Warfare in the Norman Mediterranean 9781783275212

Table of contents :
Front cover
Contents
Illustrations
Contributors
Introduction
Part I: Re-examining the Narratives
1: Greek and Latin Sources for the Norman Expansion in the South:Their Value as “Military Histories” of the Warfare in the Mediterranean Sea
2: “Conquest in Their Blood”:Hauteville Ambition, Authorial Spin,and Interpretative Challenges in the Narrative Sources
3: “The Arts of Guiscard”:Trickery and Deceit in the Norman Conquests of Southern Italy and Outremer, 1000–1120
Part II: Cultural Representation and Diffusion
4: A Gift to the Normans:the Military Legacy of Sicilian Islam
5: Norman Battle Tactics in the Mediterranean Theatre of Operations: Fighting Lombards,Greeks, Arabs, and Turks c.1050–c.1100
6: Venetian Reactions to the Normans of Southern Italy under Robert Guiscard:from Enmity to Congeniality
Part III: Policies of Conquest, Consolidation,and Expansion
7: The Norman Kingdom of Sicily:Projecting Power by Sea
8: Norman Participation in the First Crusade: a Re-examination Luigi R
9: Strategy, the Norman Conquest of Southern Italy, and the First Crusade
10: Disaster in the Delta? Sicilian support for the Crusades and the Siege of Alexandria, 1174
Bibliography
Index

Citation preview

Warfare in Norman med.qxp_layout 1 21/05/20 21:36 Page 1

GEORGIOS THEOTOKIS is Lecturer at Ibn Haldun University, Istanbul. Contributors: Matthew Bennett, Daniel P. Franke, Michael S. Fulton, Serban V. Marin, David Nicolle, Francesca Petrizzo, Luigi Russo, Charles D. Stanton, Georgios Theotokis, James Titterton. Cover image: carved capital depicting two Norman soldiers, in the Cloisters of the Cathedral of Monreale, Palermo. Photo courtesy of Dr David Nicolle (photo taken in 1976).

GENERAL EDITORS: Matthew

Bennett, Anne Curry, Stephen Morillo

THEOTOKIS (ed.)

Warfare in History

WARFARE IN THE NORMAN MEDITERRANEAN

The kingdom of Sicily plays a huge part in the history of the Norman people; their conquest brought in a new era of invasion, interaction and integration in the Mediterranean, However, much previous scholarship has tended to concentrate on their activities in England and the Holy Land. This volume aims to redress the balance by focusing on the Hautevilles, their successors and their followers. It considers the operational, tactical, technical and logistical aspects of the conduct of war in the South throughout the eleventh and twelfth centuries, looking also at its impact on Italian and Sicilian multicultural society. Topics include the narratives of the Norman expansion, exchanges and diffusion between the “military cultures”of the Normans and the peoples they encountered in the South, and their varied policies of conquest, consolidation and expansion in the different operational theatres of land and sea.

WARFARE in the NORMAN MEDITERRANEAN EDITED BY GEORGIOS THEOTOKIS

warfare in history

Warfare in the Norman Mediterranean

warfare in history issn 1358-779x

Series editors Matthew Bennett, Royal Military Academy, Sandhurst, UK Anne Curry, University of Southampton, UK Stephen Morillo, Wabash College, Crawfordsville, USA This series aims to provide a wide-ranging and scholarly approach to military history, offering both individual studies of topics or wars, and volumes giving a selection of contemporary and later accounts of particular battles; its scope ranges from the early medieval to the early modern period. New proposals for the series are welcomed; they should be sent to the publisher at the address below. Boydell & Brewer Limited, PO Box 9, Woodbridge, Suffolk ip12 3df Previously published volumes in this series are listed at the back of this volume

Warfare in the Norman Mediterranean

Edited by

Georgios Theotokis

the boydell press

© Contributors 2020 All rights reserved. Except as permitted under current legislation no part of this work may be photocopied, stored in a retrieval system, published, performed in public, adapted, broadcast, transmitted, recorded or reproduced in any form or by any means, without the prior permission of the copyright owner

First published 2020 The Boydell Press, Woodbridge ISBN 978-1-78327-521-2 eISBN 978-1-78744-855-1 The Boydell Press is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. 668 Mt Hope Avenue, Rochester, NY 14620–2731, USA website: www.boydellandbrewer.com A CIP catalogue record for this book is available from the British Library The publisher has no responsibility for the continued existence or accuracy of URLs for external or third-party internet websites referred to in this book, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate

This publication is printed on acid-free paper Cover image: carved capital depicting two Norman soldiers, in the Cloisters of the Cathedral of Monreale, Palermo. Photo courtesy of Dr David Nicolle (photo taken in 1976).

Typeset by www.thewordservice.com

Contents List of Illustrations vii List of Contributors xv Introduction 1 Part I Re-examining the Narratives 1 Greek and Latin Sources for the Norman Expansion in the South: Their Value as “Military Histories” of the Warfare in the Mediterranean Sea 11 Georgios Theotokis 2 “Conquest in Their Blood”: Hauteville Ambition, Authorial Spin, and Interpretative Challenges in the Narrative Sources 35 Francesca Petrizzo 3 “The Arts of Guiscard”: Trickery and Deceit in the Norman Conquests of Southern Italy and Outremer, 1000–1120 55 James Titterton Part II Cultural Representation and Diffusion 4 A Gift to the Normans: the Military Legacy of Sicilian Islam 79 David Nicolle 5 Norman Battle Tactics in the Mediterranean Theatre of Operations: Fighting Lombards, Greeks, Arabs, and Turks c.1050­–c.1100 133 Matthew Bennett 6 Venetian Reactions to the Normans of Southern Italy under Robert Guiscard: from Enmity to Congeniality 151 Şerban V. Marin Part III Policies of Conquest, Consolidation, and Expansion 7 The Norman Kingdom of Sicily: Projecting Power by Sea Charles D. Stanton 8 Norman Participation in the First Crusade: a Re-examination Luigi Russo

177 195

vi contents

9 Strategy, the Norman Conquest of Southern Italy, and the First Crusade 211 Daniel P. Franke 10 Disaster in the Delta? Sicilian Support for the Crusades and the Siege of Alexandria, 1174 225 Michael S. Fulton Bibliography 239 Index 251

Illustrations Maps 1 Map of the major towns and cities in Italy in the eleventh century xviii 2 Map of the major towns and cities in Sicily in the eleventh century xix 3 Map of the Middle East in the middle of the twelfth century xx Figures 2 “Conquest in Their Blood”: Hauteville Ambition, Authorial Spin, and Interpretative Challenges in the Narrative Sources. Francesca Petrizzo 1 The Hauteville siblings: a simplified family tree 34 4

A Gift to the Normans: the Military Legacy of Sicilian Islam. David Nicolle Helmets 1 Helmet with one-piece bowl and riveted cross-frame, reportedly found at Chamosen in Switzerland, Islamic early tenth to twelfth century, although the cross-pieces and brow band may have been added later in Europe (Schweizerisches Landes Museum, Zurich) 110 2 Helmet with one-piece bowl, Islamic early tenth to twelfth century, reportedly found at Raqada in central Tunisia (Museum of Islamic Studies, Raqada) 110 3 Helmet with one-piece bowl and chiselled decoration, Iran, eighth–ninth century (Furusiyah Art Foundation, inv. R-815, London) 110 4 Helmet found in the wreck of a Sassanian or early Islamic merchant ship off Bandar Rig on the Persian Gulf coast of Iran (Museum of Islamic Archaeology, Tehran) 110 5 Helmet reportedly found in eastern Iran or Afghanistan, probably late tenth century AD, shown without its decorative finial (private collection) 110 6a-b Helmet made from at least four riveted plates; at one time regarded as Fatimid tenth to twelfth century, but more likely a later medieval European great helm with its face-plate removed (Museum of Islamic Archaeology, Kayrawan) 110

viii illustrations

7–12 Weapons 7 Sword from the wreck of an Islamic ship, found off Agay, western Islamic, tenth century (Museum of Underwater Archaeology, Saint-Raphael) 111 8 Short sword with ivory-covered hilt, Sicily twelfth– thirteenth century (Furusiyya Art Foundation, inv RB-133, London) 111 9 Short sword excavated at Liétor, Andalusian, ninth–tenth century (after Navaro Palazon) 111 10 Short sword with an illegible Arabic inscription on the blade, probably Sicily, twelfth–thirteenth century (private collection) 111 11 Sword pommel, Sicily or al-Andalus, twelfth–early thirteenth century (Furusiyya Art Foundation, inv RB-93, London) 111 12 Bronze double-headed axe, reportedly found near Toulon and initially thought to be early medieval Islamic; but probably a fake (after Lacam) 111 13–14 Panel painting 13 Painted ceiling panel of a cavalryman with a kite-shaped shield, Siculo Islamic, c.1140–43 AD (in situ over the southern side-aisle of the Capella Palatina, Palermo) 112 14a-d Painted panels of mounted warriors, three with elongated but flat-bottomed shields, Siculo Islamic, c.1140–43 AD (in situ in the muqarnas ceiling of the Capella Palatina in Palermo) 112 15–25 Ivory panels 15 Warrior with a short-sleeved mail shirt or hauberk on a carved ivory oliphant, partially covered by a later silver rim, Sicily, late eleventh century (on loan to the Victoria & Albert Museum, London) 16a-b Carved ivory box showing armed men with short-sleeved mail shirts or hauberks, Sicily, eleventh century (Museum für Islamische Kunst, Berlin) 17 Carved ivory oliphant showing a warrior with a fulllength mail hauberk , Sicily, late eleventh century (Musée Crozatier, Le Puy-en-Velay) 18 Carved ivory oliphant showing a man wielding a mace, southern Italy or Sicily, eleventh century (Museum of Fine Arts, inv. 57.58L, Boston)

113 113 113 113

illustrations ix

19a-c Carved ivory box showing armed men with short-sleeved mail shirts or hauberks, Sicily, 1050–1100 AD (Metropolitan Museum of Art, inv. 17.190.241, New York) 114 20 Carved ivory oliphant showing a warrior with a mail shirt, Sicily, late eleventh century (Metropolitan Museum of Art, inv. 04.3.177, New York) 114 21 Carved ivory plaque showing a man wearing a full mail hauberk with long sleeves and perhaps a mail coif, Sicily, southern Italy or Andalusia, late twelfth century (National Museum of Antiquities, Ravenna) 114 22a-b Carved ivory plaques showing men wearing full mail hauberks with long sleeves and perhaps separate mail coifs, Sicily, southern Italy or Andalusia, late twelfth-early thirteenth century (Hermitage Museum, St. Petersburg) 115 23a-c Carved ivory plaques showing guardsmen with separate mail coifs and probably armour-lined and fabric-covered coats, from al-Humayma, Jordan, probably made in Khurasan, before 750 AD (Archaeological Musuem, Aqaba) 115 24a-d Carved ivory plaques showing scenes from the Life of Christ with armoured men in largely Byzantine style [a and d] and armed but unarmoured men with Islamicstyle tiraz bands around their upper sleeves [b–c], from Cathedral Altar, first half of twelfth century, southern Italy: a) Herod’s guards; b) guards at Holy Sepulchre; c) guards at Crucifixion; d) Massacre of the Innocents (Cathedral Museum, Salerno) 116 25 Philistines shown in provincial Byzantine style, one apparently with a form of breastplate [left], on “Rome Casket”, carved ivory box, Sicily or southern Italy, late twelfth century (Palazzo di Venezia Museum, Rome) 116 26–28 Ivory chess pieces 26a–b “Charlemagne’s chess set”, carved ivory chess knight with kite-shaped shield, Sicily or southern Italy, eleventh century (Bibliothèque Nationale Cabinet des Medailles, Paris) 27a–b “Charlemagne’s chess set”, carved ivory chess knight with round shield, Sicily or southern Italy, eleventh century (Bibliothèque Nationale Cabinet des Medailles, Paris) 28a–b “Charlemagne’s chess set”, carved ivory chess pawn or infantryman with a long but flat-bottomed shield, Sicily or southern Italy, eleventh century (Bibliothèque Nationale Cabinet des Medailles, Paris)

117 117

117

x illustrations

29–40 Stone and stucco carvings 29a–c Carved relief of huntsmen, one clearly wearing a form of quilted “soft” armour [a], southern Italy, late eleventh century (in situ above southern portal, Church of San Benedetto, Brindisi) 30a–f Carved relief showing siege of a city, southern Italy, very late eleventh or very early twelfth century: a–c) defenders; d–f ) attackers from the right wearing lamellar armour. Note that the attackers from the left wear mail armour (in situ above north door, Church of San Nicola, Bari) 31 Carved relief of a horseman wearing a small lamellar cuirass, probably part of Pharoah’s army, southern Italy, very late eleventh or very early twelfth century (in situ west front, Church of San Nicola, Bari) 32 Carving of a merman with a mace and round shield, southern Italy, 1175–1200 AD (in situ on a column inside the Cathedral, Bitonto) 33a–f Carved capitals showing a variety of warriors, some with African features, and with a variety of European, Byzantine, and Islamic styles of weaponry, armour, and costume, Sicily, c.1189 AD (in situ Cathedral Cloisters, Monreale) 34a–g Carved reliefs showing assorted demons in combat with spears, maces, and an early form of guisarme axe [d], southern Italy, late twelfth century (in situ Cathedral, Barletta) 35a–b Ambone di Nicodemo showing huntsmen or warriors with a mace and a composite bow, stone and stucco relief carving on pulpit, southern Italy, first half of twelfth century (in situ church of Santa Maria in Valle Porclaneta, Rosciolo dei Marsi) 36 Carved capital showing a typical northern Italian cavalryman and an infantryman with a segmented helmet plus long but flat-bottomed shield, northern Italy, mid-twelfth century (in situ Cathedral, Parma) 37a–b Figures on carved capitals, identified as Islamic by the embroidered tiraz bands on their upper sleeves, southern Italy, early thirteenth century (in situ Cathedral, Matera) 38a–b Carved relief of huntsmen, one with a long-sleeved mail hauberk, helmet, and composite bow [a] and one wearing only a kilt-like garment [b], either Siculo-Islamic eleventh century from an earlier building or Christian Sicily early twelfth century (in situ above north door, Church of La Martorana, Palermo) 39 Painted stucco statuette of a guardsman wearing a quilted qalansawah hat and a mail hauberk, from the Umayyad

118

118

119 119

119 120

120

121 121

121

illustrations xi

Palace at Khirbat al-Mafjir, Palestine, early eighth century (Rockefeller Museum, Jerusalem) 122 40a–d Relief carvings of shields symbolising elements of the Fatimid army, decorated plus a sword for the caliphal guards [a], elongated with flat bases for the infantry [b–c] and round for the cavalry [d], Egypt c.1087 AD (in situ Bab al-Nasr, Cairo) 122 41–54 Manuscripts and manuscript fragments 41 Painted papyrus fragment of a cavalrymen with an apparently kite-shaped shield, probably from Fustat, Egypt, tenth or eleventh century (Musée du Louvre, inv. MA 0125, Paris) 42 Drawing on paper of a faris cavalryman with a small round shield, from Fustat, Egypt, probably tenth century (Österreichische Nationalbibliothek, Ms. ACh. Vindob. 11416, Vienna) 43 Painted paper fragment showing horsemen greeting above the carnage of battle [note shield and severed limbs], the visible horsemen apparently wearing a lamellar cuirass, from Fustat, Egypt, probably early twelfth century (Keir Collection, inv. I.8, London) 44 Painted paper fragment of a turbaned warrior with a large shield and two javelins, from Fustat, Egypt, probably early twelfth century (Museum of Islamic Art, inv. 13801, Cairo) 45a–c Painted paper believed to show the Fatimid garrison of Asqalon emerging to confront Crusader foes who are wearing clearly identifiable European military equipment [not shown here], from Fustat, Egypt, probably mid-twelfth century (British Museum, Department of Oriental Antiquities, London) 46 Painted paper showing unarmoured Arab or Berber cavalrymen, from Fustat, Egypt, probably mid-twelfth century (private collection; present location unknown) 47a–c Manuscript illustrations in the Tabsira by Mardi Ibn ‘Ali Ibn Mardi al-Tarsusi, Egypt, c.1171 AD: a) crossbow forming part of the release mechanism of a counterweight trebuchet; b) crossbow mounted inside a shield; c) crossbow modified to shoot incendiary grenades (Bodleian Library, Ms. Hunt 264, Oxford) 48a–f Gospels made in Damietta, Coptic Egypt, 1179/80 AD: a) guard of Pilate with a pair of javelins; b) soldiers at Crucifixion with round and elongated flat-based shields; c) the Betrayal showing swords, a probably mace, small

123

123

123 123

124 124

124

xii illustrations

49

50a–b

51a–b 52

53a–b

54a–d

round and elongated flat-bottomed shield; d) Herod’s guards, one with a pair of javelins; e) soldier before Joseph of Arimathea, with bow, quiver, shield, sword, and pair of javelins; f ) soldier with a pair of javelins casting lots for Christ’s clothes (Bibliothèque Nationale, Ms. Copte 13, ff. 82v, 83v, 79r, 5r, 131r, 274v, Paris) Ruler’s guard, perhaps a Muslim, judging by the tiraz bands around his upper sleeves, Leges Langobardorum from Benevento, southern Italy, eleventh century (Archives of Badia della Santissima Trinità, Abbey of La Cava, Subiaco) Pharoah’s army in the Red Sea, including these cavalrymen with long-hemmed, long-sleeved mail hauberks, Exultet Roll, southern Italy, eleventh century (Cathedral Archives, Roll 2, Gaeta) Guards with probable small forms of lamellar cuirass, at Dedication to Duke and Emperor, Exultet Roll, southern Italy, eleventh century (Museo Civico, Pisa) Guard of the Norman ruler, perhaps identified as a Muslim by his beard and pointed hat, and with a large elongated shield, Regestum di Sant Angelo in Formis, southern Italy, c.1150 AD (Library, Ms. Reg. 4, Abbey of Monte Cassino) Synopsis of Histories by John Skylitzes, Sicily, late twelfth– early thirteenth century: a) Arab emir wearing long-skirted lamellar cuirass fleeing from Bardas Phocas; b) one of the Arab soldiers attacking Edessa, wearing small form of lamellar cuirass (Biblioteca Nacional, Cod. 005-3. N2, ff. 136v & 208r, Madrid) Liber ad homorem Augusti by Peter of Eboli, Sicily or southern Italy, very late twelfth or early thirteenth century: a) unarmoured Muslim and Christian archers; b) Christian and Muslim crossbowmen, two with helmets, one with a turban; c) archer with a composite bow and a quiver at the siege of Salerno; d) crossbowman with brimmed hat or helmet at the siege of Naples (Burgerbibliothek, Cod. 120, ff. 131r, 117r and 109r, Bern)

125

126

126 126

126

126

127

55–60 Ceramics 55 Ceramic fragment showing a soldier with a long-hafted mace and perhaps mail beneath his coat or tunic, Egypt or Iraq, ninth-tenth century (Benaki museum, inv. 227, Athens) 128 56 Ceramic bowl showing horsemen with small round shields and one with a sword, from Sabra al-Mansuriyah, Tunisia, 950–1050 AD (Museum of Islamic Arts, Kayrawan) 128

illustrations xiii

57a–b Ceramic plaques from Sabra palace showing bearded Arab or Berber soldiers on foot and an apparently “moon faced” cavalryman, perhaps indicating a Turk, Tunisia, mid-eleventh century (Bardo Museum, Tunis) 128 58 Lustre-ware ceramic fragment showing a turbaned infantryman with a spear, an elongated kite-shaped shield and perhaps indicating mail or other protection over his left arm, from Fustat, Egypt, tenth–eleventh century (Victoria & Albert Museum ceramic study collection, London) 128 59 Turbaned soldier with a straight sword and a large, elongated but flat-bottomed shield, on a lustre-ware ceramic bowl, Egypt or Iran, twelfth century (Keir Collection, inv. 151, London) 128 60a–b Figure of horseman with a helmet, shield, and sword, on the rim of the Vaso de Tavira terracotta bowl, IslamicAndalusian southern Portugal, eleventh century (Museu Municipal, Tavira) 129 60c–d Figure of horseman with a turban and spear, on the rim of the Vaso de Tavira terracotta bowl, Islamic-Andalusian southern Portugal, eleventh century (Museu Municipal, Tavira) 129 60e–f Figure of probable infantryman with a large shield, on the rim of the Vaso de Tavira terracotta bowl, IslamicAndalusian southern Portugal, eleventh century (Museu Municipal, Tavira) 129 61 Wall painting 61a–b Soldiers with long but in one case clearly flat-bottomed shields [a] at the Crucifixion, wall painting, central Italy, twelfth century (in situ San Paolo fuori la Mura, Rome) 130 62–64 Metalwork 62a–d Silver plate found at Perm-Molotov in Siberia, Christian Iraq, Iran or Central Asia, probably seventh century: a) angel with a mace guarding the Holy Sepulchre, either wearing a full mail hauberk or covered with feathers; b) shepherd armed with a mace; c–d) soldiers armed with maces at the Crucifixion (Hermitage Museum, St Petersburg) 130 63 Soldier or huntsman with a composite bow on a bronze door panel, southern Italy, c.1179 AD (in situ Cathedral, Ravello) 131 64a–b Men armed with shields and maces, perhaps engaged in a judicial duel, on a bronze door panel, southern Italy, late twelfth century (in situ Cathedral, Trani) 131

xiv illustrations

65–66 Mosaics 65 Man apparently armed with a long-handled mace in the Legend of King Arthur, mosaic, southern Italy, ninth– tenth century (in situ Church of the Pantocrator, Otranto) 131 66 Crossbowman struck down by an opponent’s arrow during Turkish attack on crusader-held Antioch, northern Italy, early eleventh century (in situ Church of San Colombano, Bobbio) 131 8 Norman Participation in the First Crusade: a Re-examination Luigi Russo, 1 Parchment compiled by order of Bohemond, 1096 (Courtesy of Nicolaian Museum of Bari) 196 The editor, contributors and publishers are grateful to all the institutions and persons listed for permission to reproduce the materials in which they hold copyright. Every effort has been made to trace the copyright holders; apologies are offered for any omission, and the publishers will be pleased to add any necessary acknowledgements in subsequent editions.

Contributors Dr Matthew Bennett, FSA (London), FRHistS, taught at The Royal Military Academy Sandhurst (1984–2014) and is currently a Visiting Research Fellow at Southampton University and a Visiting Lecturer at Winchester University. The focus of his research is the ethos and practice of warfare in the High Middle Ages, especially chivalry, largely through the medium of Old French literature. The military aspects of the Norman Conquest(s) have also been a major interest for almost forty years. His publications include: Campaigns of the Norman Conquest, The Cambridge Illustrated Atlas of Warfare: the Middle Ages 768–1497, with N. H. Hooper, Medieval World at War and Medieval Hostageship c.700– c.1500: Hostage, Captive, Prisoner of War, Guarantee, Peacemaker, with K. Weikert (Routledge). Daniel P. Franke is Assistant Professor of History at Richard Bland College of William and Mary in Petersburg, Virginia. His research focuses on the social, cultural, and economic factors that influence mobilisation, strategy, and operations in medieval warfare. He is the author of several articles on the crusades and the Hundred Years’ War, and is co-editing the Oxford Handbook of Medieval Military History. He is also currently completing his first monograph, The Rise and Fall of the House of Ufford: War, Family, and the State in Fourteenth-Century England, a study of the military activities of the earls of Suffolk under the later Plantagenet kings. Michael S. Fulton is a history instructor at Langara College, Vancouver. He completed his PhD at Cardiff University before serving as a postdoctoral research fellow with the Zinman Institute of Archaeology at the University of Haifa and then as a visiting scholar with the history department of the University of British Columbia. His research deals primarily with the history and archaeology of the Near East during the period of the Crusades, focusing on the nature and development of various conflicts, the use of the landscape by military planners, and the evolution of fortifications and siege warfare. He is the author of the books Artillery in the Era of the Crusades and Siege Warfare during the Crusades.

xvi Contributors

Șerban V. Marin is a Romanian medievalist. His research area focuses on the manner in which the Venetian chronicles represent various events in the past, mostly those related to Byzantium, the Muslims, and the classical crusades. He was awarded a PhD in 2009 from the University of Bucharest for his research considering the myth of the origins of Venice as presented in the Venetian chronicles. He is currently head of department at the National Archives of Romania. David Nicolle lectured in art history at Yarmouk University, Jordan, from 1983 to 1987. Since then he has continued to write and carry on research as an independent scholar attached to Nottingham University as an honorary research fellow. He now has over one hundred books to his name, while he has also worked for television in the UK, the USA, Germany and Syria as an advisor and contributor. Francesca Petrizzo is a postdoctoral tutor at the Institute for Medieval Studies, University of Leeds. She specialises in the history of kin, culture, and identity at war during the central Middle Ages, with a special focus on the Normans in the South in the eleventh and twelfth centuries. She is currently preparing a book on the life and cultural legacy of Tancred, the southern Norman and youngest of the leaders of the First Crusade, building on a Leverhulme Trust-funded project. Luigi Russo is Associate Professor of Medieval History at the European University of Rome. His main research interests include the history of the crusading movement and the history of the Normans in Italy and the Latin East. He is a member of the Board of Administration of CESN (Centro Europeo di Studi Normanni, Ariano Irpino), a Fellow of OUEN (Office universitaire d’Etudes normandes, Caen), a Scientific Committee Member of the Carnet Mondes normands médiévaux (http://mnm.hypotheses.org/a-props/equipe), the editor of the ‘Dizionario Biografico degli Italiani’, and an Advisor for the American Academy in Rome (http://www.aarome.org/people/advisors). Charles D. Stanton is a retired U.S. naval officer and former airline pilot, and earned his doctorate in medieval history at Fitzwilliam College of the University of Cambridge under Professor David Abulafia. He has written extensively on the maritime history of medieval Europe with a special focus on the Norman conquest of southern Italy and galley operations in the Mediterranean. He was awarded the 2006 Denis Bethell Prize of the Haskins Society for a paper on the subject. His body of work includes numerous essays published in respected scholarly journals and two well-received monographs: Norman Naval Operations in the Mediterranean (2011) and Medieval Maritime Warfare (2015). A third book, Roger of Lauria (c.1250–1305), ‘Admiral of Admirals’, was published in October 2019.

Contributors xvii

Georgios Theotokis is an Assistant Professsor at Ibn-Haldun University, Istanbul. He received his PhD in History from the University of Glasgow (2010), with a specialisation in medieval military history. He has published numerous articles and monographs on the history of conflict and warfare in Europe and the Mediterranean in the medieval and early modern periods, including Norman Campaigns in the Balkans 1081–1108 (2014), Byzantine Military Tactics in Syria and Mesopotamia in the Tenth Century (2018), and Twenty Battles that Shaped Medieval Europe (2019). He is also the editor of The Military History of the Mediterranean Sea (2018). James Titterton is a postdoctoral fellow at the Leeds Arts and Humanities Research Institute at the University of Leeds. He received his PhD in July 2019 from the University of Leeds; his thesis, entitled ‘Trickery and Deception in Medieval Warfare, c.1000–c.1330’, analyses the use and perception of deceitful tactics in medieval chronicles. His other publications include articles on chivalric language in John Barbour’s epic Scots poem The Bruce, honour in medieval tournament culture and the use of Vegetius’s De re militari in William of Malmesbury’s Gesta regum Anglorum.

Map 1:  Major towns and cities in Italy in the eleventh century

Map 2:  Major towns and cities in Sicily in the eleventh century

Map 3:  The Middle East in the middle of the twelfth century

Introduction After it had become pleasing to the Mighty King who orders the seasons as well as kingdoms that the shores of Apulia, for so long possessed by the Greeks, should no longer be occupied by them, the people of the Normans, distinguished by their warlike knights, should enter and rule Italy, after expelling the Greeks. (William of Apulia, The Deeds of Robert Guiscard, Book I)

In his monumental The Making of Europe, Robert Bartlett wrote that “one of the more striking aspects of the expansionary activity of the tenth to thirteenth centuries was the movement of western European aristocrats from their homelands into new areas where they settled and, if successful, augmented their fortunes”. He was referring to the medieval aristocratic expansion from the core of Europe’s old “Carolingian lands” into its periphery – eastern and southern Europe and, of course, the Middle East. Undoubtedly, therefore, in any history of the Norman people the kingdom of Sicily, the most cosmopolitan of all their conquests, demands an important place. Whether we consider the year 1017 as a cut-off point marking the advent of the Norman presence in Italy and Sicily, thus inaugurating a new era of invasion, interaction, and integration in the Mediterranean, is a moot point. Whatever we decide, the millennial anniversary is significant, and while the idea behind this collective volume was conceived in 2017, the moment offers an ideal opportunity to explore the story of the Norman exploits in the South about a thousand years ago. Regrettably, the bibliography on the topic of warfare in the Norman South has been overshadowed by the achievements of their cousins in the North. Only in the last two decades or so has there been a steady increase in the number of works that focus on the strategies and battle tactics, military organisation and equipment, expansion and consolidation of power, negotiation and diplomacy of the Normans in the Mediterranean Some of the established scholars who have been working on these topics for many years, like Bennett,1 Stanton,2 and Russo,3 have been kind enough to partici1

“Norman Conquests: A Strategy for World Domination?” Journal of Medieval Military History 15 (2017), pp. 91–101; “Norman Naval Activity in the Mediterranean, 1060–1108”, Anglo-Norman Studies 15 (1992), 41–58. 2 Norman Naval Operations in the Mediterranean (Woodbridge, 2011); “Naval Power in the Norman Conquest of Southern Italy and Sicily”, Haskins Society Journal 19 (2008), 120–36. 3 I Normanni del Mezzogiorno e il Movimento Crociato (Bari, 2014); Boemondo. Figlio del Guiscardo e principe di Antiochia (Avellino, 2009).

2 introduction

pate in this collective volume. To their contributions I have to add my own monograph, a study that examines the clash of two distinct Christian military cultures: the bold and chivalrous Normans versus the cautious and calculating Byzantines.4 It was the research into this topic that sparked my interest in the Norman gens in the Mediterranean and stirred my fascination with the warrior ethos and the fighting tactics of these newcomers to Italy and the South. The aim of the present volume is to build on the momentum of the last two decades and try to bring the Hautevilles, their successors, and their followers back into the spotlight of European and Mediterranean history. It is our intention to deviate somewhat from the “fashionable” narratives of the so-called “New Military History” and to reintegrate the operational, tactical, technical, and equipment aspects of the conduct of warfare in the South in the eleventh and twelfth centuries, while at the same time not ignoring its impact on the wider multicultural society of Italy and Sicily. The volume is divided thematically into three parts in order to pinpoint different aspects of the history of warfare in the Mediterranean in the eleventh and twelfth centuries. Part I focuses on a critical re-examination of the narratives of Norman expansion in the South. In chapter 1 Georgios Theotokis ponders how heavily modern historians should depend on medieval texts to recreate the actual tactics and strategies of the Normans in Italy and Sicily during their expansion between the 1010s and the 1070s. He asks numerous complex questions that have to do with the chroniclers’ knowledge of army numbers, battle tactics, the topography of battlefields etc.; in a sense, how far can we go in questioning our sources to “flesh out” the details of a military campaign and/or a clash between army units, and what kind of stumbling blocks does a modern historian have to overcome in the process? While recognising the hazards of relying on medieval authors – frequently dependent on ethnographic tropes and derivative set-piece battles – Theotokis argues persuasively that a close reading and balancing of these sources can allow one to recreate an accurate picture of the Normans’ military campaigns in the South. In chapter 2 Francesca Petrizzo challenges the picture presented by the sources for the Norman conquest of the South – which appears to be that of a ragtag band of brave warriors who take over Southern Italy, despite their status as underdogs amid isolation, illness, and famine. She argues that a close reading of the texts and cross-referencing with the scant documentary sources enables a more complex interpretation of the issue. Petrizzo focuses on the sources’ narratives of the methods and challenges of the Hauteville invasion in order to discuss authorial bias and its impact on the usefulness of narrative sources for the purposes of military history. It is her contention 4

The Norman Campaign in the Balkans, 1081–1108 AD (Woodbridge, 2014).



introduction 3

that the chroniclers had an ambivalent attitude towards Hauteville military ambition, both celebrating their warlike nature and seeking to extenuate it by presenting them as the troubled and unlikely victors in an embattled landscape. By analysing the inner contradictions within the texts themselves and comparing the treatment of the conquests of Sicily and southern Italy by the chroniclers, Petrizzo seeks to contextualise and deepen our attitude towards the narrative sources of the Norman conquest, and to reaffirm but also problematise their value as sources for military history. As cunning and deceit have long been recognised as essential features of Normanitas, one might expect their enemies to emphasise the Normans’ more dubious habits. Yet the Normans appear to have embraced this reputation, regarding cunning as a virtue to be celebrated. James Titterton’s chapter 3 focuses on the military cunning displayed by leading members of the Hauteville family: Robert Guiscard, Roger of Sicily, Bohemond of Antioch, and Tancred. Eleventh- and twelfth-century chronicles are replete with examples of the ruses that these commanders employed in Italy, Sicily, and the Levant, which ranged from ambushes and night attacks to more elaborate deceptions involving bribery, disguise, and “fake corpses”. His analysis compares and contrasts how these ruses were represented by authors sympathetic to the Hautevilles – particularly William of Apulia and Ralph of Caen – with those representations by authors who were ambivalent or outright hostile towards them, such as Anna Comnena, Albert of Aachen, and Amatus of Montecassino. Titterton demonstrates in a colourful and comprehensive manner the ambiguous moral nature of military ruses in this period, arguing that, being neither inherently licit nor illicit, they could be praised as exemplars of military skill and prudence, or condemned as treacherous and cowardly, depending on the sympathies of a particular author and his or her audience. Part II offers a view into the exchanges and diffusion between the military cultures of the Normans and the peoples they encountered in the South. Recent scholarship has made clear the fact that the Muslims of Sicily and southern Italy played a significant part in the spread of early medieval Islamic military technology from the Islamic world (Maghrib, Spain, and the Middle East) into western Europe. It is also well known that the Norman rulers of Sicily and southern Italy employed and made very effective use of Muslim troops from Sicily, while many aspects of their recruitment, organisation, skills, and motivation are also well understood. But the only aspect of their tactics, and by extension their equipment, that has been studied in more detail has been their archery. David Nicolle offers an exciting new approach to the topic of interaction between different military cultures in the central Mediterranean (Sicily, southern Italy) by focusing on the diffusion of military technology, military organisation, and the tactics of the warriors who operated in that region. He contends that, in order to understand the military heritage of Sicilian Islam, and thus of the Sicilian

4 introduction

Muslims’ contribution to Siculo-Norman armies, it is necessary to trace not only where this heritage originated but also how it was further developed during the period of Islamic rule in Sicily – and, indeed, in parts of southern Italy. Nicolle’s well-grounded study demonstrates that, despite the disparity of documentary evidence about the military technology and organisation of the Sicilian Muslims, a remarkable tribute to the effectiveness of the Sicilian Muslim military heritage is the fact that these troops still operated in a similar way to their ancestors of earlier centuries, throughout and beyond the Norman period. During the eleventh century the Normans gained a strong reputation for their performance on the battlefield. While they were distinguished for their craftiness and cunning spirit, they were also known for their cruelty, bloodthirstiness, and destruction as they conquered throughout Europe. In chapter 5, Matthew Bennett focuses on the narratives that portray the Normans’ battle tactics and their “invincibility” in the battlefield. Admittedly, it is not difficult to find information on the topic, for the image makers (or propagandists?) of the early Norman expansion in Italy and Sicily have glorified their supposed superiority in the field of battle. What is far more challenging, however, is to make sense of narratives that portray the Normans as warriors par excellence, and to compare them to the narratives of their enemies (Anna Comnena is a typical example). Bennett’s contribution thus concentrates on the Norman battle techniques portrayed in a wide range of narrative accounts (Latin, Greek, and Arab) and exactly how the Norman style of combat adapted to cope with many other battle techniques. Bennett maintains that, despite the violent dialectic in the battlefields of the Mediterranean between the Norman and other military cultures, historians should treat what are essentially propagandist representations of Western military superiority with great caution. Yet, he seeks to build a full picture of their talents in the field of battle that – together with other chapters in this volume which explain the Normans’ skills in politics, diplomacy, siege, and naval warfare – would demonstrate the varied reasons leading eventually to the Normans becoming one of the leading Mediterranean powers of the High Middle Ages. Complementing Bennett’s portrayal of the Normans and their fame as warriors, Şerban V. Marin’s study in chapter 6 explores the political and cultural representation of the Normans among the Venetians in the crucial period of the first Norman invasion of the Balkans (1081–84), emphasising the three main types of Venetian representation (scenarios) in their chronicles of their confrontations with the Normans in this period . Although the Normans were regarded as great enemies of Venetian interests in the Adriatic region and the Balkans, one cannot deny that, at the same time, the Venetian chronicles present a feeling of congeniality towards them and high esteem for their military skills, along with an increasing aversion towards the Comneni emperors of Byzantium. Marin attempts to sketch the changes in attitudes



introduction 5

and the cultural impact of the Normans in Venice, and the way the Norman “echoes” were portrayed in the Venetian chronicles of later centuries. Part III explores the varied policies of conquest, consolidation, and expansion in the different operational theatres (both on land and at sea) in which the Normans are mentioned the eleventh and twelfth centuries. In chapter 7 Charles D. Stanton examines the ways in which the Norman Kingdom of Sicily projected its sea power through the length and breadth of the Mediterranean during the contentious twelfth century. His is a unique analysis of the manner in which the Norman kings exercised their maritime might, which contrasted sharply with their more mercantile Italian competitors. Stanton’s chapter examines the distinctive development of Norman naval policy by charting its evolution through six identifiable phases: conquest, consolidation, expansion, deterrence, adventurism, and finally atrophy. In so doing, he delineates each stage with discussion of specific military expeditions or operations of the period in question and how each sovereign influenced or modified naval policy from the 1130s to the end of the century. Stanton’s well-founded and meticulous chapter demonstrates that the development of Norman naval power in the central Mediterranean was, by virtue of its exceptional geographic environment and innovative architects, decidedly idiosyncratic. Chapter 8, by Luigi Russo, deals with the complex question of who undertook the journey to Jerusalem, especially with regard to the call for the First Crusade. Was crusading a truly popular movement that embraced all social groups of medieval Europe? Was it a spontaneous act or, rather, a well-organised military operation? Who took part in the crusades, and how were they recruited? Russo’s chapter aims to reignite discussion of the participation in the First Crusade of Norman contingents from southern Italy under Bohemond Hauteville. His focus, however, is not limited to Bohemond’s motives in taking the Cross in the summer of 1096. Rather, he tries to disentangle the complex aristocratic networks of kinship and solidarities that dominated southern Italian politics after the death of Robert “Guiscard” Hauteville in 1085, and to evaluate – as far as possible – how far these affected the shaping of Italo-Norman policies during the First Crusade and Bohemond’s subsequent expedition against Byzantium in 1107–8. Russo’s scrutiny of the motives of the Italo-Norman aristocrats who followed Bohemond to the Middle East reveals the peculiarity of Bohemond’s crusade contingent, in the sense that these departing knights were not following any well-established family strategy of pilgrimage, as in other places in western Europe, and that they signed up from other parts of Italy that were not under Bohemond’s seigneurial control, which proves that aristocratic kinship did not play a major role in the recruitment of leading Italo-Norman peers on the eve of the First Crusade. Daniel P. Franke’s contribution in chapter 9 attempts to tackle another crucial topic that has often puzzled historians: that of the strategic connec-

6 introduction

tions between the Norman conquest of southern Italy and the First Crusade. At the heart of the issue lie a series of complex questions about the theory and practice of warfare in the Middle Ages, including the definition of “strategy” and the extent to which we can recover a sense of medieval commanders’ decision-making processes. Taking into account the conquest of Sicily, Bohemond of Taranto’s command decisions, the relationship of the Principality of Antioch to the Duchy of Apulia, and finally the creation of the Kingdom of Sicily, Franke argues that a broader definition of strategy as the often improvised personal or corporate use of military force to solve problems is necessary in order to understand what happened in the late eleventh century. While the papacy’s conception of “crusade” was certainly impacted upon by the Norman conquest of Sicily, successive popes did not steadily develop crusade doctrine from the treaty of Melfi in 1059, nor did Norman lords regard fighting Muslim powers as significantly more important than consolidating their power against rival Norman, Lombard, and Byzantine claimants. Further, southern Norman participation in the First Crusade did not reflect a larger Norman strategic plan but, rather, the personal strategic goals of the main leadership, Bohemond of Taranto and Tancred of Lecce. Franke argues that it is not until the regency of Adelaide and the rule of her son King Roger II in the 1120s and 1130s that we see the clear shape of a defined Norman strategy, demonstrating that “strategy” is best understood, in Moltke’s classic formula, as a system of expedients, and that strategic thought often follows, rather than shapes, the reality that it seeks to control. Staying in the Middle East, in chapter 10 Michael S. Fulton’s study explores a momentous episode in the Norman participation in the crusades during the twelfth century. As the Latin principalities of the Levant enjoyed limited military support from the Norman Kingdom of Sicily through most of the twelfth century, what is often noted as the exception is the Sicilian attack on Alexandria in 1174. Typically presented as an effort to aid the Frankish principalities, the attack was reportedly planned in coordination with Amalric of Jerusalem and Shiite conspirators in Egypt. The reality, however, may have been somewhat different. As Fulton asserts, most of what we know about this episode comes from Muslim sources, and most details can be traced back to the propagandistic efforts of Saladin. By inflating the threat posed to Alexandria and, by extension. to Muslim Egypt, Saladin added to the public image that he was cultivating: one of a warrior dedicated to the fight against the invading Christians and a devout defender of Sunni Islam. By implicating the Franks of the Levant in the attack, and associating it with a Fātimid conspiracy that had miscarried months earlier, Saladin further enlarged the significance of the defeat. In reality, Fulton maintains that there is little to corroborate claims that the Sicilians were acting in coordination with either the Kingdom of Jerusalem or Shiite dissidents in Egypt. This may have been a far more traditional raid than has previously



introduction 7

been supposed, one similar to others conducted by Sicilian forces against neighbouring ports along the coast of the Delta. It is clear from the range of concerns and issues raised in this volume that Norman warfare in the South had a momentous impact not only on the individual but at all levels of medieval Mediterranean society, regardless of ethnic, regional, religious, or professional affiliation. Who were these normanni who settled in (and/or conquered) southern Italy and Sicily, and how were they portrayed in the various contemporary and later histories of the Norman expansion? These “images of warfare” of the so-called Norman conquest of the Mezzogiorno are revisited through a critical re-examination of the effectiveness of the narrative sources for the purposes of military history. Yet the Norman military culture in the Mediterranean did not emerge ex nihilo, but was, rather, the conscious imitation and adaptation of different aspects of military technology, military organisation, and the tactics of armies that had operated in that region for centuries prior to their arrival, like the Byzantines and the Muslims from North Africa. It is this “violent dialectic” in the different fields of battle in the Mediterranean, and the way that each military culture adapted to its enemies, that helps to build a more comprehensive picture of Norman warfare in the Mediterranean, from the ways in which the Norman style of combat adapted to cope with many other battle techniques, to the Normans’ skill in achieving their objectives in war “by other means”. The final part of this volume helps to disentangle some complex questions about the Norman strategy of expansion throughout the eleventh and twelfth centuries, both on land and at sea. A critical point in this respect is that of the strategic connections between the Norman conquest of southern Italy and the First Crusade: did the Norman decision makers have any succinct concept of “grand strategy”, and what sense can modern historians make of it? Whether we tend to see the southern Norman contingents of the First Crusade as part of a larger normannitas that sought to conquer the world, or whether we view the Norman “grand strategy” as reactive and unsophisticated (a sort of “crisis management on a grand scale”), the only thing we can be sure of is that the ultimate emergence of a unified strategy within a medieval state like the Kingdom of Sicily did not mean that it was inevitable or not heavily dependent on battlefield outcomes and personal leadership. The same cannot be said about the naval policy of the Norman Kingdom of Sicily and the way that each ruler influenced or modified it during the contentious twelfth century, for which we have a much clearer picture of the distinct phases in the building of Norman naval power in the central Mediterranean. It is these divergent images of war, strategies, battle tactics, and diplomacy on a grand scale that dominate the pages of this collection of essays on the Warfare in the Norman Mediterranean. Ultimately, the volume aims

8 introduction

to assess, and in some cases to confirm and in others to revise, the established paradigm of the development of Norman warfare in the South, and to stimulate academic discussion of this overlooked topic.

Part I

Re-examining the Narratives

1

Greek and Latin Sources for the Norman Expansion in the South: Their Value as “Military Histories” of the Warfare in the Mediterranean Sea

Georgios Theotokis

The eleventh century is a period of profound political, social, and cultural upheaval in Italy, owing largely to the so-called “invasion” of the Normans who, initially offering their services to the highest bidder, ended up carving out their own polities in Apulia, Campania, and Sicily. This chapter will concentrate on military affairs in southern Italy between 1017, the year of the famous meeting with the Lombard rebel Melus, and 1077, which ended with the siege of Naples by the joint forces of Robert Guiscard and Richard of Capua. The chief aim of the chapter is to examine the Greek and Latin narrative sources for this period, strictly from a military perspective, and to reach some conclusions regarding their value as sources for the Norman expansion in the South and the history of warfare in the Mediterranean in the eleventh century. The major questions that are raised are: to what extent are the figures they provide for army sizes reliable, both in absolute numbers and in the ratios given between cavalry and infantry? What is our chroniclers’ knowledge of the local geography where the military operations took place, and to what extent – if at all – were they familiar with the terrain of the battles or sieges, or with the campaign routes of the armies they describe? How accurate and detailed are their descriptions of castles and fortifications, and how far do their narratives permit the accurate reconstruction of chains of events, especially in regard to the battlefield manoeuvres of armies in action? Do we have enough information to reconstruct the Norman strategy of expansion in the different but interconnected operational theatres of Italy and Sicily?

12

georgios theotokis

An Introduction to the Narrative Sources The primary sources that are examined for “military” evidence on the Norman expansion in the Mediterranean include mainly those authors who wrote in Latin, like Amatus of Montecassino, William of Apulia, and Geoffrey Malaterra. The first of these three is the author of the earliest of the three substantial narrative accounts of the conquest of southern Italy by the “peoples beyond the Alps” in the eleventh century, from its earliest stages in the 1010s to the death of Richard I of Capua on 5 April 1078. The History of the Normans1 was probably completed shortly after the death of Richard I, hence around 1078/79. William of Apulia wrote a poem in hexameter (“epic verse”) about The Deeds of Robert Guiscard,2 probably between 1096 and 1099, although the poem is not exclusively concerned with the life of the duke of Apulia and Calabria. William seems to have been a member of Roger Borsa’s court and was probably a layman, although his full identity remains elusive. He was obviously well informed about the Byzantine Empire, but his contemporary sources remain difficult to identify. Finally, Geoffrey Malaterra is the third chronicler to commemorate the conquest of Italy and Sicily by the Normans, and indeed the only one whose focus is Roger Hauteville. Malaterra notes that he was a “newcomer” to the region and that he almost certainly had come to Sicily, sometime after 1091, at the request of Count Roger, who wished to re-establish the power and influence of the Latin Church on the island. His The Deeds of Count Roger of 1

M. Champollion-Figeac, ed., L’Ystoire de li Normant et la chronique de Robert Viscart, par Aimé, moine du Mont-Cassin (Paris, 1835). A modern translation of the work is The History of the Normans, trans. Prescott Dunbar (Woodbridge, 2004). The main secondary works for Amatus’ life and work are: Kenneth B. Wolf, Making History: The Normans and Their Historians in Eleventh-Century Italy (Philadelphia, 1995), pp. 87–122; Emily Albu, The Normans in their Histories: Propaganda, Myth and Subversion (Woodbridge, 2001), pp. 106–44; Ferdinand Chalandon, Histoire de la domination normande en Italie et en Sicile, 2 vols. (Paris, 1907), pp. xxxi–xxxiv. 2 Guillaume de Pouille, La Geste de Robert Guiscard, ed. Marguerite Mathieu (Palermo 1963). For this study I used Graham Loud’s translation, which can be found online: https://bit.ly/2KcEtD7. The main secondary works on William are Chalandon, Domination normande, pp. xxxviii–xl; Wolf, Making History, pp. 123–42; Huguette Taviani-Carozzi, La Terreur du monde, Robert Guiscard et la conquête normande en Italie (Paris, 1996), pp. 20–23; Albu, The Normans in Their Histories, pp. 106–44; Graham A. Loud, “Anna Komnena and her Sources for the Normans of Southern Italy”, in Church and Chronicle in the Middle Ages. Essays Presented to John Taylor, ed. Ian N. Wood and Graham A. Loud (London, 1991), pp. 41–57 [repr. in Graham A. Loud, Conquerors and Churchmen in Norman Italy (Aldershot, 1999)]; Paul Brown, “The Gesta Roberti Wiscardi: A ‘Byzantine’ history?”, Journal of Medieval History 37 (2011), 162–79. See also: Laetitia Boehm, “Nomen gentis Normannorum: Der Aufstieg der Normannen in Spiegel der Normannischen Historiographie”, I Normanni la loro espansione in Europa nell’ alto medioevo (Settimane di studio del Centro italiano di studi sull’altomedioevo, 16, Spoleto, 1969), 623–704.



greek and latin sources

13

Calabria and Sicily and of his Brother Duke Robert Guiscard was completed in the closing years of the eleventh century.3 There are three surviving sets of annals from Bari, which are important for the study of eleventh-century southern Italy. The Bari Annals are a compilation of entries for the tenth and early eleventh centuries that conclude with a lengthier discussion of the Norman attack on Apulia in 1041–43. The Annals of Lupus Protospatharius have short entries for the years 855–1102, with the work providing useful chronological information for the career of Robert Guiscard. The so-called “Anonymous Bari Chronicle” is another, related set of annals which continues until 1118.4 Finally, we should include the work by Leo Marsicanus and the continuators of his Chronica Monasterii Casinensis, a chronicle that covers the history of Monte Cassino from the sixth century up to 1072.5 There is a scarcity of narrative sources in Greek that detail the gradual decline of Byzantine power in Italy and the eventual loss of the province of Longobardia between the 1040s and 1060s. Works by John Skylitzes, Ioannes Zonaras, and Kekaumenus offer next to nothing from the Roman perspective, confirming – once more – the strategic primacy of the operational theatre of Anatolia and the Balkans for the Byzantines, as compared to those of Italy and the Adriatic Sea.6 There are a few hagiographical sources in Greek concerning eleventh-century Byzantine Italy, like The Lives of St Philaretus the Younger and St Bartholomew the Younger, but the details they provide on military matters are minimal.7

3

Geoffrey Malaterra, The Deeds of Count Roger of Calabria and Sicily and of his Brother Duke Robert Guiscard, ed. and trans. Kenneth B. Wolf (Ann Arbor, 2005). The main secondary works on Malaterra are Ernesto Pontieri, I normanni nell’Italia meridionale (Naples, 1964); Chalandon, Domination normande, pp. xxxvi–vii; Wolf, Making History, pp. 143–71; Taviani-Carozzi, La Terreur du monde, pp. 17–20; Albu, The Normans in Their Histories, pp. 106–44. 4 This paragraph is based on Loud’s introduction to his modern translation of these texts, which can be found online: https://bit.ly/2KcEtD7. The Latin texts of both The Bari Annals and the work of Lupus Protospatharius can be found in Monumenta Germaniae Historica, Scriptores V (Hanover, 1844) pp. 52–63. The Anonymous Bari Chronicle can be found in Ludovico A. Muratori, Rerum Italicarum Scriptores (Milan, 1724), 5:147–56. 5 Chronica Monasterii Casinensis, MGH, SS, vol. 34. 6 John Skylitzes, A Synopsis of Byzantine History, 811–1057, trans. John Wortley (Cambridge, 2010); Kekaumenus, Strategikon, ed. Vasilij Wassiliewsky and Victor Jernstedt (1896, repr. Amsterdam, 1965); Ioannes Zonaras, Annales, ed. E. Weber, CSHB, vols. 41, 42.1; 42.2 (Bonn, 1841–97). 7 “De Sancto Philareto iuniore nili monachi sermo” and the “Vita S. Bartholomaei fundatoris magni monasterii Sancti Salutoris”, in Vitae sanctorum siculorum ex antiquis Graecis Latinisque monumentis, ed. Ottavio Gaetano and Pietro Salerno, 2 vols. (Panormos, 1647), 2:112–27 [St Philaretos] and 2:136–45 [St Bartholomew].

14

georgios theotokis

The Strengths and Weaknesses of the Principal Narrative Sources The first sign of insurrection against Byzantine authority after the arrival of the first Norman contingent in Italy was reported in the Capitanata, in northern Apulia, in the late spring of 1017. As Amatus of Montecassino notes of this initial stage of the Norman expansion, the Norman troops that were employed by the rebellious Lombard forces numbered only 250,8 with their role therefore undoubtedly being auxiliary. In order to quash this insurrection, the Byzantines received many elite units from the Balkans, which were led by the new Catepan, Tornikios Kontoleon. However, these troops did not arrive earlier than December (1017). Although William of Apulia talks about two battles in 1017, one in May – “a season most suitable for making war”9 – and another in late June, Amatus erroneously mentions five battles for the 1017–18 period, possibly because he noted certain small-scale skirmishes as battles in order to praise the Normans, reporting one at a place called Vaccarizza where, allegedly, out of three thousand Normans only some five hundred survived the onslaught.10 Regrettably, from the sparse chronicle information at our disposal it is impossible to reconstruct in detail the chain of events of these three battles and to understand the exact role played by the Normans in these military operations. Leo Marsicanus comments that Melus, who was the leader of the insurrection, recruited a large number of men “who either hated the Greeks or followed him personally”;11 hence, we presume that the Normans formed only a small but elite unit of the rebel Lombard army. Neither the Latin nor the Greek sources give us any hints about the topography of the battlefields or the numbers involved, except some general comments such as “a large Greek force”12 or “a great multitude of people”13 when referring to the armies under Byzantine command. We are also not able to piece together how the battles unfolded or to establish any battle formations, except to determine the outcome. The first major pitched battle of the Normans as an auxiliary force in Melus’ rebellion took place in early October 1018 at Cannae, on the right bank of the river Ofanto some ten kilometres from the Adriatic coast.14 The rebels had to face the regrouped Byzantine army under the command of a new general, Basil Boioannes, one of the most prestigious commanders that the Empire ever had in Italy, who commanded elite units from the mainland 8

Amatus, 1.22. William of Apulia, pp. 4–5. 10 Amatus, 1.23. 11 Chronic. Casin., cols. 628–29. 12 William of Apulia, pp. 4–5. 13 Malaterra, 1.21. 14 William of Apulia, p. 6. Lupus Protospatharius places it in October 1019 and not 1018. He is obviously mistaken, however: Lupus Protospatharius, a.c.1019. 9



greek and latin sources

15

15

and contingents of the famous Varangian Guard. Unfortunately, the role of the Normans in the battle is once again unknown, and the only information we get from Leo Marsicanus is that they suffered greatly, with only ten out of 250 surviving the battle.16 After Cannae, Basil Boioannes decided to consolidate his victory by building a series of fortified settlements on the north-western border of Apulia in order to control the passes leading to the mainland and the coastal cities; the latter included Troia, Fiorentino, Montecorvino, Dragonara, Civitate, and Melfi.17 Amatus gives a brief description of the fortifications of Melfi, along with the topography of the region; he reports that the settlement was “protected on all sides since it was situated on a high hill surrounded by several rivers”, although he indicates that “the perimeter wall of the town was not very high and that the defenders had to rely more on their machines and fortifications”.18 As Boioannes lacked experienced and loyal troops to man these castles, a Norman garrison under Byzantine pay, and of unknown numerical strength, was permanently established at the strong strategic fortress of Troia in 1019.19 In the aftermath of Cannae we also see the attempts by the Lombard principalities to absorb the independent coastal cities of Amalfi, Gaeta, and Naples, and the continuous struggle by the Lombard princes for preeminence over each other. As Amatus and William of Apulia note, many Normans under Rainulf, the future count of Aversa, were employed for siege operations by Gaimar III of Salerno and Pandulf IV. Other Normans were installed by the German Emperor Henry II at Comino, north of the Terra Sancti Benedicti, sometime in 1022 where, according to Leo Marsicanus and William of Jumièges, their ranks included leaders like Toustain (or Thorsteinn) of Begue, Gilbert, Osmund Drengot, Asclettin, Walter of Casiny, and Hugh Falluca.20 15

Chronic. Casin., col. 627. Our source differentiates between the Scandinavian element of the Varangian Guard and the Rus that had been in the service of the Byzantine emperors since the early tenth century: Georgios Theotokis, “Rus, Varangian and Frankish Mercenaries in the Service of the Byzantine Emperors (9th–11th c.) – Numbers, Organisation and Battle Tactics in the Operational Theatres of Asia Minor and the Balkans”, Byzantina Symmeikta 22 (2012), 126–56. 16 Chronic. Casin., col. 628. William of Apulia notes that after the battle of Cannae they [the Normans] were afraid of their numerous and powerful enemies; so small were their numbers”, p. 6. 17 Graham Loud, The Age of Robert Guiscard, Southern Italy and the Norman Conquest (London, 2000), p. 68. 18 Amatus, 2.19. 19 Francisco Trinchera, Syllabus graecarum membranarum (Napoli, 1865), 18:18–20. 20 Amatus, 1.31; The Gesta normannorum ducum of William of Jumièges, Orderic Vitalis and Robert of Torigni, ed. and trans. Elisabeth M. C. van Houts (Oxford, 1992–95), 7.30; Chronic. Casin., col. 632.

16

georgios theotokis

The numbers of men under Rainulf ’s command swelled following the death of Duke Robert of Normandy in 1035 and the minority years of his son William. A great opportunity presented itself in 1030, when Sergius IV of Naples installed Rainulf and his men as his suzerain in the newly fortified town of Aversa – some fifteen kilometres north of Naples – with the right to levy tribute on the surrounding fertile area.21 Even though the Normans had managed to acquire a permanent base in Italy, they were still working as mercenaries, and there must have been other independent groups living off the land in Campania – in fact, the chronicles of the monasteries of Montecassino and St Vincent on Volturno make frequent note of this, but they do not mention any specific names or places.22 William of Apulia suggests that after his establishment at Aversa Rainulf sent envoys to Normandy to “seduce” new recruits to the “beauties and fertility” of Apulia.23 In the twenty years between the battle at Cannae (1018) and their participation in the Byzantine expedition in Sicily (1038), the Normans took part in two major siege operations, outside Capua (1024–26) and Naples (1027), conducted by the joint forces of Gaimar and Pandulf. Yet they were still not the main players in the political upheaval against Byzantine authority in Apulia; rather, they were taking the side of the highest bidder, which – ironically – would include the Byzantines as well. Unfortunately, chronicle material about the course of both sieges is scarce. Leo Marsicanus reports that the siege of Capua ended in May 1026 and that it lasted for one and a half years, thus placing its beginning in the last months of 1024.24 Amatus also writes that Gaimar “took the city [of Capua] through the efforts of the citizens rather than by force of arms”.25 Finally, we are even less informed when it comes to the siege of the city of Naples by Pandulf ’s troops, which were led by Rainulf; any attempt to piece together the evidence about the role of the Normans in that operation is futile.26 During the reign of the Byzantine Emperor Michael IV (1034–41) a powerful military expedition was sent to Sicily to bring the island back under imperial rule and to re-establish “order” in the Tyrrhenian sea. It was launched in the summer of 1038 and led by the famous George Maniaces. This rather heterogeneous army consisted of troops furnished from the mainland, which probably would have meant a contingent consisting mainly of Greeks, Bulgarians, Vlachs, and units from the Varangian Guard, while auxiliary Italian troops would have also been raised from the theme 21

Amatus, 1.42; William of Apulia, p. 7. Chronic. Casin., cols. 653–54; Chronicon vulturnense del monaco Giovanni, ed. Vincenzo Federici, 3 vols. (Rome, 1925–38), 3.78. 23 William of Apulia, p. 7. 24 Chronic. Casin., col. 652. 25 Amatus, 1.34. 26 Chronic. Casin., col. 652. 22



greek and latin sources

17

27

of Longobardia. Furthermore, Constantinople called for its vassals in Italy to send reinforcements under William and Drogo Hauteville. Skylitzes is our best source for the military operations in Sicily until the stalemate in 1040, reporting the location of two pitched battles, and the number of casualties sustained by the Sicilians in the second (some five thousand men in all) and the number towns (thirteen) captured by Maniaces before his recall to the capital, accused of treason.28 We know nothing apart from the initial Byzantine success in taking Messina and the defeat of a Muslim army close to Syracuse that led to the capture of the city, with Malaterra reporting the exaggerated number of sixty thousand for the Arab besieging force. However, the role of the Normans is obscure in both Latin and Greek narratives of the campaign, although their small number – three hundred according to Amatus, five hundred according to Skylitzes, probably mounted men – would have confined them to small-scale operations of an auxiliary nature.29 Indeed, it is Malaterra who notes that the Normans were responsible for reversing the course of the siege of Messina by Maniaces’ troops when they counter-attacked “unusually fiercely” after the retreating “Greeks” left the battlefield to the enemy.30 The Norman establishment in the strategic fortress-town of Melfi in the Apulian–Campanian borders in 1041 was an event with profound long-term consequences for the political status quo in the area. Malaterra recorded that the Normans were only some five hundred strong when they established themselves there, and were heavily outnumbered by their enemies in the field, who were around sixty thousand (local Italian troops).31 In the short term, however, the Byzantine Catepan Michael Doukeianos reacted sharply and confronted the united Lombard–Norman forces in two pitched battles. Both battles, at Olivento (17 March 1041) and Ofanto (4 May 1041), ended with the Byzantine forces in shameful retreat. The failure to pacify the Lombards and restore order in Longobardia led to the replacement of Doukeianos with Basil Boioannes, the son and namesake of the former Catepan Basil. Let me underline once again that the role of the Normans in this early stage of their expansion in Apulia was auxiliary and they were far from playing a leading role in the political and military outcome of the 27

The sources are not clear about the origin of the units that were dispatched to Sicily under Maniaces: Malaterra, 1.7; 1.8; William of Apulia, pp. 7–8; Skylitzes, pp. 375–80; Amatus, 2.8; Chronic. Casin., cols. 669–70; Lupus Protosp., a.c. 1038. However, we can draw some parallels with the comments regarding the last dispatch of reinforcements that fought the Lombard rebels at Cannae in 1018. 28 Skylitzes, pp. 380–82. 29 Malaterra, 1.7; Amatus, 2.8; Chronic. Casin., col. 669; William of Apulia, pp. 7–8; Skylitzes, p. 400. 30 Malaterra, 2.7. 31 Malaterra, 1.7.

18

georgios theotokis

insurrection. The new catepan was swift to renew hostilities with the rebels, and the third and final battle took place at Montepeloso on 3 September of the same year, ending with the final defeat of the Byzantine army. William of Apulia recorded that the Normans had assembled at Olivento some seven hundred knights and five hundred (probably local Italian) foot soldiers, while the author of The Bari Annals provides the number of two thousand for the Normans and eighteen thousand for the “Greeks”; according to Amatus, the Normans were outnumbered by the forces under Byzantine command by, allegedly, one hundred to one.32 William is our main source for the battle at Olivento in March, mentioning the battle formations of the Norman army, which was deployed – apparently – in three acies, with the foot soldiers stationed at the flanks and their ranks “stiffened” by Norman knights, who presumably dismounted to fight their battle in the same fashion that would become prevalent in England and Normandy in the following century.33 William is also careful to mention that the cavalry in the centre was projected slightly forward, probably intending to deliver the first blow to the Byzantines, while the footmen were to protect their flanks from any outflanking manoeuvres by the enemy. Unfortunately, next to nothing can be pieced together about the course of this battle, or the ones that followed at Ofanto (May) and Montepeloso (September), except for the outcome, which was in favour of the “Gauls”. However, we do know that in September the new Catepan Boioannes received reinforcements from Sicily that included an elite regiment of the Varangian Guard.34 The author of The Bari Annals notes that many troops from the themes of “Anatolikon and Thrakesion, the Obsequiani, and Russians” perished at Ofanto in May, but I have doubts as to whether these troops would have reached Apulia so quickly after the first Byzantine defeat at Olivento in March.35 Rather, it is more likely that they fought at Montepeloso in September and the author was simply misinformed or confused. The same source puts the numbers of the opposing armies at Montepeloso at seven hundred for the Normans and ten thousand for the “Greeks” as “they [eyewitnesses?] say truthfully”.36 32

Amatus, 2.21; William of Apulia, p. 9; The Bari Annals, a.d. 1041. Stephen Morillo, Warfare Under the Anglo-Norman Kings, 1066–1135 (Woodbridge, 1994), pp. 150–74; J. F. Verbruggen, The Art of Warfare in Western Europe During the Middle Ages: From the Eighth Century to 1340 (Woodbridge, 1997), pp. 106–8. 34 Amatus, 2.26; William of Apulia, pp. 10–11. 35 The Bari Annals, a.d. 1041; Lupus Protosp., a.d. 1041. Skylitzes also reports on units of “Pisidians” and “Lycaonians” from the Anatolian themes of the Empire, although he is adamant that Boioannes did not receive any reinforcements from Sicily that summer, p. 401. 36 The Bari Annals, a.d. 1042. 33



greek and latin sources

19

We have scant information about the events that followed these three pitched battles in 1041. The Anonymous Bari Chronicle records a five-day siege of Bari in January 1043, which was led by Gaimar and included the Normans from Melfi. However, their attempt was swiftly abandoned,37 which is a striking indication of the Normans’ inability to conduct largescale siege operations against strongly fortified cities like Bari. Amatus briefly talks about the siege of Capua by Gaimar of Salerno in 1047, with the latter again having a number of Normans under his command.38 Amatus notes that by the middle of the century the Normans were such a vital asset to both Gaimar and Pandulf that “no one could drive him [the Count of Teano] out nor defend him without the aid of the Normans”.39 By 1047–48 almost the entire mainland area of northern and western Apulia belonged to the Normans, including Bovino, Lavello, Venosa, Montepeloso, and Materra, while in the next two years they began their incursions further to the south and east, reaching as far as Lecce and Scribla, an area where Robert Guiscard first established himself after his arrival in 1046/47. However, their overall numbers by the middle of the century remain unknown. The sources note that when they fought the Byzantines in 1041 they were between five and seven hundred, which is quite a reasonable number for two decades of almost continuous fighting and recruitment from parts of France. Nevertheless, the fact that many territories in the north and west of Apulia surrendered to William Hauteville does not necessarily imply that this was because of the Normans’ numerical strength or part of a well-thought plan. The Normans were still divided, with the two most powerful groups being those at Aversa and Melfi, and other smaller bands operating mostly in Capitanata and northern Campania. The best chance that the Byzantines and the papacy had to stop this systematic erosion of their territories by the Normans presented itself in 1053, in one of the most crucial confrontations in medieval Italian history. Pope Leo IX’s defeat at Civitate, apart from the obvious political consequences that it had on all the political powers of southern Italy, also opened the way to the Normans for further conquests in all directions. By the end of 1055 large areas of the “heel” of Otranto were under their strategic control. Once again, our main source for the events that unfolded on 17 June 1053 is William of Apulia, who refers to a “large force” of Italian troops that was supported by seven hundred Swabians (Amatus notes three hundred “Germans”) under a certain Werner and a certain Albert; the latter were clearly elite infantry who, according to William, were “not versed in horsemanship, who fought rather with the sword than with the lance”.40 Our 37 38 39 40

Anon. Bar., a.d. 1043; Chalandon, Domination normande, p. 107. Amatus, 3.4. Amatus, 3.5. William of Apulia, pp. 18–19; Amatus, 3.37.

20

georgios theotokis

sources also detail the names of the southern Italian leaders who actively participated in the anti-Norman coalition: Duke Atenulf of Gaeta, his brother Count Lando of Aquino, and the counts of Teano, Guardia, and Campomarino from the Biferno valley on the Adriatic coast.41 Faced with this threat, however, the Normans were also forced to unite. Humphrey of the Hautevilles had overall command of the army, succeeding his brother Drogo as leader of the Melfi Normans, and having with him Peter and Walter, the sons of Amicus, and the Hautevilles’ principal competitors, the Beneventan Normans, Gerald of Buonalbergo, Count Richard of Aversa, and Robert Guiscard from Calabria. William estimated their number at “3,000 horsemen and a few foot soldiers”, a force which seems reasonable for a coalition of different bands of Norman mercenaries fighting for their very survival.42 Both accounts by Amatus and William of Apulia agree that the Normans were suffering from hunger, and both also add that the Norman forces were in fact so lacking that they “by the example of the Apostles took the heads of grain, rubbed them in their hands, and ate the kernels”; some may have also cooked them over the fire beforehand.43 Both William and Amatus describe how the Normans divided their forces into three main divisions (acies): the centre commanded by Humphrey, the right wing by Richard of Aversa, and the left entrusted to Robert Guiscard. Opposite him, Humphrey had the Swabian infantry, which was used as a defensive shield against the Norman cavalry charge.44 Conversely, the Italians “stood all crowded together on the other side because they neglected to draw up a battle line in the proper manner”, thus resembling more of a mob than a disciplined army. Richard’s cavalry units directly attacked the Italians on the enemy’s left wing, which melted away and was pursued by the advancing Norman horsemen “like doves with a hawk in pursuit”. While the pursuit was in progress, the rest of the Norman cavalry had already engaged the enemy, who, according to William, chose to retreat, with the exception of the Swabians, who resisted vigorously and refused to leave their position. At this crucial point our sources report Richard’s return from the pursuit of the Italians to attack the Swabian infantry. Robert Guiscard is said to have completed the encirclement with an attack to their right flank.45 The sources are clear that, despite Norman inferiority in numbers as compared to the papal army, the key to victory lay in the use of their traditional heavy cavalry charge against a heterogeneous infantry army, like the papal one, which proves that even a heavily armed, well-trained, and disciplined unit of infantry cannot withstand/repel a sustained 41 42 43 44 45

William William William William William

of of of of of

Apulia, Apulia, Apulia, Apulia, Apulia,

p. 20. p. 19. p. 19; Amatus, 3.40. pp. 19–20; Amatus, 3.40; Chronic. Casin., col. 690. pp. 20–21; Amatus, 3.40.



greek and latin sources

21

heavy cavalry attack unless it is itself supported by units of archers and cavalry. In this case, the Swabians could stand their ground and deny the Normans their advantage in mobility and numbers, but as their flanks became exposed upon the retreat of their Italian allies, and because they lacked any archer or cavalry support, their encirclement and eventual annihilation became inevitable. Crossing over to Sicily, before the first major pitched battle between the Normans and the Arabs in the summer of 1061, Malaterra – who is our primary source for the Norman expansion in Sicily – reports that Roger Hauteville (Guiscard’s younger brother) had landed on Sicily for a reconnaissance mission with some sixty knights in the summer of 1060.46 For the main invasion operation in the following summer (1061), once again Malaterra notes that Roger had been given 160 knights, before being handed the task of taking the strategic port of Messina at the eastern tip of the island opposite the Calabrian coast with some three hundred knights.47 What is clear at this point is the lack of a sufficient number of troops for a major expansion outside the main theatre of operations, namely Apulia and Calabria. The first pitched battle in the Normans’ quest to bring the island of Sicily under their control took place at the heart of the island and close to the fortress of Castrogiovanni, on the banks of the river Dittaino (summer 1061). Malaterra mentions that the Norman army comprised seven hundred knights, while Amatus gives the number of one thousand for cavalry and the same for infantry; yet Guiscard would have undoubtedly left some of his men to garrison Messina, hence the number seven hundred could be closer to the truth.48 The Muslims allegedly had fifteen thousand horsemen and a hundred thousand infantry, the latter number given by Amatus surely being an exaggeration.49 Malaterra’s number of fifteen thousand for the Muslim cavalry corresponds to the number given by Amatus, but the former is also careful to distinguish between the local Sicilian Muslims and reinforcements that had arrived from Tunisia (“Africans”). The Normans did not deploy their army in three separate battalions forming one attack wave as in their victory at Civitate but, rather, in

46

Malaterra, 2.1. Geoffrey uses the number sixty twice in his first two books: once to identify the number of Slavs in Robert Guiscard’s service (1.18); and for the number of knights serving under Roger when the latter was sent to Calabria against “thousands of foes” (1.19). There is a correlation here with the biblical: “It is King Solomon carried on a throne, surrounded by sixty of Israel’s best soldiers” (Song of Solomon, 3:7). 47 Malaterra, 2.10. 48 Malaterra, 2.17; Amatus, 5.23. Guiscard’s seven hundred cavalry is also reported by Ibn al-Athir. See Michele Amari, Storia dei Musulmani di Sicilia (Florence, 1854), 5.IIIi, n. 1, p. 75. 49 Malaterra, 2.17; Amatus, 5.23; Chronic. Casin., 3.15, pp. 377–79.

22

georgios theotokis

echelons, with Roger chosen to command the first attack wave and Robert following him with the second. The Muslims had also formed three battle lines.50 Sadly, the course of the battle is unknown to us, but it is suggested that the Norman cavalry once again charged upon their enemies in their usual manner, forcing the Muslims to retreat to the castle of Castrogiovanni with some ten thousand casualties, according to Malaterra.51 The outcome of the battle, although it did not bring any significant territorial gains, was a tremendous morale boost for the Normans. Muslim reinforcements from northern Africa, accompanied by the regrouped Sicilian Muslims, marched against the Norman strongholds at the centre of the island in June 1063 and faced them on the banks of the river Cerami, some ten kilometres from Roger’s base at Troina. After a standstill of three days the Normans won a confrontation at the castle of Cerami, where Roger Hauteville’s nephew, Serlo, commanding only thirtysix knights, forced an enemy force of about three thousand cavalry men –a number surely exaggerated by Malaterra – and many infantry to retreat.52 After this initial success Roger’s force of a hundred knights engaged the enemy in two battles (vanguard and rear-guard). However, the regrouped Muslim army managed to repel the first Norman attack and move against the rear-guard, which was commanded by Roger. At this point, according to Malaterra, who is our only source for this battle, the divine intervention of St George along with Roussel of Bailleul’s exhortations saved the day for the Normans, who counter-attacked and forced their enemies to retreat. The last major pitched battle of the decade, fought against the Muslims, took place at Misilmeri (1068), some twelve kilometres south-east of the capital, Palermo. The information given by Malaterra is sparse, but we can reconstruct the main chain of events. After launching a plundering expedition in the Palermo area, Roger’s cavalry force came upon a sizeable mixed North-African and Sicilian army at Misilmeri, which was arranged in battle order, waiting for their arrival. We do not know the exact size of the two armies but, as usual, the Normans must have been outnumbered many times over. Roger did not hesitate this time, as at Cerami, and after arranging his army’s battle lines, and enjoying the element of surprise, launched an attack upon the enemy. Once again, the Muslims were unable to withstand a Norman cavalry attack, and Malaterra tells us that hardly anyone survived to carry the news to Palermo.53 The siege of Bari began at the end of the summer of 1068. None of the chroniclers provides an estimate of the numbers of either of the opposing armies, except that the duke “filled the sea with ships brought by the 50 51 52 53

Malaterra, 2.17. Ibid.; Amatus, 5.23. Malaterra, 2.33. Malaterra, 2.41.



greek and latin sources

23

Calabrians”, underlining the help provided by the Calabrian Greeks in the siege of the Apulian capital. The Byzantines were quick to understand that they were impregnable behind their high walls, as the city was “wealthy and strongly defended”, and thus they were reluctant to offer the Normans what they really wanted: to engage them in a pitched battle outside the city. Therefore, Guiscard ordered his fleet to block the entrance to the city’s port, as both of our main sources – William of Apulia and Malaterra – confirm.54 Malaterra also notes the formation of a sort of bridge linking the two sides of the port, with the ships being connected to each other by a large chain: “He had two bridges built, placing one on the shore on each side [of the city], projecting way out into the sea and attached to the ships on each flank by ropes, so that, if the Bariots should direct any attack against the ships, then his soldiers could bring them help speedily.”55 During the blockade of Bari, William of Apulia reports that the Normans deployed a number of siege engines “of different types”: “he [Guiscard] prepared a wooden tower to overtop the walls, on each side of which he placed stone-throwers, along with every sort of siege engine which might knock down the walls”.56 The chronicler also refers to an urgent mission that managed to slip out of Bari to Constantinople to request the help of Emperor Romanus IV.57 The siege continued for more than two and a half years, a clear sign of Guiscard’s decisiveness and ability to keep large numbers of troops in the field for long periods. However, the accounts of our sources provide no further details regarding the course of the siege, nor of any land or naval engagements between the two sides.58 Once Bari had capitulated on 16 April 1071, Robert Guiscard ordered his army to move to Reggio – the capital of Calabria – while his brother was already on his way to Sicily on a raid to divert the Muslims’ attention away from Palermo. We do not have any numbers, either for the Muslim garrison of Palermo or for the Norman besieging army, but we know that Guiscard ordered all of his troops that had taken part in the Bari campaign to follow him to Reggio, while he also asked for reinforcements from Bari and the surrounding areas.59 Furthermore, we know from the account of Amatus that Richard of Aversa also sent his son Jordan with two hundred knights to assist the duke in his operation against Palermo, as a sign of gratitude for the latter’s help in tackling a rebellion organised by William of Montreuil in the same year; however, this force turned back before reaching

54 55 56 57 58 59

William of Apulia, Malaterra, 2.40. William of Apulia, William of Apulia, Malaterra, 2.43. William of Apulia,

p. 27; Malaterra, 2.40. p. 27. p. 31. p. 33.

