Studies in the Archaeology of the Medieval Mediterranean

Studies in the Archaeology of the Medieval Mediterranean The Medieval Mediterranean Peoples, Economies and Cultures, 400–1500 Managing Editor Hugh Kennedy SOAS, London Editors Paul Magdalino, St. Andrews David Abulafia, Cambridge Benjamin Arbel, Tel Aviv Larry J. Simon, Western Michigan University Olivia Remie Constable, Notre Dame Frances Andrews, St. Andrews vOLUME 86 Studies in the Archaeology of the Medieval Mediterranean Edited by James G. Schryver Leiden • BOSTON 2010 Cover illustration: Polychrome lead glaze bacino from the Church of San Gavino of Porto Torres. Produced in North Africa (Tunisia) in the second half of the eleventh century. © Photograph by Michelle Hobart This book is printed on acid-free paper. Library of Congress Cataloging-in-Publication Data Studies in the archaeology of the medieval Mediterranean / edited by James G. Schryver. p. cm. — (The medieval Mediterranean : peoples, economies and cultures, 400–1500, ISSN 0928-5520 ; v. 86) Includes bibliographical references and index. ISBN 978-90-04-18175-5 (hardback : alk. paper) 1. Mediterranean Region— Civilization 2. Mediterranean Region—Antiquities. 3. Civilization, Medieval. 4. Mediterranean Region—History—476-1517. 5. Excavations (Archaeology)— Mediterranean Region. 6. Archaeology and history. I. Schryver, James G. II. Title. III. Series. DE71.S78 2010 909’.09822—dc22 2010024389 ISSN 0928-5520 ISBN 978 90 04 18175 5 (harback) ISBN 978 90 04 18724 5 (e-book) Copyright 2010 by Koninklijke Brill Nv, Leiden, The Netherlands. Koninklijke Brill Nv incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and vSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill Nv provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. CONTENTS Editor’s Preface .................................................................................... List of Contributors ............................................................................. List of Figures and Maps .................................................................... vii xi xv Introduction .......................................................................................... 1 Chapter One A Port Without a Harbour: Reconstructing Medieval al-Mina ............................................................................ Tasha Vorderstrasse 15 Chapter Two The Phantom Baronies of the Western Mani ..... Jon van Leuven 41 Chapter Three A Seigneury on the Eastern Borders of the Kingdom of Jerusalem: The Terre de Suète ................................ Cédric Devais 71 Chapter Four Merchants, Monks, and Medieval Sardinian Architecture ...................................................................................... Michelle Hobart 93 Chapter Five The forma urbis of Aleppo (Syria) during the Middle Ages ..................................................................................... 115 Giulia Annalinda Neglia Chapter Six Ethno-Archaeological Approaches to Medieval Rural Settlement in Spain and Morocco ..................................... 155 Johnny De Meulemeester Chapter Seven The Future of venice’s Past and the Archaeology of the North-Eastern Adriatic Emporia during the Early Middle Ages .................................................................... 175 Sauro Gelichi List of Works Cited ............................................................................. 211 Index ...................................................................................................... 235 EDITOR’S PREFACE When Julian Deahl first approached me with the idea for this book, the intent was to provide readers outside of archaeology with a sense of the rich variety of work that medieval archaeologists are conducting in different areas around the Mediterranean basin. That original intent still holds true, especially in terms of the geographical variety and approaches of the seven case studies included in this volume. However, as I will explain in more detail in my introductory chapter, the work as a whole also addresses some key issues within the larger field of Medieval Archaeology in important ways. This volume is meant to be read as a series of separate, individual case studies that overlap and connect at a general level with one another in terms of a shared, interdisciplinary approach to aspects of the medieval Mediterranean world that is either heavily grounded in or incorporates archaeology and archaeological evidence. These studies are also united in their attempt to move beyond discussions of the ‘tyranny of the text’ and debates about which scholars, types of evidence, or academic disciplines are most relevant or can tell us the most about the past. Instead, they offer a variety of approaches attesting to the benefits of an integrative, interdisciplinary methodology to the past that critically examines relevant data from numerous sources, whether these are archaeological, architectural, cartographic, textual, or otherwise. Where the chapters of this volume differ is in the geographical location of the studies, spread around the Mediterranean basin, and in the components they comprise. Allowing the variety on this interdisciplinary theme to shine through is intentional, and will hopefully more readily allow the reader to understand the richness of ‘archaeological’ studies being conducted on the medieval Mediterranean world. Naturally, there are a number of important challenges that an editor faces in editing any multi-authored volume. Perhaps foremost is that of trying to allow each of the authors to express their ideas in their own way, while also making them speak with the same voice. With this in mind, the authors and I have massaged the various chapters over their various drafts both in terms of larger issues, such as overall argument, and in terms of much smaller ones, such as the consistent usage of the series comma. Some readers may undoubtedly think we did not viii editor’s preface go far enough to either side of this balance, but I hope that they will at least understand the kind of balance we sought to achieve. As editor, the other major area of intervention on my part has been in terms of bibliography. The “List of Works Cited” was compiled with an eye to making it as easy as possible for those who may not have immediate access to these sources, but may still wish to consult them, to track these sources down. As a result, I have sought to keep the citations in the footnotes to their more basic form while at the same time providing more of the information one might need to locate the sources in the works cited section at the end of the volume. Thus, information such as the expanded titles of conference proceedings and the series information for many sources can be found in the latter. In formatting the “List of Works Cited,” I have used WorldCAT as the guide for capitalization (especially of foreign titles), punctuation, and spelling, again with an eye to providing the reader with all of the relevant information needed to locate specific sources. This is especially the case with books that are only held by a handful of libraries, or are spelled phonetically in WorldCAT differently than the author spelled them in their chapter. In the cases where these differences prove significant for tracking the sources down, I have included the spelling necessary to locate them in WorldCAT in [ ]. Thus, readers looking for a quick reference should find the footnotes more than sufficient. However, those readers seeking to interlibrary loan a source will want to consult the “List of Works Cited.” In addition, with the same goal of clarity and ease of location, I have repeated the names of authors in certain subsequent references in this list, whereas I have used idem and ibidem in the footnotes. On a very sad note, I would like to note the passing of Johnny De Meulemeester at the beginning of 2009, as this book was being reviewed. At the same time, I am extremely grateful to his daughter Ann, for allowing me to continue to include his chapter in this book. Publishing his chapter in a book dedicated to bringing the variety and vibrancy of archaeological work being carried out around the Mediterranean basin to an English-speaking audience arguably much more familiar with work in continental Europe and the North Atlantic is a fitting way to honor the desires of one of the giants in Medieval Archaeology, and one of the few who seemed to be able to bridge the gap between the Mediterranean and more northern climes. editor’s preface ix On a final, happy note, I would like to thank the authors for their hard work and patience throughout the editing process. We all benefited greatly from the thorough commentary of the anonymous external reviewer and express our most heartfelt thanks to them. In addition, I would like to thank the two student assistants, Jozette Allen and Bailey Jelle (now Kinsky), who helped with various stages in the editing process. Lastly, as many other authors and editors in this series have done, I would like to thank Marcella Mulder of Brill for her guidance throughout the process. LIST OF CONTRIBUTORS Jimmy Schryver (University of Minnesota, Morris) teaches Ancient and Medieval Art and Archaeology and specializes in the art, archaeology, and society of Frankish Cyprus. In this area, he is currently revising for publication his PhD thesis, Spheres of Contact and Instances of Interaction in the Art and Archaeology of Frankish Cyprus, 1191–1359. His most recent publications include “Colonialism or Conviviencia in Frankish Cyprus?” In Boundaries in Depth and in Motion. Edited by I. William Zartman, 133–159 (University of Georgia Press, 2010). He has participated in numerous excavations around the Mediterranean and is currently the assistant director of the Petra Garden and Pool Complex Excavations in Petra, Jordan. Tasha vorderstrasse (Netherlands Institute for the Near East, Leiden) is a specialist in the archaeology, history, and art of the Late Antique and medieval Middle East. Her recent publications include “Archaeology of Medieval Lebanon: An Overview,” Chronos 20 (2009): 103–128 and (with B. P. Muhs) “Collecting Egyptian Antiquities in the Year 1838: Reverend William Hodge Mill and Robert Curzon, Baron Zouche,” The Journal of Egyptian Archaeology 94 (2008): 223–245. She has participated in excavations and surveys throughout the Middle East, most recently at Ziyaret Tepe in Turkey and Dvin in Armenia. She also works on archaeological material from museum collections including the site of al-Mina, which she published as Al-Mina: A Port of Antioch from Late Antiquity to End of the Ottomans. PIHANS CIv. Leiden: NINO (2005). Jon van Leuven (University of Gothenburg) is a researcher and translator in Greek archaeology from ancient to medieval times with special focus on the Peloponnese and Aegean islands, as well as a former editor of the Journal of Prehistoric Religion. Apart from many publications on early cultures, he has primarily studied Mediterranean maps and their relation to texts and excavations. He now conducts field surveys in Messenia and Laconia and is preparing a commentary on relevant Byzantine and later authors, such as Sphrantzes. xii list of contributors Cédric Devais, member of the Institut Français du Proche-Orient (IFPO) between 2001 and 2005, is a specialist of the history of the Crusades and of the archaeology of the Kingdom of Jerusalem. He has published several works on this subject, such as “Montréal, forteresse d’Outre Jourdain et ville croisée,” Histoire Médiéval (2005) and “Expression du pouvoir aux frontières du royaume de Jérusalem: Terre de Suète et Oultre-Jourdain au XIIème siècle,” BEO (2006–2007). Recently, he directed a survey of the hinterland of Montreal castle (Shawbak, Jordan). He is currently responsible for relations with North Africa/ Middle East at the French Ministry of Culture and Communication and leads the archaeological survey of the territory of Shawbak and medieval Petra. Michelle Hobart (The Cooper Union, New York City) is a field archaeologist and scholar of central Italian and Mediterranean medieval settlements and material culture. Her current archaeological projects include the hilltop town of Capalbiaccio, where she has been director since 2008; and this summer she will be co-directing a new excavation of the small monastery of San Pietro d’Asso in the province of Siena. Dr. Hobart’s recent publications include “Capalbiaccio (GR) nel tempo: dalla preistoria all’età moderna. Le indagini archeologiche dagli anni ’70 al nuovo progetto di ricerca,” in Archeologia Medievale XXXvI; and “The Peruzzi and their Urban Enclaves: Preserving medieval fortifications in a changing Communal Florence,” in Archeologia Medievale XXX. Dr. Hobart is currently editing a volume provisionally entitled, Companion to Medieval Sardinia (Brill, 2013). Giulia Annalinda Neglia (Bari Polytechnic University) teaches Typological and Morphological Characters of Architecture at the Faculty of Architecture. In July 2003, she earned her PhD in Architectural Design for the Mediterranean Countries with a thesis on Aleppo (advisor Prof. Attilio Petruccioli). For her research on Mediterranean—Islamic cities, she has received several scholarships from international research centres, such as the DAAD (Deutscher Akademischer Austauschdienst), the Max van Berchem Foundation, and the Aga Khan Program for Islamic Architecture at MIT. Her recent publications include Medina. Essays on the Urban Landscapes of South-Eastern Mediterranean Cities (PolibaPress, Bari, 2009) and Aleppo. Processes of Formation of the Medieval Islamic City (PolibaPress, Bari, 2009). list of contributors xiii Professor Dr Johnny De Meulemeester (University of Ghent, Fellow of the Society of Antiquaries), taught the medieval archaeology of Europe and the Mediterranean World. Sadly, he passed away on January 17, 2009, in Ghent. Among other things, he will be remembered for the generosity with which he gave his advice, time, and support to students and colleagues alike. He was one of the original ‘Fathers’ of Ruralia in its earliest days, then went on to become its first Secretary, and finally, the organization’s President. Johnny is most known for his work on medieval castles. He was active in excavations in Belgium, Ireland, Jordan, Spain, and elsewhere. Sauro Gelichi (University of Ca’ Foscari, venice) teaches Medieval Archaeology and Archaeological Theory and Methods. He is currently director of the journal Archeologia Medievale. His most recent publications include Constructing Post-medieval archaeology in Italy. A new Agenda. Proceedings of the International Conference (Venice, 24th and 25th November, 2006) (Florence, 2007); A Town through the Ages: the 2006–2007, Archaeological Project in Stari Bar (Florence, 2008); and L’isola del vescovo. The Archaeological Excavations nearby the Comacchio Cathedral (Florence, 2009). He has excavated in many countries in the Mediterranean Region (Syria, Turkey, Tunisia) and is currently the director of the Stari Bar Project in Montenegro and of many other Italian excavations (venetian Lagoon, Comacchio, Nonantola). LIST OF FIGURES AND MAPS Vorderstrasse Figure 1: The archaeological site of al-Mina, located in the modern province of the Hatay in southeastern Turkey .......... 27 Van Leuven Map 1: The Mani peninsula and its vicinity, c. 1250, including main Frankish strongholds and uncertainties discussed in the text, with other places mostly of older or later date ......... Hobart Figure 1: Polychrome lead glaze bacino from the Church of San Gavino of Porto Torres. Produced in North Africa (Tunisia) in the second half of the eleventh century. Figure 2: Map of Sardinia, with the four Giudicati and names of all churches that have or had bacini. Figure 3: Façade of the Church of Santa Giusta at Oristano. The church has been heavily restored and rebuilt in parts. Figure 4: Façade of the Church of San Nicolò at Ottana. The bacini on the façade are replicas placed during the restoration. Figure 5: Church of San Platano at villaspeciosa: a quintessential Sardinia hybrid structure with Byzantine spolia, bi-chromatic masonry, typically Pisan roundels, and cavities for bacini. Figure 6: Map of the western Mediterranean production centres, cities, and monasteries mentioned in the text. Figure 7: Fragment of a CoMn bacino from Tunisia, 1200–1250, from the Church of San Lorenzo in Cagliari. Figure 8: Polychrome lead glaze bacino on San Nicolò of Trullas in Semestene, produced in eastern Sicily during the second half of the eleventh century. Figure 9: Luster ware and glazed blue on white slip bacino from the Church of Santa Susanna in Busachi. Spanish production (valencia), 1300–1350. 69 xvi list of figures and maps Figure 10: Luster ware and glazed blue on white slip bacino from the Church of Santa Susanna in Busachi. Spanish production (valencia), 1300–1350. Figure 11: Façade of the Church of Santa Barbara in Sassari, consecrated in the 1270s. The church had more than seventy bacini from at least three different Italian production centers. The pottery predates the construction of the church. Figure 12: Protomajolica bacino from the Church of Santa Barbara in Sassari. Production of Campania, second half of the thirteenth century. Figure 13: Spiral ware bacino from the Church of San vito in San Priamo. Southern Italian production, first half of the thirteenth century. Figure 14: Monochrome bacino without graffiti from the Church of San vito in San Priamo. Southern Italian production, first half of the thirteenth century. Figure 15: Detail of the façade of the Church of Santa Trinità at Saccargia in Condrongianus. Note that it has been completely rebuilt during the first half of the twentieth century. The bacini are primary-colored plates—no attempt to replicate the originals was made. Figure 16: Façade and bell tower of the Church of Santa Trinità at Saccargia in Condrongianus. Figure 17: Polychrome lead glaze bacino from San Nicolò of Trullas in Semestene. Eastern Sicilian production, eleventh century. Note how the bacino is built into the masonry. Figure 18: Façade of the Church of San Pietro di Sorres at Borutta. Figure 19: The Church of Santa Giusta at Oristano (heavily restored). The last two stories of the bell tower are completely rebuilt ex novo. Table 1: List of churches with surviving bacini or recesses for bacini in each Giudicato. Neglia Figure 1: Plan of Aleppo at the time of the Hellenistic foundation ....................................................................................... 119 Figure 2: The Roman plan secundum coelum ................................ 122 list of figures and maps Figure 3: The first Roman plan secundum naturam ..................... Figure 4: The second Roman plan secundum naturam ............... Figure 5: Byzantine fortifications and urban fabric ...................... Figure 6: Process of transformation of the Byzantine cathedral into the al-Halawiyya madrasa .................................................... Figure 7: Relation between the traces of pre-Islamic Aleppo and the layout of the Ayyubid and Mamluk urban walls ....... Figure 8: Traces of the colonnaded street in the urban fabric of the central suq ............................................................................ Figure 9: Zengid and Ayyubid fortifications and urban fabric ... Figure 10: The urban fabric around the al-Umari mosque (the Roman theatre) ....................................................................... Figure 11: Process of nodalization of the building structure of the Bayyada neighbourhood: traces of the Roman plan secundum coelum ............................................................................ Figure 12: Process of nodalization of the building structure of the Bab Quinnasrin neighbourhood: traces of the first Roman plan secundum naturam ................................................................ Figure 13: Process of nodalization of the building structure of the Bab Quinnasrin neighbourhood: traces of the second Roman plan secundum naturam ................................................. Figure 14: Relation between the areas served by cul-de-sacs and the urban fabric in the Bab Quinnasrin area ............................ xvii 123 126 128 130 131 135 139 144 145 148 149 151 Gelichi Figure 1: Towns in northern Italy during the early Middle Ages (after Ward-Perkins 1988) .................................................. Figure 2: Main towns in the “venetia et Histria” during the early Middle Ages ........................................................................... Figure 3: Schematic image of the history of archaeological research in venice .......................................................................... Figure 4: venice, Saint Mark’s bell tower, foundation—1885 (after Boni) ...................................................................................... Figure 5: The glass workshop of Torcello, venice. Polish excavations 1961–1962 (after Leciejewicz, Tabaczyńska, Tabaczyński 1977) .......................................................................... Figure 6: venice between history and archaeology ...................... 177 179 181 182 183 184 xviii list of figures and maps Figure 7: Towns in North-Eastern Italy during the early Middle Ages including the main settlements in the venetian Lagoon ................................................................................ Figure 8: Excavations in the centre of venice (after Gelichi 2006) ................................................................................................. Figure 9: Excavations in the centre of venice (after Gelichi 2006) ................................................................................................. Figure 10: Archaeology in the venetian Lagoon: main topics ... Figure 11: venetian Lagoon, S. Francesco del Deserto, Late Antique waterfronts .............................................................. Figure 12: venetian Lagoon, S. Lorenzo di Ammiana, plan of the excavations (after Canal) ........................................................ Figure 13: venetian Lagoon, Torcello, hypothesis concerning settled areas during the eighth-ninth centuries AD ................. Figure 14: Location of Comacchio .................................................. Figure 15: Goods traded by Comaclenses inhabitants .................. Figure 16: Eighth-century amphorae from Comacchio ............... Figure 17: Comacchio, valle Ponti—villaggio San Francesco 1996, wooden infrastructures ....................................................... Figure 18: Comacchio in the early Middle Ages ........................... Figure 19: Cittanova in the early Middle Ages .............................. Figure 20: venetian Lagoon, S. Pietro di Castello—venice, evaluation of the archaeological resources ................................. 185 187 188 189 192 194 195 198 199 200 201 202 203 209 INTRODUCTION1 Currently, readers who are searching for works on archaeology in the Middle Ages, and especially those students and scholars who are limited to material published in English, will be hard-pressed to gain an understanding of the range of archaeological studies being carried out that focus on the medieval Mediterranean. In general, they will find instead that the archaeology of the classical Mediterranean and the medieval north Atlantic as a whole are much more accessible. unfortunately, this state of affairs masks an exciting variety of work that continues to be carried out on all phases of the medieval Mediterranean world. Studies in the Archaeology of the Medieval Mediterranean addresses this issue by presenting a broad range of studies in English that illustrate the variety and vitality of work being conducted around the Mediterranean basin and focusing on the medieval period. Moreover, this volume seeks to promote the benefits of an interdisciplinary approach to piecing together the various complex entities that make up this area. The chapters in this volume thus focus, each in their own way, on the problem of trying to understand the complexity, not of the Mediterranean basin as a whole, but of some of the individual pieces that comprise it. Every one of them is both an argument for and a potential model of interdisciplinary research in their particular area. Each of the authors notes this in relation to their own case study, and all of them stress the care that must be taken in implementing such a methodology. The studies in this volume were purposely chosen to represent a geographic range spanning the Mediterranean basin from Spain and Morocco to Jordan and Syria. chronologically, their foci cover periods ranging from the ninth and tenth centuries to today. As a whole, they are meant to display the benefits of an interdisciplinary approach that includes or is founded upon archaeology to non-archaeologists. As the title implies, the intent is that this work will be of principal interest to anyone seeking to understand more about medieval archaeology in 1 In addition to that of Johnny de Meulemeester, I would like to dedicate this book to the memory of the late robert t. Farrell. His influence on my view and practice of Medieval Archaeology has come to the fore repeatedly during the course of bringing this work to fruition. 2 introduction the Mediterranean basin. In addition to archaeologists and historians, it will also be of interest to scholars in other fields, such as historical geography, as indicated by our external reviewer. certainly, readers are invited to take away whatever pieces of information strike them as most relevant to their own scholarly pursuits. In putting these archaeological studies together, it has also become clear that the sum of these parts has its own contribution to make in terms of the larger field of Medieval Archaeology. In this regard, the above-mentioned students and scholars investigating the state of Medieval Archaeology as a whole will rather quickly be led to form a number of impressions that this book as a whole serves to dispel. The three that the current volume shows to be most misleading include the following: that there is little to no work being carried out in this area by archaeologists; that the focus of work being carried out by European archaeologists (in northern Europe) is mostly concerned with removing archaeology out from under the shadow of the text; and that when archaeologists do take an interdisciplinary approach to re-creating the past, they limit themselves to historical and literary texts. The latter observation could also lead to the further false impression that these, especially the former, are the only other sources that can be of use to archaeologists. The first impression is perhaps the most easily explained. For one, to my knowledge, there is no major work in English that can serve as an entry into the field of Medieval Archaeology in and around the Mediterranean basin. useful collections of studies do exist of course, for example, in the form of the various SOMA volumes (Symposium on Mediterranean Archaeology) and in the Journal of Mediterranean Archaeology, with each publication’s respective focus on postgraduate and postdoctoral research. For the most part, however, each country, and this holds true for continental Europe as well, has its own journal in its own language, which focuses, quite understandably, on the scholarship related to its own medieval heritage.2 Thus, aside from the 2 For example, Archeologia Medievale; Archéologie Médiévale; Histoire médiévale et archéologie; Arqueología y territorio medieval. Books include Isabelle catteddu, Archéologie médiévale en France: le premier moyen âge, Ve–XIe siècle (Paris, 2009); Joëlle Burnouf, Archéologie médiévale en France: le second moyen âge, XIIe–XVIe siècle (Paris, 2008); S. Gelichi, Introduzione all’archeologia medievale: storia e ricerca in Italia (rome, 1997). Meanwhile, the Journal of Medieval Archaeology is decidedly focused on the British Isles. other works in English include Günter P. Fehring, The Archaeology of Medieval Germany: An Introduction (London, 1991); r. Francovich introduction 3 works just noted and the occasional article in journals like Antiquity or Levant, the English-reading student of medieval archaeology is quickly led to the impression (quite false, of course) that there is a lot more work being done in northern Europe and the British Isles and that archaeology in general in the Mediterranean basin is in short supply after the classical period. This lull appears to remain until after the millennium and the advent of the crusades when work such as that exemplified by cédric devais’ chapter in the present volume joins a growing number of other works and scholars, including myself, who are using an interdisciplinary approach to try and unravel the complexities of the Latin East.3 The same is true of the majority of medieval conferences as well, though here, one could argue that were a group or individual to host them in Mediterranean cities, the balance of papers would shift (although holding them in the depth of winter might also cause even more northern scholars seeking enlightenment, or even simply sunlight, to wish to attend). The most recent attempt (university of Aarhus Press, 2007) to provide a comprehensive introduction to medieval archaeology throughout Europe, The Archaeology of Medieval Europe, recognizes and specifically aims to overcome these national silos, wherein scholarship is focused on one nation or the medieval period of one nation.4 In the forward to this book, the need to “provide students across the continent with a basis on which to build an appreciation of the European dimension [of Medieval Archaeology]” is made explicit.5 In addition, the editors paired authors from different regions of Europe together for each chapter in order to create a more comprehensive approach. The and r. Hodges, Villa to Village: The Transformation of the Roman Countryside in Italy, c. 400–1000 (London, 2003). 3 I am assuming an average reader curious about the subject and not including publications such as the papers from the conferences at Château Gaillard or the very rare studies of Italian castles from the tenth and eleventh centuries, in English, that require more determined effort to track down. on the archaeology of the crusades, see r. Ellenblum, Frankish rural settlement in the Latin kingdom of Jerusalem (cambridge, 2002); Adrian J. Boas, Crusader archaeology: The Material Culture of the Latin East (London, 1999); J. Schryver, “Spheres of contact and Instances of Interaction in the Art and Archaeology of Frankish cyprus, 1191–1359” (Phd dissertation, cornell university, 2005); idem, “colonialism or conviviencia in Frankish cyprus?,” in Boundaries in Depth and in Motion, ed. I. William Zartman (Athens, GA, 2010), pp. 198–230. 4 James Graham-campbell and Magdalena Valor, eds., The Archaeology of Medieval Europe, vol. 1, Eighth to Twelfth Centuries AD (Aarhus, 2007). 5 Ibid., 11. 4 introduction introductory nature of the work naturally results in a broad (and in this particular case thematic) approach that does not focus on any one area such as the Mediterranean. However, the resulting text does make reference to various studies and sites in the Mediterranean and as such serves, perhaps, as a gateway into more focused scholarship such as that found in the journals mentioned above and that serves as the focus of the present volume. naturally, the abundance of sources on archaeology in medieval Ireland, Great Britain and the northern World in English is to be expected. However, a perusal of these leads to a second impression that is inaccurate and that the chapters in this volume serve to dispel. A great number of these would leave the reader feeling that the majority of medieval archaeologists are principally concerned with removing themselves out from under the shadow of the text and ceasing to accept the role of ‘handmaiden of history’.6 others would indicate that medieval archaeologists are monolithically interested only in the economy or identity in the Middle Ages, while still often feeling the spectre of the text haunting them.7 certainly, the issues related to how material culture (archaeology) and textual information (typically history) can be used separately and even combined to study the past are extremely important, and they deserve to be explored deeply.8 The same is true for the issues of economy and identity as well. However, 6 Some of the key works on this topic include david Austin and Leslie Alcock, From the Baltic to the Black Sea: Studies in Medieval Archaeology (London, 1990); david Alban Hinton, ed., 25 Years of Medieval Archaeology (Sheffield, 1983). other topics of discussion, some of which also involve Mediterranean archaeologists, include the economy in the Middle Ages and identity in the Middle Ages. 7 on economy, see r. Hodges, Dark Age Economics: The Origins of Towns and Trade AD 600–1000 (London, 1982); M. Anderton, ed., Anglo-Saxon Trading Centres: Beyond the Emporia (Glasgow, 1999); d. Hill and r. cowie, eds., Wics. The Early Mediaeval Trading Centres of Northern Europe (Sheffield, 2001); t. Pestell and K. ulmschneider, eds., Markets in early medieval Europe: Trading and Productive Sites, 650–850 (Macclesfield, chesire, uK, 2003). Most recently, see, r. Hodges, Goodbye to the Viking? Re-reading Early Medieval Archaeology (London, 2006). Sauro Gelichi, in his contribution to the present volume, notes that this discussion “has been, and essentially remains, entirely north-European.” For a thorough and well-written review of the recent debate surrounding archaeology and identity in medieval Europe, focused in part on German scholarship, see Florin curta, “Some remarks on Ethnicity in Medieval Archaeology,” Early Medieval Europe 15, no. 2 (2007), 159–185. 8 J. Moreland, “Method and theory in medieval archaeology in the 1990’s,” Archeologia Medievale 18 (1991), 7–42; idem, “The Middle Ages, Theory and Post-Modernism,” Acta Archaeologica 68 (1997), 163–182; idem, Archaeology and Text (London, 2001). introduction 5 as the studies in the Aarhus volume show at a more general level, and those in this work show at a more specialized one, there is a great deal of work that is currently being carried out that does not focus on these issues, or that has moved away from the debate on history vs. archaeology while still recognizing the challenges that it highlights. Thus, as we hope to show in the following chapters, the spectre of the text and the sense that archaeologists are in conflict and competition with it that one receives upon reading a good deal of recent works on medieval archaeology does not dominate the entire field as one might all to easily be led to believe.9 regarding the last impression that the present work serves to dispel, the studies presented in this volume show that much can indeed be done by carefully combining archaeology with other sources such as historical and literary ones. Moreover, although the authors demonstrate the importance of integrating historical and archaeological data, they also indicate that there are numerous other types of data that can be and should be carefully considered as part of an interdisciplinary approach (architectural, cartographic, religious, and scientific, to name a few). In his treatments of this issue in the 1990s, as well as most recently in his 2001 book, Archaeology and Text, John Moreland has noted that it is exactly this kind of broadly integrative approach that can aid the field of Medieval Archaeology in becoming more theoretically robust, while also presenting a solution to the issue of combining textual and material (and I would argue other kinds of information as well) more securely.10 Put simply, the more pieces of the puzzle that we are able to assemble, the clearer the picture of the past that we produce will be. or, as Moreland himself noted, “[a]ny attempt to understand the middle ages must use the full range of evidence which exists from that past.”11 9 Among the many examples that exist, see david Austin, “The ‘proper study’ of medieval archaeology,” in From the Baltic to the Black Sea: Studies in Medieval Archaeology, eds. d. Austin and L. Alcock (London, 1990), pp. 9–41; Francovich and Hodges, Villa to Village; Hodges, Goodbye to the Vikings? 10 Some of the key works on Medieval Archaeology and Theory are r. Hodges, “Method and theory in medieval archaeology,” Archeologia Medievale 8 (1982), 7–37; Moreland, “Method and theory,” 7–42; idem, “The Middle Ages,” 163–182; c. tilley, Material Culture and Text: The Art of Ambiguity (London, 1991). See also Moreland, Archaeology and Text. 11 Moreland, “The Middle Ages,” 180. 6 introduction The medieval Mediterranean was a complex world, composed of numerous and varied societies each with their own traditions and contexts. As a result, our efforts to piece together an understanding of this world, no matter what field of research we are working in, must be both nuanced and at their core able to handle and to respond to this complexity. This does not mean that our methodologies need to be complicated in ways that make them incomprehensible to others or unworkable, but it does mean that they need to be able to bridge the traditional boundaries that still exist between various academic disciplines.12 This notion should neither be surprising nor controversial as it is something that is true whether one studies the Middle Ages in the Mediterranean basin or elsewhere. An important theme with direct bearing on this issue of interdisciplinary approaches to the past that emerges across the studies in this volume is the recognition that whatever sources are being incorporated as part of a particular interdisciplinary approach, they must all be analysed carefully and rigorously, both on their own merits and in reference to one another. This is as true before we combine them, as it is during the process of combining them, and as it is after we have combined them. Archaeology and Text tasha Vorderstrasse sets the tone for the volume by exploring how archaeological and textual sources can be used together and demonstrating how beneficial the practice of doing so can be, without worrying about which one will dominate the discussion. As mentioned, however, none of the authors ignore or deny the fact that the integration of textual and material evidence can be complex. As these studies show, no matter how welcoming the additional information provided by other sources may seem, it must be examined cautiously. This also includes our own (archaeological) data, as many of the authors point out. In “A Port without a Harbour: reconstructing al-Mina,” Vorderstrasse begins with a general discussion of the challenges germane to histori- 12 For a thorough discussion of this issue from various points of view, see Eberhard W. Sauer, ed., Archaeology and Ancient History: Breaking Down the Boundaries (London, 2004). introduction 7 cal archaeology. As indicated, this stems from the problem of how to combine archaeological and textual (historical) sources of information. For example, how does one draw upon evidence that they (through no fault of their own) may not fully understand? In the particular case of al-Mina, this task is made very difficult by the fact that there is very little textual information found on the site itself, and that the name of the site in the medieval period remains open to question. Part of Vorderstrasse’s sober approach to the complexities of bringing archaeological and historical source material together to solve this problem includes a discussion of the range of reactions to taking, or not taking, an interdisciplinary approach (from integrationist to nonintegrationist approaches), the various reasons behind these reactions, and a discussion of the notion that ‘textual information needs to be evaluated in light of archaeology and vice versa’. In her particular case, she sets the tone for the arguments of the other studies in this volume by advocating the view that scholars must combine archaeological and textual evidence in order to create the most complete picture of al-Mina that is possible. By doing so, she is able to discuss the seeming contradictions presented in the various sources regarding the navigability of the orontes without having to ignore or make light of them. In addition, she is able to tease out a picture of the sort of inhabitants who would have lived at al-Mina as well as the types of installations that would have been present there during the medieval period. The Benefits of an Interdisciplinary Methodology Following Vorderstrasse’s introduction and lead, but focusing on the implications and benefits of an interdisciplinary and integrative approach to their own specific case studies, the next three chapters form the central portion of this volume. These chapters provide convincing illustrations of both the need to approach the subject of the medieval Mediterranean from a truly interdisciplinary perspective that actively seeks to integrate information and data from various sources and the benefits of doing so. In addition, they each stress the need for critical analysis during the entire process. An important part of the aforementioned need often derives from the veritable tangle of evidence that invariably confronts scholars trying to unravel it and weave together an understanding of this period around 8 introduction the Mediterranean. An example of just how tangled this evidence can become, including becoming tied to modern politics, is presented by Jon van Leuven in his chapter entitled, “The Phantom Baronies of the Western Mani,” which discusses the location of twelve baronies in the Principality of Achaea. Laying out his study, he reminds the reader that research in this area of the world is often inhibited by present-day ethnic tensions, which on the one hand negatively affect the clarity of the medieval issues and problems being explored, and on the other can even contribute to the physical destruction of evidence. In addition to these barriers are those of language, access to the remains, and varying scholarly traditions. In tune with the overall approach being championed in this volume, van Leuven suggests that “interdisciplinary collaboration offers a dual weapon against them: combining the approaches of specialists, as is often wished…” Yet, he also recognizes the difficulties in carrying out such a process as his thought continues with “and evaluating each other’s approaches, which is seldom welcomed.” His suggested response to these difficulties is not to give up, but to sharpen this weapon through “closer study of both existing facts and alternative interpretations.” In other words, he too argues for an interdisciplinary approach based on careful scrutiny of our sources. After laying out the problems surrounding these baronies and explaining their complexity, he provides a strong example of how a systematic, problem-oriented archaeology, such as that called for by Sauro Gelichi in his chapter focusing on Venice at the end of the book, is actually exactly what is needed to solve the issues such as those scholars face in medieval Greece. The Principality of Achaea was founded in the Peloponnese by ‘Frankish’ adventurers after the Fourth crusade and the resulting capture of constantinople. Scholars have long debated the location of certain baronies within this principality. In part, this is due to the challenges mentioned above, but in part it is due to problems in the use and interpretation of the main textual source for these baronies, the Chronicle of Morea. By combining a critical reading of the various versions of this text with other chronicles, maps, travellers’ accounts, and the physical remains, he attempts to “demythologize and disinter the remnants of Frankish nobility throughout the Peloponnese.” Part of this involves arguing for specific locations of baronies, such as Grítsena. But just as importantly, part of this involves admitting when we still do not possess enough information to make a definitive decision, as in the case of the location of Grand Magne. introduction 9 A bit more to the east, a further example of the benefits of an interdisciplinary approach that combines archaeological and textual research is provided by cédric devais in his chapter entitled, “A Seigneury on the Eastern Borders of the Kingdom of Jerusalem: the Terre de Suète.” devais uses a combination of archaeological and historical sources to raise new questions concerning this frontier zone of the Latin Kingdom of Jerusalem and its role in the politics of the area. An important part of his methodology is a nuanced view that is able to recognize changes occurring in the area over time. This type of view is essential to furthering our understanding of these crusader states. combining textual and material evidence ranging from chronicles to the actual caves and other sites fought over by the Franks and Syrians, he provides an important addition to our knowledge of Frankish/ crusader settlement outside of the Kingdom of Jerusalem and argues, convincingly, against the idea of a loose settlement in the Terre de Suète based solely on the collection of taxes. In addition to sketching out its borders and history, devais also notes that a careful study of the sources reveals a shift that occurred (from condominium or joint authority to Frankish domination) in the power structure of the area after the first two decades of its existence. This change, devais argues, led to a change in function as the area then began to fulfill a role as borderland or march for Jerusalem. At the other end of the Mediterranean, on the island of Sardinia, Michelle Hobart echoes this theme of ‘improved vision’ in her study of medieval Sardinian architecture. In her chapter “Merchants, Monks, and Medieval Sardinian Architecture,” she explores the complexities of medieval church construction on the island and proposes that the incorporation of bowls, known as bacini, into the facades of some of these churches served as a marker for certain groups who would visit the island. once again, we are given a sense that the nuanced understanding laid out for us is only possible because of an interdisciplinary approach that critically examines the available data from different points of view. In her particular case, Hobart combines the perspectives of material culture, architectural history, and socio-economic history to increase our understanding of “Sardinia’s evolving settlement patterns, the island’s role in Mediterranean trade, and its relationship with the Islamic world.” The thread with which she weaves these perspectives together is a type of glazed ceramic vessel known as ‘bacini’, which were used to decorate the façades of numerous medieval Italian churches. After taking the reader through a discussion of the various 10 introduction production centres of bacini in the western Mediterranean, Hobart sets their use in the façades of Sardinian churches against a regional background and the actions of players such as the Holy See and the mercantile city of Pisa. The end result, is the recognition of numerous referrals to the latter city and a reinterpretation of the role of bacini as a signifier of Pisan presence on the island and the role of individual merchants in certain Sardinian communities. Medieval Archaeology over Time The final three chapters examine the relationship of the medieval Mediterranean with other time periods, and in the case of the last two, with contemporary society. This examination is extremely important to the field at present as a constant challenge facing medieval archaeologists (and medievalists in general) is making their work relevant to policy makers and others in the modern world. In addition to the general mantra that ‘those who ignore the past are doomed to repeat it’, this section displays how it is that research in medieval archaeology can help us to plan for the future. All three demonstrate how the medieval past is connected to both the past and the present, and Prof. Gelichi, as will be explained further below, presents a particularly strong case for its relation to the future as well. In her study entitled, “The forma urbis of Aleppo (Syria) during the Middle Ages,” Giulia Annalinda neglia reinforces the idea that continuity could play a strong role in the past, even as change also makes its appearance. Her chapter combines various forms of evidence that are only rarely brought together through her incorporation of the study of town planning. As such, she provides a striking example of the kind of jumping-off point that can be constructed when built upon a foundation that results from the combination of various forms of evidence (historical, archaeological, architectural, and cartographic). By focusing on the visible or standing building archaeology, neglia also provides an exciting example of the kind of work that can be carried out with material culture from a nuanced point of view above ground. Her work shows that using an interpretative methodology allows us to compensate for the lack of historical and archaeological data on the structure of the urban fabric of Aleppo in pre-Islamic times, and to integrate the existing sectorial and discontinuous data on the structure of its urban fabric through time. introduction 11 As a result, despite the fact that information about the various phases of evolution of the urban structure of Aleppo is not available in more traditional historical and archaeological sources, which typically focus on the city’s monuments, she is able to trace the development of the city from the Hellenistic period through the Mamluk period. In addition, she is able to offer a conclusion that challenges previous views that saw the Arabs as destroying the geometry and order of the classical city, but instead is able to recognize that “the urban fabric of medieval Aleppo derives from the sum of various urban plans of roman origin, superimposed one over the other.” In addition, she notes that this process of overlay had already begun even before the Arabs conquered Syria. As they added their own phases to the urban fabric of the city, and proactively modified the urban fabric from within to bring it more in line with their needs, its complexity increased and it continued to grow into the palimpsest that we see today, and which she has unravelled. Johnny de Meulemeester’s chapter, “Ethno-archaeological approaches of medieval rural settlement in Spain and Morocco,” also takes us beyond the traditional understanding of what it means to be interdisciplinary. It not only bridges the gulf that too often exists between historical period archaeologists and anthropologists, but also makes a convincing argument for the application of modern ethnographic data to our study of the past. The boundary between past and present is one that is perhaps the most difficult to cross, and de Meulemeester shows us how this can be done successfully. As a result, he is able to use this bridge to provide us with a potential sense of what fortified granaries might have meant to the inhabitants of the Valle de ricote, the kind of thing that is always so elusive for archaeologists. de Meulemeester begins with a discussion of the differences between ethnology and archaeology. He continues with a discussion of the conditions necessary for combining the two approaches: historical continuity, geographical context, and socio-economic context. Although his chapter is a study of granaries in Murcia in Spain and the Atlas Mountains in Morocco, he also argues for an ethno-archaeology that takes account of context and that this is to be derived from the study of material remains. using this approach, de Meulemeester relates the results of fieldwork in Spain and Morocco and weaves together the excavation of a fortified granary of Berber type dating to the thirteenth century in Murcia and the social functions of its more modern descendants in the Atlas Mountains. 12 introduction Sauro Gelichi provides a nice finish to this collection by bringing these studies of the past into the present. In “The Future of Venice’s Past,” he presents an interesting and insightful call for goal-oriented archaeology in the area of Venice and environs. It is a call for revamping the methodology currently being used so that the data ultimately produced can also be used in a more helpful way. reminding us of the need to look ahead with the interdisciplinary and integrative approaches we are presenting, he also describes the ideal kind of data that would be produced from such an approach and that would also be able to be combined with other sources in order to aid the field (in general) in moving forward. Gelichi’s central discussion focuses on the appearance and function of Venice in the ninth and tenth centuries. He asks, for example, whether Venice can be defined as an ‘emporia’ or compared to the emporia of northern Europe. At the same time as he is seeking to broaden the debate about the European economy in the early medieval period in this chapter, he is also examining the nature of archaeology in the Venetian Lagoon. As part of this examination, Gelichi points out the misuse of archaeology in collaborations (or what he describes as the use of it well below its potential). However, as fits the overall mood of this volume, he does not simply complain about the lack of use of archaeological data, but seeks to find better uses of this data. Like van Leuven, Gelichi indicates that we need to challenge each other as we try to piece together the past. He is also just as critical in his examination of the field itself, however, pointing out that although there is good quantity of evidence, this evidence is not always published in a timely fashion and is not always of direct relevance to the origins of Venice or the questions mentioned above. In part this has to do with the fact that the archaeology being carried out currently is not aimed at areas that promise to enlighten us about the early medieval period. Gelichi’s proposed solution to this disconnect is that we re-focus our research so that it becomes problem-oriented, that is focused around enlightening us about the origins of the city. As such, his chapter provides a nice endnote that looks forward to the future of Medieval Mediterranean Archaeology. As outlined above, the present volume draws examples of work from around the Mediterranean basin to demonstrate both the variety of archaeological and archaeologically-oriented studies being carried out and the benefits each of these studies has enjoyed through the use of an interdisciplinary approach. The combination of textual or his- introduction 13 torical sources with archaeological and still other sources is not always straightforward, it is true. In addition, equally critical care must be taken with both the evidence in other fields as well as that in one’s own. However, as the studies in this volume show, the results are more than worth the extra effort. cHAPtEr onE A Port WItHout A HArBour: rEconStructInG MEdIEVAL al-Mina1 tasha vorderstrasse Introduction The questions of how archaeology and history can be brought together in order to reconstruct a site that dates to the historic period, but is unknown in the texts, are numerous and complex. The reconstruction of the port activities at al-Mina, one of the ports of antioch, presents a good example of the challenges involved in just such a process. it is, in many ways, an atypical site for historical archaeologists precisely because of the lack of textual information. There was very little textual information found on the site itself and the name of the site in the medieval period remains open to question. in addition, when alMina was excavated in 1936 and 1937 by sir charles leonard woolley, the excavators found no sign of any harbour installations and none have been discovered in subsequent investigations.2 Thus, in order 1 This article is based upon my dissertation on the site of al-Mina, tasha vorderstrasse, “a port of antioch under byzantium, islam, and the crusades: acculturation and differentiation at al-Mina, a.d. 350–1268” (phd dissertation, university of chicago, 2004). a revised version of my dissertation has now been published, see tasha vorderstrasse, Al-Mina: A Port of Antioch from Late Antiquity until the end of the Ottomans (leiden, 2005) and this book, as well as my article written on al-Mina, tasha vorderstrasse, “a port of antioch: late antique al-Mina,” in Antioche de Syrie: Histoire, images et traces de la ville antique (2004), pp. 363–372, addresses some of the issues discussed here. 2 charles leonard woolley, “excavations near antioch in 1936,” The Antiquaries’ Journal 17 (1937), 1–15; arthur lane, “Medieval Finds at al Mina in north syria,” Archaeologia 87 (1938), 19–78; charles leonard woolley, “The excavations at al Mina, sueidia. ii,” Journal of Hellenic Studies 53 (1938), 133–170; idem, A Forgotten Kingdom: Being a Result of the Results Obtained from the Excavations of Two Mounds Atchana and Al Mina in the Turkish Hatay (london, 1953); hatice pamir and shin’ichi nishiyama, “The orontes delta survey: an archaeological investigation of ancient trade stations/settlements,” Ancient West and East 1.2 (2002), 294–314; vorderstrasse, “a port of antioch,” pp. 363–72; hatice pamir, “The orontes delta survey,” in The Amuq Valley Regional Projects Volume I: Surveys in the Plain of Antioch 16 chapter one to reconstruct the port, scholars are dependent upon what is known about other similar ports that have been excavated and upon textual sources concerning the region that can inform us about the population. These provide a picture that can be further clarified by other textual sources, primarily from egypt, that can provide us with detailed information about daily harbour activities. This chapter examines how one can combine the various pieces of available evidence to reconstruct al-Mina’s port activities, which were a key aspect of the city and probably the reason why it was founded. i will begin with a general discussion about the challenges of historical archaeology and how these challenges have been met by both archaeologists and historians, providing a general overview of issues in historical archaeology, particularly from the roman world. The evidence from egypt, because of its excellent state of preservation of all forms of material, will receive further attention. after this overview, i will focus specifically on al-Mina and its particular issues. Challenges of Historical Archaeology historical, as opposed to prehistoric, archaeology is a discipline within archaeology that presents its own special challenges due to the presence of textual information. ideally, the two types of evidence can be combined together to enhance the conclusions from either source. while this statement may seem obvious, the presence of texts related to a particular site or problem has profound, and sometimes challenging, implications for our interpretation of the material culture. The challenge they present revolves around how scholars can integrate texts with archaeological data and used it in order to produce a more complete understanding of a particular site, region, or cultural group. The textual information available to historical archeologists provides them with a great deal of additional information that is not always as forthcoming from the material record. examples of this information include an indication of the language a particular population spoke, their religion(s), diet, and other cultural information. in some cases, texts can provide archaeologists with very specific information about the names of particular individuals and their families. Further, textual and Orontes Delta, Turkey. 1995–2002, ed. K. aslihan Yener (chicago, 2005), pp. 67–98; vorderstrasse, Al-Mina. a port without a harbour 17 documentation often presents us with information about the governments under which individuals lived, the name of their rulers, and contemporary schools of thought. This is in contrast to the prehistoric periods, where it is more difficult to reconstruct what is known about the population and who they were. as a result, a great deal of information about the life of a particular population is missed and there is less information about the cultural forces that would have affected individuals. when we turn to the historic periods, however, we are suddenly confronted with the presence of texts that can describe in detail events that occurred in a particular region. but despite the presence of texts, gaps often still remain in our understanding of a particular region. texts provide information about individuals, but it is often restricted to information for elites and by elites even in periods where there are large numbers of texts.3 Further, we are affected by the vagaries of preservation of texts such as the fact that in some periods more material exists than in others. other obstacles include the fact that texts concentrate on particular facets of life that their authors found important, ignoring aspects that scholars today might find more interesting. Therefore, the written evidence must be carefully examined and its limitations understood. it is important to judge the relative value of the writer as a historical source by examining the nature of the work, subject and scope, as well as its biases. This provides an indication of what the text can tell us about a particular site or region.4 other challenges exist for historical archaeologists as well. a historian may know a great deal about the political events that shaped a particular region, but very little about the non-elite (usually unnamed) members of the population who lived there and whose buildings the archaeologist is excavating. The degree to which the socio-political events may have shaped a particular population can sometimes be 3 For historians who have tried to construct histories of non-elites, see, among others, emmanuel le roy ladurie, The Peasants of Languedoc, trans. John day (urbana, 1974); idem, Carnival: A People’s Uprising at Romans 1579–1580, trans. Mary Feeney (london, 1979); carlo ginzburg, The Cheese and the Worms: The Cosmos of a Sixteenth-Century Miller, trans. John and anne tedeschi (baltimore, 1982). 4 william richard biers, Art, Artefacts, and Chronology in Classical Archaeology (london, 1992), p. 61; vorderstrasse, Al-Mina, p. 8; Janet Johnson, “response,” Bulletin of the American Society for Papyrology 42 (2006), 268–270. Johnson refers to and provides extensive quotes from the unpublished dissertation of d. whitcomb, “trade and tradition in Medieval southern iran” (phd dissertation, university of chicago, 1979), pp. 200–204. 18 chapter one difficult to assess in the light of these types of texts. There are, however, other primary texts (tablets, ostraca, papyri, and inscriptions) which were composed by individuals and then buried in their houses or, in the case of inscriptions, inscribed on objects such as buildings, graves, and stelae that can provide this information. inscriptions can provide cultural information about individuals within a population,5 while texts found in houses can provide us with information about not only who individuals were, but also where they lived. in some instances, entire neighborhoods and groups of families can be reconstructed.6 The important caveat to this, however, is that one must be certain the individuals mentioned in the text actually lived in that particular house and that the house was not being re-used later as a rubbish dump where the texts were deposited or that the texts were not deposited in the house of someone else for safe-keeping.7 although current excavations are revealing more texts in their domestic contexts, textual finds that can be associated with an individual homeowner are rare because the majority of texts are not found in archaeological contexts.8 There have been attempts to overcome this 5 There are numerous studies that have been done reflecting certain aspects of a population on the basis of inscriptions. see, among others, charlotte roueché, Aphrodisias in Late Antiquity: The Late Roman and Byzantine Inscriptions Including Texts from the Excavations at Aphrodisias Conducted by Kenam T. Erim (london, 1988); Thomas heine nielsen et alii, “athenian grave Monuments and social class,” Greek, Roman and Byzantine Studies 30 (1989), 411–420; charlotte roueché, Performers and Partisans at Aphrodisias in the Roman and Late Roman Periods: A Study Based on Inscriptions from the Current Excavations at Aphrodisias in Caria (london, 1993); patrick amory, “names, ethnic identity, and community in Fifth- and sixth-century burgundy,” Viator 25 (1994), 1–30; onno M. van nijf, The Civic World of Professional Associations in the Roman Near East (amsterdam, 1997). 6 brian p. Muhs, “The girls next door: Marriage patterns among the Mortuary priests in early ptolemaic Thebes,” Journal of Juristic Papyrology 35 (2005), 169–194. 7 The excavators of euhemaria (see below), report that in fact the most fruitful areas that they excavated were the houses and streets that contained rubbish. They report that the papyri, for example, dated over the period of several centuries and were mixed. see bernard pyne grenfell, arthur surridge hunt, and david george hogarth, Fayûm Towns and their Papyri (london, 1900), p. 44. 8 elizabeth c. stone, “texts, architecture and ethnographic analogy: patterns of residence in old babylonian nippur,” Iraq 43 (1981), 19–33; eadem, Nippur Neighborhoods (chicago, 1987); colin a. hope, “Three seasons of excavation at ismant elgharab in the dakhleh oasis, egypt,” Mediterranean Archaeology 1 (1988), 160–178; idem, “dakhleh oasis project: ismant el-Kharab 1988–1990,” Journal of the Society for the Study of Egyptian Antiquities (1997), 157–176; idem, “The Find context of the Kellis agricultural book,” in The Kellis agricultural account book: (P. Kell. IV Gr. 96), ed. roger s. bagnall and colin a. hope, (dakhleh oasis project Monograph) 7 (oxford, 1997), pp. 5–16; Klaas a. worp, albert rijksbaron, and John l. sharpe, a port without a harbour 19 obstacle by reconstructing the contents of houses found in older excavations, but these can be problematic. terry wilfong was successful in attributing papyri to a house found in the course of the excavations at Medinet habu because the material had been recorded in enough detail.9 peter van Minnen, on the other hand, was not successful in his attempts to associate archaeological objects with a particular house at Karanis (although he knew the papyri and some of the other objects that had been found there), in part because he was not able to find a map that showed the particular house he had chosen to study and in part because many of the objects had since disappeared.10 similar problems were encountered by the present author during an attempt to identify the archaeological context of the coins at the site of roman euhemaria. The site was excavated in 1898–1899 by bernard grenfell and arthur hunt, who ignored the archaeological context for most of the material, stating simply that various objects, such as coins, were found all over the site. Their publications do mention an archive from the house of lucius bellenus gemellus and an archive of grain receipts from an oven in another house, but they do not provide plans of either house. Furthermore, grenfell and hunt only published some of the papyrological material.11 The Kellis Isocrates Codex: (P. Kell. III Gr. 95), (dakhleh oasis project Monograph) 5 (oxford, 1997); colin a. hope, “ostraka and archaeology of ismant el-Kharab,” in Greek ostraka from Kellis: O. Kellis, nos. 1–293, eds. Klaas a. worp and colin a. hope, (dakhleh oasis project Monograph) 13 (oxford, 2004), pp. 10–15; colin a. hope, with a contribution by gillian e. bowen, “The archaeological context,” in Coptic Documentary Texts from Kellis Vol. I. P. Kell. V. (P. Kell Copt. 10–52; O. Kell. Copt. 1–2), eds. iain gardner, anthony alcock, and wolf-peter Funk, (dakhleh oasis project Monograph) 9 (oxford, 1999), pp. 96–124; robert p. stephan and arthur verhoogt, “text and context in the archive of tiberianus (Karanis, egypt; 2nd century ad),” Bulletin of the American Society for Papyrology 42 (2006), 190. For discussions of the dialogue between papyri and archaeology in general, see roger s. bagnall, Reading Papyri, Writing Ancient History (london, 1995), pp. 1, 10, 13, and 15; richard alston, Soldier and Society in Roman Egypt: A Social History (london, 1996), pp. 9 and 11; richard alston, The City in Roman and Byzantine Egypt (london, 2002), pp. 1–2; trainos gagos, Jennifer e. gates, and andrew wilburn, “Material culture and texts of graeco-roman egypt: creating context, debating Meaning,” Bulletin of the American Society for Papyrology 42 (2006), 174, 183, and 185. 9 terry wilfong, “The archive of a Family of Moneylenders from Jême,” Bulletin of the American Society for Papyrology 27 (1990), 169–181. 10 peter van Minnen, “house-to-house enquiries: an interdisciplinary approach to roman Karanis,” Zeitschrift fur Papyrologie und Epigraphik 100 (1994), 239 and 247. 11 grenfell, hunt, and hogarth, Fayûm Towns, pp. 44, 261–262, and 331; Joseph grafton Milne, “The coins,” ibid., pp. 64–73; bernard pyne grenfell and arthur surridge hunt, “englische ausgrabungen im Faijûm,” Archiv für Papyrusforschung und 20 chapter one robert stephan and arthur verhoogt were confronted with the problem that the archaeologists digging at Karanis found levels where nothing was recorded and then subsequent levels where a large number of objects were found. This suggested to them that the excavators made mistakes, particularly as the recorded find spots and published plans were not consistent, and because the papyri and ostraca, which were found in the house, actually dated to two distinct periods.12 what these various examples indicate, however, is that even when find spots are recorded, the reconstruction of the archaeological context of the material is not always straightforward. other challenges and difficulties exist for scholars attempting to link historical events to the archaeological record. Many of these stem from the fact that with the exception of some coins, documents, and inscribed material, archaeological objects can rarely be dated to an exact year or ascribed to a specific person. even destruction layers are notoriously difficult to link with historically known events; securely dated examples such as Mount vesuvius’ destruction of pompeii and herculaneum remain extremely rare.13 when archaeologists try to date the occupation of a particular site that is attested in textual sources, they often use those texts to make statements about when a particular site was inhabited. This can also lead to difficulties, however, as has been shown at sites such as at samarra in Mesopotamia. on the basis of the textual evidence, occupation was assumed to have occurred verwandte Gebiete 1 (1901), 217; nicolas hohlwein, “euhéméria du Fayoum,” Journal of Juristic Papyrology 3 (1949), 66; idem, “le vétéran lucius belllienus gemellus, gentleman-Farmer au Fayoum,” Études de Papyrologie 8 (1957), 69–91; Étienne bernard, Recueil des inscriptions grecques du Fayoum, vol. 2, La “Méris” de Thémistos (cairo, 1981), no. 136; gagos, gates, and wilburn, “Material culture and texts,” p. 177; tasha vorderstrasse, “ptolemaic silver in roman egypt: evidence from papyri, hoards and excavations”, Dutch Studies Published by NELL 6, no. 2 (2009), 97–132. 12 stephan and verhoogt, “text and context,” pp. 190 and 197–199. stephan and verhoogt’s study of the material from the house of claudius tiberianus is ongoing and will be published in a forthcoming book, The House of Claudius Tiberianus. Text and Artifact from House C/B 167 in Roman Karanis. 