian along with his subjects and then moved to Constantinople, or built the city himself and built its churches and temples, including this one. He is depicted inside it sitting on his throne. It is also said that Aṣaf b. Barkhiyā, the son of the prophet Solomon's maternal aunt – peace be upon him – built it. Ibn Baṭṭūṭa says this but God alone knows what is closest to the truth since its construction certainly resembles the work of the *jinn* and humans are generally unable to do such things.38

In this passage, al-Tamgrūtī associates the building with Constantine, rather than with his son Constantius II (r. 337–361), who founded the first church on the site, or Justinian (r. 527–565), who founded the sixth-century building which the Ottomans converted into a mosque. This creates a satisfying, although factually inaccurate, parallel between Constantine's Hagia Sophia and Süleyman the Magnificent's Süleymaniye Mosque, designed by the Ottoman architect Sinan who spent his professional life in dialogue with Hagia Sophia.<sup>39</sup> Al-Tamgrūtī recognises this dynamic, but places the Süleymaniye below Hagia Sophia, saying:

There are other mosques which try to imitate this most great mosque but fail to achieve this. The closest to it is the Süleymaniye, the mosque which the sultan Süleyman built and where he is buried … Hagia Sophia is a stronger, more magnificent and powerful building while the Süleymaniye is more ornate, pleasant and spacious and God alone knows the

<sup>37</sup> Contrast the traditional Ottoman focus on its Islamic past to the neglect of its Christian history, noted by Akyürek in this volume.

<sup>38</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 91. 39 See Necipoğlu, 'Challenging the past'.

difference between their founders with respect to Islam and unbelief, and upon each rests the imprint of its founder's vision.40

Al-Tamgrūtī's alternative myth that the church is of Solomonic origin is a common trope linking places to the archetypal Qurʾanic ruler and pre-Islamic prophet, and explaining apparently superhuman construction through Solomon's legendary ability to command the *jinn*. His reference to a cousin of Solomon called Aṣaf b. Barkhiyā may be a corruption of Ottoman lore in which Asafiya is the name given to the wife of Constantine, the namesake of the building.<sup>41</sup> As he often does, al-Tamgrūtī cites a Maghribī authority, Ibn Baṭṭūṭa, for this story but his account also repackages the Ottoman, and earlier Byzantine, conceptualisation of Justinian's Hagia Sophia and the Süleymaniye as emulating Solomon's temple in the new Jerusalem.<sup>42</sup>

Furthermore, Sinan's autobiography contains a comparison of Frankish prisoners working away on the Süleymaniye to 'Solomon's demons', which plays on the shared name (Solomon/Süleyman) of the prophet and the Ottoman sultan,43 an obvious comparison to which al-Tamgrūtī would have had exposure during his stay. Al-Tamgrūtī thereby gives the building a layered past which interweaves shared Christian and Muslim monotheistic tropes and likely draws on the stories told about Hagia Sophia in Constantinople itself. What is missing, however, is any explicit account of the church's transformation into a mosque after the Ottoman conquest, in keeping with the general elision of this phase in Constantinople's history.

After summarising the legends about the foundation of Hagia Sophia, al-Tamgrūtī gives his description of its architecture and decoration which, I would argue, combines his eye-witness experience with his awareness of extant literary descriptions of the great mosques of the Islamic west, and above all the great mosque of Córdoba, emblem of the western Umayyad caliphate. In al-Idrīsī's twelfth-century geographical work, for instance, the entry on Córdoba is entirely dominated by a lengthy description of the great mosque's superlative size, beauty, abundance of marble columns and fine decorative carving, woodwork and gold mosaics.<sup>44</sup> In the centuries after Córdoba's fall to the Castilians in 1236, the lost Umayyad great mosque became an intensely nostalgic *lieu de memoire* for Andalusī Muslim migrants to the Maghrib and indigenous Maghribīs alike, which they 'visited' through textual descriptions which preserved the mosque in a perpetual present.

Aḥmad al-Maqqarī, who was of distant Andalusī ancestry, gathered together many such passages describing Córdoba and its great mosque while he was in the Saʿdī

<sup>40</sup> Literally, 'the mantle of its founder's heart'. Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 93. Al-Tamgrūtī's comparison of the buildings is similar to Evliya Çelebi's assessment, although the latter considered the Süleymaniye superior. Necipoğlu, 'Challenging the past', 174.

<sup>41</sup> See Yerasimos, *Légendes d'empire*. I am grateful to Suna Çağaptay for directing me to this source.

<sup>42</sup> Necipoğlu, 'Challenging the past', 173–174. Ibn Baṭṭūṭa does mention this legend. Gibb, *The Travels of Ibn Battuta*, vol. II, 509.

<sup>43</sup> Morkoç, 'Reading architecture from the text', 33.

<sup>44</sup> Al-Idrīsī, *Opus Geographicum*, 575–578.

sultanate and reproduced them in his *Nafḥ al-ṭīb*, a large compendium containing a history of al-Andalus, its people and its literature, and a biography of one of its most famous literary figures, Ibn al-Khaṭīb of Granada. The section on the great mosque starts by saying, 'As for the great mosque of Córdoba, its fame is such that lengthy descriptions are unnecessary … Some historians say there is no greater building in the lands of Islam, and no more marvellous structure or fine construction.'<sup>45</sup> Several descriptions, nonetheless, follow enumerating the structure and listing its richest decoration and furniture.

The length of the courtyard from east to west was 128 cubits46, and its extent from the *qibla* to the north was 105 cubits, the width of the galleries around the courtyard was 10 cubits. Its surface area was 33,150 cubits. It had nine gates, three on the west, east and north sides of the courtyard, four [entering] the prayer hall, two to the east and two to the west, two leading to the women's enclosures from the courtyard galleries. It contained 1,293 pillars, all of them marble and the door into the *maqṣūra* was gold, the walls of the *miḥrāb* and surrounding area were likewise covered in gold mosaics. The *maqṣūra* lamp holders were pure silver. The height of the tower today … is 73 cubits to the top of the open baldachin … on the top of which are gold and silver apples whose circumference is three and a half spans. Two are made of pure gold and the other of silver. Beneath each is a stand fashioned like a lily of most wonderful craft with small gold pomegranates at the base.<sup>47</sup>

The passages al-Maqqarī cites also mention the famous Qurʿān of ʿUthmān, the *minbar* made of precious woods, and a massive chandelier suspended in the main *maqṣūra*  dome, purportedly able to hold 1,054 lights that were lit in Ramadan, among other wonders.48 It is likely that al-Tamgrūtī's descriptions of the physical characteristics of Hagia Sophia and the Süleymaniye were informed by such accounts of the lost wonders of Córdoba. Although he makes no explicit comparison between Hagia Sophia, the great mosque of Córdoba, or the great mosques of Marrakesh and Fes that he would have been personally familiar with, he moves through the Ottoman mosque space in a way which echoes the conventions used in descriptions of the Umayyad mosque.

At the same time, his description is a sensory one, rooted in his personal engagement with the space. In this sense, it chimes with Selen Morkoç's suggestion that pre-modern Islamic architecture should be understood as an occasion, or a perpetual series of occasions, encompassing construction, completion and ongoing use, rather than as buildings *per se*, and that contemporary textual accounts reflect this experiential sensibility.49 Western Islamic mosques had a hypostyle T-plan, derived originally from the Umayyad great mosque of Damascus, and Hagia Sophia may have been the first centrally planned domed mosque al-Tamgrūtī ever visited, making this marvel a central feature of his description. While the endless vistas created by countless rows of marble columns engendered wonder in the Córdoban

<sup>45</sup> Al-Maqqarī, *al-Nafḥ al-ṭīb*, vol. I, 545.

<sup>46</sup> *Dhirāʿ*, literally a 'fore-arm', roughly two-thirds of a metre.

<sup>47</sup> Al-Maqqarī, *al-Nafḥ al-ṭīb*, vol. I, 547–548.

<sup>48</sup> Al-Maqqarī, *al-Nafḥ al-ṭīb*, vol. I, 551.

<sup>49</sup> See Morkoç, 'Reading architecture from the text'.

mosque, al-Tamgrūtī uses his personal experience of Hagia Sophia's vast dome space in relation to human and animal dimensions to create similar amazement.

[Hagia Sophia] is one of the most amazing and great buildings of ancient construction and one cannot capture it in description or realise what it is like without seeing it … In the middle of the mosque there is mighty, spacious dome so extremely high in the air that when a pigeon flies under its arc, a person sitting below would think it a sparrow. The soaring, golden dome was the first thing that we saw on land. Its circumference is more than a hundred paces, and likewise its arc from base to zenith. It rises from massive piers, that were constructed by carving huge, extraordinary, pieces of hewn rock which resemble the sheer sides of mountain peaks, and from columns of veined, coloured marble, which soar in the air and are so huge that the arms of two men cannot stretch around them. Around this mighty central dome, there are other domes beyond the piers and columns which have lower ceilings than the central one and a floor on another level where people pray and can look over the central dome from a gallery running around it, erected on iron pillars separate from the wall. I looked from it myself one day and saw men under the dome who looked like mere children because of the height.<sup>50</sup>

Although the decorative elements al-Tamgrūtī identifies are to some extent conventional, marble panels and gold decoration signal the virtuosity of Hagia Sophia in his description just as marble columns and gold mosaics define the great mosque of Córdoba.