24

georgios theotokis 60

the rendezvous point. Amatus numbers Guiscard’s fleet as at least fifty (ten “gatti” and forty “other ships”),61 while Lupus Protospatharius takes it up to fifty-eight.62 This was an important increase on the Norman’s fleet capacity that probably came from all the captured Bariot ships, especially if we compare it with the naval operation at Messina only twelve years earlier, when Robert had only thirteen transport ships. Perhaps Guiscard had a plan to overcome Bari first in order to obtain more ships for the siege of Palermo. The mariners who manned the Norman ships were Calabrians, Bariots, and, according to William of Apulia, captive Greeks who were forced to serve in the fleet, while the marines were Normans.63 Malaterra also notes a strategic move by Robert, who, in order to secure his flanks from any relief army, installed a force of Norman knights in the region of Cerami and Castrogiovanni to harass the enemy; this force was later ambushed – and its leader, Serlo, killed – by a Muslim force of some seven hundred cavalry and two thousand infantry.64 The siege of Palermo lasted for five months and, although the chroniclers’ accounts are contradictory, it is clear that the city was blockaded by land and sea,65 and that there were sorties followed by sharp engagements between Norman and Palermitan detachments outside the city walls.66 As time went on, however, hunger and disease started to overcome the besieged population.67 William of Apulia refers to a naval engagement between the Normans and a mixed Kalbite–Zirid fleet of unknown size outside the port of Palermo, with the Normans inflicting several casualties on the Muslims and forcing them to retreat back into the port.68 Finally, Guiscard entered the city by applying a simple trick: he diverted the enemy to a particular spot in the city while an elite unit climbed the walls elsewhere. In the 1070s two sieges conducted by the Normans took place in Italy, one at Salerno in 1076 and the other at Naples a year later. At Salerno the Norman army was composed of “three different peoples, Latins, Greeks and Saracens, and he [Guiscard] ordered many ships to come to Salerno to guard the harbour”, although Amatus gives no specific number for these contingents.69 As in many other cases before, the main strategy was to force a blockade and defeat the enemy by famine and disease. However, the 60 61 62 63 64 65 66 67 68 69

Amatus, 6.11. Amatus, 6.14. Lupus Protosp., a.d. 1071. William of Apulia, p. 34. Malaterra, 2.46. Malaterra, 2.45; William of Apulia, pp. 34–35. Amatus, 6.16; Malaterra, 2.45; William of Apulia, pp. 33–36. Amatus, 6.17. William of Apulia, p. 34. Amatus, 8.14.



greek and latin sources

25

sources disagree on the way that the Normans took the city: Amatus notes that some Norman knights scaled the wall and exposed a part that was left unguarded, thus forcing an entry; William of Apulia highlights the deprivations inflicted upon the population of the city, which, in the eighth month of the siege, chose to capitulate.70 Following the taking of Salerno, Richard of Capua ordered his forces to assemble and march against Naples in the early spring of 1077, while asking Robert’s help for a naval blockade. According to Amatus, Guiscard sent help in the form of a naval squadron of unknown numerical strength from Amalfi and Calabria.71 We are informed that there were frequent attacks on the city by the Normans, both by land and by sea, which were repulsed, and equally numerous attempts by the defenders to counter-attack and face the Normans outside the city walls; Amatus even talks about two hundred marines caught by the people of Naples while sleeping in their ships.72 Even in this case, the Normans waited for famine to force the defenders to consider surrendering the city to them. Nevertheless, the siege was prolonged for many more months and it was Richard’s death that forced the Normans of Capua largely to abandon the operation in April 1078.73 Some conclusions Although less celebrated than Duke William’s famous conquest of England in 1066, the story of the Normans’ eleventh-century expansion into Italy and subsequent attacks on a teetering Byzantine Empire has long fascinated scholars. The first thing that strikes home with the narrative sources of the period in question is the imbalance in terms of their origin: the Latin sources, like William of Apulia, Geoffrey Malaterra, and Amatus of Montecassino, are by far the most detailed, as compared to the Greek ones, like Skylitzes, Zonaras, or Kekaumenus. This is not very surprising, considering that Italy was never a primary theatre of operations for the Byzantine governments (one exception is Constantine IV’s reign), especially during the turbulent period of political upheaval and court conspiracies that followed Basil II’s death in 1025 – thus confirming the tendency of Byzantine historiographers to take an interest mostly in the main political developments and military events that pertained to the capital and the central authority, and in the issues that posed a threat to the Empire’s existence.74 70

Amatus, 8.24; William of Apulia, p. 38; Romuald of Salerno, a.d. 1076; Lupus Protosp., a.d. 1077; Chronic. Casin., col. 779. 71 Amatus, 8.25. 72 Ibid. 73 Amatus, 8.32; Chronic. Casin., 3.45, p. 423. 74 Anthony Kaldellis, Streams of Gold, Rivers of Blood: The Rise and Fall of Byzan-

26

georgios theotokis

What exactly are the limitations of our primary sources for the Normans’ establishment in Italy and Sicily, and what challenges do they pose for a modern researcher? In many cases, several of the clerical sources give a narrative of battles, sieges, or entire campaigns that is incomplete simply because reporting these events – as a war correspondent would do today – was not their objective. While it is undoubtedly an oversimplification to say that every clerical source was ignorant of military affairs – bearing in mind the value of the accounts of Orderic Vitalis or William of Poitiers for AngloNorman history – the accounts of all our Latin sources for the period up to Robert Guiscard’s invasion of Illyria provide only the most basic outline of the battles and sieges of the period. The sole exception is William of Apulia’s description of the battle at Civitate, for which we are able to reconstruct the basic stages of the battle, including the formations of the opposing armies and their leaders. As a layman, William was very much aware of the importance of certain military factors in campaigns, such as the composition of forces, battle plans, and siege equipment as, for example, in the sieges of Bari and, later, Dyrrhachium. Other chroniclers, like Malaterra, Amatus, and Orderic Vitalis, rather composed heroic speeches made on the eve of battle – like Roger’s speeches at Cerami (1063) and Misilmeri (1068), written by Malaterra – no doubt to enhance the Norman avidas dominationis in the ears of their audience. The accounts of chroniclers who reported on major geopolitical events like battles and sieges would have been affected by five factors: (a) the chroniclers’ inherent tendency to exaggerate; (b) their reliance on oral testimonies, which always bore the risk of inflation and/or miscalculation; (c) the time when the chronicler was writing his/her work (keeping in mind that all three of our main Latin sources were writing in the second half of the eleventh century); (d) his/her experience in military matters (a dismounted knight might be counted as infantry by an inexperienced chronicler);75 and, finally, (e) his/her biases and sympathies. Therefore, the frequency and accuracy of the numbers provided by our chroniclers should be read with caution: although they provide a significant number of troop estimates for the military operations of the Normans in Italy and Sicily, these estimates tium, 955 A.D. to the First Crusade (New York, 2017), pp. 155–271; John F. Haldon, “Approaches to an Alternative Military History of the Period ca. 1025–1071”, in The Empire in Crisis(?): Byzantium in the 11th century (1025–71), ed. Vassiliki N. Vlyssidou (Athens, 2003), pp. 45–74; Michael Angold, “Belle Époque or Crisis? (1025–1118)”, in The Cambridge History of the Byzantine Empire, c.500–1492, ed. Jonathan Shepard (Cambridge, 2008), pp. 583–626. 75 The Strategikon (c.600) advises commanders to deploy their troops in varying divisions and spread their men, so that the enemy cannot simply calculate the size of the army by counting the number of divisional standards etc., Maurice’s Strategikon: Handbook of Byzantine Military Strategy, trans. George T. Dennis (latest edn 2010; Philadelphia, 1984), 9:102–3.



greek and latin sources

27

are not nearly enough for us to get a full picture of the events – not only because (medieval) people were notoriously bad at estimating large groups, but also because western European commanders were rarely concerned with accurate statistics and it would certainly have been exceptionally difficult for them to have a close idea of the effectives under their command (as is the case even for modern commanders). This is due to the fact that most feudal or pre-feudal armies did not have official establishments of the kind found in the Byzantine and Arab states;76 western European medieval armies simply consisted of whatever number appeared for any given campaign to fulfil their obligations to serve, or their contracts as mercenaries. A final reason why any medieval numbers should be viewed with caution includes the general tendency in our narrative sources to exaggerate their estimates of enemy armies in an apparent attempt to magnify their victories over those enemies, or to minimise criticism of a defeat. There is a remarkable passage in Ibn-Khaldun’s Muqaddimah (written in 1377) where the fourteenth-century Arab historian condemns the uncritical use of sources which produce gross exaggerations because of “the common desire for sensationalism […] which leads to failure to exercise self-criticism about one’s errors and intentions”.77 John France has compared the medieval armies of the high Middle Ages to “an onion”, in that they were formed by a professional “solid core” around which the rest of the force of levies was attached. Historical data shows that these professional soldiers – the royal and baronial familiae – rarely exceeded a few hundred (or, in the case of Edward I’s familia, less than a couple of thousand),78 and that these units certainly played a key role in the expansion of the Norman principalities in southern Italy between the 1040s and 1070s, and in the suppression of baronial rebellions in Apulia in 1067, 1072, and 1078; indeed, the speed with which Robert Guiscard reacted to the rebellions of 1067 and 1078/79 suggests that he mobilised his stipendiary troops, which, for the 1079 campaign against Geoffrey of Conversano, included less

76

Haldon, Warfare, State and Society, ch. 4; Warren Treadgold, “Standardized Numbers in the Byzantine Army”, War in History 12 (2005), 1–14; Clifford E. Bosworth, “Recruitment, Muster and Review in Medieval Islamic Armies”, in War, Technology and Society in the Middle East, ed. V. J. Parry and Malcolm E. Yapp (London, 1975), pp. 59–77. 77 Ibn-Khaldun, The Muqaddimah: An Introduction to History, ed. Franz Rosenthal and Nessim J. Dawood (Princeton, 1967), pp. 11–13; Hans Delbrück, Numbers in History (London, 1913), pp. 11–23. Delbrück’s methods have come under criticism by Bernard S. Bachrach, “Early Medieval Military Demography: Some Observations on the Methods of Hans Delbrück”, in The Circle of War in the Middle Ages, ed. Donald Kagay and L. J. Andrew Villalon (Woodbridge, 1999), pp. 3–20. 78 Michael Prestwich, Armies and Warfare in the Middle Ages: The English Experience (London, 2006), pp. 38–48. More numbers are provided in John France, Western Warfare in the Age of the Crusades, 1000–1300 (London, 1999), pp. 128–38.

28

georgios theotokis 79

than five hundred knights. Thus, we can assume with some degree of safety that whenever we come across estimates of numbers of elite troops (usually knights), these would have been stipendiaries (members of the familia or other bands of mercenaries) and they would, most likely, have operated in relatively low numbers of a few hundred.80 Contrary to this, other levies that formed around the core of a familia and accompanied them in campaigns – mostly sieges – would have increased the size of a medieval army exponentially and, consequently, would have been almost impossible to be reported by the sources in any way other than as “infinite” or “incalculable”.81 Therefore our narrative sources report relatively reasonable numbers for the Norman armies operating in Italy, Sicily, and – later – in the Balkans; but that is not the case when they describe the size of the armies of their enemies – the Byzantines, the Lombards, and the Sicilian Muslims. Although we know that the Normans would not have greatly exceeded five or seven hundred effectives in the first few “adventurous” decades in the South, the narrative sources are keen to report them as fighting (and prevailing) against overwhelming numbers of enemies: five hundred against sixty thousand “Italians” in 1041; seven hundred against fifteen thousand at Castrogiovanni in 1061; 136 knights against three thousand cavalry and “infinite” infantry at Cerami in 1063. According to Ovidio Capitani, Geoffrey Malaterra was undoubtedly the one who promoted the introduction of those clichés about Norman “prowess” universally accepted not only today, but also by their contemporaries.82 It was he and William of Apulia who did their utmost to create an overall “heroic” picture of the Norman Gesta in southern Italy. We may consider two factors that could help us to determine the possible upper limits of an army’s size. First, the saying that an army travels on its 79

Georgios Theotokis, The Norman Campaigns in the Balkans, 1081–1108 (Woodbridge, 2014), pp. 38–39; John France, Victory in the East: A Military History of the First Crusade (Cambridge, 1997), pp. 124–25. 80 Further examples include Guiscard’s sixty Slavs and Roger’s sixty knights operating in Calabria in the middle of the 1050s [Malaterra, 1.18 and 1.19]; the robbing of a group of rich merchants travelling from Amalfi to Melfi in 1057, which secured Robert the services of a hundred knights [Malaterra, 1.23–25]; and the three hundred juvenes led by Roger in a plundering expedition against Gerace, in Sicily, after the Norman victory at Castrogiovanni in 1061 [Malaterra, 1.21]. Malaterra uses the terms sui and fideles, instead of familia, but there can be no doubt that these were household troops. 81 For Guiscard’s invasion of Illyria in the summer of 1081, Malaterra describes the size of the Norman expeditionary force and makes a substantial distinction between the 1,300 “fully armed knights” and the “poorly armed mob” [imbecile vulgus] that made up the bulk of the army: Malaterra, 3.24. 82 Ovidio Capitani, “Specific Motivations and Continuing Themes in the Norman Chronicles of Southern Italy in the Eleventh and Twelfth Centuries”, in The Normans in Sicily and Southern Italy (Lincey Lectures 1974), ed. Ovidio Capitani et al. (Oxford, 1977), p. 6.



greek and latin sources

29

stomach has been true in any period in human history; hence, logistical constraints would have been as acute a factor as any in determining the size of a campaigning army, especially considering that in post-Civitate Italy the Normans were the aggressors and they would have undoubtedly felt very acutely the constraints of logistical management, namely the business of holding together and, above all, feeding a host on campaign.83 On top of that, we should also consider the socio-economic foundations of the operational theatres under discussion and the capabilities of these economies to support large expeditionary armies living off the land; Malaterra vividly reports the great famine and subsequent pestilence that swept the whole of the south of Italy in the spring of 1058, which was caused by the Normans’ harrowing of the land and their looting of the foodstuffs.84 It would have been in the interests of a medieval commander to raise a smaller number of good, wellequipped troops and get them into hostile territory as quickly as possible; therefore, the number of 160 knights for the amphibious invasion of Sicily by Roger in 1061 sounds reasonable enough, as well as his campaigning with no more than three hundred effectives in an island far away from his logistical bases in the south of Italy – a pool of troops which had to be replenished every spring due to combat losses, manning captured fortresses, and losing men to disease, famine, desertion etc.85 In comparison, in the numbers given by Amatus of Montecassino for Duke William’s invasion of England in 1066 our source reports the logistically unsustainable – if not impossible – one hundred thousand knights, ten thousand archers and “countless” foot soldiers accompanying the duke across the English Channel.86 The size of the terrain does not give a very detailed picture, as compared to other evidence, but the features of the battlefield – what modern histo83

John H. Pryor (ed.), Logistics of Warfare in the Age of the Crusades: Proceedings of a Workshop Held at the Centre for Medieval Studies, University of Sydney, 30 September to 4 October 2002 (Aldershot, 2006); Bernard S. Bachrach, “Logistics in Pre-Crusade Europe”, in Feeding Mars: Logistics in Western Warfare from the Middle Ages to the Present, ed. John A. Lynn (Boulder CO, 1993), pp. 57–78; John Haldon, “The Organisation and Support of an Expeditionary Force: Manpower and Logistics in the Middle Byzantine Period”, in Byzantium at War (9th–12th c.), ed. Nicolas Oikonomides (Athens, 1997), pp. 111–50. 84 Malaterra, 1.27. On the impact of campaigning armies on the early medieval economies, see Guy Halsall, Warfare and Society in the Barbarian West, 450–900 (London, 2003), pp. 119–32; Timothy Reuter, “The Recruitment of Armies in the Early Middle Ages: What Can We Know?”, in Military Aspects of Scandinavian Society in a European Perspective, A.D. 1–1300, ed. Anne N. Jorgensen and Birthe L. Clausen (Copenhagen, 1997), pp. 32–37. 85 Roger abandoned his wife at Troina (his forward base in central Sicily) in 1062 – after a failed uprising against his rule by the local population, supported by the neighbouring Muslims – in order to “set off for Calabria and Apulia to secure replacements for the horses which they had lost”: Malaterra, 2.31. 86 Amatus, 1.3.

30

georgios theotokis

rians have identified as “tactical geography” – allow us to evaluate information gathered from the narrative sources. Therefore, the second factor that we should consider when determining the size of armies is the topography of the battlefield, thus turning battlefield archaeology into an invaluable ally. The terrain could be used as a historical tool to deconstruct events, actions, and outcomes of military engagements; for example, the size of a battlefield – if we are able to establish precisely where it was – could allow us to check information gathered from the narrative sources and, possibly, eliminate inflated numbers such as twenty or forty thousand, although this method cannot be more specific.87 Regrettably, another major drawback in our chroniclers’ work is their lack of knowledge of the geography and topography of the regions where the events they describe took place. They were all relatively unfamiliar with the local geography of Apulia, Calabria, and Sicily, largely drawing their information from oral tradition and eyewitness accounts. For example, to Malaterra’s list of the Calabrian cities that surrendered to Robert Guiscard between 1057 and 1059 – namely Cosenza, Martirano, Nicastro, Maida, and Canalea88 – we can add his reports about several unidentified locations, like Narencio, Mount Nichifolo, Guillimaco, and Carbonaria.89 Malaterra’s narrative is largely characterised by vague comments like “He [Roger] traversed the highest peaks and the deepest valleys of Calabria,90 or “within a short period of time he [Roger] had secured the allegiance of eleven very strong fortresses”91 – referring to the period following the fall of Reggio. The same can be seen in Geoffrey’s second book, which deals with the period after 1041, where, once again, the narrative is full of unidentified places and even provides us with an anecdote as to how Messina got its name.92 In a similar fashion, Amatus’ accounts of battlefields and city fortifications are vague and inadequate. Unfortunately, there is no description of the fortifications of the Apulian, Calabrian, or Sicilian capitals and their surrounding regions, and not even a hint as to whether these had been recently rebuilt or reinforced to withstand any upcoming Norman siege. 87

Verbruggen, Art of Warfare, pp. 7–9; Russel S. Harmon, Francis H. Dillon III, and John B. Garver Jr., “Perspectives on Military Geography”, in Studies in Military Geography and Geology, ed. Douglas R. Caldwell, Judy Ehlen, and Russell S. Harmon (London, 2004), pp. 7–20; Peter Doyle and Matthew R. Bennett, “Terrain in Military History: An Introduction”, in Fields of Battle, Terrain in Military History, ed. Peter Doyle and Matthew R. Bennett (London, 2002), pp. 1–7; in the same volume: Kerry Cathers, “‘Markings on the Land’ and Early Medieval Warfare in the British Isles”, pp. 9–17. 88 Malaterra, 1.18. 89 Malaterra, 1.25; 1.26; 1.29; 1.33. 90 Malaterra, 1.21. 91 Malaterra, 1.36. 92 Malaterra, 2.1. More examples in: 2.4 [Clibanus]; 2.5 and 2.6 [Praoli and at the island of St Jacinto]; 2.14 [the “plain of Maniaces”]; 2.16 [St Felice].



greek and latin sources

31

One exception is the rather more detailed description of the fortifications of Melfi, a town less than 120 kilometres from Amatus’ birthplace, Salerno. Perhaps we can also add to this his topographical details about the plain of Naples, where Amatus reports on the “great swampy plain which produces many fruits”, while he also writes that the “level ground makes travel easy, and in many places there is running water”.93 Finally, William of Apulia’s account is even more disappointing and lacks even the most basic topographical information for the areas where major battles or sieges took place. His narrative simply provides us with the name of the location and often – although not always – that of a local river. Therefore, when it comes to scrutinising our narrative sources over their knowledge of local topography, we can say that it all comes down to the chroniclers’ eyewitness accounts and the places they had visited themselves, with any additional details about towns, castles, fortifications, or battlefields being quite unusual. Dating is another weak point in our primary sources for this period, with Amatus and William only rarely providing us with any dates in their accounts but, rather, with months of the year or seasons. The lack of dating or even the indication of seasons is a major drawback that affects the “histories” of this period. Other contemporary works that have a thematic rather than chronological structure include Orderic Vitalis and the Byzantine historians Skylitzes and Attaleiates, whose material is organised according to royal reigns. Loud argues that this kind of historical writing in the form of “histories” rather than “annals” “can pose problems for the reader seeking to integrate the account into a framework of contemporary events”94 – a nice way to encapsulate the pains of modern historians when faced with authors who do not follow a clear chronological pattern simply because this was not important to them. However, this view would not have been shared by the fourth-century sophist and historian Eunapius, who, at the beginning of his Universal History that continues the work of Dexippus, includes a detailed discussion of the perils of placing too great an emphasis on chronological specificity, concluding rather that absolute precision (dating by seasons or years) is disruptive, and opting for dating by imperial reigns instead.95 Conversely, Malaterra’s dating could be seen as his strong point, since he generally follows a neat chronological order and provides a date for all major events that he examines in his work. A case in point, where the problem of establishing precise dating for events such as sieges and battles becomes more acute, is Amatus’ narrative of 93

Amatus, 7.15. A History of the Normans, p. 36. 95 Dexippi, Eunapii, Petri Patricii, Prisci, Malchi, Menandri. Historiarum quae supersunt: Accedunt eclogae Photii ex Olympiodoro, Candido, Nonnoso et Theophane, et Procopii sophistae panegyricus, Graece et Latine, Prisciani panegyricus, CSHB, ed. Immanuel Bekker (Bohn, 1829), pp. 59–61. 94

32

georgios theotokis

the Norman victory at Castrogiovanni and the siege of Bari. In Book 5 of the History of the Normans Amatus devotes attention to Robert Guiscard’s invasion and early expansion on the island of Sicily in the 1060s, but it is only at the end of the book that we see a digression into the siege and capture of the Byzantine capital of Longobardia, Bari, between 1068 and 1071. In fact, there was a gap of almost seven years between the two events. However, this is not the only case of this kind of chronological digression in Amatus, with Loud highlighting several cases where the author completely disregards any chronological precision in his description of events.96 I suggest that Amatus is using a technique akin to what modern scholars call “narrative manipulation”, where an author regularly varies the narrative or chronological order to perform an analepsis, what we may call in simple terms a flashback, or a prolepsis, in equally simple terms a flash-forward. In his literary reading of the battles and sieges in Procopius’ Wars, Whately argues that this kind of narrative variation or manipulation can often be seen as a literary technique employed by skilful authors like Procopius, and it serves to shift emphasis from one side to another, highlight key actors and events as the author sees fit, look ahead or backwards in time, and, more generally, manipulate audience expectations by making the narrative more lively and interesting.97 Overall, one should not be too negative about how valuable the primary material such as Amatus’ History or Malaterra’s Deeds is to the modern historian of the Norman expansion in the Mediterranean. This chapter has brought out a number of “deficiencies” in the aforementioned material regarding the military operations of the Normans in southern Italy and Sicily: lack of army numbers, insufficient description of battles and army manoeuvres, scant dating of major (military) events, etc. However, before we rush to dismiss the value of our sources for the military history of the region in the eleventh century, we have to bear in mind that they, like their Muslim counterparts,98 were decidedly not professional military historians, and we should not expect them to provide us with any specialised military insight. Rather, we should bear in mind that “what is described in a battle description depends on unconscious cultural and conscious intellectual decisions about what is important to describe […] the way ancient authors describe the details of battle can tell us about the mental rigging of the societies in which they lived …”.99 Therefore, our medieval authors’ accounts 96

A History of the Normans, pp. 36–37. Conor Whately, Battles and Generals: Combat, Culture, and Didacticism in Procopius’ Wars (Leiden, 2016), especially pp. 1–37, 70–75, and 117–21. 98 For comparison, see Carole Hillenbrand, The Crusades: Islamic Perspectives (Edinburgh, 1999), pp. 31–33. 99 John E. Lendon, “The Rhetoric of Combat: Greek Military Theory and Roman Culture in Julius Caesar’s Battle Descriptions”, Classical Antiquity 18 (1999), 273–76. There is a recent monograph which evaluates Procopius as a military historian, 97



greek and latin sources

33

are invaluable to a (military) historian in a broader sense: they provide details of the tactics used by the Normans to subdue the local communities of Italy and Sicily, and the local urban and rural reactions to their methods; their interactions with the local Greek, Lombard, and Muslim populations; the political, social, and religious background of the different urban societies in southern Italy and Sicily; and the role of religious enthusiasm in the Norman conquests, especially in Muslim Sicily.100

emphasising the author’s classicising manner, and cultural adherents and approaches to battle: Whately, Battles and Generals, pp. 1–37. See also Kagan’s monograph, which focuses on the battle narratives of both Ammianus Marcellinus and Caesar, and examines the narrative techniques and their means of expressing causality: Kimberly Kagan, The Eye of Command (Ann Arbor, 2006). 100 Graham A. Loud and Alex Metcalfe (eds), The Society of Norman Italy (Leiden, 2002); Albu, The Normans in their Histories, pp. 106–44.

Fig 1:  The Hauteville siblings: a simplified family tree

2

“Conquest in Their Blood”: Hauteville Ambition, Authorial Spin, and Interpretative Challenges in the Narrative Sources*

Francesca Petrizzo

Fili denique Tancredi naturaliter [...] semper dominationis avidi [...]. (Geoffrey Malaterra, De rebus gestis, II.38)

The historian of the Norman conquest of the South finds in its three main chroniclers, Amatus of Montecassino, William of Apulia, and Geoffrey Malaterra, both a precious resource and a problematic one. All written between 1080 and 1100, and therefore within a short time of the completion of the conquest of the Mezzogiorno and during the last years of the conquest of Sicily, these sources present their readers with both the value of their testimony and its flawed origin. Written to bear witness to and, in two cases, openly celebrate the Norman takeover, the sources inevitably bear the imprint of their stated or unstated but self-evident goals, and using them for the purposes of military history demands a careful work of unpacking their themes, bias, and aim. This chapter will demonstrate how this can be done, by focusing as a case study on the sources’ treatment of the Hautevilles’ coming to southern Italy and to Sicily, and what this can tell us about the challenges of mining the chronicles for information about the military aspects of the Norman conquest. The choice of looking at the treatment of the Hautevilles in particular – as opposed to an examination of the depiction of the entire Norman *

This chapter is dedicated to the memory of my grandfather, Renzo Leporatti (1920– 2017). I would like to thank Prof. Graham Loud for his comments on an early draft of this work.

36

francesca petrizzo

invasion – has been made so as to overcome a particular difficulty that it is fundamental to keep in mind when examining these three chroniclers collectively: the challenges of reconciling these works’ differences from each other and of analysing them together. While the Ystoire de li Normant, Gesta Roberti Wiscardi and De rebus gestis Rogerii et Roberti were written during a relatively short time-frame, they appear to be wholly independent of each other: their transmission histories differ greatly and they are marked by distinct literary styles, aims, and geographical and ethnographical focuses. While their time and place of origin and theme make them indispensable for any study of the Norman conquest, they do not sit naturally next to each other as part of the same historiographical discourse. Nonetheless, it is possible to find a point of contact between them in their presentation of the coming to Italy of eight of the twelve sons of Tancred of Hauteville, an obscure Cotentin knight whose kin group came to dominate among the Normans in the South.1 This chapter will analyse how the three main chronicles depict the Hauteville conquest of the South, highlighting how they create a coherent picture of the sons of Tancred as a ragtag band of desperate people, lost in a hostile landscape, plagued by poverty and hunger, but nonetheless chosen by God to supplant, depending on the focus of the chronicle, the Lombards, the Byzantines, or the “Saracens”.2 Their portrayal is informed by the chroniclers’ depiction of the Normans at large as a naturally warlike, predatory, ambitious people whose grasping behaviour plagues Italy, but whose dominance is ultimately justified in God’s plan and earned through hard labour in the Southern wars. This depiction will be problematised by discussing the way in which the sources contradict themselves, by showing how they actually offer us information denying this image of hardship and labour – an image which can also be questioned by turning to the charter evidence, which shows the Hautevilles both successfully and quickly expanding their dominions, providing for their family, and marrying upwards from the very start into the Lombard aristocracy. I will then demonstrate how this influences the depiction of the military aspects of the Hauteville conquest and, especially, its hardships, using the conquest of Sicily as a useful yardstick against which to compare the depiction of the takeover of the mainland South. First I will highlight the differences between the sources, in order to differentiate between the tropes and aims employed by each one, and the 1

See fig. 1 for the Hauteville family tree. The term is here employed in quotation marks to reproduce the language of the Latin chroniclers, who often use it to refer to Muslim enemies of the Normans, in order to highlight the ideological implications of war waged against non-Christians. The loaded nature of the term has been discussed by Nicholas Morton in his work Encountering Islam on the First Crusade (Cambridge, 2016), pp. 15–19. 2



“ conquest

in their blood ” 37

way their styles affect their approach to military information. I will then move on to show their surprisingly coherent depiction of the Hauteville conquest; this will be followed by a problematisation and contextualisation of that depiction. Finally, the lines of argument will be drawn to show the ways in which we can unpack these narrative sources to provide us with their testimony for the purposes of military history, despite their part in legitimising and often whitewashing the Hautevilles’ role. Three Texts When dealing with the narrative accounts of the Norman conquest of the South it is first of all necessary to evaluate them not just as historical sources but as literary documents in their own right. The first layer of difficulty for the historian comes from their internal poetic and narrative coherence as works characterised by precise stylistic rules and literary goals which give differing slants to their treatment of the military aspects of the conquest. Amatus of Montecassino: Chanson in Translation Both the eldest and in many ways the broadest of the sources, Amatus presents us with an “insurmountable” difficulty: that of surviving only in translation.3 Written at the abbey of Montecassino by an erstwhile bishop and prolific author of works in prose and poetry, the Ystoire de li Normant enjoyed immediate influence over the Montecassino environment, and Leo Marsicanus (or Guido, if he was the author of this revision) integrated sections of it into the second draft of the Montecassino chronicle.4 However, the Latin version is lost: the work survives only in an early fourteenthcentury vernacular translation, one which we must suppose is fairly liberal, given the frequent and open interventions of the translator.5 The most recent edition, by Michèle Guéret-Laferté, has done painstaking work in differentiating between the translations and the interpolations; nonetheless, the true tone and style of Amatus’ work are inevitably lost to us.6 3

Corinna Bottiglieri, “Etnonimi e senso di identità nella Storia dei Normanni di Amato di Montecassino: alcune osservazioni”, in Sprache und Identität im frühen Mittelalter, ed. Walter Pohl and Bernhard Zeller (Vienna, 2012), pp. 219–337, p. 223. 4 Michèle Guéret-Laferté, “Introduction”, in Aimé du Mont-Cassin, Ystoire de li Normant, ed. Michèle Guéret-Laferté (Paris, 2011), pp. 9–230, 28–36; Graham Loud, “Introduction”, in Amatus of Montecassino, The History of the Normans, trans. Prescott Dunbar, rev., intr., and notes by Graham Loud (Woodbridge, 2004), pp. 20–22. 5 Guéret-Laferté, “Introduction”, in Aimé du Mont-Cassin, Ystoire de li Normant, 38–63; Loud, “Introduction”, pp. 23–36. 6 This edition will henceforth be quoted as Amatus, Ystoire, using the author’s Latin name, but respecting the fact that the Historia Normannorum is lost to us, and that the Ystoire is in many ways its own literary product.

38

francesca petrizzo

The work as it survives in translation fits thematically with the then developing trend of the chanson de geste genre, and it portrays the conquest as the heroic work of men chosen by God to rid Italy of unworthy peoples. Amatus also tells the most pious version of the descent of the Normans in the South: according to him, in 1017 forty Norman pilgrims on the way back from Jerusalem stopped at Salerno to defend it from Muslim attack, thus fulfilling their role as milites Christi and first demonstrating the benefits of Norman influence in Southern Italy.7 Apparently intended from the very start to cast the Normans in a positive light, Amatus’ text is immediately memorable for two different but complementary reasons: on the one hand, its sophisticated construction and, on the other, its dizzying flexibility in point of view. France’s estimation of Amatus’ chronicle as a work “with a problem at its heart” in terms of narrative is entirely fair: the Ystoire performs continuous, and apparently breezy, switches of sides.8 Dedicated to the immensely influential abbot Desiderius, architect of the at least partial conciliation between Montecassino and the Normans, the chronicle has at its heart the desire to restate Montecassino’s importance and praise those who pay rightful homage to it, while damning those who stand in its way.9 It is in the service of this purpose that the chronicle pays equal attention, down to the number of chapters dedicated to them, to Robert Guiscard and Richard of Capua, a Norman prince closely allied with Montecassino, seeking to present them as equal even after Guiscard’s establishment as de facto ruler of the Mezzogiorno.10 Indeed, the chronicle chooses as its ending point Richard’s death, finishing with a list of the gifts made by him and Guiscard to Montecassino, highlighting them as the worthy patrons who jointly rule the South and pay 7

Amatus, Ystoire, I. 17–18. John France, “The occasion of the coming of the Normans to Southern Italy”, in Latin Expansion in the Medieval Western Mediterranean, ed. Eleanor A. Congdon (Farnham, 2013), pp. 89–207, 98, first published in Journal of Medieval History 17 (1991), 185–205. The issue of the arrival of the Normans in the South is further discussed, with differing interpretations, by Graham Loud, The Age of Robert Guiscard: Southern Italy and the Norman Conquest (New York, 2000), pp. 60–67 and H. Hoffmann, “Die Anfänge der Normannen in Süditalien”, Quellen und Forschungen 49 (1969), 95–144, who argue that this happened after Muslim attacks on Salerno in the 990s, and by Elisabeth van Houts, who supports Raoul Glaber’s account, which makes the case for the intervention of duke Richard and his relationship with the papacy (Van Houts, “Quelques observations sur des liens entre la Normandie, l’Angleterre et l’Italie au début du XIe siècle”, in 911–2011, Penser les mondes normands médiévaux. Actes du colloque international de Caen et Cerisy (29 septembre–2 octobre 2011), ed. David Bates and Pierre Bauduin (Caen, 2016), pp. 129–46). 9 Amatus, Ystoire, Dedication, pp. 235–38. 10 Guéret-Laferté, “Introduction”, in Aimé du Mont-Cassin, Ystoire de li Normant, pp. 74–85. 8



“ conquest

in their blood ” 39

homage to the abbey, despite the disparity between them. This is achieved by a painfully narrow focus on Campania: while dedicating a few chapters to the conquest of Sicily and mentioning Guiscard’s preparations for his Balkan campaigns, the work, which concludes before Guiscard’s death, is chiefly concerned with Montecassino’s immediate area of influence and the relationship between the Normans and Lombards. It is in order to justify the Normans’ supplanting of Lombard power and fighting off papal influence that Amatus achieves truly stunning narrative evolutions: first underlining the greed of the Lombard princes, but acknowledging their prestige and secular importance when they marry into the Normans; bemoaning Norman greed and pitying the abbey that suffers it, then acknowledging God’s will in the defeat of the pope at their hands; quickly turning from praising Gisulf of Salerno, Guiscard’s brother-in-law, to tarring him with the blackest of brushes when the two clash and Gisulf is removed from power.11 It is in attempting to maintain a balance between Richard and Guiscard that Amatus’ most subtle work is performed: glossing over the self-evident greed in their frequent rivalries in order to praise them once again as Montecassino’s proudest sons.12 Continually concerned with Montecassino’s good, and the ultimate picture of the Mezzogiorno as having achieved balance under its new rulers, Amatus does not conceal facts as much as keep them in reserve until they are needed: thus his initial picture of the Normans selflessly fighting against the “Saracens” and serving Guaimar of Salerno out of love for a good prince (as opposed to the sinisterly portrayed Pandulf IV of Capua) is belied when he candidly admits that the Normans eventually rebel against Atenulf, Lombard ruler of Benevento, when he ceases the cash payments he has clearly been making all along, and that Guaimar had indeed hired them.13 This has an inevitable influence on Amatus’ portrayal of military affairs: armies can shrink or increase in number, and famine become prosperity, according to the usefulness of the moment, as I shall describe below. At once subtle and candid, ruled by the overwhelming need of Cassino and its patrons, Amatus’ Ystoire is one of the best-informed but most subtly manipulative of the accounts; something which lends unique tints to his portrayal of the Hautevilles. William of Apulia: Medieval Epic We know very little of William of Apulia: he may have been a monk, just as easily a layman, was most probably French but not definitely Norman.14 He wrote between 1095 and 1099, and his work is dedicated to both Roger Borsa, 11 12 13 14

Amatus, Ystoire, I.34, 43–34; IV.19; III.19, 39–41; IV.23, VIII. Amatus, Ystoire, VIII.35–36. Amatus, Ystoire, II.27; III.25. Marguerite Mathieu, “Introduction”, in Guillaume de la Pouille, La geste de Robert

40

francesca petrizzo

Robert Guiscard’s son and duke of Apulia, and to Pope Urban II.15 Unlike Amatus’ expansive Ystoire, his Gesta Wiscardi has an overtly narrow focus: the life and conquests of Robert Guiscard. To do this, William highlights primarily Apulia, and the Normans’ relationship with the Byzantine Empire; some coverage is given to the conquest of Sicily, but here the narration is concluded by the conquest of Palermo in 1072, after which Guiscard largely ceased to concern himself with the enterprise directly.16 William’s work is a poem in epic hexameters, whose style is not always brilliant, but constantly informed by the most refined Latin literature.17 William lends the classical gloss liberally to his verses, employing classical names for peoples (Galli for the Normans, for example, a useful two-syllable alternative which can serve well his poetic scansion needs).18 Unlike Amatus, William is comfortable in depicting the Normans as taking part in southern Italian wars without the outright character of a holy fight: his Normans join Melus’ rebellion, rather than rescuing Salerno from the Muslims.19 TavianiCarozzi has argued that this denotes William’s greater comfort – writing as he does fifteen years after Amatus – with a less glorious origin tale for Norman power in southern Italy; but I would rather argue that this fits in with the overall epic and heroic theme of the poem.20 The Gesta Wiscardi is driven forward by people’s quest for a worthy leader, and what a worthy leader inevitably does to push his people to expansion. While the most prominent such leader is Guiscard, he is by no means the only one: his son Bohemond, Melus himself, and Romanus IV Diogenes of the Byzantines are all described as worthy leaders of men, whose military prowess and sense of justice bring their people to victory and happiness.21 William finds himself making much of Roger Borsa’s brave behaviour during a storm at sea and then fighting while wounded, attempting to endow him with martial qualities in the absence of outright military victories, and the Guiscard, ed., trans., comm. and intr., Marguerite Mathieu (Palermo, 1961), pp. 2–96, 17–25. This edition will henceforth be referred to as William of Apulia, Gesta Wiscardi. 15 William of Apulia, Gesta Wiscardi, “Prologus”, p. 98. 16 William of Apulia, Gesta Wiscardi, III, lines 340–45. 17 Armando Bisanti, “Composizione, stile e tendenze dei Gesta Roberti Wiscardi di Guglielmo il Pugliese”, Archivio Normanno-Svevo 1 (2009), 87–132. 18 Marguerite Mathieu, “Introduction”, in Guillaume de la Pouille, La geste de Robert Guiscard, 66–70. 19 William of Apulia, Gesta Wiscardi, I, lines 15–34. 20 Huguette Taviani-Carozzi, “Le mythe des origins de la conquête normande en Italie”, in Cavalieri alla conquista del Sud: Studi sull’Italia normanna in memoria di Léon-Robert Ménager, ed. Errico Cuozzo and Jean-Marie Martin (Roma, 1998), pp. 57–89. 21 William of Apulia, Gesta Wiscardi, I, lines 55–56, acknowledges Melus as the first leader of the Normans in Italy; book V goes into detail about Bohemond’s guerrilla warfare in the Balkans at the head of part of Guiscard’s army; for Diogenes, III, lines 10–100.



“ conquest

in their blood ” 41

rebellion of Diogenes against his stepchildren is justified by the fact that he is a brave warrior, while they are indolent rulers.22 In this perspective, the Normans joining the fight to support the unjustly deposed Melus are fighting an honourable fight; one which echoes throughout the work. For a work dedicated to Urban II, the poem contains very little in terms of references to and discussions about holy war: the action is driven forward by the ambition and desire to own southern Italy first, Sicily and the Balkans later, of the great warriors, both Norman and Greek, who crowd the scene.23 This inevitably dictates the pacing and the attitude to war shown throughout the narration: William can go into the details of military tactics when he chooses to, as he does for example with the siege of Bari, or of brilliant stratagems, as when Guiscard passes himself off as dead in order to infiltrate a fortress (in itself a well-established trope); but we must perceive these as reflections on the good leader’s strategic genius and, as such, individual virtues subject to the suspicion of flattery.24 Much as Amatus’ greatest caveat comes from his fondness for seeking balance at all costs between the different forces at play in his narration, William’s narration is ruled by the overwhelming need to represent the war for the Mezzogiorno as one between the competing military drives of the chiefs of the peoples involved – and thus a highly personal struggle in which different men of competing virtue vie for the ultimate prize. Geoffrey Malaterra: Family History As highlighted in the introduction to this chapter, one of the most immediate difficulties in comparing the three narrative sources for the Norman conquest of the South lies in their different focus. Despite their differences, Amatus and William still overlap in their dealing with the conquest from the start. Geoffrey narrows the focus to a personal history of conquest on the part of the Hauteville family. Writing for Roger I of Sicily at the end of the eleventh century, Geoffrey was a Benedictine monk who had recently come down to southern Italy from beyond the Alps, and was possibly, but not certainly, himself Norman.25 From a conventional prologue to his work in which he excuses himself for his rustic style, he then moves to pen a lengthy and apparently 22

William of Apulia, Gesta Wiscardi, V, lines 156–76; III, lines 80–120. Book III, lines 198–201, makes the most explicit, yet still very brief, reference to fighting for the faith in the Sicilian theatre, and as done by Roger, very much a minor character in the Gesta Wiscardi. 24 William of Apulia, Gesta Wiscardi, II, lines 480–542; II, lines 338–52. 25 Marie-Agnès Lucas-Avenel, “Introduction”, in Geoffroi Malaterra, Histoire du Grand Comte Roger et de son frère Robert Guiscard, ed. and trans. Marie-Agnès Lucas-Avenel (Caen, 2016), pp. 19–25. Lucas-Avenel, as Malaterra’s editor, is an important voice of dissent against his possible Norman origin: see “Le recit de Geoffroi Malaterra ou la légitimation de Roger, Grand Comte de Sicile”, Anglo-Norman Studies 24 (2011), 169–92. 23

42

francesca petrizzo

well-informed history of the descent of the Hautevilles into Italy, beginning with an account of Tancred of Hauteville and his numerous family, then dealing with Guiscard, and finally Roger and his thirty-year conquest of Sicily.26 While Orderic Vitalis’ mention of Malaterra’s chronicle suggests that he was at least known in Normandy, we have no certainty that he used or saw his text.27 While the style of the work, inspired by the Roman historians in general and Sallust in particular, makes it at a first reading the most “friendly” and familiar to a modern reader as a historical text (excepting the occasions when the chronicler lapses into poetry, which underline the literary aspects of the work), its focus both severely limits and sharply slants its outlook. While it ostensibly deals with the deeds of both Robert and Roger, it clearly caters to its patron more than to his elder brother.28 The conquest of Sicily presented unique difficulties and required a continuous effort over more than thirty years; this meant that both the enterprise and its principal actor, Roger, appeared seldom in the chronicles of Amatus and William, which were mostly concerned with dealings on the mainland.29 The reverse is true of Malaterra’s work: constantly following the evolving conquest of Sicily, it rarely strays beyond Calabria, Roger’s base on the mainland. This offers a completely different perspective on Guiscard’s reign: as the story of the often embattled but ultimately collaborative enterprise of two brothers, one of whom had carved himself a firm position on the edges of the family dominions while the other’s ambitions brought him into the Byzantine Empire, it describes the Hauteville expansion as a family endeavour, mostly detached from other participants. While Amatus dwells on the tension between established Lombard power and Norman newcomers, and William makes much of the epic struggle between Norman 26

Two editions of the text have been used here: for books I and II, Geoffroi Malaterra, Histoire du Grand Comte Roger et de son frère Robert Guiscard. Vol. I. Livres I & II, ed. and trans. Marie-Agnes Lucas-Avenel (Caen, 2016); and for books III and IV Gaufredo Malaterra, De rebus gestis Rogerii Calanroae et Siciliae Comitis et Roberti Guiscardi ducis fratris eius, ed. Ernesto Pontieri (Bologna, 1928). The second volume of the LucasAvenel edition is not yet published; hence both will be shortened to the Latin title De rebus gestis. 27 Orderic Vitalis, Ecclesiastical History, ed. and trans. Marjorie Chibnall (Oxford, 1969–80), II, p. 100. 28 Malaterra, De rebus gestis, Epistola. 29 Amatus, Ystoire, VI.21 does report Guiscard endowing Roger with most of Sicily, but he had previously put the accent on Geoffrey Ridel, a non-Hauteville Norman, V.18. William, Gesta Wiscardi, III, lines 194–203, acknowledges Roger’s prowess and general virtue and his dedicating his youth to the conquest of Muslim Sicily, but he is not particularly interested in it; and while Malaterra (De rebus gestis, III.42) puts great stress on the promise made by Roger to Guiscard to look after Borsa’s succession, William of Apulia depicts the duke entrusting his heir to Robert of Loritello and “his dear friend Gerard” (Gesta Wiscardi, IV lines 190–200).



“ conquest

in their blood ” 43

and Byzantine warriors, for Malaterra the Hauteville conquest hinges on complex kin dynamics of cooperation and betrayal, on the background of the relentless, piecemeal, and fatiguing conquest of the island of Sicily. This flattening of the Norman invasion to the Hautevilles alone makes it at once more intimate and more visceral as a narration, often a borderline civil war between brothers, or father and son in the case of Roger and his ambitious bastard son Jordan.30 The consequences for the usefulness of the De rebus gestis for military matters are stark: while Malaterra, as we shall see below, builds a believable picture of the techniques and aims of the conquest of Sicily, his dealing with military help, food and material supplies, and alliance and reward are all contingent on his ultimate aim to show Roger as both the conqueror of Sicily and the often slighted but ultimately faithful underdog in his dealings with Guiscard. Amatus, William, and Geoffrey wrote remarkably different, complementary but distinct accounts of the conquest, with distinct narrative styles, aims, and points of view. Such accounts converge, however, in one matter: the sheer difficulty of the enterprise for the embattled, outnumbered, often starving but ultimately victorious Hautevilles. Eight Hungry Normans Kenneth Wolf and Emily Albu, in their analyses of Norman historiography, make a case for covert criticism in even the most apparently favourable of sources for the Norman conquests in Europe and southern Italy.31 By using a complex reference system woven out of classical quotations and ancient historical tropes, they show how even those historians who seem closest to the Norman point of view appear to consider them an example of greedy, ferocious, quintessentially violent people, thus fundamentally injecting their praise with an undercurrent of pessimism.32 It is their contention – which Wolf bases on an examination of the Southern sources alone, while Albu ranges across the entire spectrum, both geographical and literary, of Norman historiography – that by referring to authors such as Lucan, Sallust, and Tacitus the Norman historians intended to present their subjects as both dangerous and incorrigible, and their advent as essentially negative.33 Here, however, I would put forward the case for a different interpretation of the 30

Malaterra, De rebus gestis, I.23–25; II.23; III.36; IV.16. Kenneth Baxter Wolf, Making History: The Normans and Their Historians in EleventhCentury Italy (Philadelphia, 1995); Emily Albu, The Normans in Their Histories: Propaganda, Myth, and Subversion (Woodbridge, 2001). 32 Wolf, Making History, cf. chs. 4, 5, 6, Conclusion; Albu, The Normans in Their Histories, cf. chs. 3, 4. 33 See footnote 30; interestingly, Albu does not make use of Malaterra. 31

44

francesca petrizzo

reasons and ends for their employment by the chroniclers of the Southern conquest, especially in reference to the Hautevilles. In his article about the origins of the myth of the gens Normannorum, Loud acknowledges that there is surprising consistency throughout the sources for the Norman conquest and their portrayal of the invaders as a violent, warlike, but providential people whose arrival has been predisposed by God and whose victory is inevitable.34 I propose that we need to see such arrival not only as divinely ordained but also as essential to the development of Italy: in other words, that the Normans were not only instruments of God but agents for positive change. The vision of “barbaric”, more “primitive” people as a refreshing change from older, established, but more “corrupt” civilisations was well established in classical Latinity: thus for republican writers Rome itself had triumphed over effeminate peoples, and would lose its original honesty at its peril; and imperial historians such as Tacitus would use the barbaric “innocence” of the Germans as a foil for the perceived decadence of their own times.35 Medieval writers had explicitly done this as well: thus Paul the Deacon, in his history of the Lombards, cast them as the thunderous, troubling, but necessary coming of a people to replace the dying Romans.36 Amatus, William of Apulia, and Malaterra, all of them cultured and accomplished Latin writers, unsurprisingly appear to have used the same model; and if we consider that they, and their colleagues from France and England, tapped into the same reservoir of classical tropes, it is hardly surprising that the resulting portrait of the Normans as a people should appear coherent. Thus we see Amatus describe the Normans as “riding around happily”, “joyously plundering”, with an odd sort of innocence well reconciled to his vision of them as “incapable of restraining themselves, as other people do”.37 This chimes with William of Apulia’s description of the Normans: a fierce people from the North, who have come to conquer in France but then descend to Italy, drawn to promises of reward, keen for riches and gifts in a land supposed to be wealthy, whose strength and hope lies in the discord of others.38 In comparison to the Lombards, “greedy for dominance, constantly squabbling with each other”, whose fall from grace is embodied 34

Graham Loud, “The Gens Normannorum: Myth or Reality?”, Anglo-Norman Studies 4 (1982), 104–16. 35 See the complaint raised by Sallust in his work about the corruption of the late Republic in the De bello Iugurthino in the Loeb Classical Library volume Sallust, with a trans. John C. Rolfe (Cambridge MA, 1985) and Tacitus’ Germania in Cornelius Tacitus, ii.2: Germania, Agricola, Dialogus de Oratoribus, ed. Erich Koesterman (Leipzig, 1970). 36 Paolo Diacono, Storia dei longobardi. Testo latino a fronte, trans. Antonio Zanella (Cividale del Friuli, 1991). 37 Amatus, Ystoire, II.20; III.7–8–9; III.18. 38 William of Apulia, Gesta Wiscardi, I, lines 1–10, 28–40, 162–64.



“ conquest

in their blood ” 45

by the despicable Pandulf of Capua, Amatus frames the Normans as the people destined by God to rule over that part of Italy, and his description of the warring principalities is corroborated by the Gesta Wiscardi; and while William of Apulia does acknowledge the valour of certain Greek commanders, his depiction of Byzantine palace intrigue makes it easy to see how Guiscard’s victories were “fated by God”.39 The undeniable negative aspects of Norman rule, such as upheaval and looting, are ascribed to innate Norman greed; but Amatus shows his comfort with unlikely agents of God’s will by once upholding the Muslims as such, thus making a Christian people like the Normans infinitely more palatable in his narration.40 For William, Norman greed and Norman warlike prowess exist as part of a manifest destiny that is slowly revealed: thus for him the Normans, at first unaccustomed to victory or battling at sea, progressively grow into their role, learning how to triumph and coming to use bolder tactics.41 The Hautevilles are the perfect representatives of the people thus described: for Malaterra, “conquest comes naturally to them”, and they are impatient at the prospect of removing to Italy, outright mutinous if denied; for Roger, once safely established in Calabria, it is unthinkable to gaze at Sicily and not immediately want it; his bastard son Jordan, constantly ambitious, chafes at the bit; and his brother Guiscard seeks to expand his dominions beyond the Adriatic.42 The natural ambition of William’s Guiscard leads him first to fight and nearly murder his brother Humphrey, and then to break his word and usurp rule from Humphrey’s son Abelard, obtaining a papal investiture as duke in the meanwhile; and while in the narration this is all rather glided over as the natural evolution of a younger brother who rises to his rank as leader of the Normans in the South, it shows a constant, barely contained ambition.43 Amatus’ Guiscard, too, is an indefatigable conqueror: 39

Amatus, Ystoire, I.35–41, IV.35, and III.38, where this morale is unsubtly revealed in a dream; William of Apulia, Gesta Wiscardi, eloquently describes the evils of Italy under the constantly skirmishing Lombard princes in book I, lines 135–64, and the long and cruel ballet of power between emperors in Byzantium and commanders in Apulia dominates the background from the end of book I through to the end of book III. William’s complex, fraught attitude towards the Byzantine empire was analysed by Paul Brown in his “The Gesta Roberti Wiscardi, A Byzantine History”, Journal of Medieval History 37 (2011), 162–79. 40 Amatus, Ystoire, III.16, 23; V.8. 41 William of Apulia, Gesta Wiscardi, I, lines 2–5, 520–23 for God’s plan for the Normans and their beginning to place their hopes in the sons of Tancred, and lines 290–91 for the end of the Normans’ fear of fighting the Greek; II, lines 34–37 for the turn of the wheel on Tancred’s sons, followed by the raising of the Normans’ hopes after unexpectedly defeating the Germans at lines 284–85; at III, lines 100–5, the divine call for the Normans to open the way to the Holy Sepulchre is made explicit, and they begin to gain confidence in naval warfare at lines 132–38. 42 Malaterra, De rebus gestis, II.1, 28; IV.2; III.13. 43 William of Apulia, Gesta Wiscardi, II, lines 309–19, 400–5, 451–53.

46

francesca petrizzo

preceded by his brothers, William and Drogo, whose enterprises range from Campania to Sicily, he follows in their footsteps and builds his own power base, eventually coming to be referred to purely as “le duc” (which we can imagine to be the translation of a Latin dux).44 This insistence on the warrior prowess of the Normans in general and the Hautevilles in particular makes the chronicles rich in military detail, each in its own way: thus Amatus of Montecassino describes many of the Normans’ raids at the expense of Lombard and Muslim cities, and their destruction of the livelihood of others; William of Apulia dwells on the brilliant tactics and occasional stratagems of Guiscard, and the ability of his son Bohemond in guerrilla warfare in unconquered territory, and on more prosaic but no less useful details such as specific techniques to refloat beached boats; the second, third, and fourth books of Malaterra’s chronicle are a long narration of a progressive, exhausting, but eventually successful war of attrition against the Sicilian Muslims, and can linger over technical detail on naval warfare, or the peculiarity of the homing pigeons of the armies of Palermo for the communication of war news.45 But the description of the Hautevilles as sheer war machines, albeit chosen by God, would of course not be enough to redeem them fully in the chronicles dedicated to them, and here a fundamental and profoundly deceitful aspect of their description comes into play: the shading of their depiction as violent conquerors through highlighting the suffering endured by them in the accomplishment of their warlike deeds. If the Hautevilles have conquest in their blood, they also shed a lot of it in its pursuit and achievement. A vivid physical quality brightens the description of the descent of the Normans into Italy for both Amatus and William: a painful quality of fear, loneliness, and loss. Defeated in many battles, the Normans have “nowhere to go”; few and alone, hopeless and leaderless, they loot, not just out of greed, but from sheer, immediate, terrible hunger.46 They are constantly outnumbered: against the Byzantines; against the Calabrians; against the pope; against the emperor; at least at the start, against the Muslims.47 They are a ragtag band, driven by an unholy talent for slaughter, but who pay a terrible price for it. 44

Amatus, Ystoire, II.46 to IV.20, when he receives the princess of Salerno Sichelgaita as bride “like a prince”. 45 Amatus, Ystoire, for example II.20, III.10, V.10, VI.4; William of Apulia, Gesta Wiscardi, for example IV, lines 378–424 for the sudden victory at Durazzo, V, lines 50–79, V, lines 235–54; Malaterra, De rebus gestis, for example II.8, 42, III.1, 15, 31, IV.2. 46 Amatus, Ystoire, I.33, III.40, IV.1; William of Apulia, Gesta Wiscardi, I, lines 104–17, II, lines 138–41, 180–82; Malaterra, De rebus gestis, I.21. 47 Amatus, Ystoire, I.22, 29, II.21, III.39; William of Apulia, Gesta Wiscardi, I, lines 93–94, 254–59, II, lines 142–53; Malaterra, De rebus gestis, I.19.



“ conquest

in their blood ” 47

Once more, the Hautevilles are the most emblematic of the Normans: if their countrymen have suffered in the conquest of southern Italy, they too have done so, as much and more. Thus Amatus describes the untimely death of William Iron-arm, early Hauteville hero; the death of Drogo, betrayed and assassinated; Humphrey perishing of disease.48 William accompanies him, and both writers dwell on the suffering of Robert Guiscard at the hands of his own family as well as those of his enemies.49 Norman greed can separate brothers as well: thus Robert, who finds his brothers unwilling to give him land, is forced into the mountains to live like a bandit, stealing, kidnapping, looting.50 But if we are to believe Malaterra, once Robert himself was established, he did the same to Roger: and thus we find the youngest Hauteville stealing horses, begging for land, his pacts constantly betrayed by his older brother, alone, ragged, friendless, and unhappy.51 While the Hautevilles always appear to eventually come together (Humphrey reconciling with Guiscard on his deathbed, and Guiscard and Roger eventually reaching a good compromise), their greed separates them; their wars impose a physical toll on them; and in general they seem to struggle on as unlikely heroes in a panorama of hostility and physical discomfort.52 While the Normans are shown setting fire to harvests and taking cities by starving them, they seem to endure blows as bad as the ones they have inflicted.53 The Hautevilles head the charge against the enemies of the Normans, but are also the target of their most ferocious hatred; they fully establish Norman dominance in the South, but they are betrayed by their own people; they come to Italy “as soon as age allows”, but sometimes perish as soon as they get there.54 Their children, nephews, younger brothers, in-laws all take part in the strife; and often they die in it, leaving behind bitter mourning and violent vengeance as the only sop to the pain, for, as Amatus describes, the Normans do not weep, but set off to avenge their dead.55 While they do eventually achieve the highest honours, pain is never far from the Hautevilles, and the writers who celebrate them are keen to highlight it. 48

Amatus, Ystoire, II.35; III.19–22; IV.2. William of Apulia, Gesta Wiscardi, II, lines 20–26, 286–93, 364–80 see the emotionally charged demise of William Iron-arm, Drogo, and Humphrey. 50 Amatus, Ystoire, II.46, III.7–11; William of Apulia, Gesta Wiscardi, lines 297–363. William puts more of a straightforward heroic spin on Guiscard’s origins, immediately showing him as cunning and resourceful in the face of hardship. 51 Malaterra, De rebus gestis, I.15, 23, 25–26, II.23. 52 William of Apulia, Gesta Wiscardi, II, lines 364–82; Malaterra, De rebus gestis, I.29, II.43. In the latter chapter Guiscard is anguished by the fact that Roger is his last surviving brother, whom he does not want to lose at any cost. 53 Amatus, Ystoire, VI.4, III.40; William of Apulia, Gesta Wiscardi, II, lines 115–19. 54 Malaterra, De rebus gestis, I.11, 15. 55 Malaterra, De rebus gestis, I.12–13, II.23, II.46; Amatus, Ystoire, III.31. 49

48

francesca petrizzo

In this way, each according to their own narrative canon, the three writers appear to show us a remarkably coherent and ultimately positive picture: that of the Hautevilles as “good barbarians”, but also good leaders; the ones incited by an innate, unbeatable ambition and greed, but also chosen by God to accomplish His will in southern Italy; the heralds of a people talented in war, but that has to fight overwhelming odds; the bearers of famine, but also its sufferers. In an overall balancing act, the chroniclers seem to present the Hautevilles as the complex, not-wholly positive, but divinely appointed heroes of a quest for which they paid their price in blood, a military conquest fraught with logistical issues but ultimately successful, for “Victory in war is not in numbers, or horses, or armies, or weapons, but with him to whom Heaven grants it”.56 But can we accept such a view? Four Problems Part of the importance of the three main narrative sources for the Norman conquest resides in their unique status: we have few competing sources, both narrative and documentary, for the years of the first occupation, which means that their approach to information is hard to check with external testimonies. Nonetheless, this part of the chapter will highlight how the scant but existing charters for this period, and the internal contradictions of the sources themselves, can help us to clarify and deepen the information available on four specific aspects of the Norman conquest discussed above: numbers and support structure, isolation, food and supplies, and finally the uniqueness of the Normans’ situation in the Mezzogiorno. The idea of the Normans as few, desperately outnumbered, and unlikely victors is fundamental to their sympathetic presentation in the narrative sources. For Amatus, the ranks of Byzantines first, Germans later, who fight the Normans are numberless, swarms of bees; William’s Normans descend into battle even though they are few; Roger’s army at Cerami is outnumbered by thousands of Muslim warriors.57 Numbers in a battle are fundamental to describing the size and strength of the participants; by highlighting the paucity of the Norman forces, the historians make their ultimate victory more improbable and divine favour more evident. On the one 56

William of Apulia, Gesta Wiscard4i, II, lines 146–7. See n. 46; Malaterra, De rebus gestis, II.32. For an analysis of numbers in the narrative sources, see Graham Loud, “Betrachtungen über die normannische Eroberung Süditaliens”, in Forschungen zur Reichs-, Papst- und Landesgeschichte: Peter Herde zum 65. Geburtstag von Freunden, Schülern und Kollegen dargebracht, ed. Karl Borchardt and Enno Bünz (Stuttgart, 1998), pp. 115–31, cf. 123–24, reprinted in Conquerors and Churchmen in Southern Italy (Aldershot, 1999). 57



“ conquest

in their blood ” 49

hand, that the Normans were not many, and not all Normans, is as far as possible concrete fact: Ménager’s inventory of the Normans in the South mentions only a few thousand men, many of whom were probably not in Italy in the time of the earliest battle, and many of whom did not come from Normandy.58 The second point is acknowledged by William of Apulia, who describes the Normans replenishing their ranks by taking in any stray man.59 That the “Norman” invaders were therefore not an overwhelming force seems to be quite likely. But that the odds they faced should be so dire is in doubt. This is something which we can glean by comparing the way the battles for southern Italy, Sicily and the Balkans are treated in Amatus and William, and how Cerami stands apart from the rest of the conquest of Sicily in Malaterra. While Amatus waxes poetic about uncounted multitudes for the clashes between Normans and papal or imperial forces, his estimates for landing parties and raids into Sicily become suddenly sober, mentioning precise, believable forces such as eighty-five knights, two hundred Muslims, and similarly low, compelling numbers, accompanied by solvable supply problems, in his sporadic but recurring mentions of the Sicilian campaign.60 The same goes for William of Apulia when he is talking about Sicily (and thus we have the engaging, detailed, lengthy but unexaggerated description of the siege of Palermo) or the dogged but ultimately not imbalanced guerrilla warfare conducted by Bohemond in the Balkans.61 Here the difference in theatre is fundamental: the Normans as Christian warriors fighting the “Saracens”, or taking war beyond the lines of the by then heretical Byzantine Empire, are one thing; quite another are the interloping Normans, crashing into a theatre where they seek to supplant long-established powers out of nothing but ambition. We might expect the difference between the two campaigns to be more pronounced, and for the chroniclers to make more hay out of the Hautevilles’ “holy wars” against non-Christians and heretics, but here again we run up against a problem: the innate Hauteville greed. In Malaterra, who as a chronicler principally devoted to the conquest of Muslim Sicily should have an easier time selling the idea of the heroic Hautevilles, we still find the taint of Hauteville ambition to be explained away. Far from waging righteous holy war against all Muslims, Roger often cooperates with them; he first invades with Guiscard at the explicit invitation of one Muslim warlord, Ibn al-Thumna, thus fighting for him as his brothers had fought 58

Louis-Robert Ménager, “Appendice. Inventaire des familles normandes et franques émigrées en Italie méridionale et en Sicile (XIe–XIIe siècles)”, in Roberto il Guiscardo e il suo tempo. Atti delle prime giornate normanno-sveve, 1973 (Bari, 1991), pp. 260–390. 59 William of Apulia, Gesta Wiscardi, I, lines 165–68. 60 Amatus, Ystoire, V.15–18; VI.15–16. 61 William of Apulia, Gesta Wiscardi, III, lines 204–339, beginning of book V.

50

francesca petrizzo

with Lombard princes; he arranges exceptions for those Muslims who want to convert to Christianity to keep their unlawful wives; while he employs shock-and-awe tactics, he does not lose himself in needless slaughter, and, more than anything, he literally cannot bear the thought of peace.62 The Christianisation of Sicily is a by-product of Roger’s continuous, “bee-like” activity in the service of his own expansion, and not his primary aim.63 Holy war in the Hauteville dominions seems to be a collateral result rather than an intended goal. For Roger to come through fully as the hero of the piece, despite the realistic piecemeal campaigns described by Malaterra, he still needs to be given credit for one desperate but ultimately successful pitched battle; hence Cerami, which may well have been atypical of the rest of the campaign in style of combat, but whose description as a glorious stand sits at odds with the slow but constant progress made by Roger in the rest of the work.64 The idea that the Hautevilles were desperately isolated in their new land is also fairly easily disproved. In all three narrative sources the descent of the numerous sons of Tancred is a progressive achievement over possibly as much as twenty years, but Malaterra offers a possible explanation: the men discussed among themselves who would go and who would stay, and they would head south as soon as they came of age.65 Going further, he describes those who leave as offering to look after the children of those who stay, a fact borne out by Serlo, son of Serlo, serving with his uncle Roger in Sicily.66 While it is true that resources were not always plentiful in the South, and that when Drogo refused Guiscard land he did so not out of spite but because he genuinely did not have any to give, Italy appears to have been an appealing prospect for all of them and, far from being isolated, they had some form of continuous communication with the motherland. Drogo felt confident enough in his position in southern Italy to found an abbey, Venosa, in which he and his brothers were to be buried.67 Self-evidently successful in southern Italy, the brothers had a close enough relationship with their family of origin to arrange for the coming not only of their male siblings as they came of age, but of the female ones too. We find Fressenda, daughter of Tancred, married to Richard of Aversa, thus joining the Hautevilles with the first Norman kin group to establish itself in the 62

Malaterra, De rebus gestis, II.3–4, II.14, IV.6, II.41. Malaterra, De rebus gestis, II.32. 64 Malaterra, De rebus gestis, II.33. 65 Malaterra, De rebus gestis, I.4, 5, 11. 66 Malaterra, De rebus gestis, I.11, II.5. 67 Recueil des actes des ducs normands d’Italie, ed. Léon-Robert Ménager (Bari, 1980), n. 1, 20–22, edits the forged foundation charter for Venosa, which still attests to the perceived importance of this family foundation, see Hubert Houben, Die Abtei Venosa und das Mönchtum im normannisch-staufischen Süditalien (Tübingen, 1995), p. 137. 63



“ conquest

in their blood ” 51

Mezzogiorno; and a nameless sister of theirs married to a Roger, other names unknown, by whom she mothered Geoffrey of Conversano.68 Organising the betrothal and passage south of women of marriageable age was no mean feat of both communication and security; it is hard to imagine the Hautevilles doing so in a panorama of isolation, devastation, and overwhelmed numbers. Nor were their sisters alone in getting married: their brothers immediately, and efficiently, began marrying into the Lombard aristocracy. Little attention is paid to this by Malaterra, whose subject, Roger, conducted a completely separate marriage campaign with fellow Normans first, northern Italians later; and William of Apulia understandably emphasises Guiscard’s marriage to Sichelgaita, given his choice of focus.69 But Amatus, as part of a separate strategy to legitimise Hauteville domination in the South, shows us how they immediately married into the Lombard dynasties: William Iron-arm, Drogo, Humphrey, and William the Younger all married into the houses of the dukes of Sorrento first, and of their cousins, the princes of Salerno, later.70 While these marriages probably originated in the Lombards’ desire to tie themselves to the mercenary forces of the Normans – and only with Sichelgaita and Guiscard can we see the full affirmation of the Hautevilles as a kin group with recognisable land-holdings and inherited titles validated by an external authority and not simply by their own sheer force of arms – the early dates of their first Lombard alliances demonstrate that from the very beginning the Hautevilles were reasonably safe bets for advantageous kin unions, an image hardly reconcilable with that of desperate brigands.71 It is undoubted that the conquest of the South was hardly a walk in the park: just as the Normans burned harvests as a strategy of war, they were 68

Richard Bünemann, Robert Guiskard, 1015–1085. Ein Normanne erobert Süditalien (Köln, 1997), p. 247; Wolfgang Jahn, Untersuchungen zur normannischen Herrschaft in Süditalien, 1040–1100 (Frankfurt am Main, 1989), pp. 234–35, Amatus, Ystoire, II.45. 69 Roger married first Judith of Évreux, then Eremburga of Mortain, and finally Adelaide del Vasto; for William’s treatment of Robert’s marriage to Sichelgaita, and the legitimacy this conferred on him, see Gesta Wiscardi, II, lines 430–41. 70 Amatus, Ystoire, II.29, 35, III.34, IV.22. Huguette Taviani-Carozzi, in her La principauté lombarde de Salerne (9e–11e siècle): pouvoir et société en Italie lombarde méridionale (Rome, 1991), controversially presents this as a form of adoption into the kin group (cf. 686–720). 71 For further discussion of Hauteville marriage strategies, and Lombard–Norman intermarriage in the Mezzogiorno, see Catherine Heygate, “Marriage Strategies among the Normans of Southern Italy in the Eleventh Century”, in Norman Expansion: Connections, Continuities and Contrasts, ed. Keith J. Stringer and Andrew Jotischky (London, 2013), pp. 165–86, Aurélie Thomas, “La carrière matrimoniale des fils de Tancrède de Hauteville en Italie méridionale: rivalités fraternelles et stratégies concurrentes”, in Les stratégies matrimoniales (IXe–XIIIe siècle), ed. Martin Aurell (Turnhout, 2013), pp. 89–100 and Francesca Petrizzo, “Band of Brothers: Kin Group Dynamics Among the Hautevilles and Other Noble Norman Families in the Mezzogiorno and Syria, c. 1030–­c. 1140”, PhD thesis (University of Leeds, 2018), pp. 109–35.

52

francesca petrizzo

sometimes denied pay; their frequent raids bespeak their need of supplies; supplies, and the best way to procure them, are the object of numerous, detailed, and seemingly plausible chapters in Malaterra.72 Given the unstable nature of Drogo’s rule, and his inability to give Guiscard land and support – beyond pointing him in the general direction of as-yet-unconquered Calabria – Guiscard’s earliest times in Italy were not easy.73 But, far from being unheard-of-hardship, this seems to have been sound Hauteville family strategy: Robert of Loritello, Guiscard’s “beloved” nephew, encountered hard times and needed help in his occupation of the Abruzzi, but this, happening as it did as a later date under the aegis of established Guiscardian power, is given very different treatment in Amatus.74 Ultimately, the Hautevilles grappled with each other in a continuous mechanism of what we might call embattled negotiation: Roger, stranded in Sicily where supplies were scarce and he had to walk back from battle carrying his saddle after his horse had been killed, probably suffered from Guiscard reneging on his promise to give him warhorses, and Guiscard did not enjoy coming down to Italy and finding that he had to fend for himself.75 Both seem to have been determined to obtain as much as possible from their cooperation with their family, while at the same time carefully treading a line in which none of their rebellions ever went too far for reconciliation to be possible, in order to find a better deal. So it was with Roger and his ambitious firstborn, Jordan: well into the 1080s, the father appears to not have trusted his son, forbidding him from entering any fortified city during his absence from Sicily, while at the same time putting him at the head of the army he left to guard the island.76 The frontier of Hauteville expansion required self-evident, continuous, and delicate renegotiation of the power balance both within and outside the family. But, despite the chroniclers’ keenness to show their almost hopeless difficulties, if we combine the description of their embattled life with their successful bid for Sicily, their seemingly continuous contact with home, the marriages they arranged for their sisters, the bases they were able to provide for their brothers and nephews, and the prestigious marriages they themselves managed to achieve early on, we are led to a different picture of the struggles they encountered during the conquest of southern Italy – one that has a noticeable impact on our use of the narratives of their coming as sources for military history.

72

Malaterra, De rebus gestis, I.27; II.31–32, 35; III.14. Amatus, Ystoire, II.45; III.7–8. Drogo’s succession by acclamation to William Ironarm suggests still a kind of military dominion rather than a territorial one. 74 Amatus, Ystoire, VII.25, 30. 75 Malaterra, De rebus gestis, I.25;, II.28–31. 76 Malaterra, De rebus gestis, IV.16. 73



“ conquest

in their blood ” 53

Conclusion: Three Texts, Revisited Amatus of Montecassino, William of Apulia, and Geoffrey Malaterra provide both invaluable and problematic testimonies for the military historian of the Norman conquest of the South. They write richly textured, detail-crowded histories of the way the Normans arrived in the Mezzogiorno, the battles they fought, and the alliances they wrought. In this endeavour they provide us with in-depth testimony of details of the fighting, as varied and complex as naval technology, siege strategies, and stratagems. However, all three sources require several layers of interpretative work before they are usable for historical purposes: the refined products of well-educated writers, all dedicated to precise patrons, they carry with them both the inevitable slant of the authors’ bias and the stylistic and narrative constraints and twists of their accomplished literary styles. This chapter has sought to analyse them collectively by focusing on their treatment of the Hautevilles’ arrival in the South, an intersecting subject on which the three texts provide a surprisingly coherent perspective as a wrought and hopeless enterprise eventually validated by God’s plan for the Normans in Italy. Through an analysis of the sources’ internal contradictions and the contribution of the documentary sources, this chapter has problematised the issue, seeking to achieve a more plausible view of the event and to show a way in which to approach the narrative sources’ authorial and stylistic bias for the purposes of compiling a military history of the Normans in the South. Far from proving an insurmountable obstacle, the sources’ spin and perspective has proved to be a valuable key to historical work: another way into the exploration of the unique challenges, and ideological choices, of those writers who sought to legitimise the unexpected and successful military conquest of the Hautevilles.

3

“The Arts of Guiscard”: Trickery and Deceit in the Norman Conquests of Southern Italy and Outremer, 1000–1120

James Titteron

The title of this chapter is taken from Ralph of Caen’s spurious account of how Tancred, nephew of Bohemond of Taranto, evicted Raymond of SaintGilles’s men from the citadel of Antioch after the capture of the city from the Turks during the First Crusade in 1098: So when these things were being done, discord stirred up the servants of Tancred and Raymond; soon it rose from the servants to their lords. Tancred barely, oh barely, restrained his passions, so that he would not appease his anger with the slaughter of the Provençals. But reason came to him, that he should forbid any Christian blood to be spilled: better, he advised himself, to have recourse to the arts of Guiscard, by which that glorious man became known to the 1 whole world.

Wearing swords under their cloaks, Tancred and his men entered the citadel under a pretence of peace (the guards being unaware of the hostility that had arisen between them and their lord) and drove the guards out.

1

Ralph of Caen, Tancredus, ed. Edoardo D’Angelo (Turnhout, 2011), p. 84: “Haec dum ita geruntur, Tancredi Raimundique ministros discordia agitat; moxque a ministris surgit ad dominos. Vix, ah, uix compescit Tancredus animos, quin Prouincialium strage iram leniat. Sed occurrit uiro ratio, quae sanguinem uetet fundi Christianum: melius ipsa ad Wiscardi monet artes recurrere, per quas orbi gloriosus innotuit.” All translations are the author’s unless otherwise noted.

56

james titterton

This story is almost certainly an invention.2 At best, it is a confused version of events recorded by other chroniclers of the First Crusade. The anonymous author of the Gesta Francorum said that the Turkish commander of the citadel received Raymond’s banner as a token of surrender following the crusaders’ victory over Karbughā, the Turkish ruler of Mosul, on 28 June 1098, only to exchange it for Bohemond’s when he returned to the city that same day.3 Raymond of Aguilers, whom one would have expected to have recorded such an outrage against his patron, Raymond of Saint-Gilles, simply said: [Bohemond] violently expelled the duke’s [Godfrey of Bouillon] men, as well as those of the count of Flanders and the count of Saint-Gilles, saying that he had sworn to the Turk who handed over 4 the city that he alone would possess it.

Alternatively, Ralph of Caen may have invented the whole story in order to give Tancred, the subject of his chronicle, a more prominent role in the narrative. Yet the very fact that he chose to invent a story that emphasised Tancred’s cunning and mastery of deceit, what he termed artes Wiscardi, illustrates the high regard in which the Norman aristocracy held these qualities. Such seemingly underhand behaviour was worthy of memorialisation, even if it never actually happened. When Norman and Anglo-Norman historians described the characteristics that typified the Norman gens, cunning and an aptitude for deception were often included alongside military might, greed, and a predilection for fighting among themselves. When imagining William the Conqueror’s deathbed speech, Orderic Vitalis called the Normans a people who “covet rebellions, who long for insurrections, and are eager to commit every sin”.5 According to William of Malmesbury, “they are a people accustomed to war, and almost unable to live without battle, they actively run to engage their enemy, and where strength has not succeeded, they corrupt them with 2

For the division of Antioch after the siege, see Thomas S. Asbridge, The Creation of the Principality of Antioch 1098–1130 (Woodbridge, 2000), p. 36; John France, Victory in the East: A Military History of the First Crusade (Cambridge, 1994), pp. 310–11; Steven Runciman, The First Crusade (Cambridge, 1980), p. 158. 3 Gesta Francorum et aliorum Hierosolimitanorum, ed. Rosalind Hill (London, 1962), p. 71. 4 Raymond of Aguilers, “Historia Francorum qui ceperunt Iherusalem”, in Recueil des historiens des croisades: Historiens occidentaux tome troisième (Paris, 1866), pp. 231–310, quotation at p. 262: “Etenim consequenter homines ducis, et Flandrensis, et comitis Sancti Aegidii, violenter de castello expulit, dicens se jurasse illi Turco qui civitatem reddidit quod ipse solus haberet eam.” 5 Orderic Vitalis, The Ecclesiastical History, ed. Marjorie Chibnall, 6 vols. (Oxford, 1969–80), 4:82: “rebelliones enim cupiunt, seditiones appetunt, et ad omne nefas prompti sunt”.



“ the

arts of guiscard ” 57

6

trickery and wealth”. Geoffrey Malaterra, writing at the behest of Roger I of Sicily, presented a mixture of the Normans’ positive and negative qualities: “They are a most astute people, vengeful of injuries, despising their ancestral fields in the hope of winning more elsewhere, greedy for profit and domination, pretenders and dissemblers in everything, holding a certain balance between liberality and avarice”.7 Scholars have long been aware of this trope, but little has been written about its manifestation in chronicle narratives.8 The major exception is Emily Albu’s article on ruses in Anna Komnene’s Alexiad. Albu argued that Anna Komnene “cannot help but admire the genius for deception” that the Normans exhibited because the Byzantines and Normans possessed a “shared love of theatrical wiles”.9 Not only does Anna Komnene record the ruses that her father, Alexios I Komnenos, employed during his military career, she also relishes stories of Norman ruses, particularly the account of how Bohemond of Taranto escaped from Antioch in September 1104 to seek aid in western Europe. According to Anna Komnene, in order to reach Rome without being intercepted by the Byzantine fleets, Bohemond circulated a rumour that he had died. Bohemond then had himself carried to Soudi, Antioch’s port, in a coffin, which was loaded onto a ship. While the ship was out at sea Bohemond walked about the deck, but when they approached land he would lie in the coffin, holding a dead rooster to simulate the smell of a putrefying corpse, while his sailors made a show of mourning their loss.10 This story is 6

William of Malmesbury, Gesta regum Anglorum, ed. R. A. B. Mynors, R. M. Thomson, and M. Winterbottom, 2 vols. (Oxford, 1998), 1:461: “Gens militiae assueta et sine bello pene uiuere nescia, in hostem impigre procurrere, et ubi uires non successissent, non minus dolo et pecunia corrumpere.” 7 Geoffrey Malaterra, De rebus gestis Rogerii Calabriae et Siciliae comitis et Roberti Guiscardi ducis fratris eius, ed. Ernesto Pontieri (Bologna, 1927), p. 10: “Est quippe gens astutissima, injuriarum ultrix, spe alias plus lucrandi patrios agros vilipendens, quaestus et dominationis avida, cuiuslibet rei simulatrix ac dissimulatrix, inter largitatem et avaritiam quoddam medium habens.” 8 For studies on the concept of the gens Normannorum, see Emily Albu, The Normans in Their Histories: Propaganda, Myth, and Subervision (Woodbridge, 2001); R. H. C. Davis, The Normans and their Myth (London, 1976); Ewan Johnson, “Origin Myths and the Construction of Medieval Identities: Norman Chronicles 1000–1100”, in Texts and Identities in the Early Middle Ages, ed. Richard Corradini, Rob Meens, Christina Pössel, and Philip Shaw (Wien, 2006), pp. 153–64; Graham Loud, “The ‘Gens Normannorum’ – Myth or Reality?”, in Proceedings of the Battle Conference on Anglo-Norman Studies IV, ed. R. Allen Brown (Woodbridge, 1982), pp. 104–16; Nick Webber, The Evolution of Norman Identity 911–1154 (Woodbridge, 2005). 9 Emily Albu, “Bohemond and the Rooster: Byzantines, Normans, and the Artful Ruse”, in Anna Komnene and her Times, ed. Thalia Gouma-Peterson (Abingdon, 2014), pp. 157–68, quotation at p. 165. 10 Anna Komnene, The Alexiad, trans. E. R. A. Sewter, rev. Peter Frankopan (London, 2009), pp. 329–31.

58

james titterton

unique to the Alexiad (although there are parallels in the Norman histories, which will be discussed below) and it is told with evident admiration for Bohemond’s cleverness and daring.11 While Albu draws on a few examples from other primary sources in her study, her focus is almost entirely on the Alexiad. This chapter will take a broader perspective and consider the place of trickery in the major chronicles of Norman military activity in Italy and the Holy Land. These dramatic narratives, which focus on the deeds of remarkable individuals such as Robert Guiscard, Roger of Sicily, and Bohemond, describe numerous ruses and stratagems. They also offer a valuable insight into contemporary attitudes towards battlefield conduct: what was and was not considered an acceptable way for combatants to behave. As will be seen, the chroniclers were not always consistent in their depiction of military trickery. What could be admired in one individual, at one time, could be condemned in another. Varieties of Trickery One of the key characteristics attributed to the gens Normanorum by their historians was their skill in warfare, and this included their use of stratagems. Whether they were gathering intelligence, laying siege to a stronghold, or fighting in open battle, the Normans were depicted employing a variety of tricks and ruses to gain an advantage over their enemies. Medieval commanders needed to be able to locate their enemies, assess their strengths and weaknesses, and manoeuvre against them effectively, all of which required reliable information.12 The Normans, campaigning in the polyglot lands of southern Italy and Sicily, were able to exploit this multilingual culture to gather this kind of intelligence. In 1086, Roger I of Sicily 11

Albu, “Rooster”, pp. 163–65. See J. R. Alban and Christopher Allmand, “Spies and Spying in the Fourteenth Century”, in War, Literature, and Politics in the Late Middle Ages, ed. Christopher Allmand (Liverpool, 1976), pp. 73–101; Susan B. Edgington, “Espionage and Military Intelligence during the First Crusade, 1095–99”, in Crusading and Warfare in the Middle Ages: Realities and Representations. Essays in Honour of John France, ed. Simon John and Nicholas Morton (Farnham, 2014), pp. 75–86; Yuval Noah Harari, “Knowledge, Power and the Medieval Soldier”, in “In Laudem Hierosolymitani”: Studies in Crusades and Medieval Culture in Honour of Benjamin Z. Kedar, ed. Iris Shagrir, Ronnie Ellenblum, and Jonathan Riley-Smith (Aldershot, 2007), pp. 345–56; J. O. Prestwich, “Military Intelligence under the Norman and Angevin Kings”, in Law and Government in Medieval England and Normandy: Essays in Honour of Sir James Holt, ed. George Garnett and John Hudson (Cambridge, 1994), pp. 1–30; Bastian Walter, “Urban Espionage and Counterespionage during the Burgundian Wars (1468–1477)”, Journal of Medieval Military History 9 (2011), 132–45; James P. Ward, “Security and Insecurity, Spies and Informers in Holland during the Guelders War (1506–1515)”, Journal of Medieval Military History 10 (2012), 173–96. 12



“ the

arts of guiscard ” 59

prepared to assault the Muslim-held city of Syracuse in Sicily, sending ahead a certain Philip, son of Gregory, to make a reconnaissance: Faithfully fulfilling their command, he sailed among the Saracen fleet by night as if he were one of them: for he and all the sailors who went with him were most fluent in their language and in Greek. And when everything had been prudently examined, returning, he 13 reported that it was prepared for battle.