13 william Y. adams, “The archaeologist as detective,” in Variation in Anthropology: Essays in Honor of John C. McGregor, eds. d. w. lathrop and J. douglas (urbana, 1973), p. 24; biers, Art, Artefacts, and Chronology, p. 62; anthony Macelrae snodgrass, An Archaeology of Greece: The Present State and Future Scope of a Discipline (berkeley, 1992), p. 42; charles e. orser, “can there be an archaeology of the great Famine?,” in ‘Fearful Realities:’ New Perspectives on the Famine, eds. chris Morash and richard hayes (dublin, 1996), p. 79. a port without a harbour 21 between 836–883 ad, but it has since been discovered on the basis of archaeological remains that the site was occupied after this period.14 despite these challenges, or perhaps as a result, archaeologists and historians have a variety of opinions about how archaeology and history should be combined. They have reacted differently to using each other’s data, proposing solutions ranging from integrationist approaches (using both types of evidence together) to non-integrationists ones (using each in isolation) and various grades in between. archaeologists working in the historical periods have usually adopted the integrationist approach, sometimes at the expense of the archaeology, by privileging the texts. The textual historians, in contrast, have often advocated a non-integrationist approach arguing that one cannot use archaeological data because it is based on unproven assumptions rather than actual evidence. They criticize archaeologists for drawing too heavily upon historical and linguistic evidence that they do not fully understand.15 14 geza Fehérvári, “near eastern wares under chinese influence,” in Pottery and Metalwork in T’ang China: Their Chronology and External Relations, ed. william watson (oxford, 1976), p. 27; Jessica rawson, Michael s. tite, and Michael J. hughes, “The export of tang Sancai wares: some recent research,” Transactions of The Oriental Ceramic Society 52 (1987–1988), 42; alistair northedge, “Friedrich sarre’s die Keramik von samarra in perspective,” in Continuity and Change in Northern Mesopotamia from the Hellenistic to the Early Islamic Period, eds. Karin bartl and stefan r. hauser (berlin, 1996), pp. 229 and 238. 15 Moses Finley, The Use and Abuse of History (london, 1975), p. 93; philip rahtz, “new approaches to Medieval archaeology part 1,” in 25 Years of Medieval Archaeology, ed. david alban hinton (sheffield, 1983), pp. 12–13; peter h. sawyer, “english archaeology before the conquest: a historian’s view,” ibid., p. 46; christopher J. arnold, “archaeology and history: The shades of confrontation and cooperation,” in Archaeology at the Interface: Studies in Archaeology’s Relationships with History, Geography, Biology, and Physical Science, eds. John l. bintliff and christopher F. gaffney (oxford, 1986), p. 32; John F. Moreland, “restoring the dialectic: settlement patterns and documents in Medieval central italy,” in Archaeology, Annales, and Ethnohistory, ed. a. bernard Knapp (cambridge, uK, 1992), p. 112; a. bernard Knapp, “archaeology and the annales: time, space, and change,” ibid., p. 2; björn berglund, “historical archaeology—a challenge for archaeological Thought,” in Method and Theory in Historical Archaeology. Papers of the ‘Medieval Europe Brugge 1997’ Conference, eds. guy de boe and Frans verhaeghe (Zellik, be, 1997), p. 19; david b. small, “Monuments, laws, and analysis: combining archaeology and text in ancient athens,” in Methods in the Mediterranean: Historical and Archaeological Views in Text and Archaeology, ed. david b. small (leiden, 1995), pp. 143–144; ian Morris, Archaeology as Cultural History: Words and Things in Iron Age Greece (oxford, 2000), p. 26. 22 chapter one several examples from al-Mina and the surrounding region demonstrate the problems that can occur when archaeologists try to over-use historical evidence. First, in the course of excavations at al-Mina, the excavator, sir leonard woolley, found a female marble head. The head was in bad condition and woolley suggested it came from the statue of a city goddess, smashed by ptolemy iii when he invaded the region.16 however, there is no evidence of what type of statue the head came from and no evidence that the site was ever destroyed in this period. in another example, wachtung djobadze used the combined evidence of a tile inscribed with the name barlaam, a greek bread stamp naming barlaam, and details from the life of st. barlaam to claim that the monastery he excavated was, in fact, the monastery of st. barlaam. it is true that barlaam is mentioned in the two inscriptions, that the monastery of st. barlaam was (according to texts) located on Mt. Kasius, and that georgian speakers lived there.17 nevertheless, there is no evidence from the monastery to indicate that it was dedicated to st. barlaam, and while this reconstruction is certainly possible, it is only one possibility. it was this problem that the present author attempted to avoid in a discussion of the ruins of a church at daphne. although there was literary evidence for the names of several monasteries in the area, it was argued that it could not be determined if any of these possibilities were actually the name of this church.18 while the amount of texts may be overwhelming to the archaeologist in some instances, the sheer number of archaeological reports may be daunting for a textual historian. some archaeologists present large catalogues of objects found in books or articles, sometimes without any analysis, which is often meant to be forthcoming, but does not always appear. when textual historians use archaeological material in order to support their arguments they must assess the archaeology in the same way that the archaeologists assess the texts. There are also difficulties in combining sources when the historical and archaeological information conflict with each other. in anthony woolley, A Forgotten Kingdom, p. 180; vorderstrasse, Al-Mina, p. 16. wachtang djobadze, Archaeological Investigations in the Region West of Antioch on-the-Orontes (stuttgart, 1986), pp. 5, 203, and 209. 18 tasha vorderstrasse, “archaeology of the antiochene region in the crusader period,” in East and West in the Medieval Mediterranean I: Antioch from the Byzantine Reconquest until the end of the Crusader Principality, eds. Krijnie ciggaar and Michael Metcalf, (orientalia lovaniensia analecta) 147 (leuven, 2006), pp. 321–323, n. 11. 16 17 a port without a harbour 23 King’s examination of romano-celtic religion, for example, the author ran into such difficulties. The texts in roman gaul indicate that the romans banned the druidic religion and it died out. according to the textual evidence, the romans suppressed the druids for political reasons (the druids allegedly formed the focus of resistance against roman rule) and moral ones (the druids performed human sacrifices which repulsed the romans). The implication of the textual evidence is that one should expect to see fewer traces of celtic religion due to the banning of the religion’s officiates. instead, the majority of sites show no major changes, but rather continuity from the pre-roman and into the roman period as the temples became romanized. This left King with the problem of trying to reconcile this apparent contradiction: the texts suggested change and total replacement, while the archaeological material argued for continuity and romanization. either the druids were not the mainstream proponents of the celtic religion, or the imperial decrees were less harsh than they appeared. as the first explanation is unlikely, it appears that the anti-druid decrees recorded in the texts were unenforceable.19 These examples illustrate an important point: texts are often written to present the author’s view and/ or agendas, and not to reflect archaeological ‘reality’. consequently, textual information needs to be evaluated in light of archaeology and vice versa. archaeologists need to be careful that the texts do not overwhelm the archaeology and affect their understanding of a particular site or region.20 in response to the over-reliance of some archaeologists on texts, the historians’ criticism of these archaeologists’ use of these, and the difficulties that can be encountered when trying to combine archaeology and texts, other archaeologists have adopted the non-integrationist approach. in an extreme case, christopher arnold ignored all textual sources in his examination of the material culture changes during the 19 anthony King, “The emergence of romano-celtic religion,” in Early Roman Empire in the West, eds. Thomas blagg and Martin Millett (oxford, 1990), pp. 232–233. 20 Marcus rautmann, “archaeology and byzantine studies,” Byzantinsche Forschungen 15 (1990), 144; ian hodder, Theory and Practice in Archaeology (london, 1992), pp. 5 and 19; r. g. collingwood, The Idea of History (oxford, 1994), p. 262; J. Keith Jenkins, Re-Thinking History (london, 1997), pp. 5, 7–8, 15, 24, 28, 32–33, and 39; anders andrén, Between Artifacts and Texts: Historical Archaeology in Global Perspective, trans. alan crozier (london, 1998), p. 3; timthoy insoll, The Archaeology of Islam (oxford, 1999), pp. 4–5; vorderstrasse, Al-Mina, p. 8. 24 chapter one transition from roman to saxon britain. he claimed the documentary history did not assist in reconstructing the archaeology of a site and confused the dating of the objects. he further claimed that his study avoided the problems of textual evidence and proved that historical archaeology can stand on its own.21 such an approach, while an interesting exercise, is perhaps somewhat disingenuous since arnold is clearly well versed in the textual history of the period. Therefore, even though he claims not to be using the textual sources, he is clearly aware of them and this undoubtedly affected his understanding of the archaeology. Most archaeologists try to provide a context for their material if they are working in the historic periods and arnold is an unusual case. ignoring the texts means that only part of the picture is examined and, as a result, the objects cannot be fully understood. while there are archaeologists who are perhaps over-reliant on texts and a few who ignore them altogether, others have suggested that the dichotomy between textual history and archaeology is unreal as both the material data and the texts constitute forms of material culture. These scholars point out that although it is important for archaeologists to use the textual evidence, one must not become too reliant upon it. The written record, therefore, can be used as one of the various pieces of evidence, depending upon its relevance to the particular type of analysis.22 The present chapter advocates the view that it is necessary to combine both archaeological and textual evidence in order to create a better understanding of the material culture at al-Mina and create a more complete picture. The question that then remains, is the degree to which the textual evidence should be used in archaeological discourse. The wealth of textual evidence can overwhelm and dominate the archaeological discourse and it is important to determine which texts are important for understanding the archaeology and which are 21 christopher J. arnold, Roman Britain to Anglo-Saxon England: An Archaeological Study (london, 1984), pp. 4 and 165. 22 John c. barrett, “aspects of the iron age in atlantic scotland: a case study in the problems of interpetation,” Proceedings of the Society of Antiquaries of Scotland 111 (1981), 206; david austin, “The ‘proper study’ of Medieval archaeology,” From the Baltic to the Black Sea: Studies in Medieval Archaeology, eds. david austin and leslie alcock (london, 1990), p. 14; Moreland, “restoring the dialectic,” p. 115; graeme barker, A Mediterranean Valley: Landscape Archaeology and Annales History in the Biferno Valley (london, 1995), pp. 2–3; Johnson, “response,” pp. 268–270, citing whitcomb. a port without a harbour 25 not. certain types of textual sources, for example, such as works of philosophy or semantics are not helpful for the context of archaeological discourse. texts that concern daily life, such as documentary papyri, can help to understand how the population functioned. For example, the Menches archive from ptolemaic egypt provides an understanding of how a local village scribe entertained visiting government officials,23 while part of the archive of Theophanes gives valuable information about daily life in the antiochene region as seen through the activities of a visiting egyptian official.24 it is important to keep in mind, however, that these examples should be seen (where possible) within their larger context as they only reflect a specific situation. The activities of Menches, for instance, cannot be extrapolated as being true for every village scribe, nor can the objects consumed by Theophanes be seen as typical for every individual in antioch. in particular, Theophanes, as a visitor to the region, should be seen in that light. Thus his consumption of foodstuffs such as cheese may be a reflection of his own personal taste rather than any sort of typically antiochene culinary or gastronomic norm. now that these theoretical issues have been considered and the examples cited above have provided an indication of the dialectic that exists between archaeologists and historians, i will 23 arthur verhoogt, Regaling Officials in Ptolemaic Egypt: A Dramatic Reading of Official Accounts from the Menches Papers (leiden, 2005), p. 2. 24 p. ryl. iv, 627 and 629, in Catalogue of the Greek and Latin Papyri in the John Rylands Library Manchester, vol. 4, Documents of the Ptolemaic, Roman and Byzantine Periods, eds. colin henderson robertson and eric gardner turner (Manchester, uK, 1952). see also bärbel Kramer and John c. shelton, eds., Das Archiv des Nepheros und Verwandte Texte. Teil I. Archiv des Nepheros. Papyri aus der Trierer und der Heidelberger Papyrussammlung (Mainz am rhein, 1987), p. 59. The Theophanes archive has been been used in several studies on consumption in the roman world. see robert i. curtis, Garum and Salsamenta: Production and Commerce in Materia Medica (leiden, 1991), p. 141, n. 176; hans-Joachim drexhage, “garum und garumhandel im römischen und spätantiken Ägypten,” Münstersche Beiträge zur antiken Handelsgeschichte 12, no. 1 (1993), 31; idem, “der handel, die produktion und der verkehr von Käse nach den griechischen papyri und ostraka,” Münstersche Beiträge zur antiken Handelsgeschichte 15, no. 2 (1996), 33, 36, and 39; idem, “ein Monat in antiochia. lebenshaltungskosten und ernährungsverhalten des Theophanes im payni (26. Mai-24. Juni) ca. 318 n. chr.,” Münstersche Beiträge zur antiken Handelsgeschichte 17, no. 1 (1998), 1–10; philip Mayerson, “enigmatic Knidion: a wine measure in late roman/byzantine egypt?,” Zeitschrift fur Papyrologie und Epigraphik 141 (2002), 209; philip Mayerson, “Qualitative distinctions for ἔλαιον (oil) and ψωμίον (bread),” Bulletin of the American Society for Papyrology 39 (2002), 104. a new study on Theophanes, John F. Matthews, The Journey of Theophanes: Travel, Business, and Daily Life in the Roman East (new haven, 2006), appeared too late to be consulted by the author for this article. 26 chapter one turn to the main subject of consideration here: the site of al-Mina and how one can understand the site as a ‘riverine’ seaport within its historical archaeological context. Al-Mina as a River Seaport of Antioch when one is attempting to reconstruct a particular site and how it functioned, its geographical function is of prime importance. The case of al-Mina is no exception to this rule. The archaeological site of alMina is located in the modern province of the hatay in southeastern turkey (Fig. 1). although the site is a Mediterranean port of the city of antioch (modern antakya), it is located 1.8 km inland on the orontes river rather than on the coast. Following s. dominic ruegg, who in his study of the roman port of Miturnae used the term “river seaport” to define a settlement that is located on a river upstream from the sea and that continued to function as a port, al-Mina should be classified as a ‘riverine’ seaport. one reason for building a port on the river rather than the coast was that it could offer a safe anchorage for ships from storms, thereby enhancing its ability to act as a gateway between antioch and the Mediterranean from the late roman period onwards.25 25 louis brunot, La mer dans les traditions et les industries indigènes à Rabat & Salé (paris, 1920); ernle bradford, Mediterranean: Portrait of a Sea (london, 1971), p. 37; Michel ponisch, Lixus. Le quartier des temples (rabat, 1981), pp. 10 and 13; david J. blackman, “ancient harbors in the Mediterranean part 2,” International Journal of Nautical Archaeology and Underwater Exploration 11, no. 2 (1982), 188; Michel ponisch, “lixus: information archeologiques,” in Aufstieg und Niedergang der Römische Welt 2. 10. 2 (berlin, 1982), pp. 821 and 837; Yehuda Karmon, “geographical components in the study of the ancient Mediterranean,” in Harbour Archaeology, ed. avner raban (london, 1985), pp. 1–2; armando Montanari, “a Modern perspective: The recent development of port cities in southern europe,” in Mediterranean Cities: Historical Perspectives, eds. irad Malkin and robert l. hohlfelder (totowa, 1988), p. 166; s. dominic ruegg, “Minturnae: a roman seaport on the garigliano river, italy,” in Archaeology of Coastal Changes, ed. avnar raban (oxford, 1988), p. 209; alan K. bowman, “post-roman imports in britain and ireland: a Maritime perspective,” in External Contacts and the Economy of Late Roman and Post-Roman Britain, ed. Kenneth dark (woodbridge, 1996), p. 101; glenn e. Markoe, Peoples of the Past: Phoenicians (london, 2000), pp. 187–188; geoffrey e. rickman, “portus in perspective,” in ‘Roman Ostia’ Revisted: Archaeological and Historical Papers in Memory of Russell Meiggs, eds. anna galina Zevi and amanda claridge (london, 1996), pp. 282–283; tiziano Mannoni, “anciennes et nouvelles méthodes dans l’archéologie des échanges et du commerce,” Exchange and Trade in a port without a harbour 27 Antioch Seleucia-in-Pieria Samandag/Suwaidiyya Al-Mina Orontes river Figure 1: The archaeological site of al-Mina, located in the modern province of the hatay in southeastern turkey. 28 chapter one riverine ports have not been studied in as much detail as coastal ports, but there are some that can be compared to al-Mina. The position of al-Mina, located on a river and several kilometres inland, is similar to the phoenician/roman ports of lixus and rabat-chella-sale in modern Morocco, as well as the corsican carthaginian/roman port of alalia or aleria on the island of sardinia. all of these riverine ports were located several kilometres inland from the coast, and their builders presumably chose these locations because they provided a safe refuge for ships from an inhospitable coastline.26 in addition to its being chosen because it provided a safe and sheltered port, the location of al-Mina seems to have been selected with attention to its function as a gateway for trade. two studies conducted on the depth of the river have indicated that just upstream from alMina, the orontes river became very shallow. provided that the river Medieval Europe, p. 7; Jonathon M. wooding, “long distance imports and archaeological Models for exchange and trade in the celtic west ad 400–800,” ibid., pp. 43–44; richard alston, “trade and city in roman egypt,” in Trade, Traders, and the Ancient City, eds. helen parkins and christopher smith (london, 1998), p. 171; John Kenyon davies, “ancient economies: Models and Muddles,” ibid., p. 228; Mark humphries, “trading gods in northern italy,” ibid., pp. 204 and 222; Jeremy patterson, “trade and traders in the roman world: scale, structure, and organisation,” ibid., p. 162; peregrine horden and nicholas purcell, The Corrupting Sea: A Study of Mediterranean History (oxford, 2000), pp. 392–393; silvia orvietani busch, Medieval Mediterranean: The Catalan and Tuscan Coasts, 1100–1235 (leiden, 2001), pp. xi and 257; pamir and nishiyama, “The orontes delta survey,” p. 295; vorderstrasse, Al-Mina, pp. 11–12. ports need not necessarily be selected due to their favourable geographical location, however. The site of amalfi, which became very prosperous in the medieval period, was not located at an ideal site for a harbour. There was a short beachfront, no natural harbour, and bad local winds. see horden and purcell, The Corrupting sea, pp. 112, 117, 121–122, and 133. 26 louis brunot, La mer, pp. 65 and 78; henri basset and evariste levi-provencal, Chella: Une nécropole mérinide (paris, 1923), pp. 3–4; ponisch, Lixus, p. 13; idem, “lixus: informations,” pp. 821 and 837; idem, “territoires utiles du Maroc punique,” in Phönizier im Westen, ed. hans georg niemeyer (Mainz, 1982), pp. 437–438; Jean boube, “la circulation monetaire a sala a l’epoque preromaine,” in Lixus. Actes du colloque (rome, 1992), p. 265; Maurice lenoir, “lixus a l’epoque romaine,” ibid., p. 271; noé villaverde vega, “aportaciones a la cronologia de la antigüedad tardía en Mauretania tingitana: datos de los vajilias africanas,” ibid., p. 346; edward lipiński, “sala,” in Dictionnaire de la civilisation phénicienne et punique, ed. e. lipiński (turnhout, 1992), p. 385; Michel ponisch, “lixus,” ibid., pp. 265–266; pierre rouillard, “Maroc,” in La civilisation phénicienne et punique. Manuel de recherche, ed. véronique Krings (leiden, 1994), p. 781; Jean Jehasse and laurence Jehasse, Aleria rediviva/ Aleria ressuscité. 40 ans de découvertes archéologiques (ajaccio, corsica, Fr, 1997); Markoe, Peoples of the Past, pp. 187–188. i would like to thank dr. walter Kaegi of the university of chicago for suggesting that i look at these ports in north africa and sardinia. a port without a harbour 29 was the same depth in antiquity, this would have meant that while ships (small trading vessels rather than large boats) would have been able to travel up river to al-Mina to offload their goods, they would not have been able to sail very far beyond it.27 The question of whether or not the orontes river was navigable in antiquity from the Mediterranean to antioch is one that has occupied scholars for decades.28 it is also one that illustrates the challenges of using history and archaeology together. in this case, the main challenge is that while the geography argues that the river never could have been navigable, the texts suggest that perhaps it was. The earliest source mentioned as possible evidence for the navigability of the river between the Mediterranean and antioch is the gourob papyrus (P. Petrie ii, 45 and iii, 144) from the third century bc.29 Francis piejko perhaps best summed up the state of this problematic text, writing that the papyrus is “a forbidding aspect of unrestored lacunae and some mistaken reconstructions.”30 The papyrus describes how ptolemy iii and his navy traveled first to seleucia-in-pieria and 27 naval intelligence division, Turkey (great britain, 1942), p. 102; vorderstrasse, “a port of antioch,” pp. 363–72; vorderstrasse, Al-Mina, pp. 12–13. i would like to thank tony J. wilkinson and Jesse casana for providing me this information about their work done on the depth of the orontes river in 2000. 28 denis van berchem, “le port de séleucie de piérie et l’infrastructure logistique des guerres parthiques,” Bonner Jahrbücher 185 (1985), 68–69; pamir and nishiyama, “The orontes delta survey,” p. 295 and n. 5; vorderstrasse, “a port of antioch,” pp. 363–72; vorderstrasse, Al-Mina, pp. 12–13. 29 For editions of the papyrus, see John p. Mahaffy, The Flinders Petrie Papyri with Transcriptions, Commentaries and Index. Part II (dublin, 1893); John p. Mahaffy and Josiah gilbart smyly, The Flinders Petrie Papyri with Transcriptions, Commentaries and Index. Part III (dublin, 1905); ulrich wilcken, Grundzüge und Chrestomathie der Papyruskunde. Erster Band: Historischer Teil Zweite Hälfte: Chrestomathie (leipzig, 1912), p. 1; Maurice holleaux, Études d’épigraphie et d’histoire grecques, vol. 3, Lagides et Séleucides (paris, 1942), pp. 281–315; Francis piejko, “episodes from the Third syrian war in a gourob papyrus, 246 b.c.,” Archiv für Papyrusforschung und verwandte Gebiete 36 (1990), 13–27. For the historical context of the papyrus, see edwyn robert bevan, The House of Ptolemy. A History of Egypt under the Ptolemaic Dynasty (chicago, 1958), pp. 198–200; hans hauben, “l’expedition de ptolémée iii en orient et la sédition domestique de 245 av. J.-c.,” Archiv für Papyrusforschung und verwandte Gebiete 36 (1990), 29–37; günther hölbl, Geschichte des Ptolemäerreiches: Politik, Ideologie und religiöse Kultur von Alexander dem Großen bis zur römischen Eroberung (darmstadt, 1994), p. 47; werner huß, Ägypten in hellenistischer Zeit: 323–30 v. Chr. (Munich, 2001), pp. 338–352. 30 piejko, “episodes from the Third syrian war,” p. 13. The uncertainty about the papyrus makes it particularly unfortunate that photos of it have never been published. 30 chapter one then arrived at antioch, but after this description the text becomes very broken. although the text, as reconstructed by piejko, states that ptolemy iii and his navy traveled on the orontes by ship and then arrived at antioch, it does not actually state that they went the entire way by ship.31 in his first study of the issue, glanville downey initially argued that one could not state that the orontes was navigable on the basis of this fragmentary text.32 in his second book written two years later, however, he claimed that the naval force sailed up the orontes from seleucia-in-pieria to antioch, taking time to admire the beauty of the orontes valley from the ship.33 it is unclear why downey changed his mind about the reliability of the papyrus. another scholar, denis van berchem, also argued that the gourob papyrus cannot be used to reconstruct the navigability of the river.34 The uncertainty of the text supports van berchem and downey’s initial view that this text cannot be used to reconstruct the navigability of the river. The sources for the navigability of the orontes river in the roman period are strabo and pausanius. strabo (ca. 64 bc–19 ad) indicated that the river was navigable between the city of antioch and the sea, noting that one could voyage from the sea to antioch in one day.35 strabo, following earlier geographers, was particularly interested in describing coastlines and their harbours, islands, and navigable rivers. although he based his account of the earlier geographer posidonius, a native of the region, it is important to remember that strabo did not personally visit the roman province of syria.36 pausanius, writing in the second century ad claimed that a roman emperor (probably tiberius) wished to have ships sail up the river to antioch and had 31 ibid., coll. 2, lines 23–25 and col. 3, lines 1–19. i would like to thank nico Kruit, formerly of leiden university for his help in reading this edition and discussing the passage with me. 32 glanville downey, A History of Antioch in Syria from Seleucus to the Arab Conquest (princeton, 1961), pp. 18–19 and 89–90, n. 13. 33 idem, Ancient Antioch (princeton, 1963), p. 51. downey is clearly using bevan’s translation for this reconstruction. see bevan, The House of Ptolemy, p. 199. see also piejko, “episodes from the Third syrian war,” p. 25, n. 18. he noted that bevan misinterpreted this passage. 34 van berchem, “le port de séleucie de piérie,” p. 68. 35 strabo, Geography, volume 7, 16. 2. 7, ed. and trans. horace leonard Jones (cambridge, Ma, 1983). 36 edward charles leonard van der vliet, Strabo over Landen, Volken en Steden (assen, 1977), pp. 121 and 139; van berchem, “le port de séleucie de piérie,” p. 68; daniela dueck, Strabo of Amasia: A Greek Man of Letters in Augustan Rome (london, 2000), pp. 16, 40, 43, and 186. a port without a harbour 31 a suitable channel dug into which the orontes river was diverted in order to allow ships to move freely.37 an examination of the terrain between the orontes river delta and antioch, however, indicates that the construction of such a channel would be impossible. van berchem suggested that the construction of the special channel might have been connected with the dredging of the orontes, as mentioned in the Theodosian Code.38 nevertheless, the roman evidence is somewhat vague and does not really prove that the river was navigable. The only early byzantine author who is cited in the argument that the orontes river was navigable from antioch to the Mediterranean is libanius.39 in his Oration, libanius stated that: what is of greatest importance, however, is that the river as it flows to the sea after leaving the city, is not impassable to ships because of rocks . . . furnishing the means of transportation through itself for the varieties of wood which are bought from everywhere.40 libanius provided the most detailed description about the navigability of the orontes, and unlike the other authors, his work focused solely on antioch. libanius is an important source because unlike strabo and pausanius, he was a native of the city and would have actually seen the orontes river for himself. it is important to remember, however, that libanius’ oration was written with the purpose of proclaiming the importance of the city of antioch. Therefore, his work is somewhat biased and it is clear from his description that he wished to present the navigability of the orontes river as a positive. if he was correct, this does, of course, suggest that the river was navigable. however, libanius 37 pausanias, Description of Greece, book 8. 39. 3, ed. and trans. william h. s. Jones (cambridge, Ma, 1961). 38 see Theodosian Code 10. 23. 1, trans. clyde pharr, The Theodosian Code and Novels and the Sirmondian Constitutions (princeton, 1952). see also van berchem, “le port de séleucie de piérie,” p. 68; christian habicht, Pausanias’ Guide to Ancient Greece (berkeley, 1998), pp. 1, 8, 21, 24, 28, and 66; vorderstrasse, Al-Mina, p. 43, n. 165. as the present author has already noted, van berchem’s hypothesis seems more likely than pharr’s suggestion that the text means that the orontes river was cleared of pirates. 39 downey, A History of Antioch, p. 18; van berchem, “le port de séleucie de piérie,” pp. 69–70. 40 libanius, Oration Xi, 262, trans. glanville downey, “libanius’ oration in praise of antioch (oration Xi),” Proceedings of the American Philosophical Society 103 (1959), 652–686; georgios Fatouros and tilman Krischer, Libanios. Antiochikos (Or. XI). zur heidnischen Renaissance in der Spätantike (vienna, 1992); albert Francis norman, Antioch as a Centre of Hellenic Culture as Observed by Libanius (liverpool, 2000). 32 chapter one may have exaggerated in order to enhance the status of antioch, or he may have passed over the fact that parts of the river may not have been navigable. in other words, he may have been referring to traffic that could have occurred at the orontes river mouth or near the city of antioch, and glossed over the fact that the river was not navigable between these two points. certainly the early islamic41 and crusader42 evidence indicates that the river was not navigable. Yaqut (d. 1229) describes baggage animals travelling between antioch and port st. symeon,43 and the anonymous author of the Codex Vaticanus Arabicus 286 describes how building stones were transported from suwaidiya (port st. symeon) to antioch by carts and animals.44 This text also provides an idea of how the goods were transshipped in the medieval period and certainly suggests that at least at that point the river was not navigable. Future geomorphological studies along the orontes river will hopefully further clarify our understanding of its navigability. if they indicate that the river was navigable in the past as some of the texts suggest, then new ideas will have to be formulated about why al-Mina was founded at this particular spot along the orontes river. until that point, based on a combination of the available sources (archaeological, geographical, and historical), it seems logical to conclude that the harbour of al-Mina was placed deliberately at a point on the river where it was not possible to travel any further inland by boat. such a location would also explain why the site continued to be inhabited for hundreds of years: it was in a beneficial location. Reconstructing the Harbour of al-Mina and its Population when reconstructing the daily activities occurring at the port of alMina, one is faced with many of the same challenges outlined above. 41 al-Muqaddasī, Description of Syria including Palestine, trans. guy le strange, (The library of the palestine pilgrims’ text society) 3 (london, 1896), p. 82. 42 lane, “Medieval Finds,” 23. 43 ibid. 44 vatican city, codex vaticanus arabicus 286: 114, r. 13, trans. and ed. ignazio guidi, “una descrizione araba di antiochia,” Rendiconti della Reale Accademia dei Lincei, Classe di Scienze Morali, Storiche e Filologiche, ser. 5, vol. 6 (1897), 137–161; olga de lébédew, Codex 286 du Vatican: Récits de voyages d’un arabe (st. petersburg, 1902); william Franklin stinespring, The Description of Antioch in Codex Vaticanus Arabicus 286 (phd dissertation, Yale university, 1932). a port without a harbour 33 as a result, it becomes even more necessary to take an integrated approach to all of the available evidence. although even this process does not reveal an exact picture, it does allow us to bring certain pieces of the puzzle into focus. as mentioned, al-Mina is barely visible in the textual record. For example, there is information about the region in textual sources, but these rarely focus on daily life. Most texts refer only to antioch or some of the monasteries in the region. nor do they provide detailed descriptions about ports and how they functioned, even when the name of ports are mentioned. in addition, since we do not know the name of the site during all the relevant periods, it is impossible to know if these texts actually mention the site itself. arthur lane, who first studied the medieval ceramics from the site, also used vague literary descriptions to claim that the site was called bytyllion, suwaidiyya, and port st. symeon in the late antique and medieval periods. in the late roman period, lane suggested that the name of the site of al-Mina was bytyllion, based upon a short reference in John Malalas. There is more literary evidence for the port of suwaidiya and port st. symeon, which appears in various texts in a variety of languages. These descriptions are usually short, however, and do not provide many clues that could prove this port was al-Mina. Furthermore, absolutely no evidence was uncovered at the site to indicate what it was called in the medieval period. at the same time, the site seems too small to be the one the texts describe as port st. symeon or suwaidiya, although it might have been part of it. The exceptions to this general lack of primary information about the region are the archive of Theophanes in the late roman period (see above), inscriptions from the region, and some references from the cairo geniza. The majority of the inscriptions are not medieval, however, and there are a small number of texts that can help reconstruct information about the population.45 despite these problems, the textual sources that do date to the medieval period can help to provide information about the harbour of al-Mina in that they provide information about the types of activities 45 lane, “Medieval Finds,” 19; Josette elayi, “al-Mina sur l’oronte à l’époque perse,” in Studia Phoenicia V: Phoenicia and the East Mediterranean in the First Millennium B.C., ed. edward lipiński (leuven, 1987), pp. 264–265 and 264, n. 84; vorderstrasse, “a port of antioch,” pp. 363–72; vorderstrasse, Al-Mina, pp. 43, 93, 112–113, 143, and 159–160. For a detailed discussion of the region, see ibid., pp. 37–41, 64–67, 88–92, 104, 106–111, and 136–139. 34 chapter one might have occurred at al-Mina. while we cannot prove that these did occur here, they at least provide us with an indication of the types of individuals that may have lived at the site, the languages they might have spoken, and the type of religious convictions they might have had. based upon historical information about al-Mina, one can presume that until the end of the crusader period, the majority of people living at the site were christians. prior to the islamic conquest, the majority of individuals in the antiochene region spoke greek or syriac. syriac seems to have predominated in the countryside, but people still spoke greek there as well. after the conquest, however, greek became less common and individuals spoke both syriac and arabic. arabic became increasingly widespread, and even after the Middle byzantine occupation of the region, it remained in use, along with greek (which became newly resurgent thanks to the influence of byzantine rule). in the crusader period, most of the population probably continued to speak arabic, although some probably also learned to speak old French. The evidence points to an ethnically mixed population in the area in the time of the crusades, in addition to greek, arabic, and syriac speakers, as well as crusaders, there were also armenians, georgians, persians, and italians. in the face of little evidence, it is necessary to assume that at al-Mina one would have found a similarly mixed population that would have depended on the particular time period under examination.46 The single inscription from the site provides tantalizing evidence that seems to fit in with our understanding of the region in this period, but it remains one object. if the site is re-excavated in the future, our picture of al-Mina could easily change. nonetheless, it is necessary to try to understand the data as it appears at this point in time, rather than to wait for the understanding of material culture to be expanded, recognizing that this understanding may change.47 at the moment, the textual evidence from the region, combined with the single inscription from the site, 46 vorderstrasse, Al-Mina, pp. 40–41, 67, 90–93 (92–93 discuss the only inscription from the site, which dates to the Middle byzantine period, is written in arabic, and seems to be the gravestone of a christian woman), 104, 109–111, and 139. 47 teresita Majewski and Michael J. o’brien, “The use and Misuse of nineteenth century english and american ceramics in archaeological analysis,” Advances in Archaeological Method and Theory 11 (1987), 172. a port without a harbour 35 provides us with a model of the sort of inhabitants who would have lived at al-Mina, although we do not know what their standard of living would have been. The next question, however, is where and how they lived. There is little architectural evidence from the site, and as no harbour installations have been discovered, one is left with the difficulty of determining what was happening there during the medieval period. a combination of the historical and archaeological evidence can help us to tease out the information about what type of port we would have expected to see at al-Mina. al-Mina does not seem to have been a large site, and therefore we should not necessarily expect to find all the facilities that would be common to larger ports. it is probable that in addition to houses for the population who lived there (however limited in numbers the population might have been), there would have been buildings where goods would have been stored and mooring sites for the ships. The fact that the port was on the river rather than the coast suggests that any lighthouses would have been located closer to the sea rather than at the harbour of al-Mina itself. it is also unclear whether such a small port site would have necessarily had or needed inns and facilities for the repair of ships.48 The nearby riverine seaport of roman derhush on the orontes, located in modern syria, shows how simple the port installations might have been. This site only consisted of a few buildings and a wall for the anchorage for ships. similar installations existed at the punic and roman riverine port of lixus in Morocco (where the trapezoid basin also formed a jetty). The fact that Frank broeze, “port cities: The search for an identity,” Journal of Urban History 11 (1985), 210; david J. blackman,“bollards and Men,” in Mediterranean Cities, pp. 11–13; John peter oleson, “The technology of roman harbours,” The International Jorunal of Nautical Archaeology and Underwater Exploration 17, no. 2 (1988), 147 and 155; geoffrey e. rickman, “The archaeology and history of roman ports,” International Journal of Nautical Archaeology and Underwater Exploration 17, no. 3 (1988), 261–262; susan Mosher stuard, “ancillary evidence for the decline of Medieval slavery,” Past and Present 149 (1995), 20; John r. leonard, “harbor terminology in the roman periploi,” in Res Maritimae, vol. 1, eds. stuart swiny, robert l. hohlfelder, and helena wylde swiny (atlanta, 1997), p. 191; birgit Kulessa, “The harbour suburb and its significance for the urban development of stralsund,” in Maritime Topography and the Medieval Town, eds. Jan bill and birthe l. clausen (copenhagen, 1999), 4:72 and 77; alston, The City, p. 338; sean a. Kingsley, “decline in the ports of palestine in late antiquity,” in Recent Research in Late Antique Urbanism, ed. l. lavan (portsmouth, ri, 2002), p. 69; vorderstrasse, Al-Mina, p. 12. 48 36 chapter one lixus was larger in scale and contained a more sizeable population49 indicates that the size of the population did not necessarily affect the type of harbour installations used. a large population might still use fairly simple harbour installations as easily as a smaller one. while the examples of lixus and derhush are interesting ones, it is unfortunate that no medieval riverine ports have been excavated, particularly in syria. More evidence needs to be gathered before we can make definitive statements about their appearance. at the same time, however, evidence from the ottoman and early modern period provides us with some indication of what types of installations might have been present at al-Mina in the medieval period. The antiochene area attracted orientalists throughout the nineteenth century who had hoped to visit the city of antioch and see its classical and christian ruins.50 as a result, artists also visited the area. one of them, william henry bartlett, drew a view of an orontes ferry. The orontes ferry consisted of two sticks of wood anchored at either side of the river (at one side the stick is seen to have been driven into a pile of ruins) with a rope running between them. The ferryman went back and forth across the river by means of the rope. in this drawing, there is no attempt to depict any port installations (assuming any existed) and the ferry seems to have been for men and their livestock.51 still, it is possible that such ferries could have existed at the port of al-Mina. when richard pococke visited the orontes river in 1765, he found a large basin along the river which seems similar to the installations at derhush or lixus, and which could have been part of a riverine port system.52 however, it is difficult to know if he is referring to the ottoman port of al-Mina or another port that has since disappeared. ponisch, “lixus,” fig. 9; ignace penã, “un puerto fluvial romano en el orontes,” Liber Annus 45 (1995), 343 and 350; vorderstrasse, Al-Mina, p. 12, fig. 2 (cd-roM/ Maps). 50 see downey, A History of Antioch, pp. 664–674, who provides a list of travellers who visited the city of antioch and provided descriptions and drawings. he does not provide a list of those who were active in the region rather than the city nor does he indicate if the individuals in his list visited other sites in the area. 51 william henry bartlett, John carne, william purser, and Thomas allom, Syria, the Holy Land, Asia Minor, etc. Illustrated, in a series of Views drawn from Nature by W. H. Bartlett, William Purser, etc. With descriptions of the Plates [3 volumes in 1] (london, 1836–1838). 52 richard pococke, A Description of the East and Some Other Countries, vol. 2, Observations on Palestine or the Holy Land, Syria, Mesopotamia, Cyprus, and Candia (london, 1743–1745), pp. 186–187; vorderstrasse, Al-Mina, pp. 141–142. 49 a port without a harbour 37 still, his comments provide an indication that there was a similar sort of installation as at the much earlier sites of lixus and derhush in the area. woolley mentions that in 1882 the site consisted of several storehouses and a customs house, although he does not mention any port installations. Martin hartmann, who visited the antiochene region in the late 1890s, mentions that at al-Mina (which in this period was called el-eskele) there was a place where boats could be tied to the river bank with a rope as well as warehouses and other buildings. when woolley visited the site in 1936 and 1937, this area seems to have been reduced to a single warehouse, customs officer’s house and a few cottages.53 clearly the harbour installations at ottoman and early modern al-Mina were very basic. whether or not this is an indication of the earlier harbour activities, however, is unclear. it could suggest that medieval al-Mina had equally simple port installations, or simply that the site had become smaller through time and the port installations had become increasingly simple as it decreased in its importance. textual sources can also provide us with an idea of the sort of activities that might have occurred at a port, no matter how simple it was. The most informative sources come from roman and byzantine egypt, and while one cannot assume that the situation in egypt and syria was identical, these sources can at least provide us with an idea of what might have happened there. The textual evidence for large ports indicates that a wide variety of goods would have been traded, and that large numbers of individuals from various ethnic groups and different associated professions (including soldiers, merchant families, slaves, ship owners, and prostitutes) were involved in trade and its associated industries.54 again, one would expect that at a small port Martin hartmann, “das liwa haleb (aleppo) und ein teil des liwa dschebel bereket,” Zeitschrift der Gesellschaft für Erdkunde zu Berlin 29 (1894), 159, 511–512, n. 1, and taf. 3; woolley, “The excavations at al Mina,” 4; lane, “Medieval Finds,” 24–25; woolley, A Forgotten Kingdom, p. 172, Thomas alan sinclair, Eastern Turkey: An Architectural and Archaeological Survey (london, 1990), 4:304; werner arnold, Die arabischen Dialekte antiochiens (wiesbaden, 1998), p. 346; vorderstrasse, Al-Mina, pp. 140–142. 54 For papyrological evidence for port and trading activities from the roman period, see the evidence from egypt’s eastern desert: O. Claud. 224–242, esp. O. Claud. 225, 227, 232, 238, and 241–242, in Mons Claudianus, ostraca graeca and latina II. O. Claud. 191 à 416, eds. Jean bingen, adam bülow-Jacobsen, walter e. h. cockle, Francis Kayser, and wilfried van rengen (cairo, 1997); olaf e. Kaper and willemina Z. wendrich, “east and west in roman egypt: an introduction to life on the Fringe,” in Life on the Fringe: Living in the Southern Egyptian Deserts During the Roman and 53 38 chapter one such as al-Mina the variety of ethnic groups and professions associated with the harbour would not be as great. one would expect that there would probably be merchant families who lived at the site, such as hartmann observed when he visited in the 1890s, when he met the owner of seven warehouses and a government representative.55 but again, there were not a large number of individuals living there. one can assume that at medieval al-Mina, too, there would have been merchants who lived at the site as well as officials who dealt with customs issues. There may have also have been some limited installations for the sailors as well. but in the absence of any local or site specific medieval evidence one cannot, of course, state for certain what would have been at the site. Conclusion This chapter has adopted the viewpoint that it is necessary to use the texts in order to provide a general cultural context for the material objects because such a context is very hard to determine from the objects alone. as the examples in the overview and the case study have demonstrated, the historical and archaeological sources should be integrated together, wherever this is possible. The challenges presented by archaeology of the historic periods can also present opportunities. as has already been noted, the combination of historical and Early Byzantine Periods, ed. olaf e. Kaper (leiden, 1998), pp. 1–3; adam bülowJacobsen, “traffic on the road between coptos and the red sea,” ibid., pp. 68 and 70. For the activities of the merchant family of nicanor, see O. Brux. 7, in Ostraka aus Brüssel und Berlin, ed. paul viereck (berlin, 1922); Jean bingen, ed., with Marianne lewuillon-blume and Jan Quaegebeur, Au temps où on lisait le Grec en Égypte (brussels, 1977); O. Petr. 220–304, in John gavin tait, ed., “ostraca in prof. w. M. Flinders petrie’s collection at university college london,” in Greek Ostraca in the Bodleian Library at Oxford and Various Other Collections, vol 1, eds. John gavin tait and claire préaux (london, 1930); O. Bodl. II. 1970, in John gavin tait and claire préaux, eds., Greek Ostraca in the Bodleian Library at Oxford and Various Other Collections, vol. 2, Ostraca of the Roman and Byzantine Periods (london, 1955); alexander Fuks, “notes on the archive of nicanor,” Journal of Juristic Papyrology 5 (1951), 207–209 and 212; david Meredith, “The Myos hormos road: inscriptions and ostraca,” Chronique d’Égypte 31 (1956), 356–357; steven e. sidebotham, Roman Economic Policy in the Erythra Thalassa. 30 B.C.–A.D. 217 (leiden, 1986), pp. 11, 14–15, 17–18, 48, 50–51, 58, 66, 83, 85–87, and 91; Kai ruffing, “das nikanor-archiv und der römische süd– und osthandel,” Münstersche Beiträge zur antiken Handelsgeschichte 12, no. 2 (1993), 2–3 and 8. 55 hartmann, “das liwa haleb,” pp. 159, 511–512, n. 1, and taf. 3; arnold, Die arabischen Dialekte, p. 346; vorderstrasse, Al-Mina, pp. 142–143. a port without a harbour 39 archaeological evidence allows us to better understand how a particular site or region functioned in a particular period. in extremely well documented instances, from egypt and Mesopotamia, we can even know who lived in a particular house and reconstruct some of their activities. al-Mina is an interesting example for study because it is a medieval site where there are only a few texts that provide limited information about the site. This provides us with a good example about how the gaps in texts and archaeology can be overcome. in the prehistoric periods, we would be left speculating about who might have lived in the site, what religion they might have had and what languages they could have spoken. since we have textual evidence for the region, however, we can speculate instead about the types of people who might have lived at the site. equally, however, we are confronted with challenges when examining what types of port installations might have been found at the harbour of al-Mina. as no harbour installations have survived and the texts that describe the port installations in the region are vague (except for the ottoman period), we are left using parallels from similarly located sites in order to have a better idea of how al-Mina might have been constructed. it is only through a combination of different types of evidence that we can gain some idea of what al-Mina might have looked like and who might have lived there. chapter two the phantoM baronies oF the western Mani Jon van leuven et jusqu’à l’extrémité du cap ténare on marchait dans l’empire des illusions. pouqueville1 greece in the Middle ages was the scene of a struggle between what, already then, were generalized about as east and west—the longest one in history with fairly clear limits in space and time. The conflict, mainly an outgrowth of the early crusades and lasting over half a millennium, owed more to the country’s intercontinental exposure than to her internal assets. nonetheless, she acquired the character of a promised Land for pugnacious peoples, rich in competition within her borders. a similar questing, contentious spirit is still alive among students of the region, whose most mysterious, medieval monuments i want to revisit here. The ethnic turbulence of the medieval eastern mediterranean has posed obstacles to research. not only did human misunderstanding create complex problems for us to solve, but the survival of material evidence has also been relatively precarious. The primary sources come from diverse cultures, and the secondary literature reflects different traditions of scholarship, as well as barriers of language and of physical access to the remains. such difficulties blur our definition of reality and can turn the truth about the past—distant in both intellectual and temporal dimensions—into an imaginary fortress inhabited by ghostly theories, valid today and vanished tomorrow. potential fantasies in the sphere of science demand scrutiny as methodical as do ‘phantom kingdoms’ in the world of warlords.2 interdisciplinary collaboration offers a dual weapon against them: combining the approaches of specialists, f.-c.-h.-L. pouqueville, Voyage de la Grèce, 5 vols., 2nd ed. (paris, 1827), 5:594. see william miller, The Latins in the Levant (London, 1908), e.g. pp. 149, 175, and 244, on such disembodied domains in northern Greece during the thirteenth century. 1 2 42 chapter two as is often wished, and evaluating each other’s approaches, which is seldom welcomed. more specifically, our acquaintance with life in Greece during the ‘Latin’ heyday of 1200–1400 is based largely on a single lost document, the Chronicle of Morea, through its two derived versions in Greek and french, besides two descendants in aragonese and italian.3 This textual quadruped has itself been a ‘sacred cow’ for ideologically discordant bibliophiles who, two centuries ago, began to milk it of revelations that have not dried up yet. These efforts involve several approaches— linguistic, historical, archaeological, cartographic4—which have not always been consistent and comprehensive. The same applies to other evidence from those years, and the net result is a picture with coherence in broad outlines but controversy over key details. Data have been eclipsed in quantity by commentary, so vast by now that its best claim to a survey needs extensive updating.5 an omen of continued conjecture is that CM has aroused more attention to what it ostensibly meant than to what it actually said. explanation of its contents is concerned ever less with exact terminology, and tends to focus on extracting cultural implications, or reconstructing events and conditions, from supposedly solid information. Debate has been stimulated by the discovery in CM of countless errors or contradictions, with conceivable causes that vary from the author’s ignorance to interference by copyists and translators. The risk is that fundamental inaccuracies can lead to historical illusions, some of which are the subject of this inquiry. in other words, our interdisci- 3 Chronicle of Morea hereafter cited as CM and, for the four versions, GCM (with line numbers) for the Greek, FCM and ACM (paragraphs) for the french and aragonese, and ICM for the italian, respectively. sources and major related references can be found in e.g. rené Bouchet, Chronique de Morée: Introduction, traduction et notes (paris, 2005), among other recent translations of GCM. translations into english will be the rule in this essay, showing accents only where most relevant. The versions’ origins and dates, beginning in the 1320s, are not critical for my present arguments, but seem best explicated by michael Jeffreys, “The chronicle of the morea: priority of the Greek version,” Byzantinische Zeitschrift 68 (1975), 304–350. 4 The last is emphasized in a preliminary study of the topics treated here: Jon van Leuven, “Blank spots and grey zones: medieval mapping in southern Greece,” in Eastern Mediterranean Cartographies, eds. Giorgos tolias and Dimitris Loupis (athens 2004), pp. 243–262. 5 antoine Bon, La Morée Franque: Recherches historiques, topographiques et archéologiques sur la Principauté d’Achaïe, 1205–1430 (paris, 1969). a recent attempt by teresa shawcross, The Chronicle of Morea: historiography in Crusader Greece (oxford, 2009), still neglects most of the issues to be discussed here. the phantom baronies of the western mani 43 plinary weaponry must be sharpened with closer study of both existing facts and alternative interpretations. it is nearly 800 years since, according to CM, an army of ‘frankish’ adventurers, after the fall of constantinople in the fourth crusade, appropriated most of Greece’s southern mainland, the peloponnese (morea), establishing a dozen baronies and further holdings that became the basis of the so-called principality of achaea for a couple of centuries—with consequences ranging from castle parapets to population clashes that have echoed down to our day.6 Thus arose a restricted and rather rapid type of incastellamento with feudal elements, essentially for subduing the denizens and supplanting their Byzantine hegemony. comparison to such developments elsewhere is dependent, though, on the historical geography i will address. a summary of widely embraced views about this establishment is as follows. The twelve baronies were listed in CM without clear locations, which have been deduced chiefly from later toponyms. They divided up the peloponnese except for its southern and eastern fringes, which belonged to native folk and Venetian settlers. separated by some 30 kilometres on average, the baronial centres could have functioned as pairs—like a favourite frankish tactic of pincer positioning, soon to be employed in besieging corinth. roughly from north to south, they were pátras and Vostítsa, Kalávryta and chalandrítsa, Ákova and Karýtaina, níkli and Velígosti, Kalamáta and Grítsena, Geráki and passavá. Their nucleus was Kalamáta which, together with a fortress named arcadía on the west coast, formed the hereditary estate of the prince of the realm. other strongholds were added, not least during the reign of william ii Villehardouin, who in about 1250 built three castles surrounding the southern mountains: mistrá (near sparta), Grand magne (máini or máina), and Beaufort (Léftro). These served to control the melings, one of the autonomous slavic peoples who had immigrated in the sixth century. The last two places have been hardest to situate, but are normally sought on the west coast of the mani peninsula, well north of cape matapan (tainaron). They constituted the main frankish presence on this promontory, which had to be passed by 6 on the baronies, see David Jacoby, “The encounter of two societies: western conquerors and Byzantines in the peloponnesus after the fourth crusade,” American Historical Review 78 (1973), 873–906. 44 chapter two pilgrims, purchasers, and paladins sailing to or from the orient, and which therefore eventually thrived on piracy. The Byzantines had previously kept a naval station in the western mani, one of whose earliest fortresses stood on a headland now called tigani. This was long the preferred candidate for Grand magne’s location, another suspect being the adjacent plateau of cabo Grosso. however, the castle has increasingly been argued to lie farther north at the bay of oitylon. as for Beaufort or Léftro, overwhelming opinion puts it on the acropolis at Leuktron, an ancient town outside modern stoupa. These identifications are currently regarded by scholars as more or less firmly supported by CM and other documents, as well as by some inspection of the sites. They also seem plausible in geopolitical terms. mistrá, Grand magne at oitylon, and Léftro at Leuktron, as seen on a map, appear to enclose the taygetos range’s higher central massif, which was inhabited from coast to coast by the melings. moreover, the latter two sites were desirable for other reasons: Grand magne was not far from the barony of passavá, and Léftro lay in a province named Gisterne (Gistérna) with the richest land in the peninsula. Their advantages can be confirmed by the subsequent activities of Byzantines, Venetians, catalans, turks, and other powers in the area. ironically, this strategic scheme endured for barely a dozen years, until 1262 when the franks started to lose the peloponnese as an outcome of constantinople’s return to Greek rule. such is the general perspective; we shall examine a number of its weaknesses and try to remedy them with alternatives. in doing so, greater care must be taken with the reliability of sources and assumptions—particularly about names, places, and dates. many medieval records like CM and later discussions of the peloponnese are the products of people who had little or no demonstrable personal contact with the region, the mani, or scientific method. contemporary texts were commonly vague about locations and directions, suggesting either unfamiliarity with them or trust in readers’ knowledge. while most of the toponyms are ascertainable, the remainder invite as much theorizing about the writers themselves as about the places referred to. on the other hand, toponyms should raise similar questions about their local or external origins, instead of being treated as fixed features of the linguistic and cultural landscape. Like ghosts drifting through walls, they cannot be explained by only one kind of data, and require interdisciplinary openness for compatibility with all kinds. the phantom baronies of the western mani 45 The list of baronies was formulated in 1209 and gradually revised. with more foresight than hindsight, it indicates a secure network of knightly abodes, to be filled in by conquering territory. in the process, superior accommodations and safe fortifications were needed; several baronies explicitly possessed castles from the outset. such building was attributed by CM to a few of them—Ákova and Karýtaina, Geráki and passavá—at a time four decades later, although possibly only because various ruling families were inspired to renovate their premises around 1250, when the southeastern peloponnese accrued to the realm.7 even so, CM is strangely reticent about the baronies until mid-century, and Grítsena disappears from the annals after its foundation. nor were william ii’s three castles given distinct baronial status, in spite of their purportedly equal importance. it is here that we begin to encounter ‘phantoms’ in the power structure which may prove as elusive in position as in nature. The haunted house in the northwest Grítsena is the lone barony with a still disputable location. By consensus, it lay in an area that GCM called Lákkos or (plural) Lákkoi, the upper part of the messenian plain north of Kalamáta. its supposed purpose was defense against invasions, like one that had come from that direction in 1205 and included soldiers from Lákkos. obscurely owned until about 1275, Grítsena then seemingly devolved for a few decades upon a family who had lost the northerly barony of Kalávryta to the Greeks. all this is based on a maze of allusions in texts, and strains credibility. The plain was well protected and presumably exploited by other baronies, and further castles were built close by towards the end of the century.8 no mention was made of Grítsena in the many accounts of movements through the area, even when the only incursion across the adjacent mountains during those years was repelled—in 1264 at the battle of makry plagi, a pass just north of Grítsena’s alleged domain. Thus, rather than considering its necessity 7 for such changes, see Bon, La Morée Franque, pp. 82–84. among other indications during the intervening years, Ákova and passavá created a marital bond implying that they already had substantial estates. 8 see martine Breuillot, Châteaux Oubliés de la Messénie Médiévale (paris, 2005). Like everyone else, the author leaves the specific location of Grítsena undecided. 46 chapter two or sufficiency, scholars have transformed the barony into a shadow looming over the plain, whose sole visible role would have been ultimately to provide a title for two members of the tournay family. a happier hypothesis was adumbrated already by william Leake in 1846. he put Grítsena somewhat south of Kalamáta, guarding the central taygetos mountains. one of his reasons was that slavs lurked on the opposite eastern side, which had earned the range an almost exclusive designation in CM, “the zygós of the melings” (a ‘yoke’ connoting both unity and adversity). The other was that the term lákkoi, meaning hollow places, suited the deep gorges or glens that characterize the northwest mani. Leake suggested that the barony was ceded to the Greeks in 1262 with the rest of Laconia; he was aware that this area traditionally belonged to Laconia—as did the intervening mountains—and that Grítsena had even been identified with sparta in the chronicle of ‘Dorotheos’, a narrative filtered through two centuries from GCM.9 although i believe it to be correct, Leake’s view was also inadequate in several ways. first, he recognized a crucial point without pursuing it far: the location of frankish strongholds in the mani depends on where the slavs are assumed to have lived, and this is a tricky topic as will be seen. second, he oversimplified frankish strategy and missed an additional argument for his placement of Grítsena: the need to stop any enemy from the south, not just melings from the east. william ii’s castles in the peninsula arose a generation later, and could easily be circumvented to attack Kalamáta. had this happened in 1264 when the franks were at makry plagi, they might have lost a two-front war. perhaps precisely due to Grítsena, it was not achieved until 1292, and then only by a band of sly slavic neighbors. Third, Leake did not seek confirmation of Grítsena’s presence in archaeological or toponymic data. while the former are too limited even today in the mani, the latter can be adduced from diverse sources, and names similar to ‘Grítsena’ may attest its diffusion in the same area.10 9 william martin Leake, Peloponnesiaca: A supplement to Travels in the Morea (London, 1846), pp. 138, 142, 153, and 162. he was not, however, persuaded by a confusion of ‘Lákkoi’ with Laconia, as some other scholars have been, e.g. o. markl, Ortsnamen Griechenlands in ‘fränkischer’ Zeit (Graz-cologne, 1966), pp. 34 and 40. 10 a toponym of form ‘Grist-/crost-’ occurs southeast of Kalamáta on portolans and later maps, from the carte pisane (c. 1250) to e.g. camocio (1569); see Giorgos tolias, The Greek Portolan Charts (athens, 1999), with references. an instance from subsequent population records is ‘Griciagni’: e.g. Kostas Komis, Plithismos kai oiki- the phantom baronies of the western mani 47 These issues should have preoccupied antoine Bon, the baronies’ master ghost-buster, who devoted thousands of words to Grítsena and Lákkos/Lákkoi. instead, he spurned any connection between them and the mani, relying on later documents to support the above consensus. his basic objection was that the franks felt no need to subjugate the mani until william ii’s day.11 it was a dubious claim, at least since slavs from the zygós had taken part in the invasion of 1205, and defense of the realm took high priority from the start (FCM 130, GCM 1975). self-protection did not require subjugation in an era before ‘world policemen’, even if a regional one first appeared in william ii. neither he nor his predecessors spent much time at Kalamáta, but the reason was not that they could dispense with it. They were confident that no foe could approach it, despite the relatively vulnerable peninsula to its south. Bon and Leake shared another oversight that has pervaded the literature as a whole: Lákkos and Lákkoi are assumed to be synonymous. This is convenient, but catastrophic for the problem at hand.12 There is no doubt that Lákkos meant a place near makry plagi, because ACM says that the franks gathered at ‘los Lacos’ before the battle. But the aragonese translator was treating the singular as a plural, just as he did the reverse with skortá, a domain still farther north; possibly for the same reason, his version of the barony list located ‘Gresena’ somewhere between Kalamáta and skortá. The real plural Lákkoi occurs only with Grítsena in the GCM version—the corresponding part of smoi tis Manis, 15–19ou aionas (Yannina, 1995), pp. 290–291. cf. ‘Koúrtzeni’, a monastery between Kalamáta and sparta: p. christathakis, To monastiri tis Kourtzenis sto Lada Messenias (Kalamata, 1989); it also owned lands with designations (e.g. ‘i lachída sti Loúka’) resembling the GCM passage about Grítsena (see below), although the place itself was inhospitable for a castle, as experienced even by J. Buchon, La Grèce Continentale et la Morée (paris, 1843), p. 439. 11 Bon, La Morée Franque, pp. 71–72 and 420–421. he stressed the irrelevance of a village Lákkos in the area favoured by Leake. This was adduced by anthony Kriesis, “on the castles of Zarnata and Kelefa,” Byzantinische Zeitschrift 56 (1963), 308–316, who vaguely agreed with Leake but did not pursue the issue of Grítsena. The village’s name actually refers to a huge cave beneath it; such limestone faults are as characteristic lákkoi as the gorges in the northwest mani. 12 see e.g. Jean Longnon and peter topping, Documents sur le régime des terres dans la Principauté de Morée au XIV e siècle (paris, 1969), p. 243, who thus equate several different places, although in the right area—to be identified below—which their own archives show to have been initially called Lákkos. This is a less common type of toponym than has been alleged; see e.g. D. Georgacas and w. mcDonald, “place names of southwest peloponnesus,” Peloponnesiaka 6 (1963–68), passim. 48 chapter two FCM being omitted, and that of ICM contracted to ‘Laco Grisco’. hence, there are no textual grounds for denying that Lákkoi was not only a quite separate place, but also a sizeable one, since GCM calls it a ‘district’ with a word unique in the list, used elsewhere for large territories. such a loose definition of the barony would not be surprising in the northwest mani, as its terrain is bewilderingly rugged and the franks initially had no knowledge of it to delimit the need for defense. nevertheless, Lákkos has been misleadingly equated with the upper messenian plain, which had another name in CM—the ‘Vale of Kalámi’—and could hardly be described as a lákkos.13 The two have been thought identical because of a universal failure to distinguish between ‘Lacos’, where the franks gathered in 1264, and their route from there to makry plagi. according to FCM and GCM, they went through Kalámi, but had gathered in an area with two campsites, Koprinítsa and a well-watered village called moundrá. These names long fooled scholars into tracing the army’s movements farther north, and Bon himself had partly corrected the error when the answer emerged. The area was a basin at the northwest end of the plain, together with a ridge on its west.14 still called Lákkos by turkophiles like evliya celebi in 1668 and Bernard Brue in 1715, it had become known as the Bogáz—again meaning a gorge—when Leake crossed its “small plain” in 1805, as did many later foreigners.15 in sum, a thorough review of the evidence shows that the singular Lákkos correlates as perfectly with geography as does the plural Lákkoi in the northwest mani. 13 The kalámi (cane) still grows along the rivers in the plain. a new discussion of these names upholds the equation, as well as the confusion with Lákkoi: Breuillot, Châteaux Oubliés, pp. 72–76. 