The floor of this mosque is entirely covered with marble panels and all its inner walls are lined with marble. The interior is decorated with all sorts of ornamentation and filigree work of varied manufacture and diverse colours, each unlike the other but of outstanding beauty and marvellous design. Everywhere molten gold had been poured and stamped with octagons, hexagons and squares and gilded vegetal and line patterns, leading to the judgement that it must have been fabulous when it was completed: a marvellous and glittering sight, were it not for the effect of time upon its beauty.<sup>51</sup>

Al-Tamgrūtī also pays considerable attention to Hagia Sophia's 'lamps of gold and glass' and refers to their lighting during Ramadan, as well as mentioning the *minbar* 'entirely hewn from one block of pure marble with a gilded dome at the top' and the *miḥrāb* also 'hewn from marble' with 'copies of the Qurʾān on stands' around it.<sup>52</sup> His inventory of the furniture thus mirrors that in descriptions of the great mosque of Córdoba. The one aspect which cannot be compared is extant Byzantine Christian figural decoration. In a rare acknowledgement of the building's transformation from a church to a mosque, al-Tamgrūtī admires these images and offers his own identifications of them but implicitly criticises the Ottomans for allowing them to remain in a mosque.<sup>53</sup>

<sup>50</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 91–92.

<sup>51</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 92.

<sup>52</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 92–93.

<sup>53</sup> By this time, of course, the great mosque of Córdoba had itself become a church full of images and crosses.

Inside the mosque there were also a variety of images and crosses, including depictions of the angels Gabriel, Michael, ʿAzraʾīl, Isrāfīl and others, and depictions of the prophets on the first floor, including John, Zakariah, Mary with her son Jesus on her lap and Jesus's cradle, and other frivolities of the unbelievers. When the Muslims entered [the church], they destroyed all the images of the cross and some others too but they left others.54

Hagia Sophia receives the lion's share of al-Tamgrūtī's attention, marking it as Constantinople's premier religious building, but he does provide some information on the Süleymaniye, the most important mid-sixteenth century Ottoman imperial building project in the city. Al-Tamgrūtī pays most attention to the importation of its rare columns from different parts of the empire, something common to a number of western Islamic status buildings but also of considerable interest to Ottoman authors.55 The provenance of the stone used in the Süleymaniye has been carefully analysed by Kolay and Çelik using archival documents, and historians have pondered the symbolic meaning of the acquisition of the four large red Aswan granite columns from separate locations.56 Al-Tamgrūtī's account says:

Four marble pillars were brought from Alexandria on two ships and one sank with two pillars aboard and the other made it safely with two pillars which they placed in [the mosque]. I myself met a man from Monastir who said he was present in Alexandria when they extracted these pillars, which they took out via a breach they made in the walls as they were not able to get them through the gates.57

He therefore recounts a story to be found in Muṣtafā Celālzade's *Tabaqāt ül-memālik* and presumably well-known in Constantinople but attributes it to a man from Monastir, *i.e*. a Maghribī eye witness.<sup>58</sup> He does not comment on the columns from sites in Constantinople which received more Ottoman attention, pointing again to his selective interpretation of his experience and its appropriate presentation for his Maghribī audience. In his telling, the episode carries little imperial symbolism, beyond the fact of re-use, and he does not allude to the Ottoman inheritance of ancient empires or their Mediterranean territories. Conversely, al-Tamgrūtī's reference to the loss of two columns at sea implies a failed attempt at acquisition which the Saʿdī sultan, whose own palace was filled with Carrera marble and other expensive imported materials, would no doubt have relished.

## **Conclusion**

Al-Tamgrūtī's account provides a window into the mentality of a Saʿdī official from Morocco visiting Constantinople in the last decade of the sixteenth century. While

<sup>54</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 93.

<sup>55</sup> Morkoç, 'Reading architecture from the text', 32–33, 42–46.

<sup>56</sup> Aktuğ-Kolay and Çelik, 'Ottoman stone acquisition', 264; Morkoç, 'Reading architecture from the text', 33.

<sup>57</sup> Al-Tamgrūtī, *Al-Nafḥa al-miskiyya*, 93.

<sup>58</sup> Aktuğ-Kolay and Çelik, 'Ottoman stone acquisition', 264.

the *riḥla* as a whole has been dismissed as a poor piece of Arabic literature and may sometimes appear naïve, its author deserves more credit for agency in crafting his account to interweave his political role and the literary tradition of which he was part. In keeping with Aḥmad al-Manṣūr's religio-political agenda, al-Tamgrūtī minimises his praise of the Ottomans themselves and uses every opportunity to allude to the noble Arab ancestry of the Saʿdī dynasty. This is most striking in his positioning of Constantinople within early Islamic Arab rather than Ottoman history and his emphasis on the *longue durée* Christian-Muslim relationship not its contemporary Ottoman manifestation.

His topographical description shows his familiarity with the conventions of the genre and balances citational and experiential material to introduce the city as a whole to his elite Maghribī audience. There is much that impresses him: Constantinople's size, its maritime and commercial facilities and many of its buildings, including Hagia Sophia, the Süleymaniye, the Topkapı Saray and the shrine of Abū Ayyūb. However, Hagia Sophia stands out as the emblem of the city, a function fulfilled by the great mosque in descriptions of Córdoba circulating in Saʿdī Morocco and collated by Aḥmad al-Maqqarī in the early seventeenth century.

Al-Tamgrūtī tailors the basic building blocks from descriptions of the great mosque of Córdoba to capture his experience of a radically different building in a way which celebrates and subverts it. Hagia Sophia emerges as a masterpiece but of Byzantine rather than Ottoman architecture, an identity represented visually in its Christian mosaics. His weaving of legends likely garnered from the contemporary Ottoman city with citations from Maghribī sources adds to the impression of a building hovering between its Christian and Muslim phases, a living testament to Byzantine genius that the Ottoman Süleymaniye cannot compete with. For Maghribī readers aware of the Córdoban Umayyad narrative, wonder thereby jostles with a sense that the Ottomans, unlike the Umayyads of the past, are not truly caliphal, a status which Aḥmad al-Manṣūr claimed as his own both as a descendant of the Prophet and an heir to the Umayyad caliphal tradition.

### **Bibliography**

Aktuğ-Kolay, İ. and S. Çelik, 'Ottoman stone acquisition in the mid-sixteenth century: The Süleymaniye complex in Istanbul', *Muqarnas* 23 (2006): 251–272.

Antrim, Z., *Routes and Realms: The Power of Place in the Early Islamic World* (New York and Oxford, 2012). Cory, S., *Reviving the Caliphate in Early Modern Morocco* (Aldershot, 2013).


## Chapter 19

## Beirut as a palimpsest: Conflicting present pasts, materiality and interpretation

*Assaad Seif*

This conflation of above and below endorses the view that the city possesses a multilayered structure, a palimpsestic nature, in this vision of a verticalized world […] where cultural, as well as individual, repressed fears and desires are lodged, there is a sense of haunting and spectrality.<sup>1</sup>

#### **Introduction**

Beyond its description as a 'multi-layered record',<sup>2</sup> the palimpsest, as a metaphor, has been widely used since it was coined by Thomas De Quincey in the early nineteenth century, in his seminal work *The Palimpsest*. 3 This palimpsest trope was later borrowed to describe landscapes and their chronological superimpositions<sup>4</sup> or to explain the layering of historical events and processes,<sup>5</sup> as well as the reconstruction of urban imaginaries.<sup>6</sup> Furthermore, as a metaphor of chronological erasure and superimposition, it was extensively used by archaeologists to represent the effect of differential temporalities on the archaeological record and layers<sup>7</sup> or to connect it with the concept of time scale, through what was described as 'time perspectivism'.<sup>8</sup> Others, in an attempt to enlarge its applicability, proposed new concepts that can be

<sup>5</sup> Crang, 'Envisioning urban histories', 443.

<sup>1</sup> Arias, 'Haunted places, haunted spaces', 154.

<sup>2</sup> *Oxford English Dictionary*, 95.

<sup>3</sup> Dillon, 'Reinscribing De Quincey's palimpsest', 243; Dillon, *The Palimpsest*.

<sup>4</sup> Lowenthal, *The Past is a Foreign Country*, 57.

<sup>6</sup> Huyssen, *Present Pasts*, 7.

<sup>7</sup> Binford, 'Behavioral archaeology and the Pompeii premise', 197; Bailey, 'Time perspectives'; Bailey and Galanidou, 'Caves, palimpsests and dwelling spaces'; Malinski-Buller *et al*., 'Making time'.

<sup>8</sup> Bailey, 'Concepts'; Bailey, 'Time perspectives'; Bailey, 'Time perspectivism'; Sullivan III, 'Time perspectivism and the interpretive potential of palimpsests'; Holdaway and Wandsnider, 'Time in archaeology: an introduction'.

socketed into the main metaphor to expand its use and explanatory dimension; thus the concept of brecciation was introduced in order to better illustrate the presence and co-existence of various types of juxtapositions at a given site.<sup>9</sup> Moreover, the notion of a 'cumulative palimpsest' was introduced in order to explain the non-reversibility of the archaeological record<sup>10</sup> or to introduce the archaeological assemblages of a palimpsestic nature into the realm of behavioural archaeology.<sup>11</sup>

There is no doubt that the aforementioned studies and thorough analysis have provided solid grounds for archaeologists to use and thereby adopt the palimpsest metaphor in the study of the archaeological record. However, through the study of Beirut as a palimpsest, this chapter aims to shed light on the interactions between unseen factors and driving forces that, at the urban and archaeological levels, produced the palimpsestic nature of this city. It will tackle the palimpsest metaphor at the second level, thus shifting focus from the metaphor itself to the essential paradigms lying behind the observation and interpretation of past materiality, paying close attention to their ontological and epistemological dimensions.