An even more elaborate deception was employed ahead of Robert Guiscard’s attack against Palermo in 1068. In what Amatus of Montecassino termed “a great act of cunning” (une grant soutillesce), Guiscard sent one “Peter the Deacon” to the city as a messenger, to thank the emir for gifts he had sent in hope of buying Guiscard’s friendship. Peter, “who understood [the Saracen language] and spoke like the Saracens”, was instructed to only listen, and to observe what he saw in the city, presumably so that the citizens would speak candidly in his presence, thinking that they would not be understood.14 Peter reported back “that the city was desolate and that [the people] of the city were like a body without a soul”: an easy target for an attack.15 Graham Loud speculated that this Peter “was probably a Greek Christian”: he was certainly somebody for whom Arabic was not his first language, otherwise the deception would not have worked.16 Sound intelligence helped the Normans to deceive their enemies on a grand strategic scale, misdirecting their enemies about their intentions and out-manoeuvring them. Following the failed campaign against Palermo of 1064, Guiscard and Roger of Sicily attempted to take the citizens by surprise through misdirection in July 1071: So the duke, when the supplies and other things prepared for the expedition had been assembled, supported by his brother, whom he had sent ahead, came to Catania where the count was, pretending he was going to attack Malta, as if he did not care about Palermo. But encouraged by his brother, leaving there with many horsemen, 17 with ships and plenty of foot soldiers, he came to Palermo. 13

Malaterra, De rebus, p. 86: “Qui jussa fideliter complens, de nocte inter classem Saracenorum, ac si unus ex ipsis esset, circumnavigat: nam et lingua eorum, sicut et graeca, ipse et nautae omnes, qui cum ipso processerant, peritissimi erant. Omnibusque prudenter circumspectis, rediens, certamen paratum renuntiat.” 14 Amatus of Montecassino, Ystoire de li Normant, ed. Michèle Guéret-Laferté (Paris, 2011), p. 404: “liquel entendoit et parloit molt bien coment li sarrazin.” 15 Amatus, Ystoire, p. 404: “Et Pierre fait assavoir a lo duc coment la cité est asoutillié, et ceuz de la cité sont comme lo cors san l’arme.” 16 Amatus of Montecassino, The History of the Normans, trans. Prescott N. Dunbar, rev. Graham A. Loud (Woodbridge, 2004), p. 142, n. 38. 17 Malaterra, De rebus, p. 52: “Dux igitur, commeatibus et caeteris quae expeditioni congruebant apparatis, fratrem, quem praemiserat, subsecutus, apud Cathaniam, ubi

60

james titterton

Unfortunately, as is often the case in chronicle narratives, Malaterra does not elaborate on precisely how Guiscard “pretended” to go to Malta. It may indicate some kind of manoeuvre in which the Norman fleet sailed east, then doubled back and made for Palermo, a counter-espionage campaign in which Guiscard spread false rumours that he intended to sail for Malta, or a mixture of the two. Malaterra is more explicit about the strategy Roger I of Sicily used to cross the straits of Messina in May 1061. A large fleet, sent by “Belcamet” (possibly Ibn-Hawwas), ruler of Castrogiovanni and Agrigento, prevented the Normans from crossing.18 And so count Roger, seeing the enemy was opposing his army on the far side of the straits and not moving, turned to cunning tricks, as he was accustomed to do [...] He gave advice to the duke: that, while he remained with the army, he should show himself to the enemies; meanwhile [Roger], with a hundred and fifty knights, going to Reggio, taking ships from there under cover of darkness, 19 crossing over, without the enemy’s knowledge, would invade Sicily.

Malaterra says that Guiscard initially rejected the plan, fearing that Roger would be killed, but it was eventually put into effect, albeit with a larger advance party: Roger crossed the straits with three hundred knights, landed at Tremestieri and sacked Messina.20 The Norman chroniclers were quite comfortable in depicting their subjects attacking an unsuspecting enemy, whether by night or an ambush. Indeed, Malaterra states that Roger of Sicily’s son Jordan sought to make his name by carrying out a daring ambush. In 1077, during the siege of Trapani on the west coast of Sicily, he led a raid on a nearby peninsula where the citizens were grazing their cattle “so that he might acquire a knightly reputation for himself in the sight of others”.21 Without speaking to comes erat, venit, fingens se Maltam debellatum ire, quasi de Panormo non curans. Sed a fratre cohortatus, magno equitatu, cum navalibus peditumque copiis ab inde progrediens, Panormum venit.” 18 For the identification of Belcamet with Ibn-Hawwas, see Geoffrey Malaterra, The Deeds of Count Roger of Calabria and Sicily and of his Brother Duke Robert Guiscard, trans. Kenneth Baxter Wolf (Ann Arbor MI, 2005), p. 87, n. 6. 19 Malaterra, De rebus, p. 32: “Comes itaque Rogerius, videns hostes ex altera ripa contra suum exercitum adjacere et nusquam promoveri, ad callida argumenta, ut solitus erat [...] convertitur, consiliumque duci dedit: ut, ibidem cum exerictu remanens, sese hostibus ostentaret; ipse interim, cum centum quinquaginta militbus Regium usque progrediens, navibusque sibi abinde sub nocturna umbra contraductis, mare, nescientibus hostibus, transiens, Siciliam invaderet.” 20 Malaterra, De rebus, p. 32. 21 Malaterra, De rebus, p. 63: “ut hanc sibi famam militariter prae caeteris acquireret”.



“ the

arts of guiscard ” 61

his father, “he landed upon the peninsula by night and left the ships; before dawn he was concealed in a certain hollow on the peninsula to make an ambush”.22 When the citizens came out with their herds, Jordan sprang the trap, defeated the citizens, and captured their cattle. According to Malaterra: “because of this the city was compelled to surrender, as we said, as if it was greatly terrified”.23 In 1103 Tancred was appointed regent of Edessa, following the capture of Baldwin II, count of Edessa, at Raqqa. In 1105 the city was besieged by Chökürmish, the Turkish governor of Mosul. Albert of Aachen reported that Tancred broke the siege with a dawn assault on the Turkish camp: And lo! all the citizens and knights, having gathered together in one place, resolved to fight, and to issue from the city at first light arrayed in divisions and bearing arms, to hurry to the Turks’ camp in silence, until each man would make an uproar with horns and trumpets [tubis et cornibus], so that they could attack the unprepared enemy unexpectedly while they were still sunk in sleep and safely dreaming, and so, since they were unable to rush for their 24 arms, they could cut them to pieces in a quick slaughter.

This tactic of attacking an enemy at dawn, particularly the blowing of tubi to cause panic, is reminiscent of Gideon’s victory over the Midianites in Judges 7: “When they were making a noise in a circle around the camps in three places, and had broken the water jars, holding the torches in their left hands, sounding the trumpets [tubi] in their right, they cried: ‘A sword for the Lord and Gideon!’.”25 Far from being a shameful or cowardly act, the night attack is framed here as a holy action, performed by a righteous commander against the forces of pagan oppression. Another tactic employed by the Normans, often in conjunction with an ambush, was feigning flight: pretending to run away in order to draw an enemy away from a defensible position or into a trap. The most famous use 22

Malaterra, De rebus, p. 63: “de nocte insulam pervadit, applicans navibus digreditur; in concavitate quadam eiusdem insulae, antequam illucescat, occultatur insidiis”. 23 Malaterra, De rebus, p. 63: “Hac re urbs, quasi maxime territa, ad deditionem, ut diximus, compulsa est.” 24 Albert of Aachen, Historia Ierosolimitana, ed. Susan B. Edgington (Oxford, 2007), p. 698: “Et ecce conuocati in unum ciues et milites constituerunt prelium, et ab urbe primo diluculo in armis et turmis procedere, ad castra Turcorum cum silentio properare, dum appropriantes fortiter in tubis et cornibus tumultuarent, hostes adhuc sopore depressos et secure somniantes subito improuisos inuaderent, ac sic minime ad arma contendere ualentes celerrima strage detruncarent.” 25 Biblia sacra iuxta vulgatam versionem, ed. Robert Weber, Roger Gryson, and Bonifatius Fischer, 5th edn (Stuttgart, 2007) Judges 7:20: “Cumque per gyrum castrorum in tribus personarent locis, et hydrias confregissent, tenuerunt sinistris manibus lampades, et dextris sonantes tubas, clamaveruntque : Gladius Domini et Gedeonis.”

62

james titterton

of this tactic (in English historiography at least) was at Hastings in 1066, but it was not a uniquely Norman stratagem: Turkish horsemen used it very frequently, as did other European cavalry during this period.26 The Normans in the South are depicted using both variants. During his attack on Messina in 1061, Roger of Sicily lured the citizens away from their fortifications by feigning flight: Rushing to the city gates, they went out to make a great assault upon those who were attacking them. But the count, most astute and skilled in military matters, having been attacked, first pretended to be afraid. When he had led them a long way from the city, he rushed fiercely upon them and put them to flight. When those in the rear had been killed, he pursued the fugitives with a hateful and 27 threatening gaze all the long way to city gates.

This is what we might term the “Hastings variant” of this tactic: luring the enemy into a disadvantageous position before unexpectedly turning back to attack. In c.1057 Richard, the Norman count of Aversa, defeated Guaimar IV, prince of Salerno, by luring him into an ambush with a feigned flight. According to Amatus of Montecassino, Richard had failed the day before to extort tribute from Guaimar, whose men had driven him away with arrows: And the count’s knights, when they saw that the prince had come out of the city, fraudulently began to flee. And the men of Salerno, who were dressed in linen clothing, followed them to the place where they had laid the ambush. And those who were lying in ambush saw the men of Salerno, they rushed upon them and they 28 were unable to flee.

26

For the debate on the feigned flight at Hastings, see Bernard S. Bachrach, “The Feigned Retreat at Hastings”, Mediaeval Studies 33 (1971), 344–47; David Bates, William the Conqueror (New Haven CT, 2016), p. 242; R. Allen Brown, “The Battle of Hastings”, in Anglo-Norman Warfare: Studies in Late Anglo-Saxon and Anglo-Norman Military Organization and Warfare, ed. Matthew Strickland (Woodbridge, 1992), pp. 161–81, quotation at p. 176; Stephen Morillo, “Hastings: An Unusual Battle”, in The Battle of Hastings: Sources and Interpretations, ed. Stephen Morillo (Woodbridge, 1996), pp. 219–28, quotation at p. 225. 27 Malaterra, De rebus, p. 30: “Huius urbis cives, quorum plurima multitudo erat, hostes suos fines pervasisse cognoscentes, plurimum indignati, maxime quod paucos numero videbant, urbis portas maximo impetu prosilientes, ipsos occupatum vadunt. Porro comes, ut semper astutissimus et militia callens, primo timore simulato, cum eos longius ab urbe seduxisset, impetu facto, acerrime super eos irruens, in fugam vertit. Sicque extremos quosque cedendo, usque in portam civitatis longo reditu fugientibus, visu odibili comminator, reduxit.” 28 Amatus, Ystoire, pp. 337–38: “Et li chevalier del conte, quant il virent que lo prince estoit issut de la cité, par fraude commencerent a fouir. Et ceauz de la cité de Salerne,



“ the

arts of guiscard ” 63

We see here the importance of what Stephen Morillo called “the expectation of cowardice” in carrying out such ruses: the citizens fell for the trick because they had already put the Normans to flight the day before and so presumed that they had done so again.29 Malaterra includes an illuminating account of a feigned flight used against the Normans, which resulted in the death of Roger of Sicily’s sonin-law, Hugh of Gircé. In 1075 Roger had left Hugh as his regent on Sicily while he returned to the mainland to attend to his affairs in Calabria. Hugh was instructed not to leave the city of Catania, lest he be attacked by the Muslim ruler of Syracuse, Benarvet (Ibn al-Wird).30 According to Malaterra: “But the young man’s spirit, burning to do knightly deeds and eager for praise, paying little heed to what he had been commanded, began to strive towards this end: that he might accomplish some noble deed, which would merit knightly praise, before the count returned.”31 He went to Troina and brought Jordan, Roger’s son (see above), back to Catania with him. Benarvet then laid an ambush outside Catania and sent forward thirty horsemen to lure the Normans into the trap: “But Hugh and Jordan, burning to do knightly deeds, rousing their men to battle, came out of the city with great fury; and sending ahead thirty picked knights to watch for ambushes, they hastened to the attack more incautiously.”32 The scouts failed to spot the ambush and Hugh was killed in the fighting, along with most of his men. What is noteworthy about this account is that Malaterra does not condemn the Muslims for employing deceitful tactics. All his censure is reserved for Hugh, whom he calls incautius for pursuing the Muslim horsemen, and who disobeyed Roger’s explicit command. It demonstrates that Western chroniclers were capable of depicting enemy tactics with equanimity: neither condemning nor praising them for employing subterfuge but accepting that such tactics were common in war. It was a commander’s responsibility to anticipate and avoid them, which Hugh of Gircé failed to do. When a stronghold proved too strong to assault, the Normans were prepared to use bribery to incite treachery within the garrison. In 1068 Guiscard reportedly captured the Calabrian town of Montepeloso (modern liquel estoient vestut de dras de lin, les secutoient jusques au lieu out estoit fait l’esguait. Et cil qui faisoient l’esgait virent cil de Salerne, il lor corurent sus et cil non porent fuir.” 29 Stephen Morillo, “Expecting Cowardice: Medieval Battle Tactics Reconsidered”, Journal of Medieval Military History 4 (2006), 65–73. 30 Malaterra, Deeds of Count Roger, p. 139, n. 11. 31 Malaterra, De rebus, p. 62: “Sed juvenilis animus, militia fervens et laudis avidus, quod sibi interdictum erat minus servans, ad hoc niti coepit, ut ante reditum comitis aliquod nobile facinus, unde militarem laudem mereretur, perpetraret.” 32 Malaterra, De rebus, p. 62: “Hugo vero et Jordanus, militia ferventes, suos ad certamen cohortando, urbe cum magno impetu digrediuntur; trigintaque electos milites insidias speculatum mittentes, ipsi incautius insequi accelerant.”

64

james titterton

Irsina) by bribing the commander: “he beguiled Godfrey, the keeper of that castrum, with promises, giving it over to him, and promising that he would give him more things and a stronger castrum”.33 These bribes were not necessarily material things, either. In 1094 Roger Borsa, duke of Apulia, persuaded the Greek citizens of Rossano, in southern Calabria, to shift their allegiance from his rebellious brother-in-law, William of Grandmesnil, to him by promising to remove the Latin bishop whom he had imposed upon them the previous year.34 Probably the most famous incident of a Norman using bribery to enter a stronghold is Bohemond’s capture of Antioch in June 1098.35 The Gesta Francorum is very frank about his methods: There was a certain emir of Turkish race named Pirrus, who had entered into a very close friendship with Bohemond. Bohemond often urged him, having sent one messenger after another to him, to receive him within the city in the most friendly manner; he promised him the freedom of the Christian religion, and 36 committed to make him wealthy with much honour.

Later chroniclers of the crusade, perhaps offended by Bohemond’s familiarity with Pirrus and the apparent ease with which he converted to Christianity, added pious details to the story.37 Fulcher of Chartres depicted Pirrus receiving divine visions, while Robert the Monk presented an extended dialogue between Pirrus and Bohemond, with Bohemond implausibly cast as a scholarly evangelist of the Gospel.38 Ralph of Caen depicts Pirrus as an Armenian Christian who had 33

William of Apulia, La Geste de Robert Guiscard, ed. Marguerite Mathieu (Palermo, 1961) (hereafter, Apulia), p. 156: “promissis decipit hujus / Custodem castri Godefridum, dans sibi quaedam, / Pluraque pollicitus castrumque valentius illo.” I have followed Prescott N. Dunbar’s convention of leaving the term castrum untranslated throughout as, in the context of southern Italy, it may refer to either a free-standing castle or a fortified village. See Amatus, History, p. 38. 34 Malaterra, De rebus, p. 100. 35 For a complete study of this incident, see Yuval Noah Harari, Special Operations in the Age of Chivalry 1100–1550 (Woodbridge, 2007), pp. 53–73. 36 Gesta Francorum, p. 44: “Erat quidam ammiratus de genere Turcorum cui nomen Pirus, qui maximam amicitiam receperat cum Boamundo. Hunc sepe Boamundus pulsabat nuntiis adinuicem missis, quo eum infra ciuitatem amicissime reciperet; eique christianitatem liberius promittebat, et eum se diuitem facturum cum multo honore mandabat.” 37 See Joshua C. Birk, “The Betrayal of Antioch: Narratives of Conversion and Conquest during the First Crusade”, Journal of Medieval and Early Modern Studies 41 (2011), 463–86; Robert Levine, “The Pious Traitor: Rhetorical Reinventions of the Fall of Antioch”, Mittellateinisches Jahrbuch 33 (1998), 59–80; Rebecca L. Slitt, “Justifying Cross-Cultural Friendship: Bohemond, Firuz, and the Fall of Antioch”, Viator 38 (2007), 339–49. 38 Fulcher of Chartres, Historia Hierosolymitana (1095–1127), ed. Heinrich Hagenmeyer



“ the

arts of guiscard ” 65

been wronged by the Turkish governor of Antioch and sought to aid his co-religionists.39 The most plausible explanation, however, is that Bohemond was continuing the tradition of his relatives and employed bribery where violence had failed. Of all the possible ruses and tricks, disguise is probably the most fundamental. Malaterra recorded a romantic story of how Guiscard was able to use disguise to acquire supplies early in his career. Besieged at his stronghold of San Marco Argentano by local Calabrian forces in c.1053, Guiscard dispatched a band of Slavs who were in his service to go raiding in a nearby valley and bring back some much-needed supplies.40 According to Malaterra, Guiscard secretly left his bed and joined the foragers, disguised “in the poor clothes and the kind of shoes that they wore”.41 On their return to San Marco Argentano, the Slavs were attacked by the Calabrians. Guiscard saved the situation in suitably heroic fashion, revealing his true identity and inspiring them to drive off their attackers. He then returned to his fortress in triumph, shouting “Guiscard!” to identify himself to the garrison. Malaterra includes a noteworthy addendum to this story: Nevertheless he was reprimanded by many of his men, because he had dared to do such things, and he was warned that he ought not to behave so any more lest fortune, who now smiled upon him, if 42 he tested her, should leave him in a worse situation at a later date.

This concurs with the lesson in prudence that Malaterra read in the death of Hugh of Gircé: while such an adventure may have been an admirable deed that was remembered at the Norman court (and certainly embellished in the telling), it was reckless for a noble commander to put himself in danger, and certainly not to be imitated. Arguably the most famous ruse in all the Norman chronicles is the “fake corpse” trick. William of Apulia first attributed it to Robert Guiscard in his Latin epic, Gesta Roberti Wiscardi. According to William, Guiscard captured his first stronghold in Calabria by pretending that one of his men had died and needed to be buried in the monastery, located within the castrum: (Heidelberg, 1913), pp. 230–31; Robert the Monk, Historia Iherosolimitana, RHC Occ. 3 (1866), pp. 796–81. 39 Ralph of Caen, pp. 59–60. 40 For the presence of Slavic settlers in Italy during this period, see Graham Loud, The Age of Robert Guiscard: Southern Italy and the Norman Conquest (Abingdon, 2013), p. 216. 41 Malaterra, De rebus, p. 16: “Sicque, lecto parato, cum jam collocatus esset, de nocte, nullo sciente, consurgens, vili veste et scarpis, quibus pro calceariis utuntur, ad similitudinem abeuntium sese aptans, illis medius jungitur.” 42 Malaterra, De rebus, p. 17: “Redarguitur tamen plurimum ab ipsis, quod talia praesumpserit, et ne ulterius praesumat, admonetur, ne forte fortuna, quae nunc arrisit, postmodum, si temptetur, in pejus cedat.”

66

james titterton This man, like a corpse, was placed upon a bier and a silk cloth spread over him, having been commanded to hide his face, as it is the Normans’ custom to cover up corpses. Swords were placed underneath the bier. The body was brought to the entrance of the monastery to be buried, the fiction of the dead man deceived the 43 men who could not be deceived by the tricks of the living.

Once they were within the walls, the “funeral party” drew the hidden swords and seized the castrum. This story has attracted a great deal of scholarly comment, not least because it appears in other texts, where it is attributed to other cunning individuals. Its earliest appearance is in Dudo of Saint-Quentin’s De moribus et actis primorum Normanniae ducum, where it is employed by the villainous Hasting, a pagan raider and spiritual ancestor to the Norman people, to sack the city of Luna.44 Otto of Freising attributed it to Roger II of Sicily and, as discussed above, Anna Komnene claimed a variant of the trick was employed by Bohemond of Taranto.45 There is another, seemingly distinct, tradition that ascribed it to Harald Sigurtharson, also known as Harald Hardrada, king of Norway (1046–66). According to the Morkinskinna, the earliest surviving compendium of Norse kings’ sagas (compiled c.1220), Sigurtharson used this ruse to enter an unnamed Muslim city on Sicily during his exile, when he served the Byzantine Empire as a member of the elite Varangian Guard.46 As this compilation was written down long after their first composition, it is difficult to determine what this saga’s original sources were. Theodore M. Andersson and Kari Ellen Gade suggested that the descriptions of Harald’s Byzantine service may have been based on oral testimony, as does Sigfus Blöndal, who described the spectacular adventures as “much-expanded self-justifications, originally told by Haraldr [sic] and blown up by his flatterers”.47 This 43

Apulia, p. 150: “Qui cum, quasi mortuus, esset / Impositus feretro, pannusque obducere cera / Illitus hunc facie iussus latitante fuisset, / Ut Normannorum velare cadavera mos est, / Conduntur feretro sub tergo corporis enses; / Ad monasterii subhumandum limina corpus / Fertur, et ignaros fraudis quos fallere vivi / Non poterant homines, / defuncti fictio fallit.” Reading cera as saeta here. 44 Dudo of Saint-Quentin, De moribus et actis primorum Normanniae ducum, ed. Jules Lair (Caen, 1865), pp. 132–35. 45 Otto von Freising and Rahewin, Die Taten Friedrichs oder richtiger Cronica, ed. Franz-Josef Schmale, trans. Adolf Schmidt (Darmstadt, 1965), p. 198. 46 Morkinskinna: The Earliest Icelandic Chronicle of the Norwegian Kings 1030–1157, trans. Theodore M. Andersson and Kari Ellen Gade (London, 2000), pp. 1, 141–43. This saga was also included by the Icelandic author Snorri Sturlson (d. 1241) in his collection of kings’ sagas, the Heimskringla. See Snorri Sturlson, Heimskringla: History of the Kings of Norway, trans. Lee M. Hollander (Austin TX, 1964), p. 585. 47 Morkinskinna, p. 63; Sigfus Blöndal, The Varangians of Byzantium: An Aspect of Byzantine Military History, trans. Benedikt S. Benedikz (Cambridge, 1978), p. 66.



“ the

arts of guiscard ” 67

fails to explain the similarities between the story in this saga and Dudo’s chronicle, which predated Harald’s exile in Byzantium by some thirty years. While it is possible that both tales have their origins in a lost, oral motif from Scandinavian saga traditions, there is no textual evidence for this. The striking similarities between these two tales must lead one to prefer Dudo as the originator of the tale and the saga as the imitator. It may have been transmitted orally, via Scandinavians in Byzantine service, from Normans in southern Italy, as Klaus Rossenbeck has suggested, or by Scandinavian scholars who travelled south and encountered the story in a Continental chronicle.48 This would explain why the saga depicts a Muslim garrison permitting a Christian burial party to enter their city: it is a legacy from Dudo’s original that becomes a nonsense when applied to Harald’s historical campaign. As well as being the most famous, this is also arguably the least plausible Norman ruse: it is highly unlikely that Guiscard, or anybody else, ever actually employed it. Yet its very survival, across multiple texts from different sources, tells us that it was a story that contemporaries enjoyed hearing. Furthermore, the subtle differences between Dudo’s text and William of Apulia’s Gesta demonstrate how the admirable aspect of the story (the commander’s cunning) was separated from the more questionable elements. Dudo’s Hasting is a monster who murders the bishop performing his own funeral mass, sacks the city and puts the Christian inhabitants to the sword. William of Apulia, by contrast, is careful to avoid any suggestion that Guiscard committed sacrilege. In the Gesta, the fake corpse never actually enters the monastery. Nor is any great slaughter inflicted: “they were all captured, and Robert established his first garrison in the castrum. The monastery was not destroyed, the monastic flock was not expelled from there.”49 The essential story is the same, but William of Apulia has removed the morally troubling elements. All that remains is an irreverent display of Guiscard’s cunning. Admirable Cleverness or Contemptible Deceit? Whether all these ruses occurred exactly as described is arguably less important than the fact that the chroniclers chose to include them at all. Their very presence suggests that their audience enjoyed stories of cunning and could admire those who used guile to outwit a foe. This is not to say that the chroniclers were wholly comfortable with trickery. They occasionally 48

Klaus Rossenbeck, Die Stellung der Riddarassogur in der altnordischen Prosaliteratur: Eine Untersuchung an hand des Erzählstils (Bamberg, 1970), pp. 78–79. 49 Apulia, p. 150: “omnes capiuntur, et illic / Praesidium castri primum, Roberte, locasti. / Non monasterii tamen est eversio facta, / Non extirpatus grex est monasticus.”

68

james titterton

express unease about some of the Normans’ ruses, and sometimes outright condemnation. For example, Amatus of Montecassino called them “malicious” (maliciouz) for their tactic of luring the starving citizens of Palermo into ambushes with loaves of bread during the siege of 1071.50 A repeated theme in the narratives is the Normans’ habit of adhering to the letter of an agreement while breaking it in spirit. According to Amatus of Montecassino, Robert, duke of Apulia, earned his cognomen ‘Guiscard’ (which means ‘wily’) after capturing Peter of Tira, the governor of the Calabrian city of Bisignano, in c.1048.51 Guiscard, who had recently taken possession of San Marco Argentano, which lies across the river Crati from Bisignano, had arranged to meet Peter for talks outside the city walls. He then convinced Peter that their retinues should withdraw to prevent any quarrels breaking out. Once he was isolated, Guiscard seized Peter and dragged him away, while the Calabrians fled back to the city in terror. Peter was then compelled to pay a large ransom for his release.52 The two major accounts of this incident include details that could be interpreted as justifying Guiscard’s actions. Malaterra states that Peter was captured “on a certain day, since no treaty prevented it”: no explicit agreement had been made, no oaths had been sworn, so Guiscard had not violated his word.53 Amatus of Montecassino depicted Guiscard begging Peter’s forgiveness once he had brought him to San Marco Argentano: Then Robert knelt down, and folded his arms, and asked for mercy. And he confessed that he had sinned, but Peter’s wealth and his own poverty had compelled him to do it: “But you are my father: however, as you are my father, it is right that you should aid your 54 poor son”.

This scene may be intended to be humorous, but it may also be intended to highlight the lengths to which poverty had driven Guiscard. Earlier in his text, Amatus compared his poverty to that of the Israelites during their exodus from Egypt: And because of [his poverty] he lacked everything, except that he had an abundance of meat, so Robert lived on the mountain as the sons of Israel lived in the desert. They ate the meat in moderation,

50

Amatus, Ystoire, p. 429. Loud, Age of Guiscard, pp. 111–13. 52 Amatus, Ystoire, pp. 316–17; Malaterra, De rebus, pp. 17–18. 53 Malaterra, De rebus, p. 18: “quadam die, nullo foedere interposito”. 54 Amatus, Ystoire, p. 315: “Puiz Robert va agenoillié, et ploia les bras, et requist misericorde. Et confessa qu’il avoit fait pechié, mes la richesce de Pierre et la povreté soe lui avoit fait constraindre a ce faire. ‘Mes tu es pere: mes que tu me es pere, covient que aide a lo filz povre.’” 51



“ the

arts of guiscard ” 69

he sated himself with the taste of all manner of meat, and Robert 55 drank the water of a pure fountain that was there.

Again, the reference to the Israelites eating a mesure may be an implicit criticism of Guiscard, suggesting that he lived immodestly or greedily, but it is difficult to be certain about Amatus’ intentions. In 1083 Guiscard acquired control of the Calabrian stronghold of Santa Severina through a very literal interpretation of an agreement. Guiscard’s nephews, Abelard and Herman, count of Canne, had rebelled against his rule while he was away campaigning in the Balkans. Upon his return, Guiscard captured Herman at Canne and imprisoned him. Abelard agreed to surrender Santa Severina, in return for which Guiscard promised to free Herman when they came to a certain fortress named Gargano. Abelard remained with Guiscard for a while, eventually demanding to know when Guiscard would fulfil his half of the bargain: “Then lo! the duke revealed the trick: he stated he would not go to the place that he had named for seven years.”56 The rebellion ended with both Abelard and Herman leaving Italy to go into exile, taking service with the Byzantine Emperor Michael VII Doukas. Guiscard was not the only Norman capable of this kind of subterfuge. Roger of Sicily played a similar trick on the people of Gerace in 1062. At that time, Roger was at war with his brother Guiscard over what he perceived to be an inadequate share of the family’s lands in Calabria. During the conflict, Guiscard was captured by the Greek citizens of Gerace. Roger arrived before the city and demanded that the citizens surrender Guiscard to him, claiming that he wished to torture him to death himself. The citizens agreed, but under certain conditions: And so, when they had received counsel, doubtful whether the things that they had heard from the count had been said because he wanted to rescue his brother or because he truly hated him, they demanded an oath from the duke to declare, if he was released by them and evaded his brother’s threats, if the count should spare his 57 life, that he would not build any sort of castle in their city. 55

Amatus, Ystoire, p. 314: “Et come ce fust cose que toutes chozes lui failloient, fors tant solement qu’il avoit abundance de char, coment li filz de Israel vesquirent en lo desert, ensi vivoit Robert en lo mont. Ceaux menjoient la char a mesure, cestui se [sacioit] o une savour toutes manieres de char, et lo boire de’stui Robert estoit l’aigue de la pure fontainne.” I have accepted the editor’s suggested use of sacioit to fill a lacuna in the MS. 56 Malaterra, De rebus, p. 60: “Cum ecce dux dolum aperit: infra septem futuros annos se illuc, quo dixerat, non iturum asserit.” 57 Malaterra, De rebus, p. 38: “Sicque, consilio accepto, dubii utrum ne ea, quae a comite audierant, ex industria, ut fratrem eripiant, aut certe ex vero odio dicerentur, juarmentum a duce expetunt, ne, si ab ipsis dimissus, fraternas minas evadat, dum sibi vita comes fuerit, castellum aliquod infra civitatem suam firmare faciat.”

70

james titterton

Roger received Guiscard, who agreed to divide Calabria with him. Roger went on to construct a fortress at Gerace to extort money from the citizens: But when the Greeks objected, on account of the oath which the duke had made, the count answered: “Since half of Gerace is mine, the duke will not be able to fortify his part, lest it violate the terms of his oath; but I can build anything I like in my part, as it does 58 not violate any vow or promise.”

Once again, a Norman deceived an enemy but did not technically break his word. This had the advantage of making the chronicler’s subject appear clever, while also defending them from accusations of oath breaking or outright faithlessness. Another idea that reoccurs throughout the Norman chronicles is that of commanders succeeding “by force or by craft” (“vel armis vel arte”), with the implication that either is valid. This phrase suggests a certain defensiveness on the part of the chroniclers, as if they anticipated that their audience might disapprove of trickery. For example, when Malaterra described how Roger of Sicily formulated his stratagem to cross the straits of Messina in 1061 he wrote: “It was as if he had read: ‘What does it matter whether the palm of victory is won by arms or tricks?’”, which is a quotation from the fourthcentury poet Prudentius’s Psychomachia.59 This is itself probably a reworking of a quotation from Vergil’s Aeniad, which the Trojan warrior Coroebus uses to instruct his men to dress in the armour of dead Greeks: “Trickery or force, who would ask which in the case of an enemy?”60 This sentiment, lent authority by its classical pedigree, appears frequently in William of Apulia’s Gesta. Describing the capture of Montepeloso (see above), he wrote that: “because the duke was not strong in arms, he took the castrum by craft”.61 Likewise, when he found that he could not storm the Byzantine city of Dyrrachion (modern Durrës, on the coast of Albania) in 1082, he bribed a Venetian nobleman to admit his men to the city by night: “So the duke took Dyrrachion for himself, and because he was unable to conquer through arms, he subjugated it through craftiness to achieve victory.”62 On other occasions, 58

Malaterra, De rebus, p. 39: “Geracensibus vero juramentum sibi a duce factum objicientibus, comes respondit: ‘Cum medietas Geracii, mea sit, dux in sua parte juramenti sui ordinem, ne violetur, servare poterit; me vero, qui in mea parte quodlibet faciam, nec votum, nec promissio aliqua redarguit.’” 59 Malaterra, De rebus, p. 32: “ac si legisset: ‘Quid refert? Armis contingat palma dolisve.’” Prudentius, Prudentius, trans. H. J. Thomson, 2 vols. (Cambridge MA, 1949), 1:316. 60 Vergil, Aeneid, 2.386–91. 61 Apulia, p. 156: “dux wuod non evalet armis / Arte capit castrum”. 62 Apulia, p. 230: “Sic sibi Dirachium dux subdidit, atque quod armis / Vincere non potuit, victoria subiugat artis.”



“ the

arts of guiscard ” 71

William of Apulia used it to describe Guiscard’s general strategy: “He seized the palm of victory either by craft or by arms, he considered both in the same manner, because a cunning mind often accomplishes what violence cannot.”63 See also his description of how Guiscard put down a rebellion in Apulia and Calabria in early 1079: He overcame them all, either by craft or by arms; some he won over, some he enticed with sweet words, others he broke in battle; cunning and bold, he knew both methods; he seized some castles, others, which could not be obtained by warlike violence, he persuaded to 64 surrender with charming words.

William of Apulia’s choice of vocabulary is important. He consistently uses ars, which has connotations of skill, craft, and general cleverness, as the alternative to arma. He uses the same word to describe Guiscard’s construction of a dam to raise the level of the river Glykys in western Greece to refloat his fleet in the winter of 1084–85: “he made the difficult task easy through ars”.65 The same word is used for a feat of engineering as for his aptitude for bribery and deceit. Other chroniclers display a similar ambiguity in their vocabulary when describing ruses and other forms of cunning. According to Malaterra, Roger of Sicily was “most astute” (astutissimus) for luring the citizens of Messina out with a feigned flight.66 After Guiscard had captured Peter of Tira outside Bisignano, Malaterra said: “When the Calabrians, a most fearful people, had learned of this display of Guiscard’s cunning [calliditatus] and other, similar incidents, they all trembled before him: indeed, they said nobody could be compared to him in arms and ingenuity [ingenium], if not in power.”67 In the Tancredus, Ralph of Caen described a force of crusaders who raised many banners in order to trick the Turks into thinking that they were more numerous than they truly were as being “as prudent as

63

Apulia, p. 148: “Si contingebat sibi palma vel arte vel armis, / Aeque ducebat, quia quod violentia saepe / Non explere potest, explet versutia mentis.” 64 Apulia, p. 194, 196: “vel arte vel armis / Omnes exsuperat; monitis quam dulcibus illos / Allicit, hos bello domitat; versutus et audax / Novit utrosque modos; adimit sua castra quibusdam, / Quosdam blanditiis verborum commovet ultro / Tradere, quae nequeunt violento marte parari.” 65 Apulia, p. 248: “Dux, qui difficilem facilem facit arte laborem, dum fluvium solitis cognovit egere fluentis, namque meatus aquae brevis arta fauce fluebat, / multos afferri palos et ab amnis utraque / Margine configi connexos vimine iussit, / Et multis multa praecisis arbore ramis / Composuit crates, et arenis desuper implet.” 66 Malaterra, De rebus, p. 30. 67 Malaterra, De rebus, p. 18: “Tali calliditate et huic similibus Calabrenses de Guiscardo compertis, genus formidolosissimum, omnes ante eum tremebant: quippe cui neminem assimilari posse armis et ingenio, sed neque viribus, dicebant.”

72

james titterton 68

they were bold.” While it is not entirely appropriate to compare Amatus of Montecassino’s vocabulary to the other chronicles, as we lack the original Latin text, it is still noteworthy that Guiscard employing Peter the Deacon as a spy was described as “une grant soutillesce”, whereas the Greeks were described as “accustomed to overcome their enemies through malicious argument and subtle treachery”.69 This is in keeping with the descriptions of other peoples in these chronicles, where the authors regularly use negative language to describe actions that are objectively similar to those performed by the Normans. Describing a sea battle between Guiscard and a Venetian fleet off Dyrrachion in September 1081, Malaterra claimed that the Venetians requested a truce, offering to surrender the next morning. Instead, the Venetians spent the night strengthening their ships for battle: “But the duke, ignorant of the deception [dolus], sending the greater men of his army at first light to receive their surrender, as they had promised, ordered that they all be brought to him, all safe and unharmed.”70 Rather than surrendering as expected, the Venetians attacked the Norman ambassadors and drove them back. In his account of the battle of Dorylaion (1 July 1097), fought between the armies of the First Crusade and Qilij Arslān, sultan of Rūm, on the crusaders’ march across Anatolia, Ralph of Caen depicted a Turkish force luring Godfrey of Bouillon to fight on a nearby mountain: A rampart stood in the duke’s path, and the ground receding under him divided the mountain from the rampart. Those who trusted in trickery [dolus] more than arrows fled there. The duke, thinking that those who were fleeing wanted to seize the rampart to fight, being ignorant of their fraud [fraus] and wicked craft [ars iniquia] turned the face of battle upon them and in a swift assault freed that hillock which had previously been burdened with enemies; thenceforth he pursued those he had compelled to flee across the plain to the mountain with the slaughter of many, both enemies and his 71 own men.

68

Ralph of Caen, p. 82: “tam prudens quam audax”. Amatus, Ystoire, p. 248: “Et que li Grex molt de foiz, par maliciouz argument et o subtil tradement, avoient usance de veinchere lor anemis.” 70 Malaterra, De rebus, p. 73: “Dux vero, doli ignarus, summo diluculo potentiores exercitus sui, qui eos seipsos, sicut promiserant, dedentes susciperent, mittens, salvis omnibus, quae ipsorum erant, usque ad se deducere praecipit.” 71 Ralph of Caen, p. 32: “Stabat in occursu ducis agger, agerque sub ipso / retrocedentem dirimebat ab aggere montem. Huc ubi fisa dolis gens plusquam fisa sagittis / Diffugium fecit, ratus aggere uelle potiri / Ad pugnam refugos, dux, fraudis et artis iniquae / Ignarus, belli faciem conuertit in illos, / Et celeri incursu tumulum prius hoste grauatum / Libera; inde fugam ad montem per plana coactam / Strage premit multa, hostis scilicet atque suorum.” 69



“ the

arts of guiscard ” 73

Malaterra describes how the Christians of Sicily were ashamed that the town of Catania had been betrayed to the Muslims of Syracuse by Benthumen, a Muslim appointed to govern the city by Roger of Sicily: “Because this deed resounded throughout the island the Christians were much ashamed that such an abominable fraud [nefanda fraus] should have been found among them.”72 There is a clear double standard at work here. When the Normans deceived somebody, it was by ars and calliditas. When Greeks or Muslims were deceptive, it was dolus and fraus. It is difficult to determine to what extent, if any, the Norman soldiers themselves knew or cared about such definitions, or how they themselves would have justified their use of bribery, ambush, and deceit to achieve their conquests. Ralph of Caen offers us the nearest thing we have to a defence of the “arts of Guiscard”. He depicts a confrontation between Tancred and Arnulf of Chocques, who was (temporarily) acting as papal legate to the crusading army, following the sack of Jerusalem (15 July 1099). Ralph of Caen places a grand speech in Arnulf ’s mouth, in which he accuses Tancred of wrongfully dividing the booty from the Temple Mount among his men, when it ought to have been given to the Church: If you had paid proper attention, o son of the marquis, you would have spared these lands as if they were heaven: this land, I say, unique among lands, most like the heavens, if the earth contains anything like the heavens. But one ought to be kind to a Guiscard: for he followed in his forefathers’ footsteps. Who, amidst embraces, amidst kisses, threw his own cousin tumbling from the walls? Guiscard, of course! Who was borne alive into Montecassino like one dead, living as if to be buried? Surely it was Guiscard! Who first poured hot water, then cold, upon his nephew, whom he had enticed into friendship? The same Guiscard! Yet he is said have been a founder of churches, not a destroyer: he did not despoil 73 but furnished many.

Arnulf refers to some of Guiscard’s worst characteristics: his willingness to betray family and enemy alike and his aptitude for deceit, symbolised by the story of the “fake corpse”. This cannot be dismissed as anti-Norman 72

Malaterra, De rebus, p. 75: “Quod factum cum per totam insulam personuisset, Christiani quidem, eo quod tam nefanda fraus inter ipsos reperta sit, plurimum erubescebant.” 73 Ralph of Caen, pp. 112–13: “Quod si satis, o Marchisida, attendisses, huic saltem pepercisses, quasi in terris celo: huic, inquam, unico in terris, si quid terra celis simile habet, celorum simillimo. At indulgendum est Wiscardidae: secutus est enim patrum suorum uestigia. Quis inter amplexus, inter oscula compatrem suum a menibus rota deiecit? Nempe Wiscardus! Quis uiuus pro mortuo, incolumis pro tumulando, in montem Cassinum perlatus est? Vtique Wiscardus! Quis nepotem suum, ad concordiam elicitum, prius calida mox gelida perfudit? Idem Wiscardus! Qui tamen fertur ecclesiarum fundator, non subuersor; nec denudasse, at multas ornasse.”

74

james titterton

prejudice, inserted into the mouth of a straw man to be refuted by Tancred’s counter-arguments. Arnulf was Ralph of Caen’s tutor. The prologue to the Tancredus is addressed to him: Ralph asked him to read it and “prune” the superfluous passages.74 Moreover, although he offered a defence of his actions and of Guiscard’s reputation, Tancred eventually agreed to submit to Arnulf ’s authority and pay seven hundred marks in compensation for the lost treasure. Arnulf is portrayed here as being in the right. So why did Ralph of Caen choose to present Guiscard in such a negative light? Elsewhere in the Tancredus he refers to him in glowing terms: “For who does not esteem the valour of Guiscard, before whose conquering standards, it is said, in one day the Greek and German emperors trembled?”75 It is possible that this exchange provided an opportunity for Ralph both to acknowledge Guiscard’s dubious reputation and to defend it, without seeming to be personally critical of him or his family. Tancred’s defence of Guiscard, although clearly a literary construct invented by the chronicler, may nevertheless suggest how the Normans themselves responded to criticism of their morally dubious behaviour: You all heard him, there is no need for an external witness, with what persuasive force he savaged my ancestry: a man who has not seen any prince arise from his own lineage disparaged such a prince as Guiscard, second only to Alexander in boldness. The deeds of Guiscard are known throughout the world. It is impossible to disparage them, unless one always tried to paint white on black and 76 black on white[.]

The appeal here is not to classical texts or biblical precedent, as one might expect from a literate author like Ralph of Caen. Rather, he depicts Tancred appealing to secular concepts of lineage, honour, and fame. Tancred dismisses the attacks on Guiscard as motivated by jealousy because Arnulf ’s family is of little repute. He appeals to his peers, assembled to try the case, that his family honour had been attacked, “savaged”, by Arnulf ’s words. He claims that it is impossible to disparage Guiscard’s deeds because they are so well and so widely known. Although they appear in a prolix Latin chronicle, such sentiments would not have been out of place in a chanson de geste or romance. At no point does he deny any of the specific charges laid against 74

Ralph of Caen, p. 5. Ralph of Caen, p. 6: “Quis enim Wiscardi probitatem non probet, cuius signa sub uno, ut aiunt, die Grecus Alemannusque imperator tremuerunt uictricia?” 76 Ralph of Caen, p. 114: “Audistis ipsi, non est externo opus teste, qua ui persuadente genus meum corroserit: Wiscardo, secundae ab Alexandro audaciae, detraxit, tanto principi homo, de cuius sobole quispiam principem non uidit. Wiscardi acta nota sunt orbi, non est qui possit detrahere, nisi qui semper studuit candidum in nigra, nigrum in candida colorare.” 75



“ the

arts of guiscard ” 75

Guiscard. He implies that his boldness and the glory he won overshadowed the dubious methods he used to acquire them. Or, in the words of Geoffrey Malaterra: “What does it matter whether the palm of victory is won by arms or tricks?”77

77

Malaterra, De rebus, p. 32: “ac si legisset: ‘Quid refert? Armis contingat palma dolisve.’”

Part II

Cultural Representation and Diffusion

4

A Gift to the Normans: the Military Legacy of Sicilian Islam David Nicolle

It is well known that the Norman rulers of Sicily and southern Italy employed and made very effective use of Muslim troops from Sicily. Certain aspects of their recruitment, organisation, skills, and motivation are also well understood. But the only aspect of their tactics, and by extension their equipment, which has so far been studied in detail concerns archery.1 However, these Sicilian “Saracens” served as other types of infantry and as light cavalry, as well as providing the Normans with highly regarded military engineers. It has also been suggested that the Muslims of Sicily and southern Italy played a part in the spread of early medieval Islamic military technology to western Europe.2 As is well known, even after the forcible transfer of Muslims from Sicily to the Italian mainland by the Normans’ successors, this community continued to play a significant role for many decades (fig. 37a–b). This chapter hopes to explain quite what the Muslims of Sicily had to offer their eleventh-century Norman conquerors in terms of military technology, military organisation, and tactics. Where and how these aspects of Siculo-Muslim military tradition evolved are also important questions, because Islamic Sicily – though prosperous and culturally flourishing – remained a relatively small frontier province of the early medieval Islamic world. How much of what the Normans inherited was a local development, how much from the neighbouring Maghrib (North Africa), and how much from most distant regions of the Islamic world – to east and west? This remains an unanswered question. What is clear is that the military traditions

1

Giovanni Amatuccio, “Saracen Archers in Southern Italy” (https://bit.ly/2RYe1lG). David Nicolle, “Jawshan, Cuirie and Coat-of-Plates: An Alternative Line of Development for Hardened Leather Armour”, in A Companion to Medieval Arms and Armour, ed. David Nicolle (Woodbridge, 2002), pp. 179–221 and pls. XIII-1 to XIII-45. 2

80

david nicolle

of Islamic Sicily were rooted in the first two centuries of Islamic history in North Africa, the Middle Eastern heartlands, and even Iran. The Muslim army which invaded Sicily in 827 AD is said to have consisted of ten thousand infantrymen and seven hundred cavalry in about one hundred ships.3 Its leadership was largely Arab, and Arabs formed the elite of this force.4 However, the majority were Berbers, mainly from the Huwwarah tribe. Other soldiers included political exiles from Al-Andalus in the West and Khurasanis from the East.5 Berbers continued to play the major military role throughout what proved to be a hard-fought and prolonged campaign;6 however, chroniclers also emphasised the role of scholars and religious figures who accompanied the first invading army.7 Setting off from Sousse in Ifriqiya (now largely Tunisia), this Muslim army established a bridgehead at Mazara, on the western tip of Sicily, after a crossing which supposedly took three days. This might, in fact, mean that transporting the army to Sicily took three days, rather than the fleet being at sea for three days.8 The invasion of Sicily has been described as an early and ambitious example of a major campaign of conquest launched by a provincial governor, in this case the Aghlabid ruler of Ifriqiya, rather than by the caliph. It also provided an outlet for the aggressive enthusiasm of the jund army in Ifriqiya, while bringing a satisfactory amount of loot to the Aghlabid government and giving that newly established dynasty political and religious legitimacy.9 Unlike the conquest of the Iberian peninsula over a century earlier, the conquest of Sicily took a long time and largely consisted of sieges, but once the main city of Palermo was taken in 831 AD,10 Sicily became an amirate or province of the Aghlabid state. Yet, the last bastions of Byzantine resistance were not taken until 965 AD, by which time Sicily had passed from Aghlabid to Fatimid and then to local Kalbite rule. Just under a century later, the first Norman raid landed on the island. Islamic invasion was followed by Arab and Berber colonisation. Initially, Muslim settlement in Sicily was military and administrative. Meanwhile, many Christian prisoners, Byzantine and Sicilian, were reportedly sent to Islamic territories. Subsequently, the Muslim settlement of Sicily became 3

William E. Granara, “Political Legitimacy and Jihad in Muslim Sicily 217/827– 445/1053”, PhD thesis (University of Pennsylvania, 1986), p. 62. 4 Granara, “Political Legitimacy”, pp. 67–68. 5 Michele Amari, Storia dei Musulmani di Sicilia, 2 vols. (Catania, 1933), vol. 1, pp. 394–95. 6 Aziz Ahmad, A History of Islamic Sicily (Edinburgh, 1975), p. 22. 7 Granara, “Political Legitimacy”, pp. 94–96. 8 Granara, “Political Legitimacy”, pp. 62–63. 9 Granara, “Political Legitimacy”, p. 65. 10 Haven C. Krueger, “The Italian cities and the Arabs before 1095”, in A History of the Crusades, ed. Kenneth M. Setton (Madison, 1969), vol. 1, p. 44.



a gift to the normans 81

agricultural, with archaeological evidence indicating that the process was not only substantial but also systematic, resulting in a revival of both agriculture and urbanisation during the Islamic period,11 while the evidence of surviving place names indicates that Muslim settlement was heaviest south-west of a line between Cefalu and Catania. The newcomers included Andalusians, Egyptians, Khurasanis and black, sub-Saharan Africans as well as people from Ifriqiya.12 Within a short space of time substantial numbers of local Sicilians converted to Islam, and the new province of Siqiliya eventually had a notably mixed population including the existing Orthodox Christian Sicilians and ex-Byzantine Greeks, Latin Catholic “Lombards” or mainland Italians, and a substantial Jewish community.13 Furthermore, there is evidence that the Arabic-speaking settlers included Arab Christians from Egypt and Al-Andalus.14 Here it should also be borne in mind that the Islamic armies of Al-Andalus included both local Andalusians equipped in essentially western European style and more recently arrived Berber troops dressed and equipped in North African style (fig. 60a–e). By the time of the Norman invasion in the second half of the eleventh century, an estimated two-thirds of the population of Sicily had converted to Islam, with most of the Christian minority living in the north-east.15 Even so, the remaining Christians seem to have become musta‘rib, or Arabised, and hence suspect in the eyes of Norman and subsequent Christian settlers from the European mainland.16 Nor was Sicilian Islam particularly orthodox or mainstream. According to Ibn Hawqal writing in the ninth century, knowledge of Islam among most Sicilian Muslims was rudimentary.17 Even in the later tenth century there were complaints that Sicilians often appeared in their mosques armed, which more educated believers regarded as reprehensible,18 while the “folk” character in Sicilian Islam may have increased under Norman rule during the twelfth century. In the meantime Shi’a Muslims had found refuge in

11

Leonard C. Chiarelli, “Sicily during the Fatimid Age”, PhD thesis (University of Utah, 1986), pp. 89–90. 12 Henri Bresc, “Mudejars des Pays de la Couronne d’Aragon et Sarrasin de la Sicile Normande: Le Problème d’Acculturation”, in Jaime I y su época (X Congreso de Historia de la Corona de Aragón) (Zaragoza 1975), vol. III (Zaragoza 1980), pp. 52–53. 13 Ahmad, Islamic Sicily, p. 22. 14 Bresc, “Mudejars des Pays”, p. 58. 15 Graham A. Loud, The Age of Robert Guiscard: Southern Italy and the Norman Conquest (London, 2000), p. 147. 16 Chiarelli, “Sicily during the Fatimid Age”, p. 114. 17 Bresc, “Mudejars des Pays”, p. 53. 18 Judhari, Abu ‘Ali Mansur al-, Vie de l’Ustadh Jaudhar [Sirat al-ustadh Judhar], trans. Marius Canard (Algiers, 1958), p. 102.

82

david nicolle

Sicily during a time of persecution in North Africa,19 when the Hammadid dynasty of eastern Algeria threw off its allegiance to the Shi’a Fatimid caliphate in Egypt. Within a few decades there were also serious tensions between established Sicilian Muslims, including local converts, and recent arrivals from North Africa.20 Such divisions within Sicilian society would, of course, greatly assist the Norman invaders. For most of the time the political leadership of Muslim Sicily was dependent upon Ifriqiya in North Africa, or upon Egypt. Indeed politically, economically, and culturally, Siqiliya was part of the Maghrib. Although small, Islamic Sicily also influenced Islamic North Africa, particularly during the era of Fatimid rule (909–48 AD), which was, paradoxically, also a time of internal strife.21 Although the political situation worsened in Sicily under the Fatimids’ successors, this last century also saw a cultural flowering which carried over into the Norman era.22 Underpinning the urban and cultural achievements of Islamic Sicily were major changes in rural areas. The great latifundia estates were broken up into smaller farms by the Muslim conquerors, opening up opportunities for the servile rural population, if they converted to Islam.23 Free, and now able to defend their new status, the soon largely Muslim rural population of western Sicily nevertheless mostly lived in undefended villages in the valleys.24 Sicilian cities and towns similarly expanded considerably under Islamic rule.25 In these urban centres, Muslims and Christians normally lived in separate or at least distinct quarters,26 with Christians continuing to play a prominent role in local government.27 In fact Sicily had a distinctly military character under Islamic rule, as the geographer al-Muqqadasi wrote in the later tenth century; “Sicily, the fertile island whose people never tire of fighting the jihad”.28 Perhaps the most famous military leader to come out of Sicily during this period was the Fatimid general Abu al-Hasan Jawhar ibn ‘Abd Allah. His family background remains obscure, but according to the chronicler al-Taghri Birdi,

19

Graham A. Loud, “The Norman Conquest of Sicily”, in A History of the Crusades, ed. Kenneth M. Setton (Madison, 1969), vol. 1, p. 57. 20 Loud, The Age of Robert Guiscard, p. 147. 21 Chiarelli, “Sicily during the Fatimid Age”, abstract ii. 22 Chiarelli, “Sicily during the Fatimid Age”, pp. 60–64. 23 Chiarelli, “Sicily during the Fatimid Age”, p. 101. 24 Loud, The Age of Robert Guiscard, p. 183. 25 Gina Fasoli, “Le Città Siciliane dall’Istituzione del Tema Byzantino alla Conquista Normanna”, Archivio Storico Siracusano 2 (1956), 61–81, also in G. Fasoli (ed.), Scritti di Storia Medievale (Bologna 1974), pp. 342–43. 26 Fasoli, “Le Città Siciliane”, p. 352. 27 Chiarelli, “Sicily during the Fatimid Age”, p. 60. 28 Loud, “The Norman Conquest of Sicily”, p. 58.



a gift to the normans 83

writing in the fifteenth century but drawing upon earlier sources, Jawhar was a freedman of the Fatimid caliph and was known by various epithets including al-Siqilli (the Sicilian). Others have suggested that Jawhar might have been of Armenian origin but raised as a slave in Sicily.29 After Sicily ceased to be part of the Shi’a Fatimid caliphate in 948 AD, its people’s enthusiasm for military exploits declined.30 Nevertheless, by the time of the Norman invasion, Sicily was strongly fortified.31 Sicily being strategically located, the military history of the Islamic island had an important naval dimension, largely consisting of coastal raiding rather than attacking merchant ships at sea.32 Whether the armed man found in the wreck of a small Islamic vessel off the coast of Provence and dated to this period was such a raider, or jihadi (fig. 7), remains unknown, of course.33 What is clear is that Muslim shipbuilders developed a significantly greater horse-transporting capability than their European and Byzantine rivals. The large size of some Andalusian and Fatimid ships caused astonishment at the time, so it is not surprising to read that Arab tarida specialist horse-transporting galleys could carry up to forty horses by the late tenth century.34 Only nineteen years after invading Sicily, another substantial force of western Muslims attacked Rome. According to Bishop Prudenzio of Troyes they came in sixty-three ships which landed at Ostia carrying five hundred horses. This was a major assault and is said to have wiped out the pope’s Scholae – a Byzantine term for an elite military force.35 Later a different 29

Seta B. Dadoyan, The Fatimid Armenians: Culture and Political Interaction in the Middle East (Leiden, 1997), p. 83. 30 Chiarelli, “Sicily during the Fatimid Age”, pp. 60–64. 31 Fasoli, “Le Città Siciliane”, p. 348; for a study of the sort of fortification that the Muslims themselves faced during campaigns against the southern Italian mainland, see Jean-Marie Martin and Ghislaine Noyé, “Guerre, fortifications et habitats en Italie méridionale du Ve au Xe siècle”, in Guerre, fortification et habitat dans le monde méditerranéen au Moyen Âge. Colloque organisé par la Casa de Velázquez et l’Ecole Française de Rome, Madrid, 24–27 novembre 1985, ed. André Bazzana (Madrid, 1988), pp. 225–36, and Carlo G. Mor, “La Difesa militare della Capitanata ed i Confini della Regione al Principio de Secolo XI”, in Studies in Italian Medieval History Presented to Miss E. M. Jamison, Papers of the British School at Rome, ed. Philip Grierson, John Bryan Ward-Perkins (Rome, 1956), pp. 29–36. 32 Pierre Guichard, “Animation maritime et developpement urbains des côtes de l’Espagne orientale et du Languedoc au Xe siècle”, in Occident et Orient au Xe siècle: Actes du IXe Congrès de la Société des historiens médiévistes de l’enseignement supérieur public (Dijon, 2–4 juin 1978) (Paris, 1979), pp. 187–207. 33 Guichard, “Animation maritime”, pp. 206–7. 34 Matthew Bennett, “Norman Naval Activity in the Mediterranean c.1060–1108”, AngloNorman Studies XV (1993), 49; John H. Pryor, “From Dromon to Galea: Mediterranean Bireme Galleys AD 500–1300”, in The Age of the Galley, ed. Ralph Gardiner (London, 1995), p. 107. 35 Nicola Cilento, “I Saraceni nell’Italia Meridionale nei Secoli IX e X”, Archivio Storico per la Province Napoletane 38 (1959), 111.

84

david nicolle

term, masnada, was used for the papacy’s small standing army, which then consisted of mercenaries and others, led by Rome’s aristocracy.36 The term masnada, as it was used in Italy and the Iberian peninsula, remains problematical, although it is usually thought to share the same Latin root as the medieval French word mesnie meaning a lord’s military retinue. However, it might be linked to the Arab words misnad or musanada meaning a support, or mustanid meaning trust or reliance upon. Arab and Islamic influence upon military terminology in southern Italy would not be surprising. By the 840s AD there were increasing numbers of Arabs and other Muslims in southern Italy, most notably in Apulia on the Adriatic coast, but also on the south-western coast of the mainland.37 They included raiders, settlers, and mercenaries. Local rulers like the Lombard prince of Salerno and Duke Benevento seemed unable to protect the area’s wealthy monasteries, perhaps because, once the Lombard elite established itself in southern Italy after being defeated by Charlemagne in the North, they ceased to be warrior leaders and instead became “guardians of the law”.38 While the conquest of Sicily pressed ahead, there were repeated Islamic attacks on the Italian mainland. In 925 and 928 AD raids launched from Sicily struck Byzantine-ruled Otranto, and in 994 AD another group attacked Matera. In 1002 AD Bari was targeted yet again, this time by “Saracen” raiders who may not all have been Muslim, but were commanded by an ex-Christian renegade called Luca who was known as Qa’id (commander) Safi.39 Medieval Christian sources usually give the impression that Muslim colonists on the European mainland were simply belligerent ghazis, raiding far and wide and taking advantage of quarrels between local Christian powers. In fact, some of these Islamic enclaves became centres of trade and of economic production, particularly timber for shipbuilding.40 During this turbulent period the, so-called, Saracens started establishing colonies on the south Italian mainland, including an autonomous amirate around Bari from 847 to 871 AD.41 Despite the destruction of this outpost in south-eastern Italy, a smaller but longer-lasting Muslim colony was estab36

Peter Partner, The Lands of St Peter (London, 1972), p. 195. Barbara M. Kreutz, Before the Normans, Southern Italy in the Ninth and Tenth Centuries (Philadelphia, 1991), pp. 24–25. 38 Huguette Taviani-Carozzi, La Principauté Lombarde de Salerne, IXe–XIe siècle (Rome, 1991), p. 442. 39 Cristian Guzzo, L’arrivo dei Normanni nel Meridione d’Italia tra fonti d’epoca e storiografia contemporanea (Tuscania, 2014), p. 25. 40 Philippe Sénac, Musulmans et Sarrasins dans le sud de la Gaule (VIIIe–XIe siècle) (Paris, 1980), pp. 101–6. 41 Paolo Cammarosano, Storia dell’ Italia medievale dal VI all’ XI secolo (Rome, 2001), p. 210; a detailed history of this little-known outpost of early medieval Islamic power on the mainland of Europe was written by Giosué Musca, L’Emirate di Bari 847–871 (Bari, 1992). 37



a gift to the normans 85

lished on the west coast, at the mouth of the river Garigliano, around 881 AD.42 Its inhabitants acknowledged the authority of the Aghlabid rulers of Ifriqiya and also established good relations with the Christian Italian coastal cities of Gaeta, Amalfi, and Naples. In fact the rulers of Gaeta had initially welcomed them as a buffer against the ambitions of the Lombard Count Lando III of Capua. The role of Muslim mercenaries in mainland southern Italy during the ninth and tenth centuries is an interesting phenomenon (fig. 49). The first local leader to summon Muslim assistance seems to have been Duke Andrea of Naples when, during his dispute with the prince of Benevento in 827 AD, he invited so-called Saracens to defend his city as auxiliaries.43 Eight years later Naples again hired Arab mercenaries, or perhaps more correctly “paid allies”, who seem to have come from Sicily.44 While this almost became an established policy, the harbours of Naples, Gaeta, and Amalfi were said to have served as safe havens from which Muslim ghazis could launch their naval jihad of razzia raids.45 At one time there even appears to have been an alliance, although not necessarily a formal one, between the Italian cities of Naples, Amalfi, Gaeta, Sorrento, Conza, Acerenza, and Saracens who almost certainly came mostly from Sicily.46 Later in the ninth century significant numbers of Muslim troops were in the service of Bishop Athanasius II of Naples, who later became Duke Athanasius of Naples from 848 until his death in 898 AD, serving alongside local Neapolitan troops and sometimes a small force sent by the Byzantine emperor in Constantinople.47 It was during the time of Bishop-Duke Athanasius that a Muslim community lived between the harbour and fortified walls of Naples, causing the papal chronicler Anastasius the Librarian (c.810–c.875 AD) to complain that Christian Naples had become a refuge for predatory Saracens, just like Palermo and “Africa” (Ifriqiya).48 Outside Salerno, Saracens were even raising crops at those times of year when naval raiding was not possible,49 and around the year 900 AD they were clearly permitted to enter Salerno if unarmed.50 42

Graham A. Loud, “Southern Italy in the Tenth Century”, in The New Cambridge Medieval History, ed. Timothy Reuter (Cambridge, 1999), vol. 3, p. 626. 43 Cilento, “I Saraceni”, p. 111. 44 Kreutz, Before the Normans, p. 20. 45 Armand O. Citarella, “The Relations of Amalfi with the Arab World before the Crusades”, Speculum 42 (1967), 305. 46 Cammarosano, Storia dell’ Italia medievale, p. 209. 47 Jules Gay, L’Italie Meridionale et l’Empire Byzantin (Paris, 1904), p. 137. 48 Bruno Figliuolo, “Amalfi e il Levante nel Medioevo”, in I comuni italiani nel Regno Crociato di Gerusalemme, ed. Benjamin Kedar (Genoa, 1986), p. 579. 49 Figliuolo, “Amalfi”, pp. 579–80. 50 F. Hirsch and M. Schipa, La Longobardia Meridionale (570–1077). Il Ducato di Benevento, Il Principato di Salerno (Rome, 1968), p. 151.

86

david nicolle

Radelchis, prince of Benevento, is said to have been the first Lombard ruler to recruit Saracen mercenaries in the 840s, although Siconulf the Lombard ruler of Salerno also did so a little later. The first troops involved came from North Africa but were not those who attacked Rome and Ostia in 846 AD.51 Later recruits included men from Al-Andalus and all seem to have been described as auxiliatores.52 The background to such recruitment was a civil war which had broken out following the death of Prince Sicard of Benevento in 839 AD. His former treasurer, Radelchis, retained control of the city of Benevento while Sicard’s surviving brother, Siconulf, established his authority in Salerno. Both men soon lost control of their Muslim allies or mercenaries, who thereupon seized a coastal base for themselves.53 The military influence of Siconulf ’s Saracen troops was such that there was a fear that he might himself convert to Islam; a fear earlier expressed about his brother Sicard.54 These stories were probably hostile propaganda, yet Siconulf of Salerno apparently permitted his Muslim supporters to establish a colony called Agliarini, not far from Latina, from where they campaigned against Radelchis of Benevento and Landulf of Capua.55 Also in the 840s AD, some Saracen mercenaries arrived in Apulia,56 raided far afield, then seized Bari and established an Islamic government which offered a somewhat tenuous allegiance to the ‘Abbasid caliph’s governor of Egypt. It survived for almost quarter of a century as an Islamic amirate. Indeed Bari and Taranto soon had significant Muslim populations, served by the usual Islamic facilities including mosques and markets. With the fall of the amirate of Bari and the rise of the Shi’a Fatimid caliphate in North Africa during the first decade of the tenth century, campaigning by Sunni Muslims on the Italian mainland was virtually cut off from rest of the Sunni world, especially after the loss of the colony at the mouth of the river Garigliano in 915 AD. Subsequent actions on the Italian mainland were on a smaller scale and few of those people involved came from Islamic Sicily.57 The failure of the Muslims’ colonies on the Italian mainland is less surprising than the fact that they endured for as long as they did. This was partly because of the Muslims’ high military and naval capabilities, but also because campaigns against them were largely undertaken by small Italian states with little help from outside.58 For example, the first assault upon the 51

Kreutz, Before the Normans, p. 30. Musca, L’Emirate di Bari, pp. 23–24. 53 Loud, The Age of Robert Guiscard, p. 16. 54 Cammarosano, Storia dell’ Italia medievale, p. 208. 55 Cristian Guzzo, L’ Esercito Normanno nel Meridione d’ Italia: Battaglie, Assedi ed Armamenti del Cavalieri del Nord (1016–1194) (Brindisi, 2013), p. 20. 56 Cilento, “I Saraceni”, p. 113. 57 Granara, “Political Legitimacy”, p. 91. 58 Norman Daniel, The Arabs and Medieval Europe (London, 1975), pp. 76–77. 52



a gift to the normans 87

Muslim colony at the mouth of river Garigliano in 903 AD was by Prince Atenulf I of Capua. It failed, not least because these settlers were helped by forces sent by the Hypatus or duke Docibilis I of Gaeta.59 Twelve years later it took a broad coalition of the papacy, the Byzantines, the northern Italian Kingdom of Italy, the local Lombard states, plus the cities of Naples, Salerno, and Gaeta, to overcome the fortified Garigliano colony. Thereafter Muslim raiding continued on a smaller scale into the early tenth century, but by this time many of the raiders – Arabs and Berbers – began to settle down in various parts of southern Italy. According to local traditions and some written records, they married local women, adopted local customs and eventually converted to Christianity.60 In fact, there were still communities of identifiable Saracen origin in southern Italy when the Normans first appeared in the early eleventh century.61 In Reggio di Calabria about 13 per cent of the names in fifty documents from the bishopric of Oppido Mamertina, dating from after 1050 AD, were of Arab origin.62 Although these individuals were now Christian, they are likely to have been descended from earlier settlers rather than being recent arrivals from Sicily. The Italian mainland was not, of course, alone in being targeted by revived Islamic raiding in the late tenth and eleventh centuries. Thus, in 1019 AD some Muslim raiders were captured during an unsuccessful attack near Narbonne in southern France. Twenty notably tall captives were sent to Limoges, where the abbot kept two in his own service while giving away the rest as slaves. According to a chronicler, these prisoners did not speak Saracenic (Arabic) but a dialect which sounded like the barking of puppies.63 Perhaps the unfortunates were Berbers. Meanwhile there was a revival of maritime jihad targeting Italy, even attacking the Saracens’ old ally Amalfi.64 This culminated in an assault upon Salerno in 1016 AD which prompted the Lombard Prince Gaimar III of Salerno (sometimes confusingly referred to as Gaimar IV) to continue his recruitment of warriors from northern France. The Normans had arrived. During this period local aristocracies were taking control across much of southern Italy. Some focused on mercantile matters, others on controlling agricultural productions, while a few had significant political ambitions.65 A similar process was taking place within Islamic Sicily, which saw fragmen59

Loud, “Southern Italy in the Tenth Century”, p. 627. Haroun K. Sherwani, “Incursions of the Muslims into France, Piedmont and Switzerland”, Islamic Culture 5 (1931), 85. 61 Cristian Guzzo, L’ Esercito Normanno nel Meridione d’ Italia, p. 20. 62 Loud, The Age of Robert Guiscard, p. 52; Jeremy Johns, Arabic Administration in Norman Sicily. The Royal Diwan (Cambridge, 2002), p. 31. 63 Sherwani, “Incursions of the Muslims”, p. 109. 64 Loud, The Age of Robert Guiscard, pp. 28–29. 65 Giuseppe Galasso, “Social and Political Developments in the Eleventh and Twelfth 60

88

david nicolle

tation verging on anarchy. Writing over two centuries later, the Arab chronicler Ibn al-Athir described a Sicilian crisis which started in 388 AH (the Islamic year from 3 January to 23 December 998 AD), when the amir, who ruled the island under Fatimid suzerainty, fell victim to partial paralysis. He put one of his sons, Ja‘far, in charge but another brother, ‘Ali, rose in revolt, “supported by the Berbers and ‘abid (black slave troops)”. Ja‘far sent a jund (local army corps) against ‘Ali (1015 AD), killing many Berbers and ‘abid. Ja‘far then expelled all Berber troops to Ifriqiya and slaughtered the ‘abid “without exception” (nevertheless, soldiers with exaggerated African features still appeared in Siculo-Norman art; figs. 33c and perhaps 64b). From then on Ja‘far’s jund (provincial army) was recruited exclusively from Sicilians, greatly reducing its size, which “lighted the lusts of its inhabitants against the rulers”. Within a short time Ja‘far was besieged within his own palace, in 1019 AD. The situation became progressively worse, with several changes in ruler, until a group of Sicilian notables went to al-Mu‘izz, the Zirid ruler of Ifriqiya, demanding that he take responsibility for Sicily, otherwise they would hand the island over to the Christians. But other Sicilian leaders opposed a Zirid takeover, saying, “you have called in a stranger to govern you. By God, all this cannot end well”, and decided to fight. Al-Mu‘izz’s army was defeated, lost eight hundred men and re-embarked for Ifriqiya. This was followed by the fragmentation of Sicily, as several regional senior qa’id army officers proclaimed their independence and set about fighting each other.66 A Byzantine attempt to take advantage of the chaos and reconquer the island narrowly failed in 1037 AD. Thereafter none of the competing qa’ids were strong enough to dominate all Sicily and a few years later one of them sought assistance from the Norman, Roger of Calabria.67 Worse was to come and, in the light of such turbulent events, the Sicilian Muslims’ failure against the Norman invaders seems hardly surprising.68 Even in battle there was more to the Normans’ success and the Sicilians’ failure than differences in tactics and military equipment, although these did play a significant role.69 Islamic armies of this period were not all recruited and organised in the same ways, although there were established ideals, largely based upon ‘Abbasid practices in Islam’s Middle Eastern heartlands. In distant or frontier provinces such as North Africa or Sicily, financial constraints and Centuries”, in The Normans in Sicily and Southern Italy. Lincei Lectures 1974 (Oxford 1977), pp. 48–49. 66 Pierre Guichard, L’ Espagne et la Sicile Musulmanes aux XIe et XIIe siècles (Lyon, 1991), pp. 46–47. 67 Loud, The Age of Robert Guiscard, p. 148. 68 Chiarelli, “Sicily during the Fatimid Age”, pp. 67–69. 69 Giovanni Amatuccio, “Fino alle mura di Babilonia, Aspetti militari della conquista normanna del Sud”, Rassegna Storica Salernitana 30 (1998) (https://bit.ly/2TmrJeP).



a gift to the normans 89

a more limited pool of potential recruitment meant that such ‘Abbasid ideals could not be achieved. Hence the Aghlabids of Ifriqiya sometimes had to resort to press-gangs to enlist sufficient men for their local jund.70 Armies and recruitment had also changed over time. For example, the force which invaded Spain in the eighth century was largely illiterate and only superficially Muslim, whereas the army which invaded Sicily in the ninth century included many who could read and write, alongside recognised scholars.71 Indeed, this Aghlabid force was a microcosm of North African Islamic society, being very mixed but dominated by the trained jund, religiously motivated but also including mercenaries.72 The larger number were Berbers, although Arabs formed the elite. It also seems that many of the gazi volunteers left Sicily after its conquest appeared secure, seeking outlets for their religious enthusiasm elsewhere. Meanwhile Aghlabid Sicily, as a centre of jihad, saw the instillation of a new military elite. The organisation of the Aghlabid army was largely based upon the Arab tribes who had settled in Ifriqiya in the late seventh and eighth centuries.73 Its troops were probably paid at clearly defined times, with cavalry receiving twice as much as infantry because of the greater cost of their horses and equipment.74 In addition to Berbers and Arabs there were other troops such as the saqaliba and ‘abid. The former were of supposedly Slav origin and came from Europe. Other “white” troops included the fata and the mawali, although the latter would be disbanded by Ibrahim II (875– 902 AD) because they rebelled, reportedly being replaced by the ‘abid, black soldiers mostly of sub-Saharan African slave origin. Black African troops had long been recruited into Islamic armies, for example by the Tulunid governors of Egypt (868–905 AD),75 under whose rule they played a major role which continued to a lesser extent under the Ikhshidid governors of Egypt (936–969 AD).76 However, the establishment of the Aghlabid dynasty in Ifriqiya diverted much of the trans-Saharan trade in such slave recruits.77 70

John F. P. Hopkins, Medieval Muslim Government in Barbary, until the Sixth Century of the Hijra (London, 1958), p. 72. 71 Granara, “Political Legitimacy”, p. 105. 72 Granara, “Political Legitimacy”, pp. 94–96. 73 Mohamed Talbi, L’Emirat Aghlabide 184–296, 800–909. Histoire Politique (Paris, 1966), passim. This provides a detailed study of the Aghlabid army, though focusing on political events and recruitment rather than on details of structure or equipment. 74 Hopkins, Medieval Muslim Government in Barbary, pp. 71–72. 75 Beshir Ibrahim Beshir, “The Fatimid Caliphate 386/996–487/1094” (PhD thesis, London University 1970) 38–44; Beshir Ibrahim Beshir, “Fatimid Military Organization”, Der Islam 55 (1978), 40–41. 76 Kizobo O’Bweng-Okwess, “Le Recrutement des Soldats Négro-Africains par les Musulmans du VIIIe au XIIe Siècle”, Journal of Oriental and African Studies 1 (1989), 25. 77 Daniel Pipes, “Black Soldiers in Early Muslim Armies”, International Journal of African Historical Studies 13 (1980), 94.

90

david nicolle

Initially the ‘abid appear to have been employed as sword bearers and bodyguards by the first Aghlabid governor.78 Later these ultra-loyal black Africans became an infantry elite in the Aghlabid army from the later ninth century onwards,79 and were also found in positions of high command. The early Fatimid dynasty, which had its capital and powerbase in Ifriqiya from 909 to 973 AD, fielded an army largely consisting of Berbers, mainly from the Kutama tribe (figs. 56 and 57). Almost for the first time, and to the surprise of their opponents, the previously fragmented clans or extended families of the Berber tribes were moulded into an army divided into seven parts. These asba‘ (sing. sub‘) “sevenths” each formed an askar regiment under a muqaddam officer.80 This still basically tribal system persisted until shortly after the Fatimid conquest of Egypt, with only a relatively small number of household troops forming an elite formation close to the person of the Fatimid caliph.81 Like their Aghlabid predecessors, the Fatimids recruited substantial numbers of sub-Saharan African soldiers of slave origin. These ‘abid are said to have normally been armed with swords but, as foot soldiers, must surely have used spears and perhaps other weapons during battle.82 Other troops whom the early Fatimids either inherited from the Aghlabids or continued to recruit in small numbers included the rum, who are usually interpreted as Anatolian, ex-Byzantine Greeks or other Orthodox Christians,83 and the saqaliba white slave-recruited ghulams, supposedly of Slav or at least northern European origin. Their numbers were always small, forming an elite body of administrators and military commanders,84 their previous role as a ruler’s bodyguard having been given to Kutama Berbers.85 Subsequently, the Fatimids also recruited Arab soldiers from Syria, many of them veterans of the effective and well-equipped Hamdanid forces of what are now the Syrian–Turkish frontier provinces.86 Of tribal although not necessarily genuinely Bedouin origin, such Arabs included fast-moving and manoeuvrable, fully trained, spear- and sword-armed cavalry (fig. 46),87 some of whom were fully armoured (figs. 43 and 45a). 78

Hopkins, Medieval Muslim Government, p. 72. O’Bweng-Okwess, “Le Recrutement”, p. 25. 80 Michael Brett, The Rise of the Fatimids: The World of the Mediterranean and the Middle East in the Fourth Century of the Hijra, Tenth Century CE (Leiden, 2001), p. 90. 81 Beshir, The Fatimid Caliphate, pp. 74–76. 82 Beshir, “Fatimid Military Organization”, p. 38. 83 Yaacov Lev, “The Fatimid Army, A.H. 358–427/968–1036 C.E.: Military and Social Aspects”, Asian and African Studies 14 (1980), 169–71. 84 Beshir, The Fatimid Caliphate, pp. 34–38; Beshir, “Fatimid Military Organization”, p. 41. 85 Hopkins, Medieval Muslim Government, p. 73. 86 Lev, “The Fatimid Army”, pp. 171–72, 176–79. 87 Beshir, The Fatimid Caliphate, p. 52. 79



a gift to the normans 91

All these types of troops served in greater or lesser numbers in Sicily when the island was under Fatimid rule. Similarly, Sicilians served in the Fatimid army outside their own island. Indeed a substantial Christian contingent campaigned under Jawhar, who himself had a Sicilian background, during the Fatimid conquest of Egypt in 969 AD. Although the Fatimids would lose control of Sicily, they were occasionally recognised as overlords by the island’s competing war-lords and undoubtedly remained a major source of influence, not least in military matters. As the Fatimid army changed following the conquest of Egypt, so its influence upon Sicily surely also changed. The army of the Zirid dynasty in Ifriqiya (972–1152 AD), which played a major role in Sicilian affairs during the last century of Islamic rule, was at first much the same as the preceding Fatimid regional army or jund. Its main elements were Berbers of the Sanhaja tribe, who also formed the senior officer corps, plus the local jund militia and “black” ‘abid.88 Within the Zirid military establishment the habashi may have been Abyssinians of unfree origin, while their close association with the largely Byzantine Orthodox Christian rum might indicate that they may themselves have been of Christian Ethiopian origin.89 To add further confusion, the much smaller number of white European slave-recruited soldiers were also sometimes referred to as ‘abid.90 Meanwhile, on the Italian mainland during the tenth century, numbers of Greek-speaking ex-Byzantine soldiers also converted to Islam, perhaps the most significant group being led by a man recorded as “Bomar” (Abu Amir?) and operating in the Basilicata where, for some considerable time, they held the stronghold of Pietrapertosa near Tricarico.91 Across the straits in Sicily, the island was nominally ruled by the Kalbite dynasty from 948 to 1053 AD. Here, as elsewhere, elites changed over time. Hence the original Arab military elite had, to some extent, been replaced by Kutama and other Berbers following the overthrow of the Aghlabids by the Fatimids. They, in turn, would later lose their dominant position. In general, however, the tenth century saw a remarkable fusion of Arabs, Berbers, and Sicilian converts in Sicilian Islamic society,92 and in Sicilian armies (figs. 15–20 and 26–27). Ibn Hawqal, who visited Kalbite Sicily in the mid-tenth century, stated that the largest quarter of Palermo was inhabited by saqalibah.93 He was, 88

Hady R. Idris, La Berbérie Orientale sous les Zirides Xe–XIIIe siècles (Paris, 1962), p. 529. 89 Idris, La Berbérie Orientale, p. 530 n. 98. 90 Idris, La Berbérie Orientale, p. 530. 91 André Guillou, “Inchiesta sulla popolazione greca della Sicilia e della Calabria nel Medio Evo”, Rivista Storica Italiana 75 (1963), 56. 92 Chiarelli, “Sicily during the Fatimid Age”, p. 123. 93 Chiarelli, “Sicily during the Fatimid Age”, pp. 56–57.