14 Bon, La Morée Franque, pp. 357–359; his identification of Koprinítsa was conjectural, and he wrongly compared a spring there with one in GCM, which actually placed it at the other campsite, where it still exists at mándra. The same mistake was made by Longnon and topping, Documents, p. 243. The full answer was published at the same time by p. B. papadopoulos, Frangoi, Enetoi kai Tourkoi stin Peloponnisou 1204–1821 (athens, 1969), pp. 60–61. 15 william martin Leake, Travels in the Morea, 3 vols. (London, 1830), 1:482–483; cf. Bon, La Morée Franque, p. 440, n. 4. The basin, at most 10 km long nw-se and 7 km wide e-w, may have been named not for its shape but after one of the river gorges on its west (Langada) and east (isari) sides. it corresponds to ancient stenyklaros, ‘narrow land’, the northern half of the upper messenian plain: mattias natan Valmin, La Messénie Ancienne (Lund, 1930), p. 84. the phantom baronies of the western mani 49 more can be learned about Grítsena by examining the passages in question.16 GCM 1941–45 says: next was likewise listed sir hugues de Léle to have eight knights’ fees in Vostítsa; he dropped his surname, he was named de cherpiní. to sir Luc were given only four fees, to hold the district of the Lákkoi of Grítsena. FCM 126 has a lacuna embracing nearly all of “Léle . . . Luc . . . Lákkoi,” as if averse to alliteration, and says merely: “sir hugues de . . ., the barony of Grite with four fees.” The word ‘Grite’ is unclear in the manuscript (now in Brussels), but obviously corresponded to Grítsena, and agrees with the known fact that FCM was an abridgment of the original Chronicle. modern editors surmised that it meant ‘Grite’ was synonymous with Vostítsa, a remote barony on the north coast, and that FCM had ignored Grítsena. This was refuted by Bon, although he added a new distraction. The name Léle reminded him of helos, a place in southern Laconia, which he therefore proposed that the baron of Vostítsa had also owned and the author of GCM had mistaken for a surname. Yet the place appears consistently as héleos in the oldest GCM manuscript, and distortion of it seems even less probable than a man who held lands at opposite ends of the peloponnese. however, what prompted Bon was a related passage in ACM 119–120: and to sir Guy were given as a barony, on the coasts of Lakedaimon and the gulf of corinth, twelve fees, and on the coast of Lakedaimon he built a castle, called Lello. and soon sir Guy died and left a son named sir hugo, who having been born in a village called cherpini was named sir hugo cherpini, and who built the castle of Vostítsa. Bon inferred that hugues had inherited the divided barony and a surname like cherpini from his father. But this leaves further issues open. why did the castle of Lello not occur later in CM, despite recorded visits to héleos? why did this barony get eight fees in GCM and twelve in ACM? moreover, all these passages are bracketed by mention of the baronies of Kalávryta and passavá, so why did ACM shift its mention 16 These translations are mine, but that of GCM borrows heavily from harold Lurier, Crusaders as Conquerors: The Chronicle of Morea (new York, 1964), p. 127. 50 chapter two of Grítsena to a later point in the narrative? and who was sir Luc, the ethereal baron of the northwest mani yearning for a surname?17 a sensible scenario would be the following. Guy de cherpini was expected to receive a barony with twelve fees, but he died and left two heirs, at least one being a son. They founded a castle on a coastal island in Laconia, thus acquiring the epithet ‘de l’isle’.18 Before long, hugues departed with a right to eight of the fees and settled at Vostítsa, disavowing his epithet. The other was Luc, who kept the remaining four fees and became the baron of Grítsena. The listing of hugues first, and with more fees, explains why GCM did not repeat his surname(s) for Luc, and why it credited the latter with ‘only’ four fees.19 That they were relatives is supported by ICM, which calls them Ugon de Lels and Luca de serpi—names conventionally blamed on wrong copying of GCM, as is a similar pair in the chronicle of Dorotheos. in ACM, Luc was identified uniquely as just ‘a knight’ and Gresena as ‘the knighthood or barony’, signs of a diffidence that may account for its shift in the list, if not also for the lacuna in FCM. The castle of Lello, too, can be traced through an interdisciplinary approach including the physical evidence. one of the very few coastal isles in the peloponnese with fortifications of likely medieval date is in the northwest mani, off the little plain of Kardamyli. it held a small castle, later wrecked by the turks, with a monastery and a variable name.20 if this was where Luc de l’isle began his baronial career, he would surely have conquered better quarters on the mainland. in particular, atop the prominent acropolis a mile away, fortified since the homeric age, he could survey most of the peninsula’s west coast, look down on its largest river, and guard the oldest route across the peaks 17 Luc or Lucas has been the accepted transcription of GCM ‘Loúka’. rare among the franks, such an appellation would have suited a gasmoule, with a Greek mother fond of saint Luke—if not a pun on ‘Lákkoi’ for want of information. But the surname is what interests me below. 18 J. a. c. Buchon, Chroniques Étrangères (paris, 1840), p. xxi, the first editor of GCM, read ‘Léle’ as ‘l’Île’; others have preferred ‘Lille’. The surname ‘de l’isle’ was soon attested not far away: Bon, La Morée Franque, p. 115. 19 Grants of at most one-third of a holding were also the rule in ordinary transfers of land: Jacoby, “The encounter of two societies,” 886. 20 for this and other local features, see Dikaios Vayiakakos, “Kardamon—Kardamyli—Kardamyla,” Athina 81 (1995), 173–240, esp. 203–227. The isle has been recorded or mapped as prastóz (1600s), La madóna (1700s), panagía (1800s), and merópi (1900s). the phantom baronies of the western mani 51 of taygetos.21 he might not have faced much traffic or conflict, since CM mentions movement only around those mountains, never through them. Yet the overlying ridge offered an ideal view of the foothills and gorges that gave rise to the district’s label Lákkoi. a baronial centre near Kardamyli would have been useful to the principality’s pioneers in several respects, but whether it had to erect a substantial castle is unknowable at present, and the frequent scholarly references to a fortress of Grítsena are hypothetical. so is the barony’s longevity, although it need not have declined into a homeless title as soon as 1262, when Leake thought it lapsed. its supposed later owners, the tournay family, continued to uphold its raison d’être, if we can judge by the sudden appearance in FCM of Jean de tournay, “the lord of Grite” and the “finest knight in the morea,” to help save Kalamáta from both slavs and catalans in 1292. conceivably Jean also had a more personal commitment to Grítsena. while FCM says he was married, it reveals that he was tempted to court a pirate’s daughter— and four years later, according to ACM, he wedded a descendant of hugues, the putative relative of Luc de l’isle. The latter’s progeny, in turn, may have moved to a castle named Lille or Lisle which enters FCM at that time, slightly west of Kalamáta.22 all these are details which, once more, have lacked a coherent interpretation, granted that mine is tentative. an arguable legacy of Grítsena is the name of a village, Goúrnitsa, on the ridge behind the Kardamyli acropolis. such toponyms are often thought to be slavic, but this one has no roots ruling out the connection.23 it might have arisen from either the barony’s name or some attribute thereof—for instance, the tournay family. Their surname (‘Durná’ in GCM, or ‘toúrna’ in the turin manuscript), spelled ‘Gurna’ in ICM, could have changed to ‘Goúrna’ in popular usage, 21 see e.g. Yannis pikoulas, I notia Megalopolitiki chora (athens, 1988), p. 222; Valmin, La Messénie Ancienne, pp. 50, 195, and 199. This route also lay midway between its main alternatives, through Kampos and milia, allowing joint control. on local topography, see peter hartleb, Die Messenische Mani (Berlin, 1989). 22 This place was near no island, despite conjectures that its name meant one formed by local waters: e.g. Leake, Travels, 1:365; Bon, La Morée Franque, pp. 410–411; Breuillot, Châteaux Oubliés, p. 111. significantly, no such name occurred in GCM. 23 see e.g. max Vasmer, Die Slaven in Griechenland (Berlin, 1941), p. 167. on the area’s etymology and history, see Dikaios Vayiakakos, “i ardouvista—androuvista— megali chora tis exo manis,” Lakonikai Spoudai 13 (1996), 5–172. 52 chapter two yielding ‘Goúrnitsa’. as it happens, a goúrna is a trough or reservoir, analogous to the local lákkoi. indeed, if the name’s evolution included a standard diminutive ‘Gournítsa’, the last word in security would be at hand: the place became known as ‘Little tournay’. as a sidelight on this eminence grise of a barony, it is noteworthy that ‘Grítsena’ may not be quite the right toponym. The form in GCM 1945 is a genitive plural and, while not unparalleled for a single place (as in GCM 4701 for Kréstena), could owe to a term like ‘the Gritsénoi’ for people associated with the place (as in equally old towns like Zoúrtsa of the Zourtsánoi). a home such as ‘Grítsa’ might then be implied, also matching family names recorded not far away in space or time. which name arose first would have a bearing on the question of whether a barony was seen as colonizing a tract of land or, for example, as confiscating an estate from native landholders.24 The specter of the robber barons regardless of what else can be gleaned about Grítsena, there are other objections to the preceding analysis and its location of that barony in the mani given the conventional perspective outlined previously. The northwest mani was allegedly dominated by slavs throughout the frankish period, despite plenty of Greek church activity. how would a foreign force, not big enough to be documented as such, have established a base like Grítsena in defiance of the melings, even after their defeat in 1205? while opposed to barons of the crusader kind, they had wealthy leaders according to GCM. notorious tax-dodgers already three centuries earlier as described by the Byzantine emperor constantine Vii, they were the snatchers of the castle at Kalamáta in 1292, and “murderers and robbers” in Dorotheos’ subsequent chronicle: who dared to provoke them before the advent of the ambitious william ii? and if Grítsena still held them in check at Kardamyli, why should he have built the castle of Léftro to control them at Leuktron, only five miles down the coast? for such family names, see Komis, Plithismos kai oikismoi tis Manis, pp. 290– 291. on landholders see Jean Longnon, “The frankish states in Greece, 1204–1311,” in A History of the Crusades, eds. r. L. wolff and h. w. hazard (philadelphia, 1963), 2:235–274, esp. 237. 24 the phantom baronies of the western mani 53 one barony’s resurrection thus seems to make spooks of these longaccepted spheres of influence. Their own credentials ought therefore to be inspected, starting with the slavs. much has been written about slavic settlement in the southern peloponnese, but not many aspects are relevant here. The most striking is that we have insignificant evidence for slavs in the northwest mani prior to the last years of the thirteenth century. no clear archaeological traces of them exist, only a few personal names in church inscriptions can be connected with them, and toponyms of debatably slavic origin are impossible to date so exactly.25 The bulk of that century is not illuminated by older finds, which may attest slavs who were to be driven out by the franks—even if their descendants returned, as the ancient messenians did in the wake of the spartans who exiled them. artefacts outside a chronological window that was perhaps a mere dozen years in a case like Léftro (1250–62) readily succumb to the ‘art of facts’. all sources, including constantine Vii as we shall see, are compatible with Leake’s guess: the melings lived essentially on the east side of the taygetos range, until the zenith of frankish rule. nothing is heard of their relation to the principality of achaea before william ii built mistrá at the outlet of a major gorge from the central zygós.26 GCM then stresses their economic dependence on access to the plains, obviously meaning the eastern lowlands. once the tide turned in 1262, the melings, as allies of the Byzantines, must have been encouraged to expand westward across the mountains, for three main reasons: the Despotate of the morea was developing in Laconia, the northwest mani was thinly peopled, and pressure had to be put on the frankish heartland of Kalamáta.27 subsequently ever more melings in the area 25 This has not been appreciated by advocates of the theories criticized here; see e.g. michalis Kordosis, “i slaviki epoikisi stin peloponniso,” Dodoni 10 (1981), 381– 429, esp. 414–419. immigration from the east is claimed as early as eleventh-century inscriptions at milea, just west of a pass across the mountains; but such testimony is sporadic, cryptic, and normally on portable items, obscuring their origins or the directions in which people were moving. 26 on its topography, see Leake, Travels, 1:128–129, 3:2–5; idem, Peloponnesiaka, pp. 163–164; a. philippson, Die griechischen Landschaften: Der Peloponnes, vol. 3, teil 2 (frankfurt, 1959), pp. 426ff. mistrá is in fact the only place immediately linked by CM with the term ‘zygós’, yet scholars persist in asserting that the zygós was chiefly or solely the west side of taygetos, with a view to the later Zygos mentioned below. 27 This movement was accompanied, if not preceded, by a revival of Greek religious influence in the northwest mani: nikolaos Drandakis, “paratiriseis stis toichografies tou 13ou aiona pou sozontai sti mani,” in 17th International Byzantine Congress (new York, 1986), pp. 683–712. 54 chapter two are to be expected—though not necessarily enough to make a quick end of the franks at, say, Grítsena. in time, some portion of the western taygetos was sure to be adopted by melings as a new home with old nomenclature, and the district mostly south of ancient Leuktron did acquire the nostalgic name Zygos. This has been projected back into the frankish period on an assumption that melings always lived there, but the slow course of their arrival speaks for itself. The primary testimony of slavic infiltration is a Venetian report in 1278 of recent commercial robberies.28 two occurred at a “castle in sclavonian parts,” which is believed to mean the vicinity of Leuktron— because one concerned a cargo from coron, across the gulf, to the village of Kastania behind Leuktron. The castle had a Byzantine-appointed governor, who apparently collaborated with thieves from arduvista (modern exochori), a village in the hills behind Kardamyli. Their chieftain, michalis spanos, also robbed some merchants of a ship, while he was not said to be either a pirate or a slav. These attributes, however, belonged to similarly named men in various sources, and have led to an impression that the same man and/or family was involved, holding sway over the northwest coast for more than a century.29 another victim was a trader going to ‘cisterna’, robbed by its Byzantine governor, mesopotamitis, who was accompanying him to Zarnata. This is a well-known fortress in a bowl-shaped valley (at modern Kampos), and should suffice for identifying the ‘cistern’ in question, not far north of arduvista. The governor’s name aptly signified a place ‘between rivers’, for the valley is bounded by the gorges of the Kardamyli and sandava torrents. scary as the report sounds, and it covers the entire aegean for about eight years, only a few percent of its hundreds of incidents involved the mani peninsula. traders were so safe, if not stupid, that the same Venetian let himself be robbed twice at the aforementioned castle. at least in the northwest, ethnic diversification was seemingly creating a balance of power between robber barons; and their territorial distribution can account for a corresponding one (the so-called kapetánoi) 28 Gareth morgan, “The Venetian claims commission of 1278,” Byzantinische Zeitschrift 69 (1976), 411–438. 29 see e.g. Bon, La Morée Franque, pp. 505–506; Vayiakakos, “ardouvista,” 11–16. for further perspective on the name, see h. eideneier, ed., Spanos: eine byzantinische Satire in der Form einer Parodie (Berlin, 1977). the phantom baronies of the western mani 55 which was to enter recorded history four centuries later.30 The balance doubtless extended temporarily to any franks lingering there, as nothing was said of enclaves like Grítsena—whose presence near Kardamyli and its harbour may nonetheless be reflected in the silence about any local outlet to the sea for the presumed pirates of arduvista. however, a conflict does exist between these fleshed-out phantoms and their counterparts farther south. first, some of the above names have parallels in CM. according to consensus, as we have partly seen, the castle of Beaufort (Léftro) was built in 1250 at Leuktron, in the province of Gisterne (Gistérna). moreover, Gistérna had slavic inhabitants in 1262 (GCM 4592), and Gisterne had a slavic ruler named spany in 1296 (FCM 823). hence, scholars have assumed the following equations: Beaufort became the Venetian report’s castle—which it called Belforte or Belfonte—while Gistérna was cisterna and spany was spanos. secondly, the same province in 1262 was purportedly remembered as Kinstérna by the Byzantine historian pachymeres, about 1300, who depicted it as “very elongated and full of goods,” suiting the littoral environment of Leuktron. apart from the quartercentury gaps in this sequence of building, robbing, and writing, the references seem reassuringly consistent. Unfortunately, the details disagree. The description of Kinstérna was not demonstrably connected with the mani, is not unique to Leuktron, and does not recognizably correspond to the district identified above as cisterna. neither does it fit Gistérna, which GCM 3035 specified the castle of Léftro as being ‘near’, not within. Gistérna must therefore have been a relatively limited area or a distinct site (barring a useless statement like ‘troy is near Greece’); in fact, FCM 823 represents Gisterne as a castle. what pachymeres mentioned was “the whole province around Kinstérna,” not Kinstérna itself—and no castle nearby, even though he was writing about castles that were ceded to the Byzantines in 1262. Besides, a castle at Leuktron is unlikely to have been specified as ‘near’ cisterna, since the two were separated by the district of arduvista. nor was spanos, the strongman of arduvista, said to wield any power over cisterna or “sclavonian parts.” if he was a slav, he must have been a meling, but then he was not spany, for the melings’ domain did not include Gistérna (GCM 4592). 30 see e.g. Leake, Travels, 1:315; pouqueville, Voyage, 5:584–599. 56 chapter two in other words, the sources clash if the assumed equations are taken seriously. The contradictions might be avoidable by postulating an appropriate process of change in borders, castles and owners during the second half of the century. however, this would complicate matters, for instance by imagining that cisterna passed back and forth between slavic and Greek rule. The literature already abounds in conjecture, ranging from an idea that the whole province was named for a cistern on the acropolis at Leuktron, to its implications for the theme organization of the Byzantine empire.31 a simpler, and occasionally voiced, alternative is that such names were attached to more than one place. conflation of data can never be so simple when the resultant paradoxes are ignored, just as the preceding case of Lákkos showed. Thus, the physical analogy of certain localities with cisterns will be encountered again below. pachymeres’ description is attributable to his political interests and unfamiliarity with the mani, but it could also fit a different stretch of the west coast to be examined later. Likewise spanós, meaning ‘needy’ or ‘beardless’ in Greek, was a widespread appellation. castles called Beaufort or Belforte were not uncommon among both the crusaders and Venetians—such as a Belforte built in 1274 at the head of the adriatic—and they had Greek equivalents, as we shall also see. The franks themselves built another Beaufort in 1296, in the mountains north of Kalamáta. FCM says that its garrison came partly from Gisterne, whose ruler spany sent a ship with 200 soldiers. This peculiar pact could have been negotiated through franks still holding out at the nearby castle of Beaufort (Léftro). The shipment suggests, too, that Gisterne had more organized manpower, and lay somewhere farther down the coast, than the northwest mani, which was well within marching distance of those mountains. moreover, it can scarcely have been the same slavs who attacked Kalamáta in 1292 and assisted the franks four years later, since the former’s names and home are known to have differed. in any event, the castle of Belforte in our first robbery report may not have been at Leuktron, for no good harbour exists there to land a cargo. possibly the castle was at one of 31 e.g. nikos nikoloudes, “The Theme of Kinsterna,” in Porphyrogenita: essays on the history and literature of Byzantium and the Latin East in honour of Julian Chrysostomides, eds. J. chrysostomides and charalambos Dendrinos (London, 2003), pp. 85–89. see Jon van Leuven, “The theme system,” in Encyclopedia of Greece and the Hellenic Tradition, ed. Graham speake (London, 2000), 2:1614. the phantom baronies of the western mani 57 the ports farther north, with a similar name such as elephero (from eléfthero, ‘free’?) that can be found on maps by the 1600s.32 if a castle did occupy the acropolis at Leuktron during the late 1200s, the foregoing circumstances allow a local Greek or slavic origin. The ruins on top display techniques too ordinary to trace, and indicate a fort that was little more than a pile of pebbles, in comparison to frankish ones it might have imitated. its single gateway, inner keep—or perhaps a mere platform—and modest cistern were scant insurance against a siege. The walls, only 2–3 feet thick, are of small uncut stones and cooking-pot sherds with some loopholes (later gunports?). nothing like a road up the steep slope is detectable, access to the adjacent sea is poor (Leake could not even land there), and the view of the nearby highlands is restricted. it would have been a scenic, yet silly, investment for the prince of the morea, and almost no student of the spot has ascribed it to him.33 This, however, has not prevented the persuasion that it was Beaufort (Léftro), the culmination of william ii’s contributions to medieval military architecture as mentioned above. arguments for the latter theory commenced when CM was rediscovered, with a fact that has remained most compelling: the name Léftro coincides phonetically with Leuktron, suggesting continuity. But this can be dismissed, for GCM does not say that the castle adopted an older name, in contrast to what it says about mistrá and implies about maíni. indeed, no proof exists that ‘Leuktron’ or other toponyms survived in the northwest mani through the earlier middle ages. They reappeared slowly after 1300, notably in the documents of italians who dealt with the area, and could have been introduced by either learned Greeks or the ever more westerners who were acquainted with ancient sources.34 These included the geographies of strabo and ptolemy, e.g. christos Zacharakis, A Catalogue of printed maps of Greece, 1477–1800, 2nd ed. (athens, 1992), no. 2072. 33 exceptions are Bon, La Morée Franque, pp. 504; panayiotis Katsafados, Ta Kastra tis Mainis (athens, 1992), pp. 429–430. The ascription cannot be traced before early nineteenth-century impressions (e.g. Buchon, stackelberg). several observers have called the fort Venetian, without any evidence. a pre-frankish Byzantine origin is entertained by Kristian molin, Unknown Crusader Castles (London, 2001), pp. 217–9. one may also wonder why william ii next built a wooden castle near coron to harass the Venetians if he had a base at Leuktron, just across the bay: marin sanudo, Istoria di Romania, trans. e. papadopoulou (athens, 2000), p. 113. 34 cf. van Leuven, “Blank spots” (above, n. 4), esp. pp. 249–250. such people were numerous already in the milieu that produced CM: compare e.g. n. G. wilson, Scholars 32 58 chapter two whose spelling ‘Leuktron’ was preferred to ‘Leuktra’—which should have survived at the site if any name did, as pausanias recorded it last there. The Chronicle of Morea exhibits much less erudition, or concern with hellenization, among both the authors and protagonists. at best, the castle might have been dubbed ‘Léftro’ for its ‘bright’ stonework, by someone aware of a classical Greek word that also underlay the name ‘Leuktron’; and ‘Beaufort’ was a visual term, hardly an aural mimicking of ‘Léftro’ as Bon declared. Likewise, only classicizing maps showed such a name until the 1600s. in any case, the name was not decisive, or Léftro’s location at Leuktron agreed upon, until well into the twentieth century. interpretation of CM gradually convinced scholars that Léftro simply would not fit anywhere except on the acropolis. a basic reason was the statement in FCM (207, 218) that william ii built the castles of mistrá, Grand magne and Beaufort (Léftro) “around the mountains of the slavs to better contain them,” and in GCM (3040) that he thereby “subdued the slavic lands and had them under his will.” assuming that the mountains involved were the central taygetos, a role for the adjacent site of Leuktron inevitably won approval. Yet this is fallacious. according to both FCM and GCM, william subdued the melings completely by building only the first two castles. in GCM, where a lacuna exists in FCM, he built mistrá even before he considered subduing the melings, while ICM says he built it as a ‘brake’ on them. contrarily, ACM maintains that he wanted to possess the entire peloponnese from the start, including “the mountain of the slavs,” and that he subdued them before building any of the castles, the first two being mere betterments. as for Beaufort (Léftro), we shall explore its role in GCM below. But FCM does not say why it was added, and ICM mentions it with no name. ACM seemingly appends it to a list of strongholds that william built next, with a name ‘Loriot’ translating ‘Beau’ into Greek oraío. These incongruities alone show, on the one hand, that the aim of subduing the melings was neither necessary nor sufficient to account for william’s activities in the mani. in addition, the castles in question did not belong to separate baronies, whereas even his predecessors might be expected to have filled the peninsula with fiefs if it was so of Byzantium (London, 1983), pp. 229–272, with Jeffreys, “chronicle” (above, n. 3), 345–349. The name of ancient Leuktra itself was attributed to external influence by legends: pausanias 3.26.4–6. on place-name discontinuity, see a. Bon, Le Péloponnèse byzantin (paris, 1951), pp. 24–61. the phantom baronies of the western mani 59 thick with thieves. on the other hand, william’s own strategic capabilities have evidently been exaggerated. he himself did not enhance them by sailing away in 1249 to join the seventh crusade, or by his military adventures in the late 1250s. his initiative is slighted by GCM in two respects. first, subduing the melings did not occur to him until urged by his council and the people of mistrá. secondly, he got the idea of building Léftro from some melings, who pointed out that he could thus “have the whole zygós under his will.” he only believed them and gave the order; they built and named the castle; then he walked about, the author scoffs, “as if he . . . was lord of it all.” But this was just the beginning of the tangle of textual allusions claimed to favour the placement of Léftro at Leuktron. They were further complicated not least by the connection of Léftro with Grand magne, whose unknown location had to be determined simultaneously for coherence, and which interested historians more anyway. The related literature is highly suppositional and seldom discussed, yet underlies present opinion about the castles. its logic typically focuses on three statements. in GCM the castle of Léftro was built “near Gistérna.” pachymeres described “the whole province around Kinstérna.” and the Byzantine polymath Gregoras, about 1350, counted “máini around Leuktra” among the finest cities in the peloponnese. comparing these phrases, one could loosely infer that máini was near, if not equivalent to, the “province of Kinstérna.” taking for granted that Léftro lay at Leuktron, the consequence was that Grand magne should be in the south, not far from the lower end of the western Zygos. here seemed a final solution to how william ii created a ghetto for the melings. slavs were never welcome in Greece, and the emperor Basil the Bulgar-slayer had gained an implicit successor in Bill the meling-container. while the exact site of máini remained debatable, the extreme south of the peninsula was increasingly ruled out as too distant for control of the Zygos. four ruins in the south are old and imposing enough to have frankish associations: on tigani and cabo Grosso, and turkish fortresses at porto Kayio and oitylon. The fourth is biggest and northernmost, so the quest for Grand magne has tended to stop there. sadly, none of those places has yielded enough, if any, archaeological traces of the franks to prove who is right. it is as though some robber baron took up treasure-hunting to get the last laugh on the logicians.35 35 for the arguments in question, see primarily p. fourikis, “paratirisis eis ta 60 chapter two The flaws in their deductions have already been noted in part. even if all the assumed equations between names are accepted, the notion that the semi-barren western Zygos was more than a post-medieval refuge of melings has yet to be justified. But what Gregoras said next is relevant too: máini “was once called cape tainaron.” how he could have confused the tip of the peninsula with a district well to the north has never been elucidated. whether the far south was as isolated to the franks as to us, and whether Grand magne’s sole purpose was to threaten the melings, are no less dubious. for example, when william built it, he also had transport by sea, including two Venetian ships to stop piracy (FCM 190, 197; GCM 2788). his route to choose a site for Grand magne, probably from mistrá, is ambiguously summarized in GCM: “The prince himself rode out on horseback, while the people of the land advised him, and he passed passavá and journeyed to máini.” This is taken to mean that he rode straight through the main pass to oitylon. But it sounds longer and more leisurely than that, apart from not indicating whether he stopped there. traversing the haunt of the melings in a narrow defile ought to have provoked some confrontation. GCM does not even mention the pass as such, despite having several terms for creepy corridors in the mountains (diábata, kleisoúres, and the dróngos of the melings themselves). The verb for ‘journeyed’ recurs in GCM often with the adverb ‘straightway’, yet not here—and it is used for sea as well as land travel. Bastions in the underworld william’s option was to follow the east coast southward, perhaps shipping his horses as the franks had done previously on rough shores (FCM 112). he would have arrived at the excellent harbour of porto Kayio. The site of its later turkish castle visibly agrees with the summary’s continuation: “There he found an awesome crag on a prom- toponymia ton chronikon tou moreos,” Athina 40 (1928), 26–59; petros Kalonaros, To Chronikon tou Moreos (athens, 1940), esp. pp. 125–129; s. Kougeas, “peri ton meligon,” Pragmateiai tis Akademias Athinon 15:3 (1950), 1–34; Kriesis, “on the castles” (above, n. 11); Bon, La Morée Franque, pp. 504–505; John malcolm wagstaff, “further observations on the location of Grand magne,” Dumbarton Oaks Papers 45 (1991), 141–148. some additional instances will be cited below. the phantom baronies of the western mani 61 ontory.” This phrase’s origin has been disputed, as it also fits the alternative places for Grand magne to varying degree. But in fact it was a virtual copy of the porto Kayio site’s characterization in sailing manuals.36 The word for a ‘crag’ in GCM, normally denoting a cave (spelaion), plausibly owes to the overhang of a cliff, although it could have meant one of the deep caves at porto Kayio.37 confirmation of william’s choice comes from an incident in 1292. The catalan admiral roger de Loria sailed out of the aegean “straight to the cape where lies the castle of Grand magne,” which by then had been surrendered to the Byzantines. he raided its port, and according to his biographer three decades later, the landing occurred at porto Kayio.38 porto Kayio (Quaglio) was already named by italian sailors, but ‘maíni’ had long since arisen for much of the peninsula, and been attached to the Byzantine fortress at tigani. That william selected this name for his castle, as GCM adds, would thus be natural. ACM says that the castle was built “near the sea on the side opposite coron which is called mayna,” followed by mistrá “on the side of the land”— which may be a contrast between the peninsula and the interior, and a verification that Grand magne lay across the mountains on the east side. however, the catalan raid might refer to some port on the west coast, to which the name of porto Kayio was often transplanted by subsequent maps. Yet maina on the west and porto Kayio on the east soon became the rule in cartography, while it was not stabilized by clear records of castles until the turkish ones of the 1500s. exceptions to it reflected growing indiscrimination and confusion about topography in the southern mani, where porto Kayio could 36 see e.g. Bacchisio r. motzo, ed., Il Compasso da Navigare (cagliari, 1947), p. 36, compiled about 1250–65. This does not mention a castle, which may not have been built yet or, if so, considered relevant. also a. Delatte, Les portulans grecs (Liège, 1947), p. 57, by the mid-1400s. This mentions oitylon too, but with small headlands and a castle not on a promontory. piri reis, Kitab-i bahriye, ed. e. Z. okte (istanbul, 1988), pp. 645–658, noted the same landmark at porto Kayio by 1520, and a ruined castle manya at the end of the peninsula. 37 see e.g. Bon, La Morée Franque, p. 422 and pl. 151:1. another cave was explored with awe in 1447 by Cyriac of Ancona: later travels, ed. and trans. edward w. Bodnar and clive foss (cambridge, mass., 2003), pp. 318–321, apparently on the hill above the later castle. 38 FCM 762 and e.g. Buchon, Chroniques, p. 375. This simple correlation has suffered astonishing neglect by scholars, but several (e.g. Leake, miller) have supposed that Grand magne lay at porto Kayio. The greater detail in FCM also suggests that the franks had an abiding interest in the port, as Grand magne merited. 62 chapter two even be given a name like oitylon, and cape matapan mixed up with cabo Grosso.39 at porto Kayio the franks would have completed a chain of four castles spaced out along the east side of the taygetos range, including the contemporary “zygós of the melings”—a veritable wall between their heartland and potential invasions from constantinople. Likewise, when the Byzantines captured william ii and exchanged him in 1262 for the fortresses of monemvasia, máini and mistrá, they would have been outlining a triangular spearhead into the underbelly of Laconia, through which the peloponnese was to be reconquered. These were arguably smarter and simpler strategies than the usually imputed ones focusing on a ring of fortresses around the local slavs. Grand magne’s location may still be left open (as my map concedes), because the likeliest alternative at oitylon is in a larger neighborhood that has not been examined exhaustively. at most we can expect that, if the castle turns up there, it will have amazing implications for the medieval mind. its conjectural site, Kelefa, lies only ten kilometres—a third of the average noted above—from passavá. how the sanguine franks could have been frightened enough by the melings to set two major fortresses so close together, instead of just strengthening passavá, has not been explained. nor does the site resemble a promontory as GCM said, unless seen from directions that william did not arrive from and would have had to fantasize about. it also requires strategic thinking, by both him and the Byzantines who took over the castle, of a calculated kind that was out of character and off the record. all this is demonstrated by a tendentious attempt to find Grand magne by invoking ‘criteria’ that enshrine several of the dogmas i am discussing.40 see e.g. maps in antonis tantoulos, “i entypi chartografisi tis manis (16os–18os ai.), istorikes kai geografikes plirofories,” in Mani: Témoignages sur l’espace et la société. Voyageurs et expéditions scientifiques (XV e–XIX e s.), ed. Yannis saitas, institute for neohellenic research, national hellenic research foundation (athens, 1996), pp. 53–71. Yet Battista agnese’s map in 1554 apparently marked a pre-turkish castle maini at porto Kayio itself, in addition to maina on the west: see Bon, La Morée Franque, pl. 8. 40 wagstaff, “further observations,” 141–148. This oft-cited study also contained diverse errors, e.g. in its treatment of documents (as at p. 143, n. 24, 29, and 36; p. 144, n. 38 and 41; p. 145, n. 52; and p. 147, n. 84), and in confusing Gistérna with Grítsena. it has been elaborated in a British academy lecture by wagstaff (15/2/2007), also wrongly ascribing the Loria raid (see above, n. 38) to a later prince of achaea. 39 the phantom baronies of the western mani 63 additional hints have guided the search on paper. first, the position of Beaufort (Léftro) is thought to be confirmed by texts. FCM 207 states that “another castle was built near the sea toward the west between Kalamáta and Grand magne, and is called Beaufort in french and Lefftro in Greek.” The specification of Léftro in GCM 3035 was to “build a castle on the coast near Gistérna.” taken as complementary descriptions of the same event, these are customary grounds for the theory that Léftro lay at Leuktron, in a province that occupied much of the western littoral down to Grand magne. But they tell us no more than that Léftro was somewhere on the west coast of the mani, as long as we do not know where Grand magne was; for instance, Léftro might be at oitylon if Grand magne was at tigani. and they may tell us less, since the statements need not both be believed. They are so different that one of them could be an amendment by somebody in the production line who professed better knowledge, even if wrongly. in that case, scholarship points to FCM, a work of more pedantic revisionism than geographical expertise.41 among the french writer’s habits was to elevate the frankish prince’s status. as a further illustration, GCM gives the impression that he built those castles at the same time as other baronies like passavá were (re)constructing theirs, whereas FCM stresses repetitively that they did so only afterward, thus subordinating them to william and presenting him as a visionary— not as a myopic manufacturer of strongholds in a district dominated by passavá, which he himself had found peaceful on a recent visit (GCM 2984). The detailed directions quoted above would flatter his planning, yet have very few parallels elsewhere in FCM, and violate its norm of abridgment; the sentence in the manuscript is marred by misspellings, as if even the writer hardly believed it. whether it does contain errors is subject to proof, but it cannot be used as a formula for discovering fortresses. a second excursion brings us back to what preceded the GCM quotation. why did the melings, whose mountainous zygós william 41 see Jeffreys, “chronicle” (above, n. 3), 334–347. FCM must have been particularly vulnerable to external influence by knowledge of ancient Leuktron’s location (cf. n. 34). shawcross, The Chronicle (above, n. 5), 102, claims on linguistic grounds that FCM was “the work of an individual who was thoroughly familiar with the peloponnese, and consequently able to use correctly both Greek and french names in referring to places,” exemplifying with FCM 207 itself; but language is not the same thing as topography. such overinterpretation does not mean, however, that FCM cannot be proved right on other grounds (see below, after n. 46). 64 chapter two had subdued, suggest that he could “have the whole zygós under his will” by building Léftro? and what did they mean by the whole zygós, if not their own? These issues have been neglected, but are resolvable. to begin with, the ‘zygós’ was not simply a range of the melings. initially the term occurred as a plural belonging to them (GCM 1718), although only in the oldest manuscript and not subsequently. This combination was apparently thought obsolete, and similarly the final occurrence of ‘zygós’, for a different area, was omitted from the next manuscript (GCM 7236). otherwise, the author carefully qualified the zygós associated with mistrá as being that of the melings. Yet they, at least, must have continued to distinguish it from a further part of the taygetos, and called both parts the ‘whole zygós’ (like the branches of the alps). analogously, the Byzantine history of (pseudo-)sphrantzes, from 1475–1575, ascribed to the western mani a southern and an ‘outer’ northern zygós, divided at the pass of oitylon.42 such a distinction was inheritable from Byzantine predecessors. according to constantine Vii about 950, the slavs “towards Lakedaimon and helos” comprised two groups who, attracted to the “difficult” peninsula, had “settled upon the flanks of this same mountain, the melings in one part, and in the other part the ezerites.” he is commonly taken to mean the west and east sides of the taygetos range, due to the aforementioned assumption—which he is also regarded as justifying—that the melings later lived on the west side. however, this reasoning is not only circular but geographically and linguistically gratuitous. The emperor in constantinople was facing, and clearly referring to, the east side of the range (he even called it pentadaktylos, the five peaks, a profile visible only from the east), and approximately to its northern and southern halves. evidently the melings were expansive and pushed the ezerites farther south, rather than back to helos, an exposed littoral where only Greeks thrived by CM times. while CM never names the ezerites, it often speaks collectively of slavs, and implies the persistence of a group distinct from the melings (e.g. GCM 4592, 4605, 4662; FCM 333). 42 This ascription is conventionally read with no clear direction or division, as it initially mentions a ‘Leuktron’, which can then be equated with the ancient site; see e.g. Kalonaros, Chronikon, p. 74; Vayiakakos, “ardouvista,” 8 and 58. But otherwise it places Leuktron in the far south of the peninsula, agreeing even at that late date with Gregoras, and with my conclusions below about Léftro. readings of sphrantzes require further correction. the phantom baronies of the western mani 65 ezerites moving down the range would have finally spilled across the hills to the west side, ironically ending up in a refuge to which earlier slavs themselves had pushed Greeks. The emperor went on to describe “the city of maíni . . . on the cape . . . beyond ezerón towards the coast,” as an enclave of Greeks unlike the local inhabitants. seen from his direction, it corresponded to the fortress of tigani—just beyond a small plain, known for its main town of Kíta in post-medieval times.43 here the ezerites could have prospered in another lowland environment, the most fertile area in the southern mani. a catchment for mountain torrents, it once held a lake, although erosion has impoverished it.44 effectively a giant cistern, the Kíta plain can well be identified with Gistérna in GCM. The slavic castle of Gisterne in FCM might have been a complex, surveying the plain from the west, whose crude ruins survive at ano poula, the north half of cabo Grosso.45 Judging by the imperial taxes, the city of maíni was richer than the ezerites, but they in turn were much wealthier than the highland melings—and may still have been so in the thirteenth century. The distinction between northern and southern slavs, in the taygetos 43 Constantine Porphyrogenitus, ‘De administrando imperio’, ed. G. moravcsik, trans. r. Jenkins, rev. ed. (washington, D.c., 1967), pp. 233–237. cf. anna avramea, “Le magne byzantin: problèmes d’histoire et de topographie,” in Eupsychia: Mélanges offerts à Hélène Ahrweiler (paris, 1998), pp. 49–62, with a similar viewpoint, but taking the text to mean that the entire area was ‘maíni’ and Greek. The Kíta plain (almost the size of Lákkos noted above) and its history are best described by Dimitrios Dimitrakos-mesisklis, Oi Nyklianoi, vol. 1 (athens, 1949), which supports the present interpretation in most respects, although clinging to the traditional logic cited in note 35 above. 44 Leake, Travels, 1:287–288; Katsafados, Kastra, pp. 276–277; maps such as Zacharakis, Catalogue, no. 1194. The name ‘ezerites’ itself may mean lake-dwellers: e.g. Vasmer, Die Slaven, pp. 317–18. The Kita area also acquired many built cisterns and derived toponyms: Dikaios Vayiakakos, “peri ta paratainaria tis manis toponymia,” Athina 68 (1965), 169–248, esp. 185–189. its medieval role has been obscured by a notion that Gistérna lay at a place called ‘Kisternes’ with cisterns on cape matapan, whose rejection has abetted the theory that Gistérna was nowhere in the southern mani. 45 see Katsafados, Kastra, pp. 159–170 and 395–462; Kalonaros, Chronikon, p. 126. They and some others think this complex was Grand magne. memories of a castle called Kisternes on or near cabo Grosso were often reported later, e.g. by sir william Gell, Narrative of a Journey in the Morea (London, 1823), pp. 264–265. That the complex was slavic would also explain many similar ruins in the sangias hills which, as i suggest, had been crossed by the ezerites; for a recent tour of these i am grateful to p. Katsafados and J. chapman. Their own crudity opposes an attribution to earlier Greek migrants by e.g. Katsafados, ibid., pp. 34ff. 66 chapter two and tainaron areas respectively, continued at least until the historian chalkokondyles in the late 1400s. This perspective on slavs in the mani, with a second group as potential Greek allies, has a radical impact on the prevalent meling-centred account of the region, and it answers the preceding questions about GCM: the melings, in order to divide william ii’s attention, or possibly to thwart kindred slavs who had become rival robber barons, were advising him to bring the ezerites “under his will” too. where he built the castle of Beaufort (Léftro) “near Gistérna” is thus open to fresh investigation. The most conspicuous candidate, overlooking ano poula across a flat no-man’s-land, stands 1–2 kilometres away at the south end of cabo Grosso. frankish and enemy castles even closer together were paralleled at nauplion only a few years before (GCM 2872–3). named by local legend, the Kastro tis Oraias (castle of the Beautiful one) or Oriókastro (literally Beaufort) has largely been dismantled by the later stone-recycling maniots, like other unpossessed edifices in the area. still, it is recognizable as a substantial fortress in engravings of the 1680s.46 with a fine port below at modern Gerolimena, it is “near the sea toward the west” as FCM says—and “between Kalamáta and Grand magne” if the latter was at, for example, porto Kayio as supported above. These two sites are not much farther apart on the map than those at oitylon and passavá, but the landscape between them is rougher than a pass. Ultimately a nest of pirates and bandits, it was to enjoy the highest murder rate in Greece and the highest population density in the peninsula.47 The franks, while thereby definitely enclosing all “the mountains of the slavs,” might have found the extreme south to be uncontrollable even with a pair of strongholds, especially once Grand magne was e.g. Katsafados, Kastra, pp. 294–299, who thinks these depict ano poula. Bon, La Morée Franque, pp. 503–504, regarded the Kastro tis Oraias itself as Grand magne. many of their arguments are applicable to Léftro instead. This site also corresponds to ancient hippola, marked as a castle on maps like that cited in note 44. it can be recognized, too, as the “companion castle” near old maina seen by a passing pilgrim in 1323: mario esposito, ed., Itinerarium Symonis Semeonis (Dublin, 1960). piri reis, Kitab-i bahriye, pp. 646 and 658, mapped a harbour at Gerolimena named “Vitiruna” recalling “Léftro”—and no site at ancient Leuktron or oitylon. 47 see e.g. V. panagiotopoulos, Plithismos kai oikismoi tis Peloponnisou, 13–18 aionas (athens, 1987), pp. 179ff. That william ii mastered the intervening gorges by building such a pair of castles was cleverly suggested by Jean Bory de saint-Vincent, Relation du voyage de la commission scientifique de Morée (paris, 1837–38), 2:315. 46 the phantom baronies of the western mani 67 ceded to the Byzantines. These scarcely needed to ask for Léftro as well, and perhaps its garrison fled soon afterward. The southwestern littoral could have been what pachymeres called “the whole province around Kinstérna,” which scholars have claimed that he included in the cession, although this depends on how the text is read.48 nothing indicates that Gistérna was ever william’s to cede anyway, and FCM states that the ruler of Gisterne owned “the other castles around” by 1296. however, the Byzantine promotion of Greek immigration to the area must have forced the ezerites to hellenize or to merge with their meling brethren in the north. They may also have suffered in 1335 from an attack at two ports in the western mani—possibly those flanking cabo Grosso—by a turkish fleet that recorded it as the “land of ispen” (from Gisterne and/or spany), which has hitherto been attributed to the older name of spanos in the north. whatever the life span of Beaufort (Léftro), the toponym lasted. combinations with the name of Grand magne, such as ‘Leuktra mainis’ for the principal place in the southern mani, became traditional to both texts and maps for the next four centuries. The Kastro tis Oraias is also a plausible inspiration, through reports from sailors past the cape, for the earliest type of ptolemaic map made in the west, which showed the ancient Leuktron displaced to the south end of the peninsula.49 its value as an isolated fort was perhaps only intended to be symbolic for a foreign prince. today, tottering on the sunset edge of a spectacular precipice, it is a fittingly far-flung emblem of efforts to demythologize and disinter the remnants of frankish nobility throughout the peloponnese. while more remains to be said on these matters, i hope to have established the essential validity of some previous speculation about locations—not only in the northwest mani for Grítsena, but also in the southern mani for Beaufort (Léftro), Gistérna, Grand magne and/ or slavs.50 such proposals by prominent scholars long pointed toward 48 The claim was attractively refuted by hélène ahrweiler, Byzance et la Mer (paris, 1966), p. 355. Yet cf. the ‘taenarian district’ in Leake, Travels, 1:303. The franks’ departure certainly stimulated local Greek church painting: see sharon Gerstel, “art and identity in the medieval morea,” in The Crusades from the Perspective of Byzantium and the Muslim World, eds. angeliki e. Laiou and roy parviz mottahedeh (washington, D.c., 2001), pp. 263–285, esp. 270–273. 49 e.g. Zacharakis, Catalogue, no. 1626 (Bologna, 1477) and its progenitor (c. 1410). This was seemingly a temporary reversal of the influence remarked in note 34 above. 50 notably in the glosses of works by Jean Buchon, John schmitt, and Jean Longnon. 68 chapter two the unorthodox yet unobstructed picture of the medieval mani that has resulted here from coordinating analysis and synthesis of ideas and data. The ‘skeletons in the closet’ of theory should be rehabilitated with circumspection, and our real fallacies renounced, lest the phantom baronies fade like fugitives from injustice. Acknowledgements Being indebted to many people and institutions for discussions, materials and facilities during research on mani history, i take this opportunity to extend them my gratitude and no responsibility for errors. special thanks go to Diana wright and pierre mackay (seattle), patrick Leigh fermor (Kardamyli), Björn forsén (athens), peter mackridge (oxford), philipp Billion (marburg), the Dumbarton oaks staff, the taygetos mountain-man Konstantinos Bourlakos, and my faithful jousting partner John chapman (maniguide) who made the map. the phantom baronies of the western mani 69 map 1: The mani peninsula and its vicinity, c. 1250, including main frankish strongholds and uncertainties discussed in the text, with other places mostly of older or later date. chapter three a seigneurY on the eastern borders oF the KingdoM oF JerusaleM: the TERRE DE SUèTE cédric devais various historical and archaeological works undertaken over the last twenty years have made it possible to understand the latin states of the levant from a new angle. This new understanding has altered the idea of a society in which a handful of westerners, locked up in their cities and their fortresses, dominated a crowd of muslim peasants. although it is true that the franks never represented the majority of the population, they nevertheless formed a considerable part of it, perhaps even a quarter.1 it seems that a large number of them chose to settle in the countryside alongside the local christian communities rather than in the towns, and the several villages of the Kingdom of Jerusalem that have been studied in depth tend to corroborate this idea.2 however, the borders of the Latin states, Oultre Jourdain, and even more so the Terre de Suète (Sawad), have benefited only slightly from this revival of academic interest in Frankish settlement.3 It is quite obvious that 1 Prawer’s estimations give a Frankish settlement of 120,000, maybe 150,000 people, for a total population of 500,000. J. Prawer, Histoire du royaume latin de Jérusalem, 2 vols., C.N.R.S. (Paris, 1970). 2 R. Ellenblum, “Construction Method in the Frankish Rural Settlement,” in The Horns of Hattīn, ed. B. Z. Kedar (London, 1992), pp. 168–189; R. Ellenblum, Frankish rural Settlement in the Latin Kingdom of Jerusalem (Cambridge, 1996). 3 The situation is far from being uniform for the land of Oultre Jourdain. The archaeological work carried out by Prof. vannini has greatly enlightened us on the Frankish presence in the Petra region. Nevertheless, much remains unclear, in particular concerning the northern part of the Seigneury of Oultre Jourdain (the Amman and Madaba area). A. Walmsley, “The Middle Islamic and Crusader Periods,” in Jordan an Archaeological reader, ed. R. B. Adams (London, 2008), pp. 506–507. Several articles discuss the question of the medieval population of Petra. See, for example, G. vannini and C. Tonghini, “Medieval Petra: Settlement of the Crusader and Ayyubid Periods,” American Journal of Archaeology 101/3 (July 1997), 534–5; Ritterhausgesellschaft <Bubikon>, die Kreuzzüge Petra—eine Spurensuche (Bubikon: ritterhausges, 2006). in addition, a team from acor, directed by patricia Bikai, found remains of a crusader village in Beidha, in the petra area. The results are to be published in adaJ. Stephan Schmidt has made a similar discovery in the Wadi farasa, also in petra, along with that of a crusader cemetery with several headstones. 72 chapter three the prestigious classical past of the cities of the decapolis, due to their remarkable state of conservation, has largely overshadowed the medieval past of the area.4 Moreover, the modern geopolitical situation of the ancient Terre de Suète is particularly uncomfortable as it is currently located between israel, Jordan, and syria, which serves to make certain sites virtually inaccessible. compared to the western parts of the latin east, this vast area of the Kingdom of Jerusalem still remains virtually unknown. however, surveys, excavations, and the re-evaluation of sources have gradually improved our knowledge of the particular structures of this lordship, how its borders were organized and by whom. The study of the Terre de Suète promises to provide a better understanding of the frankish presence in these distant dependencies during the twelfth century. The geography of Suète reveals a part of the nature of this seigneury. indeed, whilst the texts define it clearly from the rest of the latin kingdom, its boundaries with the muslim world were blurred and moving. to the south, the Wadi Zarka separated it from the Oultre Jourdain,5 while to the west, the Jordan river and lake tiberias were the borders with galilee and Bessan.6 The eastern and northern boundaries7 were vague and fluctuated throughout the twelfth century following the advances and the withdrawals of the crusaders.8 at its greatest extent, around 1160, Suète covered all the ajlun area (Jebel au’f ), the golan and the hauran as far as dera’a, and the plain of meidan. in spite of these vast dimensions, it never constituted an independent barony and during its entire existence remained a part of galilee.9 The frankish 4 for example, the mamluk parts of the interior of the citadel at Bosra, which were demolished in order to return the theatre to its classical phase. 5 e. Strehlke, Tabulae Ordinis Theutonici (Berlin, 1869), charter 3, p. 3. 6 The Assises du royaume de Jérusalem divide galilee in two: galilee proper being the zone to the east of the Jordan river. a. a. Beugnot, “assises du royaume de Jerusalem, assises de la haute cour,” Recueil des Historiens des Croisades, Lois (paris, 1841), 1:422: “la baronnie de la princee de galilee deit c. chevaliers; la devise: de la terre sa’ le flum Jordain lX; de la terre de la le flum Xl chevaliers.” 7 ibn Jubayr notes that the border between the two countries was marked by a large tree on the golan, known as the ‘oak of the scales’. ibn Jubayr, “Voyages,” Recueil des Historiens des Croisades, Historiens Orientaux (paris, 1884), 3:446. 8 in 1166, the franks lost a fortified cave in the ajlun area. Then, beginning in 1184, Saladin began building Qala’at ar-rabbadh/ajlun. however, this dating is contested by alan Walmsley who believes the castle building started after the defeat at hattīn. See a. Walmsley, “The middle islamic and crusader periods,” pp. 506 and 511. 9 according to tibble, dera’a was independent from galilee, although nothing indicates that the town formed a complete lordship. S. tibble, Monarchy and Lordship in the Latin Kingdom of Jerusalem 1099–1291 (oxford, 1989), p. 51. a seigneury on the eastern borders 73 chronicles often refer to it as the fringes of the kingdom, and in fact, its configuration, its evolving character, and its proximity to the enemy made it a veritable march or borderland. The conquest of Sawad10 began during the winter of 1100 when tancred de hauteville, a norman from italy, began to covet the lands on the eastern shore of the Sea of galilee after receiving tiberias from godefrey of Bouillon11 and conquering Bessan and Samaria. This area was in the hands of an emir who was a vassal of damascus, nicknamed with great scorn, “the fat peasant” by albert d’aix.12 unable to withstand the invasion, the emir initially yielded to the requirements of de hauteville. Then, once tancred withdrew to galilee, he hastened to rescind his agreements and appealed to duqaq, the emir of damascus, to help him fight against the franks. Several campaigns and the support of godefrey were necessary before the Grossus Rusticus would admit to frankish pre-eminence in his lands. in may 1100, the two crusader princes, tancred and godefrey, once again crossed the Jordan river into the Sawad for a major campaign, leading 200 knights and around 1000 soldiers without meeting any resistance on their way. The expedition, which at the outset appeared to have been a simple military parade, turned into a strategic withdrawal when duqaq swooped down upon the rear guard led by tancred and inflicted serious damage.13 humiliated by this defeat, the franks carried out such harsh reprisals in the region that duqaq was forced to accept a truce and the abandonment of a large part of his south-western territories. it was the first stage of a conquest that would continue over the next two decades, which were to be marked by continuous changes, divisions, and re-divisions of the Sawad and the Jebel ajlun. Throughout this phase, which was marked by a relative balance between the two forces, neither of the belligerents was capable of supplanting his adversary.14 it is by this name, the black land, that the eastern chroniclers refer to the Terre de Suète. 11 tibble rightly pointed out that the town of tiberias was not taken by tancred, but by godefrey, who built the fortress. tibble, Monarchy and Lordship, p. 11. 12 albert d’aix, “liber christianae expeditionis,” Recueil des Historiens des Croisades, Historiens Occidentaux (paris, 1879), 4:517: “Hic princeps appellatus est a Gallis Grossus Rusticus prae nimia pinguique corpulentia et vili persona.” 13 albert d’aix, “liber christianae expeditionis,” p. 517. 14 details of these events are reported by deschamps and rey. e. g. rey, “notes sur les territoires possédés par les francs à l’est du lac de tibériade,” Mémoire de la société nationale des antiquaires de France 41 (1880), 86; p. deschamps, La Défense du royaume de Jèrusalem, étude historique, géographique et monumentale, 2 vols. (les châteaux des croisés en terre Sainte) 2 (paris, 1939). 10 74 chapter three in 1105, hugue de Fauquenberg, lord of saint-omer and tancred’s successor as prince of galilee, tried to gain an advantage over the atabeg by building the powerful citadel of al’âl on the golan heights. however, the attempt quickly turned into a catastrophe. tughtekin,15 perfectly conscious of the threat constituted by a fortress established less than two days march from his capital, reacted quickly and on december 24, 1105, seized the stronghold before it was even equipped to withstand a siege. but, judging that his prize was too exposed to a Frankish counterattack, he did not try to keep the castle, preferring instead to destroy it and to remove its garrison into slavery in damascus.16 after this defeat, the crusaders never tried to build castles north of the yarmouk river again. during the followings years, it was a cave-stronghold on the southern side of the yarmouk, the Caves de Suète in the vicinity of ancient abila, that became the epicentre of the bitter rivalry between Jerusalem and damascus for control of this area.17 constantly taken and retaken, it shows the indecisiveness of the franks and Seljuks. This situation lasted until 1118 when tughtekin, having broken the truce once again, was camping on the edges of the yarmouk and learned of the death of Baldwin i. taking advantage of the troubles in Jerusalem, he seized the Caves de Suète one more time, which the eastern chroniclers named el-habis. however, his plans were thwarted by the vigorous reaction of the new king of Jerusalem, Baldwin of Bourg or the ‘king count’, as he was called by the muslims.18 once he had ensured his title, Baldwin entered the lands of Suet and expelled the Syrians from el-habis. Then, pushing his advantage as far as possible, he seized the city of dera’a in the hauran, which was known on the frankish side as ‘the city of Bernard of etampes’.19 This success marked a watershed in the conquest of the Terre de Suète because the atabegs of damascus were never again The atabeg tughtekin took over damascus after the death of duqaq in 1103. it had no less than 200 men. abou’l modaffer, “extraits du miràt ez Zémân,” Recueil des Historiens des Croisades, Historiens Orientaux (paris, 1884), 3:529–530; ibn al athir, “Kamel al-tewarikh,” Recueil des Historiens des Croisades, Historiens Orientaux (paris, 1872), 1:229–230. 17 The fortress-cave changed hands several times during the decade. deschamps, La Défense du royaume de Jèrusalem, pp. 99–116. 18 So named by the eastern writers because he had previously been the count of edessa. 19 it would appear thereafter with the name “civitas de Bernardi de Stampi” in the latin texts. William of tyre, Historia rerum in partibus transmarinis gestarum (turnhoult, 1986), p. 328. 15 16 a seigneury on the eastern borders 75 able to contest the latin domination of this land.20 understanding the benefits to be gained here, Baldwin never ceased consolidating and trying to extend his possessions between the Jordan and the hauran mountains, which henceforth were a veritable march. for the franks as for the Syrians, this area was extremely attractive because of its economic and strategic potential. for the crusaders, these regions were in addition a guaranteed source of precious human support, as they had found when they arrived in Jerusalem, due to the many eastern christian communities settled there.21 The agricultural richness of the country appears in both the archaeological sources and in the treaties determining the division of the harvest of Jebel au’f and Sawad.22 on this subject, the translator of William of tyre notes that it is “un pays seins et délitable, pleinteis de vin, de froument, d’uile et de bonnes pastures à bestes.”23 This division of the agricultural production perhaps led some historians to see only a sort of condominium where christians and muslims would have exerted joint authority over the country. if we can agree with this point of view for the years before 1118/9, for the years following, it is no longer acceptable as the franks enjoyed complete power and organized the country in accordance with the standards of the rest of the latin kingdom. in 1121, following the loss of Jerash, where the Seljuks had tried once more to establish a stronghold in the ancient temple of Zeus, the turks no longer had any castles in the ajlun mountains or Sawad, whereas the crusaders maintained fortresses and garrisons there. Syrian citadels were from now on pushed out eastwards toward Salkhad and Bosra in the hauran, facing dera’a and el-habis. a very clear line of demarcation then existed that 20 When nur al-din took damascus in 1154, the frankish military supremacy would be tested once again, and with the ascension of Saladin, the disproportionate strengths of the forces would be highlighted. n. elisséeff, Nūr ad-dīn, un grand prince musulman de Syrie au temps des Croisades (511–569 h./1118–1174) (damascus, 1967). m. c. lyons and d. e. p. Jackson, Saladin, the Politics of the Holy War (cambridge, 1982). 21 The kings of Jerusalem invited the christians from transjordan to settle freely in their lands in order to repopulate them. 22 The area is called ‘Basan’ in the Bible, which means fertile. The arabic word ‘sawad’ means black earth, and by analogy, fertile land. r. p. abel, Geographie de la Palestine (paris, 1967), 1:275. 23 translator of William of tyre, “les eracles,” recueil des Historiens des Croisades, Historiens Occidentaux (paris, 1844), 1:1105. The “livre des deux jardins” notes that after ferroukh Shah took hold of el-habis in 1182, he captured no less than twenty thousand head of cattle from the enemy. abou’l modaffer, “mirat ez Zéman,” recueil des Historiens des Croisades, Historiens Orientaux (paris, 1884), 3:398. 76 chapter three passed through dera’a and the golan heights, very close to the present day division between syria and israel. it is from this point of view of a moving border that it is necessary to review the latin occupation and use of the Terre de Suète. The crusaders quickly understood the strategic interest of this 500 km long strip stretching from the golan heights to the red Sea.24 direct or indirect control of this 80 to 100 km wide region gave the kingdom a strategic depth of which it had been completely deprived at its beginnings.25 The Terre de Suète and Oultre Jourdain formed a protective glacis, isolating the heart of the realm from its hostile neighbours. With these lordships, the marching time of an army coming from Syria was almost doubled, time that could also be used to rally troops for the defence of the kingdom. Such a rampart sanctified, to use a fashionable term, the western territories, namely of galilee, Samaria, and Judea.26 This opening towards the east offered the franks the possibility of taking advantage of effective natural defences. indeed, the great Syrian desert that separates mesopotamia from the levant, although not impassable for caravans, is a frightening barrier able to weaken the mightiest of armies. But in contrast to Oultre Jourdain, Suète never reached the desert.27 it remained a frontier, forming a true march in the traditional meaning of this term.28 for nearly sixty years, until the collapse of the first Kingdom of Jerusalem, it played the role 24 c. devais, “l’expression du pouvoir aux frontières du royaume de Jérusalem: terre de Suète et oultre-Jourdain au Xiième siècle,” Bulletin d’Etudes Orientales 57 (2008), 19. 25 r. ellenblum, “Three generations of frankish castle-building in the latin Kingdom of Jerusalem,” in Autour de la Premiere Croisade, ed. m. Balard (paris, 1996), pp. 517–551. 26 Whilst tancred de hauteville was attempting to take Sawad, godefrey of Bouillon was exploring the south of transjordan and meeting the tribal chiefs. These two actions probably intervened in this sense, even though the coasts were still largely in the hands of the Syrians and egyptians. The two men were trying to keep their acquisitions safe and whole by protecting their eastern flank. albert d’aix “liber christianae expeditionis,” p. 517. 27 in order to have done so, the franks would have had to take possession of the hauran and the azraq oasis, which they were never able to do in spite of the local insurrections. in the next century, the Caves de Suète nevertheless appear in a missive that lists the frankish possessions then in the hands of the ayyubids. p. deschamps, “etude sur un texte énumérant les possessions musulmanes dans le royaume de Jérusalem vers l’année 1239,” Syria 23 (1942–1943), 86. 28 Several western authors use the terms ‘fringes’ or ‘marches’, and sometimes ‘wild lands’, to refer to the east bank of the Jordan. a seigneury on the eastern borders 77 of sentinel by containing the Seljuk and ayyubid armies.29 in fact, the muslim troops frequently ran up against the frankish ones on the golan heights or on the plain of the hauran.30 later on, the occupation of these lands prevented any union between the two great forces of the muslim world in the eastern mediterranean: egypt and Syria. The fortresses of Oultre Jourdain and the Terre de Suète, effectively barred the passage to Syria from egypt and the arabian peninsula along the hejaz road.31 The occupation played a major role after the fall of damascus in 1154 and even more after 1172 when egypt and Syria were joined together under one crown.32 Thanks to their eastern possessions, the franks constantly threatened communications using the Jordanian roads between damascus and mecca or cairo. This defensive aspect was doubled by an offensive value of primary importance. cities and fortresses were used as rallying points for troops launched against the hauran, the leja, or the caravans following the axes of the hejaz and the Wadi Sihan, and in case of defeat they were used as refuges for the retreating army.33 Several expeditions were prepared in gadara, one of the former cities of the decapolis, to loot the rich southern Syrian lands and to besiege Salkhad and Bosra. The crusaders posted in gadara, corsy, and dera’a were able to plunder the rich wheat belt,34 and their raids quickly became a means of exerting pressure that could affect the policy of the sovereigns of damascus by threatening them with starvation. in the complex political 29 no fort within the Terre de Suète appears in the list of citadels taken by Saladin. nevertheless, it is almost certain that the franks lost that country after the destruction of the army at hattīn in 1187. 