In addition, it will attempt to understand the interplay of all these different factors within the social realm to produce subjective meanings through the various narratives proposed by different actors. For this purpose, three case studies from Beirut's post-war reconstruction experience will be examined and discussed. Through these case studies an attempt to bring together both theoretical and more tangible aspects of the palimpsest metaphor will be made by bridging the social facts to the social actions in a real-world scenario.

This examination will include five monuments. Three have been newly built: the al-Amin Mosque, the related Hariri Memorial and the campanile of the Maronite Saint George's Cathedral. The two other monuments to be discussed have been recently restored along with archaeological excavations: the Zawiyat Ibn 'Arraq al-Dimashqi and Saint George's Greek Orthodox Cathedral.

#### **Beirut: Cities within a city**

The Lebanese [religious] communities are in fact true nations and it is only relatively recently, as has already been pointed out, that the word community has been applied to them to designate them. For a long time, they were designated by that of the nation which was much better suited to their real nature.<sup>12</sup>

In order to understand Beirut beyond or, rather, beneath its cosmopolitanism, we need to comprehend the multifaceted nature of the compositional socio-political and sectarian fabric of its communities.

<sup>9</sup> Bartolini, 'Critical urban heritage', 531.

<sup>10</sup> Bailey, 'Time perspectives', 204; Prera, 'Cumulative palimpsest effect', 35.

<sup>11</sup> Way, 'Behavioral approach to cumulative palimpsests', 59.

<sup>12</sup> De Vaumas, 'La répartition confessionnelle au Liban et l'équilibre de l'Etat Libanais', 554.

Etienne De Vaumas,<sup>13</sup> in his above quote, offers a succinct definition of the different communities/confessions in Lebanon in general. Sélim Abou<sup>14</sup> concurs with this when he speaks about the communities and the ethnic composition of the population of Beirut. Nevertheless, this can be easily extrapolated to the entire Lebanese fabric. While trying to put it in its conceptual and theoretical framework, Abou explains that the Lebanese communities or confessional groups are separate nations according to the German sociological definition which defines the nation by its basic ethnological traits – race, language, religion, territory and economic interests.<sup>15</sup> However, Lebanon was founded under the auspices of the French mandate according to the French paradigm, which is inclined to the '*vivre ensemble*' (living together) principle, according to which the population (or the group) eventually has to assume together a common historical heritage. It is here where the complexity of the Lebanese paradox lies. This paradigmatic paradox which penetrated the structure of the Lebanese constitution and produced the 'miracle'<sup>16</sup> of the political system in Lebanon, where everything is divided equally between the confessional communities. However, the Lebanese territory, and in the present case that of Beirut, remains roughly divided according to dominant communities with little sign of a move towards that common historical heritage.<sup>17</sup> Charbel Nahas<sup>18</sup> expresses it very well when he says that:

the constitutional text which establishes the principle of the confessional sharing of public institutions is revealing. It poses confessionalism from the start in a problematic and almost discursive form. By its formulation, it implicitly refers to the Western model of the State from which the articles of the Constitution, in general, are inspired (the Lebanese Constitution of 1926 is indeed, on many points, similar to the French Constitution of the Third Republic) and on the other hand, to an image of Lebanese society (made up of communities) given as reality itself.

He adds that this Lebanese constitution 'establishes a justification for "political confessionalism", manifestly perceived as contrary to both the constitutional model

<sup>13</sup> Etienne de Vaumas (1914–1974) was a French Jesuit priest, a geographer and a prehistorian. He worked in the Levant, mainly Lebanon, where he published several works on the geology, human geography, demography and prehistory of Lebanon. His work had a great impact on Lebanese scholars working in the domains of prehistoric archaeology, geology and human geography.

<sup>14</sup> Sélim Abou (1928–2018) was a renowned Lebanese Jesuit priest, writer, philosopher and anthropologist who had a major impact on the social and behavioural sciences in Lebanon. He founded the Faculty of Letters and Human Sciences at the Saint Joseph University in Beirut, of which he was the Dean before serving as Rector of the University.

<sup>15</sup> Abou, 'La composition ethnique', 96–97.

<sup>16</sup> In Arabic, more specifically in the spoken Lebanese dialect, miracle is *ʿajībeh*; this word has also a pejorative meaning when used to mean 'abnormal' in the biological sense. This pejorative meaning is what we intend to convey by the word 'miracle' in the present text, thus playing on the sarcastic shift between both meanings.

<sup>17</sup> Abou, 'La composition ethnique', 97.

<sup>18</sup> Charbel Nahas (b. 1954) is a Lebanese scholar, critical thinker, university professor and twice minister. He is the founder of the left wing 'Citizens in a State' movement, a platform from which he intervenes and debates on the socio-economic functions of the state, its institutional functioning, the symbolism and legitimacy of powers, as well as on social history.

and *a fortiori* to the constitutional provisions, namely the equality of citizens before the state and the political reality'.<sup>19</sup>

In this sense, Abou asserts that communities living in Beirut identify themselves primarily by religion,<sup>20</sup> while explaining that the distinction between Christian and Muslim groups is significant especially from the point of view of political ideology.<sup>21</sup> Furthermore, this community-based sectarianism is anchored in the socio-political system in Lebanon to such an extent that the sovereignty of the state stops at the doorstep of the religious space – the community's territory.

In fact, this sectarianism, derived from long-established systems of confessional feudalism, is taken to be the main constituent of the paradigmatic structure of Lebanon. Consequently, the struggle between communities is a kind of an internal driving force which maintains the structural cohesion of the system. As Ahmad Beydoun22 clearly states: 'We believe that it was not war that led to the formation of communities, though it might have consecrated them and altered their orientation and development. It was rather that these groups wage wars as the only available means of renewing their internal structures and their relationship with other stakeholders.'<sup>23</sup> Thus, the struggle between these entities has continued and will remain for as long as the same socio-political paradigms continue to rule the country.

In this sense, the 1975 civil war (1975–1990) was a product of all these internal driving/conflicting forces which tore the country apart. Nevertheless, when the war ended, the same dynamics of conflict continued in a different register, using different tools.

### **Post-war Beirut, sectarianism and the struggle for the re-appropriation of Beirut's past**

As soon as the armed conflicts ended, preliminary work to clear the rubble from Beirut's city centre began, making way for the city's new development master plan, officially presented to the public in September 1991.<sup>24</sup> In December of the same year, Law 117 gave the municipality of Beirut the right to create real estate companies and to entrust them with the implementation of the Urban Master Plan.<sup>25</sup> In 1992, immediately after Prime Minister Rafik Hariri took office, the real estate company SOLIDERE was set up.<sup>26</sup>

<sup>21</sup> Abou, 'La composition ethnique', 100.


<sup>19</sup> Nahas, 'Le confessionnalisme au Liban, du fonctionnement discursif et idéologique vers une position du problème', 4–5.

<sup>20</sup> Abou, 'La composition ethnique', 98.

<sup>22</sup> Ahmad Beydoun (b. 1943) is a Lebanese scholar, critical thinker, historian, sociologist and university professor. His publications deal in particular with the affairs of Lebanese society and its political system and issues involving the Arabic language and culture.

<sup>26</sup> SOLIDERE is a 'private Lebanese Company for the Development and Reconstruction of Beirut Central

Although this was seen as the beginning of a new golden age for Beirut and Lebanon after fifteen years of a bloody civil war, the components of the conflict that drove the country into the civil war did not end when it finished.<sup>27</sup> It was only the nature of the conflict which changed. The ingredients, constituents and players, that is, the communities and their political and religious leaders, remained the same. The power struggle merely shifted from one dynamic to another.

In this scene of post-war Beirut, each politico-religious community aimed to gain or to regain distinction in the cityscape – a struggle in which each community strove to expand its territory in the city or even to gain new territory, be it physical or its territory of influence. It was more or less a slow soft power struggle which used heritage and its components to ground their different narratives in the heart of the city. This was done by various means: building new sacred spaces, restoring others from the past, or even using material remains of the past to create new narratives. All of these tactics were means of seeking to legitimise the way in which each community anchored its own past in the city. The Shiites, as well as the Sunnis, the Maronites and the Orthodox have each tried in a different way to reposition themselves within the urban space as an act of affirming sovereignty. SOLIDERE has operated in the same way through what has been described as the rehabilitation of the urban space; Rodolphe al-Khoury has described this as 'a covert warfare in the project of architecture'.<sup>28</sup>

## **Zawiyat Ibn 'Arraq al-Dimashqi: The claim for a share within the capital**

In a brilliant article, Howayda Al-Harithy<sup>29</sup> retells in fine detail the story of the discovery of this structure in April 1992. She dissects the differing attitudes within the communities involved, analysing the social, cultural, economic and political phenomena related to the narratives which have been woven around this monument.<sup>30</sup>

This domed structure (Figure 19.1) was discovered during the demolition of the Ottoman Suqs area by SOLIDERE in order to build the New Beirut Souks – a construction project designed by the architect Jad Tabet and approved in 1999.<sup>31</sup> The intention of this project is to 'modernise' the souks which flourished in pre-war Beirut at the same

District (better known by its French acronym Solidère), with controversial ties to leading politicians, including the billionaire former prime minister, late Rafik Hariri', Fricke, 'Forever nearing the finish line', 164.