92

david nicolle

however, shocked by what he regarded as the poor behaviour of those volunteers who garrisoned or perhaps merely inhabited the coastal defence ribats. Rather than being centres of piety and jihad, Ibn Hawqal wrote that they were full of hypocrisy, fractious idlers, evildoers, men of bad conduct, people of sedition, and trash.94 Perhaps these ribats were actually centres of Sunni Islam, and thus suspect to the Shi’a Fatimid caliphate who were still the Kalbite governors’ suzerains, their inhabitants mostly being ascetics rather than warriors.95 Interestingly, Ibn Hawqal also complained that these men ate too many onions, which harmed their brains and confused their senses96 – a passion for onions having similarly been noted a decade or so earlier among the northern Iranian infantry of Buwayhid Iran and Iraq.97 There may also have been continuing tensions and jealousies between the Arabised Sicilian converts to Islam and a still identifiable “old Arab” elite. During the Kalbite period the Greek-speaking Orthodox Christians were rarely mentioned in a military context, but the early years of the Norman invasion showed that they were still willing to fight. For example, Geoffrey Malaterra described how the Greeks of Troina turned upon the Normans and received help from their Muslim neighbours: With the city now cut in two [Count Roger’s troops having won control of part of it], the Greeks built a barricade for their protection between themselves and the Normans. The Saracens from the neighbouring castra [the garrison of the fortified citadel], who were about 5,000 in number, were overjoyed to hear that the Greeks were fighting with our troops and moved rapidly to help them. Their 98 assistance greatly benefited the Greeks.

New eyes brought a fresh view of Sicily’s distinctive culture at the time of the Norman invasion, as when the garrison of Messina confidently emerged from its fortified city to confront the first Norman invaders in 1061 AD. They were nevertheless defeated with apparent ease. Again according to Geoffrey Malaterra: The people of Messina thought that because some of his [Count Roger’s] men had re-embarked on the ships they could easily defeat his divided forces. Cavalry and infantry left the city and marched out to attack him ... When Count Roger realised that they were 94

Loud, The Age of Robert Guiscard, p. 17 n. 11. Chiarelli, “Sicily during the Fatimid Age”, pp. 56–57. 96 Chiarelli, “Sicily during the Fatimid Age”, pp. 56–57. 97 Tanukhi, Muhassin Ibn ‘Ali al-, The Table-Talk of a Mesopotamian Judge, trans. and ed. David S. Margoliouth (London, 1922), pp. 95–96. 98 Geoffroi Malaterra [Gaufredo Malaterra], De Rebus Gestis Rogerii Calabriae et Siciliae Comitis et Roberti Guiscardi Ducis fratris eius, Rerum Italicarum Scriptores. Raccolta degli Storici Italiani, ed. Ernesto Pontieri, vol. 5 (Bologna, 1927–28), Book 2, part 29. 95

a gift to the normans 93



advancing against him, he sent out ahead his nephew Serlo ... with instructions that if they wished to flee, as indeed they did, they should be allowed to do so. He himself pursued them [the Muslims] at great speed while they attempted to flee, and intercepted them to 99 such effect that scarcely one among the whole multitude escaped.

Subsequently Roger “set off to attack the city at daybreak, knowing its forces to be much depleted. But although those who now survived in Messina were few in number, they and their women along with them defended their towers and ramparts as though for life itself.”100 Faced with such determined resistance, Roger sailed back to Calabria. The organisation of local military forces in Sicily had been traditional, with the island divided into iqlim districts, each with its own jund or a sub-unit of the jund, particularly in the most Islamised western part of the island.101 After defeating those Byzantines still holding out in northwestern Sicily, the Fatimid Caliph al-Mu‘izz ordered the island’s Muslim authorities to build fortified places, either citadels or castles, in every district as a defence against Byzantine counter-attacks. He also tried to get “all Muslims” – by whom he probably meant all members of the jund – to live in these fortifications.102 In 975 AD a new governor named Abu’l-Qasim had the recently abandoned Byzantine citadel of Rometta restored and installed one of his black slaves, perhaps one of the elite ‘abid, as its commander.103 Next, Abu’l-Qasim expelled the last Byzantine defenders from Messina and followed them across the Straits to Calabria, while his brother took another force to raid Byzantine Apulia. Meanwhile the German Emperor Otto III had been trying to replace Byzantine authority in southern Italy with his own. This resulted in a clash with the Muslim raiders, forcing the latter to pull back. The next meeting between the Siculo-Muslim army and Otto’s imperial army, at the battle of Capo Colonna (14 July 982 AD), had a different outcome. Initially the Germans broke through the Muslim lines, causing disorder and near panic when the amir Abu’l-Qasim was killed. Yet the Sicilians regrouped, counter-attacked and forced Emperor Otto to withdraw to Rome, while Otto’s allies, Landulf IV of Capua and Pandulf, the deposed prince of Salerno, were both killed.104 Regrouping and then achieving victory after the death of a commander was extremely rare in medieval warfare, so the

99

Malaterra, De Rebus, Book 2, part 5. De Rebus, Book 2, part 6. 101 Ahmad, Islamic Sicily, p. 38. 102 Chiarelli, “Sicily during the Fatimid Age”, pp. 51–52. 103 Chiarelli, “Sicily during the Fatimid Age”, p. 58. 104 Chiarelli, “Sicily during the Fatimid Age”, p. 60; Loud, “Southern Italy in the Tenth Century”, p. 643. 100 Malaterra,

94

david nicolle

battle of Capo Colonna was a tribute to Siculo-Muslim and Fatimid discipline and military cohesion. The local units established in Sicily around this time, each under its own qa’id commander-governor, were still in existence when the Normans invaded the island. In fact they continued to exist under Norman rule.105 Another feature which would survive into the Norman period was the iqta‘, a sort of fief, which had been introduced into Sicily by the Aghlabids.106 It would be further refined by the Fatimids in Egypt, where the iqta‘ jayshi or “army iqta‘” eventually had two forms. The older was linked to individual soldiers of various origins,107 whereas a newer form of iqta‘ was, according to the later medieval Egyptian historian al-Maqrizi, given to someone who collected revenues from a specified area.108 This money then went directly to the government. Then there was the Fatimid iqta‘ i’tidad, which was not attached to an individual soldier but was administered through an officer, with its revenues intended for a specified number of troops.109 Whether these systems were used in Fatimid and Kalbite Sicily is unclear, but their similarity with aspects of Siculo-Norman military administration, costume, and equipment (fig. 14) suggests that they were. The similarity between Fatimid Egyptian and Norman Sicilian military structures was also apparent in the later Fatimid diwan al-juyush “army ministry”. This was a lesser organ of government, having limited but defined duties, including responsibility for the jara‘id register of the names and fiefs of iqta‘ holders following the Norman reforms of 1074 AD. Mustering an army obviously varied according to the urgency of a situation. Nevertheless, when Islam was on the offensive and jihad campaigns or ghaza raiding were an almost annual occurrence, traditional patterns developed. For example, the Kitab al-Anwa, an Andalusian agricultural manual written in Arabic and Latin in 961 AD, stated that enlistment for summer campaigns normally began on 28 February, while it was considered safe for ships go to sea from 13 April onwards. Rulers and military commanders were identified by emblems of sovereignty or rank. Those for subordinate rulers and governors were given by those more senior, such as when the Fatimid caliphs of Cairo continued to send drums and banners to Zirid governors of Ifriqiya and Sicily.110 The military rank of qa’id has already been mentioned, while within armies such as those of the Zirids each ‘irafa unit was led by an ‘arif or lower ranking officer. 105

Bresc, “Mudejars des Pays”, p. 53. Islamic Sicily, pp. 23 and 38. 107 Caude Cahen, “L’Administration Financière de l’armée Fatimide d’après al-Makhzumi”, Journal of the Economic and Social History of the Orient 15 (1972), 171–73. 108 Cahen, “L’Administration Financière”, 172–73. 109 Cahen, “L’Administration Financière”, 172–73. 110 Idris, La Berbérie Orientale, pp. 516, 522. 106 Ahmad,



a gift to the normans 95

The more specific term of ‘anbar seems to have been used for an officer of ‘abid, black African origin.111 Detailed information is largely missing for the armies of Islamic Sicily, but it seems highly likely that they would have been essentially the same as better-documented forces in the Middle East, such as the garrison of Tarsus in Cilicia shortly before it fell to the Byzantines in 965 AD. Here resident cavalry units had infantry units attached to them while the trainee ghulams (elite soldiers of unfree origin) were under the charge of a respected older man referred to as their shaykh.112 Meanwhile, within the Fatimid military system, men could move from cavalry to infantry and vice versa, with resulting changes to their status and pay being noted in the registers of the diwan al-jaysh.113 There is an abundance of information about military training in medieval Islamic armies, but some of it was merely to provide an educated reader with the sort of knowledge that a pious Muslim ought to have. Thus the highly traditional manual written by Badr al-Din Ibn Juma’h early in the fourteenth century harks back to an ideal of military training and weaponry from the earliest decades of Islam. He therefore maintained that when a ruler organised archery competitions and horse races he was merely wasting money from the public treasury.114 The whole subject of medieval Islamic furusiyya military training manuals is also complicated by the fact that authors from the thirteenth to fifteenth centuries often copied, simplified, or slightly updated earlier texts which had been written for ‘Abbasid caliphs in the ninth and tenth centuries. Nevertheless, modern scholarship is identifying early texts embedded within later works.115 Thus we have a reasonable idea of what was taught in the Fatimid Caliphate’s hujariyya training barracks,116 this being based upon ‘Abbasid hujariyya training establishments – at least as far as the Fatimids’ more constrained resources would allow.117 Some even more limited versions of this training would also have been found in provincial capitals. Nonmilitary texts, such as works on geography and commerce, also include military information. That is how we know that in tenth-century Tarsus some sibyan younger soldiers under training were equipped with weapons suitable to their age and physical size. This included crossbows as well Persian bows 111

Idris, La Berbérie Orientale, pp. 529–30. Clifford E. Bosworth, “Abu ‘Amr ‘Uthman al-Tarsusi’s Siyar al Thughur and the Last Years of Arab Rule in Tarsus (Fourth/Tenth Century)”, Graeco-Arabica 5 (1993), 191–94. 113 Cahen, “L’Administration Financière”, p. 168. 114 Badr al-Din Ibn Juma’h Muhammad Ibn Ibrahim, Mustanad al-Ajnad fi ‘Alat al-Harb, ed. Usamah Nasir al-Naqshbandi (Baghdad, 1983), p. 115. 115 Shihab al-Sarraf, “Mamluk Furusiyah Literature and its Antecedents”, Mamluk Studies Review 8 (2004), 141–200. 116 Cahen, “L’Administration Financière”, p. 170. 117 Beshir, “Fatimid Military Organization”, pp. 46–48. 112

96

david nicolle

of composite construction. Meanwhile sibyan of supposedly Yemeni tribal origin were issued with Arab bows, which were probably larger, all-wood weapons for infantry use.118 Around the same time in neighbouring Syria the poet Kushajim al-Sindi, born in Palestine of Indo-Persian parents, described how hunters learned how to hit a moving target by shooting at a stuffed animal mounted on a cart which was then rolled downhill.119 Another feature which demonstrated the authority of Islamic governments was the fact that carrying weapons, even by soldiers, was strictly controlled inside towns. Thus, when the man who would become the first Fatimid caliph had to flee across North Africa early in the tenth century, he and his party reached Setif in what is now eastern Algeria. Here they were refused entry until they deposited their weapons with the gate guardians. They feared that the weapons would be lost or seized by the Aghlabid authorities, but in fact they were handed back when the travellers left after a few days.120 A century or so later the Zirid authorities in Ifriqiya attempted to maintain a similar degree of control over their towns where an urban police, militia, or gendarmerie called the hars patrolled at night with dogs and trumpets. They enforced “lights out” after the town gates were closed. It was also illegal to leave town after dark.121 Perhaps even more astonishing for north European visitors to Norman Sicily was the barid postal service which the Normans had inherited from their Muslim predecessors. It included a pigeon-post between Sicily and Ifriqiya, via the island of Pantellaria.122 Not all fighting men in Islamic Sicily were professional soldiers, the ghazi volunteers ranging from religiously motivated enthusiasts to booty-seeking adventurers. In fact some historians suggest that many of those who raided the Italian mainland were already on the margins of “respectable” Islamic society, because of either their ethnic origins or previous lives. The motivation of Muslim mercenaries recruited by various southern Italian rulers is similarly unknown. They came from many areas including Sicily, Libya, Crete, and Al-Andalus.123 Were they looking money – which seems unlikely, given the traditionally higher pay given to professional soldiers in the medieval Islamic world than in western Europe or the Byzantine Empire?124 Were 118

Bosworth, “al-Tarsusi’s Siyar al Thughur”, p. 193. Muhammad M. Ahsan, Social Life under the Abbasids 176–289 AH, 786–902 AD (London, 1979), p. 223. 120 Marius Canard, “L’autobiographie d’un chamberllan du Mahdi ‘Obeidallah le Fatimide”, Hesperis (1952), 300. 121 Idris, La Berbérie Orientale, p. 527. 122 Idris, La Berbérie Orientale, p. 527. 123 Lindsay S. G. Matheson, “The Norman Principality of Capua (1058–1098) with particular reference to Richard I (1058–1078)” (PhD thesis, Oxford University 1974), p. 9. 124 Eliyahu Ashtor, Histoire des Prix et des Salaires dans l’orient médiéval (Paris 1969), passim. 119



a gift to the normans 97

some of them disgraced men who needed to find military employment outside Islamic territory? Or had some of them been sent by Islamic governments seeking an alliance with one of the south Italian states? Those who remained in southern Italy eventually integrated into local society,125 which suggests that they already felt themselves to be exiles. By the ninth and tenth centuries there were clear differences in the military equipment used in the western and eastern regions of the Islamic world, partly because of the survival of local traditions and partly because earlier styles of combat persisted in the West. There was also an exchange of ideas, fashions, and weapons across the frontier with Christian Europe. Nevertheless, the eastern, central, and western regions of the early medieval Islamic world shared a conscious Muslim identity, so that newer Middle Eastern fashions carried high status in many aspects of Islamic culture. Long-distance trade in weapons, armour, and horse harness was also a notable feature of Islamic civilisation during this period. In such a culturally conservative society it is not surprising that there was continuing interest in the tactics, skills and military gear of early Islamic heroes. Thus, according to Badr al-Din Ibn Juma’h, the Prophet Muhammad had both a dir‘ mail hauberk and a jubbah, which is usually interpreted as a mail-lined, cloth-covered garment with integral padding.126 He also mentioned that two dir‘ hauberks could be worn at once, which was confirmed in furusiyya training manuals.127 During the period under consideration the poet Ibn Hani’, writing for the Fatimid caliph al-Mu’iz (953–975 AD), was scathing about the military of his own Andalusian homeland: They do not know how to charge their well-bred horses and they are unable to endure the melée and the intercrossing of lances. They pull out of its scabbard a sword with a fearsome point, but in their hands the steel becomes white iron [soft iron]. Their mail coats never get smeared with blood in war, but they are on the battlefield like servants with the shits ... Their blades do not come out of the scabbards of their swords and their dynasty waddles softly, like a flirty woman with a flexible waist ... The difference between them and you is like the difference between the hard lances of nab’ [wood] and the feeble reeds of breakable 128 flutes.

125

Matheson, The Norman Principality of Capua, p. 9. Badr al-Din Ibn Juma’h, Mustanad al-Ajnad, p. 58. 127 Badr al-Din Ibn Juma’h, Mustanad al-Ajnad, pp. 62–63. 128 Marius Canard, “L’Impérialisme des Fatimides et leur propagane”, in L’Expansion Arabo-Islamiques et ses Repercussions, ed. Marius Canard (London, 1974), p. 167. 126

98

david nicolle

Ibn Hani’ then praised Fatimid forces who march “fi washi al-dilas suwaighan (in sparkling long mail hauberks, covered with embroidered cloth” (figs. 43 and 46).129 The Muslim armies which conquered Sicily were said to be more like those of the Middle East than of Al-Andalus, while Latin European sources suggest that most of these troops had rather light equipment.130 However, other evidence points to something different where the elite or professional soldiers were concerned. Thus, during the initial Islamic conquest of Mazara, the Muslim general Asad asked his renegade Byzantine ally Euphemius to keep his men separate from the Muslims and to use “certain signs” by which they could be distinguished in battle.131 This could mean that Asad feared that his own men might attack all Byzantine soldiers, being unable to distinguish friend from foe, or that he wanted to be able to distinguish between his own Muslim troops and Euphemius’ followers, who might look quite similar (fig. 53). A combat between Muhammad Ibn Sahnun Ibn Sa’id and a Byzantine champion around the time of the Muslim invasion of Sicily also provides details about military equipment. The enemy horseman was armed with a sword, spear, and leather shield, whereas Ibn Sahnun, although riding only a mule, had a mail hauberk, sword, and spear.132 Towards the end of the tenth century, in a poetic source, another hero was said to have two swords.133 This might be a poetic fancy, but there is evidence from slightly later in the Middle East that elite Muslim cavalry could carry one sword in a belted scabbard and a second, sometimes longer sword in a scabbard attached to their saddle. In Zirid Ifriqiya only military leaders and the highest-status troops were equipped with dir‘ mail hauberks and baydah helmets, while the most common weapons were sword, long slender spear, and probably javelins, plus daggers (fig. 56 and 57a). All, or the great majority, of archers fought on foot with qaws bow and siham arrows (fig. 57b). It seems unlikely that crossbows had yet reached the area from the Middle East or Europe, but the qaws al-bunduq pellet bow was used for hunting birds.134 129

Alfred F. Von Kremer, “Uber detn Shi’itischen Dichter Abu’l Kasim Moh. Ibn Hani”, Zeitschrift der Deutschen Morgenlandischen Gesellschaft 24 (1870), 485–86; Canard, “L’Impérialisme des Fatimides”, p. 168. 130 Francesco Gabrieli, “Gli Arabi in Spagna e in Italia”, in Ordinamenti Militari in Occidente nell’alto Medioèvo, Settimane di Studio del Centro Italiano di Studi sull’alto Medioèvo 15/2 (Spoleto, 1968), pp. 717–18. 131 M. A. Ageil, “Naval Policy and the Rise of the Fleet in Ifriqiyyah from the 1st to 3rd Centuries A.H. (7th to 9th Centuries A.D.)”, PhD thesis (University of Michigan, 1985), p. 5. 132 Michele Amari, Biblioteca Arabo-Sicula (Turin-Rome, 1880–81), 1:310–11, Arabic text p. 187; Michele Amari, Storia dei Musulmani di Sicilia (Catania, 1933), 1:408. 133 Amari, Storia dei Musulmani di Sicilia, 2:392. 134 Idris, La Berbérie Orientale, pp. 532–3.



a gift to the normans 99

Other evidence for the use of armour in Islamic Sicily, and in some cases the lack of it, is found in European accounts of the Norman conquest. For example, Geoffrey Malaterra described a fight between Count Roger and a Muslim leader from Messina: Count Roger rode ... in front of his companions, eyes intently scanning all around him. He was unarmed except for his shield and the sword hanging from his belt – an armiger [squire] followed with his armour [armour normally being put on only when battle was imminent]. When by the light of the moon he observed the enemy’s arrival, he had gone too far in front of his squire to take his armour from him, indeed it was possible that the latter ... had fled. So he [Roger] put on speed and charged his enemy, armed only with a sword. He killed him with a single blow, cutting him in half. The [obviously unarmoured] body lay in two pieces – the [slain enemy’s] 135 horse and personal effects he gave to one of his men.

While a lack of armour might prove fatal, so could the wearing of it, as when Ibn al-Wird was drowned in Syracuse harbour in 1085 AD during a sea battle with the Normans. Ibn al-Wird was first hit by a javelin, then faced a Norman boarding party. As the wounded man attempted to leap aboard a neighbouring Islamic ship he fell into the sea and sank,136 pulled down by the iron of his otherwise unspecified armour; “in mare cum pondere ferri demurgitur” (figs. 15–20, 26 and 27).137 Geoffrey Malaterra offers a more detailed insight into the sort of armour worn by the Sicilian Islamic military elite when describing the death of “Arcadius of Palermo” during the battle of Cerami in 1063 AD. This man is usually assumed to be the qadi or chief judge of Palermo, although I think him more likely to have one of the senior qa’id military officers. Whoever “Arcadius” was, he was clearly wearing more armour than most others in the Sicilian army. Supposedly he was also slain by Count Roger in person: Comes [Roger] ergo, aciei suae praevius exhortator, Archadium de Palerna, suam aciem nostris exprobando prompissime antecedentem et splendenti clamucio – quo pro lorica utuntus – armatum, certamine inito, fortissimo congressu hastili robore dejectum, ceteris metum incutiens, interfecit. Erat enim inter suas militia praeclarissimus, cui etiam neminem armis resistere posse putabant; et clamucium quo undutus etar nullis armis poterat violari, nisi ab imo in superius impingendo inter duo ferrea, quo per juncturas concat138 enata sunt, ingenio potius quam vi vitiaretur. 135

Malaterra, De Rebus, Book 2, part 4. Gabrieli, “Gli Arabi in Spagna”, pp. 717–18; Loud, The Age of Robert Guiscard, p. 172; Malaterra, De Rebus, Book 4, part 2. 137 Malaterra, De Rebus, Book 4, part 2, lines 19–20. 138 Geoffroi Malaterra, Histoire du Grand Comte Roger et son frère Robert Guiscard, ed. 136

100

david nicolle

The French translation by M.-A. Lucas-Avenel, who also edited the text, rendered the difficult terminology of the Muslim champion’s armour as follows: Archadius ... etain arme d’un clamusium etincelant, qui tient lieu de broigne a ces gens ... on pensait meme que personne ne pouvait jui resister au combat; et aucune arme ne pouvait forces le clamusium dont il etait revetu, a moins de le disloquer en usant plus d’habilete que le force, par des coups de bas en haut, entre deux plaques en fer, a l’endroit ou elles etaient fixees par 139 des rivets.

Lucas-Avenel suggests that the word clamusium may have been altered by copyists of the manuscript, but might have stemmed from the Greek klamis. However, I believe that the editor is probably incorrect in following earlier scholars in assuming that the armour in question was a leather jerkin covered with riveted metal scales.140 Most apparent illustrations of scale armour in this period are, in reality, either forms of lamellar cuirass (fig. 24a–b, 31, 33d–f and 53a) or simple renderings of mail (figs. 14a, 50). There is little evidence of such scale cuirasses being used in the Islamic world at this date, while the evidence for their use in earlier centuries is dubious. Similarly, evidence for scale-lined armour in the later medieval Islamic armoury does not predate the fourteenth or, at most, the thirteenth century.141 Furthermore, the original Latin text merely states that the iron elements are linked to one another. Only the mention of an upwards thrust getting between the elements might suggest downwards overlapping scales, rather than the normally upwards overlapping elements of a lamellar cuirass. In any case, an upwards thrust could also have slid between the horizontally overlapping lamellar elements. Taking all this evidence into consideration, it seems likely that the unfortunate Arcadius of Palermo was wearing a normal lamellar jawshan of the type worn across the eastern and central regions of medieval Islam and which, though rarer in the western regions (fig. 30a left, 30b and 30d–f ), is known to have been exported as far as Al-Andalus in the tenth century and trans. Marie-Agnés Lucas-Avenel, Vol. 1, Books I and II (Caen, 2016), Cap. XXXIII, 341 and 343. 139 Malaterra, Histoire, Cap. XXXIII, 340 and 342. 140 Philippe Contamine, La Guerre au Moyen Age (2nd edn, Paris, 1986), pp. 320–25; Flori, J., Chevaliers et chevalerie au Moyen Age (Paris, 1998), p. 102; Olivier Renaudeau, “Problèmes d’interpretation du costume d’après du costume d’après la Broderie de Bayeux”, in La Tapisserie de Bayeux: l’art de broder l’histoire. Actes du colloque de Cerisyla-Salle, ed. Pierre Bouet, François Neveux and Brian J. Levy (Caen, 2004), pp. 243–53. 141 David Nicolle, Late Mamluk Military Equipment. Collection Travaux et Études de la Mission Archéologique Syro-Française, Citadelle de Damas (1999–2006): Volume III (Damascus, 2011), pp. 42–86; Nicolle, “Jawshan, Cuirie and Coat-of-Plates”, passim.



a gift to the normans 101

Here, the mention of al-jawashin al-tinsiyah wa al-khurasaniyah sab’miya’h qata’at (jawshans from Tinis/Tinnis and Khurasan [eastern Iran] of seven hundred elements) is particularly interesting.142 Seven hundred elements are too few for the rings of even a small mail hauberk, and are probably too many for a leather lamellar cuirass, in which the lamellae are normally large. So it seems likely to refer to a metallic lamellar cuirass. Taking the average dimensions of a single metal lamel from this period, as well as the normal overlap of each horizontally and vertically, perhaps with some additional lamellae for upper-arm protection, seven hundred seems be a suitable number of elements for a full-length cavalry cuirass as shown in eastern Islamic art of the ninth to eleventh centuries. Similarly it would be too many for the limited, abdomen and upper-body covering type more common in twelfth to thirteenth century Islamic art.143 Another form of armour of eastern Islamic origin was also spreading westward during this period: the mail-lined, quilted, and fabric-covered kazaghand, which may have differed from the jubbah primarily in shape, size, and weight. It was described as being like a khaftan in shape, although padded with qaz or kazh silk waste for use in war, when it was worn beneath mail armour. Only a few years later the kazaghand was specifically stated to be both padded and lined with mail.144 It reached the Middle East by the late eleventh or very early twelfth century,145 perhaps reaching Sicily and maybe even southern Italy before the Norman invasion. Thereafter the mail-lined and quilted armour was known in central and western Europe as the jazrain hauberk, jazerant, gasigan, jazerenc, and other variations on the term.146 Unfortunately the theory of a Sicilian role in the spread of this form of mail armour is weakened by the fact that the first mention of a ghiazzerina in

142 Ibn

al-Khatib, Lisan al-Din, Kitab ‘Amal al-’Alam. Histoire de l’Espagne Musulmane, ed. Évariste Lévi-Provençal (Rabat, 1934), pp. 118–19; Sayyid M. Imamuddin, “Commercial Relations of Spain with Iraq, Persia, Khurasan, China and India in the Tenth Century AC”, Islamic Culture 35 (1961), 179; Sayyid M. Imamuddin, Some Aspects of the Socio-Economic and Cultural History of Muslim Spain, 711–1492 AD (Leiden, 1965), p. 130. 143 Maqqari, Ahmad Ibn Muhammad al-, Analectes sur l’Histoire et la Litterature des Arabes d’Espagne, ed. Reinhart Dozy, Gustave Dugat, Ludolf Krehl, and William Wright, vol. 2 (Leiden, 1858–61), p. 382; Maqqari, Ahmad Ibn Muhammad al-, The History of the Mohammedan Dynasties in Spain, trans. Pascual de Gayangos (London, 1840–43; reprinted: London, 2002), p. 158 [also as “Muslims Accounts on Warfare in al-Andalus (Spain)” (https://bit.ly/2CQYvNW)]; Sayyid M. Imamuddin, “Commercial Relations of Spain with Ifriqiyah and Egypt in the Tenth Century AD”, Islamic Culture 38 (1964), 12. 144 Assadullah S. Melikian-Chirvani, “The Westward Journey of the Kazhagand”, Journal of the Arms and Armour Society 11 (1983), 9. 145 Melikian-Chirvani, “Westward Journey”, 16–18. 146 Melikian-Chirvani, “Westward Journey”, 25–8.

102

david nicolle

central Italy dates from the very early fourteenth century, being found in the Nuova Cronica history of Florence by Messer Filippo Villani.147 There is similarly traditional and contemporary information about helmets. Badr al-Din Ibn Juma’h writes of the Prophet’s baydah “egg-shaped” helmet and his mighfar mail coif as two separate items,148 while there is enough pictorial evidence to suggest that the hood or large hat associated with Muslim troops during the early centuries may have been protective, either being thickly padded or even lined with mail (figs. 26a–b and 39; perhaps figs. 21, 22a–b and 23a–c,). This form of headgear was widespread enough for it to seemingly be used as a means of identifying “infidels” in much early medieval Christian art (figs. 25, 33b and 51a–b). Later an account of the battle of Haydaran between the Zirid army and the Banu Hilal, written by Ibn al-Athir over a century after the event, recorded that the Zirid ‘abid infantry were so well armoured that they seemed invulnerable. So the leader of the Banu Hilal told his men to “aim for their eyes”, which suggests that the ‘abid wore mail coifs or aventails which covered their faces except for the eyes, precisely as shown in eastern Islamic art of the period.149 Shields might be the area where there is the strongest evidence for a mutual influence in military styles between the Islamic world and Italy, including or via Sicily. Once again Badr al-Din Ibn Juma’h provides traditional information, although even this can be surprising, as when he states that a turs shield, which was normally of wood, could either be made of, or more likely could include, elements of steel.150 Metal shields are normally thought to have been developed in later centuries as a defence against firearms, although one shield entirely covered with metal segments was found in a perhaps twelfth-century context in Islamic central Asia. Meanwhile the daraqa shield was traditionally made of leather or rawhide, while a larger form of leather shield of Saharan origin, widely used in North Africa and Al-Andalus, was the lamt, traditionally made of antelope skin.151 Some shields would better be described as mantlets, designed to be rested on the ground (figs. 36 right, 61a and probably 61b), although they could also be carried in battle. Whether the tariqa form of elongated shield was really known in the days of the Prophet, as sometimes claimed, seems unlikely. Or perhaps the term tariqa, which came to mean the so-called “Norman” shield with an elongated lower part tapering to a point or rounded tip (figs. 13, 14a, 26a, 30b–f, 33b, 36 left, possible 41, 45a–c, 58 and perhaps 52), was 147

Mario Scalini (private correspondence on the form of armour known in medieval Italian sources as the ghiazzerina, May 1983). 148 Badr al-Din Ibn Juma’h, Mustanad al-Ajnad, pp. 58 and 62. 149 Michael Brett, Fitnat al Qayrawan. A Study of Traditional Arabic Historiography (PhD thesis, London University, 1970), pp. 17–18. 150 Badr al-Din Ibn Juma’h, Mustanad al-Ajnad, p. 61 n. 1850. 151 Idris, La Berbérie Orientale, p. 533.



a gift to the normans 103

used for any form of elongated shield in earlier centuries. Infantry lined up with their shields formed into a shield wall had, according to tradition, been used since the days of the Prophet, and by the eleventh century, if not earlier, tariqa shields (pl. tawariq) were used by Fatimid infantry to form such a shield wall.152 Indeed, by the twelfth century the Middle Eastern januwiyya “Genoese” shield was a particularly large shield with a flattened base, designed to rest upon the ground as a mantlet, but could also be held like an ordinary shield (figs. 14b–d, 28a–b, 36 right, 40a, 48b–c, 59, 61a and probably 61b).153 The spear was a ubiquitous weapon which came in various lengths and had shafts of various materials, although the Arabs were still apparently known for long spears (figs. 42 and 46) with bamboo shafts. Javelins similarly came in a variety of weights or lengths, with an assortment of heads or blades (figs. 16b, 44, 48a and d–f ). Swords and daggers were almost as ubiquitous, although an almost chance account of an assassination attempt in the early eleventh century provides unexpected information. Here the would-be killer used a specifically Berber form of slender, pointed dagger called a yafurt or yafrut, a weapon which may have survived into modern times as the Berber tafrut.154 The only difference between the written Arabic letters “y” and “t” is that the former has two dots below the stroke of the letter while the latter has two dots above. Might a transcriber writing a technical term of Berber origin have simply made a mistake? The mace was a more widespread weapon in the early medieval Islamic world than in most of Europe, where it came into common use only as a result of the adoption of more rigid armour from the late thirteenth century onwards. Perhaps the greater role of maces, as distinct from simple clubs, in wealthier or more sophisticated early medieval Islamic armies also pointed to a wider used of semi-rigid lamellar cuirasses, as suggested in art from the eastern and central Islamic lands. Here the earliest form of true mace was the dabus, which, again according to tradition, was already known in the seventh century,155 although in reality it may have been adopted a century or so later. Thereafter the dabus became widespread across almost the entire medieval Islamic world (figs. 18, 32, 33a and c, 34b–d, 35a and 64b). Other, heavier types were in common use in the eastern and central regions by at least the ninth century (figs. 55 and 62a–b) and in southern Italy by the 152

Beshir, The Fatimid Caliphate, p. 74 n. 210. Tarsusi, Murda Ibn ‘Ali Murda al- [Mardi Ibn ‘Ali Mardi al-], “Un traité d’armurerie composé pour Saladin”, Bulletin d’Etudes Orientales 12 (1947–48), ed. and trans. Claude Cahen, 137 [trans.], 114 [Arab.]; David Jacoby, “The Supply of War Materials to Egypt in the Crusader Period”, Jerusalem Studies in Arabic and Islam 25 (2001), 106–7. 154 Denise Jacques-Meunié, “Le Nom Berbère d’un poignard Maghrébin au XIe siècle d’après un texte Arabe de l’Egypte”, Journal Asiatique 250 (1962), 614. 155 Badr al-Din Ibn Juma’h, Mustanad al-Ajnad, passim. 153

104

david nicolle

ninth or tenth century (fig. 65), in some cases being specifically associated with a capability to break lamellar armour.156 Although archery was almost entirely an infantry affair in Islamic Sicily, there were references to archers on horseback who probably formed a small elite of mounted infantry who would still dismount to fight (fig. 33f ). The early tradition of Arab-Islamic archery was on foot in support of, and while protected by, other infantrymen with spears and shields. This even had religious sanction, with religious scholars like Badr al-Din Ibn Juma’h praising the pre-eminence of archery “in the way of God”.157 Bows of composite construction were the usual forms for warfare (figs. 33f, 35b, 38a, 54a left and perhaps right, 57b, 63 and 64a), and an early form, incorporating deer or wild goat horn as well as wood and sinew, was used in Al-Andalus, and almost certainly also North Africa and Sicily. Simple wooden bows continued to be used in several parts of the medieval Islamic world, although mainly for hunting,158 and it seems that the bows of Saharan, sub-Saharan African, and Nubian peoples were of simple construction. They may also have been large and, in the case of Nubian bows, were probably descended from the type of bow used in ancient Egypt.159 The large wooden bows used by “Ethiopian” troops in Islamic Egyptian forces probably fell into this category.160 The question of the crossbow is more complicated (figs. 47a–c, 54b and 54c, and 66). Such weapons had been known in late Roman times and survived as hunting weapons in southern Europe and a few other places through the early medieval period. Large crossbows spanned by winches and other mechanical devices also continued to be known, not just in theoretical treatises but in the reality of siege warfare in the post-Roman states. These, however, were probably frame mounted, or at least were too heavy to be operated by one man. It is the reappearance of hand-held crossbows as war weapons, carried and operated by one man or by a team of shooter, loader, and shield carrier, that has most attracted the attention of medieval military historians. Normally known in the Islamic world as the qars al-rijl “foot bow” or qaws al-rikab “stirrup bow”, these war weapons would lead to a revolution in armour, and naval and siege warfare during the late medieval period. Contrary to what is generally believed, they reappeared in signifi156

Shihab al-Sarraf, “Close Combat Weapons in the Early ‘Abbasid Period: Maces, Axes and Swords”, in Companion to Medieval Arms and Armour, ed. David Nicolle (Woodbridge, 2002), pp. 152–67, figs. XII 1–9, 53–73. 157 Badr al-Din Ibn Juma’h, Mustanad al-Ajnad, pp. 53–54. 158 William F. Paterson (private correspondence on Nubian, Seljuk and Italian bows, Jan. 1983). 159 Ibid. 160 Leo VI, The Taktika of Leo VI, ed. and trans. George T. Dennis (Washington DC, 2010), p. 477.



a gift to the normans 105

cant numbers in the Islamic world before they did in Europe. Also, they first reappeared in the eastern or central provinces, primarily in defences of fortified places, most notably in tenth-century Tarsus, where over half of the city’s towers were defended by qisiyy al-rijl (sing. qaws al-rijl).161 The earliest evidence of the use of crossbows in the western regions of Islam was during the siege of Barbastro in northern Spain by Ibn Hud al-Judhami, ruler of Saragossa, in 1046 AD. According to Ibn ‘Idhari, this siege involved a corps of six hundred rumat ‘akkara or archers using the ‘akkara, a type of crossbow which became typical of the westernmost regions of medieval Islam.162 There is little military archaeological evidence from the western Islamic regions during this period, and what does exist shows both similarities with and differences from the equally sparse archaeological evidence from further east. Perhaps the best known, though still debatable, object is a helmet from Chamosen in Switzerland (fig. 1). Scholars have suggested that its decoration appears to be ninth- to tenth-century Carolingian, while the helmet itself has little in common with late Roman or early Byzantine helmets,163 being forged from a single piece of iron or steel. For these reasons an Islamic origin was suggested for the basic helmet, not because others of similar construction were then known from the Islamic world, but because of its technological sophistication and the close trading contacts between western Europe and the Islamic world.164 This theory also relied upon iconographic evidence for the spread and use of such round, one-piece helmets, not least in the probably eleventh-century Sicilian carved ivory chess pieces now in the Bibliotheque Nationale’s cabinet des Medailles in Paris (fig. 27a–b). Documentary evidence for one-piece domed helmets in the early Islamic world is relatively abundant. They were known as baydah, “egg” or “egg shaped” helmets and were described in some detail in the early Arabic Taj al-‘Arus dictionary. However, the early baydah could also be made from segments riveted together or to a frame. Taken as a whole, the term baydah probably referred to a shape rather than a method of construction, being differentiated from pointed helmets, which, at this period, seem all to have been of segmented construction.165 Since the study of the Chamosen helmet was published in 1930, two or perhaps three comparable helmets have been identified. The most obviously similar comes from Ifriqiya and is now in the Museum of Islamic Studies at Raqada, near Kayrawan (fig. 2). Unfortunately there is no further information which could help with its dating, although it was at one time labelled as 161

Bosworth, “al-Tarsusi’s Siyar al Thughur”, 188. Évariste Lévi-Provençal, Histoire de l’Espagne Musulmane (Paris, 1950–67) 3:93 n. 3. 163 Eduard A. Gessler, “Der Kalotten-Helm von Chamoson”, Zeitschrift für Historische Waffen- und Kostümkunde 3 (1930), 122. 164 Gessler, “Der Kalotten-Helm”, p. 123. 165 Gessler, “Der Kalotten-Helm”, p. 127. 162

106

david nicolle

possibly Hafsid (1229–1574 AD). The helmet, with a low-domed, one-piece bowl and chiselled decoration, from Iran and dated to the eighth to ninth centuries, is quite well known (fig. 3), but more recently a highly decorated helmet appeared on the international antiques market where it was initially dismissed as a fake (fig. 5). Reportedly found in eastern Iran or Afghanistan, its decoration is certainly “exotic”, probably rooted in the Buddhist culture of pre-Islamic Afghanistan, but with an Arabic inscription. The basic steel helmet within the decoration is straightforward and I have provisionally dated this helmet to the end of the tenth century.166 More recently still, the excavation of an early medieval ship found off the south-western coast of Iran produced a helmet (fig. 4), a probable decorative finial from this helmet, a fragment of mail armour, and other items including ceramic jars. Provisionally dated to the late Sassanian period,167 I nevertheless suggest that the decoration of the helmet and finial are so similar to those on the abovementioned helmet from eastern Iran or Afghanistan that the wreck and its contents are more like date from the early Islamic period.168 Sadly, another helmet from Tunisia which was once thought to be medieval Islamic is, in fact, almost certainly a later medieval European great helm, but with its face-plate removed (fig. 6a–b). Swords found in early medieval Islamic contexts in the western Islamic regions seem to combine features from the early Islamic Middle East and early medieval Europe. For example, a sword and the bronze chape from its scabbard were found with a hoard of ninth-century Umayyad Andalusian coins at an unspecified location near Cordova. The tang of the hilt has been “restored” incorrectly in modern times, making the grip longer than it probably was originally. However, the blade is 79 cm long, 5.5 cm wide for most of length, with a flattened diamond section and no fuller groove.169 Perhaps more relevant to the history of Islamic Sicily are a sword and the remains of a leather dagger sheath found in the wreck of two ships, one an Islamic merchant vessel dating from the tenth century, the second a small boat which probably sank at the same time as the larger ship (fig. 7). The sword was actually found on the remains of a man in the 166 David

Nicolle, unpublished report for the owner of this helmet. Tofighian, “Pazushi dar bar-rasi bastanshenasi zir ab suwahil Bandar-e Rig (Ganaveh)”, Pazhohesh-ha-ye Bastanshenasi Iran (Archaeological Researches of Iran, Journal of Department of Archaeology, Faculty of Art and Architecture, Bu-Ali Sina University) 4/6 (2014), 121–38 and 16–17 [English Abstract]; Hossein Tofighian, Farhang K. Nadooshan, and Seyyed M. Mousavi, “Sasanians in the Persian Gulf According to Archaeological Data”, Sasanika Archaeology 4 (https://bit.ly/2sYEMYa), 1–5. 168 David Nicolle, “One-piece Sasanian and Early Islamic Helmets”, in Crowns, Hats, Turbans and Helmets. The Headgear in Iranian History, Volume 1: Pre-Islamic Period, ed. K. Maksymiuk and G. Karamian (Siedlice-Tehran 2017), pp. 223–53. 169 Alberto Canto García, “Una Espada de Época Omeya del Siglo IX D.C.”, Gladius 21 (2001), 183. 167 Hossein



a gift to the normans 107

smaller boat. These wrecks lay off the coast of Agay, near Saint-Raphael, and, given the dating, may have been associated with the nearby Islamic colony at Fraxinetum.170 The closest parallels with the objects on board are found in eastern and southern Spain,171 in what was the Mashriq region of Al-Andalus. The sword is 0.88 m long and appears to incorporate a broadened extension to the tang, which would have mirrored the shape of the pommel. If this is so, then this sword, dating from the ninth to the very early eleventh century, is technologically akin to two twelfth-century Islamic swords from a cave in Gibraltar which I believe to date from the mid-twelfth century.172 The shaping of the end of a sword’s tang to be the same as that of the overall pommel might be a specifically western Islamic fashion. It might subsequently be seen during and perhaps after the Norman Kingdom of Sicily, firstly in a short sword from the twelfth or early thirteenth century which has an ivory-covered guard, grip, and pommel (fig. 8). The total length of this weapon is 61 cm, the blade being 44 cm long.173 Again, the tang appears to expand into a disc within the pommel. An earlier, even shorter short sword was found during the excavation of Lietor in Al-Andalus, and dates from the ninth or tenth centuries (fig. 9). Here the blade is approximately 34 cm long and has an ordinary tang without any disc or otherwise shaped expansion.174 Both these short blades may be evidence that early Arab-Islamic infantry, as distinct from cavalry, continued the tradition of the short Roman gladius stabbing sword.175 Al-Jahiz actually criticised the Arabs for their pride in such weapons; “You also boast of the length of your rumh [spear] and the shortness of the sayf [sword], yet it is vainglorious of the infantryman to boast of the shortness of his sword ...”.176

170

A. G. Visquis, “Présence Sarrazine en rade d’Agay au Xme Siècle”, in Rencontre d’Archéologie Sous-Marine de Fréjus, Saint-Raphael (Saint-Raphael, 1975) no page numbers. 171 Daniel Brentchaloff, Philippe Sénac, “Note sur l’épave sarrasine de la rade d’Agay (Saint-Raphaël, Var)”, Archéologie islamique 2 (1991), 71. 172 David Nicolle, “Two Swords from the Foundation of Gibraltar”, Gladius 22 (2002), 147–200; David Nicolle, “Talismanic Swords from the 12th Century Maghrib”, Graeco-Arabica 7–8 (1999–2000), 421–32; David Nicolle, “Medieval Swords of Morocco and al-Andalus: History, Construction and Decoration”, Graeco-Arabica 7–8 (1999–2000), 413–20. 173 Bashir Mohamed (ed.), L’art des chevaliers en pays d’Islam: collection de la Furusiyya Art Foundation (Paris 2007), pp. 40–41. 174 Julio Navaro Palazón and A. Robles Fernandéz, Liétor (Murcia 1996), pp. 91–93, pl. 147. 175 Marius Canard, “L’Expansion Arabe: le problème militaire”, in Settimane di Studio del Centro Italiano di Studi sull’alto Medioèvo, XII (Spoleto, 1965), p. 47; also in Marius Canard, L’Expansion Arabo-Islamique et ses répercussions (London, 1974). 176 Jahiz, Abu ‘Uthman ‘Amr Ibn Bahr al-, Al-Bayan wa’l-Tabyin, ed. Hasan al-Sundubi (Cairo, 1947), p. 14.

108

david nicolle

The blade of a second sword with a tang which expands into a disc also exists in a private collection and has yet to be published (fig. 10). It has an undeciphered Arabic or pseudo-Arabic inscription stamped into the blade and is, in my opinion, probably from Sicily, dating from shortly after the fall of the Norman kingdom. Its total length is 89.4 cm, the blade being 73.7 cm long with a maximum width of 5.6 cm. The disc which would have formed part of the pommel had a maximum width of 5.9 cm, with one rivet through it, while a lower rivet almost goes through the top of the fuller groove. A sword pommel, which, judging by its decoration, could be from twelfth- or early thirteenth-century Sicily or al-Andalus, would clearly not have been intended for a sword with such an broadened disc-shaped tang (fig. 11). More dubious is a double-headed axe with a pseudo-Arabic inscription that was found near Toulon (fig. 12). According to J. Lacam it was uncovered by a local inhabitant when planting fruit trees, and was lying beneath a stone slab which was itself 80 cm below ground level. The axe had the fragmentary remains of a handle, approximately 85 cm long, which was said to look like acacia wood, and the centre of the blade was pointing east. According to the Departmental Archaeological Centre, the Arabic inscription on the blade made no sense and was described as being in decorative pseudo-Kufic, presumably designed by someone who did not fully understand Arabic.177 Nevertheless, J. Lacam suggested that the axe might be a tabarzin axe from the early Fatimid period.178 Here it is perhaps worth noting that the Persian term tabarzin, which is now normally translated as “saddle-axe”, may originally have simply meant “large axe”.179 Conclusions Following the Norman conquest of Sicily, Sicilian Muslim troops were employed in substantial numbers throughout and beyond the Norman period. These men included elite guardsmen who served close to the ruler and were entrusted with highly significant duties such as guarding the Treasury. Other men were summoned as and when required, some serving as lightly armoured cavalry, others as “mounted infantry” archers, although not, apparently, as true “horse-archers” in the style of steppe nomads. A greater number served as ordinary infantry archers, again highly mobile and lightly armoured, if armoured at all. Siege engineers and military engineers in general formed a different sort of elite. 177 Jean

Lacam, Les Sarrazins dans le haut Moyen-Age français (Histoire et archéologie) (Paris, 1965), pp. 163–64. 178 Lacam, Les Sarrazins, p. 166. 179 Sarraf, “Close Combat Weapons”, pp. 162–67.



a gift to the normans 109

The relatively little documentary evidence from the Norman, Hohenstaufen, and Angevin twelfth and thirteenth centuries suggests that such troops continued to be equipped in distinctive styles, which were rooted into their own Siculo-Muslim military heritage. However, this also changed over the decades. Such changes or developments largely seem to have resulted from the increasing isolation of Sicilian Muslim troops from the wider Islamic world, especially after this community was exiled to the southern Italian mainland. Nevertheless, even here the Italian Muslims – as they should now be more properly called – retained their distinctive military culture, tactics, equipment, and perhaps even costume to a remarkable degree, until the destruction of the so-called Lucera colony at the close of the thirteenth century. The fact that descendants of the pre-Norman Sicilian Muslim military classes continued to be enlisted for so long, and continued to campaign in such a traditional manner, surely stands as testimony to the effectiveness of their military traditions. Running parallel to this story was the influence that Sicilian Muslim traditions of arms, armour and tactics had upon the development of such aspects of military culture within medieval Italy. They were themselves, of course, rooted in and to some extent continuing to reflect on-going technological developments from the broader Islamic world. Tenuous as it might seem, there was thus some degree of Sicilian Muslim influence upon the evolution of medieval European arms and armour as a whole, simply because Italy would exert such a profound military-technological influence north of the Alps from the thirteenth to the fifteenth centuries AD.

110

david nicolle

1

2

3

4

5

6a

6b

Helmets 1 Helmet with one-piece bowl and riveted cross-frame, reportedly found at Chamosen in Switzerland, Islamic early tenth to twelfth century, although the cross-pieces and brow band may have been added later in Europe (Schweizerisches Landes Museum, Zurich) 2 Helmet with one-piece bowl, Islamic early tenth to twelfth century, reportedly found at Raqada in central Tunisia (Museum of Islamic Studies, Raqada) 3 Helmet with one-piece bowl and chiselled decoration, Iran, eighth–ninth century (Furusiyah Art Foundation, inv. R-815, London) 4 Helmet found in the wreck of a Sassanian or early Islamic merchant ship off Bandar Rig on the Persian Gulf coast of Iran (Museum of Islamic Archaeology, Tehran) 5 Helmet reportedly found in eastern Iran or Afghanistan, probably late tenth century AD, shown without its decorative finial (private collection) 6a-b Helmet made from at least four riveted plates; at one time regarded as Fatimid tenth to twelfth century, but more likely a later medieval European great helm but with its face-plate removed (Museum of Islamic Archaeology, Kayrawan)

a gift to the normans 111



7

9

8

10

11

12 Weapons 7 Sword from the wreck of an Islamic ship, found off Agay, western Islamic, tenth century (Museum of Underwater Archaeology, Saint-Raphael) 8 Short sword with ivory-covered hilt, Sicily twelfth–thirteenth century (Furusiyya Art Foundation, inv RB-133, London) 9 Short sword excavated at Liétor, Andalusian, ninth–tenth century (after Navaro Palazon) 10 Short sword with an illegible Arabic inscription on the blade, probably Sicily, twelfth–thirteenth century (private collection) 11 Sword pommel, Sicily or al-Andalus, twelfth–early thirteenth century (Furusiyya Art Foundation, inv RB-93, London) 12 Bronze double-headed axe, reportedly found near Toulon and initially thought to be early medieval Islamic; but probably a fake (after Lacam)

112

david nicolle

13

14a

14b

14c

14d

Panel Painting 13 Painted ceiling panel of a cavalryman with a kite-shaped shield, Siculo Islamic, c.1140–43 AD (in situ over the southern side-aisle of the Capella Palatina, Palermo) 14a-d Painted panels of mounted warriors, three with elongated but flat-bottomed shields, Siculo Islamic, c.1140–43 AD (in situ in the muqarnas ceiling of the Capella Palatina in Palermo)



a gift to the normans 113

16a

15

16b

17 18

Ivory Panels 15 Warrior with a short-sleeved mail shirt or hauberk on a carved ivory oliphant, partially covered by a later silver rim, Sicily, late eleventh century (on loan to the Victoria & Albert Museum, London) 16a-b Carved ivory box showing armed men with short-sleeved mail shirts or hauberks, Sicily, eleventh century (Museum für Islamische Kunst, Berlin) 17 Carved ivory oliphant showing a warrior with a full-length mail hauberk, Sicily, late eleventh century (Musée Crozatier, Le Puy-en-Velay) 18 Carved ivory oliphant showing a man wielding a mace, southern Italy or Sicily, eleventh century (Museum of Fine Arts, inv. 57.58L, Boston)

114

david nicolle

19a 19b

19c

20

21

19a-c 20 21

Carved ivory box showing armed men with short-sleeved mail shirts or hauberks, Sicily, 1050–1100 AD (Metropolitan Museum of Art, inv. 17.190.241, New York) Carved ivory oliphant showing a warrior with a mail shirt, Sicily, late eleventh century (Metropolitan Museum of Art, inv. 04.3.177, New York) Carved ivory plaque showing a man wearing a full mail hauberk with long sleeves and perhaps a mail coif, Sicily, southern Italy or Andalusia, late twelfth century (National Museum of Antiquities, Ravenna)

a gift to the normans 115



22b

22a

23a

23b

22a-b 23a-c

23c

Carved ivory plaques showing men wearing full mail hauberks with long sleeves and perhaps separate mail coifs, Sicily, southern Italy or Andalusia, late twelfth-early thirteenth century (Hermitage Museum, St. Petersburg) Carved ivory plaques showing guardsmen with separate mail coifs and probably armour-lined and fabric-covered coats, from al-Humayma, Jordan, probably made in Khurasan, before 750 AD (Archaeological Musuem, Aqaba)

116

david nicolle

24a

24b

24c

24d

25

24a-d

25

Carved ivory plaques showing scenes from the Life of Christ with armoured men in largely Byzantine style [a and d] and armed but unarmoured men with Islamic-style tiraz bands around their upper sleeves [b–c], from Cathedral Altar, first half of twelfth century, southern Italy: a) Herod’s guards; b) guards at Holy Sepulchre; c) guards at Crucifixion; d) Massacre of the Innocents (Cathedral Museum, Salerno) Philistines shown in provincial Byzantine style, one apparently with a form of breastplate [left], on “Rome Casket”, carved ivory box, Sicily or southern Italy, late twelfth century (Palazzo di Venezia Museum, Rome)



a gift to the normans 117

26a

26b

27a

28a

27b

28b

Ivory Chess Pieces 26a–b “Charlemagne’s chess set”, carved ivory chess knight with kite-shaped shield, Sicily or southern Italy, eleventh century (Bibliothèque Nationale Cabinet des Medailles, Paris) 27a–b “Charlemagne’s chess set”, carved ivory chess knight with round shield, Sicily or southern Italy, eleventh century (Bibliothèque Nationale Cabinet des Medailles, Paris) 28a–b “Charlemagne’s chess set”, carved ivory chess pawn or infantryman with a long but flat-bottomed shield, Sicily or southern Italy, eleventh century (Bibliothèque Nationale Cabinet des Medailles, Paris)

118

david nicolle

29c 29b

29a

30a 30b

30c 30d 30e 30f

Stone and Stucco Carvings 29a–c Carved relief of huntsmen, one clearly wearing a form of quilted “soft” armour [a], southern Italy, late eleventh century (in situ above southern portal, Church of San Benedetto, Brindisi) 30a–f Carved relief showing siege of a city, southern Italy, very late eleventh or very early twelfth century: a–c) defenders; d–f ) attackers from the right wearing lamellar armour. Note that the attackers from the left wear mail armour (in situ above north door, Church of San Nicola, Bari)

a gift to the normans 119



31

33a

32

33b

33c

33d 33e

31 32 33a–f

33f

Carved relief of a horseman wearing a small lamellar cuirass, probably part of Pharoah’s army, southern Italy, very late eleventh or very early twelfth century (in situ west front, Church of San Nicola, Bari) Carving of a merman with a mace and round shield, southern Italy, 1175–1200 AD (in situ on a column inside the Cathedral, Bitonto) Carved capitals showing a variety of warriors, some with African features, and with a variety of European, Byzantine, and Islamic styles of weaponry, armour, and costume, Sicily, c.1189 AD (in situ Cathedral Cloisters, Monreale)

120

david nicolle

34a

34b

34c

34d

34f

34e

34g

35b 35a

34a–g 35a–b

Carved reliefs showing assorted demons in combat with spears, maces, and an early form of guisarme axe [d], southern Italy, late twelfth century (in situ Cathedral, Barletta) Ambone di Nicodemo showing huntsmen or warriors with a mace and a composite bow, stone and stucco relief carving on pulpit, southern Italy, first half of twelfth century (in situ church of Santa Maria in Valle Porclaneta, Rosciolo dei Marsi)



a gift to the normans 121

36

37a

38a

36 37a–b 38a–b

37b

38b

Carved capital showing a typical northern Italian cavalryman and an infantryman with a segmented helmet plus long but flat-bottomed shield, northern Italy, mid-twelfth century (in situ Cathedral, Parma) Figures on carved capitals, identified as Islamic by the embroidered tiraz bands on their upper sleeves, southern Italy, early thirteenth century (in situ Cathedral, Matera) Carved relief of huntsmen, one with a long-sleeved mail hauberk, helmet, and composite bow [a] and one wearing only a kilt-like garment [b], either SiculoIslamic eleventh century from an earlier building or Christian Sicily early twelfth century (in situ above north door, Church of La Martorana, Palermo)

122

david nicolle

39

40a

40b

40c

40d

39 40a–d

Painted stucco statuette of a guardsman wearing a quilted qalansawah hat and a mail hauberk, from the Umayyad Palace at Khirbat al-Mafjir, Palestine, early eighth century (Rockefeller Museum, Jerusalem) Relief carvings of shields symbolising elements of the Fatimid army, decorated plus a sword for the caliphal guards [a], elongated with flat bases for the infantry [b–c] and round for the cavalry [d], Egypt c.1087 AD (in situ Bab al-Nasr, Cairo)



a gift to the normans 123 41 42

43

44

Manuscripts and Manuscript Fragments 41 Painted papyrus fragment of a cavalrymen with an apparently kite-shaped shield, probably from Fustat, Egypt, tenth or eleventh century (Musée du Louvre, inv. MA 0125, Paris) 42 Drawing on paper of a faris cavalryman with a small round shield, from Fustat, Egypt, probably tenth century (Österreichische Nationalbibliothek, Ms. ACh. Vindob. 11416, Vienna) 43 Painted paper fragment showing horsemen greeting above the carnage of battle [note shield and severed limbs], the visible horsemen apparently wearing a lamellar cuirass, from Fustat, Egypt, probably early twelfth century (Keir Collection, inv. I.8, London) 44 Painted paper fragment of a turbaned warrior with a large shield and two javelins, from Fustat, Egypt, probably early twelfth century (Museum of Islamic Art, inv. 13801, Cairo)

124

45a

david nicolle

45c

45b

44

47a

47b

45a–c

46 47a–c

47c

Painted paper believed to show the Fatimid garrison of Asqalon emerging to confront Crusader foes who are wearing clearly identifiable European military equipment [not shown here], from Fustat, Egypt, probably mid-twelfth century (British Museum, Department of Oriental Antiquities, London) Painted paper showing unarmoured Arab or Berber cavalrymen, from Fustat, Egypt, probably mid-twelfth century (private collection; present location unknown) Manuscript illustrations in the Tabsira by Mardi Ibn ‘Ali Ibn Mardi al-Tarsusi, Egypt, c.1171 AD: a) crossbow forming part of the release mechanism of a counterweight trebuchet; b) crossbow mounted inside a shield; c) crossbow modified to shoot incendiary grenades (Bodleian Library, Ms. Hunt 264, Oxford)

a gift to the normans 125



48b

48a

48c

48f

48d 48e

48a–f

Gospels made in Damietta, Coptic Egypt, 1179/80 AD: a) guard of Pilate with a pair of javelins; b) soldiers at Crucifixion with round and elongated flat-based shields; c) the Betrayal showing swords, a probably mace, small round and elongated flat-bottomed shield; d) Herod’s guards, one with a pair of javelins; e) soldier before Joseph of Arimathea, with bow, quiver, shield, sword, and pair of javelins; f )- soldier with a pair of javelins casting lots for Christ’s clothes (Bibliothèque Nationale, Ms. Copte 13, ff. 82v, 83v, 79r, 5r, 131r, 274v, Paris)

126

david nicolle

49 50a

51a

51b

53a 49

50b

52

53b

Ruler’s guard, perhaps a Muslim, judging by the tiraz bands around his upper sleeves, Leges Langobardorum from Benevento, southern Italy, eleventh century (Archives of Badia della Santissima Trinità, Abbey of La Cava, Subiaco) 50a–b Pharoah’s army in the Red Sea, including these cavalrymen with longhemmed, long-sleeved mail hauberks, Exultet Roll, southern Italy, eleventh century (Cathedral Archives, Roll 2, Gaeta) 51a–b Guards with probable small forms of lamellar cuirass, at Dedication to Duke and Emperor, Exultet Roll, southern Italy, eleventh century (Museo Civico, Pisa) 52 Guard of the Norman ruler, perhaps identified as a Muslim by his beard and pointed hat, and with a large elongated shield, Regestum di Sant Angelo in Formis, southern Italy, c.1150 AD (Library, Ms. Reg. 4, Abbey of Monte Cassino) 53a–b Synopsis of Histories by John Skylitzes, Sicily, late twelfth–early thirteenth century: a) Arab emir wearing long-skirted lamellar cuirass fleeing from Bardas Phocas; b) one of the Arab soldiers attacking Edessa, wearing small form of lamellar cuirass (Biblioteca Nacional, Cod. 005-3. N2, ff. 136v & 208r, Madrid



a gift to the normans 127

54a

54c

54a–d

54b

54d

Liber ad homorem Augusti by Peter of Eboli, Sicily or southern Italy, very late twelfth or early thirteenth century: a) unarmoured Muslim and Christian archers; b) Christian and Muslim crossbowmen, two with helmets, one with a turban; c) archer with a composite bow and a quiver at the siege of Salerno; d) crossbowman with brimmed hat or helmet at the siege of Naples (Burgerbibliothek, Cod. 120, ff. 131r, 117r and 109r, Bern)

128

david nicolle

55 56

57b

57a

58

59

Ceramics 55 Ceramic fragment showing a soldier with a long-hafted mace and perhaps mail beneath his coat or tunic, Egypt or Iraq, ninth-tenth century (Benaki museum, inv. 227, Athens) 56 Ceramic bowl showing horsemen with small round shields and one with a sword, from Sabra al-Mansuriyah, Tunisia, 950–1050 AD (Museum of Islamic Arts, Kayrawan) 57a–b Ceramic plaques from Sabra palace showing bearded Arab or Berber soldiers on foot and an apparently “moon faced” cavalryman, perhaps indicating a Turk, Tunisia, mid-eleventh century (Bardo Museum, Tunis) 58 Lustre-ware ceramic fragment showing a turbaned infantryman with a spear, an elongated kite-shaped shield and perhaps indicating mail or other protection over his left arm, from Fustat, Egypt, tenth–eleventh century (Victoria & Albert Museum ceramic study collection, London) 59 Turbaned soldier with a straight sword and a large, elongated but flat-bottomed shield, on a lustre-ware ceramic bowl, Egypt or Iran, twelfth century (Keir Collection, inv. 151, London)



a gift to the normans 129

60a

60b

60c

60d

60e

60a–b 60c–d 60e

60f

Figure of horseman with a helmet, shield, and sword, on the rim of the Vaso de Tavira terracotta bowl, Islamic-Andalusian southern Portugal, eleventh century (Museu Municipal, Tavira) Figure of horseman with a turban and spear, on the rim of the Vaso de Tavira terracotta bowl, Islamic-Andalusian southern Portugal, eleventh century (Museu Municipal, Tavira) Figure of probable infantryman with a large shield, on the rim of the Vaso de Tavira terracotta bowl, Islamic-Andalusian southern Portugal, eleventh century (Museu Municipal, Tavira)

130

david nicolle

62c

61a

61b

62a

62b

62d

61 Wall painting 61a–b Soldiers with long but in one case clearly flat-bottomed shields [a] at the Crucifixion, wall painting, central Italy, twelfth century (in situ San Paolo fuori la Mura, Rome) 62–64 Metalwork 62a–d Silver plate found at Perm-Molotov in Siberia, Christian Iraq, Iran or Central Asia, probably seventh century: a) angel with a mace guarding the Holy Sepulchre, either wearing a full mail hauberk or covered with feathers; b) shepherd armed with a mace; c–d) soldiers armed with maces at the Crucifixion (Hermitage Museum, St Petersburg)



a gift to the normans 131

63

64a

65 63

64b

66

Soldier or huntsman with a composite bow on a bronze door panel, southern Italy, c.1179 AD (in situ Cathedral, Ravello) 64a–b Men armed with shields and maces, perhaps engaged in a judicial duel, on a bronze door panel, southern Italy, late twelfth century (in situ Cathedral, Trani) 65–66 Mosaics 65 Man apparently armed with a long-handled mace in the Legend of King Arthur, mosaic, southern Italy, ninth–tenth century (in situ Church of the Pantocrator, Otranto) 66 Crossbowman struck down by an opponent’s arrow during Turkish attack on crusader-held Antioch, northern Italy, early eleventh century (in situ Church of San Colombano, Bobbio)

5

Norman Battle Tactics in the Mediterranean Theatre of Operations: Fighting Lombards, Greeks, Arabs, and Turks c.1050–c.1100 Matthew Bennett The light-haired people place a great value on freedom. They are bold and undaunted in battle. Daring and impetuous as they are, they consider any timidity and even a short retreat a disgrace. They calmly despise death as they fight violently hand-to-hand combat either on horseback or on foot. If they are hard pressed in cavalry actions, they dismount at a single prearranged sign and line up on foot. Although only a few against many horsemen, they do not shrink from the fight. They are armed with shields, lances and short swords slung from their shoulders. They prefer fighting on foot and rapid charges. Whether on foot or on horseback, they draw up for battle, not in any fixed measure and formation, or in regiments or divisions, but according to tribes, their kinship with one another, and common interest. Often, as a result, when things are not going well and their friends have been killed, they will risk their lives fighting to avenge them. In combat they make the front of their battle line even and dense. Either on horseback or on foot they are impetuous and undisciplined in charging, as if they were the only people in the world who are not cowards. They are disobedient to their leaders. They are not interested in anything that is at all complicated and pay little attention to external security and their own advantage ... When it comes to a cavalry battle they are hindered by uneven and wooded terrain. They are easily ambushed along the flanks and rear of their battle line, for they do not concern themselves with scouts or other security measures. Their ranks are easily broken by simulated flight and a sudden turning back against them. Attacks by night by archers often inflict damage, since they are very disorganised in setting up camp. Above all, in warring against them one must avoid engaging in pitched battles, especially in the early stages. Instead, make use of 1 well-planned ambushes, sneak attacks and stratagems. 1

Maurice’s Strategikon: Handbook of Byzantine Military Strategy, trans. George T. Dennis (Philadelphia, 2010 [1984]), XI, 3, pp. 119–20.

134

matthew bennett

This assessment of the warrior nations of western Europe was made by the author of a “handbook of Byzantine military strategy”. This may well have been the Emperor Maurice (582–602) himself, victor over the Sassanid Persians and reformer of the Byzantine army, or someone of his circle.2 Although a stereotypical ethnological depiction of many different peoples, this outline sufficed its purpose and: “When the emperor Leo VI compiled his Tactical Constitutions at the beginning of the tenth century, he had little to add to Maurice’s organization.”3 From a methodological point of view, it might be questioned whether such an overview satisfies modern standards of a historical analysis. Its legacy was very powerful, though, even through the period of Norman expansion in the eleventh century and in such a sophisticated history as the Alexiad. This is usually attributed to Anna Komnene, daughter of the Emperor Alexios I (1081–1118); but it has been persuasively argued that her husband, Nicephorus Bryennius (1062–1137), an experienced soldier, contributed substantially to the work.4 Anna’s described the ethnic group which she anachronistically called the Kelts, otherwise known by commentators in western Europe – then and now – as the Franks, in very similar terms. The trope has a long history, being intimately interwoven with the description of Gauls and Germans as fierce at the onset and then tiring, brave but disliking heat, and fierce but undisciplined. Anna’s quote about the Frankish knight being capable of boring his way through the walls of Babylon with the impetuosity of his charge is so well known as to be a cliché. In Byzantine military manuals this virtue was always contrasted with the criticism that the Westerners lacked staying power. It is likely, of course, that the second event was the outcome of the first. A charge à l’outrance inevitably led to cavalry falling into disorder; which is why all authorities argued against it. Nor was this tactical weakness unknown in the West. The mid-twelfth-century Old French Rule of the Knights’ Templar was quite explicit that cavalry manoeuvres should be slow and controlled, save for the last burst.5 Although this referred specifically to the disciplined military monks of the Order, such knowledge was not confined to them. Their rule laid down instructions for reforming after a charge, by falling back on the banners of troops and squadrons (to use the language of a later era). The problem of recovering cavalry for further use in battle was also a timeless one. Horses, once pushed to the gallop, ended up “blown” and in 2

Dennis, Maurice’s Strategikon, pp. xi–xii. Dennis, Maurice’s Strategikon, p. xiii. 4 James Howard-Johnston, “Anna Comnena and the Alexiad”, in Alexios I Komnenos, ed. Margaret Mullet and Dion Smythe (Belfast, 1996), pp. 260–301. 5 The Rule of the Templars: The French Text of the Rule of the Order of the Knights Templar, trans. J. M. Upton-Ward (Woodbridge, 1992), pp. 181–87 [Appendix by Matthew Bennett]. 3



norman battle tactics 135

need of rest and recuperation before being re-engaged in combat. In this case, it was even more important to make best use of the first charge. The Normans have a fierce reputation in war. As Frankish charging cavalry they epitomised the Western way of battle fighting through which they overcame many and varied opponents. This is a very one-eyed view of what made them good warriors, though. A full picture of their talents would necessarily involve an assessment of their skills in politics, diplomacy, siege, and naval warfare; but since these are being covered by other authors in this volume, it falls to me to focus on battlefield tactics. It is not difficult to find information on this topic, for many historians of warfare, and indeed the Normans themselves, have glorified their supposed superiority in this field. The mid-twelfth-century vernacular poet Wace summed this up in just a few lines: “The Normans are a very fierce race, I do not know another like them. They are bold and chivalrous knights, conquerors wherever they go.”6 What is more difficult, however, is to make sense of the narratives available. It should be emphasised that battle technique was a cultural artefact: a sense of identity, masculinity, and mastery of weaponry made concrete; hence the focus on the lance attack. In this way, the Normans were able to substantiate their boast of moral and tactical dominance over all other nations. It used to be thought that only Westerners used the couched lance, but recent studies have proved that the technique was much more widely diffused.7 What mattered to the ruler, general, and employer – who was usually one and same person – was that “national characteristics” and traditional weaponry could be used to best effect both on campaign and in combat. In that regard, the Norman style had to be able to cope with many other battle techniques. How this worked out is the subject of this chapter. Evidence will be drawn from a wide range of narrative accounts: those concerned with the conquest of southern Italy and Sicily; Byzantine chronicles such as the Alexiad and earlier manuals dealing with how to counter western-type charging cavalry; crusader chronicles describing the Latins’ military contacts with the Byzantines, Arabs, Turks, and other steppe nations; and Arab accounts. The conclusions to be presented fall into three categories. First, historians should be wary of believing what are essentially propagandist and ethnic6

“En Normandie a gent mult fiere/ jo ne sai gent de tel manière/ chevaliers sunt proz et vaillant / par totes terres conquerant”, Le Roman de Rou de Wace, ed. A. J. Holden (Paris, 3 vols. 1970–73), Vol. 2, lines 9114–35. 7 Mamuka Tsurtsumia, “Couched Lance and Mounted Shock Combat in the East, The Georgian Experience”, Journal of Medieval Military History 12 (2014), 81–108; David Nicolle, “The Impact of the European Couched Lance on Muslim Military Tradition”, Journal of the Arms and Armour Society 10 (1980), 6–40; Kelly DeVries, Medieval Military Technology (Peterborough, 1992), pp. 13–14.

136

matthew bennett

stereotype representations of Western superiority over other nations. Second, that, despite this cautionary approach, culturally developed military skills and tactics could and did create a violent dialectic on the battlefield. Third, that in the fascinating interplay between such factors as the troop types available to a commander, generals’ decisions based on (in)experience, and sheer luck, the lance attack remained a crucial aspect of battle outcomes; but one that could not be relied upon to guarantee success. Norman Style of Warfare Although it might be assumed that the Norman way of fighting is well understood, it is still necessary to explore some issues with modern interpretations. Clearly, armoured cavalry played a major role in combat. Typically, fully equipped warriors wore a long mail hauberk (and the richer among them also had mail sleeves and lower-leg covering). A conical helmet with a wide nasal protected the head.8 The distinctive “tear-drop” shield was long enough to provide protection from chest to calf; when carried on horseback it covered the entire left-hand side of the rider and also protected the horse. When the knight leaned forward with his lance couched (held tightly in his right armpit) he was very difficult to hit and hurt. Furthermore, he had a firm “seat”, secure in a deep saddle with a high pommel and cantle, and long stirrups, with his feet thrust forward.9 This made him into a projectile combining man and horse that struck fear into the hearts of other horsemen and footmen alike. Although literary and artistic portrayals favour the socially elite mounted warrior, in reality a Norman army was a combined-arms force which derived its strength from a variety of weapon and troop types. As is well known and depicted on the Bayeux Tapestry, missile-men played a crucial role in the victory of Hastings in 1066. Massed archery with a range of up to two hundred metres was deadly against an unarmoured enemy and – as towards the end of that battle – to men exhausted after a long fight and unable to raise their shields in defence. Additionally, there was the recently developed hand-held crossbow, which lacked the carry of the selfbow but, close up, was capable of penetrating mail and even shields.10 It was probably this weapon which broke up the Varangians at the end of 8

See: The Bayeux Tapestry, ed. David M. Wilson (London, 1985) is still probably the best edition. Plate 24 shows the knights’ armour in the scene where Duke William is knighting Earl Harold (detail p. 22) and mounted knights can be viewed from both sides. 9 Ibid., charging knights, Pl. 53, also a foundering horse showing the box saddle, girth and stirrups, Pl. 66. 10 Archers are shown throughout the battle scenes in the Tapestry (usually in the lower border) and, in detail, Pl. 60, where three bowmen are depicted with wide, braced



norman battle tactics 137

the battle of Dyrrachium (1081). Bows of both types were also essential in keeping mounted archers at bay when the Normans encountered them in the armies of Byzantium and on crusade. The third component of the tactical mix was close-order footmen equipped with spears and other hand weapons. Since these were regarded as lower status by society at the time it is hardly surprising that they do not feature prominently in depictions of warfare. So, for example, although the written accounts of Hastings describe the role of spearmen in the initial attack, they are not portrayed in the Bayeux Tapestry.11 This could be explained in narrative terms as that it was the combination of archery and mounted charges which proved decisive in the victory. Yet a solid body of spearmen provided a secure base around and behind which the cavalry could regroup, rest, reform, and, if necessary, launch another attack. It should not be forgotten, of course, that knights could dismount and fight on foot as well. Indeed, this was a common tactic in Anglo-Norman warfare during the twelfth century when fighting against the French and other opponents within the British Isles, especially the Scots, noted for their impetuous foot charges.12 As will be shown, such dismounting also had a role in the warfare of the Mediterranean lands. Another factor, which is often neglected, is the role of non-knightly mounted troops. These were usually known in Latin texts as armigeri (where originally the armiger was a shield bearer fighting in support of his master). What causes confusion among historians is that the same word is used to describe more lightly armed and/or lower-status mounted troops.13 The distinction is easier to observe in the Old French vernacular: chevalier (knight), écuyier (squire) and serjeant (ordinary soldier derived from servant). Also, it can be difficult to analyse the role of these auxiliary mounted troops on the battlefield. It is usually assumed that the serjeants formed secondary and other supporting ranks to the cavalry formation. This can be supported by the only detailed evidence on the organisation of a cavalry force in the era: the Rule of the Knights Templar. Here is specifically stated that serjeants – being more lightly armed – were permitted to withdraw from the fight legs. That the single armoured figure is crossbowman is proved by his wearing armour, holding four bolts in his left hand and a close-legged stance. 11 The Gesta Guillelmi of William of Poitiers, ed. and trans. R. H. C. Davis and Marjorie Chibnall (Oxford, 1998), Bk. ii, ch. 16, pp. 126–27: “Pedites in fronte locavit saggitis armatos et balistis, item pedites in ordine secundo, firmiores et loricatos.” 12 For example at the Battle of the Standard, fought near Northallerton in 1138, where Norman knights – dismounted and intermixed with archers – beat off the attack of both the Galwegian foot and mounted Scottish knights. See: John Beeler, Warfare in England 1066–1189 (New York, 1966), pp. 86–92. 13 Matthew Bennett, “The Status of the Squire: the Northern Evidence”, in The Ideals and Practice of Medieval Knighthood, ed. C. Harper-Bill and R. Harvey (Woodbridge, 1986), vol. I, pp. 1–11.

138

matthew bennett 14

if wounded. Also, the squires’ role was to accompany the Brother Knights at a distance with spare horses and lances in order to rescue their masters if they were unhorsed or hurt.15 It should be remembered – despite modern misconceptions – that not all squires could expect the social promotion of knighthood and could remain what were, essentially, body servants for an entire career. Normans versus Lombards The Lombards entered Italy in 565 and quickly acquired a reputation as ferocious warriors, fielding an army of heavy charging cavalry supported by bow-armed foot soldiers. However, once they had settled down to rule their newly conquered lands they seemed to lose their military vigour. When the Franks began probing attacks into the peninsula in the mid-eighth century they began to be seen as victims rather than victors. Charlemagne’s campaigns were largely directed at capturing the rich, fortified cities of the northern plains. His advance on Rome for his coronation in 800 was unopposed. As a result, in Frankish minds and military mythology the Lombards became regarded as pushovers, to be feared only for their treacherous behaviours rather than for manly virtues. When the Normans first appeared in southern Italy they were mostly escaping stricter controls on entrepreneurial warfare in the Norman duchy. Enterprising individuals gradually seized power in a number of cities, eventually establishing a stranglehold that made a takeover possible. What is interesting, from a military viewpoint, is the Lombard rulers of the area put up so little effective resistance; by 1000 they seem to have been despised by the Byzantines and Normans alike. They seem to have become civilianised to the degree that they preferred to hire mercenaries, which is why the Normans were in Italy in the first place. Initially, the Normans began to build castles with which to threaten Lombard cities as a kind of protection racket. Then their leaders married into the Lombard aristocracy and very soon they became legitimate rulers. Once they possessed these power bases they turned their firepower upon the remaining Byzantine possessions. Robert Guiscard, who arrived in Italy in about 1046, first set himself up as a robber baron in Calabria. The extortionate methods by which the Normans made themselves wealthy produced hatred from the Lombard population, and in August 1051 his older halfbrother Drogo was assassinated, at the same time as other leading Normans were killed. 14 15

The Rule of the Templars, 420, p. 113. The Rule of the Templars, p. 183.



norman battle tactics 139

Meanwhile, the Byzantine Catepan Argyrus was instructed by his emperor to make an alliance with Pope Leo IX. The pope spent Christmas 1052 with the German emperor, Henry III, seeking aid for a campaign, and was provided with several hundred (possibly seven hundred) Swabian knights. Leo mostly had to depend upon Italian allies, though, and aimed to march into Apulia to join forces with Argyrus by midsummer. Robert Guiscard responded quickly, gathering an alleged three thousand knights and a few infantry to prevent this junction. Intercepting the pope’s army near Civitate on 17 June 1053, Robert initially attempted negotiations. The Normans were short of supplies and also perhaps nervous of challenging papal authority; but they needed a decision, and quickly. It is never a good idea to pick a fight with desperate men. The papal forces were utterly routed by the hammer blows of the Norman cavalry charges. The pope, now lacking military support, turned to the Normans for aid, and in 1059 Guiscard was appointed duke of Apulia and Calabria and future duke of Sicily, receiving a papal banner (just as Duke William did) to recognise his authority and justify the legitimacy of his cause. Case Study 1: Civitate, 1053 – Normans versus Lombards and Imperials There is some evidence both of where this battle was fought and the relative size of the armies.16 Apparently, the papal army consisted of about six thousand men. Many of these were Lombard levies, stiffened by German armoured warriors numbering three, five or seven hundred (according to three different sources) under the command of Werner and Albrecht. They stood on the right of the line. William of Apulia, who provides the most detailed account, claims that they were “not well versed in horsemanship”.17 This is actually unlikely, as German milites had been fighting on horseback for centuries, and indeed had defeated the Hungarians at the battle of the Lech in 955. In fact, the German knights also dismounted to fight at the siege of Damascus (1148), “as is the custom of the Teutons when a desperate crisis occurs”.18 Faced by the Norman cavalry, and perhaps not trusting in their Lombard, allies they may have felt safer on foot. They are described as carrying long swords, although it is possible that they actually wielded axes. On the other flank, “The Italians all stood crowded together ... because they did not know how to arrange their troops in proper battle order.” They 16

For a recent and detailed study: Charles D. Stanton, “The Battle of Civitate: a Plausible Account”, Journal of Medieval Military History 11 (2013), 25–55. 17 William of Apulia, pp. 18–19. [I used Graham Loud’s translation, which can be found online: https://bit.ly/2KcEtD7] 18 William, Archbishop of Tyre, A History of Deeds Done Beyond the Sea, trans. Emily A. Babcock and A. C. Krey (New York, 1943), vol. II, XVII. 4, p. 189.