30 it is not our purpose here to list all the skirmishes, but the battle of the plains of Buthaia and of Bouser can be cited as examples. 31 The caravan intercepted by renaud de châtillon was not composed entirely of peaceful pilgrims; it also comprised numerous troops. c. oman, The Art of War in the Middle Ages (london, 1991), p. 258; B. hamilton, The leper king and his heirs: Baldwin IV and the Crusader Kingdom of Jerusalem (cambridge, 2000), pp. 78, 116, and 226; lyons and Jackson, Saladin, pp. 6, 48, and 218. 32 hamilton, The leper king and his heirs, pp. 63–83, esp. 78 and 186–210; lyons and Jackson, Saladin, pp. 201–220. 33 This was the case during the expedition of 1148. 34 ibn Qalanisi, one of the better informed eastern authors for our period, reports that the constant frankish skirmishes to the south of damascus caused serious shortages and serious inflation of the price of wheat. The latins could thus strangle the economy of the city at will. ibn Qalanisi, Damas de 1075 à 1154, trans. r. le tourneau (damascus, 1952), pp. 39–40. 78 chapter three game that started around 1130 the capacity of the latins to starve damascus became a mighty weapon. This role of march played by the Terre de Suète implied a solid force in charge of the military operations. The “Book of Jean d’ibelin” in the Assises du royaume de Jérusalem informs us of this presence through the calculation of the knights due by each lordship of the realm. in case of an uprising, galilee had to provide some hundred knights to the king’s host which was one of the highest quotas due in the kingdom. of these hundred men, sixty came from galilee itself and forty came from the eastern possessions of the principality. Still according to ibelin, only sixty knights came from Oultre Jourdain, which was much larger, and one hundred from ascalon, the other vast border lordship. This high number of men deployed across the region clearly implies a function of limes, and completely refutes the idea of a condominium. The presence of these warriors, some of whom held lordships, surely illustrates their desire to control the area and to create a protective shield for Samaria and galilee. The expeditions against the hauran also reflect frankish expansionism, made possible by their installation in the Terre de Suète.35 every opportunity to push the conquest eastwards was exploited, even if it meant damaging essential alliances as it did in the 1140s by breaking the treaty with unar of damascus.36 at this time, the threat from aleppo was growing and unar of damascus (ainard according to William of tyre) tried to preserve the peace in spite of the blindness of the crusaders. regardless of this friendly disposition, the franks did not hesitate to support the insurrection of the emir of the hauran in 1144 who tried, albeit unsuccessfully, to seize Bosra and Salkhad. The ambiguous attitude of the emirs of the hauran during the whole twelfth century must be emphasized. it seems that as early as 1104, the governor of Bosra contacted the crusaders with the aim of forming an alliance that would emancipate him from the tutelage of damascus;37 an attempt that his successors renewed in 1126 and 1144, 35 in addition, they reflect a certain lack of foresight in so far as unar, atabeg of damascus, was a precious support against Zenghi of aleppo. 36 m. yared-riachi, La Politique extérieure de la principauté de Damas (damascus, 1997), p. 197. unar appears in the frankish chronicles under the name ainard. 37 yared-riachi, La Politique extérieure, p. 112; ibn Qalanisi, Damas de 1075 à 1154, p. 56: “[. . .] puis s’étant mis en rapport avec Baudouin, roi des francs, ils [aitegîn l’alépin, gouverneur de Bosra] allèrent le trouver pour lui demander assistance et a seigneury on the eastern borders 79 almost going as far as to open the gates of their towns to the Western invaders.38 This land was largely populated by christians and druse, and we might wonder if the new Sunni power established in damascus in 1126 did not result in the hauran being pushed towards a latin alliance. The emir al-tuntash, the instigator of the rebellion of 1147, was a converted armenian.39 his former religion, as well as that of a large fraction of his people, was close to that of the franks and could have been one of the factors that drove him towards this new loyalty. in any case, it is hard to accept any attempt by the franks to conquer places so far to the east, without also acknowledging that they were already well established on the eastern banks of Jordan river. to attack a citadel more than one hundred kilometres away from their lines of defence, right in the middle of the muslim lands, would have been suicidal. although all these attempts failed, they reflect the unstable character of the Terre de Suète, which was a buffer zone intended to absorb shocks on the one hand and to penetrate as far as possible into the rich muslim country to the south of Syria on the other. from at least 1161, the latin military power was reinforced by a contingent of brethren of the order of the Knights templar. William of tyre mentioned that in 1166 a cave fortress entrusted to the guard of the knights attacked and taken by the troops of nur al-din, without any real resistance from the brethren, an event which seriously outraged their contemporaries.40 King amaury i, who had already set out to help the stronghold, was furious at the knights’ lack of tenacity and ordered that the twelve brethren guilty of cowardice be hanged immediately.41 twelve seems a high number when compared to the order’s great classical fortresses, which generally counted around fifty demeurèrent quelques temps avec lui, au milieu des francs, l’incitant à marcher sur damas et le poussant à exercer des ravages dans les provinces qui en dépendaient.” 38 deschamps, La Défense du royaume. 39 yared-riachi, La Politique extérieure, p. 112; g. dédéyan, “un emir arménien du hawrân entre la principauté turque de damas et le royaume latin de Jérusalem (1147),” in Dei Gesta per Francos, eds. m. Balard, B. Z. Kedar, and J. riley-Smith (Burlington, 2001), p. 179. 40 William of tyre, Historia rerum, p. 879. it is probable that this was the same position as that taken by the crusaders in 1139. it was located at the fringes of the mountain of galaad, near ajlun. Surveys have been carried out in the area of the iron mines located there, which attest to a twelfth century occupation. 41 William of tyre does not specify what happened to the cave. did amaury recapture it? We do not know. 80 chapter three brethren. This new cave fortress constituted a post of utmost importance for the latins, as illustrated by the number of brethren and by the severity of the royal reaction. The acquisition of the amman citadel and of a third of the Balq’a region through a donation, undoubtedly from philippe de milly before 1166, put the ‘white mantels’ at the head of a vast territory on both sides of the Wadi Zarka, on the borderline of the realm.42 These lands were probably very interesting for another reason: they were linked to both the Terre de Suète and the land of Oultre Jourdain, and this conferred a certain degree of autonomy to them.43 Besides, they contained important resources such as iron mines, plantations of sugar cane, wheat and other cereals, and the production centres of olive oil. We might thus question the real intentions of the order. Being master of a huge border area, did it plan to create a sort of seigneury following the example of the ammanus between the lands of antioch and of little armenia? it is difficult to know for certain, but the geographical concentration of their power brings parallels to mind. unfortunately, the present lack of information does not allow any other conclusions to be drawn about the templar presence in the ajlun and Balq’a areas. in any case, reinforced by the templar garrisons, the crusaders could count on a mighty force far stronger than that which was necessary simply to collect taxes. contrary to what the presence of such military strength would suggest, there were no major fortifications in the Terre de Suète. Whilst other border zones were covered by Cracs, as in the upper galilee or in the transjordan, the Terre de Suète appeared desperately empty. The unfortunate attempt of 1105 put a serious damper on the building enthusiasm of the Westerners. They did not renew the mistake made at Qasr Berdawil and later constructions were much more modest.44 after that, the crusaders chose to settle in easily defendable, inexpensive, and naturally inaccessible positions. on their side, the Syrians chose to strengthen caves and ancient buildings.45 all the strongholds 42 amaury i confirms the donation of amman to the order in a charter dated 17 January 1166. J. delaville le roulx, “chartes de terre Sainte,” Revue de l’Orient Latin 11 (1905–1908), 183–187. 43 This was a rich area. Sugar cane was grown and there were many orchards. in addition, there were iron mines in the ajlun area. 44 Several times, however, the eastern and latin chroniclers make reference to forts or praesidium in the region without saying anything more about them. 45 This is the case, for example, at the temple of Bel in palmyra, in Bosra, Jerash, and azraq. on the subject of the construction of frontier fortresses, see the article a seigneury on the eastern borders 81 known in this area were of this kind, and they were not unusual in the middle east. in fact, the reuse of temples or theatres was frequent, as was the occupation of caves transformed into impressive bastions. The caves of Shayzar46 or tiron, the ancient temple of Baalbek, the temple of Bel in palmyra,47 and the theatre of Bosra are noteworthy examples. There could be several explanations for the absence of major construction works, above all the vicinity of the hostile territories that made perilous any extended construction time. The ease with which Thugtekin took over Qasr Berdawil showed early on the uselessness of building a fortress in the golan, a vast open area deprived of any natural defence. learning from this failure, the belligerents quickly concluded that to retreat toward the yarmouk river, in the heart of hilly country, would be more profitable than to stay in the easily accessible lowlands. additionally, deciding in favour of caves spared them from the colossal expenses required for the construction of a castle. The large investment of money and human resources for such uncertain results must have given the franks and their adversaries pause. Thus, they used caves and the remains of Byzantine cities such as gadara, Jerash, corsy, and, further in the south, amman. in each case, the ancient city acted as the foundation of the new occupation of the area (occupation that had actually never really stopped, even if it suffered a serious reduction in terms of population between the eighth and ninth centuries).48 moreover, it should be pointed out that the cave dwellings were highly developed and well adapted to the harsh conditions of transjordan, as was noted by the german pilgrim, Thietmar, during by ronnie ellenblum about Vadum Jacob: r. ellenblum, “frontier activities: the transformation of a muslim Sacred Site into the castle of Vadum iacob,” Crusades 2 (2003), 83. 46 Which were taken by a single warrior, a spicy scene which was described by osama ibn munqidh, emir of Shayzar. See h. derenbourg, “autobiographie d’ousama,” Revue de l’Orient latin 2 (1894), 399. 47 on this subject, see p. deschamps, “la cave de tyron, une grotte forteresse des croisés au liban,” Mélanges syriens offerts à M. Dussaud (paris, 1939), 2:873–882. 48 most archaeological work shows a certain degree of stability in the early days of islam, then, towards the second half of the eighth century, the forts begin to collapse. father piccirilo’s work at madaba and um ar-rasas, and alastair northedge’s work at amman sheds light on the evolution of transjordanian towns in these periods. a. northedge, Studies on Roman and Islamic ‘Ammān 1. History, site and architecture, (British academy monographs in archaeology) 3 (oxford, 1992); see also p. canivet, La Syrie de Byzance a l’Islam: VIIe–VIIIe siecles (paris, 1992). 82 chapter three his journey in 1217.49 previously, byzantine sources had mentioned that the region served as a refuge for people from west of the river who crossed eastwards to find security in the caves of the transjordan.50 habis on the yarmouk, which in arabic means ‘the hermit’, also referred to a Byzantine hermitage. a second place named habis, also consisting of caves, is located in the village of mansura near Shawbak in southern Jordan, and it is strikingly similar to the Caves de Suète, with cultural material indicating a crusader/ayyubid presence.51 even if these positions were less impressive than the Cracs, their importance must not be underestimated. The caves held by the templars did not achieve very far-reaching results, but the Caves de Suète efficiently stopped several attacks, such as that of 1158, and, according to William of tyre and some arab chroniclers, ensured frankish power in these regions.52 Their strength came from their garrison, at times composed of almost one hundred soldiers, and from the ability of the army to rescue them at very short notice.53 a large part of this power was used against the Bedouin tribes, and military actions were more often directed against looters, such as the renegades who terrorised the country in 1138, than against real armies.54 in either case, powerful walls were not necessary to protect the caves since these groups were unable to sustain a proper siege. finally, these structures were perfectly adapted to the control of the borderlands. easily defendable and needing but few structures (other than wells, basins and outer walls), these improvised forts fulfilled the role that was required of them, and did so for a very low cost in comparison with a real castle. indeed, the Caves de Suète, dug into a cliff at the bottom of a deep wadi and linked 49 Thietmar, “le pèlerinage de maître Thietmar,” in Croisades et pélerinages: récits, chroniques et voyages en Terre Sainte, XIIe–XVIe siècle, ed. d. régnier-Bohler (paris, 1997), p. 948. See also Jacques de Vitry, “historia ierosolymitana,” in S. de Sandoli, Itinera Hierosolymitana Crucesignatorum (Jerusalem, 1983), 3:316, on the subject of Bosra of the trachonitis, where it states that the “populus etiam illius regionis in speluncis habitans, et in traconibus habens domicilia, commoratur in cavernis.” 50 This was the case in particular at the time of the great Sassanid invasion of 606. B. Bagatti, Ancient Christian Villages of Galilee (Jerusalem, 2001), p. 259. 51 montreal/Shawbak hinterland survey, ifapo, 2003/2005. 52 William of tyre, Historia rerum, p. 841. at the time of the siege by Baldwin iV, in 1182, to recapture the fortress taken by Saladin, the besiegers would be head to head with the king for three weeks. See also ibid., p. 1028. 53 for purely military questions, see r. c. Smail, Crusading warfare 1097–1193 (cambridge, 1995); p. contamine, La Guerre au Moyen âge (paris, 1992); oman, Art of War; h. delbrück, Medieval Warfare (lincoln, ne, 1990). 54 William of tyre, Historia rerum, p. 682. a seigneury on the eastern borders 83 to the outside by a narrow shepherd’s path, could be used immediately as they were.55 conceived for the purpose of defence in co-ordination with a field army, these strange fortresses were strong enough to hold this rich agricultural land criss-crossed by numerous trade routes.56 These routes, at the edge of the steppe lands, constituted an additional attraction for the newcomers. Since antiquity, the transjordan has been crossed by major caravan routes connecting the mediterranean to Arabia Felix, the persian gulf, and from there, the far east.57 By asserting themselves in the transjordan and in Terre de Suète, the franks entered this vast system of exchange in which huge convoys transported costly spices, silk, or precious pottery to the coast. two of these routes converged in the vicinity of dera’a, which gave rise to one of the most important markets of the meidan plain.58 The hejaz road crossed Syria, passing through aleppo and damascus, continuing south to aqaba, and thence, through its port, to the holy cities of medina and mecca. from the red Sea ports, trade ships could reach yemen, africa, and india. a second route left the levantine coast for lower mesopotamia following the Wadi Sihan.59 By settling along these roads, and more particularly in dera’a, the franks could obtain substantial income from the goods conveyed to the hejaz and along the Wadi Sihan; sometimes they plundered the caravans crossing their lands and also prevented the passage of the hajj pilgrims. minor routes conveyed the goods to the coast and the big cities of the inner country. These generally followed the roman routes through dera’a, d. nicolle, “ain al habis,” Archéologie Médiévale 18 (1988), 115–140. J. prawer, The World of the Crusaders (Jerusalem, 1972), p. 121. 57 in our day, pilgrims coming from the caucasus and the Balkans still use the same routes to reach the holy places in the hejaz, stopping along the way at improvised markets to finance their journey. 58 This market seems to have existed up until the 1930s, since the 1937 Guide Bleu still described it. Several travellers noted this market, which attracted traders from the whole oriental world. amongst them were Jean de mandeville, “le livre de messier Jean de mandeville,” in Croisades et pelerinages, p. 1401: “il [le Jourdain] a un assez long cours souterrain jusqu’à une belle et grande plaine nommée par les sarrasinois Meldan, c’est-à-dire en français foire ou marché, parce qu’on y tient souvent des foires.” See also William of tyre, Historia rerum, p. 726; Jean de Wurzburg, “descriptio terrae Sanctae,” in S. de Sandoli, Itinera Hierosolymitana Crucesignatorum (Jerusalem, 1983), 2:289. 59 This route is also still much in use today, with convoys of trucks carrying trade between Jordan and iraq. The route is effectively the same, passing through mafraq and azraq to the centre of iraq. 55 56 84 chapter three capitolias, Jerash, gadara, and on to tiberias, acre, and tyre.60 by imposing themselves over this whole region, the Franks were able to control a major trade network and, thus, to levy substantial taxes on both travellers and trade goods.61 another element of some importance attracted the crusaders to the east of the Jordan river: the fertility of the northern Jordanian and southern syrian landscapes. The predation of the crusaders on the wheat reserves of Syria has already been mentioned above. The entire zone stretching from the Sea of galilee to the Jebel druze was, and continues to be, incredibly rich, supporting extensive cereal production, as well as orchards and olive trees. historical sources never fail to emphasize the abundance of fruit in the region, along with wheat and cattle of every kind, an abundance that has been confirmed by archaeological discoveries.62 The name ‘Sawad’, explained above, referred to the very fertile, black earth of the region. furthermore, sugar was produced in large quantity in the Jordan Valley and in neighbouring wadis, thanks to the vast sugar cane plantations. These were described by various Western travellers who discovered the delights of this unknown plant,63 and several archaeological sites along the Jordan Valley bear witness to this industry.64 The production, broadly intended for export, 60 r. p. abel, Geographie de la Palestine, 2 vols. (paris, 1967); r. dussaud, Topographie de la Syrie antique et médiévale (paris, 1927). for periods slightly later but still connected to our subject, see rafat el majali and abdul rahim mas’ad, “trade and trade routes in Jordan in the mameluk era (a.d. 1250–1516),” Studies in the History and Archaeology of Jordan 3 (1987), 311. ibn Jubayr describes a caravan that left damascus for acre in September 1184. See ibn Jubayr, “Voyages,” p. 446. 61 ibn Jubayr even notes the cost of passage in the area of Banyas, which was as much as one dinar per person. ibn Jubayr, “Voyages,” p. 447. 62 foucher de chartres describes the richness in livestock of the oriental lands. it allowed the crusaders to survive during their first year in the levant. foucher de chartres and Jeanne ménard, Histoire de la croisade: le récit d’un témoin de la première croisade, 1095–1106 (paris, 2001), p. 102. 63 Jacques de Vitry, “historia ierosolymitana,” pp. 320, 356, and 358. 64 Several sugar production sites have come to light in the Jordan Valley: abu arabi ash-Shamali, tall fandi al-Janubi, and deir ‘alla all attest an ayyubid-mamluk presence. also the excavations at tell hesban by Bethany Walker uncovered a large quantity of sugar jars, clearly destined for export. B. Walker, “The islamic Qusur of tell hisban: preliminary report on the 1998 and 2001 Seasons,” ADAJ 47 (2003), 443. Bethany Walker was kind enough to inform me of a crusader phase at tell hesban. See also p. Birgitte-porëe, “les moulins et fabriques à sucre de palestine et de chypre,” in n. coureas and riley-Smith, Cyprus and the Crusades (nicosia, 1994), pp. 377–510. however, it is important to exercise caution when using this last source. cf. m.-l. Von Wartburg, “The archaeology of cane Sugar production: a Survey of a seigneury on the eastern borders 85 guaranteed a substantial income to the owners of the plantations and refineries.65 finally, one should add the fishing industry in the Sea of galilee to the economic activities of the Terre de Suète. The wealth of the region and the defences set up by the franks are sufficient to quash the idea of a loose settlement based solely on the collecting of taxes. The evidence points to the fact that the latins were not content to rule from afar, if only because, as we have seen, they needed to guarantee the security of their borders. one of the arguments often put forward against the notion of a western occupation of the region is the distant and dangerous nature of the area. however, other areas of the kingdom, such as montreal or the area around ascalon and tyre, were no less exposed than the Terre de Suète and were settled by significant numbers of new people. in our case, by 1120, the north of Jordan and the south of Syria, which was broadly inhabited by eastern christians, did not represent a more dangerous area than upper galilee. on the contrary, it seems that the newcomers benefited from the kindness and even the support of a large part of the local population, as in montreal or petra, who allowed them to establish colonies. The franks could have been satisfied with collecting only their share of the taxes,66 but all the indications suggest that they actually settled. The annexation of these lands accompanied by the settlement of a frankish population at the gates of the Kingdom of Jerusalem, was most probably determined by the intention of securing the borders and exploiting the lands, as was the case in ascalon.67 Thus, numerous casaux appeared in the region, set up between the lower Jordan river and the plain of meidan. There are around thirty mentioned in various cartularies of the latin east. among these charters, one dated to 1127 mentions the casal of Jerraz in the twenty years of research in cyprus,” Antiquaries Journal 18 (2001), 305–335, esp. n. 15; eadem, “cane Sugar production Sites in cyprus. real and imagined,” Report of the Department of Antiquities, Cyprus (2000), 381–401, esp. 397, n. 70. 65 in the thirteenth century, there was a sugar known as ‘montreal’, coming from the transjordan. deschamps, La Défense du royaume de Jèrusalem, p. 49. 66 Which was clearly the case in some places, but this form of power was not adopted everywhere. ibn Jubayr, “Voyages,” p. 447. 67 William of tyre, Historia rerum, p. 707. according to William of tyre, the lords of the region favoured the creation of agricultural communities in the area in order to pacify the region. 86 chapter three Jordan valley,68 which can be identified as Jerash/gerasa.69 at that time gerasa was gifted to the abbey of our lady of Josaphat. This act illustrates perfectly the increase of the power of the crusaders in the area. This town, which had been conquered in 1121 and immediately abandoned because it was thought to be too dangerous, appears to have been more secure less than ten years later, proving the power of the franks in the ajlun region. These casaux formed dense groups of four or five villages at a distance of less than a dozen kilometres from each other, all of which were established around the ancient cities of the decapolis70 or the fortresses. moving from east to west, corsy71 comes first, on the eastern shore of lake tiberias, gadara is a little further south, then habis on the yarmouk river, further east is Beit ras/ capitolias and finally dera’a in the plain of the hauran. even though a casal is not apparent in the immediate environs, it seems likely that a sixth group existed. it was centred around the templar caves, most probably the very same that had been used by thieves in 1138.72 a survey by Joseph green found some mills in the Kufranja Valley near ajlun. Several sugar presses have also been found in the same region, dating to the twelfth and thirteenth centuries. furthermore, during a survey, the present author was able to check on the existence of a remarkable complex of caves dominating access to the Jordan Valley that is very similar to the complex of el-habis.73 each of these groups 68 h. f. delaborde, “chartes de la terre Sainte provenant de l’abbaye notre dame de Josaphat,” Bibliotheque des Ecoles Francaises de Rome et d’Athenes 19 (1880), p. 46, charter 18; r. röhricht, Regesta regni Hierosolymitani, (MXCVII–MCCXCI) (innsbruck, 1893), charter 159. 69 during the excavations of the temple of Zeus, Jacques Seigne found evidence of fighting that he dated to the first quarter of the twelfth century, followed by a reoccupation of the site. These finds corroborate our sources. The temple, refurbished into a fort, was indeed attacked in the 1120s and used thereafter. 70 towns that are mentioned as being bishoprics depending upon either nazareth/ Scytopolis (the seat of the archbishop having been moved to nazareth) or Bosra. a Provinciale dating to 1180 names four towns in the region dependant on nazareth: capitoile (capitolias), gadirom (gadara), pelon (pella), and chricppus (probably corsy); and two dependant on Bosra: adraon (dera’a), and philiople (fiq). “patriarcats de Jerusalem et d’antioche,” in Itinéraires à Jérusalem et descriptions de la Terre Sainte rédigés en français aux XIe, XIIe & XIIIe siècles, eds. henri Victor michelant and gaston reynaud (geneva, 1882), pp. 12–13 and 17. 71 Bellarmino Bagatti recognized the site in 1971 and noted the remains of a crusader building. Bagatti, Ancient Christian Villages, p. 67. 72 William of tyre, Historia rerum, p. 681. 73 This group of caves might be those of the templars of 1166. a seigneury on the eastern borders 87 of villages seems to have been controlled by at least one lord, only two of which have, until now, been noted by historians. Willelmus de Sueta, who deschamps had already mentioned in 1935,74 signed several notarised acts, proving the existence of a suzerainty over the whole country. more recently, tibble has noted the presence of a certain Bernard Derat,75 corresponding almost certainly to the town of dera’a, the civitas de Bernardus de Stampis of William of tyre.76 in fact, at least three others put their names on charters in the twelfth century: in 1158, Renierius de Corsie77 and Pontus de Sira78; in 1169, Guido de Beteras.79 to these five lords can be added the masters of the Caves de Suète. The existence of these lords dividing up the province points to a well structured organization in this part of galilee. each group of casaux was associated with a lord, with the notable exception of gadara, which in any case could quite easily have been attributed to a knight who has remained unknown. The whole of the territory was thus distributed according to the then current principles of vassaldom. moreover, the existence of these knights suggests that the power of the princes of galilee was not exercised exclusively by way of their fort at habis, but was equally reliant on the feudal links based on the fiefdoms. Thus, we find the forty knights listed in the Assises du royaume de Jérusalem, amongst whom some at least must have had possessions in the Terre de Suète. These lords refute the hypothesis that the franks never settled east of lake tiberias and that they were satisfied simply with a modus vivendi that guaranteed them the taxes of the region. in fact, they opted for direct control of the area. The existence of such an organization raises the question of the population of this border zone. Were these lands colonized in the twelfth 74 p. deschamps, “une grotte forteresse des croisés au-delà du Jourdain,” Journal Asiatique 227 (1935), 285–299. 75 tibble, Monarchy and Lordship, p. 52. 76 William of tyre, Historia rerum, p. 728. The author points out that the town is commonly known thus, but that its proper name is adraat. 77 delaville le roulx, “chartes de terre Sainte,” p. 182. 78 g. Bresc-Bautier, Cartulaire du chapitre du Saint-Sépulcre de Jérusalem (paris, 1984), charter 81, p. 190, and charter 83, p. 191. 79 delaville le roulx, “chartes de terre Sainte,” charter 3, p. 185. moreover, a certain girardus de Beteras appears in 1168 when a donation is made, indicating the existence of a lineage holding this fiefdom. See röhricht, Regesta regni Hierosolymitani, charter 454, p. 119. 88 chapter three century?80 unfortunately, we do not have any precise sources, such as those that describe the population of the Mahomérie, to help us. nevertheless, archaeological work81 and toponymic studies have supplied some important information on the state of the country at that time. Four groups are of particular interest because of the work that has been carried out there: gadara and the eastern shores of lake tiberias, and the areas of ajlun, dera’a, and beit ras (capitolias)/irbid. if one follows ronnie ellenblum’s reasoning on the subject of western colonization, these zones are particularly interesting in so far as they have remained essentially christian until recently. The villages of ajlun, Kufranja, and el hosn (irbid)82 still had a christian majority at the beginning of the twentieth century.83 evidently, the existence of these communities is not enough to justify the presence of franks in the region. nevertheless, a group of indications, beginning with the toponyms, tend to show that settlement did take place. The ajlun area, as already mentioned, is a bastion of Jordanian orthodoxy. The result of the toponymic studies there tends to indicate a latin presence: in the immediate vicinity of the castle there is an area called Kufranja, ‘the village of the franks’, and a local wadi is named Wadi al-franji, ‘the wadi of the franks’. These names, coupled with the archaeological work, would seem to bear witness to the settlement of franks in the vicinity. The survey carried out by Jonathan mabry and gaetano palumbo also showed that following a decrease in the eleventh century, the population of this zone increased significantly from the twelfth to the fourteenth century before slumping again in the ottoman period.84 80 ronnie ellenblum believes, based on the presumed absence of eastern christians that the franks never attempted to settle in the region. ellenblum, Frankish Settlement, p. 276. 81 alan Walmsley notes that the lack of material found during survey does not reflect the real settlement during the eleventh/twelfth century. a. Walmsley, “The middle islamic and crusader periods,” p. 519. 82 a fortress named Hisn was seized by the crusaders in 1118. This location is probably the modern hosn. The city is built on the top on a hill, but unfortunately, as yet, there is no indication of a military building there. J. m. mouton, Damas et sa principauté sous les Saljoukides et les Bourides (486–549/1076–1154) (cairo, 1994), p. 52. 83 Guide Bleu (edition hachette, 1937), p. 368. 84 J. mabry and g. palumbo, “Wadi yabis Survey,” in Archaeology of Jordan, vol. 2.1, Field Reports. Sites A-K., eds. d. homes-fredericq and J. B. hennessy (amman, 1989), p. 91. Thirteen of the surveyed sites were occupied from the twelfth to thirteenth centuries. a seigneury on the eastern borders 89 The remains of craft industries such as mills,85 sugar factories, and surface material, notably ceramics,86 attest to this occupation. What is more, it was probably in this area that the templars settled after they were forced out of their cave fortresses in 1166. during the Wadi yabis survey, the present author was able to explore a group of large caves very similar to those of el-habis.87 The re-population of this zone by the franks, and no doubt by eastern christians, could therefore have accompanied the establishment of the order of the temple and benefited from its protection in the heart of a region that was still largely christian.88 This is not an isolated example. other archaeological sources lend support to what can be seen in ajlun. ayyubid/ mamluk material found in the various excavations at gadara89 attests to the occupation of the site in the twelfth century, which corroborates with the written sources. moreover, the department of antiquities of Jordan found a crusader treasure in this town.90 finally, and perhaps most importantly, the fourth-century, five-aisled basilica was still in use in the twelfth century. it was changed into a mosque only at the end the twelfth century by the simple addition of a minaret. This discovery, made by a german team,91 clearly shows the existence of a christian community during the crusader period, and the fact that it 85 Joseph a. green, “The Water mills of the ajlun-Kufranja Valley: the relationship of technology, society and settlement,” Studies in the History and Archaeology of Jordan 5 (1995), 757. 86 t. Weber, “listib,” in Archaeology of Jordan, vol. 2.2, Field Reports. Sites L-Z, eds. d. homes-fredericq and J. B. hennessy (leuven, 1989), p. 368. 87 This group blocked the route leading to the ghor. it dominates a fertile valley where the remains of a pressure pipeline are still visible. even though the surface material was very limited, a few fragments of pottery at the foot of the caves would seem to confirm a twelfth- and thirteenth-century presence. 88 The ajlun mosque, dated to the thirteenth century, was built over a church, and the village of mar elias has remained an important pilgrimage site for eastern communities to this day. 89 collections of the gadara museum. 90 it has not yet been published, but the local department of antiquities was kind enough to inform us of this find. 91 excavations and work continue under the auspices of the Jordanian department of antiquities and the german protestant institute for archaeology, amman. This large site will certainly produce new discoveries in the future. for a general overview of the site, see d. homes-fredericq and J. B. hennessy, “um Qeis,” in Archaeology of Jordan, vol. 2.2, Field Reports. Sites L-Z, eds. d. homes-fredericq and J. B. hennessy (leuven, 1989), p. 597. 90 chapter three was replaced at the end of the twelfth or beginning of the thirteenth century indisputably implies a rupture in the region.92 two hypotheses, possibly complementary, can be presented to explain these developments. The construction of the ajlun citadel by the ayyubids and then the defeat of the franks at hattīn no doubt resulted in the departure of the latin population and some of the eastern christians from the country. at the same time, the religious domination of the new conquerors93 manifested itself in their transformation of churches into mosques, sometimes simply by the construction of a minaret and the destruction of the crosses, the symbol of their adversaries. This was a phenomenon also seen at montreal. during a survey in the golan, Bagatti noted the presence of a ruined fortification and frankish church at corsy.94 lastly, the sources show that at Saint george, the order of notre-dame of Josaphat received a church from William of Bures.95 This is of particular interest insofar as, according to ronnie ellenblum, a church has been considered as often forming the centre of a latin settlement. This casal in the vicinity of dera’a, far to the east and almost on the borders of the emirs of Bosra, had a frankish settlement. it is likely that the wealth of the country and even more so of the markets that were held close by at meidan, encouraged the colonists who were also supported by the order. finally, during their survey in the north of Jordan, lenzen and mcQuitty found evidence of frankish presence in the area of Beit ras/irbid.96 Thus, the Terre de Suète was more than a simple condominium. This status, described by the eastern sources, represented only one stage in the existence of the seigneury. The advance of the crusaders around 1120 clearly put an end to this system in favour of a feudal organization of territory whereby the lords and templars divided the land amongst themselves under the domination of the princes of tiberias. 92 The minaret of the ajlun mosque belongs to the same period, as does that at dera’a. an identical change can be seen at al-Bira. d. pringle, “magna mahomeria (al-Bira): the archaeology of a frankish new town in palestine,” in Crusade and Settlement, ed. p. W. edbury (cardiff, 1985), p. 149. in dera’a the mosque re-used Byzantine stones and capitals. an inscription on the minaret is dated from 1253. 93 lyons and Jackson, Saladin. 94 Bagatti, Ancient Christian Villages, pp. 64–67. 95 delaborde, “chartes de la terre,” pp. 40 and 46, charters 14 and 18. 96 c. J. lenzen and e. a. Knauf, “Beit ras/capitolias, a preliminary evaluation of the archaeological and textual evidence,” Syria 64 (1987), 21. a seigneury on the eastern borders 91 however, there is nothing to suggest that the seigneuries of dera’a, corsy, and Beit ras were independent of galilee, a fact confirmed by the donations made by William of Bures in the area of dera’a (casal Saint-george and Saint Job). additionally, even though the region was on the borders of the Kingdom of Jerusalem, everything seems to suggest that there were considerable settlement activities. if the medieval periods are not ignored in the current archaeological excavations, they should throw some light on the nature of the populations settled in the region. in spite of the absence of any major known structures, the Terre de Suète seems, nevertheless, to have been a protective march for the kingdom, as well as a land exploited by the latins, and not merely shadowy space where only nominal power was exercised by the franks. chapter Four Merchants, MonKs, and Medieval sardinian architecture* Michelle hobart This chapter evaluates medieval Sardinia from three perspectives: material culture, architectural history, and socio-economic history. This interdisciplinary approach yields the benefit of a better understanding of Sardinia’s evolving settlement patterns, the island’s role in mediterranean trade, and its relationship with the islamic world.1 as a point of departure, the focus of this study is placed on one particular architectural element, using it to connect these topics and further understand the island’s role in mediterranean trade. With special attention to arab culture and its ‘non-visible’ presence along the mediterranean coasts, this chapter proposes that Sardinia might have played a wider role in medieval maritime trade than previously believed. Bacini The term ‘bacini’ (sing. ‘bacino’) was first used by gian Battista passeri in 1758 to describe decorative pottery on the churches of pesaro, italy. The term has since come to indicate vessels, generally of an open form, that perform a decorative function, predominantly on the facades of religious buildings (fig. 1).2 Bacini are glazed utilitarian ceramic plates that were originally imported to continental italy from different * This chapter is dedicated to graziella Berti. 1 for an introduction to medieval Sardinia, see rossana martorelli, ed., Città, territorio, produzione e commerci nella Sardegna medievale (cagliari, 2002); robert rowland, The Periphery in the Center: Sardinia in the Ancient and Medieval Worlds, (Bar international Series) 970 (oxford, 2001), pp. 151–155; francesco floris, Storia della Sardegna (rome, 1999), pp. 152–156; raimondo turtas, Storia della chiesa in Sardegna: dalle origini al Duemila (rome, 1999), p. 971; laura galoppini, Sardegna e Mediterraneo: dai Vandali agli Aragonesi. Antologia di fonti scritte (pisa, 1993). 2 graziella Berti and liana tongiorgi, I bacini ceramici medievali delle chiese di Pisa (rome, 1981). 94 chapter four islamic west-Mediterranean centres and subsequently produced on the peninsula.3 They were first introduced to continental italy in the late tenth century and their durability and wide availability made them an ideal architectural element for many ecclesiastical structures that were built before and during the Gregorian reformation. The use of bacini as architectural elements is also found in spain, albania, Yemen, and iraq,4 although the origin of the practice is unclear. during the Middle ages, bacini were also used as domestic wares in high-end residences or large monasteries. There are two commonly accepted methods for dating bacini. The first typifies the bacini found on churches for which accurate dates of construction are known. This method assumes that bacini were incorporated into the masonry at the time the churches were constructed. additionally, the recent discovery and identification of medieval pottery-production centres in the Mediterranean has not only led to an increased accuracy (within about 25 years) in the dating of bacini, but it has also provided valuable information about the particular geographic regions from whence the pottery originated. Bacini and the Dating of Medieval Sardinian Churches Most of the medieval churches on sardinia have no surviving documents regarding their construction and, in the past, dating them was based exclusively on stylistic terms. on the other hand, in pisa, documents on the construction of churches abound. in 1981, a pioneering book by Graziella Berti and liana tongiorgi on the medieval churches of pisa that are decorated with bacini offered a comprehensive survey of all the major types of islamic and italian bacini produced in the Mediterranean, ranging from the end of the tenth to the fourteenth century.5 This text provided precise dates for each of these types, 3 Graziella Berti and sauro Gelichi, “trasmissioni di tecnologie nel Medioevo: tendenze e linee di ricerca attuali,” in Circolazione di tecnologie, maestranze e materie prime nelle produzioni ceramiche del mediterraneo dal medioevo all’età moderna (albisola, 2001), pp. 23–41. 4 see [Atti:] 26. Convegno internazionale della ceramica. I bacini murati medievali: problemi e stato della ricerca, Albisola, 28–30 maggio 1993 (albisola, 1996); hereafter I bacini murati medievali. The entire conference was dedicated to presence of bacini in the Mediterranean. 5 Berti and tongiorgi, I bacini ceramici medievali. Berti and tongiorgi analysed twenty-three churches in the maritime republic decorated with imported bacini. This merchants, monks, and medieval sardinian architecture 95 based on the age of the church and the accepted understanding that the ceramics were inserted at the time of construction. By extension, this system of dating pinpointed exactly when these products were circulating throughout the Mediterranean. riccardo francovich, one of the fathers of Medieval archaeology in central italy, suggested that the aforementioned methodology might be inverted in order to date the many churches of sardinia that are decorated with bacini, which have hardly any surviving documents. The project outlined in this chapter began in the 1990s as an experiment that sought to put francovich’s suggestion into practice. it was an opportunity to put to use this new discipline to show how archaeology could contribute to the historical socio-economic debate on a wider scale. prior to this research project, the appearance of pottery in the archeological record had often been used to date phases of occupation, but never before had a decorative ceramic architectural detail been used as one of the leading indicators for a historical discussion. indeed, sardinia is an ideal setting in which to analyse cultural exchange via bacini, as it has the largest number of architectural bacini in the Mediterranean.6 of the roughly one hundred and seventy churches built on sardinia between the eleventh and the fourteenth century, over sixty were decorated with ceramic bacini (fig. 2). approximately ninety bacini have been identified in situ on these churches and a total of seven hundred and fourteen bacini recesses were counted (table 1; the recess is the empty space left by the bacino once it has fallen out or been destroyed). The majority of the surviving bacini are no longer in situ. in some cases they were removed, restored, and are now kept in the cagliari museum or a local warehouse. however, when the bacini have been removed from their original context, it is impossible to establish their profile and the nature of their glaze. since a great many of the recesses are now empty, much information has been lost. all the same, the extant bacini provide a glimpse into the variety of bacini that were in use on medieval sardinia and they allow for certain conclusions to be drawn regarding churches lacking any other documentation. pottery, which had been very rarely studied before, contributed to the identification of most types of ceramics circulating in the Mediterranean since the end of the tenth century. Generations of archaeologists have used this book to date sealed medieval layers in excavations all over the Mediterranean. from the thirteenth century onwards, imports were increasingly substituted with local pisan and italian productions. 6 see I bacini murati medievali. 96 chapter four Dating Medieval Sardinian Churches Without Extant Bacini as mentioned above, many sardinian churches have lost their original pottery. nonetheless, the types of bacini that likely decorated these buildings can still be inferred. if the currently accepted dates for the foundation of sardinian churches are used, the types of the missing bacini can be inferred from those types used on contemporary churches with extant bacini. for instance, at the time the first churches with bacini appear, only islamic products were available in sardinia and the Mediterranean (italian potters did not use glazing techniques until the early thirteenth century). Buildings with empty recesses that are dated to the late eleventh or twelfth century, either on the basis of style, masonry techniques, or documentation, must likewise have contained islamic bacini produced either in north africa, sicily, or spain.7 These churches are san simplicio in olbia, santa Maria del regno in ardara, san Giovanni Battista in orotelli, san pietro in silki at sassari, san Michele di ploaghe in salvenor, nostra signora di talia in olmedo, san Giorgio in usini, santa Maria di sibiola in serdiana, santa lucia in serdiana, santa Giusta in oristano (fig. 3), sant’antioco di Bisarcio in ozieri, san nicolò in ottana (fig. 4), san platano in villaspeciosa (fig. 5), and san pietro of sorres at Brutta. Bacini of Sardinia The technique of glazing ceramics is the great innovation of islamic potters.8 a sharp boundary distinguishes the manufacturers of eastern and western islamic artifacts. The boundary of the eastern islamic world of the Middle ages fell along the italian adriatic coastline, a region with strong ties to Byzantium. The western islamic world consisted of coastal north africa (tunis and Morocco), sicily, parts of southern iberia, and the Maiorchine islands. at this stage, the islamic 7 Graziella Berti, “i bacini islamici del Museo nazionale di san Matteo, pisa: vent’anni dopo la pubblicazione del Corpus,” in Studi in onore di Umberto Scerrato, eds. Maria vittoria fontana and Bruno Genito (naples, 2003), pp. 121–151. 8 Berti and Gelichi, “trasmissioni di tecnologie nel Medioevo”; abdelaziz daoulatli, “la production vert et brun en tunisie du iX au Xii siecle,” in Le Vert et le Brun (Marseilles, 1995), pp. 69–89; Graziella Berti and sauro Gelichi, “Mille chemins ouverts en italie,” in Le Vert et le Brun, pp. 128–163; Marilyn Jenkins, Medieval Maghribī Ceramics. A reappraisal of the Pottery Production of the Western regions of the Muslim World (phd dissertation, new York university, institute of fine arts, 1978); to mention a few. merchants, monks, and medieval sardinian architecture 97 pottery from medieval sardinia, with the exception of a few fragments from egypt, originated from western islamic centres and continental italy.9 Tunisia The earliest islamic bacini found in sardinia are on the church of san Gavino, and have been associated with tunisian products on the basis of their ornamental patterns, technique, and shapes. These bacini were most likely produced in the kairouan area and exported from the port of Mahdia, the first capital of the fatimids, founded in 909 on the eastern coast of tunisia (fig. 6).10 trading centres such as the ports of tunis (carthage) and Mahdia exported pottery well into the thirteenth century. The bacini of san Gavino are also related to those that appear in pisa at the church of san pietro a Grado. These bacini are monochrome white and black on turquoise green and they are categorized as “tin or lead glazed bacini with yellow boli on a green background.”11 The early tunisian products were monochromatic, bichromatic, or polychromatic and differed on the basis of their technique, shape, and decoration. The glazing technique required double firing: the first to the biscotto, which lacked any decoration (with the rare exception of small incisions or stamps executed before the first firing); the second firing took place after the application of the glazes. These could be of two types. The first type was a lead-based fritta mixture (lead glaze), with or without colour. in both cases, the glaze was always transparent. When the glaze lacked transparency it was due to the fact that the temperature was not sufficiently elevated or there was a disruption during the process. The second type was the same lead glaze mixed with an opaque tin glaze, with or without colour. The degree of opacity was dictated by the quantity of tin covering the biscotto. 9 Berti, “i bacini islamici.” even in pisa, only a small amount of pottery came from Byzantium and the nearby areas. These all date to the twelfth century. for sardinia’s domestic use, see daniela rovina, La Sezione Medieval del Museo “G.A. Sanna” di Sassari (piedimonte Matese, 2000), p. 121. 10 Graziella Berti, “la ceramica tunisina ‘a cobalto e manganese’ in toscana,” in Ceramica in blu: diffusione e utilizzazione del blu nella ceramica (albisola, 2002), pp. 89–102; daoulatli, “la production vert et brun en tunisie,” pp. 69–76. 11 for the first type, see Berti and tongiorgi, I bacini ceramici medievali, pl. 112. for the second type, see ibid., pls. 86, 88, 89, 90, 95, 96, and 97. 98 chapter four a later tunisian type of bacini, very popular among pisan merchants, was decorated with cobalt and manganese (coMn) (fig. 7). This type of bacino is found on numerous churches in both pisa and sardinia, as well as in many excavations.12 The use of these bacini in pisa is precisely dated between 1175 and 1250.13 starting in the late twelfth century this type of dish was imported not only for architectural decoration, but also for domestic use. in sardinia, the juxtaposition of this type of pottery with italian products can help date a church more precisely. Sicily The earliest sicilian glazed bacini used in pisa came from the newly established islamic workshops in palermo, which produced polychromatic and lead-glazed wares beginning in the late tenth and early eleventh century. in sardinia, the mineralogical analysis of six pieces from the church of san niccolò of trullas in semestene indicates that they also come from eastern sicily14 and were manufactured in the second half of the eleventh century (fig. 8).15 other known sicilian centres of ceramic production were Marsala, syracuse, and agrigento.16 The more well-known ‘Gela ware’, named after the town Gela and typified by its whitish slip-glaze is absent in sardinia.17 Gela ware is no longer 12 This type has been found in tuscany, liguria, and lombardy. see Graziella Berti and laura cappelli, Lucca, ceramiche medievali e post-medievali: Museo nazionale di Villa Guinigi, (ricerche di archeologia altomedievale e Medievale) 19–20 (florence, 1994), pp. 131–138; Graziella Berti, “pisa: ceramiche e commerci (2 meta’ X-meta’ Xiv s.),” in Il Congresso nazionale di archeologia medievale, ed. sauro Gelichi (florence, 1997), p. 347; Graziella Berti and liana tongiorgi, I bacini ceramici del Duomo di San Miniato le chiese di Pisa (Genoa, 1981); Berti, “la ceramica tunisina.” 13 The attribution to tunisia is based both on mineralogical and chemical analyses. see Berti and tongiorgi, I bacini ceramici medievali; Berti and tongiorgi, I bacini ceramici del Duomo; Berti, “i bacini islamici.” 14 tiziano Mannoni kindly provided this information. Graziella Berti, “ceramiche islamiche del Mediterraneo occidentale usate come ‘bacini’ in toscana, in sardegna e in corsica (secoli Xi–Xiii),” in Età di Federico II nella Sicilia Centro-meridionale (agrigento, 1991), pp. 99–114, 266–276, and 329–333. 15 Michelle hobart, catalog nos. 16 and 18–19, in Moriscos. Echi della presenza islamica in Sardegna, eds. Maria francesca porcella, Marcella serreli, luisa degioannia, and antonia Giulia Maxia (cagliari, 1993), pp. 32–33. 16 alessandra Molinari, “l’italy du sud,” in Le Vert et le Brun, pp. 119–123. 17 Berti and Gelichi, “Mille chemins ouverts en italie,” pp. 146–147; stefania fiorilla, Gela: le cramiche medievali dai pozzi di Piazza S. Giacomo (Messina, 1996). merchants, monks, and medieval sardinian architecture 99 considered an islamic production, since it belongs to the tradition of southern italian protomajolica. nevertheless, identifying the tunisian or sicilian production centre of a particular bacino is difficult without the help of mineralogical analysis. When dealing with islamic products from these areas, both the shape and the technique are very similar. Spain (the region of Valencia) The islamic technique for glazing pottery was adopted in the West sometime between the eleventh and thirteenth century, at roughly the same time continental italian and sardinian churches were being embellished with islamic bacini. Thus, spanish and italian bacini also began to appear on churches from the late twelfth through the thirteenth century, and continental european bacini gradually replaced those from north africa and sicily.18 The work of islamic potters from regions of present-day spain was exported to both continental italy and sardinia. valencian lusterware is found at most coastal settlements beginning in the fourteenth century. in sardinia, two different types of lusterware were used for bacini: one with a blue glaze on a white slip and the other with a blue glaze on a turquoise ground (fig. 9). Both date to the same period of production (first half of the fourteenth century) of another spanish ceramic, the pula type.19 The latter was used for bacini fairly frequently,20 but never as such in sardinia.21 according to 18 Berti and Gelichi, “trasmissioni di tecnologie nel Medioevo”; Berti and Gelichi, “Mille chemins ouverts en italie.” 19 rovina, La Sezione Medieval, p. 72; francesca porcella, “il ‘fondo pula’ e gli affiliati,” in La corona d’Aragona: un patrimonio comune per Italia e Spagna (secc. XIV–XV) (cagliari, 1989), pp. 365–373, pls. 605–665. 20 hugo Blake, “la ceramica medievale spagnola e la liguria,” in Atti, 5 Convegno internazionale della ceramica (albisola, 1972), pp. 66–79; otto Mazzucato, I “bacini” a Roma e nel Lazio, 2 vols. (rome, 1973–1977), 1:51–52; sergio nepoti, “ceramiche tardo medievali spagnole ed islamiche orientali nell’italia centro settentrionale adriatica,” in Segundo Coloquio Ceramica medieval del Mediterraneo occidental (Madrid, 1986), p. 357, pl. 7; franco d’angelo, “le ceramiche spagnole ‘tipo pula’ delle chiese dello steri di palermo,” in Il servizio da tavola in ceramica (albisola, 1985), pp. 77–84; Berti and tongiorgi, I bacini ceramici medievali, pp. 270–271; hugo Blake, “The ceramic hoard from pula (prov. cagliari) and the pula type of spanish lusterware,” in Secundo Coloquio Internacional de Ceramica medieval en el Mediterraneo occidental (Madrid, 1986), pp. 365–407. 21 rovina, La Sezione Medieval, p. 72. 100 chapter four the present state of research, the valencian type was used for bacini in sardinia at Busachi in the campidano, in the south near cagliari (figs. 9–10),22 orosei, and possibly on the island of Molara, both part of the Gallura region on the eastern side of sardinia. fragments of valencian lusterware have also been salvaged from surfaces in the castle of san Gavino in Monreale, in santa chiara in cagliari,23 and san nicola in sassari.24 Southern Italy in italy, protomajolica, spiral ware, and monochrome lead ware were produced in the south (campania and apulia), while incised and monochrome-slipped ware was produced in the north (liguria). nearly a quarter of the bacini documented in sardinian architecture can be classified as protomajolica, which proves the commercial importance of southern italy and its many centres of pottery production in the thirteenth century.25 sufficient information exists to identify some of the workshops, but not enough to reconstruct the complete picture. The history of protomajolica is fascinating and complex. its production was once wrongly attributed to the Middle east.26 Ballardini was the first to discover that italian workshops had produced protomajolica. subsequent investigations have fully confirmed his hypothesis.27 in hobart, catalog no. 77, in Moriscos, pp. 61–63. donatella salvi, “la ceramica medievale e post-medievale,” in Santa Chiara. Restauri e Scoperte, ed. aldo ingegno (cagliari, 1993), pp. 133–151. 24 daniela rovina, “ceramiche graffite medievali e post medievali dal san nicola di sassari e altri siti della sardegna centro settentrionale,” in La ceramica graffita (albisola, 1989), pp. 201–209; eadem, “il duomo di sassari: recenti indagini archeologiche,” in Sassari, le origini (sassari, 1989), pp. 161–172. 25 Graziella Berti, Michelle hobart, and francesca porcella, “‘protomaioliche’ in sardegna,” in La protomaiolica e la maiolica arcaica dalle origini al trecento (albisola, 1993), pp. 153–167. 26 protomajolica has been found in Greece and along the turkish coast since the beginning of the twentieth century. see david Whitehouse, “The Medieval Glazed pottery of lazio,” Papers of the British School of Rome 35 (1967), 40–86; idem, “The Medieval pottery of rome,” in Papers in Italian Archaeology 1. The Lancaster Seminar: Recent Research in Prehistoric, Classical, and Medieval Archaeology, (Bar supplementary series) 41 (oxford, 1978), pp. 41 and 475–493; idem, “7. The bacini of ss. Giovanni e paolo, rome; 8. Bacini at Gaeta,” in Medieval Lazio: Studies in Architecture, Paintings and Ceramics, eds. david andrews, John osborne, and david Whitehouse, (Bar international series) 125 (oxford, 1982), pp. 347–371. 27 Gaetano Ballardini, “lo stile arcaico: italia Meridionale—il X corso di storia e tecnica delle ceramiche,” Faenza 25 (1937), 96–97. 22 23 merchants, monks, and medieval sardinian architecture 101 the 1980s, the excavation at san lorenzo Maggiore in naples showed that the protomajolica pottery was present from the mid-thirteenth through the entire fourteenth century.28 excavations on the opposite adriatic coast, at Brindisi, revealed that the pottery manufactured there dated to the first half of the thirteenth century and had been disseminated from the areas around salento and Basilicata all the way to Greece (Gastouni, corinth, and Merbaka) and the near east (al-Mina, hama, and atlit).29 protomajolica bacini were used in both northern and southern sardinia, yet not all the pottery shared the same provenance. for example, in sassari various types of protomajolica bacini were imported simultaneously from different markets and used together on the same building. san nicola, sassari’s cathedral bell tower, displays every possible type available on the island, including coMn from north africa, lusterware from spain, and both southern and northern italian vessels. a similarly rich combination can be found on the small rural church of santa Barbara outside of sassari (fig. 11), which contains incised slip ware together with a protomajolica (fig. 12), all dating to the 1270s. a piece of north african coMn on san nicola’s campanile suggests that the building could be several decades older than santa Barbara. in any case, the bacini on these two structures are particularly interesting because they establish a rather precise chronology for monuments for which there is no written documentation and they show the wide range of markets from which sassari was buying. spiral ware has been discovered in various locations in italy and north africa: liguria, lazio, campania, sicily, and tunisia (carthage).30 Many years of disagreement over its dating came to an end when sealed excavations proved that spiral ware had been manufactured since the first half of the thirteenth century.31 Mineralogical analyses make it clear that this rather common ceramic decorated with spirals, was not 28 Maria vittoria fontana and Giovanna ventrone vassallo, La ceramica medievale di San Lorenzo Maggiore in Napoli, 2 vols. (naples, 1984). 29 stella patitucci uggeri, “protomaiolica Brindisina,” Faenza 65 (1979), 241–255; pietro riavez, “atlit—protomaiolica ceramiche italiane nel Mediterraneo orientale,” in 2. Congresso nazionale di archeologia medievale, ed. Gian pietro Brogiolo (florence, 2000), pp. 444–450. 30 Mazzucato, I bacini a Roma e nel Lazio; Gaetano Ballardini, “le ceramiche architettoniche di roma e del lazio (a proposito di un volume di Mons. alberto serafini),” Faenza 16 (1928), 55–65. 31 Whitehouse, “The Medieval pottery of rome.”; Berti and cappelli, Lucca, pp. 141–158; lidia paroli, “spiral Ware,” in Archeologia Urbana a Roma: il progetto della 102 chapter four manufactured in one centre. at least three different areas of production in southern italy have already been identified: lazio, campania, and sicily. although only a few mineralogical analyses have been conducted on sardinian bacini, a generic link between them and those produced in these areas of italy seems absolutely obvious. spiral ware has been found in sardinia in small quantities and it has been suggested that it might have been imported by ligurian merchants (fig. 13).32 however, the possibility that spiral ware was imported from pisa should not be excluded. Thus, the dates of the spiral ware and related pottery on sardinian churches likely conform to the dates established for similar vessels in continental italy. almost every continental workshop that produced double-fired ceramics without ingobbio also made monochromes. such products generally contained no decoration, except for the colour obtained by the lead glaze, although some workshops used minor decorative elements, such as small incisions or stamps on the biscotto. a coat of lead or tin glazing mixture with colour was applied over the biscotto before its second firing. This technique was used in both islamic and italian centres. Mineralogical analyses are imperative when hardly any distinguishing element exists, which perhaps explains why monochromes have often been ignored in typological reconstructions. Morphological distinction is generally the rule, but the glaze application on the surfaces also helps to identify centres of production. Monochromes, when coloured, were commonly green and yellow and rarely dark brown, and they changed according to the firing process and the amount of metal agents added (iron, manganese, or copper). The sources and diffusion of green lead glaze in italy during the Middle ages was summarized by stella patitucci uggeri (apulia)33 and Maria vittoria fontana (naples).34 early monochrome products exist in regions under islamic dominance or influence and in sicily from the Crypta Balbi. 3, Il giardino del Conservatorio di S.Caterina della Rosa, ed. daniele Manacorda (florence, 1985), pp. 237–238. 32 rovina, La Sezione Medieval, p. 79. 33 stella patitucci uggeri, La ceramica medievale pugliese: alla luce degli scavi di Mesagne (Mesagne, 1977), pp. 92–102; eadem, La Protomaiolica: bilancio e aggiornamenti (florence, 1997). 34 fontana and ventrone vassallo, La ceramica medievale di San Lorenzo Maggiore in Napoli, 1:68–71. merchants, monks, and medieval sardinian architecture 103 second half of the twelfth century.35 continental italian monochromes were finally imported to rome by the beginning of the twelfth century and can be found in various locations in lazio.36 The excavation of the crypta Balbi in rome confirms that the earliest green lead glaze vessels came from southern italy. Northern Italy (Liguria) ceramics made with a special slip (engobled or slipped) technique originated and developed independently from those discussed above. Their surface is prepared with a whitish earthen slip, over which the lead glaze is applied during the second firing. The slip is necessary for certain types of ornament such as ‘graffito’, where portions of the slip are removed with fine points after the design is outlined, creating a dark-on-light design since the lead glaze is in direct contact with the impasto, which is generally darker. slipped bacini on sardinian churches are decorated with a ligurian technique known as graffita arcaica tirrenica (Gat). excavations in liguria (where evidence of production exists in the remnants of kilns around savona) date the earliest of these products to the late twelfth century.37 These are the earliest-known examples of a slippedware technique produced in italy. The introduction and diffusion of this technical process in the different regions of the peninsula is the subject of some of the most interesting research currently underway.38 although the theory that Gat originated in the eastern 35 franco d’angelo, “le ceramiche rinvenute nel convento di san francesco d’assisi a palermo e il loro significato,” Sicilia Archeologica 2 (1974), 65–73; idem, “le ceramiche spagnole,” pp. 141–152; antonino ragona, “le fornaci medievali scoperte ad agrigento,” Faenza 51 (1966), 83–89; idem, “le fornaci trecentesche per ceramiche invetriate scoperte a sciacca nel 1971,” Faenza 61 (1975), 3–6. 36 stella pattitucci uggeri, “saggio stratigrafico nell’area di san pietro degli schiavoni a Brindisi. relazione preliminare 1975–1976,” Richerche e Studi 9 (1976), 150 and pl. 13a–b; Whitehouse, “The Medieval Glazed pottery of lazio,” 40–67; idem, “The Medieval pottery of rome,” pp. 36–37. 37 Marco Milanese, “lo scavo archeologico di castel delfino (savona),” Archeologia Medievale 9 (1982), 106–107; rita lavagna and carlo varaldo, “la graffita arcaica tirrenica di produzione savonese alla luce degli scarti di fornace dei secoli Xii e Xiii,” in La ceramica graffita (albissola, 1989), pp. 119–130; hugo Blake, “The medieval incised slipped pottery of north-west italy,” in La ceramica medieval nel mediterraneo occidentale (florence, 1986), pp. 322–328; Berti and cappelli, Lucca, pp. 151–168. 38 Berti and Gelichi, “trasmissioni di tecnologie nel Medioevo.” 104 chapter four Mediterranean has been dismissed, it is possible that the technique spread to many places simultaneously, with liguria at one end and venice at the other.39 in sardinia, slipped bacini with and without graffiti are found in both the north and south. The church of santa Barbara, near sassari, which was consecrated in 1270–1279, is the only building with slipped bacini for which there is a precise date of construction. since slipped bacini were produced on the continent throughout the thirteenth century,40 the dates for santa Barbara confirm the italian origin of its bacini. Monochrome slipped ware without graffiti was apparently produced in the same workshops that made graffito slipped ware, as both have similar shapes and are made out of the same clays (fig. 14). These less-elaborately decorated, un-incised vessels suggest that there was a demand for simpler ornament. however, in sardinian churches, the two types are used interchangeably. Integration, substitutions, “antiquarian” bacini, and modern replicas a unique practice, apparent so far only in sardinia, is the substitution of old pottery for lost bacini.41 for example, the campanile of san pietro at Quartu sant’elena has a bacini recess filled with a seventeenth-century plate from Montelupo fiorentino, which was inserted at an even later date,42 while the sanctuary of Bonacattu has eighteenth- and nineteenth-century pottery randomly arranged on its façade.43 More radical restorations took place in sardinia during the 39 sauro Gelichi, “la ceramica ingubbiata medievale nell’italia nord-orientale,” in La ceramica medievale nel Mediterraneo occidentale (florence, 1986), pp. 353–407; Graziella Berti and tiziano Mannoni, “ceramiche medievali del Mediterraneo occidentale: considerazioni su alcune caratteristiche tecniche,” in A cerâmica medieval no Mediterrâneo ocidental (Mértola, pt, 1991), pp. 163–173. 40 Berti and tongiorgi, I bacini ceramici medievali, pp. 277–282. 41 in pisa most bacini have been detached from the churches and are now part of the Museum of san Matteo collection. These bacini were replaced by exact copies so that one can still admire the chromatic play of the façade of these early pisan churches. see Berti, “pisa.” 42 fausto Berti and Gianna pasquinelli, Antiche maioliche di Montelupo, secoli XIV– XVIII (pontedera, it, 1984), p. 83, pl. a, and p. 90. 