<sup>27</sup> Delatolla, 'The Lebanese Civil War and post-conflict power sharing'.

<sup>28</sup> al-Khoury, 'The postwar project', 186.

<sup>29</sup> Howayda Al-Harithy is a professor of Architecture at the American University of Beirut. Her research into Islamic art and architecture focuses on the Mamluk period, engaging theoretical models of interpretation, particularly post-structuralist models, as analytic tools as to the production of architectural and urban space. Since the 1990s, she has focused on the theoretical debate on heritage construction and consumption related to identity-building and post-war reconstruction.

<sup>30</sup> Al-Harithy, 'Weaving historical narratives', 215–230.

<sup>31</sup> SOLIDERE, Annual Report, 32.

*Fig. 19.1: Zawiyat Ibn 'Arraq al Dimashqi as restored and integrated within the Beirut Souks (Assaad Seif).*

location.<sup>32</sup> Consequently, SOLIDERE implemented a systematic demolition of the old Mamluk and Ottoman souks in order to clear the way for this new construction project. This systematic destruction also operated in the different areas of the city centre in such a way that, according to critic Saree Makdisi,<sup>33</sup> 'in the months since reconstruction officially began in earnest in summer 1994, more buildings have been demolished than in almost twenty years of artillery bombardment and house-to-house combat'.<sup>34</sup>

Upon the discovery of this monument, the Shiite community rushed to the area, claiming the building as part of their religious heritage, thus sparing it from imminent destruction. This was based on the identification of the monument as being the *zawiya* of Muhammad Khidr al-'Iraqi. Furthermore, in order to ascertain its religious and sacred aspect, the collective delirium of the crowd, among whom several stories of the paranormal were told about the shrine and people's encounters with it, was taken advantage of by the Shiite religious and political authorities.<sup>35</sup>

A few weeks later, the Sunni Faculty of Islamic Studies in Beirut, after researching in their archives, on the basis of the genealogy of the said Ibn 'Arraq, presented tangible evidence through certified documents and issued the following statement: 'The zawiya that was found is that of Muhammad bin 'Ali 'Abd al-Rahman bin 'Arraq, Shams al-Din Abu 'Ali al-Kanʿāni al-Dimashqi, known as Shaykh al-Islam, and that noble great scholar is of the Shafiʿi jurists',<sup>36</sup> thus reinstating the monument with the Sunni community.<sup>37</sup> Further studies of the structure carried out later revealed evidence of the use of the *zawiya* as a burial-place under the Ottomans, when it hosted the tomb of the Ottoman rulers of Beirut, and then a religious school, before it was transformed into a shop within the suqs. After an intense debate between the two

<sup>32</sup> The construction project was then entrusted to Rafael Moneo.

<sup>33</sup> Saree Makdisi is a professor of English and Comparative Literature at UCLA. Of Palestinian and Lebanese descent, he has several publications on contemporary Arab politics and culture.

<sup>34</sup> Makdisi, 'Laying claim to Beirut', 622.

<sup>35</sup> Al-Harithy, 'Weaving historical narratives', 215–218.

<sup>36</sup> Al-Harithy, 'Weaving historical narratives', 218.

<sup>37</sup> Tens of metres to the west of the Zāwiyat Ibn 'Arrāq, there was another Zāwiya, that of al Imām Abd el Rahmān al Ouzāi from the Sunni community. It was destroyed by SOLIDERE without any reaction, neither from the Dar al-Fatwa, nor from the Sunni Faculty of Islamic Studies. A new space was later arranged for prayers in the new mall built at the location of the Zāwiya. The new Zāwiya space was inaugurated in 2015 under the auspices of the Sunni Mufti. A marble plate indicating the Zāwiya and the name of the Imām is fixed on the wall for indication purposes.

Islamic sects, the monument was delivered to the Sunni General Directorate for the Islamic Waqfs of the Dar al-Fatwa in Beirut.<sup>38</sup>

The *zawiya* was used as an ideological object in a 'tug-of-war' between Shiites and Sunnis, each claiming the centrally placed ancestor for their own. Given that all Muslim places of worship in the Beirut city centre belonged to the Sunni community, the Shiites saw in it a potential route to claiming a piece inside the heart of the capital.

Despite the history of this building, which had a narrative of its own (*zawiya*, grave, madrasa, then shop), Al-Harithy recounts the new narratives woven around it, starting with the socio-political sectarian dispute over its ownership, going on to the new narrative constructed by SOLIDERE, which uses the monument as a token of historicity in its mall-like development, thus serving its economic ends, and, finally, the narrative created by Moneo,<sup>39</sup> the architect-designer of the souks, in which the monument becomes a purely architectural object of design serving the style and the volumetric equilibrium in a signature architectural project. Al-Hatithy expresses very well this architectural oversimplification by stating that 'the zāwiya of Ibn 'Arraq restored only as a domed cube and articulated as a freestanding building in the open space, as if it were a fountain, playing the role of a landmark at the entrance of the souks'.<sup>40</sup>

From a religious space to an architectural cubic structure with a dome, the only feature to remind the visitor of its original function was a small copper plate attached to the wall with these few words: 'Zawiyat Ibin Arrak Al Dimashki Ash-Shafii. Prayers. Hijri 878–933, Gregorian 1473–1527'. Recently, a new printed panel was affixed to the front of the monument with a larger text explaining its nature. However, the text is flattened as much as possible in order to merge with the newly created urban architectural landscape while fitting within the framework of the fallacious SOLIDERE motto, 'the ancient city of the future'.<sup>41</sup>

Within the same semiotic mechanisms explained by Al-Harithy<sup>42</sup>, in which the monument becomes a sign that escapes contextuality, we propose to look at these mechanisms as taking the form of a structuralist war of signs and significations between two communities and an economic project inaugurated by SOLIDERE, in which the latter sought a way out through the secularisation of the signified space. This is done within a very strict 'value from fit' system<sup>43</sup> in which the only accepted decisions are those corresponding to the aims, plans and orientations of the new order created by SOLIDERE. This system belongs within a wider political programme which was intended by the consecutive Hariri governments for the future of the city and the country.

<sup>38</sup> Al-Harithy, 'Weaving historical narratives', 221.

<sup>39</sup>  Moneo, 'The Souks of Beirut', 263–273.

<sup>40</sup> Al-Harithy, 'Weaving Historical Narratives', 226.

<sup>41</sup> As stated by Nāser Chammā'a, SOLIDERE Chairman: 'At SOLIDERE, we have adopted a process of development and reconstruction that aims to foster a sense of urban wholeness by incorporating the past into our plans. This is reflected by our slogan. Beirut, The Ancient City of the Future, a slogan that governs our approach to city planning'. Quoted in Llewellyn, 'The big build-up'.

<sup>42</sup> Al-Harithy, 'Weaving historical narratives', 215.

<sup>43</sup> Camacho *et al*., 'Moral value transfer from regulatory fit', 498–499.

## **The al-Amin Mosque, Rafik Harari Memorial and the Saint George Maronite Cathedral: A behind-the-scenes power struggle and the construction of the city's history**

The Muhammad al-Amin Mosque is a new structure, funded entirely by late Prime Minister Rafik Hariri, built at the corner of Rue Emir Béchir and the Place des Martyrs, the main and most important public space in Beirut (Figure 19.2).<sup>44</sup> Adding to the provocative nature which the building possesses by virtue of its imposing volume dominating the public space at the centre of the city, this mosque, already planned in the 1940s, went through a myriad of political issues and transformations.

Frank Mermier<sup>45</sup> has analysed the circumstances which accompanied the project and the Mosque's construction.<sup>46</sup> The site of the Muhammad al-Amin Mosque was occupied by the Zawiyat Abi al-Nasr, destroyed after the start of the civil war in 1975. The *zawiya* was founded in 1853 by Sheikh Muhammad Abu al Nasr al-Yafi. The land was given to Sheikh Muhammad by the sultan in order to restore the sectarian balance between Muslims and Christians, having been preceded by a donation of land which he had just made to the sisters of the Lazarist congregation to thank them for their medical action. They built a convent there which later became the Lazariyyé building.<sup>47</sup>

The decision to transform the Zawiyat Abi al-Nasr into a mosque, namely the Muhammad al-Amin, was taken in 1933, when the Abu al-Nasr family donated the land to the new mosque. In the 1940s, the construction project for the Muhammad al-Amin Mosque began. In 1945 the mosque's committee was created and it was then registered as a waqf in 1948. In the same year, the Sunni Beirut court decided to transfer ownership of the land to the administration of the waqf under the guardianship of the Dar al-Fatwa. The aim was for the Muhammad al-Amin Mosque to replace the al-ʿUmari Mosque, the oldest mosque in Beirut, as the great mosque of the capital.<sup>48</sup>

From the early 1950s the committee of the mosque began to raise funds for the future building project. In 1965 the committee was transformed into an official association (the Association of the Muhammad al-Amin Mosque), accredited by the Lebanese Ministry of Interior. From then on, the association registered in its name the plots of land it managed to acquire. Further land was added to the plots by donations from Christian and Muslim families alike. The remaining plots were bought from the Jewish banker Safra.<sup>49</sup>

<sup>44</sup> This central place is marked by the public executions of the nationalists who were fighting for Lebanon's independence from the Ottoman rule in May 1916, hence its name Martyrs' Square. Since, it has witnessed all the main social and political manifestations that influenced the major changes in the country's political scenery.