140

matthew bennett

faced the best of the Norman cavalry, under Richard of Aversa, while Robert Guiscard led the reserve. Although the Norman numbers are uncertain, they are claimed to have had “3,000 cavalry and a few infantry”.19 Crucially, they were veteran, battle-hardened troops. The battle may be swiftly described, for on the papal left flank the Lombards ran away, chased by Richard’s men. In contrast, on the right wing, the Germans put up a stout resistance. It took a combination of a flank attack by Robert’s reserve and the return of Humphrey’s victorious troops to finish them off. They do seem to have been massacred. They may even be represented by nineteenth- and twentiethcentury discoveries in the area of the bodies of men and horses .20 Many of the men were very tall, just as the Germans were described, with long, fair hair, and looking down on the shorter Normans before the battle. Civitate was one of a series of battles in which Norman knights defeated opposing infantry, which has tempted some historians to see this an inevitable triumph of a new-style charging cavalry. While there is something in this analysis it is important to understand that the other two battles, Hastings (1066) and Dyrrachium (1081), also had important individual features which determined their outcome. Rather, the Norman commanders on each occasion should be commended for finding the right way to defeat their enemy. It is fair to say that Civitate was the easiest of the victories in that it largely depended upon the success of the mounted knights; but disciplined foot soldiers (even if in this case dismounted knights) were quite capable of holding them at bay. Normans versus Muslims The conquest of Sicily began soon afterwards, although increasingly this was the responsibility of Robert’s younger brother Roger.21 In order to support this invasion, the Normans launched a campaign of capturing ports around the coast of southern Italy. These provided both ships and, more importantly, skilled sailors, mostly Greek, with which to man them. After a couple of incursions the first serious attempt on Sicily took place in 1061, on the invitation of a Muslim war-lord. This led to the capture of Messina. In operational terms the most significant feature of the invasion was the use of specially designed transports to carry over the horses of the knights. These were constructed rather like modern roll-on-roll-off-ferries, with a double sternpost, which were backed onto the beach by means of oars. Then two doors were opened and a ramp was set up to allow the 19

William of Apulia, p. 19. Stanton, “The Battle of Civitate”, p. 51. 21 See: M. Bennett, “Norman Naval Activity in the Mediterranean c.1060–1108”, AngloNorman Studies 15 (1993), 41–58, esp. 48–50. 20



norman battle tactics 141

horses to walk ashore. The vessels used could carry only a dozen or so at a time, requiring two journeys to transport approximately 450 mounts. Once established, Roger called over three hundred more.22 The value of allowing the chevaliers their own horses was demonstrated in battle against the Muslims. The arrival of Guiscard with more troops raised numbers to two thousand (although these may not have all been cavalry). There were subsequent land victories: on the northern side of the island outside Palermo in 1068, and finally and decisively outside Taormina in 1079. In both these battles the Muslims are accorded huge forces in the tens of thousands, against just a few hundred Normans. As usual, the Normans’ victory is ascribed to their impetuous cavalry charge, although in the latter case it was admitted that solid Muslim infantry repelled three charges, and the battle was not won until the Normans dispersed the enemy cavalry.23 It should be pointed out, though, that these victories were only in support of the more significant siege operations which were aimed at the main ports of the region. The two best examples, both completed in 1071, also included sea battles. On the eastern coast of Apulia, Bari was still holding out under Byzantine control following a siege of three years. Since this was a promontory fortress, with only a narrow neck of land linking it to the mainland, siege operations were restricted and supplies were still able to get through to the defenders. In response, Guiscard created a blockade all around the headland with a semi-circle of ships bound together. This construction repelled the attacks of a Byzantine relieving fleet, eventually resulting in the city’s surrender. As soon as Bari had fallen, and with a fleet strengthened by captured ships, the Normans set out for Palermo. Here again they were challenged from the sea by a Muslim fleet from Zirid North Africa; but once again the Norman ships outfought the enemy, leading to the capture of that city, too. The sources do not allow for a very good understanding of the nature of the Muslim forces which the Normans encountered, although a recent study by Paul Brown has made a determined effort to wring some sense out of them, in his description of a battle in 1061.24 One problem is the stereotyped way in which the Norman chroniclers write about Muslims; and in this manner they parallel the Saracens of the chansons de geste. For example, Amatus attributes numbers in the tens and hundreds of thousands, while the Normans and their allies are given reasonable numbers of a thousand or so. Even Malaterra’s fifteen thousand is probably too large in comparison to seven hundred knights and three hundred Muslim allies. All that can be 22

Bennett, “Norman Naval Activity”, p. 51. Paul Brown, Mercenaries to Conquerors: Norman Warfare in the Eleventh- and Twelfthcentury Mediterranean (Barnsley, 2016) p. 108. 24 Brown, Mercenaries to Conquerors, pp. 86–114. 23

142

matthew bennett

said about the battle, which took place near Enna, is that Roger drew up his forces in two wings. Against them, Ibn al-Hawas apparently deployed his army in three lines; but it was still routed by the Norman cavalry charge. In 1063 another encounter against al-Hawas and his Zirid allies led to a similar rout.25 Brown is inclined to attribute the Norman superiority to their deployment in wedge formation. However, this is to put too much weight on the Latin term cuneus, which also means unit or troop.26 Normans versus Byzantines The final area of conflict to be considered is between the Normans and Byzantines in north-western Greece, in what is now called Albania.27 Robert’s ambitions now turned to the Byzantine Empire itself. In 1071 the Seljuk Turks inflicted a defeat on the Greeks at Manzikert, capturing their emperor. This precipitated a civil war and encouraged Guiscard to press claims via marriage between his daughter and a claimant to the throne. When this failed, in May 1081 he launched a seaborne invasion against Dyrrachium (modern Durres, Albania). The fleet may have numbered 150 ships, with horse transports carrying 1,300 knights and their horses, and the larger vessels equipped with leather-covered wooden siege towers. Anna, the historian daughter of Alexios Komnenos (1081–1118), who had just become emperor, describes the events in detail. Robert’s fleet suffered heavy damage from a storm and was then defeated by the Venetian allies of the Greeks, leaving his army cut off, but he pursued the siege determinedly. Alexios brought up a relief by land; but it was a composite force with little coherence. When he attacked on 17 October, the only troops which fought well were his Anglo-Scandinavian Varangian Guard, which actually drove back the Norman cavalry. But, as at Hastings, combined actions by knights and missile-men defeated even these redoubtable fighters. Dyrrachium held out until February 1082, when it fell through treachery. Rebellion in Italy drew Robert away, allowing the Greeks to recover the city the following year; but in 1084 he was back. This time he had worked out the tactics by which to defeat the Venetians and, surrounding nine of their large vessels with his smaller craft, sank seven and captured two off Corfu. Guiscard died the following year, leaving his redoubtable son Bohemond to carry on the fighting, but Roger (his son 25

Brown, Mercenaries to Conquerors, pp. 94–96. Brown, Mercenaries to Conquerors, p. 96. The modern tendency to read cuneus as always meaning a wedge formation (usually of cavalry) needs to be challenged. See also p. 139 for alleged Byzantine use of “hostile wedges”. 27 Georgios Theotokis, The Norman Campaigns in the Balkans, 1081–1108 (Woodbridge, 2014), now provides the most detailed and authoritative account of the topic. 26



norman battle tactics 143

by his second marriage) inherited the Italian lands. Bohemond went on to become a great crusader and, as duke of Antioch, attacked Dyrrachium again in 1107–8. Emperor Alexios had learned not to confront the Normans in battle, and had also rebuilt his fleet significantly, so he adopted the tactics of blockading the besiegers so that Bohemond had to sue for peace. Bohemond died in 1111. Case Study 2: Durazzo, 1081 – Normans versus Byzantines This is the only detailed account of a Norman army fighting the Byzantines. The sources are good on both sides, especially because Anna, the historian daughter of Alexios I Komnenos, describes warfare very well. As discussed above, this was probably because she learned from her husband, Nicephorus Bryennius, an experienced general who wrote his own (now largely lost) history of the period. The Norman Army It is important to emphasise that the diversity of the Norman forces is often underestimated and misrepresented. Yes, they did possess heavy charging cavalry, but the assumption that they therefore must have been impetuous, requiring representation as obligatory chargers, for example, is mistaken. Rather, they presented professionalism and discipline on the battlefield. After all, the observation of Usamah ibn Munqhid, a well-known Muslim commentator, was that “The Franks are most cautious in warfare”.28 Admittedly this related to the Latin Kingdom of Jerusalem, where manpower was notoriously short, but it is a useful corrective to the carelessly assumed former representation. Anna Komnene writes approvingly of how professional Guiscard’s approach to war was: Not being satisfied with the men who had served in his army from the beginning and had experience in battle, he formed a new army made up of recruits without any consideration of age. From all quarters of Lombardy and Apulia he gathered them, over and under age, pitiable objects who had never seen armour, even in their dreams, but then clad in breastplates and carrying shields, awkwardly drawing bows to which they were completely unused and falling flat on the ground when they were allowed to march. Naturally these things provided an excuse for incessant trouble in Lombardy. Everywhere one could hear the lamentation of men and the wailing of women who shared in the bad fortune of their menfolk, for one mourned a husband unfit for military service, another a son who knew nothing of war a third a brother who was just a farmer or 28

Usama ibn Munqhid, The Book of Contemplation, ed. and trans. Paul M. Cobb (London, 2008), p. 25.

144

matthew bennett occupied in some other job ... Yet, however unused to soldiering they were .... he trained them daily and hammered his recruits into 29 a disciplined force.

This passage combines both an assumed superiority over the barbarian that is common in Byzantine histories with an appreciation of how dangerous an enemy the Normans had made themselves. It also serves to provide an excuse for the hero of the work, Alexios I, when he leads his army to defeat. Certainly the Norman troops were good enough to manoeuvre and fight effectively at Dyrrachium in 1081. Dyrrachium is the best-recorded battle of the era. It was fought between an invading force led by Robert Guiscard and a Byzantine imperial army under Alexios Komnenos I. Robert was besieging the city and Alexios sought to defeat him by clever tactics. He had with him three units of Guard cavalry, Turks under Tatikios, some 2,800 Manicheans (Bogomils) – a group of heretics who seemed to have served as mercenaries, and were presumably infantry – the emperor’s bodyguard, made up of Franks led by a man called FitzHumbert, the Varangian Guard under Wulfr (Nampites), and, perhaps, others not listed. And the Emperor was planning a sudden night attack from both sides upon Robert’s entrenchments. He commanded the whole native army to march by way of the salt-pits and attack from the rear, and he did not object to their undertaking this longer march 30 as it would add to the unexpectedness of their attack.

Robert seems to have anticipated this manoeuvre by “leaving his tents empty and crossing the bridge overnight”. He intended to attack from the front when he knew that the enemy were in position. He led the centre, with his son Bohemond in command on the left and Count Amicus (presumably) on the right. In response, Alexios also led from the centre, with generals commanding on both flanks. The Varangians were deployed in the centre and behind them a body of archers, the members of which were able to pass through and shoot at will. These appear to be the troops who are also called peltasts, which suggests lightly armoured troops. Meanwhile, the Turks arrived in the Norman rear, soon supported by a sally by the Byzantines inside the city. Robert feinted with the cavalry force, but this was masked off by the infantry. Then his right flank, under Amicus, launched a cavalry charge against the Varangian left. However, they were repulsed and ran back into the sea, up to their necks. Encouraged by their success, the Varangians marched against the Norman line, apparently rushing forward in 29

Anna Comnena, The Alexiad, trans. Elizabeth A. S. Dawes (Cambridge: Ontario, 2000), I. xiv, p. 27. 30 Alexiad, IV. vi, pp. 145–46.

norman battle tactics 145



some disarray. Robert’s response was to “order a detachment of infantry to fall upon them ... resulting in a massacre”. I have previously conceived this attack as being a combination of Norman cavalry to pin the Varangians in place and crossbow-armed infantry to shoot down even the best-armoured Scandinavian warriors. The attack may also have involved the well-trained spearman discussed above. This seems to have resulted in a general rout, as Duke Robert advanced upon the deserted imperial tent to plunder it along with the Byzantine baggage train. Alexios narrowly escaped death during this pursuit, purportedly turning around to unhorse and kill a Norman who was chasing him.31 Bohemond against the Byzantines Later battles between these two enemies were conducted by Guiscard’s son Bohemond and Alexios. What is particularly interesting from a tactical point of view is that the Byzantines are depicted as looking for ways to nullify the impact of the Norman cavalry. According to Anna: [Alexios] knew besides from his previous battles with Robert that the first onset of the Frankish cavalry upon their opponents was quite irresistible, he judged it would be best to have an attack by missiles made first upon the enemy by a small picked body of peltasts … But the Emperor dreading the irresistible first shock of the Latin cavalry hit upon a new device. He had wagons built, smaller and lighter than the ordinary ones, and four poles fixed to each, in these he placed heavy infantry so that when the Latins came dashing down at full gallop upon the Roman phalanx, the heavyarmed infantry should push the wagons forward and thus break the 32 Latins’ line.

Unfortunately for this plan, Bohemond got wind of it and deployed his forces so as to attack the flanks of the Byzantine force and rout it. The vehicles which Anna mentions are very interesting. When I first encountered them they were described as “scythed chariots”; but they were clearly something very different! In fact they were what came to be known as plaustrella, which were actually carts – not wagons – the difference being that having a two-wheel base, rather than four, made them very manoeuvrable, the four poles projecting from the back of the vehicle enabling it to be swung around rather like a wheel-barrow. Projecting in the front – although Anna does not say this – were a number of spears, usually four or six, creating a very dangerous obstacle for cavalry. They 31 32

Alexiad, IV. viii, p. 151. Ibid., V. iv, p. 87.

146

matthew bennett

were used by the Italian militias against Frederick Barbarossa in 1160 (and indeed much later).33 This plan having failed, Alexios attempted the same employment but with caltrops (a four-spiked metal device the size of a man’s hand).34 Called tribolos by the Greeks, they were designed so that there was always a spike sticking upwards, however they fell on the ground: For he prepared iron caltrops, and on the eve of the day on which he expected a battle, he had them spread over the intermediate part of the plain, where he guessed the Frankish cavalry would make their fiercest onslaught, thus aiming to break the first irresistible attack of the Latins by piercing the feet of their horses. And he ordered the Roman spearmen who held the front line, to ride forward at a measured pace in order not to be lamed by the caltrops, and to part to either side and then turn; the light-armed troops were to send a heavy shower of darts on the Franks from a distance, and the left and right wings were to fall upon them in a vehement charge. These indeed were my father’s plans but they did not escape Bohemond. For this is what happened: whatever plans my father made against him in the evening, the Frank knew by the morning. So he skilfully modified his plans in accordance with what he had been told, and engaged in battle but did not, as was his custom, begin with a frontal attack, but forestalling the Emperor’s intention, he raised the din of battle on either flank, bidding the front ranks keep still for a time. Then the battle became a hand-to-hand fight, the soldiers of the Roman army turned their backs to the Latins and had not even the courage to look them in the face again, as they had been thoroughly frightened beforehand 35 by their previous defeat.

What is interesting about these descriptions is that Anna is quite prepared to admit the moral and physical superiority of the Normans – actually matching their own self-image! Perhaps this is made possible by the knowledge that in a third encounter Alexios was to be victorious. In fact, this achieved by the simple and more conventional stratagem of a feigned flight and an ambush: [Bohemond] again followed his usual mode of procedure and thinking the Emperor was where he saw the imperial ensigns in the middle of 33

Plaustrella: Aldo A. Settia, “Passato e future nell ‘orizzonte tecnico’ di Guido da Vigevano c.1280–1350”, in Quaderno Sism 2016. Future Wars. Storia della distopia militare, ed. by Virgilio Ilari (Milan, 2016), pp. 93–108, esp. pp. 98–99. Also, as ribalds, 100 carts, each with ten spears, in the English battle-line at Crécy, 1346: Thom Richardson, The Tower Armoury in the fourteenth century (Leeds, 2016) pp. 136–37. 34 Mamuka Tsurtsumia, “ΤΡΙΒΟΛΟΣ: a Byzantine Landmine”, Byzantion 82 (2012), 415–22, incuding photograph of a surviving medieval caltrop, p. 422. 35 Alexiad, V. iv, p. 88.



norman battle tactics 147 the line, he dashed down upon this deception like a whirlwind. After a short resistance his opponents turned their backs and he rushed 36 after them in mad pursuit as in our previous descriptions.

Only then did Alexios move, launching a surprise counter-attack on the Normans’ camp. Meanwhile a body of archers (peltasts) were sent against the disordered chargers with orders: [Alexios] having detached other brave men, and a goodly number of peltasts he ordered them to ride quickly after Bryennius [the Norman second in command], and when they overtook him, not to start a close fight, but rather aim at the horses from a little distance and direct showers of arrows upon them. They did thus overtake the Franks and showered arrows upon the horses so that the horsemen were reduced to great difficulties. For every Frank is invincible both in attack and appearance when he is on horseback, but when he comes off his horse, partly due to the size of his shield, and partly to the long curved peaks of his shoes and a consequent difficulty in walking, then he becomes very easy to deal with and a different 37 man altogether, for all his mental energy evaporates, as it were.

However, Bohemond twice charges and defeats Byzantine troops with the usual charge. Only the adept action of an Uze auxiliary in seizing the Bohemond’s standard and pointing it towards the ground stemmed the Normans’ impetuosity and caused them to mill about in confusion. Also, interestingly, and on a subsequent occasion, the Normans, when acting as a rearguard, proved capable of countering the threat of archery. “Bohemond, with his superb knowledge of military tactics, ordered his men to stand firm in serried ranks protecting themselves shield to shield.” It is not quite clear if these men are dismounted, although the context makes that most likely. Remember too that the long shield also served to protect the unarmoured horses of the riders.38 Normans versus Turks Normans serving as Byzantine mercenaries found themselves fighting Turkish horse-archers when attempting to set themselves up as rulers in Asia Minor in the 1070s; but most of all in the campaigns of the First Crusade and in the early twelfth century in Syria.39 As we have seen, Bohe36

Alexiad, V. vi, p. 92. Alexiad, V. vi, p. 92. 38 Alexiad, V, vi, p. 171. 39 John France, Victory in the East: a Military History of the First Crusade (Cambridge, 1994). 37

148

matthew bennett

mond had experience of steppe tactics during his conflict with Alexios. He knew how dangerous they were when they directed their archery against the knights’ horses. Therefore it was crucial that he had a command role in the crusader host at the time of the battle of Gök Su (Dorylaeum) in June 1097. His instructions to form camp in order to provide a base and to disrupt the attacks of the Turks worked well, allowing a second crusader column to outflank the enemy and attack them in the rear, prompting a rout.40 He also played a role in the defeat of Kerbogha’s army outside Antioch a year later, taking charge of the mounted reserve. This battle has been completely reinterpreted by John France, who points out that most of the crusaders were dismounted and therefore only able to charge the enemy camp, while the small number of mounted knights covered their flank.41 This was also a victory for desperation, since the crusaders were in a fight-or-die situation and the Muslim force was not as coordinated or enthusiastic for the fight. The early crusade chronicles contain some good studies of the tactical devices open to the Latins when confronted with a mobile enemy. Between them it is possible to discern a standard response built around a cautious deployment and the utilisation of a reserve; for example, when Bohemond led an expedition to garner supplies during the long siege of Antioch. There was also a more serious confrontation against a relief force led by Ridwan of Aleppo in early February 1098, known as the Lake Battle. Bohemond led out seven hundred knights at night in order to surprise the advancing Muslims. John France makes good use of chronicle accounts to provide a detailed analysis of the encounter.42 Dividing his small force into six squadrons, Bohemond hid five of them behind a hill in order to ambush the Aleppan line march. These drove two squadrons of the Muslim cavalry into the following mass and, despite being outnumbered, held them up until Bohemond delivered the final blow with his reserve, scattering the enemy. Two important tactical considerations influenced the outcome. First, that the wet weather reduced the effectiveness of the Turkish horse archers (so negating the risk to the crusaders’ unarmoured horses). Second, that the Christian knights attacked “with lances forward” (erectis hastis) – the couched lance technique – and this both terrified and overthrew the already discomforted Muslim forces. Shock tactics delivered by disciplined and coordinated squadrons certainly proved invincible in the hands of an expert general such as Bohemond. This does not mean that they were universally successful. Indeed the greatest defeat suffered in the early decades of the crusader states was the

40 41 42

France, Victory in the East, pp. 143–96, esp. 173–81, incl. four maps. France, Victory in the East, pp. 282–93, and map p. 289. France, Victory in the East, pp. 245–51 and map p. 249.



norman battle tactics 149 43

Field of Blood in 1118. This was fought between Roger of Salerno, the Norman ruler of the city, and the Turkish war-lord Il-ghazi of Mardin leading a force of veteran horse archers. When the Turks employed their normal stratagem of withdrawing in the face of the Frankish heavy cavalry, the Antiochenes pursued too rashly and were surrounded by superior forces and exterminated.44 In the end, leadership and good tactics depended upon the quality of the commander and the troops at his disposal. When they were put in an impossible situation, all the bravery in the world was to no avail. While the Normans could be world beaters in many circumstances, the harsh reality of being outthought, outmanoeuvred, and outfought left any fantasies of invincibility far behind.

43

Nicholas Morton, The Field of Blood: The Battle for Aleppo and the Remaking of the Medieval Middle East (New York, 2018); Thomas Asbridge, The Creation of the Principality of Antioch 1098–1130 (Woodbridge, 2000). 44 France, Victory in the East, pp. 69–81 and map p. 78. Also, in more detail: Thomas Asbridge, “The Significance and Causes of the Battle of the Field of Blood”, Journal of Medieval History 23(1997), 301–16.

6

Venetian Reactions to the Normans of Southern Italy under Robert Guiscard: from Enmity to Congeniality

Şerban V. Marin

This chapter considers the manner in which some specific events involving the Normans in southern Italy at the time of Robert Guiscard are represented in the 237 Venetian chronicles that cover the period in question. These chronicles are preserved either in manuscript (215)1 or – more rarely (22) – in edited form. The texts were written between the thirteenth and eighteenth centuries, so they can be viewed rather as secondary sources. However, they express very well the perspective of Venetian society on certain events of the past. Usually, the Normans appear in the narratives of the Venetian chroniclers when Robert Guiscard threatened Byzantium under its ruler Emperor Alexios I Comnenus (1081–1118). The majority of the chronicles narrate the Byzantine emperor’s appeal to doge Domenico Selvo (1071–84) to participate in the defence of Romania against Robert (1081). Short versions Scenario A According to most Venetian chronicles (ninety-eight, or 41.35 per cent),2 the event unfolded very simply: 1

Located at the Biblioteca Nazionale Marciana (126), the library of Museo Civico Correr (69), the library of the Querini-Stampalia Foundation (5), the Archivio di Stato di Venezia (4), all of which are in Venice, and the Österreichische Nationalbibliothek in Vienna (11). 2 Brevis, 363; A latina, 89; M 37, 17a–17b; Correr 1499, 12a, col. 2–12b, col. 1; pseudoE. Dandolo, 52; M 2545, 26b, col. 1; PD 392c, 8a, col. 2; Correr 1013, [41b]–[42a]; M

152 – – – –

şerban v. marin Alexios is under attack by Robert Guiscard; the emperor makes an appeal to the Venetians; Doge Domenico Selvo assists the Byzantines; 3 the emperor owes his gratitude to the Venetians.

In this scenario the most common text of the confrontation with Robert Guiscard is as follows (according to chronicle pseudo-E. Dandolo): Ancora nel tempo del dicto [Duxe], Alexio imperator de Gretia mandò a Venesia domandando secorso et aida a çiò che contrastar podese a Ruberto Guiscardo del’insulla de Cecillia re, el qual era andado a dapnnificar alcuna parte de Romania. Alora meser lo Duxe cum conscentimento del povolo quello sì socorse de navillii bem armadi, i quali a la tornada soa, cum grande utelle et honor del dicto imperador, fu reccevudi in Veniesia cum salvamento, dela qual cosa l’imperador molto ringraciando, obligandossi a li Venetiani perpetualmente. (Also in the time of the mentioned [doge], Emperor Alexios of Greece sent [a mission] to Venice in order to ask for assistance and help to enable him to oppose Robert Guiscard the king of the island of Sicily, who had come to inflict damage upon a part of Romania. Hence, with the consent of the people, messer the doge assisted him with well-equipped ships, which on their return, with great utility and honour of the mentioned emperor, were received in Venice safe and sound, for which reason the emperor was grateful, being for ever obliged to the Venetians.) Z 18, 65b, col. 2; Filippo, [33a, col. 1–2]; M 2548, 11a, col. 1; M 2549, 15b, col. 1–15b, col. 2; M 89, 17a, col. 1; M 2556, 35; M IX 28, 11a, col. 2–11b, col. 1; D. Contarini, 69a, col. 1–69a, col. 2; F 6117, 52b, col. 1; M 104, 73b; M 319, 90b; M 38, 16a, col. 1–16a, col. 2; M 559, 31b; Vitturi, 13a, col. 2; Cicogna 2116, 18b, col. 2; Cicogna 2413, [14b]; M 2559, 14a, col. 2–15, col. 1 [=14b, col. 1]; Gradenigo 53, 9b, col. 2; ASV 59, 24a; ASV 61, 15a–b; M 2546, [18a–b]; DDR 121, 14b; M 788, 20a; ASV 60, 9a; Antonio, 13a; PD 482c, 25b; Correr 760, 9a, col. 1; Correr 873, 89b, col. 1; DDR 449, 12b; F 3458**, 37b; Cicogna 589, 29b–30a; Correr 710, 18a; Marco, 38b; M 2571, 53b–54a; M 2576, 17a–b; M 723, [78a]; Rosso, 20a; pseudo-Rotta, 9b–10a; M 125–3, 226b; M 555, 22b, col. 2–23a, col. 1; pseudo-Barbaro, 105a–106a, 106b–107a; Cicogna 1982, 9a, col. 1; M 322, 40a–b; pseudo-Abbiosi, 14a; M 2568 bis, 202a; M 2567, 25b, col. 1–25b, col. 2; PD 312c, [26b]; Cicogna 3712, 95a; M 728, 10a, col. 2; M 68, 66a; Cicogna 3725, 65b; M 513, 74a; M 514, 32b–33a; M 39, 19b; M 2581, 41a; Agostini-Tiepolo, 74a; Querini 16, 20b; Sansovino, 554; MG 327, 220; PD 378c, 35a; Veniera 791, 61b; pseudo-Trevisan, 52a, col. 2; Savina, 32a; M 793, 56b–57a (although placed under Doge Vitale Falier); Cicogna 3556–3, [17b]; M 2550, 61b; M 91 tert, 608b; Correr 1421, 97a–b; Correr 1307; MG 106, 40a–b; Cicogna 2831, 19b–20a; M 2395, 125b–26a; Contarena, 28b–29a; M 2669, 102b; M 44, 21b–22a; pseudo-Donà, 16b; M 54, 118a; M. Contarini, 19a–b; M 70, [32a]; M 75, 33b; M 102, 28a; M 66 bis, 84a; Correr 1456, 104b–105a; Sivos, 34b; Veniera-Flangini, 9a; M 64, 150b; M 74, 59b; Cicogna 2302, 36; M 1669, 555; Cicogna 2832, 66. 3 This latter component of the events is absent in some of the chronicles following scenario A: Brevis (which, instead of the honours and obligations assumed by Alexios, refers to the position of protosevastos ascribed to the doge); M 104; M 319 (which, instead of obligations, speaks of a “fraternal amor” between the two allies); ASV 60; Antonio; Marco; M 125–3; pseudo-Abbiosi; M 513; M 514; Sivos.



venetian reactions to the normans 153

Among the chronicles that follow this scenario, Barbaro, Savina, and M 74 provide longer versions, while using the same sequence of events. Three chronicles that relate this scenario replace Alexios with Nikephorus.4 As for M 70 and M 75, Robert and Guiscard are regarded as two different individuals. In the case of M 1669, the confrontation with the Normans takes place only in Puglia, while Sansovino and Correr 1307 place the action in Dalmatia rather than the Balkans. Scenario B A somewhat longer version of this scenario (in seventy-two of the chronicles, or 30.38 per cent)5 provides details about the composition of the Venetian fleet. M 2569 narrates as follows: In questo tempo lo Imperador de Constantinopoli lo mando a domandare socorso al doxe de Venexia per poder contrastare contra lo Re Roberto de Sicilia, loqual Re Roberto jera uegnudo per damnificar la Romania. Et aldido lo ditto doxe la ditta dimandaxon del subito lui fexe aparechiar vna grandissima armada che fo de naue XXXVJ[,] galie XIIIJ, VIIIJ tarete tute ben jn ponto. Et sapiando lo ditto Re Roberto la ditta armada del subito lui manda a dir alo Imperador, che lui voleua far paxe cun lui a che modo che lui volesse. (At this time, the emperor of Constantinople sent to ask for assistance from the doge of Venice in order to be able to oppose King Robert of Sicily, which King Robert had come to inflict damage upon Romania. And when the said doge heard the mentioned request, he immediately put in order a very great fleet, which was of 36 ships, 14 galleys, 9 tarete all of them well equipped.

4

Sansovino; Cicogna 3556–3; Correr 1307. M 2564, 63a; M 2569, 42a; Cigotto, 130b–31a; M 2034, 133b, col. 1–133b, col. 2; M 2563, 5a; Cicogna 2117, 23b; M 2560, 53a–b; M Z 20, 42a; Cicogna 592, 45a; ASV 58, 105b; M 2566, 7b–8a; Correr 1337, 52a–62b; F 6147, 105b–106a; Cicogna 3753, 92a, col. 1; Sanudo, 477; M 324, 30b–31a; M 541, 15b–16a; M 550, 62a–b; M 52, 98b; Correr 1046, 14a, col. 2; Correr 760 bis, 111b; Cicogna 1899, 46a; WL 74–3, 48b–49a; PD 380c, fasc. 4, 23a; Pigno, 151b–52a; Querini 13, 49a–b; M 2544, 33a, col. 1–33a, col. 2; Cavalli, 50a; M 51, 47a, col. 1–2; Zancaruolo, 14b; Gussoni, 179–80; PD 236c, 56a–b; Querini 15, 37a; F 6234, 53a; M 46, 21b; M 2573, 71a; M 2543, 20b–21a; Cicogna 2123, 14a; P. Cornaro, 22a; Cicogna 1982 bis, 114a; Correr 1327, 20a; Cicogna 3599, 70a–b; M 45, 25a–b; M 2568, 24b; M 47, 22b, col. 2–23a, col. 1; M Z 21, 10a; Cicogna 1898, [92a–b]; M 53, 114a; pseudo-Erizzo, 13b–14a; pseudo-Navagero, 961–62; Correr 1045, 48b (a shorter description); Grandis, 119a; T. Donato, 54b; M 728 bis, 140b–41a; Cicogna 351, 57b (a shorter description); M 2580, 116a; A. Valier, 110b; M 1568, 113b; Cicogna 3675, XXXIX b; M 393, [17b]; M 1586, 11b–12a; M 327, 56a; Correr 1305, 26a; Cicogna 590, 35b; M 2046, 115b–16a; Cicogna 2234, 293; Lio, 35b; M 43, 17a; M 2676, 39b–40a; M 1577, 94–95; M 2028, 13b–14a; Querini 36, 119b, col. 2. 5

154

şerban v. marin

And having information about this fleet, King Robert sent immediately to the emperor to tell him that he wanted to conclude peace with him on his terms.)

Usually there are about thirty-six ships, fourteen galleys, and the so-called bregantini   6 or tarete,7 numbering nine.8 In fact, the preponderance of ships over galleys may imply that this fleet fulfilled a provisioning rather than a military role. The account of events is more or less the same, with the exception that, in comparison with the previous version, these chronicles refer to the fact that when confronted with such a mobilisation King Robert immediately agreed to conclude peace on the emperor’s terms; thus, the Norman seems to be rather fearful of Venetian intervention in the conflict,9 so much so that a military confrontation is completely absent. In any event, this scenario could be regarded as the beginning of a process of humanising the Norman commander. Unlike in scenario A, the people’s consent to the doge’s endeavour in assisting Alexios is absent. Conversely, in addition to the honours received and the obligations undertaken by the emperor, some of the chronicles that adopt the more developed scenario B mention several holy relics presented to the Venetians on this occasion.10 According to M 2564: e da puo fado paxe lo Inperador chon lo re Ruberto chiamado Vischardo lo lezenzia larmada de Venizianj[,] e mando chon quela moltj chorpi santj e molte reliquie a donar a i Venizianj hoferandose senpre a tutj sue chomandj e che senpre luj e i suo sozersorij i sarja hubligadj. 6

M 2564; Cigotto; M 2034; M 2563; Cicogna 2117; M 2560; M Z 20; M 2566; Correr 1337; F 6147; M 541; M 550; Cicogna 1899; WL 74–3; PD 380c; Pigno; Gussoni; Querini 15; F 6234; M 46; Cicogna 2123; M 45; Cicogna 1898; Correr 1045; T. Donato; Cicogna 351; M 2580; M 393; M 1586; M 327; Correr 1305; Cicogna 590; M 2046; Cicogna 2234; M 43. 7 M 2569; Cicogna 592; Cicogna 3753; M 324; Correr 1046; M 2544; M 51; Zancaruolo; PD 236c; M 2543; Correr 1327; M 2568; M 47; M Z 21; M 53; Grandis; A. Valier; Cicogna 3675; M 2028 (where “noue tarette” becomes “naue tarette”); Querini 36. The tarete becomes carache in Sanudo. Other versions speak of galeotte/galiazze (ASV 58; Querini 13; P. Cornaro; Cicogna 3599; pseudo-Erizzo; M 1577), gombarizze (pseudoNavagero), or simply “altri nauilij” (M 52; Correr 760 bis; Cavalli; Cicogna 1982 bis; M 728 bis; M 1568; Lio). See also a combination of “36 naui tarete con altre nauilij pizoli” in M 2573. In M 2676 the term for this type of boat is in a lacuna. 8 The number is eight in Cicogna 3753; M 324; M 46; M 2543; Correr 1327; M 53; Grandis; A. Valier; M 2676; Querini 36; three in PD 380c; six in Cicogna 2123; no number in M 52; Correr 760 bis; M 2573; Cicogna 3675. 9 He is even said to have accepted peace “lagrimando” (weeping), according to M 550; Gussoni and M 393. 10 Among the chronicles in this scenario, this detail is absent in: M 2569; Cicogna 592; Sanudo; M 52; Correr 1046; Correr 760 bis; M 2544; Cavalli; Zancaruolo; PD 236c; Querini 15; F 6234; M 2543; Cicogna 1982 bis; Correr 1327; M 2568; M 47; M Z 21; Correr 1045; M 728 bis; Cicogna 351; M 1568; Cicogna 3675; M 327; Lio; M 2028.



venetian reactions to the normans 155 (and after the peace had been concluded between the emperor and King Robert called Guiscard, he dismissed the fleet of the Venetians and sent many holy bodies and many relics to be donated to the Venetians, offering himself to all their commands and that he and his successors would be for ever oblige.)

Strangely, M 393 refers to the holy relics offered by the emperor to the Venetians but fails to mention the honours received by the former. Combining scenarios A and B, which actually, despite the small differences mentioned above, are almost similar, results in a consistency of 71.73 per cent. None of these chronicles provides any details about the naval confrontation. It is not clear if there was an open struggle between the Venetians and Robert Guiscard, as the chronicles seem first and foremost to reduce the entire action to the conclusion of peace and, moreover, to emphasise simply the prompt assistance afforded by the Venetians to the Byzantines. Other (short) scenarios In a few other cases, beside scenarios A and B, Alexios’ gratitude is absent, as the context refers solely to the devastating defeat of the allies.11 One should also note the complete absence of Robert as a character in these cases. According to M 67: In questo tempo essendo questo Prencipe in lega con Alessio Imperator de Romania contra Normani et essendo con le sue armade à Durazzo per deffender quelo et essendo apizzade tutte le armade insieme et facendo fatti d’arme[,] subito Alesio con la sua armada fugi uerso la Morea[.] Vedendo il ditto Dose questo uolto la sua armada et fuggi uerso li litti Veneti. (At this time, this prince being in alliance with Emperor Alexios of Romania against the Normans and being with all his fleets at Durazzo in order to defend the mentioned [Alexios] and all the fleets being gathered together and making deeds of war, Alexios with his fleet fled immediately towards Morea[.] Seeing this, the said doge suspended his fleet and ran away towards the Venetian lagoons.)

Curti notes in summary that: “Le armi Venete contro Roberto Guiscardo et à fauore di Alessio Imperatore, auendo auuta la peggio […]. [The Venetian fleets [fought] against Robert Guiscard and in favour of Emperor Alexios, having the worst […].]”12 An extremely short version of this is also provided by Querini 63, which mentions only the conflict between the emperor of Constantinople and the Normans, and by Cicogna 3276-42, which notes the victory at Durazzo, ignoring any Byzantine involvement in the affair.13 11 12 13

M 67, 147a; M 2541, 117a; F 6166, 275b; pseudo-Barba, 24b. Curti, [15a]. Querini 63, 95b; Cicogna 3276-42, 10.

156

şerban v. marin

A particular case is M 2572, which provides a general description of the Venetian geopolitical position in the midst of the conflicts that involved the Normans, the Kingdom of Hungary, and Byzantium.14 Completely original descriptions are also provided by the thirteenthcentury Annales Venetici Breves: “Sciendum est, quia Dominicus Silvius dux ducavit annos XII et fecit bellum cum Ruberto Viscardo, unde fuit disperses Petrus Ursiulus et Henricus, filius eius et Dominicus Ursiulus, frater eius, ab imperatore traditi fuerunt et ibi mortui sunt in ipsa capcione”,15 and then referring again to: “Anno Domini millesimo octuagesimo, indicione quarta, exierunt Venetici contra Robertum Viscardum cum navibus in bello.”16 Further, there is the case of six texts which mention only a conflict between the Venetians and the inhabitants of Durazzo, without any reference to the Sicilians or to the Byzantines.17 The long version Scenario C The remaining chronicles (thirty-three, or 13.92 per cent), which include those accidentally regarded as more authoritative among the Venetian historical works, offer a much more extended version of the confrontation with the Normans.18 Most surpass the simplicity of scenarios A and B by including the context related to the first Norman intrusion into the Balkans. Among these, the majority refer to the situation that unfolded in the Byzantine court when Emperor Michael VII Doukas Parapinakes (1067–78) was dethroned by Nikephorus III Botaniates (1078–81). These amount to 11.39 per cent,19 while 14

M 2572, 8b–9a. Annales Venetici Breves, 86. This information is only replicated in chronicle Magno, 132a (see below, scenario C). 16 Annales Venetici Breves, 86. 17 MG 249, 43a; Cicogna 1983, [12b]; Cicogna 1900, [14b]; M 31, 92a; F 6821, 61a; Correr 1032, 48b. According to F 6821: “El ditto doxe rompete guerra a queli de Duraxo a tanto chel morite assaisime persone [...].” 18 Extensa, 216; Venetiarum Historia, 79; Monaci, 79–80; Cicogna 3518, 64b–65a; Biondo, 6; M 796, 45b–46a; Sabellico, 95, 97–98; P. Dolfin, 238a–39a; Marcello, XXX; Caroldo, 96–97; pseudo-Zancaruolo, clj b–clij a; F 6211, 18a; M 60, 11a; Cicogna 2754, 13b; Magno, 127a–28b; Giustinian, 19b; PD 391c, [16b]–[17a]; Doglioni, 67–68; Cicogna 3556–7, 9b; M 58, 8b–9a; M 59, 19b–20a; M 91 bis, 429a; F 6235, 75a; P. Morosini, 91; Vianoli, 160–61; M 2592, 15b; M 61, libro 4, 6–7; M 1999, 14a; F 6566, 21b; Diedo, 52; M 1833, 12b; Laugier, 394–97; M 974, 206–7. 19 Extensa; Venetiarum Historia; Monacis; M 796; Sabellico; P. Dolfin; Caroldo; pseudoZancaruolo; F 6211; M 60; Cicogna 2754; Magno; Giustinian; PD 391c; Doglioni; 15

venetian reactions to the normans 157

20

9.28 per cent regard the episode as allowing the Robert Guiscard’s intervention in favour of the former emperor as the necessary pretext for expansion into the Balkans while Nikephorus appealed to the Venetian doge. In this circumstance, the positions of the Venetians and the Normans are presented as definitively antagonistic. The same chronicles insist on the later coup d’état of Alexios Comnenus against Nikephorus – an event that did not shift the alliances, so that the Normans remained as the main opponents of Venice and Byzantium. The incursion of the Normans into Venetian areas of interest gave some of these chroniclers the opportunity to make a digression regarding the origins and early history of the newcomers (3.80 per cent21). Five chronicles even refer to the setting up of the Duchy of Normandy under Rollon.22 Their number increases to a certain extent when referring to the Normans in southern Italy and Sicily (7.59 per cent23), or to their relationship with the papacy and the Western Empire (6.33 per cent24). Particular attention should be paid to Cicogna 3556–7; M 58; M 59; F 6235; P. Morosini; Vianoli; M 2592; M 61; M 1999; Diedo; Laugier; M 974. 20 Extensa; Venetiarum Historia; M 796; Sabellico; Caroldo; pseudo-Zancaruolo; M 60; Cicogna 2754; Magno; PD 391c; Doglioni; Cicogna 3556–7; M 58; M 59; F 6235; P. Morosini; Vianoli; M 2592; M 61; Diedo; Laugier; M 974. 21 Sabellico, 93; pseudo-Zancaruolo, clj a; Giustinian, 19b; P. Morosini, 90; M 61, libro 4, 4; M 1999, 14a; Diedo, 51; Laugier, 392; M 974, 205. 22 M 796; Sabellico; pseudo-Zancaruolo; P. Morosini; M 61; Diedo; M 974. 23 Southern Italy (Extensa, 208, 212, 213, 215; Cicogna 3518, 65a; Sabellico, 93–95; P. Dolfin, 228a–b, 230b–31a, 232b–33a, 234a–b, 237b; pseudo-Zancaruolo, clj a, clj a–clj b; Magno, 102b, 111a, 111b, 112a, 117a, 118a, 122b, 124b, 127a, 129a, 131a–b; Giustinian, 19b; Doglioni, 65, 66; P. Morosini, 90, 90–91; M 61, libro 4, 5–6; M 1999, 14a; M 1577, 93; F 6566, 21a; Diedo, 51–52; M 1833, 12a–b; Laugier, 391–93; M 974, 205; one could add here chronicle pseudo-Erizzo, 13a from scenario B) and Sicily (Extensa, 213; Biondo, 7; Sabellico, 93, 94; PD, 232b, 234a; pseudo-Zancaruolo, clj a; Magno, 111b, 112a, 117a, 118a; Giustinian, 19b; P. Morosini, 90; M 61, libro 4, 6; M 1999, 14a; M 1833, 12a; M 974, 205). Beside Robert Guiscard, who is present in all these versions, further Norman characters from the Hauteville family are only episodically mentioned: William the Iron Arm (two chronicles: Sabellico, 94, 95; M 61, libro 4, 6); Drago (four chronicles: Extensa, 208; P. Dolfin, 228a; Magno, 102b; M 61, libro 4, 6); Humphrey (one chronicle: M 61, libro 4, 6); Roger Bosson, Robert Guiscard’s brother, the first count of Sicily (five chronicles: Extensa, 213; P. Dolfin, 232b, 234a; pseudo-Zancaruolo, clj a; Magno, 111b, 112a, 118a; M 974, 205–6). The arrival of the Normans in southern Italy is also present in seven other chronicles that follow scenarios A or B (M 2034, 133a, col. 2; PD 380c, fasc. 4, 23a; Querini 13, 49a; P. Cornaro, 81b; Cicogna 3599, 70a; pseudo-Erizzo, 13a; M 1577, 93), when referring to D. Contarini’s dogeship, although the information is condensed in just one phrase. According to PD 380c: “Jn questo tempo domentre che algune generation de zente coe Normani, Greci & Saratini [sic] dischoresse per la Pugia & dauali de grande afitiun, […].” 24 Extensa, 212, 213; Cicogna 3518, 65a; P. Dolfin, 232b–33a, 234b; Caroldo, 97; pseudoZancaruolo, clj a; Magno, 112a, 128b, 129a, 131a–b; Giustinian, 19b; Doglioni, 65; M 31,

158

şerban v. marin

Cicogna 3518, which shifts the entire episode to take place in favour of the Western Empire and Rome itself instead of Byzantium – even the name of the emperor becomes Alexander – and, unlike the other chronicles, practically invents an original history of Robert’s family in southern Italy.25 As a whole, the presence of the Normans prior to the confrontation with Alexios Comnenus – that is, during the dogeships of Domenico Flabanico (1032–43),26 Domenico Contarini (1043–71),27 and Domenico Selvo himself 28 – makes an appearance in 10.55 per cent of the Venetian chronicles. This proportion remains at roughly the same level when the chronicles narrate the Normans’ approach to Venetian areas of interest; I do not refer here to the five chronicles29 that mention a supposed Venetian incursion into Puglia but, rather, to the mention of the Norman menace in Dalmatia in seventeen cases (7.17 per cent30). References to the Normans increase exponentially only when they deal with Guiscard’s campaign in the Balkans once the direct Venetian interest in these developments is already acute. Scenario C provides some more specific details about the resulting military confrontation near Durazzo. Doge Selvo’s personal participation in the battle is very well documented in scenario C, which is presented in many chronicles. An additional detail is that in this scenario some chronicles resupply the structural composition of the Venetian fleet as presented in scenario B.31 91b; P. Morosini, 90; M 61, libro 4, 5, 6, 7; Diedo, 52; M 1833, 12a–b; Laugier, 393, 394, 395; M 974, 205–6. 25 Cicogna 3518, 64b–65a. 26 Extensa, 208; P. Dolfin, 228a–b; Magno, 102b. 27 Extensa, 212, 213; P. Dolfin, 230b–31a, 232b–33a, 234a–b; pseudo-Zancaruolo, clj a–clj b; Magno, 111a, 111b, 112a; Doglioni, 65; M 31, 91b; P. Morosini, 90–91; M 1577, 93; F 6566, 21a; M 974, 198. See also pseudo-Erizzo (scenario B), 13a. 28 Extensa, 215; Monaci, 79; Biondo, 6; Sabellico, 92–95; P. Dolfin, 237b; M 60, 10b–11a; Cicogna 2754, 13b; Magno, 117a, 122b, 124b, 127a, 129a, 131a–b; Giustinian, 19b; PD 391c, [16b]; Doglioni, 66–67; M 58, 8b; M 59, 19b; M 61, libro 4, 4–6; M 1999, 14a; Diedo, 51–52; M 1833, 12a–b; Laugier, 391–93; M 974, 204–6. 29 Doglioni, 65; P. Morosini, 90; F 6566, 21a; M 1669, 555; M 974, 198. 30 Extensa, 215; Monaci, 79; Sabellico, 92–93; P. Dolfin, 237b–38a; Marcello, XXX; F 6211, 18a; M 60, 10b–11a; Cicogna 2754, 13b; Magno, 121b, 128a, 129a; PD 391c, [16b]; M 58, 8b; M 59, 19b; M 2572, 8b; M 61, libro 4, 4, 6; F 6566, 21b; Diedo, 51; M 974, 204. 31 Cicogna 3518 (bregantini; in addition, this particular chronicle provides the names of all the captains of this fleet); pseudo-Zancaruolo (bregantini); F 6211 (not referring to brigantine/tarete); M 60 (only the galleys and navies are mentioned); Cicogna 2754 (only the galleys and navies are mentioned); Magno (eight tarete); PD 391c (only the galleys and navies are mentioned); M 58 (not referring to brigantine/tarete); M 59 (not referring to brigantine/tarete); F 6235 (galeotte); P. Morosini (carache; the number of galleys is eighteen instead of fourteen); Vianoli (carache; the number of galleys is eighteen instead of fourteen); M 61 (carache; the number of galleys is eighteen instead of fourteen); M 974 (carache; the number of galleys is eighteen instead of fourteen).



venetian reactions to the normans 159

Unlike the previous scenarios, where Alexios is presented as the Venetians’ ally, in scenario C it is Nikephorus who is acting alongside the Venetians. However, this version applies only in the beginning, and, following the coup d’état against the latter, Alexios enters the scene in this narrative as well. This time more details about the development of military affairs are provided. According to some chronicles, many Venetians were killed in battle and taken prisoner.32 The festive narration of scenarios A and B is diminished, and the Normans are now presented as fierce opponents. The impression that this was a difficult struggle is underlined by various expressions.33 Thus, the initial victory of the allies was not an easy one, as Caroldo underlines: “la fortuna si dimostrò, et all’una, et all’altra parte, molto varia (the chance proved to be very various for one side and the other)”. The same idea is relayed by Giustinian, who mentions: “si combattè molte hore da vna parte, & dall’altra con grande vccisione (both sides fought each other for many hours with much killing)”, and M 61: “e lungo tempo fù dubioso il conflitto pari essendo la risolutione di non cedere, et il desiderio

32

Extensa; Venetiarum Historia; Monaci; Biondo; M 796; Sabellico; P. Dolfin; Marcello; Caroldo; pseudo-Zancaruolo; Giustinian (“con vna spauenteuole vccisione”); M 2592; F 6566; Diedo. 33 “circa Durachium bellum commisit asperimum” (Venetiarum Historia); “Namque Normannae congressa classi victoriam sed cruentissimam obtinuit” (Biondo); “nec diu cunctatus in eum Silvius ferociter invehitur” (Sabellico); “sed longe diuerso exitu pugnatum” (Marcello); “si venne appresso Durazzo a battaglia crudele” and “si combattè molte hore da vna parte, & dall’altra con grande vccisione” (Giustinian); “l’hebbe uittoria ma con gran taiada” (Agostini-Tiepolo and Querini 16, although placed under Vitale Falier); “e hauè uittoria dopo l’hauer combatudo per molte hore con ocision de molti da una banda e da l’altra” (Savina); “la vittoria fù de’ Greci, & di Venetiani, tutto che sanguinosa” and “vi rimasero Venetiani perdenti con gran stragge, & rouina” (Doglioni); “li aquisto honorata vittoria laqual costo molto a Uenicianj” and “la giornata fu sanguinosa da luna et laltra parte” (Correr 1307); “in fiera battaglia vicino à Durazzo” (P. Morosini); “restarono le due armate Veneta, e Greca inuolte in vna calamità reciproca, & in vna strage, e sconfitta comune” (Vianoli); “et ala fine de una crudelissima bataglia l’armata de Venetiani runpero quela del Signor Ruberto et deti Normandi” (M 2592); “Seguirono due sanguinosi incontri presso Durazo” and “[…], fù tanto il dolore del Doge Seluo, e de suoi che doppo una lunga ed’ostinata contesa andarono finalmente: i Normandi in piega, lasciando à nostri una segnolata uittoria” (M 1999); “e alla città di Durazzo che da quelli era assediato fu fatta una crudele e teribile giornata” and “la rotta non fusse così grande, ma bene sanguinolente, dall’una e l’altra parte” (F 6566); “et auè vittoria, doppo d’auer combattudo per molte ore, con l’occision de molti da una banda, e da l’altra” (M 74); “Allettati di nuovo i Veneziani dalle promesse de’ Greci spinsero vigorose forze in loro ajuto, seguendo sanguinoso incontro” (Diedo). There is also a more metaphorical description in P. Morosini: “e l’armata così Venetiana, come Greca, con strage non minore superata, e mal trattata; cambiarono le speranze di felice vittoria, in totale disperatione dell’impresa”.

160

şerban v. marin

di uincere (and the battle was uncertain for long time, the determination not to surrender and the desire to win being the same)”. The advantages of one side or the other are also emphasised by Giustinian: “erano superiori di legni Normandi, ma i Greci, & i Venetiani di valore, & di isperientia nelle guerre da mare (the Normans were superior in number of ships, but the Greeks and the Venetians in value and experience in the naval battles)”. The Normans are thus not only humanised but presented as a force to be reckoned with. Chronicle M 61 goes so far as to exalt their “eroiche” (heroical) deeds. Although enemies, the Normans seem to be regarded rather with a feeling of congeniality. The first Norman–Venetian war From all the above versions a question arises about the representation of the past: did the Venetians act under the command of the Byzantine emperor, as simple mercenaries? In many cases, the manner of the description in the chronicles leads to such a conclusion. This is chiefly the result of the terms humble (“humele”) and humility (“humelitade”), used by twenty chronicles in relation to the doge when acting on behalf of the Empire,34 but also of the fact that in the end, according to forty-two chronicles, the emperor dismissed (“licenziò”) the Venetian fleet put at his disposal.35 Directly36 or indirectly,37 twenty-four chronicles are absolutely clear that the fleet was put under Alexios’ command.38 Finally, another hint could be that, according to most chronicles, Domenico Selvo did not participate in person, even though his direct presence in the action is noted in

34

Correr 1499; Correr 1013; M Z 18; Filippo; M 2549; M 2556; D. Contarini; M 38; Vitturi; M 2559; ASV 61; DDR 121; PD 482c; M 555; M 322; pseudo-Erizzo; Savina; Contarena; M 2669; M 44. This, however, could just be a simple copying error of the term “utile”. 35 M 2564; Cigotto; M 2034; M 2563; Cicogna 2117; M 2560; Cicogna 3518; M Z 20; ASV 58; M 2566; Correr 1337; F 6147; Cicogna 3753; M 324; M 541; Cicogna 1899; WL 74–3; PD 380c; Pigno; Querini 13; Gussoni; M 46; M 2573; P. Cornaro; Cicogna 3599; M 45; Cicogna 1898; M 53; pseudo-Erizzo; pseudo-Navagero; Magno; Grandis; T. Donato; M 2580; Cicogna 590; M 2046; Cicogna 2234; M 43; M 2676; M 1577; Querini 36; M 974. 36 M IX 28; Cigotto; Cicogna 2117; M 2560; Correr 1337; F 6147; WL 74–3; Pigno; Gussoni; Cicogna 1982; Cicogna 1898; T. Donato; M 728. 37 Correr 1013; Filippo; M 2556; M 2564; M 2034; F 6117; M 38; Correr 760; Cicogna 1899; PD 380c; M 44. 38 Due to a copying error, according to Cavalli, it was the emperor himself who prepared the fleet.



venetian reactions to the normans 161

forty-three chronicles (19.41 per cent)39. Usually, the presence of the doge is confined to simply sending the fleet to fight on behalf of Alexios, and nothing more. Meanwhile, leaving aside the particular case of Cicogna 3518, which involves an extremely detailed presentation of all the captains of the fleet, with a certain Paulo Buora as captain-in-chief, there are twenty-eight exceptions (11.81 per cent) referring to Antonio Orio as commander of the Venetian fleet.40 In regard to the composition of the Venetian fleet, beside the data offered in scenario B – and supplemented by some in scenario C – there are three further cases which mention only twenty-two navilii.41 Regarding the location of the battle, the city of Durazzo is mentioned by fifty-six chronicles.42 Thus, the proportion of chronicles referring to the siege of Durazzo is 23.63 per cent. In addition to the battles for this city on the modern Albanian coast, other geographical locations are mentioned in scenario C: Otranto as the place of gathering of the Norman forces in order to cross over to the Balkans (seven chronicles)43 – sometimes Taranto is also mentioned in this respect44

39

Extensa; Venetiarum Historia (“et subsequenter personaliter est progressus”); Biondo (only indirectly, “Duxitque in Guiscardum Sylvius classem, […]”); M 796; Sabellico (“Numerosa itaque classe Venetus Princeps in Normanos movit”); M 788; P. Dolfin; Correr 710; M 2576; Caroldo; pseudo-Rotta; pseudo-Zancaruolo; F 6211; M 60; Cicogna 2754; Magno; Agostini-Tiepolo; Querini 16; Giustinian; Sansovino (only indirectly); PD 391c; Savina; M 793 (under V. Falier); Doglioni; Cicogna 3556–7; M 58; M 59; M 67; F 6235; P. Morosini; M 2541; F 6166; Vianoli; M 2592; M 61; pseudo-Barba; M 1999 (only indirectly); F 6566; M 74; Diedo; M 1833; Laugier; M 974. 40 Cicogna 2413; ASV 59; DDR 449; M 723; Cicogna 2123; M 2567; PD 312c; Cicogna 3712; M 68; M 2580; Cicogna 3725; M 1568; MG 327; Veniera 791; M 793 (although under V. Falier); pseudo-Trevisan; M 2046; Correr 1421; Cicogna 2234; Lio; M 2395; M 43; pseudo-Donà; M 54; Correr 1456; Veniera-Flangini; M 64; Cicogna 2302. 41 Pseudo-Trevisan (but only on the margin of the text); M 125–3; M 2567. 42 Extensa; Venetiarum Historia; Monaci; M 319; Biondo; M 796; Sabellico; ASV 60; P. Dolfin; M 2567; Marcello; Caroldo; MG 249; pseudo-Barbaro; Cicogna 1983; Cicogna 1900; M 2567; pseudo-Zancaruolo; F 6211; M 60; Cicogna 2754; Magno; Giustinian; PD 391c; pseudo-Trevisan; Savina; Doglioni; M 31; Cicogna 3556–3; Cicogna 3556–7; M 58; F 6821; M 59; M 67; Correr 1032; Correr 1307; F 6235; Querini 63; P. Morosini; M 2541; F 6166; M 70; M 75; Vianoli; M 2592; M 61; pseudo-Barba; Sivos; M 1999; F 6566; M 74; Diedo; Cicogna 3276-42; M 1833; Laugier; M 974. 43 Extensa; Venetiarum Historia; Monaci; Sabellico; Caroldo; pseudo-Zancaruolo; Magno. 44 Sabellico; M 2592.

162

şerban v. marin 45

– Valona (eight texts), Albania,46 and Greece,47 more specifically, Corfu,48 Morea,49 Epirus,50 or “nell’acque trà l’Isole di Corfù, e Ceffalenia [sic]”.51 Beside these, the most frequently cited geographical location in scenarios A and B (125 chronicles) is the general term of Romania, in relation to the place where Robert invaded. Sometimes, due to obvious error, Romania becomes Romagna52 or even Roma.53 Conversely, some chronicles neglect to mention any geographical determination in the Balkan area,54 while M 1669 refers to Puglia, and Sansovino, Correr 1307, M 2572 and even Magno refer to Dalmatia, as already mentioned. In sum, the narrative of diplomatic–military events at the times of Doge Domenico Selvo is present in 217 chronicles (91.56 per cent), and is entirely ignored in only 20 chronicles.55 The second Venetian–Norman war The subsequent naval battle in 1084 against the Norman fleet was a surprising defeat for the Venetians. This appeared to be an event that the Venetians wished to forget, and one which their chroniclers chose to conceal. However, it is included in fifty-five chronicles (23.21 per cent, almost one quarter),56 45

Extensa; Venetiarum Historia; Monaci; Sabellico; P. Dolfin; pseudo-Zancaruolo; Magno; M 61. 46 Pseudo-Barbaro. 47 Pseudo-Barbaro; M 2572. 48 Magno. 49 M 67; M 2541; F 6166; pseudo-Barba. 50 M 1833. 51 Diedo; Cicogna 3276-42; M 974. 52 Correr 1499; M IX 28; M 2034; M 559; Correr 760; Cicogna 589; Cicogna 2123; Cicogna 1982; F 6234; M 728; Cicogna 351; Agostini-Tiepolo; M 91 tert; M. Contarini; M 66 bis. 53 Cicogna 3518; M 550. 54 Annales Venetici Breves; M 2563; M 541; Marco; Querini 15; M 125–3; Correr 1327; M 2568 bis; M 46; Rosso; M 2568 bis; Correr 1045; M 513; M 514; M 327; M 39; Sansovino; M 327; Correr 1305; Cicogna 590; M 2572; M 2676; Curti. 55 Canal; M IX 28 bis; Cicogna 2113; M 2565; M 798; M 2555; M 2037; M 628a; M 87; Cicogna 2815; M 793; M 303; M XI 77; Alberegno; pseudo-Alberega; Fabris; M 80; Correr 1306; M 2602; Fr. Falier. 56 Annales Venetici Breves, 86 (although, on this occasion, it only refers to the death of three members of the Orseolo family, and not to the defeat itself ); Extensa, 216, 218; Venetiarum Historia, 79; Monaci, 79–80; Biondo, 6, 6–7; M 796, 46a, 46b; Sabellico, 97–98, 98; ASV 60, 9a; P. Dolfin, 238b–39a, 240b; Marcello, XXX; M 2571, 54a; Caroldo, 97; MG 249, 43a; pseudo-Barbaro, 106b–107a; Cicogna 1983, [12b]; Cicogna 1900, [14b]; pseudo-Zancaruolo, clj b–clij a, clij b; pseudo-Erizzo, 15a; F 6211, 18b; M 60, 11a; Cicogna 2754, 13b; Magno, 128a–b, 132a, 134a; Giustinian, 19b, 20a; Sansovino,



venetian reactions to the normans 163

especially those in scenario C, which are once again more documented and more sober than scenarios A and B. Among them, only 7.17 per cent point to the location of this battle as being near Sasno island.57 The Norman victory over the Venetians and Greeks is reported to have taken place under either Domenico Selvo58 or the subsequent doge, Vitale Falier (1084–96).59 As for the latter doge, six chronicles present him as victorious.60 What is more surprising, however, is that there are no less than twentyone texts that mention two or more defeats inflicted to the Venetians,61 while, even more astonishingly, fifteen chronicles62 refer exclusively to the Norman victory under Selvo’s dogeship, leaving the previous Venetian victory aside. This maritime disaster is regarded as the reason why Domenico Selvo was forced by the Venetian people to resign from the dogeship (found in 24.89 per cent of the texts),63 even though this is in contradiction to 554, 555; Savina, 32b; Doglioni, 67, 68; M 31, 92a; Cicogna 3556–3, [17b]; Cicogna 3556–7, 9b, 10a; M 58, 9a; F 6821, 61a; M 59, 20a–b; M 67, 147a; Correr 1032, 48b; Correr 1307, 98a; F 6235, 75b, 76a; P. Morosini, 91, 93; M 2541, 117a; F 6166, 275b; M 70, [32a]; M 75, 33b; Vianoli, 161, 164; M 2592, 15b–16a; M 61, libro 4, 6–8; pseudo-Barba, 24b; Sivos, 34b–35a; M 1999, 14a, 14b; F 6566, 21b; Curti, [15a]; M 74, 59b; Diedo, 52–53, 53; Cicogna 3276-42, 10; M 1833, 12b–13a, 13a; Laugier, 397–99, 406–7; M 974, 207, 207–8, 211. 57 Extensa; Monaci; Biondo; M 796; P. Dolfin; pseudo-Zancaruolo; pseudo-Erizzo; F 6211; M 60; Cicogna 2754; Magno; Sansovino; Doglioni; M 58; M 59; F 6235; M 1999. 58 Annales Venetici Breves; Extensa; Venetiarum Historia; Monaci; Biondo; M 796; Sabellico; ASV 60; P. Dolfin; Marcello; M 2571; Caroldo; MG 249; pseudo-Barbaro; Cicogna 1983; Cicogna 1900; pseudo-Zancaruolo; Magno; Giustinian; Sansovino; Savina; Doglioni; M 31; Cicogna 3556–3; Cicogna 3556–7; F 6821; M 67; Correr 1032; Correr 1307; F 6235; P. Morosini; M 2541; F 6166; M 70; M 75; Vianoli; M 2592; M 61; pseudo-Barba; Sivos; M 1999; F 6566; Curti; M 74; Diedo; Cicogna 3276-42; M 1833; Laugier; M 974. 59 Extensa; Monaci; Biondo; M 796; Sabellico; P. Dolfin; pseudo-Zancaruolo; pseudoErizzo; F 6211; M 60; Cicogna 2754; Magno; Giustinian; Sansovino; Cicogna 3556–3; Cicogna 3556–7; M 58; M 59; F 6235; P. Morosini; Vianoli; M 1999; Diedo; Cicogna 3276-42; M 1833; Laugier; M 974. 60 M 67; M 2541; F 6166; pseudo-Barba; Sivos. One could add here M 793, which, erroneously, places the entire action under V. Falier, although the manner of presenting the events follows closely scenario A. 61 Extensa; Biondo; M 796; Sabellico; P. Dolfin; pseudo-Zancaruolo; Magno; Giustinian; Sansovino; Doglioni; Cicogna 3556–3; Cicogna 3556–7; F 6235; P. Morosini; Vianoli; M 1999; Diedo; Cicogna 3276-42; M 1833; Laugier; M 974. 62 ASV 60; MG 249; Cicogna 1983; Cicogna 1900; M 31; F 6821; M 67; Correr 1032; Correr 1307; M 2541; F 6166; M 70; M 75; pseudo-Barba; Curti. 63 Extensa, 216; Venetiarum Historia, 79; M 2545, 26b, col. 1–26b, col. 2; Monaci, 80; F 6117, 52b; M 796, 46a; Sanudo, 477; M 2546, [18b]; P. Dolfin, 239a; Marcello, XXX; M 2571, 54a; Caroldo, 97; MG 249, 43a; M 125–3, 226b; pseudo-Barbaro, 107a; Cicogna 1983, [12b]; Cicogna 1900, [14b]; M 2568 bis, 202a; M 2567, 25b, col. 2; pseudo-Zancaruolo, clij a; pseudo-Erizzo, 14a–b; Magno (referring to V. Falier), 132b;

164

şerban v. marin

chronicles in scenario A, which notes that the decision to help the Byzantines was taken with the people’s consent. The deposal of this doge is not always related to the defeat itself (as thirty-seven chronicles do explicitly relate it64), but sometimes (in the remaining twenty-two texts) the people’s reproach refers simply to the assistance given to the Byzantines in the conflict.65 In these latter cases, the forced resignation of Selvo seems to be worthy of mention, but the defeat itself is not. Another conclusion that could be drawn from this strange adulation of the Venetians – patricians and commoners alike – for Robert Guiscard is that we are able to discern the beginnings of some sort of collegiality towards the Normans that seems to grow in parallel with a gradual distancing from the traditional alliance with Byzantium.

M 514, 33a; M 2581, 41a; Agostini-Tiepolo, 74a; Querini 16 (referring to V. Falier: “et come el tornò, ch’l fosse priuado, la uerità ė ch l’hebbe uittoria ma con gran taiada”; on the contrary, for the campaign under Selvo, it mentions “et il Dose tornò à Venetia con grand’honor et trionfo”); Giustinian, 19b; Cicogna 2815, 72b (although it does not mention the campaign itself ); pseudo-Trevisan, 52a, col. 2; Savina, 32b; Doglioni, 67–68; M 31, 92a; Cicogna 3556–3, [17b]; Cicogna 3556–7, 9b; F 6821, 61a; M 67, 147a–b; M 91 bis, 429a; Correr 1032, 48b; Correr 1307, 98a; F 6235, 75b; Querini 63, 95b; P. Morosini, 91–92; M 2541, 117a; F 6166, 275b; M 70, [32a]; M 75, 33b; M 2602, 195a (although any mention of the war is absent); Vianoli, 161; M 2592, 16a; M 61, libro 4, 7–8; Sivos, 34b–35a; M 1999, 14a; F 6566, 21b; Curti, [15a]; M 74, 59b; Diedo, 53; Cicogna 3276-42, 10; Laugier, 399; M 974, 208. Without referring to the doge’s resignation, M 1833, 13a still notes that: “e torna a Venezia in sommo discredito”. Among them, several chronicles note the two versions associated with the end of Selvo’s dogeship, using generic formulae like “alcuni dicono” or “trouassi che”, suggesting that his dogeship would have been a successful one. 64 M 2545; F 6117; M 796; M 2546; Marcello; M 2571; Caroldo; MG 249; pseudoBarbaro (adding the derogatory expression “un Dose cosi uil, et effeminado”); Magno (but referring to V. Falier); M 2581; Giustinian; Savina; Doglioni; M 31; Cicogna 3556–3; Cicogna 3556–7; F 6821; M 67; M 91 bis; Correr 1307; F 6235; P. Morosini; M 2541; F 6166; M 70; M 75; Vianoli; M 2592; M 61; Sivos; M 1999; F 6566; M 74; Diedo; Cicogna 3276-42; Laugier; M 974. 65 Extensa; Venetiarum Historia; Monaci; Sabellico; Sanudo; P. Dolfin; M 125–3; Cicogna 1983; Cicogna 1900; M 2568 bis; M 2567; pseudo-Zancaruolo; pseudoErizzo; M 514; M 31; Cicogna 2815; pseudo-Trevisan; F 6821; Correr 1032; Querini 63 (only “per lo odio di Allessio Imperatore”); M 2602; Curti. Among them, Sanudo offers a more precise reason for this, mentioning the mercantile connection with the Normans; P. Dolfin refers to the many Venetians killed in action against the Normans, while Cicogna 1983, Cicogna 1900, M 31, F 6821, and Correr 1032 also refer to the many victims at the siege of Durazzo as the reason behind the doge’s resignation.



venetian reactions to the normans 165 The Matrimonial Alliance with Byzantium

Apparently unrelated to the topic, a number of chronicles (sixty-four, or 27 per cent66) refer to the wedding of Doge Domenico Selvo with a Byzantine princess, with some (thirty-one, or 13.08 per cent 67) emphasising that this was at the request of the Byzantine emperor. What we are interested in when referring to this event is that a number of texts (twenty-four, or 10.13 per cent)68 place the wedding in the context of the confrontations with the Normans. Going further, some of these chronicles establish a direct connection between the two episodes, while others seem only to suggest it by placing this matrimonial alliance immediately after the conclusion of peace, presenting it as a result of the emperor’s gratitude to the Venetians. The Strengthening of the Venetian–Byzantine Alliance The memory of the Venetian–Byzantine cooperation – fruitful or not – was maintained. This is proved by the fact that when Alexios accepted the request of the following doge, Vitale Falier, to receive the title of Duke of Dalmatia (and Croatia) (this episode is represented in most of the Venetian chroni-

66

Extensa, 215; Venetiarum Historia, 78–79; M 2545, 26b, col. 1; Monaci, 79; M 2034, 133b; F 6117, 52b, col. 1; M 796, 45a; Cicogna 3753, 92a, col. 1; Sabellico, 92; Sanudo, 477; M 324, 31a; M 2546, [18a]; P. Dolfin, 237a; PD 380c, fasc. 4, 23a–b; Querini 13, 49b; M 51, 47a, col. 1; Correr 710, 18a; M 2571, 53b; M 2573, 71a; Caroldo, 95; P. Cornaro, 82a; pseudo-Barbaro, 105b, 106a–106b, 108a; Correr 1327, 20a; Cicogna 3599, 70b; pseudo-Zancaruolo, clj b; M 53, 114a; pseudo-Erizzo, 14a; pseudo-Navagero, 962; Magno, 118a; Correr 1045, 48b; Grandis, 119a; A. Valier, 110b; M 2581, 40b; AgostiniTiepolo, 73b; Querini 16, 20a; Giustinian, 19a–b; Cicogna 2815, 72a; Savina, 32b, 32b; Doglioni, 65–66; Cicogna 3556–3, [17b]; M 1586, 12a; M 327, 56a; M 67, 146b–47a; M 91 bis, 428b; F 6235, 75a–b; Querini 63, 95b; P. Morosini, 91; M 2541, 116a–b; F 6166, 274b–75a; M 70, [31b]–[32a]; M 75, 33a–b; Vianoli, 160; M 2592, 15b; M 61, libro 4, 3–4; pseudo-Barba, 24a; Sivos, 34b; M 2676, 40a; Fr. Falier, 71–72; M 1577, 95; Querini 36, 119b, col. 2; M 74, 59b, 59b–60a; Laugier, 390–91; M 1669, 555; M 974, 203–4, 206, 208. Beside these, one should note the cases of Cicogna 3556–7, 9b; Diedo, 53; and Cicogna 3276-42, 10, which, although not mentioning the wedding episode per se, name the emperor as the doge’s “cugnato”. 67 Venetiarum Historia; M 2545; M 2034; F 6117; M 796; Cicogna 3753; M 324; M 2546; PD 380c; Querini 13; M 51; M 2571; M 2573; Caroldo; P. Cornaro; Cicogna 3599; M 53; pseudo-Erizzo; pseudo-Navagero; Magno; Correr 1045; Grandis; A. Valier; M 2581; M 1586; M 327; M 91 bis; M 2676; Fr. Falier; M 1577; Querini 36. 68 Cicogna 3753; Sanudo; M 324; PD 380c; Querini 13; M 2573; P. Cornaro; pseudo-Barbaro; Cicogna 3599; M 53; pseudo-Erizzo; pseudo-Navagero; Correr 1045; Grandis; Savina; M 1586; M 327; F 6235; P. Morosini; M 2676; M 1577; Querini 36; M 74; M 1669.