43 a. cameirana and G. Bozzano, “la terraglia nera ad albisola all’inizio dell’ottocento,” in [Atti:] III Convegno internazionale della ceramica (albisola, 1970), merchants, monks, and medieval sardinian architecture 105 first quarter of the twentieth century; the bacini in ploaghe were eliminated during restoration, while modern replicas replaced old ceramics in santa trinità of saccargia at condrongianus (fig. 15), san nicolo at ottana, and santa Giusta at oristano. not many bacini on the facades of sardinian churches have survived the test of time. Between 1050 and 1300, over sixty churches were decorated with bacini. of these, only sixteen still contain bacini in situ (table 1). of the roughly ninety surviving bacini, some are still in good shape, others have been reduced to fragments, and all have been dated and assigned a provenance.44 Merchants in the western Mediterranean first began dealing in islamic glazed pottery in the late tenth century, but the earliest bacini found on sardinian churches (from tunis and sicily) date from the second half of the eleventh century. This represents a delay of about fifty years from the time bacini first appeared in continental italian churches, particularly those of pisa.45 as the islamic glazing technique started to be used in continental europe, bacini from both spain and italy start to appear in sardinia. By the fourteenth century, almost exclusively spanish bacini decorate the last churches with this type of ornament. it is with the arrival of the aragonese and their occupation of the entire island of sardinia at the beginning of the fourteenth century that trade and church construction practically cease. The Architecture of Sardinia sardinia’s medieval architecture differs from that of any other region along the Mediterranean coast. a regional and vernacular architecture—isolated both literally and metaphorically—it has been largely neglected by international scholars. on the other hand, indigenous historians have worked extensively on these buildings and focused on the connections and parallels to similar ecclesiastical architecture in pp. 61–114; frederico Marzinot, Ceramica e ceramisti di Liguria (Genoa, 1979), p. 280, nos. 323, 324. These were taches noir, white, and blue. 44 Michelle hobart, Sardinian Medieval Churches and their Bacini: Architecture Embedded with Archaeology (phd dissertation, new York university, institute of fine arts, 2006). in only two churches, the poor condition of the pottery has not permitted a precise attribution, but in both cases the likeliest source is early fourteenth-century spain. 45 Berti and tongiorgi, I bacini ceramici del Duomo. The church of san piero a Grado, along the coast of the old (and now buried by silt) harbour of pisa has a large number of bacini that are among the first to appear in italy in the late tenth century. 106 chapter four continental europe.46 in the 1950s, raffaello delogu and those who followed him claimed that sardinia’s medieval architecture originated with the arrival of new settlers that created a new “architectural language”.47 at first, sardinian workshops used tuscan models that were products of the renaissance in ecclesiastical architecture that took place between the eleventh and fourteenth century as power shifted from the Byzantine to the latin church. This is clearly evident in sardinia’s earliest medieval churches: san Gavino at porto torres, santa Maria at ardara, and santa trinita di saccargia at condrongianus (fig. 16).48 a second generation of buildings incorporated an insular, local taste, transforming these continental paradigms into a regional style, which resulted in a unique and eclectic architectural language. Bacini are but one of the decorative elements that resulted from this marriage. in the second half of the eleventh century, the holy see made a conscious effort to increase religious fervour in sardinia, so as to better control the island. it was no accident that the commission of san Gavino, the first major building project of the Gregorian reform in sardinia, was emblematic of architectural revisions that were happening contemporaneously on the mainland. however, the builders of san Gavino did not come from rome, but directly from pisa’s cathedral workshop. The architectural forms of this first generation of sardinian religious buildings were dominated by tuscan romanesque sources. in more than one instance, continental workshops built sardinian churches soon after they had completed similar ones on the continent. pisan merchants in sardinia, particularly in the north, were actively seeking to settle and gain favours from local administrators; in offering masons from their workshops, they subtly imposed their own architectural idiom on the island. not only did they make rich use of bacini, but they introduced the masonry technique known as muro a sacco: two parallel walls filled in between with rubble, into which the 46 raffaello delogu, L’architettura del medioevo in Sardegna (rome, 1953); roberto coroneo, Architettura romanica dalla metà del mille al primo ‘300 (nuoro, 1993). 47 delogu, L’architettura del medioevo, pp. 132–133, 241–247. delogu’s theory is confirmed by coroneo and serra. 48 documents show the invitation of the maestro de pedra to construct san Gavino at porto torres. The same atelier was sent to build santa Maria at ardara and perhaps santa Giusta near oristano. aldo sari, “il romanico nel guidicato di torres tra Xi e Xiii secolo,” in La civiltà giudicale in Sardegna nei secoli XI–XIII: fonti e documenti scritti (sassari, 2002), pp. 440–447; fernanda poli, La Basilica di San Gavino a Porto Torres (sassari, 1997), p. 64. merchants, monks, and medieval sardinian architecture 107 bacini and corbel tables were inserted during construction (fig. 17).49 coloured bands of stones, roman spolia, corbel tables, blind arcades, and similar windows also lent these buildings a tuscan character. although many of these elements appeared in other parts of central italy, pisa seems the most likely transmitter. Medieval sardinian churches were commonly single-nave stone structures, but there were also more elaborate plans with aisles and transepts.50 a different architectural source (i.e. Marseilles, via rome) was used for the double-nave churches, the so-called ‘victorine churches’, which have a distinctive plan and were restored or built ex novo during similar initiatives of the Gregorian reform (fig. 5).51 This type of plan is particular to the southern area of the island and seems to have been used to honor local saints, patrons, or martyrs. after the first pisa-driven generation of building on sardinia, churches like san pietro di sorres at Borutta (fig. 18), sant’antioco di Bisarcio in ozieri, santa Giusta at oristano (fig. 19), and santa Maria at Bonarcado—all of which are adorned with bacini—exhibited greater variety, as well as ties with other tuscan workshops.52 in sardinia, the early christian tradition of incorporating local spolia is evident in both san Gavino at porto torres and santa Giusta, near oristano, just to mention a few. however, the practice did not prevent masons from designing new capitals when old ones were no longer available, as can be seen in the latter. indeed, they introduced new architectural elements, such as half columns set against walls, a practice originating in pisa. elsewhere, spolia are the only significant adornments besides bacini, as at the church of san platano in villaspeciosa, where a Byzantine marble slab was placed in the centre of the façade (fig. 5). 49 Giovanna Bianchi, “trasmissione dei saperi tecnici ed analisi dei procedimenti costruttivi,” Archeologia dell’Architettura 1 (1996), 53–64; alma casula, “la chiesa di san Giorgio di oliastreto in agro di usini,” in La civilta giudicale in Sardegna (sassari, 2002), pp. 519–536, photo 2. 50 Gabriela frulio, “tecniche costruttive della sardegna medievale: il monumento come fonte per la conoscenza,” in La civilta giudicale in Sardegna (sassari, 2002), pp. 495–496. 51 The victorines were perceived as the most loyal order to rome. pier Giorgio spanu, “i possedimenti vittorini del priorato cagliaritano di san saturno. il santuario del martire efisio a nora,” in Citta, territorio, produzione e commerci, pp. 65–103. 52 sari, “il romanico nel guidicato di torres.” 108 chapter four The constant referral to pisa and pisan features in sardinian architecture might have served an ulterior purpose. Bacini may have been a means through which the mercantile class from pisa could display its power in sardinia. The glazed ceramic vessels that these merchants brought to sardinia must have been considered an exotic import and its supplier an investor who would facilitate the construction of churches for the benefit of the locals. Bacini could thus be considered ornaments by which pisan merchants could assert their role in the community and claim ownership in a complex and competitive socio-political environment. But, sardinia’s churches would also have been seen by foreigners who came to sardinia for natural resources or stopped at the island on the way to the Balearic islands, spain, sicily, or north africa.53 in this manner, the pisans could underscore their privileged status in the Mediterranean. This might not have been evident to sardinians who had never been to pisa, but it would have been obvious to those who traveled and exchanged goods in the Mediterranean. finally, the common assumption that in sardinia bacini were used as a cheap alternative for costly materials is unfounded. an ornament that seems simple to our eyes today may have been appreciated quite differently in the Middle ages. a similar argument has been made for a distant cousin of the bacini, namely Byzantine glazed tiles. analysing the historical transformation of this architectural adornment, anthony cutler has convincingly traced its evolution from the Greco-roman tradition to the post-roman era, when the geometrical patterns and colour variations of tiles were highly valued in their own right. The buildings in which glazed tiles were used as ornament reflected the aesthetic choice of the emerging rulers. as cutler points out, glazed tiles were also used in buildings such as the Mosque of cordoba, where there were no financial restraints.54 similarly, bacini were used experimentally on the duomo of pisa, among the largest Western cathedrals 53 Graziella Berti, catia rizzo, and Marco tangheroni, Il mare, la terra, il ferro: ricerche su Pisa medievale (secoli VII–XIII) (ospadelatto (pisa), 2004); david abulafia, “The pisan Bacini in the Medieval Mediterranean economy: an historian’s point of view,” in Papers in Italian Archaeology 4, (Bar international series) 246 (oxford, 1987), pp. 287–302. 54 anthony cutler, “tiles and tribulations: a community of clay across Byzantium and its adversaries,” in A Lost Art Rediscovered. The Architectural Ceramics of Byzantium, eds. sharon e. J. Gerstel and Julie a. lauffenburger (Baltimore, 2001), pp. 159–169. merchants, monks, and medieval sardinian architecture 109 built in the eleventh century.55 Thus, bacini were viewed as objects of prestige, not—as some have implied—an inexpensive construction material. Monks and Merchants in Sardinia after 1000 ad, the socio-economic context of sardinia changed quickly as it became a common rest-stop for Mediterranean travelers and traders in the island’s raw materials. This influx of foreigners, combined with the threat of invasion, the rising influence of mercantile families, the displacement of the Byzantines by the roman catholic church, the emergence of new cities, and finally, the arrival of diverse monastic orders, all shaped the social context for the unique, simple, and elegant local architecture of the period. at the turn of the first millennium, sardinia was (and still is) divided into four regions (Giudicati) that practically corresponded to the dioceses, each having a Giudice, or governor.56 When, precisely, the four areas became Giudicati is a contentious subject. While some historians believe that the division into Giudicati was a phenomenon that began with the island’s detachment from Byzantium, others claim that the island was reorganized for defensive reasons due to arab incursions in the eighth century, although one theory does not necessarily exclude the other.57 it is worth noting that despite frequent references to arab attacks on sardinia, very little evidence of their actual occurrence has come forth. however, there are records of arab incursions against cagliari, the most important southern port on sardinia, during the eighth century. it appears that a strong local resistance opposed these attacks until 752, when the umayyad abd er rhaman and his brother occupied cagliari and enforced the jizah (a tax imposed on 55 piero sanpaolesi, Il Duomo di Pisa e l’architettura romanica toscana dalle origini (pisa, 1975), p. 165. With the exception of speir, where the measurements are just slightly superior. 56 each Giudicato had different curatorie, administrative centres where the curator was the royal representative of the crown. The Giudice was considered a king, rex Sardiniae. see coroneo, Architettura romanica, p. 67. 57 rowland, The Periphery in the Center, pp. 151–153, 155; floris, Storia della Sardegna, pp. 152–156; Milia, “la civiltà giudicale,” in Il Medioevo: dai giudicati agli Aragonesi, ed. rafael conde y delgado de Molina and Massimo Guidetti (Milan, 1987), pp. 200–201. 110 chapter four the conquered).58 The ninth century also witnessed attempts at the occupation of the island by small and newly formed islamic groups from north africa and spain. Though an exercitus Sardinia—most likely a Byzantine legacy and by then in the hands of the local judex— was able to fend off the repetitive attacks, it is likely that during this period Muslim settlers were integrated with local communities, despite their alleged attempts to take over. in any case, during the ninth and tenth century sardinia seems to have enjoyed political autonomy and ethnic diversity, whether that diversity came about through hostile actions or not. at present, the numerous excavations conducted throughout sardinia have revealed sparse traces of mosques and other islamic settlement.59 This fact is ironic, considering the supposed centuries of Muslim presence in sardinia, especially along the coast. if the material record is blindly accepted, without regard for the aforementioned documentation of arab attacks, one hypotheses is that these so-called arab attacks may have constituted more of an excuse for continental mercantile families and the church to settle in sardinia. once again, it would appear that history has been written by those who were eager to justify their priorities as they engaged in conquest. archaeology has a great role to play here and at this time seems to reveal few traces of islamic physical presence on sardinia, perhaps due to the only recent interest in sardinia’s islamic past. however, the possibility that Muslim settlers were so assimilated into sardinian society that it is impossible to identify their presence in the archaeological record should not be excluded. after the schism with Byzantium in 1054, a major initiative of the roman church’s Gregorian reform was to reorganize land all over europe. The church focused on reclaiming its authority over continental territories that had become feuds of distant landlords or bishops and lands that had been forgotten or ignored by the imperial centres of Byzantium and rome.60 in sardinia, the church initiated an energetic campaign to restore and return to latin rite the religious 58 he was the nephew of the abd er rhaman, who died at tours in the 730s in a battle against the carolingians. floris, Storia della Sardegna, pp. 146–147. 59 Moriscos, pp. 28–30. 60 robert Bartlett, The Making of Europe: Conquest, Colonization, and Cultural Change, 950–1350 (princeton, 1993), pp. 243–250. merchants, monks, and medieval sardinian architecture 111 buildings that had been abandoned during the Greek Byzantine era.61 The influence of the roman church was secured in the form of donations and privileges exchanged between them and local entities. for instance, pope alexander ii (1060–1073) sent the first papal delegate to the synod in cagliari62 and desiderius, abbot of Monte cassino and future pope victor iii (1086–1087),63 aggressively expanded ecclesiastical properties by claiming to have received them as donations and listing them as part of the monastery’s possessions in the Chronica Monasterii Cassinensis in 1058.64 This same policy was followed by other monasteries, who added land and possessions to their holdings through inheritance or acquisition in order to gain better command over surrounding territory after the break with Byzantium.65 although the church was indeed motivated to instill religious reform in sardinia, one could argue that its actions were likewise prompted by the fear of not participating in the local competition over land, as numerous territorial concessions had been made by the Giudicati to lay individuals, such as merchants from pisa and Genoa.66 similar losses had already occurred in nearby corsica, where independent bishops and lay landlords from the mainland had appropriated for themselves most of the property on the island, a phenomena witnessed all over europe.67 it seems quite evident that the local donor of land was aware—at least during the early stages of this process—that he could lose control over his property to ambitious individuals, while an institution like the church would guarantee a reciprocal relationship and follow higher standards than an individual landlord. for example, the donation made by the judex Mariano i in 1082 of the church of 61 raimondo turtas, “la chiesa sarda dalle origini fino al periodo spagnolo,” in Storia della Sardegna, eds. Manlio Brigaglia, attilio Mastino, and Gian Giacomo ortu (rome, 2006), 1:116–130. 62 turtas, Storia della chiesa in Sardegna, p. 971. 63 ibid., p. 207; Giles constable, The Reformation of the Twelth Century (cambridge, 1996). 64 among the many other campanian churches and lands mentioned were those in sardinia. coroneo, Architettura romanica, p. 119. 65 see the discussion of san vincenzo al volturno in chris Wickham, Early Medieval Italy: Central Power and Local Society, 400–1000 (london, 1981); federico Marazzi and paolo delogu, San Vincenzo al Volturno: cultura, istituzioni, economia (Montecassino, 1996); richard hodges, Light in the Dark Ages. The Rise and Fall of San Vincenzo al Volturno (london, 1997). 66 Bartlett, The Making of Europe, p. 293. 67 turtas, Storia della chiesa in Sardegna, p. 197; Geneviève Moracchini-Mazel, Les monuments paléochrétiens de la Corse (paris, 1967). 112 chapter four san Michele of plaiano to the opera of santa Maria of pisa explicitly states that “no lay or religious person (bishop or archbishop as opposed to an ‘institution’) may benefit from this donation.”68 on the other hand, merchant families had very close ties with the Giudici (often marriages occurred or were arranged between the two factions) and could offer more resources in exchange for land holdings. obviously, it was preferable for the church that the Giudici make concessions to its potential allies, such as the tuscan administration of the opera of santa Maria in pisa, which had overseen the construction of the duomo, rather than the hard-to-control, non-religious institutions or powerful mercantile families, a situation that seems to have occurred frequently.69 however, there were no rules in place for such donations and every area changed according to different circumstances. The Giudici seem to have been well aware of how to play one group against the other, so as to maintain some form of control over their own land. in 1065, concerned by the number of concessions and donations given to pisa and Genoa, abbot desiderius sent twelve monks from Monte cassino to the Giudice Barisone i of torres (in the north) to establish a monastery. a decade later, in 1073, pope Gregory vii (1073–1085) wrote to the four Giudici of sardinia, reminding them of their obedience and dependence on the church.70 Gregory vii seemed threatened and realized that he had to accelerate initiatives by sending more monks to the island. That same letter contains the earliest documented donation of a church to a monastic order on the island.71 in 1087, desiderius, the powerful Bishop of Monte cassino and by then pope victor iii, pursued the same policy, urging the archbishop of cagliari (in the south) and other prelates to restore extant churches and construct new ones in sardinia in order to rouse religious fervor. coroneo, Architettura romanica, p. 117. on the relationship with the opera of s. Maria of pisa and the Giudicati, see rowland, The Periphery in the Center, p. 158, n. 113. 70 h. e. J. cowdrey, The Age of Abbot Desiderius: Motecassino, the Papacy, and the Normans in the Eleventh and Early Twelfth Centuries (oxford, 1983), pp. 10–11, 65, and 70. 71 coroneo, Architettura romanica, pp. 117–121; renata serra, La Sardegna (Milano, 1989), p. 402. a donation was an official document that offered a particular order or bishop a church to enhance the religious life in a Giudicato. a financial compensation was then offered to the church for its “services” for the development of the area. 68 69 merchants, monks, and medieval sardinian architecture 113 churches were lacking and in need of attention after such a long period of neglect by Byzantine and local rulers.72 soon after, more concessions were made by sardinians to other continental religious institutions. various monastic orders began arriving in sardinian towns and the countryside to build churches ex novo. These orders—the Benedictine, the victorine, the camaldulensis, the vallumbrosian, the cistercian, and the franciscan—settled in different regions of the island. The french victorines occupied most of the south; the franciscans the southwest; and the camaldulenses and vallumbrosians the north. Most of these orders had close ties with pisa and some with Genoa.73 There were also other monastic orders scattered throughout sardinia, such as the Basilian, those from Monte cassino, and the Girosolimitani. The sprawl of churches over all of sardinia, a process that was characteristic of religious expansion during the Middle ages, was supported by strong local administrations that, at first, welcomed and invited foreigners to invest in the island. These solicitations provided infrastructure to rural areas and contributed to agricultural ameliorations. it is worth re-emphasizing the power of the opera of santa Maria of pisa in this religious reorganization of sardinia. in the first half of the eleventh century, pisa was building one of the most impressive cathedrals in western europe and was clearly showing signs of independence from rome. in that vein, it is interesting to note that some sardinian martyrs and saints were actually held in pisan churches and not housed on the island or transferred to rome.74 it was pisa’s strong economic position, a result of its successful mercantile exchanges, that fueled this religious expansion into sardinia. While it is difficult to speculate about the role of the merchant class in ecclesiastical construction, certainly trade both benefitted and necessitated the evolution brought about by the church, although the two factions had a more conflictual relationship. What is remarkable about the bacini as a social indicator is that in both surviving architecture and the archaeological record, islamic bacini do not appear until after 1050, when competition between turtas, Storia della Chiesa in Sardegna, pp. 183, 189. ibid., pp. 180–192 and 216–221; rowland, The Periphery in the Center, p. 167. 74 The relics of saint Gemiliano were preserved in s. Matteo in pisa, yet venerated in sestu (with bacini) and in samassi in sardinia. sanpaolesi, Il Duomo di Pisa, pp. 94–101. 72 73 114 chapter four monks and merchants began in sardinia. Why, therefore, on an island where there was likely an islamic presence from the eighth century, does a clearly islamic form of decoration only appear with the arrival of continental settlers? Beginning in the second half of the tenth century, first with the church of san piero a Grado and later the duomo of pisa, continental architects incorporated elements of islamic visual culture into their buildings. during the interim between the fall of Byzantium and the rise of the roman church, pisa used arab symbols as a reflection of their power and the breadth of their growing maritime republic. pisa may also have been hinting at their privileged relationship with north africa and possibly sardinia. What is interesting is that when pisan monks and merchants arrived in sardinia and began constructing and re-constructing churches, they used the same islamic symbols, namely bacini, to represent themselves in a region where there was probably already an established islamic presence. Thus, the bacini represents an odd colonial interplay, whereby the conquering nation, in this case pisa, re-represents the culture of the repressed as a symbol of their own dominion. chapter Five the FORMA URBIS oF aleppo (Syria) during the Middle ageS1 giulia annalinda neglia Introduction like many cities in Syria, aleppo shows a continuity of growth from the hellenistic to the medieval islamic city, via the roman and Byzantine city. Moreover, its urban structure continues to recall such urban models even today. it is possible to decipher this palimpsest structure and the traces of each phase of its urban development through a structural interpretation of the building fabric. This can be done using a research methodology based on notions of process typology deriving from the theories developed in italy in the 1960s by Muratori and his school.2 This interpretative method rests on the conviction that it is possible to ‘read’ the traces of human organization of a unique urban or territorial entity in the structure of its building fabric. Such traces are not necessarily archaeological remains but they may instead correspond to traces of urban plans that have remained unaltered through time along the original plot boundaries—for example, new walls constructed on ancient foundations—and, in the surrounding countryside, property boundaries, channels, routes, or even rows of trees that are still legible today in the structure of the urban and agrarian fabric. in other words, the reading of the urban fabric that derives from this analytical method, as well as the interpretation of its forms in different historical periods of human organization, is partially independent of the 1 This article is based on doctoral studies carried out at the icar department of Bari polytechnic, as part of the phd programme in Architectural Design for Mediterranean Countries. The title of my doctoral thesis is Città del Mediterraneo: Aleppo, forme e tipi della città intra moenia. [Mediterranean cities: Aleppo. Forms and types of the city intra moenia], and my supervisor was prof. attilio petruccioli. 2 See especially g. caniggia, Lettura di una città: Como (rome, 1963); g. cataldi, p. iacono, and a. Merlo, “la geometria di firenze e il progetto matrice della città e del territorio,” Firenze Architettura 1 (2000), 4–17. 116 chapter five discovery of archaeological remains, since it is based on the presupposition that the history of a city is inscribed in its urban fabric. The reading of the different phases of human organization of medieval aleppo has been conducted as an historical, spatial, and temporal identification of the type-morphological process that invested the fabric of the medieval city within the perimeter of the Mamluk walls and determined the transformation of its building types and urban structures from those inherited from hellenistic-roman plans. it has been necessary to adopt such a methodology, since historical and archaeological sources do not document the different phases in the evolution of the urban structure of medieval aleppo. instead, they tend to refer almost exclusively to the history and archaeology of its monuments. Thus, they both leave a void in terms of information on the growth and transformation of the urban fabric in the Middle ages, and hide the nuances of its development under a hypothesis of homogeneous urban growth from aleppo’s hellenistic foundation to the Mamluk city. using an interpretative methodology allows us to compensate for the lack of historical and archaeological data on the structure of the urban fabric of aleppo in pre-islamic times and to integrate the existing sectorial and discontinuous data on the structure of its urban fabric through time. The reading of the urban fabric of aleppo has been carried out in different stages and entails relating the orthogonal alignments of the building fabric (in order to find planned phases of urban development) to the spontaneous routes in the urban fabric (in order to find traces of spontaneous expansion subsequent to planning). furthermore, the structure of the urban fabric has been measured using different units of measurement (such as the greek foot or the roman actus and centuria or the Mamluk cubitus), corresponding to the different phases of urban development, in order to have a metrical confirmation of our hypotheses. reading the city in horizontal sections and dividing the urban fabric on the basis of its different phases of development, it has been possible to find within the urban organism all the phases in the process of superimposition and transformation that have determined its form and structure, even though their boundaries seem indistinguishable today. in particular, it has been possible to carry out a structural analysis of aleppo, since the passage from one form to another, in the different phases of human development, came about gradually and without the ruptures caused by long periods of abandon. My analysis has used the french cadastral surveys and maps from the 1930s, which chart the forma urbis of aleppo 117 the form of aleppo before its modernization. The findings from this cartography have been compared with those from other documents, including the following: maps of aleppo drawn at different scales, in order to corroborate evidence on an urban, territorial and neighbourhood level; thematic maps, in order to corroborate evidence from previous studies on aleppo; historical maps of aleppo; and historical photographs and images at various scales. The data from this cartographic analysis have been constantly compared with data of a historico-archaeological nature in the available bibliographical material and corroborated on location with photographic and metrical surveys of the urban fabric. finally, this information has been compared with data on the forms of Syrian cities that have structures similar to that of aleppo. from this interpretation of the urban fabric of aleppo within the perimeter of the Mamluk walls, the permanence of traces of planning dating to roman times stands out in the medieval urban fabric. These traces are not immediately legible as they derive from territorial plans from different phases, closely dependent on the form of the site and territory of aleppo, and have determined an apparently unplanned urban form that, over a long period, was linked to the ‘spontaneous’ structure of medieval islamic cities. however, the application of an interdisciplinary methodology enables one to untangle these traces and organize them into a coherent narrative of the city’s history. The Hellenistic-roman city in order to reconstruct the medieval urban fabric of aleppo in its different phases of formation, building densification, and nodalization, one needs to begin by deciphering the structures of pre-islamic urban and territorial planning that represent its substratum. The first significant phase in the human organization of the urban fabric and territory of aleppo corresponds to the hellenistic period between 301 and 281 Bc, when Seleucus i founded the settlement of Beroea (the greek name for aleppo) near the tell el-akabé, the site of the pre-historic settlement of aleppo.3 3 See J. Sauvaget, “le ‘tell’ d’alep,” in mélanges syriens offerts à monsieur rené Dussaud par ses amis et ses élèves, ed. académie des inscriptions & belles-lettres (france) (paris, 1939), 1:59–63. 118 chapter five The plan of Beroea in its Seleucid form4 was one of a mono-directional city, whose blocks (47.2m × 124m)5 were perpendicular to the main route (via recta), which was laid out in an east-west direction and linked two existing sites: the acropolis, connected to the encircling walls, and the tell (fig. 1). during the roman period, Syrian territory as a whole underwent different planning phases, and was organized according to roman agricultural planning methods. in this phase there were few colonies founded ex novo. instead, the romans simply enlarged and reorganized existing cities. however, these were at times so extensive as to be comparable to new foundations based on imperial urban canons, but with variations resulting from pre-existing natural and human conditions. in aleppo, three different planning phases date from the roman period.6 These phases can be traced in the orthogonal alignments 4 for more on the history and urban structure of Seleucid cities in Syria, see J.-c. Balty, “apamée et la Syrie du nord aux époques hellénistique et romaine,” Revue du Monde Musulman et de la Méditerranée 62 (1994), 15–26; e. frézouls, “observation sur l’urbanisme dans l’orient syrien,” Annales Archéologiques Arabes Syriennes. Revue d’Archéologie et d’Histoire 21 (1971), 231–243; J. d. grainger, The cities of Seleukid, Syria (oxford, 1990); h. lacoste, “la restitution du plan antique d’apamée in Syrie,” Bulletin de l’Académie Royale du Belgique, classe de Beaux-Arts 43 (1960), 53–62; p. leriche, “les fortifications grecques et romaines en Syrie,” in Archéologie et Histoire de la Syrie II. La Syrie de l’époque achéménide à l’avènement de l’Islam, eds. J.-M. dentzer and W. orthmann (Saarbrücken, 1989), pp. 267–282; idem, “le phénomène urbain dans la Syrie hellénistique,” Bulletin d’études orientales 52 (2000), 99–125; p. Monceaux and l. Brossé, “chalcis ad Belum: notes sur l’histoire et les ruines de la ville,” Syria 6 (1925), 339–350; f. e. peters, “city planning in greco–roman Syria: Some new considerations,” Damaszener Mitteilungen 1 (1983), 269–277; J. Sauvaget, “esquisse d’une histoire de la ville de damas,” Revue des études islamiques 8 (1934), 421–480; idem, “le plan de laodicée-sur-Mer,” Bulletin d’études orientales 4 (1934), 81–114; idem, “le plan de laodicée-sur-Mer (note complémentaire),” Bulletin d’études orientales 6 (1936), 51–52; idem, “le plan antique de damas,” Syria 26 (1949), 314–358; e. Will, “les villes de la Syrie à l’époque hellénistique et romaine,” in Archéologie et Histoire de la Syrie II, pp. 223–250. 5 The unit of measurement needed to reconstruct the form of the city in this phase of urban development is the greek foot. using this measurement we can obtain the dimension of the streets (8ft and 16ft wide, or 2.36m and 4.72m) and the dimension of the blocks of the settlement (160ft × 420ft, or 47.2m × 124m). These dimensions were first verified by e. Wirth in: h. gaube and e. Wirth, Aleppo: historische und geographische Beiträge zur baulichen Gestaltung, zur sozialen Organisation und zur wirtschaftlichen Dynamik einer vorderasiatischen Fernhandelsmetropole (Wiesbaden, 1984), and differ from those found by J. Sauvaget in: Sauvaget, “le plan de laodicéesur-Mer,” pp. 81–114, which were 46m × 120m. 6 references for aleppo in this phase of urban development are found in h. gaube, “aleppo zwischen alexander dem großen und der arabischen eroberung,” in Damaskus-Aleppo: 5000 Jahre Stadtentwicklung in Syrien, ed. M. fansa (Mainz, 2000), pp. 101–107; S. guyer, “la madrasa al–halâwiyya à alep,” Bulletin de l’Institut the forma urbis of aleppo figure 1: plan of aleppo at the time of the hellenistic foundation. 119 120 chapter five of the building and agrarian fabric by measuring the urban fabric, analysing the route systems, and consulting historical sources. in the first phase of reorganization, the romans followed the direction of the plan and the north-south orientation of the hellenistic city, though they changed its structure. The orthogonal alignments of the urban fabric in this arrangement, recognisable in the cadastral records, are widespread in the walled city, especially in the fabric to the east of the citadel in an area that, up to now, has not been considered as part of a systematic urban development prior to the ayyubid period.7 from the measurements of the building fabric and an analysis of the route systems using the dimension of the actus (35.6m), one finds that the urban structure changes suddenly to the east of the hellenistic agora, which was transformed into the roman forum. Beyond this area, it is no longer possible to detect the rhythm of the hellenistic urban structure. for example, the width of the two blocks immediately to the east of this area is 71m, corresponding to two acta. The same dimensional rhythm may be read in the lay-out of many routes within the urban fabric. an analysis of the structure of the territory around aleppo reveals the presence of the same homogeneous and isotropic structure, based on a 710m × 710m module and its sub-divisions, in both the rural areas and some of the routes to the south and east of the city boundaries. The main axes in laying out this urban and territorial framework seem to have been the intersection between the hellenistic via recta, which connects the Bab antakia to the citadel, and the route along the eastern boundary of the first roman block next to the roman forum. in this phase of urban development, the via recta had become the roman decumanus and was transformed into a colonnaded street, while the second route probably became the cardus. Français d’Archéologie Orientale 11 (1914), 217–231; Sauvaget, “le ‘tell’ d’alep,” pp. 59–63; idem, Alep, essai sur le développement d’une grande ville syrienne, des origines au milieu du XIXe siècle (paris, 1941); idem, “l’enceinte primitive de la ville d’alep,” Mélanges de l’Institut Français de Damas 1 (1929), 133–159; idem, “Les perles choisies” d’Ibn ach-Chihna (Beyrouth, 1933. reprint, frankfurt am Main, 1993); idem, “Les trésor d’or” de Sibt Ibn al-’Ajami (Beyrouth, 1950). 7 See idem, Alep, essai sur le développement; gaube and Wirth, Aleppo; a. gangler, Ein traditionelles Wohnviertel im Nordosten der Altstadt von Aleppo (tübingen, 1993). none of them have considered the area east of the citadel as an urban feature before the ayyubid and Mamluk periods. the forma urbis of aleppo 121 also in this phase, the territory of aleppo was organized according to roman planning methods (centuriatio) and was oriented secundum coelum—that is, with a north-south, east-west orientation—in an isotropic manner (fig. 2). due to the close relation between the urban and the territorial structure legible in this planning phase, it is plausible that the encircling walls of the hellenistic city were demolished. in an empire in which safety was assured by the Pax romana, the defensive city walls were now unnecessary and would have been replaced by defensive structures on a territorial level along the eastern borders of the empire. one can assume that this phase corresponds to the period between the first century Bc and the first century ad when, after the conquest of Syria in 64 Bc, the transformation of Syrian territory began. This first roman planning phase thus represents a kind of ‘squaring’ of the city and territory of aleppo and thereby the establishment of a structural grid for successive urban and territorial planning phases. The structure of aleppo in pre-islamic times is also made up of other forms and ‘markers’ that were superimposed on this scheme and that become visible through a structural reading of the urban fabric. Because these forms and structures are derived from different territorial planning phases and are closely dependent on the character of Syrian territory and the configuration of the urban site, they are not immediately legible. an examination of the orthogonal alignments of the urban fabric reveals that, to the south of the great Mosque, starting from one of the routes of the hellenistic settlement, the direction of the urban fabric changes suddenly. The axis of Bab Quinnasrin street deviates by 18° from the north-south orientation.8 The same deviation is noticeable in relation to the axis of Bab al-nasr street, and is also present to the south and east of the citadel, and outside the Bab antakia (fig. 3). one can assume that work carried out by the romans between the first and second century ad included the construction of a major road 8 This rotation of the urban fabric of aleppo was first noted by eugen Wirth and then by anette gangler in their books on aleppo. See gaube and Wirth, aleppo, pp. 124–126; gangler, Ein traditionelles, p. 25. Wirth attributed a deviation by 23°–32° of the streets south of the citadel and a deviation by 23° to the west of the Bab al-nasr axis to urban development in the Mamluk period when this area was enclosed within the new city walls. Starting from Wirth’s consideration, gangler attributed this rotation to the permanence of what she defined as “very ancient urban structures,” which were nevertheless not chronologically determined. 122 chapter five figure 2: The roman plan secundum coelum. the forma urbis of aleppo figure 3: The first roman plan secundum naturam. 123 124 chapter five system linking aleppo to other roman cities in Syria. What one reads on the cadastral map are the traces of the urban fabric organized in these phases of urban development along the axis of a main route following the direction of Syrian territory.9 Since the romans did not use degrees in their computations, the main axis (cardus) of this new territorial system, based on a 710m × 710m module, may be described as the hypotenuse of a right-angle triangle with a 1:3 ratio (whose catheti correspond to 1 and 3 centuriae and the modules of the planning grid secundum coelum), which has its origin at the intersection between the cardus and decumanus of the earlier planning phase and coincides with Bab al-nasr street and with the main route of the settlement in the neighbourhood outside the walls to the north of the city (alJabbaleh, al-Qawas, Qastal Musht, al-Bassatina, Qastal harami). This axis is set at a distance of two acta from Bab Quinnasrin Street, the gateway to chalcis. The decumanus in this system coincides, outside the walls, with a route that starts from the Bab antakia, the gateway to antioch. an orthogonal of the decumanus coincides with the axis of the entrance to the citadel. The layout of the city deriving from the construction of these routes marks the first major hiatus in the urban growth of aleppo in that it is not continuous with the existing fabric, but has rotated it secundum naturam, that is, according to the direction of Syrian territory (fig. 3). in the territory and urban fabric of aleppo one can still find traces of a further plan secundum naturam. a reading of the orthogonal 9 K. Miller, Itineraria Romana: römische Reisewege an der Hand der Tabula Peutingeriana (rome, 1916); idem, Tabula Peutingeriana. Weltkarte des Castorius genannt die Peutingerische Tafeln Farben des Originals herausgegeben und eingeleitet von Dr. Konrad Miller (ravensburg, 1888). to trace the history of aleppo in this phase the Tabula Peutingeriana and the Itinerarii Antonini can be used: they trace the role of aleppo as a centre for roman viability. on a map from the Tabula Peutingeriana depicting the route network of territories known by the romans before the third century ad, aleppo is called Bethina and is represented by a symbol of two towers. to many scholars, this symbol indicates the nodal points of the roman road system. This data in conjunction with the information given by the Itinerarii Antonini, represents an important source for definition of the role played by aleppo in roman times. indeed, the city had an important role as a point of intersection between the caravan routes and roman planned routes. a reconstruction of the road system in Syria during roman times, according to the information derived both from the Tabula Peutingeriana and the Itinerarii Antonini, has been made by e. Mouterde and a. poidebard. See e. Mouterde and a. poidebard, Le limes de Chalcis; organisation de la steppe en haute Syrie romaine (paris, 1945); t. Bauzou, “les routes romaines de Syrie,” in Archéologie et Histoire de la Syrie II, pp. 205–221. the forma urbis of aleppo 125 alignments of the urban fabric reveals that within the walled Mamluk city, the urban fabric deviates by about 10° from the north-south orientation along the walls—which are not orthogonals and do not have the orientations we shall discuss below (fig. 4). This further deviation is not as immediately evident as the one described above, since it can be easily confused with the spontaneous fabric. it is, however, too widespread and extensive along the boundary walls to be casual, spontaneous, or unplanned. Starting from the network secundum coelum, one can identify the direction of the main axis of this territorial system (of 710m × 710m), which deviates by about 10° to the west from the north-south axis, and the cardus of which is an axis passing through the Bab antakia. near this axis, corresponding to the replanned urban fabric of the tell, there are archaeological remains dating from the roman period and corresponding to the wall alignments that follow this framework. The decumanus in this system may have been an axis passing through the Bab antakia. This planning phase seems to have been determined in part by the need to adapt the agrarian organization of the territory, which would later be integrated in the urban structure, to the main direction of the valley of the Quweiq river—in other words, to the centuriation of the valley. The derivation of the structure of the later urban fabric of islamic aleppo from this roman agrarian plan makes it easy to verify the actus as a unit of measurement. in particular, the perimeter of the Mamluk walled city, the boundaries of which were aligned along the traces of roman-Byzantine structures, coincides with a square of four centuriae (fig. 4). The traces of these three different roman plans10 represent the substratum of the structure and fabric of medieval islamic aleppo. The process of medievalization of this roman substratum began already in the Byzantine period through the construction of a new circuit of walls 10 The palimpsest of signs that makes up the territory of aleppo in the roman period is not an anomaly. The superimposition of several centuriations in the same territory is a sequence typical of roman territorial planning, verifiable in other Syrian and Mediterranean areas. traces of superimposed centuriae are charted for southern-central italy in cataldi, iacono, and Merlo, “la geometria di firenze,” pp. 4–17; g. chouquer, structures agraires en italie centro-méridionale: cadastres et paysage ruraux (rome, 1987). in Syria, traces of centuriae have been found in damascus by M. dodinet, J. leblanc, J.-p. vallat, and f. villeneuve, “le paysage antique en Syrie: l’exemple de damas,” syria 67 (1990), 339–355; and in homs by W. J. van liere, “ager centuriatus of the roman colonia of emesa (homs),” Les annales archéologiques de syrie: revue d’archéologie et d’histoire 8 and 9 (1959), 55–58. 126 chapter five figure 4: The second roman plan secundum naturam. the forma urbis of aleppo 127 and the consequent phase of urban contraction and building densification (of the building fabric). The medieval byzantine City The beginning of the process of contraction and medievalization of the urban fabric of aleppo can be dated to the sixth century ad. This phase belongs to the reign of Justinian and is chronologically later than the earthquake of 526 ad and the persian invasion of 540 ad led by Khosrau i, events that mark the first moment of arrest in the urban growth, which had been virtually uninterrupted in the long roman phase. The work of urban and architectural renewal following the persian invasion consisted of rebuilding the cathedral, restoring the colonnaded street, rebuilding the defensive system of the citadel, and rebuilding the fortifications and reinforcing them with a system of double walls and trenches in the weakest areas to the south of the citadel.11 it is likely that it was after these historical events that the medieval encircing walls were built. The lines of these walls would remain virtually unchanged from the Byzantine reconstruction of the city to the Mamluk period, systematically reinforcing the roman defensive system, which until then had been only temporary and had consisted of palisades and trenches. Through a combined reading of the written historical sources and the different roman layouts of the urban fabric of aleppo, one can hypothesize that part of the Byzantine defensive system of double walls coincided with and therefore was built on traces of a roman fortification, which took the form of a castellum situated on the south side of the acropolis.12 The main axis of this fortification coincides with the decumanus of the second roman planning phase, while its counteraxis corresponds to the entrance structure of the fortified ayyubid citadel (the hellenistic-roman acropolis). The Byzantine trench (Kandak al-rum), on the other hand, coincided with the plot boundaries on the eastern and northern side of the square of four centuriae, which represents the city boundary of the third roman planning phase of the territory of aleppo (fig. 5). in this phase, Jewish neighbourhoods 11 See Sauvaget, alep. for a reconstruction of the building phases of the aleppo city walls, see idem, “l’enceinte primitive,” 133–159. 12 128 chapter five figure 5: Byzantine fortifications and urban fabric. the forma urbis of aleppo 129 formed near the Bab al-nasr (the ancient Bab al-yahud) and around the synagogue, thereby demonstrating the effective renewal of building activity and the beginning of the process of densification of the building fabric within the new circuit of walls. The traces of this phase of urban reorganization, which are legible through a structural reading of the building fabric, especially in the Bashite neighbourhood, are useful in verifying the position of the Byzantine circuit of walls within the urban fabric. in this phase, the most important example of the beginning of the process of medievalization at the heart of the ancient city is the construction of the cathedral near the roman forum (the agora of the hellenistic settlement). The first nucleus of the Byzantine cathedral was based on the typology of Byzantine martyria found in the Syrian settlements of Bosra, Sergiopolis, Zeugma, and Seleucia.13 This first centrally-planned building was reorganized under Justinian following a longitudinal plan with three domes along the main axis, a nave and two aisles, an apse to the east, and a portico to the west. The area of the roman forum not occupied by the cathedral was used as a cemetery. one can reconstruct the position and role of this building within the urban fabric through a morphological and dimensional reading of the building fabric around what is now the great Mosque, as well as from historical and archaeological data. from such a reading, morever, it emerges that the nodal axis of the cathedral was placed at 71m from the axis of the colonnaded street. This measurement seems to confirm the hypothesis of an earlier roman reorganization of the fabric of the hellenistic plan at the heart of the city (fig. 6). The medieval islamic City When the arabs conquered aleppo, they took over a city and a territory that already had a disordered form that reflected the co-existence of the various superimposed roman plans (fig. 7). They occupied the traces of the roman agrarian and urban organization, and starting from this organization developed the new features of the medieval islamic city. as a result, in order to describe the transition from the romanByzantine to the islamic organization of aleppo, one needs to examine 13 See guyer, “la madrasa”; e. herzfeld, inscriptions et monuments d’alep, vol. 2 (cairo, 1954). 130 chapter five figure 6: process of transformation of the Byzantine cathedral into the al-halawiyya madrasa. the forma urbis of aleppo 131 figure 7: relation between the traces of pre-islamic aleppo and the layout of the ayyubid and Mamluk urban walls. 132 chapter five a time span of around four centuries. The image that emerges from a morphological reading of the building fabric and an examination of the historical data indicates that long after the Muslim conquest, and more especially for the whole of the umayyad period (636–750), there were no substantial changes made to the structure of the urban fabric. after this period, however, there began a slow process of transformation that would assume visible form only in the tenth century, following a change in the social and civic order of the population of aleppo and the region. one can identify at least three macro-phases in the type-morphological process that characterized the transition from the structure of the roman-Byzantine city to the medieval islamic city, and these will guide our structural reading of the building fabric of medieval aleppo. First Phase The first phase corresponds politically with the arab conquest of aleppo and historically with the period from 636 to 837 ad, during the umayyad and abbassid caliphates. in 636–37, the city was captured by Muslim troops under the command of Khalid ibn al-Walid. The occupation came about peacefully and left intact the existing civic structure, which at that point was already completed formed. The umayyad caliph, who had established his capital in damascus, came from the very same Byzantine culture of which he was the direct heir. This meant that the urban layout of the Byzantine city remained virtually unchanged for a long time. during the umayyad caliphate, the importance of aleppo and its consequent urban growth were limited as it did not have a prominent political or administratative role. in this period the structure of the urban fabric of aleppo did not undergo any radical changes. This was neither because the umayyads did not have the cultural wherewithal to change the urban organization of the conquered cities, nor because they considered the hellenisticroman city as the maximum expression of urban civilization. instead, the reason for this apparent immobility can be attributed to the fact that the umayyads came from roman-Syrian cities, which served as both the centres of their caliphates and their cultural points of reference. even though there was no change in the urban form in this phase— and it seems that in the very early islamic period the pre-islamic features of the urban structure of aleppo remained unchanged too—a the forma urbis of aleppo 133 series of minor modifications were carried out at an architectural level. These changed the consolidated sense of the Byzantine city and are legible at the level of the urban fabric. for example, the first mosque of aleppo (al-Shuabiya) was built near the Bab antakia, at the end of the roman colonnaded street and encroaching on the roman monumental arch.14 This operation, while representing a mere change in function of a monumental building, brought about considerable morphological changes in the urban fabric at the centre of the city, altering the structure of the main street axis of the city, which was now blocked at both ends. on the Bab antakia side, the colonnaded street was blocked at one end by the presence of the mosque and at the opposite end by the presence of the citadel. from this moment onwards, the process of transformation of the colonnaded street into a suq began: the space of the colonnades and the central lane were initially encroached upon by commercial building units that transformed the main street axis of the city into two, three, and four parallel commercial roads. This process of transformation of the colonnades corresponds to a phenomenon common in other hellenistic-roman Syrian cities.15 Since the commercial vocation of the colonnaded street was already affirmed in the imperial age, the congestion of its colonnades with shops corresponds to the formalization of a phenomenon that had already been widespread for centuries.16 14 The use of such a small structure as the main place of prayer in the city shows that for many years the Muslim troops were limited to a small number of soldiers who did not need a large space for prayer. See Sauvaget, alep, pp. 74–75. 15 See Sauvaget, “esquisse d’une histoire,” 421–480; Sauvaget, “le plan de laodiceesur-Mer,” 81–114. 16 See J. c. david and M. al-dbiyat, “la ville en Syrie et ses territoires: héritages et mutations,” bulletin d’études orientales 52 (2000), 22–23. in this article, the authors refer to M. rodinson’s thesis on the origin of the suq in his preface to p. chalmeda, El señor del zoco en España: Edades media y moderna, contribución al estudio de la historia del mercado (Madrid, 1973). according to robinson, the akkadian word ‘suq’, of Semitic origin, evoking narrowness, vaguely indicated streets and public routes in general. Therefore, the alexandrian translators mainly translated ‘agora’ as ‘suq’. Moreover, Semitic languages of the hellenic east, especially aramaic, used the word ‘suq’ to translate the hellenistic-roman concept of agora-forum. if the word ‘suq’ was indeed used to translate the term ‘agora’ as early as the third century Bc, then the suq was already the place par excellence of concentrated commercial and urban activity; the nerve centre of municipal life and business transactions, as well as a meeting-place for urban dwellers. Such a hypothesis would place the origin of the term at least a thousand years prior to the Muslim conquest. 134 chapter five one can reconstruct this process through a morphological reading of the building fabric of the main axis of the central suq of aleppo as it is today, on the cadastral survey, and through a comparative analysis with other examples in Syria. although the lack of historicoarchaeological data prevents an exact understanding of its phases, the process of congestion and commercialization that invested the main route network of aleppo is similar to that which invested the colonnaded streets of other Syrian cities such as laodicea, apamea, and palmyra, even though in the last two cases this process was crystallized at an early stage of development. looking at the case of palmyra, it is apparent that in late antiquity a series of shops with entrances from the side lanes of the colonnaded street had already encroached on the central lane, thereby maker it narrower. likewise, in aleppo, a reading of the structure of the urban fabric reveals that the shops of the suq conserve—in the rhythmic dimensions of the width of the shops and the thickness of the walls—the dimensions of the original intercolumnal structure and the diameter of the columns of the colonnaded street, in spite of the large amount of reorganization that took place over time in this part of the city (fig. 8). We cannot say exactly what happened to the building fabric in this phase, but it was certainly not one of urban expansion. after the conquest of antioch in 636, aleppo was cut off from the main international trade routes and its period of economic decline began. almost a century had passed since the capture of aleppo before the need was felt to build the first congregational mosque. in 715 ad, however, when the Muslim presence in aleppo increased, a larger mosque was needed. The great Mosque was built on the site previously occupied by the hellenistic agora, the roman forum, and Byzantine cathedral cemetery, while the heart of the city centre was chosen for the site of the friday Mosque. in this way, the former site of political activity and the sacred place of christianity, was replaced by a building symbolizing the new political and religious power. The construction of the great Mosque on the site of the hellenistic agora (later the site of the roman forum) was facilitated by the fact that throughout the Middle east the qibla faces south and so it was easy for Muslims to adapt pre-islamic urban structures and specialized buildings for worship, while at the same time avoiding any significant reorganization of the urban fabric and surrounding buildings. in the case of aleppo, an analysis of the morphology of the urban fabric and a comparison with similar cases in the same region reveal the forma urbis of aleppo figure 8: traces of the colonnaded street in the urban fabric of the central suq. 135 136 chapter five that the congregational Mosque occupied the southern part of the roman forum, a large empty space at the centre of the urban organism, using as its main entrance one of the entrances to the forum on the east side of the courtyard of the great Mosque (the sahn). later, in the abbassid period (750–865), a new entrance was opened on the west side of the sahn, while at the same time a secondary route along the axis of the suq was built. in the Zengid period (1128–1174), the entrance to the al-halawiyya madrasa—built by adapting the remains of the Byzantine cathedral—was opened on this new route (fig. 6). during the abbassid caliphate, the process of spontaneous development of the urban fabric, typical of the medieval city, was accentuated. in a phase of urban growth in which commerce began to become the most important aspect of the city, transferring the functions of the agora to the suq and the specialized buildings around it seems to correspond to the initial phase in a process of monumentalization of the agora that would eventually turn the suq into grand architecture on an urban level in the ottoman period. in this phase, the first khans for housing merchants and their wares were built next to the suq. The prototype of the khan is to be found in the agora, the function and structure of which is reproduced on a smaller scale in this new building type. With the specialiazation of three sides of the great Mosque as part of the suq, commercial activity became closely linked to public and religious activity. This is demonstrated by the permanence on the site of the ancient agora of the qaisariya, a covered and open market for the sale of fabrics that continued to function within the enclosure of the great Mosque until the twelfth century. The architecture of this building may derive from the adaptation of the space and form of the roman basilica.17 another phenomenon in the permanence and re-use of the places and spaces of the roman-Byzantine city in medieval aleppo is represented by the musalla, a large, open-air square for prayer, complete with a mihrab. Surprisingly, in aleppo the musalla was not built outside the city walls, as was common, but within them in the south-east of the city, near the citadel.18 although a lack of archaeological data makes it difficult to determine the dimensions and form of the aleppo musalla precisely, an interpretation of the morphology of the urban 17 18 See Sauvaget, Alep, pp. 79–80. ibid., 76. the forma urbis of aleppo 137 fabric suggests that this place of prayer coincided with the large empty space at the foot of the citadel. The existence of a vacant area within the roman castellum enclosure may have justified the choice of such an unusual site. on the level of the urban fabric, the abbassid period does not seem to have been one of expansion; indeed this was a period of stasis for the whole of Syria. once the capital was relocated from damascus to Baghdad, the region found itself in a marginal position with respect to the centre of the empire. in 877, aleppo fell to ahmad ibn tulun, founder of the tulunid dynasty. however, it was retaken by the caliph, and from 936–37 was in the hands of Muhammad al-ikhshid, who elected the head of the arab Kilab tribe as governor. in 944, the city was conquered by Sayf al-dawla of the hamdanids, who chose aleppo as his residence. it was during these years that, for the first time since the advent of islam, aleppo became a state capital (and the residence of the hamdanid governor). The city grew in importance politically and especially culturally due to the literary pursuits in the palace built by the governor outside the city walls. second Phase This flourishing cultural and urban period was soon interrupted by a series of historical events that, from a morphological point of view, would lead to the construction of the second ‘contracted form’ of aleppo. This form corresponds historically with the period between the tenth century, with urban renewal followed by invasions in 962 ad, and the twelfth century. This is perhaps the darkest period in the history of the city since the Muslim conquest. it consisted of half a century of turmoil due to a combination of internal politics, Byzantine invasions, Beduoin raids, and successive attempts by the fatimids to create a breach in order to invade iraq. from this moment on, the city no longer grew according to a homogeneous and isotropic grid system, but in terms of different neighbourhoods. These were distinctly separate and self-contained, and were derived from a medieval morphological process that affected the urban fabric, both residential and public. on an urban level, the city began to be subdivided into self-sufficient districts, the features of which seem to follow a model diametrically opposed to the widespread models of roman planning. 138 chapter five in 962, the Byzantine emperor nicephorus phocas conquered the city and left it in ruins after a week of burning. during the invasion the city walls were so badly damaged that they had to be partially rebuilt, especially the corner to the south-east of the citadel. in 1085, the sheriff’s citadel was built to the south of the city on one of the highest hills in the urbanized area. in 1095, the city became part of the Seljuk empire. from this moment, especially after the Zengids and nur al-din came to power, the city flourished once again. during the fifty years’ reign of the Zengids (1128–1181), considerable changes were made to the urban structure of aleppo. The city was reorganized on both an architectural and an urban-fabric level. The congregational Mosque was rebuilt, as were the wooden roofs of the suq and citadel mosque. Meanwhile several madrasas, a new hospital, and a law court were newly built (fig. 9). Many churches were converted into mosques and the system of water channels was reorganized, resulting in an increase in the number of public and private baths. interventions on an urban level in this phase of human development of aleppo took the form of the partial reconstruction of the city walls,19 the rebuilding of the citadel, the specialization of the suq, and the beginning of the formation of a fabric of specialized buildings along the extra moenia routes. one of the most important operations completed by nur al-din was that of reinforcing the defensive system and building the Bab al-faraj gate, which conditioned the layout of some routes, especially the one linking it to the Bab al-nasr gate— which ran tangent to the ancient polarity of the roman theatre—and the congregational Mosque (fig. 9). on an architectural level, the process of building reorganization that took place in this phase is exemplified by the transformation of the Byzantine cathedral into the al-halawiyya madrasa. The construction of the madrasa on the site of the cathedral symbolized the triumph of islam over christianity, even though it did not completely cancel out the original structures of the constantinian basilica. The direction for prayer was rotated by 90° in order to adapt the building to the new religious exigencies. 19 See Sauvaget, Alep, pp. 140–141 and pl. 58; idem, “l’enceinte primitive,” 133–159. the forma urbis of aleppo figure 9: Zengid and ayyubid fortifications and urban fabric. 139 140 chapter five The brightest period in the urban history of this phase begins with the ayyubid reign (1207–1260), when the city once again came to play a central role on a trans-national level due to its geographical position as an important node in the network of international trade routes. in the fifty years of ayyubid domination the fortifications were systematically reorganized or else replaced and a new defensive structure was built around the extra moenia neighbourhoods in alignment with the Byzantine trenches (Kandak al-rum). The citadel was newly fortified and rebuilt as a royal city with palaces, fountains, and mosques. The acropolis was used as a quarry for this construction, which resulted in its sides being modelled into the form we recognize today. in this phase too, the specialization of the suqs began, with their subdivision on the basis of their main commercial functions. The leather suq (el-dabbagha) and the soap suq (el-Masabén), for example, were founded in neighbourhoods specializing in the production of these goods. Beginning in this period, the still sparse urban fabric of the classical planning phases began to be congested and medievalized by a residential structure that no longer followed the geometric rules of roman cadastral parcelling, but instead those of the internal organization of the clan. The return to a clan type of organization initiated a process that modified from within the structure of the urban fabric. henceforth, it began to be subdivided for defence purposes as well as on the basis of the new settlement logic into self-sufficient neighbourhoods, separated by gates and provided with primary services such as baths, mosques, and neighbourhood suqs. although this transformation of the building fabric has been interpreted in the past as a degeneration of the classical plan, i would like to suggest that we shift our viewpoint and instead see it as the (proactive) result of new social exigencies and, more especially, the need to adapt existing urban structures that were already complex and apparently confused. Thus, a process of medievalization of the intra moenia urban fabric began with the settlement of new social groups in the roman urban and territorial configurations. yet the introverse and self-enclosed nature of the type of urban space generated by this process seems to date to pre-classical aggregates that disappeared with the advent of hellenistic-roman planning, or perhaps survived on the periphery of inhabited areas, and enters into a dialectical relation with the sparser the forma urbis of aleppo 141 fabric of the roman planning model.20 The phenomenon of fragmentation of the urban fabric in self-enclosed, discrete aggregates is the same not only in the medievalization of the fabric of hellenistic-roman cities in Syria, but throughout the whole medieval arab world. on a neighbourhood level, the cul-de-sac system (blind alleys that give access to residential buildings) is derived from a process of transformation of the ancient courtyard house lots and their subdivision into more than one housing unit following the resurgence of urban growth. This resulted in the need for more housing space, and the consequent increase in building density. This process is similar to the medievalization that invested the residential fabric based on aggregates of courtyard houses in italy and led to the complete introversion of the building fabric around the structure of the cul-de-sac, centre of the social life of the clan.21 a further process typical of this phase is represented by the formation of extra moenia neighbourhoods (hadér) inhabited by turks and tradesmen and located outside the perimeter of the city walls along the main caravan routes, which were already flanked by buildings in the pre-islamic period. Third Phase The third phase of the ‘contracted form’ of aleppo corresponds to the Mamluk period (1260–1516), immediately after the city was captured by the Mongols, when its fortifications were destroyed and its principal monuments, together with whole neighbourhoods, burned to the ground. aleppo would take a long time to recover from this catastrophe, which represents the only significant interruption in its urban growth in 5000 years of urban history. The initial period of Mamluk domination of aleppo cannot therefore be considered as among its 20 See e. Wirth, Das orientalisches stadt im islamischen Vorderasien und syrien (Mainz, 2000). on the basis of J. Schmidt’s studies, Wirth codified the compact model of spontaneous aggregation of the islamic city, comparing pre-industrialized Syrian cities to Middle eastern cities from the Bronze age. Both city types share the same compact organization of their residential fabric that makes them inaccessible except via a system of cul-de-sacs. 21 See caniggia, Lettura di una città; g. caniggia and g. l. Maffei, Composizione architettonica e tipologia edilizia: 1. Lettura dell’edilizia di base (padua, 1979). for example, cul-de-sac route systems are found in non-islamic areas such as como and polignano a Mare in italy. 