<sup>45</sup> Frank Mermier is an anthropologist, specialist in Arab-Muslim societies, mainly in Yemen and Lebanon. His research themes focus on urban anthropology in the Arab Middle East, the study of the city and the memory of the Lebanese civil war.

<sup>46</sup> Mermier, 'La mosquée Muhammad al-Amîn'.

<sup>47</sup> Mermier, 'La mosquée Muhammad al-Amîn', sections 9–10.

<sup>48</sup> Mermier, 'La mosquée Muhammad al-Amîn', sections 13–14.

<sup>49</sup> Mermier, 'La mosquée Muhammad al-Amîn', section 15.

*Fig. 19.2: Al-Amin Mosque (Assaad Seif).*

From its foundation and until the outbreak of the 1975 civil war, the mosque's committee was subject to all kinds of rivalries present within the Sunni community of Beirut. The mosque project was, therefore, part of the political interest/concern of prime ministers and family leaders alike. It reflected the inner political struggles as well as alliances and coalitions of the Beirut Sunni families and political leaders on the national, regional and even international levels, while playing on the rivalries of the Arab powers of the time. It was affected by the Egyptian-Saudi rivalry from the birth of the Nasser regime, with each side cultivating allegiances within the Lebanese Sunni leadership. To this was added another conflict in the aftermath of the war, namely the Syrian-Saudi rivalry, through the takeover by the Ahbash of the association of the mosque and the site, where they installed a prayer place for their followers. Headed by some members of the Sunni families on the board of the association, the Ahbash is a Sufi religious movement involved in local politics supported, at the time, by the Syrian regime (through their intelligence services)<sup>50</sup>. The Ahbash confronted the representatives of official Sunni Islam, such as the Dar al-Fatwa, or the Sunni Islamist parties, such as that of the Muslim Brotherhood. They directed their criticism at Rafik Hariri, targeting his ties to Wahhabi leaders in Saudi Arabia. In order to put

<sup>50</sup> Founded in the mid-1980s, Al Aḥbāsh are also known as the Association of Islamic Charitable Projects (AICP).

an end to the Ahbash's control of the future mosque and its domineering position in the Beirut cityscape, the Mufti and Rafik Hariri used a decree to officially dissolve the Muhammad al-Amin mosque association at the end of August 2002. The property was then transferred to the Dar al-Fatwa. Rafik Hariri declared, shortly before the dissolution of the association, that 'it had become clear that a group had become involved in the project to build the mosque for political ends'.<sup>51</sup>

The latest episode in the power struggle over this project within the Sunni community was with the Saudi Prince al-Walid Bin Talal. The latter has both Saudi and Lebanese nationalities, his maternal grandfather being Riad al-Solh, ex-prime minister of Lebanon and one of the fathers of Lebanese independence. The Lebanese media made use of Bin Talal's political ambitions and spread rumours about his potential candidacy for the post of Prime Minister of Lebanon in 2002. In March 2002, Bin Talal gave two million US dollars to the Dar al-Fatwa for the purchase of extra land in order to widen the perimeter of the al-Amin Mosque. When Rafik Hariri found out, in a very skilful move, he brushed aside his competitor by undertaking to build the mosque at the expense of the Hariri Foundation. He donated twenty million US dollars to build a place of worship which would disturb the layout of the Place des Martyrs.52 On 3 October 2003, the foundation stone for the construction of this monument was laid on the site by the Mufti and Prime Minister Rafik Hariri.<sup>53</sup>

Through a process of legitimisation and enforcement of control over the city by means of financing this ostentatious monument of faith, Rafik Hariri transformed this architecture into a locus of a socio-political action. It was a skilful *tour de force* by means of which Hariri imposed not only his leadership and hegemony over his Sunni political rivals, but also stamped his name for ever on the nation's capital.

On 14 February 2005 Prime Minister Rafik Hariri was assassinated, before the mosque was completed. Immediately and without any hesitation, both the urban and religious authorities decided that Hariri's tomb would be placed next to the mosque that he had built for the glory of his community in Beirut. Burying Hariri next to the Muhammad al-Amin Mosque perpetuated the dynamic he had initiated in Beirut whereby his gradual accumulation of power was imbricated with both SOLIDERE, with its vision for the centre of Beirut, and the Dar al-Fatwa, the sole decisionmaking religious authority for the Sunni community. Later, the burial place would be redesigned and transformed into a memorial (Figure 19.3a).

Rafik Hariri's memorial represents the climax of the martyrdom process of the Sunni community, in the metaphorical sense, grounding their legitimacy in Beirut through the physical presence of Hariri's grave next to the al-Amin Mosque. In this sense, Hariri as the 'Martyr of the Nation' has overshadowed all the previous national martyrs who gave this location of the city its name of 'Martyrs' Square' (Figure 19.3b).

<sup>51</sup> Mermier, 'La mosquée Muhammad al-Amîn', sections 15–27.

<sup>52</sup> Mermier, 'La mosquée Muhammad al-Amîn', sections 33–34.

<sup>53</sup> Mufti Qabbāni had already officially announced the launching of the project on 29 October 2001 during a speech he gave in Dar al-Fatwa to an audience of Sunni figures and the Saudi Ambassador.

*Fig. 19.3a: The Hariri Memorial Grave near Muhammad al-Amin Mosque (I.S. Douaihi. Reprinted with permission from the photographer).*

*Fig. 19.3b: The Hariri Memorial, and its position in Martyrs' Square with the Martyrs' Statue in the centre (Dia Mrad, cropped with permission from the photographer).*

W h e t h e r a s a p l a c e of memory, support and m e d i a t i o n o f re l i g i o u s experience and faith54 or a physical, immovable property,<sup>55</sup> the presence of both the mosque and the memorial in the heart of the city with their heavily embedded symbolic charge became a territorial marker forever anchored in the cityscape – a new heritage in the making.

*Fig. 19.4: The St. George Maronite Cathedral in the 1890s (unknown photographer).*

## **The Maronite Cathedral of Saint George: The silent rival**

Unsurprisingly, the imposing volume of the al-Amin Mosque, with its ostentatious Ottoman character and style, its slender 'pencil' minarets and its large central dome,<sup>56</sup> recalling Sinan domed mosques,<sup>57</sup> impinged on the nearby Maronite Cathedral of Saint George. Hardly modest in its own design, the Maronite Cathedral drew its inspiration from another venerable prototype, the Basilica of Santa Maria Maggiore in Old Rome.<sup>58</sup> In 1894, when the cathedral was consecrated (Figure 19.4), it caused a stir, perhaps comparable to that provoked by the Muhammad al-Amin Mosque.<sup>59</sup> It took precedence, by its dimensions, over all the other places of worship in the city, Christian or Muslim. At that time, the building of the cathedral was felt as a provocation by the Greek Orthodox and Sunnis alike, who consider themselves as the city-dwellers of Beirut *par excellence*. <sup>60</sup> The cathedral was restored in the 1990s, respecting the volumetric envelope imposed by the urbanism law, but appeared squashed after the construction of the Al-Amin Mosque.

As a reaction, especially to the mosque's high minarets, the Maronite Archeparch of Beirut ordered the erection of a campanile on the south-western corner of the church with a clear architectural programme: while the campanile had to be compatible with the cathedral's architecture, its height had to surpass that of the al-Amin Mosque minarets.<sup>61</sup> The architect applied the programme as prescribed by the Archeparchy, with the total height of the campanile (Figure 19.5) surpassing the minaret by about 1 m.<sup>62</sup>


<sup>56</sup> Vloeberghs, *Architecture, Power and Religion in Lebanon*, 260–262.

<sup>57</sup> Kuban, 'The style of Sinan's domed structures'.

<sup>58</sup> Stanton, 'Terrain vague'.

<sup>59</sup> Stanton, 'Terrain vague'.

<sup>60</sup> Mermier, 'La mosquée Muhammad al-Amîn', section 12.

<sup>61</sup> Makarem, 'La croix de la cathédrale Saint-Georges des maronites bientôt visible à des kilomètres'.

<sup>62</sup> Hury, 'Le campanile de Saint-Georges des maronites en construction à Beyrouth'.

This unconscious Freudian phallic competition leaves us with nothing but stupefaction as to how far the latent power struggle between the different communities can go.