166

şerban v. marin 69

cles – 90.30 per cent)

69

the reason for it is sometimes (in 18.57 per cent)70

Extensa, 217; A latina, 90; Venetiarum Historia, 80; M 37, 17b; Correr 1499, 12b, col. 1; pseudo-E. Dandolo, 52–53; M 2545, 26b, col. 2; PD 392c, 8b, col. 1; Correr 1013, [42b]; M Z 18, 65b, col. 2–66a, col. 1; Filippo, [33b, col. 1]; M 2548, 11a, col. 1; M 2549, 15b, col. 2; M 89, 17a, col. 2; Monaci, 80; M 2556, 35; M IX 28, [11b, col. 1]; M IX 28 bis, [126a]; M 2564, 63a–b; M 2569, 42a; Cigotto, 131b; D. Contarini, 69a, col. 2; M 2034, 134a, col. 1; F 6117, 52b, col. 1; M 104, 73b; M 2563, 5a; M 38, 16a, col. 2; Cicogna 2117, 23b; M 2560, 53b; Cicogna 3518, [65b]; M Z 20, 42a–b; M 2565, 34b; Cicogna 592, [34b]; M 559, 31b; Vitturi, 13a, col. 2–13b, col. 1; Cicogna 2116, 19a, col. 1; Cicogna 2413, [15a]; M 2559, 15, col. 1 [=14b, col. 1]; Gradenigo 53, 9b, col. 2–10a, col. 1; ASV 59, 24a; ASV 58, 105b; M 2566, 8a; Correr 1337, 52b; F 6147, 106b; Cicogna 3753, 92a, col. 2; Sabellico, 98; ASV 61, 15b; Sanudo, 477–78; M 324, 31a; M 2546, [18b]; DDR 121, 14b; M 541, 16a; M 788, 20b; M 550, 62b; Antonio, 13a; M 52, 99a; ASV 60, 9a; P. Dolfin, 239a–b; PD 482c, 26a; Correr 1046, 14a, col. 2–14b, col. 1; Correr 760, 9a, col. 1–9a, col. 2; Correr 760 bis, 111b; Cicogna 1899, 46b; WL 74–3, 49a; PD 380c, fasc. 4, 23b; Correr 873, 89b, col. 1–89b, col. 2; Pigno, 152a–b; Querini 13, 49b; DDR 449, 12b; M 2544, 33a, col. 2; F 3458**, 37b–38a; Cicogna 589, 30a; M 798, xii b; Cavalli, 50b; M 51, 47a, col. 2; Correr 710, 19a; M 2555, 14a–b; Marcello, XXX; Zancaruolo, 14b; Gussoni, 180–81; PD 236c, 56b; Querini 15, 37b; F 6234, 53a–b; M 46, 22a; M 2571, 54a; M 2576, 17b; M 2573, 71b; Caroldo, 98; M 723, 78a; Rosso, 20b; M 2543, 21a; MG 249, 43a–b; Cicogna 2123, 14b; P. Cornaro, 82a–b; pseudo-Rotta, 10a; M 125–3, 226b; M 555, 23a, col. 1; pseudo-Barbaro, 108a; Cicogna 1982, 9a, col. 1–9a, col. 2; Cicogna 1982 bis, 114b; Cicogna 1983, [12b]; Cicogna 1900, [14b]; M 2037, 157b; M 322, 40b; Correr 1327, 20b; Cicogna 3599, 70b; pseudo-Abbiosi, 14b; M 45, 25b; M 2568, 24b; M 2568 bis, 202a; M 2567, 25b, col. 2–26a, col. 1; M 47, 23a, col. 1; pseudo-Zancaruolo, clij b; M Z 21, 10a; Cicogna 1898, [93a]; M 628a, 76b; M 53, 114b; PD 312c, [27a]; Cicogna 3712, 95a; pseudo-Erizzo, 14b; pseudo-Navagero, 962; F 6211, 18b; M 60, 11a; Cicogna 2754, 13b; Magno, 133b; Correr 1045, 49a (only as a chapter title); Grandis, 119a–b; T. Donato, 55a; M 728, 10a, col. 2–10b, col. 1; M 728 bis, 141a; Cicogna 351, 57b; M 68, 66b; M 2580, 116a; Cicogna 3725, 66a; M 513, 74a; M 514, 33a; A. Valier, 111a; M 87, 8b; M 39, 20a; M 2581, 41a; Giustinian, 18b–20a; M 1568, 114a; Sansovino, 555; MG 327, 221; Cicogna 2815, 72b; Cicogna 3675, XXXIX a; PD 378c, 35b; PD 391c, [17a]; Veniera 791, 62a; pseudo-Trevisan, 72a, col. 2–72b, col. 1; Savina, 32b; M 793, 56b; Doglioni, 68; M XI 77, 18b–19a; M 31, 92a; Cicogna 3556–3, [17b]; Cicogna 3556–7, [9b]; M 1586, 12a; Alberegno, 19a–b; M 58, 9a; M 327, 56a; M 2550, 62a; F 6821, 61a; Correr 1305, 26a; Cicogna 590, 36a; M 2046, 116a; M 59, 20a; pseudo-Alberega, 359b; M 91 bis, 429b; M 91 tert, 608b; Correr 1032, 49a; Correr 1421, 97b; Correr 1307, 98a–b; MG 106, 40b–41a; F 6235, 75b–76a; Querini 63, 95b; Fabris, C 22a; Cicogna 2831, 20a; Cicogna 2234, 294; Lio, 36b; P. Morosini, 92; M 2395, 126a; Contarena, 29a; M 2669, 102b–103a; M 43, 17a–b; M 44, 22a; pseudo-Donà, 17a; M 54, 118b; M. Contarini, 19b; M 80, 89b; M 70, [32a–b]; M 75, 33b; M 102, 28b; M 2592, 16a; M 61, libro 4, 8; M 66 bis, 84a–b; Correr 1456, 105a; Sivos, 35b; M 1999, 14a–b; M 2676, 40a; Fr. Falier, 73; M 1577, 96; M 2028, 14a; Querini 36, 120a, col. 1; Veniera-Flangini, 9b; M 64, 151a; F 6566, 21b; Curti, [15a]; M 74, 60a; Diedo, 53; Cicogna 3276-42, 10; M 1833, 13a; Laugier, 403–6; Cicogna 2302, 36–37; M 974, 211; Cicogna 2832, 66–67. 70 M 2564; Cigotto; Cicogna 2117; M 2560; M 2565; ASV 58; M 2566; Correr 1337; F 6147; Cicogna 3753; M 324; Cicogna 1899; WL 74–3; Pigno; Querini 13; M 798; M 51; Gussoni; M 2573; Caroldo (referring to “Narentani et Slavi” instead of “Normani”); P. Cornaro; pseudo-Barbaro; M 2037; Cicogna 3599; M 45; Cicogna 1898; M 628a; M 53;



venetian reactions to the normans 167

clearly noted: “per lo sochorso che lor i auea dado chontra el re de Zezilia chome nuj avemo dito per avantj (for the assistance that they had given against the king of Sicily, as we mentioned above)” (according to M 2564). Other cases (21.10 per cent)71 refer to the Normans only indirectly, generally mentioning the services provided by the Venetians to Alexios: “per li seruisij che llo aueua reçeuudo (for the services that he had received)” (according to Correr 1499). The total proportion of sources that mention Alexios’ obligations in the context of the Venetian–Byzantine alliance is 39.66 per cent. In all cases, the Normans are once again invoked in the chronicles. What is curious is that eight of them, which now refer to previous services against the Normans, had not actually provided any details regarding the confrontation with the Normans itself.72 The death of Robert Guiscard (1085) is not very well represented, not being an event of direct interest to the Venetians, but it is still included, albeit in only nine chronicles.73 The third Venetian–Norman war Contrary to historical reality, the Venetian chronicles place Robert Guiscard as present also at a later event, namely the subsequent anti-Byzantine campaign of his son, Bohemund of Tarent, during the dogeship of Ordelafo Falier (1102–17). As a whole, this new military confrontation is noted in 72.46 per cent of the 236 Venetian texts that cover this period.74 pseudo-Erizzo; pseudo-Navagero; Grandis; T. Donato; A. Valier; M 1586; Alberegno; pseudo-Alberega; M 91 bis; Fabris; M 80; M 61; M 2676; M 1577; Querini 36; M 974. 71 Correr 1499; M 2545; Correr 1013; M Z 18; Filippo; M 2548; M 2549; M 2556; D. Contarini; M 2034; F 6117; M 2563; M 38; M Z 20; Vitturi; Cicogna 2116; M 2559; ASV 61; M 2546; DDR 121; M 541; M 788; M 550; PD 482c; PD 380c; Correr 710; M 46; M 2571; M 2576; Rosso; pseudo-Rotta; M 555; M 322; F 6211; M 39; M 2581; PD 378c; Savina; M 58; M 2550; Correr 1305; Cicogna 590; Contarena; M 2669; M 44; M. Contarini; M 70; M 75; M 66 bis; M 74. 72 M 2565; M 798; M 2037; M 628a; Alberegno; pseudo-Alberega; Fabris; M 80. 73 Extensa, 219; Monaci, 80; Biondo, 7; P. Dolfin, 243a; pseudo-Zancaruolo, clij b; Magno, 134a; Cicogna 2815, 72b; Doglioni, 68; M 61, libro 4, 7. In addition, Magno refers to the controversy between Roger Bosson and Bohemund. 74 Annales Venetici Breves, 90; Brevis, 363; Extensa, 226; A latina, 93; Venetiarum Historia, 89; M 37, 18a; Correr 1499, 13a, col. 1; pseudo-E. Dandolo, 55; A. Morosini, 5; M 2545, 28a, col. 1–28a, col. 2, 29b, col. 1 (two different Norman campaigns); PD 392c, 8b, col. 2–9a, col. 1; Correr 1013, [44b]; M Z 18, 66b, col. 1; Filippo, [35a, col. 2]–[35b, col. 1]; M 2548, 11b, col. 1; M 2549, 16b, col. 1; M 89, 18a, col. 1; Monaci, 95; M 2556, 37; M IX 28, 12a, col. 1; M 2564, 64b; M 2569, 43a; Cigotto, 137a–b; D. Contarini, 69b, col. 2; M 2034, 135b, col. 1–135b, col. 2; F 6117, 53a, col. 2 (two different campaigns); M 104, 73b; M 2563, 5b; M 319, 91a; M 38, 17a, col. 1; Cicogna 2117, 25a; M 2560, 54 bis a; Cicogna 3518, 66b–67a; M Z 20, 43b; Cicogna 592, 46b; M 559, 33a; Vitturi, 14a, col. 1; Cicogna 2116, 19b, col. 2; Cicogna 2413, [16a]; M 2559, 15a, col. 2; Gradenigo 53, 10a, col. 2; ASV

168

şerban v. marin

Indeed, in the vast majority of the chronicles that mention him in this new context Robert is rightfully designated as Bohemund’s father. In some cases, however, he appears to be still alive: either the two Normans (Robert as king of Sicily and Bohemund as prince of Antioch) cooperate against the same Alexios I (eleven chronicles),75 or Robert acts on his own, completely substituting for his son as opponent of the new Venetian–Byzantine alliance (seventeen texts).76 What is more, according to six of the works that follow this scenario, the family relationship between the two Norman leaders is reversed, with Robert being regarded as Bohemund’s son.77 It seems that the chroniclers here find it easier to present one and the same enemy (Robert) versus one and the same ally (Alexios). 59, 25a; ASV 58, 106a; M 2566, 8b–9a; Correr 1337, 55a; F 6147, 109b; Cicogna 3753, 95a, col. 1–95a, col. 2; Sabellico, 105 (but under Doge Vitale Michiel I, 1096–1102), 124 (still under V. Michiel I), 125–26, 126–27; ASV 61, 16a; Sanudo, 482; M 324, 33b–34a; M 2546, [19b], [20b] (two different campaigns); DDR 121, 15b; M 541, 17b; M 550, 64a; M 788, 21b; Antonio, 13b; M 52, 100a; ASV 60, 9a; P. Dolfin, 251a–b; PD 482c, 27a; Correr 1046, 14b, col. 2; Correr 760, 9b, col. 1; Correr 760 bis, 112a; Cicogna 1899, 47b; WL 74–3, 51a–b; PD 380c, fasc. 4, 24b; Correr 873, 90a, col. 2; Pigno, 154b; Querini 13, 51a; DDR 449, 13b; M 2544, 34a, col. 1; F 3458**, 39a–b; Cicogna 589, 31b; Cavalli, 51a–b; M 51, 50a, col. 2; Marco, 39a; Zancaruolo, 15b; Gussoni, 186; PD 236c, 57b; Querini 15, 38a–b; M 46, 22b; M 2571, 56b–57a (two different campaigns); M 2576, 18b; M 2573, 75a–b; M 723, [79a]; Rosso, 21a; M 2543, 23a; Cicogna 2123, 16a; P. Cornaro, 83b–84a; pseudo-Rotta, 10b; M 125–3, 227a; M 555, 24a, col. 1; pseudo-Barbaro, 115a–b; Cicogna 1982, 9b, col. 1; Cicogna 1982 bis, 115a; M 322, 41a–b; Correr 1327, 21a; Cicogna 3599, 71a–b; pseudo-Abbiosi, 15a; M 45, 27a; M 2568, 25a; M 2568 bis, 203a; M 2567, 27b, col. 2; M 47, 23b, col. 2–24a, col. 1; pseudo-Zancaruolo, cl viiij b; M Z 21, 11a; Cicogna 1898, [96b]; M 53, 117b; PD 312c, 27b; Cicogna 3712, 96a; pseudo-Erizzo, 19a; pseudoNavagero, 964; Magno, 176a–b; Grandis, 122a–b; T. Donato, 60a–b; M 728, 10b, col. 2; M 728 bis, 141b; M 68, 66b; M 2580, 117a; Cicogna 3725, 66b; M 514, 35a–b, 37b–38a (two different campaigns); A. Valier, 115b–16a; M 39, 20b–21a; M 2581, 45a, 53a–b (two different campaigns); M 1568, 115a; MG 327, 231; Cicogna 2815, 76b–77a; Cicogna 3675, XXXX b; PD 378c, 36a; Veniera 791, 62b; pseudo-Trevisan, 53b, col. 2; Savina, 34b; M 793, 57a; Doglioni, 76–77; M 31, 96b (the opponents are not mentioned at all and the battles take place in Dalmatia); Cicogna 3556–3, [18a]; Cicogna 3556–7, 10b (the battle takes place in Puglia); M 1586, 13a; M 2550, 63b; Correr 1305, 27a; Cicogna 590, 36b; M 91 tert, 609a; Correr 1421, 98b; MG 106, 43a; F 6235, 81b (the opponents are not mentioned and the battle takes place in Puglia); Querini 63, 96b; Cicogna 2831, 20b; Cicogna 2234, 297; Lio, 39b; P. Morosini, 100; M 2395, 127a–b; Contarena, 30b; M 2669, 104a–b; M 44, 23a; pseudo-Donà, 17b; M 54, 126a; M. Contarini, 20b; M 102, 29b; M 61, libro 4, 17; M 66 bis, 85a; Correr 1456, 106a; M 2676, 42a–b; M 1577, 102–3; M 2028, 15b; Querini 36, 122a, col. 2–122b, col. 1; Veniera-Flangini, 10a; M 64, 160a; M 74, 64a; Cicogna 2302, 39; M 1669, 556–57; Cicogna 2832, 69. 75 Cicogna 3753; M 324; M 2573; M 2568 bis; M 53; pseudo-Erizzo; pseudo-Navagero; Grandis; A. Valier; M 2676; Querini 36. 76 M 2545; F 6117; ASV 58; M 2546; Querini 13; M 2571; P. Cornaro; pseudo-Barbaro; Cicogna 1982 (the expression is confusing: “Bonamente e so fiolo di Ruberto e Guiscardo”); Correr 1327; Cicogna 3599; M 514; M 2581; Savina; M 1577; M 74; M 1669. 77 M 2545; F 6117; M 2546; M 2571; M 514; M 2581.



venetian reactions to the normans 169 The Representation of Robert Guiscard

It is also interesting to note the various titles used by the Venetian chroniclers for Robert, who is mostly regarded as king of Sicily, rather than duke of Apulia. It seems that, for the Venetian texts, Sicily was a more notable political entity. Robert’s political title: 78

79

– Robert as king: 70.89 per cent; among them, only five chronicles mention him simply as king (without any geographical designation), while the great majority of the rest refer to him as king of Sicily; 80 – Robert as duke/count/senior: 9.70 per cent; 78

A latina; Venetiarum Historia; M 37; Correr 1499; pseudo-E. Dandolo; M 2545; PD 392c; Correr 1013; M Z 18; Filippo; M 2548; M 2549; M 89; M 2556; M IX 28; M 2564; M 2569; Cigotto; D. Contarini; M 2034; F 6117; M 104; M 2563; M 38; Cicogna 2117; M 2560; Cicogna 3518; M Z 20; M 2565; Cicogna 592; M 559; Vitturi; Cicogna 2116; Cicogna 2413; M 2559; Gradenigo 53; ASV 59; ASV 58; M 2566; Correr 1337; F 6147; Cicogna 3753; ASV 61; Sanudo; M 324; M 2546; DDR 121; M 541; M 788; M 550; Antonio; M 52; ASV 60; P. Dolfin; PD 482c; Correr 1046; Correr 760; Correr 760 bis; Cicogna 1899; WL 74–3; PD 380c; Correr 873; Pigno; Querini 13; DDR 449; M 2544; F 3458**; Cicogna 589; M 798 (only later, during V. Falier); Cavalli; M 51; Zancaruolo; Gussoni; PD 236c; F 6234; M 46; M 2571; M 2573; M 723; Rosso; M 2543; Cicogna 2123; P. Cornaro; pseudo-Rotta; M 125–3 (only later as “Ruberto”); M 555; pseudo-Barbaro (only later as “Re de Sicilia”); Cicogna 1982; Cicogna 1982 bis; M 2037 (only later, under V. Falier); M 322; Correr 1327; Cicogna 3599; pseudo-Abbiosi; M 45; M 2568; M 2568 bis; M 2567; M 47; Cicogna 1898; M 628a (only later, under V. Falier); M 53; PD 312c; Cicogna 3712; pseudo-Erizzo; pseudo-Navagero; Correr 1045; Grandis; T. Donato; M 728; M 728 bis; Cicogna 351; M 68; M 2580; Cicogna 3725; M 513; M 514; A. Valier; M 39; M 2581; Agostini-Tiepolo (the strange formula of “il Conte Roberto Re di Sicilia”); Querini 16 (“il Conte Ruberto Re di Sicilia”); M 1568; MG 327; Cicogna 3675; PD 378c; Veniera 791; pseudo-Trevisan; Savina (the combined designation of “Re de Cicilia e duca della Puglia e de Calauria”); M 793 (only later, under V. Falier); M 393; M 1586; Alberegno (only later, under V. Falier); M 327; M 2550; Correr 1305; Cicogna 590; M 2046; pseudoAlberega (only later, under V. Falier); M 91 tert; Correr 1421; MG 106; Fabris (only later, under V. Falier); Cicogna 2831; Cicogna 2234; Lio; M 2395; Contarena; M 2669; M 43; M 44; pseudo-Donà; M 54; M. Contarini; M 80 (only later, under V. Falier); M 2602 (only later); M 102; M 66 bis; Correr 1456; M 2676 (only noted in the margin of the text); M 1577; M 2028; Querini 36; Veniera-Flangini; M 64; M 74; Cicogna 2302; Cicogna 2832. 79 Venetiarum Historia; M 788; P. Dolfin; M 2576; M Z 21. 80 P. Dolfin (“Conte di Puglia”); Correr 710 (“el Conte Ruberto de Sicilia”); Caroldo (“Prencipe di Taranto et di Puglia”); pseudo-Barbaro (only later as “Duca de Calabria, et Pugia”); F 6211 (“Duca de Puglia de Normani”; then, “Duce de Normani”); M 60 (“Duca de Normani”); Cicogna 2753 (“Duca de Puglia de Normani”; then, “Duca de Normani”); Magno (“Conte de Apulia”; then, “Ducha de Normani”); Agostini-Tiepolo (the combined formula of “il Conte Roberto Re di Sicilia”); Querini 16 (“il Conte Ruberto Re di Sicilia”); Sansovino (“Duca di Puglia”); PD 391c (“Duca de Normani”); Savina (the combined designation of “Re de Cicilia e duca della Puglia e de Calauria”); Doglioni (“Duca di Puglia, & di Calabria”); M 58 (“Duca de Normani”); M 59 (“Re” [erased and replaced later by] “Duca di Puglia de Hormani” [sic]; then, “Duca de

170

şerban v. marin 81

– Robert with no political title: 12.24 per cent.

Robert’s name: 82

– Robert as Guiscard: 76.37 per cent. Sometimes (that is, in ten chronicles), Guiscard is highlighted as a nickname/sobriquet by the use of terms 83 like “cognominato”, “chiamado”, or “ditto”. The chronicle Doglioni notes: “cognominato Guiscardo (che significa bellicoso) (nicknamed Guiscard [which means warrior])”, thus providing an etymological explanation. Beside “Viscardo” and its variations, originating in the Latin Hormani”); F 6235 (“Duca di Puglia, et Calabria”); M 2572 (“Prencipi di Puglia”); M 70 (“Signori di Normani”); M 75 (“Signori de Normani”); M 2592 (“Signor Ruberto”); Sivos (“Duca di Puglia, et delle genti Normani”); M 74 (“Re di Sicilia, e Duca della Puglia, e de Caluuria” [sic=a]). 81 Annales Venetici Breves; Brevis; Extensa; M 319; Monaci; Biondo; M 796; Sabellico; Marco; Querini 15 (only “de Sicilia”); pseudo-Zancaruolo; Giustinian (only as “capitano”); Cicogna 2815 (only later as “Roberto”); Cicogna 3556–3 (only as “capitano de Normandi”); M 67; P. Morosini; M 2541; M 2572; F 6166; Vianoli; M 61; M 1999; F 6566; Curti; Diedo; M 1833; Laugier; M 1669; M 974. 82 Annales Venetici Breves; Brevis; Extensa; A latina; Venetiarum Historia; M 37; Correr 1499; pseudo-E. Dandolo; M 2545; PD 392c; Correr 1013; M Z 18; Filippo; M 2548; M 2549; M 89; Monaci; M 2556; M IX 28; Cigotto; D. Contarini; M 2034 (only later as “Vyscharo”); F 6117; M 104; M 319; M 38; Cicogna 2117; M 2560; Cicogna 3518; Cicogna 592 (only later as “Vischardo”); Biondo; M 559; Cicogna 2116; Cicogna 2413; M 2559; Gradenigo 53; ASV 59; M 2566 (only later as “Viscardo”); Correr 1337; F 6147; Cicogna 3753; Sabellico; ASV 61; M 324; M 2546; DDR 121; M 541 (only later as “Uischardo”); M 788; M 550; Antonio; M 52; P. Dolfin; PD 482c; Correr 760; Correr 760 bis; Cicogna 1899 (only later as “Guischardo”); WL 74–3; PD 380c (only later as “Uischardo”); Correr 873; Pigno; DDR 449; M 2544 (only later as “Viscardo”); F 3458**; Cicogna 589; Cavalli; M 51; Marco; Gussoni; PD 236c (only later as “Viscardo”); Querini 15; F 6234; M 46 (only later as “Uiscardo”); M 2571; M 2576; M 2573; Caroldo; M 723; Rosso; M 2543; Cicogna 2123; pseudo-Rotta; M 555; pseudo-Barbaro; Cicogna 1982; Cicogna 1982 bis; M 322; Correr 1327; pseudo-Abbiosi; M 45 (only later as “Viscondo”); M 2568 (only later as “Vischardo”); M 2568 bis (the formula is: “Uiscardo Re de Cecilia ditto Ruberto Viscardo”); M 2567; pseudo-Zancaruolo; M Z 21; Cicogna 1898; M 53; PD 312c; Cicogna 3712; pseudo-Erizzo; pseudo-Navagero; F 6211; M 60; Cicogna 2753; Magno; Grandis; T. Donato; M 728; M 728 bis; Cicogna 351; M 68; M 2580; Cicogna 3725; M 513; M 514; A. Valier; M 2581; Giustinian; M 1568; Sansovino; MG 327; PD 378c; Veniera 791; pseudo-Trevisan; Savina; M 793 (only later, under V. Falier); M 393; Doglioni; Cicogna 3556–3; M 1586 (only later as “Viscardo”); M 58; M 2550; Correr 1305 (only later as “Uiscardado”); Cicogna 590 (only later as “Uiscardo”); M 2046; M 59; M 67 (only later as “Guiscardo”); M 91 tert; Correr 1421; MG 106; F 6235; Cicogna 2831; Cicogna 2234; Lio; P. Morosini; M 2541 (only later as “Guiscardo”); M 2395; Contarena; M 2669; M 43; M 44; M 2572 (only later as “Guiscardo”); pseudo-Donà; M 54; M. Contarini; F 6166 (only later as “Guiscardo”); M 70; M 75; M 102; Vianoli; M 61; pseudo-Barba (only later); M 66 bis; Correr 1456; M 1999; M 2676; M 2028 (only later as “Viscardo”); Querini 36; Veniera-Flangini; M 64; F 6566; Curti; M 74; Diedo; Laugier; Cicogna 2302; M 1669; M 974; Cicogna 2832. 83 M Z 20; M 2566; M 541; Cicogna 1899; M 46; M 2568 bis (although, in this latter case, it seems that the nickname is Robert); Doglioni; Correr 1305; M 2046; M 974.



venetian reactions to the normans 171 form of “Viscardus”, the nickname of Guiscard takes different forms. 84 85 86 Thus, it becomes “Cischardo”, “Vicardo”, “Viscondo”, the curious 87 88 “Uiscardado”, “Beltrando”, culminating in the highly derogatory 89 “Bastardo”, found in no less than twenty-nine cases. Sometimes, not so often (sixteen chronicles), the sobriquet of Guiscard is placed before the 90 name Robert; 91

– Robert without Guiscard: 12.66 per cent. In seven texts the reader is under the impression that “Robert” and “Guiscard” are two distinct 92 individuals; – “Robert without Robert”, meaning an absence of first name. This principally occurs in those cases where the character is mentioned later, during the mission to Alexios to secure the title of Dalmatia and Croatia, under 93 the dogeship of Vitale Falier, but also in some cases when referring to the 94 Norman “condottiere” in the context of the war against Byzantium. These instances are found in seventeen chronicles. 84

Cicogna 2117. Cicogna 2413; DDR 449; MG 327. 86 M 45. 87 Correr 1305. 88 PD 312c; pseudo-Donà. 89 M 37; M 89; M 559; Gradenigo 53; ASV 59; Correr 873; F 3458**; Cicogna 589; M 723; Cicogna 2123; Correr 1327 (using the formula “Ruberto Viscardo Re de Sicilia qual fu bastardo”); Cicogna 3712; M 68; M 2580; Cicogna 3725; Veniera 791; M 793; M 2046; M 91 tert; Correr 1421; MG 106 (“Bastando” – sic); Cicogna 2831; Cicogna 2234 (“Ruberto bastardo del Rè dell’isola di Cicilia”); M 2395 (“Ruberto bastardo del Rè di Sicilia”); M 102; Correr 1456; Veniera-Flangini; Cicogna 2832. Three of these chronicles apply this formula later, when referring to Bohemund, who was also regarded as “Bastardo”, see Gradenigo 53, 10b, col. 2 (“Bonamonte fio bastardo del Re Rubertto”); Correr 873, 90a, col. 2 (“Bonamonte bastardo del Re Ruberto”); MG 106, 43a (“Bonamonte fio bastardo del Re Ruberto”). 90 M 104; Cicogna 2116; M 2559; ASV 61; Antonio; PD 482c; Rosso; pseudo-Rotta; M 555; pseudo-Abbiosi; M 513; PD 378c; M 2550; M 2669; M. Contarini; M 66 bis. 91 M 2564; M 2569; M 2563; M Z 20; M 796; ASV 58; Sanudo; ASV 60; Correr 1046; Querini 13; Correr 710; Zancaruolo; P. Cornaro; M 125–3 (only later as “Ruberto”); Cicogna 3599; M 47; Correr 1045; M 39; Agostini-Tiepolo; Querini 16; Cicogna 2815 (only later as “Roberto”); Cicogna 3675; PD 391c; M 327; M 80; M 2592; Sivos; M 2676 (only later as “Ruberto”); M 1577; M 1833. 92 Vitturi (“a Uischardo et a Ruberto Re dj Zezilia”); Cicogna 2831 (“Re Ruberto Bastardo, et Re del isola de Sezilia”); M 2572 (“i Guiscardi del sangue Normano Prencipi di Puglia”); M 70 (“Roberto, & Guiscardo signori de Normani”); M 75 (“Roberto, et Guiscardo signori de Normani”); M 102 (“Rè Ruberto Bastardo, et Rè dell’jsola de Sezilia”); Cicogna 2832 (“Re Ruberto bastardo, et Re de l’isolo [sic=a] de Sezilia”). 93 M 2565 (“lo Re de Cecilia”); M 2037 (“Re di Cicilia”); M 628a (“lo Re de Cecilia”); Alberegno (“lo Re de Cecilia”); pseudo-Alberega (“lo Re di Cicilia”); Fabris (“il Re de Cicilia”); M 80 (“il Re di Cicilia”). For this event, see above. 94 Biondo (“Guiscardus”, without Robert); M 798 (“lo Re de Cecilia”); M 67 (“Viscardo”, without Robert); pseudo-Alberega (“lo Re di Cicilia”); Fabris (“il Re 85

172

şerban v. marin

The name of Robert itself is sometimes changed: “Uberto” in Cigotto and ASV 60; “Caruberto” in M 104 and pseudo-Abbiosi; “Chruderto” in Antonio. However, these name changes are not as frequent as “Guiscard”. Ten other chronicles speak exclusively about “Normanni” in the context of these events, without referring to any particular individual.95 As a matter of fact, Robert’s Norman ethnicity is quite rarely emphasised, although there are cases when he is indubitably presented as “de nation Normando (of Norman nation)”,96 while M 2572 refers to “i Guiscardi del sangue Normano (the Guiscards of Norman blood)”, and M 1999 specifies: “Ruberto Guiscardo della discendenza de Normandi gente di Francia (Robert Guiscard of descent from the Norman people from France)”. In contrast, one should note the case of six chronicles97 where, although the episode of the Norman–Byzantine conflict under Domenico Selvo is absent, the character of Robert Guiscard is inserted prior to this (meaning under Doge Domenico Contarini) and is presented as follows (according to M 31): “in questo tempo un certo zentilomo Italian il quale haueua nome Ruberto di Gismondo (at this time, a certain Italian gentleman who was named Robert of Gismondo)”, so that he is regarded as a native Italian. It is interesting to note in how many chronicles the Normans are presented with their ethnic name. The number is forty-eight.98 In comparison with this figure, however, the designation of Robert as “king of Sicily” is much more conspicuous in the Venetian chronicles, as noted above. As a whole, the Venetian chronicles that refer to the Normans in southern Italy naturally emphasise the position of Serenissima in the Norman–Byzantine conflict. While the defeat in 1085 is merely concealed in the texts, the preliminary victory over Robert Guiscard in the Ionian sea is usually underlined, thus promoting Venice at the core of the events as the political entity that, by interfering in affairs, imposed peace between the Normans and the Byzantines. de Cicilia”); M 2541 (“Viscardo”, without Robert); M 2572 (“Guiscardo”, without Robert); M 80 (“il Re di Cicilia”); F 6166 (“Viscardo”, without Roberto); pseudoBarba (“Viscardo”, without Roberto). 95 Marcello; Cicogna 3556–7; M 67; Correr 1307; Querini 63; M 2541; F 6166; pseudoBarba; Cicogna 3276-42, and M 1833. The same occurred in the first instance for M 796, but later the king’s name was used. 96 Pseudo-Barbaro; Savina; F 6566; M 74; M 974. 97 MG 249, 43a; Cicogna 1983, [12a]; Cicogna 1900, [14a]; M 31, 91b; F 6821, 60b; Correr 1032, 48b. 98 Extensa; Venetiarum Historia; Monaci; Biondo; M 796; Sabellico; P. Dolfin; PD 380c; Marcello; Caroldo; pseudo-Barbaro; pseudo-Zancaruolo; F 6211; M 60; Cicogna 2754; Magno; Giustinian; Sansovino; Cicogna 2815; PD 391c; Savina; Doglioni; Cicogna 3556–3; Cicogna 3556–7; M 58; M 59; M 67; Correr 1307; F 6235; Querini 63; P. Morosini; M 2541; M 2572; F 6166; M 70; M 75; Vianoli; M 2592; M 61; pseudo-Barba; Sivos; M 1999; F 6566; M 74; Diedo; Cicogna 3276-42; M 1833; Laugier.



venetian reactions to the normans 173

Equally, one could conclude that the representation of the Normans and Byzantines changes. Although enemies, the Normans are gradually regarded with sympathy. In contrast, the tone in relation to the Byzantines becomes more and more deprecatory, emphasising their cowardice, or at least their passivity, in front of the Normans. This serves to imply Venice’s position as the saviour of the Eastern Empire, and also to prepare the ground for the completely negative representation of Byzantium in the following century, when the enmity of Manuel I Comnenus towards the Venetians would materialise in open conflict between Venice and Byzantium and the disaster of the Venetian fleet in the Aegean archipelago in 1172.

Part III

Policies of Conquest, Consolidation, and Expansion

7

The Norman Kingdom of Sicily: Projecting Power by Sea Charles D. Stanton

In the twelfth century, the Norman Kingdom of Sicily rose to rival the great maritime republics of the Italian peninsula for dominance of the Mediterranean. But it did so in a manner quite dissimilar to that of Genoa, Pisa, or Venice. The Norman realm’s unique approach to the development and employment of naval power has much to do with how the kingdom was created and the mindset of its creators. Those who founded it were, at their core, mounted men-at-arms and conquerors. The Normans who migrated into the Mezzogiorno (southern Italy and Sicily) in the eleventh century were indigent knights – soldiers of fortune – offering their swords for rent in return for plunder.1 When their activities inevitably evolved from mercenary enterprise and organised brigandage to overt conquest, the Normans adopted sea power as a pragmatic means for effecting the seizure of Sicily – the island that would ultimately become the core of their kingdom. Their first ships were commandeered vessels of commerce that they used to literally ferry their armies across the Strait of Messina.2 Once the conquest was completed, they deployed their burgeoning fleets almost entirely to defend their acquisition and aggrandise its hegemony to include the whole of the central Mediterranean. 1

R. Allen Brown, The Normans (Woodbridge, 1984), p. 79; Marjorie Chibnall, The Normans (Oxford, 2000), p. 75; John France, “The Occasion of the Coming of the Normans to Italy”, Journal of Medieval History 17 (1991), 185–205; Einar Joranson, “The Inception of the Career of the Normans in Italy – Legend and History”, Speculum 23 (1948), 353–96. 2 Charles D. Stanton, Norman Naval Operations in the Mediterranean (Woodbridge, 2011), pp. 27–46; Donald P. Waley, “‘Combined Operations’ in Sicily, A.D. 1060–1078”, Papers for the British School at Rome 22 (1954), 118–25, esp. 120–21.

178

charles d. stanton

In comparison, the north Italian maritime republics had adopted an altogether different modus operandi for employing their naval capability. And they did so to achieve vastly contrasting objectives, also rooted in their origins. They were born as mercantile communities wedded to the sea. While their citizens may have begun as freebooters and coastal traders, these innate seafarers quickly embraced maritime commerce on a grand scale as their principal occupation. And the nature of that commerce was overwhelmingly the transportation of desired goods from their sources to markets where they could be sold at a significant profit or exchanged for products in demand elsewhere in the Mediterranean world. As a consequence, the primary purpose of their fleets was to expand their markets and extend their trade routes. Routes to the Orient became particularly lucrative, especially after the onset of the crusading movement in the late eleventh century.3 Accordingly, the Italian mercantile communes of Amalfi, Genoa, Pisa, and Venice established a set of west–east trade routes presided over by commercial colonies in the East called fondaci. These trade routes and fondaci provided the lifeblood of their survival. It was to protect them that the Italian maritime powers developed their fleets of war.4 The emergence of Venice as a pre-eminent power in the medieval Mediterranean provides a case in point. Following the Latin conquest of Constantinople spearheaded by La Serenissima during the Fourth Crusade in 1204, the noted maritime historian John E. Dotson has observed, “Venice established a chain of naval bases that formed the backbone of their naval power from that time forward.”5 In contrast, the Norman Kingdom of Sicily did not seek commerce in distant lands. Due to its geographic advantage at the centre of the Middle Sea, it let commerce come to it. It levied tolls on traders who transited its waters and sought shelter on its shores. Sicily’s status as a breadbasket of the Mediterranean since classical times made it a “Mecca” for merchants seeking to capitalise on its bountiful wheat harvests. And the Norman kingdom parlayed that same wheat surplus into an immensely profitable gold-forgrain traffic with Muslim North Africa.6 It was chiefly to defend its remunerative dominance of the central Mediterranean that the Norman Kingdom

3

Archibald R. Lewis and Timothy Runyan, European Naval and Maritime History, 300–1500 (Bloomington, 1985), pp. 63–66. 4 David Abulafia, The Two Italies: Economic Relations between the Norman Kingdom of Sicily and the Northern Communes (Cambridge, 1977), p. 53; Stanton, Norman Naval Operations, pp. 189–95. 5 John E. Dotson, “Foundations of Venetian Naval Strategy from Pietro II Orseolo to the Battle of Zonchio, 1000–1500”, Viator 32 (2001), 113–15, esp. 120. 6 David Abulafia, “The Norman Kingdom of Africa and the Norman Expeditions to Majorca and the Muslim Mediterranean”, Anglo-Norman Studies 7 (1985), 26–39, esp. 35–36; Two Italies, pp. 48, 86.



the norman kingdom of sicily 179

of Sicily initially fashioned a strong naval capability. It later used its formidable fleets to project its power throughout the Great Sea for plunder and profit as well as to discourage invasion by a host of covetous adversaries which included the Holy Roman Empire under the German kings, the Byzantine Empire and the Italian maritime republics themselves. Within that context, this discussion will briefly track the idiosyncratic development of naval power under the Norman kings from consolidation to expansion to defence through deterrence to adventurism and, finally, atrophy. Consolidation The Norman mostly responsible for the post-conquest consolidation of the realm was Roger II de Hauteville, its founder and first king. His uncle was the great Robert Guiscard who had led the subjugation of southern Italy, while his father was none other than Roger I de Hauteville, the “Great Count” who had completed the seizure of Sicily after a dogged thirty-year campaign.7 Even before winning his crown, Roger II was indoctrinated in the strategic advantages of Sicily’s fortuitous geographical situation by his chief counsellor, a Greek Melkite named George of Antioch who had served as finance minister to the Zirid princes of Mahdiyah in Ifriqiya (the old Roman Province of Africa).8 George doubtless apprised his young protégé that the key to increased wealth and power was dominance of the central Mediterranean, a feat that could be accomplished only by seizing control of the shores both to the north and south of the island.9 Taking advantage of George’s expertise in the Maghreb (coastal Africa west of Egypt), Roger began with a few furtive attempts to increase Norman influence in North Africa. In 1118 he dispatched twenty-four warships to aid the governor of Gabes in a dispute with the Zirid ruler of Mahdiyah in the hope of winning a foothold on the continent, but the mission resulted in a humiliating retreat.10 In 1123, he sent a three hundred-ship armada led in person by Christodoulos (his first minister at the time) and George of Antioch to conquer Mahdiyah itself, but the two bureaucrats lacked military experience and failed to take basic tactical precautions. The expedition 7

Hubert Houben, Roger II of Sicily: A Ruler between East and West, trans. G. A. Loud and Diana Milburn (Cambridge, 2002), pp. 14–24; Graham A. Loud, The Age of Robert Guiscard: Southern Italy and the Norman Conquest (Harlow, 2000), pp. 146–72. 8 Al-Maqrizi, Kitab al-tarikh al-kabir al-muqaffa li-Misr, trans. Jeremy Johns in Arabic Administration in Norman Sicily (Cambridge, 2002), pp. 80–82. 9 Stanton, Norman Naval Operations, pp. 70, 114. 10 Al-Nuwayri, Biblioteca Arabo-Sicula, trans. Michele Amari (2 vols, Turin, 1880–81), II, p. 154; Ibn al-Athir, The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Ta’rikh, trans. D. S. Richards (3 pts, Aldershot, 2006–8), I, pp. 185–86.

180

charles d. stanton

was a disastrous debacle from the start.11 Roger persisted though, retaking Malta (previously conquered by his father) in 1127.12 He probably would have pursued his African aspirations then, had not a propitious opportunity to capture control of the coastline to the north intervened that same year. Duke William of Apulia died that summer without a direct descendant, leaving his duchy to his cousin Roger as his designated heir. Roger promptly proceeded north with a small squadron of ships to take possession of his inheritance.13 His claim was immediately contested by an eclectic coalition of adversaries: dissident Norman nobles (like Robert II of Capua and Rainulf of Caiazzo) bent on establishing their own independent fiefdoms and Pope Honorius II who contended that only the papacy had the right to invest the duke of Apulia, based on Robert Guiscard’s submission to the Holy See in 1059 and 1080.14 But Roger cornered Honorius at Benevento in 1128 and forced the pope to recognise him as duke.15 Following the passing of Honorius in 1130, Roger then inveigled papal pretender Anacletus II into anointing him king of Sicily in return for his military support.16 Innocent II, the competing claimant to the throne of St Peter, however, rejected that investiture with the backing of the Holy Roman Emperor, Lothair III of Germany, who considered the Mezzogiorno part of his own “God-given” empire.17 The rebel Norman nobles joined the union.18 Roger, nevertheless, intuitively understood that the unwieldy alliance was unlikely to hold for long, so he dealt with it by employing a mixture of patience, Machiavellian statecraft, and the prudent use of naval power. It is the deft deployment of his fleet that is germane to the topic at hand. Roger, through his adept 11

Ibn al-Athir, Chronicle of Ibn al-Athir, I, pp. 245–46; At-Tijani, Biblioteca AraboSicula, II, pp. 68–74. 12 Alexander of Telese, Alexandri Telesini abbatis Ystoria Rogerii regis Sicilie, Calabrie atque Apulie, ed. Ludovica de Nava (Rome, 1991), bk. I, ch. 4, p. 8. See also Ferdinand Chalandon, Histoire de la Domination Normande en Italie et en Sicile (2 vols., Paris, 1907), I, p. 377; Houben, Roger II of Sicily, p. 41. 13 Alexander of Telese, Ystoria Rogerii regis Sicilie, bk. I, chs. 4–5, p. 8. See also Chalandon, Histoire de la Domination Normande, I, pp. 385–87. 14 Alexander of Telese, Ystoria Rogerii regis Sicilie, bk. I, chs. 8–10, pp. 10–12. See also Houben, Roger II of Sicily, pp. 44–45. 15 Alexander of Telese, Ystoria Rogerii regis Sicilie, bk. I, ch. 15, pp. 14–15. See also Houben, Roger II of Sicily, pp. 46–47. 16 Chronica Ignoti Monachi Cisterciensis S. Maria de Ferraria, ed. Augustus Gaudenzi (Naples, 1888), anno 1130, p. 18; Falco of Benevento, Chronicon Beneventanum, ed. Edoardo D’Angelo (Florence, 1998), p. 106; Regesta pontificum Romanorum ab condita ecclesia ad annum post Christum natum MCXCVIII, ed. P. Jaffé, S. Loewenfeld, F. Kaltenbrunner, and P. Ewald (Leipzig, 1885–88), no. 8411. See also Houben, Roger II of Sicily, pp. 51–52. 17 Ian S. Robinson, The Papacy 1073–1198, Continuity and Innovation (Cambridge, 1990), pp. 442–46. 18 Houben, Roger II of Sicily, pp. 61–63.



the norman kingdom of sicily 181

“amiratus amiratorum” (“admiral of admirals” – George of Antioch), tasked the realm’s ships and seamen with executing three basic military operations: ferrying Sicilian troops across the Strait of Messina, blockading enemy ports, and securing “fortress Sicily” from seaborne invasion. From the moment Roger stood aboard his flotilla of seven galleys before the port of Salerno in the summer of 1127 to demand his rightful place as ruler of Norman Italy, to July 1139 when he finally won recognition as king of Sicily from Pope Innocent II at Mignano, Norman fleets transported Sicilian armies with thousands of Saracen soldiers to the mainland Mezzogiorno on numerous occasions. For instance, when Pope Honorius II placed Roger under anathema and cobbled together a coalition of barons to oppose him, the young count had his ships convey from Sicily a land force of two thousand knights, three thousand infantrymen and fifteen hundred archers in the spring of 1128 to enforce his claim.19 A formal papal investiture as duke of Apulia resulted on 22 August near Benevento. Roger returned the following spring with three thousand horse and six thousand foot on forty to sixty ships to mop up the remaining resistance in Apulia.20 Most notably, in the summer of 1134 he landed an army of some seven thousand on sixty vessels at Salerno to quell yet another revolt, ultimately leading it to a resounding victory over the rebels’ Pisan allies at Fratta near Ravello.21 And, of course, once these fleets had delivered their cargo of combatants, they did not simply anchor idly offshore. In Apulia, Roger used them to help reduce Brindisi in 1128 and Bari in 1129.22 In the spring of 1131 it was Campania’s turn. George of Antioch blockaded Amalfi and occupied Capri along with several other gulf islands, while Roger reduced Ravello himself.23 Amalfi’s eventual capitulation compelled Naples to do the same.24 Such operations were repeated with regularity throughout the ebb and flow of Roger’s campaign to consolidate the mainland portions of his kingdom, ending with another successful blockade of Bari in September 1139.25 To be sure, King Roger unavoidably suffered reverses along the way. The alliance of adversaries whom he faced often outnumbered his forces appreciably 19

Romuald of Salerno, Romualdi Salernitani Chronicon, ed. Carlo Garufi (Rerum Italicarum Scriptores, 2nd edn, Città di Castello, 1935), p. 216. 20 Romuald of Salerno, Romualdi Salernitani Chronicon, p. 218. 21 Bernardo Maragone, Gli annales Pisani di Bernardo Maragone, ed. Michele Lupo Gentile (Rerum Italicarum Scriptores, 2nd edn, Bologna, 1936), p. 10. 22 Alexander of Telese, Ystoria Rogerii regis Sicilie, bk. I, ch. 12, pp. 12–13; Romuald of Salerno, Romualdi Salernitani Chronicon, p. 216. 23 Alexander of Telese, Ystoria Rogerii regis Sicilie, bk. II, chs. 7–11, pp. 26–28; Chronica de Ferraria, p. 18; Falco of Benevento, Chronicon Beneventanum, p. 108; Romuald of Salerno, Romualdi Salernitani Chronicon, pp. 218–19. 24 Falco of Benevento, Chronicon Beneventanum, p. 122. 25 Falco of Benevento, Chronicon Beneventanum, pp. 226–30; Romuald of Salerno, Romualdi Salernitani Chronicon, p. 226.

182

charles d. stanton

– especially the land contingents. He suffered a very serious setback at Nocera in 1132, for instance, at the hands of his arch-enemy Rainulf of Caiazzo.26 The same rebel Norman lord humiliated him once again at Rignano in October 1137.27 And he never even attempted to confront Lothair III when he invaded the realm with a huge army that same year. There simply was no advantage to risking a pitched land battle against a vastly superior force. Thus, Roger did what he always did in such circumstances: he retreated to “fortress Sicily”, where his fleet could protect him from invasion while he awaited the inevitable collapse of the fractious coalition that opposed him.28 Expansion Finally having solidified control of southern Italy, Roger II was free to extend his influence onto the Maghreb coast and realise his strategy of establishing Norman sway over the central Mediterranean. He did so by deploying his fleets in the same manner that he had used to quell opposition on the mainland Mezzogiorno. The weapon that made the Maghreb vulnerable to Norman naval power, however, was wheat. North Africa had suffered a series of persistent famines since the mid-eleventh century when the Zirid emir of Qayrawan, al-Mu‘izz ibn Badis, declared independence from his Fatimid overlords, the caliphs of Cairo. The latter responded by unleashing Bedouin hordes from Upper Egypt, who shut the Zirids up in Mahdiyah and inflicted widespread devastation upon the region.29 No longer able to produce the foodstuffs needed to feed their people, the Zirids sought Sicily’s abundant supply of wheat and, thus, became supplicants of its Norman king.30 Roger, advised by his shrewd first minister (George of Antioch), knew to convert this tributary relationship into Norman dominance using supremacy over the surrounding seas. The Normans began in 1135 by seizing the island of Jerba opposite Gabes in the gulf of the same name, despite the fact that it was nominally in Mahdiyah’s sphere of influence.31 In 1142 George of Antioch ratcheted up the

26

Alexander of Telese, Ystoria Rogerii regis Sicilie, bk. II, chs. 29–35, pp. 36–40; Chronica de Ferraria, p. 19; Falco of Benevento, Chronicon Beneventanum, pp. 134–42; Romuald of Salerno, Romualdi Salernitani Chronicon, p. 220. 27 Falco of Benevento, Chronicon Beneventanum, pp. 196–98; Romuald of Salerno, Romualdi Salernitani Chronicon, p. 225. 28 Romuald of Salerno, Romualdi Salernitani Chronicon, p. 222. 29 Abulafia, “Norman Kingdom of Africa”, 27–28; Jamil Abun-Nasr, A History of the Maghrib during the Islamic Period (Cambridge, 1987), pp. 68–71. 30 Abulafia, “Norman Kingdom of Africa”, p. 33. 31 Ibn al-Athir, Chronicle of Ibn al-Athir, I, pp. 321–22; At-Tijani, Biblioteca AraboSicula, II, p. 55.



the norman kingdom of sicily 183

pressure by conducting a raid on the port of Mahdiyah itself with a twentyfive-ship flotilla, ostensibly in retribution for delinquent grain remittances. The submissive response of the Zirid emir, Abu’l-Hasan al-Hasan ibn Ali, made his subservience to Roger clear.32 Thus emboldened, Roger dispatched an expedition in 1143 to take Tripoli, another port of Zirid-ruled Ifriqiyah. The undertaking failed when the surrounding Bedouin tribes rallied to the city’s defence, but Norman fleets took out their frustration by raiding the Maghreb coast all the way to what is now western Algeria that same year and the year after.33 And in 1145 the Normans conquered the island of Kerkenna just off the Tunisian coast facing Sfax. It too had previously fallen within Mahdiyah’s purview.34 The following year they returned to Tripoli, this time with an armada of one hundred to two hundred ships under the personal direction of George of Antioch. Abetted by internal strife, they quickly gained the city’s submission.35 The fall of Tripoli engendered a ripple effect that caused other ports between Tripoli and Tunis to spontaneously submit. The ruler of Gabes (a freedman called Yusuf ), for instance, actually wrote to Roger unsolicited, offering to become his amil or vassal. This triggered a punitive response from al-Hasan, which then prompted Yusuf ’s subsequent demise and provided Roger a convenient pretext to crown his African strategy.36 In 1148 he dispatched George with an armada of between 250 and 300 vessels to subjugate Mahdiyah itself. Famine-induced food shortages made withstanding a prolonged siege impossible, so al-Hasan abandoned the city without a struggle. George soon added both Susa and Sfax to Norman control.37 Having achieved dominion over the Maghreb coast from Tripoli to Cape Bon and beyond, the Normans had at last won supremacy over the Sicilian Channel – a vital east–west trade route. Given the state of nautical technology at the time – i.e., ships (especially galleys) being heavily dependent upon the surrounding shores for navigation, food, water, and shelter – traffic could now flow through the centre of the “Great Sea” untrammelled by the authority of the king of Sicily only with great difficulty.38 32

Ibn al-Athir, Chronicle of Ibn al-Athir, I, p. 365; At-Tijani, Biblioteca Arabo-Sicula, II, pp. 75–76. 33 Chronica de Ferraria, anno 1143, p. 27; Ibn al-Athir, Chronicle of Ibn al-Athir, I, pp. 366–67. 34 Ibn al-Athir, Chronicle of Ibn al-Athir, I, p. 378. 35 Chronica de Ferraria, anno 1146, p. 28; Ibn al-Athir, Chronicle of Ibn al-Athir, I, p. 380; At-Tijani, Biblioteca Arabo-Sicula, II, p. 60. 36 Ibn al-Athir, Chronicle of Ibn al-Athir, II, pp. 13–14. 37 Ibn al-Athir, Chronicle of Ibn al-Athir, II, pp. 18–20; At-Tijani, Biblioteca AraboSicula, II, pp. 76–78. 38 Abulafia, “Norman Kingdom of Africa”, 35–36; Stanton, Norman Naval Operations, pp. 108–10.

184

charles d. stanton

At the same time, the Normans under Roger II and George of Antioch were magnanimous masters who granted their conquered African territories abundant autonomy, while turning such Muslim port cities as Tripoli, Mahdiyah, and eventually Bona into cosmopolitan trading entrepôts. It was in these markets that sub-Saharan gold brought in by caravan was traded for grains and such finished European products as textiles, engendering a flourishing north–south commerce. All of this, of course, caused the Norman Kingdom of Sicily to become the wealthy envy of the Mediterranean world.39 Defence and Deterrence The richer Roger’s realm became, the more the Byzantine Emperor and Holy Roman Emperor felt that they were entitled to it. Accordingly, the primary role of the Norman fleet became defence. This is not, however, to imply that the fleet was used to patrol Norman waters with the intention of preventing amphibious assaults on the kingdom’s soil. Limitations in nautical technology – poor line of sight visibility (six to ten miles) and slow closing speeds (less than ten knots) – made sighting and interdicting enemy fleets at sea problematic in the extreme.40 A more efficacious tactic was to set a coastal watch so that royal forces could react quickly to an enemy as he came ashore, which is exactly what the twelfth-century Chronica Ignoti Monachi Cisterciensis Santa Mariae de Ferraria claimed Roger did, complete with fortifications dotting the shorelines at regular intervals (about every three miles).41 King Roger, however, preferred to prevent the fleets of his foes from even approaching his realm, so he resorted to two basic stratagems. The first was diplomacy, which he successfully deployed on several occasions to detach the era’s maritime powers – i.e., Genoa, Venice, and Pisa – from any imperial alliance. In 1136, for instance, he sabotaged an incipient antiNorman league that Venice was attempting to engineer with both Byzantine Emperor John II Komnenos and Holy Roman Emperor Lothair III by peeling away the Venetians with generous trade privileges.42 The following year he crippled Lothair’s invasion of the Regno (the traditional appellation 39

Ibn abi Dinar, Biblioteca Arabo-Sicula, II, p. 296. See also Abulafia, “Norman Kingdom of Africa”, p. 36. 40 John Pryor and Elizabeth Jeffreys, The Age of ΔPOMΩN (Dromōn), The Byzantine Navy ca 500–1204 (Leiden, 2006), pp. 388–89; Stanton, Norman Naval Operations, p. 268. 41 Chronica de Ferraria, anno 1151, pp. 26–27. See also Houben, Roger II of Sicily, p. 156. 42 Gottlieb Lukas Friedrich Tafel and Georg Martin Thomas (eds), Urkunden zur älteren Handels- und Staatsgeschichte der Republik Venedig (2 vols., Vienna, 1856), I, docs LXV–LXVI, pp. 172–75. See also Abulafia, Two Italies, pp. 88, 143; Camillo Manfroni, Storia della Marina Italiana (3 vols, Livorno 1897–1902), I, pp. 189–90.



the norman kingdom of sicily 185

for the Kingdom of Sicily) by suborning the emperor’s seaborne allies, the Pisans, with promises of cash and commercial concessions. Without the Tuscan republic’s sea power, the imperial campaign soon collapsed.43 The second and perhaps most effective method of defence was deterrence. That’s where the daunting Norman fleet came in. The Norman kings used their fleets to project power and discourage attack. Their primary warship was the galea, a distinctly Italian refinement of the Byzantine dromōn, which had dominated naval warfare in the Mediterranean since the fifth century.44 Byzantine emperor Leo VI (886–912) succinctly described the standard dromōn in Constitution XIX of his Taktika, a treatise on military tactics and strategy: Let each dromōn be of good length and proper size with two oar-banks, as they are called, one below and one above. Let each row have at least twentyfive benches for rowers to sit on. All told, therefore, there should be twentyfive benches below and, likewise, twenty-five above, making a total of fifty. On each one of them let two oarsmen be seated, one on the right and one on the left, so that all the oarsmen, who also serve as soldiers, those above 45 and below, come to one hundred men.

By the late tenth century the dromōn had evolved into a fully decked bireme (two files of oarsmen, one above the other, on each side), measuring roughly 31.25m (102ft 6in) in length and 4.4m (14ft 5in) of beam.46 Its primary means of propulsion was a standard ousia (crew complement) of 108 rowers, augmented as necessary by lateen sails on a foremast and a main mast, while twin steering oars mounted on either side of the stern provided directional control. Weaponry included a chain-mounted spur on the bow along with siphons for spewing Greek fire, installed on a parapeted forecastle and possibly on fortified platforms amidships called kastelloma.47 By comparison, the galea adopted by the Norman kings in the twelfth century must have been considerably longer (about 39.5m or 129ft 8in) and slightly wider amidships (around 4.6m or 15ft 2in).48 This is because this craft employed an innovative oarage system called alla sensile, whereby the rowers all sat two per thwart (bench) on the main deck instead of on two 43

Chronica de Ferraria, anno 1137, pp. 21–22; Falco of Benevento, Chronicon Beneventanum, p. 188; Bernardo Maragone, Annales Pisani, p. 11; Romuald of Salerno, Romualdi Salernitani Chronicon, p. 227. 44 Pryor and Jeffreys, Dromōn, pp. 123–24. 45 Leo VI (Byzantine Emperor), The Taktika of Leo VI: Text, Translation, and Commentary, trans. George T. Dennis (Washington DC, 2010), Constitution 19, para. 8, pp. 504–7. 46 Pryor and Jeffreys, Dromōn, p. 205. 47 Charles D. Stanton, Medieval Maritime Warfare (Barnsley, 2015), pp. 14–15; Norman Naval Operations, pp. 228–29. 48 John Pryor, “A View from a Masthead: The First Crusade from the Sea”, Crusades 7 (2008), 87–151, esp. 130–31.

186

charles d. stanton

levels, one above the other. Such a configuration would have allowed each man the overhead room to manipulate his individual oar from a stand-sitstroke position, theoretically permitting greater power, which would have translated into greater speed. The increased interscalmium (stroke distance) this would have required (at least 1.2m as opposed to less than a metre) explains the galley’s added length. Furthermore, the lack of a lower deck would have provided increased cargo capacity below deck, which would have meant that these galleys could carry more water and provisions, engendering greater endurance.49 In other words, the galea possessed the speed and range to conduct the sort of long-distance, quick-strike missions needed to keep the kingdom’s enemies off balance. Its armament was probably an assortment of projectile-launchers like the bow-ballista, rather than Greek fire siphons.50 The fleet, of course, also included an array of support vessels. Merchantmen, often called simply naves, were conscripted from conquered ports in accordance with a quota system to carry provisions.51 Sagittae (“arrows”), small swift galleys with as few as sixteen oars, were used to scout enemy ports and run messages.52 Refitted dromōns called chelandia were employed as horse transports.53 There were even huge oared fighting ships called catti or gatti, which were believed to be high-profile triremes with castles, capable of carrying dozens of infantrymen.54 Many of these ships were acquired through feudal obligation from subject port cities, but the galeae, the mainstay of the fleet, were most likely purpose built in the royal arsenals of Palermo and Messina. And they were crewed through the datium marinariorum, a levy of mariners from the realm’s lordships, monasteries, and townships.55 All of these component parts were assembled under an amiratus to provide what David Abulafia termed “a naval cordon sanitaire” around the island kingdom and project Norman sea power to 49

Pryor and Jeffreys, Dromōn, pp. 423–44. Stanton, Norman Naval Operations, pp. 230–36; “The Naval Strategy of Roger II, King of Sicily”, in New Interpretations in Naval History: Selected Papers from the Fifteenth Naval History Symposium, ed. Maochun Yu (Annapolis, 2009), pp. 14–33, esp. 26–27. 51 Pryor, “The Mediterranean Round Ship”, in Cogs, Caravels and Galleons: The Sailing Ship, 1000–1650, ed. Robert Gardiner (London, 1994), pp. 59–76; Stanton, Norman Naval Operations, pp. 238–41. 52 Willy Cohn, Die Geschichte der normannisch-sicilischen Flotte unter der Regierung Rogers I. und Rogers II. (1060–1154) (Breslau, 1910), pp. 95–96; Manfroni, Storia della Marina italiana, I, p. 457; John Pryor, “From Dromōn to Galea: Mediterranean bireme galleys AD 500–1300”, in The Age of the Galley, Mediterranean Oared Vessels since PreClassical Times, ed. Robert Gardiner and John Morrison (London, 1995), pp. 101–16, esp. 108; Stanton, Norman Naval Operations, pp. 241–42. 53 Pryor and Jeffreys, Dromōn, pp. 322–25. 54 Stanton, Norman Naval Operations, pp. 237–38. 55 Stanton, “Naval Strategy of Roger II”, p. 26. 50



the norman kingdom of sicily 187

discourage attack, particularly from the remnants of the Western and Eastern Roman Empires.56 Keeping Byzantium at Bay Aside from Lothair’s invasion of the Regno in 1137, the most aggressive imperial threat to the sovereignty of the Kingdom of Sicily in the first decades of its existence came from Constantinople. The Byzantine emperor considered the Normans to have wrested the historically Greek-controlled themes of Calabria, Apulia (Longobardia), and Sicily from the empire by force of arms, and he was determined to retrieve those provinces in the same manner. Perhaps the best example occurred during the Second Crusade. In 1146 Lothair’s successor, King Conrad III of Germany, arranged the marriage of his adopted daughter, Bertha of Sulzbach, to Manuel I Komnenos to cement their alliance against the Norman Kingdom of Sicily.57 The advent of the Second Crusade led by Louis VII of France provided Roger with a propitious reprieve from imminent attack.58 So, when Manuel recalled troops from his garrisons in mainland Greece to bolster Constantinople’s defences in anticipation of the depredations of Louis’ crusader army, Roger pounced.59 He sent his fleet on an extended raid of the Peloponnesus all the way to the Isthmus of Corinth and back. In the process, it captured Corfu and ravaged such cities as Methone, Neapolis, Monemvasia, Nafplion, Negroponte, Athens, Thebes, and Corinth itself.60 When Manuel attempted to liberate the fortress of Kerkyra on Corfu the next year, Roger dispatched George of Antioch with another fleet of sixty ships on a raid that reached all the way into the Bosporus, where, in the summer of 1149, it fired flaming arrows at the Byzantine imperial palace itself and set the

56

David Abulafia, The Great Sea, A Human History of the Mediterranean (London, 2011), pp. 318–22. 57 John Kinnamos, Deeds of John and Manuel Comnenus, trans. Charles Brand (New York, 1976), bk. II, ch. 4, pp. 36–38; Otto of Freising, Gesta Frederici: Ottonis et Rahewini Gesta Friderici I. Imperatoris, ed. Georg Waitz (Monumenta Germaniae Historica, Scriptores Rerum Germanicarum, Hanover-Leipzig, 1912), bk. I, ch. XXIV, pp. 37–43. 58 Ibn al-Athir, Chronicle of Ibn al-Athir, I, pp. 372–3; Odo of Deuil, De Profectione Ludovici VII in Orientem, trans. Virginia Berry (New York, 1948), bk. I, pp. 6–7. 59 Niketas Choniates, O City of Byzantium! Annals of Niketas Choniates, trans. Harry Magoulias (Detroit, 1984), p. 36; John Kinnamos, Deeds of John and Manuel Comnenus, bk. II, chs. 12–3, pp. 58–61. See also Chalandon, Histoire de la Domination Normande, II, p. 135. 60 Chronica de Ferraria, anno 1147, p. 28; Niketas Choniates, O City of Byzantium!, pp. 43–45; Romuald of Salerno, Romualdi Salernitani Chronicon, p. 227. See also Houben, Roger II of Sicily, p. 84.

188

charles d. stanton

surrounding communities ablaze.61 The Regno never suffered an assault by imperial forces of any sort during Roger’s reign. During the rule of Roger’s son and heir, William I, the Byzantines (in league with some Norman rebels) did, however, make serious inroads into royal lands on the east coast of the Italian peninsula, which culminated in a strangling siege of Brindisi in 1156.62 But the Norman king responded with a devastating counter-attack. He used his ample fleet to blockade the Byzantine flotilla in the harbour while his ground forces crushed the besieging Greek army with a constricting envelopment. Nearly none escaped, including the Greek commanders.63 In retaliation against the Emperor Manuel, who had sponsored the attack, William directed another punitive raid on Constantinople the following year to dissuade any future imperial aggression against his realm. Numbering a staggering 140 galeae along with twenty-four transport dromōns, according to Pisan chronicler Bernardo Maragone, the Norman fleet defeated a Greek squadron and sacked Negroponte before once again penetrating the Dardanelles.64 The Byzantine chronicler Niketas Choniates reported that it went on to parade before the walls of Constantinople, firing silver-shafted arrows at the Blachernae palace situated at the north-western end of the Golden Horn.65 The fact that Manuel signed a thirty-year peace pact with William the next year testifies to the raid’s efficacy.66 Adventurism While King William I ably defended the Regno as represented by the Mezzogiorno, he was less diligent in maintaining the kingdom’s possessions on the Maghreb coast. Through a combination of neglect and poor governance, the encroaching Almohads of Marrakech were able to strip all Norman North Africa from the realm by early 1160, when the garrison of Mahdiyah finally capitulated after a gruelling four-month land–sea siege.67 This loss of Norman dominance over the south shore of the Sicilian Channel would 61

Andrea Dandolo, Chronica Per Expensum Descripta, ed. Ester Pastorello (Rerum Italicarum Scriptores, vol. XII, Bologna, 1938–42), p. 243; Ibn al-Athir, Chronicle of Ibn al-Athir, II, p. 32; John Kinnamos, Deeds of John and Manuel Comnenus, bk. III, ch. 5, pp. 81–82. See also Manfroni, Storia della Marina Italiana, I, p. 203–5. 62 John Kinnamos, Deeds of John and Manuel Comnenus, bk. IV, chs. 2–10, pp. 107–23. 63 Annales Casinenses, ed. George Pertz (Monumenta Germaniae Historica, Scriptores, XIX, Hanover, 1866), pp. 303–20, anno 1156, p. 311; John Kinnamos, Deeds of John and Manuel Comnenus, bk. IV, chs 10–13, pp. 122–29; Romuald of Salerno, Romualdi Salernitani Chronicon, pp. 239–40. 64 Bernardo Maragone, Annales Pisani, p. 17. 65 Niketas Choniates, O City of Byzantium!, p. 57. 66 Annales Casinenses, anno 1158, p. 311. 67 “Hugo Falcandus’, La historia o liber de regno Sicilie e la epistola ad Petrum Panormitane



the norman kingdom of sicily 189

presage the decline of Sicilian maritime supremacy in the central Mediterranean. The squandering of Norman naval power on ill-advised adventurism by his successor, William II, would accelerate the process. Shortly after reaching his majority in 1171, William “the Good” initiated a practice of launching large-scale naval expeditions every five or six years to far-flung corners of the Mediterranean with the apparent purpose of projecting the kingdom’s power. There seems, however, to have been no overarching strategy or precise policy goals. As a result, these adventures achieved little while serving as a significant drain on the royal coffers. The first was in 1174, when William II dispatched an armada of some two hundred ships, including war galleys and supply dromōns, to participate in a coordinated attack on Alexandria with the army of King Amalric of Jerusalem (see chapter 10 in this volume). The exploit, undertaken at the request of some Fatimid malcontents seeking the overthrow of Ayyubid rule in Cairo under Saladin, was a spectacular fiasco. Amalric died of dysentery weeks before the Sicilians came ashore in Egypt on 28 June, so the Latin kingdom’s supporting ground force never showed up. Subsequently, Tancred of Lecce, the king’s cousin and overall commander, failed to set up a fortified encampment. As a result, the citizens of Alexandria sortied out of the city’s gates in force on the third night of the siege to butcher many of the besiegers and burn their siege machines. A few nights later, Saladin’s soldiers finished the job by overrunning the camp and driving the surviving Sicilians back to their ships in disarray.68 Unfazed, King William launched another grand enterprise in 1181 – this one consisting of 184 galleys, 45 horse transports, and 40 supply ships, according to the Annales Pisani.69 The objective was ostensibly to suppress Almoravid pirates based in the Balearics off the east coast of the Iberian peninsula, but contemporary annalists record few incidents of Almoravid corsair activity in the central Mediterranean at the time. Moreover, the west Italian maritime powers (Genoa and Pisa) declined to participate. Genoa even renewed its truce with the sultan of Majorca during that same timeframe.70 The Sicilian fleet’s admiral, Walter of Moac, made a unilateral attempt on the archipelago from Liguria the following spring, but the Annales Pisani testifies that tempestuous weather prevented the armada Urbis thesaurarium di Ugo Falcando, ed. G. B. Siragusa (Rome, 1904), pp. 25–28; Ibn al-Athir, Chronicle of Ibn al-Athir, II, pp. 76–77, 103–6. 68 Ibn al-Athir, Chronicle of Ibn al-Athir, II, p. 229; Bernardo Maragone, Annales Pisani, p. 61; William of Tyre, Historia rerum in partibus transmarinis gestarum, ed. R. Huygens, H. Mayer, and R. Gerhard (2 vols., Corpus Christianorum Continuatio Mediaevalis, Turnholt, 1986), II, bk. 21, ch. 3, p. 963. See also Chalandon, Histoire de la Domination Normande, II, pp. 395–97. 69 Maragone (Salem), Annales Pisani, p. 72. 70 Abulafia, Two Italies, p. 157.

190

charles d. stanton

from ever reaching its destination.71 Worst still, William of Tyre reported: “Driven by unfavourable winds, practically the entire fleet was wrecked in the vicinity of the coastal cities of Savona, Albenga and Ventimiglia, where the tumultuous waves dashed the ships on the shore.”72 The impenitent William orchestrated an even worse calamity just three years later. In 1185 he assembled a massive armada at Messina, which the Andalusian traveller and eyewitness Ibn Jubayr said numbered “three hundred sail, between galleys and dromōns”, along with a hundred supply ships.73 Eustathios, the Greek archbishop of Thessaloniki and also a contemporary observer, estimated the ground force which it carried at around eighty thousand, including five thousand knights.74 The conquest of Constantinople had been a Norman desideratum since the late eleventh century when Robert Guiscard hurled two enormous, ill-fated invasions of the Eastern Empire. William was determined to realise it. And, at first, it appeared that his forces might succeed. The immense fleet intimidated Durazzo into a precipitate surrender on 24 June.75 From there, the army marched along the old Roman Via Egnatia to Thessaloniki, while the fleet circumnavigated the Peloponnesus to arrive in mid-August.76 On the 24th, barely nine days after beginning the land–sea siege of the city, the Sicilians took Thessaloniki by storm, amid the sort of slaughter typical of the era.77 It was during the third and final phase of the campaign that it all abruptly fell apart. The land forces en route to Constantinople under undisciplined leadership were surrounded at Amphipolis and butchered.78 Without ground support, the fleet led by the hapless Tancred of Lecce failed to even breach the Bosporus. After a mere seventeen days of perfunctory raiding in the Sea of Marmara, it headed home, only to be decimated by a summer storm. Both Durazzo and Thessaloniki quickly returned to Greek control.79 71

Maragone (Salem), Annales Pisani, p. 72. William of Tyre, Historia rerum in partibus transmarinis gestarum, II, bk. 22, ch. 8, pp. 1017–18; History of Deeds Done beyond the Sea, ed. Emily Babcock and A. C. Krey (2 vols., New York, 1943), II, bk. 22, ch. 8, p. 458. 73 Ibn Jubayr, The Travels of Ibn Jubayr, trans. R. Broadhurst (London, 1952), p. 354. 74 Eustathios of Thessaloniki, The Capture of Thessaloniki, trans. J. Melville Jones (Canberra, 1988), ch. 138, p. 151. 75 Annales Casinenses, anno 1185, p. 313; Eustathios of Thessaloniki, Capture of Thessaloniki, ch. 53, pp. 64–65; La continuation de Guillaume de Tyr (1184–1197), ed. Matthew Morgan (Paris, 1982), ch. 73, pp. 82–83; Niketas Choniates, O City of Byzantium!, p. 164. 76 Eustathios of Thessaloniki, Capture of Thessaloniki, chs. 54–56, pp. 66–9. 77 Eustathios of Thessaloniki, Capture of Thessaloniki, chs. 82–86, pp. 100–5; chs. 98–106, pp. 112–21; La continuation de Guillaume de Tyr, ch. 73, pp. 82–83; Niketas Choniates, O City of Byzantium!, p. 165. 78 Niketas Choniates, O City of Byzantium!, p. 199. 79 Annales Casinenses, anno 1185, p. 313; Niketas Choniates, O City of Byzantium!, pp. 199–201. 72



the norman kingdom of sicily 191

Still undeterred, William was the first Western monarch to respond to Saladin’s devastating defeat of Christian forces at the Horns of Hattin on 4 July 1187. According to contemporary chroniclers, he sent his most adept admiral, Margaritus of Brindisi, at the head of a fleet of fifty or sixty galleys to relieve the beleaguered Latin Kingdom of Jerusalem in the summer of 1188. It saved Tripoli but failed to stop Saladin’s march up the coast of Syria.80 William himself was preparing to conduct yet another ambitious expedition in support of Christian Syria when he suddenly succumbed to an unspecified malady in Palermo on 18 November 1189.81 His penchant for profligate naval campaigns with little or no return meant that his bequest to his kingdom was a depleted royal treasury and diminished naval capability. Worse yet, he died without issue. Atrophy and Imperial Acquisition Catastrophically, William’s death without a direct male descendant invited the end of Norman rule in the Kingdom of Sicily. As it turns out, William literally sold his kingdom for his last campaign. In order to obviate the possibility of an invasion of his realm by the Holy Roman Empire under the Hohenstaufens of Germany while he pursued the conquest of Constantinople, in October 1184 William had promised the hand of his aunt Constance, the posthumous daughter of Roger II, to Henry VI of Hohenstaufen, the son of Frederick Barbarossa. William then designated Constance as his successor, should he pass without fathering a direct heir.82 In an era when patrilineal or agnatic inheritance governed royal succession throughout much of Europe, that meant that the crown of Sicily was to fall to Constance’s husband, Henry VI of the House of Hohenstaufen.83 The majority of the kingdom’s nobility, with the support of the papacy, attempted to preclude that eventuality by electing William’s cousin, Tancred

80

La continuation de Guillaume de Tyr, ch. 73, p. 83; Ibn al-Athir, Chronicle of Ibn al-Athir, II, p. 345; Itinerarium Peregrinorum et Gesta Regis Ricardi in Chronicles and Memorials of the Reign of Richard I, ed. William Stubbs (Rolls Series, Rerum Britannicarum Medii Aevi Scriptores, 38, London, 1864), I, bk. I, ch. XIV, pp. 27–28. 81 Annales Casinenses, anno 1189, p. 314; Chronica de Ferraria, anno 1189, p. 31; La continuation de Guillaume de Tyr, ch. 73, p. 83; Richard of San Germano, Ryccardi de Sancto Germano Notarii Chronica, ed. Carlo Garufi (Rerum Italicarum Scriptores, 2nd edn, VII, Bologna), Explicit Prologus, pp. 4–6. 82 Annales Casinenses, anno 1185, p. 312. See also Chalandon, Histoire de la Domination Normande, II, pp. 386–87; Houben, Roger II of Sicily, p. 172. 83 Annales Casinenses, anno 1190, p. 314; Richard of San Germano, Ryccardi de Sancto Germano Notarii Chronica, Explicit Prologus, p. 6. See also Houben, Roger II of Sicily, p. 172.

192

charles d. stanton 84

of Lecce, as the new king. This, of course, resulted in a disputed claim to the crown. His legitimacy in question from the inception, Tancred was compelled to buy support from wavering barons who sold their allegiance dearly, further sapping the royal fisc.85 To make matters much worse, before this penurious and debilitated new sovereign could solidify his tenuous hold on the throne, two of the most powerful monarchs of the age showed up at his doorstep. In the summer of 1190 Philip II Augustus of France and Richard I Lionheart of England met at Messina at the head of massive sea and land forces. Ostensibly on their way to the Third Crusade, they had come by way of Sicily at the previous invitation of William II, who had intended to join them with his wife, Joanne, the sister of Lionheart. And, of course, Tancred was in no position to refuse.86 He certainly would have done, if he could have done. The combustible combination of unruly crusaders and Richard’s well-chronicled imperiousness was an affront to Sicilian sensibilities. An explosion was inevitable. On 3 October a brawl over a loaf of bread broke out in the market and escalated into a conflagration of violence that consumed much of Messina.87 In the course of conquering the city, the crusaders set the Sicilian fleet ablaze in the harbour. Eyewitness reports from Ambroise and the Itinerarium Peregrinorum indicate that its galleys were burned to the waterline. Messina was the home port of the royal fleet and the primary naval arsenal.88 Thus, in all probability, Richard’s ruffians not only destroyed the Norman fleet almost in total, but also the means by which it could eventually be replaced: the royal arsenal. Subsequent events reveal that Norman naval capability was effectively ended on that date. In the aftermath of Messina’s savage sacking by the crusader armies, Richard and Philip extorted exorbitant reparations from Tancred. The enfeebled Norman monarch ended up siphoning off another forty thousand ounces in gold from the royal treasury to pay them off, further frustrating any hope of restoring his fleet in the near future.89 What was worse was that he had to endure his disruptive and demanding guests 84

Annales Casinenses, anno 1190, p. 314; Annales Ceccanenses, ed. George Pertz (Monumenta Germaniae Historica, Scriptores, XIX, Hanover, 1866), pp. 275–302, esp. anno 1189, p. 288. 85 Richard of San Germano, Ryccardi de Sancto Germano Notarii Chronica, anno 1190, p. 6. 86 Itinerarium Peregrinorum et Gesta Regis Ricardi, I, bk. II, ch. IX, p. 150. 87 Ambroise, History of the Holy War: Ambroise’s Estoire de la Guerre Saint, ed. and trans. Marianne Ailes and Malcom Barber (2 vols., Woodbridge, 2003), I, lines 714–36, p. 12; Itinerarium Peregrinorum et Gesta Regis Ricardi, I, bk. II, chs. XV–XVI, pp. 159–63. 88 Ambroise, Ambroise’s Estoire de la Guerre Saint, I, lines 816–17, p. 14; Itinerarium Peregrinorum et Gesta Regis Ricardi, I, bk. II, ch. XVI, p. 164. 89 Ambroise, Ambroise’s Estoire de la Guerre Saint, I, lines 981–1049, pp. 16–17; Itinerarium Peregrinorum et Gesta Regis Ricardi, I, bk. II, ch. XXI, pp. 169–70; Roger of



the norman kingdom of sicily 193

until the following spring (April 1191), when they finally departed for the Holy Land.90 Ironically, at about the same time Pope Clement III crowned Henry VI of Hohenstaufen as Holy Roman Emperor. His father, Frederick Barbarossa, had drowned in Asia Minor the previous year on his way to Palestine, making Henry successor to the German throne.91 Henry forthwith concluded treaties of alliance with both Pisa and Genoa.92 Soon thereafter he began a resolute campaign to convert his claim on the crown of Sicily into a reality. Tancred resisted as best he could, but he simply possessed neither the manpower nor the maritime assets to hold out indefinitely. Throughout the protracted struggle to retain the Regno there are no confirmed accounts of a Norman fleet engaging the enemy.93 Tancred’s death of illness on 20 February 1194 marked the end of the effort.94 When Henry made his final push to conquer the kingdom in the spring of that same year with the full support of the Pisan and Genoese fleets, not a single Sicilian ship was recorded to have set sail to oppose him.95 Conclusion There is no small irony in the fact that the very same west Italian sea powers that so lustily participated in the demise of the Norman Kingdom of Sicily were also the primary beneficiaries of the rise of Norman naval dominance in the central Mediterranean. It was, in fact, the Norman conquest of Sicily and the surrounding shores from the Muslims that permitted the easy extension of Genoese and Pisan commercial enterprise eastward – a movement accelerated and enhanced by the advent of the crusading movement. Indeed, the trade routes and mercantile colonies that the two maritime republics established in the wake of the Crusades fed their prosperity and geopolitical ascendancy for nearly two centuries.96 Nor was Venice denied the spoils of Howden, Chronica Magistri Rogeri de Houedene, ed. William Stubbs (4 vols., Rolls Series 51, London, 1868–71), III, pp. 61–66. 90 Itinerarium Peregrinorum et Gesta Regis Ricardi, I, bk. II, ch. XXVI, pp. 176–77. 91 Annales Casinenses, anno 1191, p. 314; Otto of St Blasien, Ottonis de Sancto Blasio Chronica, ed. Adolf Hofmeister (Monumenta Germaniae Historica, Scriptores Rerum Germanicarum, Hanover, 1912), p. 55; Richard of San Germano, Ryccardi de Sancto Germano Notarii Chronica, anno 1191, pp. 10–11. 92 Constitutiones et Acta Publica Imperatorum et Regum, ed. Ludwig Weiland (Monumenta Germaniae Historica, Legum Sectio IV, Vol. I, Hanover, 1893), no. 333, pp. 472–77; no. 337, pp. 479–83. 93 Stanton, Norman Naval Operations, pp. 169–71. 94 Annales Casinenses, anno 1194, p. 317; Chronica de Ferraria, anno 1193, p. 32; Richard of San Germano, Ryccardi de Sancto Germano Notarii Chronica, anno 1193, p. 16. 95 Manfroni, Storia della Marina Italiana, I, p. 264. 96 Stanton, Norman Naval Operations, pp. 174–89.

194

charles d. stanton

Norman Sicily’s success on the Middle Sea, for the source of their own great wealth and mercantile reach can be traced back to generous commercial concessions granted to the Venetians by the Byzantine Emperor Alexios I Komnenos. Desperate for maritime allies in his struggle against the rise of Norman sea power under Robert Guiscard, Alexios issued a chrysobull in 1082 that not only conceded to the Venetians an invaluable fondaco in Constantinople, but also allowed them to trade exempt from tariffs throughout the Eastern Empire, with the exception of Cyprus and Rhodes.97 Manuel I Komnenos confirmed those privileges in 1148 and expanded them to include Cyprus and Rhodes as well in order to enlist the Venetians in his own fight against Norman naval incursions.98 These sweeping concessions in the East enabled the Venetians to establish a commercial empire extending all the way into the Black Sea. Although Norman naval power developed differently, operated distinct from any other Italian-based sea power, and died a premature death, it was clearly a catalyst for Western maritime mastery in the Mediterranean, lasting well into the Ottoman era.

97

Tafel and Thomas (eds), Urkunden der Republik Venedig, I, doc. XXIII, pp. 43–54. See also Manfroni, Storia della Marina Italiana , I, pp. 124–25, 201. 98 Tafel and Thomas, Urkunden der Republik Venedig, doc. L, pp. 109–13. See also Manfroni, Storia della Marina Italiana , I, p. 201.

8

Norman Participation in the First Crusade: a Re-examination Luigi Russo

In the 1980s the Italian medievalist Raoul Manselli noted the historiographical gap that existed in general works dedicated to the crusader movement regarding “participation in the crusade, and especially in southern Italy”, lamenting the delay with which Italian scholarship took interest in this issue.1 His observation is still relevant today and forms the topic of this chapter. I will concentrate on the early phases of the participation by the Normans of southern Italy in the crusades, tackling aspects related to the recruitment of the Norman contingent led by Bohemond of Hauteville, and reflecting on the dynamics of their unexpected participation which caused unease and hostility with the Byzantine Empire, urging the West at the time for military support against the Turks.2 In my opinion, the key document in this discussion is the parchment compiled by order of Bohemond on the eve of his departure for the Holy Land, currently preserved at the Nicolaian Museum of Bari, inaugurated in 2010 (Fig. 1). The document solemnly opens by delegating to Guidelmus, catepan of Bari, the sale “de cuncta hereditate que michi in eadem civitate mea Baro pertinent” “of all my inheritance present in the same city of Bari”).3 1

Raoul Manselli, Italia e Italiani alla prima crociata (Rome, 1983), p. 7. The only relevant work that had appeared in the immediately preceding years was Ernesto Pontieri, Tra i Normanni nell’Italia meridionale (Naples, 1964), pp. 361–408. 2 On the Byzantine Empire’s request for military help, see Peter Frankopan, The First Crusade: The Call from the East (London, 2012), pp. 87–100. A good introduction to the question can be found in Jonathan Harris, “Byzantium and the First Crusade: Three Avenues of Approach”, Estudios bizantinos 2 (2014), 125–41. 3 Francesco Nitti di Vito, ed., Le pergamene di S. Nicola di Bari (1075–1194) (Bari, 1902), [Codice diplomatico barese, 5], doc. 22, pp. 41–42, lines 1–7: “Ego Boamundus, Robberti ducis filius, hanc potestatem et auctoritatem do et concedo tibi Guidelmo flammengo catepano civitatis nostre Bari ut tu, in vice mea, de cuncta hereditate que michi in eadem civitate mea Baro pertinet, videlicet casis casilibus curtis vineis vinea-

196

Fig. 1

luigi russo

Parchment compiled by order of Bohemond, 1096 (Courtesy of Nicolaian Museum of Bari)



norman participation in the first crusade 197

Ralph B. Yewdale, in the first biography of Bohemond, did not give this document the necessary attention;4 yet the list of possessions in Bari of the eldest son of Robert Guiscard provides evidence for reconstructing the early stages of the former’s presence on the crusade. Additionally, it highlights two important questions, frequently sidelined in the context of the history of the crusader movement, i.e. the questions of recruitment and of the logistics of the movement of warriors in a prolonged military campaign.5 The implications of Bohemond’s liquidation of his heritage are clear: preparing to set out to the East, he was focused on collecting as much money as he could. Jonathan Riley-Smith has estimated that the costs of participation in the crusade could amount four or five years’ income for the average medieval knight.6 Bohemond chose to sell the real estate (houses and fields) located

libus olivis terris simul et de omni causa mea que intus in Baro michi pertinet, et foris licentiam habeas vendere, vicariare, dare, et concedere, sine requisitione et contrarietate mea meorumque heredum et successorem, sive nostrorum omnium hordinatorum, et quicumque a te acceperit, firmum et stabile illi ita permaneat.” [“I Bohemond, son of the duke Robert, give and grant you Guidelmus Flammengus, catapan of our city of Bari, full right so that you can, in my place, sell, exchange, give and grant all my inheritance due in the city of Bari, that is to say houses, farmhouses, courts, vineyards, olive trees, lands and everything that belongs to me inside and outside the city of Bari, without any reservation or opposition on my part, of my heirs and successors, or of all our administrators, and whoever receives something from you, definitely receives it.”] The document is signed with the lead seal of Bohemond, which remained intact and is reproduced in table 1, section seals (doc. 4). An in-depth analysis of the document is provided by Pasquale Cordasco, “Boamundus ducis Robberti filius: cancelleria e documenti di Boemondo I, principe di Antiochia”, in “Unde boat mundus quanti fuerit Boamundus”: Boemondo I di Altavilla, un normanno tra Occidente e Oriente. Atti del Convegno internazionale di studio per il IX centenario della morte, Canosa di Puglia, 5–6–7 maggio 2011, ed. Cosimo Damiano Fonseca and Pasquale Ieva (Bari, 2015), pp. 93–206. On the catepan Guidelmo, see Giosuè Musca and Pasquale Corsi, “Da Melo al regno normanno”, in Storia di Bari. Dalla conquista normanna al ducato sforzesco, 2, ed. Francesco Tateo (Bari, 1990), pp. 5–55, especially p. 37. 4 Ralph Bailey Yewdale, Bohemond I: Prince of Antioch (1924; repr. Amsterdam, 1970), p. 36. Further details in Jean Flori, Bohémond d’Antioche: chevalier d’aventure (Paris, 2007), pp. 70–71, where she discusses in depth the total number of Bohemond’s followers. 5 Pryor rightly claims that: “Frequently, it is as though the Crusaders’ human needs, the lands through which they passed, and the means by which they did so, did not exist.” John H. Pryor, “Preface”, in Logistics of Warfare in the Age of the Crusades, ed. John H. Pryor (Aldershot, 2006), p. xii. 6 For a cost estimate and the crusaders’ financial strategies, see Jonathan Riley-Smith, The First Crusaders, 1095–1131 (Cambridge, 1997), pp. 93–105; see also Michel Balard, “La preparazione economica delle crociate”, in Il Concilio di Piacenza e le crociate, ed. Pierre Racine (Piacenza, 1995), pp. 193–200. Compare the more cautious arguments of Giles Constable, Crusaders and Crusading in the Twelfth Century (Farnham, 2008) on the variability of shared expenditure for every pilgrim, p. 141.