142 chapter five most splendid. The sultans’ most urgent concern was the restoration and renovation of the most important buildings including the great Mosque, especially in terms of replacing the wooden ceiling with a cross vault, the citadel and defence systems, and the acqueduct, which led to an increase in the number of baths and fountains. from the fourteenth century onwards, the first real expansion of the city began. Moreover, a large amount of replanning was carried out along the most important urban routes and the line of the ayyubid walls. in particular, after the enlargement of the city walls, new gates were opened, such as the Bab al-hadid to the north-east, which gave access to the route through the Bayyada neighbourhood, and the Bab al-neirab to the south-east, which represented the link with territorial routes from persia. on an architectural level, in 1318–19, the governor’s mosque, althunbugha al-Sahili, was constructed on the model of the great Mosque. located to the south-east of the ayyubid walls, the governor’s mosque marked a new role for this urban zone as it developed into the second most important mosque in aleppo. in this phase, the circuit of walls was enlarged in order to contain the city to the east of the ayyubid walls. in 1428, the walls were rebuilt along what had been the line of the Byzantine trenches (Kandak alrum) where the ayyubid dynasty had built a defensive wall for the more urbanized extra moenia neighbourhoods, encorporating them within a unified system of defence. The enlargement of the circuit of walls also encorporated the citadel, which thereafter lost its military function and took on a merely representative role. it was also in this phase that the Bayyada, ughulbak, althunbugha, al-Bustan, and Qasileh neighbourhoods were created. each new neighbourhood was given a mosque and hospitals, as was the case, for example, with the arghun al-Kamili maristan and Mankalibugha al-Shamsi mosque. The horse market was moved to below the citadel, near the specialized suqs, on the site where the musalla was probably located; the latter was no longer in use in this phase, since the proliferation of mosques had rendered it obsolete. finally, in this phase a process of monumentalization of the suq began whereby it developed from a simple commercial street into a fabric of specialized buildings. all the large buildings in the main suq in aleppo date to the second half of the fifteenth or the beginning of the sixteenth century and therefore can be attributed to this phase of urban development. This represents the final phase in the transition from the roman-Byzantine city to the medieval islamic city. at this the forma urbis of aleppo 143 time a veritable urban development took place that included the specialization of building types, the specialization of the urban and building fabric, and the monumentalization of the architecture. as mentioned above, one can trace the most important phases in the process of building densification in the Mamluk period and the nodalization of the main routes in correspondence to the main street axes by using the data that emerge from a structural reading of the building fabric and comparing them with historical data. a first instance of nodalization of the building fabric that can be ascribed to the Mamluk period is the construction of the al-umari Mosque on the site of the roman Theatre, a polarity of the urban fabric in the pre-islamic period. We have already mentioned the following developments: the laying out of the Jewish quarter, Bashita, and the opening of the Bab al-faraj in the Zengid period; the specialization of the intra moenia route linking the Bab al-faraj with the Bab al-yahud; and the older route linking the theatre with the central suq of the medina. at the beginning of the Mamluk period, the roman theatre was probably already congested and the residential fabric of the Bandara and dabbagha neighbourhoods, built at the back of it, was probably already developed. in 1328, work began on the construction of the al-umari Mosque in the western zone of the ima cavea and media cavea of the roman Theatre and would be protracted for more than a century. an analysis of the aggregate structure, compared with historical data, confirms that when the circuit of walls was built in this zone, the Bab al-faraj was rebuilt to the north of its original position and the route leading from the gate to the mosque was formed. This rebuilding turned the latter into the most important specialized pole in the Bashita neighbourhood (fig. 10). Within some of the aggregates of the existing fabric to the east of the ayyubid city that still contained the traces of the first roman plan secundum coelum, a process of edification of the unbuilt lots began that coincided with the reorganization and nodalization of some routes and the construction of specialized architecture within the residential building fabric. an example of this process can be gleaned from an analysis of the structure of the building fabric of the Bayyada neighbourhood to the north-east of the citadel, where the traces of the first roman plan are still legible. The process of nodalization of the building structure of this neighbourhood’s central nucleus began with the persistent use, in the Middle ages, of a route tangent to the north side of the citadel and 144 chapter five figure 10: The urban fabric around the al-umari mosque (the roman theatre). the forma urbis of aleppo 145 figure 11: process of nodalization of the building structure of the Bayyada neighbourhood: traces of the roman plan secundum coelum. 146 chapter five dating from the roman layout (fig. 11). today, this route is still one of the main distribution axes in the neighbourhood. close to this, the wall alignments are still legible, corresponding to the north-south, east-west orientation of the first roman plan, with a regular rhythm of 60 × 120 roman feet, equivalent to 17.7 × 35.5 metres.22 from an analysis of these traces within the structure of the urban fabric one can recreate the original structure of an aggregate of domus houses iso-oriented in an east-west direction and distributed by routes that branch off from the axis of the main route.23 in this phase, the route leading to the Bab al-hadid became a node as the locksmiths’ suq and hub of specialized buildings (such as the al-Sarawi Mosque), thereby modifying the route hierarchy of the roman layout and replacing in importance the east-west axis skirting the citadel to the north. it became both the fulcrum of one of the richest neighbourhoods in the Mamluk period and a node at its southern end due to the concentration of palaces and residential buildings of wealthy citizens. one finds a similar process of nodalization within the aggregate structure along the axis of the route leading to the Bab Quinnasrin, which dates from the first roman plan secundum naturam. here, one can read the traces of lots with well-codified dimensions and structures, originating from the second phase of the roman urban and territorial organization of aleppo, as well as traces of the process of building densification and medievalization of the fabric of the roman plan. By superimposing a grid of domus lots on the 1930s french cadastral map, one can note a widespread correspondence between the original plan and the neighbourhood structure, since along the entire Bab Quinnasrin axis the domus lots, iso-oriented for climatic reasons in an east-west direction, face onto this matrix route. in the northern part of this system, one can observe a rotation of the lots, which here face in the direction of the counter-axis of this plan, the decumanus. during this phase in the process of medievalization of the urban fabric, the rigid, yet sparse, original structure of the domus houses (60 × 120 roman feet) was considerably recast and densified, gradually 22 These dimensions correspond to those of the clima, the surface area occupied by a domus with sides measuring 17.7 × 35.5 metres. 23 The presence of residential fabric with such stratified morphological characteristics in an area outside the circuit of ayyubid walls could be justified by the existence of the defensive trench around the city, originating from the roman period, of Kandak al-rum. the forma urbis of aleppo 147 being transformed into highly specialized and hierarchical lots and aggregates. along the Bab Quinnasrin axis, one can still read in the aggregate structure some of the secondary routes and linking routes laid out in the second phase of the roman organization; some of these routes have been transformed into cul-de-sacs. for example, it is clear from an analysis of the urban fabric that a route leading to the Bab Quinnasrin—and linking the urban fabric planned in hellenistic times (the continuation southwards of Suq al-Kassabiya, a secondary route in the hellenistic plan) to a route planned in the second phase of the roman organization of aleppo (the route from the Bab Quinnasrin that skirts han fansah, a plan from the second phase of the roman organization of aleppo)—was one of the linking routes of the latter (i.e. roman) plan. after the medievalization of the fabric, it was parcelled into two aligned cul-de-sacs. a reading of the urban fabric also allows us to classify, along the main axis of the plan, some increments deriving from the process of medievalization, specialized as suq stalls and residential buildings (fig. 12). in the same area one can read traces of the third phase of the crucial human organization of the city and territory of aleppo, corresponding to the second roman plan secundum naturam, whence it derives its original structure, or at least the layout and dimensions of most of the lots of the neighbourhood building fabric. Superimposing a grid of domus lots oriented in the direction of the third roman plan on the cadastral map, one can observe a widespread correspondence between the grid and the neighbourhood structure itself. Throughout the whole area of the Bab Quinnasrin neighbourhood one can read the permanence of iso-oriented domus lots from the original plan that face in an east-west direction. in particular, the architecture of the Bab Quinnasrin, and of the khan next to it, coincides in dimension and orientation with the structure and direction of this plan. finally, some of the secondary and linking routes planned in this phase are still legible in the aggregate structure. after a process of building densification and the reorganization that partially changed the neighbourhood structure, some of these routes became cul-desacs. They are still legible, however, in the urban fabric of the Bab Quinnasrin area, despite the extensive reorganization taking place in this phase; for example, following the construction of the al-rūmī Mosque, which transformed the viability and the organization of lots in the eastern part of the Bab Quinnasrin aggregate and thereby makes a reading more difficult (fig. 13). 148 chapter five figure 12: process of nodalization of the building structure of the Bab Quinnasrin neighbourhood: traces of the first roman plan secundum naturam. the forma urbis of aleppo 149 figure 13: process of nodalization of the building structure of the Bab Quinnasrin neighbourhood: traces of the second roman plan secundum naturam. 150 chapter five on the basis of our knowledge so far, one can single out the cul-de-sac system as the most salient urban form deriving from the process of medievalization of the roman courtyard house fabric in the final building phase of medieval aleppo.24 Measured drawings of the structure of the urban fabric on an aggregate scale in the Bab Quinnasrin area show that it was organized around this system, as the heart of the urban organism and centre of the domestic space of the clan. The end of the cul-de-sac can be considered as a courtyard at an aggregate level, a semi-public space dedicated to the activities of clan life. in the fabric dating from the roman plan, which is deeper than that of the blocks dating from the hellenistic plan, and in which the specialization of the residential buildings is minor, it is easy to observe the frequent use of this distributive system, deriving from the parcellization of the original geometric route network, in order to allow access to the courtyard houses generated by processes of subdivision and building densification of the ancient domus lots. The drawings that show the relation between the areas served by cul-de-sacs and the urban fabric also show that most of the courtyard houses here are only accessible via this distributive system and consequently the aggregate is completely introverse (fig. 14). a reading of the structure of the building fabric in aleppo reveals that the zone between the old ayyubid walls and the Byzantine trenches was already densely urbanized at the time of its enclosure within the new circuit of Mamluk walls and that here there was no more space for expansion other than by complex processes of nodalization. due to the increasing specialization and subsequent reduction in extension of the residential building fabric at the heart of the city, considerable urban expansion took place to the north and north-east through the creation of extra moenia neighbourhoods along the main caravan routes. Therefore, in this phase there was an expansion, specialization, and subdivision of the extra moenia urban fabric in separate and autonomous neighbourhoods, based on the scheme of ethnic subdivision typical of the residential areas of islamic cities. 24 on the formation of cul-de-sacs, see also g. Sjoberg, The Preindustrial City. Past and Present (new york, 1965). the forma urbis of aleppo 151 figure 14: relation between the areas served by cul-de-sacs and the urban fabric in the Bab Quinnasrin area. 152 chapter five Conclusion The reconstruction of the form of aleppo has provided a linear growth model or process for the urban fabric. This is made up of the stratifications and superimpositions of different concepts. in particular, what stands out in an interpretation of the urban fabric of aleppo, contained within the perimeter of the Mamluk circuit of walls, is the permanence of traces of roman planning in the medieval urban fabric. These traces are not immediately legible, since they derive from territorial plans carried out in different phases and are closely dependent on the form of the site and territory of aleppo. This is a feature common to other cities in Syria. indeed, many of these cities were planned in hellenistic times. They were later expanded and, in the roman period, underwent extensive renewal, in many cases being completely reorganized, adding to and often superimposing on both the original nucleus of the Seleucid settlement and earlier transformations. The resulting different plans—the main orientations of which, on an urban level, derived from the nature of the site and, on a larger level, from the main routes within Syria—created an apparently unplanned urban form. over a long period, this plan merged with type-morphological processes based on a different development logic from that typical of roman urban planning, one much closer to the ‘spontaneous’ structure of medieval islamic cities. in accordance with the idea of a linear process in the development of the urban fabric in pre-modern times, in order to understand the urban structure of medieval aleppo i have tried to decipher the palimpsest of traces that influenced it. This approach allows one to compensate for the lack of historical data on the structure of the urban fabric of aleppo in pre-medieval times and to integrate the existing sectorial and discontinuous data on the structure of its urban fabric through time. from this reconstruction it emerges that the complex, stratified, and apparently chaotic form of the urban fabric of medieval aleppo derives from the sum of various urban plans of roman origin, superimposed one over the other, demonstrating the inaccuracy of the view of arabs as destroyers of the geometrical and ordered organization of the classical city. indeed, when the arabs conquered Syria they inherited a complex urban and territorial structure, which was apparently ‘disordered’, since it derived from the summation of different plans with different layouts. They occupied these existing settlements, and henceforth the medieval islamic city developed, with its own physical the forma urbis of aleppo 153 and conceptual organization. in particular, the superimposition of the layouts of the roman plans in their various phases created a complex form for the urban fabric of aleppo, especially along the main axes of the Bab Quinnasrin, Bab al-hadid and Bab al-nasr, which represented zones of specialization and development of the medieval islamic urban form. chapter Six ethno-archaeological approaches to Medieval rural settleMent in spain and Morocco Johnny de Meulemeester1 Introduction archaeology studies societies that have disappeared, while ethnology studies those still living. as a result, even when the object of the study is the same, the methods will differ. For example, the archaeologist typically resorts only to the material vestiges of a society without having access to the actors, while the ethnologist questions the actors, without always paying attention to the material vestiges.2 Ethno-archaeology3 is a study that by the observation of traditional but current practices, goes back into time to provide the archaeologist with explanatory diagrams and interpretations of the data he or she has uncovered.4 Its final purpose is to explain, through an ethnographic detour, those archaeological facts that could otherwise not be easily interpreted. it brings what is often lacking in archaeological description: volume and 1 prof. dr. J. de Meulemeester, formerly of the Medieval and post-Medieval archaeology unit (north-Western europe and Mediterranean areas), department of archaeology and ancient history of europe, ghent university, sadly passed away during the time in which this volume was being reviewed. The editor wishes to express his most heartfelt gratitude to prof. Meulemeester’s daughter, anne, for allowing his chapter to be published posthumously. although corrections suggested by the external reader have been addressed, the text remains, on the whole, the one that Johnny last submitted. 2 see o. aurenche, “les conditions de l’enquête ethnoarchéologique,” in EthnoArchéologie méditerranéenne: finalités, démarches et résultats: table ronde, eds. a. Bazzana and M.-ch. delaigue (Madrid, 1995), p. 13. 3 on ethno-archaeology, see e.g. M.-ch. delaigue, “ethnoarchéologie et histoire: l’habitat rural de la péninsule ibérique comme indicateur des modifications de peuplement,” in Ethnoarchéologie: justification, problèmes, limites (Juan-les-pins, Fr, 1992), pp. 391–407; o. aurenche, “les conditions,” pp. 13–16. 4 M.-ch. delaigue, “deux exemples d’habitat rural en andalousie orientale: approche ethno-archéologique,” in La casa hispano-musulmana: aportaciones de la arqueología. La maison hispano-musulmane: apports de l’archéologie, eds. J. Bermudez lopez and a. Bazzana (granada, 1990), p. 21. 156 chapter six life, based on a full-scale observation.5 however, in order to combine the two approaches, one must assume three conditions: historical continuity, geographical context, and socio-economic context. in terms of the first condition, as aurenche puts it, one may question to what extent all the great civilisations of the Middle east (from assyrian to ottoman) changed the rhythm of life of the semi-nomadic pastoralist of northern syria. clearly, a lot of aspects of daily life did not change or were only slightly adapted, showing historical continuity.6 The geographical context can be treated in two different ways. First, one can either compare societies distant in time, but established in the same geographical area, which is possible for example for the settlement structures in the Middle east or north africa. second, one can compare societies distant in spaces and time, but established in comparable microenvironments. islamic spain, where the christian conquest did bring some structural changes, provides an example of the blending of both. We do not have to look farther than the other side of the straits of gibraltar, where in similar geographical circumstances the rural population survives on a pattern that links geographical conditions and historical continuity. The ethno-archaeological comparison can even shed light on the socio-economic context, at least in a very general direction. Working on material culture, it acts to highlight the presence or the absence of certain techniques within the studied societies. in the case of spain (and in particular the studied area of the Murcian valle de ricote) this approach reveals that, after the christian conquest, the new leaders continued to employ similar irrigation techniques as under islamic rule and even appealed to the Mudejares (Muslims living under christian rule) to re-establish and to maintain the irrigation network. real changes came only very late with the introduction of metal (for norias or waterwheels in the nineteenth century) and concrete tubes that enable the canals to be put under ground (thus avoiding evaporation). a potential danger lies in devoting oneself to a partial ethno-archaeology that would study a technical or cultural feature while taking it out of its context and apply the resulting conclusions without precau- 5 a. Bazzana, Maisons d’al-Andalus: habitat médiéval et structures du peuplement dans l’Espagne orientale (Madrid, 1992), p. 17. 6 aurenche, “les conditions,” pp. 13–16. ethno-archaeological approaches 157 tions to a given archaeological situation. Therefore, it seems advisable to prefer a total ethno-archaeology that takes account of the context, initially defined by the means of archaeology, i.e. by the study of the material remains. examples of actual field practice make this more clear. Methodology a collaboration involving the heritage department of the Walloon region (Be), the casa de velasquez (the French school in Madrid), the unité Mixte de recherche 5648 (lyon ii/ehess) of the cnrs (Fr), and the departments of archaeology and of geography of ghent university (Be), has made possible the study of the medieval settlement of the valle de ricote, a geographical and administrative centre on the segura river in the autonomous region of Murcia (es). The original hisn (castle) of ricote controlled a relatively extensive territory, measuring a little over eight hundred square kilometres. The territory includes rural sites such as hamlets and villages, as well as defensive sites and even a small town. Moreover, it shows the presence, from the ninth century onwards, of a complex irrigation system. irrigation patterns form part of the archaeological elements in the rural landscape. From the river upwards to the line of the peaks, the valley is marked by the succession of the following elements: irrigated fields, irrigation canals, human habitats, dry culture lands, and finally pastures on the mountain slopes. to get a better understanding of the key elements of this Murcian landscape, geo-archaeological research was initiated in the valley of the awnil in the atlas Mountains in Morocco. simultaneously, this research revealed an ethno-archaeological link between the local granaries, the irrigation system, the settlement type, and its historical evolution. This investigation used different methodological paths to achieve its purpose in spain as well as in Morocco: • systematic and full open area excavation was applied to the Cabezo de la cobertera hilltop (abarán/Blanca, es), where, as was later shown, a fortified granary was located; • geomorphological and geo-archaeological studies were carried out in both landscapes to understand their development and the impact of human interference; 158 chapter six • in both study areas aerial and satellite photography was used to interpret hydraulic systems and irrigation canals in particular; • intensive fieldwalking was used to survey the irrigation systems and to analyse water collecting systems, dams, canals, partitions, qanats (underground canals), reservoirs, watermills, etc.; • interviews of local people were conducted in Morocco to find out about field distribution, crops, ownership, organization of the irrigation and its maintenance, social structures, and finally historical evolution. i. Fortified Granaries 1. In the Murcia region The central element of the study was the excavation of a fortified granary of Berber type dating to the thirteenth century.7 The summit of the cabezo de la cobertera, which towers some one hundred metres above the segura river, is approximately 40 metres long by 30 metres wide and is occupied by the remains of about thirty cells, built along both sides of narrow lanes that the inhabitants would have used to move around the settlement. These buildings are mainly of rectangular plan, averaging 4 metres to 5 metres in length by 1,5 metres to 2 metres in width. several cells have a sort of trough located at the back. The cell is entered via a narrow door. at the centre of the site, there is a water storage tank of approximately 16 cubic metres. The archeo-historical study of the valley and its different settlement elements suggest that the first occupation of the site dates back to the twelfth century and that it was still in use towards the middle of the thirteenth century, to be abandoned at the time of the Mudejar revolt (i.e. in about 1266).8 7 see F. amigues, J. de Meulemeester, and a. Matthys, “archéologie d’un grenier collectif fortifié hispano-musulman: le cabezo de la cobertera (vallée de rio segura/Murcie),” in Castrum 5, ed. andré Bazzana (Madrid, 1992), pp. 347–359; J. de Meulemeester, “Même problème, même solution: quelques réflexions autour d’un grenier fortifié,” in Le village médiéval et son environnement. Études offertes à JeanMarie Pesez, eds. l. Feller, p. Mana, F. piponnier, and Jean-Marie pesez (paris, 1998), pp. 97–112. 8 J. de Meulemeester, “le valle de ricote et le développement des recherches de la région wallonne à l’étranger,” in Mélanges d’Archéologie médiévale. Liber amicorum en hommage d’André Matthys, eds. J. Meulemeester and p. gillet (liège, 2006), pp. 46–55. ethno-archaeological approaches 159 at the time of its discovery, this type of settlement was unknown in spanish archaeology and could not be associated with any medieval islamic or christian text explaining its use. For the moment, it remains the only one of its kind in the archaeology of the islamic peninsula. Thus, it could only be interpreted through comparison with the collective granaries of Berber origin that are known as agadir in southern Morocco. 2. In the Atlas Mountains a. Berber granaries from the Maghreb to central europe There are various solutions for the long-term conservation of grain in north africa. The processes of ensilage vary between nomadic, seminomadic, and sedentary societies. in order to illustrate this variety, i would like to consider the example of southern tunisia.9 generally, nomads living under a tent leave their reserves of grain close to the production fields in an area of ensilage entrusted to a remunerated guard recruited from one harvest to another. They use silos and above-ground containers indiscriminately. The guard digs the silo in the shape of cylinder, the depth and diameter of which hardly exceed a metre. nomads also use large baskets in esparto manufacture that copy the shape of a large earthenware jar. They leave the baskets behind at the ensilage area where they are entrusted to the guard. The guard takes care of the seed grain as well as of the consumption grain, not only against all depredations from where ever they come, but also, indirectly, against the improvidence of the women, who have only within range, under their tent or their hut, just what it is necessary to nourish the family.10 The fear for female wasting was noted for other areas too. as noted by a tunisian semi-sedentary “every one prefers to have his grain well sheltered and aired well, far from the women.”11 9 a. louis, “la conservation à long terme des grains chez les nomades et semisédentaires du sud de la tunisie,” in Les techniques de conservation des grains à long terme: leur rôle dans la dynamique des systèmes de cultures et des sociétés, eds. M. gast and F. sigaut (paris, 1979), 1: 205–214. 10 J. despois, “les greniers fortifiés de l’afrique du nord,” Les Cahiers de Tunisie, 1, no. 1 (1953), 58. 11 louis, “la conservation,” p. 208. 160 chapter six semi-nomads preserve the grain for direct consumption in their residence. The remainder is stored in collective granaries composed of superimposed cells, called ghorfa. These initially represent only one ‘upper room’, added on to the roof of the house. in periods of disquiet, sedentary societies employ baskets that are stored in the court or the terrace of the house, which functions as the collective granary. When security is no longer a primary concern, they employ domestic granaries of the ghorfa type, annexed to the residence.12 This type of domestic granary survived in the andalusian mountain villages under the name of cámara, a room which is located on the upper floor.13 in south Morocco, the collective granary is called an agadir. The agadir is a building set apart from the village where the Berbers store their harvest and other goods. These granaries are often also fortresses, and thus may be situated in steep-sided or other naturally defensible locations. These often serve the whole canton, and become a catalyst for the development of a common law, trade, and market.14 Many examples of these collective granaries exist in north africa. They are known by the name of ksar (tripolitania and in southern tunisia),15 guelaâ (algerian aurès),16 kasba (the sahara),17 aqrar (northern Morocco).18 some of these dominate the villages, others, isolated from the daily habitation, crown a rock piton. They are best preserved and best studied in Morocco.19 The need for building granaries initially comes from causes of climatic origin. in semi-arid areas where the majority of the granaries 12 s. Ferchiou, “conserves céréaliers et rôle de la femme dans l’économie familiale en tunisie,” in Les techniques de conservation, 1:190–197. 13 M.-ch. delaigue, Capileira, village andalous. Un habitat montagnard à toits plats (oxford, 1988), pp. 75–76. 14 r. Montagne, “un magasin collectif de l’anti-atlas: l’agadir des ikounka,” Hespéris 9 (1929), 145–266. 15 louis, “la conservation,” pp. 205–214; a. ayoub and J. l. le Quellec, “gasr alhag. un grenier fortifié dans la djeffara libyenne,” in Les techniques de conservation, 2:3–18. ‘Ksar’ typically refers to a Berber village as a whole, which consists of attached houses and contains other buildings, among which would be collective granaries. 16 despois, “les greniers fortifiés,” 38–60. 17 r. capot-rey, “greniers domestiques et greniers fortifiés au sahara. le cas du gourara,” Travaux de l’Institut de Recherches Sahariennes 14 (1956), 139–159. ‘Kasbah’ typically refers to the citadel or defensible portion of a settlement. 18 Montagne, “un magasin collectif,” 145–266; c. pereda roig, Los hórreos colectivos de Beni Sech-Yel (cueta, es, 1939). 19 despois, “les greniers fortifiés,” 38. ethno-archaeological approaches 161 are established, the harvest of cereals cultivated on dry ground or only irrigated with some surface waters is very irregular. in good years, the harvest can be of an incredible abundance, but often it is poor or non-existent. it is thus necessary to store food for the periods of shortage.20 The continual threat of plundering is added to the climatic elements. This threat came from the central state that wanted to eliminate the spirit of freedom of the granaries’ users and the independence of their small autonomous states. But in times of bad harvests, the threat came especially from neighbours in search of food. The grain must be defended!!! indeed, the worst enemy of the granary and the institution that it represents was the central authority, the makhzen.21 a tribe was only subjected after the destruction of its granary, the symbol of its economic independence and cohesion, and often its last effective area for a defensive stand. For example, in the nineteenth century, the turks destroyed all the ksour (sing. ksar) of the djebel nefousa in tripolitania following a revolt.22 other ‘threats’ or causes of decline of a granary included the pacification of an area (for example by French colonial rule), which removed the insecurity that had been the granary’s reason for being. Finally, the granary today remains for many villages a place of encounter and social life, like pubs in rural european areas.23 in tripolitania and southern tunisia, the ksar is a strengthened collective granary that could also be used as collective refuge on certain occasions. it is made of several hundreds of cells superimposed on two to five levels. These cells, called ghorfa, are laid out in a quadrilateral pattern around a central courtyard in order to form an enclosure that one reaches only through a protected porch. The ksar can comprise several granaries, each one organized around its own central courtyard. The ghorfa is barrel vaulted, contrary to the structure of cells in other areas. The rooms are small, narrow, low, and long. They appear like small corridors, on average 3m to 5m long and 1,20m to 1,50m wide. Their height seldom exceeds 1,20m. one enters via a small door approximately 0,70m high. The one who deposits his grain in the cell 20 ibid., 47. a. adam, “ la maison et le village dans quelques tribus de l’anti-atlas,” Hespéris 37 (1950), 330–331. 22 despois, “les greniers fortifiés,” 56. 23 ibid., 58; d. J. Jacques-Meunié, “greniers collectifs,” Hespéris 36 (1949), 135. 21 162 chapter six is either owner or tenant. if one does not have a ghorfa, one preserves his grain in a container installed in the central court. also specific to this area are the ksour of the plains, which belong to semi-nomads without villages. The cells of these granaries surround vast courtyards where the pastors can gather their cattle. some of the ghorfa on the ground floor are used as shops or dwellings. in tripolitania, the granary of gasr al-hag takes the shape of an amphitheatre. it is established on a light slope without any notable natural defence. This absence is compensated for by watch towers that control the approaches to the granary. entry into the granary is also only possible through a single door 2m high by 1,20m wide that gives access to a rectangular anteroom. This is flanked by two small, side bedrooms that are used for resting, for reading the Koran, and for depositing objects and valuable documents, which are kept in wooden trunks. proceeding through the anteroom, one reaches the circular, central court. a staircase leads to the upper floor of the cells. a system of hooks makes it possible to pull grain up to the ghorfa of the two lower floors; in the absence of a hook, one uses wooden scales. The granary is made up of one hundred and fifty cells, laid out on four levels. an opening of 0,80m by 1m gives access to the ghorfa, which are 3m long by 2m wide by 2m high. in the back wall, an opening of approximately 20cm allows for ventilation. oral tradition dates the gasr al-hag granary to the fourteenth century, a date generally advanced for the construction of the ksour of the tripolitanian djebel nafusa as well as those of southern tunisia. The ksar represents here an advanced form of the kalaa, the citadel-refuge. The latter differs from the ksar by the dimensions of its spaces: it is not a question of small cells for stock, but of broader and higher rooms that could also be used as dwellings. Thanks to inscriptions, one of these refuges has been dated to the last decade of the eleventh century. architectural analysis makes it possible to conclude that these larger constructions were compartmentalized later, when the refuge became a collective granary. This transfer of function must have occurred in the fourteenth century. The agadir of southern Morocco is an establishment of a tribe or clan where each head of a family manages an individual closed cell for which he has the key. The store and its dependencies are placed under the guard of a gatekeeper who supervises the comings and goings of the users and prohibits the entry of foreigners. perhaps the term of agadir is related to the phoenician term gadir, i.e. ‘wall’. ethno-archaeological approaches 163 The granary-court could have begun as a single enclosure. often the courtyard remains empty. elsewhere, it is almost entirely occupied by other stock pens. one is tempted to believe that this granary-court originally answered the needs of pastoral tribes, such as the need for holding animals. it is well known that this courtyard-plan surrounded by small rooms is shared by many caravanserais in the east and that this disposition makes it possible for caravans to shelter animals, people, and goods. The possibility of plundering and war effected the development of some of the granaries as quasi-fortifications that included watch towers and were located on precipitous places of difficult access. These agadirs present a narrow central alley, with several stages of cells aligned on each side. each of these cells would have had the same grain forms and dimensions. The flat roof is made of beaten earth. often, the projecting flagstones fixed at progressive heights in the frontages are used as hand- and footholds to reach the upper floors. sometimes, tree trunks allow one to reach the higher cells. all the cells open towards the interior of the store and are closed by a wooden door. There is no other opening to the outside, except for some ventilation holes. The dependencies of the communal granary vary in number and importance. in addition to the gatekeepers room, there can be a mill, a forging mill, a stable, a cattle shed, one or two rooms for extra guards, a meeting room for the local dignitaries, and even a small mosque. a large agadir may also contain several cisterns. in terms of the distribution of space, the selected ground is divided into equal parcels and the floors are drawn by lot. a family can occupy several cells, and each cell is built and maintained by the family that owns it. They must maintain it in good condition as its degradation is likely to bring harm to the neighbouring cells. The parts of the building that are held in common by the community are also built and maintained in common. The owners choose a guard or gatekeeper who is attached permanently to the building and whose subsistence they ensure. The agadir is sacred, like the tomb of a saint or a mosque. no bad action, robbery, lie, adultery, or murder should take place in the store. it is an inviolable space and it is also a place of asylum. This sacred character can emanate from several sources. For example, it can be due to the maraboutic protection to which the agadir is often dedicated. however, it appears probable that the substrate of this belief is older than islam. grain is the sacred object par excellence because it 164 chapter six is the germ of life. it is therefore not unreasonable to think that the grain preserved in the agadir communicates to the building its magical capacity and the sacred inviolability that it generates. in the past, when contracts or agreements were often formerly concluded in the granary, they would have undoubtedly been covered by its protection. The origin of the Moroccan collective granary remains obscure and the time and the place of its appearance are still unknown. north africa alone testifies with certainty to its use. nevertheless, the organization and the operation of the agadir are tied to common law. The oldest known charters of these fortresses date from the sixteenthseventeenth centuries. however, the excavation of the Cabezo de la Cobertera modifies this chronology. in the current state of knowledge, the collective store appears to be specific to the countries of the Mediterranean basin, although other areas also knew comparable institutions. apart from the Cabezo de la Cobertera, the system of the agadir (defensive granary or defensive structure turned granary) has not been recognized yet in islamic spain where the rural refuge par excellence was the albacar,24 which was a type of large bailey. studies on rural hispano-Muslim society25 and on fortifications (husun, sing. hisn)26 have advanced the probable existence of some sites—e.g. uxò,27 Monte Marinet,28 Magdalena,29 silla, and Montroy30—where installations without a habitation function could probably have been used as storage rooms.31 24 a. Bazzana, p. cressier, and p. guichard, Les châteaux ruraux d’ Al-Andalus. Histoire et archéologie des husun du sud-est de l’ Espagne (Madrid, 1988), pp. 10–32; Bazzana, Maisons d’ al-Andalus, pp. 338–340. 25 see Bermudez lopez and Bazzana, eds., La casa hispano-musulmana. 26 see Bazzana et alii, Les châteaux ruraux, pp. 10–32. 27 ibid., 212–216 and 256; a. Bazzana, “Maisons rurales du shark al-andalus. essai de typologie,” in La casa hispano-musulmana, p. 248; Bazzana, Maisons d’ al-Andalus, p. 260. 28 Bazzana, “Maisons rurales,” pp. 248 and 370; a. Bazzana and p. guichard, “archéologie extensive dans la région valencienne (espagne),” in Castrum 2, ed. ghislaine noyé (rome, 1988), p. 18; Bazzana, Maisons d’ al-Andalus, p. 261. 29 Bazzana, Maisons d’ al-Andalus, p. 261. 30 Bazzana et alii, Les châteaux ruraux, pp. 121–122; for silla, see Bazzana, Maisons d’ al-Andalus, pp. 261–262. 31 Bazzana, Maisons d’ al-Andalus, p. 346. ethno-archaeological approaches 165 in the rest of spain, the only correspondence with the collective granary on the level of defensive value and protective philosophy is in the catalan sagreres.32 These are circular enclosures around a church. pierre Bonnassie calls the sacraria (in catalan sagrera) “a sacred protected space, almost always with a radius of thirty steps, which surrounds the church and profits from the same safeguard the church itself.”33 The role of the sagrera in the formation of the catalan village seems to have been central, and in the eleventh century it becomes an essential element in the evolution of the landscape. The sacraria also reveals the evolution from granary-refuge towards village. The system developed in the catalan countryside gave rise to the circular villages so widespread in the languedoc-roussillon region of France. other more northern areas experienced a similar development.34 The cellaria—in german keller or gaden—of the alsace, switzerland, germany, or transylvania (romania) form the only examples of village storerooms preserved until our days, but in the Middle ages, the system was much more widespread. The Gadenkirchen of transylvania and central germany, from the start, were built to store reserves.35 it is sufficient to replace the marabout and the central houses of the Berber granary with a church to obtain the same overall plan. in fact, between the arabo-Berber collective granary and the christian cemetery-refuge there are no fundamental differences. B. tazlaft and its granaries The investigation of Moroccan collective granaries began in the province of ouarzazate in 2000. The project brought together archaeologists, a geomorphologist for the study of the landscape, and Moroccan anthropologists who carried out the oral element of the research.36 32 p. Bonnassie, La Catalogne du milieu du Xe siècle à la fin du XIe siècle (toulouse, 1975); idem, “les sagreres catalanes: la concentration de l’habitat dans le ‘cercle de paix’ des églises (xie siècle),” in L’environnement des églises et la topographie religieuse des campagnes médiévales, eds. M. Fixot and e. Zadora-rio (paris, 1994), pp. 68–79. 33 Bonnassie, “les sagreres catalanes,” pp. 68–79. 34 M. Fixot and e. Zadora-rio, L’église, le terroir (paris, 1989); d. Baudrel and J.-p. cazes, “les villages ecclésiaux dans le bassin de l’aude,” in L’environnement des églises, pp. 80–97. 35 h. hinz, “Wehrkirchen und Burgenbau,” Château Gaillard: études de castellologie médiévale, IX–X (caen, 1982), pp. 118 and 120. 36 since 2000, this project forms part of a bilateral agreement of cooperation between Morocco and Belgium (Ministry of the Walloon region—direction de 166 chapter six some of the collective granaries studied remain in use. The collective granary contributes in a very special way to the social bond that forms the heart of the community, starting at the time of the harvest and continuing through the year with the management and distribution of the grain. The investigation concentrated on the evidence regarding the socioeconomic environment as registered in the landscape. it is important to note that a granary will only exist where complex and well-managed irrigation systems function to allow arable farming. The village of tazlaft, located 15 kilometres upstream of the ksar of aït Ben adou, a unesco World heritage site, was chosen as the study area especially for its position at the junction of two valleys. The larger of these is occupied by an old caravan route, connecting Marrakech to the sahara desert. archaeologists working on earthen buildings have been interested in the typical collective granaries of the Berber communities since the 1930s because these constructions and their building techniques are characteristic of the atlas Mountain region.37 a large-scale survey was carried out by the French-Moroccan archaeologist djamila JacquesMeunié between 1941 and 1954, during the course of a number of long stays in the area. This work represents a remarkable undertaking in view of the period and the limited means then available.38 at the time Jacques-Meunié was already anxious about the abandonment and destruction that threatened these monuments. it was clear that the study of these collective granaries would become increasingly difficult as they gradually lost their function. in tazlaft, the survey brought to light several cliff granaries that have to be linked to the period in which semi-nomads inhabited the area, before the actual settled village developed and became permanent. These semi-nomadic people left but few traces of their occupation and use of the land, but their presence is demonstrated by the existence of several cliff granaries along the length of the assif (river) l’archéologie de la division du patrimoine, namur) directed by the author and by philippe Mignot (MrW). see ph. Mignot, J. de Meulemeester, M. Boussalh, and M. Jlok, “Medieval rural settlement in Murcia (spain) and present-day rural settlement in the Moroccan atlas,” in European landscapes and lifestyles: the Mediterranean and beyond, eds. Z. roca et alii (lisbon, 2007), pp. 131–145. 37 r. Montagne, Villages et kasbas berbères: tableau de la vie sociale des berbères sédentaires dans le sud du Maroc (paris, 1930). 38 d. J. Jacques-Meunié, Greniers-Citadelles au Maroc, 2 vols. (paris, 1951). ethno-archaeological approaches 167 Marghane. each of these consists of a series of cells, beside and/or on top of one another, cut into the rock. access to these granaries is extremely difficult. one of them is still accessible via a dangerous goat path. it is made up of some sixty cells, oval in plan and arranged in three galleries on two different levels. at the time when the nomadic lifestyle was giving way to a settled one, villages were built around public spaces and facilities, among which the granary remained important. indeed, the collective granary came to be a relatively substantial focal point. a first granary of this type was built a short distance from the village on a hillock that overlooks the village and the assif Marghane. of this first granary, only a few fragments of walls remain. archaeological analysis, without the benefit of trial trenches or full-scale excavations, suggests that the granary was divided into cells measuring about 5m by 2,5m, with corridors between 1m and 1,5m wide. according to local oral tradition, this granary was destroyed by another tribe in the course of a war some two hundred to three hundred years ago. This ruined granary was replaced by the current one located in the centre of the village. a sample of wood from the original floor of the upper storey of the current granary has been radiocarbon-dated to the first half of the seventeenth century (c13/c12, Beta analytic, Miami), which confirms the oral tradition about the destruction of the hilltop granary. The current granary is a four-sided structure measuring about 19m by 21m and comprising two storeys and four corner towers. on top of this structure, there is a terrace that allows the accommodation of some additional cells. The granary contains about sixty cells. despite the survival of certain defensive features—for example, the corner towers or the singular entrance—several aspects of its role as a refuge for the village community have been lost, and its remaining function is merely that of a collective storage facility. it has no water tank, no designated accommodation for its guardian, no real benches at the entrance, no religious installations, and no meeting rooms for the village council. upon entering through the door, one arrives in a corridor from which, to the left, one can gain access to the staircase that leads up towards the terrace. in the middle of the corridor stand the containers reserved for the guardian of the granary. The arrangement of the cells is centred around the square central yard of the granary, which serves at the same time as a source of light and an impluvium. The cells are long and narrow, each being of almost the same dimensions (ca. 6,50m by 1,50m). although the doors are sometimes decorated 168 chapter six with incised and painted geometric designs, unfortunately here in tazlaft, those that survive are in a very poor state of preservation. There are still small bits of paper stuck with cow dung to the lintel above the door, on which various verses from the Koran can be read: “glory to god, who in his magnanimity gave us the rain which saved us from drought.”39 a piece of paper is stuck to the lintel each year while the harvest is being stored in the granary. Both the decorations and the papers are regarded as talismans. each door is furnished with a wooden lock, for which the key has wooden teeth; in only one case is the key made of iron. The interior of the cell is divided into compartments using low walls built of mud bricks. The size of the compartments varies according to the needs and desires of each occupant. The height of the ceiling, which reaches 2,80m on the first floor, allows for the installation of an extra storage level against the back wall of the cell. Finally, the ownership of the cells reflects part of the history and the social structure of the village. Those who own a cell seem to belong to the families that trace their ancestry to the period preceding the destruction of the hilltop granary. it is conspicuous that while the actual chief of the village is most likely descended from a leading family of the tribe that conquered it and destroyed the granary in the seventeenth century, he has no rights or ownership in the current village granary! This chief, who did everything to welcome us into the village and to get the granary restored by our team, even came into conflict with some of the owners because he gave us entrée to the building without consulting them, and thereby was seen as trespassing or overstepping his prerogatives. ii. Hydraulic networks 1. In the Murcia Region The other focus of investigation in this study concerned the issue of irrigation.40 traces of medieval agrarian hydraulic systems are still 39 This sentiment is shared by numerous verses in the Koran: 35:9, 42:28, and 43:11. similar thoughts can also be found in the following: 14:32, 16:65, 35:27, 36:33, and 50:9. 40 a. Bazzana, J. de Meulemeester, and a. Matthys, “Quelques aspects du peuplement médiéval du valle de ricote (Murcie/espagne),” in Rural Settlements in Medieval Europe, Medieval Europe Brugge 1997 Conference, eds. g. de Boe and F. verhaeghe (Zellik, Be, 1997), pp. 39–54. ethno-archaeological approaches 169 discernible in the valle de ricote. our study investigated the structures surviving from two settlement periods: the first corresponding to the Muslim presence, and the second corresponding to the series of changes that followed the christian conquest. irrigation structures constitute an integral element of the archaeological record that is recognizable in the rural landscape of today. The area of the regadio (irrigation) is clearly distinguishable, forming the huerta (the garden) at the bottom of the valley, which is the fertile zone dominated by orchards. cereal crops and vines are traditionally grown higher up, in the secano, or dry agricultural area. Thus, starting from the bottom, between the river and the line of the peaks, the medieval land use structure of the valley consisted of the following: the irrigated areas; the line of acequias; then the zone of human residential accommodation, above which were the secano; and finally, pastures on the upper slopes and in the mountains. in the Middle ages, the Muslim farmers were using four different technological approaches for the recovery of water: – ponds used for the storage of water from springs; – wells from which water was drawn using a container attached to a rope, or perhaps strung over a beam; – water wheels used for raising water such as the noria, with its vertical wheel installed on a canal, or on the river itself; – various systems of acequias or channels that transported water diverted from a river by means of a dam. according to the most recent historical research, most of the development of the regadio took place at the height of the islamic period, around the ninth and the tenth centuries. at a later stage, after the thirteenth century and during the christian period at the end of the Middle ages, there was a significant phase of expansion in valencia and Murcia, the nature and scale of which has led some authors to write of a ‘boom’ in hydraulic engineering works from 1480 onwards.41 The extensive networks of regadio, which were built at that stage by a more centralized power, are organized so as to cover the region and 41 J. lagardère, “droit des eaux et des installations hydrauliques au Maghreb et en andalus aux xie et xiie siècles dans le Mi‘yar d’al-Wansharisi,” Cahiers de Tunisie, 37–38, nos. 145–148 (1988–1989), 83–122. 170 chapter six as such they are therefore rather different from the more locally-based systems that had been set up by the Muslims. There are two main points to be made in relation to these water management systems. The first is in connection with the hydraulic engineering works: the dams, river diversions, canalizations, local irrigation networks, channels, and conduits. once built, these mark the countryside definitively, introducing what Miquel Barceló has described as the ‘principle of rigidity’, by which he observed that the installation of such a framework precludes further modification of details of the landscape, or any other partial transformation of the organization of agrarian space.42 Thus, in cieza, abaràn/Blanca, or ricote, certain elements of the tenth to thirteenth century islamic hydraulic system remain recognisable in spite of the superimposition of another system during the christian period. neither the feudal conquest of the thirteenth century nor the successive phases of depopulation and new colonization that took place in the fourteenth-fifteenth centuries completely erased the traces of the earlier arrangement. The second point is that it is important to be familiar with the basic principles governing the organization of irrigated areas in the medieval islamic context. every means of securing access to water was exploited, and the technology used was generally straightforward. it required no major investment and would normally not call for specialist expertise. it was the product of generations of accumulated experience. one of the simplest and most widely used techniques involved the diversion of some of the water from a river into an irrigation canal by means of a dam upstream of the area to be irrigated. Because the natural fall of the river was greater than that of the canal, the difference in altitude was sufficient to bring the water via a series of connected channels, built of earth, towards the fields. here, simple but effective systems of sluices allowed the selective watering of specific landholdings. Most importantly, an even fuller appreciation of the two key elements of this Murcian landscape may be obtained through ethnoarchaeological research. The remains that have been identified in the ricote valley were part of a system very similar to that which is still in use in the atlas Mountains of Morocco. 42 M. Barceló, “el disenõ de espacios irrigades en al-andalus,” in El agua en zonas áridas, arqueología e historia (almeria, es, 1989), pp. 8–50. ethno-archaeological approaches 171 2. In the Atlas Mountains a. The hydraulic network at tazlaft a preliminary reconnaissance of the valley of the assif awnila and assif Marghane, to the north and to the west of tazlaft, respectively, reveals settlement patterns similar to those of all the other valleys of the area. although differences emerge as a result of local topography, these relate more particularly to the geomorphology, which changes as one climbs up the valley. to the north of tazlaft, rocks eroded from the cliffs which stand above the thirty-two degree slopes of the steep-sided valley, fall to join the constantly-moving screen below. The slopes and the cliffs serve only as rough grazing for herds of goats and flocks of sheep and any attempt at improvement by humans is likely to be crushed in situ by falling rock. at the foot of the slope, the bed of the river is made up of three parts: the river bed itself, a lower terrace which floods each year and on which the water table remains sufficiently high to allow an annual cycle of cultivation without irrigation, and an upper terrace where perennial plants and trees are grown with the help of an irrigation system. The waters of the river cut sinusoid furrows in the riverbed, eroding material from the concave bank and depositing it on the convex bank. The main irrigation channel is cut along the edge of the upper terrace nearest to the foot of the slope. The water supply for this channel may be obtained via a dam in the river, by recovery of surface waters, or through a combination of the two. The hydraulic network at tazlaft begins with a dam in the assif awnila at the point where it emerges from the village of tassarda. a second canal of lesser importance serves to water an area of ground on the inner bank of the curving river. it also receives its water by means of a dam in the river and is located at the edge of the lower terrace and at the foot of the upper terrace. This canal was the result of a private initiative and responsibility for its maintenance lies with the owner of the irrigated area. in contrast, the main canal is managed by the village assembly who appoints a water guardian responsible for controlling all irrigation. each landowner is entitled to open a sluice gate on the canal in order to water his land, and the length of time for which the sluice gate may remain open is determined on the basis of the relative extent of the area to be irrigated. The water guardian (in arabic, amin) is informed of any infringements. The informer might, for example, be a neighbour directly affected by the infringement. The village assembly 172 chapter six has the power to impose punishment for infringements, which may take the form of a fine payable in cash. Before payment of the fine, the culprit might be required to organize a full meal for the members of the assembly, which might actually prove to be even more costly than the monetary fine itself. The main channel follows the right bank of the river all the way to the platform on which the village of tazlaft is built. here, it leaves the course of the assif awnila and, in the form of a qanat, it passes at a depth of about 20 metres under the present day village. it reemerges into the open air to the west of the village. at intervals of about 20 metres, maintenance and ventilation hatches provide access. The diameters of the mouths of these hatches have been significantly widened by the passage of rainwater. The main channel also serves to irrigate the terraces on the left bank of the assif Marghane, to the west of the village. a series of secondary and tertiary canals distribute water around the various parcels of land. Finally, the remaining water—which is almost stagnant by this stage—is returned to the assif Marghane. in general, the water of the assif Marghane—a name that actually means “the salted river”—is of a lower quality than that of the assif awnila. This difference in quality is the reason why water is brought from the awnila for the irrigation of the land on the left bank of the Marghane. The village community maintains the canal. twice a year a list of repairs is drawn up, and appropriate remedial work is assigned to each family. special maintenance may also be required at times of flooding. The canal must be cleaned out twice a year. Thus, the time devoted to the maintenance of the hydraulic system amounts every year to a period between fifteen days and two months. despite the lack of direct proof, we consider it more than likely that the irrigation system was originally made up of a number of channels that had been built as a result of private initiatives. it is in fact possible that irrigation of the upper terrace was adopted over time as improvements were made in the techniques of canal construction that made it possible to take water from much further upstream in the river and to calculate the requisite angle of slope for the canal, so as to allow an adequate water supply even as far away as the other side of the village of tazlaft. however, it seems clear that a construction such as that of the main tazlaft canal goes beyond the bounds of private initiative and can only have been managed at the level of the entire village. every local farmer is entitled to put a few stones in the river in order to ethno-archaeological approaches 173 divert some of its water to irrigate the crops in his adjacent landholdings. however, the irrigation of an extensive area, the construction of several kilometres of channels, the digging—as in tazlaft—of a part of the system in the form of a qanat in a tunnel passing under the village, and the construction of a dam far upstream of the village all imply a thorough familiarity not only with the area but also with the principles of hydraulic engineering. our work has also revealed when this system was put into place. part of the first village, including the location of the first mosque, is now occupied by fields and irrigated by the new system, while parcellation and the old canal’s pattern show a different irrigation system. The conquest of the first village and the destruction of the initial hilltop granary resulted in a shift of the village location and in the irrigation of the freed area. Conclusion The comparison of both areas investigated in our study clearly demonstrates not only a transfer of technology between countries, but also the transfer of a way of life between similar landscapes. over the last decennium, the subject matter became undoubtedly archaeological: in spain, citrus fruit production on the plateaus—using fossilised water sources!—has led to the abandonment of the valley fields and of the use of the irrigation systems. in Morocco, for other reasons, the villages are losing their independent approach to subsistence and to production, and with this loss the associated systems, described above, are disappearing. This comparative study provides a means by which we can more clearly understand the way in which the latter community might have functioned during the medieval period. chapter Seven the future of venice’S paSt and the archaeology of the north-eastern adriatic eMporia during the earlY Middle ages sauro gelichi1 Were there Italian Emporia in the Early Middle Ages? The debate about the archaeology of early medieval emporia has been, and essentially remains, entirely north-european, due to both its greater visibility in those regions and the intrinsic nature of archaeology in those countries.2 of course, the development of these investigations has produced records that have been immediately used for a more general discussion of the economy in europe and the Mediterranean from late antiquity to the carolingian age, following in the wake of a critical revision of the views of the well-known belgian historian pirenne.3 developed during the first decade of the last century, the pirenne thesis has been variously discussed and disputed over time.4 however, due to its overall persuasive nature, it continues to be one of the few examples that attempts to explain the developments and mechanisms of the economy in early medieval europe in general terms. lying in the background of this theory and linked to pirenne’s view of europe as a centre of the carolingian economy is the question of towns and their function in the early medieval period. This is a question to which pirenne had also made a particular contribution in terms of a reassessment of their role and function, at least before università ca’ Foscari—venezia. r. hodges, Dark Age Economics: the origins of towns and trade AD 600–1000 (london, 1982); M. anderton, ed., Anglo-Saxon trading centres: beyond the emporia (glasgow, 1999); d. hill and r. cowie, eds., Wics. The Early Mediaeval Trading Centres of Northern Europe (sheffield, 2001); t. pestell and K. ulmschneider, eds., Markets in early medieval Europe: trading and productive sites, 650–850 (macclesfield, cheshire, uK, 2003). 3 r. hodges and d. Whitehouse, Mohammed, Charlemagne and the origins of Europe: archaeology and the Pirenne thesis (london, 1983). 4 h. pirenne, Mahomet et Charlemagne (Brussels, 1937). 1 2 176 chapter seven the year 1000, with some exceptions that include, in italy, venice.5 it is not appropriate to discuss here the reasons for a view so little predisposed towards the phenomenon of urban development which, at least in italy, was of limited entity and minor importance,6 as has been shown by italian scholars in the past.7 it only remains to wonder, beyond begging the question on the basis of non-archaeological records, about the nature of the role of at least some of these towns (or presumed towns) in that crucial period (between the seventh and ninth century) that saw the final collapse of the economic systems of the ancient world and the beginning of new trading systems.8 in the early medieval period, barring a few caveats, italy remains essentially a land of towns;9 and it is here that, together with a few vanished towns and many surviving ones, certain ‘new’ towns make their appearance. This fact has clearly been known for some time, yet scarcely analysed in its entirety. if one takes a quick look at the map of urban sites in northern italy in the early medieval period one sees immediately that while some areas show a decline of towns (the southern piedmont region, for example) others compensate for this with the creation of new, inhabited urban centres (the north-east coast on the adriatic, for example) (Fig. 1).10 it is perhaps not by chance that at the heart of this flourishing of ‘new towns’ there lies the lagoon where, as we know, venice was to rise. But what are these ‘new towns’? What do they have in common? What is their purpose? and lastly, can they be defined as ‘emporia’ 5 h. pirenne, Medieval Cities; their origins and the revival of trade (princeton, 1925). 6 g. p. Brogiolo and s. gelichi, La città nell’alto medioevo italiano: archeologia e storia (rome, 1998). 7 c. violante, La società milanese in età precomunale (Bari, 1953). 8 in terms of the collapse of the economic systems of the ancient world, i am in agreement with the views published in B. Ward-perkins, The Fall of Rome: and the end of civilization (oxford, 2005). on the question of the “various early medieval economies,” see the timely observations expressed in J. moreland, “concepts of the early medieval economy,” in The Long Eighth Century, eds. i. hanson and c. Wickham (leiden, 2000), pp. 1–34. 9 s. gelichi, “The cities,” in Italy in the early Middle Ages: 476–1000, ed. c. la rocca (oxford, 2002), pp. 168–174. 10 For the distribustion of urban sites in northern italy in the early medieval period, see B. Ward-perkins, “The towns of northern italy: rebirth or renewal?,” in The rebirth of towns in the West AD 700–1050, eds. r. hodges and B. hobley (london, 1988), pp. 16–27 and fig. 6. extensive near perfect little or none some considerable N New towns emerging or emerged by c.800 Towns continuously settled, with indication of Roman grid survival: Areas of surviving centuriation 10 17 1 20 24 o cin Ti 61 58 41 Turin 69 35 31 30 2 5 Venice 7 64 42 45 Po 54 Adige 50 74 40 56 27 23 70 28 51 44 59 19 46 12 36 53 63 52 43 68 38 37 39 4 65 Verona 22 16 57 8 Milan 48 71 21 62 72 14 73 Aquileia 49 13 11 55 32 26 Ravenna 18 29 60 67 25 Luna 3 0 47 the future of venice’s past 9 66 Land over 500m Towns abandoned or radically decayed by c.800 33 34 15 6 100km 177 Figure 1: towns in northern italy during the early middle ages (after Ward-perkins 1988). 178 chapter seven or anyway compared to the emporia of northern europe, as certain scholars have been suggesting for some time? The term ‘emporium’ (but above all the concept of ‘emporium’) as a place for the redistribution of goods, essentially belongs, as mentioned above, to the north european historiography debate.11 While there may not be uniformity of opinion as to its meaning, following hodges, one may define an emporium and distinguish it from other contemporary settlements on the basis of these features: the role played by the merchandise that passed through there (type and quantity); its spread and the kind of materials used to construct it; and lastly, whether it was also a centre of production.12 do these features belong to the ‘new towns’ in the northern adriatic area just mentioned, venice included? That archaeology may prove a decisive tool in shedding light on these questions is shown by the investment in resources that much of north-european archaeology has devoted to the analysis and study of these sites. one cannot say the same for italy, where research in the venetian lagoon and, more generally, in the north-east coastal region on the adriatic has followed quite different directions and issues (Fig. 2). it may also be interesting to try to understand whether archaeology has been systematically conducted in these areas and if so, towards which topics it has been directed. The present author remains convinced that this subject is of enormous potential but has, as yet, been little developed. Where work has been conducted, it has centred on venice and its lagoon. returning to archaeology in the venetian lagoon and reflecting upon these sources is therefore also fundamental for reviving and broadening the debate about the european economy in the early medieval period, which currently runs the risk of overlooking the importance of places or of assessing them on the basis of utterly inadequate records. The main purpose of this chapter, therefore, is to reflect upon these issues while discussing the nature of earlier archaeology in the venetian lagoon (analysing the past); evaluating its present approaches (discussing the present); placing its potential in a broader regional picture (The other venices); and lastly, proposing a different strategy for the future (planning the Future). 