## **The Greek Orthodox Cathedral of Saint George: The heritage myth, between fiction and reality**

Located to the east of the Place de l'Étoile, at the heart of the city centre, the Greek Orthodox Cathedral of Saint George was heavily affected by the civil war. In 1994, the restoration project of the cathedral allowed the AUB Archaeological Museum team to excavate underneath the church (Figure 19.6). Two major articles explaining the discoveries and the museum project built over the conserved remains were published.<sup>63</sup> The head of the excavation mission, Leila Badre, in her article published in 2016, explained that the excavations had two major objectives: 'to (1) investigate the historical sequence and continuous use of the location as a religious site even before the construction of the cathedral; and (2) locate the Byzantine Anastasis Church, which—according to ancient texts—was located near the famous Law School of Justinian at Berytus. These two objectives were in fact clearly and tightly linked (the Anastasis Church being part of the continuous religious occupation of the site)'.<sup>64</sup>

The way the objectives are presented gives the impression that, on the one hand, the archaeological mission had solid

*Fig. 19.5: The new campanile built to the southwest corner of the St. George Maronite Cathedral (Assaad Seif).*

*Fig. 19.6: St. George Orthodox Cathedral main western entrance (James Gallagher, CC BY 2.0).*

<sup>63</sup> Badre, 'The Greek Orthodox Cathedral'; Skaf and Makaroun Bou Assaf, 'Une nouvelle approche pour la préservation in situ des mosaïques'.

<sup>64</sup> Badre, 'The Greek Orthodox Cathedral', 74.

information about the subterranean vestiges waiting to be unearthed and, on the other, the archaeologist/mission members were digging with the intention of finding specific structures with specific functions from specifically indicated periods. This far-fetched assumption goes beyond positivism. Or perhaps when someone writes an article *a posteriori*, they tend to mix temporalities, thus past and present become one entity, a present-past, where the hidden intentions are revealed in the form of confirmed truth.

According to the article, the archaeological excavations were carried out on a quarter of the total area of the church. Although it yielded a stratification dating from the Hellenistic period until the modern times, the text describes and analyses the Hellenistic and Roman remains briefly as compared to those of the Byzantine and the medieval periods. For instance, the Hellenistic period was represented by a small domestic structure with finds ranging from household material to a few oil lamps; the Roman strata revealed part of the Cardo Maximus.<sup>65</sup> As for the discoveries of the Byzantine and the medieval periods, the author gave extensive descriptions and made many allusions in order to bring the narrative as close as possible to the intended objectives declared in the article's preamble.

In this context, the text mentions the discovery of a Byzantine capital with crosses on its four sides (no photo is included), in addition to an oil lamp with a cross stamped on it (also without a photo) and a fragment of a mosaic floor orientated east–west with typical motifs of the fifth century AD. However, the interpretation was solely based on the mosaic floor in order to advance the idea that this site was a church, and very plausibly the Anastasis church, on the basis of the following arguments:<sup>66</sup> '(1) It dates to the fifth century AD, which is contemporary with the time of the law school. (2) Its motifs are typical of those used in the floors of churches during the same period. (3) Its traditional ecclesiastical east-west orientation is similar to that of the present cathedral.' Curiously, the capital and the oil lamp, mentioned to accentuate the religious function of the space were not taken as evidence for the church's existence. They were probably discovered in a dump or in an unreliable context which cannot be related to the claim that this was a religious structure.

If we take the above arguments one by one, we find that they are insufficiently substantiated.<sup>67</sup> Regarding the mosaic's orientation, we know that the Roman and subsequently the Byzantine city grid orientation is east–west/north–south. Thus, the claimed mosaic could very easily belong to a residential space orientated along the city's grid, rather than to a religious space, namely a Byzantine church (Figure 19.7). Furthermore, the orientation argument of the mosaic is not relevant, since no one can prove its religious nature without having the apse (among other features); a counter-argument which Badre herself uses later on the same page.<sup>68</sup> Moreover, none of the other churches discovered in Lebanon dating from the same period are orientated strictly east–west; they always have an orientation shift which can reach

<sup>65</sup> Badre, 'The Greek Orthodox Cathedral', 74.

<sup>66</sup> Badre, 'The Greek Orthodox Cathedral', 76.

<sup>67</sup> We will not comment on the proposed dating argument since it is not of relevance.

<sup>68</sup> Badre, 'The Greek Orthodox Cathedral', 76.

thirty to thirty-five degrees from the main east–west axis.<sup>69</sup> Thus the claimed east–west orientation cannot be taken as a proof.

As for the ivy vine heart-shaped leaf decorations, many other similar mosaics with similar motifs have been discovered in other – secular – buildings dating from the same period, disproving the notion of its exclusive use inside religious buildings. In the nearby Jemayzeh area, a mosaic with this type of decoration was discovered during the excavation of a bathhouse tepidarium dating from the same period (Figure 19.8).<sup>70</sup> Other examples have also been discovered in Beirut, namely at the civic basilica of the city excavated by Loffray.<sup>71</sup> Furthermore, other scholars and Byzantine mosaic experts agree that there is no substantial evidence which establishes the religious association of the Late Roman/Byzantine mosaics discovered underneath the Saint George Cathedral.<sup>72</sup>

From the medieval period, remains of a church of the late twelfth to early thirteenth century AD were discovered. Badre proposes that the church would have been the medieval Saint George's,

*Fig. 19.7: The claimed Anastasis mosaic inside the Crypt Museum (SOLIDERE, CC BY 2.0).*

*Fig. 19.8: The ivy vine heart-shaped leaves decorations of the tepidarium of the bath house (Assaad Seif).*

and suggests a continuity in the use of this church throughout the Mamluk period.<sup>73</sup> She then claims a link between this church and the one mentioned by the Chevalier d'Arvieux. Dating from the seventeenth century, the latter represents the earliest mention of the Greek Orthodox Church of Saint George inside the walls of the city.<sup>74</sup>

In fact, most of the references regarding the Mamluk period indicate a Church of Saint George outside the city, whose location is known today at the site of the al-Khidr Mosque at the eastern entrance to the city.<sup>75</sup> The only source suggesting such

<sup>69</sup> See plans in Hélou, *Les mosaïques protobyzantines du Liban* and Chehab*, Les mosaïques du Liban*.

<sup>70</sup> This mosaic has been re-integrated into the building erected on the location of the discovery and it is open to the public for visits, Seif, 'Beyrouth, vingt ans d'archéologie urbaine'.

<sup>71</sup> Loffray, 'Forum et monuments de Béryte', 46–47, figs. 3–4.

<sup>72</sup> Hélou, *Les mosaïques protobyzantines du Liban*, 72.

<sup>73</sup> Badre, 'The Greek Orthodox Cathedral', 79.

<sup>74</sup> d'Arvieux, *Mémoires*, 146; Badre, 'The Greek Orthodox Cathedral', 79.

<sup>75</sup> Al Khidr and Saint George are commonly conflated among Christians and Muslims in the Levant, Miller II, *Baal, St. George, and Khidr*. Moukarzel, *La ville de Beyrouth*, 90.

a church inside the city is Pierre de Pennis (*c*. 1350), who mentions the existence of two churches inside the city, the Church of the Saviour and the Church of the Blessed George.<sup>76</sup> Nevertheless, there is no indication in his text as to whether the Church of the Blessed George is that of the Melkite (*i.e.* Greek Orthodox) or another community. Moreover, there is no indication as to the location of this church in the text of de Pennis. Denys Pringle pushes the interpretation even further by proposing that this church could also have been the precursor of the Maronite Saint George's Cathedral which the Maronites held from the Latin Bishop of Beirut, as mentioned in a papal confirmation of 1184.<sup>77</sup> Knowing that Pringle's argument could be contested, and given that the fifteenth-century Druze chronicler Salih Ibn Yahya mentions that 'the Cypriots had churches in Beirut and a group of merchants who lived there and had shops and bars',<sup>78</sup> we cannot preclude the possibility that the church mentioned by Pierre de Pennis could have been one of the Cypriot churches in Beirut.<sup>79</sup>

Consequently, the argument claiming that the excavated medieval remains are those of a Melkite Saint George's lacks sufficient proof since no one can prove yet that the medieval church discovered underneath the current Greek Orthodox Saint George's Cathedral is related to the Orthodox community, as interpreted by Badre.<sup>80</sup> After describing the medieval church, Badre goes on to give an account of the excavated evidence of the eighteenth-century remains of the Church of Saint George, at which time the church was part of a bigger complex and a convent.<sup>81</sup>

All these discoveries are now part of 'the Crypt Museum' and, it is claimed, reveal the use of the site from the Hellenistic period until modern Beirut.<sup>82</sup> Although there

<sup>76</sup> de Pennis, 'Libellus de Pierre de Pennis de Locis Ultramarinis', 380 ; Moukarzel, *La ville de Beyrouth*, 743.

<sup>77</sup> Hiestand, *Papsturkunden*, 304, no. 127 after Pringle, *Churches of the Crusader Kingdom of Jerusalem*, 116, 119. Pringle's interpretation could be contested, since according to the pilgrim Joos van Ghistele, who visited Beirut in 1481, the Maronites held their services in the crypt of the Franciscan Church of the Saviour. 'This proves that the Maronites did not have their own church in Beirut, so they shared the Franciscan church.', Moukarzel, 'Beirut's Church of the Savior under the Mamluks', 113.

<sup>78</sup> Ṣāliḥ Ibn Yaḥya, *Kitāb tārikh Bayrūt wa akhbār al-umarā' al-buḥturiyyīn min bani al Gharb [li-Ṣāliḥ Bin Yaḥya]*, 35.