198

luigi russo

in the city of Bari, hoping for a decisive injection of cash to finance his expedition to the East.7 Scarcity of money would have been one of the most pressing problems for Bohemond during the journey, as compared to other, better-off leaders able to draw on more substantial resources. For example, Robert Curthose, duke of Normandy, received ten thousand silver marks from his brother William Rufus, king of England, in exchange for the temporary sale of the Duchy of Normandy,8 while Godfrey of Bouillon sold the family castle to Bishop Otbert of Liège upon payment of 1,300 silver marks and three gold ones with a right to redeem.9 Bohemond’s shortage of cash would, inevitably, have had an influence on the course of events. The financial limitations of Guiscard’s first-born are not surprising in light of his difficult situation in southern Italy vis-à-vis his half-brother Roger Borsa (legitimate heir of Robert Guiscard), against whom he had fought at length in the immediately preceding years – a fact which determined the heterogeneity of Bohemond’s possessions.10 In the late 1080s and early 1090s, Bohemond controlled Giovinazzo, Taranto, Oria, Otranto, and Gallipoli, alongside half of Bari, without boasting the princely title of Taranto that modern historiography wrongly attributes to him.11 His patrimony, therefore, was 7

Despite the alienation of the possessions in Bari, a later document clarifies that Bohemond still retained some jurisdiction over the town, see Le pergamene di S. Nicola di Bari, doc. 32, pp. 55–57 (October 1100), where the barinorum critis Nicola judged a cause “per auctoritatem gloriosi nostri domini Boamundi”, p. 57, lines 60–61. 8 For further details on Robert, see Judith A. Green, “Robert Curthose Reassessed”, in Anglo-Norman Studies XXII. Proceedings of the Battle Conference, ed. Christopher Harper-Bill (Woodbridge, 2000), pp. 95–116; William M. Aird, Robert Curthose, Duke of Normandy (c.1050–1134) (Woodbridge, 2008), pp. 153–67; on the negotiations between the two brothers, see David Crouch, Normans. The History of a Dynasty (London, 2002), pp. 146–52. 9 The clauses of the agreement are in Reineri monachi Sancti Laurentii Leodiensis, Triumphale Bulonicum, ed. Wilhelm Arndt, MGH SS 20:584: “Lotharingiae dux celebris Godefridus […] pro mille ac trecentis argenti et tribus auri marcis Oberto, Leodiensi episcopo, idem castrum contradidit, vadii tamen nomine, alterutrum scilicet tali conditione firmata; quatinus si repatriasset dux, ipse facultatem haberet redimendi, sin vero minime reverti contigisset, eadem ad retinendum facultas fratri eius Eustachio Bononiensi cederet comiti; qui redemptione si supersedisset, tum vero in ius ac ditionem Sanctae Mariare Sanctique Lamberti aeterna hereditate libera et rata possessione transiret”; the transactions are also reported by Chronicon Sancti Huberti Andaginensis, ed. Ludwig Conrad Bethmann and Wilhelm Wattenbach, MGH SS 8:615; see also Laurentii de Leodio gesta episcoporum Virdunensium et abbatum S. Vitoni, ed. Georg Waitz, in MGH SS 10:498; and The Ecclesiastical History of Orderic Vitalis, ed. Marjorie Chibnall, 6 vols. (Oxford, 1969–80), 5:208. 10 Luigi Russo, Boemondo. Figlio del Guiscardo e principe di Antiochia (Avellino, 2009), pp. 43–56. 11 For Apulia, see Wolfgang Jahn, Untersuchungen zur normannischen Herrschaft in Süditalien (1040–1100) (Frankfurt am Main, 1989), pp. 146–57, especially pp. 153, 156–57;



norman participation in the first crusade 199

far from constituting a political and jurisdictional entity with defined borders. Instead, it was the result of a series of aggressions, held together by Bohemond’s political and military initiatives that exploited the power vacuum which characterised the Apulian political setting at least until the early decades of the twelfth century, when the rise of Roger II guaranteed the latter’s uncontested rule in the whole of southern Italy.12 Regardless, the property liquidation in Bari would have afforded temporary relief of the pressing financial demands of an undertaking as expensive as the crusade,13 especially when one bears in mind that Bohemond had to pay the cost of ferrying the knights of his entourage across the Adriatic Sea14 – an expense not at all negligible if the testimony in the Annals of the chronicler Lupus Protospatharius of Bari that they were more than five hundred is reliable.15 In the council of the princes participating in the First Crusade, the Norman sometimes complained about Bohemond’s precarious financial situation, declaring his lack of money.16 His economic deficit was see also the excellent essay of Vera von Falkenhausen, “Taranto”, in Itinerari e centri urbani nel Mezzogiorno normanno-svevo. Atti delle decime Giornate normanno-sveve, ed. Giosuè Musca (Bari, 1993), pp. 451–75, especially pp. 453–54. On the alleged princely title of Taranto, see Russo, Boemondo, pp. 50–51 and related notes. 12 An analysis of the formation of the Norman dominions in Puglia and Basilicata can be found in Jahn, Untersuchungen zur normannischen Herrschaft, pp. 355–60. On political events, see Hubert Houben, Ruggero II di Sicilia. Un sovrano tra Oriente e Occidente (Rome, 1999), pp. 41–77. 13 Flori, Bohémond d’Antioche, p. 70: “La croisade, en effet, coûte très cher.” 14 Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. Rosalind Hill and Roger A. Baskerville Mynors (London, 1962), p. 8: “Hi omnes transfretaverunt ad Boamundi famulatum, et applicuerunt Bulgariae partibus.” [“All of these crossed over in the service of Bohemond and landed in Bulgaria.”] According to the Hystoria de via et recuperatione Antiochiae atque Ierusolymarum, ed. Edoardo d’Angelo (Tavarnuzze, 2009), p. 20: “Alii vero, qui ante transierunt mare, promiserunt Boamundo, quod eum [Boemondo] expectarent in Bulgaria.” 15 Lupi Protospatarii Annales, ed. Georg H. Pertz, MGH SS 5:62: “subito inspiratione Dei Boamundus cum aliis comitibus et plus quam 500 equitibus, facientes sibi signum crucis super pannos in humero dextro, reliquerunt obsidionem [of Amalfi]” [“through the inspiration of God, Bohemond with other counts and more than five hundreds knights, fixing the sign of the cross on the right shoulder of their robes, abandoned the siege (of Amalfi)”]. Some doubts about the number of Bohemond’s followers were expressed by Riley-Smith, The First Crusaders, p. 100: “The following he [Bohemond] recruited does not look particularly impressive.” A confirmation of the number of ItaloNorman followers, however, comes from the Annales Cavenses, ed. Fulvio delle Donne (Rome, 2011) [Rerum Italicarum scriptores, 9], p. 40 (s.a. 1096): “Gens multa Normannorum et Italorum mare transierunt.” Brilliant reflections on the use of numbers in historical and epic sources can be found in the essay of the late Jean Flori, “Des Chroniques aux chansons de geste: l’usage des nombres comme élément de typologie”, Romania 117 (1999), 396–422. 16 In that regard, see Raymond d’Aguilers, Liber, ed. John Hugh Hill and Laurita L. Hill (Paris, 1969), chap. 7, p. 53: “Accedebat etiam et alia exercitui calamitas, quod Boimundus

200

luigi russo

mitigated by the contributions of the basileus Alexios I Komnenos during the early stages of the journey. Economic support continued as long as the agreement between the Western forces and the Byzantine authorities remained active – that is, until the conquest of Antioch in the summer of 1098. In receiving this help, the crusading armies were allowed to focus on the strategic and logistical fields, a focus which had assisted Byzantium in coping with numerous military threats in the past.17 The Norman Participation in the First Crusade According to the most recent studies, those who decided to join the crusade did so for a variety of reasons. Therefore the previously held idea that partic​​ ipation in the crusade functioned as a “safety valve” for knights in search of material gains and personal advantages, suggested by the leading French medievalist Georges Duby in a famous essay more than fifty years ago, is too simplistic.18 There was no single reason to take the cross; rather, there were a wide range of reasons to be weighed carefully by those responding to the periodic appeals of ecclesiastical authorities in the medieval West.19 The Normans of southern Italy at the time of the First Crusade are a good example, as I will show on the basis of an analysis of the principal knights following Bohemond: qui clarissimus factus fuerat in Hispania discessurum se aiebat, eo quod propter honorem venerit, et homines et equos suos inopia deperire conspiceret, nec esse divitem se dicebat, cui ad tam longam obsidionem rei familiaris opes sufficerent …”. Christopher Tyerman (ed.), Chronicles of the First Crusade (London, 2012), pp. 153–54: “To add to our misfortunes, Bohemund now famous for his brilliant service in Hispania, threatened to depart, adding that honour had brought him to his decision because he saw his men and horses dying from hunger; moreover, he stated that he was a man of limited means whose personal wealth was inadequate for a protracted siege.” 17 John Haldon, Warfare, State and Society in the Byzantine World, 565–1204 (London, 1999). 18 Georges Duby, “Dans la France du nord-ouest au XIIe siècle: les ‘Jeunes’ dans la société aristocratique”, Annales E.S.C. 19 (1964), 835–46, especially p. 844: “La ‘jeunesse,’ on le voit, ce groupe de turbulence prolongée, exclu par tant de conditions sociales du corps des hommes établis, des pères de famille, des chefs de maison, cette marge instable qui suscita et soutint à la fois les entreprises de la croisade, l’engouement pour les tournois, la propension au luxe et au concubinage, exerça une influence décisive sur la démographie de la noblesse de cette région et sur l’évolution de ses patrimoines” [italics are mine]. Equally misleading is the idea of​“un’esaltazione collettiva mistico-cavalleresca” at the heart of the crusader movement hypothesised by Pontieri, Tra i Normanni nell’Italia meridionale, p. 381. Criticisms of this model have been formulated by Jonathan Riley-Smith, “Early Crusaders to the East and the Costs of Crusading, 1095–1130”, in Cross-Cultural Convergences in the Crusader Period. Essays Presented to Aryeh Grabois on his Sixty-Fifth Birthday, ed. Michael Goodich, Sophia Menache, and Sylvia Schein (New York, 1995), pp. 237–57. 19 For further details, see Luigi Russo, I crociati in Terrasanta. Una nuova storia (1095– 1291) (Rome, 2018), pp. 19–21.



norman participation in the first crusade 201 My lord Bohemond went home to his own land and made careful preparations for setting out on the way to the Holy Sepulchre. Thereafter he crossed the sea with his army, and with him went Tancred son of the Marquis, Richard of the Principality and Ranulf his brother, Robert of Anse, Herman of Cannes, Robert of Sourdeval, Robert Fitz-Toustan, Humphrey Fitz-Ralph, Richard son of Count Ranulf, the count of Russignolo and his brothers, Boel of Chartres, Aubré of Cagnano and 20 Humphrey of Monte Scaglioso.

The discussion will be based on the aforementioned list of the main Norman crusaders reported by the anonymous chronicler of the Gesta Francorum et aliorum Hierosolimitanorum to have followed the Italo-Norman army until Antioch.21 A well-established historiographical tradition has recognised in the participants of this first wave of crusaders under Bohemond the less rich, powerful, and integrated representatives of aristocratic families than those following the prestigious Hauteville clan. As such, they were more attuned to the type of recruitment promoted by Bohemond once he became aware of Pope Urban II’s appeal from Clermont (November 1095),22 an evaluation to which I will return at the end of this chapter. Understanding the importance of regional differences in the initial decision to participate in the crusader movement is key. While a fair share of

20

Gesta Francorum et aliorum Hierosolimitanorum, 1, pp. 7–8: “Denique reversus iterum in terram suam dominus Boamundus diligenter honestavit sese ad incipiendum Sancti Sepulchri iter. Tandem transfretavit mare cum suo exercitu, et cum eo Tancredus Marchisi filius, et Richardus princeps, ac Rainulfus frater eius, et Rotbertus de Ansa, et Hermannus de Canni, et Rotbertus de Surda Valle, et Robertus filius Tostani, et Hunfredus filius Radulfi, et Ricardus filius comitis Rainulfi, et comes de Russignolo cum fratribus suis, et Boello Carnotensis, et Alberedus de Cagnano, et Hunfredus de Monte Scabioso.” A slightly different list is found in the Chronica monasterii Casinensis, ed. Hartmut Hoffmann, MGH SS 34:476–77. For a thorough analysis, see Luigi Russo, “The Monte Cassino Tradition of the First Crusade: From the Chronica monasterii Casinensis to the Hystoria de via et recuperatione Antiochiae atque Ierusolymarum”, in Writing the Early Crusades. Texts, Transmission and Memory, ed. Marcus Bull and Damien Kempf (Woodbridge, 2014), pp. 53–62. 21 More details on the author in the Italian edition of the text: Le gesta dei Franchi e degli altri pellegrini gerosolimitani, ed. Luigi Russo (Alessandria, 2003), pp. 7–12. See also Kenneth Baxter Wolf, “Crusade and Narrative. Bohemond and the Gesta Francorum”, Journal of Medieval History 17 (1991), 207–16. 22 Evelyn Jamison, “Some Notes on the ‘Anonymi Gesta Francorum,’ with Special Reference to the Norman Contingent from South Italy and Sicily in the First Crusade”, Studies in French Language and Medieval Literature Presented to Professor Mildred K. Pope (Manchester, 1939), pp. 183–208; Bruno Figliuolo, “Ancora sui Normanni d’Italia alla prima crociata”, Archivio storico per le province napoletane 54 (1986), 1–16; more recently Luigi Russo, “I Normanni e il movimento crociato. Una revisione”, in Il Papato e i Normanni. Temporale e spirituale in età normanna, ed. Edoardo d’Angelo and Claudio Leonardi (Florence, 2011), pp. 163–74.

202

luigi russo

kin groups – mostly of the Franco-Flemish area23 –had engaged in long and expensive journeys to the Holy Land on the basis of a series of traditions which had gradually become an integral part of their heritage, and which were further fuelled by papal declamations, instead, we see that Italo-Norman participation in the crusade was limited to a quite small number of knights who were not particularly important politically vis-à-vis other continental European lords. Precisely because Norman participation was deprived from the outset of the support of the most eminent families linked to Roger Borsa and Count Roger I, it was characterised by political fragility destined to limit their contribution to the birth of Outremer. This precarious political situation would have had a considerable influence on the nature of the recruitment of the Norman army of southern Italy. An analysis of the profiles of the principal knights following Bohemond immediately highlights the almost total absence of aristocrats related to the court of Count Roger I, except for Robert of Sourdeval, who appears in some documents of the 1090s immediately before setting out to the East.24 Indeed, another fellow of the Sicilian count, Roger of Barnesville – one of the most important landowners in Sicily, according to the sources25 – did not join Bohemond but instead the army led by Robert Curthose, duke of Normandy, who arrived in southern Italy in the winter of 1096–97, crossing the Adriatic at the beginning of spring.26 This confirms the fact that the entire operation scarcely garnered interest from among the most trusted members of the Sicilian court, who did not support the crusade at all.27 23

The memory of their contribution to the conquest of Jerusalem is preserved in a series of genealogical and historical texts from Flanders: Genealogia regum Francorum comitumque Flandriae, ed. Ludwig C. Bethmann, MGH SS 9:308; Genealogia Flandrensium comitum (Flandria generosa), MGH SS 9:323 (n.**). See, for example, Nicholas L. Paul, “Crusade, Memory and Regional Politics in Twelfth-Century Amboise”, Journal of Medieval History 31 (2005), 127–41. 24 Léon-Robert Ménager, “Inventaire des familles normandes et franques émigrées en Italie méridionale et en Sicile (XIe–XIIe siècles)”, in Hommes et institutions de l’Italie normande, ed. Léon-Robert Ménager, 4 (London, 1981), p. 346. Julia Becker (ed.), Documenti latini e greci del conte Ruggero I di Calabria e Sicilia (Rome, 2013), doc. 40, pp. 166–68; doc. 60, pp. 232–34 (the date proposed by the modern editor, 1097, is incorrect: at that time Robert was in the East and no longer in Sicily). See also Alan V. Murray, “How Norman was the Principality of Antioch? Prolegomena to a Study of the Origins of the Nobility of a Crusader State”, Family Trees and the Roots of Politics: The Prosopography of Britain and France from the Tenth to the Twelfth Century, ed. Katharine S. B. Keats-Rohan (Woodbridge, 1997), pp. 349–59, especially p. 356. 25 Léon-Robert Ménager (ed.), Recueil des actes des ducs normands d’Italie, 1046–1127, I, Les premiers ducs, 1046–1087 (Bari, 1981), docs. 52–53, pp. 181–84; Documenti latini e greci del conte Ruggero I di Calabria e Sicilia, doc. 39, pp. 162–64; doc. 40, pp. 166–68; doc. 52, pp. 205–7; doc. 60, pp. 232–34. 26 Aird, Robert Curthose, Duke of Normandy, pp. 168–70. 27 Ménager, “Inventaire des familles normandes et franques”, pp. 353–54. Willelmi



norman participation in the first crusade 203

Going back to those knights who set off with Bohemond, an examination of the warriors listed in the various sources at our disposal – i.e. the Chronicle of Monte Cassino, Gesta Francorum, Hystoria de via, and to a lesser extent Ralph of Caen – points to their provenance from the existing regions of Campania and Basilicata (exceptions are Herman of Cannes, Humphrey Fitz-Ralph, Aubré of Cagnano from Puglia, and Robert of Sourdeval, who owned properties in Molise), in which Bohemond did not hold any territorial possessions.28 Rather, recruitment seems to have taken place partly within the extended Hauteville family network, with particular reference to the second-rate members or those less deeply rooted in southern Italy. Suffice it to note in this respect Richard of the Principality and his two older siblings, who did not have significant career prospects within their family,29 or Herman of Cannes, the youngest son of the late Humphrey, duke of Puglia (1057), who had revolted several times in the past against Guiscard and whose room for manoeuvre in Italy was limited.30 In addition, the career of Robert, son of Gerard, deserves mention. He was the second-born of Gerard, count of Buonalbergo, as well as a cousin of Bohemond on his mother’s side. Robert played the role of constable to Bohemond during the crusade and remained in the Latin East until the death of his elder brother; he then returned to the family’s possessions, where he reappeared alongside his nephew Giordanus, sharing with him the management of their common inheritance until at least 1119.31 The political profile of Geoffrey of Monte Tyrensis archiepiscopi chronicon, ed. Robert Burchard Constantijn Huygens (Turnhout, 1986), Corpus Christianorum, Cont. Med., 63–63A, I, p. 191 (Rogerus de Barnevilla); Albert of Aachen, Historia Ierosolimitana, ed. Susan B. Edgington (Oxford, 2007), pp. 198–99, 234–35, 288–91. On this, see also Luigi Russo, “Oblio e memoria di Boemondo d’Altavilla nella storiografia normanna”, Bullettino dell’istituto storico per il medio evo 106 (2004), pp. 137–65. 28 The presence of Boel of Chartres and members of the Rossignolo family remains enigmatic, as underlined by Jamison, “Some Notes on the ‘Anonymi Gesta Francorum’”, pp. 297–98. 29 Radulphus Cadomensis Tancredus, ed. Edoardo d’Angelo (Turnhout, 2011) (Corpus Christianorum, Cont. Med., 231), chap. 154, p. 44: “Ricardus de Principatu, nec genere minor nec animis, cognatas partes et lingua simul animabat, et lancea. Is, comitis Wilhelmi filius, Wiscardi nepos, relicta fratri Tancredo Syracusa, Boamundum secutus amitialem suum, sociaverat Tancredum.” Further details in George T. Beech, “A Norman-Italian Adventurer in the East. Richard of Salerno, 1097–1112”, Anglo-Norman Studies XV (1993), 26–40. 30 See Raoul Manselli, “Altavilla, Ermanno d’”, and “Altavilla, Unfredo d’”, Dizionario biografico degli Italiani, 2 vols. (Rome, 1960), p. 542 and pp. 548–49. Herman appears among the subscribers for a donation in the name of Guiscard – already dead – to the monastery of Sant’Angelo of Mileto, see Recueil des actes des ducs normands d’Italie, 1046–1127, doc. 60, pp. 212–15. 31 Jamison, “Some Notes on the ‘Anonymi Gesta Francorum’”, pp. 293–94; Jonathan Riley-Smith, “The Motives of the Earliest Crusaders and the Settlement of Latin Palestine, 1095–1100”, The English Historical Review 98 (1983), 721–36, especially p. 727.

204

luigi russo

Scaglioso is very different (note that his father, Humphrey, had already died during the crusade, fighting on the front line of the Hauteville clan):32 as the first-born, he inherited the most important share of his father’s estate, Monte Scaglioso in Basilicata. Based on this, it can be assumed that his response to the crusade was inspired by the prospects of the East.33 The same applies to another crusader, Robert Fitz-Toustan, nephew of Ralph II count of Molise, whose donations to the church of Santa Sofia of Benevento are testament to his significant wealth.34 Given the above, Bohemond’s recruitment during the siege of Amalfi was effected according to his extended network of kin relationships rather than seigneurial bonds. Another hypothesis put forward here is that the decision of some knights to join Bohemond was based on his reputation for military prowess, acquired during the military campaigns of 1081–84 against the Byzantine Empire in which he had played a decisive role despite the final defeat of the Normans.35 Last, but not least, the participation of Robert of Anzi in the ItaloNorman army remains highly problematic. He appears alongside Oria citizens who were besieged by Bohemond in 1091,36 while after the conquest of Antioch (1098) he was linked to Lorraine’s army led by Godfrey of Bouillon. 32

Humphrey of Montescaglioso appears in the list of benefactors of the abbey of Venosa, the burial place of the Hauteville dynasty. See Il “libro del capitolo” del monastero della SS. Trinità di Venosa (Cod Casin. 334): una testimonianza del Mezzogiorno normanno, ed. Hubert Houben (Galatina, 1984), p. 99. 33 Errico Cuozzo, “La contea di Montescaglioso nei secoli XI–XIII”, Archivio storico per le province napoletane 103 (1985), 7–37, especially p. 28. 34 Ménager, “Inventaire des familles normandes et franques”, Addenda, pp. 9–10; Evelyn Jamison, “The Administration of the County of Molise in the Twelfth and Thirteenth Centuries”, The English Historical Review 44 (1929), 529–59, especially p. 558. 35 The reference work is Georgios Theotokis, The Norman Campaigns in the Balkans, 1081–1108 (Woodbridge, 2014), pp. 137–84; but also Luigi Russo, “L’Expansion normande contre Byzance (XIème–XIIème siècles). Réflexions sur une question toujours ouverte”, in 911–2011: Penser les mondes normands médiévaux, ed. David Bates and Pierre Bauduin (Caen, 2016), pp. 147–62. 36 Romualdus archiepiscopus Salernitanus, Chronicon, ed. Carlo Alberto Garufi (Bologna, 1909–1935) (Rerum Italicarum scriptores, 7:1), p. 199: “Eodem anno [1091] dum obsideretur Oria ciuitas a Boamundo, auxilio Roberti de Anse, Oritani dissipauerunt obsidionem eius, et ipso Boamundo fugam petente cunctum eius, apparatum et signa ceperunt” [“army, and with him went Tancred son of the Marquis, Richard of the Principality and Ranulf his brother, Robert of Anse, Herman of Cannes, Robert of Sourdeval, Robert Fitz-Toustan, Humphrey Fitz-Ralph, Richard son of Count Ranulf, the count of Russignolo and his brothers, Boel of Chartres, Aubré of Cagnano and Humphrey of Monte Scaglioso”]. On the conflation of Robertus de Anse with Roberto of Anzi, see Jamison, “Some Notes on the ‘Anonymi Gesta Francorum’”, pp. 294–95, and also Edoardo d’Angelo, “Prolegomena to a New Edition of Lupus Protospatharius’s Annales”, Latin Culture in the Eleventh Century. Proceedings of the Third International Conference on Medieval Latin Studies. Cambridge, September 9–12, 1999, ed. Michael W.



norman participation in the first crusade 205

He is later reported as receiving rents near Arsur, ultimately becoming permanently integrated into the nobility of the Kingdom of Jerusalem. His career shows that the rise of some Norman knights in Bohemond’s army was episodic, linked to circumstances of the moment, and certainly not a result of their enduring loyalty to Guiscard’s first-born.37 The fluidity of crusaders’ careers has been emphasised by Alan Murray in multiple studies.38 Bohemond in the Occident The peculiar characteristics of Norman crusading recruitment in southern Italy are confirmed by a fact scarcely noted by scholars. Once back in the West, Bohemond, the new prince of Antioch, renewed his propaganda efforts in France, this time for the recruitment of knights for the military campaign against Byzantium (1106). His strategy and commitment did not find a parallel in the Italian peninsula, where, in the year 1107, he began the ill-fated siege of the port of Durazzo in a renewed war effort against the Byzantine Empire.39 In fact, the importance of southern Italy had been overshadowed during Bohemond’s stay in the West. All his efforts were concentrated in the French regions he encountered, culminating in his prestigious wedding with Constance, daughter of the Capetian King Philip I, held in Chartres (May 1106).40 Further, Bohemond attempted to replicate a similar strategy in England, but was cordially yet firmly denied by King Henry I when he requested to visit the Anglo-Norman court. This proves that Norman recruitment strategies were explicitly more concentrated outside the Italian peninsula, where Bohemond did not think it possible to convince Herren, Christopher James McDonough, and Ross G. Arthur (Turnhout, 2002), pp. 167–85, especially pp. 184–85. 37 Albert of Aachen, Historia Ierosolimitana, pp. 500–3; Murray, How Norman Was the Principality of Antioch?, p. 353; Alan V. Murray, “The Army of Godfrey of Bouillon, 1096–1099: Structure and Dynamics of a Contingent on the First Crusade”, Revue belge de philologie et d’histoire 70 (1992), 301–29, especially p. 328. 38 Murray, The Army of Godfrey of Bouillon, pp. 301–29. 39 For further details, see Russo, Boemondo. Figlio del Guiscardo e principe di Antiochia, pp. 177–95; Theotokis, The Norman Campaigns in the Balkans, pp. 200–14. The lack of support for Bohemond in Italy is rightly noted by John France, “The Normans and Crusading”, The Normans and their Adversaries at War, ed. Richard P. Abels and Bernard S. Bachrach (Woodbridge, 2001), pp. 87–101, especially p. 96. 40 Luigi Russo, “Il viaggio di Boemondo d’Altavilla in Francia (1106): un riesame”, Archivio storico italiano 163 (2005), 3–42 (repr. and enlarged in Russo, I Normanni del Mezzogiorno, pp. 101–42). Sources: Chronicon ignoti civis Barensis, ed. Ludovico Antonio Muratori (Milan, 1724) (Rerum Italicarum scriptores, 5), col. 155 (s.aa. 1105–6); Hystoria de via et recuperatione Antiochiae atque Ierusolymarum, pp. 135–36; Fulcheri Carnotensis, Historia Hierosolymitana (1095–1127), ed. Heinrich Hagenmeyer (Heidelberg, 1913), pp. 464–67.

206

luigi russo

an adequate number of followers for his ambitious plans against Byzantium, his main foe.41 All of the above took place in the aftermath of the military conquests of Edessa, Antioch, Jerusalem, and Tripoli, in the formative years of the Outremer’s political structures.42 Institutional fluidity promoted the rise of those aristocratic kin groups that turned out to be more dynamic in a highly competitive context like the Latin East in which the chronic lack of knights to be employed against the surrounding enemies – Muslims or Byzantines – was felt with great urgency.43 In this context in the Latin East the most influential Norman household of southern Italy was not a key player. It was therefore precisely this nature of the early Italo-Norman participation, disconnected from wider networks of kinship, that did not allow the Normans to put down roots in the Holy Land. To the contrary, in Capetian France it is clear that the ruling dynasty had adopted a strong commitment to recruitment for the crusades (even though it was impossible for King Philip I to participate himself, because he had been excommunicated by the pope on account of his bigamy), which is demonstrated by the sustained engagement of aristocratic families of the Île-de-France in the crusades throughout the twelfth century (and beyond).44 41

The Ecclesiastical History of Orderic Vitalis, 6:68: “Qui antequam Gallias attingeret legatos suos in Angliam direxerat, et de adventus sui causa in Ausoniam regi mandaverat, et quod ad curiam eius transfretare vellet insinuaverat. At contra providus rex, metuens ne sibi electos milites de dicione sua subtraheret, mandavit ei ne discrimen hibernae navigationis subiret, presertim cum ipse rex in Neustriam ante aximorum celebria transfretaret, ibique satis secum colloqui valere” [Before reaching Gaul he [Bohemond] had sent his envoys to England, to inform the king of why he had come to southern Italy, and indicate that he wished to cross the sea to visit his court. But the prudent king, fearing that he might tempt away his best knights, advised him instead not to risk a winter voyage, particularly as he himself would be crossing to Normandy before Easter and could confer with him there]. On King Henry I, see Judith A. Green, Henry I. King of England and Duke of Normandy (Cambridge, 2009). 42 Alan V. Murray, “Women in the Royal Succession of the Latin Kingdom of Jerusalem (1099–1291)”, in Mächtige Frauen? Königinnen und Fürstinnen im Europäischen Mittelalter (11.–14. Jahrhundert), ed. Claudia Zey (Ostfildern, 2015), pp. 131–62. Note 6 provides a definition of the term Outremer to include Jerusalem, Antioch, Tripoli, and Edessa, p. 134. 43 On this, see Joshua Prawer, The Crusaders’ Kingdom. European Colonialism in the Middle Ages (London, 1972); Thomas S. Asbridge, The Creation of the Principality of Antioch, 1098–1130 (Woodbridge, 2000); Monique Amouroux-Mourad, Le Comté d’Edesse: 1098–1150 (Paris, 1988). 44 For the gathering in Paris (February 1096) of the members of the Capetian royal court to decide on their participation in the crusade, see Guitberti abbatis Sanctae Mariae novigenti historia quae inscribitur Dei Gesta per Francos, ed. Robert Burchard Constantijn Huygens (Turnhout, 1996) (Corpus Christianorum. Cont. Med., 127 A), p. 133. Cf. Marcus Bull, “The Capetian Monarchy and the Early Crusade Movement: Hugh of Vermandois and Louis VII”, Nottingham Medieval Studies 50 (1996), 25–46; see



norman participation in the first crusade 207

All of the above is confirmed by the dynastic events that occurred in the Principality of Antioch founded by Bohemond and where, for at least thirty years, there was a significant presence of Normans from southern Italy among the ruling groups.45 The premature death of Bohemond II in battle (February 1130) marked a clear-cut political and institutional vacuum in favour of Raymond of Poitiers, son of the Duke William IX of Aquitaine, who had participated in the crusade of 1101. Linked to the Norman court of England, Raymond was invited to the Latin East by King Fulk of Jerusalem, who wanted to address the issue of the regency of the Antiochene principality while contributing decisively to the removal of the Italo-Normans from the political entities established in the aftermath of the First Crusade.46 In fact, Raymond married the young Constance in 1136 after a long tug of war with Bohemond II’s widow, Alice of Jerusalem, becoming prince of Antioch (1136–49).47 This paved the way for a new political era which saw the marginalisation of the Italo-Norman families from the Antiochene political scene, a fact attested by their progressive disappearance from the sources, beginning in the 1140s, mirroring the changing balance of political power.48 also James L. Naus, “The French Royal Court and the Memory of the First Crusade”, Nottingham Medieval Studies 55 (2011), 49–78. On Philip’s I excommunication, see Georges Duby, Le Chevalier, la femme, et le prêtre. Le mariage dans la France féodale (Paris, 1981), pp. 7–26, especially pp. 7–10. 45 Luigi Russo, “La fine dell’espansione: i Normanni ad Antiochia (1098–1130)”, Representations of Power at the Mediterranean Borders of Europe (12th–14th centuries), ed. Ingrid Baumgärtner, Mirko Vagnoni, and Megan Welton (Florence, 2014), pp. 121–37. The best source for the Principality of Antioch in the early twelfth century is Galterii cancellarii bella Antiochena, ed. Heinrich Hagenmeyer (Innsbruck, 1896), pp. 61–115. 46 On the relationship between Raymond and the court of King Henry I Beauclerc, see John Oswald Prestwich, “The Military Household of the Norman Kings”, The English Historical Review 96 (1981), 1–37, especially p. 8. For an overview, see Malcolm Barber, The Crusader States (London, 2012), pp. 152–53, 167–69, and Rudolf Hiestand, “Antiochia, Sizilien und das Reich am Ende des 12. Jahrhunderts”, Quellen und Forschungen aus Italienischen Archiven und Bibliotheken 73 (1993), 70–121, especially pp. 83–86. 47 Thomas Asbridge, “Alice of Antioch: A Case Study of Female Power in the Twelfth Century”, The Experience of Crusading, II: Defining the Crusader Kingdom, ed. Peter Edbury and Jonathan Phillips (Cambridge, 2003), pp. 29–47. 48 Jean-Marie Martin, “Les Structures féodales normanno-souabes et la Terre Sainte”, Il Mezzogiorno normanno-svevo e le crociate. Atti delle quattordicesime Giornate normanno-sveve, ed. Giosuè Musca (Bari, 2002), pp. 225–50: “au total, les noms italo-normands semblent se raréfier après 1140 et pratiquement disparaître avant la fin du XIIe siècle”, p. 236. On the Antiochene aristocracy, see the standard work of Claude Cahen, La Syrie du Nord à l’époque des croisades et la principauté franque d’Antioche (Paris, 1940), pp. 534–46; Andrew D. Buck, The Principality of Antioch and its Frontiers in the Twelfth Century (Woodbridge, 2017), pp. 128–63. See also France, “The Normans and Crusading”, p. 97: “From the perspective of Antioch, relationships with Byzantium and Jerusalem were far more important than any sense of being part of a Norman Empire […].”

208

luigi russo

Conclusions To conclude this necessarily brief survey, the recruitment of the ItaloNorman contingent of Bohemond represents a unicum.49 The composition of his retinue, with the lack of participation by large segments of the aristocratic families recently established in Italy and linked to the heirs of Robert Guiscard, his son Roger Borsa, and his brother, Roger I, strongman of southern Italy in the late eleventh century, provides an interesting springboard to reinvestigate the question of the crusaders’ motivations, a popular theme in recent historiography of the crusader movement. Retracing the careers of the principal knights who joined the first-born of Guiscard (except for Tancred), based on the extant patchy sources, there are two considerations that emerge in my opinion: on the one hand, there is the provisional nature of the response to the call for the crusade in Norman Italy, which was the result of individual choices unrelated to broader family strategies often followed by other pilgrims from European regions such as the Île-deFrance or Champagne; on the other hand, there is the discrepancy between the geographical origin of the majority of the departing knights and the area in which Bohemond exercised direct control (i.e. the land of Otranto and central Apulia), which underlined the low relevance of the seigneurial bonds on the choice of Italo-Norman knights in southern Italy to became crusaders in the late eleventh century. Elsewhere, these ties were proved to be very strong.50 Bohemond drew on the bonds of kin relationships as his recruitment strategy, but his position as an outsider did not allow him to extend a call that could convince the leading members of the Hauteville clan, closely linked to his half-brother Roger Borsa and his uncle Roger I, to join him. Nevertheless, the crusade was attractive for a segment of the Italo-Norman aristocracy: this is demonstrated by Roger of Barnesville, who, although not linked to the eldest son of Robert Guiscard, chose to set out to Jerusalem later, and Geoffrey of Monte Scaglioso, who responded to Bohemond’s appeal even though he was the heir to his ancestral lands. However, the story of Robert of Anzi is perhaps the most emblematic of all. Formerly an enemy of Bohemond, Robert chose to follow him to the Holy Land. He then broke off to join Lorraine’s army led by the House of Bouillon, and finally became a feudal lord in the Kingdom of Jerusalem, thus demonstrating the precarious nature of his connection with Guiscard’s first-born. 49

On the peculiarity of the Normans, see Luigi Russo, “Bad Crusaders? The Normans of Southern Italy and the Crusading Movement in the Twelfth century”, Anglo-Norman Studies 38 (2016), 169–80. 50 Riley-Smith, The First Crusaders, p. 105: “Although many different factors must have played a part in the response of each individual, the decision to crusade usually took place within the contexts provided by lordship and especially kindred.”



norman participation in the first crusade 209

The choice to take the Cross by some among the Italo-Norman aristocracy must therefore be related to the particular context of southern Italy in the late eleventh century. On the one hand, it reflects the establishment of the leading Norman kin groups in southern Italy who did not have significant ties with Bohemond, himself distanced from his father’s heritage a few years earlier.51 On the other hand, for those who responded to the call, the choice to set out for the liberation of the Holy Sepulchre was an individual choice in which personal ambition and economic gain were inextricably tied. In sum, we can conclude that it is extremely difficult to provide an all-encompassing reason why some knights chose to participate in the First Crusade – a decision that would influence their careers as much as it would lead some to their deaths.

51

Russo, Oblio e memoria di Boemondo d’Altavilla; Pietro Dalena, “‘Guiscardi coniux Alberada’: donne e potere nel clan del Guiscardo”, in Roberto il Guiscardo tra Europa, Oriente e Mezzogiorno, ed. Cosimo Damiano Fonseca (Galatina, 1990), pp. 157–80.

9

Strategy, the Norman Conquest of Southern Italy, and the First Crusade* Daniel P. Franke

In the early 1200s the chronicler Ibn al-Athir, surveying the events of the past century, stated that the Norman conquest of southern Italy was part of a coordinated counter-offensive against Islam.1 In al-Athir’s telling, what became the First Crusade was initially meant to help Roger of Sicily pursue further conquests, preferably an invasion of Ifriqiya, modern-day Tunisia. But Roger found that plan ridiculous, supposedly indicating his opinion through loud flatulence. “If you are determined to wage holy war on the Muslims,” he said, “then the best way is to conquer Jerusalem. You will free it from their hands and have glory.” The genesis of the First Crusade, in this telling, lay in the Norman conquest of Sicily and southern Italy. Whether this anecdote has any grounding in historical reality or not, it represents a widely held belief, dating to the medieval period, that the Norman conquest of southern Italy was integral to the First Crusade. Whether in terms of military strategy, personnel, geopolitical objectives, the movement of peoples, or the ideology of holy war, the activities of the Normans in Italy between 1000 and 1091 and in the eastern Mediterranean between 1196 and 1119 are assumed to have some kind of connection.

* I would like to thank Peter W. Sposato, University of Indiana Kokomo, for his

feedback and critiques of my lines of argument. I would also like to specially thank Francesca Petrizzo, Leverhume Fellow at Università degli Studi di Roma Tor Vegata for similarly giving me critical feedback, and for allowing me to read her unpublished dissertation “Band of Brothers: Kin Dynamics of the Hautevilles and Other Normans in Southern Italy and Syria, c.1030–c.1140” (University of Leeds, 2018). Any errors of interpretation and fact are my own. 1 Ibn al-Athir, al-Kamil fi’l-Ta’rikh, trans. D. S. Richards (Aldershot, 2005), opening chapter, pp. 13–14.

212

daniel p. franke

Historians have long puzzled over these issues, and their answers have reflected the state of the scholarship at any given moment. Generations of historians, such as David Douglas and Matthew Bennett, have tended to see the southern Norman contingents of the First Crusade as part of a larger “Normanitas” that sought to conquer the world, or large parts of it. Charles Homer Haskins, surveying the history of the Normans of the South, stated that their story “would be conclusive proof of the creative power of the Norman genius for conquest and administration”.2 Other specialists such as Graham Loud, Elisabeth Van Houts, David Bates, and Kenneth Baxter Wolf have de-emphasised such abstract organising principles and, instead, focused on the evolving and complex relationship of the southern Normans with the Reform movement in the Catholic Church and Normans’ own sense of identity.3 Others, such as Paul Chevedden and, to a lesser extent, Andrew Latham, have viewed the Norman conquest and the First Crusade as part of the same larger, coordinated strategy of counter-offensives against Mediterranean Muslim powers, with the various armies being ultimately directed by the “central intelligence” of a nascent papal monarchy that had the capacity to wage war.4 This line of enquiry goes back at least to Augustin Fliche, who in 1938 postulated that the Normans filled a “gap” in the defence of “Christendom” from Islam, since the German empire could no longer be relied upon.5 At the heart of these enquiries lie complex questions about the theory and practice of warfare in the Middle Ages, including the definition of “strategy” and the extent to which we can recover a sense of medieval commanders’ decision-making processes. Recent work by Georgios Theotokis, Paul Brown, John Hosler, and Lawrence Freedman has begun to reshape the way we approach these issues.6 When this is combined with new studies of Norman family networks and state formation, 2

David Charles Douglas, The Norman Achievement 1050–1100 (Berkeley, 1969); Matthew Bennett, “Norman Conquests: A Strategy for World Domination?”, Journal of Medieval Military History 15 (2017), 91–102; Charles Homer Haskins, The Normans in European History (Boston, 1915), p. 192. 3 Graham A. Loud, The Age of Robert Guiscard: Southern Italy and the Norman Conquest (Harlow, 2000); Elisabeth van Houts, The Normans in Europe (Manchester, 2000); Kenneth Baxter Wolf, Making History: The Normans and Their Historians in EleventhCentury Italy (Philadelphia, 1995); David Bates, The Normans and Empire (Oxford, 2013). 4 See Paul Chevedden, “A Crusade from the First: The Norman Conquest of Islamic Sicily, 1060–1091”, Al-Masaq 22 (2010), 191–225; Chevedden, “Pope Urban II and the Ideology of the Crusades”, in The Crusader World, ed. Adrian Boas (London, 2016), pp. 7–53; Andrew Latham, Theorizing Medieval Geopolitics: War and World Order in the Age of the Crusades (New York, 2012), especially chapter 4, “Religious War: The Wars of the Corporate-Sovereign Church”, pp. 106–29. 5 Augustin Fliche, “Les origins de l’action de la papauté en vue de la croisade”, Revue d’histoire ecclésiastique 34 (1938), 765–75. 6 Georgios Theotokis, The Norman Campaigns in the Balkans, 1081, 1108 (Woodbridge, 2014); Paul Brown, Mercenaries to Conquerors: Norman Warfare in the Eleventh- and TwelfthCentury Mediterranean (Barnsley, South Yorkshire, 2016); John D. Hosler, “Reframing the



strategy, conquest, and the first crusade 213

we have powerful analytical tools for reassessing the question of strategy, the Norman conquest of southern Italy, and the First Crusade. Taking into account the conquest of Sicily, Bohemond of Taranto’s career, and finally the creation of the Kingdom of Sicily, it is argued here that a broader definition of strategy as the often improvised personal or corporate use of military force to solve problems is necessary in order to understand what happened in the late eleventh century. While the papacy’s conception of “crusade” was certainly impacted upon by the Norman Conquest of Sicily, successive popes did not base their crusade strategy and doctrine on the treaty of Melfi in 1059, nor did Norman lords regard fighting Muslim powers as significantly more important than consolidating their power against rival Norman, Lombard, and Byzantine claimants. Further, southern Norman participation in the First Crusade did not reflect a larger Norman strategic plan but, rather, the personal strategic goals of the main leadership, particularly Bohemond of Taranto. This is borne out through an examination of Bohemond’s strategic thought as far as we are able to recover it. Not until the regency of Adelaide and the rule of her son King Roger II in the 1120s and 1130s do we see the clear shape of a defined Norman strategy, demonstrating that “strategy” is best understood, in Moltke’s classic formula, as a system of expedients, and that strategic thought often follows, rather than shapes, the reality that it seeks to control. Strategy The challenge to an enquiry on strategy and the crusades is setting effective (and persuasive) parameters of what is meant by strategy and where to look for it. Classical definitions of strategy include Clausewitz’s terse dictum in Book 3 Chapter 1 of On War: “Strategy is the use of the engagement for the purpose of the war.”7 Colin Gray has defined it somewhat more extensively as “neither policy nor armed combat; rather it is the bridge between them … The strategist must relate military power (strategic effect) to the goals of policy … It is essentially different from military skill or political competence.” Gray goes on to say that “strategy is perilously complex by its very nature. Every element or dimension can impact all others. The nature of strategy is constant throughout history but its character continually evolves with changes in technology, society, and political ideas.”8 Conversation on Medieval Military Strategy”, The Journal of Medieval Military History 16 (2018), 189–206; Lawrence Freedman, Strategy: A History (Oxford, 2013). 7 Carl von Clausewitz, On War, ed. and trans. Michael Howard and Peter Paret (Princeton, 1976) p. 177. 8 Colin S. Gray, “Why Strategy Is Difficult”, in Strategy and History: Essays on Theory and Practice (London, 2006), pp. 74–80; 77.

214

daniel p. franke

It does not take much imagination to grasp the problems with applying Clausewitz’s, and more particularly strategic theorists like Gray’s, definitions to the issue of the southern Normans and the First Crusade. Words and phrases such as “goals of policy” and “the purpose of the war” denote the very modern relationship between war and the state that Clausewitz took for granted, but which we would have to individualise and “miniaturise” to the disparate collection of Norman lords who effected the conquest of southern Italy. Further, since in the First Crusade the papacy had no direct control of any of the armies involved, it does not at first glance make sense to discuss papal strategy (or even grand strategy). It is no wonder that modern theorists and scholars have frequently looked at medieval warfare and thrown up their hands in frustration. Fortunately, the time is long past when medieval strategy was considered to be the unicorn of military history and it is now widely accepted that medieval generals and political leaders thought and operated in terms that were “strategic”, even though the word itself may not have been used.9 As John Hosler has written, medieval strategic thought was original and did not rely on the often doctrine-bound assumptions of modern analysts.10 Quoting John Gooch, Hosler defines strategy as “a method of solving problems by the application of military force”.11 This definition does not require a set of clearly articulated strategic principles or a “blueprint”, something which Geoffrey Parker has also suggested applies to Habsburg Spain, and it also accords with classical analyses of strategy like Peter Paret’s: “Strategic thought is inevitably highly pragmatic … The history of strategic thought is a history not of pure but of applied reason.”12 As Hosler has argued: We should immediately dispense with the modernist demand for classicalstyle manuals and stop asking the question, “Were there medieval authors who wrote texts like De re militari?” More pertinent questions include “Were

9

For a beginning inquiry in this direction, though lacking any informed appreciation of medieval military history, see Beatrice Heuser, “Strategy Before the Word: Ancient Wisdom for the Modern World”, The RUSI Journal 155 (2010), 36–43. 10 Hosler, “Reframing the Conversation”, p. 190. 11 Hosler, “Reframing the Conversation”, p. 190; John Gooch, “History and the Nature of Strategy”, in The Past as Prologue: The Importance of History to the Military Profession, ed. Williamson Murray and Richard Hart Sinnreich (Cambridge, 2006), pp. 133–50; 139. Gooch goes on to say: “Like all historical subjects, its study requires the reconstruction of events and the re-creation of processes. It also requires that historians determine as accurately as possible the circumstances in which events happen and exert their influence to shape outcomes.” 12 Geoffrey Parker, “The Making of Strategy in Habsburg Spain: Philip II’s ‘Bid for Mastery,’ 1556–1598”, in The Making of Strategy: Rulers, States, and War, ed. Williamson Murray et al. (Cambridge, 1994), pp. 115–50; 115; Peter Paret (ed.), Makers of Modern Strategy from Machiavelli to the Nuclear Age (Princeton, 1986), p. 3.



strategy, conquest, and the first crusade 215 there medieval authors who wrote about strategy?” followed by, ‘What sort of 13 texts did they write?” and “How did they intend to transmit their ideas?”

This kind of un-doctrinaire approach to strategy is paralleled by more recent studies, particularly Lawrence Freedman’s massive 2013 tome. A core contention of the book is that, rather than a grand plan that coordinates resources and foresees and directs complex events, strategy is better understood as a short-term solution to a problem, that will in due course be replaced when new problems emerge.14 More recently, Matthew Bennett has argued that “strategy” can be effectively applied to individual and collective actions and decisions, and that the “essence of medieval strategy” was embodied in Norman leadership as a “combination of vision, ideology, family, and diplomacy with the physical elements of men and their ships and fortresses which enabled relatively small groups of warriors to conquer territories and set up ruling dynasties in a truly remarkable way from around 1050 onwards.”15 The Conquests, the Church, and the Crusade When we turn to the narrative of the Normans’ conquest of the South, it quickly becomes apparent that, fractious and distributed as they were, there was some kind of central idea animating and connecting their activities, occasionally even coordinating them.16 It is questionable as to whether there was actually a strategy in the military sense, although Matthew Bennett has argued that there was.17 But just as important for understanding the strategic connections between the Norman conquest and the First Crusade is the role of the Catholic Church as architect of crusading – what today would be called in the United States “grand strategy” and in other countries “national strategy”. Scholars are broadly agreed as to what “grand strategy” is: “the intellectual architecture that lends structure to foreign policy; it is the logic that helps states navigate a complex and dangerous world”, according to Hal Brands. According to Hew Strachan, “Grand strategy … 13

Hosler, “Reframing the Conversation”, p. 193. See Freedman, Strategy, particularly ch. 16 “The Revolution in Military Affairs”, ch. 17 “The Myth of the Master Strategist”, and ch. 38 “Stories and Scripts”. Typical is one of Freedman’s concluding remarks that “much strategy is about getting to the next stage rather than some ultimate destination. Rather than think of strategy as a three-act play, it is better to think of it as a soap opera …” (p. 628). 15 Bennett, “Norman Conquests”, p. 102. 16 Recent research by Francesca Petrizzo confirms this interpretation, emphasising the diversity of policies and agendas among the disparate Norman princes, with the main Hauteville family providing the only unifying force: Petrizzo, “Band of Brothers”, p. 65 (see n. 1) 17 Bennett, “Norman Conquests”, n. 12. 14

216

daniel p. franke

can entertain ambitions and goals which are more visionary and aspirational than pragmatic and immediate. It is as much a way of thinking as a way of doing.”18 These are both useful definitions for assessing the influence of the Norman campaigns on papal strategic thought. Arguments for a Norman influence on Urban II’s crusade go back to at least Erdmann’s seminal study.19 Beyond “influence”, arguments for a papal grand strategy are perhaps most prominently found in Paul Chevedden’s writings and in Andrew Latham’s study of medieval geopolitics.20 Chevedden has maintained for some time now that what we call the First Crusade was the logical development from a process that began with the Norman conquest of Sicily, or more specifically with the treaty of Melfi in 1059: This alliance was the first in a series of bilateral partnerships that linked the religious and secular spheres of Western Christendom in a common enterprise: the reconquest of former Christian territory and the assimilation of this territory back into Christendom. From 1059 onward, the war with Islam 21 would be approached in new and different ways.

In combatting what is at times a strawman (crusade historians are a bit more nuanced than he gives them credit for, and recently it has been strongly argued that the crusade indulgence actually was an innovation), Chevedden does good service in reminding historians that history is as much the passage of time as a moment in time – the long history of sanctified and redemptive violence versus the preaching of a single sermon at Clermont. There are, however, two problems with Chevedden’s argument that affect our topic here.22 First, the treaty of Melfi was not really about Sicily. It was about the lands that Guiscard had already conquered and which were fractious and unsettled at best, in their loyalty both to him and to the reform movement. Graham Loud makes this point very clearly: 18

Hal Brands, What Good is Grand Strategy? Power and Purpose in American Statecraft from Harry S. Truman to George W. Bush (Ithaca, 2014), p. 1; Hew Strachan, The Direction of War: Contemporary Strategy in Historical Perspective (Cambridge, 2012), p. 235. 19 Carl Erdmann, The Origin of the Idea of Crusade, trans. Marshall W. Baldwin and Walter Goffart (Princeton, 1977), pp. 120–23. 20 See n. 3, Chevedden, “A Crusade from the First”, and “Pope Urban II and the Ideology of the Crusades”; Latham, Theorizing Medieval Geopolitics, ch. 4. Latham follows Riley-Smith in not seeing the Norman campaigns as true crusades, but emphasises those elements that portray Urban II as promoting a “war of liberation”: Theorizing Medieval Geopolitics, p. 124. 21 Chevedden, “A Crusade from the First”, p. 214. 22 Chevedden, “A Crusade from the First”, pp. 192–201; for the crusade indulgence, see Ane L. Bysted, The Crusade Indulgence: Spiritual Rewards and the Theology of the Crusades, c.1095–1216 (Leiden, 2014).



strategy, conquest, and the first crusade 217 It can be seen from this that while there was a defensive element, that Robert would respect papal territory, the real stress in the oath was the military help that he would in future render to his overlord; helping the pope “to hold the Roman papacy securely”, and ensuring that after his death he would, if required, help the reformers at Rome … to elect a pope of their choosing. Such a provision was directed against the Roman nobility, who were still, despite their defeat in 1058–9, a very real threat to the position of the reformers … This issue was of crucial importance to the very survival of the reformed papacy, and was far more significant to them than any powers 23 that the popes might seek to exercise over southern Italy …

Chevedden overplays the importance of Sicily in the document, appearing as it does in one line – “dux Apulie et Calabrie et utroque subveniente futurus Sicilie” – and mistakes the rest of the document as a blueprint for reconquest in the “war against Islam”, rather than an attempt by the papacy to lock down what was a very unstable situation in southern Italy. Robert swore, after all, to protect the Church “contra omnes homines”, not from “Saracens” specifically.24 Guiscard and his brother Roger were themselves also concerned about the overall situation and stood to profit from this alliance with the papacy. Indeed, the chronicler Geoffrey Malaterra describes the years 1058 and 1059 as being particularly chaotic for the brothers, who faced a rupture between themselves, the outbreak of famine and epidemics, and the revolt of the Calabrians who massacred the Norman garrison of Nicastro. Before the treaty of Melfi, which was concluded in the summer of 1059, Roger had managed to win a major victory over an insurgent army at San Martino, after which Calabria “while not yet completely obedient to [him], nonetheless trembled from his proximity and did not dare provoke him as often as before”.25 Guiscard, meanwhile, was focused on the capture of Reggio de Calabria, which he accomplished in the autumn of 1059. Malaterra writes that it was the capture of Reggio that led to Guiscard becoming duke, omitting any mention of Melfi – perhaps from a determination to see the title deriving from conquest, and the treaty simply amounting to papal recognition of a fait accompli.26 The second problem with Chevedden’s argument is that the expression of crusading ideas that he sees as the origin point for crusading – the treaty of Melfi in 1059 and the subsequent Norman invasion of Sicily in 1060–61 – was not new either. The goal liberanda Christianitas, the use 23

Loud, The Age of Robert Guiscard, p. 189. The Treaty of Melfi is given in full in Loud, The Age of Robert Guiscard, pp. 188–89, and in Le Liber Censuum de l’Église Romaine, ed. M. Paul Fabre (Paris, 1889), vol. 1, no. 163, p. 422. 25 Geoffrey Malaterra, The Deeds of Count Roger of Calabria and Sicily and of his Brother Duke Robert Guiscard, trans. Kenneth Baxter Wolf (Ann Arbor, 2005), I:20–32. 26 Malaterra, I:34, capture of Reggio, and I:35, assumption of title of duke. 24

218

daniel p. franke

of sacred banners, and an early form of crusading indulgence all appear in Leo IX’s ill-fated campaign against the Normans in 1053, culminating at the complete papal defeat at Civitate.27 The German troops that Leo had recruited for this campaign were apparently promised absolution for their sins and relief from penance, according to Amatus of Montecassino. The Normans, attempting to negotiate with the pope, promised to render the church tax every year, and appealed to the imperial banner they had received from Henry III, by way of legitimising their actions, although their negotiations fell on deaf ears.28 Writing to the Byzantine Emperor Constantine Monomachos after Guiscard’s cavalry had destroyed the papal army, Leo urged him to help restore “the patrimony of the Roman Church” and assured the emperor that his intention to pursue “the liberation of Christendom” was undiminished.29 The papacy’s attitude toward the Normans changed once the treaty of Melfi had been concluded. Theotokis points out that from the pontificate of Alexander II (1061–73), with whom Guiscard and the other Norman princes had to deal during the first phase of the conquest of Sicily, “Rome now favoured the fomenting of divisions among [Richard of Capua and Robert Guiscard] as the most efficient way to exert some sort of control over the Norman lands”.30 And of course Gregory VII’s proposed crusade against Guiscard in 1074 has been widely discussed.31 The subsequent history of the conquest of Sicily only underscores how difficult it is to claim that that process was a kind of blueprint for what followed in 1095–99.32 27

Erdmann, The Origin of the Idea of Crusade, pp. 120–25. Amatus, III:39 for negotiations; III:23 and 40 for promises of absolution and remission. See also Loud, The Age of Robert Guiscard, pp. 118–21. 29 The most recent comprehensive account with the sources and interpretations of the campaign and battle of Civitate is Brown, Mercenaries to Conquerors, pp. 62–69. For Leo IX’s letter: “De restituendis ecclesiae Romanae patrimoniis hortatur”, in R. Jaffe, Regesta Pontificum Romanorum, 2nd edn, vol. 1, #4333; “ab hac nostra intentione liberandae Christianitatis non deficiemus”, in Patrologiae Latinae 143, col. 779b, Leo IX to Constantine Monomachos, cols. 777–81. 30 Theotokis, Norman Campaigns, p. 138. 31 See H. E. J. Cowdrey, “Pope Gregory VII’s ‘Crusading’ Plans of 1074”, in Outremer: Studies in the History of the Crusading Kingdom of Jerusalem Presented to Joshua Prawer, ed. Benjamin Z. Kedar, H. E. Meyer, and R. C. Smail (Jerusalem, 1982), pp. 27–40. 32 See Amatus, Book V and Malaterra, Book II; also Brown, Mercenaries to Conquerors, pp. 86–114, particularly after Guiscard’s death in 1085, p. 109. See Alex Metcalfe on the complexity of ethnic and linguistic identity in Sicily both before, during, and after the Norman conquest, Muslims and Christians in Norman Sicily: Arabic Speakers and the End of Islam (London, 2003). Nicholas Morton has argued that the First Crusade armies gradually came to appreciate the complexity of Levantine society, a kind of complexity which the southern Normans had much experience in navigating, but which would have made the creation and application of strategy very difficult.. See Nicholas Morton, Encountering Islam on the First Crusade (Cambridge, 2017). 28



strategy, conquest, and the first crusade 219

Inasmuch as the southern Normans influenced papal grand strategy regarding military operations against Muslim regimes and the freedom of the Christian Church, they did so first and foremost because the Normans themselves threatened that freedom, and only secondarily because they offered the possibility of a new front against Muslim powers. The record of papal–Norman interactions between 1053 and 1096 clearly demonstrates that it is wide of the mark to see the papacy as doing anything more than making lemonade out of lemons (if the expression can be excused) in its dealings with the Hauteville clan. Bohemond and the First Crusade Recruitment for the crusade in Norman dominions should also give us pause in assuming a close strategic connection between the two, since there is precious little evidence that there was any kind of coordination between the various lords and the papacy (chapter 8 in this volume). In fact, the opposite seems to have been the case, as Bohemond’s departure on crusade seriously compromised his uncle’s and half-brother’s siege of Amalfi. According to the Gesta Francorum, the preaching and mobilisation in parts of France and the Empire had made no impression on Bohemond until the actual arrival of Frankish forces, at which point, “inspired by the Holy Ghost”, he decided on impulse to join them. Malaterra also supports the idea that Bohemond’s decision was on impulse: he “was always looking for ways to subject [the Byzantine Empire] to his authority … The war-oriented young men of the duke’s and count’s armies, attracted by the novelty of the thing … hurried eagerly to Bohemond.” So, with him went most of Duke Roger Borsa and Count Roger’s cavalry, as well as various other lords. Count Roger had no choice but to lift the siege.33 At this point one should probably be cautious about giving Bohemond too much credit for a coherent strategy beyond a general goal of carving out an independent principality somewhere, somehow, at some point in the East. Yet, following Freedman’s analytical framework, there can be no doubt that Bohemond was pursuing a strategy of his own, even if it was one of personal aggrandisement at the expense of what modern analysts would identify as a proto-state interest, that is, the governments of his uncle Count Roger I of Sicily or his half-brother Roger Borsa Duke of Apulia. Theotokis argues convincingly that the crusade presented “a unique opportunity to escape the relentless pressure” that his half-brother Roger Borsa had subjected him to since Guiscard’s death in 1085, when as the older brother

33

Gesta Francorum et aliorum Hierosolimitanorum, ed. Rosalind Hill (London, 1967), pp. 7–8; “Mox Sancto commotus Spiritu”; Malaterra, 4:24.

220

daniel p. franke

from a previous marriage Bohemond could not be expected to simply accept Borsa’s claim to be Guiscard’s successor.34 With opportunity came the new problem of how to repair what would seem to have been the irreparably damaged relationship with the Byzantine Empire. This necessitated a new strategy, of which the Gesta Francorum and Ralph of Caen’s Gesta Tancredi emphasised different parts. The Gesta Francorum claims that Bohemond emphasised restraint, discipline, and forbearance in his dealings with Byzantines. Upon reaching Macedonia, he called a council of war and ordered that those in his army be “good and courteous”, since they were traversing Christian lands.35 Ralph of Caen instead suggests that Bohemond was overwhelmingly concerned with projecting strength, since he had no confidence that the Byzantines would believe that he came with peaceful intentions. This concern was seemingly justified at the Vardar river, when Byzantine forces attempted to oppose Tancred’s crossing directly while attacking Bohemond’s troops when he and Tancred were separated by the river. Both Ralph and the Gesta agree that Tancred’s energetic response saved the day for the Normans.36 What Ralph does not record is the massacre of a castle of “heretics” before the Vardar, which apparently was the cause of the Greek attack.37 In realising his goals of a principality in the eastern Mediterranean, Bohemond was not off to a promising start. His strategy, however, took concrete shape with his oath to Alexios, given in full in Anna Komnene’s Alexiad: in it, Bohemond, while having to promise to be Alexios’ man, secured the promise of the lands that he wound up taking and holding without the Byzantines anyway – Antioch and many of the surrounding towns. Realising this goal would now depend on his battlefield abilities, which had already been tested in southern Italy and the Balkans.38 This is not to suggest that strategy is simply the sum of one’s tactics and operations, but the attainment of strategic goals does depend on successfully employing military force at the tactical level. So, Bohemond’s battlefield performance must also be taken into consideration. Those who have made a study of the First Crusade have long tended to praise Bohemond and his army as being “the most experienced and the most suited for what lay ahead”.39 In framing his discussion of Bohemond, John France says “Bohemond alone had experience of eastern 34

Theotokis, Norman Campaigns, p. 185. Gesta Francorum, 8, “omes ut boni et humiles essent; et ne depredarentur terram istam quia Christianorum erat …”. 36 The Gesta Tancredi of Ralph of Caen: A History of the Normans on the First Crusade, ed. Bernard S. Bachrach and David S. Bachrach (Aldershot, 2005), chs. 2–7, 23–28. 37 Gesta 7–8: “intravimus Palagoniam, in qua erat quoddam hereticorum castrum”. 38 See Theotokis, passim. For the treaty with Alexios, see Anna Komnene, The Alexiad, trans. E. R. A. Sewter, rev. Peter Frankopan (New York, 2009), pp. 385–96. 39 Theotokis, Norman Campaigns, p. 187. 35



strategy, conquest, and the first crusade 221

affairs and had commanded a big army in a major campaign and in open battle”.40 Yet Bohemond’s performance during the First Crusade, particularly early on at Doryleum, suggests someone whose grasp of tactics and operations was at worst very uneven, and at best needed to mature over time. This is not the place to re-fight Doryleum, which has been expertly analysed by France, Theotokis, and others, but it is noteworthy that the vanguard, where Bohemond, Tancred, and Robert of Normandy were, was beyond immediate supporting distance of the main army – a fundamental aspect of operational design that most medieval militaries, whether tenthcentury Byzantines or twelfth-century Germans, always attended to.41 It was also not the last time Bohemond would be caught by surprise with inadequate reconnaissance. By the time the armies had reached Antioch, his performance had improved to the point that he defeated Ridwan of Aleppo handily by taking the initiative and using the terrain to his advantage. If he was not already regarded as such, this victory established him “as the true military leader of the First Crusade”,42 and demonstrated that he had the tactical ability to secure larger strategic objectives, even if it was inconsistently manifested. A final point to consider in assessing Bohemond’s role as a strategic linchpin between the southern Normans and the First Crusade is the history of the principality of Antioch in the early twelfth century – a sort of coming full circle from the question of family loyalties and group identity with which this chapter began. Despite frequent assumptions to the contrary, it is in fact quite difficult to find any indication that the Norman campaigns in the eastern Mediterranean amounted to some kind of continuation of the conquest of southern Italy, or even Sicily. These conclusions are broadly supported by what one might term the “Asbridge School” of crusades studies, comprised of major studies of the principality by Thomas Asbridge and his former student, Andrew Buck.43 40

John France, Victory in the East: A military history of the First Crusade (Cambridge, 1994), p. 84. 41 For example, the “Strategy” treatise, ch. 20, or the “Campaign Organization” treatise, ch. 7, in George Dennis’s Three Byzantine Military Treatises (Washington DC, 1985), discuss guarding against surprise attacks while marching in enemy territory, the latter advising how to split the army into two mutually supporting parts in the event the terrain was too restrictive to manoeuvre. During the Third Crusade, Barbarossa’s vanguard proved crucial in absorbing initial Seljuk probes and developing the battle space. See Historia de Expeditione Friderici, manoeuvres in Anatolia on 14 May 1190, in Quellen zur Geschichte des Kreuzzuges Kaiser Friedrichs I., ed. A. Chroust, MGH SRG n.s. 5 (Munich: MGH, 1989), p. 81. 42 Theotokis, Norman Campaigns, p. 193. 43 Thomas S. Asbridge, The Creation of the Principality of Antioch 1098–1130 (Woodbridge, 2000); Andrew D. Buck, The Principality of Antioch and its Frontiers in the Twelfth Century (Woodbridge, 2017).

222

daniel p. franke

Asbridge reconstructs in considerable detail the complex manoeuvres that resulted in the creation of the principality. These are far too complicated to recount here, but the key point is that, aside from his guile in securing possession of Antioch itself, Bohemond often had precious little to do with the actual subjugation of the surrounding land to Frankish control – this was shared by major crusade leaders such as Baldwin and Tancred, as well as Robert of Normandy and host of lesser counts. Only upon their leaving the region to continue to Jerusalem or elsewhere did Bohemond assume control of many of these areas, and he had mixed success in expanding his power base.44 In any case, his success was short lived, as he was captured in August 1100 by Turkish forces in a battle that Ralph of Caen says “he never should have entered”.45 Four years later, Bohemond was back, only to suffer a disastrous defeat at Harran in May 1104, where he and Baldwin of Edessa were caught unarmed and unprepared, with no scouts deployed, by Muslim forces from Mosul; apparently only Tancred had taken proper precautions.46 After this disaster, Bohemond returned to Apulia to organise troops and money for another campaign.47 He brought more than simply military plans with him, however: while in prison he had come to the realisation that biblical exegetes had been misinterpreting the famous dream of Nebuchadnezzar in the Book of Daniel. The stone that smashes the statue in the king’s dream was not Christ, as was often thought but, rather, the warriors of the First Crusade. As Jay Rubenstein has recently argued, it was an idea that wound up transforming the consciousness of western Europe, and, one could say, the matrix from which grand strategy was produced.48 His preaching tour has been discussed by Rubenstein, and Theotokis has convincingly argued that Bohemond’s plan to launch his new crusade army into the Balkans, rather than back into his principality, was a shrewd appreciation of the strategic challenges he faced. If the goal was to halt Byzantine attempts to retake Antioch and its surrounding cities, striking at the heart of the Byzantine Empire was much more likely to achieve results than expending his forces in the cockpit of the Levant.49 Yet Theotokis

44

Asbridge, Creation of the Principality, ch. 1, “The Birth of the Principality”. Gesta Tancredi, ch. 141, p. 157. 46 Gesta Tancredi, chs. 148–49, pp. 164–65. 47 Fulcher of Chartres, A History of the Expedition to Jerusalem 1095–1127, trans. Frances Rita Ryan (New York, 1969), Book II:26–27. 48 See Jay Rubenstein, Nebuchadnezzar’s Dream: The Crusades, Apocalyptic Prophecy, and the End of History (Oxford, 2018), pp. 6 and 7–20. 49 Jay Rubenstein, “The Deeds of Bohemond: Reform, Propaganda, and the History of the First Crusade”, Viator 47 (2016), 113–35; Georgios Theotokis, “Bohemond of Taranto’s 1107–8 Campaign in Byzantine Illyria – Can It Be Viewed as a Crusade?” Rosetta 11 (2012), 72–81. 45



strategy, conquest, and the first crusade 223

also supports a contention of the present chapter, that it is almost certain that Bohemond’s crusade against Christian Byzantium, sanctioned by Pope Paschal II and so debated by historians, was a continuation of previous papal policy, specifically Gregory VII’s 1074 call for aid against Guiscard and his 1080 call for intervention in Byzantium, both of which involved sacred banners and the absolution of sins.50 Inasmuch as there was a connection between the papacy’s crusading ideas and the Normans, the Normans were a means to develop ideas that were already in play. Conclusion: Roger II and After In terms of the crusade, the Normans were the focus of papal strategy for holy war before they became the exponents of it, and it seems unlikely that they contributed greatly to the development of the crusading idea or crusading strategy, beyond being unreliable sons of the Church who were proficient in the use of violence. In this, Carl Erdmann’s words are still relevant, if somewhat flippant: “The historic role of the Normans is equally singular: the very people against whom the papal crusade was first directed shortly became the principal exponents of the crusading idea. In them, a crude passion for bullying seems to have mingled in a most peculiar way with religious zeal.”51 The Norman conquest of the South and the subsequent southern Norman participation in the First Crusade demonstrate clearly Lawrence Freedman’s contention that strategy tends to be far more short term and contingent than military historians, and military personnel, like to admit.52 Further, the limitations of communications technology and infrastructure being what they were, strategic entropy was all but inevitable before the early modern age, and military historians need to be particularly careful of extrapolating coherent institutionalised strategies from, as Moltke said, “a system of expedients”.53 Such strategies tend to follow events, not shape them. It should therefore come as no surprise that it is in the life and reign of Roger II, king of Sicily from 1130 to 1154, that we can see the assertion of a coherent, somewhat unified strategy for the Norman Mediterranean, although one that did not deliberately make a target of the Byzantine Empire. A powerful impetus toward the formulation of this modern-looking strategy seems to have been Roger’s mother Adelaide, who had, during Roger’s minority, established herself as a formidable diplomat and leader with a breadth of 50

Theotokis, “Bohemond”, p. 80. Erdmann, The Origin of the Idea of Crusade, p. 125. 52 See note 15. 53 Daniel J. Hughes (ed.), Moltke on the Art of War: Selected Writings (New York, 1993), p. 124. 51

224

daniel p. franke

vision that included the eastern Mediterranean and a shrewd avoidance of the internecine warfare that plagued much of Roger Borsa’s rule.54 In the process of unifying southern Italy, Roger II methodically seized any holdings in southern Italy to which the Normans of Antioch had connections – Taranto, Otranto, and Brindisi among others.55 The weakness of the papacy also helped to create the framework for unified strategy: in the privilege of 27 September 1130, Anacletus II gave the young king the kind of freedom of action that Guiscard, Borsa, and others had longed for – choice of successor to the throne, extensive control over church appointments, and the pope accountable for upholding the terms of the treaty.56 An examination of the southern Normans and the First Crusade in terms of strategy reveals a complex situation that defies attempts to derive easy answers and satisfying meta-narratives from it. If anything, as a case study of strategy it demonstrates the validity of much recent scholarship on the subject that seeks to complicate the habitual (and in the United States the standard) pedagogy of strategy as a neat, overarching paradigm within which one harmonises the ends of conflict with the methods and resources by which one pursues that conflict, and where the failure to do so will likely result in military defeat. Instead, what emerges from the papacy, the crusade, and the Normans of the South is something more akin to Freedman’s portrayal of strategy as short term, oriented toward solving particular problems and in constant need of revision or creation in the face of new problems. Such strategy was the province of individual leaders, without, as Hosler argues, there being any requirement that the leader’s strategy be committed to writing. Simply belonging to the same kin group or religious group did not guarantee a unified goal or method to achieve that goal. And, as so often in history, the ultimate emergence of a unified strategy within a medieval state, as eventually happened in the Kingdom of Sicily, did not mean that it was inevitable or not heavily dependent on battlefield outcomes and personal leadership. As for the papacy, we should be careful to apply these same lessons to the evolution of crusade doctrine and practice, and we should avoid assuming that it is a truth universally acknowledged that a corporate religious body in need of a military must be in possession of a modern strategy. The Normans of the South render this simple picture far more complicated.

54

Francesca Petrizzo, “Band of Brothers”, pp. 138–43. For these campaigns see Alexander of Telese’s account of his campaigns in “The History of King Roger”, in Roger II and the Making of the Kingdom of Sicily, ed. Graham A. Loud (Manchester, 2012), pp. 69–78. 56 Hubert Houben, Roger II of Sicily: A Ruler Between East and West (Cambridge, 2002), pp. 52–3; and see also his discussion “Mediterranean Policy”, pp. 76–86. 55

10

Disaster in the Delta? Sicilian support for the Crusades and the Siege of Alexandria, 1174*

Michael S. Fulton

Following the considerable contributions of certain south-Italian Normans to the success of the First Crusade (see chapter 8 in this volume), and the establishment of the Latin principalities in the Levant thereafter, individuals from this region seem to have played a fairly limited role in the history of the crusades and the Latin East. Whether this was, as John France has observed when viewing the First Crusade from a Norman point of view, the “last gasp of that great expansionist drive”, or, as Graham Loud has concluded from a different perspective, that “the crusade had no part in the Sicilian tradition”, it is hard to deny that the Kingdom of Sicily played a remarkably small part in the affairs of the Latin principalities through most of the twelfth century.1 What is commonly regarded as the moment of exception to this apparent apathy took place in 1174 when a Sicilian fleet attacked Alexandria (chapter 7 in this volume). Cast by historians as part of a broader crusading campaign, which was meant to coordinate with the forces of the Latin Kingdom of Jerusalem and a Shiite uprising in Cairo, this purported effort to capture Alexandria failed when the Sicilians found

* My thanks to Niall Christie for reading and commenting on the first draft of this

chapter. 1 John France, “The Normans and Crusading”, in The Normans and Their Adversaries at War: Essays in Memory of C. Warren Hollister, ed. Richard P. Abels and Bernard S. Bachrach (Woodbridge, 2001), pp. 87–101; Graham A. Loud, “Norman Italy and the Holy Land”, in The Horns of Hattin, ed. Benjamin Kedar (Jerusalem, 1992), pp. 49–62. See also Helene Wieruszowski, “The Norman Kingdom of Sicily and the Crusades”, in A History of the Crusades: Volume II, the Later Crusades, 1189–1311, ed. Robert Lee Wolff and Harry W. Hazard, gen. ed. Kenneth M. Setton (Madison, 1969), pp. 3–42.