11 see, for example, d. a. hinton, “metalwork and the emporia,” in Anglo-Saxon trading centres: beyond the emporia, ed. m. anderton (1999), pp. 24–31. 12 hodges, Dark Age Economics, pp. 47–65 and 104–129. the future of venice’s past 179 Figure 2: main towns in the “venetia et histria” during the early middle ages. 180 chapter seven Analysing the Past if one analyses in perspective the kind of archaeology undertaken in the lagoon up to the 1990s (Fig. 3), one sees that it can be broken down into two main stages. The first, perhaps over-emphasised by subsequent archaeological reviews, is shown in the investigations that giacomo Boni managed to undertake, towards the end of the nineteenth century, into the foundations of saint mark’s bell tower (Fig. 4).13 The second consists of the excavations made by polish archaeologists at the start of the 1960s on the island of torcello, in the vicinity of the large church buildings of santa Fosca and santa maria assunta (Fig. 5).14 These two events, far apart in time and located in two rather different environments (urban in the first case, but not the second), are isolated episodes in the context of italian archaeology. in particular, the polish excavations are noteworthy for their innovative methodology, the tools used, and the historical issues they were meant to deal with, especially if one considers that these tools were applied to study a post-antique context. unfortunately, this project had no more than a modest and irregular follow-up and was not able to typify and direct archaeology in the lagoon in subsequent years.15 however, at the same time, it did have a modest influence on the rise of italian medieval archaeology.16 For a long time, archaeological research focused on the early medieval period was ignored as being of no use to the subjects typically investigated in the field of classical archaeology (for example, the study of roman or greek origins) or else was only taken up again for the purpose of certifying just those topics.17 it should not seem surprising, therefore, that until now archaeology has contributed little to the history of the city and, above all, has provided only feeble support in defining that very problem of origins that has always been the real hub around which much of local history turns. even so, beyond the limits 13 g. Boni, “il muro di fondazione del campanile di s. marco a venezia,” Archivio Veneto 29/2 (1885), 354–368. 14 l. leciejewicz, e. tabaczyńska, and s. tabaczyński, Torcello: scavi, 1961–62 (rome, 1977). 15 l. leciejewicz, ed., Torcello: nuove ricerche archeologiche (rome, 2000). 16 s. gelichi, Introduzione all’archeologia medievale: storia e ricerca in Italia (rome, 1997), pp. 70–78. 17 see, for example, W. dorigo, Venezia origini: fondamenti, ipotesi, metodi (milan, 1983). the future of venice’s past 181 Figure 3: schematic image of the history of archaeological research in venice. of the lagoon, venice has always been recognized as being central to the picture of the economic history of the early middle ages in europe and the mediterranean (supra) (Fig. 6). in this context, it is the task of archaeology to shed more light on the subject than historical, literary, and diplomatic sources have been able to do thus far.18 in recent years, non-selective archaeology on a large scale has been practised, thanks to the commitment of the two soprintendenze that regularly follow the excavations and restoration of historical monuments in the city centre and on all the lagoonal islands.19 The inspection process has also been extended to the city’s underwater heritage. Thus, a new period of archaeological research seems to have opened in venice: it remains only to ask oneself, at the service of which projects and with what results. 18 see r. hodges, Towns and trade in the age of Charlemagne (london, 2000), pp. 59–64; m. mccormick, Origins of the European economy: communications and commerce, A.D. 300–900 (cambridge, 2001), pp. 529–531. 19 see l. Fozzati et alii, “archeologia delle fondamenta veneziane,” in Tra due elementi sospesa: Venezia, costruzione di un paesaggio urbano, ed. e. concina (venice, 2000), pp. 134–166; m. de min, “edilizia altomedievale e medioevale nel territorio lagunare. nuovi dati conoscitivi dai cantieri di restauro,” ibid., pp. 98–133; idem, “venezia e il territorio lagunare,” in Ritrovare restaurando: rinvenimenti e scoperte a Venezia e in laguna (venice, 2000), pp. 15–25. 182 chapter seven Figure 4: venice, saint mark’s bell tower, foundation—1885 (after Boni). Discussing the Present The lagoonal islands feature a collection of settlements of different types (Fig. 7). some of these have resulted in inhabited areas still alive today, the main one being the city of venice; while others are places partly (e.g. torcello) or completely abandoned (e.g. san lorenzo di ammiana). This diversity influences not only the quantity but also the quality of archaeological deposits. Whatever the object of one’s archaeological research, scholars undoubtedly have to take account of this difference. archaeological research up to now has operated in both types of site. The city of venice, still a vital centre, could only be tackled according the future of venice’s past 183 Figure 5: The glass workshop of torcello, venice. polish excavations 1961–1962 (after leciejewicz, tabaczyńska, tabaczyński 1977). 184 chapter seven Figure 6: venice between history and archaeology. to the principles of urban archaeology: archaeology which sets itself the task of studying the city’s history, but which may also be useful in analysing the periods prior to its formation. essentially similar features are found in other sites such as murano and Burano, which are distinguished from venice only by their lesser extent as settlements and the location of fewer monuments in their historic centres, but in which the deposits are subject to the same kind of stress and behaviour. outside of this context, the abandoned or partly abandoned sites outline different situations. These are generally places that had fewer settlements, both in terms of density and often also duration. among them are places inhabited only in late antiquity and reoccupied in the late middle ages (san Francesco del deserto), others which were colonized rather late (san giacomo in paludo, not before the twelfth century), and yet others which show continuity up to the middle ages (san lorenzo di ammiana). many of these places may be important for studying certain topical aspects of lagoonal settlement, such as the phenomenon of monachism in the late middle ages, or the stages the future of venice’s past 185 Figure 7: towns in north-eastern italy during the early middle ages including the main settlements in the venetian lagoon. 186 chapter seven of military occupation of the lagoon during the austrian period.20 in addition, many of these places may be of considerable interest in understanding the stages of formation of lagoonal settlements in the early middle ages. The archaeology carried out in recent years in the lagoon is notable for the number of controlled operations, but has been undertaken without any kind of selectivity. as a result, a variety of sites have been excavated and presumably recorded in the same way. unfortunately, there is not enough data available regarding quantity and quality to verify the effectiveness of this work throughout the lagoonal islands. however, some evaluations are possible concerning the city of venice, a geographic zone which, in the incidence of operations, is one of the most excavated places in the whole area. This work may be assessed according to three aspects: how much has been excavated, where it has been excavated, and what has been excavated.21 as no absolute quantitative data has been provided in terms of cubic metres of earth archaeologically removed, an idea of quantity may only be expressed through the number of excavations carried out. These amount to 35 in just the historic centre of venice and reach almost 100 if the entire lagoon is included.22 on the basis of this data one may rightly affirm that venice has been one of the most heavily excavated cities of italy during the last ten years. however, if one considers the effect of this activity in terms of historical knowledge, one realizes that this assessment needs a marked reappraisal, as only less than half of the excavations are denoted by preliminary reports without a following final report (41%), and more than half the operations (59%) have come to scholars’ attention solely thanks to fleeting references or newspaper articles (Fig. 8). restricting our attention to the city centre alone, the topographic location of the excavations is also of some importance (Fig. 9). By far 20 s. gelichi et alii, “isola di san giacomo in paludo (laguna nord, venezia): gli scavi delle campagne del 2003 (sgp03a e sgp03b),” Quaderni di Archeologia del Veneto 20 (2004), 160–177; s. gelichi et alii, “identity marks, organization of spaces and characteristics of consumption of an island of the venetian lagoon between the later middle ages and the modern age,” in Constructing Post-medieval Archaeology in Italy: a New Agenda, eds. s. gelichi and m. librenti (Florence, 2007), pp. 97–108. 21 s. gelichi, “venezia tra archeologia e storia: la costruzione di un’ identita urbana,” in Le città italiane tra la tarda antichità e l’alto Medioevo, ed. a. augenti (Florence, 2006), pp. 151–186. 22 l. Fozzati et alii, “archeologia delle fondamenta veneziane,” p. 135. the future of venice’s past 187 Figure 8: excavations in the centre of venice (after gelichi 2006). 188 chapter seven Figure 9: excavations in the centre of venice (after gelichi 2006). the future of venice’s past 189 Figure 10: archaeology in the venetian lagoon: main topics. the great majority of these operations are carried out in areas which, on the basis of other sources, we may surmise to be outside (and sometimes far from) settlement areas of the early middle ages (between the ninth and tenth centuries). Therefore, if one’s aim were that of studying the formation of the first venetian settlement between rivoalto and olivolo, it is clear that even when published, these excavations would probably be of little relevance. still, this does not mean that they cannot shed light on other topics such as patterns of urban expansion or the features of venetian material culture in the late middle ages and post middle ages. Finally, if we analyse what has been excavated (Fig. 10), it becomes fairly clear that most operations have uncovered the construction and settlement phases of ecclesiastical buildings such as churches and monasteries (32%), followed by canal embankments (18%). other topics such as houses (13%), or buildings connected to goods, trading and production sites (3%) are encountered less frequently. in this case the data are not so surprising if one only considers the high density of religious buildings subject to restoration or the need for continual reclamation work in the canals, which naturally allows also for the investigation of the canal banks and their development over time. paleoenvironmental research must be considered separately, as at present it occurs in a relatively high quantity (21%) and is often selective, with specific coring operations sometimes programmed and 190 chapter seven carried out.23 it is not by chance, therefore, that it is in this field that the best results have been recorded with regard to knowledge of the transformations the lagoon has undergone from ancient times until the present day. The nature of the excavations together with the scarcity of publication clearly explains why such a striking accumulation of operations has yielded no corresponding new interpretation of the city’s topography during its period of origin, with the exception of those deriving from paleoenvironmental analyses. instead, careful observation in fact reveals that some recent attempts at a summary of residential buildings and canal embankments essentially derive from written sources and attempt to interpret the archaeological record (which again, is almost completely non-existent, also with regard to residential buildings) on the basis of these, with a view to confirming them.24 even the volumes that dorigo has dedicated to venice may be included, partially at least, in this kind of production. The first, written when archaeological research in the lagoon was not able to count on scientific investigations, uses the archaeological data (whether one agrees with it or not) only to confirm the existence and nature of a stable occupation during roman times.25 The most recent is one of those excellent products of historical topography, in which the meticulous reconstruction of material assets (types of building construction, topographic collocation of construction, and town plan of dwellings) is mainly based upon what may be drawn from written sources (or the remains, generally rather late, of church buildings).26 if the churches and monasteries, therefore, are the sites most thoroughly investigated, it is not surprising to find yet again that the archaeology of medieval architecture (although a stratigraphic analysis is not used with regard to constructions) is the field that provides the most complete, organic a. J. ammerman and c. e. mcclennen, eds., Venice before San Marco: recent studies on the origins of the city (colgate, 2001). 24 For a summary of residential buildings, see m. Bortoletto, “cenni sull’edilizia minore veneziana alla luce di alcuni rinvenimenti archeologici,” Archeologia delle Acque, 2/1 (2000), 9–20. For a summary of embankments, see idem, “de canalibus, rivis, piscinisque: primi passi verso un’ archeologia idronomastica veneziana,” in l. Fozzati et alii (2000), pp. 136–152. For a study of the written sources, see the aforementioned work, W. dorigo, L’edilizia abitativa nella “Civitas Rivoalti” e nella “Civitas Veneciarum” (secoli XI–XIII) (venice, 1993). 25 W. dorigo, Venezia origini. 26 idem, Venezia romanica: la formazione della città medioevale fino all’età gotica (venice, 2003). 23 the future of venice’s past 191 summaries. This is true both for defining the planimetric development of buildings and for attempting some more general interpretation of the features of building techniques.27 it is probable that for studying the formation of lagoonal settlements up to the ninth century, archaeology in venice is of little significance (certainly no more than in other places). From this viewpoint, the archaeology practised so far in the city seems to have eluded the crucial points of settlement prior to and contemporary with the creation of the rivoalto. different issues affect the abandoned or quasi-abandoned sites. in these cases there has been only a moderate invasion of later building where it is not altogether absent. one may therefore presume to have available archaeological deposits which are better conserved and may be investigated over a wider area. Three places in particular have been excavated and, in part, published. The first, san Francesco del deserto, has revealed very interesting traces of occupation from the period of late antiquity (Fig. 11).28 The sequences of this period, together with those of torcello of the same period, are the best available at present concerning the history of the lagoon (inasmuch as they do not concern dwelling sites but only embankment constructions). The pottery classifications dating back to between the fifth and seventh centuries, certainly contemporary with the creation of these embankments, indicate a picture of the circulation of pottery and amphorae that appears to offer a different scenario from that reconstructed through the use of historiography and also through a summary reading of the famous passage by procopius.29 in any case, the settlement does not appear to have persisted (at least in the areas examined) and so carrying on research at this site seems to be more promising for understanding the nature of late roman sites, rather than those of the early middle ages. 27 For a more general interpretation, see de min, “edilizia altomedievale,” pp. 98–133. For a discussion of the planimetric development of san lorenzo di castello, see idem, “venezia: chiesa di san lorenzo di castello: un esempio di scavo correlato al restauro architettonico,” in Ritrovare restaurando. Rinvenimenti e scoperte a Venezia e in laguna (venice, 2000), pp. 40–47. 28 idem, “edilizia altomedievale,” pp. 98–133; idem, Venezia e il territorio lagunare, pp. 15–25. 29 see e. grandi, “le ceramiche fini da mensa dalla laguna veneziana. i contesti di san Francesco del deserto e torcello,” in La circolazione delle ceramiche nell’Adriatico tra Tarda Antichità ed Altomedioevo: III Incontro di Studio Cer.am.Is sulle ceramiche tardoantiche ed altomedievali, eds. s. gelichi and c. negrelli (mantua, 2007), pp. 127–153. 192 chapter seven Figure 11: venetian lagoon, san Francesco del deserto, late antique waterfronts. the future of venice’s past 193 The second site is san lorenzo di ammiana, excavated on two occasions by different teams. in this case, only the publication of research undertaken by canal in the 1980s and some brief indications of the 1991 excavation are available (Fig. 12).30 This place, in a really hopeless state of conservation, would appear to have preserved a better stratigraphic progression of about 4 metres in thickness lacking any interruption from the late roman period to the late middle ages. it is clear that this site offers a considerable opportunity for early medieval archaeology in the lagoon taking into consideration the objective difficulties of access to the location and of excavation. however, these evaluations are only possible in intermediate form, through a fragmented and scarcely coherent archaeological record. Furthermore, this record is partly conditioned by the identification of this place with the castle of Castratium, of which remains have been recognised in certain wall structures (pilasters). regarding torcello, it may seem rather surprising that the excavations carried out so far have revealed little about that ‘emporion mega’ (as constantine porphyrogenitus still calls it at the start of the ninth century), while the findings relate mainly to the date and architectonic sequence of the baptistery or the church of santa maria assunta.31 if, as has been recently suggested, we must post-date the workshop for glass production brought to light, between 1961–62, opposite the church of santa Fosca, “to not earlier than the 9th century,” it seems we may gain an understanding of something more than just the emporium. But we may be greatly disappointed concerning other aspects of the site, such as the dwellings or the plan of the settlement.32 The explanation for this is easily found in the location of the various investigations, almost all of which were made near the church buildings (Fig. 13). it is not by chance then, that the only workshop discovered dates prior to the construction of the church of santa Fosca to which 30 l. Fersuoch et alii, “indagini archeologiche a san lorenzo di ammiana (venezia),” Archeologia Veneta 12 (1989), 71–82; g. p. Brogiolo and s. gelichi, Nuove ricerche sui castelli altomedievali in Italia settentrionale (Florence, 1996), pp. 44–49. 31 on the excavations, see leciejewicz, ed., Torcello; m. Bortoletto, “murano, mazzorbo e torcello: tre siti a confronto. indagini archeologiche nella laguna nord di venezia,” Archeologia delle Acque 1/1 (1999), 55–74. For constantine porphyrogenitus’s comments, see De Administrando Imperio, pp. 27–28. For santa maria assunta, see de min, Venezia e il territorio, pp. 23–25. 32 l. leciejewicz, “italian-polish research into the origin of venice,” Archaeologia Polona 40 (2002), 58. 194 chapter seven Figure 12: venetian lagoon, san lorenzo di ammiana, plan of the excavations (after canal). the future of venice’s past 195 Figure 13: venetian lagoon, torcello, hypothesis concerning settled areas during the eighth–ninth centuries ad. 196 chapter seven it now lies near, and that the only information known concerning dwellings relates to the periods preceding the building of a span of the cathedral of santa maria assunta. overall, the archaeology of these three sites, which perhaps lend themselves better than those in the centre of venice to an understanding of the nature of lagoonal settlement before the ninth century, has also been disappointing. This may be partly due to the nature of the stratification (san Francesco del deserto), and partly also to strategies (san lorenzo di ammiana and torcello). on the whole, non-selective archaeology such as that undertaken in recent years does not seem to be the best way of shedding light on the question of the origins of venice. The Other Venices archaeology in venice is not only possible, as has been amply demonstrated by the number of excavations that are effected every year in the city and in the lagoon, but it is also useful. however, for this archaeology to really be of service to history it needs to be guided by a problem-oriented approach. over time, many research aims have been proposed for the city. some are false problems or issues of little relevance, such as that of the existence of a ‘permanent settlement’ even in roman times. it reminds one of the need, plausible in the laudatory aims of the Chronicle of giovanni diacono, to restore to the city a sheen of nobility through reclaiming its ‘classical’ aspect (not to mention those who have tried to find traces of aegean presence there). others are more intriguing issues, such as those concerned with understanding the formation of lagoonal settlement in the early middle ages through an examination of its topographic position, its features, and the quality and nature of its economy. however, the present author remains convinced that for a full understanding of these issues they must not only be freed from many stereotypes, but must also be linked to episodes and aspects that distinguished the entire northern adriatic area between the fifth and tenth centuries.33 This area was one of many s. gelichi, “Flourishing places in north-eastern italy: towns and emporia between late antiquity and the carolingian age,” in Post-Roman Towns, Trade and Settlement in Europe and Byzantium, vol. 1, The Heirs of the Roman West, ed. J. henning (Berlin, 2007), pp. 77–104, in terms of both settlement sites and economic connotations. 33 the future of venice’s past 197 emerging new towns and venice is nothing but the outcome of a long competition, which was not restricted only to the lagoonal area. up until the ninth century, the destiny of these places was not yet determined. archaeological research in the area of comacchio (the po delta) (Fig. 14) is increasingly demonstrating the role of this emporium during the course of the eighth century, bringing out of isolation that remarkable record of the Capitolare for trade along the po, drawn up by the inhabitants of comacchio with the longobard king liutprand, probably in 715.34 Furthermore, the people of comacchio did not only transport salt, as rather cursory attempts have been made to show, but also other kinds of merchandise coming partly from the eastern mediterranean, such as oil, wine and spices (Fig. 15).35 This central role has been archaeologically confirmed not only by the conspicuous presence of amphorae (the mineral and petrographic analyses of which indicate that they come from both southern italy and the eastern mediterranean area) (Fig. 16) but also by the remains of infrastructures found on the site of valle ponti (wharves) (Fig. 17). Thus, comacchio represents a genuine port at the convergence of communication routes linking the adriatic coast to the internal marshland (to the south) and the main branch of the po (to the north) (Fig. 18).36 it is clear that these records oblige us to review the rather summary positions taken concerning not only the po valley trade of the eighth century, but also the might of the longobard kings and, more generally, the economic stagnation of the eighth century.37 34 see l. m. hartmann, Zur Wirtschaftsgeschichte Italiens im frühen Mittelalter: Analekten (gotha, 1904); m. montanari, “il capitolare di liutprando: note di storia dell’ economia e dell’ alimentazione,” in La civilta comacchiese e pomposiana dalle origini preistoriche al tardo medioevo (Bologna, 1986), pp. 461–475. 35 s. gelichi, “The eels of venice. The long eighth century of the emporia of the northern region along the adriatic coast,” in 774: Ipotesi su una transizione, ed. s. gasparri (turnhout, 2008), pp. 81–117. 36 s. gelichi et alii, “Comacchio tra iv e X secolo: territorio, abitato e infrastrutture,” in IV Congresso Nazionale di Archeologia Medievale, eds. r. Francovich and m. valenti (Florence, 2006), pp. 114–123; s. gelichi et alii, “. . . igne castrum combussit. . . . comacchio tra la trada antichità e l’alto medioevo,” Archeologia Medievale 32 (2006), 19–48; s. gelichi, ed., Comacchio e il suo territorio tra la Tarda Antichità e l’Altomediovo (Ferrara, 2007), pp. 363–685; idem, ed., L’isola del vescovo. Gli scavi archeologici intorno alla Cattedrale di Comacchio. The Archaeological Excavations nearby the Comacchio Cathedral (Florence, 2009). 37 regarding the might of the longobard kings, see r. Balzaretti, “cities, emporia and the monasteries: local economies in the po valley, c. ad 700–875,” in Towns in transition: urban evolution in late antiquity and the early Middle Ages, eds. n. christie and s. t. loseby (aldershot, 1996), pp. 213–234. regarding the more general issue of 198 chapter seven Figure 14: location of comacchio. the future of venice’s past 199 Figure 15: goods traded by Comaclenses inhabitants. The extent to which such a powerful role is also evident in the venetian lagoon at around the same period is still awaiting archaeological proof. unfortunately, the indiscriminate gathering of data seems to confuse, rather than clarify, the picture. at the same time, some stereotypes and prejudices need to be abandoned. The much-vaunted Byzantine venice appears increasingly as an ambiguously contoured idea, also at an institutional level.38 Therefore, in order to reconstruct lagoonal the economic stagnation of the eighth century, see c. Wickham, “overview: production, distribution and demand, ii,” in The Long Eighth Century. Production, Distribution and Demand, eds. i. l. hansen and c. Wickham (leiden, 2000), pp. 345–377; idem, Framing the early Middle Ages: Europe and the Mediterranean 400–800 (oxford, 2005); s. gelichi, “una discussione con chris Wickham,” Storica 34 (2006), pp. 134–147. 38 s. gasparri, “venezia fra l’italia bizantina e il regno italico: la civitas e l’assemblea,” in Venezia: itinerari per la storia della citta, eds. s. gasparri, g. levi, and p. moro (Bologna, 1997), pp. 61–82. 200 chapter seven Figure 16: eighth-century amphorae from comacchio. the future of venice’s past 201 Figure 17: comacchio, valle ponti—villaggio san Francesco 1996, wooden infrastructures. settlement patterns between the fifth and the eighth century, we should look more at contemporary villages along the po river valley (and perhaps also at the great emporia of northern europe) rather than at the architecture of surviving eastern Byzantine (and arab) towns. From this point of view, a remarkable site is that of cittanova eraclea, where another buried venice has been sought and, to some amazement, has not been found (Fig. 19). in fact, this settlement, established according to tradition by the emperor heraclius, reflects the ways of living and building in the lagoon in the early middle ages more accurately than archaeologists have initially realized. archaeological surveys and the few excavations made have shown an organized construction of an area along a canal (dwellings on areas of land surrounded by ditches and with wharves leading out onto the water) with a settlement centre of an institutional nature (seat of the duke and the bishop) located to the northeast.39 only in this latter part do brick 39 d. calaon, “cittanova (ve): analisi gis,” in V Congresso nazionale di archeologia medievale, eds. r. Francovich and m. valenti (Florence, 2006), pp. 216–224. 202 chapter seven Figure 18: comacchio in the early middle ages. the future of venice’s past Figure 19: cittanova in the early middle ages. 203 204 chapter seven buildings appear to be found. my impression is that a similar situation would also emerge in torcello, if excavations were only carried out in the areas not occupied by the churches. many historians are well aware of these issues, but they ask no more of archaeologists than the confirmation of their ideas that are based on written records. archaeology of this kind has little significance; it generally finds the answers that are already known and tends to fix the information that the archaeologist, often by chance, has found onto a predetermined procedure. historians and archaeologists of the early middle ages are amazed that a city like venice has not brought back to light the evidence, in material records of the early middle ages, that its position and role in history would decree. one also wonders whether, indeed, the archaeological information may help us to have a better understanding of these periods, and therefore whether it is wise to have to relinquish it. although they are of a fragmentary nature, the publications in fact afford a glimpse of considerable potential, at present scarcely or not at all utilized. indeed, many themes that are exquisitely archaeological, for instance topographic subjects, are still exclusively based on written sources, even though recent examples have shown that the same items of information say different things, if one only looks at them from a different angle.40 one has the impression, moreover, that the birth and development of venice (a ‘winner’ site) cannot be understood without also analysing the ‘loser’ sites. in the eighth and ninth centuries, the whole of the north-east coastal region, as we have seen, was an area of intense activity and competition. This competition developed on two levels: one, of a local nature, concerning the venetian lagoon and the nearby areas whose diverse destinies are recorded in detail in written records which, in describing the successive shifts of power (from cittanova to metamauco, from metamauco to rivoalto), help us to understand the economic logic behind them (from a society whose fortunes depended on landed property, to one that shifted its interests to trade); and a second level, concerning all the stretch of land from the venetian lagoon 40 For example, a. J. ammerman, “venice before the grand canal,” Memoirs of the American Academy in Rome, 48 (2003), 141–158, where the author discusses the topography of venice in the eighth century. the future of venice’s past 205 to ravenna, where other centres (among which comacchio stands out) seem interested in playing the same game. The archaeology of ‘loser’ sites is, as we have seen, extremely rewarding because it enables us to better construe the original nature of these places, which, moreover, have not undergone the great transformations of a surviving town. The research carried out so far at cittanova, as well as that now underway in comacchio, are an undeniable demonstration of this. however, one may also return to the venetian lagoon and consider a different archaeological approach. Planning the Future The Future of London’s Past is a book that marked a turning point in urban archaeological research and the concept of preventive archaeology in europe.41 it meant that an archaeological map of a city centre was transformed from a simple instrument of historical knowledge into a model for designing and planning research. understanding and defining how much, where, and what was conserved of our archaeological heritage came to mean also obtaining the most suitable tools for working with awareness on its future. The london project became an example throughout europe and, although only at a later date, also in italy. here, during the early eighties, the first map of urban archaeological risk was published, to be followed only by other sporadic episodes.42 This means that the question of envisaging the archaeological potential of a site in order to take protective action is not yet felt to be a useful instrument. instead, the responsibility for continuing to preside over all urban archaeology is 41 see m. Biddle and d. hudson, The future of London’s past: a survey of the archaeological implications of planning & development in the nation’s capital (Worcester, 1973). 42 h. andersson, “The era of town inventories. a kind of evaluation,” Medieval Europe 1992—Pre-printed Papers, vol.1, Urbanism (york, 1992), pp. 15–26; p. hudson, Archeologia urbana e programmazione della ricerca, l’esempio di Pavia (Florence, 1981); g. p. Brogiolo, ed., Archeologia urbana in Lombardia. Valutazione dei depositi archeologici e inventario dei vincoli (modena, 1984); s. gelichi, a. alberti, and m. librenti, Cesena: la memoria del passato. Archeologia urbana e valutazione dei depositi (Florence, 1999). 206 chapter seven delegated to rescue archaeology, fittingly called ‘reactive archaeology’ by alain schnapp.43 The laws in force in the field of town planning and archaeological practice also make it difficult to find suitable planning means that render urban change compatible with protective needs.44 The shift from a concept of archaeological assets as pure heritage (the manufacture, the monument) to one in which the value lies in awareness (the context: ‘monumentality versus research’, to use a fitting expression suggested by martin carver) demands different ways of relating to the past.45 Furthermore, this extension in terms of the type and chronology of the concept of a cultural asset and therefore also of an archaeological asset has made it even more difficult to manage and conserve them. Faced with the ‘misfortunes of abundance’, it is necessary to choose not to extend in a totally simplistic fashion a general, indiscriminate control. The danger in doing so lies in the risk of creating an impasse due to excess in the accumulation of data, the weakness of routine practice in diagnostic procedures, and the futility of redundancy in the repetition of investigated cases.46 one of the expressions most frequently associated with venice is that it is an extraordinary city. nobody can doubt this claim if they only think of the uniqueness and special nature of the site. But, does this mean that her archaeology is also extraordinary, or rather, does not follow the rules? does it mean that venice cannot be investigated and discovered by applying to her the same diagnostic tools that are used in all other cities? i previously raised the issue of why the archaeology of this city has contributed little, so far, to solving the question of origins; and in particular why it has little significance, at present, for understanding those very centuries, the ninth and tenth, that were undoubtedly the crucial moment in the rise of the settlement around the rivoalto on one hand 43 a. schnapp, “archeologia urbana, archeologia preventiva,” Archeologia urbana e centro storico di Napoli (naples, 1984), p. 25. 44 g. p. Brogiolo, “attori, autori e ‘fruitori’ del progetto archeologico,” in Archeologia e urbanistica, ed. a. ricci (Florence, 2002), pp. 305–318; s. gelichi, “città pluristratificate: la consistenza e la conservazione dei bacini archeologici,” ibid., pp. 61–76. 45 m. carver, Archaeological value and evaluation (mantua, 2003), p. 40. 46 a. ricci, I mali dell’abbondanza, Considerazioni impolitiche sui beni culturali (rome, 1996); m. carver, “on archaeological value,” Antiquity 70, no. 67 (1996), 45–56. the future of venice’s past 207 and the affirmation of venetian dominance in the adriatic on the other. indeed, historians and archaeologists of the early middle ages in the mediterranean and europe have, up to now, only been able to count on numismatic evidence, together with the development of religious institutions interpreted as an expression of economic resources.47 This interpretation would surely be more rewarding if consideration were also taken of the archaeological data relative to such institutions and not only the information, which is still very debatable, deriving from hagiographic sources and chronicles.48 in part, i have indirectly answered this question when considering the features of venetian archaeology in which i identified the elements of its intrinsic weakness (supra). one may go further, however, and assess this problem on a narrower scale of detail and on sites not lacking in important archaeological investigation, to try and understand also whether the results obtained could have been predicted. an example is the island of olivolo, near the rivoalto, a diocese from the beginning of the eighth century and also the subject of an important archaeological report published in detail. in the 1980s, an extensive stratigraphic investigation was carried out on the island on the exterior of the apse of the church of san pietro in castello.49 This excavation revealed a sequence relating to a permanent settlement from the beginning of the fifth century up to the seventh century. subsequently, however, the archaeological record seems to come to a halt and to start up again with the late and post-medieval periods (even these are scarcely represented). This is an area, then, that from the beginning of late antiquity was densely inhabited and in the early middle ages became a large open space, perhaps a market garden. This occurred just at the time when the diocese was established at olivolo. such a situation, which might at first seem incongruent, may find a logical explanation when we consider once again the location of this case in point (immediately outside the present day church). 47 r. hodges, Towns and trade; m. mccormick, Origins of the European economy, pp. 530–531. 48 see also gelichi, “venezia tra archeologia,” pp. 151–186. 49 s. tuzzato, “venezia. gli scavi a san pietro di castello (olivolo). nota preliminare sulle campagne 1986–1989,” Quaderni di Archeologia del Veneto 7 (1991), 92–103; idem, “le strutture lignee altomedievali a olivolo (s. pietro di castello—venezia),” Studi di archeologia della X Regio in ricordo di Michele Tombolani (rome, 1994), pp. 479–485; s. tuzzato et alii, “san pietro di castello a venezia. nota preliminare dopo la campagna 1992,” Quaderni di Archeologia del Veneto 9 (1993), 72–80. 208 chapter seven it is likely that it was the reorganization carried out on the island to make way for the new workshops connected with the bishop’s residence that caused a shift of the settlement core, the traces of which remain today. But this situation could perhaps have been predicted. in fact, on the site of castello a series of written records (transfers of property) clearly indicate, at least in the twelfth-thirteenth centuries, the presence of a number of buildings to the south-west of the church but never behind it, where a market garden was marked. The early medieval dwellings, then, of both the clergy and of other social groups in some way connected with the diocese must have developed in the area between the bell tower and the present Fondamenta di Quintavalle. The comparison of such information with the prg (general town planning scheme) ‘città antica di venezia’ (‘ancient venice’), which shows in this area the historic buildings still in existence, would confirm this hypothesis. Besides this, the system of bridges, dating to at least the twelfth-thirteenth century, suggests viability located entirely on the western side of the island towards the internal canal of san pietro. in addition, there seems to be sign of an internal landing-place, of a type already existing in the twelfth century, near the church of san sergio and Bacco (Fig. 20). This, then, should be the ideal place to investigate the development of canal embankments and landing-places, possible boat-building yards, buildings for service or trade, and ultimately, dwellings. The assemblage of a whole series of data of various sorts (and not necessarily, or solely, archaeological) directs us, in the case of olivolo, towards identifying specific areas of differing archaeological potential (in terms of quantity and quality of the conserved deposit). This is an aspect that will have to be taken into consideration if, in the future, our resources are to be channelled in an appropriate, effective manner. The general impression one has, therefore, is that venice appears to be an archaeologically complex, yet a quite ‘normal’ place. it is these very characteristics that make it possible to apply the methods and procedures that are now standard for other cities to this city and to the lagoonal islands. in the last two years, a project of this sort has been initiated for the lagoonal area. archaeological data, as well as all the information that could be indirectly related to the deposits have been entered into a gis program with the aim of measuring them using a series of parameters such as chronology, extension, depth, quality, and the future of venice’s past 209 Figure 20: Venetian Lagoon, S. Pietro di Castello—Venice, evaluation of the archaeological resources. 210 chapter seven intelligibility.50 to do this, we have acted upon a very different series of parameters, using information relating to construction (foundation, destruction, and reconstruction of church buildings and dwellings); to the topographic distribution of production, and areas of transformation and distribution (boatyards, arsenals, furnaces, charcoal stores, warehouses, granaries, etc.); and to the ancient hydrography, with particular reference to the filling in of canals. in addition, a survey is being taken of all the archaeological gaps linked, for instance, to the work of remaking embankments and fondamenta, to the (few) vaults, and to the places occupied by wells.51 The non-archaeological sources to which we must turn in order to obtain complete results are, therefore, numerous (from archives to chronicles to maps), and are kept in various organizations and institutes. The aim is to construct means that are able to evaluate the amount and nature of what is conserved and, at the same time, indicate to us the various resources of sites and areas so as to better orientate future archaeological work according to the aims and projects it has set for itself. venice has been waiting for a long time for archaeology worthy of the place. unfortunately, the archaeological history of our country, being scarcely interested in the growth of the multi-period aspect of this subject and tied to a very selective concept in which the middle ages carry no weight, has had a negative influence on the development of research in the lagoon. in recent years, the situation has seen an inversion of this tendency, but paradoxically the results do not seem worthy of the ample commitment. a return to normality may therefore be advisable. to paraphrase an old expression linked to the concept of urban archaeology, one must move from ‘archaeology in venice’ to ‘archaeology for venice’. 50 51 carver, Archaeological Value and Evaluation, pp. 61–64. gelichi, “venezia tra archeologia,” pp. 151–186. liSt of WorKS cited Primary Sources abou’l Modaffer Youssof ibn Kizoghlou. “extraits du Miràt ez Zémân.” in Recueil des Historiens des Croisades, Historiens Orientaux, vol. 3, 511–70. paris, 1884. Al-Muqaddasī. Description of Syria including Palestine. Trans. Guy Le Strange. The Library of the Palestine Pilgrims’ Text Society 3. London, 1896. Albert d’Aix. “Liber christianae expeditionis pro ereptione, emmendatione et restitutione sanctae Hierosolymitanae ecclesiae.” In Recueil des Historiens des Croisades, Historiens Occidentaux, Vol. 4, 265–713. Paris, 1879. Assises de Jerusalem. In Recueil des Historiens des Croisades, Lois, Vol. 1, Assises de la Haute cour, ed. (Le Comte) A. A. Beugnot. Paris, 1841. Bartlett, William Henry, John Carne, William Purser, and Thomas Allom. Syria, the Holy Land, Asia Minor, etc. Illustrated, in a series of Views drawn from Nature by W. H. Bartlett, William Purser, etc. With descriptions of the Plates. [3 volumes in 1] London, 1836–1838. Bernard, Étienne. Recueil des inscriptions grecques du Fayoum, Vol. 2, La “Méris” de Thémistos. Bibliothèque d’étude 79. Cairo, 1981. Bingen, Jean, Adam Bülow-Jacobsen, Walter E. H. Cockle, Francis Kayser, and Wilfried van Rengen, eds. Mons Claudianus, ostraca graeca and latina II. O. Claud. 191 à 416. Documents de Fouilles de l’IFAO 32. Cairo, 1997. Bory de Saint-Vincent, Jean. Relation du voyage de la commission scientifique de Morée dans le Péloponnèse, les Cyclades et l’Attique, Vol. 2. Paris, 1837–38. Buchon, J. Chroniques étrangères relatives aux expéditions françaises pendant le XIIIe siècle, publiées pour la première fois, élucidées et traduites. Paris, 1840. ——. La Grèce continentale et la Morée. Voyage, séjour et études historiques en 1840 et 1841. Paris, 1843. Ciriaco d’Ancona, Edward W. Bodnar, and Clive Foss. Later travels. The I Tatti Renaissance library 10. Cambridge, MA, 2003. Delaborde, H. F. Chartes de la Terre Sainte provenant de l’Abbaye de Notre Dame de Josaphat. Bibliothèque des Écoles Françaises d’Athènes et de Rome 19. Paris, 1880. Delaville Le Roulx, J. “Chartes de Terre Sainte.” Revue de l’Orient Latin 11 (1905– 1908): 181–191. Derenbourg, H. “Autobiographie d’Ousama.” Revue de l’Orient Latin 2 (1894): 327–565. Downey, Glanville, trans. “Libanius’ Oration in Praise of Antioch (Oration XI).” Proceedings of the American Philosophical Society 103 (1959): 652–686. Fatouros, Georgios and Tilman Krischer. Libanios. Antiochikos (Or. XI). zur heidni­ schen Renaissance in der Spatantike. Vienna, 1992. Foucher de Chartres and Jeanne Ménard. Histoire de la croisade: le récit d’un témoin de la première croisade, 1095–1106. Paris, 2001. Gell, William. Narrative of a Journey in the Morea. London, 1823. Great Britain, Naval Intelligence Division. Turkey. 1942. Grenfell, Bernard Pyne, Arthur Surridge Hunt, and David George Hogarth. Fayûm Towns and their Papyri. London, 1900. Guidi, Ignazio. “Una descrizione araba di Antiochia.” Rendiconti della Reale Accademia dei Lincei, Classe di Scienze Morali, Storiche e Filologiche Ser. 5, Vol. 6 (1897): 137–161. 212 list of works cited Holleaux, Maurice. Études d’épigraphie et d’histoire grecques, Vol. 3, Lagides et Séleucides. Paris, 1942. Ibn al Athir, “Kamel al-Tewarikh.” In Recueil des Historiens des Croisades, Historiens Orientaux, Vol. 1, 189–744. Paris, 1872. Ibn Jubayr. “Extrait du Voyage d’Ibn Djobeir.” In Recueil des Historiens des Croisades, Historiens Orientaux, Vol. 3, 441–56. Paris, 1884. Ibn Qalanisi [Ibn-al-Qalānisī, Ibn al-Ḳ alānisī]. Damas de 1075 à 1154. Trans. R. Le Tourneau. Damascus, 1952. Jacques de Vitry. “Historia Ierosolymitana seu Orientalis.” In Itinera Hierosolymitana Crucesignatorum (saec. XII–XIII), Vol. 3, Tempore recuperationis Terrae Sanctae (1187–1244), ed. S. De Sandoli, 297–391. Publications of the Studium Biblicum Franciscanum: Collectio maior 24. Jerusalem, 1983. Jean de Mandeville. “Le Livre de messire Jean de Mandeville.” Translated and Annotated by Christiane Deluz. In Croisades et pélerinages: récits, chroniques et voyages en Terre Sainte, XIIe–XVIe siècle, ed. D. Régnier-Bohler, 1391–1495. Paris, 1997. Jean de Wurzburg. “Descriptio Terrae Sanctae.” In Itinera Hierosolymitana Crucesignatorum (saec. XII–XIII). Vol. 2. Tempore Regum Francorum (1100–1187), ed. S. De Sandoli, 225–295. Publications of the Studium Biblicum Franciscanum: Collectio maior 24. Jerusalem, 1980. Klaas A. Worp, Albert Rijksbaron, and John L. Sharpe. The Kellis Isocrates Codex: (P. Kell. III Gr. 95). Dakhleh Oasis Project Monograph 5. Oxford, 1997. Kramer, Bärbel and John C. Shelton, eds. Das Archiv des Nepheros und Verwandte Texte. Teil I. Archiv des Nepheros. Papyri aus der Trierer und der Heidelberger Papyrussammlung. Aegyptiaca Treverensia 4. Mainz am Rhein, 1987. Leake, William Martin. Peloponnesiaca: a supplement to Travels in the Moréa. London, 1846. ——. Travels in the Morea. With a map and plans. 3 vols. London, 1830. Lébédew [Lebedev], Olga de. Codex 286 du Vatican: Récits de voyages d’un arabe. St. Petersburg, 1902. Mahaffy, John P. The Flinders Petrie Papyri with Transcriptions, Commentaries and Index. Part II. Dublin, 1893. Mahaffy, J. P. and Josiah Gilbart Smyly. The Flinders Petrie Papyri with Transcriptions, Commentaries and Index. Part III. Dublin, 1905. Miller, K. Tabula Peutingeriana. Weltkarte des Castorius: genannt die Peutingerische Tafeln. Farben des Originals herausgegeben und eingeleitet von Dr. Konrad Miller. Ravensburg, 1888. “Patriarcats de Jérusalem et d’Antioche.” In Itinéraires à Jérusalem et descriptions de la Terre Sainte rédigés en franÇais aux XIe, XIIe & XIIIe siècles, eds. Henri Victor Michelant and Gaston Raynaud, 9–19. Geneva, 1882. Pausanias. Description of Greece. Edited and trans. by William H. S. Jones. Cambridge, MA, 1961. Pharr, Clyde, trans. The Theodosian Code and Novels and the Sirmondian Constitutions. Princeton, 1952. Pococke, Richard. A Description of the East and Some Other Countries. [2 volumes in 3] London, 1743–1745. Pouqueville, F.-C.-H.-L. Voyage de la Grèce. 6 vols. 2nd ed. Paris, 1826–1827. Recueil des Historiens des Croisades, Historiens occidentaux. 5 vols. Paris, 1844–95. Recueil des Historiens des Croisades, Historiens orientaux. 5 vols. Paris, 1872–1906. Recueil des Historiens des Croisades, Lois. 2 vols. Paris, 1841–43. Reis, Piri. Kitab­i bahriye. Edited by E. Z. Okte. Istanbul, 1988. Robertson, Colin Henderson and Eric Gardner Turner, eds. Catalogue of the Greek and Latin Papyri in the John Rylands Library Manchester, Vol. 4, Documents of the Ptolemaic, Roman and Byzantine Periods. Manchester, 1952. list of works cited 213 Röhricht, R. Regesta regni Hierosolymitani, (MXCVII–MCCXCI) [1097–1291]. Innsbruck, 1893. Sanudo, Marin. Istoria di Romania [Historia tēs Rōmanias]. Trans. E. Papadopoulou. Athens, 2000. Stinespring, William Franklin. “The Description of Antioch in Codex Vaticanus Arabicus 286.” PhD diss., Yale University, 1932. Strabo. Geography. Edited and trans. by Horace Leonard Jones. Cambridge, MA, 1983. Strehlke, E. Tabulae Ordinis Theutonici. Berlin, 1869. Tait, John Gavin, and Claire Préaux, eds. Greek Ostraca in the Bodleian Library at Oxford and Various Other Collections. 3 vols. London, 1930. Thietmar. “Le Pèlerinage de Maître Thietmar.” Translated and Annotated by Christiane Deluz. In Croisades et pélerinages: récits, chroniques et voyages en Terre Sainte, XIIe– XVIe siècle, ed. D. Régnier-Bohler, 928–958. Paris, 1997. Verhoogt, Arthur. Regaling Officials in Ptolemaic Egypt: A Dramatic Reading of Official Accounts from the Menches Papers. Papyroloica Lugduno-Batava 32. Leiden, 2005. Viereck, Paul, ed. Ostraka aus Brüssel und Berlin. Berlin, 1922. Wilcken, Ulrich. Grundzüge und Chrestomathie der Papyruskunde. Erster Band: Historischer Teil Zweite Hälfte: Chrestomathie. Leipzig, 1912. William of Tyre. Chronicon. Corpus Christianorum (Continuatio Mediaevalis 63–63 A). Instrumenta Lexicologica Latina Series A, Fasciculus 32, ed. R. C. B. Huygens. Turnhoult, 1986. ——. Historia Rerum in Partibus Transmarinis Gestarum. In Recueil des Historiens des Croisades, Historiens Occidentaux, Vol. 1. Paris, 1844. Secondary Sources [Atti:] 26. Convegno internazionale della ceramica. I bacini murati medievali: problemi e stato della ricerca, Albisola, 28–30 maggio 1993. Albisola (Savona): Centro ligure per la storia della ceramica, 1996. Abel, R. P. Géographie de la Palestine. 3rd ed. 2 vols. Paris, 1967. Abulafia, D. “The Pisan Bacini in the Medieval Mediterranean Economy: an Historian’s Point of View.” In Papers in Italian Archaeology 4, eds. Caroline Malone and Simon Stoddart, 278–302. BAR International Series 246. Oxford, 1987. Adam, A. “La maison et le village dans quelques tribus de l’Anti-Atlas.” Hespéris 37 (1950): 289–362. Adams, William Y. “The Archaeologist as Detective.” In Variation in Anthropology: Essays in Honor of John C. McGregor, eds. D. W. Lathrop and J. Douglas, 17–30. Urbana, 1973. Ahrweiler, Hélène. Byzance et la mer: la marine de guerre, la politique et les institutions maritimes de Byzance aux VIIe–XVe siècles. Bibliothèque byzantine 5. Paris, 1966. Alston, Richard. “Trade and City in Roman Egypt.” In Trade, Traders, and the Ancient City, eds. Helen Parkins and Christopher Smith, 168–202. London, 1998. Alston, R. Soldier and Society in Roman Egypt: A Social History. London, 1996. ——. The City in Roman and Byzantine Egypt. London, 2002. Amigues F., J. De Meulemeester, and A. Matthys. “Archéologie d’un grenier collectif fortifié hispano-musulman: le Cabezo de la Cobertera (Vallée de Rio Segura/ Murcie).” In Castrum 5. Archéologie des espaces agraires méditerranéens au Moyen Âge: Actes du colloque de Murcie (Espagne) tenu du 8 au 12 mai 1992, ed. André Bazzana, 347–359. Collection de la Casa de Velázquez 55—Collection de l’École française de Rouen 105. Madrid, 1992. Ammerman, A. J. “Venice before the Grand Canal.” Memoirs of the American Academy in Rome 48 (2003): 141–158. 214 list of works cited Ammerman, A. J. and C. E. McClennen, eds. Venice before San Marco. Recent Studies on the Origins of the City. Hamilton, N.Y., 2001. Amory, Patrick. “Names, Ethnic Identity, and Community in Fifth- and Sixth-Century Burgundy.” Viator 25 (1994): 1–30. Andersson, H. “The Era of Town Inventories. A kind of evaluation.” In Medieval Europe 1992—Pre-printed Papers, Vol. 1, Urbanism, 15–26. York, 1992. Anderton, M., ed. Anglo-Saxon Trading Centres: Beyond the Emporia. Glasgow, 1999. Andrén, Anders. Between Artifacts and Texts: Historical Archaeology in Global Perspective. Trans. Alan Crozier. London, 1998. Arnold, Christopher J. Roman Britain to Anglo-Saxon England: An Archaeological Study. London, 1984. Arnold, C. J. “Archaeology and History: The Shades of Confrontation and Cooperation.” In Archaeology at the Interface: Studies in Archaeology’s Relationships with History, Geography, Biology, and Physical Science, eds. John L. Bintliff and Christopher F. Gaffney, 32–39. British Archaeological Reports International Series 300. Oxford, 1986. Arnold, Werner. Die arabischen Dialekte antiochiens. Wiesbaden, 1998. Aurenche, Olivier. “Les conditions de l’enquête ethnoarchéologique.” In EthnoArchéologie méditerranéenne: finalités, démarches et résultats: table ronde, eds. A. Bazzana and M.-Ch. Delaigue, 13–16. Collection de la Casa de Velázquez 54. Madrid, 1995. Austin, David. “The ‘Proper Study’ of Medieval Archaeology.” In From the Baltic to the Black Sea: Studies in Medieval Archaeology, eds. David Austin and Leslie Alcock, 9–42. London, 1990. Austin, David and Leslie Alcock. From the Baltic to the Black Sea: Studies in Medieval Archaeology. One World Archaeology 18. London, 1990. Avramea, Anna. “Le Magne byzantin: problèmes d’histoire et de topographie.” In Eupsychia: Mélanges offerts à Hélène Ahrweiler, eds. Hélène Ahrweiler and Michel Balard, 49–62. Byzantina Sorbonensia 16. Paris, 1998. Ayoub, A. and J. L. Le Quellec. “Gasr al-Hag. Un grenier fortifié dans la Djeffara libyenne.” In Les techniques de conservation des grains à long terme: leur rôle dans la dynamique des systèmes de cultures et des sociétés, Vol. 2, eds. M. Gast and F. Sigaut, 3–18. Paris, 1981. Bagatti, B. Ancient Christian Villages of Galilee. Jerusalem, 2001. Bagnall, Roger S. Reading Papyri, Writing Ancient History. London, 1995. Ballardini, G. “Le ceramiche architettoniche di Roma e del Lazio (a proposito di un volume di Mons. Alberto Serafini).” Faenza 16 (1928): 55–65. ——. “Lo stile arcaico: Italia Meridionale—il X Corso di storia e tecnica delle ceramiche,” Faenza 25 (1937): 96–97. Balty, J.-C. “Apamée et la Syrie du nord aux époques hellénistique et romaine.” Revue du Monde Musulman et de la Méditerranée 62 (1994): 15–26. Balzaretti, R. “Cities, Emporia and the Monasteries: Local Economies in the Po Valley, c. AD 700–875.” In Towns in Transition: Urban Evolution in Late Antiquity and the Early Middle Ages, eds. N. Christie and S. T. Loseby, 213–234. Aldershot, Hants, England, 1996. Barceló, M. “El disenõ de espacios irrigades en al-Andalus.” In El agua en zonas áridas, arqueología e historia: actas del I Coloquio de Historia y Medio Físico, Almería, 14–15–16 de diciembre de 1989, 8–50. Almeria, ES, 1989. Barker, Graeme. A Mediterranean Valley: Landscape Archaeology and Annales History in the Biferno Valley. London, 1995. Barrett, John C. “Aspects of the Iron Age in Atlantic Scotland: A Case Study in the Problems of Interpetation.” Proceedings of the Society of Antiquaries of Scotland 111 (1981): 205–219. list of works cited 215 Bartlett, Robert. The Making of Europe: Conquest, Colonization, and Cultural Change, 950–1350. Princeton, 1993. Basset, Henri and Évariste Lévi-Provençal. Chella: Une nécropole mérinide. Paris, 1923. Baudrel, D. and J.-P. Cazes. “Les villages ecclésiaux dans le bassin de l’Aude.” In L’environnement des églises et la topographie religieuse des campagnes médiévales: actes du IIIe Congrès international d’archéologie médiévale, Aix-en-Provence, 28–30 septembre 1989, eds. Michel Fixot and Elisabeth Zadora-Rio, 80–97. Documents d’archéologie française 46. Paris, 1994. Bauzou, T. “Les routes romaines de Syrie.” In Archéologie et Histoire de la Syrie, Vol. 2, La Syrie de l’époque achéménide à l’avènement de l’Islam, eds. J.-M. Dentzer and W. Orthmann, 205–221. Saarbrücken, 1989. Bazzana A. and M.-Ch. Delaigue, eds. Ethno-Archéologie méditerranéenne: finalités, démarches et résultats: table ronde. Collection de la Casa de Velázquez 54. Madrid, 1995. Bazzana, A. Maisons d’al­Andalus: habitat médiéval et structures du peuplement dans l’Espagne orientale. Collection de la Casa de Velázquez 37. Madrid, 1992. ——. “Maisons rurales du Shark Al-Andalus. Essai de typologie.” In La casa hispano­ musulmana: aportaciones de la arqueología. La maison hispano­musulmane: apports de l’ archéologie, eds. J. Bermudez Lopez and A. Bazzana, 247–267 and 370–372. Granada, 1990. Bazzana, A., P. Cressier, and P. Guichard. Les châteaux ruraux d’ Al­Andalus. Histoire et archéologie des Husun du sud­est de l’ Espagne. Publications de la Casa de Velázquez. Série Archéologie 11. Madrid, 1988. Bazzana, A. and P. Guichard. “Archéologie extensive dans la région valencienne (Espagne).” In Castrum 2. Structures de l’habitat et occupation du sol dans les pays méditerranéens: les méthodes et l’apport de l’archéologie extensive, ed. Ghislaine Noyé, 3–28. Collection de l’ École Française de Rome 105:2—Publications de la Casa de Velázquez. Série Archéologie 9. Rome, 1988. Bazzana, A., J. De Meulemeester, and A. Matthys. “Quelques aspects du peuplement médiéval du Valle de Ricote (Murcie/Espagne).” In Rural Settlements in Medieval Europe, Medieval Europe Brugge 1997 Conference, eds. Guy de Boe and F. Verhaeghe, 39–54. I.A.P. rapporten 6. Zellik, BE, 1997. Berglund, Björn. “Historical Archaeology – A Challenge for Archaeological Thought.” In Method and Theory in Historical Archaeology. Papers of the ‘Medieval Europe Brugge 1997’ Conference, eds. Guy de Boe and Frans Verhaeghe, 15–21. Instituut voor het Archaologisch Patrimonium 3. Zellik, BE, 1997. Bermudez Lopez, J. and A. Bazzana, eds. La casa hispano­musulmana: aportaciones de la arqueología. La maison hispano­musulmane: apports de l’ archéologie. Granada, 1990. Berti, Fausto and Gianna Pasquinelli. Antiche maioliche di Montelupo, secoli XIV– XVIII: Pontedera, Manifattura Pasquinuccio Pasquinucci, 21 ottobre–4 novembre 1984: Pisa, Palazzo Lanfranchi, 8–22 novembre. Pontedera, IT, 1984. Berti, Graziella. “Ceramiche Islamiche del Mediterraneo occidentale usate come ‘bacini’ in Toscana, in Sardegna e in Corsica (secoli XI–XIII).” In L’eta di Federico II nella Sicilia Centro­meridionale: citta, monumenti, reperti: atti delle Giornate di studio: Gela 8–9 Dicembre 1990, ed. S. Scuto, 99–114, 266–76, 329–33. Palermo, 1991. ——. “I bacini Islamici del Museo Nazionale di San Matteo, Pisa: vent’anni dopo la pubblicazione del Corpus.” In Studi in Onore di Umberto Scerrato: per il suo settan­ tacinquesimo compleanno, eds. U. Scerrato, M. V. Fontana, and B. Genito, 121–151. Naples, 2003. ——. “Pisa: Ceramiche e commerci (2 meta’ X- meta’ XIV s.).” In Il Congresso nazi­ onale di archeologia medievale: auditorium del Centro studi della Cassa di risparmio di Pisa (ex Benedettine): Pisa, 29–31 maggio 1997, ed. S. Gelichi, 346–51. Florence, 1997. 216 list of works cited ——. “La ceramica tunisina ‘a cobalto e manganese’ in Toscana.” In [Atti:] 35. Convegno internazionale della ceramica, 2002: Ceramica in blu: diffusione e utilizzazione del blu nella ceramica: Savona, 31 maggio-1 giugno, 89–102. Albisola (Savona), 2002. Berti, G. and L. Cappelli. Lucca, ceramiche medievali e post-medievali: Museo nazionale di Villa Guinigi. Ricerche di archeologia altomedievale e medievale 19–20. Florence, 1994. Berti, G. and S. Gelichi. “Mille chemins ouverts en Italie.” In Le Vert et le Brun: de Kairouan a Avignon, ceramiques du Xe au XVe siecle, eds. Chapelle de la Vieille Charite (Marseille, France), Musees de Marseille, and Reunion des musees nationaux (France), 128–163. Marseille, 1995. ——. “Trasmissioni di tecnologie nel Medioevo: tendenze e linee di ricerca attuali.” In [Atti:] 32. Convegno internazionale della ceramica, 1999: Circolazione di tecnologie, maestranze e materie prime nelle produzioni ceramiche del mediterraneo dal medioevo all’eta moderna. 33. Convegno internazionale della ceramica, 2000: la ceramica come indicatore socio-economico, 23–41. Albisola (Savona), 2001. Berti, G. and T. Mannoni. “Ceramiche medievali del Mediterraneo occidentale: considerazioni su alcune caratteristiche tecniche.” In A cerâmica medieval no Mediterrâneo ocidental, Lisboa, 16–22 Novembro 1987, 163–173. Mértola, PT: Campo Arqueológico de Mértola, 1991. Berti, G., Catia Renzi Rizzo, and Marco Tangheroni. Il mare, la terra, il ferro: ricerche su Pisa medievale (secoli VII–XIII). Collana Percorsi 12. Ospadelatto (Pisa), 2004. Berti, G. and L. Tongiorgi. I bacini ceramici del Duomo di S. Miniato: (ultimo quarto del XII secolo). Genoa, 1981. ——. I bacini ceramici medievali delle chiese di Pisa. Rome, 1981. Berti, G., M. Hobart, and F. Porcella. “ ‘Protomaioliche’ in Sardegna.” In [Atti:] 23. Convegno internazionale della ceramica “La Protomaiolica e la Maiolica Arcaica dalle Origini al Trecento”: Albisola, 25–27 maggio, 1990, 153–167. Albisola (Savona), 1993. Beugnot, A. A. See Assises de Jerusalem. Bevan, Edwyn Robert. The House of Ptolemy. A History of Egypt under the Ptolemaic Dynasty. Chicago, 1958. Bianchi, Giovanna. “Trasmissione dei saperi tecnici ed analisi dei procedimenti costruttivi.” Archeologia dell’Architettura 1 (1996): 53–64. Biddle, M., D. Hudson, and C. M. Heighway. The Future of London’s Past. Worcester, 1973. Biers, William Richard. Art, Artefacts, and Chronology in Classical Archaeology. London, 1992. Bingen, Jean with Marianne Lewuillon-Blume and Jean Quaegebeur, eds. Au temps où on lisait le Grec en Égypte. Brussels, 1977. Birgitte-Porëe, P. “Les Moulins et Fabriques a Sucre de Palestine et de Chypre: Histoire, Geographie et Technologie d’une Production Croisee et Medievale.” In Cyprus and the Crusades: Papers Given at the International Conference ‘Cyprus and the Crusades’, Nicosia, 6–9 September, 1994, eds. N. Coureas and J. Riley-Smith, 377–510. Nicosia: Society for the Study of the Crusades and the Latin East and the Cyprus Research Center, 1995. Blackman, David J. “Ancient Harbors in the Mediterranean Part 2.” International Journal of Nautical Archaeology and Underwater Exploration 11.2 (1982): 185–211. Blackman, D. J. “Bollards and Men.” In Mediterranean Cities: Historical Perspectives, eds. Irad Malkin and Robert L. Hohlfelder, 7–20. Totowa, 1988. Blake, Hugo. “La ceramica medieval e spagnola e la Liguria.” In [Atti:] 4 Convegno internazionale della ceramica, Albisola, 28 maggio–3 giugno 1971 published together in one volume with [Atti:] 5 Convegno internazionale della ceramica, Albisola, 31 maggio–4 giugno 1972, 55–105. Albisola (Savona), 1972. list of works cited 217 Blake, H. “The Ceramic Hoard from Pula (prov. Cagliari) and the Pula Type of Spanish lusterware.” In Segundo Coloquio Internacional de Cerámica Medieval en el Mediterráneo Occidental [Toledo, 1981], ed. Juan Zozaya, 365–407. Madrid, 1986. ——. “The Medieval Incised Slipped Pottery of North West Italy.” In La ceramica medievale nel Mediterraneo occidentale: Siena 8–12 ottobre 1984, Faenza 13 ottobre 1984, International Medieval Archeology Colloquium, 317–352. Florence, 1986. Boas, Adrian J. Crusader archaeology: The Material Culture of the Latin East. London, 1999. Bon, Antoine. La Morée franque: recherches historiques, topographiques et archéologiques sur la Principauté d’Achaïe (1205–1430). Paris, 1969. ——. Le Peloponnese byzantin jusqu’en 1204. Paris, 1951. Boni, G. “Il muro di fondazione del campanile di S. Marco a Venezia.” Archivio Veneto 29.2 (1885): 354–368. Bonnassie, P. La Catalogne du milieu du Xe siècle à la fin du XIe siècle. Toulouse, 1975. ——. “Les sagreres catalanes: la concentration de l’habitat dans le ‘cercle de paix’ des églises (XIe siècle).” In L’environnement des églises et la topographie religieuse des campagnes médiévales: actes du IIIe Congrès international d’archéologie médiévale, Aix­en­Provence, 28–30 septembre 1989, eds. Michel Fixot and Elisabeth ZadoraRio, 68–79. Documents d’archéologie française 46. Paris, 1994. Bortoletto, M. “Murano, Mazzorbo e Torcello: tre siti a confronto. Indagini archeologiche nella laguna nord di Venezia.” Archeologia delle Acque 1.1 (1999): 55–74. ——. “Cenni sull’edilizia minore veneziana alla luce di alcuni rinvenimenti archeologici.” Archeologia delle Acque 2.1 (2000): 9–20. ——. “De canalibus, rivis, piscinisque: primi passi verso un’archeologia idronomastica veneziana,” 136–152. [Part of ] Fozzati L. et alii. “Archeologia delle fondamenta veneziane.” In Tra due elementi sospesa: Venezia, costruzione di un paesaggio urbano, ed. L. Anglani, 134–166. Venice, 2000. Boube, Jean. “La circulation monetaire à Sala à l’époque préromaine.” In Lixus. Actes du colloque organisé par l’Institut des sciences de l’archéologie et du patrimonie de Rabat avec le concours de l’école française de Rome, Larache, 8–11 novembre 1989, 255–265. Collection de l’École français de Rome 166. Rome, 1992. Bouchet, René. Chronique de Morée: introduction, traduction et notes. Roue à livres 46. Paris, 2005. Bowman, Alan K. “Post-Roman Imports in Britain and Ireland: A Maritime Perspective.” In External Contacts and the Economy of Late Roman and Post­Roman Britain, ed. Kenneth Dark, 97–108. Woodbridge, 1996. Bradford, Ernle. Mediterranean: Portrait of a Sea. London, 1971. Bresc-Bautier, G. Le Cartulaire du chapitre du Saint­Sépulcre de Jérusalem. Documents relatifs à l’histoire des Croisades 15. Paris, 1984. Breuillot, Martine. Châteaux oubliés de la Messénie médiévale. Paris, 2005. Broeze, Frank. “Port Cities: The Search for an Identity.” Journal of Urban History 11 (1985): 209–225. Brogiolo, G. P. “Attori, autori e ‘fruitori’ del progetto archeologic.” In Archeologia e urbanistica: International school in archaeology, Certosa di Pontignano, Siena 26 gennaio–1 febbraio 2001, ed. A. Ricci, 305–318. Quaderni del Dipartimento di archeologia e storia delle arti, Sezione archeologica, Università di Siena 53–54. Firenze, 2002. Brogiolo, G. P. and S. Gelichi. La città nell’alto medioevo italiano: archeologia e storia. Rome, 1998. ——. Nuove ricerche sui castelli altomedievali in Italia settentrionale. Florence, 1996. Brogiolo, G. P., ed. Archeologia urbana in Lombardia. Valutazione dei depositi archeo­ logici e inventario dei vincoli. Modena, 1984. Brunot, Louis. La mer dans les traditions et les industries indigènes à Rabat & Salé. École Superieure de langue Arabe et de dialectes Berbères de Rabat 5. Paris, 1920. 218 list of works cited Bülow-Jacobsen, Adam. “Traffic on the Road between Coptos and the Red Sea.” In Life on the Fringe: Living in the Southern Egyptian Deserts During the Roman and Early Byzantine Periods, ed. Olaf E. Kaper, 63–74. (Leiden, 1998). Burnouf, Joëlle. Archéologie médiévale en France: le second moyen âge, XIIe–XVIe siècle. Paris, 2008. Busch, Silvia Orvietani. Medieval Mediterranean: The Catalan and Tuscan Coasts, 1100–1235. The Medieval Mediterranean 32. Leiden, 2001. Calaon, D. “Cittanova (VE): analisi GIS.” In IV Congresso nazionale di archeologia medievale: Scriptorium dell’Abbazia, Abbazia di San Galgano (Chiusdino, Siena), 26–30 settembre 2006, eds. Riccardo Francovich and Marco Valenti, 216–24. Borgo San Lorenzo (FI) [Florence, Italy], 2006. Cameirana, A. and G. Bozzano. “La terraglia nera ad Albissola all’inizio dell’ottocento.” In [Atti:] 3 Convegno internazionale della ceramica, Albisola, 31 maggio–2 giugno 1970: La terraglia nera ad Albisola all’inizio dell’800, 61–114. Albisola (Savona), 1970. Caniggia, G. Lettura di una città: Como. Rome, 1963. Caniggia, G. and G. L. Maffei. Composizione architettonica e tipologia edilizia. Architettura e urbanistica 34. Venice, 1979. Canivet, P. La Syrie de Byzance à l’Islam: VIIe–VIIIe siècles: actes du colloque international, Lyon, Maison de l’Orient Méditerranéen, Paris, Institut du Monde Arabe; 11–15 septembre 1990. Paris, 1992. Capot-Rey, R. “Greniers domestiques et greniers fortifiés au Sahara. Le cas du Gourara.” Travaux de l’Institut de Recherches Sahariennes 14 (1956): 139–159. Carver, M. “On Archaeological Value.” Antiquity 70.67 (1996): 45–56. ——. Archaeological Value and Evaluation. Mantua, 2003. Casula, A. “La chiesa di San Giorgio di Oliastreto in agro di Usini.” In La civiltà giudicale in Sardegna nei secoli XI–XIII: fonti e documenti scritti. Atti del convegno nazionale, Sassari, Aula Magna dell’Università, 16–17 marzo 2001; Usini, chiesa di Santa Croce, 18 marzo 2001, 519–536. Sassari, 2002. Cataldi, G., P. Iacono, and A. Merlo. “La geometria di Firenze e il progetto matrice della citta e del territorio.” Firenze Architettura 1 (2000): 4–17. Catteddu, Isabelle. Archéologie médiévale en France: le premier moyen âge, Ve–XIe siècle. Paris, 2009. Chalmeda, P. El ”señor del zoco” en España: edades media y moderna; contribución al estudio de la histroia del mercado. Madrid, 1973. Chouquer, G. Structures agraires en Italie centro-méridionale: cadastres et paysage ruraux. Collection de l’Ecole française de Rome 100. Rome, 1987. Christathakis, P. To monastiri tis Kourtzenis sto Lada Messenias. Kalamata, 1989. Collingwood, R. G. The Idea of History. Oxford, 1994. Constable, Giles. The Reformation of the Twelth Century. Cambridge, 1996. Contamine, P. La guerre au Moyen Âge. 3rd ed. Nouvelle Clio 24. Paris, 1992. Coroneo, Roberto. Architettura romanica dalla meta del mille al primo ‘300. Nuoro, 1993. Cowdrey, H. E. J. The Age of Abbot Desiderius: Montecassino, the Papacy, and the Normans in the Eleventh and Early Twelfth Centuries. Oxford, 1983. Curta, Florin. “Some Remarks on Ethnicity in Medieval Archaeology.” Early Medieval Europe 15.2 (2007): 159–185. Curtis, Robert I. Garum and Salsamenta: Production and Commerce in Materia Medica. Studies in Ancient Medicine 3. Leiden, 1991. Cutler, A. “Tiles and Tribulations: A Community of Clay Across Byzantium and its Adversaries.” In A Lost Art Rediscovered. The Architectural Ceramics of Byzantium, eds. Sharon E. J. Gerstel and Julie A. Lauffenburger, 159–169. Baltimore, 2001. D’Angelo, Franco. “Le ceramiche rinvenute nel convento di San Francesco d’Assisi a Palermo e il loro significato.” Sicilia Archeologica 2 (1974): 65–73. list of works cited 219 ——. “Le ceramiche spagnole ‘tipo Pula’ delle chiese dello steri di Palermo.” In Il servizio da tavola in ceramica: atti [del] XV Convegno internazionale della ceramica, Albisola, 27–31 maggio 1982, 77–84. Albisola, 1985. Daoulatli, A. “La production vert et brun en Tunisie du IX au XII siecle.” In Le Vert et le Brun: de Kairouan à Avignon, céramiques du Xe au XVe siècle, eds. Chapelle de la Vieille Charité (Marseille, France), Musées de Marseille, and Réunion des musées nationaux (France), 69–89. Marseille, 1995. David, J. C. and M. al–Dbiyat. “La ville en Syrie et ses territoires: héritages et mutations.” Bulletin d’études orientales 52 (2000): 17–27. Davies, John Kenyon. “Ancient economies: Models and Muddles.” In Trade, Traders, and the Ancient City, eds. Helen Parkins and Christopher Smith, 225–256. London, 1998. De Meulemeester, J. “Même problème, même solution: quelques réflexions autour d’un grenier fortifié.” In Le village médiéval et son environnement. Études offertes à Jean-Marie Pesez, eds. L. Feller, P. Mana, F. Piponnier, and Jean-Marie Pesez, 97–112. Histoire ancienne et médiévale 48. Paris, 1998. ——. “Le Valle de Ricote et le développement des recherches de la Région wallonne à l’étranger.” In Mélanges d’Archéologie médiévale. Liber amicorum en hommage d’André Matthys, eds. J. De Meulemeester and P. Gillet, 46–55. Les Cahiers de l’Urbanisme 2006. Liège, 2006. De Min, M. “Edilizia altomedievale e medioevale nel territorio lagunare. Nuovi dati conoscitivi dai cantieri di restauro.” In Tra due elementi sospesa: Venezia, costruzione di un paesaggio urbano, ed. L. Anglani, 98–133. Venice, 2000. ——. “Venezia e il territorio lagunare.” In Ritrovare restaurando. Rinvenimenti e scoperte a Venezia e in laguna, ed. Soprintendenza per i beni ambientali e architettonici di Venezia, 15–25. Cornuda, 2000. ——. “Venezia. Chiesa di San Lorenzo di Castello: un esempio di scavo correlato al restauro architettonico.” In Ritrovare restaurando. Rinvenimenti e scoperte a Venezia e in laguna, ed. Soprintendenza per i beni ambientali e architettonici di Venezia, 40–47. Cornuda, 2000. Dédéyan, G. “Un Emir armenien du Hawran entre la principaute turque de Damas et le royaume latin de Jerusalem (1147).” In Dei Gesta per Francos, eds. M. Balard, B. Z. Kedar, and J. Riley-Smith, 179–185. Burlington, 2001. Delaigue, M.