<sup>79</sup> The Cypriots lived in Beirut between 1350 and 1365. In 1365, Sultan Al-Ashraf Shaaban expelled them from the Mamluk Sultanate, along with other Franks, because of the attack and pillaging by the King of Cyprus Pierre I on Alexandria for three days. For this reason, the existence of any church (other than the Church of the Saviour which was in the possession of the Franciscans) was not mentioned in the writings of European pilgrims who visited Beirut after 1370, the year in which the Europeans concluded a peace treaty with the Mamluk Sultan (see on this topic, Moukarzel, 'The relations between Beirut and Cyprus during the fourteenth and fifteenth centuries', 57–71).

<sup>80</sup> Indeed, the presence of the Melkite community in Beirut at that period cannot be proven. If one existed, it did not have an imposing religious presence. Francesco Suriano, the head of the Franciscan Monastery in Beirut between 1480 and 1483, does not mention the presence of the Melkite community in the city or touch upon any disagreement with them, at a time when conflicts between the Franciscans and the Melkites in Jerusalem were at their climax (Suriano Treaties on the Holy Land, after Moukarzel, *La ville de Beyrouth*, 89–97, 751).

<sup>81</sup> Badre, 'The Greek Orthodox Cathedral', 79; Davie, 'Le couvent Saint-Georges de Bayroût al-Qadîmat'.

<sup>82</sup> Skaf and Makaroun Bou Assaf, 'Une nouvelle approche pour la préservation in situ des mosaïques'.

are a few remains and objects from the different Islamic periods exhibited, in general the layers which obstructed the Byzantine and medieval remains were removed to clear the way and reveal the 'more important' continuous Christian use of the space. These 'special' remains are lit individually with the chronological sequence clearly indicated so that the visitor witnesses a smooth continuity of the use of space by the Orthodox community of Beirut. This programme is not explicitly outlined, but rather implied through the careful arrangement of the discoveries along the visitors' route. In case the visitor did not understand the message intended by the whole *mise-en-scène*, Badre clearly states that there is 'a visual link between the past inside the crypt and the present in the upper cathedral, [through] a glass square-shaped floor [which] was installed to unify periods'. To clarify matters further, a film is projected at the end of the visit 'inviting visitors to further explore the historical evolution of the earliest cathedral in Beirut from the Byzantine period to the present day'.<sup>83</sup>

In her conclusion, Badre claims that the site 'is *believed* to have housed the Anastasis Church until an earthquake in AD 551 destroyed most of the city'. Later, she confirms that 'the *reoccupation* [of the site] for religious purposes' took place during the Crusader era after a short period of abandonment. She then directly links it with the earlier phases of the site which would eventually house the actual Saint George's Cathedral. Finally, she expresses her ideologically biased position when she says that the archaeological work she describes helped in 'the reunification of the community after the civil war, and a reawakening of Lebanese culture'. She goes on to praise the Crypt Museum, 'the crowning jewel of the restoration of this church', where the different vestiges were staged as a 'rich physical representation of this ancient land', amplifying her claims to incorporate Beirut and even Lebanon as a whole; she may even insinuate the nation state itself.<sup>84</sup>

## **The search for the Berytus Law School and the location of the Anastasis Church**

The search for the famous Beirut Law School has been a never-ending marathon between scholars and archaeologists working in and on Beirut. The early scholars who worked on the problem based their interpretations on fragmentary information;<sup>85</sup> more recent archaeologists who have subsequently worked in Beirut have taken for granted most of what Collinet and Loffray proposed by copying them in their new research and publications.

The elements used in the attempts to locate the Law School are: (1) the texts of a funerary inscription discovered near the Saint George Cathedral of the Maronites86 and

<sup>83</sup> Quotations from Badre, 'The Greek Orthodox Cathedral', 92.

<sup>84</sup> Quotations from Badre, 'The Greek Orthodox Cathedral', 97.

<sup>85</sup> Collinet, 'Beyrouth, centre d'affichage et de dépôt des constitutions impériales'; Collinet, *Histoire de l'école de droit de Beyrouth*; Jalabert, 'Inscriptions grecques et latines de Syrie', Loffray, 'Forum et monuments de Béryte', Loffray, 'Forums et monuments de Béryte (suite) II. Le niveau médiéval'.

<sup>86</sup> Jalabert, 'Inscriptions grecques et latines de Syrie', 72.

(2) the seminal work on the Law School of Beirut published by Collinet.<sup>87</sup> Collinet's book recounts the story of Zacharias Scholasticus, a student of the Law School under the famous Severus of Antioch, a monk and theologian who was Patriarch of Antioch from 512 until his death in 538.<sup>88</sup> In Zacharias story it was mentioned that praying in the nearby Anastasis church was a daily ritual for the Law School students. Consequently, it was argued that the discovery of one of these two monuments would be crucial to locating the other. Moreover, in his book Collinet uses the funerary inscription discovery location and Zacharias story in order to propose a potential location for the Law School and the Anastasis alike.<sup>89</sup> Since Badre is interested in proving the perpetual religious use of the Saint George Orthodox Cathedral location, the Anastasis became her prime target. Consequently, she used the above-mentioned arguments in order to give ground for her proposals.<sup>90</sup>

The fragmentary funerary inscription – the first evidence – which was first published in 1906 by Jalabert, 'had already disappeared by 1925'.<sup>91</sup> In this fragmentary and eroded inscription, the dubious reading prompted Collinet to propose the name of a famous law professor: Patricius.<sup>92</sup> On the basis of the assumptions that this funerary inscription was both found *in situ* (which no one can confirm) and would originally have been located in the cemeteries next to the Byzantine Church of the Anastasis, Collinet proposed to locate the Law School in the area of the discovery, that is, in the vicinity of Martyrs' Street, today Emir Bechir Street, more precisely, to the south of the great Roman Beirut–Tripoli road, which extends to Gouraud Street. He also accepted the proposition that this Law School could have been built against the outer side of the city walls in that area.<sup>93</sup>

Later, Loffray gave another location for the Anastasis and the Law School by suggesting that the conclusions of Collinet, based on the text of Jalabert, were 'distorted by a material error', namely, the confusion between the Saint George of the Maronites and Saint George of the Orthodox. He also claimed that R. F. Mouterde indicated to him the precise location where the funerary inscription was discovered (which no one can confirm either). Consequently, he proposed the location of the Anastasis on the premises of the Greek Orthodox Saint George's Cathedral.<sup>94</sup>

As a result, all the proposed locations are based purely on interpretations and speculations with no substantial evidence in favour of one or the other. Nevertheless, Leila Badre adopted the one that corresponds better to her model, and alluded to the presence of the Anastasis beneath the Greek Orthodox Cathedral without any concrete

<sup>87</sup> Collinet, *Histoire de l'école de droit de Beyrouth*.

<sup>88</sup> Collinet, *Histoire de l'école de droit de Beyrouth*, 48–49.

<sup>89</sup> Collinet, *Histoire de l'école de droit de Beyrouth*, 61–75.

<sup>90</sup> Badre insists that the Anastasis 'was likely located where the cathedral (Saint George) stands today', Badre, 'The Greek Orthodox Cathedral', 76.

<sup>91</sup> Hall, *Roman Berytus*, 207.

<sup>92</sup> Collinet, *Histoire de l'école de droit de Beyrouth*, 73, 136–137 ; Hall, *Roman Berytus*, 208.

<sup>93</sup> Collinet, *Histoire de l'école de droit de Beyrouth*, 71–73.

<sup>94</sup> Loffray, 'Forum et monuments de Béryte', 33–35.

proof. In the light of these claims and through the repetition of this narrative – without its being questioned by archaeologists and scholars – the public today is convinced of this story. More dangerously, the bishopric makes every effort to sustain these beliefs through brochures, websites and other media.<sup>95</sup> In fact, the exact location of the Anastasis and the Law School still cannot be confirmed, with the result that the claims proposed by Badre's influential 2016 article can only be wishful thinking.

## **The cherry on top**

In addition to all the scenography inside the museum, the bishopric has erected an inscribed Roman column, discovered in the nearby excavations (Figures 19.9–19.10), on the left side of the western entrance to the cathedral. When I asked the priest responsible for the church about the column,<sup>96</sup> he claimed that it bears the name of Quartus, the first Orthodox bishop of Beirut during the first century AD.<sup>97</sup> Inspection of the inscription shows that it does bear the name Quartus, but an accurate translation suggests a totally different interpretation.

The Latin text reads:

PATRIAE DULCISSIMAE P. MARIUS QUARTUS EQUO PUBLICO ORNATUS SUA PECUNIA FECIT

The English translation:<sup>98</sup>

To the sweetest homeland P[ublius] Marius Quartus Adorned with the equestrian order Has made with his funds

In reality, it was a column established in a public place as a dedication by a certain Publius Marius Quartus in order to honour his 'sweetest homeland'. Nevertheless, driven by the obsession with proving their ancient ancestry and with 'rooting' their existence in the city since early Christianity, the bishopric and the people (including archaeologists) who pivot in their orbit, whether consciously or unconsciously, have succumbed to an interpretative, frenetic delusion, transforming the archaeological evidence into a fantasised reality which satisfied their aspirations and aims. This confirms, once again, that the different communities are only able to conceive their past, and consequently, its materiality, through a pure, linear continuum between ancestors and descendants.<sup>99</sup> In this sense, the column is presented as an

<sup>95</sup> Some of the sites where the information is circulating: en.wikipedia.org/wiki/Saint\_George\_Greek\_ Orthodox\_Cathedral; www.facebook.com/SaintGeorgeCathedralBeirut/; lebanonuntravelled.com/ saint-georges-cathedral-beirut/; www.beirut.com/l/6502.