226

michael s. fulton

themselves unsupported. It seems, however, that there may be another way to interpret this episode, one which conforms more to the Sicilians’ modus operandi. Most of the details regarding this campaign, and the context of its planning, are found in Muslim sources, the majority of which can be traced back to the propagandistic concerns of Saladin, who was still defining his position as ruler of Egypt at this time. Saladin had an interest in seeing what might otherwise be characterised as a fairly traditional raid presented as a more serious attack, one undertaken in coordination with others who sought to remove him from power in Egypt. This allowed him to legitimise his position and portray himself as a champion of Sunni Islam. When looking at the participants of the First Crusade and its successes, it is hard to overlook the contributions of Bohemond of Taranto, as well as those of his nephew, Tancred. In the decades following the crusade, men of similar southern-Italian Norman ancestry would fill out the baronage of the principality that Bohemond and Tancred established around Antioch;2 direct support from the rulers of southern Italy, however, was limited. Bohemond returned to Europe in 1104, where he gathered support for another crusade. This he directed against the Byzantines in the Balkans in 1107. Despite being launched from southern Italy, Bohemond’s crusade was not widely supported in the region; although he enjoyed the blessing of the pope, neither Roger I of Sicily nor Duke Borsa of Apulia, his uncle and half-brother respectively, openly supported his efforts.3 The traditional acceptance of a lack of interest shown by Sicily towards the affairs of the Latin principalities of the Levant has been pinned on William of Tyre’s remarks regarding Baldwin I of Jerusalem’s repudiation of Adelaide of Sicily. The marriage of Baldwin and Adelaide in 1113 had brought much-needed cash to the Kingdom of Jerusalem, but scandal led to an annulment a few years later, at which point Baldwin kept the dowry. This, according to William, led Adelaide’s son, Roger II of Sicily, and his heirs to conceive “a mortal hatred against the kingdom and its peoples”, noting that no reconciliation had yet been effected when he was composing this part of his history, probably in the 1170s.4 Accordingly, this has served as a convenient way of explaining the lack of coordination and military cooperation between the two kingdoms in the years that followed.5 In reality, although there were obvious family ties between southern Italy and the principality of Antioch, and there were important economic links between the Kingdom of Sicily and the Latin principalities, it seems that “crusading” 2

For others who accompanied Bohemond on crusade from southern Italy, see Loud, “Norman Italy and the Holy Land”, pp. 49–50, 52. 3 France, “The Normans and Crusading”, pp. 93–96. 4 William of Tyre, Cronique ed. R. B. C. Huygens (London, 1986), 11.29, p. 542. 5 For example, Steven Runciman, A History of the Crusades, 3 vols. (Cambridge, 1951–54), 2:105.



disaster in the delta? 227

was never particularly popular in southern Italy in the decades following the First Crusade.6 Whether this was due to preoccupation or disinterest is another debate, but what is certain is that the Sicilians had the capability to involve themselves, had they so chosen or had circumstances allowed. The offensive capacity of the Sicilians was displayed by Roger II, and in 1154 Sicilian forces sacked Tinnis, at the mouth of the easternmost of the four main branches of the Nile delta.7 Fātimid Egypt was at this time in decline and an internal struggle for power led to its further weakening over the following decade. By 1164 Amalric of Jerusalem and Nūr al-Dīn, the Sunni Turkic ruler of Syria, had begun to actively intervene in Egypt, where their forces fought to gain influence and, eventually, control over the riches of the region. In January 1169 Nūr al-Dīn’s deputy, Shīrkūh, had successfully gained effective control of Egypt and was named vizier by the Shiite Fātimid caliph. Shīrkūh died in March of the same year, at which point he was succeeded by his nephew, Saladin. As vizier, Saladin would suppress the Fāimid caliphate in September 1171, bringing Egypt back into the fold of Sunni Islam. From this point, Saladin ruled in the name of Nūr al-Dīn, until the Syrian ruler’s death in May 1174. Even before Saladin had supressed the Fātimids, however, he faced challenges to his authority. In the summer of 1169 a plot was hatched by officials of the Fātimid palace. The plan involved the conspirators calling on the Franks for support. It was hoped that the arrival of a Frankish army would force Saladin to confront the invaders, allowing the conspirators to rise up in his absence and seize Cairo, at which point they would move against Saladin’s rear, trapping the vizier between two armies. Saladin became aware of the scheme and, at an opportune moment, had the ringleader executed; he also used the occasion to remove the rest of the eunuchs from the caliphal palace. This caused the revolt of the Sudanese and Armenian regiments in Cairo, the core of the Fātimid army, who saw Saladin’s actions and the influence of his Syrian forces as a threat to their own influence. Saladin’s ability to discover the plot and defeat the uprising that followed provided the opportunity to rid himself of his rivals and secure his rule over Egypt.8 But it was not only enemies within Egypt that Saladin had to confront. 6

Loud, “Norman Italy and the Holy Land”, pp. 51–53. Ibn al-Qalānisī, The Damascus Chronicle of the Crusades [Taʾrīkh Dimashq], trans. Hamilton Alexander Rosskeen Gibb (London, 1932), pp. 321–22; Ibn al-Athīr, The Chronicle of Ibn al-Athīr [al-Kāmil fi al-taʾrīkh], trans. Donald Sidney Richards, 3 vols. (Aldershot, 2006–8), 2:65. 8 Abū Shāma, Livre des deux Jardins [Kitāb al-Rawḍatayn], ed. and trans. in RHC Or 4–5 (Paris, 1898–1906), 4:145–48, 172–73; History of the Patriarchs of the Egyptian Church, ed. and trans. Oswald Hugh Ewart Khs-Burmester, et al. 4 vols. (Cairo, 1942–74), pt. 9, 3.2:109–11; Ibn al-Athīr, 2:179–80; Anne-Marie Eddé, Saladin, trans. Jane Marie Todd (Cambridge MA, 2011), pp. 40–43; Andrew S. Ehrenkreutz, Saladin (Albany, 1972), pp. 76–79. 7

228

michael s. fulton

In September 1168 an alliance had been concluded between the Byzantines and the Kingdom of Jerusalem. Before a coordinated effort could be arranged, however, Amalric opted to invade, marching his forces into Egypt in late October. His failure to secure Cairo before Shīrkūh’s arrival with a Syrian army essentially gifted Egypt to his rival. This led Amalric to send out appeals for help to the princes of Europe, including William II of Sicily, which did not go unnoticed by Muslim sources.9 Manuel Comnenus was the only one to reply positively, honouring the agreement that had been concluded the previous year, and a Byzantine fleet arrived at Acre in the late summer of 1169. The combined force then set out for Egypt and before the end of October, just two months after Saladin’s suppression of the Fātimid conspiracy and decimation of the Egyptian army, Damietta was placed under siege. According to William of Tyre, the Byzantine fleet included 150 beaked galleys, sixty horse-transports, and ten or twelve dromones that carried supplies, including arms and siege engines, a detail confirmed by Bahā’ al-Dīn ibn Shaddād.10 The fleet sacked Tinnis before continuing on to Damietta. The land force arrived on 27 October and the fleet, which followed, anchored slightly upstream from the mouth of the river three days later. It seems that Saladin was aware of the Byzantines’ intentions; six Egyptian vessels had met the Byzantine fleet off Cyprus in July, four of them escaping to spread news of their encounter. Anticipating an attack against Damietta, Saladin dispatched his nephew, Taqī al-Dīn ‘Umar, and his uncle, Shihāb al-Dīn Mahmūd al-Hārimī, to command the city’s defence, while he remained at Cairo. Upon its arrival, the Christian army elected to pause for a few days, inadvertently allowing Muslim reinforcements to stream into the city. The siege lasted about fifty days and involved the use of a number of engines, including artillery and at least one siege tower. The siege failed to progress and terms were finally reached between the two sides, allowing the Christian army to withdraw in peace, and the Franks were back in Palestine in time for Christmas.11 Damietta was a sensible target for the Frankish–Byzantine force. The city commanded the mouth of the main eastern branch of the Nile delta and, as Ibn al-Athīr remarked, it was a sensible base from which to conquer the rest 9

William of Tyre, 20.12, p. 925; Ibn al-Athīr, 2:183. William of Tyre, 20.13–15, pp. 926–29; Bahāʾ al-Dīn ibn Shaddād, The Rare and Excellent History of Saladin [al-Nawādir al-sultāniyya wa al-maḥāsin al-yūsufiyya], trans. Donald Sidney Richards (Aldershot, 2001), p. 45. See also Niketas Choniates, O’ City of Byzantium, trans. Harry J. Magoulias (Detroit, 1984), p. 91. 11 William of Tyre, 20.15–16, pp. 929–33; Abū Shāma, 4:149, 150–55, 173–74; Bahāʾ al-Dīn, p. 46; Ibn al-Athīr, 2:183–84; Niketas Choniates, pp. 91–96; John Kinnamos, Deeds of John and Manuel Comnenus, trans. Charles. M. Brand (New York, 1976), 6.9, pp. 208–9; Ehrenkreutz, p. 80; Eddé, p. 44. 10

disaster in the delta? 229

12

of Egypt. Tinnis, which controlled access up the lesser eastern branch, was the closest to Palestine but was a less defensible port, making it an attractive target for raiders but a less advantageous foothold. The same might be said of Rosetta, at the mouth of the main western branch of the delta, which was also farther from Palestine. Alexandria, which was a more defensible city, was Egypt’s most active Mediterranean port.13 Although it stood at the mouth of what was once a main channel of the delta, it was by this point connected to the main Rosetta artery by little more than a shallow canal. Despite its strength and wealth, Alexandria was thus a less attractive staging point for an invasion of Egypt, as larger vessels would not have been able to sail up the river and support land forces as they moved against Cairo. It seems that it was Damietta’s strength and the navigability of the river between it and Cairo that made the city an attractive target. Following in the path of the Frankish–Byzantine attack of 1169, Damietta was besieged in 1218–19 during the course of the Fifth Crusade and was taken by the crusading force led by Louis IX of France in 1249. By comparison, Alexandria was arguably never attacked by a crusading army during the twelfth and thirteenth centuries – this depends on one’s definition of “crusading”; the possible exceptions were the Sicilian attack of 1174 and Amalric’s siege of the city in 1167. Niall Christie has suggested that, in addition to the city’s dubious tactical value, Alexandria’s importance as a flourishing point of exchange may have persuaded crusading armies to look elsewhere for targets.14 Despite the image of the resolute mujāhid that Saladin cultivated, he kept the port open to European traders. In 1173 he allowed the Pisans to establish a colony there, which included a church.15 In a letter written in 1174–75 detailing Saladin’s accomplishments, al-Qādī al-Fādil, head of the Egyptian chancery, noted that the trade with these European enemies was bringing arms and war

12

Ibn al-Athīr, 2:183. For overviews of the development of Alexandria’s defences, see Christophe Benech, “Recherches sur le tracé des murailles antiques d’Alexandrie”, in Alexandrina 3, ed. Jean-Yves Empereur (Cairo, 2009), pp. 401–45; Kathrin Machinek, “Der Wandel der Stadtbefestigung Alexandrias vom Mittelalter bis in die Gegenwart”, in “Vmbringt mit starcken turnen, murn”. Ortsbefestigungen im Mittelalter, ed. Olaf Wagener (Frankfurt am Main, 2010), pp. 431–50; Machinek, “Aperçu sur les fortifications médiévales d’Alexandrie: Histoire, architecture et archéologie”, in La guerre dans le Proche-Orient médiéval, ed. Mathieu Eychene and Abbès Zouache (Cairo, 2015), pp. 363–94. 14 Niall Christie, “Cosmopolitan Trade Centre or Bone of Contention? Alexandria and the Crusades, 487–857/1095–1453”, Al-Masaq 26.1 (2014), 49–61. 15 Michele Amari (ed.), I diplomi arabi del R. Archivio Fiorentino (Florence, 1863), no. 2.7, pp. 257–61; Subhi Labib, “Egyptian Commercial Policy in the Middle Ages”, in Studies in the Economic History of the Middle East, ed. M. A. Cook (Oxford, 1970, repr. Abingdon, 2014), pp. 63–77, at p. 66. 13

230

michael s. fulton 16

materials into Egypt. To further secure this lucrative trade centre, Saladin ordered the development of its defences in 1171, the same year in which refortification efforts began in Cairo. Perhaps in part to inspect the progress of these works, Saladin visited Alexandria the following year. In 1174 Saladin faced another conspiracy, hatched by a collection of Shiite figures who hoped to gain greater influence by deposing him and restoring the Fātimid caliphate. The plot was discovered in the spring and Saladin once more used the opportunity to cleanse Egypt of his opponents, executing the dissidents and exiling the remaining elements of the Fātimid army.17 Not long after, on 15 May, Nūr al-Dīn, Saladin’s nominal lord, died in Damascus. The Franks had used this opportunity to move against the frontier town of Banyas. Hearing this, Saladin mustered his forces in the name of marching to the relief of the Damascenes; however, before he and his army had left Egypt, news arrived that a truce had been concluded between Jerusalem and Damascus.18 Outraged by this peace, quite likely because Frankish aggression gave him a pretext to assert his hegemony over Muslim Syria, Saladin kept his army mobilised at Fāqūs. It was here that he had halted when he received word regarding the truce; it was here that he was informed that King Amalric of Jerusalem had died on 11 July; and it was here, on 30 July, that Saladin learned that the Sicilians had landed at Alexandria. The Sicilian fleet was first spotted around noon on Sunday 28 July 1174. In a letter he sent to an emir in Syria describing the attack, Saladin notes that although the lookouts were surprised, it was known that the Sicilian force was mustering, warnings having been received from both the Maghreb and Byzantium. In a slightly later letter written by al-Fādil, it is expressed that the Sicilians had been building their strength ever since the failed Byzantine–Frankish siege of Damietta five years earlier.19 Sicilian ships continued to arrive through the afternoon. Ibn al-Athīr, al-Maqrīzī, and others repeat the figures provided in the letter sent out by Saladin after the siege: two hundred galleys, thirty-six horse-transports, six ships loaded with siege engines and materials, and forty transport vessels loaded with men and cargo.20 It seems that the Sicilians waited until the next day, Monday 29 July, to disembark.21 The Alexandrians were initially tempted to oppose a landing, 16

Abū Shāma, 4:170–80. Ibn al-Athīr, 2:218–20; al-Maqrīzī, A History of the Ayyūbid Sultans of Egypt [al-Sulūk], trans. Ronald J. C. Broadhurst (Boston, 1980), p. 47; Eddé, pp. 58–59; Ehrenkreutz, pp. 112–15. 18 Abū Shāma, 4:160–63; al-Maqrīzī, p. 49. 19 Abū Shāma, 4:164, 177–78. 20 Abū Shāma, 4:167; Ibn al-Athīr, 2:229; Bahāʾ al-Dīn, p. 50; al-Maqrīzī, p. 49. See also Malcolm C. Lyons and D. E. P. Jackson, Saladin: The Politics of the Holy War (Cambridge, 1982), p. 77, n. 21. 21 See Lyons and Jackson, Saladin, p. 76. 17



disaster in the delta? 231

but it was decided that it would be too dangerous to risk a defeat at a distance from the town walls. The first attack forces landed near the classical lighthouse, occupying the peninsula north of the town, which divided the city’s two harbours, and pushed a group of Muslim skirmishers back into the city. The Sicilian vessels then rowed into the main harbour, where they seized some ships while the Muslims scuttled others to prevent their capture. It was at this point the Sicilians also gained a foothold on the shore beyond the northern peninsula. Fighting continued through the day and in the evening the invaders erected their tents, which numbered three hundred. The next morning, Tuesday 30 July, set-piece siege operations began. The Sicilian army is said to have numbered about fifty thousand, including thirty thousand fighters, of whom fifteen hundred or so fought on horseback. The Muslim sources also claim that three siege towers and an equal number of trebuchets were deployed against the defences of Alexandria. While the former may have been equipped with rams, the latter supposedly threw black stones brought from Sicily.22 The defenders of Alexandria are described as being few in number, although some reassurance came with the arrival of a few troops from the surrounding region. It was not until Tuesday 30 July that word of the attack reached Saladin. From Fāqūs, Saladin sent reinforcements to Alexandria, as well as Damietta. Meanwhile, the attackers advanced their engines, which were positioned within a couple of hundred metres of the sea. Taking a proactive stance, on either Tuesday or Wednesday the defenders launched a sally through a postern in front of the Sicilians’ engines and managed to burn their siege towers. Fighting continued until the afternoon of Wednesday 31 July, at which point the Sicilians, having suffered considerable losses and the destruction of their siege engines, became discouraged. When news arrived that Saladin was approaching, it reinvigorated the defenders, leading them to launch another attack that evening. Although it was dark by this point, their charge reached the besiegers’ camp. As fighting took place among the Sicilians’ tents, some of the attackers fled to the sea, pursued by parties of defenders. Efforts were made on behalf of the Sicilian naval forces to put certain vessels to sea, taking them out of danger, although some were apparently lost. A party of knights, without their horses, attempted to make a stand on a hill, but they were all eventually killed or captured by noon the following day. The extent of the attackers’ defeat was clear to all that morning, Thursday 1 August, and the Sicilians who had made it back to their ships set sail for home, leaving Alexandria to their stern. Much of the goods from the attackers’ camp, which had been so quickly overrun, were captured by the victorious defenders.23 22

Abū Shāma, 4:164–66; Ibn al-Athīr, 2:229; al-Maqrīzī, p. 49; Lyons and Jackson, p. 76. Abū Shāma, 4:166–67; Ibn al-Athīr, 2:229–30; al-Maqrīzī, pp. 49–50; Bahāʾ al-Dīn, p. 50.

23

232

michael s. fulton

When interpreting the events of this attack, historians have generally taken a fairly enthusiastic approach. For example, Helene Wieruszowski accepted that William II dedicated himself to preparing for an attack against Egypt from about 1171, and from 1173 negotiations were conducted between Shiite dissidents, the Kingdom of Jerusalem and the Kingdom of Sicily, with the aim of removing Saladin from power in Egypt. According to this interpretation, all parties were aware of a growing rift between Saladin and Nūr al-Dīn, and perhaps also growing opposition to Saladin in Egypt, but neither the Sicilians nor the Franks learned about the collapse of the Fātimid conspiracy before the attack was launched. It is also proposed that the Sicilian commanders may have been unaware of Amalric’s death when they reached Alexandria on 28 July 1174. Upon their arrival, the progress of the attackers was inhibited when the inhabitants of Alexandria sank their ships in order to deny the Sicilians entry to the harbour. The details of the events which followed are then summarised.24 Two decades later, Donald Matthew offered a similar overview of events, again emphasising the coordination between Sicilian, Frankish, and Fātimid parties and the ignorance of the Sicilians regarding the discovery of the Shiite plot and Amalric’s death.25 For Matthew, this interpretation fell into line with his broader presentation of William II as an “enemy of Islam”, if not a crusader. In more recent years, historians have shown somewhat less enthusiasm when examining this episode. Graham Loud seems to downplay the Alexandria campaign in order to emphasise that the Sicilians’ focus was directed far more often towards the Byzantines, while William’s support for the Franks was more obvious in the interest he dedicated to the Holy Land churches of southern Italy.26 John France likewise devotes less attention than might be expected in his study of Norman crusading efforts.27 If Loud and France have helped to better contextualise this episode, it remains to examine the campaign itself more closely. There are numerous issues with the standard interpretation of events. These include: the significance of the Fātimid plot; the suggestion that this was to be a joint venture with Frankish forces; the details of the attack, their original source and the manner of their presentation by certain figures; and how the presented narrative fits into the broader context of events. When considering the Fātimid plot of 1174, certain details are suspiciously similar to those of the intrigue of 1169. On both occasions the conspirators planned to draw the Franks into an invasion of Egypt, which would lead Saladin to either march out against them, allowing the conspirators to rise up and seize control of Cairo and then move against Saladin’s rear; 24 25 26 27

Wieruszowski, pp. 34–36. Donald Matthew, The Norman Kingdom of Sicily (Cambridge, 1992), p. 279. Loud, “Norman Italy and the Holy Land” pp. 51, 61. France, “The Normans and Crusading”, p. 100.



disaster in the delta? 233

or, if Saladin dispatched a force to confront the Franks, the plotters would rise up and seize Saladin himself. It is not inconceivable that the same plan was attempted twice; however, the similarities are enough to warrant a degree of suspicion. The first plot miscarried in the summer of 1169 and was followed by the Byzantine–Frankish attack against Damietta three months later. In this instance, although there are suggestions that the conspirators were in correspondence with the Franks,28 neither contemporary sources nor modern historians suggest that the invasion was linked to the Shiite plot. Five years later, when another attack took place three and a half months after the public crucifixions of conspirators began on 6 April 1174, it may have been too convenient not to draw a direct link this time. According to Muslim sources, the Franks were involved to some degree in the Fātimid plot of 1174. They suggest that Zayn al-Dīn Ibn Najā (known as Ibn Nujiyya) betrayed the conspirators’ plan to Saladin, who was then in a position to act against the plotters when it best suited his interests. When considering the Franks, Ibn al-Athīr notes that a Frankish envoy was sent with gifts for Saladin, but his true objective was to coordinate with the conspirators through a local Egyptian Christian. Saladin, however, discovered the true mission of this envoy from his own spies among the Franks and assigned his own local Christian operative to gain the envoy’s trust and discover more.29 Whether or not there is any truth in this anecdote, it emphasises the prevalence and effectiveness of the clandestine intelligence networks that operated in the region. Half a century earlier, in what is perhaps the most famous anecdote of this type, it is said that ‘Izz al-Dīn Mas‘ūd, who ruled Aleppo on behalf of his father, Āqsunqur al-Bursuqī, learned of his father’s death from the Franks before word reached him from his father’s men in Mosul.30 Even if room is left for considerable embellishment, it seems inconceivable that the Franks were unaware that the Fātimid plot had miscarried by the end of April 1174 – it is more likely that they did not know about it until it collapsed than that they knew of its existence but not its failure. When Saladin finally moved against the conspirators, it is widely reported that he had the leading figures crucified in prominent positions around Cairo.31 Ibn al-Athīr notes that the Franks ceased any plans to invade Egypt when they learned of this, but news did not reach the Sicilians.32 Although possible, it is doubtful whether reports of the public crucifixions had not arrived at the Sicilian court by late July 1174, two and a half months after they had begun. Setting aside the part that the Fātimid plot played in the Sicilians’ planning of the Alexandria campaign, there is the role, or lack thereof, 28 29 30 31 32

Ibn al-Athīr, 2:179. Ibn al-Athīr, 2:219. Ibn al-Athīr, 1:262. Al-Maqrīzī, p. 47; Ehrenkreutz, p. 114. Ibn al-Athīr, 2:220.

234

michael s. fulton

played by the Franks to consider. It is widely assumed that this was meant to be a joint Sicilian–Frankish venture, not unlike the Byzantine– Frankish campaign against Damietta in 1169. There is very little evidence, however, to support this. The Franks had envisioned conquering Egypt from about the time they captured Jerusalem in 1099,33 and the region’s conquest was discussed directly with the Byzantines in the summer of 1168, when an alliance and joint attack against Egypt was negotiated.34 As these negotiations went on at the Byzantine court, back in Palestine the Franks actively divided up parts of Egypt, allocating who would receive what once the conquest was completed. The Hospitallers, for example, who were among the leading supporters of the campaign, were to receive the important frontier town of Bilbays, additional properties in the other large towns, rents in Egypt amounting to 150,000 bezants, and other benefits.35 The Franks ended up invading that autumn, choosing not to wait for Byzantine support, and ultimately gained little. The following year, the Franks again divvied up Egypt ahead of their attack against Damietta, this time acting in conjunction with their Byzantine allies.36 Conspicuously, there is little evidence to suggest that a similar division of potential plunder was undertaken in 1174, implying that the Franks had no real plans to invade Egypt in force that year. When considering the general lack of military support sent from Sicily to the Levant, such a joint venture would have been exceptional. Despite direct appeals to William II in 1169 and again in 1171, the Sicilians seem to have sent no more military aid than the other princes of western Europe who received similar letters. It was, rather, the Byzantines who showed the most willingness to help, renewing their pledge of support following a visit by Amalric himself to Constantinople in 1171.37

33

For example, 1100: William of Tyre, 9.16, 10.4, pp. 441, 456–57; Hans E. Mayer (ed.), Die Urkunden der lateinischen Könige von Jerusalem, 4 vols. (Hanover: MGH, 2010) [hereafter UKJ], no. 7, 1:103–4; Jonathan Riley-Smith et al. (eds), Revised Regesta Regni Hierosolymitani Database (http://crusades-regesta.com/) [hereafter RRR], no. 26; 1104: Reinhold Röhricht (ed.), Regesta Regni Hierosolymitani (Innsbruck, 1893) [hereafter RRH], no. 43, p. 8; UKJ, no. 29, 1:137–44; RRR, no. 80; 1159: UKJ, no. 249, 1:457–59; RRR, no. 623. 34 William of Tyre, 20.4, pp. 915–17; UKJ, nos. 330, 331, 334, 2:571–73, 275–77; RRR, nos. 798, 799, 807. 35 William of Tyre, 20.5, pp. 917–18; RRH, no. 452, p. 118; UKJ, no. 336, pp. 578–82; RRR, no. 809. See also Jonathan Riley-Smith, The Knights Hospitaller in the Levant, c.1070–1309 (Basingstoke, 2012), pp. 33–34. 36 UKJ, nos. 340, 341, 342, 2:588–96; RRH, nos. 465, 466, p. 122; Reinhold Röhricht (ed.), Regesta Regni Hierosolymitani, Additamentum (Innsbruck, 1904), no. 451a, p. 28; RRR, nos. 838, 839, 840. 37 William of Tyre, 20.12, 23–24, pp. 926, 944–46; John Kinnamos, 6.10, p. 209; RRH, no. 464, pp. 121–22; UKJ, no. 350, 2:610–12; RRR, nos. 837, 870.



disaster in the delta? 235

Regardless of the events leading up to the Sicilians’ arrival at Alexandria in 1174, the details of the attack that followed are also suspect. Most information is provided by Muslim sources; the relative silence of their Christian counterparts, even those based in Egypt, is in itself suspicious. William of Tyre describes the siege as follows: During the first year of this king’s reign [that of Baldwin IV], about the beginning of August, King William of Sicily sent a fleet of two hundred ships to attack Alexandria. With a splendid force of both cavalry and infantry, it sailed down to Egypt. During the stay of five or six days made before that city, through the lack of caution displayed by the governors and leaders, both the infantry and cavalry force sustained great losses by death and capture and 38 were finally obliged to retire in confusion.

William confirms the duration of the attack and its failure, but adds little more, and at no point hints that there was ever any discussion that this was to be a joint Sicilian–Frankish venture. By comparison, William’s account of the siege of Damietta five years earlier is immensely more detailed.39 In their slightly later accounts the contemporary historians Bahā’ al-Dīn and Ibn al-Athīr provide similarly brief overviews.40 A far more detailed version of events, that outlined above, has been preserved in the history of Abū Shāma, who made use of the earlier history of ‘Imād al-Dīn, which includes a letter sent by Saladin to an emir in Syria following the attack. The issue with the details of the attack, as found in Abū Shāma’s account, is a practical one, exposing what appear to be clear exaggerations. For example, there is the time that it took the Sicilians to erect their siege machinery to consider. It seems to be implied that work assembling the engines began on Monday 29 July, that they were brought into action the following day, and that they were destroyed on Wednesday 31 July. It may have been possible to erect sizable trebuchets within about twenty-four hours – the suggestion that they threw imported ammunition may indicate that these were early counterweight trebuchets. Saladin was able to set up what were probably engines of a comparable size within a similar window of time during his campaigns in Frankish Palestine and western Syria in 1187–88 and at Jaffa in 1192.41 However, erecting three siege towers, even if assembled from prefabricated components, would have been a far greater challenge: siege towers typically took weeks to construct. It took elements of the First Crusade three days to reassemble Godfrey of Bouillon’s siege tower after moving it to a new position one night during the siege of 38

William of Tyre 21.3, p. 963, trans. Emily A. Babcock and A. C. Krey, A History of Deeds Done Beyond the Sea, 2 vols. (New York, 1976), 2:399–40. 39 William of Tyre 20.13–17, pp. 926–33. 40 Bahāʾ al-Dīn, pp. 45–46; Ibn al-Athīr, 2:183–84. 41 See Michael S. Fulton, Artillery in the Era of the Crusades: Siege Warfare and the Development of Trebuchet Technology (Leiden, 2018), pp. 153–74, 179, 200–2.

236

michael s. fulton 42

Jerusalem in 1099. Two years later, during the Franks’ siege of Caesarea, their siege tower was still incomplete when forces stormed the city about two weeks after the siege had begun.43 The second siege tower built by the Anglo-Norman forces at Lisbon in 1147 probably took about a month to build, while that built by Frankish–Damascene forces against Banyas in 1140 seems to have been put together within about three weeks.44 That three siege towers were erected in little more than a day, or even two, seems a dubious claim. Saladin probably exaggerated many of the particulars in order to impress the readers of the letters he sent out after the siege. Comparable embellishments can be found in other letters. In that noted above, which Saladin’s chancellor in Egypt, al-Qādī al-Fādil, sent to the caliph in Baghdad, it is claimed that a thousand ships and a land force numbering 200,000 men attacked Damietta in 1169. Similar letters were sent by Saladin or those writing on his behalf after his failed siege of Kerak in 1184 and following his capture of Ascalon in 1187.45 In each, accuracy is clearly sacrificed in order to place greater emphasis on Saladin’s zeal in service of the military jihad. It was this role, as protector and defender of the faith, that gave his rule greater legitimacy. While this was initially used to justify his importance in Egypt, it would remain the cornerstone of his propaganda following Nūr al-Dīn’s death, as he set his sights on expanding his realm beyond the boundaries of Egypt, often at the expense of fellow Muslims. It seems that Saladin was quick to appreciate the similarities between the circumstances in 1174 and those that had accompanied the Frankish– Byzantine attack against Damietta in 1169. Muslim sources were aware that the Franks were sending out desperate appeals for help to Europe’s leading monarchs, including William II.46 By associating his enemies with the Franks, Saladin could further legitimise his actions against them. He was able to justify his suppression of the Fātimid military in 1169 by claiming that not only was it an obstacle to the authority of the Abbasid caliph in Baghdad, but that many among its ranks were acting treacherously by conspiring with the Franks; the violence that he unleashed was thus in

42

For this well-known episode, see John France, Victory in the East: A Military History of the First Crusade (Cambridge, 1994), pp. 348–49. 43 Fulcher of Chartres, Historia Hierosolymitana, ed. Heinrich Hagenmeyer (Heidelberg, 1913), 2.9.1–4, pp. 400–2. 44 For the siege of Lisbon, see De Expugnatione Lyxbonensi, ed. and trans. Charles W. David (New York, 1936), pp. 124–78; see also Randall Rogers, Latin Siege Warfare in the Twelfth Century (Oxford, 1992), pp. 183–88. For the siege of Banyas, see William of Tyre, 15.8–10, pp. 685–88; Ibn al-Qalānisī, pp. 260–61; see also Fulton, Artillery in the Era of the Crusades, pp. 112–14. 45 Abū Shāma, 4:170–80, 251–53; History of the Patriarchs, pt. 9, 3.2:129–31. 46 Ibn al-Athīr, 2:183.



disaster in the delta? 237

defence of Sunni orthodoxy.47 Although the plot of 1174 was hatched with intentions of restoring the Fātimid caliphate, the attack made against Alexandria that followed was far less sensational than that against Damietta in 1169. The parallels between the two, however, were too clear for Saladin to pass up. By exaggerating the scale and significance of the Sicilian attack, and linking it with the failed plot earlier in the year, he drove home his message: he was the defender of Sunni orthodoxy in the Levant, the bastion of Islam keeping Egypt out of the hands of the Christians and Shiite heretics alike. In reality, the Sicilian attack may have been much more of a raid than a concerted effort to seize and hold Alexandria. Sicilian forces had raided the coast of Egypt in the past and would twice raid Tinnis before the end of the decade, briefly occupying the town on the latter occasion. To Saladin’s credit, historians have accepted the size of the force as he presents it and, almost without exception, have cast it as a campaign of conquest, not unlike the Frankish–Byzantine assault on Damietta five years earlier. If the numbers of men, machines, and ships are set aside, we can consider that the “besiegers” were defeated only a few days after they arrived; that it was a sally by the defenders, unsupported by a relief army, that drove off the assailants; that the governor of Alexandria was willing to wait two days before informing Saladin of the force’s arrival; and that Saladin, once informed, sent only a portion of his army to Alexandria, while another was sent to Damietta. The lack of urgency exhibited by the Muslim commanders and the apparent strength of Alexandria’s local defenders, relative to their besiegers, suggests that this “siege” was rather a “raid”. Although the Sicilians may have had ambitions of entering and even sacking the city, it is hard to see how they could have hoped to hold it. Unlike the Frankish–Byzantine campaigns, among which this episode is typically grouped, it was commercial motivations, rather than territorial ones, that probably guided the Sicilian commanders. As in the decades before the attack on Alexandria, Sicilian forces at times raided the coast of Egypt, seizing wealth and then withdrawing without making any attempt to hold the lands they briefly occupied. William II used the rhetoric of crusade, much as Saladin used that of jihad, for political ends, but neither one compromised his position in support of these causes. It was not until Saladin had taken Jerusalem in 1187 that Sicilian forces directly supported the Latin principalities of Palestine by furnishing a significant military force to aid their defence. A fleet assisted with the defence of Tripoli and helped to safeguard Tyre, while its main action was undertaken off the coast of Margat, where it harassed Saladin’s army as it passed along a narrow stretch of the coastal road in the summer of 1188.48 47

Abū Shāma, 4:170–80. See also Eddé, pp. 42–43. ʿImād al-Dīn al-Isfahānī, Conquête de la Syrie et de la Palestine par Saladin [Kitāb al-fath al-qussī fi al-fath al-qudsī], trans. Henri Massé (Paris, 1972), pp. 125–26; Ibn al-Athīr, 2:345. 48

238

michael s. fulton

The relationship between Norman Sicily and the Frankish Levant remains to be studied in closer detail; however, the lack of military support provided by the former is noteworthy. Although the campaign against Alexandria in 1174 is often cast as the exception, it seems to have conformed much more to tradition than is often supposed. It is doubtful whether the attack would have received the same amount of attention had it not been for the letter preserved first by ‘Imād al-Dīn and then by Abū Shāma; the relative silence of contemporaries is significant. Likewise, the obvious motivations behind the composition of this singular source must be taken into consideration. When weighed against the surrounding evidence, including the lack of urgency shown by Muslim commanders and the absence of evidence that the Franks had any intentions of taking part in the expedition, Saladin’s portrayal of events fits a model found in other politically motivated letters that contain clear exaggerations. Rather than describing a mere raid, even if a larger one than normal, it was in Saladin’s interest to exaggerate the magnitude of the attack and hence to present himself as a defender of Islam, the protector of Egypt against the Fātimid heretics and Christian infidels. He inflated the raid, presenting it as a much more significant siege, in order to fulfil the image he was cultivating. As for the Sicilians, this was one of a number of operations they conducted along the Egyptian coast during the second half of the twelfth century – their aims were almost certainly more commercial than religious or territorial.

Bibliography Primary Sources Abū Shāma, Livre des deux Jardins [Kitāb al-Rawdatayn], ed. and trans. in RHC Or 4–5 (Paris, 1898–1906). Aimé du Mont-Cassin, Ystoire de li Normant, ed. M. Guéret-Laferté (Paris, 2011). Albert of Aachen, Historia Ierosolimitana, ed. and trans. S. B. Edgington (Oxford, 2007). Alexander of Telese, Alexandri Telesini abbatis Ystoria Rogerii regis Sicilie, Calabrie atque Apulie, ed. L. De Nava (Rome, 1991). Al-Maqrīzī, A History of the Ayyūbid Sultans of Egypt, trans. R. J. C. Broadhurst (Boston, 1980). Al-Maqrizi, Kitab al-tarikh al-kabir al-muqaffa li-Misr, trans. J. Johns in Arabic Administration in Norman Sicily (Cambridge, 2002). Al-Nuwayri, Biblioteca Arabo-Sicula, ed. and trans. M. Amari (2 vols., Turin, 1880–81), II, 110–60. Amari, Michele (ed.), I diplomi arabi del R. Archivio Fiorentino (Florence, 1863). Amatus of Montecassino, The History of the Normans, trans. P. N. Dunbar, revised G. A. Loud (Woodbridge, 2004). Amatus of Montecassino, Ystoire de li Normant, ed. M. Guéret-Laferté (Paris, 2011). Ambroise, History of the Holy War: Ambroise’s Estoire de la Guerre Saint, ed. and trans. M. Ailes and M. Barber (2 vols., Woodbridge, 2003). Annales Casinenses, ed. G. Pertz (Monumenta Germaniae Historica, Scriptores, XIX, Hanover, 1866), 303–20. Annales Ceccanenses, ed. George Pertz (Monumenta Germaniae Historica, Scriptores, XIX, Hanover, 1866), 275–302. At-Tijani, Biblioteca Arabo-Sicula, ed. M. Amari (2 vols., Turin, 1880–1), II, 41–81. Badr al-Din Ibn Juma’h Muhammad Ibn Ibrahim (ed. Usamah Nasir al-Naqshbandi), Mustanad al-Ajnad fi ‘Alat al-Harb (Baghdad, 1983). Bahāʾ al-Dīn ibn Shaddād, The Rare and Excellent History of Saladin [al-Nawādir al-sultāniyya wa al-mahāsin al-yūsufiyya], trans. D. S. Richards (Aldershot, 2001). Becker, Julia (ed.), Documenti latini e greci del conte Ruggero I di Calabria e Sicilia (Rome, 2013). Choniates, Niketas, O City of Byzantium, trans. H. J. Magoulias (Detroit, 1984). Chronica Ignoti Monachi Cisterciensis S. Maria de Ferraria, ed. A. Gaudenzi (Naples, 1888). Dandolo, Andrea, Chronica Per Expensum Descripta, ed. E. Pastorello (Rerum Italicarum Scriptores, vol. XII, Bologna, 1938–42). De Expugnatione Lyxbonensi, ed. and trans. C. W. David (New York, 1936).

240 bibliography Dudo of Saint-Quentin, De moribus et actis primorum Normanniae ducum, ed. J. Lair (Caen, 1865). Eustathios of Thessaloniki, The Capture of Thessaloniki, trans. J. Melville Jones (Canberra, 1988). Fabre, M. Paul (ed.), Le Liber Censuum de l’Église Romaine (Paris, 1889). Falcandus, Hugo, La historia o liber de regno Sicilie e la epistola ad Petrum Panormitane Urbis thesaurarium di Ugo Falcando, ed. G. B. Siragusa (Rome, 1904). Falco of Benevento, Chronicon Beneventanum, ed. E. D’Angelo (Florence, 1998). Fulcher of Chartres, Historia Hierosolymitana 1095–1127, ed. H. Hagenmeyer (Heidelberg, 1913). Gesta Francorum et aliorum Hierosolimitanorum, ed. and trans. R. Hill and R. A. Baskerville Mynors (London, 1962). Guillaume de la Pouille, La geste de Robert Guiscard, ed. and trans. M. Mathieu (Palermo, 1961). History of the Patriarchs of the Egyptian Church, ed. and trans. O. H. E. Khs-Burmester et al. 4 vols (Cairo, 1942–74). Hystoria de via et recuperatione Antiochiae atque Ierusolymarum, ed. E. d’Angelo (Tavarnuzze, 2009). Ibn abi Dinar, Biblioteca Arabo-Sicula, ed. M. Amari (2 vols., Turin, 1880–1881), II, pp. 273–97. Ibn al-Athir, The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Ta’rikh, trans. D. S. Richards (3 pts, Aldershot, 2006–8). Ibn al-Khatib, Lisan al-Din (ed. E. Lévi-Provençal), Kitab ‘Amal al-‘Alam. Histoire de l’Espagne Musulmane (Rabat, 1934). Ibn al-Qalānisī, The Damascus Chronicle of the Crusades, trans. H. A. R. Gibb (London, 1932). Ibn Jubayr, The Travels of Ibn Jubayr, trans. R. Broadhurst (London, 1952). ʿImād al-Dīn al-Isfahānī, Conquête de la Syrie et de la Palestine par Saladin, trans. H. Massé (Paris, 1972). Itinerarium Peregrinorum et Gesta Regis Ricardi in Chronicles and Memorials of the Reign of Richard I, ed. W. Stubbs (Rolls Series, Rerum Britannicarum Medii Aevi Scriptores, 38, London, 1864). Judhari, Abu ‘Ali Mansur al-, Vie de l’Ustadh Jaudhar [Sirat al-ustadh Judhar], trans. M. Canard (Algiers, 1958). Kinnamos, John, Deeds of John and Manuel Comnenus, trans. C. M. Brand (New York, 1976). Komnene, Anna, The Alexiad, trans. E. R. A. Sewter, revised by P. Frankopan (London, 2009). Leo VI, The Taktika of Leo VI: Text, Translation, and Commentary, trans. George T. Dennis (Washington DC, 2010). Loud, Graham A. (ed. and trans.), Roger II and the Making of the Kingdom of Sicily (Manchester, 2012). Malaterra, Gaufredo, De rebus gestis Rogerii Calanroae et Siciliae Comitis et Roberti Guiscardi ducis fratris eius, ed. E. Pontieri (Bologna, 1928). Malaterra, Geoffrey, The Deeds of Count Roger of Calabria and Sicily and of his Brother Duke Robert Guiscard, trans. K. Baxter Wolf (Ann Arbor MI, 2005).



bibliography 241

Malaterra, Geoffroi, Histoire du Grand Comte Roger et de son frère Robert Guiscard, ed. and trans. M.-A. Lucas-Avenel (Caen, 2016). Maqqari, Ahmad Ibn Muhammad al- (ed. R. Dozy, G. Dugat, L. Krehl and W. Wright), Analectes sur l’Histoire et la Litterature des Arabes d’Espagne, vol. 2 (Leiden 1858–61). Ménager, Léon-Robert (ed.), Recueil des actes des ducs normands d’Italie, 1046–1127, I, Les premiers ducs, 1046–1087 (Bari, 1981). Morkinskinna: The Earliest Icelandic Chronicle of the Norwegian Kings 1030–1157, trans. T. M. Andersson and K. E. Gade (London, 2000). Odo of Deuil, De Profectione Ludovici VII in Orientam, ed. and trans. V. G. Berry (New York, 1948). Orderic Vitalis, The Ecclesiastical History, ed. M. Chibnall, 6 vols. (Oxford, 1969– 80). Otto of Freising, Gesta Frederici: Ottonis et Rahewini Gesta Friderici I. Imperatoris, ed. G. Waitz (Monumenta Germaniae Historica, Scriptores Rerum Germanicarum, Hanover-Leipzig, 1912). Otto of St Blasien, Ottonis de Sancto Blasio Chronica, ed. A. Hofmeister (Monumenta Germaniae Historica, Scriptores Rerum Germanicarum, Hanover-Leipzig, 1912). Paolo Diacono, Storia dei longobardi. Testo latino a fronte, trans. Antonio Zanella (Cividale del Friuli, 1991). Prudentius, Prudentius, trans. H. J. Thomson, 2 vols. (Cambridge MA, 1949). Radulphus Cadomensis Tancredus, ed. E. d’Angelo (Turnhout, 2011). Ralph of Caen, Tancredus, ed. E. D’Angelo (Turnhout, 2011). Ralph of Caen, The Gesta Tancredi of Ralph of Caen: A History of the Normans on the First Crusade, trans. B. S. Bachrach and D. S. Bachrach (Aldershot, 2005). Raymond of Aguilers, “Historia Francorum qui ceperunt Iherusalem”, in Recueil des historiens des croisades: Historiens occidentaux tome troisième (Paris, 1866), 231–310. Regesta pontificum Romanorum ab condita ecclesia ad annum post Christum natum MCXCVIII, ed. P. Jaffé, S. Loewenfeld, F. Kaltenbrunner and P. Ewald (Leipzig, 1885–88). Romuald of Salerno, Romualdi Salernitani Chronicon, ed. C. A. Garufi in Rerum Italicarum Scriptores, Volume VII.1 (Città di Castello, 1935). Skylitzes, John, A Synopsis of Byzantine History, 811–1057, trans. J. Wortley (Cambridge, 2010). Sturlson, Snorri, Heimskringla: History of the Kings of Norway, trans. L. M. Hollander (Austin TX, 1964). Tanukhi, Muhassin Ibn ‘Ali al-, ed. and trans. D. S. Margoliouth in The Table-Talk of a Mesopotamian Judge (London, 1922). Tarsusi, Murda Ibn ‘Ali Murda al- [Mardi Ibn ‘Ali Mardi al-], ed. and trans. C. Cahen in “Un traité d’armurerie composé pour Saladin”, Bulletin d’Etudes Orientales 12 (1947–48) 103–260. Three Byzantine Military Treatises, ed. and trans. G. Dennis (Washington DC, 1985). Vie et miracles du pape S. Léon IX, ed. A. Poncelet in Analecta Bollandiana, vol. 25 (Brussels, 1906), pp. 258–97. Walter the Chancellor, The Antiochene Wars, trans. T. S. Asbridge and S. B. Edgington (Aldershot, 1999).

242 bibliography Willelmi Tyrensis archiepiscopi chronicon, ed. R. Burchard and C. Huygens (Turnhout, 1986). William of Malmesbury, Gesta regum Anglorum, ed. R. A. B. Mynors, R. M. Thomson and M. Winterbottom, 2 vols. (Oxford, 1998). William of Tyre, A History of Deeds Done Beyond the Sea, trans. E. A. Babcock and A. C. Krey. 2 vols. (New York, 1976). William of Tyre, Cronique, ed. R. B. C. Huygens (London, 1986). William of Tyre, Historia rerum in partibus transmarinis gestarum, ed. R. Huygens, H. Mayer and R. Gerhard (2 vols., Corpus Christianorum Continuatio Mediaevalis, Turnholt, 1986). William of Tyre, La continuation de Guillaume de Tyr (1184–1197), ed. M. Morgan (Paris, 1982).

Secondary Sources Abels, Richard P. and Bachrach, Bernard S. (eds), The Normans and their Adversaries at War (Woodbridge, 2001). Ahmad, Ahmed, A History of Islamic Sicily (Edinburgh, 1975). Ahsan, M. M., Social Life under the Abbasids 176–289 AH, 786–902 AD (London, 1979). Alban. J. R. and Allmand, Christopher, “Spies and Spying in the Fourteenth Century”, in War, Literature, and Politics in the Late Middle Ages, ed. Christopher Allmand (Liverpool, 1976), pp. 73–101. Albu, Emily, “Bohemond and the Rooster: Byzantines, Normans, and the Artful Ruse”, in Anna Komnene and Her Times, ed. T. Gouma-Peterson (Abingdon, 2014), pp. 157–68. Albu, Emily, The Normans in Their Histories: Propaganda, Myth and Subversion (Woodbridge, 2001). Amari, Michelle, Storia dei Musulmani di Sicilia (Catania, 1933). Angold, Michael, “Belle Époque or Crisis? (1025–1118)”, in The Cambridge History of the Byzantine Empire, c. 500–1492, ed. J. Shepard (Cambridge, 2008), pp. 583–626. Asbridge, Thomas S., The Creation of the Principality of Antioch 1098–1130 (Woodbridge, 2000). Bachrach, Bernard S., “Early Medieval Military Demography: Some Observations on the Methods of Hans Delbrück”, in The Circle of War in the Middle Ages, ed. D. Kagay and L. J. Andrew Villalon (Woodbridge, 1999), pp. 3–20. Bachrach, Bernard S., “Logistics in Pre-Crusade Europe”, in Feeding Mars: Logistics in Western Warfare from the Middle Ages to the Present, ed. J. A. Lynn (Boulder CO, 1993), pp. 57–78. Bachrach, Bernard S., “The Feigned Retreat at Hastings”, Mediaeval Studies 33 (1971), 344–47. Bates, David, The Normans and Empire (Oxford, 2013). Bates, David, William the Conqueror (New Haven CT, 2016). Bennett, Matthew, “Norman Conquests: A Strategy for World Domination?” Journal of Medieval Military History 15 (2017), 91–102.



bibliography 243

Beshir, B. J. (Beshir Ibrahim Beshir), “Fatimid Military Organization”, Der Islam 55 (1978) 37–57. Beshir, B. J. (Beshir Ibrahim Beshir), “The Fatimid Caliphate 386/996–487/1094” (PhD thesis, London University, 1970). Birk, Joshua C., “The Betrayal of Antioch: Narratives of Conversion and Conquest during the First Crusade”, Journal of Medieval and Early Modern Studies 41 (2011), 463–86. Birkenmeier, John W., The Development of the Komnenian Army 1081–1180 (Leiden, 2002). Bisanti, Armando, “Composizione, stile e tendenze dei Gesta Roberti Wiscardi di Guglielmo il Pugliese”, Archivio Normanno-Svevo 1 (2009), 87–132. Bliese, John R. E., “The Courage of the Normans – A Comparative Study of Battle Rhetoric”, Nottingham Medieval Studies 35 (1991), 1–26. Blöndal, Sigfus, The Varangians of Byzantium: An Aspect of Byzantine Military History, trans. B. S. Benedikz (Cambridge, 1978). Bosworth, Clifford E., “Abu ‘Amr ‘Uthman al-Tarsusi’s Siyar al Thughur and the Last Years of Arab Rule in Tarsus (Fourth/Tenth Century)”, Graeco-Arabica 5 (1993) 183–95. Bosworth, Clifford E., “Recruitment, Muster and Review in Medieval Islamic Armies”, in War, Technology and Society in the Middle East, ed. V. J. Parry and M. E. Yapp (London, 1975), pp. 59–77. Bottiglieri, Corinna, “Etnonimi e senso di identità nella Storia dei Normanni di Amato di Montecassino: alcune osservazioni”, in Sprache und Identität im frühen Mittelalter, ed. Walter Pohl and Bernhard Zeller (Vienna, 2012), pp. 219–337. Brands, Hal, What Good is Grand Strategy? Power and Purpose in American Statecraft from Harry S. Truman to George W. Bush (Ithaca NY, 2014). Bresc, Henri, “Mudejars des Pays de la Couronne d’Aragon et Sarrasin de la Sicile Normande: Le Problème d’Acculturation”, in ed. anon., X Congreso de Historia de la Corona de Aragon; Jaime I y su Epoca (Zaragoza 1975), vol. III (Zaragoza, 1980), pp. 51–60. Brett, Michael, “Fitnat al Qayrawan. A Study of Traditional Arabic Historiography” (PhD thesis, London University, 1970). Brett, Michael, The Rise of the Fatimids: The World of the Mediterranean and the Middle East in the Fourth Century of the Hijra, Tenth Century CE (Leiden, 2001). Brown, Paul, “The Gesta Roberti Wiscardi: A ‘Byzantine’ history?” Journal of Medieval History 37 (2011), 162–179. Brown, Paul, Mercenaries to Conquerors: Norman Warfare in the Eleventh- and Twelfth-Century Mediterranean (Barnsley, 2016). Brown, R. Allen, “The Battle of Hastings”, in Anglo-Norman Warfare: Studies in Late Anglo-Saxon and Anglo-Norman Military Organization and Warfare, ed. M. Strickland (Woodbridge, 1992), pp. 161–81. Buck, Andrew D., The Principality of Antioch and its Frontiers in the Twelfth Century (Woodbridge, 2017). Bünemann, Richard, Robert Guiskard, 1015–1085. Ein Normanne erobert Süditalien (Köln, 1997). Bysted, Ane L., The Crusade Indulgence: Spiritual Rewards and the Theology of the Crusades, c. 1095–1216 (Leiden, 2014).

244 bibliography Cahèn, Claude, “L’Administration Financière de l’armée Fatimide d’après al-Makhzumi”, Journal of the Economic and Social History of the Orient 15 (1972), 163–82. Cahèn, Claude, La Syrie du Nord à l’époque des croisades et la principauté franque d’Antioche (Paris, 1940). Canard, Marius, “L’autobiographie d’un chamberllan du Mahdi ‘Obeidallah le Fatimide”, Hesperis 32 (1952) 279–329. Canard, Marius, “L’Expansion Arabe: le problème militaire”, in ed. anon., Settimane di Studio del Centro Italiano di Studi sull’alto Medioèvo XII (Spoleto 1965) 37–63; also in M. Canard, L’ Expansion Arabo-Islamique et ses répercussions (London, 1974). Chevedden, Paul, “A Crusade from the First: The Norman conquest of Islamic Sicily, 1060–1091”, Al-Masaq 22 (2010), 191–225. Chevedden, Paul, “Canon 2 of the Council of Clermont (1095) and the Crusade Indulgence”, Annuarium Historiae Conciliorum 37 (2005), 253–322. Chevedden, Paul, “Pope Urban II and the Ideology of the Crusades”, in The Crusader World, ed. A. Boas (Abingdon, 2015), pp. 7–53. Chevedden, Paul, “The Islamic View and the Christian View of the Crusades: A New Synthesis”, History 93 (2008), 181–200. Chiarelli, L. C., “Sicily during the Fatimid Age” (PhD thesis, University of Utah, 1986). Christie, Niall, “Cosmopolitan Trade Centre or Bone of Contention? Alexandria and the Crusades, 487–857/1095–1453”, Al-Masaq 26 (2014), 49–61. Cilento, N., “I Saraceni nell’Italia Meridionale nei Secoli IX e X”, Archivio Storico per la Province Napoletane, n.s. 38 (1959), 109–22. Citarella, Armand O., “The Relations of Amalfi with the Arab World before the Crusades”, Speculum 42 (1967) 299–312. Clausewitz, Carl von, On War, ed. and trans. M. Howard and P. Paret (Princeton NJ, 1976). Cowdrey, H. E. J., “Pope Gregory VII’s ‘Crusading’ Plans of 1074”, in Outremer: Studies in the History of the Crusading Kingdom of Jerusalem Presented to Joshua Prawer, ed. B. Z. Kedar, H. E. Meyer and R. C. Smail. (Jerusalem, 1982), pp. 27–40. Cowdrey, H. E. J., The Age of Abbot Desiderius: Montecassino, the Papacy, and the Normans in the Elevenths and Early Twelfth Centuries (Oxford, 1983). Cowell, Andrew, The Medieval Warrior Aristocracy (Cambridge, 2007). Dadoyan, Seta B., The Fatimid Armenians: Culture and Political Interaction in the Middle East (Leiden, 1997). Davis, R. H. C., The Normans and their Myth (London, 1976). Delbrück, Hans, Numbers in History (London, 1913), pp. 11–23. Douglas, David Charles, The Norman Achievement 1050–1100 (Berkeley CA, 1969). Doyle, Peter and Bennett, Matthew R., “Terrain in Military History: An Introduction”, in Fields of Battle, Terrain in Military History, ed. P. Doyle and M. R. Bennett (London, 2002), pp. 1–7. Eddé, Anne-Marie, Saladin, trans. J. M. Todd (Cambridge MA, 2011). Edgington, Susan B., “Espionage and Military Intelligence during the First Crusade, 1095–99”, in Crusading and Warfare in the Middle Ages: Realities and Representations. Essays in Honour of John France, ed. S. John and N. Morton (Farnham, 2014), pp. 75–86.



bibliography 245

Ehrenkreutz, Andrew S., Saladin (Albany NY, 1972). Erdmann, Carl, The Origin of the Idea of Crusade, trans by M. W. Baldwin and W. Goffart (Princeton NJ, 1977). Fliche, Augustin, “Les origins de l’action de la papauté en vue de la croisade”, Revue d’histoire ecclésiastique 34 (1938), 765–75. Flori, Jean, Bohémond d’Antioche: chevalier d’aventure (Paris, 2007). Flori, Jean, Chevaliers et chevalerie au Moyen Age. Vie Quotodienne (Paris, 1998). France, John, “The Occasion of the Coming of the Normans to Southern Italy”, in Latin Expansion in the Medieval Western Mediterranean, ed. E. A. Congdon (Farnham, 2013), pp. 89–207; first published in Journal of Medieval History 17 (1991), 185–205. France, John, Victory in the East: A Military History of the First Crusade (Cambridge, 1994). France, John, Western Warfare in the Age of the Crusades, 1000–1300 (London, 1999). France, John, “The Normans and Crusading”, in The Normans and Their Adversaries at War: Essays in Memory of C. Warren Hollister, ed. R. P. Abels and B. S. Bachrach (Woodbridge, 2001), pp. pages. Frankopan, Peter, “Turning Latin into Greek: Anna Komnene and the Gesta Roberti Wiscardi”, Journal of Medieval History 39 (2013), 80–99. Freedman, Lawrence, Strategy: A History (Oxford, 2013). Fulton, Michael S., Artillery in the Era of the Crusades: Siege Warfare and the Development of Trebuchet Technology (Leiden, 2018). Gabrieli, Francesco, “Gli Arabi in Spagna e in Italia”, in ed. anon., Ordinamenti Militari in Occidente nell’alto Medioèvo, Settimane di Studio del Centro Italiano di Studi sull’alto Medioèvo XV/2 (Spoleto 1968), pp. 701–20. Galdi, Amalia, “Culti e agiografie d’età normanna in Italia meridionale”, in People, Texts and Artifacts: Cultural Transmission in the Medieval Norman Worlds, ed. D. Bates, E. D’Angelo and E. van Houts (London, 2018), pp. 89–104. Gooch, John, “History and the Nature of Strategy”, in The Past as Prologue: The Importance of History to the Military Profession, ed. W. Murray and R. H. Sinnreich (Cambridge, 2006), pp. 133–50. Guzzo, Cristian, L’arrivo dei Normanni nel Meridione d’Italia tra fonti d’epoca e storiografia contemporanea (Tuscania, 2014). Guzzo, Cristian, L’Esercito Normanno nel Meridione d’Italia: Battaglie, Assedi ed Armamenti del Cavalieri del Nord (1016–1194) (Brindisi, 2013). Haldon, John F., “Approaches to an Alternative Military History of the Period ca. 1025–1071”, in The Empire in Crisis(?): Byzantium in the 11th century (1025–71), ed. V. N. Vlyssidou (Athens, 2003), pp. 45–74. Haldon, John F., “The Organisation and Support of an Expeditionary Force: Manpower and Logistics in the Middle Byzantine Period”, in Byzantium at War (9th–12th c.), ed. N. Oikonomides (Athens, 1997), pp. 111–50. Halsall, Guy, Warfare and Society in the Barbarian West, 450–900 (London, 2003). Harari, Yuval Noah, “Knowledge, Power and the Medieval Soldier”, in “In Laudem Hierosolymitani”: Studies in Crusades and Medieval Culture in Honour of Benjamin Z. Kedar, ed. I. Shagrir, R. Ellenblum, and J. Riley-Smith (Aldershot, 2007), pp. 345–56.

246 bibliography Harari, Yuval Noah, Special Operations in the Age of Chivalry 1100–1550 (Woodbridge, 2007). Harmon, Russel S., Dillon III, Francis H. and Garver Jr, John B., “Perspectives on Military Geography”, in Studies in Military Geography and Geology, ed. D. R. Caldwell, J. Ehlen, and R. S. Harmon (London, 2004), pp. 7–20. Heygate, Catherine, “Marriage Strategies Among the Normans of Southern Italy in the Eleventh Century”, in Norman Expansion: Connections, Continuities and Contrasts, ed. K. J. Stringer and A. Jotischky (London, 2013), pp. 165–86. Hillenbrand, Carole, The Crusades: Islamic Perspectives (Edinburgh, 1999). Hoffmann, Hartmut, “Die Anfänge der Normannen in Süditalien”, Quellen und Forschungen 49 (1969), 95–144. Hosler, John D., “Reframing the Conversation on Medieval Military Strategy”, The Journal of Medieval Military History 16 (2018), 189–206. Houben, Hubert, Die Abtei Venosa und das Mönchtum im normannisch-staufischen Süditalien (Tübingen, 1995). Houben, Hubert, Roger II of Sicily: A Ruler Between East and West (Cambridge, 2002). Jacoby, David, “The Supply of War Materials to Egypt in the Crusader Period”, Jerusalem Studies in Arabic and Islam 25 (2001), 102–32. Jacques-Meunié, Djinn, “Le Nom Berbère d’un poignard Maghrébin au XIe siècle d’après un texte Arabe de l’Egypte”, Journal Asiatique 250 (1962), 613–18. Jahn, Wolfgang, Untersuchungen zur normannischen Herrschaft in Süditalien, 1040– 1100 (Frankfurt am Main, 1989). Jamison, Evelyn, “Some Notes on the ‘Anonymi Gesta Francorum’, with Special Reference to the Norman Contingent from South Italy and Sicily in the First Crusade”, in Studies in French Language and Medieval Literature Presented to Professor Mildred K. Pope (Manchester, 1939), pp. 183–208. Johns, Jeremy, Arabic Administration in Norman Sicily. The Royal Diwan (Cambridge, 2002). Johnson, Ewan, “Normandy and Norman Identity in Southern Italian Chronicles”, Anglo-Norman Studies 27 (2004), 85–100. Johnson, Ewan, “Origin Myths and the Construction of Medieval Identities: Norman Chronicles 1000–1100”, in Texts and Identities in the Early Middle Ages, ed. R. Corradini, R. Meens, C. Pössel and P. Shaw (Wien, 2006), pp. 153–64. Kagan, Kimberly, The Eye of Command (Ann Arbor, 2006). Kaldellis, Anthony, Streams of Gold, Rivers of Blood: The Rise and Fall of Byzantium, 955 A.D. to the First Crusade (Oxford, 2017). Kaldellis, Anthony, Streams of Gold, Rivers of Blood: The Rise and Fall of Byzantium, 955 A.D. to the First Crusade (New York, 2017). Kreutz, Barbara M., Before the Normans: Southern Italy in the Ninth and Tenth Centuries (Philadelphia PA, 1991). Latham, Andrew, Theorizing Medieval Geopolitics: War and World Order in the Age of the Crusades (New York, 2012). Lev, Yaacov, “The Fatimid Army, A.H. 358–427/968–1036 C.E.: Military and Social Aspects”, Asian and African Studies 14 (1980), 165–92. Levine, Robert, “The Pious Traitor: Rhetorical Reinventions of the Fall of Antioch”, Mittellateinisches Jahrbuch 33 (1998), 59–80.



bibliography 247

Loud, Graham, “Anna Komnena and Her Sources for the Normans of Southern Italy”, in Church and Chronicle in the Middle Ages: Essays Presented to John Taylor, ed. I. Wood and G. A. Loud (London, 1991), pp. 41–58. Loud, Graham, “Betrachtungen über die normannische Eroberung Süditaliens”, in Forschungen zur Reichs-, Papst- und Landesgeschichte: Peter Herde zum 65. Geburtstag von Freunden, Schülern und Kollegen dargebracht, ed. K. Borchardt and E. Bünz (Stuttgart, 1998), pp. 115–31, cf. 123–24, reprinted in Conquerors and Churchmen in Southern Italy (Aldershot, 1999). Loud, Graham, “The Gens Normannorum: Myth or Reality?”, Anglo-Norman Studies 4 (1982), 104–16. Loud, Graham, The Age of Robert Guiscard, Southern Italy and the Norman Conquest (London, 2000). Loud, Graham, and Metcalfe, Alex (eds), The Society of Norman Italy (Leiden, 2002). Lucas-Avenel, Marie-Agnès, “Le recit de Geoffroi Malaterra ou la légitimation de Roger, Grand Comte de Sicile”, Anglo-Norman Studies 24 (2011), 169–92. Lyons, Malcolm C. and D. E. P. Jackso, Saladin: The Politics of the Holy War (Cambridge, 1982. Manselli, Raoul, Italia e Italiani alla prima crociata (Rome, 1983). Matthew, Donald, The Norman Kingdom of Sicily (Cambridge, 1992). Mayer, Hans E. (ed.), Die Urkunden der lateinischen Könige von Jerusalem, 4 vols. (Hanover, 2010). Ménager, Léon-Robert, “Inventaire des familles normandes et franques émigrées en Italie méridionale et en Sicile (XIe–XIIe siècles)”, in Hommes et institutions de l’Italie normande, ed. L.-R. Ménager (London, 1981), pp. 260–390. Metcalfe, Alex, Muslims and Christians in Norman Sicily: Arabic Speakers and the End of Islam (London, 2003). Morillo, Stephen, “Expecting Cowardice: Medieval Battle Tactics Reconsidered”, Journal of Medieval Military History 4 (2006), 65–73. Morillo, Stephen, “Hastings: An Unusual Battle”, in The Battle of Hastings: Sources and Interpretations, ed. S. Morillo (Woodbridge, 1996), pp. 219–28. Morillo, Stephen, Warfare Under the Anglo-Norman Kings, 1066–1135 (Woodbridge, 1994). Morton, Nicholas, Encountering Islam on the First Crusade (Cambridge, 2017). Murray, Alan V., “How Norman was the Principality of Antioch? Prolegomena to a Study of the Origins of the Nobility of a Crusader State”, in Family Trees and the Roots of Politics: The Prosopography of Britain and France from the Tenth to the Twelfth Century, ed. K. S. B. Keats-Rohan (Woodbridge, 1997), pp. 349–59. Musca, Giosue, L’Emirato di Bari 847–871 (Bari, 1992). Nicolle, David, “Jawshan, Cuirie and Coat-of-Plates: An Alternative Line of Development for Hardened Leather Armour”, in A Companion to Medieval Arms and Armour, ed. D. Nicolle (Woodbridge, 2002), pp. 179–221. Nicolle, David, “Medieval Swords of Morocco and al-Andalus: History, Construction and Decoration”, Graeco-Arabica 7–8 (1999–2000), 413–20. Nicolle, David, “Talismanic Swords from the 12th Century Maghrib”, GraecoArabica 7–8 (1999–2000), 421–32. Nicolle, David, “Two Swords from the Foundation of Gibraltar”, Gladius 22 (2002), 147–200.

248 bibliography O’Bweng-Okwess, Kizobo, “Le Recrutement des Soldats Négro-Africains par les Musulmans du VIIIe au XIIe Siècle”, Journal of Oriental and African Studies 1 (1989), 24–29. Paret, Peter (ed.), Makers of Modern Strategy from Machiavelli to the Nuclear Age (Princeton NJ, 1986). Paul, Nicholas L., “A Warlord’s Wisdom: Literacy and Propaganda at the Time of the First Crusade”, Speculum 85 (2010), 534–66. Prestwich, John O., “Military Intelligence under the Norman and Angevin Kings”, in Law and Government in Medieval England and Normandy: Essays in Honour of Sir James Holt, ed. G. Garnett and J. Hudson (Cambridge, 1994), pp. 1–30. Prestwich, Michael, Armies and Warfare in the Middle Ages: The English Experience (London, 2006). Pryor, John H., “From Dromon to Galea: Mediterranean bireme galleys AD 500–1300”, in The Age of the Galley, ed. R. Gardiner (London, 1995), pp. 101–16. Pryor, John H. (ed.), Logistics of Warfare in the Age of the Crusades: Proceedings of a Workshop Held at the Centre for Medieval Studies, University of Sydney, 30 September to 4 October 2002 (Aldershot, 2006). Reuter, Timothy, “The Recruitment of Armies in the Early Middle Ages: What Can We Know?”, in Military Aspects of Scandinavian Society in a European Perspective, A.D. 1–1300, ed. A. N. Jorgensen and B. L. Clausen (Copenhagen, 1997), pp. 32–37. Riley-Smith, Jonathan, The First Crusaders, 1095–1131 (Cambridge, 1997). Riley-Smith, Jonathan, The Knights Hospitaller in the Levant, c.1070–1309 (Basingstoke, 2012). Rogers, Randall, Latin Siege Warfare in the Twelfth Century (Oxford, 1992). Röhricht, Reinhold (ed.), Regesta Regni Hierosolymitani, Additamentum (Innsbruck, 1904). Rossenbeck, Klaus, Die Stellung der Riddarassogur in der altnordischen Prosaliteratur: Eine Untersuchung an hand des Erzählstils (Bamberg, 1970). Rubenstein, Jay, “The Deeds of Bohemond: Reform, Propaganda, and the History of the First Crusade”, Viator 47 (2016), 113–35. Rubenstein, Jay, Nebuchadnezzar’s Dream: The Crusades, Apocalyptic Prophecy, and the End of History (Oxford, 2018). Runciman, Steven, A History of the Crusades (Cambridge, 1951–54). Runciman, Steven, The First Crusade (Cambridge, 1980). Russo, Luigi, “Oblio e memoria di Boemondo d’Altavilla nella storiografia normanna”, Bullettino dell’istituto storico per il medio evo 106 (2004), 137–65. Russo, Luigi, Boemondo. Figlio del Guiscardo e principe di Antiochia (Avellino, 2009). Sarraf, S. al-, “Close Combat Weapons in the Early ‘Abbasid Period: Maces, Axes and Swords”, in Companion to Medieval Arms and Armour, ed. D. Nicolle (Woodbridge, 2002) 149–78, figs. XII 1–123. Sarraf, S. al-, “Mamluk Furusiyah Literature and its Antecedents”, Mamluk Studies Review 8 (2004), 141–200. Slitt, Rebecca L., “Justifying Cross-Cultural Friendship: Bohemond, Firuz, and the Fall of Antioch”, Viator 38 (2007), 339–49. Smith, R. Upsher Jr, “‘Nobilissimus’ and Warleader: The Opportunity and the



bibliography 249

Necessity Behind Robert Guiscard’s Balkan Expeditions”, Byzantion 70 (2000), 507–26. Stanton, Charles D., “Anonymous Vaticanus: Another Source for the Normans of the South?” The Haskins Society Journal 24 (2012), 79–93. Stanton, Charles D., Medieval Maritime Warfare (Barnsley, 2015). Stanton, Charles D., Norman Naval Operations in the Mediterranean (Woodbridge, 2011). Strachan, Hew, The Direction of War: Contemporary Strategy in Historical Perspective (Cambridge, 2012). Talbi, M., L’Emirat Aghlabide 184–296, 800–909. Histoire Politique (Paris 1966). Taviani-Carozzi, Huguette, “Le mythe des origins de la conquête normande en Italie”, in Cavalieri alla conquista del Sud: Studi sull’Italia normanna in memoria di Léon-Robert Ménager, ed. E. Cuozzo and J.-M. Martin (Roma, 1998), pp. 57–89. Taviani-Carozzi, Huguette, La principauté lombarde de Salerne (9e–11e siècle): pouvoir et société en Italie lombarde méridionale (Rome, 1991). Taviani-Carozzi, Huguette, La Terreur du monde, Robert Guiscard et la conquête normande en Italie (Paris, 1996). Theotokis, Georgios, “Bohemond of Taranto’s 1107–8 Campaign in Byzantine Illyria – Can It Be Viewed as a Crusade?” Rosetta 11 (2012), 72–81. Theotokis, Georgios, “Geoffrey Malaterra as a Military Historian for the Norman Expansion in Italy and Sicily – Strengths and Weaknesses in His Narrative”, Mediterranean Chronicle 2 (2012), 105–16. Theotokis, Georgios, “Rus, Varangian and Frankish Mercenaries in the Service of the Byzantine Emperors (9th–11th c.) – Numbers, Organisation and Battle Tactics in the Operational Theatres of Asia Minor and the Balkans”, Byzantina Symmeikta 22 (2012), 126–56. Theotokis, Georgios, “The Norman Invasion of Sicily, 1061–1072: Numbers and Military Tactics”, War in History 17 (2010), 381–402. Theotokis, Georgios, The Norman Campaigns in the Balkans 1081–1108 AD (Woodbridge, 2014). Thomas, Aurélie, “La carrière matrimoniale des fils de Tancrède de Hauteville en Italie méridionale: rivalités fraternelles et stratégies concurrentes”, in Les stratégies matrimoniales (IXe–XIIIe siècle), ed. M. Aurell (Turnhout, 2013), pp. 89–100. Tounta, Eleni, “The Perception of Difference and the Differences of Perception: The Image of the Norman Invaders of Southern Italy in Contemporary Western Medieval and Byzantine Sources”, Byzantina Symmeikta 20 (2010), 111–42. Treadgold, Warren, “Standardized Numbers in the Byzantine Army”, War in History 12 (2005), 1–14. Van Houts, Elisabeth, “Quelques observations sur des liens entre la Normandie, l’Angleterre et l’Italie au début du XIe siècle”, in 911—2011, Penser les mondes normands médiévaux. Actes du colloque international de Caen et Cerisy (29 septembre – 2 octobre 2011), ed. D. Bates and P. Bauduin (Caen, 2016), pp. 129–46. Verbruggen, J. F., The Art of Warfare in Western Europe During the Middle Ages: From the Eighth Century to 1340 (Woodbridge, 1997). Walter, Bastian, “Urban Espionage and Counterespionage during the Burgundian Wars (1468–1477)”, Journal of Medieval Military History 9 (2011), 132–45.

250 bibliography Ward, James P., “Security and Insecurity, Spies and Informers in Holland during the Guelders War (1506–1515)”, Journal of Medieval Military History 10 (2012), 173–96. Webber, Nick, The Evolution of Norman Identity 911–1154 (Woodbridge, 2005). Whately, Conor, Battles and Generals: Combat, Culture, and Didacticism in Procopius’ Wars (Leiden, 2016). Wieruszowski, Helene, “The Norman Kingdom of Sicily and the Crusades”, in A History of the Crusades: Volume II, the Later Crusades, 1189–1311, ed. R. L. Wolff and H. W. Hazard (Madison, 1969). Wolf, Kenneth B., Making History: The Normans and Their Historians in EleventhCentury Italy (Philadelphia, 1995). Yewdale, Ralph Bailey, Bohemond I: Prince of Antioch (Amsterdam, 1970).

Index Albanian coast  161 Alexandria  6, 189, 225, 229, 230–38 Alexios I Comnenus, Byzantine emperor 57, 134, 142–48, 151–60, 165–68, 171, 194, 220 Alice of Jerusalem  207 al–Qādī al–Fādil  229, 236 Amalfi  15, 25, 28 (n. 80), 85, 87, 178, 181, 199 (n. 15), 204, 219 Amalric of Jerusalem  6, 189, 227–34 amiratus 186 amiratus amiratorum [“admiral of admirals” – George of Antioch]  181 Antioch  55, 56 (n. 2), 57, 64–65, 143, 148, 168, 200–7, 220–26 Principality of  6, 207, 221, 226 Apulia  1, 13–23, 30, 45 (n. 39), 84, 86, 93, 139, 141, 143, 181, 187, 222 Duchy of  6, 11, 40, Aubré of Cagnano  201, 203 Bari  13, 18–19, 22–23, 24–26, 32, 41, 84, 86, 141, 181, 195, 197–99 amirate of  86 Nicolaian Museum of  195 Bari Annals  13, 18–19, 22–23 Boel of Chartres  201, 201 (n. 20), 203, (n. 28), 204 (n. 36) Bohemond of Taranto  3, 5–6, 46, 49, 55–58, 64–65, 143–48, 168, 195–209, 196 (n. 3), 198 (n. 7), 199 (n. 15), 213, 219–23, 226 Boioannes, Basil  14–15, 17–18, 18 (n. 35) Bribery  3, 63–5, 71–73 Cairo  225, 227–30, 232–33 Ayyubid ruler of  189 Fatimid Caliph of  94, 182, Cannae, Battle of  14–17 Cerami, Battle of  22–28, 48–50, 99 Chartres 205 Civitate Battle of  (1053), 19, 21, 26, 29, 139–40, 218, 218 (n. 29) castle of  15 Constance of France  205, 207 Constance of Sicily  191 crossbow  104–5, 136, 145

Damietta 228–37 Domenico Selvo  151–52, 158, 162–63, 165, 172 Drogo Hauteville  17–20, 46–47, 50–52, 138 dromōn 185 Dudo of Saint Quentin  66–67 Dyrrachium Battle of (1081)  137, 140, 144 siege of (1081–2)  142 siege of (1107–8)  143 Edessa  61, 206, 206 (n. 42) Egypt  6, 68, 81, 179, 182, 189, 226–38 Ancient 104 ‘Abbasid governor of  86 Ikhshidid governors of  89 Tulunid governors of  89 Fatimid conquest of  90, 91, 94 Fatimids  6, 80, 82, 83, 86, 88, 90–98, 103, 108, 182, 189, 227–28, 230–38 feigned flight  62, 62 (n. 26), 63, 71, 146 fondaco 194 Fulcher of Chartres  64 galea  185–86, 188 Gallipoli 198 Genoa  177–78, 184, 189, 193 gens Normannorum  2, 44, 56, 57 (n. 8), 58 Geoffrey of Monte Scaglioso  208 George of Antioch  179–84, 187 Giovinazzo 198 Godfrey of Bouillon  56, 72, 198, 204, 235 Gregory VII, Pope  218, 223 hauberk  97–8, 101, 113–14, 121–22, 130, 136 Henry I of England  205 Henry II, Holy Roman Emperor  15 Henry III, Holy Roman Emperor  139, 218 Henry VI, Holy Roman Emperor  191, 193 Herman, count of Canne  69, 201, 203, 203 (n. 30) Humphrey Fitz-Ralph  201, 203, 204 (n. 36) Humphrey of the Hautevilles  20, 45, 47, 51, 140, Humphrey of Monte Scaglioso  201, 204 (n. 32 & 36)

252 index Ifriqiyah 183 Intelligence, Military  58–59, 233 iqta‘ 94

Pisa  177, 178, 181, 184, 185, 189, 193

javelin(s)  98, 103, 123–25 jawshan 100 Jerusalem  5, 38, 208, 211, 222, 230 Kingdom of  6, 143, 191, 205, 208, 225, 226, 228, 232 capture of, (1099)  73, 202 (n. 23), 206, 234, 236 capture of, (1187)  237

Ralph II count of Molise  204 Ralph of Caen  3, 56, 64, 71–75, 203, 220, 222 Raymond of Aguilers  56 Raymond of Poitiers  207, 207 (n. 46) Raymond of Saint-Gilles  55–56 ribat 92 Richard of Aversa  20, 140 Richard I of Capua  11–12, 25, 38, 39, 62, 218 Richard I of England  192 Richard son of Count Ranulf  201, 203 Robert Curthose  198, 202 Robert Fitz-Toustan  201, 204, 204 (n. 36) Robert of Anse  201, 204 (n. 36) Robert of Sourdeval  201–3, 204 (n. 36) Roger Borsa, duke of Apulia  12, 39, 64, 198, 202, 208, 219 Roger II, King of Sicily  6, 66, 179, 182, 184, 191, 199, 213, 223–24, 226–27

Kalbite Dynasty of Sicily  80, 91, 92, 94 fleet 24 Kekaumenus  13, 25 lamellar  100, 101–14, 118–19, 123, 126 Leo IX, Pope  19, 139, 218 Leo VI, Byzantine emperor  134, 185 Louis VII of France  187 Louis IX of France  229 Maghreb  179, 182–83, 188, 230 Mahdiyah  179, 182, 183–84, 188 Maniaces 16–17 Manuel I Comnenus, Byzantine emperor 173, 187–88, 194, 228 Melfi castle of  15, 17, 31, Norman Principality of  19, 20, Treaty of  6, 213, 216–18 Messina  17, 21, 24, 29, 60, 62, 70, 71, 92, 93, 99, 140, 177, 186, 190, 192 Michael IV, Byzantine emperor  16 Michael VII, Byzantine emperor  69, 156 Misilmeri, Battle of  22, 26

qa’id  84, 88, 94, 99

Saladin  6, 189, 226–37 Sasno, Battle of  163 Serlo  22, 24, 50, 93 Skylitzes  13, 17, 18 (n. 35), 31, 126 Tancred of Lecce  3, 6, 36, 42, 45 (n. 41), 50, 55, 61, 73, 74, 189–93, 201, 208, 220–22, 226 Taranto  86, 161, 198, 224 Tripoli  183–4, 191, 206, 237 Urban II, Pope  40–41, 201, 216

Nikephorus III Botaneiates, Byzantine emperor  156, 157, 159 Nūr al-Dīn  227, 230, 232, 236

Varangian Guard  15, 15 (n. 15), 16, 18, 66, 136, 142, 144–45 Vitale Falier  159 (n. 33), 163, 165, 171

Ofanto, Battle of  14, 17, 18 Olivento, Battle of  17, 18 Otbert of Liège  198 Otranto  19, 84, 161, 198, 208, 224

William Hauteville  19 William II of Sicily  189, 192, 228, 232, 234, 236–37 William IX of Aquitaine  207

Philip I of France  205, 207 (n. 44) Philip II of France  192

Zirids  94, 182 Zonaras  13, 25

Warfare in History

The Battle of Hastings: Sources and Interpretations, edited and introduced by Stephen Morillo Infantry Warfare in the Early Fourteenth Century: Discipline, Tactics, and Technology, Kelly DeVries The Art of Warfare in Western Europe during the Middle Ages, from the Eighth Century to 1340 (second edition), J.F. Verbruggen Knights and Peasants: The Hundred Years War in the French Countryside, Nicholas Wright Society at War: The Experience of England and France during the Hundred Years War, edited by Christopher Allmand The Circle of War in the Middle Ages: Essays on Medieval Military and Naval History, edited by Donald J. Kagay and L.J. Andrew Villalon The Anglo-Scots Wars, 1513–1550: A Military History, Gervase Phillips The Norwegian Invasion of England in 1066, Kelly DeVries The Wars of Edward III: Sources and Interpretations, edited by Clifford J. Rogers The Battle of Agincourt: Sources and Interpretations, Anne Curry War Cruel and Sharp: English Strategy under Edward III, 1327–1360, Clifford J. Rogers The Normans and their Adversaries at War: Essays in Memory of C. Warren Hollister, edited by Richard P. Abels and Bernard S. Bachrach The Battle of the Golden Spurs (Courtrai, 11 July 1302): A Contribution to the History of Flanders’ War of Liberation, 1297–1305, J.F. Verbruggen War at Sea in the Middle Ages and the Renaissance, edited by John B. Hattendorf and Richard W. Unger Swein Forkbeard’s Invasions and the Danish Conquest of England, 991–1017, Ian Howard Religion and the Conduct of War, c.300–1215, David S. Bachrach Warfare in Medieval Brabant, 1356–1406, Sergio Boffa Renaissance Military Memoirs: War, History and Identity, 1450–1600, Yuval Harari The Place of War in English History, 1066–1214, J.O. Prestwich, edited by Michael Prestwich War and the Soldier in the Fourteenth Century, Adrian R. Bell German War Planning, 1891–1914: Sources and Interpretations, Terence Zuber The Battle of Crécy, 1346, Andrew Ayton and Sir Philip Preston

The Battle of Yorktown, 1781: A Reassessment, John D. Grainger Special Operations in the Age of Chivalry, 1100–1550, Yuval Noah Harari Women, Crusading and the Holy Land in Historical Narrative, Natasha R. Hodgson The English Aristocracy at War: From the Welsh Wars of Edward I to the Battle of Bannockburn, David Simpkin The Calais Garrison: War and Military Service in England, 1436–1558, David Grummitt Renaissance France at War: Armies, Culture and Society, c.1480–1560, David Potter Bloodied Banners: Martial Display on the Medieval Battlefield, Robert W. Jones Alfred’s Wars: Sources and Interpretations of Anglo-Saxon Warfare in the Viking Age, Ryan Lavelle The Dutch Army and the Military Revolutions, 1588–1688, Olaf van Nimwegen In the Steps of the Black Prince: The Road to Poitiers, 1355–1356, Peter Hoskins Norman Naval Operations in the Mediterranean, Charles D. Stanton Shipping the Medieval Military: English Maritime Logistics in the Fourteenth Century, Craig L. Lambert Edward III and the War at Sea: The English Navy, 1327–1377, Graham Cushway The Soldier Experience in the Fourteenth Century, edited by Adrian R. Bell and Anne Curry Warfare in Tenth-Century Germany, David S. Bachrach Chivalry, Kingship and Crusade: The English Experience in the Fourteenth Century, Timothy Guard The Norman Campaigns in the Balkans, 1081–1108, Georgios Theotokis Welsh Soldiers in the Later Middle Ages, 1282–1422, Adam Chapman Merchant Crusaders in the Aegean, 1291–1352, Mike Carr Henry of Lancaster’s Expedition to Aquitaine, 1345–1346: Military Service and Professionalism in the Hundred Years War, Nicholas A. Gribit Scotland’s Second War of Independence, 1332–1357, Iain A. MacInnes Military Communities in Late Medieval England: Essays in Honour of Andrew Ayton, edited by Gary P. Baker, Craig L. Lambert and David Simpkin The Black Prince and the Grande Chevauchée of 1355, Mollie M. Madden Military Society and the Court of Chivalry in the Age of the Hundred Years War, Philip J. Caudrey