-C. Capileira, village andalou. Un habitat montagnard à toits plats. BAR International Series 466. Oxford, 1988. ——. “Deux exemples d’habitat rural en Andalousie orientale: approche ethnoarchéologique.” In La casa hispano-musulmana: aportaciones de la arqueología. La maison hispano-musulmane: apports de l’ archéologie, eds. J. Bermudez Lopez and A. Bazzana, 21–45. Granada, 1990. ——. “Ethnoarchéologie et histoire: l’habitat rural de la péninsule Ibérique comme indicateur des modifications de peuplement.” In Ethnoarchéologie: justification, problèmes, limites, 391–407. Juan-les-Pins, FR, 1992. Delatte, Armand. Les portulans grecs. Bibliothèque de la Faculté de philosophie et lettres de l’Université de Liège 107. Liège, 1947. Delbrück, H. Medieval Warfare. Lincoln, NE, 1990. Delogu, R. and A. Mezzetti. L’architettura del medioevo in Sardegna. Rome, 1953. Deschamps, P. “Une Grotte forteresse des Croisés au-delà du Jourdain.” Journal Asiatique 227 (1935): 285–299. ——. “La Cave de Tyron, une grotte forteresse des Croises au Liban.” In Melanges Syriens offerts a M. Rene Dussaud, secretaire perpetuel de l’Academie des inscriptions et belles­lettres, par ses amis et ses eleves. 2 vols. Ed. Academie des inscriptions & belles-lettres (France), Vol. 2, 873–882. Bibliotheque archeologique et historique 30. Paris, 1939. 220 list of works cited ——. La Défense du royaume de Jèrusalem, étude historique, géographique et monumentale. 2 vols. Bibliothèque archéologique et historique 34. Les Châteaux des Croisés en Terre Sainte 2. Paris, 1939. ——. “Etude sur un texte enumerant les possessions musulmanes dans le royaume de Jerusalem vers l’annee 1239.” Syria 23.1–2 (1942–1943): 86–104. Despois, J. “Les greniers fortifiés de l’Afrique du Nord.” Les Cahiers de Tunisie 1.1 (1953): 38–60. Devais, C. “L’Expression du pouvoir aux frontières du royaume de Jérusalem: Terre de Suète et Oultre-Jourdain au XIIème siècle.” Bulletin d’Etudes Orientales 57 (2008): 19–30. Dimitrakos-Mesisklis, Dimitrios [Dēmētrakos-Mesisklēs, Dēmētrios V.]. Oi Nyklianoi, Vol. 1. Athens, 1949. Djobadze, Wachtang. Archaeological Investigations in the Region West of Antioch onthe-Orontes. Stuttgart, 1986. Dodinet, M., J. Leblanc, J.-P. Vallat, and F. Villeneuve. “Le paysage antique en Syrie: l’exemple de Damas.” Syria 67 (1990): 339–355. Dorigo, W. Venezia Origini: fondamenti, ipotesi, metodi. Milan, 1983. ——. L’edilizia abitativa nella “Civitas Rivoalti” e nella “Civitas Veneciarum” (XI e XIII secolo). Prolusioni [Venice], 1993. ——. Venezia romanica: la formazione della città medioevale fino all’età gotica. Venice, 2003. Downey, Glanville. A History of Antioch in Syria from Seleucus to the Arab Conquest. Princeton, 1961. ——. Ancient Antioch. Princeton, 1963. Drandakis, Nikolaos. “Paratiriseis stis toichografies tou 13ou aiona pou sozontai sti Mani.” In The 17th International Byzantine Congress: Major papers, Dumbarton Oaks/Georgetown University, Washington, D.C., August 3–8, 1986, 683–712. New Rochelle, NY, 1986. Drexhage, Hans-Joachim. “Garum und Garumhandel im römischen und spätantiken Ägypten.” Münstersche Beiträge zur antiken Handelsgeschichte 12.1 (1993): 27–55. ——. “Der Handel, die Produktion und der Verkehr von Kase nach den griechischen Papyri und Ostraka.” Munstersche Beitrage zur antiken Handelsgeschichte 15.2 (1996): 33–41. ——. “Ein Monat in Antiochia. Lebenshaltungskosten und Ernahrungsverhalten des Theophanes im Payni (26. Mai–24. Juni) ca. 318 n. Chr.” Munstersche Beitrage zur antiken Handelsgeschichte 17.1 (1998): 1–10. Dueck, Daniela. Strabo of Amasia: A Greek Man of Letters in Augustan Rome. London, 2000. Dussaud, R. Topographie de la Syrie antique et médiévale. Paris, 1927. Eideneier, H., ed. Spanos: eine byzantinische Satire in der Form einer Parodie. Berlin, 1977. El Majali, Rafat and Abdul Rahim Mas’ad. “Trade and Trade Routes in Jordan in the Mameluk Era (AD 1250–1516).” Studies in the History and Archaeology of Jordan 3 (1987): 311–316. Elayi, Josette. “Al-Mina sur l’Oronte à l’époque perse.” In Studia Phoenicia V: Phoe­ nicia and the East Mediterranean in the First Millennium B.C. Proceedings of the Conference held in Leuven from the 14th to the 16th of November 1985, ed. Edward Lipiński, 249–266. Orientalia Lovaniensia Analecta 22. Leuven, 1987. Elisséeff, N. Nūr ad­Dīn, un grand prince musulman de Syrie au temps des Croisades (511–569 h./1118–1174). Damascus, 1967. Ellenblum, R. “Construction Method in the Frankish Rural Settlement.” In The Horns of Hattīn, ed. B. Z. Kedar, 168–189. London, 1992. ——. Frankish Rural Settlement in the Latin Kingdom of Jerusalem. Cambridge, 1996. ——. “Three generations of Frankish castle-building in the Latin Kingdom of Jerusalem.” In Autour de la Première Croisade: Actes du Colloque de la Society for list of works cited 221 the Study of the Crusades and the Latin East: Clermont-Ferrand, 22–25 juin 1995, ed. M. Balard, 517–551. Paris, 1996. ——. “Frontier Activities: the Transformation of a Muslim Sacred Site into the Castle of Vadum Iacob.” Crusades 2 (2003): 83–97. Esposito, Mario, ed. [Symon Simeonis, and Mario Esposito]. Itinerarium Symonis Semeonis ab Hybernia ad Terram Sanctam. Scriptores Latini Hiberniae 4. Dublin, 1960. Fehérvári, Geza. “Near Eastern Wares Under Chinese Influence.” In Pottery and Metalwork in T’ang China: Their Chronology and External Relations, ed. William Watson, 27–33. Oxford, 1976. Fehring, Günter P. The Archaeology of Medieval Germany: An Introduction. Studies in Archaeology. London, 1991. Ferchiou, S. “Conserves céréaliers et rôle de la femme dans l’économie familiale en Tunisie.” In Les techniques de conservation des grains à long terme: leur rôle dans la dynamique des systèmes de cultures et des sociétés, eds. Marceau Gast, François Sigaut, and Corinne Beutler, Vol. 1, 190–197. Paris, 1979. Fersuoch, L. et alii. “Indagini archeologiche a San Lorenzo di Ammiana (Venezia).” Archeologia Veneta 12 (1989): 71–96. Finley, Moses. The Use and Abuse of History. London, 1975. Fiorilla, S. Gela: le ceramiche medievali dai pozzi di Piazza S. Giacomo. Monumenti di Sicilia. Messina, 1996. Fixot, M. and E. Zadora-Rio. L’ église, le terroir. Monographie du CRA 1. Paris, 1989. Floris, F. Storia della Sardegna. Rome, 1999. Fontana, Maria Vittoria and Giovanna Ventrone Vassallo. La ceramica medievale di San Lorenzo Maggiore in Napoli: atti del Convegno La Ceramica medievale di San Lorenzo Maggiore in Napoli, nel quadro della produzione dell’Italia centro-meridionale e i suoi rapporti con la ceramica islamica. Napoli, Basilica di San Lorenzo Maggiore, 25–27 giugno, 1980. 2 vols. Naples, 1984. Fourikis, P. “Paratirisis eis ta toponymia ton Chronikon tou Moreos.” Athina [Athēna: syngramma periodikon tēs en Athēnais Epistēmonikēs Hetaireias] 40 (1928): 26–59. Fozzati L. et alii. “Archeologia delle fondamenta veneziane.” In Tra due elementi sospesa: Venezia, costruzione di un paesaggio urbano, ed. L. Anglani, 134–166. Venice, 2000. Francovich, R. and R. Hodges. Villa to Village: The Transformation of the Roman Countryside in Italy, c. 400–1000. London, 2003. Frézouls, E. “Observation sur l’urbanisme dans l’orient syrien.” Annales Archéologiques Arabes Syriennes. Revue d’Archéologie et d’Histoire 21 (1971): 231–243. Frulio, Gabriela. “Tecniche costruttive della Sardegna medievale: il monumento come fonte per la conoscenza.” in La civiltà giudicale in Sardegna nei secoli XI–XIII: fonti e documenti scritti. Atti del convegno nazionale, Sassari, Aula Magna dell’Università, 16–17 marzo 2001; Usini, chiesa di Santa Croce, 18 marzo 2001, 485–496. Sassari, 2002. Fuks, Alexander. “Notes on the Archive of Nicanor.” Journal of Juristic Papyrology 5 (1951): 207–216. Gagos, Trainos, Jennifer E. Gates, and Andrew Wilburn. “Material Culture and Texts of Graeco-Roman Egypt: Creating Context, Debating Meaning.” Bulletin of the American Society for Papyrology 42 (2006): 169–188. Galoppini, Laura. Sardegna e Mediterraneo: dai Vandali agli Aragonesi. Antologia di fonti scritte. Materiali didattici ETS 44. Pisa, 1993. Gangler, Anette, Matthias Bückle, and Heinz Gaube. Ein traditionelles Wohnviertel im Nordosten der Altstadt von Aleppo. Tübingen, 1993. Gasparri, S. “Venezia fra l’Italia bizantina e il regno italico: la civitas e l’assemblea.” In Venezia: itinerari per la storia della città, eds. S. Gasparri, G. Levi, and P. Moro, 61–82. Bologna, 1997. 222 list of works cited Gaube, H. “Aleppo zwischen Alexander dem Großen und der Arabischen Eroberung.” In Damaskus-Aleppo: 5000 Jahre Stadtentwicklung in Syrien, ed. M. Fansa [Beate Bollmann], 101–107. Archäologische Mitteilungen aus Nordwestdeutschland 28. Mainz am Rhein, 2000. Gaube, H. and E. Wirth. Aleppo: historische und geographische Beiträge zur baulichen Gestaltung, zur sozialen Organisation und zur wirtschaftlichen Dynamik einer vorderasiatischen Fernhandelsmetropole. 2 vols. Beihefte zum Tübinger Atlas des Vorderen Orients 58. Wiesbaden, 1984. Gelichi, S. “La ceramica ingubbiata medievale nell’Italia nord-orientale.” In La ceramica medievale nel Mediterraneo occidentale: Siena 8–12 ottobre 1984, Faenza 13 ottobre 1984, International Medieval Archeology Colloquium, 353–407. Florence, 1986. ——. Introduzione all’archeologia medievale: storia e ricerca in Italia. Rome, 1997. ——. “Città pluristratificate: la consistenza e la conservazione dei bacini archeologici.” In Archeologia e urbanistica: International school in archaeology, Certosa di Pontignano, Siena 26 gennaio–1 febbraio 2001, ed. A. Ricci, 61–76. Quaderni del Dipartimento di archeologia e storia delle arti, Sezione archeologica, Università di Siena 53–54. Florence, 2002. ——. “The Cities.” In Italy in the Early Middle Ages: 476–1000, ed. Cristina La Rocca, 168–174. Oxford, 2002. ——. “Una discussione con Chris Wickham.” Storica 34 (2006): 134–147. ——, ed. Comacchio e il suo territorio tra la tarda antichita e l’altro medioevo. Ferrara, 2007. ——. “Venezia tra archeologia e storia: la costruzione di un’identita urbana.” In Le citta italiane tra la tarda antichita e l’alto Medioevo: atti del Convegno (Ravenna, 26–28 febbraio 2004), ed. A. Augenti, 151–186. Biblioteca di archeologia medievale 20. Florence, 2006. ——. “Flourishing places in North-Eastern Italy: towns and emporia between late antiquity and the Carolingian age.” In Post-Roman towns, trade and settlement in Europe and Byzantium, Vol. 1, The Heirs of the Roman West, ed. J. Henning, 77–104. Millennium-Studien 5. Berlin, 2007. ——. “The eels of Venice. The long eighth century of the emporia of the northern region along the Adriatic Coast.” In 774: ipotesi su una transizione: atti del Seminario di Poggibonsi, 16–18 febbraio 2006, ed. S. Gasparri, 81–117. Seminari internazionali del Centro interuniversitario per la storia e l’archeologia dell’alto Medioevo 1. Turnhout, 2008. ——, ed. L’isola del vescovo. Gli scavi archeologici intorno alla Cattedrale di Comacchio. The Archaeological Excavations nearby the Comacchio Cathedral. Florence, 2009. ——. “La disparition e les reventes. In margine a Framing the Early Middle Ages e al lungo VIII secolo dell’economia italiana.” Storica: Forthcoming. —— et alii. “Isola di San Giacomo in Paludo (laguna nord, Venezia): gli scavi delle campagne del 2003 (SGP03a e SGP03b).” Quaderni di Archeologia del Veneto 20 (2004): 160–177. —— et alii. “. . . igne castrum combussit . . . Comacchio tra la Trada Antichità e l’Alto Medioevo.” Archeologia Medievale 32 (2006): 19–48. —— et alii. “Comacchio tra IV e X secolo: territorio, abitato e infrastrutture.” In IV Congresso nazionale di archeologia medievale: Scriptorium dell’Abbazia, Abbazia di San Galgano (Chiusdino, Siena), 26–30 settembre 2006, eds. Riccardo Francovich and Marco Valenti, 114–123. Borgo San Lorenzo (FI) [Florence, Italy], 2006. —— et alii. “Identity marks, organization of spaces and characteristics of consumption of an island of the Venetian lagoon between the Later Middle Ages and the Modern Age.” In Constructing Post-Medieval Archaeology in Italy: A New Agenda: Proceedings of the International Conference (Venice, 24th and 25th November 2006), eds. S. Gelichi and M. Librenti, 97–108. Borgo San Lorenzo (FI) [Florence, Italy], 2007. list of works cited 223 Gelichi, S., A. Alberti, and M. Librenti. Cesena, la memoria del passato: archeologia urbana e valutazione dei depositi. Biblioteca di archeologia medievale 16. Florence, 1999. Georgacas, D. and W. McDonald. “Place names of Southwest Peloponnesus.” Peloponnesiaka 6 [1967] (1963–68). Gerstel, Sharon E. J. and Julie A. Lauffenburger, eds. A Lost Art Rediscovered. The Architectural Ceramics of Byzantium. Baltimore, 2001. Gerstel, S. “Art and identity in the medieval Morea.” In The Crusades from the Perspective of Byzantium and the Muslim World, eds. Angeliki E. Laiou and Roy Parviz Mottahedeh, 263–285. Washington, D.C., 2001. Ginzburg, Carlo. The Cheese and the Worms: The Cosmos of a Sixteenth-Century Miller. Trans. John and Anne Tedeschi. Baltimore, 1982. Grainger, J. D. The cities of Seleukid, Syria. Oxford, 1990. Graham-Campbell, James and Magdalena Valor, eds. The Archaeology of Medieval Europe, Vol. 1, Eighth to Twelfth Centuries AD. Aarhus, 2007. Grandi, E. “Le ceramiche fini da mensa dalla laguna veneziana. I contesti di San Francesco del Deserto e Torcello.” In La circolazione delle ceramiche nell’Adriatico tra tarda antichità e altomedioevo: III Incontro di studio CER.AM.IS, eds. S. Gelichi and C. Negrelli, 127–153. Mantua, 2007. Green, Joseph A. “The Water Mills of the Ajlun-Kufranja Valley: the relationship of technology, society and settlement.” Studies in the History and Archaeology of Jordan 5 (1995): 757–65. Grenfell, Bernard Pyne and Arthur Surridge Hunt. “Englische Ausgrabungen im Faijûm.” Archiv für Papyrusforschung und verwandte Gebiete 1 (1901): 216–218. Guyer, S. “La madrasa al–Halâwiyya à Alep.” Bulletin de l’Institut Français d’Archéologie Orientale 11 (1914): 217–231. Habicht, Christian. Pausanias’ Guide to Ancient Greece. Berkeley, 1998. Hamilton, B. The leper king and his heirs: Baldwin IV and the Crusader Kingdom of Jerusalem. Cambridge, UK, 2000. Hartleb, Peter. Die messenische Mani: eine Studie zum Wandel in der Peripherie Griechenlands. Institut für Geographische Wissenschaften, Freie Universität, Abhandlung 45. Berlin, 1989. Hartmann, Ludo Moritz. Zur Wirtschaftsgeschichte Italiens im frühen Mittelalter: Analekten. Gotha, 1904. Hartmann, Martin. “Das Liwa Haleb (Aleppo) und ein Teil des Liwa Dschebel Bereket.” Zeitschrift der Gesellschaft für Erdkunde zu Berlin 29 (1894): 142–188. Hauben, Hans. “L’expedition de Ptolémée III en Orient et la sédition domestique de 245 av. J.-C.” Archiv für Papyrusforschung und verwandte Gebiete 36 (1990): 29–37. Herzfeld, Ernst. Inscriptions et monuments d‘Alep. [Matériaux pour un Corpus inscriptionum Arabicarum; Deuxième partie; Syrie du Nord]. 2 vols. Mémoires publiés par les membres de l’Institut français d’archéologie orientale du Caire 76. Cairo, 1954. Hill, D. and R. Cowie, eds. Wics: the early mediaeval trading centres of northern Europe. Sheffield archaeological monographs 14. Sheffield, 2001. Hinton, David Alban, ed. 25 Years of Medieval Archaeology. Sheffield, 1983. ——. “Metalwork and the emporia.” In Anglo-Saxon Trading Centres: Beyond the Emporia, ed. M. Anderton, 24–31. Glasgow, 1999. Hinz, H. “Wehrkirchen und Burgenbau.” In Château-Gaillard: études de castellologie médiévale, IX–X. Actes des colloques internationaux tenus à Basel (1978) et à Durham (1980), 117–144. Caen, 1982. Hobart, M. In Moriscos. Echi della presenza e della cultura islamica in Sardegna. Catalogo, (catalogue), M. F. Porcella and M. Serreli, 32–33 and 61–62. Pinacoteca Nazionale di Cagliari, 1993. 224 list of works cited Hodder, Ian. Theory and Practice in Archaeology. London, 1992. Hodges, Richard. Dark Age Economics: The Origins of Towns and Trade A.D. 600– 1000. London, 1982. ——. “Method and theory in medieval archaeology.” Archeologia Medievale 8 (1982): 7–37. ——. Light in the Dark Ages. The Rise and Fall of San Vincenzo al Volturno. London, 1997. ——. Towns and Trade in the Age of Charlemagne. London, 2000. ——. Goodbye to the Viking? Re-reading Early Medieval Archaeology. London, 2006. Hodges, R. and D. Whitehouse. Mohammed, Charlemagne & the origins of Europe: archaeology and the Pirenne thesis. London, 1983. Hohlwein, Nicolas. “Euhéméria du Fayoum.” Journal of Juristic Papyrology 3 (1949): 63–99. ——. “Le veteran Lucius Belllienus Gemellus, Gentleman-Farmer au Fayoum.” Études de Papyrologie 8 (1957): 69–91. Hölbl, Günther. Geschichte des Ptolemäerreiches: Politik, Ideologie und religiöse Kultur von Alexander dem Großen bis zur römischen Eroberung. Darmstadt, 1994. Homes-Fredericq, D. and J. B. Hennessy. “Um Qeis.” In Archaeology of Jordan, Vol. 2.2, Field Reports. Sites L-Z, eds. D. Homes-Fredericq and J.B. Hennessy, 597– 611. Leuven, 1989. Hope, Colin A. “Three Seasons of Excavation at Ismant el-Gharab in the Dakhleh Oasis, Egypt.” Mediterranean Archaeology 1 (1988): 160–178. ——. “Dakhleh Oasis Project: Ismant el-Kharab 1988–1990.” Journal of the Society for the Study of Egyptian Antiquities (1997): 157–176. ——. “The Find Context of the Kellis Agricultural Book.” In The Kellis agricultural account book: (P. Kell. IV Gr. 96), ed. Roger S. Bagnall and Colin A. Hope, 5–16. Dakhleh Oasis Project Monograph 7. Oxford, 1997. ——. with a contribution by Gillian E. Bowen. “The Archaeological context.” In Coptic Documentary Texts from Kellis, Vol. 1, P. Kell. V. (P. Kell Copt. 10–52; O. Kell. Copt. 1–2), eds. Iain Gardner, Anthony Alcock, and Wolf-Peter Funk, 96–124. Dakhleh Oasis Project Monograph 9. Oxford, 1999. ——. “Ostraka and Archaeology of Ismant el-Kharab.” In Greek ostraka from Kellis: O. Kellis, nos. 1–293, ed. Klaas A. Worp and Colin A. Hope, 10–15. Dakhleh Oasis Project Monograph 13. Oxford, 2004. Horden, Peregrine and Nicholas Purcell. The Corrupting Sea: A Study of Mediterranean History. Oxford, 2000. Hudson, P. Archeologia urbana e programmazione della ricerca: l’esempio di Pavia. Florence, 1981. Humphries, Mark. “Trading gods in northern Italy.” In Trade, Traders, and the Ancient City, eds. Helen Parkins and Christopher Smith, 203–224. London, 1998. Huß, Werner. Ägypten in hellenistischer Zeit: 323–30 v. Chr. Munich, 2001. Insoll, Timthoy. The Archaeology of Islam. Oxford, 1999. Jacoby, David. “The encounter of two societies: Western conquerors and Byzantines in the Peloponnesus after the Fourth Crusade.” American Historical Review 78 (1973): 873–906. Jacques-Meunié, D. J. “Greniers collectifs.” Hespéris 36 (1949): 97–137. ——. Greniers-Citadelles au Maroc. 2 vols. Publications de l’Institut des Hautes Etudes Marocaines 52. Paris, 1951. Jeffreys, Michael. “The Chronicle of the Morea: Priority of the Greek version.” Byzantinische Zeitschrift 68 (1975): 304–350. Jehasse, Jean and Laurence Jehasse. Aleria rediviva/Aleria ressuscité. 40 ans de découvertes archéologiques. Ajaccio, Corsica, FR, 1997. Jenkins, J. Keith. Re-Thinking History. London, 1997. list of works cited 225 Jenkins, M. “Medieval Maghribī Ceramics: A Reappraisal of the Pottery Production of the Western Regions of the Muslim World.” PhD diss., Institute of Fine Arts, New York University, 1978. Johnson, Janet. “Response.” Bulletin of the American Society for Papyrology 42 (2006): 268–270. Jordanie. Guides bleus. Paris, 1937. Kalonaros, Petros. To chronikon tou Moreōs. To hellēnikon keimenon kata ton kōdika tēs Kopenchagēs meta sumplērōseōn kai parallagōn ek tou Parisinou. Athens, 1940. Kaper, Olaf E. and Willemina Z. Wendrich. “East and West in Roman Egypt: An Introduction to Life on the Fringe.” In Life on the Fringe: Living in the Southern Egyptian Deserts During the Roman and Early Byzantine Periods, ed. Olaf E. Kaper, 1–4. Leiden, 1998. Karmon, Yehuda. “Geographical Components in the Study of the Ancient Mediterranean.” In Harbour Archaeology. Proceedings of the First International Workshop on Ancient Mediterranean Harbors. Caesarea Maritima, ed. Avner Raban, 1–6. British Archaeological Reports International Series 257. London, 1985. Katsafados [Katsaphados], Panayiotis S. Ta kastra tēs Mainēs. Athens, 1992. King, Anthony. “The Emergence of Romano-Celtic Religion.” In Early Roman Empire in the West, eds. Thomas Blagg and Martin Millett, 220–242. Oxford, 1990. Kingsley, Sean A. “Decline in the Ports of Palestine in Late Antiquity.” In Recent Research in Late Antique Urbanism, ed. L. Lavan, 69–87. Journal of Roman Archaeology Supplementary Series 42. Portsmouth, RI, 2002. Knapp, A. Bernard. “Archaeology and the Annales: time, space, and change.” In Archaeology, Annales, and Ethnohistory, ed. A. Bernard Knapp, 1–21. Cambridge, UK, 1992. Komis [Komes], Kostas. Plēthysmos kai oikismoi tēs Manēs, 15os–19os aiōnas. Yannina, 1995. Kordosis, Michalis. “I slaviki epoikisi stin Peloponniso.” Dodoni [Dōdōnē: epistēmonikē epetēris tēs Philosophikēs Scholēs tou Panepistēmiou Iōanninōn] 10 (1981): 381–429. Kougeas, S. “Peri ton Meligon.” Pragmateiai tis Akademias Athinon [Pragmateiai tēs Akadēmias Athēnōn] 15.3 (1950): 1–34. Kriesis, Anthony. “On the castles of Zarnata and Kelefa.” Byzantinische Zeitschrift 56 (1963): 308–316. Krings, Véronique, ed. La civilisation phénicienne et punique. Manuel de recherché. Leiden, 1994. Kulessa, Birgit. “The Harbour Suburb and its Significance for the Urban Development of Stralsund.” In Maritime Topography and the Medieval Town: Papers from the Fifth International Conference on Waterfront Archaeology in Copenhagen, 14–16 May 1998, Studies in Archaeology and History, Vol. 4, eds. Jan Bill and Birthe L. Clausen, 71–78. Copenhagen, 1999. Lacoste, H. “La restitution du plan antique d’Apamée in Syrie.” Bulletin de l’Académie Royale du Belgique, classe de Beaux­Arts 43 (1960): 53–62. Lagardere, J. “Droit des eaux et des installations hydrauliques au Maghreb et en Andalus aux XIe et XIIe siècles dans le Mi‘yar d’al-Wansharisi.” Les Cahiers de Tunisie 37–38, nos. 145–148 (1988–1989): 83–124. Lane, Arthur. “Medieval Finds at Al Mina in North Syria.” Archaeologia 87 (1938): 19–78. Lavagna, Rita and Carlo Varaldo. “La graffita arcaica tirrenica di produzione savonese alla luce degli scarti degli scarti di fornace dei secoli XII e XIII.” In [Atti:] 19. Convegno internazionale della ceramica: La ceramica graffita, Albisola, 30 maggio– 4 giugno 1986, 119–130. Albisola (Savona), 1989. Le Roy Ladurie, Emmanuel. Carnival: A People’s Uprising at Romans 1579–1580. Trans. Mary Feeney. London, 1979. 226 list of works cited ——. The Peasants of Languedoc. Trans. John Day. Urbana, 1974. Leciejewicz, L. “Italian-Polish Research into the Origin of Venice.” Archaeologica Polona 40 (2002): 51–71. Leciejewicz, L., E. Tabaczyńska, and S. Tabaczyński. Torcello: scavi, 1961–62. Rome, 1977. Leciejewicz, L., ed. Torcello: nuove ricerche archeologiche. Rome, 2000. Lenoir, Maurice. “Lixus à l’époque romaine.” In Lixus. Actes du colloque organisé par l’Institut des sciences de l’archéologie et du patrimonie de Rabat avec le concours de l’école française de Rome, Larache, 8–11 novembre 1989, 271–287. Collection de l’École français de Rome 166. Rome, 1992. Lenzen, C. J. and E. A. Knauf. “Beit Ras/Capitolias, a Preliminary Evaluation of the Archaeological and Textual Evidence.” Syria 64 (1987): 21–46. Leonard, John R. “Harbor Terminology in the Roman Periploi.” In Res Maritimae. Cyprus and the Eastern Mediterranean from Prehistory to Late Antiquity. Proceedings of the Second International Symposium “Cities on the Sea,” Nicosia, Cyprus, October 18–22, 1994, American Archaeological Research Institute Monograph Series, Vol. 1, eds. Stuart Swiny, Robert L. Hohlfelder, and Helena Wylde Swiny, 163–200. Atlanta, 1997. Leriche, P. “Le phénomène urbain dans la Syrie Hellénistique.” Bulletin d’études orientales 52 (2000): 99–125. ——. “Les fortifications grecques et romaines en Syrie.” In Archéologie et Histoire de la Syrie II. La Syrie de l’époque achéménide à l’avènement de l’Islam, eds. J.-M. Dentzer and W. Orthmann, 267–282. Saarbrücken, 1989. Lipiński, Edward, ed. Dictionnaire de la civilisation phénicienne et punique. Turnhout, 1992. Longnon, Jean and Peter Topping. Documents sur le régime des terres dans la Principauté de Morée au XIVe siècle. Paris, 1969. Longnon, Jean. “The Frankish states in Greece, 1204–1311.” In A History of the Crusades, eds. R. L. Wolff and H. W. Hazard, Vol. 2, 235–274. Philadelphia, 1963. Louis, A. “La conservation à long terme des grains chez les nomades et semi-sédentaires du sud de la Tunisie.” In Les techniques de conservation des grains à long terme: leur rôle dans la dynamique des systèmes de cultures et des sociétés, eds. Marceau Gast, François Sigaut, and Corinne Beutler, Vol. 1, 205–214. Paris, 1979. Lurier, Harold E. Crusaders as conquerors: The Chronicle of Morea. New York, 1964. Lyons, M. C. and D. E. P. Jackson. Saladin, the Politics of the Holy War. Cambridge, 1982. Mabry, J. and G. Palumbo. “Wadi Yabis Survey.” In Archaeology of Jordan, Vol. 2.1, Field Reports. Sites A-K., eds. D. Homes-Fredericq and J. B. Hennessy, 91–97. Leuven, 1989. Majewski, Teresita and Michael J. O’Brien. “The Use and Misuse of Nineteenth Century English and American Ceramics in Archaeological Analysis.” Advances in Archaeological Method and Theory 11 (1987): 97–209. Mannoni, Tiziano. “Anciennes et nouvelles méthodes dans l’archéologie des échanges et du commerce.” In Exchange and Trade in Medieval Europe. Papers of the Medieval Europe Brugge 1997 Conference, eds. Guy de Boe and Frans Verhaeghe, 7–14. Instituut voor het Archaologisch Patrimonium 3. Zellik, BE, 1997. Marazzi, Federico and Paolo Delogu. San Vincenzo al Volturno: cultura, istituzioni, economia. Miscellanea vulturnense 3. Montecassino: Abbazia di Montecassino, 1996. Markl, Otto. Ortsnamen Griechenlands in fränkischer Zeit. Graz, 1966. Markoe, Glenn E. Peoples of the Past: Phoenicians. London, 2000. Martorelli, Rossana, ed. Città, territorio, produzione e commerci nella Sardegna medievale: studi in onore di Letizia Pani Ermini offerti dagli allievi sardi per il settantesimo compleanno. Cagliari, 2002. Marzinot, Federico. Ceramica e ceramisti di Liguria. Genoa, 1979. Matthews, John F. The Journey of Theophanes: Travel, Business, and Daily Life in the Roman East. New Haven, 2006. list of works cited 227 Mayerson, Philip. “Enigmatic Knidion: A Wine measure in Late Roman/Byzantine Egypt?” Zeitschrift fur Papyrologie und Epigraphik 141 (2002): 205–209. Mayerson, P. “Qualitative Distinctions for ἔλαιον (Oil) and ψωμίον (Bread).” Bulletin of the American Society for Papyrology 39 (2002): 101–109. Mazzucato, Otto. I “bacini” a Roma e nel Lazio. 2 vols. Rome, 1973–1977. McCormick, M. Origins of the European economy: communications and commerce, A.D. 300–900. Cambridge, 2001. Meredith, David. “The Myos Hormos Road: Inscriptions and Ostraca.” Chronique d’Égypte 31 (1956): 356–362. Mignot, Ph., J. De Meulemeester, M. Boussalh, and M. Jlok. “Medieval Rural Settlement in Murcia (Spain) and Present-Day Rural Settlement in the Moroccan Atlas.” In European landscapes and lifestyles: the Mediterranean and beyond, eds. Z. Roca et alii, 131–145. Lisbon, 2007. Milanese, Marco. “Lo scavo archeologico di Castel Delfino (Savona).” Archeologia Medievale 9 (1982): 74–114. Milia, G. “La civiltà giudicale.” In Il Medioevo: dai giudicati agli Aragonesi, ed. Rafael Conde y Delgado de Molina and Massimo Guidetti, 193–229. Storia dei Sardo e della Sardegna 2. Milan, 1987. Miller, K. Itineraria romana. Römische reisewege an der hand der Tabula peutingeriana. Rome, 1916. Miller, William. The Latins in the Levant; a history of Frankish Greece (1204–1566). London, 1908. Minnen, Peter van. “House-to-house Enquiries: An Interdisciplinary Approach to Roman Karanis.” Zeitschrift fur Papyrologie und Epigraphik 100 (1994): 227–251. Molin, Kristian. Unknown Crusader Castles. London, 2001. Molinari, A. “L’Italy du Sud.” In Le Vert et le Brun: de Kairouan à Avignon, céramiques du Xe au XVe siècle, eds. Chapelle de la Vieille Charité (Marseille, France), Musées de Marseille, and Réunion des musées nationaux (France), 119–123. Marseille, 1995. Monceaux P. and L. Brossé. “Chalcis ad Belum: Notes sur l’histoire et les ruines de la ville.” Syria 6 (1925): 339–350. Montagne, R. “Un magasin collectif de l’Anti-Atlas: l’Agadir des Ikounka.” Hespéris 9 (1929): 145–266. ——. Villages et kasbas berbères: tableau de la vie sociale des Berbères sédentaires dans le sud du Maroc. Paris, 1930. Montanari, Armando. “A Modern Perspective: The Recent Development of Port Cities in Southern Europe.” In Mediterranean Cities: Historical Perspectives, eds. Irad Malkin and Robert L. Hohlfelder, 166–185. Totowa, 1988. Montanari, M. “Il capitolare di Liutprando: note di storia dell’economia e dell’alimentazione.” In La civilta comacchiese e pomposiana dalle origini preistoriche al tardo medioevo: atti del Convegno nazionale di studi storici, Comacchio 17–19 maggio 1984, ed. Convegno nazionale di studi storici, 461–475. Bologna, 1986. Moracchini-Mazel, Geneviève. Les Monuments paléochrétiens de la Corse. Paris, 1967. Moravcsik, G., ed. Constantine Porphyrogenitus, ‘De administrando imperio’. Trans. R. Jenkins. Rev. ed. Corpus fontium historiae byzantinae 1. Washington, D.C., 1967. Moreland, John. “Method and theory in medieval archaeology in the 1990’s.” Archeologia Medievale 18 (1991): 7–42. Moreland, J. [F.] “Restoring the Dialectic: Settlement Patterns and Documents in Medieval Central Italy.” In Archaeology, Annales, and Ethnohistory, ed. A. Bernard Knapp, 112–129. Cambridge, UK, 1992. Moreland, J. “The Middle Ages, Theory and Post-Modernism.” Acta Archaeologica 68 (1997): 163–182. ——. “Concepts of the Early medieval Economy.” In The Long Eighth Century, eds. I. Hansen and C. Wickham, 1–34. The transformation of the Roman world 11. Leiden, 2000. ——. Archaeology and Text. London, 2001. 228 list of works cited Morgan, Gareth. “The Venetian claims commission of 1278.” Byzantinische Zeitschrift 69 (1976): 411–438. Morris, Ian. Archaeology as Cultural History: Words and Things in Iron Age Greece. Oxford, 2000. Motzo, Bacchisio R., ed. Il compasso da navigare; opera italiana della metà del secolo XIII. Pref. e testo del Codice Hamilton 396. Cagliari, 1947. Mouterde, E. and A. Poidebard. Le limes de Chalcis; organisation de la steppe en haute Syrie romaine. Bibliothèque archéologique et historique 38. Paris, 1945. Mouton, J. M. Damas et sa principauté sous les Saljoukides et les Bourides (486– 549/1076–1154). Cairo, 1994. Muhs, Brian P. “The Girls Next Door: Marriage Patterns among the Mortuary Priests in Early Ptolemaic Thebes.” Journal of Juristic Papyrology 35 (2005): 169–194. Neglia, Giulia Annalinda. “Città del Mediterraneo: Aleppo, forme e tipi della città intra moenia.” PhD diss., Politecnico di Bari, 2003. Nepoti, Sergio. “Ceramiche tardo medievali spagnole ed islamiche orientali nell’Italia centro settentrionale adriatica.” In Segundo Coloquio Internacional de Cerámica Medieval en el Mediterráneo Occidental [Toledo, 1981], ed. Juan Zozaya, 353–363. Madrid, 1986. Nicolle, D. “Ain al Habis.” Archéologie Médiévale 18 (1988): 115–140. Nielsen, Thomas Heine, Lars Bjertrup, Mogens Herman Hansen, Lene Rubinstein, and Torben Vestergaard. “Athenian Grave Monuments and Social Class.” Greek, Roman and Byzantine Studies 30 (1989): 411–420. Nijf, Onno M. van. The Civic World of Professional Associations in the Roman Near East. Dutch Monographs on Ancient and Archaeology 17. Amsterdam, 1997. Nikoloudes, Nikos. “The Theme of Kinsterna.” In Porphyrogenita: essays on the history and literature of Byzantium and the Latin East in honour of Julian Chrysostomides, eds. J. Chrysostomides and Charalambos Dendrinos, 85–89. London, 2003. Norman, Albert Francis. Antioch as a Centre of Hellenic Culture as Observed by Libanius. Translated Texts for Historians 34. Liverpool, 2000. Northedge, Alistair. Studies on Roman and Islamic ‘Ammān, Vol. 1, History, site and architecture. British Academy monographs in archaeology 3. Oxford, 1992. ——. “Friedrich Sarre’s Die Keramik von Samarra in Perspective.” In Continuity and Change in Northern Mesopotamia from the Hellenistic to the Early Islamic Period. Proceedings of a Colloquium held at the Seminar für Vorderasiatische Altertumskunde Freie Univeristät Berlin, 6th–9th April, 1994, eds. Karin Bartl and Stefan R. Hauser, 229–258. Berliner Beiträge zum Vorderen Orient Band 17. Berlin, 1996. Oleson, John Peter. “The Technology of Roman Harbours.” The International Jorunal of Nautical Archaeology and Underwater Exploration 17.2 (1988): 147–157. Oman, C. The Art of War in the Middle Ages. London, 1991. Orser, Charles E. “Can there be an Archaeology of the Great Famine?” In ‘Fearful Realities:’ New Perspectives on the Famine, eds. Chris Morash and Richard Hayes, 77–90. Dublin, 1996. Pamir, Hatice and Shin’ichi Nishiyama. “The Orontes Delta Survey: An Archaeological Investigation of Ancient Trade Stations/Settlements.” Ancient West and East 1.2 (2002): 294–314. Pamir, Hatice. “The Orontes Delta Survey.” In The Amuq Valley Regional Projects Volume I: Surveys in the Plain of Antioch and Orontes Delta, Turkey. 1995–2002, ed. K. Aslihan Yener, 67–98. Oriental Institute Publications 131. Chicago, 2005. Panagiotopoulos [Panagiōtopoulos], V. Plēthysmos kai oikismoi tēs Peloponnēsou: 13os–18os aiōnas. Athens, 1987. Papadopoulos, P. B. Frangoi, Enetoi kai Tourkoi stin Peloponnisou 1204–1821 [Phrankoi, Henetoi kai Tourkoi stēn Peloponnēson, 1204–1821]. Athens, 1969. Paroli, Lidia. “Spiral Ware.” In Archeologia urbana a Roma: il progetto della Crypta list of works cited 229 Balbi. 3, Il giardino del Conservatorio di S. Caterina della Rosa, ed. Daniele Manacorda, Vol. 2, 237–238. Florence, 1985. Patitucci Uggeri, S. “Saggio stratigrafico nell’area di San Pietro degli Schiavoni a Brindisi. Relazione preliminare 1975–1976.” Richerche e Studi 9 (1976): 133–199. ——. La ceramica medievale pugliese: alla luce degli scavi di Mesagne. Museo Civico di Mesagne, 1977. ——. “Protomaiolica Brindisina.” Faenza 65 (1979): 241–255. ——. La Protomaiolica: bilancio e aggiornamenti. Quaderni di archeologia medievale 2. Florence, 1997. Patterson, Jeremy. “Trade and traders in the Roman World: Scale, Structure, and Organisation.” In Trade, Traders, and the Ancient City, eds. Helen Parkins and Christopher Smith, 149–167. London, 1998. Penã, Ignace. “Un puerto fluvial romano en el Orontes.” Liber Annus 45 (1995): 343–350. Pereda Roig, C. Los hórreos colectivos de Beni Sech-Yel. Ceuta, ES, 1939. Pestell, T. and K. Ulmschneider, eds. Markets in early medieval Europe: trading and productive sites, 650–850. Macclesfield, Cheshire, UK, 2003. Peters, F. E. “City planning in Greco–Roman Syria: Some New Considerations.” Damaszener Mitteilungen 1 (1983): 269–277. Philippson, A. Die griechischen Landschaften; eine Landeskunde, Vol. 3, Der Peloponnes, Part 2, Der Westen und Süden der Halbinsel. Frankfurt am Main, 1959. Piejko, Francis. “Episodes from the Third Syrian War in a Gourob Papyrus, 246 B.C.” Archiv für Papyrusforschung und verwandte Gebiete 36 (1990): 13–27. Pikoulas, Yannis [G.A.]. I notia Megalopolitiki chora. [Hē notia Megalopolitikē chōra: apo ton 8. p. Ch. hōs ton 4. m. Ch. aiōna (symvolē stēn topographia tēs)]. Athens, 1988. Pirenne, H. Medieval cities; their origins and the revival of trade. Princeton, 1925. ——. Mahomet et Charlemagne. Brussels, 1937. Poli, Fernanda. La Basilica di San Gavino a Porto Torres: la storia e le vicende architet­ toniche. Sassari, 1997. Ponisch, Michel. Lixus. Le quartier des temples. Rabat, 1981. ——. “Lixus: Information archeologiques.” In Aufstieg und Niedergang der Römische Welt. 2.10.2, 817–849. Berlin, 1982. ——. “Territoires utiles du Maroc punique.” In Phönizier im Westen, ed. Hans Georg Niemeyer, 429–444. Madrider Beiträge 8. Mainz, 1982. Porcella, F. “Il ‘fondo Pula’ e gli affiliati.” In La corona d’Aragona: un patrimonio comune per Italia e Spagna (secc. XIV–XV): Cagliari, Cittadella dei Musei, 27 gen­ naio­1 maggio 1989, ed. Gabriella Olla Repetto, 365–377. Cagliari, 1989. Prawer, J. Histoire du royaume latin de Jérusalem. 2 vols. Paris, 1970. ——. The World of the Crusaders. Jerusalem, 1972. Pringle, D. “Magna Mahomeria (al-Bīra): the Archaeology of a Frankish New Town in Palestine.” In Crusade and Settlement, ed. P. W. Edbury, 147–168. Cardiff, 1985. Ragona, Antonino. “Le fornaci medievali scoperte ad Agrigento.” Faenza 51 (1966): 83–89. ——. “Le fornaci trecentesche per ceramiche invetriate scoperte a Sciacca nel 1971.” Faenza 61 (1975): 3–6. Rahtz, Philip. “New Approaches to Medieval Archaeology Part 1.” In 25 Years of Medieval Archaeology, ed. David Alban Hinton, 12–123. Sheffield, 1983. Rautmann, Marcus. “Archaeology and Byzantine Studies.” Byzantinische Forschungen 15 (1990): 137–165. Rawson, Jessica, Michael S. Tite, and Michael J. Hughes. “The Export of Tang Sancai Wares: Some Recent Research.” Transactions of The Oriental Ceramic Society 52 (1987–1988): 39–62. 230 list of works cited Rey, E.-G. “Notes sur les territoires possédés par les Francs à l’Est du lac de Tibériade, de la Mer Morte, et du Jourdain.” Mémoire de la société nationale des antiquaires de France 41. 5th series, Vol. 1 (1880): 86–94. Riavez, P. “Atlit—Protomaiolica. Ceramiche italiane nel Mediterraneo orientale.” In 2. Congresso nazionale di archeologia medievale: Musei civici, Chiesa di Santa Giulia, Brescia 28 settembre–1 ottobre 2000, ed. Gian Pietro Brogiolo, 444–450. Florence, 2000. Ricci, Andreina. I mali dell’abbondanza: considerazioni impolitiche sui beni culturali. Rome, 1996. Rickman, Geoffrey E. “Portus in Perspective.” In ‘Roman Ostia’ Revisted: Archaeological and Historical Papers in Memory of Russell Meiggs, eds. Anna Galina Zevi and Amanda Claridge, 281–291. London, 1996. ——. “The Archaeology and History of Roman Ports.” International Journal of Nautical Archaeology and Underwater Exploration 17.3 (1988): 257–267. Ritterhausgesellschaft <Bubikon>. Die Kreuzzüge Petra—eine Spurensuche. Bubikon: Ritterhausges, 2006. Roueche, Charlotte. Performers and Partisans at Aphrodisias in the Roman and Late Roman Periods: A Study Based on Inscriptions from the Current Excavations at Aphrodisias in Caria. Journal of Roman Studies Monographs 6. London, 1993. ——. Aphrodisias in Late Antiquity: The Late Roman and Byzantine Inscriptions Including Texts from the Excavations at Aphrodisias Conducted by Kenam T. Erim. Journal of Roman Studies Monographs 5. London, 1988. Rovina, D. “Ceramiche graffite medievali e post medievali dal San Nicola di Sassari e altri siti della Sardegna centro settentrionale.” In [Atti:] 19. Convegno internazionale della ceramica: La ceramica graffita, Albisola, 30 maggio–4 giugno 1986, 201–209. Albisola (Savona), 1989. ——. “Il Duomo di Sassari: recenti indagini archeologiche.” In Sassari le origini, eds. AA.VV., 161–172. Sassari: Gallizzi, 1989. ——. La Sezione Medieval del Museo “G. A. Sanna” di Sassari. Piedimonte Matese, 2000. Rowland, Robert J. The Periphery in the Center: Sardinia in the Ancient and Medieval Worlds. BAR International Series 970. Oxford, 2001. Ruegg, S. Dominic. “Minturnae: A Roman Seaport on the Garigliano River, Italy.” In Archaeology of Coastal Changes: Proceedings of the First International Symposium “Cities of the Sea” Past and Present, Haifa, Israel, Sept. 20–22, 1986, ed. Avnar Raban, 209–228. British Archaeological Reports International Series 404. Oxford, 1988. Ruffing, Kai. “Das Nikanor-Archiv und der römische Süd- und Osthandel.” Münstersche Beiträge zur antiken Handelsgeschichte 12.2 (1993): 1–26. Salvi, D. “La ceramica medievale e post-medievale.” In Santa Chiara. Restauri e Scoperte, ed. A. Ingegno, 133–151. Cagliari, 1993. Sanpaolesi, Piero. Il Duomo di Pisa e l’architettura romanica toscana delle origini. Cultura e storia pisana 4. Pisa, 1975. Sari, Aldo. “Il romanico nel guidicato di Torres tra XI e XIII secolo.” In La civiltà giudicale in Sardegna nei secoli XI–XIII: fonti e documenti scritti. Atti del convegno nazionale, Sassari, Aula Magna dell’Università, 16–17 marzo 2001; Usini, chiesa di Santa Croce, 18 marzo 2001, 439–457. Sassari, 2002. Sauer, Eberhard W., ed. Archaeology and Ancient History: Breaking Down the Boundaries. London, 2004. Sauvaget, Jean. “L’enceinte primitive de la ville d’Alep.” Mélanges de l’Institut Français de Damas 1 (1929): 133–159. ——. «Les perles choisies» d’Ibn ach-Chihna. Matériaux pour servir à l‘histoire de la ville d‘Alep 1. Beyrouth, 1933. Reprint, Frankfurt am Main, 1993. ——. “Esquisse d’une histoire de la ville de Damas.” Revue des études islamiques 8 (1934): 421–480. list of works cited 231 ——. “Le plan de Laodicée-sur-Mer.” Bulletin d’études orientales 4 (1934): 81–114. ——. “Le plan de Laodicée-sur-Mer (note complémentaire).” Bulletin d’études orientales 6 (1936): 51–52. ——. “Le ‘tell’ d’Alep.” In Melanges syriens offerts a monsieur Rene Dussaud, secretaire perpetuel de l’Academie des inscriptions et belles-lettres, par ses amis et ses eleves. 2 vols. Ed. Academie des inscriptions & belles-lettres (France), Vol. 1, 59–63. Bibliotheque archeologique et historique 30. Paris, 1939. ——. “Le plan antique de Damas.” Syria 26 (1949): 314–358. ——. Alep, essai sur le développement d’une grande ville syrienne, des origines au milieu du XIXe siècle. 2 vols. Paris, 1941. ——. “Les trésor d‘or” de Sibt Ibn al-’Ajami. Matériaux pour servir à l‘histoire de la ville d‘Alep 2. Beyrouth, 1950. Sawyer, Peter H. “English Archaeology Before the Conquest: A Historian’s View.” In 25 Years of Medieval Archaeology, ed. David Alban Hinton, 44–47. Sheffield, 1983. Schnapp, A. “Archeologia urbana, archeologia preventiva.” In Archeologia urbana e centro antico di Napoli: atti del convegno 1983, ed. Stefania Adamo Muscettola, 25–27. Taranto, 1984. Schryver, J. “Spheres of Contact and Instances of Interaction in the Art and Archaeology of Frankish Cyprus, 1191–1359.” PhD diss., Cornell University, 2005. ——. “Colonialism or Conviviencia in Frankish Cyprus?” In Boundaries in Depth and in Motion, ed. I. William Zartman, 198–230. Athens, GA, 2010. Serra, Renata. La Sardegna. Italia romanica 10. Milan, 1989. Sidebotham, Steven E. Roman Economic Policy in the Erythra Thalassa. 30 B.C.–A.D. 217. Mnemosyne Supplement 91. Leiden, 1986. Sinclair, Thomas Alan. Eastern Turkey: An Architectural and Archaeological Survey, Vol. 4. London, 1990. Smail, R. C. Crusading warfare 1097–1193. Cambridge, 1995. Small, David B. “Monuments, Laws, and Analysis: Combining Archaeology and Text in Ancient Athens.” In Methods in the Mediterranean: Historical and Archaeological Views in Text and Archaeology, ed. David B. Small, 143–174. Mnemosyne Supplement 135. Leiden, 1995. Snodgrass, Anthony MacElrae. An Archaeology of Greece: The Present State and Future Scope of a Discipline. Berkeley, 1992. Spanu, Piergiorgio. “I possedimenti Vittorini del Priorato Cagliaritano di San Saturno. Il santuario del martire Efisio a Nora.” In Città, territorio, produzione e commerci nella Sardegna medievale: studi in onore di Letizia Pani Ermini offerti dagli allievi sardi per il settantesimo compleanno, ed. Rossana Martorelli, 65–103. Cagliari, 2002. Stephan, Robert P. and Arthur Verhoogt. “Text and Context in the Archive of Tiberianus (Karanis, Egypt; 2nd century AD).” Bulletin of the American Society for Papyrology 42 (2006): 189–201. Stone, Elizabeth C. “Texts, Architecture and Ethnographic Analogy: Patterns of Residence in Old Babylonian Nippur.” Iraq 43 (1981): 19–33. ——. Nippur Neighborhoods. Studies in Ancient Oriental Civilization 44. Chicago, 1987. Stuard, Susan Mosher. “Ancillary Evidence for the Decline of Medieval Slavery.” Past and Present 149 (1995): 3–28. Tantoulos, Antonis. “I entypi chartografisi tis Manis (16os–18os ai.), Istorikes kai geografikes plirofories.” In Mani: Témoignages sur l’espace et la société. Voyageurs et expéditions scientifiques (XVe–XIXe s.) [Manē: martyries gia to chōro kai tēn koinōnia : periēgētes kai epistēmonikes apostoles, 15os–19os ai. : praktika sympo­ siou, Limeni Areopolēs, 4–7 Noemvriou 1993], ed. Yannis [Giannēs] Saitas, 53–71. Athens, 1996. Tibble, S. Monarchy and Lordship in the Latin Kingdom of Jerusalem 1099–1291. Oxford, 1989. 232 list of works cited Tilley, C. Material Culture and Text: The Art of Ambiguity. London, 1991. Tolias, Giorgos. The Greek portolan charts: 15th–17th centuries: a contribution to the Mediterranean cartography of the modern period. Athens, 1999. Turtas, Raimondo. Storia della chiesa in Sardegna: dalle origini al Duemila. Rome, 1999. ——. “La Chiesa Sarda dalle origini fino al periodo spagnolo.” In Storia della Sardegna, eds. Manlio Brigaglia, Attilio Mastino, and Gian Giacomo Ortu, Vol. 1, 116–139. Rome, 2006. Tuzzato, S. “Venezia. Gli scavi a San Pietro di Castello (Olivolo). Nota preliminare sulle campagne 1986–1989.” Quaderni di Archeologia del Veneto 7 (1991): 92–103. ——. “Le strutture lignee altomedievali a Olivolo (S. Pietro di Castello—Venezia).” In Studi di archeologia della X Regio in ricordo di Michele Tombolani, eds. Bianca Maria Scarfì and Michele Tombolani, 479–485. Rome, 1994. —— et alii. “San Pietro di Castello a Venezia. Nota preliminare dopo la campagna 1992.” Quaderni di Archeologia del Veneto 9 (1993): 72–80. Valmin, Mattias Natan. La Messénie Ancienne [Études topographiques sur la Messénie ancienne]. Lund, 1930. Van Berchem, Denis. “Le port de Séleucie de Piérie et l’infrastructure logistique des guerres parthiques.” Bonner Jahrbücher 185 (1985): 47–88. van Leuven, Jon. “Blank spots and grey zones: medieval mapping in southern Greece.” In Eastern Mediterranean Cartographies, eds. Giorgos Tolias and Dimitris Loupis, 243–262. Tetradia Ergasias 25/26. Athens, 2004. ——. “The theme system.” In Encyclopedia of Greece and the Hellenic Tradition, ed. Graham Speake, Vol. 2, 1614. London, 2000. Van Liere, W. J. “Ager Centuriatus of the Roman Colonia of Emesa (Homs).” Les annales archéologiques de Syrie: revue d’archéologie et d’histoire 8 and 9 (1959): 55–58. Vannini, G. and C. Tonghini. “Medieval Petra: Settlement of the Crusader and Ayyubid Periods.” American Journal of Archaeology 101.3 (July 1997): 534–5. Vasmer, Max. Die Slaven in Griechenland. Berlin, 1941. Vayiakakos, Dikaios. “Peri ta paratainaria tis Manis toponymia.” Athina [Athēna: syng­ ramma periodikon tēs en Athēnais Epistēmonikēs Hetaireias] 68 (1965): 169–248. ——. “Kardamon—Kardamyli—Kardamyla.” Athina [Athēna: syngramma periodikon tēs en Athēnais Epistēmonikēs Hetaireias] 81 (1995): 173–240. ——. “I Ardouvista—Androuvista—Megali Chora tis Exo Manis.” Lakonikai Spoudai 13 (1996): 5–172. Vega, Noé Villaverde. “Aportaciones a la cronologia de la antigüedad tardía en Mauretania Tingitana: datos de los vajilias africanas.” In Lixus. Actes du colloque organisé par l’Institut des sciences de l’archéologie et du patrimonie de Rabat avec le concours de l’école française de Rome, Larache, 8–11 novembre 1989, 337–364. Collection de l’École français de Rome 166. Rome, 1992. Violante, C. La società milanese in età precomunale. Bari, 1953. Vliet, Edward Ch. L. van der. Strabo over landen, volken en steden. Van Gorcum’s historische bibliotheek 94. Assen, NL, 1977. Von Wartburg, M.-L. “Cane sugar production sites in Cyprus. Real and imagined.” Report of the Department of Antiquities, Cyprus (2000): 381–401. ——. “The Archaeology of Cane Sugar Production: A Survey of Twenty Years of Research in Cyprus.” Antiquaries Journal 81 (2001): 305–35. Vorderstrasse, Tasha. Al­Mina: A Port of Antioch from Late Antiquity until the end of the Ottomans. Publications de l’Institut historique-archéologique néerlandais de Stamboul 104. Leiden, 2005. ——. “A Port of Antioch: Late Antique al-Mina.” In Antioche de Syrie: histoire, ima­ ges et traces de la ville antique: colloque organisé par B. Cabouret, P.­L. Gatier et list of works cited 233 C. Saliou, Lyon, Maison de l’Orient et de la Méditerranée, 4,5,6 octobre 2001, eds. Cabouret, Bernadette, Pierre-Louis Gatier, and Catherine Saliou, 363–372. Topoi supplément 5. Lyon, 2004. ——. “A Port of Antioch under Byzantium, Islam, and the Crusades: Acculturation and Differentiation at al-Mina, A.D. 350–1268.” PhD diss., University of Chicago, 2004. ——. “Archaeology of the Antiochene Region in the Crusader Period.” In East and West in the Medieval Mediterranean I: Antioch from the Byzantine Reconquest until the end of the Crusader Principality. Acta of the Congress held at Hernen Castle in May 2003, eds. Krijnie Ciggaar and Michael Metcalf, 319–336. Orientalia Lovaniensia Analecta 147. Leuven, 2006. ——. “Ptolemaic Silver in Roman Egypt: Evidence from Papyri, Hoards and Excavations.” Dutch Studies Published by NELL 6.2 (2009): pp. 97–132. Wagstaff, John Malcolm. “Further observations on the location of Grand Magne.” Dumbarton Oaks Papers 45 (1991): 141–148. Walker, B. J. and Øystein S. LaBianca. “The Islamic Qusūr of Tell Hisbān: Preliminary Report on the 1998 and 2001 Seasons.” Annual of the Department of Antiquities of Jordan 47 (2003): 443–471. Walmsley, A. “The Middle Islamic and Crusader Periods.” In Jordan an Archaeological Reader, ed. R. B. Adams, 495–537. London, 2008. Ward-Perkins, B. “The towns of northern Italy: rebirth or renewal?” In The rebirth of towns in the West, AD 700–1050: a review of current research into how, when, and why there was a rebirth of towns between 700 and 1050, eds. R. Hodges and B. Hobley, 16–27. London, 1988. ——. The fall of Rome: and the end of civilization. Oxford, 2005. Weber, T. “Listib.” In Archaeology of Jordan, Vol. 2.2, Field Reports. Sites L-Z, eds. D. Homes-Fredericq and J. B. Hennessy, 368–373. Leuven, 1989. Whitcomb, D. “Trade and Tradition in Medieval Southern Iran.” PhD diss., University of Chicago, 1979. Whitehouse, D. “The Medieval Glazed Pottery of Lazio.” Papers of the British School at Rome 35 (1967): 40–86. ——. “The Medieval Pottery of Rome.” In Papers in Italian Archaeology 1. The Lancaster Seminar: Recent Research in Prehistoric, Classical, and Medieval Archaeology, eds. H. McK. Blake, T. W. Potter, and David Whitehouse, 475–493. BAR Supplementary Series 41. Oxford, 1978. ——. “7. The bacini of SS. Giovanni e Paolo, Rome; 8. Bacini at Gaeta.” In Medieval Lazio: Studies in Architecture, Painting and Ceramics, eds. D. Andrews, J. Osborne, and D. Whitehouse, 347–371. Papers in Italian archaeology 3. BAR International Series 125. Oxford, 1982. Wickham, Chris. Early Medieval Italy: Central Power and Local Society, 400–1000. New studies in medieval history. London, 1981. ——. “Overview: production, distribution and demand, II.” In The Long Eighth Century, eds. I. Hansen and C. Wickham, 345–377. The transformation of the Roman world 11. Leiden, 2000. ——. Framing the Early Middle Ages. Oxford, 2005. Wilfong, Terry. “The Archive of a Family of Moneylenders from Jême.” Bulletin of the American Society for Papyrology 27 (1990): 169–181. Will, E. “Les villes de la Syrie à l’époque hellénistique et romaine.” In Archéologie et Histoire de la Syrie, Vol. 2, La Syrie de l’époque achéménide à l’avènement de l’Islam, eds. J.-M. Dentzer and W. Orthmann, 223–250. Saarbrücken, 1989. Wilson, N. G. Scholars of Byzantium. London, 1983. Wirth, E. Die orientalische Stadt im islamischen Vorderasien und Nordafrika: städtische Bausubstanz und räumliche Ordnung, Wirtschaftsleben und soziale Organisation. 2 vols. Mainz, 2000. 234 list of works cited Wooding, Jonathon M. “Long Distance Imports and Archaeological Models for Exchange and Trade in the Celtic West AD 400–800.” In Exchange and Trade in Medieval Europe. Papers of the Medieval Europe Brugge 1997 Conference, eds. Guy de Boe and Frans Verhaeghe, 43–50. Instituut voor het Archaologisch Patrimonium 3. Zellik, BE, 1997. Woolley, Charles Leonard. “Excavations Near Antioch in 1936.” The Antiquaries’ Journal 17 (1937): 1–15. ——. “The Excavations at Al Mina, Sueidia. II.” Journal of Hellenic Studies 53 (1938): 133–170. ——. A Forgotten Kingdom: Being a Result of the Results Obtained from the Excavations of Two Mounds Atchana and Al Mina in the Turkish Hatay. London, 1953. Yared-Riachi, M. La Politique extérieure de la principauté de Damas. Damascus, 1997. Zacharakis, Christos. A Catalogue of Printed Maps of Greece, 1477–1800. 2nd ed. Athens, 1992. index abbassid caliphate 132, 136–137 abd er rhaman 109, 110 adriatic 56, 96, 101, 175–176, 178, 196–197, 207 africa xii, 28, 83, 96, 99, 101, 108, 110, 114, 156, 159–160, 164 ahmad ibn tulun 137 ajlun 72–73, 75, 79–80, 86, 88–90 albania 94 albert d’aix 73 aleppo xii, 10–11, 78, 83, 115–153 arab conquest of 132 hellenistic-roman 117–127 Byzantine 127–129 Mamluk 11, 116–117, 120–121, 125, 127, 131, 141, 143, 146, 150, 152 medieval islamic 129–151 pre-islamic 10, 116–117, 121, 132, 134, 141, 143 urban fabric of 11, 115–153 aleppo, locations within agora 120, 129, 133–134, 136 al-umari mosque 143–144 Bab antakia 121, 124–125, 133 Bab al-faraj 138, 143 Bab al-hadid 142, 146, 153 Bab al-nasr 121, 124, 129, 138, 153 Bab Quinnasrin 121, 124, 146–151, 153 Bab al-yahud 129, 143 congregational Mosque of 134, 136, 138 extra moenia 138, 140–142, 150 friday Mosque of 134 great Mosque of 121, 129, 134, 136, 141–142 intra moenia 140, 143 khans 136 musalla 136, 142 qaisariya 136 roman castellum 127, 137 roman forum 120, 129, 133–134, 136 sahn 136 suq 133–136, 138, 140, 142–143, 146–147 alexander ii, pope 111 al-Mina xi, 6–7, 15–39, 101 amaury i, King 79–80 amman 71, 80–81, 89 amphorae 191, 197, 200 antioch xi, 15, 25–27, 29–33, 36, 80, 124, 134 antiochene 25, 34, 36–37 aqaba 83 arab presence in the Mediterranean 93–114, 141 Arabia Felix 83 arabian peninsula 77 arabic 34, 75, 82, 171 arabs 11, 129, 152 aragon, Kingdom of 105 aragonese, language 42 archaeology vii, 181, 184, 186, 189–191, 193, 196, 205–207, 210 ethno- 11, 155–173 geo- 157 historical 6–7, 15–26 35 problem-oriented 8, 12, 196 non-selective 181, 196 rescue 206 urban 184, 205, 210 visible 10 architecture architecture of aleppo 115–153 architecture of pisa 94–95, 97–98, 105, 106–109, 114 architecture of Sardinia 93–114 ecclesiastical 94, 105–106, 113, 189 armenia xi little armenia 80 armenian(s) 34, 79 ascalon 78, 85 Assises (of Jerusalem) 72, 78, 87 assyrian 156 atlas Mountains 11, 157, 159, 166, 170–171 awnila river 171–172 ayyubid(s) 76–77, 82, 84, 89–90, 120, 127, 131, 138–140, 142–143, 146, 150 bacini 9–10, 93–114 churches with 94–95, 97–98, 105, 108–109, 114 Bagatti, Bellarmino 86, 90 Baldwin i 74 236 index Baldwin of Bourg 74–75 Balearic Islands 108 Ballardini, Gaetano 100 Barceló, Miquel 170 Beaufort 43–44, 55–58, 63, 66–67 Beit Ras (see Irbid) Berber(s) 11, 158–160, 165–166 Berti, Graziella 93–94 Bon, Antoine 47 Boni, Giacomo 180, 182 border(s)/borderlands 9, 41, 56, 71–73, 76, 78, 80, 82, 121 Bosra 72, 75, 77, 78, 80–82, 86, 90, 129 Burano 184 Byzantine Church 106 Byzantium 96, 108, 109, 110–111, 114 Cabezo de la Cobertera 157–158, 164 cadastral maps and records 120, 124, 146–147 parcelling 140 survey 116, 134 canal embankments 189–191, 208, 210 Canal, Ernesto 193, 194 Cape Matapan 43, 62, 65 Capitolare 197 Carolingian(s) 110 Carolingian Age 175 economy 175 cartography 61, 117 Carver, Martin 206 castle(s) xii–xiii, 3, 41–69, 72, 74–75, 81–82, 88, 100, 157, 193 Catalan, countryside, village, etc. 165 Catalans 44, 51 Caves de Suète 74, 76, 82, 87 cellaria 165 centuriatio 121 centuriation 125 charters 85, 87’ 90, 164 Chronicle of Morea 8, 42, 58 church(es) 9–10, 22, 51–53, 67, 89, 90, 93–114, 138, 165, 180, 189–190, 193, 204, 207–208, 210 Church, The 106, 109–112, 114 Cittanova Eraclea 201, 203–205 Comacchio xiii, 197–198, 200–202, 205 Constantine VII Porphyrogenitus 52–53, 64, 193 context 6, 11, 18–20, 24–26, 38, 95, 109, 156–157, 170, 180–181, 184, 206 Corsica(n) 28, 111 Corsy 77, 81, 86, 90 Crusader period, settlement, etc. 9, 32, 34, 52, 71, 73, 82, 84, 89 Crusader(s)/Crusade(s), The xii, 3, 8, 34, 41, 43, 56, 59, 71–91 cul-de-sac(s) 141, 147, 150, 151 Cutler, Anthony 108 Damascus 73–75, 77–79, 83–84, 125, 132, 137 Decapolis 72, 77, 86 Delogu, Raffaello 106 Dera’a 72, 74–77, 83, 86–90 Deschamps, Paul 73–74, 87 Desiderius 111, 112 diocese(s) 109, 207–208 Dorigo, Wladimiro 190 early medieval emporia 12, 175–176, 178, 201 early medieval period 12, 175–176, 178, 180 economy 4, 12, 77, 175, 178, 196 economic system(s) 176 Egypt 16, 25, 37, 39, 77, 97 Egyptian(s) 25, 76 el-Habis 74–75, 82, 86–87, 89 emporium/emporia 12, 175, 178, 197 debate 175–176, 178, ensilage 159 ethnology 11, 155 excavation(s) xi–xiii, 11, 18–19, 22, 72, 84, 86, 89, 91, 95, 98, 101, 103, 110, 157–158, 164, 167, 180–181, 183, 186–190, 193–194, 196, 201, 204, 207 Ezerite Slavs 64–67 Fatimid(s) 97, 137 fieldwalking 158 Fontana, Maria Vittoria 102 Fourth Crusade 8, 43 Francovich, Riccardo 95 Frankish castles, settlements, etc. xi, 8–9, 43, 46, 52–54, 57, 59, 63, 66–67, 69, 71–91 Franks 9, 44, 46–48, 50, 53–56, 59–62, 66–67, 71–91 French, language 34, 42, 63 Gadara 77, 81, 84, 86–89 Georgian(s) 34 Georgian speakers 22 Gerasa (see Jerash) Giovanni Diacono, Chronicle of GIS 208 Giudicati 109, 111–112 glazed tiles 108 Godefrey of Bouillon 73, 76 196 index Golan Heights 72, 74, 76–77, 81, 90 granary(ies) 11, 155–173, 210 agadir 159–160, 162–164 collective 159–162, 164–167 fortified 11, 157–168 ghorfa 160–162 ksar 160–162, 166 Grand Magne 8, 43–44, 58–63, 65–67 Greece 8, 41–69, 100–101 Greek, language 22, 34, 42, 56, 58, 63, 117 Green, Joseph 86 Gregoras 59–60, 64 Gregorian Reformation 94, 106–107, 110 Gregory VII, Pope 112 Gritsena 8, 43, 45–47, 49–52, 54–55, 62, 67 harbour(s) 15–16, 28, 30, 32–33, 35–39, 55–56, 60, 66 Hattin, Battle of 72, 77, 90 Hauran 72, 74–79, 86 Hejaz Road 77, 83 Heraclius (Byzantine emperor) 201 Hodges, Richard 178 Holy See 10, 106 holy 83 Hugue de Fauquenberg 74 Hugues de Léle 49–51 India 83 Iraq 83, 94, 137 Irbid 88, 90 irrigation 156–158, 166, 168–173 hydraulic networks 168–173 qanats 158, 172–173 Islamic Spain 156, 159 Islamic world 9, 93, 96 Israel 72, 76 Italian, language 42 Italians 34, 57 Italy 93–114, 175–209 Jacques-Meunié, Djamila 166 Jean d’Ibelin 78 Jebel Au’f 72, 75 Jerash (Gerasa) 75, 80–81, 84, 86 Jerusalem 9, 74–75 Jerusalem, king(s) of 74–75 Jerusalem, Kingdom of xii, 9, 71–72, 76, 85, 90 jizah 109–110 Jordan xi-xiii, 1, 72, 82–83, 85, 89–90 Jordan River 72–73, 75–76, 79, 84–85 237 Jordan Valley 84, 86 Jordanian 77, 84, 88–89 Kandak al-Rum 127, 140, 142 Kardamyli 50–52, 54–55, 68 Khalid ibn al-Walid 132 Khosrau I 127 Koran 162, 168 Kufranja Valley 86 Late Antiquity 134, 175, 184, 191, 207 Latin East 3, 72, 85 Leake, William 46–48, 51, 53, 57, 61 Leftro 43–44, 52–53, 55–59, 63–64, 66–67 Lenzen, Cherie 90 Leuktron 44, 52, 54–59, 63–64, 66–67 Liutprand (Lombard king) 197 London 205 Longobards 197 Mabry, Jonathan 88 Mahomérie 88 Maiorchine Islands 97 Mamluk 72, 84, 89 Mani peninsula 8, 41–69 Marghane River 166–167, 171–172 medieval architecture Sardinia’s 105–106 archaeology of 190 Meidan, Plain of 72, 83, 85, 90 Meling Slavs 43–44, 46, 52–55, 58–60, 62–67 Metamauco 204 Middle East xi, xii, 81, 100, 134, 156 Middle Eastern cities 140 Mistra 43–44, 53, 57–62, 64 monastic order(s) 109, 112–113 Montreal (Shawbak) xii, 82, 85, 90 Morea 43, 51, 53, 57 Morocco 1, 11, 28, 36, 96, 155, 157–160, 162, 165, 170, 173 Mudejar(es) 156, 158 Murano 184 Muratori, Saverio 115 Murcia(n) 11, 157–158, 168–170 Nicephorus Phocas 137 North Africa 96, 99, 101, 108, 110, 114 Nur al-Din 75, 79, 138 Olivolo 189, 207–208 Orient, the 44 oriental 83, 84 Orientalists 36 238 index Orontes River 7, 26–27, 29–32, 35–36 Ottoman(s) xi, 36–37, 39, 88, 136, 156 Oultre Jourdain xii, 71–72, 76–78, 80 Procopius 191 Ptolemy III 22, 30 Ptolemy 57 Pachymeres 55–56, 59, 67 Palmyra 80–81, 134 Palumbo, Gaetano 88 papyri 18–20, 25 papyrology 19, 37 Passava 43–45, 49, 60, 62–63, 66 Passeri, Gian Battista 93 Patitucci Uggeri, Stella 102 Pausanius 30–31 Peloponnese xi, 8, 41–69 Persia 142 Persian Gulf 83 Persian invasion of 540 127 Persians 34 Petra xi–xii, 71, 85 Philippe de Milly 80 photography, aerial and satellite 117, 158 Pirenne, Henri 175 Pisa 102 merchants from 106, 108–109, 110–112, 113 Opera of Santa Maria 112, 113 planning, city and town 10, 116–118, 121, 124–125, 127, 137, 140, 142, 152 Po River delta 197 valley 197, 201 Polish archaeologists/excavations 180, 183 port(s) xi, 6, 15–39, 57, 61, 66–67, 83, 97, 109, 197 Porto Kayio 59–62, 66 Posidonius 30 pottery bacini 93–114 cobalt and manganese (CoMn) glaze 97, 101 Gela ware 98–99 graffito 103–104 incised slip ware 100, 101 lead glaze 97, 98, 100, 102 lusterware 99–100, 101 protomajolica 99, 100–101 Pula ware 99 slipped ware 103–104 spiral ware 100–102 Principality of Achaea 8, 43, 53 Qasr Berdawil 80–81 Quweiq River 125 radiocarbon dating 167 Ravenna 205 Red Sea 76, 83 regadio 169 Roger de Loria 61 rural settlement 11, 155–173 sagrera 165 Sardinia xii, 9, 28, 93–113 Sawad 71, 73, 75–76, 84 Schnapp, Alain 206 Sea of Galilee 73, 84–85 secundum coelum 121–122, 124–125, 143, 145 secundum naturam 123–124, 126, 146–149 Segura River 157–158 Seleucus I 117 Seljuks 74–75, 77 Seljuk Empire 138 Shawbak (see Montreal) Sicily 96, 98–99, 101–103, 105, 108 Spain xiii, 1, 11, 94, 96, 99, 101, 105, 108, 110, 155–157, 164, 173 Sphrantzes xi, 64 Strabo 30–31, 57 Straits of Gibraltar 156 sugar 84–85 sugar cane 80, 84 sugar factories and production 84, 86, 88 Syria xiii, 1, 10–11, 30, 35–37, 72, 76–77, 79, 83–85, 115–153, 156 Syriac 34 Syrians 9, 74–76, 80, 115–153 Tancred de Hauteville 73–74, 76 Taygetos mountains 44, 46, 51, 53–54, 58, 62, 64–65, 68 Tazlaft 165–168, 171–173 Templar(s) 79, 80, 82, 89, 90 presence in Oultre Jourdain 80, 86, 89 Terre de Suète xii, 9, 71–91 Thietmar, Magister 81 Tibble, Steven 72–73, 87 index Tiberias 73, 84 Lake Tiberias 72, 86–88 prince(s) of 90 Tongiorgi, Liana 94 topography/topographic concerns 51, 53, 61, 63, 171, 186, 190, 196, 204, 210 Torcello 180, 182–183, 191, 193, 195–196, 204 glass workshop 183, 193 towns 52, 71, 79, 81, 86, 113, 175–177, 179, 185, 201 new 176, 178, 197 Transjordan 75–76, 80–83, 85 Tripolitania 160–162 Tughtekin 74 Tunisia xiii, 96–99, 101, 159–165 Turks 44, 50, 75, 141, 161 Tyre 84–85 Umayyad caliphate 132 Unar of Damascus 78 Valencia 99–100, 169 Valle de Ricote 11, 156–157, 168, 170 Valle Ponti 197, 201 239 Venetian Lagoon 175–209 San Francesco del Deserto 184, 191, 196 San Lorenzo di Ammiana 182, 184, 193, 196 Venetians 44, 56–57 Venice xiii, 8, 12, 104, 175–209 Austrian (period) 186 San Pietro di Castello 207–208 Rivoalto 189, 191, 204, 206–207 Victor III, Pope (see Desiderius) Wadi Yabis 89 Wadi Zarka 72, 80 William II Villehardouin 43, 45–47, 52–53, 57–59, 62, 66 William of Bures 90 William of Tyre 75, 78–79, 82, 85, 87 ‘winner’ vs ‘loser’ sites 204–205 Woolley, Sir Leonard 15, 22, 37 Yarmouk River 74, 81–82, 86 Yemen 83, 94 Zengids 136, 138–139, 143 figureS figure 1: polychrome lead glaze bacino from the church of San gavino of porto torres. produced in north africa (tunisia) in the second half of the eleventh century. figure 2: Map of Sardinia, with the four giudicati and names of all churches that have or had bacini. figure 3: façade of the church of Santa giusta at oristano. The church has been heavily restored and rebuilt in parts. figure 4: façade of the church of San nicolò at ottana. The bacini on the façade are replicas placed during the restoration. figure 5: church of San platano at villaspeciosa: a quintessential Sardinia hybrid structure with Byzantine spolia, bi-chromatic masonry, typically pisan roundels, and cavities for bacini. figure 6: Map of the western Mediterranean production centres, cities, and monasteries mentioned in the text. figure 7: fragment of a coMn bacino from tunisia, 1200–1250, from the church of San lorenzo in cagliari. figure 8: polychrome lead glaze bacino on San nicolò of trullas in Semestene, produced in eastern Sicily during the second half of the eleventh century. figure 9: luster ware and glazed blue on white slip bacino from the church of Santa Susanna in Busachi. Spanish production (valencia), 1300–1350. figure 10: luster ware and glazed blue on white slip bacino from the church of Santa Susanna in Busachi. Spanish production (valencia), 1300–1350. figure 11: façade of the church of Santa Barbara in Sassari, consecrated in the 1270s. The church had more than seventy bacini from at least three different italian production centers. The pottery predates the construction of the church. figure 12: protomajolica bacino from the church of Santa Barbara in Sassari. production of campania, second half of the thirteenth century. figure 13: Spiral ware bacino from the church of San vito in San priamo. Southern italian production, first half of the thirteenth century. figure 14: Monochrome bacino without graffiti from the church of San Vito in San priamo. Southern italian production, first half of the thirteenth century. figure 15: detail of the façade of the church of Santa trinità at Saccargia in condrongianus. note that it has been completely rebuilt during the first half of the twentieth century. The bacini are primary-colored plates—no attempt to replicate the originals was made. figure 16: façade and bell tower of the church of Santa trinità at Saccargia in condrongianus. figure 17: polychrome lead glaze bacino from San nicolò of trullas in Semestene. eastern Sicilian production, eleventh century. note how the bacino is built into the masonry. figure 18: façade of the church of San pietro di Sorres at Borutta. figure 19: The church of Santa giusta at oristano (heavily restored). The last two stories of the bell tower are completely rebuilt ex novo. table 1: list of churches with surviving bacini or recesses for bacini in each giudicato. SaSSari church toWn date date (uSing Bacini) (accepted) extant Bacini origin of Bacini receSSeS production porto torreS SeMeStene ardara olBia SaSSari condrongianuS MoreS 1050–1080 1080 pre-1107 pre-1111 pre-1113 — 1100–1120 1112 1116; 1180–1200 1100–1150 3 6 8 — — 51‡ — 9 1 11‡ + 40‡ ~1 8 9 10 11 S. gavino S. nicolo di trullaS S. Maria del regno S. SiMplicio S. pietro in SilKi S. trinita di Saccargia noStra Signora di todorache* S. Michele di Salvenor noStra Signora di talia S. giorgio S. antioco di BiSarcio ploaghe olMedo uSini oZieri 1110–1130 — — — — 37 3 1 6 12 13 14 15 16 17 18 19 S. antonio aBate* S. pietro di SorreS S. Maria Maddalena* S. antonio S. BarBara S. nicola S. Stefano S. ponZiano oSSi Borutta chiaraMonti SaSSari SaSSari SaSSari Monteleone roccadoria Molara 1300–1350 1127 1130 pre-1135 1150–1160; 1170–1190 1150–1200 1170–1190 1210 1270 1270 1272 1260–1300 — — — — 31 20 — 2 ~2 2 ~2 11 10 46 1 2 1 2 3 4 5 6 7 23 7 tunisia; Sicily Sicily replicas S. italy; n. italy Spain; S. italy; n. italy Spain table 1 (cont.) cagliari 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 church toWn date date (uSing Bacini) (accepted) extant Bacini origin of Bacini receSSeS production S. lorenZo S. Maria di SiBiola S. platano S. lucia S. Maria delle graZie* S. pietro di paradiSo S. luSSorio* S. BartoloMeo S. Maria aSSunta S. BarBara S. chiara S. pietro S. pietro S. geMiliano S. priaMo S. Salvatore S. pietro S. giovanni BattiSta S. pietro di ponte S. Saturnino S. BarBara S. Michele S. Maria aSSunta S. giovanni BattiSta S. pietro S. Biagio S. guglielMo S. narciSo cagliari Serdiana villaSpecioSa Serdiana deciMuputZu Quartuccio SelargiuS ortaceSuS MaracalagoniS furtei cagliari villa San pietro villaMar SeStu San vito SeStu SettiMo villaMar Quartu Sant’elena uSSana capoterra Siddi guSpini BaruMini guaMaggiore furtei cagliari furtei 1100–1110 2 — — — — — — — — — 3 — — 2 10 — — — 12 + 1 † 1 4 — — — — — — — 1250–1275 1250–1275 1250–1300 1250–1300 pre-1247 1120–1130 pre-1141 1100-1200 (?) 1100-1200 1100-1200 1150–1200 pre-1219 post-1237 1200–1300 1257 1250–1270 1250–1270 1250–1270 1200–1250 1250–1280 1250-1300 1270–1280 1270–1290 1200–1300 1281 1280–1300 pre-1297 pre-1316 1300-1400 ~48 9 3 2 2 1 1 1 1 1 3 32 6 10 10 10 5 10 57 + 7† 2 73 2 6 2 2 2 3+ 14 tunisia; S. italy tunisia; S. italy S. italy S. italy; n. italy Spain; S. italy; n. italy S. italy Spain; S. italy table 1 (cont.) oriStano church toWn 48 49 S. giuSta S. Maria Bonarcado 50 51 52 53 54 55 56 57 58 S. Michele S. leonardo oratorio delle aniMe S. giovanni BattiSta S. Maria di Zuradile * Santuario di Bonacattu S. Severa S. antonio da padova S. SuSanna SolaruSSa MaSullaS MaSSaMa Zerfaliu MarruBiu Bonarcado ollaStra SiMaxiS Zeddiani BuSachi date (uSing Bacini) date (accepted) extant Bacini Bacini receSSeS origin of production 32‡ — 32‡ 2 replicas 1250–1280 1300–1350 1300–1350 1135–1145 1146–1147; 1242–1268 1150–1200 1180–1210 1200–1225 1200–1250 1200–1300 1242–1268 1250–1280 1200–1300 1342 — — — — — 13‡ 1 2 4 1 2 5 2 8 14‡ 1 9 5 replicas S. italy Spain Spain date (uSing Bacini) date (accepted) extant Bacini Bacini receSSeS origin of production 1300–1350 1300–1350 pre-1116 1116–1160 post-1149 1300–1350 1300–1350 — 7‡ — 5 3 3 7‡ 1 15 30 nuoro 59 60 61 62 63 church toWn S. giovanni BattiSta S. nicolo S. Maria di corte S. antonio aBate S. gavino orotelli ottana Sindia oroSei oroSei replicas Spain unidentified * churches that had bacini, but have not been documented. † The original thirteenth-century building had 57 bacini, of which 12 survive. The sixteenth-century belfry had 7 bacini, of which 1 survives (from montelupo). ‡ new bacini and recesses have been inserted as part of restoration. it is unknown whether this decoration was original.