<sup>96</sup> Personal discussions during a visit to the cathedral.

<sup>97</sup> Smith, *Smith's Bible Dictionary*.

<sup>98</sup> We would like to thank Alexander Hourani for assisting in the translation of the Latin text.

<sup>99</sup> Seif, 'Conceiving the past', 286.

*Fig. 19.9: The dedication column to the left side of the entrance of the Greek Orthodox St. George Cathedral (Assaad Seif).*

*Fig. 19.10: Close view of the dedication column showing the Latin inscription (Assaad Seif).*

additional artifice to garnish the story of this ancestral anchorage of the presentday Orthodox community in the land of Berytus.

## **Conclusion**

During the post-war reconstruction of Beirut, the different communities began encroaching on both the city's spatial and political realms in order to establish their authority and reposition themselves within the national political arena. This was done in different ways: through attempts to appropriate a monument, as in the example of Zawiyat Ibn 'Arraq al-Dimashqi; the adoption of a suitable heritage narrative, as in the case of the Saint George's Orthodox Church; or sometimes even through the production of a new heritage, as in the case of the al-Amin Mosque. These attempts at appropriation were based on the biased interpretation of archaeology and/or territorial marking in the city. We may add to this the intensive physical interventions of the development company SOLIDERE, for whom the masterplan was and remains the bible by which everything should

abide, including physical remains of an archaeological nature. In the case of SOLIDERE, we witness another type of narrative in which the latter cleansed the urban environment of its religious and ideological dimension in order to transfer monuments or sites from the religious/ideological to the secular realm.

We may approach Beirut as a metaphorical palimpsest, in order to construe meaning from the different case studies. The first case study demonstrated how SOLIDERE undertook a systematic erasure of a major part of the old text inscribed on vellum in order to create a new one that was free of the unwanted residues of war;<sup>100</sup> this is a new palimpsest in the making where the economic drive was taking over the city. Nevertheless, the Zawiyat Ibn 'Arraq al-Dimashqi represented the un-erased spot on the palimpsest which remained standing, showing through from another temporal dimension. A zone of the vellum which did not correspond to the aims of the new writer (SOLIDERE) and what it represents socio-politically. This is why in order to

<sup>100</sup>Seif, 'Conceiving the past', 284.

incorporate it into the new text written by SOLIDERE, the narrative shift established by the company operated as a metamorphosis, distorting the original appearance of this *zawiya*. This shift was taken even further by stripping down the monument and reducing it to a simple geometric form that fits within the architectural idiom of the souks area in its new incarnation.

In the case of the al-Amin Mosque, the Hariri Memorial and the Maronite Cathedral of Saint George, we can perceive the power struggle and dynamics of conflict in which the past, the present and the future are intertwined in an agglomerated ensemble. Together they embody a palimpsest that is continuously in the making through the ongoing creation of unique and highly contextualised dynamics101 that are fuelled by their cultural and spatio-temporal character. Furthermore, through its complex association with the al-Amin Mosque, the Hariri Memorial has transcended the physical space through its transformation into an untouchable sacred relic – a contemporary taboo. It is like stamping the vellum of the palimpsest with an indelible ink that cannot be erased; on the contrary, it is bound to stay and maintain its eternal presence even when new texts are written with or next to it in the near or distant future.

In the case of the Greek Orthodox Cathedral of Saint George, we witness a case of a cumulative palimpsest, where 'the successive episodes of deposition, or layers of activity, remain[ed] superimposed one upon the other without loss of evidence, but are so reworked and mixed that it is difficult or impossible to separate them out into their original constituents'.<sup>102</sup> In this case, we were concerned with another paradigmatic approach, that is, of interpretivism. We have tried to seek sense beyond the different definitions of the palimpsest and uses of the metaphor in order to delve deeper into the palimpsest's interpretational process and its diverse features, where a certain past is imposed by the archaeologist. In this case, we have witnessed the invention of present pasts and their ideological framework.

### *In fine*

Heritage is never rediscovered; it is only the objects/monuments or the material remains of past socio-cultural dynamics that are rediscovered. The interpretation of these objects creates what is commonly known as Heritage. This creation is carried out through a story-making/storytelling process in which all these ingredients are meticulously positioned according to their differential value and related significance, as Heritage is no more than a value-based construct.

The staging of physical and tangible elements of the past, that is, archaeological objects and finds from previous historical periods, creates a new memory realm that can be transformed into heritage through the acceptance of its ideological dimension/ meaning/interpretation by the society or community which adopts it as belonging to its past.

<sup>101</sup>Dillon, 'Reinscribing De Quincey's palimpsest', 249.

<sup>102</sup>Bailey, 'Time perspectives', 204.

Consequently, different components/communities within a city can create their different memories and different pasts according to different interpretations of the past and its materiality. Thus, the city becomes a patchwork of memories cast in stones, a multifaceted heritage reflecting not only the wealth of the city but also the wealth of its created pasts. These pasts most of the time conflict with each other as a reflection of the present conflicts – a Present Past built on biased interpretations of past materiality and the re-inventing of a new Heritage.

### **Bibliography**

Abou, S., 'La composition ethnique de Beyrouth', *Beirut – Crossroads of Cultures* (Beirut, 1970): 92–112. Arias, R., 'Haunted places, haunted spaces: The spectral return of Victorian London in neo-Victorian


Bailey, G., 'Time Perspectives, palimpsests and the archaeology of time', *Journal of Anthropological Archaeology* 26 (2007): 198–223.


Moukarzel, P., 'Beirut's Church of the Savior under the Mamluks', *Mamlūk Studies Review* 22 (2019): 97–119.

Moukarzel, P., 'The relations between Beirut and Cyprus during the fourteenth and fifteenth centuries' in M. Parakos and J. Chick (eds.), *Othello's Island 1: Selected Proceedings from Othello's Island: The Annual Conference on Medieval and Early Modern Studies* (London, 2019), pp. 57–71.

Nahas, C., 'Le confessionnalisme au Liban, du fonctionnement discursif et idéologique vers une position du problème', PhD thesis, École des Hautes Études en Sciences Sociales (1980).

*Oxford English Dictionary*, J. A. Simpson and E. S. C. Weiner (eds.), vol. 11 (Oxford, 1989).

Prera, A. L., 'A cumulative palimpsest effect: the multilayered meanings of late-to-terminal classic era, above-floor deposits at structure M13–1' in O. Navarro-Farr and M. Rich (eds.), *Archaeology at El Perú-Waka': Ancient Maya Performances of Ritual*, *Memory, and Power* (Tucson, 2014), pp. 34–52.

Pringle, D., *The Churches of the Crusader Kingdom of Jerusalem: A Corpus* (Cambridge, 1993).

'Saint George Greek Orthodox Cathedral' [available at: en.wikipedia.org/wiki/Saint\_George\_Greek\_ Orthodox\_Cathedral] (accessed 23 September 2021).

'Saint George Orthodox Cathedral' [available at: www.beirut.com/l/6502] (accessed 23 September 2021).

'Saint George Orthodox Cathedral of Beirut' [available at: www.facebook.com/ SaintGeorgeCathedralBeirut/] (accessed 23 September 2021).


Seif, A., 'Conceiving the past: fluctuations in a multi-value system', *Conservation and Management of Archaeological Sites 11* (2009): 282–295.

Seif, A., 'Beyrouth, vingt ans d'archéologie urbaine', *Dossiers d'Archéologie* 392 (2019): 20–23.

Skaf, I. and Y. Makaroun Bou Assaf, 'Une nouvelle approche pour la préservation in situ des mosaïques et vestiges archéologiques au Liban: La crypte de l'église Saint-Georges à Beyrouth' in A. Ben Abed, M. Demas and T. Roby (eds.), *Lessons Learned: Reflecting on the Theory and Practice of Mosaic Conservation,* Proceedings of the 9th ICCM Conference Hammamet, Tunisia, November 29-December 3, 2005 (Los Angeles, 2008), pp. 224–229.

Smith, W., *Smith's Bible Dictionary* (Peabody, 1990).

SOLIDERE, *Annual report* (Beirut, 1999), p. 32.

Stanton, M., 'Terrain vague' in T. Chakar and C. Ernsten (eds.), *Beyroutes. A Guide to Beiru*t (Beirut, 2010), pp. 25–27.

Sullivan III, A. P., 'Time perspectivism and the interpretive potential of palimpsests: theoretical and methodological considerations of assemblage formation history and contemporaneity' in S. Holdaway and L. Wandsnider (eds.), *Time in Archaeology: Time Perspectivism Revisited* (Salt Lake City, 2008), pp. 31–45.

Vloeberghs, W., *Architecture, Power and Religion in Lebanon* (Leiden, 2015).

Way, A. M., 'A behavioral approach to cumulative palimpsests: An example from Weereewaa (Lake George), Australia', *Journal of Anthropological Archaeology* 52 (2018): 59–70.